Hugging Face's logo

yecanlee
/
Mistral-7B-v0.3

Model card Files Community
Mistral-7B-v0.3 / ndarray-cache.json
yecanlee's picture
yecanlee
Add redpajama-3b instruct model weights
9203f82
raw
history blame contribute delete
148 kB
{
"metadata": {
"ParamSize": 325,
"ParamBytes": 4077395968.0,
"BitsPerParam": 4.500422426883416
},
"records": [
{
"dataPath": "params_shard_0.bin",
"format": "raw-shard",
"nbytes": 67108864,
"records": [
{
"name": "lm_head.q_weight",
"shape": [
512,
32768
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 67108864,
"byteOffset": 0
}
],
"md5sum": "bb71541d634b24d47c951efc10e0a6a7"
},
{
"dataPath": "params_shard_1.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.22.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "30c9ba84e27c213ed3c4d0ca8bb5da61"
},
{
"dataPath": "params_shard_2.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.22.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "749a7267eda5694658e7f0fcd4ff3ead"
},
{
"dataPath": "params_shard_3.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.23.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "afdd755539fbf408a3bc405e212582da"
},
{
"dataPath": "params_shard_4.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.23.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "cd963becf328656e5246d80dc0a70c4c"
},
{
"dataPath": "params_shard_5.bin",
"format": "raw-shard",
"nbytes": 30441472,
"records": [
{
"name": "lm_head.q_scale",
"shape": [
128,
32768
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.22.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.22.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 8396800
},
{
"name": "model.layers.22.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 12066816
},
{
"name": "model.layers.22.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 19406848
},
{
"name": "model.layers.23.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 19415040
},
{
"name": "model.layers.23.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 19423232
},
{
"name": "model.layers.23.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 23093248
},
{
"name": "model.layers.23.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30433280
}
],
"md5sum": "6411f2ce14c85793abf8733748f92cf6"
},
{
"dataPath": "params_shard_6.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.24.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "5c6b28d3c1572b3ededfe4161d1aaecd"
},
{
"dataPath": "params_shard_7.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.24.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "c994343930586ee7300c692818afe162"
},
{
"dataPath": "params_shard_8.bin",
"format": "raw-shard",
"nbytes": 27271168,
"records": [
{
"name": "model.layers.23.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 0
},
{
"name": "model.layers.23.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 12582912
},
{
"name": "model.layers.23.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 14155776
},
{
"name": "model.layers.23.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 22544384
},
{
"name": "model.layers.24.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 23592960
},
{
"name": "model.layers.24.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 23601152
}
],
"md5sum": "6f8b28cecfdbdaf9d51e334d469a8e55"
},
{
"dataPath": "params_shard_9.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.25.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "8983a14a512022abcb2dfa7400d39ce8"
},
{
"dataPath": "params_shard_10.bin",
"format": "raw-shard",
"nbytes": 30949376,
"records": [
{
"name": "model.layers.24.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 0
},
{
"name": "model.layers.24.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 7340032
},
{
"name": "model.layers.24.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 7348224
},
{
"name": "model.layers.24.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 19931136
},
{
"name": "model.layers.24.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 21504000
},
{
"name": "model.layers.24.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 29892608
},
{
"name": "model.layers.25.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30941184
}
],
"md5sum": "94a7fc08c2fe774a89dc07a5858b14c4"
},
{
"dataPath": "params_shard_11.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.25.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "c559398183f4e82d67456864431bfd8e"
},
{
"dataPath": "params_shard_12.bin",
"format": "raw-shard",
"nbytes": 25174016,
"records": [
{
"name": "model.layers.25.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 0
},
{
"name": "model.layers.25.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 3670016
},
{
"name": "model.layers.25.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 11010048
},
{
"name": "model.layers.25.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 11018240
},
{
"name": "model.layers.25.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 23601152
}
],
"md5sum": "9fcbb6ac218bc2be3414a84c26cd2401"
},
{
"dataPath": "params_shard_13.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.26.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "ea28414b6a5be3d77c0a79472b372a0b"
},
{
"dataPath": "params_shard_14.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.26.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "232cc1cebdf834a97bd24f9a611e3d01"
},
{
"dataPath": "params_shard_15.bin",
"format": "raw-shard",
"nbytes": 33046528,
"records": [
{
"name": "model.layers.25.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.25.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.26.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 9437184
},
{
"name": "model.layers.26.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 9445376
},
{
"name": "model.layers.26.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 13115392
},
{
"name": "model.layers.26.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 20455424
