SmolLM2-1.7b-Instruct-q4f16_0 / ndarray-cache.json
MelgarK's picture
init
f0453e0
{
"metadata": {
"ParamSize": 243,
"ParamBytes": 962793472.0,
"BitsPerParam": 4.500674339093837
},
"records": [
{
"dataPath": "params_shard_0.bin",
"format": "raw-shard",
"nbytes": 50331648,
"records": [
{
"name": "model.embed_tokens.q_weight",
"shape": [
49152,
256
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 50331648,
"byteOffset": 0
}
],
"md5sum": "aecbb590f5331e434f51e83c5b26786f"
},
{
"dataPath": "params_shard_1.bin",
"format": "raw-shard",
"nbytes": 32509952,
"records": [
{
"name": "model.embed_tokens.q_scale",
"shape": [
49152,
64
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 0
},
{
"name": "model.layers.0.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 6291456
},
{
"name": "model.layers.0.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 6295552
},
{
"name": "model.layers.0.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 14684160
},
{
"name": "model.layers.0.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 15732736
}
],
"md5sum": "5c4aea34fc63918a67ca80388cda88f0"
},
{
"dataPath": "params_shard_2.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.1.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "15e493ac51df4c838950dc2d9183aa6c"
},
{
"dataPath": "params_shard_3.bin",
"format": "raw-shard",
"nbytes": 32522240,
"records": [
{
"name": "model.layers.0.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 0
},
{
"name": "model.layers.0.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 2097152
},
{
"name": "model.layers.0.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 2101248
},
{
"name": "model.layers.0.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 8392704
},
{
"name": "model.layers.0.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 9179136
},
{
"name": "model.layers.0.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 11276288
},
{
"name": "model.layers.1.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 11538432
},
{
"name": "model.layers.1.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 11542528
},
{
"name": "model.layers.1.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 19931136
},
{
"name": "model.layers.1.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 20979712
},
{
"name": "model.layers.1.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 23076864
},
{
"name": "model.layers.1.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 23080960
},
{
"name": "model.layers.1.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 29372416
},
{
"name": "model.layers.1.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 30158848
},
{
"name": "model.layers.1.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 32256000
},
{
"name": "model.layers.10.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 32518144
}
],
"md5sum": "219f490e7c9e202eede11b245b3faa78"
},
{
"dataPath": "params_shard_4.bin",
"format": "raw-shard",
"nbytes": 28315648,
"records": [
{
"name": "model.layers.10.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.10.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.10.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 9437184
},
{
"name": "model.layers.10.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 26214400
},
{
"name": "model.layers.10.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 28311552
}
],
"md5sum": "a13e6b774a02236aaf7e619bc6dfcef6"
},
{
"dataPath": "params_shard_5.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.11.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "3cbb02f6eaaa99a34059ca7f375a32bd"
},
{
"dataPath": "params_shard_6.bin",
"format": "raw-shard",
"nbytes": 30420992,
"records": [
{
"name": "model.layers.10.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 0
},
{
"name": "model.layers.10.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 6291456
},
{
"name": "model.layers.10.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 7077888
},
{
"name": "model.layers.10.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 9175040
},
{
"name": "model.layers.11.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 9437184
},
{
"name": "model.layers.11.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 9441280
},
{
"name": "model.layers.11.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 17829888
},
{
"name": "model.layers.11.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 18878464
},
{
"name": "model.layers.11.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 20975616
},
{
"name": "model.layers.11.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 20979712
},
{
"name": "model.layers.11.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 27271168
},
{
"name": "model.layers.11.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 28057600
},
{
"name": "model.layers.11.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 30154752
},
{
"name": "model.layers.12.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 30416896
}
],
"md5sum": "5a3ffb2e92256a76b0d9e6cec873f784"
},
{
"dataPath": "params_shard_7.bin",
"format": "raw-shard",
"nbytes": 28315648,
"records": [
{
"name": "model.layers.12.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.12.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.12.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 9437184
},
{
"name": "model.layers.12.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 26214400
},
