SAE Visualizer

Index

EXPLANATIONS

Gender => concepts

INFERENCE

IFRAME INTEGRATION

Direct URL

Iframe Code

<iframe src="http://localhost:3000/embed/dictionaries/qwen3-1.7b-lorsa-8x-topk64-layer0/features/330" width="100%" height="800" frameborder="0"></iframe>

NEGATIVE LOGITS

efs-13.625

不到-12.875

cox-12.813

oze-12.313

charging-12.250

复合-12.063

仰-12.000

inge-11.813

散-11.750

egov-11.625

POSITIVE LOGITS

rics18.500

_Mund18.000

纶17.125

_Means16.875

STRU16.750

_Mercer16.625

_Fuller16.250

捅16.125

mun15.938

uelle15.875

ACTIVATION TIMES 0.0702%

Activation histogram is not available.

Logits histogram is not available.

TOPACTIVATIONS

SUBSAMPLING80%

SUBSAMPLING60%

SUBSAMPLING40%

NONACTIVATING