Hugging Face's logo

dusty-nv
/
Llama-2-7b-chat-hf-q4f16_ft-MLC

Model card Files Community
Llama-2-7b-chat-hf-q4f16_ft-MLC / ndarray-cache.json
dusty-nv's picture
dusty-nv
Upload folder using huggingface_hub
9467f39 verified
raw
history blame contribute delete
146 kB
{
"metadata": {
"ParamSize": 325,
"ParamBytes": 3388710912.0,
"BitsPerParam": 4.023154527843241
},
"records": [
{
"dataPath": "params_shard_0.bin",
"format": "raw-shard",
"nbytes": 65536000,
"records": [
{
"name": "lm_head.q_weight",
"shape": [
32000,
512
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 65536000,
"byteOffset": 0
}
],
"md5sum": "8a73677d56fcbae5b449c5d44d65055b"
},
{
"dataPath": "params_shard_1.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.24.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "48beda8fcc0855e47eb87bf1ec6ea546"
},
{
"dataPath": "params_shard_2.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.24.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "60600ed6250e27b68132c299c8b02fb4"
},
{
"dataPath": "params_shard_3.bin",
"format": "raw-shard",
"nbytes": 30829568,
"records": [
{
"name": "lm_head.q_scale",
"shape": [
32000,
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192000,
"byteOffset": 0
},
{
"name": "model.layers.24.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8192000
},
{
"name": "model.layers.24.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8200192
},
{
"name": "model.layers.24.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30744576
},
{
"name": "model.layers.24.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30752768
},
{
"name": "model.layers.24.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30796800
},
{
"name": "model.layers.24.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 30804992
}
],
"md5sum": "58470f16f8ac357260ba4239961bad06"
},
{
"dataPath": "params_shard_4.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.25.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "5388a02d6096f1ecd218de3bc82fdf85"
},
{
"dataPath": "params_shard_5.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.25.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "40238cada74432a9c82f99a69a7c80fb"
},
{
"dataPath": "params_shard_6.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.24.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.24.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.25.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.25.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.25.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.25.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.25.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.25.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "661943a599c91b237584edccc2975df4"
},
{
"dataPath": "params_shard_7.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.26.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "563cefcc1bbb76ba8c98b2904df29c4d"
},
{
"dataPath": "params_shard_8.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.26.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "b9b8274cbae64600a622844ed0de68ed"
},
{
"dataPath": "params_shard_9.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.25.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.25.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.26.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.26.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.26.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.26.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.26.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.26.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "70c3ab208329b9bdb823ca3f3bef5098"
},
{
"dataPath": "params_shard_10.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.27.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "af4dd049e894205366701379f4dafedc"
},
{
"dataPath": "params_shard_11.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.27.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "236354f1934d70c890d79726ea61d4a9"
},
{
"dataPath": "params_shard_12.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.26.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.26.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.27.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.27.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.27.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.27.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.27.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.27.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "46666ba792cb188fa0d3e9d6a2c6cec0"
},
{
"dataPath": "params_shard_13.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.28.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "0c1674024d1ac127800d4e36b2b5b576"
},
{
"dataPath": "params_shard_14.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.28.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "26920281e1fc2310e6bc9a0fd671e97b"
},
{
"dataPath": "params_shard_15.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.27.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.27.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.28.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.28.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.28.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.28.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.28.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.28.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "2d2418e22cde57499f8830b4f5774664"
},
{
"dataPath": "params_shard_16.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.29.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "e6109352e6843fb33a03dc7edc8c1fba"
},
{
"dataPath": "params_shard_17.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.29.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "edeace7405cc023c8eb050604348b52e"
},
{
"dataPath": "params_shard_18.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.28.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.28.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.29.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.29.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.29.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.29.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.29.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.29.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "cf95bcb9b100b1fbe3deae4c85439ae7"
},
{
"dataPath": "params_shard_19.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.30.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "498b6134be7db08692ee2259d3e1f2ab"
},
{
"dataPath": "params_shard_20.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.30.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "ab93eb527ea14392b662de7cdd2aa43b"
},
{
"dataPath": "params_shard_21.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.29.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.29.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.30.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.30.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.30.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.30.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.30.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.30.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "16e92515c12cc3e5f951d76cbae0791a"
},
{
"dataPath": "params_shard_22.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.31.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "c7d725ca474552623954ef6ac5bad7f9"
},
{
"dataPath": "params_shard_23.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.31.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "40bbb9762560b8ddf641a834bcec8a3e"
},
{
"dataPath": "params_shard_24.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.30.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.30.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.31.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.31.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.31.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.31.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.31.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.31.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "cb6688b2e1ba66016619a65597b42e26"
