|
{ |
|
"metadata": { |
|
"ParamSize": 533, |
|
"ParamBytes": 6649587712.0, |
|
"BitsPerParam": 3.6016642145955235 |
|
}, |
|
"records": [ |
|
{ |
|
"dataPath": "params_shard_0.bin", |
|
"format": "raw-shard", |
|
"nbytes": 311427072, |
|
"records": [ |
|
{ |
|
"name": "lm_head.q_weight", |
|
"shape": [ |
|
152064, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 311427072, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "363abc23581cac77d0bbe89f85d1034e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_1.bin", |
|
"format": "raw-shard", |
|
"nbytes": 38928384, |
|
"records": [ |
|
{ |
|
"name": "lm_head.q_scale", |
|
"shape": [ |
|
152064, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 38928384, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "0c9c8bee80dec0c8d7c5bec780e9a3b3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_2.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.33.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "a9a462c7537e5c812043dd9d1d134402" |
|
}, |
|
{ |
|
"dataPath": "params_shard_3.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31897600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.33.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.33.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 10240 |
|
}, |
|
{ |
|
"name": "model.layers.33.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28354560 |
|
} |
|
], |
|
"md5sum": "bf2639d5c342f2b01d16903cc07fd4f9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_4.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.34.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "d983ee8ca02ef7bd741bcbc894a2acdc" |
|
}, |
|
{ |
|
"dataPath": "params_shard_5.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.34.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "82081b6752cf51658ce28b4bbed6542f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_6.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32423936, |
|
"records": [ |
|
{ |
|
"name": "model.layers.33.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.33.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.34.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.34.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 7098368 |
|
}, |
|
{ |
|
"name": "model.layers.34.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 10641408 |
|
}, |
|
{ |
|
"name": "model.layers.34.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 17719296 |
|
}, |
|
{ |
|
"name": "model.layers.34.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 17729536 |
|
}, |
|
{ |
|
"name": "model.layers.34.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 17743872 |
|
} |
|
], |
|
"md5sum": "d983f635235ae24db16e512f69b087b6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_7.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.35.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "a67e5436322f5ee210da6260f6c8edaf" |
|
}, |
|
{ |
|
"dataPath": "params_shard_8.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.35.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "7f615f9e9765f6dfc90c70ff42bd19de" |
|
}, |
|
{ |
|
"dataPath": "params_shard_9.bin", |
|
"format": "raw-shard", |
|
"nbytes": 24287232, |
|
"records": [ |
|
{ |
|
"name": "model.layers.34.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.34.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 1835008 |
|
}, |
|
{ |
|
"name": "model.layers.34.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 12320768 |
|
}, |
|
{ |
|
"name": "model.layers.35.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 13631488 |
|
}, |
|
{ |
|
"name": "model.layers.35.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 13641728 |
|
}, |
|
{ |
|
"name": "model.layers.35.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 17184768 |
|
}, |
|
{ |
|
"name": "model.layers.35.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 24262656 |
|
}, |
|
{ |
|
"name": "model.layers.35.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 24272896 |
|
} |
|
], |
|
"md5sum": "ed780f6ec0e55340c4ad6708390a69d4" |
|
}, |
|
{ |
|
"dataPath": "params_shard_10.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.36.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "18d863dc4324fe9c7510de8bd9f6e884" |
|
}, |
|
{ |
|
"dataPath": "params_shard_11.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.36.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "904f5ed78ab344e57b094216e96342fa" |
|
}, |
|
{ |
|
"dataPath": "params_shard_12.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.35.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.35.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.35.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.35.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.36.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.36.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "601f8586b483729b0219bfa034ad5ead" |
|
}, |
|
{ |
|
"dataPath": "params_shard_13.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.36.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.36.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.36.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.36.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.36.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "54f3327a0fe7b9aff359554296be7e48" |
|
}, |
|
{ |
|
"dataPath": "params_shard_14.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.37.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "27dace9a824c3c90f679e01b7f0d1450" |
|
}, |
|
{ |
|
"dataPath": "params_shard_15.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.37.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "4155ac213ee67ab77b0961742cf9a786" |
|
}, |
|
{ |
|
"dataPath": "params_shard_16.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.36.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.36.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.37.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.37.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.37.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.37.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.37.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "8bb864caa5871be35d567ad7de3a47e0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_17.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.38.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "36768f2ed1f688616425d775672d05df" |
|
}, |
|
{ |
|
"dataPath": "params_shard_18.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.38.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "0d3ced20429ad8087dd343afe8394a81" |
|
}, |
|
{ |
|
"dataPath": "params_shard_19.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.37.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.37.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.37.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.37.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.38.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.38.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "0663d4a8c3e77620cbe893c0ca7988dc" |
|
}, |
|
{ |
|
