numen-tech's picture
Add weights
29b7fb6
{
"metadata": {
"ParamSize": 283,
"ParamBytes": 1656834048.0,
"BitsPerParam": 4.069857841273854
},
"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": "2e2696b64d1c0d41210499c7dd159d2c"
},
{
"dataPath": "params_shard_1.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": "7d1a759f31f63011aed7803994cab7fe"
},
{
"dataPath": "params_shard_2.bin",
"format": "raw-shard",
"nbytes": 32913408,
"records": [
{
"name": "model.embed_tokens.q_scale",
"shape": [
128256,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6156288,
"byteOffset": 0
},
{
"name": "model.layers.0.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 6156288
},
{
"name": "model.layers.0.mlp.down_proj.q_weight",
"shape": [
3072,
1024
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 6162432
},
{
"name": "model.layers.0.mlp.down_proj.q_scale",
"shape": [
3072,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 18745344
},
{
"name": "model.layers.0.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 19138560
},
{
"name": "model.layers.0.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 19924992
},
{
"name": "model.layers.0.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 19931136
},
{
"name": "model.layers.0.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 27795456
},
{
"name": "model.layers.0.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 28041216
},
{
"name": "model.layers.0.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 32759808
},
{
"name": "model.layers.1.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 32907264
}
],
"md5sum": "4fa9b44a7e38c877edc418744039c09e"
},
{
"dataPath": "params_shard_3.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": "552bd134bf881040c6fc47bf263fd538"
},
{
"dataPath": "params_shard_4.bin",
"format": "raw-shard",
"nbytes": 26750976,
"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,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 12582912
},
{
"name": "model.layers.1.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 12976128
},
{
"name": "model.layers.1.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 13762560
},
{
"name": "model.layers.1.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13768704
},
{
"name": "model.layers.1.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21633024
},
{
"name": "model.layers.1.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 21878784
},
{
"name": "model.layers.1.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 26597376
},
{
"name": "model.layers.10.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "62e1e388be748b04945573edab8a7a1c"
},
{
"dataPath": "params_shard_5.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": "576fd927246ad52ec49963eea9990c26"
},
{
"dataPath": "params_shard_6.bin",
"format": "raw-shard",
"nbytes": 26750976,
"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,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 12582912
},
{
"name": "model.layers.10.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 12976128
},
{
"name": "model.layers.10.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 13762560
},
{
"name": "model.layers.10.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13768704
},
{
"name": "model.layers.10.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21633024
},
{
"name": "model.layers.10.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 21878784
},
{
"name": "model.layers.10.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 26597376
},
{
"name": "model.layers.11.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "7f6542902c660c9a79578cee72f459a5"
},
{
"dataPath": "params_shard_7.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": "a34070f3a5df7e8c4a537e31add8f2aa"
},
{
"dataPath": "params_shard_8.bin",
"format": "raw-shard",
"nbytes": 26750976,
"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,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 12582912
},
{
"name": "model.layers.11.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 12976128
},
{
"name": "model.layers.11.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 13762560
},
{
"name": "model.layers.11.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13768704
},
{
"name": "model.layers.11.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21633024
},
{
"name": "model.layers.11.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 21878784
},
{
"name": "model.layers.11.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 26597376
},
{
"name": "model.layers.12.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "ec72d1626666807c129293671d8ef0bb"
},
{
"dataPath": "params_shard_9.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": "16c0dcb6fb9a68bb602936854f5e357d"
},
{
"dataPath": "params_shard_10.bin",
"format": "raw-shard",
"nbytes": 26750976,
"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,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 12582912
},
{
"name": "model.layers.12.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 12976128
},
{
"name": "model.layers.12.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 13762560
},
{
"name": "model.layers.12.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13768704
},
{
"name": "model.layers.12.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21633024
},
{
"name": "model.layers.12.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 21878784
},
{
"name": "model.layers.12.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 26597376
},
{
"name": "model.layers.13.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "f51ed7991f7f0e26a12c64025d222ce0"
},
{
"dataPath": "params_shard_11.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": "4424b3dd76177e6568f6eca32654d9c0"
},
{
"dataPath": "params_shard_12.bin",
"format": "raw-shard",
"nbytes": 26750976,
"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,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 12582912
},
{
"name": "model.layers.13.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 12976128
},
{