},
{
"name": "model.layers.26.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 20463616
}
],
"md5sum": "74eced314cc89b2c1000cec540af98e2"
},
{
"dataPath": "params_shard_16.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.27.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "b28120489e3d87c942f1c5dfd61368df"
},
{
"dataPath": "params_shard_17.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.27.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "4d519824a9bda455605bd13e23ccfc9b"
},
{
"dataPath": "params_shard_18.bin",
"format": "raw-shard",
"nbytes": 22036480,
"records": [
{
"name": "model.layers.26.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 0
},
{
"name": "model.layers.26.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 1572864
},
{
"name": "model.layers.26.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 9961472
},
{
"name": "model.layers.27.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 11010048
},
{
"name": "model.layers.27.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 11018240
},
{
"name": "model.layers.27.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 14688256
},
{
"name": "model.layers.27.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 22028288
}
],
"md5sum": "6ccef7aed4ac82e14130c935e269e228"
},
{
"dataPath": "params_shard_19.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.28.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "506793dc4b830df8ba3d5aacc3a2167e"
},
{
"dataPath": "params_shard_20.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.28.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "4b6d988c80db22e08ecce9fe9822bbd2"
},
{
"dataPath": "params_shard_21.bin",
"format": "raw-shard",
"nbytes": 27271168,
"records": [
{
"name": "model.layers.27.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 0
},
{
"name": "model.layers.27.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 12582912
},
{
"name": "model.layers.27.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 14155776
},
{
"name": "model.layers.27.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 22544384
},
{
"name": "model.layers.28.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 23592960
},
{
"name": "model.layers.28.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 23601152
}
],
"md5sum": "728653c562c856a0737b900bc3a72c29"
},
{
"dataPath": "params_shard_22.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.29.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "fbec58732c3ed3db93927910d7fc8b26"
},
{
"dataPath": "params_shard_23.bin",
"format": "raw-shard",
"nbytes": 30949376,
"records": [
{
"name": "model.layers.28.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 0
},
{
"name": "model.layers.28.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 7340032
},
{
"name": "model.layers.28.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 7348224
},
{
"name": "model.layers.28.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 19931136
},
{
"name": "model.layers.28.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 21504000
},
{
"name": "model.layers.28.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 29892608
},
{
"name": "model.layers.29.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30941184
}
],
"md5sum": "32045bea789a71b6356a64e3f509e88a"
},
{
"dataPath": "params_shard_24.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.29.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "4dfd868e81846c9317bb3ac6f6b77890"
},
{
"dataPath": "params_shard_25.bin",
"format": "raw-shard",
"nbytes": 25174016,
"records": [
{
"name": "model.layers.29.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 0
},
{
"name": "model.layers.29.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 3670016
},
{
"name": "model.layers.29.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 11010048
},
{
"name": "model.layers.29.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 11018240
},
{
"name": "model.layers.29.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 23601152
}
],
"md5sum": "235402ce06ff8e5053e3695edd8b0b5e"
},
{
"dataPath": "params_shard_26.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.30.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "5cc97a395c98555c0f6136a76fcd95ef"
},
{
"dataPath": "params_shard_27.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.30.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "ed76f20712c5d871e7b964eb300acc99"
},
{
"dataPath": "params_shard_28.bin",
"format": "raw-shard",
"nbytes": 33046528,
"records": [
{
"name": "model.layers.29.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.29.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.30.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 9437184
},
{
"name": "model.layers.30.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 9445376
},
{
"name": "model.layers.30.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 13115392
},
{
"name": "model.layers.30.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 20455424
},
{
"name": "model.layers.30.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 20463616
}
],
"md5sum": "a849a43df69ffe6b8261f9e296da411b"
},
{
"dataPath": "params_shard_29.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.31.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "f2335be112354c123a75efdf77a04ada"
},
{
"dataPath": "params_shard_30.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.31.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "93e3188514428d5df296f5f26f227f79"
},
{
"dataPath": "params_shard_31.bin",
"format": "raw-shard",
"nbytes": 22036480,
"records": [
{
"name": "model.layers.30.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 0
},
{
"name": "model.layers.30.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 1572864
},
{
"name": "model.layers.30.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 9961472
},
{
"name": "model.layers.31.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 11010048
