Hugging Face's logo

numen-tech
/
Qwen3-4B-Instruct-2507-abliterated-GPTQ-Int4

Text Generation
MLC-LLM
English
4-bit precision
Model card Files
xet
Community
Qwen3-4B-Instruct-2507-abliterated-GPTQ-Int4 / ndarray-cache.json
numen-tech's picture
numen-tech
Add weights
6fd01fd
raw
history blame contribute delete
180 kB
 {
"metadata": {
"ParamSize": 435,
"ParamBytes": 2074376192.0,
"BitsPerParam": 4.125578908258419
},
"records": [
{
"dataPath": "params_shard_0.bin",
"format": "raw-shard",
"nbytes": 194478080,
"records": [
{
"name": "model.embed_tokens.q_weight",
"shape": [
151936,
320
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 194478080,
"byteOffset": 0
}
],
"md5sum": "18c43dc27ac8cdc4983bb7466bb8089e"
},
{
"dataPath": "params_shard_1.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.0.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "ca0263d8d5cb3961ee8146161d14066b"
},
{
"dataPath": "params_shard_2.bin",
"format": "raw-shard",
"nbytes": 33229312,
"records": [
{
"name": "model.embed_tokens.q_scale",
"shape": [
151936,
20
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6077440,
"byteOffset": 0
},
{
"name": "model.layers.0.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 6077440
},
{
"name": "model.layers.0.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 6082560
},
{
"name": "model.layers.0.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 18534400
},
{
"name": "model.layers.0.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 18923520
},
{
"name": "model.layers.0.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 19701760
},
{
"name": "model.layers.0.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 19706880
},
{
"name": "model.layers.0.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 19707136
},
{
"name": "model.layers.0.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 27571456
},
{
"name": "model.layers.0.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 27817216
},
{
"name": "model.layers.0.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 33060096
},
{
"name": "model.layers.0.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 33223936
},
{
"name": "model.layers.1.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 33224192
}
],
"md5sum": "d6794ca5215cfbda047b7df458341815"
},
{
"dataPath": "params_shard_3.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.1.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "0d63728adfab16735c78ae0734ec180c"
},
{
"dataPath": "params_shard_4.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.1.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.1.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.1.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.1.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.1.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.1.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.1.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.1.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.1.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.1.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.10.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "caca403e0e666536f9f6290fe5f8df74"
},
{
"dataPath": "params_shard_5.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.10.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "03eed4b4d1d7161edf2538f837ea04ab"
},
{
"dataPath": "params_shard_6.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.10.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.10.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.10.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.10.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.10.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.10.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.10.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.10.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.10.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.10.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.11.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "89912817553820c793b65b62c235007c"
},
{
"dataPath": "params_shard_7.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.11.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "2f2d718f05a5f615603769013c15e77d"
},
{
"dataPath": "params_shard_8.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.11.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.11.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.11.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.11.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.11.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.11.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.11.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.11.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.11.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.11.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.12.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "92293959de446aded83c4578b97b7dcd"
},
{
"dataPath": "params_shard_9.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.12.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "6ae44508f2fba303664477c259a84180"
},
{
"dataPath": "params_shard_10.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.12.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.12.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.12.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.12.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.12.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.12.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.12.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.12.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.12.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.12.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.13.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "bb01fbeb78d30a78712ccd3844c20bc2"
},
{
"dataPath": "params_shard_11.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.13.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "91cc6f6a5ed21cda18b1ceb1bea9161e"
},
{
"dataPath": "params_shard_12.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.13.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.13.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.13.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.13.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.13.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.13.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.13.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.13.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.13.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.13.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.14.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "08a122fdde4bf411087c2134379f0a21"
},
{
"dataPath": "params_shard_13.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.14.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "02af8c697a602a3b77986eb79e1cbc05"
},
{
"dataPath": "params_shard_14.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.14.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.14.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.14.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.14.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.14.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.14.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.14.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.14.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.14.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.14.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.15.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "a98e1f94d29ed6975b45a532c55b516b"
},
{
"dataPath": "params_shard_15.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.15.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "0fd8033b7d8dd574be5fd26bf885c792"
},
{
"dataPath": "params_shard_16.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.15.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.15.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.15.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.15.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.15.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.15.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.15.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.15.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.15.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.15.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.16.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "cbbbe239b1a0fc001db0737c936fd129"
},
{
"dataPath": "params_shard_17.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.16.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "3ae6e37cb07a673a56b75d3bdd0742ae"
},
{
"dataPath": "params_shard_18.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.16.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.16.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.16.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.16.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.16.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.16.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.16.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.16.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.16.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.16.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.17.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "e5e59c8939d4506357031201ffca75a7"