"dataPath": "params_shard_20.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.38.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.38.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.38.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.38.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.38.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "8e84c3a83ebf2dc303adde3daefb8fd9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_21.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.39.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "1df20f28b6206b00afc75b8fb6543ef9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_22.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.39.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b89e4203949b4df1210f42e35e52a879" |
|
}, |
|
{ |
|
"dataPath": "params_shard_23.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.38.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.38.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.39.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.39.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.39.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.39.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.39.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "f4597db823259a0a97094dcf906a180e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_24.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.40.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b4b868207d6cb26dd3aab7f50e223b0b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_25.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.40.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "56ba1fcd9efb1998745d9c7a56ef34db" |
|
}, |
|
{ |
|
"dataPath": "params_shard_26.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.39.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.39.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.39.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.39.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.40.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.40.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "00b4a237556811863d8343b3cff292c6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_27.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.40.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.40.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.40.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.40.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.40.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "5afa4ee098f87df8b60cdcf3ff04acbf" |
|
}, |
|
{ |
|
"dataPath": "params_shard_28.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.41.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "448fbcac4caa015d3bb3d307cbb4d888" |
|
}, |
|
{ |
|
"dataPath": "params_shard_29.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.41.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "ce0c27f156e5c0a63683e83840759873" |
|
}, |
|
{ |
|
"dataPath": "params_shard_30.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.40.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.40.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.41.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.41.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.41.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.41.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.41.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "3f54c28482a3a71f284d9a3bbc8dd3d7" |
|
}, |
|
{ |
|
"dataPath": "params_shard_31.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.42.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "3d3ecfdc234ec4e1ab4e34609d4b85b5" |
|
}, |
|
{ |
|
"dataPath": "params_shard_32.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.42.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "7dfbb65d81742156d0f01ac1354357ff" |
|
}, |
|
{ |
|
"dataPath": "params_shard_33.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.41.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.41.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.41.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.41.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.42.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.42.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "b7e21f59751e3f4c42c567b8f4d0b82c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_34.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.42.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.42.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.42.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.42.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.42.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "eea818a63c69af8b3f9319a38d997fee" |
|
}, |
|
{ |
|
"dataPath": "params_shard_35.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.43.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e3329dd077014a4e853eba0adcebeb83" |
|
}, |
|
{ |
|
"dataPath": "params_shard_36.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.43.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "0ad7df7a5584d485400ebdb81e8fd035" |
|
}, |
|
{ |
|
"dataPath": "params_shard_37.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.42.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.42.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.43.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.43.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.43.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.43.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.43.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "fba0fe8eb2b8bb1d4f5040ecdfe2d9c1" |
|
}, |
|
{ |
|
"dataPath": "params_shard_38.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.44.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "08b24091beb8840e1016991034da1a09" |
|
}, |
|
{ |
|
"dataPath": "params_shard_39.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.44.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e53f27a65b6e2570a33057df341eb9e5" |
|
}, |
|
{ |
|
"dataPath": "params_shard_40.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.43.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.43.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.43.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.43.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.44.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.44.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "d40d55b2b134347ffb148ee493a5740b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_41.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.44.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.44.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.44.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.44.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.44.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "a11a8e268e6a4316b9db54ffa094559a" |
|
}, |
|
{ |
|
"dataPath": "params_shard_42.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.45.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "1aa0d11cb6e8de2f64f469ac866ae556" |
|