"name": "model.layers.13.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 13762560
},
{
"name": "model.layers.13.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13768704
},
{
"name": "model.layers.13.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21633024
},
{
"name": "model.layers.13.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 21878784
},
{
"name": "model.layers.13.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 26597376
},
{
"name": "model.layers.14.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "5593a95bf7afedde2792436400edc604"
},
{
"dataPath": "params_shard_13.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": "63a38bfeae2e72af45fd21ffe27f974f"
},
{
"dataPath": "params_shard_14.bin",
"format": "raw-shard",
"nbytes": 26750976,
"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,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 12582912
},
{
"name": "model.layers.14.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 12976128
},
{
"name": "model.layers.14.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 13762560
},
{
"name": "model.layers.14.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13768704
},
{
"name": "model.layers.14.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21633024
},
{
"name": "model.layers.14.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 21878784
},
{
"name": "model.layers.14.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 26597376
},
{
"name": "model.layers.15.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "31c866c51f139163faa5d7f9db27ce1b"
},
{
"dataPath": "params_shard_15.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": "2ca98026c8d1e5891c89d3b4113a064c"
},
{
"dataPath": "params_shard_16.bin",
"format": "raw-shard",
"nbytes": 26750976,
"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,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 12582912
},
{
"name": "model.layers.15.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 12976128
},
{
"name": "model.layers.15.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 13762560
},
{
"name": "model.layers.15.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13768704
},
{
"name": "model.layers.15.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21633024
},
{
"name": "model.layers.15.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 21878784
},
{
"name": "model.layers.15.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 26597376
},
{
"name": "model.layers.16.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "59418f093d16141b44f29accdfed9658"
},
{
"dataPath": "params_shard_17.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": "33f26b45e77ca1bbefb6e1b74bfa215d"
},
{
"dataPath": "params_shard_18.bin",
"format": "raw-shard",
"nbytes": 26750976,
"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,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 12582912
},
{
"name": "model.layers.16.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 12976128
},
{
"name": "model.layers.16.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 13762560
},
{
"name": "model.layers.16.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13768704
},
{
"name": "model.layers.16.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21633024
},
{
"name": "model.layers.16.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 21878784
},
{
"name": "model.layers.16.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 26597376
},
{
"name": "model.layers.17.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "9f5cf264493dd9e1f2004472d1357f41"
},
{
"dataPath": "params_shard_19.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": "9d147b91d9df1e605b27cd0327ad0699"
},
{
"dataPath": "params_shard_20.bin",
"format": "raw-shard",
"nbytes": 26750976,
"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,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 12582912
},
{
"name": "model.layers.17.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 12976128
},
{
"name": "model.layers.17.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 13762560
},
{
"name": "model.layers.17.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13768704
},
{
"name": "model.layers.17.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21633024
},
{
"name": "model.layers.17.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 21878784
},
{
"name": "model.layers.17.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 26597376
},
{
"name": "model.layers.18.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "940411e3b0424545b230114941e5501a"
},
{
"dataPath": "params_shard_21.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": "9d28fdd14a162faed7fe49f146c192f1"
},
{
"dataPath": "params_shard_22.bin",
"format": "raw-shard",
"nbytes": 26750976,
"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,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 12582912
},
{
"name": "model.layers.18.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 12976128
},
{
"name": "model.layers.18.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 13762560
},
{
"name": "model.layers.18.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13768704
},
{
"name": "model.layers.18.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21633024
},
{
"name": "model.layers.18.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 21878784
},
{
"name": "model.layers.18.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 26597376
},
{
"name": "model.layers.19.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "2b06a4236515ad8b0000a8aea97e31b4"
},
{
"dataPath": "params_shard_23.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": "0786aa5398b8ddcf986b131678e18106"
},
{
"dataPath": "params_shard_24.bin",
"format": "raw-shard",
"nbytes": 26750976,
"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,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 12582912
},
{
"name": "model.layers.19.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 12976128
},
{
"name": "model.layers.19.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 13762560
