|
{ |
|
"metadata": { |
|
"ParamSize": 283, |
|
"ParamBytes": 1807423488.0, |
|
"BitsPerParam": 4.500626782697164 |
|
}, |
|
"records": [ |
|
{ |
|
"dataPath": "params_shard_0.bin", |
|
"format": "raw-shard", |
|
"nbytes": 197001216, |
|
"records": [ |
|
{ |
|
"name": "model.embed_tokens.q_weight", |
|
"shape": [ |
|
128256, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 197001216, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f85c983396b413974b8d94445cc34d73" |
|
}, |
|
{ |
|
"dataPath": "params_shard_1.bin", |
|
"format": "raw-shard", |
|
"nbytes": 24631296, |
|
"records": [ |
|
{ |
|
"name": "model.embed_tokens.q_scale", |
|
"shape": [ |
|
128256, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 24625152, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.0.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 24625152 |
|
} |
|
], |
|
"md5sum": "85ff3cccf68c50cab2af14c8b4181e50" |
|
}, |
|
{ |
|
"dataPath": "params_shard_2.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.0.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "2198b97e9864fb9af7168a4cc0deea67" |
|
}, |
|
{ |
|
"dataPath": "params_shard_3.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.0.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.0.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 12582912 |
|
}, |
|
{ |
|
"name": "model.layers.0.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 14155776 |
|
}, |
|
{ |
|
"name": "model.layers.0.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.0.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.0.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 25171968 |
|
}, |
|
{ |
|
"name": "model.layers.0.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 26155008 |
|
}, |
|
{ |
|
"name": "model.layers.0.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 30873600 |
|
}, |
|
{ |
|
"name": "model.layers.1.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31463424 |
|
} |
|
], |
|
"md5sum": "386b025e860d706bddf2bfe80d83f751" |
|
}, |
|
{ |
|
"dataPath": "params_shard_4.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.1.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "a7814c631c3bb720557b60edc99d65b6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_5.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.1.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.1.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 12582912 |
|
}, |
|
{ |
|
"name": "model.layers.1.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 14155776 |
|
}, |
|
{ |
|
"name": "model.layers.1.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.1.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.1.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 25171968 |
|
}, |
|
{ |
|
"name": "model.layers.1.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 26155008 |
|
}, |
|
{ |
|
"name": "model.layers.1.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 30873600 |
|
}, |
|
{ |
|
"name": "model.layers.10.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31463424 |
|
} |
|
], |
|
"md5sum": "bcb20ccb131b409c80a047c5a4546bd4" |
|
}, |
|
{ |
|
"dataPath": "params_shard_6.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.10.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "9ee06ffba5f42727734a0b25dfa5be87" |
|
}, |
|
{ |
|
"dataPath": "params_shard_7.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.10.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.10.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 12582912 |
|
}, |
|
{ |
|
"name": "model.layers.10.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 14155776 |
|
}, |
|
{ |
|
"name": "model.layers.10.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.10.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.10.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 25171968 |
|
}, |
|
{ |
|
"name": "model.layers.10.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 26155008 |
|
}, |
|
{ |
|
"name": "model.layers.10.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 30873600 |
|
}, |
|
{ |
|
"name": "model.layers.11.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31463424 |
|
} |
|
], |
|
"md5sum": "f492a9ac7878fb03646088e513e4a503" |
|
}, |
|
{ |
|
"dataPath": "params_shard_8.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.11.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "8e530078957cbad3504436b7d045367e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_9.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.11.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.11.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 12582912 |
|
}, |
|
{ |
|
"name": "model.layers.11.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 14155776 |
|
}, |
|
{ |
|
"name": "model.layers.11.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.11.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.11.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 25171968 |
|
}, |
|
{ |
|
"name": "model.layers.11.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 26155008 |
|
}, |
|
{ |
|
"name": "model.layers.11.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 30873600 |
|
}, |
|
{ |
|
"name": "model.layers.12.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31463424 |
|
} |
|
], |
|
"md5sum": "f40c530fb0d05060bba405cf9b1a4bfc" |
|
}, |
|
{ |
|
"dataPath": "params_shard_10.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.12.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "32dbe1891076d9934f8884b4d61fc8bd" |
|
}, |
|
{ |
|
