Hugging Face's logo

mlc-ai
/
Qwen3-4B-q4f16_0-MLC

MLC-LLM
web-llm
Model card Files Community
Qwen3-4B-q4f16_0-MLC / ndarray-cache.json
riczhou's picture
riczhou
Upload folder using huggingface_hub
eb0629f verified
raw
history blame contribute delete
180 kB
 {
"metadata": {
"ParamSize": 435,
"ParamBytes": 2262920192.0,
"BitsPerParam": 4.500560626944995
},
"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": "82277c94f44a3a7153639891038ecfb0"
},
{
"dataPath": "params_shard_1.bin",
"format": "raw-shard",
"nbytes": 24314880,
"records": [
{
"name": "model.embed_tokens.q_scale",
"shape": [
151936,
80
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24309760,
"byteOffset": 0
},
{
"name": "model.layers.0.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 24309760
}
],
"md5sum": "62997375dbe823e54515758ffc385c30"
},
{
"dataPath": "params_shard_2.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": "418d3ac241b79455b50a55c2a0e7da31"
},
{
"dataPath": "params_shard_3.bin",
"format": "raw-shard",
"nbytes": 31877632,
"records": [
{
"name": "model.layers.0.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.0.mlp.down_proj.q_scale",
"shape": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.0.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.0.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.0.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.0.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.0.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.0.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.0.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.0.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.1.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "41cca8b6647090749bd278ca45ef26a6"
},
{
"dataPath": "params_shard_4.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": "5ffeef03fa740dd2dc73ee5d6ec665c5"
},
{
"dataPath": "params_shard_5.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.1.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.1.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.1.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.1.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.1.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.1.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.1.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.1.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.10.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "880c9a649b1c05c81dc9a5b829b3dab9"
},
{
"dataPath": "params_shard_6.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": "d136d092fe15e5e6f59a26ec87b53a5a"
},
{
"dataPath": "params_shard_7.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.10.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.10.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.10.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.10.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.10.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.10.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.10.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.10.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.11.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "f51dcbd9ac5d1c7eba2a87ac2a224cc7"
},
{
"dataPath": "params_shard_8.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": "eed1ce52ed3ab226cda2549dfb367bcc"
},
{
"dataPath": "params_shard_9.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.11.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.11.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.11.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.11.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.11.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.11.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.11.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.11.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.12.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "fd9d397f81d5339e1623bfb4adc16d1a"
},
{
"dataPath": "params_shard_10.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": "16c5bd2846a01b77bc3eb1b398a58999"
},
{
"dataPath": "params_shard_11.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.12.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.12.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.12.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.12.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.12.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.12.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.12.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.12.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.13.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "df0e74a8865fe1d2b9884e0d6ad410a3"
},
{
"dataPath": "params_shard_12.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": "14641923048cf475e7776702e33c4bff"
},
{
"dataPath": "params_shard_13.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.13.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.13.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.13.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.13.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.13.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.13.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.13.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.13.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.14.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "5b4dc557c86418d8ec64954af007024b"
},
{
"dataPath": "params_shard_14.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": "e43cf7054ab9d850157b686f9753ffc1"
},
{
"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": "056a78e3739b0d1ec4e82b9984f9867a"
},
{
"dataPath": "params_shard_16.bin",
"format": "raw-shard",
"nbytes": 31872512,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.14.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.14.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.14.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.14.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.14.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.14.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.14.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.14.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
}
],
"md5sum": "66e76b49e7993673143844aabccce9ad"
},
{
"dataPath": "params_shard_17.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": "fdfe6a7e4d536ecc4d90a65a42ca267e"
},
{
"dataPath": "params_shard_18.bin",
"format": "raw-shard",
"nbytes": 31872512,
"records": [
{
"name": "model.layers.15.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 0
},
{
"name": "model.layers.15.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 3112960
},
{
"name": "model.layers.15.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 3113216
},
{
"name": "model.layers.15.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 10977536
},
{