},
{
"dataPath": "params_shard_25.bin",
"format": "raw-shard",
"nbytes": 65536000,
"records": [
{
"name": "model.embed_tokens.q_weight",
"shape": [
32000,
512
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 65536000,
"byteOffset": 0
}
],
"md5sum": "66ad602d220b664a12421ccdb3ccb5d6"
},
{
"dataPath": "params_shard_26.bin",
"format": "raw-shard",
"nbytes": 22544384,
"records": [
{
"name": "model.layers.0.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 0
}
],
"md5sum": "dcd48fbf17f1481b6781d015b9558f29"
},
{
"dataPath": "params_shard_27.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.0.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "3de37e48337b58d859ff5d81e388356a"
},
{
"dataPath": "params_shard_28.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.0.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "a8802288f66b8f5014ce73d28e9fe287"
},
{
"dataPath": "params_shard_29.bin",
"format": "raw-shard",
"nbytes": 22544384,
"records": [
{
"name": "model.layers.1.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 0
}
],
"md5sum": "0d4cd037ba88a1d66c4ed8f0e4543607"
},
{
"dataPath": "params_shard_30.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.1.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "1c7a960d249bcafd4949e1863226b818"
},
{
"dataPath": "params_shard_31.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.1.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "f7d9f243335edd9e96c0b1f802f22e55"
},
{
"dataPath": "params_shard_32.bin",
"format": "raw-shard",
"nbytes": 25180160,
"records": [
{
"name": "model.layers.31.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.31.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.norm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.embed_tokens.q_scale",
"shape": [
32000,
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192000,
"byteOffset": 8404992
},
{
"name": "model.layers.0.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 16596992
},
{
"name": "model.layers.0.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 16605184
},
{
"name": "model.layers.0.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 16613376
},
{
"name": "model.layers.0.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 16657408
},
{
"name": "model.layers.0.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 16665600
},
{
"name": "model.layers.0.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 16690176
},
{
"name": "model.layers.0.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 25078784
},
{
"name": "model.layers.1.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 25086976
},
{
"name": "model.layers.1.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 25095168
},
{
"name": "model.layers.1.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 25103360
},
{
"name": "model.layers.1.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 25147392
},
{
"name": "model.layers.1.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 25155584
}
],
"md5sum": "726c9236794da499fcfe8c680aedaea4"
},
{
"dataPath": "params_shard_33.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.10.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "896cf7fd500739d23abd2f4b10fb05f0"
},
{
"dataPath": "params_shard_34.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.10.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "cfa2c83ea2f99883c9417594870f7294"
},
{
"dataPath": "params_shard_35.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.1.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.1.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.10.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.10.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.10.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.10.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.10.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.10.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "299bbaae66b907f136a2042ca8ff457e"
},
{
"dataPath": "params_shard_36.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.11.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "013f4e100284a4fb0aeca6f5705b5264"
},
{
"dataPath": "params_shard_37.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.11.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "2e48a672823e87381f1d21f97a95ca41"
},
{
"dataPath": "params_shard_38.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.10.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.10.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.11.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.11.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.11.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.11.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.11.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.11.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "304789574ca396b6445ea11705ab6ccb"
},
{
"dataPath": "params_shard_39.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.12.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "523ea66a7803d5f9e086d1ed15c66b2c"
},
{
"dataPath": "params_shard_40.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.12.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "7890805eb2391e209a180a073bbb3d33"
},
{
"dataPath": "params_shard_41.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.11.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.11.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.12.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.12.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.12.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.12.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.12.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.12.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "a0d4a1c46f58d349be00df1f935f25e1"
},
{
"dataPath": "params_shard_42.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.13.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "cca6f6db01f0c241f91eca75d78cf697"
},
{
"dataPath": "params_shard_43.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.13.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "16ea552883ac533264af52b24d67c3bc"
},
{
"dataPath": "params_shard_44.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.12.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.12.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.13.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.13.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.13.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.13.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.13.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.13.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "dc323a79de5e77e1c71dcf0f45ab012d"
},
{
"dataPath": "params_shard_45.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.14.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "4885d0033c52054a84bae00484b699f7"
},
{
"dataPath": "params_shard_46.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.14.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "86ad62c0b43828d66c49128cd15662d3"
},
{
"dataPath": "params_shard_47.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.13.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.13.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.14.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.14.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.14.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.14.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.14.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.14.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "774f91b5a387fd38207a2e37c0bc3278"
},
{
"dataPath": "params_shard_48.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.15.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "420d8ffb3abb00a2890f7c33cf8e988f"
},
{
"dataPath": "params_shard_49.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.15.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "cf67bfcb3198bfc04f65be659799b58e"
},
{
