|
{ |
|
"metadata": { |
|
"ParamSize": 507, |
|
"ParamBytes": 4175948800.0, |
|
"BitsPerParam": 3.6148726715433237 |
|
}, |
|
"records": [ |
|
{ |
|
"dataPath": "params_shard_0.bin", |
|
"format": "raw-shard", |
|
"nbytes": 368640000, |
|
"records": [ |
|
{ |
|
"name": "model.embed_tokens.q_weight", |
|
"shape": [ |
|
256000, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 368640000, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b5ed923363ccbdb6d0afe47d7c7ee7fc" |
|
}, |
|
{ |
|
"dataPath": "params_shard_1.bin", |
|
"format": "raw-shard", |
|
"nbytes": 46080000, |
|
"records": [ |
|
{ |
|
"name": "model.embed_tokens.q_scale", |
|
"shape": [ |
|
256000, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 46080000, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "0261895d9ab3ada0dc6aaf55ac3d2b0f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_2.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.0.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "7beb018aab4f20e5e387b3616f24dc7f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_3.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28349440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.0.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.0.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 7168 |
|
}, |
|
{ |
|
"name": "model.layers.0.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 20593664 |
|
}, |
|
{ |
|
"name": "model.layers.0.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 23166976 |
|
}, |
|
{ |
|
"name": "model.layers.0.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 28327936 |
|
}, |
|
{ |
|
"name": "model.layers.0.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 28335104 |
|
}, |
|
{ |
|
"name": "model.layers.0.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 28342272 |
|
} |
|
], |
|
"md5sum": "a726bad53d2c000069ad1459fd48376a" |
|
}, |
|
{ |
|
"dataPath": "params_shard_4.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.1.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f356f10b65cad83ea0701cc6b7f2d6db" |
|
}, |
|
{ |
|
"dataPath": "params_shard_5.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.1.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "48aacda4abf27e98376cb4bb88851469" |
|
}, |
|
{ |
|
"dataPath": "params_shard_6.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.0.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.0.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.0.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.0.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.1.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.1.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.1.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.1.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.1.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.1.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "584f694e4dbc65ed6271fe871b76170e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_7.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.2.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "33342c2618b39a0ad657ed065d51ebaa" |
|
}, |
|
{ |
|
"dataPath": "params_shard_8.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.2.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b5df95cc1e75ca0316852c58b3189c8e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_9.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.1.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.1.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.1.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.1.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.2.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.2.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.2.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.2.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.2.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.2.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "8646d103adf79de15fe8c4f6b5036e5c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_10.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.3.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "90911afbc88c3000715e25b1a5bfced3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_11.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.3.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "906427626c15ace8c0ab975ccd1dc58e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_12.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.2.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.2.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.2.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.2.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.3.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.3.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.3.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.3.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.3.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.3.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "957192ab7c408368138fa4e964d6e712" |
|
}, |
|
{ |
|
"dataPath": "params_shard_13.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.4.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e7354ae04d17c7efbd0571ac344b7b66" |
|
}, |
|
{ |
|
"dataPath": "params_shard_14.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.4.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "dd9312bb9f001636c79b22dbcd2fe48f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_15.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.3.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.3.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.3.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.3.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.4.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.4.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.4.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.4.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.4.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.4.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "0eb394f8e3c66211d35d6f7967c9e87f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_16.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.5.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b3580bb009469be6c1d53c55eba9b905" |
|
}, |
|
{ |
|
"dataPath": "params_shard_17.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.5.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "daf869e80a3d483dc4981b11b26be40e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_18.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.4.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.4.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.4.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.4.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.5.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.5.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.5.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.5.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.5.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.5.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "855dde99bb70a5c630f978ba4d1643ac" |
|
}, |
|
{ |
|
"dataPath": "params_shard_19.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.6.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "68129cef18cdd3e8f190d5e664935ad0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_20.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.6.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "2b90af2d198cb07056cd6d432d2dc60b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_21.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.5.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.5.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.5.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.5.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.6.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.6.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.6.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.6.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.6.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.6.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "ad199b3932bb7bf3f5b1bf70ee7d3440" |
|
}, |
|
{ |
|