},
{
"name": "model.layers.31.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 11018240
},
{
"name": "model.layers.31.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 14688256
},
{
"name": "model.layers.31.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 22028288
}
],
"md5sum": "9cfc1577492aaea3ddbd5df3724a5439"
},
{
"dataPath": "params_shard_32.bin",
"format": "raw-shard",
"nbytes": 67108864,
"records": [
{
"name": "model.embed_tokens.q_weight",
"shape": [
32768,
512
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 67108864,
"byteOffset": 0
}
],
"md5sum": "81eb4d4d027f14ed794a5d66f09b76ae"
},
{
"dataPath": "params_shard_33.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.0.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "0d03003b1cb2bcfaf5f3974542f33238"
},
{
"dataPath": "params_shard_34.bin",
"format": "raw-shard",
"nbytes": 31997952,
"records": [
{
"name": "model.layers.31.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 0
},
{
"name": "model.layers.31.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 12582912
},
{
"name": "model.layers.31.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 14155776
},
{
"name": "model.layers.31.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 22544384
},
{
"name": "model.norm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 23592960
},
{
"name": "model.embed_tokens.q_scale",
"shape": [
32768,
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 23601152
},
{
"name": "model.layers.0.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31989760
}
],
"md5sum": "966f460cb42069df5fbb85bb06273037"
},
{
"dataPath": "params_shard_35.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.0.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "c6506760adb75d333f6cd407ad769baa"
},
{
"dataPath": "params_shard_36.bin",
"format": "raw-shard",
"nbytes": 25174016,
"records": [
{
"name": "model.layers.0.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 0
},
{
"name": "model.layers.0.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 3670016
},
{
"name": "model.layers.0.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 11010048
},
{
"name": "model.layers.0.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 11018240
},
{
"name": "model.layers.0.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 23601152
}
],
"md5sum": "c889ea1a8ab39bb63e665c8ea08d322f"
},
{
"dataPath": "params_shard_37.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.1.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "586af544c6d35086fa9de3e11207822d"
},
{
"dataPath": "params_shard_38.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.1.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "13102fc98dafcf7103917f498a4b2c98"
},
{
"dataPath": "params_shard_39.bin",
"format": "raw-shard",
"nbytes": 33046528,
"records": [
{
"name": "model.layers.0.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.0.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.1.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 9437184
},
{
"name": "model.layers.1.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 9445376
},
{
"name": "model.layers.1.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 13115392
},
{
"name": "model.layers.1.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 20455424
},
{
"name": "model.layers.1.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 20463616
}
],
"md5sum": "82e5216316eb049b4c237684400c70da"
},
{
"dataPath": "params_shard_40.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.10.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "aeaf79dacfd605ae19303a4d7d4efdd8"
},
{
"dataPath": "params_shard_41.bin",
"format": "raw-shard",
"nbytes": 32505856,
"records": [
{
"name": "model.layers.1.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 0
},
{
"name": "model.layers.1.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 1572864
},
{
"name": "model.layers.1.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 9961472
},
{
"name": "model.layers.10.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 11010048
},
{
"name": "model.layers.10.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 18350080
},
{
"name": "model.layers.10.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 30932992
}
],
"md5sum": "c8954b8d33b41b7a9df1fab10d62eb18"
},
{
"dataPath": "params_shard_42.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.2.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "db0e1544fa5a765ae9c204a2c92f0c4c"
},
{
"dataPath": "params_shard_43.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.2.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "091ad8a9ad19c88e45f8d3dbffd15fde"
},
{
"dataPath": "params_shard_44.bin",
"format": "raw-shard",
"nbytes": 33046528,
"records": [
{
"name": "model.layers.10.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.10.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.2.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 9437184
},
{
"name": "model.layers.2.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 9445376
},
{
"name": "model.layers.2.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 13115392
},
{
"name": "model.layers.2.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 20455424
},
{
"name": "model.layers.2.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 20463616
}
],
"md5sum": "c7210365eeff5b0e926377e6335b8b91"
},
{
"dataPath": "params_shard_45.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.3.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "ad1c22c2da4b9cebec9c579f73510e45"
},
{
"dataPath": "params_shard_46.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.3.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "c7fcc89f18ff6bc18b68dd6ca8706858"
},
{
"dataPath": "params_shard_47.bin",
"format": "raw-shard",
"nbytes": 22036480,
"records": [
{
"name": "model.layers.2.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 0
},
{
