Files
Qwen__Qwen3-4B-Thinking-250…/fake_quant_config.json
ModelHub XC 335724feac 初始化项目,由ModelHub XC社区提供模型
Model: meteorain/Qwen__Qwen3-4B-Thinking-2507-RTN-fp4-e2m1-g128-faked-bf16
Source: Original Platform
2026-08-23 14:44:20 +08:00

2537 lines
52 KiB
JSON

{
"format": "fp4-e2m1",
"codebook": "fp4-e2m1",
"block_size": 128,
"num_modules": 252,
"num_weights": 3633315840,
"elapsed_sec": 88.056,
"quantized_modules": [
{
"name": "model.layers.0.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.201322776178131e-06,
"max_abs_err": 0.0478515625
},
{
"name": "model.layers.0.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.8370086410141084e-06,
"max_abs_err": 0.0458984375
},
{
"name": "model.layers.0.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.054790901544038e-06,
"max_abs_err": 0.02001953125
},
{
"name": "model.layers.0.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.426284587883856e-06,
"max_abs_err": 0.06640625
},
{
"name": "model.layers.0.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.378349098667968e-06,
"max_abs_err": 0.044921875
},
{
"name": "model.layers.0.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 5.554906692850636e-06,
"max_abs_err": 0.041015625
},
{
"name": "model.layers.0.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 6.256674623728031e-06,
"max_abs_err": 0.0703125
},
{
"name": "model.layers.1.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 5.377312845666893e-06,
"max_abs_err": 0.05078125
},
{
"name": "model.layers.1.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 5.800921371701406e-06,
"max_abs_err": 0.0283203125
},
{
"name": "model.layers.1.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.191730790305883e-06,
"max_abs_err": 0.0205078125
},
{
"name": "model.layers.1.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.505150966200745e-06,
"max_abs_err": 0.0546875
},
{
"name": "model.layers.1.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.027107247064123e-06,
"max_abs_err": 0.0361328125
},
{
"name": "model.layers.1.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 2.73822342933272e-06,
"max_abs_err": 0.0341796875
},
{
"name": "model.layers.1.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 3.171168600601959e-06,
"max_abs_err": 0.091796875
},
{
"name": "model.layers.2.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 5.740839696954936e-06,
"max_abs_err": 0.0439453125
},
{
"name": "model.layers.2.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.230192411749158e-06,
"max_abs_err": 0.0498046875
},
{
"name": "model.layers.2.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.671000392088899e-06,
"max_abs_err": 0.02001953125
},
{
"name": "model.layers.2.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.74190880797687e-06,
"max_abs_err": 0.0439453125
},
{
"name": "model.layers.2.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 4.111591351829702e-06,
"max_abs_err": 0.0400390625
},
{
"name": "model.layers.2.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 3.437744453549385e-06,
"max_abs_err": 0.06640625
},
{
"name": "model.layers.2.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 3.9127653508330695e-06,
"max_abs_err": 0.0625
},
{
"name": "model.layers.3.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 5.802833584311884e-06,
"max_abs_err": 0.0380859375
},
{
"name": "model.layers.3.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 5.819788384542335e-06,
"max_abs_err": 0.0302734375
},
{
"name": "model.layers.3.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.759400093869772e-06,
"max_abs_err": 0.02197265625
},
{
"name": "model.layers.3.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.816717930429149e-06,
"max_abs_err": 0.05859375
},
{
"name": "model.layers.3.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 7.263103725563269e-06,
"max_abs_err": 0.052734375
},
{
"name": "model.layers.3.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 3.9278734220715705e-06,
"max_abs_err": 0.0517578125
},
{
"name": "model.layers.3.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 4.3746426854340825e-06,
"max_abs_err": 0.076171875
},
{
"name": "model.layers.4.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.002755981171504e-06,
"max_abs_err": 0.0419921875
},
{
"name": "model.layers.4.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.1892119447293226e-06,
"max_abs_err": 0.05859375
},
{
"name": "model.layers.4.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.882107300043572e-06,
"max_abs_err": 0.01904296875
},
{
"name": "model.layers.4.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 6.069946266507031e-06,
"max_abs_err": 0.068359375
},
{
"name": "model.layers.4.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 1.1256169273110572e-05,
"max_abs_err": 0.0419921875
},
{
"name": "model.layers.4.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 4.754626388603356e-06,