"name": "model.layers.15.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 11960576
},
{
"name": "model.layers.15.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 17203456
},
{
"name": "model.layers.15.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17858816
},
{
"name": "model.layers.2.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17859072
},
{
"name": "model.layers.2.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 17864192
},
{
"name": "model.layers.2.mlp.down_proj.q_scale",
"shape": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 30316032
}
],
"md5sum": "8524f5b87f6f39a53776d94c9b23d978"
},
{
"dataPath": "params_shard_19.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": "b394788ca79e09fb688ca9a824a80499"
},
{
"dataPath": "params_shard_20.bin",
"format": "raw-shard",
"nbytes": 31877632,
"records": [
{
"name": "model.layers.2.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 0
},
{
"name": "model.layers.2.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 3112960
},
{
"name": "model.layers.2.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 3118080
},
{
"name": "model.layers.2.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 3118336
},
{
"name": "model.layers.2.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 10982656
},
{
"name": "model.layers.2.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 11965696
},
{
"name": "model.layers.2.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 17208576
},
{
"name": "model.layers.2.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17863936
},
{
"name": "model.layers.3.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17864192
},
{
"name": "model.layers.3.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 17869312
},
{
"name": "model.layers.3.mlp.down_proj.q_scale",
"shape": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 30321152
}
],
"md5sum": "6a7b23955f8a787329d19fa1c65a5f90"
},
{
"dataPath": "params_shard_21.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": "faa8d3d8b3370b25643ff795fefd374a"
},
{
"dataPath": "params_shard_22.bin",
"format": "raw-shard",
"nbytes": 31877632,
"records": [
{
"name": "model.layers.3.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 0
},
{
"name": "model.layers.3.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 3112960
},
{
"name": "model.layers.3.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 3118080
},
{
"name": "model.layers.3.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 3118336
},
{
"name": "model.layers.3.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 10982656
},
{
"name": "model.layers.3.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 11965696
},
{
"name": "model.layers.3.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 17208576
},
{
"name": "model.layers.3.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17863936
},
{
"name": "model.layers.4.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17864192
},
{
"name": "model.layers.4.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 17869312
},
{
"name": "model.layers.4.mlp.down_proj.q_scale",
"shape": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 30321152
}
],
"md5sum": "ebe07fc146da0036c1471801a3143b95"
},
{
"dataPath": "params_shard_23.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": "f89ea4ba6c72939ccec9e247621deb8a"
},
{
"dataPath": "params_shard_24.bin",
"format": "raw-shard",
"nbytes": 31877632,
"records": [
{
"name": "model.layers.4.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 0
},
{
"name": "model.layers.4.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 3112960
},
{
"name": "model.layers.4.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 3118080
},
{
"name": "model.layers.4.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 3118336
},
{
"name": "model.layers.4.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 10982656
},
{
"name": "model.layers.4.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 11965696
},
{
"name": "model.layers.4.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 17208576
},
{
"name": "model.layers.4.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17863936
},
{
"name": "model.layers.5.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17864192
},
{
"name": "model.layers.5.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 17869312
},
{
"name": "model.layers.5.mlp.down_proj.q_scale",
"shape": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 30321152
}
],
"md5sum": "a20551ad864d61c0bb4b883e7cb2e31e"
},
{
"dataPath": "params_shard_25.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": "bfabd667016c41225a523b92127ddbba"
},
{
"dataPath": "params_shard_26.bin",
"format": "raw-shard",
"nbytes": 31877632,
"records": [
{
"name": "model.layers.5.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 0
},
{
"name": "model.layers.5.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 3112960
},
{
"name": "model.layers.5.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 3118080
},
{
"name": "model.layers.5.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 3118336
},
{
"name": "model.layers.5.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 10982656
},
{
"name": "model.layers.5.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 11965696
},
{
"name": "model.layers.5.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 17208576
},
{
"name": "model.layers.5.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17863936
},
{
"name": "model.layers.6.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17864192
},
{
"name": "model.layers.6.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 17869312
},
{
"name": "model.layers.6.mlp.down_proj.q_scale",
"shape": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 30321152
}
],
"md5sum": "418b5c2f02eafb1aafb41871feb75396"
},
{
"dataPath": "params_shard_27.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": "7fe154c920e430857e03d979757dd7b1"
},
{
"dataPath": "params_shard_28.bin",
"format": "raw-shard",
"nbytes": 31877632,
"records": [
{
"name": "model.layers.6.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 0
},
{
"name": "model.layers.6.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 3112960
},
{
"name": "model.layers.6.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 3118080
},
{
"name": "model.layers.6.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 3118336
},
{
"name": "model.layers.6.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 10982656
},
{
"name": "model.layers.6.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 11965696
},
{
"name": "model.layers.6.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 17208576
},
{
"name": "model.layers.6.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17863936
},
{
"name": "model.layers.7.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17864192
},
{
"name": "model.layers.7.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 17869312
},
{
"name": "model.layers.7.mlp.down_proj.q_scale",
"shape": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 30321152