"dataPath": "params_shard_50.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.14.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.14.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.15.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.15.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.15.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.15.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.15.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.15.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "201c9b044867ebcb3710629fe6252413"
},
{
"dataPath": "params_shard_51.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.16.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "afa3a50f6b810dfb65f674d4149438bd"
},
{
"dataPath": "params_shard_52.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.16.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "410f2059701d65f4d00c2ae0199a4dd4"
},
{
"dataPath": "params_shard_53.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.15.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.15.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.16.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.16.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.16.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.16.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.16.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.16.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "4e7ae262aff83444ad1aa625cd873fb4"
},
{
"dataPath": "params_shard_54.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.17.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "728460c4729b25cd8785948ff132b7bf"
},
{
"dataPath": "params_shard_55.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.17.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "ce9f8a9d51f859b562a7f2b0d9cdbbf7"
},
{
"dataPath": "params_shard_56.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.16.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.16.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.17.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.17.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.17.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.17.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.17.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.17.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "208f836780fb1eb7bfb50ec000dcca66"
},
{
"dataPath": "params_shard_57.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.18.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "52ae29176d540516d51ba223ad64bd5b"
},
{
"dataPath": "params_shard_58.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.18.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "bfa2c4ebb16f57618e6ec269112a5210"
},
{
"dataPath": "params_shard_59.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.17.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.17.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.18.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.18.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.18.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.18.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.18.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.18.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "e504e072b193858aeaa085d1b2a65bcc"
},
{
"dataPath": "params_shard_60.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.19.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "7d9faefd8234d6ebe5c23d92359a4536"
},
{
"dataPath": "params_shard_61.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.19.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "1e9f26f7426810aa42f78bddc96bece2"
},
{
"dataPath": "params_shard_62.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.18.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.18.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.19.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.19.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.19.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.19.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.19.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.19.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "1be06ec37adc31a9a142436ed2332c60"
},
{
"dataPath": "params_shard_63.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.2.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "5530279443589eaf3869a29c62d48624"
},
{
"dataPath": "params_shard_64.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.2.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "3ffd412f8e860102b356c787829806bd"
},
{
"dataPath": "params_shard_65.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.19.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.19.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.2.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.2.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.2.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.2.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.2.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.2.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "822a5105fb181e040393fe66de19ea7b"
},
{
"dataPath": "params_shard_66.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.20.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "11bde350cf6b21380d294bbdebff67e8"
},
{
"dataPath": "params_shard_67.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.20.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "e045919e2c9cc6bd0881fa96c6e35962"
},
{
"dataPath": "params_shard_68.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.2.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.2.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.20.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.20.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.20.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.20.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.20.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.20.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "be82be90f4d15bdd17f0238f169c3e59"
},
{
"dataPath": "params_shard_69.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.21.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "3ea33c30cb8a06ac980a448bad5ac5a8"
},
{
"dataPath": "params_shard_70.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.21.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "27e4ed14b65acb5a39de23a8e1dd21f3"
},
{
"dataPath": "params_shard_71.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.20.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.20.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.21.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.21.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.21.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.21.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.21.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.21.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "faefef63042329682f8d6cac71c519d5"
},
{
"dataPath": "params_shard_72.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.22.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "e72217c45bd1f1300ea2e0c903077087"
},
{
"dataPath": "params_shard_73.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.22.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "0f31bea4f1622b74673c7495a70fa3f5"
},
{
"dataPath": "params_shard_74.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.21.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.21.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.22.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.22.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.22.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.22.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.22.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.22.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "48b3feffee045e2dac41d74541e2e75b"
},
{
"dataPath": "params_shard_75.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.23.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "feee6ebbf68d4e82a19e58a0574c56db"
},
{
"dataPath": "params_shard_76.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.23.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "d2a42d1e402c6d9d1b2bdfd47d1530f8"
