MiniMixtral_MoE / ndarray-cache.json
Kieran2828's picture
Upload 4 files
97954d6 verified
{
"metadata": {
"ParamSize": 137,
"ParamBytes": 394450304.0,
"BitsPerParam": 4.501226222315055
},
"records": [
{
"dataPath": "params_shard_0.bin",
"format": "raw-shard",
"nbytes": 18435776,
"records": [
{
"name": "model.embed_tokens.q_weight",
"shape": [
32003,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16385536,
"byteOffset": 0
},
{
"name": "model.embed_tokens.q_scale",
"shape": [
32003,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048192,
"byteOffset": 16385536
},
{
"name": "model.norm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 18433728
}
],
"md5sum": "39262692eadce1dc2cc71f107dde7dba"
},
{
"dataPath": "params_shard_1.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.0.moe.e1_e3.q_weight",
"shape": [
4,
8192,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "4b4790e347b682a2d1625d4ca8cf2fdc"
},
{
"dataPath": "params_shard_2.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.1.moe.e1_e3.q_weight",
"shape": [
4,
8192,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "96cd48c89abfb63f5e47ac29c2dea29f"
},
{
"dataPath": "params_shard_3.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.2.moe.e1_e3.q_weight",
"shape": [
4,
8192,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "2ebe7fb32b921287504c204504a3a3bf"
},
{
"dataPath": "params_shard_4.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.3.moe.e1_e3.q_weight",
"shape": [
4,
8192,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "040d60d3b1c140992c37ae6f47157639"
},
{
"dataPath": "params_shard_5.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.4.moe.e1_e3.q_weight",
"shape": [
4,
8192,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "78d5dc519482da4b6fa5a6fdbf3075fc"
},
{
"dataPath": "params_shard_6.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.5.moe.e1_e3.q_weight",
"shape": [
4,
8192,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "ddf0d5b05b33b228c73a67b12bdf5181"
},
{
"dataPath": "params_shard_7.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.6.moe.e1_e3.q_weight",
"shape": [
4,
8192,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "6393f0828fe8a7be924c6a58b122b655"
},
{
"dataPath": "params_shard_8.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.7.moe.e1_e3.q_weight",
"shape": [
4,
8192,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "747277f3a82adb4957c54b7fbcc2265e"
},
{
"dataPath": "params_shard_9.bin",
"format": "raw-shard",
"nbytes": 33138368,
"records": [
{
"name": "lm_head.q_weight",
"shape": [
32003,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16385536,
"byteOffset": 0
},
{
"name": "lm_head.q_scale",
"shape": [
32003,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048192,
"byteOffset": 16385536
},
{
"name": "model.layers.0.input_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 18433728
},
{
"name": "model.layers.1.input_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 18435776
},
{
"name": "model.layers.2.input_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 18437824
},
{
"name": "model.layers.3.input_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 18439872
},
{
"name": "model.layers.4.input_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 18441920
},
{
"name": "model.layers.5.input_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 18443968
},
{
"name": "model.layers.6.input_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 18446016
},
{
"name": "model.layers.7.input_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 18448064
},
{
"name": "model.layers.8.input_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 18450112
},
{
"name": "model.layers.9.input_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 18452160
},
{
"name": "model.layers.10.input_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 18454208
},
{
"name": "model.layers.11.input_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 18456256
},
{
"name": "model.layers.0.moe.e1_e3.q_scale",
"shape": [
4,
8192,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 18458304
},
{
"name": "model.layers.1.moe.e1_e3.q_scale",
"shape": [
4,
8192,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 20555456
},
{
"name": "model.layers.2.moe.e1_e3.q_scale",
"shape": [
4,
8192,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 22652608
},
{
"name": "model.layers.3.moe.e1_e3.q_scale",
"shape": [
4,
8192,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 24749760
},
{
"name": "model.layers.4.moe.e1_e3.q_scale",
"shape": [
4,
8192,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 26846912
},
{
"name": "model.layers.5.moe.e1_e3.q_scale",
"shape": [
4,
8192,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 28944064
},
{
"name": "model.layers.6.moe.e1_e3.q_scale",
"shape": [
4,
8192,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 31041216
}
],
"md5sum": "15fc560eff0beef7363843096cd3f83f"
},
{
"dataPath": "params_shard_10.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.9.moe.e1_e3.q_weight",
"shape": [