},
{
"dataPath": "params_shard_19.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.17.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "9a1723fb0c0382a428e2f90370af226e"
},
{
"dataPath": "params_shard_20.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.17.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.17.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.17.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.17.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.17.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.17.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.17.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.17.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.17.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.17.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.18.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "bd7b9dc78bb0507a86fb56da3b5354a1"
},
{
"dataPath": "params_shard_21.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.18.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "11f07cf9a779c82b62d50f1195248a18"
},
{
"dataPath": "params_shard_22.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.18.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.18.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.18.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.18.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.18.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.18.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.18.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.18.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.18.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.18.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.19.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "e321a3e1dc0220012f90df5ac77132fd"
},
{
"dataPath": "params_shard_23.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.19.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "843070bac8bdc161a3235efc942cb6a2"
},
{
"dataPath": "params_shard_24.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.19.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.19.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.19.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.19.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.19.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.19.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.19.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.19.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.19.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.19.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.2.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "208612e7466af9a275cfde92f42bcfc9"
},
{
"dataPath": "params_shard_25.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.2.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "86c6f662fc2b0bbd114165762a68105d"
},
{
"dataPath": "params_shard_26.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.20.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "cd0cc2d6b2ffd7f0a5d91d49c85cf194"
},
{
"dataPath": "params_shard_27.bin",
"format": "raw-shard",
"nbytes": 27920128,
"records": [
{
"name": "model.layers.2.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.2.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.2.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.2.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.2.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.2.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.2.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.2.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.2.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.2.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.20.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 27141632
},
{
"name": "model.layers.20.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27919872
}
],
"md5sum": "c4385522339a1b0834964627ced67884"
},
{
"dataPath": "params_shard_28.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.3.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "649e39b36d38ddd2795313d1f50dc53f"
},
{
"dataPath": "params_shard_29.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.20.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 0
},
{
"name": "model.layers.20.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 7864320
},
{
"name": "model.layers.20.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 8110080
},
{
"name": "model.layers.20.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 13352960
},
{
"name": "model.layers.20.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13516800
},
{
"name": "model.layers.3.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13517056
},
{
"name": "model.layers.3.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 13522176
},
{
"name": "model.layers.3.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 25974016
},
{
"name": "model.layers.3.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 26363136
},
{
"name": "model.layers.3.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141376
},
{
"name": "model.layers.3.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27146496
}
],
"md5sum": "81de97971339fb9f2bd83cde8ee6181d"
},
{
"dataPath": "params_shard_30.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.4.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "dd0fce5602115dba5f9bcd8adae0ae5c"
},
{
"dataPath": "params_shard_31.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.3.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 0
},
{
"name": "model.layers.3.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 7864320
},
{
"name": "model.layers.3.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 8110080
},
{
"name": "model.layers.3.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 13352960
},
{
"name": "model.layers.3.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13516800
},
{
"name": "model.layers.4.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13517056
},
{
"name": "model.layers.4.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 13522176
},
{
"name": "model.layers.4.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 25974016
},
{
"name": "model.layers.4.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 26363136
},
{
"name": "model.layers.4.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141376
},
{
"name": "model.layers.4.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27146496
}
],
"md5sum": "f179f716f305887a72dcdf194cfec1ab"
},
{
"dataPath": "params_shard_32.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.5.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "8940a15189428619bbebe89641b082bf"
},
{
"dataPath": "params_shard_33.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.4.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 0
},
{
"name": "model.layers.4.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 7864320
},
{
"name": "model.layers.4.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 8110080
},
{
"name": "model.layers.4.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 13352960
},
{
"name": "model.layers.4.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13516800
},
{
"name": "model.layers.5.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13517056
},
{
"name": "model.layers.5.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 13522176
},
{
"name": "model.layers.5.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 25974016
},
{
"name": "model.layers.5.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 26363136
},
{
"name": "model.layers.5.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141376
},
{
"name": "model.layers.5.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27146496
}
],
"md5sum": "8bb86df7367163b9ca7f2967dc182fc4"
},
{
"dataPath": "params_shard_34.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.6.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "36a95ba7edb9508067c47a046ac06184"
},
{
"dataPath": "params_shard_35.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.5.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 0
},
{
"name": "model.layers.5.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 7864320
},
{
"name": "model.layers.5.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 8110080
},
{
"name": "model.layers.5.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 13352960
},
{
"name": "model.layers.5.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13516800
},
{
"name": "model.layers.6.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13517056
},
{
"name": "model.layers.6.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 13522176
},
{
"name": "model.layers.6.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 25974016
},
{
"name": "model.layers.6.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 26363136
},
{
"name": "model.layers.6.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141376
},
{
"name": "model.layers.6.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27146496
}
],
"md5sum": "a0a421d93d427d97e72cd8de63b588a4"
},
{
"dataPath": "params_shard_36.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.7.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "d80ca15ba90ac34db5355ce2b3f05c09"
},
{
"dataPath": "params_shard_37.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.6.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 0
},
{