},
{
"name": "model.layers.19.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13768704
},
{
"name": "model.layers.19.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21633024
},
{
"name": "model.layers.19.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 21878784
},
{
"name": "model.layers.19.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 26597376
},
{
"name": "model.layers.2.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "6caaa08992808da9307a0e775a142e36"
},
{
"dataPath": "params_shard_25.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": "6d95d89ac3de3da280c6416f03413fdb"
},
{
"dataPath": "params_shard_26.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": "47299923dd3222bed04ffb19e656887d"
},
{
"dataPath": "params_shard_27.bin",
"format": "raw-shard",
"nbytes": 27531264,
"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,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 12582912
},
{
"name": "model.layers.2.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 12976128
},
{
"name": "model.layers.2.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 13762560
},
{
"name": "model.layers.2.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13768704
},
{
"name": "model.layers.2.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21633024
},
{
"name": "model.layers.2.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 21878784
},
{
"name": "model.layers.2.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 26597376
},
{
"name": "model.layers.20.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 26744832
}
],
"md5sum": "d2e791fd7edb58dcd59a035b07edf9c3"
},
{
"dataPath": "params_shard_28.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": "a46cf32c2e6837bb59d0d805e548bdfb"
},
{
"dataPath": "params_shard_29.bin",
"format": "raw-shard",
"nbytes": 26750976,
"records": [
{
"name": "model.layers.20.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 0
},
{
"name": "model.layers.20.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 7864320
},
{
"name": "model.layers.20.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 8110080
},
{
"name": "model.layers.20.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 12828672
},
{
"name": "model.layers.3.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 12976128
},
{
"name": "model.layers.3.mlp.down_proj.q_weight",
"shape": [
3072,
1024
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 12982272
},
{
"name": "model.layers.3.mlp.down_proj.q_scale",
"shape": [
3072,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 25565184
},
{
"name": "model.layers.3.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 25958400
},
{
"name": "model.layers.3.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "fcfd966ece62b43f0812770971460e1f"
},
{
"dataPath": "params_shard_30.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": "7bff3b9fba40e2ae130e2f70e4c93f8f"
},
{
"dataPath": "params_shard_31.bin",
"format": "raw-shard",
"nbytes": 26750976,
"records": [
{
"name": "model.layers.3.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 0
},
{
"name": "model.layers.3.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 7864320
},
{
"name": "model.layers.3.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 8110080
},
{
"name": "model.layers.3.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 12828672
},
{
"name": "model.layers.4.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 12976128
},
{
"name": "model.layers.4.mlp.down_proj.q_weight",
"shape": [
3072,
1024
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 12982272
},
{
"name": "model.layers.4.mlp.down_proj.q_scale",
"shape": [
3072,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 25565184
},
{
"name": "model.layers.4.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 25958400
},
{
"name": "model.layers.4.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "7b582c6c971838aaa6ad544d4d5f17ca"
},
{
"dataPath": "params_shard_32.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": "6c8218a0300eb9d4a1b8fd111115d91f"
},
{
"dataPath": "params_shard_33.bin",
"format": "raw-shard",
"nbytes": 26750976,
"records": [
{
"name": "model.layers.4.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 0
},
{
"name": "model.layers.4.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 7864320
},
{
"name": "model.layers.4.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 8110080
},
{
"name": "model.layers.4.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 12828672
},
{
"name": "model.layers.5.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 12976128
},
{
"name": "model.layers.5.mlp.down_proj.q_weight",
"shape": [
3072,
1024
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 12982272
},
{
"name": "model.layers.5.mlp.down_proj.q_scale",
"shape": [
3072,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 25565184
},
{
"name": "model.layers.5.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 25958400
},
{
"name": "model.layers.5.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "8adb5562ba0bfff8f353f144e89a1e32"
},
{
"dataPath": "params_shard_34.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": "2def1d90946b41ec3329e323e62772f2"
},
{
"dataPath": "params_shard_35.bin",
"format": "raw-shard",
"nbytes": 26750976,
"records": [
{
"name": "model.layers.5.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 0
},
{
"name": "model.layers.5.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 7864320
},
{
"name": "model.layers.5.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 8110080
},
{
"name": "model.layers.5.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 12828672
},
{
"name": "model.layers.6.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 12976128
},
{
"name": "model.layers.6.mlp.down_proj.q_weight",
"shape": [
3072,
1024
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 12982272
},
{
"name": "model.layers.6.mlp.down_proj.q_scale",
"shape": [
3072,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 25565184