}
],
"md5sum": "ac25cfc1b05a058f3fefb189d1be8613"
},
{
"dataPath": "params_shard_29.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": "ebba95bb572e233fba1af5d802b70900"
},
{
"dataPath": "params_shard_30.bin",
"format": "raw-shard",
"nbytes": 31877632,
"records": [
{
"name": "model.layers.7.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 0
},
{
"name": "model.layers.7.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 3112960
},
{
"name": "model.layers.7.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 3118080
},
{
"name": "model.layers.7.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 3118336
},
{
"name": "model.layers.7.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 10982656
},
{
"name": "model.layers.7.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 11965696
},
{
"name": "model.layers.7.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 17208576
},
{
"name": "model.layers.7.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17863936
},
{
"name": "model.layers.8.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17864192
},
{
"name": "model.layers.8.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 17869312
},
{
"name": "model.layers.8.mlp.down_proj.q_scale",
"shape": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 30321152
}
],
"md5sum": "f9d2a76089d3122ee1a259fdefb80083"
},
{
"dataPath": "params_shard_31.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": "3a2588bfc877ba629f4486f6be06457d"
},
{
"dataPath": "params_shard_32.bin",
"format": "raw-shard",
"nbytes": 31877632,
"records": [
{
"name": "model.layers.8.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 0
},
{
"name": "model.layers.8.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 3112960
},
{
"name": "model.layers.8.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 3118080
},
{
"name": "model.layers.8.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 3118336
},
{
"name": "model.layers.8.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 10982656
},
{
"name": "model.layers.8.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 11965696
},
{
"name": "model.layers.8.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 17208576
},
{
"name": "model.layers.8.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17863936
},
{
"name": "model.layers.9.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17864192
},
{
"name": "model.layers.9.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 17869312
},
{
"name": "model.layers.9.mlp.down_proj.q_scale",
"shape": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 30321152
}
],
"md5sum": "2ac1d1eab1e7c768a86e073a49b4c0c6"
},
{
"dataPath": "params_shard_33.bin",
"format": "raw-shard",
"nbytes": 31887872,
"records": [
{
"name": "model.layers.9.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 0
},
{
"name": "model.layers.9.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 3112960
},
{
"name": "model.layers.9.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 3118080
},
{
"name": "model.layers.9.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 3118336
},
{
"name": "model.layers.9.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 10982656
},
{
"name": "model.layers.9.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 11965696
},
{
"name": "model.layers.9.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 17208576
},
{
"name": "model.layers.9.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17863936
},
{
"name": "model.layers.15.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17864192
},
{
"name": "model.layers.15.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 17869312
},
{
"name": "model.layers.15.mlp.down_proj.q_scale",
"shape": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 30321152
},
{
"name": "model.layers.15.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31877632
},
{
"name": "model.layers.16.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31882752
}
],
"md5sum": "bc2ce18b0d482eaeb59ea01912eaaf74"
},
{
"dataPath": "params_shard_34.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": "da74d08be923228341220d3b0a941e19"
},
{
"dataPath": "params_shard_35.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.16.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.16.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.16.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.16.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.16.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.16.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.16.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.16.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.17.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "ac37caeb5ea544971f3700ed196fa206"
},
{
"dataPath": "params_shard_36.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": "188a7b29d66131b7090df9a534fc74c9"
},
{
"dataPath": "params_shard_37.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.17.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.17.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.17.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.17.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.17.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.17.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.17.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.17.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.18.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "f85d9139429903c8355ac4194aaf3540"
},
{
"dataPath": "params_shard_38.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": "8b629fab2ada8a0d34e2d66665ac8c04"
},
{
"dataPath": "params_shard_39.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.18.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.18.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.18.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.18.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.18.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.18.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.18.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.18.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.19.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "75961bdd49fbff57161b81a7f8883320"
},
{
"dataPath": "params_shard_40.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": "060093182e52ffd133ec3eafe4bc2c69"
},
{
"dataPath": "params_shard_41.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.19.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.19.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.19.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.19.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.19.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.19.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.19.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.19.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.20.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "dac032b98b2f95f10c014c0989d3fcb1"