}, |
|
{ |
|
"dataPath": "params_shard_43.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.45.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "ad8b40a89df20a3f4152faa4072a78a2" |
|
}, |
|
{ |
|
"dataPath": "params_shard_44.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.44.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.44.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.45.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.45.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.45.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.45.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.45.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "60f190f0969d0fc73162178aa6d027b8" |
|
}, |
|
{ |
|
"dataPath": "params_shard_45.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.46.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "78b884e9fd2b603ba38c1ca0524f0a61" |
|
}, |
|
{ |
|
"dataPath": "params_shard_46.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.46.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "01bbe093e4df2eb609313786632e41a0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_47.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.45.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.45.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.45.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.45.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.46.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.46.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "6603b0b2d0ee2a6e41f0e08e915348e5" |
|
}, |
|
{ |
|
"dataPath": "params_shard_48.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.46.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.46.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.46.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.46.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.46.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "6536a71700af2bdd9632ca54dbbb8e8b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_49.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.47.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "79dbe16d1fd4e01871159a12b99a81d1" |
|
}, |
|
{ |
|
"dataPath": "params_shard_50.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.47.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "1dcb3c6dbea876e4858f4acce10458e5" |
|
}, |
|
{ |
|
"dataPath": "params_shard_51.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.46.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.46.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.47.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.47.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.47.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.47.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.47.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "c8527f06d869e40f33c958d42a0fc031" |
|
}, |
|
{ |
|
"dataPath": "params_shard_52.bin", |
|
"format": "raw-shard", |
|
"nbytes": 311427072, |
|
"records": [ |
|
{ |
|
"name": "model.embed_tokens.q_weight", |
|
"shape": [ |
|
152064, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 311427072, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "62b09ea564dc148277bf8a410576bfc0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_53.bin", |
|
"format": "raw-shard", |
|
"nbytes": 38928384, |
|
"records": [ |
|
{ |
|
"name": "model.embed_tokens.q_scale", |
|
"shape": [ |
|
152064, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 38928384, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "499e9b501de9d60a2549350662f59a00" |
|
}, |
|
{ |
|
"dataPath": "params_shard_54.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.0.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "74f8dcf293adfe89c9ff501330f8aff7" |
|
}, |
|
{ |
|
"dataPath": "params_shard_55.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.0.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "d28df7b065055953c4a8323f6ec7015c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_56.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31875072, |
|
"records": [ |
|
{ |
|
"name": "model.layers.47.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.47.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.47.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.47.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.norm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.0.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28321792 |
|
}, |
|
{ |
|
"name": "model.layers.0.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28332032 |
|
} |
|
], |
|
"md5sum": "378e3373b6cbf0501e17031a549588a9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_57.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.0.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.0.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.0.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.0.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.0.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "fc1664d271dac054a9b089161754343e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_58.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.1.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "ffb5d5c14a1f5d45bc533ec3f7cbdc5b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_59.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.1.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "649da5775a6b3ff47a67ddf58b7416a1" |
|
}, |
|
{ |
|
"dataPath": "params_shard_60.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.0.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.0.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.1.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.1.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.1.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.1.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.1.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "78ab87e976b808e43f359b5d25d47c92" |
|
}, |
|
{ |
|
"dataPath": "params_shard_61.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.10.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e4813763bf489b2e5c7931c626b65be9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_62.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.10.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "38ec71791ab1d29a282dca3eccc88421" |
|
}, |
|
{ |
|
"dataPath": "params_shard_63.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.1.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.1.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.1.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.1.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.10.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.10.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "9d41871bc9d6c184b0935d120ec5c41d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_64.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.10.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.10.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.10.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.10.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.10.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "e64aa9d5f381fc727c47d4d18db3130f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_65.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.11.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "9def5a8b0cc4c334f22e70270597b0d4" |