"dataPath": "params_shard_11.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.12.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.12.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 12582912 |
|
}, |
|
{ |
|
"name": "model.layers.12.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 14155776 |
|
}, |
|
{ |
|
"name": "model.layers.12.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.12.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.12.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 25171968 |
|
}, |
|
{ |
|
"name": "model.layers.12.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 26155008 |
|
}, |
|
{ |
|
"name": "model.layers.12.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 30873600 |
|
}, |
|
{ |
|
"name": "model.layers.13.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31463424 |
|
} |
|
], |
|
"md5sum": "ef44d1d5fb8efdf07060c7bb44cf874b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_12.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.13.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "cd5ff9184f02f5c09ba93ee92ff94271" |
|
}, |
|
{ |
|
"dataPath": "params_shard_13.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.13.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.13.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 12582912 |
|
}, |
|
{ |
|
"name": "model.layers.13.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 14155776 |
|
}, |
|
{ |
|
"name": "model.layers.13.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.13.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.13.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 25171968 |
|
}, |
|
{ |
|
"name": "model.layers.13.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 26155008 |
|
}, |
|
{ |
|
"name": "model.layers.13.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 30873600 |
|
}, |
|
{ |
|
"name": "model.layers.14.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31463424 |
|
} |
|
], |
|
"md5sum": "df1c40ade6785457bd8183c23b875b76" |
|
}, |
|
{ |
|
"dataPath": "params_shard_14.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.14.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "fdaa6d2aa85e23a0555623a10cd6206a" |
|
}, |
|
{ |
|
"dataPath": "params_shard_15.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.14.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.14.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 12582912 |
|
}, |
|
{ |
|
"name": "model.layers.14.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 14155776 |
|
}, |
|
{ |
|
"name": "model.layers.14.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.14.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.14.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 25171968 |
|
}, |
|
{ |
|
"name": "model.layers.14.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 26155008 |
|
}, |
|
{ |
|
"name": "model.layers.14.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 30873600 |
|
}, |
|
{ |
|
"name": "model.layers.15.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31463424 |
|
} |
|
], |
|
"md5sum": "682549347d433856d7fc6e8c267a2f7f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_16.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.15.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "3c830b58a140473b9f82bfa66bf3d42a" |
|
}, |
|
{ |
|
"dataPath": "params_shard_17.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.15.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.15.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 12582912 |
|
}, |
|
{ |
|
"name": "model.layers.15.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 14155776 |
|
}, |
|
{ |
|
"name": "model.layers.15.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.15.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.15.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 25171968 |
|
}, |
|
{ |
|
"name": "model.layers.15.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 26155008 |
|
}, |
|
{ |
|
"name": "model.layers.15.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 30873600 |
|
}, |
|
{ |
|
"name": "model.layers.16.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31463424 |
|
} |
|
], |
|
"md5sum": "a890000a2aec0e3eb21f2fd805f919ff" |
|
}, |
|
{ |
|
"dataPath": "params_shard_18.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.16.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "9a13f94d0562fa66bc2de91ccd7d0488" |
|
}, |
|
{ |
|
"dataPath": "params_shard_19.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.16.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.16.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 12582912 |
|
}, |
|
{ |
|
"name": "model.layers.16.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 14155776 |
|
}, |
|
{ |
|
"name": "model.layers.16.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.16.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.16.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 25171968 |
|
}, |
|
{ |
|
"name": "model.layers.16.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 26155008 |
|
}, |
|
{ |
|
"name": "model.layers.16.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 30873600 |
|
}, |
|
{ |
|
"name": "model.layers.17.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31463424 |
|
} |
|
], |
|
"md5sum": "c4f4d995b6b48b0726ab38b924e0a423" |
|
}, |
|
{ |
|
"dataPath": "params_shard_20.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.17.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "fa0aed7b065c958444805e7e4d9aba34" |
|
}, |
|
{ |
|