4,
8192,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "225eec8ab6c3a421d5f563f229fcf42f"
},
{
"dataPath": "params_shard_11.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.10.moe.e1_e3.q_weight",
"shape": [
4,
8192,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "0b7d9bfd90e4239969f587c972110172"
},
{
"dataPath": "params_shard_12.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.11.moe.e1_e3.q_weight",
"shape": [
4,
8192,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "49d56d62a2c5cbeb10306793e6aa6f70"
},
{
"dataPath": "params_shard_13.bin",
"format": "raw-shard",
"nbytes": 27262976,
"records": [
{
"name": "model.layers.7.moe.e1_e3.q_scale",
"shape": [
4,
8192,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 0
},
{
"name": "model.layers.8.moe.e1_e3.q_weight",
"shape": [
4,
8192,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 2097152
},
{
"name": "model.layers.8.moe.e1_e3.q_scale",
"shape": [
4,
8192,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 18874368
},
{
"name": "model.layers.9.moe.e1_e3.q_scale",
"shape": [
4,
8192,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 20971520
},
{
"name": "model.layers.10.moe.e1_e3.q_scale",
"shape": [
4,
8192,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 23068672
},
{
"name": "model.layers.11.moe.e1_e3.q_scale",
"shape": [
4,
8192,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 25165824
}
],
"md5sum": "655bd70482a3f3d9e74580733bbc15d1"
},
{
"dataPath": "params_shard_14.bin",
"format": "raw-shard",
"nbytes": 28311552,
"records": [
{
"name": "model.layers.0.moe.e2.q_weight",
"shape": [
4,
1024,
512
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.0.moe.e2.q_scale",
"shape": [
4,
1024,
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.1.moe.e2.q_weight",
"shape": [
4,
1024,
512
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 9437184
},
{
"name": "model.layers.1.moe.e2.q_scale",
"shape": [
4,
1024,
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 17825792
},
{
"name": "model.layers.2.moe.e2.q_weight",
"shape": [
4,
1024,
512
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 18874368
},
{
"name": "model.layers.2.moe.e2.q_scale",
"shape": [
4,
1024,
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 27262976
}
],
"md5sum": "b8058afbdee107c4dd664bf5677bb64b"
},
{
"dataPath": "params_shard_15.bin",
"format": "raw-shard",
"nbytes": 28311552,
"records": [
{
"name": "model.layers.3.moe.e2.q_weight",
"shape": [
4,
1024,
512
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.3.moe.e2.q_scale",
"shape": [
4,
1024,
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.4.moe.e2.q_weight",
"shape": [
4,
1024,
512
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 9437184
},
{
"name": "model.layers.4.moe.e2.q_scale",
"shape": [
4,
1024,
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 17825792
},
{
"name": "model.layers.5.moe.e2.q_weight",
"shape": [
4,
1024,
512
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 18874368
},
{
"name": "model.layers.5.moe.e2.q_scale",
"shape": [
4,
1024,
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 27262976
}
],
"md5sum": "ea81f78fb218262780ff13b5c95662dc"
},
{
"dataPath": "params_shard_16.bin",
"format": "raw-shard",
"nbytes": 28311552,
"records": [
{
"name": "model.layers.6.moe.e2.q_weight",
"shape": [
4,
1024,
512
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.6.moe.e2.q_scale",
"shape": [
4,
1024,
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.7.moe.e2.q_weight",
"shape": [
4,
1024,
512
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 9437184
},
{
"name": "model.layers.7.moe.e2.q_scale",
"shape": [
4,
1024,
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 17825792
},
{
"name": "model.layers.8.moe.e2.q_weight",
"shape": [
4,
1024,
512
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 18874368
},
{
"name": "model.layers.8.moe.e2.q_scale",
"shape": [
4,
1024,
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 27262976
}
],
"md5sum": "2ba93f293ac038d9c735a1bcdea65dc3"
},
{
"dataPath": "params_shard_17.bin",
"format": "raw-shard",
"nbytes": 33546240,
"records": [
{
"name": "model.layers.9.moe.e2.q_weight",
"shape": [
4,
1024,
512
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.9.moe.e2.q_scale",
"shape": [
4,
1024,
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.10.moe.e2.q_weight",
"shape": [
4,
1024,
512
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 9437184
},
{
"name": "model.layers.10.moe.e2.q_scale",
"shape": [
4,
1024,
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 17825792
},
{
"name": "model.layers.11.moe.e2.q_weight",
"shape": [
4,
1024,
512
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 18874368
},
{
"name": "model.layers.11.moe.e2.q_scale",
"shape": [
4,
1024,
128
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 27262976
},
{
"name": "model.layers.0.post_attention_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 28311552
},
{
"name": "model.layers.1.post_attention_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 28313600