"name": "model.layers.2.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 1572864
},
{
"name": "model.layers.2.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 9961472
},
{
"name": "model.layers.3.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 11010048
},
{
"name": "model.layers.3.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 11018240
},
{
"name": "model.layers.3.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 14688256
},
{
"name": "model.layers.3.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 22028288
}
],
"md5sum": "dd258bef4d34e995800d042a61608122"
},
{
"dataPath": "params_shard_48.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.4.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "7baaae81f1a7b26c425924e518f55f93"
},
{
"dataPath": "params_shard_49.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.4.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "d7c81c625a773d4edb6aee5410862635"
},
{
"dataPath": "params_shard_50.bin",
"format": "raw-shard",
"nbytes": 27271168,
"records": [
{
"name": "model.layers.3.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 0
},
{
"name": "model.layers.3.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 12582912
},
{
"name": "model.layers.3.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 14155776
},
{
"name": "model.layers.3.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 22544384
},
{
"name": "model.layers.4.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 23592960
},
{
"name": "model.layers.4.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 23601152
}
],
"md5sum": "a71f0732de9d3bb9b799c800e0ddeaf4"
},
{
"dataPath": "params_shard_51.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.5.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "db8f2f744cf8aacb5c52b6b4453f4cd7"
},
{
"dataPath": "params_shard_52.bin",
"format": "raw-shard",
"nbytes": 30949376,
"records": [
{
"name": "model.layers.4.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 0
},
{
"name": "model.layers.4.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 7340032
},
{
"name": "model.layers.4.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 7348224
},
{
"name": "model.layers.4.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 19931136
},
{
"name": "model.layers.4.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 21504000
},
{
"name": "model.layers.4.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 29892608
},
{
"name": "model.layers.5.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30941184
}
],
"md5sum": "f92c44fe12c19a4e712dd5b4fc43fb81"
},
{
"dataPath": "params_shard_53.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.5.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "336aca9a103ce6348357bec4c728b00a"
},
{
"dataPath": "params_shard_54.bin",
"format": "raw-shard",
"nbytes": 25174016,
"records": [
{
"name": "model.layers.5.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 0
},
{
"name": "model.layers.5.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 3670016
},
{
"name": "model.layers.5.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 11010048
},
{
"name": "model.layers.5.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 11018240
},
{
"name": "model.layers.5.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 23601152
}
],
"md5sum": "7411bb984ff4236614362c3682657331"
},
{
"dataPath": "params_shard_55.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.6.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "9344cfcef3602ba1a353b68a9f01fcd6"
},
{
"dataPath": "params_shard_56.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.6.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "c3178a18cae8cce3b7c87b0f9dd90e85"
},
{
"dataPath": "params_shard_57.bin",
"format": "raw-shard",
"nbytes": 33046528,
"records": [
{
"name": "model.layers.5.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.5.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.6.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 9437184
},
{
"name": "model.layers.6.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 9445376
},
{
"name": "model.layers.6.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 13115392
},
{
"name": "model.layers.6.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 20455424
},
{
"name": "model.layers.6.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 20463616
}
],
"md5sum": "a9da1336e29903913693254be8cdad97"
},
{
"dataPath": "params_shard_58.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.7.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "b0ab5b4e8d57737abe0dd0556d19932c"
},
{
"dataPath": "params_shard_59.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.7.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "191fe3415d607cfba99aa70ab0e42396"
},
{
"dataPath": "params_shard_60.bin",
"format": "raw-shard",
"nbytes": 22036480,
"records": [
{
"name": "model.layers.6.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 0
},
{
"name": "model.layers.6.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 1572864
},
{
"name": "model.layers.6.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 9961472
},
{
"name": "model.layers.7.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 11010048
},
{
"name": "model.layers.7.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 11018240
},
{
"name": "model.layers.7.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 14688256
},
{
"name": "model.layers.7.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 22028288
}
],
"md5sum": "fca2cbf7b535bd77a1390f63fc7b2699"
},
{
"dataPath": "params_shard_61.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.8.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "ccf591d25494730f01758427ecfdc983"
},
{
"dataPath": "params_shard_62.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.8.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "ef077eb2ba7afc1930d51d6d2261bdcf"
},
{
"dataPath": "params_shard_63.bin",