"dataPath": "params_shard_21.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.17.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.17.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 12582912 |
|
}, |
|
{ |
|
"name": "model.layers.17.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 14155776 |
|
}, |
|
{ |
|
"name": "model.layers.17.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.17.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.17.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 25171968 |
|
}, |
|
{ |
|
"name": "model.layers.17.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 26155008 |
|
}, |
|
{ |
|
"name": "model.layers.17.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 30873600 |
|
}, |
|
{ |
|
"name": "model.layers.18.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31463424 |
|
} |
|
], |
|
"md5sum": "2e8d4c50031c02b8b378ae5bdfde56c2" |
|
}, |
|
{ |
|
"dataPath": "params_shard_22.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.18.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f1b05be16ddbc7f4d00a359f8ebf3f9b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_23.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.18.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.18.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 12582912 |
|
}, |
|
{ |
|
"name": "model.layers.18.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 14155776 |
|
}, |
|
{ |
|
"name": "model.layers.18.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.18.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.18.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 25171968 |
|
}, |
|
{ |
|
"name": "model.layers.18.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 26155008 |
|
}, |
|
{ |
|
"name": "model.layers.18.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 30873600 |
|
}, |
|
{ |
|
"name": "model.layers.19.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31463424 |
|
} |
|
], |
|
"md5sum": "1df0cc746e6146d872a5b515e07375c3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_24.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.19.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6175e09011f65326f1f489932cafd871" |
|
}, |
|
{ |
|
"dataPath": "params_shard_25.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.19.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.19.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 12582912 |
|
}, |
|
{ |
|
"name": "model.layers.19.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 14155776 |
|
}, |
|
{ |
|
"name": "model.layers.19.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.19.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.19.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 25171968 |
|
}, |
|
{ |
|
"name": "model.layers.19.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 26155008 |
|
}, |
|
{ |
|
"name": "model.layers.19.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 30873600 |
|
}, |
|
{ |
|
"name": "model.layers.2.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31463424 |
|
} |
|
], |
|
"md5sum": "10355ad31168989e9da49232e844c037" |
|
}, |
|
{ |
|
"dataPath": "params_shard_26.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.2.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "c76c0f401cbf9a94c5ad6f4ab3a130ed" |
|
}, |
|
{ |
|
"dataPath": "params_shard_27.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.20.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e8196ddf7eb4b259cbddf03ee53c7203" |
|
}, |
|
{ |
|
"dataPath": "params_shard_28.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31463424, |
|
"records": [ |
|
{ |
|
"name": "model.layers.2.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.2.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 12582912 |
|
}, |
|
{ |
|
"name": "model.layers.2.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 14155776 |
|
}, |
|
{ |
|
"name": "model.layers.2.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.2.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.2.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 25171968 |
|
}, |
|
{ |
|
"name": "model.layers.2.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 26155008 |
|
}, |
|
{ |
|
"name": "model.layers.2.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 30873600 |
|
} |
|
], |
|
"md5sum": "c3c0f52e1e86645a590d6a2ecb2922bd" |
|
}, |
|
{ |
|
"dataPath": "params_shard_29.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.3.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "bc02b78ae616ceda2785064dfc3d66d0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_30.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31463424, |
|
"records": [ |
|
{ |
|
"name": "model.layers.20.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.20.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 3145728 |
|
}, |
|
{ |
|
"name": "model.layers.20.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 11010048 |
|
}, |
|
{ |
|
"name": "model.layers.20.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 11993088 |
|
}, |
|
{ |
|
"name": "model.layers.20.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 16711680 |
|
}, |
|
{ |
|
"name": "model.layers.3.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.3.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.3.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 29890560 |
|
} |
|
], |
|
"md5sum": "d1460b589cd66f8c1f6ef5f2c4aee380" |
|
}, |
|
{ |
|
"dataPath": "params_shard_31.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.4.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "0dc7e4320ed6e59ee246b53776e89204" |
|
}, |
|
{ |
|
