scigemma_fine_tuned_quantized_MLC / ndarray-cache.json
ngoc's picture
Fine-tuned quantized Gemma 2b-it model.
3a67b3b verified
{
"metadata": {
"ParamSize": 183,
"ParamBytes": 1409830912.0,
"BitsPerParam": 4.500347711112945
},
"records": [
{
"dataPath": "params_shard_0.bin",
"format": "raw-shard",
"nbytes": 262144000,
"records": [
{
"name": "model.embed_tokens.q_weight",
"shape": [
256000,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 262144000,
"byteOffset": 0
}
],
"md5sum": "4500be88779baf43abfc8597bf8cf1fe"
},
{
"dataPath": "params_shard_1.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.0.mlp.down_proj.q_weight",
"shape": [
2048,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "a41c9e07b0194e45a28cbce3226efc36"
},
{
"dataPath": "params_shard_2.bin",
"format": "raw-shard",
"nbytes": 32772096,
"records": [
{
"name": "model.embed_tokens.q_scale",
"shape": [
256000,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 32768000,
"byteOffset": 0
},
{
"name": "model.layers.0.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 32768000
}
],
"md5sum": "f379b208c95b8217c994896c7b6f2498"
},
{
"dataPath": "params_shard_3.bin",
"format": "raw-shard",
"nbytes": 33554432,
"records": [
{
"name": "model.layers.0.mlp.gate_up_proj.q_weight",
"shape": [
32768,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 33554432,
"byteOffset": 0
}
],
"md5sum": "6d43e7a1d3f06e7ba0705abab5dfa54e"
},
{
"dataPath": "params_shard_4.bin",
"format": "raw-shard",
"nbytes": 33554432,
"records": [
{
"name": "model.layers.1.mlp.gate_up_proj.q_weight",
"shape": [
32768,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 33554432,
"byteOffset": 0
}
],
"md5sum": "c770c74a48c10970f5dcc96299b8c616"
},
{
"dataPath": "params_shard_5.bin",
"format": "raw-shard",
"nbytes": 30482432,
"records": [
{
"name": "model.layers.0.mlp.down_proj.q_scale",
"shape": [
2048,
512
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 0
},
{
"name": "model.layers.0.mlp.gate_up_proj.q_scale",
"shape": [
32768,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4194304,
"byteOffset": 2097152
},
{
"name": "model.layers.0.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 6291456
},
{
"name": "model.layers.0.self_attn.qkv_proj.q_weight",
"shape": [
2560,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2621440,
"byteOffset": 6295552
},
{
"name": "model.layers.0.self_attn.qkv_proj.q_scale",
"shape": [
2560,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 327680,
"byteOffset": 8916992
},
{
"name": "model.layers.0.self_attn.o_proj.q_weight",
"shape": [
2048,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 9244672
},
{
"name": "model.layers.0.self_attn.o_proj.q_scale",
"shape": [
2048,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 11341824
},
{
"name": "model.layers.1.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 11603968
},
{
"name": "model.layers.1.mlp.down_proj.q_weight",
"shape": [
2048,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 11608064
},
{
"name": "model.layers.1.mlp.down_proj.q_scale",
"shape": [
2048,
512
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 28385280
}
],
"md5sum": "f022830cf526bd89069bb44cfb1fb72b"
},
{
"dataPath": "params_shard_6.bin",
"format": "raw-shard",
"nbytes": 33554432,
"records": [
{
"name": "model.layers.10.mlp.gate_up_proj.q_weight",
"shape": [
32768,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 33554432,
"byteOffset": 0
}
],
"md5sum": "248b1ae95b09a03ac97a4db3627da841"
},
{
"dataPath": "params_shard_7.bin",
"format": "raw-shard",
"nbytes": 32583680,
"records": [
{
"name": "model.layers.1.mlp.gate_up_proj.q_scale",
"shape": [
32768,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4194304,
"byteOffset": 0
},
{
"name": "model.layers.1.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 4194304
},
{
"name": "model.layers.1.self_attn.qkv_proj.q_weight",
"shape": [
2560,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2621440,
"byteOffset": 4198400
},
{
"name": "model.layers.1.self_attn.qkv_proj.q_scale",
"shape": [
2560,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 327680,
"byteOffset": 6819840
},
{
"name": "model.layers.1.self_attn.o_proj.q_weight",
"shape": [
2048,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 7147520
},
{
"name": "model.layers.1.self_attn.o_proj.q_scale",
