|
{ |
|
"metadata": { |
|
"ParamSize": 533, |
|
"ParamBytes": 9234108416.0, |
|
"BitsPerParam": 5.001536828453907 |
|
}, |
|
"records": [ |
|
{ |
|
"dataPath": "params_shard_0.bin", |
|
"format": "raw-shard", |
|
"nbytes": 389283840, |
|
"records": [ |
|
{ |
|
"name": "lm_head.q_weight", |
|
"shape": [ |
|
152064, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 389283840, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "21fec9dbca9130f963de9fb5a2600f94" |
|
}, |
|
{ |
|
"dataPath": "params_shard_1.bin", |
|
"format": "raw-shard", |
|
"nbytes": 48660480, |
|
"records": [ |
|
{ |
|
"name": "lm_head.q_scale", |
|
"shape": [ |
|
152064, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 48660480, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f6e1145a931adb39368d3482afc28060" |
|
}, |
|
{ |
|
"dataPath": "params_shard_2.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.42.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "d6b0b9b3c85716761b3c968eca6a51c0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_3.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.42.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "5ac0e57616cce117dd3f0d4c14225f12" |
|
}, |
|
{ |
|
"dataPath": "params_shard_4.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.43.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "8dce9eef02ece331f97cb8b9cdcd356d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_5.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.43.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "4fb99aa9c47d77fc4a7e03c9b8477ae0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_6.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.43.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "c4cdc1471a8d862aa36fdc0a7ebd16e3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_7.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28891136, |
|
"records": [ |
|
{ |
|
"name": "model.layers.42.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.42.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 10240 |
|
}, |
|
{ |
|
"name": "model.layers.42.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 4433920 |
|
}, |
|
{ |
|
"name": "model.layers.42.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 13281280 |
|
}, |
|
{ |
|
"name": "model.layers.43.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 13291520 |
|
}, |
|
{ |
|
"name": "model.layers.43.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 13301760 |
|
}, |
|
{ |
|
"name": "model.layers.43.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 17725440 |
|
}, |
|
{ |
|
"name": "model.layers.43.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 26572800 |
|
}, |
|
{ |
|
"name": "model.layers.43.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 26583040 |
|
}, |
|
{ |
|
"name": "model.layers.43.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 26597376 |
|
} |
|
], |
|
"md5sum": "d60783305f83f83941448e82706ceac1" |
|
}, |
|
{ |
|
"dataPath": "params_shard_8.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.44.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "86e77edd47eafefb8a8456965e6bee29" |
|
}, |
|
{ |
|
"dataPath": "params_shard_9.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.44.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6c53ddb70cfd1aa9e172e48f0c725108" |
|
}, |
|
{ |
|
"dataPath": "params_shard_10.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.44.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "46677454d131ca67618e93b458f91ffa" |
|
}, |
|
{ |
|
"dataPath": "params_shard_11.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"records": [ |
|
{ |
|
"name": "model.layers.43.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.43.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.44.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.44.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.44.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.44.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.44.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.44.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "048a1298997f118cf64e7c9c6d586b17" |
|
}, |
|
{ |
|
"dataPath": "params_shard_12.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.45.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "18dddd0f1919337ebf3afc3003843a4a" |
|
}, |
|
{ |
|
"dataPath": "params_shard_13.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.45.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "29424eba7cc326df5fc52005ef6cda0c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_14.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.45.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "173754727c1255a6dffe87d5e0febf1b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_15.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"records": [ |
|
{ |
|
"name": "model.layers.44.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.44.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.45.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.45.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.45.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.45.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.45.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.45.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "448390dd622190054cee2126d71d2369" |
|
}, |
|
{ |
|
"dataPath": "params_shard_16.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.46.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "a68831bf80170147895585cb628ba20e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_17.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.46.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e94e57ffa1c7b89d7abcdccb93c4d33d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_18.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.46.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "91db90df869f4c39abf891d72a3a0ea2" |
|
}, |
|
{ |
|
"dataPath": "params_shard_19.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"records": [ |
|
{ |
|
"name": "model.layers.45.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.45.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.46.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.46.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.46.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.46.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.46.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.46.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "264bd26b4f77b8915649aaf134d5b294" |
|
}, |
|
{ |
|
"dataPath": "params_shard_20.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.47.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "736275bf52f8223b152e48f7d5c03d3b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_21.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.47.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "534cefca621192268d11f4ae4bd9a5e1" |
|
}, |
|
{ |
|
"dataPath": "params_shard_22.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.47.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "9382ba8e3686f0c608027aef4660caa7" |
|
}, |
|
{ |
|
"dataPath": "params_shard_23.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"records": [ |
|
{ |
|
"name": "model.layers.46.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.46.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.47.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.47.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.47.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.47.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.47.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.47.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "062ef2d15bc7c7476b7281e6452e0ded" |