"dataPath": "params_shard_22.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.7.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "1b197b64a0f9082b6dbc5a1253bc82b6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_23.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25076736, |
|
"records": [ |
|
{ |
|
"name": "model.layers.6.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.6.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.6.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.6.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.7.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 19915776 |
|
} |
|
], |
|
"md5sum": "0f7164911a8fec92c8e1858af6109f57" |
|
}, |
|
{ |
|
"dataPath": "params_shard_24.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.10.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "5e89239525eed82d07f7aef1e767783d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_25.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.10.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e13fbdb9e1f9eed94669cf6f92d24a93" |
|
}, |
|
{ |
|
"dataPath": "params_shard_26.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.7.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.7.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.7.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.7.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.10.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.10.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.10.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.10.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.10.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.10.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "6e61309e7031b79854d92a7c9730b2c9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_27.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.11.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b62e44f06dabb66e94acb9dda020a5a3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_28.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.11.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "cfe27bcc4c961b26fa4929a7e4468557" |
|
}, |
|
{ |
|
"dataPath": "params_shard_29.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.10.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.10.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.10.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.10.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.11.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.11.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.11.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.11.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.11.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.11.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "644cc777384147753532a9695a056750" |
|
}, |
|
{ |
|
"dataPath": "params_shard_30.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.12.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b60a8feadc3ce16463022e56a5bf9d85" |
|
}, |
|
{ |
|
"dataPath": "params_shard_31.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.12.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "237d3778e80052301d8482c0f62f043d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_32.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.11.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.11.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.11.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.11.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.12.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.12.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.12.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.12.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.12.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.12.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "e712886f98819f62ca32a27e21473bca" |
|
}, |
|
{ |
|
"dataPath": "params_shard_33.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.13.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "4873df72ffaa037065070d5d0d5e20a3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_34.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.13.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "9dcd32a83a8ce20cda9b5fe2c3edafc0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_35.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.12.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.12.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.12.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.12.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.13.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.13.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.13.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.13.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.13.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.13.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "337d81795956f2e13b1941b5ce7fde9d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_36.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.14.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "8b2c579c7bba194ac38c15f461660e23" |
|
}, |
|
{ |
|
"dataPath": "params_shard_37.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.14.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "4ced802d3ff3800797209378c912ce99" |
|
}, |
|
{ |
|
"dataPath": "params_shard_38.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.13.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.13.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.13.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.13.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.14.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.14.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.14.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.14.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.14.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.14.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "e08e6956cbdc3e642ff23f61f87b8dab" |
|
}, |
|
{ |
|
"dataPath": "params_shard_39.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.15.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "645d060d3ee8186888f4790947ccc7eb" |
|
}, |
|
{ |
|
"dataPath": "params_shard_40.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.15.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6df21e201ae3b9651502052d3e17b18c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_41.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.14.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.14.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.14.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.14.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.15.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.15.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.15.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.15.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.15.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.15.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "7d3397a45c65fe29de61e569113493d1" |
|
}, |
|
{ |
|
"dataPath": "params_shard_42.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.16.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6735ba7b87f31143812c5805f08f4c58" |
|
}, |
|
{ |
|
"dataPath": "params_shard_43.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.16.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6a5a13b2a922d1c27bbde137ce6b3fa4" |
|
}, |
|
{ |
|
"dataPath": "params_shard_44.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.15.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.15.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.15.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.15.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.16.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.16.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.16.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.16.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.16.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.16.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "6dc95280acde858c9e309150a2f7b0e9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_45.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.17.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "752f4545bd0fb5a53db8d2e5b55b72b9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_46.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.17.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "2f492b281b5441ab97336d77018812e5" |