"shape": [
2048,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 9244672
},
{
"name": "model.layers.10.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 9506816
},
{
"name": "model.layers.10.mlp.down_proj.q_weight",
"shape": [
2048,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 9510912
},
{
"name": "model.layers.10.mlp.down_proj.q_scale",
"shape": [
2048,
512
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 26288128
},
{
"name": "model.layers.10.mlp.gate_up_proj.q_scale",
"shape": [
32768,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4194304,
"byteOffset": 28385280
},
{
"name": "model.layers.10.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 32579584
}
],
"md5sum": "13491f73f28dd012cb8eebc3ec069f33"
},
{
"dataPath": "params_shard_8.bin",
"format": "raw-shard",
"nbytes": 33554432,
"records": [
{
"name": "model.layers.11.mlp.gate_up_proj.q_weight",
"shape": [
32768,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 33554432,
"byteOffset": 0
}
],
"md5sum": "e6a1202d4af336cba5ecf9e83be120ea"
},
{
"dataPath": "params_shard_9.bin",
"format": "raw-shard",
"nbytes": 33431552,
"records": [
{
"name": "model.layers.10.self_attn.qkv_proj.q_weight",
"shape": [
2560,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2621440,
"byteOffset": 0
},
{
"name": "model.layers.10.self_attn.qkv_proj.q_scale",
"shape": [
2560,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 327680,
"byteOffset": 2621440
},
{
"name": "model.layers.10.self_attn.o_proj.q_weight",
"shape": [
2048,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 2949120
},
{
"name": "model.layers.10.self_attn.o_proj.q_scale",
"shape": [
2048,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 5046272
},
{
"name": "model.layers.11.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 5308416
},
{
"name": "model.layers.11.mlp.down_proj.q_weight",
"shape": [
2048,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 5312512
},
{
"name": "model.layers.11.mlp.down_proj.q_scale",
"shape": [
2048,
512
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 22089728
},
{
"name": "model.layers.11.mlp.gate_up_proj.q_scale",
"shape": [
32768,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4194304,
"byteOffset": 24186880
},
{
"name": "model.layers.11.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 28381184
},
{
"name": "model.layers.11.self_attn.qkv_proj.q_weight",
"shape": [
2560,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2621440,
"byteOffset": 28385280
},
{
"name": "model.layers.11.self_attn.qkv_proj.q_scale",
"shape": [
2560,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 327680,
"byteOffset": 31006720
},
{
"name": "model.layers.11.self_attn.o_proj.q_weight",
"shape": [
2048,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 31334400
}
],
"md5sum": "0648344befe06162eb0c656f955ef662"
},
{
"dataPath": "params_shard_10.bin",
"format": "raw-shard",
"nbytes": 33554432,
"records": [
{
"name": "model.layers.12.mlp.gate_up_proj.q_weight",
"shape": [
32768,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 33554432,
"byteOffset": 0
}
],
"md5sum": "6856b6f4920078ccf54cee33b6f696e0"
},
{
"dataPath": "params_shard_11.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.13.mlp.down_proj.q_weight",
"shape": [
2048,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "e4e10ae0a1d7bdc290ef4ee35785bf7e"
},
{
"dataPath": "params_shard_12.bin",
"format": "raw-shard",
"nbytes": 33554432,
"records": [
{
"name": "model.layers.13.mlp.gate_up_proj.q_weight",
"shape": [
32768,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 33554432,
"byteOffset": 0
}
],
"md5sum": "0dc1f5481eb55b0a91dd7cfffa8c9fd8"
},
{
"dataPath": "params_shard_13.bin",
"format": "raw-shard",
"nbytes": 30748672,
"records": [
{
"name": "model.layers.11.self_attn.o_proj.q_scale",
"shape": [
2048,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 0
},
{
"name": "model.layers.12.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 262144
},
{
"name": "model.layers.12.mlp.down_proj.q_weight",
"shape": [
2048,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 266240
},
{
"name": "model.layers.12.mlp.down_proj.q_scale",
"shape": [
2048,
512
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 17043456
},
{
"name": "model.layers.12.mlp.gate_up_proj.q_scale",