{
"name": "model.layers.12.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 28311552
}
],
"md5sum": "de7fbc76ae42aa67aa799037f24ac52c"
},
{
"dataPath": "params_shard_8.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.13.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "3656e9d9b6b000fdc8c179288680fb44"
},
{
"dataPath": "params_shard_9.bin",
"format": "raw-shard",
"nbytes": 30420992,
"records": [
{
"name": "model.layers.12.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 0
},
{
"name": "model.layers.12.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 6291456
},
{
"name": "model.layers.12.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 7077888
},
{
"name": "model.layers.12.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 9175040
},
{
"name": "model.layers.13.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 9437184
},
{
"name": "model.layers.13.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 9441280
},
{
"name": "model.layers.13.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 17829888
},
{
"name": "model.layers.13.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 18878464
},
{
"name": "model.layers.13.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 20975616
},
{
"name": "model.layers.13.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 20979712
},
{
"name": "model.layers.13.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 27271168
},
{
"name": "model.layers.13.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 28057600
},
{
"name": "model.layers.13.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 30154752
},
{
"name": "model.layers.14.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 30416896
}
],
"md5sum": "7273399c05104f4d023c518b0594a3c2"
},
{
"dataPath": "params_shard_10.bin",
"format": "raw-shard",
"nbytes": 28315648,
"records": [
{
"name": "model.layers.14.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.14.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.14.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 9437184
},
{
"name": "model.layers.14.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 26214400
},
{
"name": "model.layers.14.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 28311552
}
],
"md5sum": "3d010bc3d3b27c74fc50b5b16260d042"
},
{
"dataPath": "params_shard_11.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.15.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "3e26ada98da0aabd7ad735334a748f35"
},
{
"dataPath": "params_shard_12.bin",
"format": "raw-shard",
"nbytes": 30420992,
"records": [
{
"name": "model.layers.14.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 0
},
{
"name": "model.layers.14.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 6291456
},
{
"name": "model.layers.14.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 7077888
},
{
"name": "model.layers.14.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 9175040
},
{
"name": "model.layers.15.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 9437184
},
{
"name": "model.layers.15.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 9441280
},
{
"name": "model.layers.15.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 17829888
},
{
"name": "model.layers.15.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 18878464
},
{
"name": "model.layers.15.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 20975616
},
{
"name": "model.layers.15.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 20979712
},
{
"name": "model.layers.15.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 27271168
},
{
"name": "model.layers.15.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 28057600
},
{
"name": "model.layers.15.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 30154752
},
{
"name": "model.layers.16.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 30416896
}
],
"md5sum": "0623839db27acf8671232a7eebde2c0b"
},
{
"dataPath": "params_shard_13.bin",
"format": "raw-shard",
"nbytes": 28315648,
"records": [
{
"name": "model.layers.16.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.16.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.16.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 9437184
},
{
"name": "model.layers.16.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 26214400
},
{
"name": "model.layers.16.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 28311552
}
],
"md5sum": "f88ff8577d267540384308cbe5c35ec9"
},
{
"dataPath": "params_shard_14.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.17.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "882d589a98d54b6b16b71c4e82d7db74"
},
{
"dataPath": "params_shard_15.bin",
"format": "raw-shard",
"nbytes": 30420992,
"records": [
{
"name": "model.layers.16.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 0
},
{
"name": "model.layers.16.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 6291456
},
{
"name": "model.layers.16.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 7077888
},
{
"name": "model.layers.16.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 9175040
},
{
"name": "model.layers.17.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 9437184
},
{
"name": "model.layers.17.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 9441280
},
{
"name": "model.layers.17.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 17829888
},
{
"name": "model.layers.17.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 18878464