"dataPath": "params_shard_32.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.3.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.3.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 3145728 |
|
}, |
|
{ |
|
"name": "model.layers.3.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 3151872 |
|
}, |
|
{ |
|
"name": "model.layers.3.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 11016192 |
|
}, |
|
{ |
|
"name": "model.layers.3.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 11999232 |
|
}, |
|
{ |
|
"name": "model.layers.3.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 16717824 |
|
}, |
|
{ |
|
"name": "model.layers.4.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.4.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 17313792 |
|
}, |
|
{ |
|
"name": "model.layers.4.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 29896704 |
|
} |
|
], |
|
"md5sum": "c98dabd5f0b7bff1744517f67b42586e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_33.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.5.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e4e7a7659dcfee0207b0770ef8a01d3e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_34.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.4.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.4.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 3145728 |
|
}, |
|
{ |
|
"name": "model.layers.4.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 3151872 |
|
}, |
|
{ |
|
"name": "model.layers.4.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 11016192 |
|
}, |
|
{ |
|
"name": "model.layers.4.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 11999232 |
|
}, |
|
{ |
|
"name": "model.layers.4.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 16717824 |
|
}, |
|
{ |
|
"name": "model.layers.5.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.5.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 17313792 |
|
}, |
|
{ |
|
"name": "model.layers.5.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 29896704 |
|
} |
|
], |
|
"md5sum": "a39c9d2897b1fa9232f9ea191432cbf9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_35.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.6.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "8f992450ffdca2c455ad94f5d0ecdb9f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_36.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.5.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.5.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 3145728 |
|
}, |
|
{ |
|
"name": "model.layers.5.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 3151872 |
|
}, |
|
{ |
|
"name": "model.layers.5.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 11016192 |
|
}, |
|
{ |
|
"name": "model.layers.5.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 11999232 |
|
}, |
|
{ |
|
"name": "model.layers.5.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 16717824 |
|
}, |
|
{ |
|
"name": "model.layers.6.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.6.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 17313792 |
|
}, |
|
{ |
|
"name": "model.layers.6.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 29896704 |
|
} |
|
], |
|
"md5sum": "4a98c643c60d02639b5c91fb23ab2cf5" |
|
}, |
|
{ |
|
"dataPath": "params_shard_37.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.7.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "212bc5c7bf2f55715a12e6a0a234783b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_38.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.6.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.6.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 3145728 |
|
}, |
|
{ |
|
"name": "model.layers.6.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 3151872 |
|
}, |
|
{ |
|
"name": "model.layers.6.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 11016192 |
|
}, |
|
{ |
|
"name": "model.layers.6.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 11999232 |
|
}, |
|
{ |
|
"name": "model.layers.6.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 16717824 |
|
}, |
|
{ |
|
"name": "model.layers.7.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.7.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 17313792 |
|
}, |
|
{ |
|
"name": "model.layers.7.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 29896704 |
|
} |
|
], |
|
"md5sum": "705ea22455c8b1108ab686600dfc605e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_39.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.8.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e32d2965b9bbf7ce3e36b781d00b85cf" |
|
}, |
|
{ |
|
"dataPath": "params_shard_40.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.7.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.7.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 3145728 |
|
}, |
|
{ |
|
"name": "model.layers.7.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 3151872 |
|
}, |
|
{ |
|
"name": "model.layers.7.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 11016192 |
|
}, |
|
{ |
|
"name": "model.layers.7.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 11999232 |
|
}, |
|
{ |
|
"name": "model.layers.7.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 16717824 |
|
}, |
|
{ |
|
"name": "model.layers.8.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.8.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 17313792 |
|
}, |
|
{ |
|
"name": "model.layers.8.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 29896704 |
|
} |
|
], |
|