"shape": [
32768,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4194304,
"byteOffset": 19140608
},
{
"name": "model.layers.12.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 23334912
},
{
"name": "model.layers.12.self_attn.qkv_proj.q_weight",
"shape": [
2560,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2621440,
"byteOffset": 23339008
},
{
"name": "model.layers.12.self_attn.qkv_proj.q_scale",
"shape": [
2560,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 327680,
"byteOffset": 25960448
},
{
"name": "model.layers.12.self_attn.o_proj.q_weight",
"shape": [
2048,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 26288128
},
{
"name": "model.layers.12.self_attn.o_proj.q_scale",
"shape": [
2048,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 28385280
},
{
"name": "model.layers.13.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 28647424
},
{
"name": "model.layers.13.mlp.down_proj.q_scale",
"shape": [
2048,
512
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 28651520
}
],
"md5sum": "ade90f9e1a6bfc0e047787c8de05dfeb"
},
{
"dataPath": "params_shard_14.bin",
"format": "raw-shard",
"nbytes": 33554432,
"records": [
{
"name": "model.layers.14.mlp.gate_up_proj.q_weight",
"shape": [
32768,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 33554432,
"byteOffset": 0
}
],
"md5sum": "fcf6e10970a5ed97502968c52ac6e36e"
},
{
"dataPath": "params_shard_15.bin",
"format": "raw-shard",
"nbytes": 32583680,
"records": [
{
"name": "model.layers.13.mlp.gate_up_proj.q_scale",
"shape": [
32768,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4194304,
"byteOffset": 0
},
{
"name": "model.layers.13.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 4194304
},
{
"name": "model.layers.13.self_attn.qkv_proj.q_weight",
"shape": [
2560,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2621440,
"byteOffset": 4198400
},
{
"name": "model.layers.13.self_attn.qkv_proj.q_scale",
"shape": [
2560,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 327680,
"byteOffset": 6819840
},
{
"name": "model.layers.13.self_attn.o_proj.q_weight",
"shape": [
2048,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 7147520
},
{
"name": "model.layers.13.self_attn.o_proj.q_scale",
"shape": [
2048,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 9244672
},
{
"name": "model.layers.14.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 9506816
},
{
"name": "model.layers.14.mlp.down_proj.q_weight",
"shape": [
2048,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 9510912
},
{
"name": "model.layers.14.mlp.down_proj.q_scale",
"shape": [
2048,
512
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 26288128
},
{
"name": "model.layers.14.mlp.gate_up_proj.q_scale",
"shape": [
32768,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4194304,
"byteOffset": 28385280
},
{
"name": "model.layers.14.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 32579584
}
],
"md5sum": "bd8734d49f26450c4a4748fd9263640c"
},
{
"dataPath": "params_shard_16.bin",
"format": "raw-shard",
"nbytes": 33554432,
"records": [
{
"name": "model.layers.15.mlp.gate_up_proj.q_weight",
"shape": [
32768,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 33554432,
"byteOffset": 0
}
],
"md5sum": "48d736a273bc723a2690c74f0fda6a4e"
},
{
"dataPath": "params_shard_17.bin",
"format": "raw-shard",
"nbytes": 33431552,
"records": [
{
"name": "model.layers.14.self_attn.qkv_proj.q_weight",
"shape": [
2560,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2621440,
"byteOffset": 0
},
{
"name": "model.layers.14.self_attn.qkv_proj.q_scale",
"shape": [
2560,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 327680,
"byteOffset": 2621440
},
{
"name": "model.layers.14.self_attn.o_proj.q_weight",
"shape": [
2048,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 2949120
},
{
"name": "model.layers.14.self_attn.o_proj.q_scale",
"shape": [
2048,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 5046272
},
{
"name": "model.layers.15.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 5308416
},
{
"name": "model.layers.15.mlp.down_proj.q_weight",
"shape": [
2048,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 5312512
},
{
"name": "model.layers.15.mlp.down_proj.q_scale",
"shape": [
2048,
512
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 22089728
},