},
{
"name": "model.layers.17.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 20975616
},
{
"name": "model.layers.17.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 20979712
},
{
"name": "model.layers.17.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 27271168
},
{
"name": "model.layers.17.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 28057600
},
{
"name": "model.layers.17.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 30154752
},
{
"name": "model.layers.18.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 30416896
}
],
"md5sum": "116ec89093a09a0bfb0cc4777fd0773c"
},
{
"dataPath": "params_shard_16.bin",
"format": "raw-shard",
"nbytes": 28315648,
"records": [
{
"name": "model.layers.18.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.18.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.18.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 9437184
},
{
"name": "model.layers.18.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 26214400
},
{
"name": "model.layers.18.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 28311552
}
],
"md5sum": "035f2d53581c402b0f271fcbeb13fed2"
},
{
"dataPath": "params_shard_17.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.19.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "0ac45a5e6a7a590d6562f6f39cc98859"
},
{
"dataPath": "params_shard_18.bin",
"format": "raw-shard",
"nbytes": 30420992,
"records": [
{
"name": "model.layers.18.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 0
},
{
"name": "model.layers.18.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 6291456
},
{
"name": "model.layers.18.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 7077888
},
{
"name": "model.layers.18.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 9175040
},
{
"name": "model.layers.19.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 9437184
},
{
"name": "model.layers.19.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 9441280
},
{
"name": "model.layers.19.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 17829888
},
{
"name": "model.layers.19.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 18878464
},
{
"name": "model.layers.19.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 20975616
},
{
"name": "model.layers.19.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 20979712
},
{
"name": "model.layers.19.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 27271168
},
{
"name": "model.layers.19.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 28057600
},
{
"name": "model.layers.19.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 30154752
},
{
"name": "model.layers.2.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 30416896
}
],
"md5sum": "c1f2555fb25f0d1d42da469dd032d63a"
},
{
"dataPath": "params_shard_19.bin",
"format": "raw-shard",
"nbytes": 28315648,
"records": [
{
"name": "model.layers.2.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.2.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.2.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 9437184
},
{
"name": "model.layers.2.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 26214400
},
{
"name": "model.layers.2.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 28311552
}
],
"md5sum": "f03dbe07f43ce086c0c90b39a3b70400"
},
{
"dataPath": "params_shard_20.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.20.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "edaa761b77a4272a858f49a90cdb9367"
},
{
"dataPath": "params_shard_21.bin",
"format": "raw-shard",
"nbytes": 30420992,
"records": [
{
"name": "model.layers.2.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 0
},
{
"name": "model.layers.2.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 6291456
},
{
"name": "model.layers.2.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 7077888
},
{
"name": "model.layers.2.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 9175040
},
{
"name": "model.layers.20.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 9437184
},
{
"name": "model.layers.20.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 9441280
},
{
"name": "model.layers.20.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 17829888
},
{
"name": "model.layers.20.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 18878464
},
{
"name": "model.layers.20.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 20975616
},
{
"name": "model.layers.20.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 20979712
},
{
"name": "model.layers.20.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 27271168
},
{
"name": "model.layers.20.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 28057600
},
{
"name": "model.layers.20.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 30154752
},
{
"name": "model.layers.21.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 30416896
}
],
"md5sum": "c34ee1948869bf0f023229e2bdedec87"
},
{
"dataPath": "params_shard_22.bin",
"format": "raw-shard",
"nbytes": 28315648,
"records": [
{
"name": "model.layers.21.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.21.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.21.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 9437184
},
{
"name": "model.layers.21.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 26214400
},
{
"name": "model.layers.21.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 28311552
}
],
"md5sum": "43db638b4a71edcfa51d3320113f3204"