},
{
"name": "model.layers.2.post_attention_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 28315648
},
{
"name": "model.layers.3.post_attention_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 28317696
},
{
"name": "model.layers.4.post_attention_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 28319744
},
{
"name": "model.layers.5.post_attention_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 28321792
},
{
"name": "model.layers.6.post_attention_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 28323840
},
{
"name": "model.layers.7.post_attention_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 28325888
},
{
"name": "model.layers.8.post_attention_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 28327936
},
{
"name": "model.layers.9.post_attention_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 28329984
},
{
"name": "model.layers.10.post_attention_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 28332032
},
{
"name": "model.layers.11.post_attention_layernorm.weight",
"shape": [
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2048,
"byteOffset": 28334080
},
{
"name": "model.layers.0.self_attn.qkv_proj.q_weight",
"shape": [
1536,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 28336128
},
{
"name": "model.layers.0.self_attn.qkv_proj.q_scale",
"shape": [
1536,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 98304,
"byteOffset": 29122560
},
{
"name": "model.layers.1.self_attn.qkv_proj.q_weight",
"shape": [
1536,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 29220864
},
{
"name": "model.layers.1.self_attn.qkv_proj.q_scale",
"shape": [
1536,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 98304,
"byteOffset": 30007296
},
{
"name": "model.layers.2.self_attn.qkv_proj.q_weight",
"shape": [
1536,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 30105600
},
{
"name": "model.layers.2.self_attn.qkv_proj.q_scale",
"shape": [
1536,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 98304,
"byteOffset": 30892032
},
{
"name": "model.layers.3.self_attn.qkv_proj.q_weight",
"shape": [
1536,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 30990336
},
{
"name": "model.layers.3.self_attn.qkv_proj.q_scale",
"shape": [
1536,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 98304,
"byteOffset": 31776768
},
{
"name": "model.layers.4.self_attn.qkv_proj.q_weight",
"shape": [
1536,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 31875072
},
{
"name": "model.layers.4.self_attn.qkv_proj.q_scale",
"shape": [
1536,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 98304,
"byteOffset": 32661504
},
{
"name": "model.layers.5.self_attn.qkv_proj.q_weight",
"shape": [
1536,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 32759808
}
],
"md5sum": "6cea3c1ce36c0eeebc1e0cd0c211e553"
},
{
"dataPath": "params_shard_18.bin",
"format": "raw-shard",
"nbytes": 12582912,
"records": [
{
"name": "model.layers.5.self_attn.qkv_proj.q_scale",
"shape": [
1536,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 98304,
"byteOffset": 0
},
{
"name": "model.layers.6.self_attn.qkv_proj.q_weight",
"shape": [
1536,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 98304
},
{
"name": "model.layers.6.self_attn.qkv_proj.q_scale",
"shape": [
1536,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 98304,
"byteOffset": 884736
},
{
"name": "model.layers.7.self_attn.qkv_proj.q_weight",
"shape": [
1536,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 983040
},
{
"name": "model.layers.7.self_attn.qkv_proj.q_scale",
"shape": [
1536,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 98304,
"byteOffset": 1769472
},
{
"name": "model.layers.8.self_attn.qkv_proj.q_weight",
"shape": [
1536,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 1867776
},
{
"name": "model.layers.8.self_attn.qkv_proj.q_scale",
"shape": [
1536,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 98304,
"byteOffset": 2654208
},
{
"name": "model.layers.9.self_attn.qkv_proj.q_weight",
"shape": [
1536,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 2752512
},
{
"name": "model.layers.9.self_attn.qkv_proj.q_scale",
"shape": [
1536,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 98304,
"byteOffset": 3538944
},
{
"name": "model.layers.10.self_attn.qkv_proj.q_weight",
"shape": [
1536,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 3637248
},
{
"name": "model.layers.10.self_attn.qkv_proj.q_scale",
"shape": [
1536,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 98304,
"byteOffset": 4423680
},
{
"name": "model.layers.11.self_attn.qkv_proj.q_weight",
"shape": [
1536,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 4521984
},
{
"name": "model.layers.11.self_attn.qkv_proj.q_scale",
"shape": [
1536,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 98304,
"byteOffset": 5308416
},
{
"name": "model.layers.0.self_attn.o_proj.q_weight",
"shape": [
1024,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 524288,
"byteOffset": 5406720
},
{
"name": "model.layers.0.self_attn.o_proj.q_scale",
"shape": [
1024,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 65536,