|
}, |
|
{ |
|
"dataPath": "params_shard_47.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.16.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.16.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.16.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.16.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.17.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.17.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.17.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.17.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.17.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.17.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "7b643bc0a869ec4eb330861215e1b922" |
|
}, |
|
{ |
|
"dataPath": "params_shard_48.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.18.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "fdb15b1d10b4ee723c257296feaedc84" |
|
}, |
|
{ |
|
"dataPath": "params_shard_49.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.18.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "afc0eee9ccad5aab4e2225ff1c984eb8" |
|
}, |
|
{ |
|
"dataPath": "params_shard_50.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.17.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.17.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.17.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.17.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.18.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.18.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.18.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.18.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.18.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.18.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "9ebe186daad13a2d0f19dc99f04a1039" |
|
}, |
|
{ |
|
"dataPath": "params_shard_51.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.19.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "4732dff7d0fd292152a754f3d2758035" |
|
}, |
|
{ |
|
"dataPath": "params_shard_52.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.19.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "5785574f2d781a553b2578b7bbe628c1" |
|
}, |
|
{ |
|
"dataPath": "params_shard_53.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.18.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.18.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.18.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.18.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.19.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.19.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.19.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.19.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.19.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.19.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "5b9f69b26e3f0482cdb6ff5cdc42a52d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_54.bin", |
|
"format": "raw-shard", |
|
"nbytes": 33186816, |
|
"records": [ |
|
{ |
|
"name": "model.layers.19.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.19.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.19.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.19.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.20.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.20.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 31712256 |
|
} |
|
], |
|
"md5sum": "93497413d03d376abf09448c82f776e6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_55.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.8.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e664ae818a87ba10626f4f04c596f93e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_56.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.8.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "dd072f4177e3d19a201612f9cedd3c77" |
|
}, |
|
{ |
|
"dataPath": "params_shard_57.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32413696, |
|
"records": [ |
|
{ |
|
"name": "model.layers.20.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.20.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 5906432 |
|
}, |
|
{ |
|
"name": "model.layers.7.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 6644736 |
|
}, |
|
{ |
|
"name": "model.layers.7.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 6651904 |
|
}, |
|
{ |
|
"name": "model.layers.7.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 27238400 |
|
}, |
|
{ |
|
"name": "model.layers.7.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 29811712 |
|
}, |
|
{ |
|
"name": "model.layers.7.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 29818880 |
|
}, |
|
{ |
|
"name": "model.layers.7.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 29826048 |
|
}, |
|
{ |
|
"name": "model.layers.8.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 29833216 |
|
}, |
|
{ |
|
"name": "model.layers.8.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 29840384 |
|
} |
|
], |
|
"md5sum": "8b3365e55c427ef83bc25fe520078443" |
|
}, |
|
{ |
|
"dataPath": "params_shard_58.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.9.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "fef9d26441470e4b3e98aede2191cea4" |
|
}, |
|
{ |
|
"dataPath": "params_shard_59.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.9.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "1f1855ecfb61b17c66072a05189ef924" |
|
}, |
|
{ |
|
"dataPath": "params_shard_60.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32861184, |
|
"records": [ |
|
{ |
|
"name": "model.layers.8.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.8.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 5160960 |
|
}, |
|
{ |
|
"name": "model.layers.8.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 5168128 |
|
}, |
|
{ |
|
"name": "model.layers.8.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 5175296 |
|
}, |
|
{ |
|
"name": "model.layers.8.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 5182464 |
|
}, |
|
{ |
|
"name": "model.layers.8.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 16978944 |
|
}, |
|
{ |
|
"name": "model.layers.8.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 18453504 |
|
}, |
|
{ |
|
"name": "model.layers.8.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 24359936 |
|
}, |
|
{ |
|
"name": "model.layers.9.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 25098240 |
|
}, |
|
{ |
|
"name": "model.layers.9.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 25105408 |
|
}, |
|
{ |
|
"name": "model.layers.9.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 27678720 |
|
}, |
|
{ |
|
"name": "model.layers.9.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 32839680 |
|
}, |
|
{ |
|
"name": "model.layers.9.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 32846848 |
|
}, |
|
{ |
|
"name": "model.layers.9.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 32854016 |
|
} |
|
], |
|
"md5sum": "ee76a82773f2d32c061fb033164b2370" |
|
}, |
|
{ |
|
"dataPath": "params_shard_61.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.20.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b4c4f33e278e02fed1a4140f11864ece" |
|
}, |
|
{ |
|
"dataPath": "params_shard_62.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.20.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "381527b3aaf80e1f5c2fe5942962bef7" |
|
}, |
|
{ |
|
"dataPath": "params_shard_63.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.21.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "dac36fab9bf1479485e31f3bf29abce7" |
|
}, |
|
{ |
|
"dataPath": "params_shard_64.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.21.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f090b441d41acfa1abdd4b29b25b4ced" |
|
}, |
|
{ |
|
"dataPath": "params_shard_65.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30259200, |
|
"records": [ |
|
{ |
|
"name": "model.layers.9.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.9.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.9.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.9.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.20.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.20.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.20.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.20.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.20.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.20.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
}, |
|
{ |
