| { |
| "metadata": { |
| "ParamSize": 445, |
| "ParamBytes": 7969843200.0, |
| "BitsPerParam": 4.500835443779134 |
| }, |
| "records": [ |
| { |
| "dataPath": "params_shard_0.bin", |
| "format": "raw-shard", |
| "nbytes": 388956160, |
| "records": [ |
| { |
| "name": "lm_head.q_weight", |
| "shape": [ |
| 151936, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 388956160, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "1b5214c8b61cb690c23b4bf33a1b050f" |
| }, |
| { |
| "dataPath": "params_shard_1.bin", |
| "format": "raw-shard", |
| "nbytes": 48619520, |
| "records": [ |
| { |
| "name": "lm_head.q_scale", |
| "shape": [ |
| 151936, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 48619520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "461317ddd8b685fe4a15917e1d9cf556" |
| }, |
| { |
| "dataPath": "params_shard_2.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.37.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "e2d824dd4d3f12eee9675a79e0813297" |
| }, |
| { |
| "dataPath": "params_shard_3.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.37.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "f9e771098bf366a928d12aa44e4a0225" |
| }, |
| { |
| "dataPath": "params_shard_4.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.37.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "aa2ce8f981d9d979cf791d3230864ac7" |
| }, |
| { |
| "dataPath": "params_shard_5.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.38.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "6ad22f33746633f0cece17d6eba2bf7c" |
| }, |
| { |
| "dataPath": "params_shard_6.bin", |
| "format": "raw-shard", |
| "nbytes": 32870400, |
| "records": [ |
| { |
| "name": "model.layers.37.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.37.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 10240 |
| }, |
| { |
| "name": "model.layers.37.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 4392960 |
| }, |
| { |
| "name": "model.layers.37.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 13158400 |
| }, |
| { |
| "name": "model.layers.37.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 13168640 |
| }, |
| { |
| "name": "model.layers.37.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 13199360 |
| }, |
| { |
| "name": "model.layers.37.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 18114560 |
| }, |
| { |
| "name": "model.layers.37.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 31221760 |
| }, |
| { |
| "name": "model.layers.38.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 32860160 |
| } |
| ], |
| "md5sum": "845e6055059aafc42c76145c487bec42" |
| }, |
| { |
| "dataPath": "params_shard_7.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.38.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "c6c8ad692ac9f588bc0177a5e3df72a2" |
| }, |
| { |
| "dataPath": "params_shard_8.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.38.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "db1af84480584136506bbd2f9cb03238" |
| }, |
| { |
| "dataPath": "params_shard_9.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.39.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "b4f26b2cb8faae5529afe518575fb876" |
| }, |
| { |
| "dataPath": "params_shard_10.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.38.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.38.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 4382720 |
| }, |
| { |
| "name": "model.layers.38.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 13148160 |
| }, |
| { |
| "name": "model.layers.38.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 13158400 |
| }, |
| { |
| "name": "model.layers.38.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 13189120 |
| }, |
| { |
| "name": "model.layers.38.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 18104320 |
| }, |
| { |
| "name": "model.layers.38.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 31211520 |
| }, |
| { |
| "name": "model.layers.39.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 32849920 |
| } |
| ], |
| "md5sum": "fd0ed8c2574cb956aaf0993abdca9f8a" |
| }, |
| { |
| "dataPath": "params_shard_11.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.39.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "fbcc50892fd30cf0914cbd955b045d2c" |
| }, |
| { |
| "dataPath": "params_shard_12.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.39.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "9193d0ab96b1a4bc53b98f57d72cd901" |
| }, |
| { |
| "dataPath": "params_shard_13.bin", |
| "format": "raw-shard", |
| "nbytes": 388956160, |
| "records": [ |
| { |
| "name": "model.embed_tokens.q_weight", |
| "shape": [ |
| 151936, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 388956160, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "978aa5a9c9b879ce22da0224cdd6ab4c" |
| }, |
| { |
| "dataPath": "params_shard_14.bin", |
| "format": "raw-shard", |
| "nbytes": 48619520, |
| "records": [ |
| { |
| "name": "model.embed_tokens.q_scale", |
| "shape": [ |
| 151936, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 48619520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "dd355a54044598e3851479fc62ab3020" |
| }, |
| { |
| "dataPath": "params_shard_15.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.0.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "0eef57244ea726a2b102836b8e00cd69" |
| }, |
| { |
| "dataPath": "params_shard_16.bin", |
| "format": "raw-shard", |
| "nbytes": 32870400, |
| "records": [ |
| { |
| "name": "model.layers.39.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.39.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 4382720 |
| }, |
| { |
| "name": "model.layers.39.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 13148160 |
| }, |
| { |
| "name": "model.layers.39.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 13158400 |
| }, |
| { |
| "name": "model.layers.39.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 13189120 |
| }, |
| { |
| "name": "model.layers.39.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 18104320 |
| }, |
| { |
| "name": "model.layers.39.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 31211520 |
| }, |
| { |
| "name": "model.norm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 32849920 |
| }, |
| { |
| "name": "model.layers.0.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 32860160 |
| } |
| ], |
| "md5sum": "8b4cd64c943a00314bf7a6b28c50e761" |
| }, |
| { |
| "dataPath": "params_shard_17.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.0.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "0eb76c35f7bf4e2eddf15b1910f7953d" |
| }, |
| { |
| "dataPath": "params_shard_18.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.0.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "d66b2c0ae05db75e39b72fd7130802c4" |
| }, |
| { |