|
}, |
|
{ |
|
"dataPath": "params_shard_66.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.11.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b55b001988c19c2bf54177701f074aff" |
|
}, |
|
{ |
|
"dataPath": "params_shard_67.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.10.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.10.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.11.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.11.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.11.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.11.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.11.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "41268f6a25347a9fd0a152e48aec96c9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_68.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.12.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "4d1d98893ae80910ba45cdb41bba0a55" |
|
}, |
|
{ |
|
"dataPath": "params_shard_69.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.12.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "47a5ddb31d32cbbde0d1e20bfcac813e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_70.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.11.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.11.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.11.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.11.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.12.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.12.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "a29c402ce7b64d4bb1acec47ff017177" |
|
}, |
|
{ |
|
"dataPath": "params_shard_71.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.12.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.12.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.12.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.12.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.12.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "9a0e2a166216e21b0e7b1954b9b06e47" |
|
}, |
|
{ |
|
"dataPath": "params_shard_72.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.13.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "a8b5efb451cd12bd415c3f8f55fb9097" |
|
}, |
|
{ |
|
"dataPath": "params_shard_73.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.13.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "3f9a4fbeaa0f78932cb795b49c9bdca5" |
|
}, |
|
{ |
|
"dataPath": "params_shard_74.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.12.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.12.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.13.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.13.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.13.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.13.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.13.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "9f533924aff3c142a3b2ab9ac7659d26" |
|
}, |
|
{ |
|
"dataPath": "params_shard_75.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.14.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e320e3739ed8b0290a278fc82de55933" |
|
}, |
|
{ |
|
"dataPath": "params_shard_76.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.14.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e3104054469b4232b1bed72322d45acc" |
|
}, |
|
{ |
|
"dataPath": "params_shard_77.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.13.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.13.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.13.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.13.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.14.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.14.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "e3ac397b8c80e46992dd52961dce0886" |
|
}, |
|
{ |
|
"dataPath": "params_shard_78.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.14.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.14.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.14.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.14.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.14.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "2c6229ae3723796a3846fbf4e76286dc" |
|
}, |
|
{ |
|
"dataPath": "params_shard_79.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28325888, |
|
"records": [ |
|
{ |
|
"name": "model.layers.14.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.14.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.15.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.15.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 11810816 |
|
}, |
|
{ |
|
"name": "model.layers.15.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 26490880 |
|
} |
|
], |
|
"md5sum": "0a8790749bff0b2e83b60218fde9ae69" |
|
}, |
|
{ |
|
"dataPath": "params_shard_80.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.2.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "fabcedfe065dcb398ad4ac976c90b7c4" |
|
}, |
|
{ |
|
"dataPath": "params_shard_81.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.2.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "06951b544ea29cec4dfe8771b6e98283" |
|
}, |
|
{ |
|
"dataPath": "params_shard_82.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.15.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.15.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.2.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.2.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.2.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.2.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.2.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "d5d2641ec48bfa8ce482547e131d8dde" |
|
}, |
|
{ |
|
"dataPath": "params_shard_83.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.3.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "60c4ce812c05bb5d3bc1e34197f19c09" |
|
}, |
|
{ |
|
"dataPath": "params_shard_84.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.3.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b33b4d183d1f9da34c8241c3c36737bd" |
|
}, |
|
{ |
|
"dataPath": "params_shard_85.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.2.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.2.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.2.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.2.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.3.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.3.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "29955531ef44c2ef37279071ef5482ed" |
|
}, |
|
{ |
|
