load_backend: loaded BLAS backend from /opt/homebrew/Cellar/ggml/0.25.3/libexec/libggml-blas.so ggml_metal_device_init: tensor API disabled for pre-M5 and pre-A19 devices ggml_metal_library_init: using embedded metal library ggml_metal_library_compile_all: compiled 'fa' library in 0.014 sec ggml_metal_library_compile_all: compiled 'mul_mv' library in 0.003 sec ggml_metal_library_compile_all: compiled 'mul_mm' library in 0.002 sec ggml_metal_library_compile_all: compiled 'quantize' library in 0.004 sec ggml_metal_library_compile_all: compiled 'softmax' library in 0.001 sec ggml_metal_library_compile_all: compiled 'norm' library in 0.001 sec ggml_metal_library_compile_all: compiled 'unary' library in 0.001 sec ggml_metal_library_compile_all: compiled 'binbcast' library in 0.004 sec ggml_metal_library_compile_all: compiled 'reduce' library in 0.004 sec ggml_metal_library_compile_all: compiled 'tri' library in 0.014 sec ggml_metal_library_compile_all: compiled 'ssm' library in 0.004 sec ggml_metal_library_compile_all: compiled 'wkv' library in 0.004 sec ggml_metal_library_compile_all: compiled 'gated_delta_net' library in 0.004 sec ggml_metal_library_compile_all: compiled 'solve_tri' library in 0.004 sec ggml_metal_library_compile_all: compiled 'rope' library in 0.004 sec ggml_metal_library_compile_all: compiled 'conv' library in 0.004 sec ggml_metal_library_compile_all: compiled 'upscale' library in 0.014 sec ggml_metal_library_compile_all: compiled 'argsort' library in 0.004 sec ggml_metal_library_compile_all: compiled 'pool' library in 0.003 sec ggml_metal_library_compile_all: compiled 'misc' library in 0.003 sec ggml_metal_library_compile_all: loaded 20 libraries from embedded data in 0.014 sec (max single = 0.014 sec) ggml_metal_rsets_init: creating a residency set collection (keep_alive = 180 s) ggml_metal_device_init: GPU name: MTL0 (Apple M1 Pro) ggml_metal_device_init: GPU family: MTLGPUFamilyApple7 (1007) ggml_metal_device_init: GPU family: MTLGPUFamilyMetal4 (5002) ggml_metal_device_init: simdgroup reduction = true ggml_metal_device_init: simdgroup matrix mul. = true ggml_metal_device_init: has unified memory = true ggml_metal_device_init: has bfloat = true ggml_metal_device_init: has tensor = false ggml_metal_device_init: use residency sets = true ggml_metal_device_init: use shared buffers = true ggml_metal_device_init: recommendedMaxWorkingSetSize = 26800.60 MB load_backend: loaded MTL backend from /opt/homebrew/Cellar/ggml/0.25.3/libexec/libggml-metal.so load_backend: loaded CPU backend from /opt/homebrew/Cellar/ggml/0.25.3/libexec/libggml-cpu-apple_m1.so llama_model_loader: loaded meta data with 22 key-value pairs and 291 tensors from /sha256-6a0746a1ec1aef3e7ec53868f220ff6e389f6f8ef87a01d77c96807de94ca2aa (version GGUF V3 (latest)) llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. llama_model_loader: - kv 0: general.architecture str = llama llama_model_loader: - kv 1: general.name str = Meta-Llama-3-8B-Instruct llama_model_loader: - kv 2: llama.block_count u32 = 32 llama_model_loader: - kv 3: llama.context_length u32 = 8192 llama_model_loader: - kv 4: llama.embedding_length u32 = 4096 llama_model_loader: - kv 5: llama.feed_forward_length u32 = 14336 llama_model_loader: - kv 6: llama.attention.head_count u32 = 32 llama_model_loader: - kv 7: llama.attention.head_count_kv u32 = 8 llama_model_loader: - kv 8: llama.rope.freq_base f32 = 500000.000000 llama_model_loader: - kv 9: llama.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 10: general.file_type u32 = 2 llama_model_loader: - kv 11: llama.vocab_size u32 = 128256 llama_model_loader: - kv 12: llama.rope.dimension_count u32 = 128 llama_model_loader: - kv 13: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 14: tokenizer.ggml.pre str = llama-bpe llama_model_loader: - kv 15: tokenizer.ggml.tokens arr[str,128256] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 16: tokenizer.ggml.token_type arr[i32,128256] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 17: tokenizer.ggml.merges arr[str,280147] = ["Ġ Ġ", "Ġ ĠĠĠ", "ĠĠ ĠĠ", "... llama_model_loader: - kv 18: tokenizer.ggml.bos_token_id u32 = 128000 llama_model_loader: - kv 19: tokenizer.ggml.eos_token_id u32 = 128009 llama_model_loader: - kv 20: tokenizer.chat_template str = {% set loop_messages = messages %}{% ... llama_model_loader: - kv 21: general.quantization_version u32 = 2 llama_model_loader: - type f32: 65 tensors llama_model_loader: - type q4_0: 225 tensors llama_model_loader: - type q6_K: 1 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_0 print_info: file size = 4.33 GiB (4.64 BPW) llama_prepare_model_devices: using device MTL0 (Apple M1 Pro) (unknown id) - 25558 MiB free init_tokenizer: initializing tokenizer for type 2 load: 0 unused tokens load: control token: 128255 '<|reserved_special_token_250|>' is not marked as EOG load: control token: 128253 '<|reserved_special_token_248|>' is not marked as EOG load: control token: 128251 '<|reserved_special_token_246|>' is not marked as EOG load: control token: 128249 '<|reserved_special_token_244|>' is not marked as EOG load: control token: 128248 '<|reserved_special_token_243|>' is not marked as EOG load: control token: 128247 '<|reserved_special_token_242|>' is not marked as EOG load: control token: 128245 '<|reserved_special_token_240|>' is not marked as EOG load: control token: 128244 '<|reserved_special_token_239|>' is not marked as EOG load: control token: 128242 '<|reserved_special_token_237|>' is not marked as EOG load: control token: 128241 '<|reserved_special_token_236|>' is not marked as EOG load: control token: 128240 '<|reserved_special_token_235|>' is not marked as EOG load: control token: 128237 '<|reserved_special_token_232|>' is not marked as EOG load: control token: 128235 '<|reserved_special_token_230|>' is not marked as EOG load: control token: 128232 '<|reserved_special_token_227|>' is not marked as EOG load: control token: 128231 '<|reserved_special_token_226|>' is not marked as EOG load: control token: 128226 '<|reserved_special_token_221|>' is not marked as EOG load: control token: 128224 '<|reserved_special_token_219|>' is not marked as EOG load: control token: 128223 '<|reserved_special_token_218|>' is not marked as EOG load: control token: 128221 '<|reserved_special_token_216|>' is not marked as EOG load: control token: 128220 '<|reserved_special_token_215|>' is not marked as EOG load: control token: 128218 '<|reserved_special_token_213|>' is not marked as EOG load: control token: 128216 '<|reserved_special_token_211|>' is not marked as EOG load: control token: 128215 '<|reserved_special_token_210|>' is not marked as EOG load: control token: 128214 '<|reserved_special_token_209|>' is not marked as EOG load: control token: 128213 '<|reserved_special_token_208|>' is not marked as EOG load: control token: 128212 '<|reserved_special_token_207|>' is not marked as EOG load: control token: 128210 '<|reserved_special_token_205|>' is not marked as EOG load: control token: 128208 '<|reserved_special_token_203|>' is not marked as EOG load: control token: 128207 '<|reserved_special_token_202|>' is not marked as EOG load: control token: 128206 '<|reserved_special_token_201|>' is not marked as EOG load: control token: 128205 '<|reserved_special_token_200|>' is not marked as EOG load: control token: 128204 '<|reserved_special_token_199|>' is not marked as EOG load: control token: 128201 '<|reserved_special_token_196|>' is not marked as EOG load: control token: 128199 '<|reserved_special_token_194|>' is not marked as EOG load: control token: 128194 '<|reserved_special_token_189|>' is not marked as EOG load: control token: 128192 '<|reserved_special_token_187|>' is not marked as EOG load: control token: 128191 '<|reserved_special_token_186|>' is not marked as EOG load: control token: 128188 '<|reserved_special_token_183|>' is not marked as EOG load: control token: 128187 '<|reserved_special_token_182|>' is not marked as EOG load: control token: 128185 '<|reserved_special_token_180|>' is not marked as EOG load: control token: 128184 '<|reserved_special_token_179|>' is not marked as EOG load: control token: 128182 '<|reserved_special_token_177|>' is not marked as EOG load: control token: 128181 '<|reserved_special_token_176|>' is not marked as EOG load: control token: 128180 '<|reserved_special_token_175|>' is not marked as EOG load: control token: 128175 '<|reserved_special_token_170|>' is not marked as EOG load: control token: 128174 '<|reserved_special_token_169|>' is not marked as EOG load: control token: 128173 '<|reserved_special_token_168|>' is not marked as EOG load: control token: 128172 '<|reserved_special_token_167|>' is not marked as EOG load: control token: 128171 '<|reserved_special_token_166|>' is not marked as EOG load: control token: 128170 '<|reserved_special_token_165|>' is not marked as EOG load: control token: 128169 '<|reserved_special_token_164|>' is not marked as EOG load: control token: 128166 '<|reserved_special_token_161|>' is not marked as EOG load: control token: 128164 '<|reserved_special_token_159|>' is not marked as EOG load: control token: 128163 '<|reserved_special_token_158|>' is not marked as EOG load: control token: 128157 '<|reserved_special_token_152|>' is not marked as EOG load: control token: 128156 '<|reserved_special_token_151|>' is not marked as EOG load: control token: 128154 '<|reserved_special_token_149|>' is not marked as EOG load: control token: 128153 '<|reserved_special_token_148|>' is not marked as EOG load: control token: 128151 '<|reserved_special_token_146|>' is not marked as EOG load: control token: 128149 '<|reserved_special_token_144|>' is not marked as EOG load: control token: 128148 '<|reserved_special_token_143|>' is not marked as EOG load: control token: 128147 '<|reserved_special_token_142|>' is not marked as EOG load: control token: 128144 '<|reserved_special_token_139|>' is not marked as EOG load: control token: 128141 '<|reserved_special_token_136|>' is not marked as EOG load: control token: 128139 '<|reserved_special_token_134|>' is not marked as EOG load: control token: 128138 '<|reserved_special_token_133|>' is not marked as EOG load: control token: 128137 '<|reserved_special_token_132|>' is not marked as EOG load: control token: 128130 '<|reserved_special_token_125|>' is not marked as EOG load: control token: 128127 '<|reserved_special_token_122|>' is not marked as EOG load: control token: 128125 '<|reserved_special_token_120|>' is not marked as EOG load: control token: 128124 '<|reserved_special_token_119|>' is not marked as EOG load: control token: 128123 '<|reserved_special_token_118|>' is not marked as EOG load: control token: 128122 '<|reserved_special_token_117|>' is not marked as EOG load: control token: 128121 '<|reserved_special_token_116|>' is not marked as EOG load: control token: 128120 '<|reserved_special_token_115|>' is not marked as EOG load: control token: 128119 '<|reserved_special_token_114|>' is not marked as EOG load: control token: 128118 '<|reserved_special_token_113|>' is not marked as EOG load: control token: 128117 '<|reserved_special_token_112|>' is not marked as EOG load: control token: 128116 '<|reserved_special_token_111|>' is not marked as EOG load: control token: 128113 '<|reserved_special_token_108|>' is not marked as EOG load: control token: 128112 '<|reserved_special_token_107|>' is not marked as EOG load: control token: 128111 '<|reserved_special_token_106|>' is not marked as EOG load: control token: 128110 '<|reserved_special_token_105|>' is not marked as EOG load: control token: 128108 '<|reserved_special_token_103|>' is not marked as EOG load: control token: 128107 '<|reserved_special_token_102|>' is not marked as EOG load: control token: 128104 '<|reserved_special_token_99|>' is not marked as EOG load: control token: 128103 '<|reserved_special_token_98|>' is not marked as EOG load: control token: 128102 '<|reserved_special_token_97|>' is not marked as EOG load: control token: 128101 '<|reserved_special_token_96|>' is not marked as EOG load: control token: 128100 '<|reserved_special_token_95|>' is not marked as EOG load: control token: 128097 '<|reserved_special_token_92|>' is not marked as EOG load: control token: 128094 '<|reserved_special_token_89|>' is not marked as EOG load: control token: 128093 '<|reserved_special_token_88|>' is not marked as EOG load: control token: 128091 '<|reserved_special_token_86|>' is not marked as EOG load: control token: 128090 '<|reserved_special_token_85|>' is not marked as EOG load: control token: 128087 '<|reserved_special_token_82|>' is not marked as EOG load: control token: 128086 '<|reserved_special_token_81|>' is not marked as EOG load: control token: 128084 '<|reserved_special_token_79|>' is not marked as EOG load: control token: 128082 '<|reserved_special_token_77|>' is not marked as EOG load: control token: 128077 '<|reserved_special_token_72|>' is not marked as EOG load: control token: 128074 '<|reserved_special_token_69|>' is not marked as EOG load: control token: 128073 '<|reserved_special_token_68|>' is not marked as EOG load: control token: 128070 '<|reserved_special_token_65|>' is not marked as EOG load: control token: 128067 '<|reserved_special_token_62|>' is not marked as EOG load: control token: 128066 '<|reserved_special_token_61|>' is not marked as EOG load: control token: 128064 '<|reserved_special_token_59|>' is not marked as EOG load: control token: 128061 '<|reserved_special_token_56|>' is not marked as EOG load: control token: 128059 '<|reserved_special_token_54|>' is not marked as EOG load: control token: 128058 '<|reserved_special_token_53|>' is not marked as EOG load: control token: 128057 '<|reserved_special_token_52|>' is not marked as EOG load: control token: 128051 '<|reserved_special_token_46|>' is not marked as EOG load: control token: 128042 '<|reserved_special_token_37|>' is not marked as EOG load: control token: 128041 '<|reserved_special_token_36|>' is not marked as EOG load: control token: 128040 '<|reserved_special_token_35|>' is not marked as EOG load: control token: 128039 '<|reserved_special_token_34|>' is not marked as EOG load: control token: 128035 '<|reserved_special_token_30|>' is not marked as EOG load: control token: 128034 '<|reserved_special_token_29|>' is not marked as EOG load: control token: 128032 '<|reserved_special_token_27|>' is not marked as EOG load: control token: 128031 '<|reserved_special_token_26|>' is not marked as EOG load: control token: 128030 '<|reserved_special_token_25|>' is not marked as EOG load: control token: 128029 '<|reserved_special_token_24|>' is not marked as EOG load: control token: 128027 '<|reserved_special_token_22|>' is not marked as EOG load: control token: 128026 '<|reserved_special_token_21|>' is not marked as EOG load: control token: 128025 '<|reserved_special_token_20|>' is not marked as EOG load: control token: 128023 '<|reserved_special_token_18|>' is not marked as EOG load: control token: 128022 '<|reserved_special_token_17|>' is not marked as EOG load: control token: 128021 '<|reserved_special_token_16|>' is not marked as EOG load: control token: 128019 '<|reserved_special_token_14|>' is not marked as EOG load: control token: 128017 '<|reserved_special_token_12|>' is not marked as EOG load: control token: 128014 '<|reserved_special_token_9|>' is not marked as EOG load: control token: 128013 '<|reserved_special_token_8|>' is not marked as EOG load: control token: 128012 '<|reserved_special_token_7|>' is not marked as EOG load: control token: 128011 '<|reserved_special_token_6|>' is not marked as EOG load: control token: 128010 '<|reserved_special_token_5|>' is not marked as EOG load: control token: 128006 '<|start_header_id|>' is not marked as EOG load: control token: 128005 '<|reserved_special_token_3|>' is not marked as EOG load: control token: 128003 '<|reserved_special_token_1|>' is not marked as EOG load: control token: 128002 '<|reserved_special_token_0|>' is not marked as EOG load: control token: 128000 '<|begin_of_text|>' is not marked as EOG load: control token: 128038 '<|reserved_special_token_33|>' is not marked as EOG load: control token: 128060 '<|reserved_special_token_55|>' is not marked as EOG load: control token: 128043 '<|reserved_special_token_38|>' is not marked as EOG load: control token: 128007 '<|end_header_id|>' is not marked as EOG load: control token: 128062 '<|reserved_special_token_57|>' is