| "dataPath": "params_shard_19.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.1.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "9431cbed4f544a28d9da8d0e46b94cd8" |
| }, |
| { |
| "dataPath": "params_shard_20.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.0.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.0.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 4382720 |
| }, |
| { |
| "name": "model.layers.0.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 13148160 |
| }, |
| { |
| "name": "model.layers.0.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 13158400 |
| }, |
| { |
| "name": "model.layers.0.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 13189120 |
| }, |
| { |
| "name": "model.layers.0.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 18104320 |
| }, |
| { |
| "name": "model.layers.0.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 31211520 |
| }, |
| { |
| "name": "model.layers.1.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 32849920 |
| } |
| ], |
| "md5sum": "20911221ae915a2c8290d524f18353c4" |
| }, |
| { |
| "dataPath": "params_shard_21.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.1.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "e31af9a0f91bb82d21e331a9d4bd3a4b" |
| }, |
| { |
| "dataPath": "params_shard_22.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.1.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "b88e202202ecbea474ddc8f338b9d7c8" |
| }, |
| { |
| "dataPath": "params_shard_23.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.2.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "6cefa67ff9e2b68562ce8fb62198c677" |
| }, |
| { |
| "dataPath": "params_shard_24.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.1.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.1.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 4382720 |
| }, |
| { |
| "name": "model.layers.1.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 13148160 |
| }, |
| { |
| "name": "model.layers.1.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 13158400 |
| }, |
| { |
| "name": "model.layers.1.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 13189120 |
| }, |
| { |
| "name": "model.layers.1.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 18104320 |
| }, |
| { |
| "name": "model.layers.1.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 31211520 |
| }, |
| { |
| "name": "model.layers.2.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 32849920 |
| } |
| ], |
| "md5sum": "ee567f243298c75040bf568adb05849b" |
| }, |
| { |
| "dataPath": "params_shard_25.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.2.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "121006c8c7a2365899f07314eef2970a" |
| }, |
| { |
| "dataPath": "params_shard_26.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.2.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "e13975e83d54977490e251d543d79c40" |
| }, |
| { |
| "dataPath": "params_shard_27.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.3.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "00c5c7ac918b27c2a8e91a78e80ad879" |
| }, |
| { |
| "dataPath": "params_shard_28.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.2.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.2.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 4382720 |
| }, |
| { |
| "name": "model.layers.2.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 13148160 |
| }, |
| { |
| "name": "model.layers.2.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 13158400 |
| }, |
| { |
| "name": "model.layers.2.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 13189120 |
| }, |
| { |
| "name": "model.layers.2.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 18104320 |
| }, |
| { |
| "name": "model.layers.2.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 31211520 |
| }, |
| { |
| "name": "model.layers.3.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 32849920 |
| } |
| ], |
| "md5sum": "ef6d3da09858aa612276cf6dcb999e7a" |
| }, |
| { |
| "dataPath": "params_shard_29.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.3.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "10d00efcc577158dc7c47295765a2427" |
| }, |
| { |
| "dataPath": "params_shard_30.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.3.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "d433ee8b5d62bbe8c3bea9d87a52cb5c" |
| }, |
| { |
| "dataPath": "params_shard_31.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.4.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "9c1afcbe1cf0bf041e29350468cb4b03" |
| }, |
| { |
| "dataPath": "params_shard_32.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.3.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.3.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 4382720 |
| }, |
| { |
| "name": "model.layers.3.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 13148160 |
| }, |
| { |
| "name": "model.layers.3.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 13158400 |
| }, |
| { |
| "name": "model.layers.3.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 13189120 |
| }, |
| { |
| "name": "model.layers.3.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 18104320 |
| }, |
| { |
| "name": "model.layers.3.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 31211520 |
| }, |
| { |
| "name": "model.layers.4.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 32849920 |
| } |
| ], |
| "md5sum": "6c56554a4f4aad88a9b23a05be3558ca" |
| }, |
| { |
| "dataPath": "params_shard_33.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.4.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "7a46e8ed7370aca1248fc69a848fe115" |
| }, |
| { |
| "dataPath": "params_shard_34.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.4.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "337c0258d86a46c0c18a523be103382a" |
| }, |
| { |
| "dataPath": "params_shard_35.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.5.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "0a02be9c901d3d78986d6e13d78f5fef" |
| }, |
| { |
| "dataPath": "params_shard_36.bin", |
| "format": "raw-shard", |
| "nbytes": 32880640, |
| "records": [ |
| { |
| "name": "model.layers.4.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.4.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 4382720 |
| }, |
| { |
| "name": "model.layers.4.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 13148160 |
| }, |
| { |
| "name": "model.layers.4.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 13158400 |
| }, |
| { |
| "name": "model.layers.4.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 13189120 |
| }, |
| { |
| "name": "model.layers.4.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 18104320 |
| }, |
| { |
| "name": "model.layers.4.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 31211520 |
