INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_18.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.18.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

emory

-0.29

å¸¸

-0.27

authority

-0.26

egen

-0.25

èī²

-0.24

dÃŃ

-0.24

éĽĨ

-0.24

vature

-0.24

aupt

-0.24

.toolStrip

-0.24

POSITIVE LOGITS

æĸ¯é¡¿

0.29

å¤Ħåľ¨

0.28

 incumb

0.27

 originals

0.27

åıĳå¸ĥåħ¬åĳĬ

0.26

Ø®Ø¨

0.26

éª¨å¤´

0.25

cob

0.25

jav

0.25

 publicly

0.24

Activations Density 4.736%

No Known Activations

This feature has no known activations.