|
"name": "model.layers.21.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27678720 |
|
}, |
|
{ |
|
"name": "model.layers.21.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 27685888 |
|
} |
|
], |
|
"md5sum": "64fdea4139376c9217f5d8e701268197" |
|
}, |
|
{ |
|
"dataPath": "params_shard_66.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.22.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "7c856a3557b6e25a1e4e9d9286aa6d9d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_67.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.22.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "1dea96260e2fd8490fe4f4a612560b56" |
|
}, |
|
{ |
|
"dataPath": "params_shard_68.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32861184, |
|
"records": [ |
|
{ |
|
"name": "model.layers.21.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.21.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 5160960 |
|
}, |
|
{ |
|
"name": "model.layers.21.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 5168128 |
|
}, |
|
{ |
|
"name": "model.layers.21.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 5175296 |
|
}, |
|
{ |
|
"name": "model.layers.21.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 5182464 |
|
}, |
|
{ |
|
"name": "model.layers.21.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 16978944 |
|
}, |
|
{ |
|
"name": "model.layers.21.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 18453504 |
|
}, |
|
{ |
|
"name": "model.layers.21.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 24359936 |
|
}, |
|
{ |
|
"name": "model.layers.22.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 25098240 |
|
}, |
|
{ |
|
"name": "model.layers.22.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 25105408 |
|
}, |
|
{ |
|
"name": "model.layers.22.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 27678720 |
|
}, |
|
{ |
|
"name": "model.layers.22.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 32839680 |
|
}, |
|
{ |
|
"name": "model.layers.22.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 32846848 |
|
}, |
|
{ |
|
"name": "model.layers.22.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 32854016 |
|
} |
|
], |
|
"md5sum": "23a28442f18a401ed5fe5240aff0162f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_69.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.23.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "8e1c2b019f3ca9631ebd4aad326f6082" |
|
}, |
|
{ |
|
"dataPath": "params_shard_70.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.23.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b725d8d0d5b98a842bcf2362a35f46c6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_71.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.22.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.22.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.22.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.22.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.23.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.23.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.23.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.23.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.23.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.23.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "c8ddd23bf46c8e4b4554313b6dee29e9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_72.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.24.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "140c2d45a959b34d3c9c89382e3c2b7b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_73.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.24.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f19f5d32ff7c5cdad893e5d7edbf020e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_74.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.23.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.23.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.23.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.23.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.24.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.24.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.24.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.24.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.24.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.24.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "dc26d068cb2923deb6523c2f358befd5" |
|
}, |
|
{ |
|
"dataPath": "params_shard_75.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.25.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "3bd17347ef6a5dee9599ac874498179e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_76.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.25.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "d9c1c509bb50ac35883fa86e2077919d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_77.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.24.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.24.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.24.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.24.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.25.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.25.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.25.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.25.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.25.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.25.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "d87a04b32465acb912b2d3eddb87692a" |
|
}, |
|
{ |
|
"dataPath": "params_shard_78.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.26.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "ddf3126a9365ebd885a589731e014d0f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_79.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.26.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "a31bb701806e294a364f4b53b7c1e02d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_80.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.25.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.25.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.25.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.25.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.26.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.26.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.26.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.26.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.26.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.26.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "e1bcebb56fdd674b87a3332891d8212e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_81.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.27.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "7500ff6bdbd560bcb81ec8b99aa0faa3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_82.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.27.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "27bd03e7bbf29c6801446edcd595df1d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_83.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.26.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.26.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.26.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.26.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.27.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.27.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.27.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.27.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.27.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.27.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "d6f525f0edac1e9227c6063caf7cfb01" |
|
}, |
|
{ |
|
"dataPath": "params_shard_84.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.28.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b1523e576a1b5ec5a14bbc6ad483d001" |
|
}, |
|
{ |
|
"dataPath": "params_shard_85.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.28.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "d6e9b4e05d5e821e4c27c2de3c66475e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_86.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.27.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.27.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.27.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.27.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.28.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.28.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.28.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.28.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.28.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.28.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "f2099a8e24361bdcfbdd2560d7647361" |