"max_abs_err": 0.03515625
},
{
"name": "model.layers.4.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 5.090760623716051e-06,
"max_abs_err": 0.076171875
},
{
"name": "model.layers.5.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 5.9362282627262175e-06,
"max_abs_err": 0.0439453125
},
{
"name": "model.layers.5.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 5.912828783039004e-06,
"max_abs_err": 0.052734375
},
{
"name": "model.layers.5.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.794049113523215e-06,
"max_abs_err": 0.0205078125
},
{
"name": "model.layers.5.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.666167908202624e-06,
"max_abs_err": 0.060546875
},
{
"name": "model.layers.5.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 9.924154255713802e-06,
"max_abs_err": 0.048828125
},
{
"name": "model.layers.5.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 5.342687018128345e-06,
"max_abs_err": 0.044921875
},
{
"name": "model.layers.5.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 5.323748609953327e-06,
"max_abs_err": 0.099609375
},
{
"name": "model.layers.6.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.310954631771892e-06,
"max_abs_err": 0.041015625
},
{
"name": "model.layers.6.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 5.938825779594481e-06,
"max_abs_err": 0.046875
},
{
"name": "model.layers.6.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.854073490103474e-06,
"max_abs_err": 0.03515625
},
{
"name": "model.layers.6.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 6.054650384612614e-06,
"max_abs_err": 0.0859375
},
{
"name": "model.layers.6.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 8.740071280044504e-06,
"max_abs_err": 0.0458984375
},
{
"name": "model.layers.6.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 5.650788807542995e-06,
"max_abs_err": 0.048828125
},
{
"name": "model.layers.6.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 5.723705726268236e-06,
"max_abs_err": 0.08203125
},
{
"name": "model.layers.7.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 5.9416552176116966e-06,
"max_abs_err": 0.052734375
},
{
"name": "model.layers.7.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.005701834510546e-06,
"max_abs_err": 0.0419921875
},
{
"name": "model.layers.7.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 7.775552148814313e-06,
"max_abs_err": 0.021484375
},
{
"name": "model.layers.7.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.903746114199748e-06,
"max_abs_err": 0.05859375
},
{
"name": "model.layers.7.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 8.53730489325244e-06,
"max_abs_err": 0.0546875
},
{
"name": "model.layers.7.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 5.71421787753934e-06,
"max_abs_err": 0.041015625
},
{
"name": "model.layers.7.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 5.725755272578681e-06,
"max_abs_err": 0.046875
},
{
"name": "model.layers.8.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.006920557410922e-06,
"max_abs_err": 0.03515625
},
{
"name": "model.layers.8.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.128035693109268e-06,
"max_abs_err": 0.03515625
},
{
"name": "model.layers.8.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 7.722680493316147e-06,
"max_abs_err": 0.0205078125
},
{
"name": "model.layers.8.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 6.167997980810469e-06,
"max_abs_err": 0.0478515625
},
{
"name": "model.layers.8.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 7.392537554551382e-06,
"max_abs_err": 0.0419921875
},
{
"name": "model.layers.8.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 5.977764885756187e-06,
"max_abs_err": 0.044921875
},
{
"name": "model.layers.8.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 6.091036084399093e-06,
"max_abs_err": 0.125
},
{
"name": "model.layers.9.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 5.781957497674739e-06,
"max_abs_err": 0.0546875
},
{
"name": "model.layers.9.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 5.878169758943841e-06,
"max_abs_err": 0.064453125
},
{
"name": "model.layers.9.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 7.968968930072151e-06,
"max_abs_err": 0.02294921875
},
{
"name": "model.layers.9.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 6.075926648918539e-06,
"max_abs_err": 0.037109375
},
{
"name": "model.layers.9.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 8.653337317809928e-06,
"max_abs_err": 0.05859375
},
{
"name": "model.layers.9.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 5.497732672665734e-06,
"max_abs_err": 0.0390625
},
{
"name": "model.layers.9.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 5.472218163049547e-06,
"max_abs_err": 0.0546875
},
{
"name": "model.layers.10.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.081360425014282e-06,