"format": "raw-shard",
"nbytes": 27271168,
"records": [
{
"name": "model.layers.7.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 0
},
{
"name": "model.layers.7.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 12582912
},
{
"name": "model.layers.7.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 14155776
},
{
"name": "model.layers.7.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 22544384
},
{
"name": "model.layers.8.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 23592960
},
{
"name": "model.layers.8.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 23601152
}
],
"md5sum": "2c8961d9ddafab3a0055ada2e2c40d8c"
},
{
"dataPath": "params_shard_64.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.9.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "e7cd0bd7cbb728b68b38bc87a33d4bd1"
},
{
"dataPath": "params_shard_65.bin",
"format": "raw-shard",
"nbytes": 30949376,
"records": [
{
"name": "model.layers.8.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 0
},
{
"name": "model.layers.8.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 7340032
},
{
"name": "model.layers.8.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 7348224
},
{
"name": "model.layers.8.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 19931136
},
{
"name": "model.layers.8.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 21504000
},
{
"name": "model.layers.8.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 29892608
},
{
"name": "model.layers.9.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30941184
}
],
"md5sum": "5e25094f172fc2c41ee14579ea89dce7"
},
{
"dataPath": "params_shard_66.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.9.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "e2913383a4700b75eff1778475066893"
},
{
"dataPath": "params_shard_67.bin",
"format": "raw-shard",
"nbytes": 25174016,
"records": [
{
"name": "model.layers.9.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 0
},
{
"name": "model.layers.9.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 3670016
},
{
"name": "model.layers.9.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 11010048
},
{
"name": "model.layers.9.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 11018240
},
{
"name": "model.layers.9.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 23601152
}
],
"md5sum": "46260a7de0b80a33ceed5ae68b6b7994"
},
{
"dataPath": "params_shard_68.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.10.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "f7422a86b9601017a8b09bb31814141a"
},
{
"dataPath": "params_shard_69.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.11.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "4859f4cea54e1307823e5307d1acf0c9"
},
{
"dataPath": "params_shard_70.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.11.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "7f831ea495fc65f89fe869b5e9577d96"
},
{
"dataPath": "params_shard_71.bin",
"format": "raw-shard",
"nbytes": 24150016,
"records": [
{
"name": "model.layers.9.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.9.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.10.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 9437184
},
{
"name": "model.layers.10.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 9445376
},
{
"name": "model.layers.10.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 13115392
},
{
"name": "model.layers.11.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 13123584
},
{
"name": "model.layers.11.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 13131776
},
{
"name": "model.layers.11.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 16801792
},
{
"name": "model.layers.11.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 24141824
}
],
"md5sum": "ff1fbc815fc5df5b30f8b033b8c75557"
},
{
"dataPath": "params_shard_72.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.12.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "661f0ac470cd7eff0e1e398e314501d6"
},
{
"dataPath": "params_shard_73.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.12.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "1f3fe11c757ad975fd94670ac1f61c44"
},
{
"dataPath": "params_shard_74.bin",
"format": "raw-shard",
"nbytes": 27271168,
"records": [
{
"name": "model.layers.11.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 0
},
{
"name": "model.layers.11.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 12582912
},
{
"name": "model.layers.11.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 14155776
},
{
"name": "model.layers.11.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 22544384
},
{
"name": "model.layers.12.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 23592960
},
{
"name": "model.layers.12.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 23601152
}
],
"md5sum": "2a0fefc229d1d5d6485cd89bb3eccdb5"
},
{
"dataPath": "params_shard_75.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.13.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "5878c1a8a51a56bc39a586f84f257050"
},
{
"dataPath": "params_shard_76.bin",
"format": "raw-shard",
"nbytes": 30949376,
"records": [
{
"name": "model.layers.12.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 0
},
{
"name": "model.layers.12.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 7340032
},
{
"name": "model.layers.12.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 7348224
},
{
"name": "model.layers.12.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 19931136
},
{
"name": "model.layers.12.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 21504000
},
{
"name": "model.layers.12.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 29892608
},
{
"name": "model.layers.13.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30941184
}
],
"md5sum": "5ea19d523b4350905fd496fe95e01d01"
},
{
"dataPath": "params_shard_77.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.13.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "7d83f936eda33a471bbcfc3316e48c1a"