| }, |
| { |
| "name": "model.layers.5.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 32849920 |
| } |
| ], |
| "md5sum": "86cea78feb99552aceb672b696272fc8" |
| }, |
| { |
| "dataPath": "params_shard_37.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.10.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "c5a4397f52cbb1d069275e3a80ca3bcf" |
| }, |
| { |
| "dataPath": "params_shard_38.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.10.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "94a48dfe2c5095c9db91a3f8c94c191c" |
| }, |
| { |
| "dataPath": "params_shard_39.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.10.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "f47d42ca076ed78cdc01fff24ca73b0e" |
| }, |
| { |
| "dataPath": "params_shard_40.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.5.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.5.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 4915200 |
| }, |
| { |
| "name": "model.layers.5.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 18022400 |
| }, |
| { |
| "name": "model.layers.10.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 19660800 |
| }, |
| { |
| "name": "model.layers.10.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 19671040 |
| }, |
| { |
| "name": "model.layers.10.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 24053760 |
| }, |
| { |
| "name": "model.layers.10.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 32819200 |
| }, |
| { |
| "name": "model.layers.10.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 32829440 |
| } |
| ], |
| "md5sum": "b1714b9ae063d1c17466793ae142a84c" |
| }, |
| { |
| "dataPath": "params_shard_41.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.11.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "5c6b57d1cfb1551fac6d82848755c639" |
| }, |
| { |
| "dataPath": "params_shard_42.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.11.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "36369d1c512ede13b1d5bcfa1c434216" |
| }, |
| { |
| "dataPath": "params_shard_43.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.11.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "277147ca82ed47f424c9a083815d49b3" |
| }, |
| { |
| "dataPath": "params_shard_44.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.10.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.10.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 4915200 |
| }, |
| { |
| "name": "model.layers.10.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 18022400 |
| }, |
| { |
| "name": "model.layers.11.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 19660800 |
| }, |
| { |
| "name": "model.layers.11.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 19671040 |
| }, |
| { |
| "name": "model.layers.11.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 24053760 |
| }, |
| { |
| "name": "model.layers.11.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 32819200 |
| }, |
| { |
| "name": "model.layers.11.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 32829440 |
| } |
| ], |
| "md5sum": "18db32a427c50cab74375bdf28c940ad" |
| }, |
| { |
| "dataPath": "params_shard_45.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.12.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "c990c9654d8fbf390be459426e23b2b8" |
| }, |
| { |
| "dataPath": "params_shard_46.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.12.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "19f5790c6b46351016e6e905c6fd46fd" |
| }, |
| { |
| "dataPath": "params_shard_47.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.12.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "c7eade0f44fe2cb9221b62775ec79c67" |
| }, |
| { |
| "dataPath": "params_shard_48.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.11.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.11.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 4915200 |
| }, |
| { |
| "name": "model.layers.11.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 18022400 |
| }, |
| { |
| "name": "model.layers.12.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 19660800 |
| }, |
| { |
| "name": "model.layers.12.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 19671040 |
| }, |
| { |
| "name": "model.layers.12.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 24053760 |
| }, |
| { |
| "name": "model.layers.12.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 32819200 |
| }, |
| { |
| "name": "model.layers.12.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 32829440 |
| } |
| ], |
| "md5sum": "a3f39d982eb3c5aa4dd527869e608ce6" |
| }, |
| { |
| "dataPath": "params_shard_49.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.13.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "d5171e7105119278bad78a8837fccae0" |
| }, |
| { |
| "dataPath": "params_shard_50.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.5.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "77520211974440bcb93a0c2fe1dca376" |
| }, |
| { |
| "dataPath": "params_shard_51.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.5.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "0b32a0edf0010b1e5cdbc9b1437116ba" |
| }, |
| { |
| "dataPath": "params_shard_52.bin", |
| "format": "raw-shard", |
| "nbytes": 28999680, |
| "records": [ |
| { |
| "name": "model.layers.12.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.12.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 4915200 |
| }, |
| { |
| "name": "model.layers.12.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 18022400 |
| }, |
| { |
| "name": "model.layers.13.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 19660800 |
| }, |
| { |
| "name": "model.layers.13.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 19691520 |
| }, |
| { |
| "name": "model.layers.5.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 24606720 |
| }, |
| { |
| "name": "model.layers.5.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 24616960 |
| } |
| ], |
| "md5sum": "70f57d5bec8a558bd2ebce2bbdecd05e" |
| }, |
| { |
| "dataPath": "params_shard_53.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.6.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "5dae845278a3d103969ef8d7cd064902" |
| }, |
| { |
| "dataPath": "params_shard_54.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.6.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "b8f88df303cc5b659c72f49fa3c40839" |
| }, |
| { |
| "dataPath": "params_shard_55.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.6.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "36bb81bcc1fc5587f1848faaf88598b1" |
| }, |
| { |
| "dataPath": "params_shard_56.bin", |
| "format": "raw-shard", |
| "nbytes": 26890240, |
| "records": [ |
| { |
| "name": "model.layers.5.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.5.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 8765440 |