},
{
"dataPath": "params_shard_23.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.22.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "b315df6b65a02a0393ec937e5b1f1ff9"
},
{
"dataPath": "params_shard_24.bin",
"format": "raw-shard",
"nbytes": 30420992,
"records": [
{
"name": "model.layers.21.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 0
},
{
"name": "model.layers.21.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 6291456
},
{
"name": "model.layers.21.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 7077888
},
{
"name": "model.layers.21.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 9175040
},
{
"name": "model.layers.22.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 9437184
},
{
"name": "model.layers.22.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 9441280
},
{
"name": "model.layers.22.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 17829888
},
{
"name": "model.layers.22.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 18878464
},
{
"name": "model.layers.22.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 20975616
},
{
"name": "model.layers.22.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 20979712
},
{
"name": "model.layers.22.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 27271168
},
{
"name": "model.layers.22.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 28057600
},
{
"name": "model.layers.22.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 30154752
},
{
"name": "model.layers.23.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 30416896
}
],
"md5sum": "048cabca1ad55caf2fd1f3f8601cd442"
},
{
"dataPath": "params_shard_25.bin",
"format": "raw-shard",
"nbytes": 28315648,
"records": [
{
"name": "model.layers.23.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.23.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.23.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 9437184
},
{
"name": "model.layers.23.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 26214400
},
{
"name": "model.layers.23.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 28311552
}
],
"md5sum": "8c0ce6a24334c6e9f8aea11707255a5b"
},
{
"dataPath": "params_shard_26.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.3.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "e1d8a04a59f3d4f5fb10f4d59b5033f2"
},
{
"dataPath": "params_shard_27.bin",
"format": "raw-shard",
"nbytes": 30420992,
"records": [
{
"name": "model.layers.23.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 0
},
{
"name": "model.layers.23.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 6291456
},
{
"name": "model.layers.23.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 7077888
},
{
"name": "model.layers.23.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 9175040
},
{
"name": "model.layers.3.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 9437184
},
{
"name": "model.layers.3.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 9441280
},
{
"name": "model.layers.3.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 17829888
},
{
"name": "model.layers.3.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 18878464
},
{
"name": "model.layers.3.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 20975616
},
{
"name": "model.layers.3.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 20979712
},
{
"name": "model.layers.3.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 27271168
},
{
"name": "model.layers.3.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 28057600
},
{
"name": "model.layers.3.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 30154752
},
{
"name": "model.layers.4.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 30416896
}
],
"md5sum": "7661f255ed30f82a9db88ab8978c5cb8"
},
{
"dataPath": "params_shard_28.bin",
"format": "raw-shard",
"nbytes": 28315648,
"records": [
{
"name": "model.layers.4.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.4.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.4.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 9437184
},
{
"name": "model.layers.4.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 26214400
},
{
"name": "model.layers.4.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 28311552
}
],
"md5sum": "c39356e169233125a690f3d301c37899"
},
{
"dataPath": "params_shard_29.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.5.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "0457ee696fcd785ee43d4dbe8438f831"
},
{
"dataPath": "params_shard_30.bin",
"format": "raw-shard",
"nbytes": 30420992,
"records": [
{
"name": "model.layers.4.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 0
},
{
"name": "model.layers.4.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 6291456
},
{
"name": "model.layers.4.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 7077888
},
{
"name": "model.layers.4.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 9175040
},
{
"name": "model.layers.5.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 9437184
},
{
"name": "model.layers.5.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 9441280
},
{
"name": "model.layers.5.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 17829888
},
{
"name": "model.layers.5.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 18878464
},
{
"name": "model.layers.5.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 20975616
},
{
