INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_27.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.27.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

%!

-0.28

ocratic

-0.27

 Ð¦

-0.26

 conceivable

-0.26

hands

-0.25

é¢ľ

-0.24

 Sector

-0.24

æ´ª

-0.24

è¿ĻäºĽä¸ľè¥¿

-0.24

æľīèī²

-0.24

POSITIVE LOGITS

.shtml

0.28

 maximum

0.27

 beyond

0.27

egin

0.26

imum

0.26

maximum

0.26

oris

0.25

tur

0.25

è¶ħè¶Ĭ

0.25

ç»¥

0.24

Activations Density 0.987%

No Known Activations

This feature has no known activations.