| }, |
| { |
| "name": "model.layers.6.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 8775680 |
| }, |
| { |
| "name": "model.layers.6.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 8785920 |
| }, |
| { |
| "name": "model.layers.6.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 13168640 |
| }, |
| { |
| "name": "model.layers.6.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 21934080 |
| }, |
| { |
| "name": "model.layers.6.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 21944320 |
| }, |
| { |
| "name": "model.layers.6.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 21975040 |
| } |
| ], |
| "md5sum": "4bd4d294610774f3f9991bdf95b4881b" |
| }, |
| { |
| "dataPath": "params_shard_57.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.7.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "8f6e9c9649ebeeed0618533801d28ad6" |
| }, |
| { |
| "dataPath": "params_shard_58.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.7.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "5805d30d8080b53aa7ddf6f180050ceb" |
| }, |
| { |
| "dataPath": "params_shard_59.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.7.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "58f794c30635c6a42a38139d7082523d" |
| }, |
| { |
| "dataPath": "params_shard_60.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.6.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.6.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.7.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.7.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.7.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.7.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.7.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.7.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "e94ebe391c23edb1a2c13e057518a55b" |
| }, |
| { |
| "dataPath": "params_shard_61.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.8.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "9725f58b1552a6b6b97ac56f6458a95e" |
| }, |
| { |
| "dataPath": "params_shard_62.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.8.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "c8feee48f1d7270c0e3562b3962ae272" |
| }, |
| { |
| "dataPath": "params_shard_63.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.8.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "1ec9f54cfa2aa69c16e4d799e3c48ed7" |
| }, |
| { |
| "dataPath": "params_shard_64.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.7.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.7.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.8.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.8.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.8.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.8.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.8.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.8.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "cd64b6c3f7f30893271692f3628848b7" |
| }, |
| { |
| "dataPath": "params_shard_65.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.9.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "0146058c589d374c3352f11e0f531b4b" |
| }, |
| { |
| "dataPath": "params_shard_66.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.9.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "41a0939dbb930801ab5cf2f67d8c7fa4" |
| }, |
| { |
| "dataPath": "params_shard_67.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.9.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "90ded6426f0cdbcdf6c3a6db4166c420" |
| }, |
| { |
| "dataPath": "params_shard_68.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.8.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.8.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.9.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.9.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.9.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.9.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.9.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.9.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "7a4268f8a14e2457ae082da579cf2d29" |
| }, |
| { |
| "dataPath": "params_shard_69.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.13.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "8ed97b937a1a3a26a5cf5c6e1f82bf1d" |
| }, |
| { |
| "dataPath": "params_shard_70.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.13.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "dc2dcabf2de397048cda325a63ba82f8" |
| }, |
| { |
| "dataPath": "params_shard_71.bin", |
| "format": "raw-shard", |
| "nbytes": 27914240, |
| "records": [ |
| { |
| "name": "model.layers.9.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.9.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.13.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.13.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.13.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.13.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| } |
| ], |
| "md5sum": "53e38b4ca72aef1acd483d8d6a1ab092" |
| }, |
| { |
| "dataPath": "params_shard_72.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.14.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "424e8f833833a9050ba41477725463b6" |
| }, |
| { |
| "dataPath": "params_shard_73.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.14.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "e643296086945b9b94d95fa18fb1081b" |
| }, |
| { |
| "dataPath": "params_shard_74.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.14.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "86ed296e38b8bc7b8eee33df00ac8703" |
| }, |
| { |
| "dataPath": "params_shard_75.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.13.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.13.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.14.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.14.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.14.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.14.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.14.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.14.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "c97958f525007d89490b0e21121321d9" |
| }, |
| { |
| "dataPath": "params_shard_76.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.15.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "da22983e1a16c40b9ce6cc9b8b1833c5" |
| }, |
| { |
| "dataPath": "params_shard_77.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.15.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "0800acbe944c59cf208e828e1b497918" |
| }, |
| { |
| "dataPath": "params_shard_78.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.15.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "dbdd17ef34e815e1c52debb7062a7a75" |
