SAE Visualizer

Index

EXPLANATIONS

space => space

INFERENCE

IFRAME INTEGRATION

Direct URL

Iframe Code

<iframe src="http://localhost:3000/embed/dictionaries/qwen3-1.7b-lorsa-8x-topk64-layer0/features/319" width="100%" height="800" frameborder="0"></iframe>

NEGATIVE LOGITS

态-14.063

泪-13.688

髡-13.375

ecd-12.750

_arb-12.563

小康社会-12.375

ormap-12.313

修-12.313

_signals-12.250

agh-12.063

POSITIVE LOGITS

_IMS16.250

ERO16.125

>:</15.250

ENDER15.000

_Monument14.875

팅14.500

悬挂14.375

errer14.375

更要14.375

imps14.313

ACTIVATION TIMES 0.0923%

Activation histogram is not available.

Logits histogram is not available.

TOPACTIVATIONS

SUBSAMPLING80%

SUBSAMPLING60%

SUBSAMPLING40%

NONACTIVATING