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

æĿĵ

-0.28

erged

-0.27

æĻ¨

-0.26

æĢ»äº§åĢ¼

-0.26

 andere

-0.26

çĽ¹

-0.25

'autres

-0.24

 other

-0.24

åħ¶ä»ĸ

-0.24

æĪĸåħ¶ä»ĸ

-0.24

POSITIVE LOGITS

asts

0.29

ables

0.28

 competit

0.26

ÑĢÐ°Ð¹

0.26

pane

0.24

iates

0.24

 margins

0.24

 Paste

0.24

amble

0.24

eties

0.23

Activations Density 0.013%

No Known Activations

This feature has no known activations.