"max_abs_err": 0.0458984375
},
{
"name": "model.layers.10.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.082946583774174e-06,
"max_abs_err": 0.0400390625
},
{
"name": "model.layers.10.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 7.814529453753494e-06,
"max_abs_err": 0.0234375
},
{
"name": "model.layers.10.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 6.291828867688309e-06,
"max_abs_err": 0.083984375
},
{
"name": "model.layers.10.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 8.264326424978208e-06,
"max_abs_err": 0.07421875
},
{
"name": "model.layers.10.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 5.71836244489532e-06,
"max_abs_err": 0.046875
},
{
"name": "model.layers.10.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 5.754830453952309e-06,
"max_abs_err": 0.0625
},
{
"name": "model.layers.11.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.097926416259725e-06,
"max_abs_err": 0.0380859375
},
{
"name": "model.layers.11.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.3139259509625845e-06,
"max_abs_err": 0.052734375
},
{
"name": "model.layers.11.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 7.98126984591363e-06,
"max_abs_err": 0.02197265625
},
{
"name": "model.layers.11.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 6.15482485954999e-06,
"max_abs_err": 0.044921875
},
{
"name": "model.layers.11.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 7.749045835225843e-06,
"max_abs_err": 0.0625
},
{
"name": "model.layers.11.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.060492978576804e-06,
"max_abs_err": 0.0546875
},
{
"name": "model.layers.11.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 6.0031288739992306e-06,
"max_abs_err": 0.072265625
},
{
"name": "model.layers.12.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.197836682986235e-06,
"max_abs_err": 0.0400390625
},
{
"name": "model.layers.12.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.086477696953807e-06,
"max_abs_err": 0.0625
},
{
"name": "model.layers.12.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 7.839434147172142e-06,
"max_abs_err": 0.0263671875
},
{
"name": "model.layers.12.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 6.057850441720802e-06,
"max_abs_err": 0.091796875
},
{
"name": "model.layers.12.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 7.370619186986005e-06,
"max_abs_err": 0.0625
},
{
"name": "model.layers.12.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.338099865388358e-06,
"max_abs_err": 0.064453125
},
{
"name": "model.layers.12.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 6.154072252684273e-06,
"max_abs_err": 0.10546875
},
{
"name": "model.layers.13.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.168503659864655e-06,
"max_abs_err": 0.0556640625
},
{
"name": "model.layers.13.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.065721208869945e-06,
"max_abs_err": 0.0546875
},
{
"name": "model.layers.13.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 7.47771855458268e-06,
"max_abs_err": 0.021484375
},
{
"name": "model.layers.13.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.778343620477244e-06,
"max_abs_err": 0.052734375
},
{
"name": "model.layers.13.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.893405952723697e-06,
"max_abs_err": 0.076171875
},
{
"name": "model.layers.13.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.5187500695174094e-06,
"max_abs_err": 0.052734375
},
{
"name": "model.layers.13.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 6.2860167417966295e-06,
"max_abs_err": 0.08984375
},
{
"name": "model.layers.14.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.125354957475793e-06,
"max_abs_err": 0.03515625
},
{
"name": "model.layers.14.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.064327862986829e-06,
"max_abs_err": 0.041015625
},
{
"name": "model.layers.14.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 7.123348041204736e-06,
"max_abs_err": 0.01953125
},
{
"name": "model.layers.14.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.626794518320821e-06,
"max_abs_err": 0.080078125
},
{
"name": "model.layers.14.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.6963134486286435e-06,
"max_abs_err": 0.0703125
},
{
"name": "model.layers.14.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.336153091979213e-06,
"max_abs_err": 0.05078125
},
{
"name": "model.layers.14.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 6.141398444015067e-06,
"max_abs_err": 0.09375
},
{
"name": "model.layers.15.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.275447958614677e-06,
"max_abs_err": 0.04296875
},
{
"name": "model.layers.15.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.0338215916999616e-06,
"max_abs_err": 0.044921875
},
{
"name": "model.layers.15.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.959560323593905e-06,