{
"name": "model.layers.15.mlp.gate_up_proj.q_scale",
"shape": [
32768,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4194304,
"byteOffset": 24186880
},
{
"name": "model.layers.15.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 28381184
},
{
"name": "model.layers.15.self_attn.qkv_proj.q_weight",
"shape": [
2560,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2621440,
"byteOffset": 28385280
},
{
"name": "model.layers.15.self_attn.qkv_proj.q_scale",
"shape": [
2560,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 327680,
"byteOffset": 31006720
},
{
"name": "model.layers.15.self_attn.o_proj.q_weight",
"shape": [
2048,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 31334400
}
],
"md5sum": "d0f02a184933faee0a170143e7fae038"
},
{
"dataPath": "params_shard_18.bin",
"format": "raw-shard",
"nbytes": 33554432,
"records": [
{
"name": "model.layers.16.mlp.gate_up_proj.q_weight",
"shape": [
32768,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 33554432,
"byteOffset": 0
}
],
"md5sum": "dbe8745bc723a4513e8d78356faa72b1"
},
{
"dataPath": "params_shard_19.bin",
"format": "raw-shard",
"nbytes": 33554432,
"records": [
{
"name": "model.layers.17.mlp.gate_up_proj.q_weight",
"shape": [
32768,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 33554432,
"byteOffset": 0
}
],
"md5sum": "19b29e912202e7bde2602fc11551fbd0"
},
{
"dataPath": "params_shard_20.bin",
"format": "raw-shard",
"nbytes": 32841728,
"records": [
{
"name": "model.layers.15.self_attn.o_proj.q_scale",
"shape": [
2048,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 0
},
{
"name": "model.layers.16.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 262144
},
{
"name": "model.layers.16.mlp.down_proj.q_weight",
"shape": [
2048,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 266240
},
{
"name": "model.layers.16.mlp.down_proj.q_scale",
"shape": [
2048,
512
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 17043456
},
{
"name": "model.layers.16.mlp.gate_up_proj.q_scale",
"shape": [
32768,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4194304,
"byteOffset": 19140608
},
{
"name": "model.layers.16.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 23334912
},
{
"name": "model.layers.16.self_attn.qkv_proj.q_weight",
"shape": [
2560,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2621440,
"byteOffset": 23339008
},
{
"name": "model.layers.16.self_attn.qkv_proj.q_scale",
"shape": [
2560,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 327680,
"byteOffset": 25960448
},
{
"name": "model.layers.16.self_attn.o_proj.q_weight",
"shape": [
2048,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 26288128
},
{
"name": "model.layers.16.self_attn.o_proj.q_scale",
"shape": [
2048,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 28385280
},
{
"name": "model.layers.17.mlp.gate_up_proj.q_scale",
"shape": [
32768,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4194304,
"byteOffset": 28647424
}
],
"md5sum": "00f72f9098ac47ac3d434bab36dc5d40"
},
{
"dataPath": "params_shard_21.bin",
"format": "raw-shard",
"nbytes": 33554432,
"records": [
{
"name": "model.layers.2.mlp.gate_up_proj.q_weight",
"shape": [
32768,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 33554432,
"byteOffset": 0
}
],
"md5sum": "ece8b367cf8860f50e31770721afaac8"
},
{
"dataPath": "params_shard_22.bin",
"format": "raw-shard",
"nbytes": 33431552,
"records": [
{
"name": "model.layers.17.self_attn.qkv_proj.q_weight",
"shape": [
2560,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2621440,
"byteOffset": 0
},
{
"name": "model.layers.17.self_attn.qkv_proj.q_scale",
"shape": [
2560,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 327680,
"byteOffset": 2621440
},
{
"name": "model.layers.17.self_attn.o_proj.q_weight",
"shape": [
2048,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 2949120
},
{
"name": "model.layers.17.self_attn.o_proj.q_scale",
"shape": [
2048,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 5046272
},
{
"name": "model.layers.2.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 5308416
},
{
"name": "model.layers.2.mlp.down_proj.q_weight",
"shape": [
2048,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 5312512
},
{
"name": "model.layers.2.mlp.down_proj.q_scale",
"shape": [
2048,
512
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 22089728