"md5sum": "1e1412220bf9b83c851c1b8de784760b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_41.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.9.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "7df4f068caebe1c39ac96e4248a206fb" |
|
}, |
|
{ |
|
"dataPath": "params_shard_42.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.8.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.8.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 3145728 |
|
}, |
|
{ |
|
"name": "model.layers.8.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 3151872 |
|
}, |
|
{ |
|
"name": "model.layers.8.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 11016192 |
|
}, |
|
{ |
|
"name": "model.layers.8.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 11999232 |
|
}, |
|
{ |
|
"name": "model.layers.8.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 16717824 |
|
}, |
|
{ |
|
"name": "model.layers.9.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.9.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 17313792 |
|
}, |
|
{ |
|
"name": "model.layers.9.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 29896704 |
|
} |
|
], |
|
"md5sum": "f84fc002938813d38940dcb61de596d4" |
|
}, |
|
{ |
|
"dataPath": "params_shard_43.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31481856, |
|
"records": [ |
|
{ |
|
"name": "model.layers.9.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.9.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 3145728 |
|
}, |
|
{ |
|
"name": "model.layers.9.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 3151872 |
|
}, |
|
{ |
|
"name": "model.layers.9.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 11016192 |
|
}, |
|
{ |
|
"name": "model.layers.9.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 11999232 |
|
}, |
|
{ |
|
"name": "model.layers.9.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 16717824 |
|
}, |
|
{ |
|
"name": "model.layers.20.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.20.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 17313792 |
|
}, |
|
{ |
|
"name": "model.layers.20.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 29896704 |
|
}, |
|
{ |
|
"name": "model.layers.20.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31469568 |
|
}, |
|
{ |
|
"name": "model.layers.21.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31475712 |
|
} |
|
], |
|
"md5sum": "4c3012eb7b4c13eea5b6cba058082a6e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_44.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.21.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "7c01090cf74f63b2893b40b076cdb5c0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_45.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.21.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.21.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 12582912 |
|
}, |
|
{ |
|
"name": "model.layers.21.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 14155776 |
|
}, |
|
{ |
|
"name": "model.layers.21.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.21.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.21.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 25171968 |
|
}, |
|
{ |
|
"name": "model.layers.21.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 26155008 |
|
}, |
|
{ |
|
"name": "model.layers.21.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 30873600 |
|
}, |
|
{ |
|
"name": "model.layers.22.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31463424 |
|
} |
|
], |
|
"md5sum": "7d9d805a044146ead8648df95969ac18" |
|
}, |
|
{ |
|
"dataPath": "params_shard_46.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.22.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f1ca9ab711f493557a70e0ed3988525d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_47.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.22.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.22.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 12582912 |
|
}, |
|
{ |
|
"name": "model.layers.22.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 14155776 |
|
}, |
|
{ |
|
"name": "model.layers.22.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.22.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.22.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 25171968 |
|
}, |
|
{ |
|
"name": "model.layers.22.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 26155008 |
|
}, |
|
{ |
|
"name": "model.layers.22.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 30873600 |
|
}, |
|
{ |
|
"name": "model.layers.23.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31463424 |
|
} |
|
], |
|
"md5sum": "1c12d6f38dcd118bdc7f463ad9991e27" |
|
}, |
|
{ |
|
"dataPath": "params_shard_48.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.23.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "684665392fc7f3206862df65f2b0dea0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_49.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.23.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.23.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 12582912 |
|
}, |
|
{ |
|
"name": "model.layers.23.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 14155776 |
|
}, |
|
{ |
|
"name": "model.layers.23.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.23.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.23.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 25171968 |
|
}, |
|
{ |
|
"name": "model.layers.23.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 26155008 |
|
}, |
|
{ |
|
"name": "model.layers.23.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 30873600 |
|
}, |
|
{ |
|