"name": "model.layers.6.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 7864320
},
{
"name": "model.layers.6.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 8110080
},
{
"name": "model.layers.6.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 13352960
},
{
"name": "model.layers.6.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13516800
},
{
"name": "model.layers.7.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13517056
},
{
"name": "model.layers.7.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 13522176
},
{
"name": "model.layers.7.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 25974016
},
{
"name": "model.layers.7.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 26363136
},
{
"name": "model.layers.7.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141376
},
{
"name": "model.layers.7.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27146496
}
],
"md5sum": "d0a53c17cc9118e000d350e652962bd7"
},
{
"dataPath": "params_shard_38.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.8.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "147547515be90bc43acd323680353cda"
},
{
"dataPath": "params_shard_39.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.7.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 0
},
{
"name": "model.layers.7.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 7864320
},
{
"name": "model.layers.7.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 8110080
},
{
"name": "model.layers.7.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 13352960
},
{
"name": "model.layers.7.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13516800
},
{
"name": "model.layers.8.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13517056
},
{
"name": "model.layers.8.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 13522176
},
{
"name": "model.layers.8.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 25974016
},
{
"name": "model.layers.8.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 26363136
},
{
"name": "model.layers.8.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141376
},
{
"name": "model.layers.8.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27146496
}
],
"md5sum": "5bade45765cf8a0183ce1133f23c77a1"
},
{
"dataPath": "params_shard_40.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.9.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "071c9eff1aed797e7e17eb5cc50e81c2"
},
{
"dataPath": "params_shard_41.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.8.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 0
},
{
"name": "model.layers.8.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 7864320
},
{
"name": "model.layers.8.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 8110080
},
{
"name": "model.layers.8.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 13352960
},
{
"name": "model.layers.8.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13516800
},
{
"name": "model.layers.9.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13517056
},
{
"name": "model.layers.9.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 13522176
},
{
"name": "model.layers.9.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 25974016
},
{
"name": "model.layers.9.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 26363136
},
{
"name": "model.layers.9.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141376
},
{
"name": "model.layers.9.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27146496
}
],
"md5sum": "a41cf3fe3540c115d3d85266dc98b0b9"
},
{
"dataPath": "params_shard_42.bin",
"format": "raw-shard",
"nbytes": 26373376,
"records": [
{
"name": "model.layers.9.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 0
},
{
"name": "model.layers.9.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 7864320
},
{
"name": "model.layers.9.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 8110080
},
{
"name": "model.layers.9.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 13352960
},
{
"name": "model.layers.9.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13516800
},
{
"name": "model.layers.20.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13517056
},
{
"name": "model.layers.20.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 13522176
},
{
"name": "model.layers.20.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 25974016
},
{
"name": "model.layers.20.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 26363136
},
{
"name": "model.layers.21.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 26368256
}
],
"md5sum": "da05ece757d7b3d2dafc7374d8ae5ca5"
},
{
"dataPath": "params_shard_43.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.21.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "aa4c3492ee7fd0210c7afa76106a61fe"
},
{
"dataPath": "params_shard_44.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.21.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.21.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.21.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.21.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.21.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.21.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.21.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.21.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.21.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.21.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.22.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "fd637a25ab5f6b056e303ba349159ff7"
},
{
"dataPath": "params_shard_45.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.22.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "f8a5cc3e434d5b99f451f1bd3129a494"
},
{
"dataPath": "params_shard_46.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.22.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.22.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.22.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.22.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.22.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.22.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.22.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.22.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.22.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.22.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.23.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "78a04757f1edcdcfec0ab0dc757580a6"
},
{
"dataPath": "params_shard_47.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.23.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "63d70fcf38a5ba59d6155610959c273a"
},
{
"dataPath": "params_shard_48.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.23.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.23.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.23.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.23.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.23.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.23.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.23.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.23.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.23.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.23.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.24.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "b1a0f5c1c3c340e789d76bbd12551ded"
},
{
"dataPath": "params_shard_49.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.24.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "5ad5577205691576c5567064b5a28ea0"
},
{
"dataPath": "params_shard_50.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.24.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.24.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.24.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.24.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.24.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.24.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.24.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.24.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.24.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.24.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.25.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "a53805b48a6f2f130aa9c28169dd9558"
},
{
"dataPath": "params_shard_51.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.25.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "051a3c93791c89ccf052ea24b3cde5c8"
},
{
"dataPath": "params_shard_52.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.25.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.25.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.25.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.25.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.25.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.25.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.25.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.25.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.25.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.25.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.26.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "5a74d6ff24ce8f0f6c934acf1931c43e"
},
{
"dataPath": "params_shard_53.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.26.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "996eec3301582fd26c97a550d5fa5b9b"
},
{