|
}, |
|
{ |
|
"dataPath": "params_shard_87.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.29.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "cd1455562c9a3d50ac2ff743b1dd760c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_88.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.29.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "a9febbc9a7c7c9f81ff2bfccf6524a0c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_89.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.28.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.28.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.28.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.28.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.29.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.29.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.29.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.29.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.29.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.29.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "bf250ff50a04169adec5b76e107f5ceb" |
|
}, |
|
{ |
|
"dataPath": "params_shard_90.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.30.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "bd097aff4fa711afd52ca446a609e864" |
|
}, |
|
{ |
|
"dataPath": "params_shard_91.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.30.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "5eb2c4cc291bc12242c62b336dd664dd" |
|
}, |
|
{ |
|
"dataPath": "params_shard_92.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.29.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.29.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.29.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.29.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.30.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.30.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.30.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.30.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.30.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.30.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "0bca995f7a8a1727d9b82aa03ef126fe" |
|
}, |
|
{ |
|
"dataPath": "params_shard_93.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.31.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "4a27b65b91c816a6d79eb2b4d086dfb2" |
|
}, |
|
{ |
|
"dataPath": "params_shard_94.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.31.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "46265f3ccedcff72c83868be7d6a5cf2" |
|
}, |
|
{ |
|
"dataPath": "params_shard_95.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.30.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.30.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.30.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.30.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.31.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.31.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.31.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.31.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.31.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.31.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "3ceaceb53a3a622e1cd2210529a964a6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_96.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.32.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "0f0b40574d8f816309999ef6ea5b50da" |
|
}, |
|
{ |
|
"dataPath": "params_shard_97.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25076736, |
|
"records": [ |
|
{ |
|
"name": "model.layers.31.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.31.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.31.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.31.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.32.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 19915776 |
|
} |
|
], |
|
"md5sum": "eaa781b636fb387f2f249c2da6c146de" |
|
}, |
|
{ |
|
"dataPath": "params_shard_98.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.32.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "0c2160cdeb6d6241de5d5b38914d2dcc" |
|
}, |
|
{ |
|
"dataPath": "params_shard_99.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.33.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "aea2d96a5563bf61f7d0c8c62d43e5b6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_100.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.33.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "3f13659b357bb2fb292f18429bf4d53f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_101.bin", |
|
"format": "raw-shard", |
|
"nbytes": 30280704, |
|
"records": [ |
|
{ |
|
"name": "model.layers.32.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.32.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.32.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.32.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.32.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.32.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.32.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.32.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 22503424 |
|
}, |
|
{ |
|
"name": "model.layers.32.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 22510592 |
|
}, |
|
{ |
|
"name": "model.layers.33.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 22517760 |
|
}, |
|
{ |
|
"name": "model.layers.33.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 22524928 |
|
}, |
|
{ |
|
"name": "model.layers.33.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 25098240 |
|
}, |
|
{ |
|
"name": "model.layers.33.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 30259200 |
|
}, |
|
{ |
|
"name": "model.layers.33.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 30266368 |
|
}, |
|
{ |
|
"name": "model.layers.33.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 30273536 |
|
} |
|
], |
|
"md5sum": "9d58714edc319f399031748beb654375" |
|
}, |
|
{ |
|
"dataPath": "params_shard_102.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.34.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "5f15bc50f8bb4db044fc469abb050b5a" |
|
}, |
|
{ |
|
"dataPath": "params_shard_103.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.34.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "934d205139a4eda876a1079f3c1bf87f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_104.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.33.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.33.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.33.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.33.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.34.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.34.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.34.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.34.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.34.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.34.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "b86d9e98db71974d282c7d3016a600a9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_105.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.35.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "00387dc339775ba5358f5a6d98965ef9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_106.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.35.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "16c4b260b1e94f4637632ca88aefc6ea" |
|
}, |
|
{ |
|
"dataPath": "params_shard_107.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.34.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.34.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.34.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.34.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.35.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.35.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.35.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.35.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.35.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.35.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "545a8bc13b4600c1e6649ca450d32eea" |
|
}, |
|
{ |
|