| }, |
| { |
| "dataPath": "params_shard_79.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.14.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.14.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.15.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.15.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.15.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.15.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.15.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.15.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "941eb223e9932f89293c558e1f9cb6d8" |
| }, |
| { |
| "dataPath": "params_shard_80.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.16.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "4eb59e941eabb7eab99f1e4f77de08a8" |
| }, |
| { |
| "dataPath": "params_shard_81.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.16.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "c7268c81bd8023fbe2b1e340a5a84e12" |
| }, |
| { |
| "dataPath": "params_shard_82.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.16.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "3c8cd792518a8504f4926e67159d322a" |
| }, |
| { |
| "dataPath": "params_shard_83.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.15.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.15.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.16.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.16.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.16.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.16.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.16.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.16.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "79e3992f0a59810a62a81baa7f3abf59" |
| }, |
| { |
| "dataPath": "params_shard_84.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.17.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "8de4950b716869ac0b255c7879a5b849" |
| }, |
| { |
| "dataPath": "params_shard_85.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.17.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "59b5cca8e21cc13e6fc2966c75a26044" |
| }, |
| { |
| "dataPath": "params_shard_86.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.17.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "88ae25e3d3b9572562e8107579e640f0" |
| }, |
| { |
| "dataPath": "params_shard_87.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.16.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.16.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.17.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.17.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.17.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.17.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.17.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.17.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "8042ccaa9ffc4356e73e1e17500688d6" |
| }, |
| { |
| "dataPath": "params_shard_88.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.18.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "0fce05773a216f1a0708701eef54ce3d" |
| }, |
| { |
| "dataPath": "params_shard_89.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.18.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "fd116533b01b19d5c3215382060460ce" |
| }, |
| { |
| "dataPath": "params_shard_90.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.18.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "5f933500f18f99f4bb8d31e811dbdc84" |
| }, |
| { |
| "dataPath": "params_shard_91.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.17.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.17.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.18.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.18.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.18.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.18.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.18.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.18.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "c30472d8a5896f7a33756b83791af614" |
| }, |
| { |
| "dataPath": "params_shard_92.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.19.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "5f7368ddbb040465d1c656e3afb01fe2" |
| }, |
| { |
| "dataPath": "params_shard_93.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.19.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "2bde3b796d90cf9b4ddcc3a7065e0c0c" |
| }, |
| { |
| "dataPath": "params_shard_94.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.19.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "19c46f9a158f24269876f095103e22bc" |
| }, |
| { |
| "dataPath": "params_shard_95.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.18.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.18.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.19.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.19.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.19.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.19.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.19.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.19.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "01f312bb4e6a350857e3d24d4729c5bd" |
| }, |
| { |
| "dataPath": "params_shard_96.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.20.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "096f9488e2a6376fa459dd7485ff501a" |
| }, |
| { |
| "dataPath": "params_shard_97.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.20.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "05df468fda12be5ea509e0609af329bc" |
| }, |
| { |
| "dataPath": "params_shard_98.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.20.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "43031118e34f7d816f66e671fd2c592a" |
| }, |
| { |
| "dataPath": "params_shard_99.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.19.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.19.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.20.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.20.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.20.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.20.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.20.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.20.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "0a75033c533c6912ff4a5a10f098f798" |
| }, |
| { |
| "dataPath": "params_shard_100.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.21.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "9d77d02866d0dd63af17f52d12270f80" |
| }, |
| { |
| "dataPath": "params_shard_101.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.21.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "3f4e3f2408bd2b6ac09771c21b7ec144" |
| }, |
| { |