not marked as EOG load: control token: 128168 '<|reserved_special_token_163|>' is not marked as EOG load: control token: 128159 '<|reserved_special_token_154|>' is not marked as EOG load: control token: 128162 '<|reserved_special_token_157|>' is not marked as EOG load: control token: 128054 '<|reserved_special_token_49|>' is not marked as EOG load: control token: 128047 '<|reserved_special_token_42|>' is not marked as EOG load: control token: 128053 '<|reserved_special_token_48|>' is not marked as EOG load: control token: 128227 '<|reserved_special_token_222|>' is not marked as EOG load: control token: 128095 '<|reserved_special_token_90|>' is not marked as EOG load: control token: 128150 '<|reserved_special_token_145|>' is not marked as EOG load: control token: 128081 '<|reserved_special_token_76|>' is not marked as EOG load: control token: 128079 '<|reserved_special_token_74|>' is not marked as EOG load: control token: 128099 '<|reserved_special_token_94|>' is not marked as EOG load: control token: 128250 '<|reserved_special_token_245|>' is not marked as EOG load: control token: 128176 '<|reserved_special_token_171|>' is not marked as EOG load: control token: 128068 '<|reserved_special_token_63|>' is not marked as EOG load: control token: 128132 '<|reserved_special_token_127|>' is not marked as EOG load: control token: 128158 '<|reserved_special_token_153|>' is not marked as EOG load: control token: 128161 '<|reserved_special_token_156|>' is not marked as EOG load: control token: 128131 '<|reserved_special_token_126|>' is not marked as EOG load: control token: 128246 '<|reserved_special_token_241|>' is not marked as EOG load: control token: 128254 '<|reserved_special_token_249|>' is not marked as EOG load: control token: 128033 '<|reserved_special_token_28|>' is not marked as EOG load: control token: 128145 '<|reserved_special_token_140|>' is not marked as EOG load: control token: 128178 '<|reserved_special_token_173|>' is not marked as EOG load: control token: 128219 '<|reserved_special_token_214|>' is not marked as EOG load: control token: 128072 '<|reserved_special_token_67|>' is not marked as EOG load: control token: 128238 '<|reserved_special_token_233|>' is not marked as EOG load: control token: 128048 '<|reserved_special_token_43|>' is not marked as EOG load: control token: 128065 '<|reserved_special_token_60|>' is not marked as EOG load: control token: 128146 '<|reserved_special_token_141|>' is not marked as EOG load: control token: 128198 '<|reserved_special_token_193|>' is not marked as EOG load: control token: 128055 '<|reserved_special_token_50|>' is not marked as EOG load: control token: 128143 '<|reserved_special_token_138|>' is not marked as EOG load: control token: 128140 '<|reserved_special_token_135|>' is not marked as EOG load: control token: 128020 '<|reserved_special_token_15|>' is not marked as EOG load: control token: 128036 '<|reserved_special_token_31|>' is not marked as EOG load: control token: 128129 '<|reserved_special_token_124|>' is not marked as EOG load: control token: 128098 '<|reserved_special_token_93|>' is not marked as EOG load: control token: 128209 '<|reserved_special_token_204|>' is not marked as EOG load: control token: 128186 '<|reserved_special_token_181|>' is not marked as EOG load: control token: 128222 '<|reserved_special_token_217|>' is not marked as EOG load: control token: 128126 '<|reserved_special_token_121|>' is not marked as EOG load: control token: 128004 '<|reserved_special_token_2|>' is not marked as EOG load: control token: 128075 '<|reserved_special_token_70|>' is not marked as EOG load: control token: 128160 '<|reserved_special_token_155|>' is not marked as EOG load: control token: 128069 '<|reserved_special_token_64|>' is not marked as EOG load: control token: 128109 '<|reserved_special_token_104|>' is not marked as EOG load: control token: 128183 '<|reserved_special_token_178|>' is not marked as EOG load: control token: 128092 '<|reserved_special_token_87|>' is not marked as EOG load: control token: 128106 '<|reserved_special_token_101|>' is not marked as EOG load: control token: 128096 '<|reserved_special_token_91|>' is not marked as EOG load: control token: 128135 '<|reserved_special_token_130|>' is not marked as EOG load: control token: 128190 '<|reserved_special_token_185|>' is not marked as EOG load: control token: 128196 '<|reserved_special_token_191|>' is not marked as EOG load: control token: 128045 '<|reserved_special_token_40|>' is not marked as EOG load: control token: 128085 '<|reserved_special_token_80|>' is not marked as EOG load: control token: 128189 '<|reserved_special_token_184|>' is not marked as EOG load: control token: 128133 '<|reserved_special_token_128|>' is not marked as EOG load: control token: 128089 '<|reserved_special_token_84|>' is not marked as EOG load: control token: 128155 '<|reserved_special_token_150|>' is not marked as EOG load: control token: 128046 '<|reserved_special_token_41|>' is not marked as EOG load: control token: 128028 '<|reserved_special_token_23|>' is not marked as EOG load: control token: 128252 '<|reserved_special_token_247|>' is not marked as EOG load: control token: 128179 '<|reserved_special_token_174|>' is not marked as EOG load: control token: 128063 '<|reserved_special_token_58|>' is not marked as EOG load: control token: 128177 '<|reserved_special_token_172|>' is not marked as EOG load: control token: 128230 '<|reserved_special_token_225|>' is not marked as EOG load: control token: 128076 '<|reserved_special_token_71|>' is not marked as EOG load: control token: 128078 '<|reserved_special_token_73|>' is not marked as EOG load: control token: 128228 '<|reserved_special_token_223|>' is not marked as EOG load: control token: 128193 '<|reserved_special_token_188|>' is not marked as EOG load: control token: 128044 '<|reserved_special_token_39|>' is not marked as EOG load: control token: 128080 '<|reserved_special_token_75|>' is not marked as EOG load: control token: 128136 '<|reserved_special_token_131|>' is not marked as EOG load: control token: 128128 '<|reserved_special_token_123|>' is not marked as EOG load: control token: 128115 '<|reserved_special_token_110|>' is not marked as EOG load: control token: 128050 '<|reserved_special_token_45|>' is not marked as EOG load: control token: 128217 '<|reserved_special_token_212|>' is not marked as EOG load: control token: 128105 '<|reserved_special_token_100|>' is not marked as EOG load: control token: 128088 '<|reserved_special_token_83|>' is not marked as EOG load: control token: 128200 '<|reserved_special_token_195|>' is not marked as EOG load: control token: 128056 '<|reserved_special_token_51|>' is not marked as EOG load: control token: 128016 '<|reserved_special_token_11|>' is not marked as EOG load: control token: 128167 '<|reserved_special_token_162|>' is not marked as EOG load: control token: 128202 '<|reserved_special_token_197|>' is not marked as EOG load: control token: 128037 '<|reserved_special_token_32|>' is not marked as EOG load: control token: 128197 '<|reserved_special_token_192|>' is not marked as EOG load: control token: 128233 '<|reserved_special_token_228|>' is not marked as EOG load: control token: 128142 '<|reserved_special_token_137|>' is not marked as EOG load: control token: 128165 '<|reserved_special_token_160|>' is not marked as EOG load: control token: 128211 '<|reserved_special_token_206|>' is not marked as EOG load: control token: 128134 '<|reserved_special_token_129|>' is not marked as EOG load: control token: 128229 '<|reserved_special_token_224|>' is not marked as EOG load: control token: 128236 '<|reserved_special_token_231|>' is not marked as EOG load: control token: 128052 '<|reserved_special_token_47|>' is not marked as EOG load: control token: 128225 '<|reserved_special_token_220|>' is not marked as EOG load: control token: 128203 '<|reserved_special_token_198|>' is not marked as EOG load: control token: 128015 '<|reserved_special_token_10|>' is not marked as EOG load: control token: 128008 '<|reserved_special_token_4|>' is not marked as EOG load: control token: 128195 '<|reserved_special_token_190|>' is not marked as EOG load: control token: 128018 '<|reserved_special_token_13|>' is not marked as EOG load: control token: 128083 '<|reserved_special_token_78|>' is not marked as EOG load: control token: 128071 '<|reserved_special_token_66|>' is not marked as EOG load: control token: 128024 '<|reserved_special_token_19|>' is not marked as EOG load: control token: 128239 '<|reserved_special_token_234|>' is not marked as EOG load: control token: 128152 '<|reserved_special_token_147|>' is not marked as EOG load: control token: 128049 '<|reserved_special_token_44|>' is not marked as EOG load: control token: 128243 '<|reserved_special_token_238|>' is not marked as EOG load: control token: 128114 '<|reserved_special_token_109|>' is not marked as EOG load: control token: 128234 '<|reserved_special_token_229|>' is not marked as EOG load: printing all EOG tokens: load: - 128001 ('<|end_of_text|>') load: - 128009 ('<|eot_id|>') load: special tokens cache size = 256 load: token to piece cache size = 0.8000 MB print_info: arch = llama print_info: vocab_only = 0 print_info: no_alloc = 0 print_info: n_ctx_train = 8192 print_info: n_embd_inp = 4096 print_info: n_embd = 4096 print_info: n_embd_out = 4096 print_info: n_layer = 32 print_info: n_layer_all = 32 print_info: n_head = 32 print_info: n_head_kv = 8 print_info: n_rot = 128 print_info: n_swa = 0 print_info: is_swa_any = 0 print_info: non_causal_type = 0 print_info: n_embd_head_k = 128 print_info: n_embd_head_v = 128 print_info: n_gqa = 4 print_info: n_embd_k_gqa = 1024 print_info: n_embd_v_gqa = 1024 print_info: f_norm_eps = 0.0e+00 print_info: f_norm_rms_eps = 1.0e-05 print_info: f_clamp_kqv = 0.0e+00 print_info: f_max_alibi_bias = 0.0e+00 print_info: f_logit_scale = 0.0e+00 print_info: f_attn_scale = 0.0e+00 print_info: f_attn_value_scale = 0.0000 print_info: n_ff = 14336 print_info: n_expert = 0 print_info: n_expert_used = 0 print_info: n_expert_groups = 0 print_info: n_group_used = 0 print_info: causal attn = 1 print_info: pooling type = -1 print_info: rope type = 0 print_info: rope scaling = linear print_info: freq_base_train = 500000.0 print_info: freq_scale_train = 1 print_info: n_ctx_orig_yarn = 8192 print_info: rope_yarn_log_mul = 0.0000 print_info: rope_finetuned = unknown print_info: model type = 8B print_info: model params = 8.03 B print_info: general.name = Meta-Llama-3-8B-Instruct print_info: vocab type = BPE print_info: n_vocab = 128256 print_info: n_merges = 280147 print_info: BOS token = 128000 '<|begin_of_text|>' print_info: EOS token = 128009 '<|eot_id|>' print_info: EOT token = 128001 '<|end_of_text|>' print_info: LF token = 198 'Ċ' print_info: EOG token = 128001 '<|end_of_text|>' print_info: EOG token = 128009 '<|eot_id|>' print_info: max token length = 256 load_tensors: loading model tensors, this can take a while... (load_mode = mmap) load_tensors: layer 0 assigned to device MTL0, is_swa = 0 load_tensors: layer 1 assigned to device MTL0, is_swa = 0 load_tensors: layer 2 assigned to device MTL0, is_swa = 0 load_tensors: layer 3 assigned to device MTL0, is_swa = 0 load_tensors: layer 4 assigned to device MTL0, is_swa = 0 load_tensors: layer 5 assigned to device MTL0, is_swa = 0 load_tensors: layer 6 assigned to device MTL0, is_swa = 0 load_tensors: layer 7 assigned to device MTL0, is_swa = 0 load_tensors: layer 8 assigned to device MTL0, is_swa = 0 load_tensors: layer 9 assigned to device MTL0, is_swa = 0 load_tensors: layer 10 assigned to device MTL0, is_swa = 0 load_tensors: layer 11 assigned to device MTL0, is_swa = 0 load_tensors: layer 12 assigned to device MTL0, is_swa = 0 load_tensors: layer 13 assigned to device MTL0, is_swa = 0 load_tensors: layer 14 assigned to device MTL0, is_swa = 0 load_tensors: layer 15 assigned to device MTL0, is_swa = 0 load_tensors: layer 16 assigned to device MTL0, is_swa = 0 load_tensors: layer 17 assigned to device MTL0, is_swa = 0 load_tensors: layer 18 assigned to device MTL0, is_swa = 0 load_tensors: layer 19 assigned to device MTL0, is_swa = 0 load_tensors: layer 20 assigned to device MTL0, is_swa = 0 load_tensors: layer 21 assigned to device MTL0, is_swa = 0 load_tensors: layer 22 assigned to device MTL0, is_swa = 0 load_tensors: layer 23 assigned to device MTL0, is_swa = 0 load_tensors: layer 24 assigned to device MTL0, is_swa = 0 load_tensors: layer 25 assigned to device MTL0, is_swa = 0 load_tensors: layer 26 assigned to device MTL0, is_swa = 0 load_tensors: layer 27 assigned to device MTL0, is_swa = 0 load_tensors: layer 28 assigned to device MTL0, is_swa = 0 load_tensors: layer 29 assigned to device MTL0, is_swa = 0 load_tensors: layer 30 assigned to device MTL0, is_swa = 0 load_tensors: layer 31 assigned to device MTL0, is_swa = 0 load_tensors: layer 32 assigned to device MTL0, is_swa = 0 create_tensor: loading tensor token_embd.weight create_tensor: loading tensor output_norm.weight create_tensor: loading tensor output.weight create_tensor: loading tensor blk.0.attn_norm.weight create_tensor: loading tensor blk.0.attn_qkv.weight create_tensor: loading tensor blk.0.attn_q.weight create_tensor: loading tensor blk.0.attn_k.weight create_tensor: loading tensor blk.0.attn_v.weight create_tensor: loading tensor blk.0.attn_q.bias create_tensor: loading tensor blk.0.attn_k.bias create_tensor: loading tensor blk.0.attn_v.bias create_tensor: loading tensor blk.0.attn_output.weight create_tensor: loading tensor blk.0.attn_output.bias create_tensor: loading tensor blk.0.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.0.ffn_gate.weight create_tensor: loading tensor blk.0.ffn_down.weight create_tensor: loading tensor blk.0.ffn_up.weight create_tensor: loading tensor blk.0.ffn_gate.bias create_tensor: loading tensor blk.0.ffn_down.bias create_tensor: loading tensor blk.0.ffn_up.bias create_tensor: loading tensor blk.1.attn_norm.weight create_tensor: loading tensor blk.1.attn_qkv.weight create_tensor: loading tensor blk.1.attn_q.weight create_tensor: loading tensor blk.1.attn_k.weight create_tensor: loading tensor blk.1.attn_v.weight create_tensor: loading tensor blk.1.attn_q.bias create_tensor: loading tensor blk.1.attn_k.bias create_tensor: loading tensor blk.1.attn_v.bias create_tensor: loading tensor blk.1.attn_output.weight create_tensor: loading tensor blk.1.attn_output.bias create_tensor: loading tensor blk.1.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.1.ffn_gate.weight create_tensor: loading tensor blk.1.ffn_down.weight create_tensor: loading tensor blk.1.ffn_up.weight create_tensor: loading tensor blk.1.ffn_gate.bias create_tensor: loading tensor blk.1.ffn_down.bias create_tensor: loading tensor blk.1.ffn_up.bias create_tensor: loading tensor blk.2.attn_norm.weight create_tensor: loading tensor blk.2.attn_qkv.weight create_tensor: loading tensor blk.2.attn_q.weight create_tensor: loading tensor blk.2.attn_k.weight create_tensor: loading tensor blk.2.attn_v.weight create_tensor: loading tensor blk.2.attn_q.bias create_tensor: loading tensor blk.2.attn_k.bias create_tensor: loading tensor blk.2.attn_v.bias create_tensor: loading tensor blk.2.attn_output.weight create_tensor: loading tensor blk.2.attn_output.bias create_tensor: loading tensor blk.2.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.2.ffn_gate.weight create_tensor: loading tensor blk.2.ffn_down.weight create_tensor: loading tensor blk.2.ffn_up.weight create_tensor: loading tensor blk.2.ffn_gate.bias create_tensor: loading tensor blk.2.ffn_down.bias create_tensor: loading tensor blk.2.ffn_up.bias create_tensor: loading tensor blk.3.attn_norm.weight create_tensor: loading tensor blk.3.attn_qkv.weight create_tensor: loading tensor blk.3.attn_q.weight create_tensor: loading tensor blk.3.attn_k.weight create_tensor: loading tensor blk.3.attn_v.weight create_tensor: loading tensor blk.3.attn_q.bias create_tensor: loading tensor blk.3.attn_k.bias create_tensor: loading tensor blk.3.attn_v.bias create_tensor: loading tensor blk.3.attn_output.weight create_tensor: loading tensor blk.3.attn_output.bias create_tensor: loading tensor blk.3.