},
{
"name": "model.layers.2.mlp.gate_up_proj.q_scale",
"shape": [
32768,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4194304,
"byteOffset": 24186880
},
{
"name": "model.layers.2.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 28381184
},
{
"name": "model.layers.2.self_attn.qkv_proj.q_weight",
"shape": [
2560,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2621440,
"byteOffset": 28385280
},
{
"name": "model.layers.2.self_attn.qkv_proj.q_scale",
"shape": [
2560,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 327680,
"byteOffset": 31006720
},
{
"name": "model.layers.2.self_attn.o_proj.q_weight",
"shape": [
2048,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 31334400
}
],
"md5sum": "ea62c260387e9e85b51f1789e65f1616"
},
{
"dataPath": "params_shard_23.bin",
"format": "raw-shard",
"nbytes": 33554432,
"records": [
{
"name": "model.layers.3.mlp.gate_up_proj.q_weight",
"shape": [
32768,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 33554432,
"byteOffset": 0
}
],
"md5sum": "9a6af9a5e4a0265090188427e1aecfc7"
},
{
"dataPath": "params_shard_24.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.4.mlp.down_proj.q_weight",
"shape": [
2048,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "981c5debf4daf3c9c656012fd4e33fc3"
},
{
"dataPath": "params_shard_25.bin",
"format": "raw-shard",
"nbytes": 33554432,
"records": [
{
"name": "model.layers.4.mlp.gate_up_proj.q_weight",
"shape": [
32768,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 33554432,
"byteOffset": 0
}
],
"md5sum": "2f6765a79b9593adf7278225413c936d"
},
{
"dataPath": "params_shard_26.bin",
"format": "raw-shard",
"nbytes": 30748672,
"records": [
{
"name": "model.layers.2.self_attn.o_proj.q_scale",
"shape": [
2048,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 0
},
{
"name": "model.layers.3.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 262144
},
{
"name": "model.layers.3.mlp.down_proj.q_weight",
"shape": [
2048,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 266240
},
{
"name": "model.layers.3.mlp.down_proj.q_scale",
"shape": [
2048,
512
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 17043456
},
{
"name": "model.layers.3.mlp.gate_up_proj.q_scale",
"shape": [
32768,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4194304,
"byteOffset": 19140608
},
{
"name": "model.layers.3.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 23334912
},
{
"name": "model.layers.3.self_attn.qkv_proj.q_weight",
"shape": [
2560,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2621440,
"byteOffset": 23339008
},
{
"name": "model.layers.3.self_attn.qkv_proj.q_scale",
"shape": [
2560,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 327680,
"byteOffset": 25960448
},
{
"name": "model.layers.3.self_attn.o_proj.q_weight",
"shape": [
2048,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 26288128
},
{
"name": "model.layers.3.self_attn.o_proj.q_scale",
"shape": [
2048,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 28385280
},
{
"name": "model.layers.4.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 28647424
},
{
"name": "model.layers.4.mlp.down_proj.q_scale",
"shape": [
2048,
512
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 28651520
}
],
"md5sum": "bf4c5b8f508dd8b6fb73689ff6b76f25"
},
{
"dataPath": "params_shard_27.bin",
"format": "raw-shard",
"nbytes": 33554432,
"records": [
{
"name": "model.layers.5.mlp.gate_up_proj.q_weight",
"shape": [
32768,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 33554432,
"byteOffset": 0
}
],
"md5sum": "9209de41b57f95ab784075aa5816bbc6"
},
{
"dataPath": "params_shard_28.bin",
"format": "raw-shard",
"nbytes": 32583680,
"records": [
{
"name": "model.layers.4.mlp.gate_up_proj.q_scale",
"shape": [
32768,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4194304,
"byteOffset": 0
},
{
"name": "model.layers.4.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 4194304
},
{
"name": "model.layers.4.self_attn.qkv_proj.q_weight",
"shape": [
2560,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2621440,
"byteOffset": 4198400
},
{
"name": "model.layers.4.self_attn.qkv_proj.q_scale",
"shape": [
2560,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 327680,
"byteOffset": 6819840
},
{
"name": "model.layers.4.self_attn.o_proj.q_weight",
"shape": [