"dataPath": "params_shard_108.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.36.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "bea068207fea0d0eb0090e588824d912" |
|
}, |
|
{ |
|
"dataPath": "params_shard_109.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.36.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "3d09cbcaedcbc0c6f236ab6c517a49da" |
|
}, |
|
{ |
|
"dataPath": "params_shard_110.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.35.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.35.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.35.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.35.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.36.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.36.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.36.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.36.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.36.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.36.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "84e7fdb44abb9ed949b0dff11777d434" |
|
}, |
|
{ |
|
"dataPath": "params_shard_111.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.37.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "5302e7279855200cee8dd8e6afb8af04" |
|
}, |
|
{ |
|
"dataPath": "params_shard_112.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.37.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b01a418e1e779867629f078a5ee46216" |
|
}, |
|
{ |
|
"dataPath": "params_shard_113.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.36.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.36.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.36.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.36.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.37.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.37.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.37.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.37.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.37.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.37.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "da17f38d6608e5005aa9219994f1148f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_114.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.38.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "30d5bcc31cf413fe914d282dc5c853e6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_115.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.38.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "7a34f7b36bbd16ec33c94043cefc6e27" |
|
}, |
|
{ |
|
"dataPath": "params_shard_116.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.37.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.37.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.37.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.37.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.38.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.38.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.38.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.38.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.38.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.38.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "a56387b5001c5e1a298447e546915f5c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_117.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.39.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6e2a2669837a606027e889aeb8e4b15d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_118.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.39.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "9f5891bbb822232aa9bb4aceffb7277e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_119.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.38.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.38.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.38.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.38.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.39.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.39.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.39.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.39.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.39.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.39.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "2911869b8494fcfb4d34011f23f75113" |
|
}, |
|
{ |
|
"dataPath": "params_shard_120.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.40.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "20c9964222061354261ec21e44d4a6bc" |
|
}, |
|
{ |
|
"dataPath": "params_shard_121.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.40.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b9c2d4f666715069571d14bc753b4791" |
|
}, |
|
{ |
|
"dataPath": "params_shard_122.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.39.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.39.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.39.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.39.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.40.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.40.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.40.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.40.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.40.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.40.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "b2c79e81d93e9122aca1a88e73297ca9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_123.bin", |
|
"format": "raw-shard", |
|
"nbytes": 20586496, |
|
"records": [ |
|
{ |
|
"name": "model.layers.41.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
1436 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 20586496, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "971a803da28bf363151ebc73bd60be33" |
|
}, |
|
{ |
|
"dataPath": "params_shard_124.bin", |
|
"format": "raw-shard", |
|
"nbytes": 41287680, |
|
"records": [ |
|
{ |
|
"name": "model.layers.41.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
28672, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 41287680, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "2ebe9c63cd212f60c2cb63c0639cd024" |
|
}, |
|
{ |
|
"dataPath": "params_shard_125.bin", |
|
"format": "raw-shard", |
|
"nbytes": 27678720, |
|
"records": [ |
|
{ |
|
"name": "model.layers.40.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.40.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.40.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.40.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.layers.41.input_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
}, |
|
{ |
|
"name": "model.layers.41.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
359 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 2573312, |
|
"byteOffset": 19922944 |
|
}, |
|
{ |
|
"name": "model.layers.41.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
28672, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5160960, |
|
"byteOffset": 22496256 |
|
}, |
|
{ |
|
"name": "model.layers.41.post_attention_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27657216 |
|
}, |
|
{ |
|
"name": "model.layers.41.post_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27664384 |
|
}, |
|
{ |
|
"name": "model.layers.41.pre_feedforward_layernorm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 27671552 |
|
} |
|
], |
|
"md5sum": "0120233613e9b39598db3eb5e3155341" |
|
}, |
|
{ |
|
"dataPath": "params_shard_126.bin", |
|
"format": "raw-shard", |
|
"nbytes": 19922944, |
|
"records": [ |
|
{ |
|
"name": "model.layers.41.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
8192, |
|
360 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 11796480, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.41.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
8192, |
|
90 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1474560, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.41.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3584, |
|
412 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 5906432, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.41.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3584, |
|
103 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 738304, |
|
"byteOffset": 19177472 |
|
}, |
|
{ |
|
"name": "model.norm.weight", |
|
"shape": [ |
|
3584 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7168, |
|
"byteOffset": 19915776 |
|
} |
|
], |
|
"md5sum": "6787fbedf5306c2d0f53db9994ce0bd1" |
|
} |
|
] |
|
} |