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.3.ffn_gate.weight create_tensor: loading tensor blk.3.ffn_down.weight create_tensor: loading tensor blk.3.ffn_up.weight create_tensor: loading tensor blk.3.ffn_gate.bias create_tensor: loading tensor blk.3.ffn_down.bias create_tensor: loading tensor blk.3.ffn_up.bias create_tensor: loading tensor blk.4.attn_norm.weight create_tensor: loading tensor blk.4.attn_qkv.weight create_tensor: loading tensor blk.4.attn_q.weight create_tensor: loading tensor blk.4.attn_k.weight create_tensor: loading tensor blk.4.attn_v.weight create_tensor: loading tensor blk.4.attn_q.bias create_tensor: loading tensor blk.4.attn_k.bias create_tensor: loading tensor blk.4.attn_v.bias create_tensor: loading tensor blk.4.attn_output.weight create_tensor: loading tensor blk.4.attn_output.bias create_tensor: loading tensor blk.4.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.4.ffn_gate.weight create_tensor: loading tensor blk.4.ffn_down.weight create_tensor: loading tensor blk.4.ffn_up.weight create_tensor: loading tensor blk.4.ffn_gate.bias create_tensor: loading tensor blk.4.ffn_down.bias create_tensor: loading tensor blk.4.ffn_up.bias create_tensor: loading tensor blk.5.attn_norm.weight create_tensor: loading tensor blk.5.attn_qkv.weight create_tensor: loading tensor blk.5.attn_q.weight create_tensor: loading tensor blk.5.attn_k.weight create_tensor: loading tensor blk.5.attn_v.weight create_tensor: loading tensor blk.5.attn_q.bias create_tensor: loading tensor blk.5.attn_k.bias create_tensor: loading tensor blk.5.attn_v.bias create_tensor: loading tensor blk.5.attn_output.weight create_tensor: loading tensor blk.5.attn_output.bias create_tensor: loading tensor blk.5.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.5.ffn_gate.weight create_tensor: loading tensor blk.5.ffn_down.weight create_tensor: loading tensor blk.5.ffn_up.weight create_tensor: loading tensor blk.5.ffn_gate.bias create_tensor: loading tensor blk.5.ffn_down.bias create_tensor: loading tensor blk.5.ffn_up.bias create_tensor: loading tensor blk.6.attn_norm.weight create_tensor: loading tensor blk.6.attn_qkv.weight create_tensor: loading tensor blk.6.attn_q.weight create_tensor: loading tensor blk.6.attn_k.weight create_tensor: loading tensor blk.6.attn_v.weight create_tensor: loading tensor blk.6.attn_q.bias create_tensor: loading tensor blk.6.attn_k.bias create_tensor: loading tensor blk.6.attn_v.bias create_tensor: loading tensor blk.6.attn_output.weight create_tensor: loading tensor blk.6.attn_output.bias create_tensor: loading tensor blk.6.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.6.ffn_gate.weight create_tensor: loading tensor blk.6.ffn_down.weight create_tensor: loading tensor blk.6.ffn_up.weight create_tensor: loading tensor blk.6.ffn_gate.bias create_tensor: loading tensor blk.6.ffn_down.bias create_tensor: loading tensor blk.6.ffn_up.bias create_tensor: loading tensor blk.7.attn_norm.weight create_tensor: loading tensor blk.7.attn_qkv.weight create_tensor: loading tensor blk.7.attn_q.weight create_tensor: loading tensor blk.7.attn_k.weight create_tensor: loading tensor blk.7.attn_v.weight create_tensor: loading tensor blk.7.attn_q.bias create_tensor: loading tensor blk.7.attn_k.bias create_tensor: loading tensor blk.7.attn_v.bias create_tensor: loading tensor blk.7.attn_output.weight create_tensor: loading tensor blk.7.attn_output.bias create_tensor: loading tensor blk.7.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.7.ffn_gate.weight create_tensor: loading tensor blk.7.ffn_down.weight create_tensor: loading tensor blk.7.ffn_up.weight create_tensor: loading tensor blk.7.ffn_gate.bias create_tensor: loading tensor blk.7.ffn_down.bias create_tensor: loading tensor blk.7.ffn_up.bias create_tensor: loading tensor blk.8.attn_norm.weight create_tensor: loading tensor blk.8.attn_qkv.weight create_tensor: loading tensor blk.8.attn_q.weight create_tensor: loading tensor blk.8.attn_k.weight create_tensor: loading tensor blk.8.attn_v.weight create_tensor: loading tensor blk.8.attn_q.bias create_tensor: loading tensor blk.8.attn_k.bias create_tensor: loading tensor blk.8.attn_v.bias create_tensor: loading tensor blk.8.attn_output.weight create_tensor: loading tensor blk.8.attn_output.bias create_tensor: loading tensor blk.8.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.8.ffn_gate.weight create_tensor: loading tensor blk.8.ffn_down.weight create_tensor: loading tensor blk.8.ffn_up.weight create_tensor: loading tensor blk.8.ffn_gate.bias create_tensor: loading tensor blk.8.ffn_down.bias create_tensor: loading tensor blk.8.ffn_up.bias create_tensor: loading tensor blk.9.attn_norm.weight create_tensor: loading tensor blk.9.attn_qkv.weight create_tensor: loading tensor blk.9.attn_q.weight create_tensor: loading tensor blk.9.attn_k.weight create_tensor: loading tensor blk.9.attn_v.weight create_tensor: loading tensor blk.9.attn_q.bias create_tensor: loading tensor blk.9.attn_k.bias create_tensor: loading tensor blk.9.attn_v.bias create_tensor: loading tensor blk.9.attn_output.weight create_tensor: loading tensor blk.9.attn_output.bias create_tensor: loading tensor blk.9.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.9.ffn_gate.weight create_tensor: loading tensor blk.9.ffn_down.weight create_tensor: loading tensor blk.9.ffn_up.weight create_tensor: loading tensor blk.9.ffn_gate.bias create_tensor: loading tensor blk.9.ffn_down.bias create_tensor: loading tensor blk.9.ffn_up.bias create_tensor: loading tensor blk.10.attn_norm.weight create_tensor: loading tensor blk.10.attn_qkv.weight create_tensor: loading tensor blk.10.attn_q.weight create_tensor: loading tensor blk.10.attn_k.weight create_tensor: loading tensor blk.10.attn_v.weight create_tensor: loading tensor blk.10.attn_q.bias create_tensor: loading tensor blk.10.attn_k.bias create_tensor: loading tensor blk.10.attn_v.bias create_tensor: loading tensor blk.10.attn_output.weight create_tensor: loading tensor blk.10.attn_output.bias create_tensor: loading tensor blk.10.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.10.ffn_gate.weight create_tensor: loading tensor blk.10.ffn_down.weight create_tensor: loading tensor blk.10.ffn_up.weight create_tensor: loading tensor blk.10.ffn_gate.bias create_tensor: loading tensor blk.10.ffn_down.bias create_tensor: loading tensor blk.10.ffn_up.bias create_tensor: loading tensor blk.11.attn_norm.weight create_tensor: loading tensor blk.11.attn_qkv.weight create_tensor: loading tensor blk.11.attn_q.weight create_tensor: loading tensor blk.11.attn_k.weight create_tensor: loading tensor blk.11.attn_v.weight create_tensor: loading tensor blk.11.attn_q.bias create_tensor: loading tensor blk.11.attn_k.bias create_tensor: loading tensor blk.11.attn_v.bias create_tensor: loading tensor blk.11.attn_output.weight create_tensor: loading tensor blk.11.attn_output.bias create_tensor: loading tensor blk.11.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.11.ffn_gate.weight create_tensor: loading tensor blk.11.ffn_down.weight create_tensor: loading tensor blk.11.ffn_up.weight create_tensor: loading tensor blk.11.ffn_gate.bias create_tensor: loading tensor blk.11.ffn_down.bias create_tensor: loading tensor blk.11.ffn_up.bias create_tensor: loading tensor blk.12.attn_norm.weight create_tensor: loading tensor blk.12.attn_qkv.weight create_tensor: loading tensor blk.12.attn_q.weight create_tensor: loading tensor blk.12.attn_k.weight create_tensor: loading tensor blk.12.attn_v.weight create_tensor: loading tensor blk.12.attn_q.bias create_tensor: loading tensor blk.12.attn_k.bias create_tensor: loading tensor blk.12.attn_v.bias create_tensor: loading tensor blk.12.attn_output.weight create_tensor: loading tensor blk.12.attn_output.bias create_tensor: loading tensor blk.12.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.12.ffn_gate.weight create_tensor: loading tensor blk.12.ffn_down.weight create_tensor: loading tensor blk.12.ffn_up.weight create_tensor: loading tensor blk.12.ffn_gate.bias create_tensor: loading tensor blk.12.ffn_down.bias create_tensor: loading tensor blk.12.ffn_up.bias create_tensor: loading tensor blk.13.attn_norm.weight create_tensor: loading tensor blk.13.attn_qkv.weight create_tensor: loading tensor blk.13.attn_q.weight create_tensor: loading tensor blk.13.attn_k.weight create_tensor: loading tensor blk.13.attn_v.weight create_tensor: loading tensor blk.13.attn_q.bias create_tensor: loading tensor blk.13.attn_k.bias create_tensor: loading tensor blk.13.attn_v.bias create_tensor: loading tensor blk.13.attn_output.weight create_tensor: loading tensor blk.13.attn_output.bias create_tensor: loading tensor blk.13.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.13.ffn_gate.weight create_tensor: loading tensor blk.13.ffn_down.weight create_tensor: loading tensor blk.13.ffn_up.weight create_tensor: loading tensor blk.13.ffn_gate.bias create_tensor: loading tensor blk.13.ffn_down.bias create_tensor: loading tensor blk.13.ffn_up.bias create_tensor: loading tensor blk.14.attn_norm.weight create_tensor: loading tensor blk.14.attn_qkv.weight create_tensor: loading tensor blk.14.attn_q.weight create_tensor: loading tensor blk.14.attn_k.weight create_tensor: loading tensor blk.14.attn_v.weight create_tensor: loading tensor blk.14.attn_q.bias create_tensor: loading tensor blk.14.attn_k.bias create_tensor: loading tensor blk.14.attn_v.bias create_tensor: loading tensor blk.14.attn_output.weight create_tensor: loading tensor blk.14.attn_output.bias create_tensor: loading tensor blk.14.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.14.ffn_gate.weight create_tensor: loading tensor blk.14.ffn_down.weight create_tensor: loading tensor blk.14.ffn_up.weight create_tensor: loading tensor blk.14.ffn_gate.bias create_tensor: loading tensor blk.14.ffn_down.bias create_tensor: loading tensor blk.14.ffn_up.bias create_tensor: loading tensor blk.15.attn_norm.weight create_tensor: loading tensor blk.15.attn_qkv.weight create_tensor: loading tensor blk.15.attn_q.weight create_tensor: loading tensor blk.15.attn_k.weight create_tensor: loading tensor blk.15.attn_v.weight create_tensor: loading tensor blk.15.attn_q.bias create_tensor: loading tensor blk.15.attn_k.bias create_tensor: loading tensor blk.15.attn_v.bias create_tensor: loading tensor blk.15.attn_output.weight create_tensor: loading tensor blk.15.attn_output.bias create_tensor: loading tensor blk.15.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.15.ffn_gate.weight create_tensor: loading tensor blk.15.ffn_down.weight create_tensor: loading tensor blk.15.ffn_up.weight create_tensor: loading tensor blk.15.ffn_gate.bias create_tensor: loading tensor blk.15.ffn_down.bias create_tensor: loading tensor blk.15.ffn_up.bias create_tensor: loading tensor blk.16.attn_norm.weight create_tensor: loading tensor blk.16.attn_qkv.weight create_tensor: loading tensor blk.16.attn_q.weight create_tensor: loading tensor blk.16.attn_k.weight create_tensor: loading tensor blk.16.attn_v.weight create_tensor: loading tensor blk.16.attn_q.bias create_tensor: loading tensor blk.16.attn_k.bias create_tensor: loading tensor blk.16.attn_v.bias create_tensor: loading tensor blk.16.attn_output.weight create_tensor: loading tensor blk.16.attn_output.bias create_tensor: loading tensor blk.16.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.16.ffn_gate.weight create_tensor: loading tensor blk.16.ffn_down.weight create_tensor: loading tensor blk.16.ffn_up.weight create_tensor: loading tensor blk.16.ffn_gate.bias create_tensor: loading tensor blk.16.ffn_down.bias create_tensor: loading tensor blk.16.ffn_up.bias create_tensor: loading tensor blk.17.attn_norm.weight create_tensor: loading tensor blk.17.attn_qkv.weight create_tensor: loading tensor blk.17.attn_q.weight create_tensor: loading tensor blk.17.attn_k.weight create_tensor: loading tensor blk.17.attn_v.weight create_tensor: loading tensor blk.17.attn_q.bias create_tensor: loading tensor blk.17.attn_k.bias create_tensor: loading tensor blk.17.attn_v.bias create_tensor: loading tensor blk.17.attn_output.weight create_tensor: loading tensor blk.17.attn_output.bias create_tensor: loading tensor blk.17.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.17.ffn_gate.weight create_tensor: loading tensor blk.17.ffn_down.weight create_tensor: loading tensor blk.17.ffn_up.weight create_tensor: loading tensor blk.17.ffn_gate.bias create_tensor: loading tensor blk.17.ffn_down.bias create_tensor: loading tensor blk.17.ffn_up.bias create_tensor: loading tensor blk.18.attn_norm.weight create_tensor: loading tensor blk.18.attn_qkv.weight create_tensor: loading tensor blk.18.attn_q.weight create_tensor: loading tensor blk.18.attn_k.weight create_tensor: loading tensor blk.18.attn_v.weight create_tensor: loading tensor blk.18.attn_q.bias create_tensor: loading tensor blk.18.attn_k.bias create_tensor: loading tensor blk.18.attn_v.bias create_tensor: loading tensor blk.18.attn_output.weight create_tensor: loading tensor blk.18.attn_output.bias create_tensor: loading tensor blk.18.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.18.ffn_gate.weight create_tensor: loading tensor blk.18.ffn_down.weight create_tensor: loading tensor blk.18.ffn_up.weight create_tensor: loading tensor blk.18.ffn_gate.bias create_tensor: loading tensor blk.18.ffn_down.bias create_tensor: loading tensor blk.18.ffn_up.bias create_tensor: loading tensor blk.19.attn_norm.weight create_tensor: loading tensor blk.19.attn_qkv.weight create_tensor: loading tensor blk.19.attn_q.weight create_tensor: loading tensor blk.19.attn_k.weight create_tensor: loading tensor blk.19.attn_v.weight create_tensor: loading tensor blk.19.attn_q.bias create_tensor: loading tensor blk.19.attn_k.bias create_tensor: loading tensor blk.19.attn_v.bias create_tensor: loading tensor blk.19.attn_output.weight create_tensor: loading tensor blk.19.attn_output.bias create_tensor: loading tensor blk.19.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.19.ffn_gate.weight create_tensor: loading tensor blk.19.ffn_down.weight create_tensor: loading tensor blk.19.ffn_up.weight create_tensor: loading tensor blk.19.ffn_gate.bias create_tensor: loading tensor blk.19.ffn_down.bias create_tensor: loading tensor blk.19.ffn_up.bias create_tensor: loading tensor blk.20.attn_norm.weight create_tensor: loading tensor blk.20.attn_qkv.weight create_tensor: loading tensor blk.20.attn_q.weight create_tensor: loading tensor blk.20.attn_k.weight create_tensor: loading tensor blk.20.attn_v.weight create_tensor: loading tensor blk.20.attn_q.bias create_tensor: loading tensor blk.20.attn_k.bias create_tensor: loading tensor blk.20.attn_v.bias create_tensor: loading tensor blk.20.attn_output.weight create_tensor: loading tensor blk.20.attn_output.bias create_tensor: loading tensor blk.20.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.20.ffn_gate.weight create_tensor: loading tensor blk.20.ffn_down.weight create_tensor: loading tensor blk.20.ffn_up.weight create_tensor: loading tensor blk.20.ffn_gate.bias create_tensor: loading tensor blk.20.ffn_down.bias create_tensor: loading tensor blk.20.ffn_up.bias create_tensor: loading tensor blk.21.attn_norm.weight create_tensor: loading tensor blk.21.attn_qkv.weight create_tensor: loading tensor blk.21.attn_q.weight create_tensor: loading tensor blk.21.attn_k.weight create_tensor: loading tensor blk.21.attn_v.weight create_tensor: loading tensor blk.21.attn_q.bias create_tensor: loading tensor blk.21.attn_k.bias create_tensor: loading tensor blk.21.attn_v.bias create_tensor: loading tensor blk.21.attn_output.weight create_tensor: loading tensor blk.21.attn_output.bias create_tensor: loading tensor blk.21.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.21.ffn_gate.weight create_tensor: loading tensor blk.21.ffn_down.weight create_tensor: loading tensor blk.21.ffn_up.weight create_tensor: loading tensor blk.21.ffn_gate.bias create_tensor: loading tensor blk.21.ffn_down.bias create_tensor: loading tensor blk.21.ffn_up.bias create_tensor: loading tensor blk.22.attn_norm.weight create_tensor: loading tensor blk.22.attn_qkv.weight create_tensor: loading tensor blk.22.attn_q.weight create_tensor: loading tensor blk.22.attn_k.weight create_tensor: loading tensor blk.22.attn_v.weight create_tensor: loading tensor blk.22.attn_q.bias create_tensor: loading tensor blk.22.attn_k.bias create_tensor: loading tensor blk.22.attn_v.bias create_tensor: loading tensor blk.22.attn_output.weight create_tensor: loading tensor blk.22.attn_output.bias create_tensor: loading tensor blk.22.