2048,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 7147520
},
{
"name": "model.layers.4.self_attn.o_proj.q_scale",
"shape": [
2048,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 9244672
},
{
"name": "model.layers.5.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 9506816
},
{
"name": "model.layers.5.mlp.down_proj.q_weight",
"shape": [
2048,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 9510912
},
{
"name": "model.layers.5.mlp.down_proj.q_scale",
"shape": [
2048,
512
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 26288128
},
{
"name": "model.layers.5.mlp.gate_up_proj.q_scale",
"shape": [
32768,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4194304,
"byteOffset": 28385280
},
{
"name": "model.layers.5.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 32579584
}
],
"md5sum": "4ba09f58095510e4cba2fb177014cf56"
},
{
"dataPath": "params_shard_29.bin",
"format": "raw-shard",
"nbytes": 33554432,
"records": [
{
"name": "model.layers.6.mlp.gate_up_proj.q_weight",
"shape": [
32768,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 33554432,
"byteOffset": 0
}
],
"md5sum": "9fd4e9c8e4490fdc8a9bd3b9897f0e66"
},
{
"dataPath": "params_shard_30.bin",
"format": "raw-shard",
"nbytes": 33431552,
"records": [
{
"name": "model.layers.5.self_attn.qkv_proj.q_weight",
"shape": [
2560,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2621440,
"byteOffset": 0
},
{
"name": "model.layers.5.self_attn.qkv_proj.q_scale",
"shape": [
2560,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 327680,
"byteOffset": 2621440
},
{
"name": "model.layers.5.self_attn.o_proj.q_weight",
"shape": [
2048,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 2949120
},
{
"name": "model.layers.5.self_attn.o_proj.q_scale",
"shape": [
2048,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 5046272
},
{
"name": "model.layers.6.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 5308416
},
{
"name": "model.layers.6.mlp.down_proj.q_weight",
"shape": [
2048,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 5312512
},
{
"name": "model.layers.6.mlp.down_proj.q_scale",
"shape": [
2048,
512
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 22089728
},
{
"name": "model.layers.6.mlp.gate_up_proj.q_scale",
"shape": [
32768,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4194304,
"byteOffset": 24186880
},
{
"name": "model.layers.6.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 28381184
},
{
"name": "model.layers.6.self_attn.qkv_proj.q_weight",
"shape": [
2560,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2621440,
"byteOffset": 28385280
},
{
"name": "model.layers.6.self_attn.qkv_proj.q_scale",
"shape": [
2560,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 327680,
"byteOffset": 31006720
},
{
"name": "model.layers.6.self_attn.o_proj.q_weight",
"shape": [
2048,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 31334400
}
],
"md5sum": "7971455c7f41a576a7f93ee5b0f73def"
},
{
"dataPath": "params_shard_31.bin",
"format": "raw-shard",
"nbytes": 33554432,
"records": [
{
"name": "model.layers.7.mlp.gate_up_proj.q_weight",
"shape": [
32768,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 33554432,
"byteOffset": 0
}
],
"md5sum": "38386a351e96f7f41e743d8bed02e694"
},
{
"dataPath": "params_shard_32.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.8.mlp.down_proj.q_weight",
"shape": [
2048,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "51d83e40bad4edebb89aa78658dbfc1b"
},
{
"dataPath": "params_shard_33.bin",
"format": "raw-shard",
"nbytes": 33554432,
"records": [
{
"name": "model.layers.8.mlp.gate_up_proj.q_weight",
"shape": [
32768,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 33554432,
"byteOffset": 0
}
],
"md5sum": "37945461998e7d29b6368e2ee512c542"
},
{
"dataPath": "params_shard_34.bin",
"format": "raw-shard",
"nbytes": 30748672,
"records": [
{
"name": "model.layers.6.self_attn.o_proj.q_scale",
"shape": [
2048,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 0
},
{
"name": "model.layers.7.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 262144
},
{
"name": "model.layers.7.mlp.down_proj.q_weight",
"shape": [
2048,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 266240
},
{
"name": "model.layers.7.mlp.down_proj.q_scale",
"shape": [
2048,
512