"max_abs_err": 0.0283203125
},
{
"name": "model.layers.15.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.300466909829993e-06,
"max_abs_err": 0.0498046875
},
{
"name": "model.layers.15.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.249875241337577e-06,
"max_abs_err": 0.076171875
},
{
"name": "model.layers.15.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.205583758855937e-06,
"max_abs_err": 0.048828125
},
{
"name": "model.layers.15.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 6.0809124988736585e-06,
"max_abs_err": 0.078125
},
{
"name": "model.layers.16.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 5.911268999625463e-06,
"max_abs_err": 0.041015625
},
{
"name": "model.layers.16.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 5.6489629969291855e-06,
"max_abs_err": 0.0380859375
},
{
"name": "model.layers.16.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.985724212427158e-06,
"max_abs_err": 0.0205078125
},
{
"name": "model.layers.16.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.6090857469826005e-06,
"max_abs_err": 0.07421875
},
{
"name": "model.layers.16.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.553138518938795e-06,
"max_abs_err": 0.080078125
},
{
"name": "model.layers.16.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.124392712081317e-06,
"max_abs_err": 0.12109375
},
{
"name": "model.layers.16.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 5.871052508155117e-06,
"max_abs_err": 0.1953125
},
{
"name": "model.layers.17.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.217421287146863e-06,
"max_abs_err": 0.037109375
},
{
"name": "model.layers.17.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 5.340928055375116e-06,
"max_abs_err": 0.0341796875
},
{
"name": "model.layers.17.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.652290721831378e-06,
"max_abs_err": 0.0283203125
},
{
"name": "model.layers.17.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.430667897599051e-06,
"max_abs_err": 0.068359375
},
{
"name": "model.layers.17.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.375770681188442e-06,
"max_abs_err": 0.1015625
},
{
"name": "model.layers.17.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.1591904341185e-06,
"max_abs_err": 0.04296875
},
{
"name": "model.layers.17.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 5.912334927415941e-06,
"max_abs_err": 0.103515625
},
{
"name": "model.layers.18.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 5.798182428407017e-06,
"max_abs_err": 0.0537109375
},
{
"name": "model.layers.18.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 5.427996256912593e-06,
"max_abs_err": 0.03125
},
{
"name": "model.layers.18.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.89846547174966e-06,
"max_abs_err": 0.0224609375
},
{
"name": "model.layers.18.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.430327291833237e-06,
"max_abs_err": 0.06640625
},
{
"name": "model.layers.18.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.245585609576665e-06,
"max_abs_err": 0.09765625
},
{
"name": "model.layers.18.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.165443210193189e-06,
"max_abs_err": 0.04296875
},
{
"name": "model.layers.18.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 5.9225058066658676e-06,
"max_abs_err": 0.072265625
},
{
"name": "model.layers.19.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.040308562660357e-06,
"max_abs_err": 0.052734375
},
{
"name": "model.layers.19.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 5.1501365305739455e-06,
"max_abs_err": 0.060546875
},
{
"name": "model.layers.19.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.5970416471827775e-06,
"max_abs_err": 0.0224609375
},
{
"name": "model.layers.19.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.532404429686721e-06,
"max_abs_err": 0.068359375
},
{
"name": "model.layers.19.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.173650945129339e-06,
"max_abs_err": 0.08203125
},
{
"name": "model.layers.19.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.213643700903049e-06,
"max_abs_err": 0.048828125
},
{
"name": "model.layers.19.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 5.941985818935791e-06,
"max_abs_err": 0.07421875
},
{
"name": "model.layers.20.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.438817308662692e-06,
"max_abs_err": 0.0419921875
},
{
"name": "model.layers.20.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 5.025427526561543e-06,
"max_abs_err": 0.0390625
},
{
"name": "model.layers.20.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.605811904591974e-06,
"max_abs_err": 0.0234375
},
{
"name": "model.layers.20.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.261394562694477e-06,
"max_abs_err": 0.064453125
},
{