"name": "model.layers.24.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31463424 |
|
} |
|
], |
|
"md5sum": "4a0753ced4ea985a0360b91a8fba84f3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_50.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.24.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "53934835b8866881b58ac2d465e5affa" |
|
}, |
|
{ |
|
"dataPath": "params_shard_51.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.24.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.24.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 12582912 |
|
}, |
|
{ |
|
"name": "model.layers.24.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 14155776 |
|
}, |
|
{ |
|
"name": "model.layers.24.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.24.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.24.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 25171968 |
|
}, |
|
{ |
|
"name": "model.layers.24.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 26155008 |
|
}, |
|
{ |
|
"name": "model.layers.24.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 30873600 |
|
}, |
|
{ |
|
"name": "model.layers.25.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31463424 |
|
} |
|
], |
|
"md5sum": "72368bd18f88ad0418b43128c9a2d935" |
|
}, |
|
{ |
|
"dataPath": "params_shard_52.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.25.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "54bab129e0aeec7c43854ee2914ed372" |
|
}, |
|
{ |
|
"dataPath": "params_shard_53.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.25.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.25.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 12582912 |
|
}, |
|
{ |
|
"name": "model.layers.25.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 14155776 |
|
}, |
|
{ |
|
"name": "model.layers.25.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.25.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.25.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 25171968 |
|
}, |
|
{ |
|
"name": "model.layers.25.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 26155008 |
|
}, |
|
{ |
|
"name": "model.layers.25.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 30873600 |
|
}, |
|
{ |
|
"name": "model.layers.26.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31463424 |
|
} |
|
], |
|
"md5sum": "25a77567530693d2df04dd34abd77a56" |
|
}, |
|
{ |
|
"dataPath": "params_shard_54.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.26.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "78487032bfa857e18fc0c0ee6240e45a" |
|
}, |
|
{ |
|
"dataPath": "params_shard_55.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.26.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.26.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 12582912 |
|
}, |
|
{ |
|
"name": "model.layers.26.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 14155776 |
|
}, |
|
{ |
|
"name": "model.layers.26.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.26.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.26.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 25171968 |
|
}, |
|
{ |
|
"name": "model.layers.26.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 26155008 |
|
}, |
|
{ |
|
"name": "model.layers.26.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 30873600 |
|
}, |
|
{ |
|
"name": "model.layers.27.input_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31463424 |
|
} |
|
], |
|
"md5sum": "2f35f6a239552f4d734abf51d7dd3eb9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_56.bin", |
|
"format": "raw-shard", |
|
"nbytes": 25165824, |
|
"records": [ |
|
{ |
|
"name": "model.layers.27.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
16384, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 25165824, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "70000cc4d6347ef7e8d89e0a211e45cb" |
|
}, |
|
{ |
|
"dataPath": "params_shard_57.bin", |
|
"format": "raw-shard", |
|
"nbytes": 31469568, |
|
"records": [ |
|
{ |
|
"name": "model.layers.27.mlp.down_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
1024 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 12582912, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.27.mlp.down_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
256 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1572864, |
|
"byteOffset": 12582912 |
|
}, |
|
{ |
|
"name": "model.layers.27.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
16384, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 3145728, |
|
"byteOffset": 14155776 |
|
}, |
|
{ |
|
"name": "model.layers.27.post_attention_layernorm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 17301504 |
|
}, |
|
{ |
|
"name": "model.layers.27.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 7864320, |
|
"byteOffset": 17307648 |
|
}, |
|
{ |
|
"name": "model.layers.27.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 983040, |
|
"byteOffset": 25171968 |
|
}, |
|
{ |
|
"name": "model.layers.27.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
3072, |
|
384 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4718592, |
|
"byteOffset": 26155008 |
|
}, |
|
{ |
|
"name": "model.layers.27.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
3072, |
|
96 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 589824, |
|
"byteOffset": 30873600 |
|
}, |
|
{ |
|
"name": "model.norm.weight", |
|
"shape": [ |
|
3072 |
|
], |
|
"dtype": "float16", |
|
"format": "f32-to-bf16", |
|
"nbytes": 6144, |
|
"byteOffset": 31463424 |
|
} |
|
], |
|
"md5sum": "ea44ff5f40b18c3091c83b62d36e2845" |
|
} |
|
] |
|
} |