},
{
"name": "model.layers.6.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 25958400
},
{
"name": "model.layers.6.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "4a75be197e1e4bc55b5a66f10a3877d1"
},
{
"dataPath": "params_shard_36.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": "ec92853afb8b3bfadfbe75d596835e06"
},
{
"dataPath": "params_shard_37.bin",
"format": "raw-shard",
"nbytes": 26750976,
"records": [
{
"name": "model.layers.6.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 0
},
{
"name": "model.layers.6.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 7864320
},
{
"name": "model.layers.6.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 8110080
},
{
"name": "model.layers.6.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 12828672
},
{
"name": "model.layers.7.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 12976128
},
{
"name": "model.layers.7.mlp.down_proj.q_weight",
"shape": [
3072,
1024
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 12982272
},
{
"name": "model.layers.7.mlp.down_proj.q_scale",
"shape": [
3072,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 25565184
},
{
"name": "model.layers.7.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 25958400
},
{
"name": "model.layers.7.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "be460dca6fe228e1e2975ada5d3a363c"
},
{
"dataPath": "params_shard_38.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": "d2bf211100d8a36a05d3571171ce1113"
},
{
"dataPath": "params_shard_39.bin",
"format": "raw-shard",
"nbytes": 26750976,
"records": [
{
"name": "model.layers.7.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 0
},
{
"name": "model.layers.7.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 7864320
},
{
"name": "model.layers.7.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 8110080
},
{
"name": "model.layers.7.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 12828672
},
{
"name": "model.layers.8.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 12976128
},
{
"name": "model.layers.8.mlp.down_proj.q_weight",
"shape": [
3072,
1024
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 12982272
},
{
"name": "model.layers.8.mlp.down_proj.q_scale",
"shape": [
3072,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 25565184
},
{
"name": "model.layers.8.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 25958400
},
{
"name": "model.layers.8.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "786f2e137fcdfd24854968d4e4df1343"
},
{
"dataPath": "params_shard_40.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": "b291287995242368deab9f2fd22789db"
},
{
"dataPath": "params_shard_41.bin",
"format": "raw-shard",
"nbytes": 26750976,
"records": [
{
"name": "model.layers.8.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 0
},
{
"name": "model.layers.8.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 7864320
},
{
"name": "model.layers.8.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 8110080
},
{
"name": "model.layers.8.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 12828672
},
{
"name": "model.layers.9.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 12976128
},
{
"name": "model.layers.9.mlp.down_proj.q_weight",
"shape": [
3072,
1024
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 12982272
},
{
"name": "model.layers.9.mlp.down_proj.q_scale",
"shape": [
3072,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 25565184
},
{
"name": "model.layers.9.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 25958400
},
{
"name": "model.layers.9.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "d3673ab0bfe96d481e356104aa96ba7c"
},
{
"dataPath": "params_shard_42.bin",
"format": "raw-shard",
"nbytes": 25970688,
"records": [
{
"name": "model.layers.9.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 0
},
{
"name": "model.layers.9.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 7864320
},
{
"name": "model.layers.9.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 8110080
},
{
"name": "model.layers.9.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 12828672
},
{
"name": "model.layers.20.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 12976128
},
{
"name": "model.layers.20.mlp.down_proj.q_weight",
"shape": [
3072,
1024
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 12982272
},
{
"name": "model.layers.20.mlp.down_proj.q_scale",
"shape": [
3072,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 25565184
},
{
"name": "model.layers.20.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 25958400
},
{
"name": "model.layers.21.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 25964544
}
],
"md5sum": "d6826537e784b8c30879fcb784ff0953"
},
{
"dataPath": "params_shard_43.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": "c04aab3cceaebb001b3ba2a16b1b9087"
},
{
"dataPath": "params_shard_44.bin",
"format": "raw-shard",
"nbytes": 26750976,
"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,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 12582912
},
{
"name": "model.layers.21.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 12976128
},
{
"name": "model.layers.21.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 13762560
},
{
"name": "model.layers.21.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13768704
},
{
"name": "model.layers.21.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21633024
},
{
"name": "model.layers.21.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 21878784
},
{
"name": "model.layers.21.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 26597376
},
{
"name": "model.layers.22.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "f4fbaf447d2bbdaf1699255ee1546683"
},
{
"dataPath": "params_shard_45.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": "c90fc9aa8909b2aecb74a9986fe1e9d6"
},
{
"dataPath": "params_shard_46.bin",
"format": "raw-shard",
"nbytes": 26750976,
"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,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 12582912
},
{