},
{
"dataPath": "params_shard_78.bin",
"format": "raw-shard",
"nbytes": 25174016,
"records": [
{
"name": "model.layers.13.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 0
},
{
"name": "model.layers.13.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 3670016
},
{
"name": "model.layers.13.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 11010048
},
{
"name": "model.layers.13.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 11018240
},
{
"name": "model.layers.13.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 23601152
}
],
"md5sum": "0ae387011fa58eb05a1d0ec85b699055"
},
{
"dataPath": "params_shard_79.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.14.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "5b350b4a922f2ba679224032f9ce6592"
},
{
"dataPath": "params_shard_80.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.14.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "e4def9ed00c2b4ced5af37d635848274"
},
{
"dataPath": "params_shard_81.bin",
"format": "raw-shard",
"nbytes": 33046528,
"records": [
{
"name": "model.layers.13.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.13.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.14.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 9437184
},
{
"name": "model.layers.14.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 9445376
},
{
"name": "model.layers.14.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 13115392
},
{
"name": "model.layers.14.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 20455424
},
{
"name": "model.layers.14.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 20463616
}
],
"md5sum": "058222e7f3c92367c318789d204eb2a4"
},
{
"dataPath": "params_shard_82.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.15.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "63a77c75d4af22efecf1b1330b55ca13"
},
{
"dataPath": "params_shard_83.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.15.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "bff33f372090eee883e2e106214fc484"
},
{
"dataPath": "params_shard_84.bin",
"format": "raw-shard",
"nbytes": 22036480,
"records": [
{
"name": "model.layers.14.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 0
},
{
"name": "model.layers.14.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 1572864
},
{
"name": "model.layers.14.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 9961472
},
{
"name": "model.layers.15.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 11010048
},
{
"name": "model.layers.15.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 11018240
},
{
"name": "model.layers.15.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 14688256
},
{
"name": "model.layers.15.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 22028288
}
],
"md5sum": "d53703d56fac339a396eaf15cc794009"
},
{
"dataPath": "params_shard_85.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.16.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "26c1c0445e7a59b39036cd293b0172c8"
},
{
"dataPath": "params_shard_86.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.16.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "5ecd8a028f9d9f646e05b90da8c52a62"
},
{
"dataPath": "params_shard_87.bin",
"format": "raw-shard",
"nbytes": 27271168,
"records": [
{
"name": "model.layers.15.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 0
},
{
"name": "model.layers.15.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 12582912
},
{
"name": "model.layers.15.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 14155776
},
{
"name": "model.layers.15.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 22544384
},
{
"name": "model.layers.16.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 23592960
},
{
"name": "model.layers.16.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 23601152
}
],
"md5sum": "1434c1759f5a3208534ad373b5758ae7"
},
{
"dataPath": "params_shard_88.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.17.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "ee83292b7fcf2c0e537a51b01a5fcc6f"
},
{
"dataPath": "params_shard_89.bin",
"format": "raw-shard",
"nbytes": 30949376,
"records": [
{
"name": "model.layers.16.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 0
},
{
"name": "model.layers.16.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 7340032
},
{
"name": "model.layers.16.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 7348224
},
{
"name": "model.layers.16.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 19931136
},
{
"name": "model.layers.16.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 21504000
},
{
"name": "model.layers.16.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 29892608
},
{
"name": "model.layers.17.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30941184
}
],
"md5sum": "932c2c8b34451c7da23edf434ac0d23f"
},
{
"dataPath": "params_shard_90.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.17.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "fcaf0b92795fc1003896d121a5b47abf"
},
{
"dataPath": "params_shard_91.bin",
"format": "raw-shard",
"nbytes": 25174016,
"records": [
{
"name": "model.layers.17.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 0
},
{
"name": "model.layers.17.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 3670016
},
{
"name": "model.layers.17.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 11010048
},
{
"name": "model.layers.17.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 11018240
},
{
"name": "model.layers.17.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 23601152
}
],
"md5sum": "ce00d2be6fb9ea90fa2de42b65a396e6"
},
{
"dataPath": "params_shard_92.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.18.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "c03ae7765fcd3f0c3daf2ffc1493aeee"
},
{
"dataPath": "params_shard_93.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.18.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "e7819aaa965f04f4729eb2423cc95fb1"
