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

å°±æ²¡æľī

-0.30

å°±ä¸į

-0.28

-oper

-0.26

çľŁåģĩ

-0.25

å°±æľīäºĨ

-0.25

 LATIN

-0.25

ä½łå°±

-0.25

imar

-0.25

éĻ·åħ¥äºĨ

-0.24

.gc

-0.24

POSITIVE LOGITS

çªĳ

0.28

af

0.27

extends

0.25

BSD

0.25

 adopt

0.25

 consistent

0.25

appe

0.24

 instead

0.24

safe

0.24

//===

0.24

Activations Density 0.002%

No Known Activations

This feature has no known activations.