FP32 mapped graph
authors: cosine=1.000000000 clipped=0
siglip: cosine=1.000000000 clipped=0
caffeine: cosine=1.000000000 clipped=0
robosign: cosine=1.000000000 clipped=0
fried_fish: cosine=1.000000000 clipped=0
cow_beach2: cosine=1.000000000 clipped=0
Tensor scales, nearest weights
authors: cosine=0.996425001 clipped=1
siglip: cosine=0.997553611 clipped=0
caffeine: cosine=0.997727986 clipped=1
robosign: cosine=0.994956063 clipped=8
fried_fish: cosine=0.997094232 clipped=52
cow_beach2: cosine=0.997382946 clipped=100
GPTQ encoder.layer0.attention.qkv: 0.88830 weighted error ratio
GPTQ encoder.layer0.attention.output: 0.18091 weighted error ratio
GPTQ encoder.layer0.mlp.up: 0.86183 weighted error ratio
GPTQ encoder.layer0.mlp.down: 0.79659 weighted error ratio
GPTQ encoder.layer1.attention.qkv: 0.79212 weighted error ratio
GPTQ encoder.layer1.attention.output: 0.15878 weighted error ratio
GPTQ encoder.layer1.mlp.up: 0.78058 weighted error ratio
GPTQ encoder.layer1.mlp.down: 0.60102 weighted error ratio
GPTQ encoder.layer2.attention.qkv: 0.87424 weighted error ratio
GPTQ encoder.layer2.attention.output: 0.41378 weighted error ratio
GPTQ encoder.layer2.mlp.up: 0.84807 weighted error ratio
GPTQ encoder.layer2.mlp.down: 0.59522 weighted error ratio
GPTQ encoder.layer3.attention.qkv: 0.87974 weighted error ratio
GPTQ encoder.layer3.attention.output: 0.51079 weighted error ratio
GPTQ encoder.layer3.mlp.up: 0.81089 weighted error ratio
GPTQ encoder.layer3.mlp.down: 0.56302 weighted error ratio
GPTQ encoder.layer4.attention.qkv: 0.83697 weighted error ratio
GPTQ encoder.layer4.attention.output: 0.46677 weighted error ratio
GPTQ encoder.layer4.mlp.up: 0.76460 weighted error ratio
GPTQ encoder.layer4.mlp.down: 0.57484 weighted error ratio
GPTQ encoder.layer5.attention.qkv: 0.82711 weighted error ratio
GPTQ encoder.layer5.attention.output: 0.48863 weighted error ratio
GPTQ encoder.layer5.mlp.up: 0.76627 weighted error ratio
GPTQ encoder.layer5.mlp.down: 0.58455 weighted error ratio
GPTQ encoder.layer6.attention.qkv: 0.80352 weighted error ratio
GPTQ encoder.layer6.attention.output: 0.44570 weighted error ratio
GPTQ encoder.layer6.mlp.up: 0.75380 weighted error ratio
GPTQ encoder.layer6.mlp.down: 0.64796 weighted error ratio
GPTQ encoder.layer7.attention.qkv: 0.77599 weighted error ratio
GPTQ encoder.layer7.attention.output: 0.45982 weighted error ratio
GPTQ encoder.layer7.mlp.up: 0.75025 weighted error ratio
GPTQ encoder.layer7.mlp.down: 0.57839 weighted error ratio
GPTQ encoder.layer8.attention.qkv: 0.76409 weighted error ratio
GPTQ encoder.layer8.attention.output: 0.42413 weighted error ratio
GPTQ encoder.layer8.mlp.up: 0.75367 weighted error ratio
GPTQ encoder.layer8.mlp.down: 0.65282 weighted error ratio
GPTQ encoder.layer9.attention.qkv: 0.74620 weighted error ratio
GPTQ encoder.layer9.attention.output: 0.43574 weighted error ratio
GPTQ encoder.layer9.mlp.up: 0.77500 weighted error ratio
GPTQ encoder.layer9.mlp.down: 0.65482 weighted error ratio
GPTQ encoder.layer10.attention.qkv: 0.75844 weighted error ratio
GPTQ encoder.layer10.attention.output: 0.44491 weighted error ratio
GPTQ encoder.layer10.mlp.up: 0.76218 weighted error ratio
GPTQ encoder.layer10.mlp.down: 0.65682 weighted error ratio
GPTQ encoder.layer11.attention.qkv: 0.73850 weighted error ratio
GPTQ encoder.layer11.attention.output: 0.37618 weighted error ratio
GPTQ encoder.layer11.mlp.up: 0.74979 weighted error ratio
GPTQ encoder.layer11.mlp.down: 0.65363 weighted error ratio
GPTQ encoder.layer12.attention.qkv: 0.75529 weighted error ratio
GPTQ encoder.layer12.attention.output: 0.44552 weighted error ratio
GPTQ encoder.layer12.mlp.up: 0.72018 weighted error ratio
GPTQ encoder.layer12.mlp.down: 0.63100 weighted error ratio
GPTQ encoder.layer13.attention.qkv: 0.74532 weighted error ratio
GPTQ encoder.layer13.attention.output: 0.43892 weighted error ratio