"name": "model.layers.5.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 20979712
},
{
"name": "model.layers.5.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 27271168
},
{
"name": "model.layers.5.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 28057600
},
{
"name": "model.layers.5.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 30154752
},
{
"name": "model.layers.6.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 30416896
}
],
"md5sum": "a082f39f4fd790e7903785b275ab8737"
},
{
"dataPath": "params_shard_31.bin",
"format": "raw-shard",
"nbytes": 28315648,
"records": [
{
"name": "model.layers.6.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.6.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.6.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 9437184
},
{
"name": "model.layers.6.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 26214400
},
{
"name": "model.layers.6.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 28311552
}
],
"md5sum": "569af8d20b5d138d71b13b77ea4f31a4"
},
{
"dataPath": "params_shard_32.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.7.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "71a77a6db3e7d729b07e46a6e0c17a67"
},
{
"dataPath": "params_shard_33.bin",
"format": "raw-shard",
"nbytes": 30420992,
"records": [
{
"name": "model.layers.6.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 0
},
{
"name": "model.layers.6.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 6291456
},
{
"name": "model.layers.6.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 7077888
},
{
"name": "model.layers.6.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 9175040
},
{
"name": "model.layers.7.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 9437184
},
{
"name": "model.layers.7.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 9441280
},
{
"name": "model.layers.7.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 17829888
},
{
"name": "model.layers.7.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 18878464
},
{
"name": "model.layers.7.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 20975616
},
{
"name": "model.layers.7.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 20979712
},
{
"name": "model.layers.7.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 27271168
},
{
"name": "model.layers.7.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 28057600
},
{
"name": "model.layers.7.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 30154752
},
{
"name": "model.layers.8.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 30416896
}
],
"md5sum": "5708b28f759c7d85a69f0df2c5bd1a4e"
},
{
"dataPath": "params_shard_34.bin",
"format": "raw-shard",
"nbytes": 28315648,
"records": [
{
"name": "model.layers.8.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 0
},
{
"name": "model.layers.8.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 8388608
},
{
"name": "model.layers.8.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 9437184
},
{
"name": "model.layers.8.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 26214400
},
{
"name": "model.layers.8.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 28311552
}
],
"md5sum": "43940eb6d48621fb383bd57283f43e03"
},
{
"dataPath": "params_shard_35.bin",
"format": "raw-shard",
"nbytes": 16777216,
"records": [
{
"name": "model.layers.9.mlp.gate_up_proj.q_weight",
"shape": [
256,
16384
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 16777216,
"byteOffset": 0
}
],
"md5sum": "c1f5652b20b772c46f50bd76fde7e56a"
},
{
"dataPath": "params_shard_36.bin",
"format": "raw-shard",
"nbytes": 30420992,
"records": [
{
"name": "model.layers.8.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 0
},
{
"name": "model.layers.8.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 6291456
},
{
"name": "model.layers.8.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 7077888
},
{
"name": "model.layers.8.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 9175040
},
{
"name": "model.layers.9.input_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 9437184
},
{
"name": "model.layers.9.mlp.down_proj.q_weight",
"shape": [
1024,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 8388608,
"byteOffset": 9441280
},
{
"name": "model.layers.9.mlp.down_proj.q_scale",
"shape": [
256,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 1048576,
"byteOffset": 17829888
},
{
"name": "model.layers.9.mlp.gate_up_proj.q_scale",
"shape": [
64,
16384
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 18878464
},
{
"name": "model.layers.9.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 20975616
},
{
"name": "model.layers.9.self_attn.qkv_proj.q_weight",
"shape": [
256,
6144
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 6291456,
"byteOffset": 20979712
},
{
"name": "model.layers.9.self_attn.qkv_proj.q_scale",
"shape": [
64,
6144
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 786432,
"byteOffset": 27271168
},
{
"name": "model.layers.9.self_attn.o_proj.q_weight",
"shape": [
256,
2048
],
"dtype": "uint32",
"format": "f32-to-bf16",
"nbytes": 2097152,
"byteOffset": 28057600
},
{
"name": "model.layers.9.self_attn.o_proj.q_scale",
"shape": [
64,
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 262144,
"byteOffset": 30154752
},
{
"name": "model.norm.weight",
"shape": [
2048
],
"dtype": "float16",
"format": "f32-to-bf16",
"nbytes": 4096,
"byteOffset": 30416896
}
],
"md5sum": "651344c1b6db233afe34e18ea3ab92c4"
}
]
}