|
}, |
|
{ |
|
"dataPath": "params_shard_24.bin", |
|
"format": "raw-shard", |
|
"nbytes": 389283840, |
|
"records": [ |
|
{ |
|
"name": "model.embed_tokens.q_weight", |
|
"shape": [ |
|
152064, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 389283840, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "d1847bf1f9f63d5248b392d08c695f9e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_25.bin", |
|
"format": "raw-shard", |
|
"nbytes": 48660480, |
|
"records": [ |
|
{ |
|
"name": "model.embed_tokens.q_scale", |
|
"shape": [ |
|
152064, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 48660480, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "4c8080b110e35084cbc54f94efc7c927" |
|
}, |
|
{ |
|
"dataPath": "params_shard_26.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.0.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "0985eee5735e2e7e2af87a3f85cd5381" |
|
}, |
|
{ |
|
"dataPath": "params_shard_27.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.0.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "0e38df8f97bb5e475b27faf0e6e6aedc" |
|
}, |
|
{ |
|
"dataPath": "params_shard_28.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.0.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "84e0ea205875f3db1fa16b182f9750ec" |
|
}, |
|
{ |
|
"dataPath": "params_shard_29.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30355456, |
|
"records": [ |
|
{ |
|
"name": "model.layers.47.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.47.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.norm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.0.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.0.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14766080 |
|
}, |
|
{ |
|
"name": "model.layers.0.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19189760 |
|
}, |
|
{ |
|
"name": "model.layers.0.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.0.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28047360 |
|
}, |
|
{ |
|
"name": "model.layers.0.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28061696 |
|
} |
|
], |
|
"md5sum": "61924c15b911411baeb4328117582155" |
|
}, |
|
{ |
|
"dataPath": "params_shard_30.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.1.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6d73613d44bb8b08b376db3ddd928954" |
|
}, |
|
{ |
|
"dataPath": "params_shard_31.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.1.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "edaceaa2b19b1de0959babf498c52446" |
|
}, |
|
{ |
|
"dataPath": "params_shard_32.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.1.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "4d7cad6104eedbe3f0b7046cb19224db" |
|
}, |
|
{ |
|
"dataPath": "params_shard_33.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"records": [ |
|
{ |
|
"name": "model.layers.0.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.0.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.1.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.1.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.1.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.1.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.1.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.1.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "44f39fe8f500f4574a68d5ee9709e8ab" |
|
}, |
|
{ |
|
"dataPath": "params_shard_34.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.2.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "22e24085402165170d3c18413acfcfdb" |
|
}, |
|
{ |
|
"dataPath": "params_shard_35.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.2.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "22108c37a8562d73d8d1d47ed33847a2" |
|
}, |
|
{ |
|
"dataPath": "params_shard_36.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.2.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "575e8c85074d36fcb384c2f5e77c5fad" |
|
}, |
|
{ |
|
"dataPath": "params_shard_37.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"records": [ |
|
{ |
|
"name": "model.layers.1.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.1.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.2.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.2.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.2.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.2.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.2.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.2.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "e29a1387d32d06633c9d1f779cda7dfc" |
|
}, |
|
{ |
|
"dataPath": "params_shard_38.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.3.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "55d3bf7db621b6bdfb73a6553c391829" |
|
}, |
|
{ |
|
"dataPath": "params_shard_39.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.3.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "3290e43b8919d28cbc676ac1f53f426c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_40.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.3.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "693aee923db022ce559083bf2bfef7e6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_41.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"records": [ |
|
{ |
|
"name": "model.layers.2.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.2.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.3.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.3.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.3.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.3.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.3.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.3.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "7022cc94afe6e3496fe6aeada766646c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_42.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.4.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "7a9535c42a9b461ba54e8aaa72e8f14b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_43.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.4.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "0828ec5a716ad1b3d81c4ad44ac12d33" |
|
}, |
|
{ |
|
"dataPath": "params_shard_44.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.4.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "da355eaaa48f79c4026f12db81786d60" |
|
}, |
|
{ |
|
"dataPath": "params_shard_45.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"records": [ |
|
{ |
|
"name": "model.layers.3.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.3.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.4.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.4.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.4.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.4.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.4.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.4.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "1a1566e402f13f50f94cea7b25cf8098" |
|
}, |
|
{ |
|
"dataPath": "params_shard_46.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.5.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "077896afc42eca087eaa53daeffd992b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_47.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.5.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "c44168fea3948c10a0c296ebe86a6f96" |
|
}, |
|
{ |
|
"dataPath": "params_shard_48.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.5.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "08a6c6e6b0497e99eb1961681e4a3bff" |