"byteOffset": 5931008
},
{
"name": "model.layers.1.self_attn.o_proj.q_weight",
"shape": [
1024,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 524288,
"byteOffset": 5996544
},
{
"name": "model.layers.1.self_attn.o_proj.q_scale",
"shape": [
1024,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 65536,
"byteOffset": 6520832
},
{
"name": "model.layers.2.self_attn.o_proj.q_weight",
"shape": [
1024,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 524288,
"byteOffset": 6586368
},
{
"name": "model.layers.2.self_attn.o_proj.q_scale",
"shape": [
1024,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 65536,
"byteOffset": 7110656
},
{
"name": "model.layers.3.self_attn.o_proj.q_weight",
"shape": [
1024,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 524288,
"byteOffset": 7176192
},
{
"name": "model.layers.3.self_attn.o_proj.q_scale",
"shape": [
1024,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 65536,
"byteOffset": 7700480
},
{
"name": "model.layers.4.self_attn.o_proj.q_weight",
"shape": [
1024,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 524288,
"byteOffset": 7766016
},
{
"name": "model.layers.4.self_attn.o_proj.q_scale",
"shape": [
1024,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 65536,
"byteOffset": 8290304
},
{
"name": "model.layers.5.self_attn.o_proj.q_weight",
"shape": [
1024,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 524288,
"byteOffset": 8355840
},
{
"name": "model.layers.5.self_attn.o_proj.q_scale",
"shape": [
1024,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 65536,
"byteOffset": 8880128
},
{
"name": "model.layers.6.self_attn.o_proj.q_weight",
"shape": [
1024,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 524288,
"byteOffset": 8945664
},
{
"name": "model.layers.6.self_attn.o_proj.q_scale",
"shape": [
1024,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 65536,
"byteOffset": 9469952
},
{
"name": "model.layers.7.self_attn.o_proj.q_weight",
"shape": [
1024,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 524288,
"byteOffset": 9535488
},
{
"name": "model.layers.7.self_attn.o_proj.q_scale",
"shape": [
1024,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 65536,
"byteOffset": 10059776
},
{
"name": "model.layers.8.self_attn.o_proj.q_weight",
"shape": [
1024,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 524288,
"byteOffset": 10125312
},
{
"name": "model.layers.8.self_attn.o_proj.q_scale",
"shape": [
1024,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 65536,
"byteOffset": 10649600
},
{
"name": "model.layers.9.self_attn.o_proj.q_weight",
"shape": [
1024,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 524288,
"byteOffset": 10715136
},
{
"name": "model.layers.9.self_attn.o_proj.q_scale",
"shape": [
1024,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 65536,
"byteOffset": 11239424
},
{
"name": "model.layers.10.self_attn.o_proj.q_weight",
"shape": [
1024,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 524288,
"byteOffset": 11304960
},
{
"name": "model.layers.10.self_attn.o_proj.q_scale",
"shape": [
1024,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 65536,
"byteOffset": 11829248
},
{
"name": "model.layers.11.self_attn.o_proj.q_weight",
"shape": [
1024,
128
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 524288,
"byteOffset": 11894784
},
{
"name": "model.layers.11.self_attn.o_proj.q_scale",
"shape": [
1024,
32
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 65536,
"byteOffset": 12419072
},
{
"name": "model.layers.0.moe.gate.weight",
"shape": [
4,
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 12484608
},
{
"name": "model.layers.1.moe.gate.weight",
"shape": [
4,
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 12492800
},
{
"name": "model.layers.2.moe.gate.weight",
"shape": [
4,
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 12500992
},
{
"name": "model.layers.3.moe.gate.weight",
"shape": [
4,
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 12509184
},
{
"name": "model.layers.4.moe.gate.weight",
"shape": [
4,
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 12517376
},
{
"name": "model.layers.5.moe.gate.weight",
"shape": [
4,
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 12525568
},
{
"name": "model.layers.6.moe.gate.weight",
"shape": [
4,
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 12533760
},
{
"name": "model.layers.7.moe.gate.weight",
"shape": [
4,
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 12541952
},
{
"name": "model.layers.8.moe.gate.weight",
"shape": [
4,
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 12550144
},
{
"name": "model.layers.9.moe.gate.weight",
"shape": [
4,
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 12558336
},
{
"name": "model.layers.10.moe.gate.weight",
"shape": [
4,
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 12566528
},
{
"name": "model.layers.11.moe.gate.weight",
"shape": [
4,
1024
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 8192,
"byteOffset": 12574720
}
],
"md5sum": "c81b7f0ed2e3e42b10a9de6ffc778b2e"
}
]
}