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

 colle

-0.28

 kidding

-0.27

æ³Ħéľ²

-0.26

canf

-0.26

æ¸Ĺ

-0.24

cheng

-0.24

 rightfully

-0.23

åħļ

-0.23

un

-0.23

Had

-0.23

POSITIVE LOGITS

åĵĳ

0.29

éĤ°

0.29

 onboard

0.27

å§ĵ

0.25

-vs

0.24

æĬĢæľ¯åĴĮ

0.24

Mus

0.24

{}\

0.24

-feature

0.24

::<

0.23

Activations Density 0.000%

No Known Activations

This feature has no known activations.