|
}, |
|
{ |
|
"dataPath": "params_shard_49.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"records": [ |
|
{ |
|
"name": "model.layers.4.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.4.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.5.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.5.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.5.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.5.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.5.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.5.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "c03fb366d36f772f6a3db6677c791a68" |
|
}, |
|
{ |
|
"dataPath": "params_shard_50.bin", |
|
"format": "raw-shard", |
|
"nbytes": 33110016, |
|
"records": [ |
|
{ |
|
"name": "model.layers.5.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.5.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.6.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.6.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 14759936 |
|
} |
|
], |
|
"md5sum": "15562e954e417542b466c7ebba736fee" |
|
}, |
|
{ |
|
"dataPath": "params_shard_51.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.10.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "eced1495325f8186d908b3e7529e94d0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_52.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.10.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b57f35e07b2150044beeaf3e93f90fa7" |
|
}, |
|
{ |
|
"dataPath": "params_shard_53.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.10.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "c0a5aa5f3315c7d21675846de81d407e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_54.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32638976, |
|
"records": [ |
|
{ |
|
"name": "model.layers.6.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.6.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 2293760 |
|
}, |
|
{ |
|
"name": "model.layers.6.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 15400960 |
|
}, |
|
{ |
|
"name": "model.layers.10.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 17039360 |
|
}, |
|
{ |
|
"name": "model.layers.10.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 17049600 |
|
}, |
|
{ |
|
"name": "model.layers.10.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 21473280 |
|
}, |
|
{ |
|
"name": "model.layers.10.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 30320640 |
|
}, |
|
{ |
|
"name": "model.layers.10.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 30330880 |
|
}, |
|
{ |
|
"name": "model.layers.10.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 30345216 |
|
} |
|
], |
|
"md5sum": "9cb91dd8ac24e1c5eb7181864b70c6b0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_55.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.11.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "83d70661cecef1a9b1a3152d01626cb3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_56.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.11.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "7d421dfd64ad40d41194dd5c8986598e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_57.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.11.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "71baccb207d951f2d0e9748c21abdce0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_58.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"records": [ |
|
{ |
|
"name": "model.layers.10.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.10.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.11.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.11.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.11.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.11.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.11.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.11.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "2225816834ab1d67bcd520f0de8f5376" |
|
}, |
|
{ |
|
"dataPath": "params_shard_59.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.12.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "a401d69c388a3dfbcefc0ae41144abcc" |
|
}, |
|
{ |
|
"dataPath": "params_shard_60.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.12.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6259aeacdc878dc30476020da693ba81" |
|
}, |
|
{ |
|
"dataPath": "params_shard_61.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.12.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "55375cd6c8f070e84b7fcf3524e6e1a0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_62.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"records": [ |
|
{ |
|
"name": "model.layers.11.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.11.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.12.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.12.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.12.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.12.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.12.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.12.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "f720babd7ddd6550db15278c589d5119" |
|
}, |
|
{ |
|
"dataPath": "params_shard_63.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.13.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "dd4dd53f144a550bea42d65a0dd5710c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_64.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.13.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "3cc2ca2d234d5f5429002466cc547f4c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_65.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.13.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "1f9b31116f28212530d75e22d91c57e3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_66.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"records": [ |
|
{ |
|
"name": "model.layers.12.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.12.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.13.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.13.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.13.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.13.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.13.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.13.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "164a1ad0cd706cea434c0010c0ab8c7d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_67.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.14.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b9fc6d022a499b65b46cd06071b28b38" |
|
}, |
|
{ |
|
"dataPath": "params_shard_68.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.14.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "2a26e6c3d7a9fb4d6f3aa86404835efa" |
|
}, |
|
{ |
|
"dataPath": "params_shard_69.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.14.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "942c9ca736c7022c71628a848827e5be" |
|
}, |
|
{ |
|
"dataPath": "params_shard_70.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.14.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.14.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.14.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.14.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.14.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.14.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "f877d80f85829cfa2ce6f2b4fd745418" |
|
}, |
|
{ |
|
"dataPath": "params_shard_71.bin", |
|
"format": "raw-shard", |
|