},
{
"dataPath": "params_shard_77.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.22.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.22.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.23.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.23.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.23.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.23.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.23.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.23.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "92c06c140535ef75db86f4f6b296cba7"
},
{
"dataPath": "params_shard_78.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.3.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "73aa286d4954a5afbb5d65078f95814d"
},
{
"dataPath": "params_shard_79.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.3.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "f5d0ba91a72da81e1a205b7789297852"
},
{
"dataPath": "params_shard_80.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.23.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.23.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.3.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.3.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.3.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.3.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.3.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.3.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "0c2f705a3547aa8bc7c973b60ea02a33"
},
{
"dataPath": "params_shard_81.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.4.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "ee3e9e8a6f1e179da2dfadb2811d3984"
},
{
"dataPath": "params_shard_82.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.4.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "64ff70b4dbb66e754385e76eb23ca2d2"
},
{
"dataPath": "params_shard_83.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.3.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.3.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.4.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.4.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.4.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.4.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.4.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.4.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "f5b1851eb6fac7108df5b5faa55219ac"
},
{
"dataPath": "params_shard_84.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.5.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "42baa3c8770dd148b38c908e31aa5460"
},
{
"dataPath": "params_shard_85.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.5.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "21099941cc25eb0c7f5a1340539aedde"
},
{
"dataPath": "params_shard_86.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.4.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.4.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.5.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.5.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.5.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.5.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.5.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.5.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "ddd4ae658562dd2894104077984a610b"
},
{
"dataPath": "params_shard_87.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.6.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "4343ff1c6a61bdeee2ea6e97491b8872"
},
{
"dataPath": "params_shard_88.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.6.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "f35142bcb000b6e4b3824e9be361a8f7"
},
{
"dataPath": "params_shard_89.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.5.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.5.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.6.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.6.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.6.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.6.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.6.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.6.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "9a391910f72cc05b8810a5df90d54aa0"
},
{
"dataPath": "params_shard_90.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.7.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "1ed1e04f3974675f43b6d54b0ad395f4"
},
{
"dataPath": "params_shard_91.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.7.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "f30f641e0b2c0b9168e104ef8be45762"
},
{
"dataPath": "params_shard_92.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.6.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.6.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.7.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.7.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.7.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.7.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.7.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.7.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "4892de8f49d048fd331163a8e0b23dac"
},
{
"dataPath": "params_shard_93.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.8.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "f5bcf7f869396fdf9fef32c899deccfd"
},
{
"dataPath": "params_shard_94.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.8.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "94929792ec1f36bb0c745595b1211ef9"
},
{
"dataPath": "params_shard_95.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.7.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.7.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.8.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.8.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.8.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.8.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.8.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.8.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "eb7319cb7a45b9c61561f83f67b580b9"
},
{
"dataPath": "params_shard_96.bin",
"format": "raw-shard",
"nbytes": 45088768,
"records": [
{
"name": "model.layers.9.mlp.gate_up_proj.q_weight",
"shape": [
4096,
11008
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 45088768,
"byteOffset": 0
}
],
"md5sum": "224c3577dd0592aa5297b7a7a95822c2"
},
{
"dataPath": "params_shard_97.bin",
"format": "raw-shard",
"nbytes": 25165824,
"records": [
{
"name": "model.layers.9.self_attn.qkv_proj.q_weight",
"shape": [
4096,
6144
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 25165824,
"byteOffset": 0
}
],
"md5sum": "437e4151d0912833420bb26f568528d8"
},
{
"dataPath": "params_shard_98.bin",
"format": "raw-shard",
"nbytes": 31034368,
"records": [
{
"name": "model.layers.8.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.8.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
},
{
"name": "model.layers.9.input_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8396800
},
{
"name": "model.layers.9.mlp.down_proj.q_weight",
"shape": [
11008,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 22544384,
"byteOffset": 8404992
},
{
"name": "model.layers.9.mlp.down_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 30949376
},
{
"name": "model.layers.9.mlp.gate_up_proj.q_scale",
"shape": [
1,
22016
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 44032,
"byteOffset": 30957568
},
{
"name": "model.layers.9.post_attention_layernorm.weight",
"shape": [
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 31001600
},
{
"name": "model.layers.9.self_attn.qkv_proj.q_scale",
"shape": [
1,
12288
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 24576,
"byteOffset": 31009792
}
],
"md5sum": "6720f59bbd4205aa5992d8fe8b84fb68"
},
{
"dataPath": "params_shard_99.bin",
"format": "raw-shard",
"nbytes": 8396800,
"records": [
{
"name": "model.layers.9.self_attn.o_proj.q_weight",
"shape": [
4096,
2048
],
"dtype": "int8",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.9.self_attn.o_proj.q_scale",
"shape": [
1,
4096
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 8388608
}
],
"md5sum": "39ee4c16115e2b0e4874c6ebfe10cb95"
}
]
}