INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Moderator

-0.27

ç²

-0.25

çļĦåĵģçīĮ

-0.25

GK

-0.25

å²Ĳ

-0.25

æĿµ

-0.24

esc

-0.24

\admin

-0.24

 inset

-0.24

esc

-0.23

POSITIVE LOGITS

then

0.26

favor

0.26

 missions

0.26

kah

0.25

 charges

0.24

 yourselves

0.23

 flash

0.23

etas

0.23

seg

0.23

iforn

0.23

Activations Density 0.001%

No Known Activations

This feature has no known activations.