"nbytes": 33110016, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.15.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.15.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 14759936 |
|
} |
|
], |
|
"md5sum": "feb4f9de281cc8aa299859bcd88ae195" |
|
}, |
|
{ |
|
"dataPath": "params_shard_72.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.6.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6d138f8845da9b288721ff66a1c2e2fe" |
|
}, |
|
{ |
|
"dataPath": "params_shard_73.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.6.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "c1b4d0d58fb1972c6a1d69b7677b7740" |
|
}, |
|
{ |
|
"dataPath": "params_shard_74.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.7.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "af8cd5d34e5b7c94fc55a6d03540f320" |
|
}, |
|
{ |
|
"dataPath": "params_shard_75.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30341120, |
|
"records": [ |
|
{ |
|
"name": "model.layers.15.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.15.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 2293760 |
|
}, |
|
{ |
|
"name": "model.layers.15.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 15400960 |
|
}, |
|
{ |
|
"name": "model.layers.6.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 17039360 |
|
}, |
|
{ |
|
"name": "model.layers.6.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 17049600 |
|
}, |
|
{ |
|
"name": "model.layers.6.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 21473280 |
|
}, |
|
{ |
|
"name": "model.layers.6.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 30320640 |
|
}, |
|
{ |
|
"name": "model.layers.7.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 30330880 |
|
} |
|
], |
|
"md5sum": "2651e2555eae3a74f9ca314cd25f27f3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_76.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.7.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "9cf3711f1f0b9fca49b3a663ec7441aa" |
|
}, |
|
{ |
|
"dataPath": "params_shard_77.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31645696, |
|
"records": [ |
|
{ |
|
"name": "model.layers.7.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.7.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 4423680 |
|
}, |
|
{ |
|
"name": "model.layers.7.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.7.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 13281280 |
|
}, |
|
{ |
|
"name": "model.layers.7.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 13295616 |
|
} |
|
], |
|
"md5sum": "775440dd1fe7a70913a6886acc6b1cc9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_78.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.8.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "74cb82f04a69562b5e4579bc75b9c097" |
|
}, |
|
{ |
|
"dataPath": "params_shard_79.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.8.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "46fd38ce6ad1a126fcb4bc4bdd5774dd" |
|
}, |
|
{ |
|
"dataPath": "params_shard_80.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.8.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "eaaffa61e9d6b02570fbf9727f8d1943" |
|
}, |
|
{ |
|
"dataPath": "params_shard_81.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32638976, |
|
"records": [ |
|
{ |
|
"name": "model.layers.7.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.7.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 2293760 |
|
}, |
|
{ |
|
"name": "model.layers.7.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 15400960 |
|
}, |
|
{ |
|
"name": "model.layers.8.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 17039360 |
|
}, |
|
{ |
|
"name": "model.layers.8.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 17049600 |
|
}, |
|
{ |
|
"name": "model.layers.8.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 21473280 |
|
}, |
|
{ |
|
"name": "model.layers.8.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 30320640 |
|
}, |
|
{ |
|
"name": "model.layers.8.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 30330880 |
|
}, |
|
{ |
|
"name": "model.layers.8.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 30345216 |
|
} |
|
], |
|
"md5sum": "5dc744b7aa713e0519026f2eba687045" |
|
}, |
|
{ |
|
"dataPath": "params_shard_82.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.9.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e0cae0641ba5830b9b815be1602fc9da" |
|
}, |
|
{ |
|
"dataPath": "params_shard_83.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.9.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "78c475a964cb63753012b4b761d474b9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_84.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.9.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "99f045bb7023be9a884eeead83a7e974" |
|
}, |
|
{ |
|
"dataPath": "params_shard_85.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.9.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.9.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.9.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.9.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.9.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.9.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "eebc4b15673e3f0bf9bdca7e4acf6c45" |
|
}, |
|
{ |
|
"dataPath": "params_shard_86.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.15.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "db46cbfe53b384de5892e4dae0d2a528" |
|
}, |
|
{ |
|
"dataPath": "params_shard_87.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.15.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "93ef9d2939ce610a0f7da0d42db0c642" |
|
}, |
|
{ |
|
"dataPath": "params_shard_88.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.16.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "d000b69430b913806b0c585d0fa72af7" |
|
}, |
|
{ |
|
"dataPath": "params_shard_89.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.16.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b0a41523c558f542560ce0e8ea41a716" |
|
}, |
|
{ |
|
"dataPath": "params_shard_90.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32471040, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.15.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.15.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.15.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.15.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.16.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.16.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 28047360 |
|
} |
|
], |
|
"md5sum": "2a46083f0e1c9f107e7342499bd9abee" |
|
}, |
|
{ |
|
"dataPath": "params_shard_91.bin", |
|
"format": "raw-shard", |
|
"nbytes": 29515776, |
|
"records": [ |
|
{ |
|
"name": "model.layers.16.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.16.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 8847360 |
|
}, |
|
{ |
|
"name": "model.layers.16.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 8857600 |
|
}, |
|
{ |
|
"name": "model.layers.16.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 8871936 |
|
}, |
|
{ |
|
"name": "model.layers.16.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 27222016 |
|
} |
|
], |
|
"md5sum": "016b09807bffd68a92976930b9d8d298" |
|
}, |
|
{ |
|
"dataPath": "params_shard_92.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.17.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e3d9dda56a5c8594e7ceedf7d4d24e14" |
|
}, |
|
{ |
|
"dataPath": "params_shard_93.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.17.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "cdee9aadd972b9e06987610dc6bbc2d5" |
|