},
{
"dataPath": "params_shard_42.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": "cd08e418aae6fa4d60ee8ccade5b6f5d"
},
{
"dataPath": "params_shard_43.bin",
"format": "raw-shard",
"nbytes": 31877632,
"records": [
{
"name": "model.layers.20.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 0
},
{
"name": "model.layers.20.mlp.down_proj.q_scale",
"shape": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.20.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.20.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.20.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.20.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.20.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.20.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.20.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.20.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.21.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "e93ac03fd1d3e0fdce8c15ff1467736c"
},
{
"dataPath": "params_shard_44.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": "8d81887c60c7204f4f718580911a5c67"
},
{
"dataPath": "params_shard_45.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.21.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.21.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.21.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.21.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.21.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.21.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.21.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.21.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.22.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "64016fbec1fbab50c4130c0e47616f5a"
},
{
"dataPath": "params_shard_46.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": "847bc145b7644775ff6f90987aed1e40"
},
{
"dataPath": "params_shard_47.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.22.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.22.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.22.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.22.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.22.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.22.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.22.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.22.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.23.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "3b2c884dba1148ecc53dbac7b22c65a5"
},
{
"dataPath": "params_shard_48.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": "d8e49d1142d7e55463bf3caa2fb245d2"
},
{
"dataPath": "params_shard_49.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.23.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.23.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.23.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.23.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.23.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.23.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.23.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.23.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.24.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "afaa88923b6e14e74c0ea92ee62af524"
},
{
"dataPath": "params_shard_50.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": "e79c10bb891c01adf141d43cf8e459e7"
},
{
"dataPath": "params_shard_51.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.24.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.24.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.24.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.24.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.24.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.24.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.24.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.24.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.25.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "3806573b9138117bcf1fe919e8bc4217"
},
{
"dataPath": "params_shard_52.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": "e10993162203267ceab1c1d6ce720866"
},
{
"dataPath": "params_shard_53.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.25.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.25.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.25.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.25.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.25.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.25.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.25.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.25.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.26.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "aafaf88b34cf3ed300d3b411e80e3c78"
},
{
"dataPath": "params_shard_54.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": "172c14871df9110f2fd91755fddea51c"
},
{
"dataPath": "params_shard_55.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.26.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.26.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.26.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.26.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.26.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.26.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.26.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.26.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.27.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "b28762ddba2402b9268593ea90b63c0e"
},
{
"dataPath": "params_shard_56.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": "07ac7ef84440d61a0de1283e4c54bf43"
},
{
"dataPath": "params_shard_57.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.27.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.27.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.27.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.27.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.27.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.27.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.27.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.27.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.28.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "0fa2e1072b332ee04aab6bd0a28a341a"
},
{
"dataPath": "params_shard_58.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": "0ff4a46620137de4fd9f4778d64f70c0"
},
{
"dataPath": "params_shard_59.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.28.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.28.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.28.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.28.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.28.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.28.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.28.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.28.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.29.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "c20cb95e1785f8d2a688da6640c6a7ef"