},
{
"dataPath": "params_shard_94.bin",
"format": "raw-shard",
"nbytes": 33046528,
"records": [
{
"name": "model.layers.17.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.17.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.18.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 9437184
},
{
"name": "model.layers.18.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 9445376
},
{
"name": "model.layers.18.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 13115392
},
{
"name": "model.layers.18.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 20455424
},
{
"name": "model.layers.18.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 20463616
}
],
"md5sum": "fd2fed5a05cf515104fc3d6c65deb932"
},
{
"dataPath": "params_shard_95.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.19.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "2603f7574720394fd1916a2e97d64532"
},
{
"dataPath": "params_shard_96.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.19.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "e660fab6d7c5359d59b614d11afeefda"
},
{
"dataPath": "params_shard_97.bin",
"format": "raw-shard",
"nbytes": 22036480,
"records": [
{
"name": "model.layers.18.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 0
},
{
"name": "model.layers.18.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 1572864
},
{
"name": "model.layers.18.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 9961472
},
{
"name": "model.layers.19.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 11010048
},
{
"name": "model.layers.19.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 11018240
},
{
"name": "model.layers.19.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 14688256
},
{
"name": "model.layers.19.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 22028288
}
],
"md5sum": "da33a7d73a0d7a579c1b610f68aa9b7c"
},
{
"dataPath": "params_shard_98.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.20.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "355a5c8633cc40ec7b42f52869039a74"
},
{
"dataPath": "params_shard_99.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.20.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "632184007b1ac33bffbdeb111a35962f"
},
{
"dataPath": "params_shard_100.bin",
"format": "raw-shard",
"nbytes": 27271168,
"records": [
{
"name": "model.layers.19.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 0
},
{
"name": "model.layers.19.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 12582912
},
{
"name": "model.layers.19.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 14155776
},
{
"name": "model.layers.19.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 22544384
},
{
"name": "model.layers.20.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 23592960
},
{
"name": "model.layers.20.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 23601152
}
],
"md5sum": "1f10d818142e8969d5229a7e001017ab"
},
{
"dataPath": "params_shard_101.bin",
"format": "raw-shard",
"nbytes": 29360128,
"records": [
{
"name": "model.layers.21.mlp.down_proj.q_weight",
"shape": [
1792,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 29360128,
"byteOffset": 0
}
],
"md5sum": "b5061fd2bedbe1dd5ec0fdfe85f21992"
},
{
"dataPath": "params_shard_102.bin",
"format": "raw-shard",
"nbytes": 30949376,
"records": [
{
"name": "model.layers.20.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 0
},
{
"name": "model.layers.20.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 7340032
},
{
"name": "model.layers.20.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 7348224
},
{
"name": "model.layers.20.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 19931136
},
{
"name": "model.layers.20.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 21504000
},
{
"name": "model.layers.20.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 29892608
},
{
"name": "model.layers.21.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30941184
}
],
"md5sum": "1b40d7f6d60dca8ea6abb765310fd0e8"
},
{
"dataPath": "params_shard_103.bin",
"format": "raw-shard",
"nbytes": 58720256,
"records": [
{
"name": "model.layers.21.mlp.gate_up_proj.q_weight",
"shape": [
512,
28672
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 58720256,
"byteOffset": 0
}
],
"md5sum": "eaf1d1f6bcfc617d0f1ff14c7bcb2c54"
},
{
"dataPath": "params_shard_104.bin",
"format": "raw-shard",
"nbytes": 25174016,
"records": [
{
"name": "model.layers.21.mlp.down_proj.q_scale",
"shape": [
448,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3670016,
"byteOffset": 0
},
{
"name": "model.layers.21.mlp.gate_up_proj.q_scale",
"shape": [
128,
28672
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 7340032,
"byteOffset": 3670016
},
{
"name": "model.layers.21.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 11010048
},
{
"name": "model.layers.21.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 11018240
},
{
"name": "model.layers.21.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 23601152
}
],
"md5sum": "c9f6f917664447f22d2845b1aeaae984"
},
{
"dataPath": "params_shard_105.bin",
"format": "raw-shard",
"nbytes": 33030144,
"records": [
{
"name": "model.layers.21.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.21.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.22.self_attn.qkv_proj.q_weight",
"shape": [
512,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12582912,
"byteOffset": 9437184
},
{
"name": "model.layers.22.self_attn.qkv_proj.q_scale",
"shape": [
128,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1572864,
"byteOffset": 22020096
},
{
"name": "model.layers.22.self_attn.o_proj.q_weight",
"shape": [
512,
4096
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 23592960
},
{
"name": "model.layers.22.self_attn.o_proj.q_scale",
"shape": [
128,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 31981568
}
],
"md5sum": "9e3dd1802550a258b91a0bce3faaecb1"
}
]
}