scigemma_fine_tuned_quantized_MLC / ndarray-cache.json
shivaylamba's picture
Fine-tuned quantized Gemma 2b-it model.
303ea53 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": "4a94d33574c77287f7a6999507a5039c"
},
{
"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": "b01723cbcd26b1df29c27b46949e4b66"
},
{
"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": "93a025b9f9ba57adbeffe43b470e463c"
},
{
"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": "c6b4d43cf2db667f72d1749e0c5f73cd"
},
{
"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": "30e4b0fa05d390e5c6c5875908b1abda"
},
{
"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": "859d6bbee571a99f925ac4a85e13944f"
},
{
"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": "c19da6a685272067182bcb79faf8e5e6"
},
{
"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": "729e62bec6f5656a7a8cbe06508ed584"
},
{
"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": "445db82f670eb539eb5a420903bc2dab"
},
{
"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": "19a1caba4a793b654ee18e285f12c2c6"
},
{
"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": "3e43cc406a305dbfff2eec8d0ac30c68"
},
{
"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": "849ad8a45274550645283f5f199514c4"
},
{
"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": "4fc350af289318618a920a78f9002514"
},
{
"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": "2379ff206c3f8188eedce0d2dbe542f9"
},
{
"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": "3167b16e1158b11316722d383959cd2b"
},
{
"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": "d586cd77dfd6ffee06fa5d090d1a4c61"
},
{
"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": "6a7f56b5dab9c4d3bc9a0a78bf41ee45"
},
{
"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": "02d41990b4210267bf98ef49ba989989"
},
{
"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": "37a304656cab5daabd788a36b79b8a61"
},
{
"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": "a17db55e83f527d0719dd3ccd0a5f7e8"
},
{
"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": "d52abb06dde1e91ceea42ca6cc396416"
},
{
"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": "2047e7380c4618430b3c36b8133f9705"
},
{
"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": "31e43150df35feb09ef605a00f1b9f60"
},
{
"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": "8761b0a028ba52653ec4145c72978f04"
},
{
"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": "b575771c2f41886d5ffd774d1fd1f6a2"
},
{
"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": "64c76a1f58fc374d7223e607604e80c1"
},
{
"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": "edc3f5501cf20f5809ba99326827f968"
},
{
"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": "63bc7b8a9cbaede6efeba0f617eaea6c"
},
{
"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": "3d45777ac7224bb09e2ecb98fc90971e"
},
{
"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": "5d384548e9bb844784a8f1d5b60cfd54"
},
{
"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": "a4a917c2b7f98be3b520bedddb075647"
},
{
"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": "06b70e0de23b76f304417c5f7b37c4b0"
},
{
"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": "72a4a4bfbaed6a5a8a862a5661d2d6c0"
},
{
"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": "45a2bad78f964d62393ae8d8ece8f5d7"
},
{
"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": "09c6bae6466f445341c468a6c742e9a3"
},
{
"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": "e548a649a34dbd4a6730fad1355f4251"
}
]
}