| "dataPath": "params_shard_102.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.21.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "f2a68bde6a5eb46ec7e80847e384801e" |
| }, |
| { |
| "dataPath": "params_shard_103.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.20.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.20.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.21.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.21.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 14776320 |
| }, |
| { |
| "name": "model.layers.21.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 19691520 |
| }, |
| { |
| "name": "model.layers.21.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 19701760 |
| }, |
| { |
| "name": "model.layers.21.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 24084480 |
| }, |
| { |
| "name": "model.layers.21.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 32849920 |
| } |
| ], |
| "md5sum": "381feba349d9c709449f29c32ea4dd75" |
| }, |
| { |
| "dataPath": "params_shard_104.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.22.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "e96690aecadc470897cf9885ff0364c0" |
| }, |
| { |
| "dataPath": "params_shard_105.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.22.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "5db1097277a901656c731b42beec88ee" |
| }, |
| { |
| "dataPath": "params_shard_106.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.22.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "fc3e01beb44fde223ab65e0f20a8261a" |
| }, |
| { |
| "dataPath": "params_shard_107.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.21.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.21.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.22.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.22.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.22.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.22.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.22.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.22.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "3e117e015f0a1d44f794b275c2aa6ec4" |
| }, |
| { |
| "dataPath": "params_shard_108.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.23.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "a16d92fe81abdfea919972090b4c2ec4" |
| }, |
| { |
| "dataPath": "params_shard_109.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.23.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "3870223a517fdc4ba216f69e0b7b5f7b" |
| }, |
| { |
| "dataPath": "params_shard_110.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.23.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "3f8334671728da14050bd88634213cd8" |
| }, |
| { |
| "dataPath": "params_shard_111.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.22.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.22.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.23.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.23.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.23.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.23.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.23.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.23.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "225ada6a6556d7ee1056f71b6a8b7061" |
| }, |
| { |
| "dataPath": "params_shard_112.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.24.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "170d0533228d236dd64435d4653bd6e0" |
| }, |
| { |
| "dataPath": "params_shard_113.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.24.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "acc84a984894963ca3608ce3ca966a70" |
| }, |
| { |
| "dataPath": "params_shard_114.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.24.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "cd76e1820afb7e720c248438a806c560" |
| }, |
| { |
| "dataPath": "params_shard_115.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.23.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.23.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.24.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.24.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.24.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.24.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.24.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.24.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "ff0dd590392e2ffcf4ae3391200ba706" |
| }, |
| { |
| "dataPath": "params_shard_116.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.25.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "819537b48c7c480abe7ed1762994fb92" |
| }, |
| { |
| "dataPath": "params_shard_117.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.25.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "4a77987892823940b0a5e5e3459277c8" |
| }, |
| { |
| "dataPath": "params_shard_118.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.25.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "3e17899c7a5d5430bfde317e08f1c63c" |
| }, |
| { |
| "dataPath": "params_shard_119.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.24.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.24.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.25.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.25.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.25.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.25.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.25.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.25.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "472e0bed1843cf201168ce7ece4ca5a6" |
| }, |
| { |
| "dataPath": "params_shard_120.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.26.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "96685b74c57e8a5e41a9c8c88243ce63" |
| }, |
| { |
| "dataPath": "params_shard_121.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.26.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "a1d3eee1c85ca56ce84bab46a14b20d7" |
| }, |
| { |
| "dataPath": "params_shard_122.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.26.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "0135ac7306a8aee9ed888f755a181f4d" |
| }, |
| { |
| "dataPath": "params_shard_123.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.25.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.25.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.26.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.26.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.26.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.26.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.26.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.26.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "d253e6f764b2bf03873d33898f2f8750" |
| }, |
| { |