"name": "model.layers.20.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.0325842241581995e-06,
"max_abs_err": 0.0625
},
{
"name": "model.layers.20.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.205663794389693e-06,
"max_abs_err": 0.05078125
},
{
"name": "model.layers.20.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 5.996919753670227e-06,
"max_abs_err": 0.06640625
},
{
"name": "model.layers.21.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.1687410379818175e-06,
"max_abs_err": 0.046875
},
{
"name": "model.layers.21.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 4.971132511855103e-06,
"max_abs_err": 0.029296875
},
{
"name": "model.layers.21.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.4224936977552716e-06,
"max_abs_err": 0.0361328125
},
{
"name": "model.layers.21.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.11908228872926e-06,
"max_abs_err": 0.0341796875
},
{
"name": "model.layers.21.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 5.857799806108233e-06,
"max_abs_err": 0.080078125
},
{
"name": "model.layers.21.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.1662776715820655e-06,
"max_abs_err": 0.044921875
},
{
"name": "model.layers.21.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 6.060807208996266e-06,
"max_abs_err": 0.080078125
},
{
"name": "model.layers.22.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 5.983277333143633e-06,
"max_abs_err": 0.044921875
},
{
"name": "model.layers.22.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 4.734947651741095e-06,
"max_abs_err": 0.10546875
},
{
"name": "model.layers.22.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.43792873233906e-06,
"max_abs_err": 0.03125
},
{
"name": "model.layers.22.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.332039563654689e-06,
"max_abs_err": 0.072265625
},
{
"name": "model.layers.22.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.27352073934162e-06,
"max_abs_err": 0.0576171875
},
{
"name": "model.layers.22.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.463576028181706e-06,
"max_abs_err": 0.06640625
},
{
"name": "model.layers.22.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 6.172966095618904e-06,
"max_abs_err": 0.06640625
},
{
"name": "model.layers.23.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.793758075218648e-06,
"max_abs_err": 0.0458984375
},
{
"name": "model.layers.23.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 5.169235464563826e-06,
"max_abs_err": 0.0380859375
},
{
"name": "model.layers.23.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.7514947659219615e-06,
"max_abs_err": 0.0302734375
},
{
"name": "model.layers.23.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.732099907618249e-06,
"max_abs_err": 0.0498046875
},
{
"name": "model.layers.23.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.715514700772474e-06,
"max_abs_err": 0.0703125
},
{
"name": "model.layers.23.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.481814125436358e-06,
"max_abs_err": 0.041015625
},
{
"name": "model.layers.23.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 6.1963219195604324e-06,
"max_abs_err": 0.0859375
},
{
"name": "model.layers.24.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.9132938733673654e-06,
"max_abs_err": 0.044921875
},
{
"name": "model.layers.24.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 5.1758347581198905e-06,
"max_abs_err": 0.033203125
},
{
"name": "model.layers.24.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.669490630883956e-06,
"max_abs_err": 0.0361328125
},
{
"name": "model.layers.24.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.9178237279411405e-06,
"max_abs_err": 0.0546875
},
{
"name": "model.layers.24.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.960771770536667e-06,
"max_abs_err": 0.0625
},
{
"name": "model.layers.24.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.5153894865943585e-06,
"max_abs_err": 0.04296875
},
{
"name": "model.layers.24.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 6.246365046536084e-06,
"max_abs_err": 0.046875
},
{
"name": "model.layers.25.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.1165278566477355e-06,
"max_abs_err": 0.044921875
},
{
"name": "model.layers.25.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 5.473897999763722e-06,
"max_abs_err": 0.03125
},
{
"name": "model.layers.25.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 7.03422847436741e-06,
"max_abs_err": 0.0263671875
},
{
"name": "model.layers.25.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.731247256335337e-06,
"max_abs_err": 0.029296875
},
{
"name": "model.layers.25.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 7.115002517821267e-06,
"max_abs_err": 0.0625
},
{
"name": "model.layers.25.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.5182052821910474e-06,
"max_abs_err": 0.05078125
},
{