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 17043456
},
{
"name": "model.layers.7.mlp.gate_up_proj.q_scale",
"shape": [
32768,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4194304,
"byteOffset": 19140608
},
{
"name": "model.layers.7.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 23334912
},
{
"name": "model.layers.7.self_attn.qkv_proj.q_weight",
"shape": [
2560,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2621440,
"byteOffset": 23339008
},
{
"name": "model.layers.7.self_attn.qkv_proj.q_scale",
"shape": [
2560,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 327680,
"byteOffset": 25960448
},
{
"name": "model.layers.7.self_attn.o_proj.q_weight",
"shape": [
2048,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 26288128
},
{
"name": "model.layers.7.self_attn.o_proj.q_scale",
"shape": [
2048,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 28385280
},
{
"name": "model.layers.8.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 28647424
},
{
"name": "model.layers.8.mlp.down_proj.q_scale",
"shape": [
2048,
512
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 28651520
}
],
"md5sum": "a837aa9bd24bee5d89a9f053b4818107"
},
{
"dataPath": "params_shard_35.bin",
"format": "raw-shard",
"nbytes": 33554432,
"records": [
{
"name": "model.layers.9.mlp.gate_up_proj.q_weight",
"shape": [
32768,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 33554432,
"byteOffset": 0
}
],
"md5sum": "78b8bb3f1930f61572ed644032c38fe4"
},
{
"dataPath": "params_shard_36.bin",
"format": "raw-shard",
"nbytes": 32583680,
"records": [
{
"name": "model.layers.8.mlp.gate_up_proj.q_scale",
"shape": [
32768,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4194304,
"byteOffset": 0
},
{
"name": "model.layers.8.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 4194304
},
{
"name": "model.layers.8.self_attn.qkv_proj.q_weight",
"shape": [
2560,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2621440,
"byteOffset": 4198400
},
{
"name": "model.layers.8.self_attn.qkv_proj.q_scale",
"shape": [
2560,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 327680,
"byteOffset": 6819840
},
{
"name": "model.layers.8.self_attn.o_proj.q_weight",
"shape": [
2048,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 7147520
},
{
"name": "model.layers.8.self_attn.o_proj.q_scale",
"shape": [
2048,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 9244672
},
{
"name": "model.layers.9.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 9506816
},
{
"name": "model.layers.9.mlp.down_proj.q_weight",
"shape": [
2048,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 9510912
},
{
"name": "model.layers.9.mlp.down_proj.q_scale",
"shape": [
2048,
512
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 26288128
},
{
"name": "model.layers.9.mlp.gate_up_proj.q_scale",
"shape": [
32768,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4194304,
"byteOffset": 28385280
},
{
"name": "model.layers.9.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 32579584
}
],
"md5sum": "b8af44078990172bbf3ec685fd7a8d85"
},
{
"dataPath": "params_shard_37.bin",
"format": "raw-shard",
"nbytes": 24195072,
"records": [
{
"name": "model.layers.9.self_attn.qkv_proj.q_weight",
"shape": [
2560,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2621440,
"byteOffset": 0
},
{
"name": "model.layers.9.self_attn.qkv_proj.q_scale",
"shape": [
2560,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 327680,
"byteOffset": 2621440
},
{
"name": "model.layers.9.self_attn.o_proj.q_weight",
"shape": [
2048,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 2949120
},
{
"name": "model.layers.9.self_attn.o_proj.q_scale",
"shape": [
2048,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 5046272
},
{
"name": "model.layers.17.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 5308416
},
{
"name": "model.layers.17.mlp.down_proj.q_weight",
"shape": [
2048,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 5312512
},
{
"name": "model.layers.17.mlp.down_proj.q_scale",
"shape": [
2048,
512
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 22089728
},
{
"name": "model.layers.17.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 24186880
},
{
"name": "model.norm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 24190976
}
],
"md5sum": "b07fe16f614531a4881acf7acbfb8dd7"
}
]
}