"dataPath": "params_shard_86.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.3.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.3.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.3.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.3.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.3.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "592d7f75b754b87411c19ba7afc7760b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_87.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.4.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "0d7aa2ac96dad9ba27cb05726146afc3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_88.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.4.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "2944529885616dd40d4849bccb8248e9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_89.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.3.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.3.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.4.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.4.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.4.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.4.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.4.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "0c36fbda47cab0d56d65c14d60c2b03e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_90.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.5.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "1efe81848ff94807822c752b6db6d1ec" |
|
}, |
|
{ |
|
"dataPath": "params_shard_91.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.5.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "38ecbb9f2a39df723f9f13a1cefb7216" |
|
}, |
|
{ |
|
"dataPath": "params_shard_92.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.4.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.4.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.4.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.4.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.5.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.5.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "dca563dbd9f601852157dba3c3fac3c2" |
|
}, |
|
{ |
|
"dataPath": "params_shard_93.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.5.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.5.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.5.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.5.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.5.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "1c8c0db9a8bde1e6aab35b46d35a25a1" |
|
}, |
|
{ |
|
"dataPath": "params_shard_94.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.6.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "72dc3c7fb1ca392eb0a8fa7ca1167df4" |
|
}, |
|
{ |
|
"dataPath": "params_shard_95.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.6.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "423755dc1e5b5177794309b1ca671bad" |
|
}, |
|
{ |
|
"dataPath": "params_shard_96.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.5.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.5.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.6.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.6.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.6.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.6.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.6.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "99fd2aa9536ff6428a674ef27a044853" |
|
}, |
|
{ |
|
"dataPath": "params_shard_97.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.7.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f99850118bd228943488ae6a3efd9eed" |
|
}, |
|
{ |
|
"dataPath": "params_shard_98.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.7.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "365870749c26fec79903eece7d3ed45f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_99.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.6.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.6.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.6.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.6.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.7.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.7.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "ab3706d80409b76ae0c72c8c4031362e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_100.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.7.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.7.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.7.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.7.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.7.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "55f37de1dda42f08820689e5d07dc498" |
|
}, |
|
{ |
|
"dataPath": "params_shard_101.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.8.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "7b01378d8a937b9c1ab36e1e563f44e7" |
|
}, |
|
{ |
|
"dataPath": "params_shard_102.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.8.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "4efa12b083901061152f0782b2b36fda" |
|
}, |
|
{ |
|
"dataPath": "params_shard_103.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.7.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.7.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.8.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.8.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.8.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.8.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.8.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "cb34100a3574bbb200a450f102f89811" |
|
}, |
|
{ |
|
"dataPath": "params_shard_104.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.9.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6de86e3f612bbbac4370cf640d0e72eb" |
|
}, |
|
{ |
|
"dataPath": "params_shard_105.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.9.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "0de928b467367ca0fce1d29d7d12c9f9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_106.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.8.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.8.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.8.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.8.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.9.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.9.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "1877fab67050ecb0f690890f08005ba8" |
|
}, |
|
{ |
|
"dataPath": "params_shard_107.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.9.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.9.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.9.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.9.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.9.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "12e9ec5aaab58cb47f8a537d5504d8f8" |
|
}, |
|
{ |
|
"dataPath": "params_shard_108.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.15.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b2580862c7b9464f404a53ee1c675a89" |
|
}, |
|
{ |
|
"dataPath": "params_shard_109.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.15.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e73a3faf17345fab3fb662ac1af02315" |
|
}, |
|
{ |
|
"dataPath": "params_shard_110.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.16.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "3f6bac6bba2e2dd37e76b4cb2ba7cd64" |