}, |
|
{ |
|
"dataPath": "params_shard_94.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.17.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "4edba327093368bbb77240e95a2174f5" |
|
}, |
|
{ |
|
"dataPath": "params_shard_95.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.17.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.17.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.17.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.17.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.17.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.17.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "6e960a3ec8d1596749ba3eb9b239f5f6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_96.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.18.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "12ffcb40bf990074dcaff3773e216ed6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_97.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.18.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e8f79f787548c0328645ffc8e3a66e99" |
|
}, |
|
{ |
|
"dataPath": "params_shard_98.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.18.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "417dec0903bd372d72defed66cada8cb" |
|
}, |
|
{ |
|
"dataPath": "params_shard_99.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.18.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.18.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.18.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.18.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.18.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.18.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "69424ca34f7840db36bf4f73e1a4c012" |
|
}, |
|
{ |
|
"dataPath": "params_shard_100.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.19.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f3445da8fa9a2426837661b6341d166c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_101.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.19.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "2512566ed0ebf67ff6be2400db9d33fd" |
|
}, |
|
{ |
|
"dataPath": "params_shard_102.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.19.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "fa76ac245b52c99013dd3aa8b1dcf997" |
|
}, |
|
{ |
|
"dataPath": "params_shard_103.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.19.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.19.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.19.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.19.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.19.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.19.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "3ba5a7c4cbeefd5519380128dd1f6339" |
|
}, |
|
{ |
|
"dataPath": "params_shard_104.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.20.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "206ac6edfa388b698a30e009f42e6fda" |
|
}, |
|
{ |
|
"dataPath": "params_shard_105.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.20.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "bb00fc2ee7062bb5b0bdd988ac22211d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_106.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.20.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "d3d9c4d81faf86c17eadc1a2d8c8becc" |
|
}, |
|
{ |
|
"dataPath": "params_shard_107.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.20.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.20.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.20.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.20.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.20.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.20.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "736d04a8f5271e5b07ffa8c6f66fc500" |
|
}, |
|
{ |
|
"dataPath": "params_shard_108.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.21.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e4b0e65ac23684136fc971f416c04518" |
|
}, |
|
{ |
|
"dataPath": "params_shard_109.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.21.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6571a990cfa409a67f2e6015ca3a04dc" |
|
}, |
|
{ |
|
"dataPath": "params_shard_110.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.21.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b82dba03c4faa3d5f18091157d8092dd" |
|
}, |
|
{ |
|
"dataPath": "params_shard_111.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.21.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.21.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.21.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.21.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.21.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.21.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "78549688f64c6712e085b98aa53326ad" |
|
}, |
|
{ |
|
"dataPath": "params_shard_112.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.22.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f779451c3cfa33002f44d3cdf004169f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_113.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.22.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "a1c98b6328371bd075cd8ae24862c86d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_114.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.22.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "eb3866802671e07fb64ef41eb15afb76" |
|
}, |
|
{ |
|
"dataPath": "params_shard_115.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.22.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.22.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.22.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.22.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.22.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.22.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "b2832263b6e6e762581125cabc26ca79" |
|
}, |
|
{ |
|
"dataPath": "params_shard_116.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.23.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "10719bd033b7e6ff40633180c8e3206d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_117.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.23.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "0d471400023ba1d6e0e7e338908c5c74" |
|
}, |
|
{ |
|
"dataPath": "params_shard_118.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.23.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "fb0e4c5d64943a7147cdfcb9fe3d8871" |
|
}, |
|
{ |
|
"dataPath": "params_shard_119.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.23.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.23.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.23.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.23.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.23.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.23.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "653e3c73140f57ebfc57a87b7f5a1509" |
|
}, |
|
{ |
|
"dataPath": "params_shard_120.bin", |
|
"format": "raw-shard", |
|
"nbytes": 33110016, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.24.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.24.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 14759936 |
|
} |
|
], |
|
"md5sum": "ff81852804d1ddd7cfa56941570c31ba" |
|
}, |
|
{ |
|
"dataPath": "params_shard_121.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.24.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6e0a1da951b825473de0f1fb9fbab79f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_122.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.24.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "5ff40f6459169f6f98d665b9d7561a24" |