| "dataPath": "params_shard_124.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.27.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "db8cdf71dcf99dc7748a60380cfccd6c" |
| }, |
| { |
| "dataPath": "params_shard_125.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.27.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "16ae459b3f0a35834c1f828ffd0ec7b5" |
| }, |
| { |
| "dataPath": "params_shard_126.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.27.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "14227a3bcaf3cb32a533416587596f43" |
| }, |
| { |
| "dataPath": "params_shard_127.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.26.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.26.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.27.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.27.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.27.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.27.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.27.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.27.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "c160cb8a05a1b5ba1980381fdc54b79b" |
| }, |
| { |
| "dataPath": "params_shard_128.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.28.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "dd82223d1cfe91eb333ec5b3da8b90ee" |
| }, |
| { |
| "dataPath": "params_shard_129.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.28.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "95b9e22a200452a839d6188bb1c020cb" |
| }, |
| { |
| "dataPath": "params_shard_130.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.28.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "6fdc353db1a61817dd2f56d13b0d63ff" |
| }, |
| { |
| "dataPath": "params_shard_131.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.27.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.27.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.28.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.28.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.28.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.28.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.28.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.28.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "180c84b4710da5efd0f0e1a3469a4960" |
| }, |
| { |
| "dataPath": "params_shard_132.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.29.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "0dacf38df2b63c4c2fe52c9bf08ea33b" |
| }, |
| { |
| "dataPath": "params_shard_133.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.29.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "33197af472c5923647d09958d72c191f" |
| }, |
| { |
| "dataPath": "params_shard_134.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.29.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "265309d97d510f23d7f369bf7b56da30" |
| }, |
| { |
| "dataPath": "params_shard_135.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.28.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.28.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.29.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.29.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 14776320 |
| }, |
| { |
| "name": "model.layers.29.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 19691520 |
| }, |
| { |
| "name": "model.layers.29.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 19701760 |
| }, |
| { |
| "name": "model.layers.29.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 24084480 |
| }, |
| { |
| "name": "model.layers.29.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 32849920 |
| } |
| ], |
| "md5sum": "2781d1d40daa5967469459b116654266" |
| }, |
| { |
| "dataPath": "params_shard_136.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.30.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "49f0170ed2ca16ad4b59ec1d1b79e423" |
| }, |
| { |
| "dataPath": "params_shard_137.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.30.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "8223c14fc150a39ee5f9df27cde3fd5f" |
| }, |
| { |
| "dataPath": "params_shard_138.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.30.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "1ab3b77a85ba4b97cae696f9aa9b0cf0" |
| }, |
| { |
| "dataPath": "params_shard_139.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.29.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.29.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.30.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.30.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.30.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.30.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.30.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.30.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "2772447562f354821e2fa06f8600171f" |
| }, |
| { |
| "dataPath": "params_shard_140.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.31.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "823b150b2a83d6d09e998c9f833e4190" |
| }, |
| { |
| "dataPath": "params_shard_141.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.31.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "39b6e8de6a163d61979496dc125ac327" |
| }, |
| { |
| "dataPath": "params_shard_142.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.31.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "99effd0cbc2b1568c09699864c6ec7d5" |
| }, |
| { |
| "dataPath": "params_shard_143.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.30.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.30.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.31.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.31.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.31.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.31.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.31.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.31.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "a347fd86b0feee429ef79f9c6ca988f7" |
| }, |
| { |
| "dataPath": "params_shard_144.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.32.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "5c39af1fe475e79d915247614da95369" |
| }, |
| { |
| "dataPath": "params_shard_145.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.32.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "ffdffd39626bf41936317197a218a073" |
| }, |
| { |
| "dataPath": "params_shard_146.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.32.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "0ba5dfb2f731184cb5152bfc7a8d2e73" |
| }, |
| { |
