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

Not in Any Lists

No Comments

Negative Logits

á»ģm

-0.26

ople

-0.26

çªĦ

-0.26

atl

-0.25

æľºåħ³

-0.25

embr

-0.24

into

-0.24

ampionship

-0.24

äººç¾¤ä¸Ń

-0.24

_abort

-0.23

POSITIVE LOGITS

 ÑģÐ¾Ñģ

0.25

ç¯

0.23

Accessory

0.23

ibia

0.23

 merit

0.23

 Cond

0.23

 Loft

0.22

çīĽ

0.22

aData

0.22

çĶŁçĮª

0.22

Activations Density 0.002%

No Known Activations

This feature has no known activations.