"name": "model.layers.25.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 6.28761745247175e-06,
"max_abs_err": 0.068359375
},
{
"name": "model.layers.26.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.3675847741251346e-06,
"max_abs_err": 0.04296875
},
{
"name": "model.layers.26.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 5.329186933522578e-06,
"max_abs_err": 0.0283203125
},
{
"name": "model.layers.26.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.990022029640386e-06,
"max_abs_err": 0.02783203125
},
{
"name": "model.layers.26.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.997808329993859e-06,
"max_abs_err": 0.0322265625
},
{
"name": "model.layers.26.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 7.119180281733861e-06,
"max_abs_err": 0.05859375
},
{
"name": "model.layers.26.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.656381628999952e-06,
"max_abs_err": 0.064453125
},
{
"name": "model.layers.26.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 6.395952823368134e-06,
"max_abs_err": 0.0703125
},
{
"name": "model.layers.27.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.220388513611397e-06,
"max_abs_err": 0.0390625
},
{
"name": "model.layers.27.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 4.927534973830916e-06,
"max_abs_err": 0.025390625
},
{
"name": "model.layers.27.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.624793059017975e-06,
"max_abs_err": 0.03125
},
{
"name": "model.layers.27.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.735781087423675e-06,
"max_abs_err": 0.0458984375
},
{
"name": "model.layers.27.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 7.053577519400278e-06,
"max_abs_err": 0.068359375
},
{
"name": "model.layers.27.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.8143585849611554e-06,
"max_abs_err": 0.041015625
},
{
"name": "model.layers.27.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 6.551741989824222e-06,
"max_abs_err": 0.06640625
},
{
"name": "model.layers.28.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.4780879256431945e-06,
"max_abs_err": 0.037109375
},
{
"name": "model.layers.28.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 5.259320005279733e-06,
"max_abs_err": 0.02880859375
},
{
"name": "model.layers.28.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 7.258705409185495e-06,
"max_abs_err": 0.0234375
},
{
"name": "model.layers.28.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 5.693392722605495e-06,
"max_abs_err": 0.0419921875
},
{
"name": "model.layers.28.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.909311650815653e-06,
"max_abs_err": 0.0703125
},
{
"name": "model.layers.28.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.96105053066276e-06,
"max_abs_err": 0.060546875
},
{
"name": "model.layers.28.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 6.704045972583117e-06,
"max_abs_err": 0.0625
},
{
"name": "model.layers.29.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 7.15238547854824e-06,
"max_abs_err": 0.0380859375
},
{
"name": "model.layers.29.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 5.073855845694197e-06,
"max_abs_err": 0.037109375
},
{
"name": "model.layers.29.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 6.731678240612382e-06,
"max_abs_err": 0.033203125
},
{
"name": "model.layers.29.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 6.136456249805633e-06,
"max_abs_err": 0.02880859375
},
{
"name": "model.layers.29.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.891063549119281e-06,
"max_abs_err": 0.09765625
},
{
"name": "model.layers.29.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 7.167830062826397e-06,
"max_abs_err": 0.056640625
},
{
"name": "model.layers.29.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 6.8822446337435395e-06,
"max_abs_err": 0.056640625
},
{
"name": "model.layers.30.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.0297402342257556e-06,
"max_abs_err": 0.0458984375
},
{
"name": "model.layers.30.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 5.450096523418324e-06,
"max_abs_err": 0.0419921875
},
{
"name": "model.layers.30.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 7.51342304283753e-06,
"max_abs_err": 0.0322265625
},
{
"name": "model.layers.30.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 6.032732926541939e-06,
"max_abs_err": 0.0400390625
},
{
"name": "model.layers.30.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.815336291765561e-06,
"max_abs_err": 0.087890625
},
{
"name": "model.layers.30.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 7.337210263358429e-06,
"max_abs_err": 0.048828125
},
{
"name": "model.layers.30.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 7.066417310852557e-06,
"max_abs_err": 0.064453125
},
{
"name": "model.layers.31.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.274240149650723e-06,