| "dataPath": "params_shard_147.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.31.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.31.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.32.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.32.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.32.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.32.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.32.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.32.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "d375c66c9b832d80a84f16cde3004aed" |
| }, |
| { |
| "dataPath": "params_shard_148.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.33.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "dcfbbf843da91af9c9cbf5cafcfde1ce" |
| }, |
| { |
| "dataPath": "params_shard_149.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.33.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "5af78ec368c7808517e74bb373c50690" |
| }, |
| { |
| "dataPath": "params_shard_150.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.33.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "83d0ee166f4aa0a568e61827eebbb345" |
| }, |
| { |
| "dataPath": "params_shard_151.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.32.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.32.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.33.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.33.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.33.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.33.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.33.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.33.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "f50bcfdcfc7690595b7c8781782c4605" |
| }, |
| { |
| "dataPath": "params_shard_152.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.34.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "39f386d1f057e89a5318e3ea682d7fc0" |
| }, |
| { |
| "dataPath": "params_shard_153.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.34.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "11be01fc79de57bdf83c21e7c642bc9a" |
| }, |
| { |
| "dataPath": "params_shard_154.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.34.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "c03c5ebe92076277be0893b6895c8399" |
| }, |
| { |
| "dataPath": "params_shard_155.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.33.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.33.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.34.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.34.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.34.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.34.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.34.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.34.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "6da30f710dc8c8c0d1931c176d3f767c" |
| }, |
| { |
| "dataPath": "params_shard_156.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.35.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "d918e063a42b105edd29a1b3747051c8" |
| }, |
| { |
| "dataPath": "params_shard_157.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.35.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "faf2a8915feb660532a54e64eace8d58" |
| }, |
| { |
| "dataPath": "params_shard_158.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.35.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "b9906872724cb6cca09d08dc126dc548" |
| }, |
| { |
| "dataPath": "params_shard_159.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.34.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.34.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.35.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.35.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.35.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.35.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.35.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.35.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "9757f9b22c2c7b8e5f60ff2f9ee01d30" |
| }, |
| { |
| "dataPath": "params_shard_160.bin", |
| "format": "raw-shard", |
| "nbytes": 35061760, |
| "records": [ |
| { |
| "name": "model.layers.36.mlp.down_proj.q_weight", |
| "shape": [ |
| 5120, |
| 1712 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 35061760, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "303da5caffd7ddc0728ccece4b773cbe" |
| }, |
| { |
| "dataPath": "params_shard_161.bin", |
| "format": "raw-shard", |
| "nbytes": 70123520, |
| "records": [ |
| { |
| "name": "model.layers.36.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 27392, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 70123520, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "70224dd36ea0ed427a76ad6e920ba28d" |
| }, |
| { |
| "dataPath": "params_shard_162.bin", |
| "format": "raw-shard", |
| "nbytes": 39321600, |
| "records": [ |
| { |
| "name": "model.layers.36.self_attn.c_attn.q_weight", |
| "shape": [ |
| 15360, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 39321600, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "2bad836567e256de8e9b788df0c0967e" |
| }, |
| { |
| "dataPath": "params_shard_163.bin", |
| "format": "raw-shard", |
| "nbytes": 32860160, |
| "records": [ |
| { |
| "name": "model.layers.35.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.35.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| }, |
| { |
| "name": "model.layers.36.input_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 14745600 |
| }, |
| { |
| "name": "model.layers.36.mlp.down_proj.q_scale", |
| "shape": [ |
| 5120, |
| 428 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4382720, |
| "byteOffset": 14755840 |
| }, |
| { |
| "name": "model.layers.36.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 27392, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 8765440, |
| "byteOffset": 19138560 |
| }, |
| { |
| "name": "model.layers.36.post_attention_layernorm.weight", |
| "shape": [ |
| 5120 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 10240, |
| "byteOffset": 27904000 |
| }, |
| { |
| "name": "model.layers.36.self_attn.c_attn.bias", |
| "shape": [ |
| 15360 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 30720, |
| "byteOffset": 27914240 |
| }, |
| { |
| "name": "model.layers.36.self_attn.c_attn.q_scale", |
| "shape": [ |
| 15360, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 4915200, |
| "byteOffset": 27944960 |
| } |
| ], |
| "md5sum": "c868345af8c582bf4683f39367373724" |
| }, |
| { |
| "dataPath": "params_shard_164.bin", |
| "format": "raw-shard", |
| "nbytes": 14745600, |
| "records": [ |
| { |
| "name": "model.layers.36.self_attn.o_proj.q_weight", |
| "shape": [ |
| 5120, |
| 640 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 13107200, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.36.self_attn.o_proj.q_scale", |
| "shape": [ |
| 5120, |
| 160 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 1638400, |
| "byteOffset": 13107200 |
| } |
| ], |
| "md5sum": "77dd478178d01ea90767f767053c229f" |
| } |
| ] |
| } |