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.22.ffn_gate.weight create_tensor: loading tensor blk.22.ffn_down.weight create_tensor: loading tensor blk.22.ffn_up.weight create_tensor: loading tensor blk.22.ffn_gate.bias create_tensor: loading tensor blk.22.ffn_down.bias create_tensor: loading tensor blk.22.ffn_up.bias create_tensor: loading tensor blk.23.attn_norm.weight create_tensor: loading tensor blk.23.attn_qkv.weight create_tensor: loading tensor blk.23.attn_q.weight create_tensor: loading tensor blk.23.attn_k.weight create_tensor: loading tensor blk.23.attn_v.weight create_tensor: loading tensor blk.23.attn_q.bias create_tensor: loading tensor blk.23.attn_k.bias create_tensor: loading tensor blk.23.attn_v.bias create_tensor: loading tensor blk.23.attn_output.weight create_tensor: loading tensor blk.23.attn_output.bias create_tensor: loading tensor blk.23.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.23.ffn_gate.weight create_tensor: loading tensor blk.23.ffn_down.weight create_tensor: loading tensor blk.23.ffn_up.weight create_tensor: loading tensor blk.23.ffn_gate.bias create_tensor: loading tensor blk.23.ffn_down.bias create_tensor: loading tensor blk.23.ffn_up.bias create_tensor: loading tensor blk.24.attn_norm.weight create_tensor: loading tensor blk.24.attn_qkv.weight create_tensor: loading tensor blk.24.attn_q.weight create_tensor: loading tensor blk.24.attn_k.weight create_tensor: loading tensor blk.24.attn_v.weight create_tensor: loading tensor blk.24.attn_q.bias create_tensor: loading tensor blk.24.attn_k.bias create_tensor: loading tensor blk.24.attn_v.bias create_tensor: loading tensor blk.24.attn_output.weight create_tensor: loading tensor blk.24.attn_output.bias create_tensor: loading tensor blk.24.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.24.ffn_gate.weight create_tensor: loading tensor blk.24.ffn_down.weight create_tensor: loading tensor blk.24.ffn_up.weight create_tensor: loading tensor blk.24.ffn_gate.bias create_tensor: loading tensor blk.24.ffn_down.bias create_tensor: loading tensor blk.24.ffn_up.bias create_tensor: loading tensor blk.25.attn_norm.weight create_tensor: loading tensor blk.25.attn_qkv.weight create_tensor: loading tensor blk.25.attn_q.weight create_tensor: loading tensor blk.25.attn_k.weight create_tensor: loading tensor blk.25.attn_v.weight create_tensor: loading tensor blk.25.attn_q.bias create_tensor: loading tensor blk.25.attn_k.bias create_tensor: loading tensor blk.25.attn_v.bias create_tensor: loading tensor blk.25.attn_output.weight create_tensor: loading tensor blk.25.attn_output.bias create_tensor: loading tensor blk.25.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.25.ffn_gate.weight create_tensor: loading tensor blk.25.ffn_down.weight create_tensor: loading tensor blk.25.ffn_up.weight create_tensor: loading tensor blk.25.ffn_gate.bias create_tensor: loading tensor blk.25.ffn_down.bias create_tensor: loading tensor blk.25.ffn_up.bias create_tensor: loading tensor blk.26.attn_norm.weight create_tensor: loading tensor blk.26.attn_qkv.weight create_tensor: loading tensor blk.26.attn_q.weight create_tensor: loading tensor blk.26.attn_k.weight create_tensor: loading tensor blk.26.attn_v.weight create_tensor: loading tensor blk.26.attn_q.bias create_tensor: loading tensor blk.26.attn_k.bias create_tensor: loading tensor blk.26.attn_v.bias create_tensor: loading tensor blk.26.attn_output.weight create_tensor: loading tensor blk.26.attn_output.bias create_tensor: loading tensor blk.26.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.26.ffn_gate.weight create_tensor: loading tensor blk.26.ffn_down.weight create_tensor: loading tensor blk.26.ffn_up.weight create_tensor: loading tensor blk.26.ffn_gate.bias create_tensor: loading tensor blk.26.ffn_down.bias create_tensor: loading tensor blk.26.ffn_up.bias create_tensor: loading tensor blk.27.attn_norm.weight create_tensor: loading tensor blk.27.attn_qkv.weight create_tensor: loading tensor blk.27.attn_q.weight create_tensor: loading tensor blk.27.attn_k.weight create_tensor: loading tensor blk.27.attn_v.weight create_tensor: loading tensor blk.27.attn_q.bias create_tensor: loading tensor blk.27.attn_k.bias create_tensor: loading tensor blk.27.attn_v.bias create_tensor: loading tensor blk.27.attn_output.weight create_tensor: loading tensor blk.27.attn_output.bias create_tensor: loading tensor blk.27.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.27.ffn_gate.weight create_tensor: loading tensor blk.27.ffn_down.weight create_tensor: loading tensor blk.27.ffn_up.weight create_tensor: loading tensor blk.27.ffn_gate.bias create_tensor: loading tensor blk.27.ffn_down.bias create_tensor: loading tensor blk.27.ffn_up.bias create_tensor: loading tensor blk.28.attn_norm.weight create_tensor: loading tensor blk.28.attn_qkv.weight create_tensor: loading tensor blk.28.attn_q.weight create_tensor: loading tensor blk.28.attn_k.weight create_tensor: loading tensor blk.28.attn_v.weight create_tensor: loading tensor blk.28.attn_q.bias create_tensor: loading tensor blk.28.attn_k.bias create_tensor: loading tensor blk.28.attn_v.bias create_tensor: loading tensor blk.28.attn_output.weight create_tensor: loading tensor blk.28.attn_output.bias create_tensor: loading tensor blk.28.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.28.ffn_gate.weight create_tensor: loading tensor blk.28.ffn_down.weight create_tensor: loading tensor blk.28.ffn_up.weight create_tensor: loading tensor blk.28.ffn_gate.bias create_tensor: loading tensor blk.28.ffn_down.bias create_tensor: loading tensor blk.28.ffn_up.bias create_tensor: loading tensor blk.29.attn_norm.weight create_tensor: loading tensor blk.29.attn_qkv.weight create_tensor: loading tensor blk.29.attn_q.weight create_tensor: loading tensor blk.29.attn_k.weight create_tensor: loading tensor blk.29.attn_v.weight create_tensor: loading tensor blk.29.attn_q.bias create_tensor: loading tensor blk.29.attn_k.bias create_tensor: loading tensor blk.29.attn_v.bias create_tensor: loading tensor blk.29.attn_output.weight create_tensor: loading tensor blk.29.attn_output.bias create_tensor: loading tensor blk.29.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.29.ffn_gate.weight create_tensor: loading tensor blk.29.ffn_down.weight create_tensor: loading tensor blk.29.ffn_up.weight create_tensor: loading tensor blk.29.ffn_gate.bias create_tensor: loading tensor blk.29.ffn_down.bias create_tensor: loading tensor blk.29.ffn_up.bias create_tensor: loading tensor blk.30.attn_norm.weight create_tensor: loading tensor blk.30.attn_qkv.weight create_tensor: loading tensor blk.30.attn_q.weight create_tensor: loading tensor blk.30.attn_k.weight create_tensor: loading tensor blk.30.attn_v.weight create_tensor: loading tensor blk.30.attn_q.bias create_tensor: loading tensor blk.30.attn_k.bias create_tensor: loading tensor blk.30.attn_v.bias create_tensor: loading tensor blk.30.attn_output.weight create_tensor: loading tensor blk.30.attn_output.bias create_tensor: loading tensor blk.30.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.30.ffn_gate.weight create_tensor: loading tensor blk.30.ffn_down.weight create_tensor: loading tensor blk.30.ffn_up.weight create_tensor: loading tensor blk.30.ffn_gate.bias create_tensor: loading tensor blk.30.ffn_down.bias create_tensor: loading tensor blk.30.ffn_up.bias create_tensor: loading tensor blk.31.attn_norm.weight create_tensor: loading tensor blk.31.attn_qkv.weight create_tensor: loading tensor blk.31.attn_q.weight create_tensor: loading tensor blk.31.attn_k.weight create_tensor: loading tensor blk.31.attn_v.weight create_tensor: loading tensor blk.31.attn_q.bias create_tensor: loading tensor blk.31.attn_k.bias create_tensor: loading tensor blk.31.attn_v.bias create_tensor: loading tensor blk.31.attn_output.weight create_tensor: loading tensor blk.31.attn_output.bias create_tensor: loading tensor blk.31.ffn_norm.weight create_tensor: loading tensor rope_freqs.weight create_tensor: loading tensor blk.31.ffn_gate.weight create_tensor: loading tensor blk.31.ffn_down.weight create_tensor: loading tensor blk.31.ffn_up.weight create_tensor: loading tensor blk.31.ffn_gate.bias create_tensor: loading tensor blk.31.ffn_down.bias create_tensor: loading tensor blk.31.ffn_up.bias create_tensor: loading tensor blk.0.attn_q.scale create_tensor: loading tensor blk.0.attn_k.scale create_tensor: loading tensor blk.0.attn_v.scale create_tensor: loading tensor blk.0.attn_output.scale create_tensor: loading tensor blk.0.ffn_gate.scale create_tensor: loading tensor blk.0.ffn_down.scale create_tensor: loading tensor blk.0.ffn_up.scale create_tensor: loading tensor blk.0.attn_q.input_scale create_tensor: loading tensor blk.0.attn_k.input_scale create_tensor: loading tensor blk.0.attn_v.input_scale create_tensor: loading tensor blk.0.attn_output.input_scale create_tensor: loading tensor blk.0.ffn_gate.input_scale create_tensor: loading tensor blk.0.ffn_down.input_scale create_tensor: loading tensor blk.0.ffn_up.input_scale create_tensor: loading tensor blk.1.attn_q.scale create_tensor: loading tensor blk.1.attn_k.scale create_tensor: loading tensor blk.1.attn_v.scale create_tensor: loading tensor blk.1.attn_output.scale create_tensor: loading tensor blk.1.ffn_gate.scale create_tensor: loading tensor blk.1.ffn_down.scale create_tensor: loading tensor blk.1.ffn_up.scale create_tensor: loading tensor blk.1.attn_q.input_scale create_tensor: loading tensor blk.1.attn_k.input_scale create_tensor: loading tensor blk.1.attn_v.input_scale create_tensor: loading tensor blk.1.attn_output.input_scale create_tensor: loading tensor blk.1.ffn_gate.input_scale create_tensor: loading tensor blk.1.ffn_down.input_scale create_tensor: loading tensor blk.1.ffn_up.input_scale create_tensor: loading tensor blk.2.attn_q.scale create_tensor: loading tensor blk.2.attn_k.scale create_tensor: loading tensor blk.2.attn_v.scale create_tensor: loading tensor blk.2.attn_output.scale create_tensor: loading tensor blk.2.ffn_gate.scale create_tensor: loading tensor blk.2.ffn_down.scale create_tensor: loading tensor blk.2.ffn_up.scale create_tensor: loading tensor blk.2.attn_q.input_scale create_tensor: loading tensor blk.2.attn_k.input_scale create_tensor: loading tensor blk.2.attn_v.input_scale create_tensor: loading tensor blk.2.attn_output.input_scale create_tensor: loading tensor blk.2.ffn_gate.input_scale create_tensor: loading tensor blk.2.ffn_down.input_scale create_tensor: loading tensor blk.2.ffn_up.input_scale create_tensor: loading tensor blk.3.attn_q.scale create_tensor: loading tensor blk.3.attn_k.scale create_tensor: loading tensor blk.3.attn_v.scale create_tensor: loading tensor blk.3.attn_output.scale create_tensor: loading tensor blk.3.ffn_gate.scale create_tensor: loading tensor blk.3.ffn_down.scale create_tensor: loading tensor blk.3.ffn_up.scale create_tensor: loading tensor blk.3.attn_q.input_scale create_tensor: loading tensor blk.3.attn_k.input_scale create_tensor: loading tensor blk.3.attn_v.input_scale create_tensor: loading tensor blk.3.attn_output.input_scale create_tensor: loading tensor blk.3.ffn_gate.input_scale create_tensor: loading tensor blk.3.ffn_down.input_scale create_tensor: loading tensor blk.3.ffn_up.input_scale create_tensor: loading tensor blk.4.attn_q.scale create_tensor: loading tensor blk.4.attn_k.scale create_tensor: loading tensor blk.4.attn_v.scale create_tensor: loading tensor blk.4.attn_output.scale create_tensor: loading tensor blk.4.ffn_gate.scale create_tensor: loading tensor blk.4.ffn_down.scale create_tensor: loading tensor blk.4.ffn_up.scale create_tensor: loading tensor blk.4.attn_q.input_scale create_tensor: loading tensor blk.4.attn_k.input_scale create_tensor: loading tensor blk.4.attn_v.input_scale create_tensor: loading tensor blk.4.attn_output.input_scale create_tensor: loading tensor blk.4.ffn_gate.input_scale create_tensor: loading tensor blk.4.ffn_down.input_scale create_tensor: loading tensor blk.4.ffn_up.input_scale create_tensor: loading tensor blk.5.attn_q.scale create_tensor: loading tensor blk.5.attn_k.scale create_tensor: loading tensor blk.5.attn_v.scale create_tensor: loading tensor blk.5.attn_output.scale create_tensor: loading tensor blk.5.ffn_gate.scale create_tensor: loading tensor blk.5.ffn_down.scale create_tensor: loading tensor blk.5.ffn_up.scale create_tensor: loading tensor blk.5.attn_q.input_scale create_tensor: loading tensor blk.5.attn_k.input_scale create_tensor: loading tensor blk.5.attn_v.input_scale create_tensor: loading tensor blk.5.attn_output.input_scale create_tensor: loading tensor blk.5.ffn_gate.input_scale create_tensor: loading tensor blk.5.ffn_down.input_scale create_tensor: loading tensor blk.5.ffn_up.input_scale create_tensor: loading tensor blk.6.attn_q.scale create_tensor: loading tensor blk.6.attn_k.scale create_tensor: loading tensor blk.6.attn_v.scale create_tensor: loading tensor blk.6.attn_output.scale create_tensor: loading tensor blk.6.ffn_gate.scale create_tensor: loading tensor blk.6.ffn_down.scale create_tensor: loading tensor blk.6.ffn_up.scale create_tensor: loading tensor blk.6.attn_q.input_scale create_tensor: loading tensor blk.6.attn_k.input_scale create_tensor: loading tensor blk.6.attn_v.input_scale create_tensor: loading tensor blk.6.attn_output.input_scale create_tensor: loading tensor blk.6.ffn_gate.input_scale create_tensor: loading tensor blk.6.ffn_down.input_scale create_tensor: loading tensor blk.6.ffn_up.input_scale create_tensor: loading tensor blk.7.attn_q.scale create_tensor: loading tensor blk.7.attn_k.scale create_tensor: loading tensor blk.7.attn_v.scale create_tensor: loading tensor blk.7.attn_output.scale create_tensor: loading tensor blk.7.ffn_gate.scale create_tensor: loading tensor blk.7.ffn_down.scale create_tensor: loading tensor blk.7.ffn_up.scale create_tensor: loading tensor blk.7.attn_q.input_scale create_tensor: loading tensor blk.7.attn_k.input_scale create_tensor: loading tensor blk.7.attn_v.input_scale create_tensor: loading tensor blk.7.attn_output.input_scale create_tensor: loading tensor blk.7.ffn_gate.input_scale create_tensor: loading tensor blk.7.ffn_down.input_scale create_tensor: loading tensor blk.7.ffn_up.input_scale create_tensor: loading tensor blk.8.attn_q.scale create_tensor: loading tensor blk.8.attn_k.scale create_tensor: loading tensor blk.8.attn_v.scale create_tensor: loading tensor blk.8.attn_output.scale create_tensor: loading tensor blk.8.ffn_gate.scale create_tensor: loading tensor blk.8.ffn_down.scale create_tensor: loading tensor blk.8.ffn_up.scale create_tensor: loading tensor blk.8.attn_q.input_scale create_tensor: loading tensor blk.8.attn_k.input_scale create_tensor: loading tensor blk.8.attn_v.input_scale create_tensor: loading tensor blk.8.attn_output.input_scale create_tensor: loading tensor blk.8.ffn_gate.input_scale create_tensor: loading tensor blk.8.ffn_down.input_scale create_tensor: loading tensor blk.8.ffn_up.input_scale create_tensor: loading tensor blk.9.attn_q.scale create_tensor: loading tensor blk.9.attn_k.scale create_tensor: loading tensor blk.9.attn_v.scale create_tensor: loading tensor blk.9.attn_output.scale create_tensor: loading tensor blk.9.ffn_gate.scale create_tensor: loading tensor blk.9.ffn_down.scale create_tensor: loading tensor blk.9.ffn_up.scale create_tensor: loading tensor blk.9.attn_q.input_scale create_tensor: loading tensor blk.9.attn_k.input_scale create_tensor: loading tensor blk.9.attn_v.input_scale create_tensor: loading tensor blk.9.attn_output.input_scale create_tensor: loading tensor blk.9.ffn_gate.input_scale create_tensor: loading tensor blk.9.ffn_down.input_scale create_tensor: loading tensor blk.9.ffn_up.input_scale create_tensor: loading tensor blk.10.attn_q.scale create_tensor: loading tensor blk.10.attn_k.scale create_tensor: loading tensor blk.10.attn_v.scale create_tensor: loading tensor blk.10.attn_output.scale create_tensor: loading tensor blk.10.ffn_gate.scale create_tensor: loading tensor blk.10.ffn_down.scale create_tensor: loading tensor blk.10.ffn_up.scale create_tensor: loading tensor blk.10.attn_q.input_scale create_tensor: loading tensor blk.10.attn_k.input_scale create_tensor: loading tensor blk.10.attn_v.input_scale create_tensor: loading tensor blk.10.attn_output.input_scale create_tensor: loading tensor blk.10.ffn_gate.input_scale create_tensor: loading tensor blk.10.ffn_down.input_scale create_tensor: loading tensor