|
}, |
|
{ |
|
"dataPath": "params_shard_123.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.25.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "9bb7e6230a361aaeeff3c8667357af29" |
|
}, |
|
{ |
|
"dataPath": "params_shard_124.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30341120, |
|
"records": [ |
|
{ |
|
"name": "model.layers.24.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.24.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 2293760 |
|
}, |
|
{ |
|
"name": "model.layers.24.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 15400960 |
|
}, |
|
{ |
|
"name": "model.layers.24.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 17039360 |
|
}, |
|
{ |
|
"name": "model.layers.24.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 17049600 |
|
}, |
|
{ |
|
"name": "model.layers.24.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 21473280 |
|
}, |
|
{ |
|
"name": "model.layers.24.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 30320640 |
|
}, |
|
{ |
|
"name": "model.layers.25.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 30330880 |
|
} |
|
], |
|
"md5sum": "39edb461b2729f3c81df2c5c5cd9e504" |
|
}, |
|
{ |
|
"dataPath": "params_shard_125.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.25.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "3ff952f957124d4cf75431b13c840017" |
|
}, |
|
{ |
|
"dataPath": "params_shard_126.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31645696, |
|
"records": [ |
|
{ |
|
"name": "model.layers.25.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.25.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 4423680 |
|
}, |
|
{ |
|
"name": "model.layers.25.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.25.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 13281280 |
|
}, |
|
{ |
|
"name": "model.layers.25.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 13295616 |
|
} |
|
], |
|
"md5sum": "55ee2d855ecc1cae8f63b2f2438be13b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_127.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.26.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "2f8b29731ae04791bd7dc4f125164260" |
|
}, |
|
{ |
|
"dataPath": "params_shard_128.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.26.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "ad16838457be241bc2bd6900b6e33662" |
|
}, |
|
{ |
|
"dataPath": "params_shard_129.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.26.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "afe3ef38c929ec72d0fc30bb699958b1" |
|
}, |
|
{ |
|
"dataPath": "params_shard_130.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32638976, |
|
"records": [ |
|
{ |
|
"name": "model.layers.25.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.25.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 2293760 |
|
}, |
|
{ |
|
"name": "model.layers.25.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 15400960 |
|
}, |
|
{ |
|
"name": "model.layers.26.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 17039360 |
|
}, |
|
{ |
|
"name": "model.layers.26.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 17049600 |
|
}, |
|
{ |
|
"name": "model.layers.26.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 21473280 |
|
}, |
|
{ |
|
"name": "model.layers.26.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 30320640 |
|
}, |
|
{ |
|
"name": "model.layers.26.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 30330880 |
|
}, |
|
{ |
|
"name": "model.layers.26.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 30345216 |
|
} |
|
], |
|
"md5sum": "1828b96cb17aa0113a6407a26e933954" |
|
}, |
|
{ |
|
"dataPath": "params_shard_131.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.27.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "7c2319ed6fb03cf0bd83debca2bbcca0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_132.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.27.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "ed7aef2fc1a130a6daa88cac52056d8d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_133.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.27.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "d831c3cc35860dea0d8b341ffc819e02" |
|
}, |
|
{ |
|
"dataPath": "params_shard_134.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.27.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.27.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.27.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.27.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.27.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.27.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "d6dc0ed8c1d55d5f5f864fb66339d525" |
|
}, |
|
{ |
|
"dataPath": "params_shard_135.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.28.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b5fa22a490c8953c37652d91a6ab116a" |
|
}, |
|
{ |
|
"dataPath": "params_shard_136.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.28.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "4e10ee74d13936236e0f737275cae376" |
|
}, |
|
{ |
|
"dataPath": "params_shard_137.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.28.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "4218104fe894cd1bc4e099ca95a91a54" |
|
}, |
|
{ |
|
"dataPath": "params_shard_138.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.28.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.28.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.28.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.28.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.28.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.28.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "b1a8ce8d0eebd35fb8ece18a921c3f62" |
|
}, |
|
{ |
|
"dataPath": "params_shard_139.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.29.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "a5c3f87825ba98f6f2d397cfa3f5b0a4" |
|
}, |
|
{ |
|
"dataPath": "params_shard_140.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.29.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b4895fa7d8ba921a5e6d070d15f174f8" |
|
}, |
|
{ |
|
"dataPath": "params_shard_141.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.29.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f66d446d5c666f3bb48df96838ec3e77" |
|
}, |
|
{ |
|
"dataPath": "params_shard_142.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.29.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.29.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.29.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.29.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.29.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.29.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "7107c7ea442802df6d2d6e0f29704afd" |
|
}, |
|
{ |
|
"dataPath": "params_shard_143.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.30.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "7cf953395ef60ce3b7880389dfe5cf10" |
|
}, |
|
{ |
|
"dataPath": "params_shard_144.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.30.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "c5a45ef5168f4b7d6170ae0095cd5341" |
|
}, |
|
{ |
|
"dataPath": "params_shard_145.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.30.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "5e3bf5350127a20ce047da9bc35bdac6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_146.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.30.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.30.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.30.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.30.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.30.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.30.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "177549074b4ce442caa8aa3cf90ef925" |
