INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_22.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.22.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

éĢ¸

-0.23

 media

-0.23

mus

-0.23

çĽ¸äº²

-0.23

èĤ¥

-0.22

 pois

-0.22

 releases

-0.22

 underscores

-0.21

lassen

-0.21

ule

-0.21

POSITIVE LOGITS

åıĬçĽ¸åħ³

0.28

 whatsoever

0.26

 weakest

0.25

èĭľ

0.25

ä¸īä½įä¸Ģä½ĵ

0.23

 excellent

0.22

ect

0.22

-fetch

0.22

è¿Ļæĺ¯æĪĳä»¬

0.22

 vnode

0.22

Activations Density 0.000%

No Known Activations

This feature has no known activations.