|
}, |
|
{ |
|
"dataPath": "params_shard_111.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.16.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "c03739411165c22a8966aa7f868c9642" |
|
}, |
|
{ |
|
"dataPath": "params_shard_112.bin", |
|
"format": "raw-shard", |
|
"nbytes": 33093632, |
|
"records": [ |
|
{ |
|
"name": "model.layers.9.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.9.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.15.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.15.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.15.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.15.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.16.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22437888 |
|
}, |
|
{ |
|
"name": "model.layers.16.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 22448128 |
|
}, |
|
{ |
|
"name": "model.layers.16.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 25991168 |
|
}, |
|
{ |
|
"name": "model.layers.16.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 33069056 |
|
}, |
|
{ |
|
"name": "model.layers.16.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 33079296 |
|
} |
|
], |
|
"md5sum": "c28e0bced64e46610135f6dbd53d842c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_113.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.17.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "71bb2e869da1b7a0c8abb3cdc363609f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_114.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.17.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "31b212eece4bb0ec668b8f3ca229a267" |
|
}, |
|
{ |
|
"dataPath": "params_shard_115.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.16.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.16.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.16.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.16.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.17.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.17.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "387f295714147e49f616da4157c27f62" |
|
}, |
|
{ |
|
"dataPath": "params_shard_116.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.17.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.17.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.17.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.17.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.17.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "96aff11460ee5f76f6e39781f38bc5d3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_117.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.18.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "3bced622a58a1476e64c64ae2f701a64" |
|
}, |
|
{ |
|
"dataPath": "params_shard_118.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.18.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "ca1db15dea43a8096af3e08986604700" |
|
}, |
|
{ |
|
"dataPath": "params_shard_119.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.17.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.17.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.18.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.18.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.18.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.18.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.18.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "e8f534160fb2baae05d85c7f672d2d3a" |
|
}, |
|
{ |
|
"dataPath": "params_shard_120.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.19.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6334469f9e8faac3eda54de9aa9a7cdc" |
|
}, |
|
{ |
|
"dataPath": "params_shard_121.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.19.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "11aebb4fde349a42256867f775634ab3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_122.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.18.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.18.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.18.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.18.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.19.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.19.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "960611856cd26ac5eaa7352214fc97b4" |
|
}, |
|
{ |
|
"dataPath": "params_shard_123.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.19.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.19.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.19.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.19.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.19.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "f24ae423ec9feaa93e568dfb19fe3007" |
|
}, |
|
{ |
|
"dataPath": "params_shard_124.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.20.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f2f1b80c1b98e6369c149c99c8249791" |
|
}, |
|
{ |
|
"dataPath": "params_shard_125.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.20.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "5807203116cbd4eee59de7430a5cdd12" |
|
}, |
|
{ |
|
"dataPath": "params_shard_126.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.19.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.19.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.20.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.20.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.20.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.20.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.20.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "160241df5664456594d9bd00b93f33a6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_127.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.21.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f7dd7d7374dc262f39cce29b19d21c3c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_128.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.21.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "648a92af9375af33a4264499cb301c80" |
|
}, |
|
{ |
|
"dataPath": "params_shard_129.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.20.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.20.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.20.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.20.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.21.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.21.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "e4985e05c7ae25db17d156ac234f4407" |
|
}, |
|
{ |
|
"dataPath": "params_shard_130.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.21.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.21.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.21.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.21.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.21.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "a0186cfb96756cc93a7eab780e5cdf86" |
|
}, |
|
{ |
|
"dataPath": "params_shard_131.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.22.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "7b6ee823b622ea7b8cab35bbc50afc18" |
