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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 seamless

-0.26

 phys

-0.25

 miss

-0.25

 those

-0.25

 laws

-0.25

 clearly

-0.24

é¼İ

-0.24

 torn

-0.24

aba

-0.24

 lost

-0.24

POSITIVE LOGITS

emarks

0.28

à¸ģà¸£à¸£à¸¡

0.27

åĭĻ

0.27

ynes

0.26

ovolta

0.25

-fat

0.25

Unload

0.25

 Porno

0.25

çº¢æĹĹ

0.24

 Dion

0.24

Activations Density 0.004%

No Known Activations

This feature has no known activations.