"dataPath": "params_shard_54.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.26.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.26.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.26.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.26.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.26.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.26.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.26.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.26.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.26.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.26.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.27.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "c0cc5038eec0b457efe546e6ece8507f"
},
{
"dataPath": "params_shard_55.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.27.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "1428155c5e309ccc9a3b31457654d0fa"
},
{
"dataPath": "params_shard_56.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.27.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.27.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.27.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.27.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.27.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.27.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.27.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.27.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.27.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.27.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.28.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "805da0930f49b6557d4a1e573c4dc2a3"
},
{
"dataPath": "params_shard_57.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.28.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "21b6231ed968f92493ec074555731aec"
},
{
"dataPath": "params_shard_58.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.28.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.28.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.28.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.28.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.28.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.28.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.28.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.28.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.28.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.28.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.29.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "aa4e7d11623e24c840a071fc8cdaacbc"
},
{
"dataPath": "params_shard_59.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.29.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "d66b1a44f337ad40738315fd234e9580"
},
{
"dataPath": "params_shard_60.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.29.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.29.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.29.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.29.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.29.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.29.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.29.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.29.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.29.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.29.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.30.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "2312d0c412037a588bb8c2ae023d47f9"
},
{
"dataPath": "params_shard_61.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.30.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "1b934e02c37dda5d1a8c7968670167d1"
},
{
"dataPath": "params_shard_62.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.30.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.30.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.30.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.30.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.30.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.30.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.30.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.30.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.30.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.30.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.31.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "92a8278371570cddd4645a35389433b2"
},
{
"dataPath": "params_shard_63.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.31.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "d6bbad0f065e21c89a5115fa89af9b5a"
},
{
"dataPath": "params_shard_64.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.31.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.31.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.31.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.31.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.31.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.31.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.31.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.31.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.31.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.31.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.32.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "9b7f04b4ce8fd79f3cc831cc94c7d972"
},
{
"dataPath": "params_shard_65.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.32.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "f0d7b70fd7364c2b392a692b86daf0c6"
},
{
"dataPath": "params_shard_66.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.32.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.32.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.32.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.32.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.32.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.32.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.32.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.32.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.32.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.32.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.33.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "6ee91ab78b87d1b4e604068761f5b2d2"
},
{
"dataPath": "params_shard_67.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.33.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "86276def7b52dd1143cddf20f5d1b985"
},
{
"dataPath": "params_shard_68.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.33.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.33.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.33.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.33.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.33.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.33.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.33.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.33.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.33.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.33.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.34.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "8bbce5959a43698a55c6f547ab8e1e69"
},
{
"dataPath": "params_shard_69.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.34.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "baeabb4f2aa7e6248d6f950922a8781d"
},
{
"dataPath": "params_shard_70.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.34.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.34.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.34.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.34.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.34.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.34.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.34.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.34.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.34.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.34.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.layers.35.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "b16b97f89161d2cc21bd93eb4f58d6e5"
},
{
"dataPath": "params_shard_71.bin",
"format": "raw-shard",
"nbytes": 24903680,
"records": [
{
"name": "model.layers.35.mlp.gate_up_proj.q_weight",
"shape": [
320,
19456
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 24903680,
"byteOffset": 0
}
],
"md5sum": "4dd1473e1ba4be0aad11c5c52e742d08"
},
{
"dataPath": "params_shard_72.bin",
"format": "raw-shard",
"nbytes": 27146752,
"records": [
{
"name": "model.layers.35.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.35.mlp.down_proj.q_scale",
"shape": [
76,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 389120,
"byteOffset": 12451840
},
{
"name": "model.layers.35.mlp.gate_up_proj.q_scale",
"shape": [
20,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 778240,
"byteOffset": 12840960
},
{
"name": "model.layers.35.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 13619200
},
{
"name": "model.layers.35.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 13624320
},
{
"name": "model.layers.35.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 13624576
},
{
"name": "model.layers.35.self_attn.c_attn.q_scale",
"shape": [
20,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 245760,
"byteOffset": 21488896
},
{
"name": "model.layers.35.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 21734656
},
{
"name": "model.layers.35.self_attn.o_proj.q_scale",
"shape": [
32,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 163840,
"byteOffset": 26977536
},
{
"name": "model.layers.35.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 27141376
},
{
"name": "model.norm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 27141632
}
],
"md5sum": "86007539ee3c5059c3001125b958930f"
}
]
}