"name": "model.layers.22.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 12976128
},
{
"name": "model.layers.22.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 13762560
},
{
"name": "model.layers.22.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13768704
},
{
"name": "model.layers.22.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21633024
},
{
"name": "model.layers.22.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 21878784
},
{
"name": "model.layers.22.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 26597376
},
{
"name": "model.layers.23.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "08fe55b6ad5639b2309942a610b203d8"
},
{
"dataPath": "params_shard_47.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": "cacc9e759c1036cbdb978db57026a18b"
},
{
"dataPath": "params_shard_48.bin",
"format": "raw-shard",
"nbytes": 26750976,
"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,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 12582912
},
{
"name": "model.layers.23.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 12976128
},
{
"name": "model.layers.23.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 13762560
},
{
"name": "model.layers.23.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13768704
},
{
"name": "model.layers.23.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21633024
},
{
"name": "model.layers.23.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 21878784
},
{
"name": "model.layers.23.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 26597376
},
{
"name": "model.layers.24.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "6ca9ff8bdd119d7521ed2b54e6ea9630"
},
{
"dataPath": "params_shard_49.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": "c6ba9f9cae1a45c8e1ad5f4ef6c36ff3"
},
{
"dataPath": "params_shard_50.bin",
"format": "raw-shard",
"nbytes": 26750976,
"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,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 12582912
},
{
"name": "model.layers.24.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 12976128
},
{
"name": "model.layers.24.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 13762560
},
{
"name": "model.layers.24.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13768704
},
{
"name": "model.layers.24.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21633024
},
{
"name": "model.layers.24.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 21878784
},
{
"name": "model.layers.24.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 26597376
},
{
"name": "model.layers.25.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "239f518bb2b060da6ecaedac7b2e7be6"
},
{
"dataPath": "params_shard_51.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": "854e30dfdf7ba9ca315204fc102463d2"
},
{
"dataPath": "params_shard_52.bin",
"format": "raw-shard",
"nbytes": 26750976,
"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,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 12582912
},
{
"name": "model.layers.25.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 12976128
},
{
"name": "model.layers.25.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 13762560
},
{
"name": "model.layers.25.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13768704
},
{
"name": "model.layers.25.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21633024
},
{
"name": "model.layers.25.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 21878784
},
{
"name": "model.layers.25.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 26597376
},
{
"name": "model.layers.26.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "7b311814e52585a710e36fa606a016a4"
},
{
"dataPath": "params_shard_53.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": "2fda2e73242085073444fc3efdb31aaa"
},
{
"dataPath": "params_shard_54.bin",
"format": "raw-shard",
"nbytes": 26750976,
"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,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 12582912
},
{
"name": "model.layers.26.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 12976128
},
{
"name": "model.layers.26.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 13762560
},
{
"name": "model.layers.26.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13768704
},
{
"name": "model.layers.26.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21633024
},
{
"name": "model.layers.26.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 21878784
},
{
"name": "model.layers.26.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 26597376
},
{
"name": "model.layers.27.input_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "74e805615f2b37eb18715ca320974c59"
},
{
"dataPath": "params_shard_55.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": "1eb3b29c9b4ac03e041ee1b8aad9888e"
},
{
"dataPath": "params_shard_56.bin",
"format": "raw-shard",
"nbytes": 26750976,
"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,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 393216,
"byteOffset": 12582912
},
{
"name": "model.layers.27.mlp.gate_up_proj.q_scale",
"shape": [
16384,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 12976128
},
{
"name": "model.layers.27.post_attention_layernorm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 13762560
},
{
"name": "model.layers.27.self_attn.qkv_proj.q_weight",
"shape": [
5120,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13768704
},
{
"name": "model.layers.27.self_attn.qkv_proj.q_scale",
"shape": [
5120,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21633024
},
{
"name": "model.layers.27.self_attn.o_proj.q_weight",
"shape": [
3072,
384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 4718592,
"byteOffset": 21878784
},
{
"name": "model.layers.27.self_attn.o_proj.q_scale",
"shape": [
3072,
24
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 147456,
"byteOffset": 26597376
},
{
"name": "model.norm.weight",
"shape": [
3072
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6144,
"byteOffset": 26744832
}
],
"md5sum": "0ee1a3c7d8effe6df65ebffe3b2570c1"
}
]
}