|
}, |
|
{ |
|
"dataPath": "params_shard_147.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.31.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "5034e22deb0c8c33f4deca4f45d3efa4" |
|
}, |
|
{ |
|
"dataPath": "params_shard_148.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.31.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "d67ff8cfafe87ff98d227207a7c3ef25" |
|
}, |
|
{ |
|
"dataPath": "params_shard_149.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.31.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b676dbf85c641047d28baeb7add9f3f5" |
|
}, |
|
{ |
|
"dataPath": "params_shard_150.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.31.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.31.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.31.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.31.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.31.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.31.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "4c311a7910b830482205905cccc15385" |
|
}, |
|
{ |
|
"dataPath": "params_shard_151.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.32.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "d00eb9730fead41279ebe70b314f0a10" |
|
}, |
|
{ |
|
"dataPath": "params_shard_152.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.32.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "d1ac34f88796bb7b10ead9c55e9a84e6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_153.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.32.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "3974c2837f3dfbafb77d555e5a226501" |
|
}, |
|
{ |
|
"dataPath": "params_shard_154.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.32.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.32.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.32.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.32.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.32.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.32.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "add2b25e780dbdf32b1715aed72b9415" |
|
}, |
|
{ |
|
"dataPath": "params_shard_155.bin", |
|
"format": "raw-shard", |
|
"nbytes": 33110016, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.33.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.33.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 14759936 |
|
} |
|
], |
|
"md5sum": "af0f6d6b2976a6c5d1110364b6637f91" |
|
}, |
|
{ |
|
"dataPath": "params_shard_156.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.33.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "a984fa1f9bafcf9bfa3c073f24edc4dc" |
|
}, |
|
{ |
|
"dataPath": "params_shard_157.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.33.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6b1ed33cee0d27d5b1937073213f740c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_158.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.34.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "0631c9875afdd146b983ff6613badf1e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_159.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30341120, |
|
"records": [ |
|
{ |
|
"name": "model.layers.33.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.33.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 2293760 |
|
}, |
|
{ |
|
"name": "model.layers.33.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 15400960 |
|
}, |
|
{ |
|
"name": "model.layers.33.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 17039360 |
|
}, |
|
{ |
|
"name": "model.layers.33.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 17049600 |
|
}, |
|
{ |
|
"name": "model.layers.33.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 21473280 |
|
}, |
|
{ |
|
"name": "model.layers.33.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 30320640 |
|
}, |
|
{ |
|
"name": "model.layers.34.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 30330880 |
|
} |
|
], |
|
"md5sum": "3ea094a6f93ab198d30df31d25e412ee" |
|
}, |
|
{ |
|
"dataPath": "params_shard_160.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.34.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f3377a6840a4ba511024c2dca66868b8" |
|
}, |
|
{ |
|
"dataPath": "params_shard_161.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31645696, |
|
"records": [ |
|
{ |
|
"name": "model.layers.34.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.34.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 4423680 |
|
}, |
|
{ |
|
"name": "model.layers.34.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.34.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 13281280 |
|
}, |
|
{ |
|
"name": "model.layers.34.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 13295616 |
|
} |
|
], |
|
"md5sum": "a1835a643b7a63e800386faf0a1b4fbc" |
|
}, |
|
{ |
|
"dataPath": "params_shard_162.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.35.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "453b76b7d6eb65187b67fdd68f97f18d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_163.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.35.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f5758cd6aa53151387e796f863619cb4" |
|
}, |
|
{ |
|
"dataPath": "params_shard_164.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.35.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "a9827dedc6b6d29456b09ef39a5a89c4" |
|
}, |
|
{ |
|
"dataPath": "params_shard_165.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32638976, |
|
"records": [ |
|
{ |
|
"name": "model.layers.34.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.34.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 2293760 |
|
}, |
|
{ |
|
"name": "model.layers.34.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 15400960 |
|
}, |
|
{ |
|
"name": "model.layers.35.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 17039360 |
|
}, |
|
{ |
|
"name": "model.layers.35.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 17049600 |
|
}, |
|
{ |
|
"name": "model.layers.35.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 21473280 |
|
}, |
|
{ |
|
"name": "model.layers.35.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 30320640 |
|
}, |
|
{ |
|
"name": "model.layers.35.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 30330880 |
|
}, |
|
{ |
|
"name": "model.layers.35.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 30345216 |
|
} |
|
], |
|
"md5sum": "da92ba883759c5185bb9d14c2c5401d1" |
|
}, |
|
{ |
|
"dataPath": "params_shard_166.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.36.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "bc0bca40b15751989caaed235db6a13b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_167.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.36.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "64c69e7a24ac133daaec8551abb6cbbd" |
|
}, |
|
{ |
|
"dataPath": "params_shard_168.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.36.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f3143e7c2a7b782ebbfb363db2bd9b3b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_169.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.36.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.36.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.36.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.36.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.36.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.36.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "9d9dc8a501a51789f40741d385522bf0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_170.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.37.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "85e65053a41abf976e4c32058d16cd84" |