|
}, |
|
{ |
|
"dataPath": "params_shard_132.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.22.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "959c34f2a64160ee1e64145e9261845c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_133.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.21.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.21.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.22.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.22.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.22.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.22.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.22.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "0168df5129b84eefe726037fdb303c02" |
|
}, |
|
{ |
|
"dataPath": "params_shard_134.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.23.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "04149e51be81f8e5b9041cbf98aa002c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_135.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.23.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "45752125c1aadb65f43ae899021221ae" |
|
}, |
|
{ |
|
"dataPath": "params_shard_136.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.22.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.22.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.22.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.22.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.23.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.23.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "5d37ca11cd95a2362ba3c28af6384923" |
|
}, |
|
{ |
|
"dataPath": "params_shard_137.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.23.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.23.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.23.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.23.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.23.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "e62eeec392148f1c5ad6d5e090db0d89" |
|
}, |
|
{ |
|
"dataPath": "params_shard_138.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.24.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e961e2e77709f665d88cbefcdd48dc74" |
|
}, |
|
{ |
|
"dataPath": "params_shard_139.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.24.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "514c674439a3d0e1d288fabf9af9a0d0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_140.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.23.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.23.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.24.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.24.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.24.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.24.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.24.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "306a50528ea0f8dac90cccac2f74dbff" |
|
}, |
|
{ |
|
"dataPath": "params_shard_141.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.25.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "c0f5195a3093bea72581488b1648d830" |
|
}, |
|
{ |
|
"dataPath": "params_shard_142.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.25.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "469f56f14f54bb0cb94348b38c4622fb" |
|
}, |
|
{ |
|
"dataPath": "params_shard_143.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.24.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.24.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.24.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.24.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.25.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.25.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "2d56fbc9deb9a02776ea0cd487b5c6ed" |
|
}, |
|
{ |
|
"dataPath": "params_shard_144.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.25.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.25.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.25.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.25.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.25.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "93da71d117e84efcde6669d05e5d0f21" |
|
}, |
|
{ |
|
"dataPath": "params_shard_145.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.26.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "13151f22dbb66a50326d1df1a2c0ac48" |
|
}, |
|
{ |
|
"dataPath": "params_shard_146.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.26.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "5b72105d1cce5083e5313d314115c041" |
|
}, |
|
{ |
|
"dataPath": "params_shard_147.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.25.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.25.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.26.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.26.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.26.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.26.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.26.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "843f28ccaa4a3d99a9c6c4ca90016c8f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_148.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.27.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f765c71c09b8a7990e410b8e401b0f8f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_149.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.27.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "083589810f93aaf2234d6693e8352fb1" |
|
}, |
|
{ |
|
"dataPath": "params_shard_150.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.26.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.26.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.26.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.26.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.27.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.27.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "5a4d26ab9d94158a201927fc6ba59220" |
|
}, |
|
{ |
|
"dataPath": "params_shard_151.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.27.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.27.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.27.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.27.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.27.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "3f1450403baba8ca9bbb99f1cc680e04" |
|
}, |
|
{ |
|
"dataPath": "params_shard_152.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.28.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b4e66d7a53143ee23591b23ec2af7df5" |
|
}, |
|
{ |
|
"dataPath": "params_shard_153.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.28.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "c837b525e024deebeddfdaaba03c5b20" |
|
}, |
|
{ |
|
"dataPath": "params_shard_154.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.27.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.27.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.28.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.28.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.28.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.28.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.28.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "c92b4e5e16c109d73b8b270bff7d9888" |