blk.10.ffn_up.input_scale create_tensor: loading tensor blk.11.attn_q.scale create_tensor: loading tensor blk.11.attn_k.scale create_tensor: loading tensor blk.11.attn_v.scale create_tensor: loading tensor blk.11.attn_output.scale create_tensor: loading tensor blk.11.ffn_gate.scale create_tensor: loading tensor blk.11.ffn_down.scale create_tensor: loading tensor blk.11.ffn_up.scale create_tensor: loading tensor blk.11.attn_q.input_scale create_tensor: loading tensor blk.11.attn_k.input_scale create_tensor: loading tensor blk.11.attn_v.input_scale create_tensor: loading tensor blk.11.attn_output.input_scale create_tensor: loading tensor blk.11.ffn_gate.input_scale create_tensor: loading tensor blk.11.ffn_down.input_scale create_tensor: loading tensor blk.11.ffn_up.input_scale create_tensor: loading tensor blk.12.attn_q.scale create_tensor: loading tensor blk.12.attn_k.scale create_tensor: loading tensor blk.12.attn_v.scale create_tensor: loading tensor blk.12.attn_output.scale create_tensor: loading tensor blk.12.ffn_gate.scale create_tensor: loading tensor blk.12.ffn_down.scale create_tensor: loading tensor blk.12.ffn_up.scale create_tensor: loading tensor blk.12.attn_q.input_scale create_tensor: loading tensor blk.12.attn_k.input_scale create_tensor: loading tensor blk.12.attn_v.input_scale create_tensor: loading tensor blk.12.attn_output.input_scale create_tensor: loading tensor blk.12.ffn_gate.input_scale create_tensor: loading tensor blk.12.ffn_down.input_scale create_tensor: loading tensor blk.12.ffn_up.input_scale create_tensor: loading tensor blk.13.attn_q.scale create_tensor: loading tensor blk.13.attn_k.scale create_tensor: loading tensor blk.13.attn_v.scale create_tensor: loading tensor blk.13.attn_output.scale create_tensor: loading tensor blk.13.ffn_gate.scale create_tensor: loading tensor blk.13.ffn_down.scale create_tensor: loading tensor blk.13.ffn_up.scale create_tensor: loading tensor blk.13.attn_q.input_scale create_tensor: loading tensor blk.13.attn_k.input_scale create_tensor: loading tensor blk.13.attn_v.input_scale create_tensor: loading tensor blk.13.attn_output.input_scale create_tensor: loading tensor blk.13.ffn_gate.input_scale create_tensor: loading tensor blk.13.ffn_down.input_scale create_tensor: loading tensor blk.13.ffn_up.input_scale create_tensor: loading tensor blk.14.attn_q.scale create_tensor: loading tensor blk.14.attn_k.scale create_tensor: loading tensor blk.14.attn_v.scale create_tensor: loading tensor blk.14.attn_output.scale create_tensor: loading tensor blk.14.ffn_gate.scale create_tensor: loading tensor blk.14.ffn_down.scale create_tensor: loading tensor blk.14.ffn_up.scale create_tensor: loading tensor blk.14.attn_q.input_scale create_tensor: loading tensor blk.14.attn_k.input_scale create_tensor: loading tensor blk.14.attn_v.input_scale create_tensor: loading tensor blk.14.attn_output.input_scale create_tensor: loading tensor blk.14.ffn_gate.input_scale create_tensor: loading tensor blk.14.ffn_down.input_scale create_tensor: loading tensor blk.14.ffn_up.input_scale create_tensor: loading tensor blk.15.attn_q.scale create_tensor: loading tensor blk.15.attn_k.scale create_tensor: loading tensor blk.15.attn_v.scale create_tensor: loading tensor blk.15.attn_output.scale create_tensor: loading tensor blk.15.ffn_gate.scale create_tensor: loading tensor blk.15.ffn_down.scale create_tensor: loading tensor blk.15.ffn_up.scale create_tensor: loading tensor blk.15.attn_q.input_scale create_tensor: loading tensor blk.15.attn_k.input_scale create_tensor: loading tensor blk.15.attn_v.input_scale create_tensor: loading tensor blk.15.attn_output.input_scale create_tensor: loading tensor blk.15.ffn_gate.input_scale create_tensor: loading tensor blk.15.ffn_down.input_scale create_tensor: loading tensor blk.15.ffn_up.input_scale create_tensor: loading tensor blk.16.attn_q.scale create_tensor: loading tensor blk.16.attn_k.scale create_tensor: loading tensor blk.16.attn_v.scale create_tensor: loading tensor blk.16.attn_output.scale create_tensor: loading tensor blk.16.ffn_gate.scale create_tensor: loading tensor blk.16.ffn_down.scale create_tensor: loading tensor blk.16.ffn_up.scale create_tensor: loading tensor blk.16.attn_q.input_scale create_tensor: loading tensor blk.16.attn_k.input_scale create_tensor: loading tensor blk.16.attn_v.input_scale create_tensor: loading tensor blk.16.attn_output.input_scale create_tensor: loading tensor blk.16.ffn_gate.input_scale create_tensor: loading tensor blk.16.ffn_down.input_scale create_tensor: loading tensor blk.16.ffn_up.input_scale create_tensor: loading tensor blk.17.attn_q.scale create_tensor: loading tensor blk.17.attn_k.scale create_tensor: loading tensor blk.17.attn_v.scale create_tensor: loading tensor blk.17.attn_output.scale create_tensor: loading tensor blk.17.ffn_gate.scale create_tensor: loading tensor blk.17.ffn_down.scale create_tensor: loading tensor blk.17.ffn_up.scale create_tensor: loading tensor blk.17.attn_q.input_scale create_tensor: loading tensor blk.17.attn_k.input_scale create_tensor: loading tensor blk.17.attn_v.input_scale create_tensor: loading tensor blk.17.attn_output.input_scale create_tensor: loading tensor blk.17.ffn_gate.input_scale create_tensor: loading tensor blk.17.ffn_down.input_scale create_tensor: loading tensor blk.17.ffn_up.input_scale create_tensor: loading tensor blk.18.attn_q.scale create_tensor: loading tensor blk.18.attn_k.scale create_tensor: loading tensor blk.18.attn_v.scale create_tensor: loading tensor blk.18.attn_output.scale create_tensor: loading tensor blk.18.ffn_gate.scale create_tensor: loading tensor blk.18.ffn_down.scale create_tensor: loading tensor blk.18.ffn_up.scale create_tensor: loading tensor blk.18.attn_q.input_scale create_tensor: loading tensor blk.18.attn_k.input_scale create_tensor: loading tensor blk.18.attn_v.input_scale create_tensor: loading tensor blk.18.attn_output.input_scale create_tensor: loading tensor blk.18.ffn_gate.input_scale create_tensor: loading tensor blk.18.ffn_down.input_scale create_tensor: loading tensor blk.18.ffn_up.input_scale create_tensor: loading tensor blk.19.attn_q.scale create_tensor: loading tensor blk.19.attn_k.scale create_tensor: loading tensor blk.19.attn_v.scale create_tensor: loading tensor blk.19.attn_output.scale create_tensor: loading tensor blk.19.ffn_gate.scale create_tensor: loading tensor blk.19.ffn_down.scale create_tensor: loading tensor blk.19.ffn_up.scale create_tensor: loading tensor blk.19.attn_q.input_scale create_tensor: loading tensor blk.19.attn_k.input_scale create_tensor: loading tensor blk.19.attn_v.input_scale create_tensor: loading tensor blk.19.attn_output.input_scale create_tensor: loading tensor blk.19.ffn_gate.input_scale create_tensor: loading tensor blk.19.ffn_down.input_scale create_tensor: loading tensor blk.19.ffn_up.input_scale create_tensor: loading tensor blk.20.attn_q.scale create_tensor: loading tensor blk.20.attn_k.scale create_tensor: loading tensor blk.20.attn_v.scale create_tensor: loading tensor blk.20.attn_output.scale create_tensor: loading tensor blk.20.ffn_gate.scale create_tensor: loading tensor blk.20.ffn_down.scale create_tensor: loading tensor blk.20.ffn_up.scale create_tensor: loading tensor blk.20.attn_q.input_scale create_tensor: loading tensor blk.20.attn_k.input_scale create_tensor: loading tensor blk.20.attn_v.input_scale create_tensor: loading tensor blk.20.attn_output.input_scale create_tensor: loading tensor blk.20.ffn_gate.input_scale create_tensor: loading tensor blk.20.ffn_down.input_scale create_tensor: loading tensor blk.20.ffn_up.input_scale create_tensor: loading tensor blk.21.attn_q.scale create_tensor: loading tensor blk.21.attn_k.scale create_tensor: loading tensor blk.21.attn_v.scale create_tensor: loading tensor blk.21.attn_output.scale create_tensor: loading tensor blk.21.ffn_gate.scale create_tensor: loading tensor blk.21.ffn_down.scale create_tensor: loading tensor blk.21.ffn_up.scale create_tensor: loading tensor blk.21.attn_q.input_scale create_tensor: loading tensor blk.21.attn_k.input_scale create_tensor: loading tensor blk.21.attn_v.input_scale create_tensor: loading tensor blk.21.attn_output.input_scale create_tensor: loading tensor blk.21.ffn_gate.input_scale create_tensor: loading tensor blk.21.ffn_down.input_scale create_tensor: loading tensor blk.21.ffn_up.input_scale create_tensor: loading tensor blk.22.attn_q.scale create_tensor: loading tensor blk.22.attn_k.scale create_tensor: loading tensor blk.22.attn_v.scale create_tensor: loading tensor blk.22.attn_output.scale create_tensor: loading tensor blk.22.ffn_gate.scale create_tensor: loading tensor blk.22.ffn_down.scale create_tensor: loading tensor blk.22.ffn_up.scale create_tensor: loading tensor blk.22.attn_q.input_scale create_tensor: loading tensor blk.22.attn_k.input_scale create_tensor: loading tensor blk.22.attn_v.input_scale create_tensor: loading tensor blk.22.attn_output.input_scale create_tensor: loading tensor blk.22.ffn_gate.input_scale create_tensor: loading tensor blk.22.ffn_down.input_scale create_tensor: loading tensor blk.22.ffn_up.input_scale create_tensor: loading tensor blk.23.attn_q.scale create_tensor: loading tensor blk.23.attn_k.scale create_tensor: loading tensor blk.23.attn_v.scale create_tensor: loading tensor blk.23.attn_output.scale create_tensor: loading tensor blk.23.ffn_gate.scale create_tensor: loading tensor blk.23.ffn_down.scale create_tensor: loading tensor blk.23.ffn_up.scale create_tensor: loading tensor blk.23.attn_q.input_scale create_tensor: loading tensor blk.23.attn_k.input_scale create_tensor: loading tensor blk.23.attn_v.input_scale create_tensor: loading tensor blk.23.attn_output.input_scale create_tensor: loading tensor blk.23.ffn_gate.input_scale create_tensor: loading tensor blk.23.ffn_down.input_scale create_tensor: loading tensor blk.23.ffn_up.input_scale create_tensor: loading tensor blk.24.attn_q.scale create_tensor: loading tensor blk.24.attn_k.scale create_tensor: loading tensor blk.24.attn_v.scale create_tensor: loading tensor blk.24.attn_output.scale create_tensor: loading tensor blk.24.ffn_gate.scale create_tensor: loading tensor blk.24.ffn_down.scale create_tensor: loading tensor blk.24.ffn_up.scale create_tensor: loading tensor blk.24.attn_q.input_scale create_tensor: loading tensor blk.24.attn_k.input_scale create_tensor: loading tensor blk.24.attn_v.input_scale create_tensor: loading tensor blk.24.attn_output.input_scale create_tensor: loading tensor blk.24.ffn_gate.input_scale create_tensor: loading tensor blk.24.ffn_down.input_scale create_tensor: loading tensor blk.24.ffn_up.input_scale create_tensor: loading tensor blk.25.attn_q.scale create_tensor: loading tensor blk.25.attn_k.scale create_tensor: loading tensor blk.25.attn_v.scale create_tensor: loading tensor blk.25.attn_output.scale create_tensor: loading tensor blk.25.ffn_gate.scale create_tensor: loading tensor blk.25.ffn_down.scale create_tensor: loading tensor blk.25.ffn_up.scale create_tensor: loading tensor blk.25.attn_q.input_scale create_tensor: loading tensor blk.25.attn_k.input_scale create_tensor: loading tensor blk.25.attn_v.input_scale create_tensor: loading tensor blk.25.attn_output.input_scale create_tensor: loading tensor blk.25.ffn_gate.input_scale create_tensor: loading tensor blk.25.ffn_down.input_scale create_tensor: loading tensor blk.25.ffn_up.input_scale create_tensor: loading tensor blk.26.attn_q.scale create_tensor: loading tensor blk.26.attn_k.scale create_tensor: loading tensor blk.26.attn_v.scale create_tensor: loading tensor blk.26.attn_output.scale create_tensor: loading tensor blk.26.ffn_gate.scale create_tensor: loading tensor blk.26.ffn_down.scale create_tensor: loading tensor blk.26.ffn_up.scale create_tensor: loading tensor blk.26.attn_q.input_scale create_tensor: loading tensor blk.26.attn_k.input_scale create_tensor: loading tensor blk.26.attn_v.input_scale create_tensor: loading tensor blk.26.attn_output.input_scale create_tensor: loading tensor blk.26.ffn_gate.input_scale create_tensor: loading tensor blk.26.ffn_down.input_scale create_tensor: loading tensor blk.26.ffn_up.input_scale create_tensor: loading tensor blk.27.attn_q.scale create_tensor: loading tensor blk.27.attn_k.scale create_tensor: loading tensor blk.27.attn_v.scale create_tensor: loading tensor blk.27.attn_output.scale create_tensor: loading tensor blk.27.ffn_gate.scale create_tensor: loading tensor blk.27.ffn_down.scale create_tensor: loading tensor blk.27.ffn_up.scale create_tensor: loading tensor blk.27.attn_q.input_scale create_tensor: loading tensor blk.27.attn_k.input_scale create_tensor: loading tensor blk.27.attn_v.input_scale create_tensor: loading tensor blk.27.attn_output.input_scale create_tensor: loading tensor blk.27.ffn_gate.input_scale create_tensor: loading tensor blk.27.ffn_down.input_scale create_tensor: loading tensor blk.27.ffn_up.input_scale create_tensor: loading tensor blk.28.attn_q.scale create_tensor: loading tensor blk.28.attn_k.scale create_tensor: loading tensor blk.28.attn_v.scale create_tensor: loading tensor blk.28.attn_output.scale create_tensor: loading tensor blk.28.ffn_gate.scale create_tensor: loading tensor blk.28.ffn_down.scale create_tensor: loading tensor blk.28.ffn_up.scale create_tensor: loading tensor blk.28.attn_q.input_scale create_tensor: loading tensor blk.28.attn_k.input_scale create_tensor: loading tensor blk.28.attn_v.input_scale create_tensor: loading tensor blk.28.attn_output.input_scale create_tensor: loading tensor blk.28.ffn_gate.input_scale create_tensor: loading tensor blk.28.ffn_down.input_scale create_tensor: loading tensor blk.28.ffn_up.input_scale create_tensor: loading tensor blk.29.attn_q.scale create_tensor: loading tensor blk.29.attn_k.scale create_tensor: loading tensor blk.29.attn_v.scale create_tensor: loading tensor blk.29.attn_output.scale create_tensor: loading tensor blk.29.ffn_gate.scale create_tensor: loading tensor blk.29.ffn_down.scale create_tensor: loading tensor blk.29.ffn_up.scale create_tensor: loading tensor blk.29.attn_q.input_scale create_tensor: loading tensor blk.29.attn_k.input_scale create_tensor: loading tensor blk.29.attn_v.input_scale create_tensor: loading tensor blk.29.attn_output.input_scale create_tensor: loading tensor blk.29.ffn_gate.input_scale create_tensor: loading tensor blk.29.ffn_down.input_scale create_tensor: loading tensor blk.29.ffn_up.input_scale create_tensor: loading tensor blk.30.attn_q.scale create_tensor: loading tensor blk.30.attn_k.scale create_tensor: loading tensor blk.30.attn_v.scale create_tensor: loading tensor blk.30.attn_output.scale create_tensor: loading tensor blk.30.ffn_gate.scale create_tensor: loading tensor blk.30.ffn_down.scale create_tensor: loading tensor blk.30.ffn_up.scale create_tensor: loading tensor blk.30.attn_q.input_scale create_tensor: loading tensor blk.30.attn_k.input_scale create_tensor: loading tensor blk.30.attn_v.input_scale create_tensor: loading tensor blk.30.attn_output.input_scale create_tensor: loading tensor blk.30.ffn_gate.input_scale create_tensor: loading tensor blk.30.ffn_down.input_scale create_tensor: loading tensor blk.30.ffn_up.input_scale create_tensor: loading tensor blk.31.attn_q.scale create_tensor: loading tensor blk.31.attn_k.scale create_tensor: loading tensor blk.31.attn_v.scale create_tensor: loading tensor blk.31.attn_output.scale create_tensor: loading tensor blk.31.ffn_gate.scale create_tensor: loading tensor blk.31.ffn_down.scale create_tensor: loading tensor blk.31.ffn_up.scale create_tensor: loading tensor blk.31.attn_q.input_scale create_tensor: loading tensor blk.31.attn_k.input_scale create_tensor: loading tensor blk.31.attn_v.input_scale create_tensor: loading tensor blk.31.attn_output.input_scale create_tensor: loading tensor blk.31.ffn_gate.input_scale create_tensor: loading tensor blk.31.ffn_down.input_scale create_tensor: loading tensor blk.31.ffn_up.input_scale done_getting_tensors: tensor 'token_embd.weight' (q4_0) (and 0 others) cannot be used with preferred buffer type CPU_REPACK, using CPU instead ggml_metal_log_allocated_size: allocated buffer, size = 4156.00 MiB, ( 4156.38 / 25559.05) load_tensors: offloading output layer to GPU load_tensors: offloading 31 repeating layers to GPU load_tensors: offloaded 33/33 layers to GPU load_tensors: CPU_Mapped model buffer size = 281.81 MiB load_tensors: MTL0_Mapped model buffer size = 4155.99 MiB ....................................................................................... llama_context: constructing llama_context llama_context: n_seq_max = 1 llama_context: n_ctx = 8704 llama_context: n_ctx_seq = 8704 llama_context: n_batch = 2048 llama_context: n_ubatch = 512 llama_context: causal_attn = 1 llama_context: flash_attn = auto llama_context: kv_unified = false llama_context: freq_base = 500000.0 llama_context: freq_scale = 1 llama_context: n_rs_seq = 0 llama_context: n_outputs_max = 2048 llama_context: n_outputs_max_per_seq = 1 llama_context: n_ctx_seq (8704) > n_ctx_train (8192) -- possible training context overflow ggml_metal_init: allocating ggml_metal_init: found device: Apple M1 Pro ggml_metal_init: picking default device: Apple M1 Pro ggml_metal_init: use fusion = true ggml_metal_init: use concurrency = true ggml_metal_init: use graph optimize = true set_abort_callback: call llama_context: CPU output buffer size = 0.49 MiB llama_kv_cache: layer 0: dev = MTL0 llama_kv_cache: layer 1: dev = MTL0 llama_kv_cache: layer 2: dev = MTL0 llama_kv_cache: layer 3: dev = MTL0 llama_kv_cache: layer 4: dev = MTL0 llama_kv_cache: layer 5: dev = MTL0 llama_kv_cache: layer 6: dev = MTL0 llama_kv_cache: layer 7: dev = MTL0 llama_kv_cache: layer 8: dev = MTL0 llama_kv_cache: layer 9: dev = MTL0 llama_kv_cache: layer 10: dev = MTL0 llama_kv_cache: layer 11: dev = MTL0 llama_kv_cache: layer 12: dev = MTL0 llama_kv_cache: layer 13: dev = MTL0 llama_kv_cache: layer 14: dev = MTL0 llama_kv_cache: layer 15: dev = MTL0 llama_kv_cache: layer 16: dev = MTL0 llama_kv_cache: layer 17: dev = MTL0 llama_kv_cache: layer 18: dev = MTL0 llama_kv_cache: layer 19: dev = MTL0 llama_kv_cache: layer 20: dev = MTL0 llama_kv_cache: layer 21: dev = MTL0 llama_kv_cache: layer 22: dev = MTL0 llama_kv_cache: layer 23: dev = MTL0 llama_kv_cache: layer 24: dev = MTL0 llama_kv_cache: layer 25: dev = MTL0 llama_kv_cache: layer 26: dev = MTL0 llama_kv_cache: layer 27: dev = MTL0 llama_kv_cache: layer 28: dev = MTL0 llama_kv_cache: layer 29: dev = MTL0 llama_kv_cache: layer 30: dev = MTL0 llama_kv_cache: layer 31: dev = MTL0 llama_kv_cache: MTL0 KV buffer size = 1088.00 MiB llama_kv_cache: size = 1088.00 MiB ( 8704 cells, 32 layers, 1/1 seqs), K (f16): 544.00 MiB, V (f16): 544.00 MiB llama_kv_cache: attn_rot_k = 0, n_embd_head_k_all = 128 llama_kv_cache: attn_rot_v = 0, n_embd_head_k_all = 128 llama_context: enumerating backends llama_context: backend_ptrs.size() = 3 sched_reserve: reserving ... sched_reserve: max_nodes = 2328 sched_reserve: reserving full memory module sched_reserve: worst-case: n_tokens = 512, n_seqs = 1, n_outputs = 1 graph_reserve: reserving a graph for ubatch with n_tokens = 1, n_seqs = 1, n_outputs = 1 llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 1, 1, 1, 1 }] is used by node 'node_985' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 1, 1, 1, 1 }] is used by node 'node_986' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_21' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_52' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_83' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_114' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_145' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_176' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_207' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_238' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_269' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_300' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_331' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_362' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_393' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_424' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_455' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_486' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_517' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_548' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_579' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_610' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_641' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_672' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_703' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_734' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_765' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_796' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_827' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_858' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_889' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_920' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_951' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_982' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' inp_tokens' [i32, ne = { 1, 1, 1, 1 }] is used by node 'embd' (GET_ROWS) resolve_fused_ops: Flash Attention enabled resolve_fused_ops: resolving fused DeepSeek V4 HC support: graph_reserve: reserving a graph for ubatch with n_tokens = 1, n_seqs = 1, n_outputs = 1 llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 1, 1, 1, 1 }] is used by node 'node_985' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 1, 1, 1, 1 }] is used by node 'node_986' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_21' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_52' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_83' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_114' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_145' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_176' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_207' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_238' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_269' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_300' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_331' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_362' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_393' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_424' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_455' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_486' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_517' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_548' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_579' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_610' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_641' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_672' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_703' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_734' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_765' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_796' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_827' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_858' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_889' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_920' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_951' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_982' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' inp_tokens' [i32, ne = { 1, 1, 1, 1 }] is used by node 'embd' (GET_ROWS) resolve_fused_ops: fused DeepSeek V4 HC pre enabled graph_reserve: reserving a graph for ubatch with n_tokens = 1, n_seqs = 1, n_outputs = 1 llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 1, 1, 1, 1 }] is used by node 'node_985' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 1, 1, 1, 1 }] is used by node 'node_986' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_21' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_52' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_83' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_114' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_145' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_176' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_207' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_238' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_269' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_300' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_331' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_362' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_393' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_424' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_455' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_486' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_517' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_548' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_579' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_610' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_641' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_672' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_703' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_734' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_765' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_796' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_827' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_858' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_889' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_920' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_951' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_982' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' inp_tokens' [i32, ne = { 1, 1, 1, 1 }] is used by node 'embd' (GET_ROWS) resolve_fused_ops: fused DeepSeek V4 HC comb enabled graph_reserve: reserving a graph for ubatch with n_tokens = 1, n_seqs = 1, n_outputs = 1 llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 1, 1, 1, 1 }] is used by node 'node_985' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 1, 1, 1, 1 }] is used by node 'node_986' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_21' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_52' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_83' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_114' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_145' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_176' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_207' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_238' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_269' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_300' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_331' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_362' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_393' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_424' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_455' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_486' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_517' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_548' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_579' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_610' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_641' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_672' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_703' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_734' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_765' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_796' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_827' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_858' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_889' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_920' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_951' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_982' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' inp_tokens' [i32, ne = { 1, 1, 1, 1 }] is used by node 'embd' (GET_ROWS) resolve_fused_ops: fused DeepSeek V4 HC post enabled graph_reserve: reserving a graph for ubatch with n_tokens = 512, n_seqs = 1, n_outputs = 512 llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 512, 1, 1, 1 }] is used by node 'node_985' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 512, 1, 1, 1 }] is used by node 'node_986' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_21' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_52' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_83' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_114' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_145' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_176' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_207' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_238' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_269' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_300' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_331' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_362' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_393' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_424' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_455' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_486' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_517' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_548' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_579' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_610' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_641' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_672' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_703' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_734' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_765' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_796' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_827' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_858' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_889' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_920' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_951' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_982' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' inp_tokens' [i32, ne = { 512, 1, 1, 1 }] is used by node 'embd' (GET_ROWS) graph_reserve: reserving a graph for ubatch with n_tokens = 1, n_seqs = 1, n_outputs = 1 llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 1, 1, 1, 1 }] is used by node 'node_985' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 1, 1, 1, 1 }] is used by node 'node_986' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_21' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_52' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_83' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_114' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_145' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_176' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_207' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_238' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_269' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_300' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_331' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_362' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_393' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_424' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_455' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_486' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_517' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_548' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_579' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_610' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_641' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_672' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_703' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_734' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_765' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_796' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_827' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_858' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_889' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_920' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_951' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 1, 1, 1 }] is used by node 'node_982' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' inp_tokens' [i32, ne = { 1, 1, 1, 1 }] is used by node 'embd' (GET_ROWS) graph_reserve: reserving a graph for ubatch with n_tokens = 512, n_seqs = 1, n_outputs = 512 llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 512, 1, 1, 1 }] is used by node 'node_985' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 512, 1, 1, 1 }] is used by node 'node_986' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_21' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_52' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_83' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_114' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_145' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_176' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_207' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_238' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_269' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_300' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_331' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_362' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_393' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_424' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_455' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_486' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_517' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_548' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_579' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_610' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_641' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_672' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_703' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_734' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_765' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_796' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_827' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_858' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_889' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_920' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_951' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8704, 512, 1, 1 }] is used by node 'node_982' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' inp_tokens' [i32, ne = { 512, 1, 1, 1 }] is used by node 'embd' (GET_ROWS) sched_reserve: MTL0 compute buffer size = 258.50 MiB sched_reserve: CPU compute buffer size = 24.51 MiB sched_reserve: graph: nodes = 998, splits = 2, input objects = 4, input tensors = 6 sched_reserve: reserve took 22.05 ms, sched copies = 1 attach_threadpool: call set_n_threads: n_threads = 8, n_threads_batch = 8 ggml_metal_library_compile_pipeline: compiling pipeline: base = 'kernel_rms_norm_mul_f32_4', name = 'kernel_rms_norm_mul_f32_4' ggml_metal_library_compile_pipeline: loaded kernel_rms_norm_mul_f32_4 0x103a12600 | th_max = 1024 | th_width = 32 ggml_metal_library_compile_pipeline: compiling pipeline: base = 'kernel_mul_mm_q4_0_f32', name = 'kernel_mul_mm_q4_0_f32_bci=0_bco=0_ne12=1_ne13=1_r2=1_r3=1' ggml_metal_library_compile_pipeline: loaded kernel_mul_mm_q4_0_f32_bci=0_bco=0_ne12=1_ne13=1_r2=1_r3=1 0x103a12e00 | th_max = 896 | th_width = 32 ggml_metal_library_compile_pipeline: compiling pipeline: base = 'kernel_rope_norm_f32', name = 'kernel_rope_norm_f32_imrope=0_is_back=0' ggml_metal_library_compile_pipeline: loaded kernel_rope_norm_f32_imrope=0_is_back=0 0x103a145c0 | th_max = 1024 | th_width = 32 ggml_metal_library_compile_pipeline: compiling pipeline: base = 'kernel_set_rows_f32_i64_f16', name = 'kernel_set_rows_f32_i64_f16' ggml_metal_library_compile_pipeline: loaded kernel_set_rows_f32_i64_f16 0x103a14dc0 | th_max = 1024 | th_width = 32 ggml_metal_library_compile_pipeline: compiling pipeline: base = 'kernel_flash_attn_ext_blk', name = 'kernel_flash_attn_ext_blk_nqptg=8_ncpsg=64' ggml_metal_library_compile_pipeline: loaded kernel_flash_attn_ext_blk_nqptg=8_ncpsg=64 0x103a15980 | th_max = 1024 | th_width = 32 ggml_metal_library_compile_pipeline: compiling pipeline: base = 'kernel_flash_attn_ext_f16_dk128_dv128', name = 'kernel_flash_attn_ext_f16_dk128_dv128_mask=1_sinks=0_bias=0_scap=0_kvpad=0_bcm=0_ns10=1024_ns20=1024_nsg=4' ggml_metal_library_compile_pipeline: loaded kernel_flash_attn_ext_f16_dk128_dv128_mask=1_sinks=0_bias=0_scap=0_kvpad=0_bcm=0_ns10=1024_ns20=1024_nsg=4 0x103a15dc0 | th_max = 704 | th_width = 32 ggml_metal_library_compile_pipeline: compiling pipeline: base = 'kernel_bin_fuse_f32_f32_f32_4', name = 'kernel_bin_fuse_f32_f32_f32_4_op=0_nf=1_rb=0_cb=0' ggml_metal_library_compile_pipeline: loaded kernel_bin_fuse_f32_f32_f32_4_op=0_nf=1_rb=0_cb=0 0x103a160c0 | th_max = 1024 | th_width = 32 ggml_metal_library_compile_pipeline: compiling pipeline: base = 'kernel_swiglu_f32', name = 'kernel_swiglu_f32' ggml_metal_library_compile_pipeline: loaded kernel_swiglu_f32 0x103a163c0 | th_max = 1024 | th_width = 32 ggml_metal_library_compile_pipeline: compiling pipeline: base = 'kernel_get_rows_f32', name = 'kernel_get_rows_f32' ggml_metal_library_compile_pipeline: loaded kernel_get_rows_f32 0x8cdc80000 | th_max = 1024 | th_width = 32 ggml_metal_library_compile_pipeline: compiling pipeline: base = 'kernel_bin_fuse_f32_f32_f32_4', name = 'kernel_bin_fuse_f32_f32_f32_4_op=0_nf=1_rb=1_cb=0' ggml_metal_library_compile_pipeline: loaded kernel_bin_fuse_f32_f32_f32_4_op=0_nf=1_rb=1_cb=0 0x8cdc80300 | th_max = 1024 | th_width = 32 ggml_metal_library_compile_pipeline: compiling pipeline: base = 'kernel_mul_mv_q4_0_f32', name = 'kernel_mul_mv_q4_0_f32_nsg=2_ne12=1_r2=1_r3=1_split=0' ggml_metal_library_compile_pipeline: loaded kernel_mul_mv_q4_0_f32_nsg=2_ne12=1_r2=1_r3=1_split=0 0x8cdc80600 | th_max = 640 | th_width = 32 ggml_metal_library_compile_pipeline: compiling pipeline: base = 'kernel_mul_mv_q6_K_f32', name = 'kernel_mul_mv_q6_K_f32_nsg=2_ne12=1_r2=1_r3=1_split=0' ggml_metal_library_compile_pipeline: loaded kernel_mul_mv_q6_K_f32_nsg=2_ne12=1_r2=1_r3=1_split=0 0x8cdc80900 | th_max = 1024 | th_width = 32 set_n_threads: n_threads = 8, n_threads_batch = 8 set_n_threads: n_threads = 8, n_threads_batch = 8 state_read_meta: cell_count = 8192, dest_seq_id = 0 set_n_threads: n_threads = 8, n_threads_batch = 8 state_read_meta: cell_count = 8192, dest_seq_id = 0 set_n_threads: n_threads = 8, n_threads_batch = 8 state_read_meta: cell_count = 8192, dest_seq_id = 0 set_n_threads: n_threads = 8, n_threads_batch = 8 state_read_meta: cell_count = 8192, dest_seq_id = 0 set_n_threads: n_threads = 8, n_threads_batch = 8 llama_perf_context_print: load time = 2309.95 ms llama_perf_context_print: prompt eval time = 0.00 ms / 11264 tokens ( 0.00 ms per token, inf tokens per second) llama_perf_context_print: eval time = 0.00 ms / 1 runs ( 0.00 ms per token, inf tokens per second) llama_perf_context_print: total time = 50999.46 ms / 11265 tokens llama_perf_context_print: graphs reused = 4 ~llama_context: MTL0 compute buffer size is 258.5000 MiB, matches expectation of 258.5000 MiB ~llama_context: CPU compute buffer size is 24.5137 MiB, matches expectation of 24.5137 MiB ~llama_context: CPU compute buffer size is 0.0000 MiB, matches expectation of 0.0000 MiB ggml_metal_free: deallocating llama_context: constructing llama_context llama_context: n_seq_max = 1 llama_context: n_ctx = 8448 llama_context: n_ctx_seq = 8448 llama_context: n_batch = 2048 llama_context: n_ubatch = 512 llama_context: causal_attn = 1 llama_context: flash_attn = auto llama_context: kv_unified = false llama_context: freq_base = 500000.0 llama_context: freq_scale = 1 llama_context: n_rs_seq = 0 llama_context: n_outputs_max = 2048 llama_context: n_outputs_max_per_seq = 1 llama_context: n_ctx_seq (8448) > n_ctx_train (8192) -- possible training context overflow ggml_metal_init: allocating ggml_metal_init: found device: Apple M1 Pro ggml_metal_init: picking default device: Apple M1 Pro ggml_metal_init: use fusion = true ggml_metal_init: use concurrency = true ggml_metal_init: use graph optimize = true set_abort_callback: call llama_context: CPU output buffer size = 0.49 MiB llama_kv_cache: layer 0: dev = MTL0 llama_kv_cache: layer 1: dev = MTL0 llama_kv_cache: layer 2: dev = MTL0 llama_kv_cache: layer 3: dev = MTL0 llama_kv_cache: layer 4: dev = MTL0 llama_kv_cache: layer 5: dev = MTL0 llama_kv_cache: layer 6: dev = MTL0 llama_kv_cache: layer 7: dev = MTL0 llama_kv_cache: layer 8: dev = MTL0 llama_kv_cache: layer 9: dev = MTL0 llama_kv_cache: layer 10: dev = MTL0 llama_kv_cache: layer 11: dev = MTL0 llama_kv_cache: layer 12: dev = MTL0 llama_kv_cache: layer 13: dev = MTL0 llama_kv_cache: layer 14: dev = MTL0 llama_kv_cache: layer 15: dev = MTL0 llama_kv_cache: layer 16: dev = MTL0 llama_kv_cache: layer 17: dev = MTL0 llama_kv_cache: layer 18: dev = MTL0 llama_kv_cache: layer 19: dev = MTL0 llama_kv_cache: layer 20: dev = MTL0 llama_kv_cache: layer 21: dev = MTL0 llama_kv_cache: layer 22: dev = MTL0 llama_kv_cache: layer 23: dev = MTL0 llama_kv_cache: layer 24: dev = MTL0 llama_kv_cache: layer 25: dev = MTL0 llama_kv_cache: layer 26: dev = MTL0 llama_kv_cache: layer 27: dev = MTL0 llama_kv_cache: layer 28: dev = MTL0 llama_kv_cache: layer 29: dev = MTL0 llama_kv_cache: layer 30: dev = MTL0 llama_kv_cache: layer 31: dev = MTL0 llama_kv_cache: MTL0 KV buffer size = 1056.00 MiB llama_kv_cache: size = 1056.00 MiB ( 8448 cells, 32 layers, 1/1 seqs), K (f16): 528.00 MiB, V (f16): 528.00 MiB llama_kv_cache: attn_rot_k = 0, n_embd_head_k_all = 128 llama_kv_cache: attn_rot_v = 0, n_embd_head_k_all = 128 llama_context: enumerating backends llama_context: backend_ptrs.size() = 3 sched_reserve: reserving ... sched_reserve: max_nodes = 2328 sched_reserve: reserving full memory module sched_reserve: worst-case: n_tokens = 512, n_seqs = 1, n_outputs = 1 graph_reserve: reserving a graph for ubatch with n_tokens = 1, n_seqs = 1, n_outputs = 1 llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 1, 1, 1, 1 }] is used by node 'node_985' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 1, 1, 1, 1 }] is used by node 'node_986' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_21' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_52' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_83' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_114' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_145' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_176' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_207' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_238' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_269' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_300' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_331' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_362' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_393' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_424' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_455' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_486' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_517' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_548' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_579' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_610' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_641' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_672' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_703' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_734' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_765' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_796' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_827' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_858' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_889' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_920' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_951' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_982' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_tokens' [i32, ne = { 1, 1, 1, 1 }] is used by node 'embd' (GET_ROWS) resolve_fused_ops: Flash Attention enabled resolve_fused_ops: resolving fused DeepSeek V4 HC support: graph_reserve: reserving a graph for ubatch with n_tokens = 1, n_seqs = 1, n_outputs = 1 llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 1, 1, 1, 1 }] is used by node 'node_985' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 1, 1, 1, 1 }] is used by node 'node_986' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_21' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_52' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_83' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_114' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_145' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_176' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_207' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_238' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_269' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_300' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_331' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_362' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_393' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_424' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_455' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_486' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_517' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_548' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_579' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_610' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_641' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_672' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_703' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_734' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_765' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_796' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_827' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_858' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_889' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_920' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_951' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_982' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_tokens' [i32, ne = { 1, 1, 1, 1 }] is used by node 'embd' (GET_ROWS) resolve_fused_ops: fused DeepSeek V4 HC pre enabled graph_reserve: reserving a graph for ubatch with n_tokens = 1, n_seqs = 1, n_outputs = 1 llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 1, 1, 1, 1 }] is used by node 'node_985' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 1, 1, 1, 1 }] is used by node 'node_986' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_21' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_52' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_83' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_114' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_145' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_176' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_207' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_238' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_269' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_300' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_331' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_362' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_393' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_424' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_455' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_486' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_517' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_548' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_579' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_610' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_641' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_672' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_703' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_734' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_765' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_796' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_827' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_858' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_889' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_920' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_951' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_982' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_tokens' [i32, ne = { 1, 1, 1, 1 }] is used by node 'embd' (GET_ROWS) resolve_fused_ops: fused DeepSeek V4 HC comb enabled graph_reserve: reserving a graph for ubatch with n_tokens = 1, n_seqs = 1, n_outputs = 1 llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 1, 1, 1, 1 }] is used by node 'node_985' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 1, 1, 1, 1 }] is used by node 'node_986' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_21' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_52' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_83' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_114' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_145' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_176' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_207' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_238' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_269' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_300' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_331' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_362' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_393' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_424' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_455' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_486' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_517' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_548' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_579' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_610' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_641' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_672' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_703' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_734' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_765' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_796' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_827' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_858' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_889' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_920' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_951' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_982' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_tokens' [i32, ne = { 1, 1, 1, 1 }] is used by node 'embd' (GET_ROWS) resolve_fused_ops: fused DeepSeek V4 HC post enabled graph_reserve: reserving a graph for ubatch with n_tokens = 512, n_seqs = 1, n_outputs = 512 llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 512, 1, 1, 1 }] is used by node 'node_985' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 512, 1, 1, 1 }] is used by node 'node_986' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_21' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_52' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_83' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_114' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_145' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_176' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_207' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_238' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_269' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_300' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_331' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_362' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_393' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_424' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_455' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_486' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_517' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_548' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_579' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_610' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_641' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_672' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_703' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_734' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_765' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_796' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_827' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_858' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_889' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_920' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_951' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_982' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_tokens' [i32, ne = { 512, 1, 1, 1 }] is used by node 'embd' (GET_ROWS) graph_reserve: reserving a graph for ubatch with n_tokens = 1, n_seqs = 1, n_outputs = 1 llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 1, 1, 1, 1 }] is used by node 'node_985' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 1, 1, 1, 1 }] is used by node 'node_986' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_21' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_52' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_83' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_114' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_145' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_176' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_207' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_238' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_269' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_300' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_331' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_362' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_393' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_424' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_455' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_486' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_517' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_548' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_579' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_610' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_641' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_672' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_703' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_734' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_765' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_796' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_827' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_858' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_889' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_920' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_951' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 1, 1, 1 }] is used by node 'node_982' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_v_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Qcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 1, 1, 1, 1 }] is used by node 'Kcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 1, 1, 1, 1 }] is used by node 'cache_k_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_tokens' [i32, ne = { 1, 1, 1, 1 }] is used by node 'embd' (GET_ROWS) graph_reserve: reserving a graph for ubatch with n_tokens = 512, n_seqs = 1, n_outputs = 512 llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 512, 1, 1, 1 }] is used by node 'node_985' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' out_ids' [i32, ne = { 512, 1, 1, 1 }] is used by node 'node_986' (GET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_21' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_52' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_83' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_114' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_145' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_176' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_207' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_238' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_269' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_300' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_331' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_362' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_393' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_424' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_455' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_486' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_517' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_548' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_579' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_610' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_641' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_672' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_703' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_734' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_765' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_796' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_827' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_858' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_889' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_920' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_951' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_kq_mask' [f16, ne = { 8448, 512, 1, 1 }] is used by node 'node_982' (FLASH_ATTN_EXT) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_v_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_v_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-0' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-1' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-2' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-3' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-4' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-5' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-6' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-7' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-8' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-9' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-10' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-11' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-12' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-13' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-14' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-15' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-16' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-17' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-18' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-19' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-20' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-21' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-22' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-23' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-24' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-25' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-26' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-27' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-28' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-29' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-30' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Qcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' inp_pos' [i32, ne = { 512, 1, 1, 1 }] is used by node 'Kcur-31' (ROPE) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l0 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l1 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l2 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l3 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l4 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l5 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l6 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l7 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l8 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l9 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l10 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l11 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l12 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l13 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l14 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l15 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l16 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l17 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l18 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l19 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l20 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l21 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l22 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l23 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l24 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l25 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l26 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l27 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l28 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l29 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l30 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' attn_inp_k_idxs' [i64, ne = { 512, 1, 1, 1 }] is used by node 'cache_k_l31 (view)' (SET_ROWS) llama_graph_n_input_tensors: input tensor ' inp_tokens' [i32, ne = { 512, 1, 1, 1 }] is used by node 'embd' (GET_ROWS) sched_reserve: MTL0 compute buffer size = 258.50 MiB sched_reserve: CPU compute buffer size = 24.26 MiB sched_reserve: graph: nodes = 998, splits = 2, input objects = 4, input tensors = 6 sched_reserve: reserve took 18.82 ms, sched copies = 1 attach_threadpool: call set_n_threads: n_threads = 8, n_threads_batch = 8 ggml_metal_library_compile_pipeline: compiling pipeline: base = 'kernel_flash_attn_ext_vec_f16_dk128_dv128', name = 'kernel_flash_attn_ext_vec_f16_dk128_dv128_mask=1_sink=0_bias=0_scap=0_kvpad=0_sparse=0_ns10=1024_ns20=1024_nsg=1_nwg=32' ggml_metal_library_compile_pipeline: loaded kernel_flash_attn_ext_vec_f16_dk128_dv128_mask=1_sink=0_bias=0_scap=0_kvpad=0_sparse=0_ns10=1024_ns20=1024_nsg=1_nwg=32 0x8cdc80c00 | th_max = 448 | th_width = 32 ggml_metal_library_compile_pipeline: compiling pipeline: base = 'kernel_flash_attn_ext_vec_reduce', name = 'kernel_flash_attn_ext_vec_reduce_dv=128_nwg=32' ggml_metal_library_compile_pipeline: loaded kernel_flash_attn_ext_vec_reduce_dv=128_nwg=32 0x8cdc80f00 | th_max = 1024 | th_width = 32 state_read_meta: cell_count = 8192, dest_seq_id = 0 set_n_threads: n_threads = 8, n_threads_batch = 8 ggml_metal_library_compile_pipeline: compiling pipeline: base = 'kernel_flash_attn_ext_vec_f16_dk128_dv128', name = 'kernel_flash_attn_ext_vec_f16_dk128_dv128_mask=1_sink=0_bias=0_scap=0_kvpad=0_sparse=0_ns10=1024_ns20=1024_nsg=4_nwg=32' ggml_metal_library_compile_pipeline: loaded kernel_flash_attn_ext_vec_f16_dk128_dv128_mask=1_sink=0_bias=0_scap=0_kvpad=0_sparse=0_ns10=1024_ns20=1024_nsg=4_nwg=32 0x8cdc81200 | th_max = 448 | th_width = 32 state_read_meta: cell_count = 8192, dest_seq_id = 0 set_n_threads: n_threads = 8, n_threads_batch = 8 state_read_meta: cell_count = 8192, dest_seq_id = 0 set_n_threads: n_threads = 8, n_threads_batch = 8 state_read_meta: cell_count = 8192, dest_seq_id = 0 set_n_threads: n_threads = 8, n_threads_batch = 8 state_read_meta: cell_count = 8192, dest_seq_id = 0 set_n_threads: n_threads = 8, n_threads_batch = 8 llama_perf_context_print: load time = 51156.97 ms llama_perf_context_print: prompt eval time = 0.00 ms / 1 tokens ( 0.00 ms per token, inf tokens per second) llama_perf_context_print: eval time = 0.00 ms / 641 runs ( 0.00 ms per token, inf tokens per second) llama_perf_context_print: total time = 74250.02 ms / 642 tokens llama_perf_context_print: graphs reused = 639 ~llama_context: MTL0 compute buffer size is 258.5000 MiB, matches expectation of 258.5000 MiB ~llama_context: CPU compute buffer size is 24.2637 MiB, matches expectation of 24.2637 MiB ~llama_context: CPU compute buffer size is 0.0000 MiB, matches expectation of 0.0000 MiB ggml_metal_free: deallocating