},
{
"dataPath": "params_shard_60.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": "a6bf2ffe5119b5ddd442b044ff2c2a40"
},
{
"dataPath": "params_shard_61.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.29.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.29.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.29.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.29.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.29.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.29.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.29.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.29.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.30.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "931699d3b7e723b6eafa2ab9c11326c1"
},
{
"dataPath": "params_shard_62.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": "2facfbebf979ef7e1153795526878eeb"
},
{
"dataPath": "params_shard_63.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.30.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.30.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.30.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.30.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.30.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.30.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.30.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.30.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.31.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "a60ea241312cdab6609e7308b2dbbf29"
},
{
"dataPath": "params_shard_64.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": "f6d5d4ca1d4ea36c5760ebe3fc0ad804"
},
{
"dataPath": "params_shard_65.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.31.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.31.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.31.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.31.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.31.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.31.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.31.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.31.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.32.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "c58b1dda2549c2800cd3458ef45f1dba"
},
{
"dataPath": "params_shard_66.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": "55d512a81712178d1f5eb730288312d1"
},
{
"dataPath": "params_shard_67.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.32.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.32.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.32.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.32.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.32.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.32.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.32.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.32.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.33.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "e280fcef343460b0c182e167513f2f8b"
},
{
"dataPath": "params_shard_68.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": "033ac4e11ae3e6f2f77a3fda342b01b1"
},
{
"dataPath": "params_shard_69.bin",
"format": "raw-shard",
"nbytes": 31877632,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.33.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.33.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.33.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.33.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.33.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.33.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.33.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.33.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
},
{
"name": "model.layers.34.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
}
],
"md5sum": "2e1ec5c4b383c57fa48f88c479399974"
},
{
"dataPath": "params_shard_70.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": "ddb9b4710cf02b018b979a66fcf07f57"
},
{
"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": "ed21d266613a257c2ab3df15206ea60f"
},
{
"dataPath": "params_shard_72.bin",
"format": "raw-shard",
"nbytes": 31872512,
"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": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 12451840
},
{
"name": "model.layers.34.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 14008320
},
{
"name": "model.layers.34.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17121280
},
{
"name": "model.layers.34.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17126400
},
{
"name": "model.layers.34.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 17126656
},
{
"name": "model.layers.34.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 24990976
},
{
"name": "model.layers.34.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 25974016
},
{
"name": "model.layers.34.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 31216896
},
{
"name": "model.layers.34.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 31872256
}
],
"md5sum": "87a077e01e2ece27a88da6e82015a535"
},
{
"dataPath": "params_shard_73.bin",
"format": "raw-shard",
"nbytes": 31882752,
"records": [
{
"name": "model.layers.35.mlp.gate_up_proj.q_scale",
"shape": [
80,
19456
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 3112960,
"byteOffset": 0
},
{
"name": "model.layers.35.self_attn.k_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 3112960
},
{
"name": "model.layers.35.self_attn.c_attn.q_weight",
"shape": [
320,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 7864320,
"byteOffset": 3113216
},
{
"name": "model.layers.35.self_attn.c_attn.q_scale",
"shape": [
80,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 983040,
"byteOffset": 10977536
},
{
"name": "model.layers.35.self_attn.o_proj.q_weight",
"shape": [
512,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 5242880,
"byteOffset": 11960576
},
{
"name": "model.layers.35.self_attn.o_proj.q_scale",
"shape": [
128,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 655360,
"byteOffset": 17203456
},
{
"name": "model.layers.35.self_attn.q_norm.weight",
"shape": [
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 256,
"byteOffset": 17858816
},
{
"name": "model.layers.35.input_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 17859072
},
{
"name": "model.layers.35.mlp.down_proj.q_weight",
"shape": [
1216,
2560
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 12451840,
"byteOffset": 17864192
},
{
"name": "model.layers.35.mlp.down_proj.q_scale",
"shape": [
304,
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1556480,
"byteOffset": 30316032
},
{
"name": "model.layers.35.post_attention_layernorm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31872512
},
{
"name": "model.norm.weight",
"shape": [
2560
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 5120,
"byteOffset": 31877632
}
],
"md5sum": "325876ad6e2f8e977d1e985270fe3d28"
}
]
}