GPTQ encoder.layer13.mlp.up: 0.71420 weighted error ratio
GPTQ encoder.layer13.mlp.down: 0.64946 weighted error ratio
GPTQ encoder.layer14.attention.qkv: 0.75193 weighted error ratio
GPTQ encoder.layer14.attention.output: 0.39102 weighted error ratio
GPTQ encoder.layer14.mlp.up: 0.71465 weighted error ratio
GPTQ encoder.layer14.mlp.down: 0.65897 weighted error ratio
GPTQ encoder.layer15.attention.qkv: 0.74435 weighted error ratio
GPTQ encoder.layer15.attention.output: 0.33622 weighted error ratio
GPTQ encoder.layer15.mlp.up: 0.60348 weighted error ratio
GPTQ encoder.layer15.mlp.down: 0.73127 weighted error ratio
GPTQ encoder.layer16.attention.qkv: 0.72789 weighted error ratio
GPTQ encoder.layer16.attention.output: 0.31962 weighted error ratio
GPTQ encoder.layer16.mlp.up: 0.44801 weighted error ratio
GPTQ encoder.layer16.mlp.down: 0.86710 weighted error ratio
GPTQ encoder.layer17.attention.qkv: 0.72428 weighted error ratio
GPTQ encoder.layer17.attention.output: 0.22745 weighted error ratio
GPTQ encoder.layer17.mlp.up: 0.34725 weighted error ratio
GPTQ encoder.layer17.mlp.down: 0.87600 weighted error ratio
GPTQ encoder.layer18.attention.qkv: 0.70456 weighted error ratio
GPTQ encoder.layer18.attention.output: 0.22486 weighted error ratio
GPTQ encoder.layer18.mlp.up: 0.37426 weighted error ratio
GPTQ encoder.layer18.mlp.down: 0.88315 weighted error ratio
GPTQ encoder.layer19.attention.qkv: 0.70328 weighted error ratio
GPTQ encoder.layer19.attention.output: 0.22913 weighted error ratio
GPTQ encoder.layer19.mlp.up: 0.31254 weighted error ratio
GPTQ encoder.layer19.mlp.down: 0.92805 weighted error ratio
GPTQ encoder.layer20.attention.qkv: 0.68895 weighted error ratio
GPTQ encoder.layer20.attention.output: 0.20516 weighted error ratio
GPTQ encoder.layer20.mlp.up: 0.28702 weighted error ratio
GPTQ encoder.layer20.mlp.down: 0.96194 weighted error ratio
GPTQ encoder.layer21.attention.qkv: 0.68356 weighted error ratio
GPTQ encoder.layer21.attention.output: 0.19830 weighted error ratio
GPTQ encoder.layer21.mlp.up: 0.26509 weighted error ratio
GPTQ encoder.layer21.mlp.down: 0.97674 weighted error ratio
GPTQ encoder.layer22.attention.qkv: 0.66627 weighted error ratio
GPTQ encoder.layer22.attention.output: 0.19612 weighted error ratio
GPTQ encoder.layer22.mlp.up: 0.26505 weighted error ratio
GPTQ encoder.layer22.mlp.down: 0.97545 weighted error ratio
GPTQ encoder.layer23.attention.qkv: 0.64709 weighted error ratio
GPTQ encoder.layer23.attention.output: 0.20789 weighted error ratio
GPTQ encoder.layer23.mlp.up: 0.26377 weighted error ratio
GPTQ encoder.layer23.mlp.down: 0.96558 weighted error ratio
GPTQ encoder.layer24.attention.qkv: 0.63514 weighted error ratio
GPTQ encoder.layer24.attention.output: 0.16593 weighted error ratio
GPTQ encoder.layer24.mlp.up: 0.29253 weighted error ratio
GPTQ encoder.layer24.mlp.down: 0.97196 weighted error ratio
GPTQ encoder.layer25.attention.qkv: 0.60877 weighted error ratio
GPTQ encoder.layer25.attention.output: 0.14492 weighted error ratio
GPTQ encoder.layer25.mlp.up: 0.36286 weighted error ratio
GPTQ encoder.layer25.mlp.down: 0.91258 weighted error ratio
GPTQ encoder.layer26.attention.qkv: 0.59795 weighted error ratio
GPTQ encoder.layer26.attention.output: 0.11493 weighted error ratio
GPTQ encoder.layer26.mlp.up: 0.52423 weighted error ratio
GPTQ encoder.layer26.mlp.down: 0.83664 weighted error ratio
GPTQ map.attention.query: 0.02027 weighted error ratio
GPTQ map.attention.kv: 0.50968 weighted error ratio
GPTQ map.attention.output: 0.02823 weighted error ratio
GPTQ map.mlp.up: 0.02461 weighted error ratio
GPTQ map.mlp.down: 0.05305 weighted error ratio
Tensor scales, GPTQ weights and bias correction
authors: cosine=0.995570808 clipped=0
siglip: cosine=0.998426036 clipped=1
caffeine: cosine=0.998132672 clipped=1
robosign: cosine=0.996655598 clipped=13
fried_fish: cosine=0.996907833 clipped=54
cow_beach2: cosine=0.997832857 clipped=103