|
}, |
|
{ |
|
"dataPath": "params_shard_171.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.37.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "0bbe50dbfc8c7244c140d7038dfe5e91" |
|
}, |
|
{ |
|
"dataPath": "params_shard_172.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.37.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "a90cffaeaf152ee6c12378aec9868edd" |
|
}, |
|
{ |
|
"dataPath": "params_shard_173.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"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": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.37.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.37.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.37.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.37.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.37.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.37.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "bca6e012ce42eb5eac30ac2dff2f6d41" |
|
}, |
|
{ |
|
"dataPath": "params_shard_174.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.38.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b37ff0a7333b40e0c6ca6e92df357675" |
|
}, |
|
{ |
|
"dataPath": "params_shard_175.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.38.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "79676c3ed23f9c8cec0980b32bddb5ec" |
|
}, |
|
{ |
|
"dataPath": "params_shard_176.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.38.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "2b0c003178fa61575a31259af76c879f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_177.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"records": [ |
|
{ |
|
"name": "model.layers.37.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.37.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.38.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.38.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.38.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.38.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.38.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.38.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "1437e0d6df954f2f861e416415fca90c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_178.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.39.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "63f1d3c8a85d2686712ff2c6d0c4d040" |
|
}, |
|
{ |
|
"dataPath": "params_shard_179.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.39.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "308e8773520c90a88504fa187e96ef00" |
|
}, |
|
{ |
|
"dataPath": "params_shard_180.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.39.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e390ea8e46997b6775068215de25b200" |
|
}, |
|
{ |
|
"dataPath": "params_shard_181.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"records": [ |
|
{ |
|
"name": "model.layers.38.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.38.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.39.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.39.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.39.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.39.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.39.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.39.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "df4ef75283bb5a517695e52a659b3a76" |
|
}, |
|
{ |
|
"dataPath": "params_shard_182.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.40.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "0d0150ec8b2ce062cede337cecddc093" |
|
}, |
|
{ |
|
"dataPath": "params_shard_183.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.40.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "dda518c624c0372f7e0d57c4842a3429" |
|
}, |
|
{ |
|
"dataPath": "params_shard_184.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.40.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "335bb46bb55efad78bb946693fc4b847" |
|
}, |
|
{ |
|
"dataPath": "params_shard_185.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"records": [ |
|
{ |
|
"name": "model.layers.39.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.39.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.40.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.40.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.40.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.40.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.40.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.40.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "35f2dc69e2d4521cf645a01d55ae0ab2" |
|
}, |
|
{ |
|
"dataPath": "params_shard_186.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.41.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "0287f43ad7df8064c3db4b7fcbd608a2" |
|
}, |
|
{ |
|
"dataPath": "params_shard_187.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.41.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "4c643266a33dbf0de87a295d76f26d04" |
|
}, |
|
{ |
|
"dataPath": "params_shard_188.bin", |
|
"format": "raw-shard", |
|
"nbytes": 18350080, |
|
"records": [ |
|
{ |
|
"name": "model.layers.41.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6f94fc5b1539a0dff380d9a4f52196d6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_189.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30345216, |
|
"records": [ |
|
{ |
|
"name": "model.layers.40.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.40.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.41.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.41.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 14755840 |
|
}, |
|
{ |
|
"name": "model.layers.41.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 19179520 |
|
}, |
|
{ |
|
"name": "model.layers.41.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28026880 |
|
}, |
|
{ |
|
"name": "model.layers.41.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28037120 |
|
}, |
|
{ |
|
"name": "model.layers.41.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 28051456 |
|
} |
|
], |
|
"md5sum": "44c5aa4aff5d116501e4b2fac2c11bd5" |
|
}, |
|
{ |
|
"dataPath": "params_shard_190.bin", |
|
"format": "raw-shard", |
|
"nbytes": 33110016, |
|
"records": [ |
|
{ |
|
"name": "model.layers.41.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.41.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
}, |
|
{ |
|
"name": "model.layers.42.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 14745600 |
|
}, |
|
{ |
|
"name": "model.layers.42.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 18350080, |
|
"byteOffset": 14759936 |
|
} |
|
], |
|
"md5sum": "7b4d44d69c89e30b172130dd12c1a1a0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_191.bin", |
|
"format": "raw-shard", |
|
"nbytes": 17039360, |
|
"records": [ |
|
{ |
|
"name": "model.layers.42.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2293760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.42.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 2293760 |
|
}, |
|
{ |
|
"name": "model.layers.42.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 15400960 |
|
} |
|
], |
|
"md5sum": "0984f63d235a587b4b6cdc66c3568ed5" |
|
} |
|
] |
|
} |