|
}, |
|
{ |
|
"dataPath": "params_shard_155.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.29.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "726cf107b85f9550aef80ff23f8643ac" |
|
}, |
|
{ |
|
"dataPath": "params_shard_156.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.29.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "2aad4f553190d841224eb9e51e948591" |
|
}, |
|
{ |
|
"dataPath": "params_shard_157.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.28.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.28.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.28.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.28.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.29.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.29.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "b9f2c3a4df5bb96020968c3ed8691d54" |
|
}, |
|
{ |
|
"dataPath": "params_shard_158.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.29.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.29.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.29.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.29.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.29.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "a6cff6b8cc076c8a9424cf76b8d9f6f2" |
|
}, |
|
{ |
|
"dataPath": "params_shard_159.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.30.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "4b9f3f18af26b0eea74d2a6171a525d6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_160.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.30.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "13f534e43d3136a4c0c2554154455ec0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_161.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.29.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.29.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.30.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.30.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.30.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.30.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.30.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "7136f234067afab6fd1040b267a798d9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_162.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.31.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b9b0b4397ddf9967083500c03aea6dc8" |
|
}, |
|
{ |
|
"dataPath": "params_shard_163.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.31.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6d666f06c2d38a9de9be31de14adb8c7" |
|
}, |
|
{ |
|
"dataPath": "params_shard_164.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31864832, |
|
"records": [ |
|
{ |
|
"name": "model.layers.30.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.30.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.30.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.30.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.31.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 28311552 |
|
}, |
|
{ |
|
"name": "model.layers.31.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 28321792 |
|
} |
|
], |
|
"md5sum": "dad93a078e725bfdea8172354aedad32" |
|
}, |
|
{ |
|
"dataPath": "params_shard_165.bin", |
|
"format": "raw-shard", |
|
"nbytes": 23617536, |
|
"records": [ |
|
{ |
|
"name": "model.layers.31.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.31.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 7077888 |
|
}, |
|
{ |
|
"name": "model.layers.31.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 7088128 |
|
}, |
|
{ |
|
"name": "model.layers.31.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 7102464 |
|
}, |
|
{ |
|
"name": "model.layers.31.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 21782528 |
|
} |
|
], |
|
"md5sum": "b5cf7acf389fe041c66baa357c2609b1" |
|
}, |
|
{ |
|
"dataPath": "params_shard_166.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28344320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.32.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 28344320, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "13851c721986a94fafa96e4012952d6a" |
|
}, |
|
{ |
|
"dataPath": "params_shard_167.bin", |
|
"format": "raw-shard", |
|
"nbytes": 56623104, |
|
"records": [ |
|
{ |
|
"name": "model.layers.32.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 56623104, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "8acc5e3d7cda3deaf1fc9759cbd06807" |
|
}, |
|
{ |
|
"dataPath": "params_shard_168.bin", |
|
"format": "raw-shard", |
|
"nbytes": 22452224, |
|
"records": [ |
|
{ |
|
"name": "model.layers.31.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.31.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 10485760 |
|
}, |
|
{ |
|
"name": "model.layers.32.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 11796480 |
|
}, |
|
{ |
|
"name": "model.layers.32.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
346 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3543040, |
|
"byteOffset": 11806720 |
|
}, |
|
{ |
|
"name": "model.layers.32.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7077888, |
|
"byteOffset": 15349760 |
|
}, |
|
{ |
|
"name": "model.layers.32.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 22427648 |
|
}, |
|
{ |
|
"name": "model.layers.32.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 22437888 |
|
} |
|
], |
|
"md5sum": "3aa7d21b124af64cf87c7af9489b7517" |
|
}, |
|
{ |
|
"dataPath": "params_shard_169.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28325888, |
|
"records": [ |
|
{ |
|
"name": "model.layers.32.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.32.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.32.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.32.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
}, |
|
{ |
|
"name": "model.layers.33.self_attn.c_attn.bias", |
|
"shape": [ |
|
7168 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14336, |
|
"byteOffset": 28311552 |
|
} |
|
], |
|
"md5sum": "910dfcce874cf11be8288a302d233114" |
|
}, |
|
{ |
|
"dataPath": "params_shard_170.bin", |
|
"format": "raw-shard", |
|
"nbytes": 28311552, |
|
"records": [ |
|
{ |
|
"name": "model.layers.33.self_attn.c_attn.q_weight", |
|
"shape": [ |
|
7168, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 14680064, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.33.self_attn.c_attn.q_scale", |
|
"shape": [ |
|
7168, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1835008, |
|
"byteOffset": 14680064 |
|
}, |
|
{ |
|
"name": "model.layers.33.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
512 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10485760, |
|
"byteOffset": 16515072 |
|
}, |
|
{ |
|
"name": "model.layers.33.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
128 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1310720, |
|
"byteOffset": 27000832 |
|
} |
|
], |
|
"md5sum": "ac55acb07cb928c5467308fecb7ecb3b" |
|
} |
|
] |
|
} |