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

Ex

-0.25

 slashes

-0.25

åħ¨æĹ¥

-0.24

ord

-0.24

Term

-0.23

blings

-0.23

åħħ

-0.22

thr

-0.22

thur

-0.22

æĮ¤

-0.22

POSITIVE LOGITS

æģ©

0.24

lie

0.23

å¸ĤæķĻèĤ²å±Ģ

0.23

æ¾ľ

0.23

 Fist

0.23

’en

0.23

lie

0.22

 Fuse

0.21

åıĪæľī

0.21

apan

0.21

Activations Density 0.017%

No Known Activations

This feature has no known activations.