"max_abs_err": 0.04296875
},
{
"name": "model.layers.31.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 5.056775080447551e-06,
"max_abs_err": 0.0361328125
},
{
"name": "model.layers.31.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 8.097029422060587e-06,
"max_abs_err": 0.02880859375
},
{
"name": "model.layers.31.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 6.395014679583255e-06,
"max_abs_err": 0.046875
},
{
"name": "model.layers.31.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.6728343881550245e-06,
"max_abs_err": 0.103515625
},
{
"name": "model.layers.31.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 7.511306648666505e-06,
"max_abs_err": 0.05859375
},
{
"name": "model.layers.31.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 7.338986506510992e-06,
"max_abs_err": 0.060546875
},
{
"name": "model.layers.32.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.101087819843087e-06,
"max_abs_err": 0.04296875
},
{
"name": "model.layers.32.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 5.10598056280287e-06,
"max_abs_err": 0.037109375
},
{
"name": "model.layers.32.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 8.094936674751807e-06,
"max_abs_err": 0.0322265625
},
{
"name": "model.layers.32.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 6.3699199017719366e-06,
"max_abs_err": 0.048828125
},
{
"name": "model.layers.32.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.488141480076592e-06,
"max_abs_err": 0.0703125
},
{
"name": "model.layers.32.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 7.560893209301867e-06,
"max_abs_err": 0.05078125
},
{
"name": "model.layers.32.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 7.357637059612898e-06,
"max_abs_err": 0.0703125
},
{
"name": "model.layers.33.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 6.1929249568493105e-06,
"max_abs_err": 0.046875
},
{
"name": "model.layers.33.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 4.146252649661619e-06,
"max_abs_err": 0.025390625
},
{
"name": "model.layers.33.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 9.177104402624536e-06,
"max_abs_err": 0.02880859375
},
{
"name": "model.layers.33.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 6.9120242187636904e-06,
"max_abs_err": 0.099609375
},
{
"name": "model.layers.33.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.371019026119029e-06,
"max_abs_err": 0.072265625
},
{
"name": "model.layers.33.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 7.623647888976848e-06,
"max_abs_err": 0.052734375
},
{
"name": "model.layers.33.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 7.210394869616721e-06,
"max_abs_err": 0.07421875
},
{
"name": "model.layers.34.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 5.741615950682899e-06,
"max_abs_err": 0.0390625
},
{
"name": "model.layers.34.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 4.765691755892476e-06,
"max_abs_err": 0.0400390625
},
{
"name": "model.layers.34.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 9.10491871763952e-06,
"max_abs_err": 0.03125
},
{
"name": "model.layers.34.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 6.580262379429769e-06,
"max_abs_err": 0.052734375
},
{
"name": "model.layers.34.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 6.623169156227959e-06,
"max_abs_err": 0.076171875
},
{
"name": "model.layers.34.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 7.4251893238397315e-06,
"max_abs_err": 0.060546875
},
{
"name": "model.layers.34.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 6.5312747210555244e-06,
"max_abs_err": 0.091796875
},
{
"name": "model.layers.35.self_attn.q_proj",
"shape": [
4096,
2560
],
"dtype": "bfloat16",
"mse": 5.816370048705721e-06,
"max_abs_err": 0.046875
},
{
"name": "model.layers.35.self_attn.k_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 5.401536782301264e-06,
"max_abs_err": 0.0341796875
},
{
"name": "model.layers.35.self_attn.v_proj",
"shape": [
1024,
2560
],
"dtype": "bfloat16",
"mse": 8.028151569305919e-06,
"max_abs_err": 0.029296875
},
{
"name": "model.layers.35.self_attn.o_proj",
"shape": [
2560,
4096
],
"dtype": "bfloat16",
"mse": 6.169641892483924e-06,
"max_abs_err": 0.08203125
},
{
"name": "model.layers.35.mlp.gate_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 7.3170535870303866e-06,
"max_abs_err": 0.08203125
},
{
"name": "model.layers.35.mlp.up_proj",
"shape": [
9728,
2560
],
"dtype": "bfloat16",
"mse": 7.652883141417988e-06,
"max_abs_err": 0.08203125
},
{
"name": "model.layers.35.mlp.down_proj",
"shape": [
2560,
9728
],
"dtype": "bfloat16",
"mse": 6.063106866349699e-06,
"max_abs_err": 0.076171875
}
],
"source_model": "Qwen/Qwen3-4B-Thinking-2507",
"dtype": "bf16",
"target_regex": ".*",
"ignore_regex": [],
"include_lm_head": false,
"note": "Dense fake quant: rounded weights are saved as normal HF tensors."
}