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

met

-0.37

 totalCount

-0.28

 seperate

-0.26

/met

-0.26

 prÃ³

-0.25

enr

-0.25

(pe

-0.25

 immigr

-0.25

Met

-0.25

vents

-0.24

POSITIVE LOGITS

recht

0.26

åħĭèİ±

0.25

dam

0.25

å¸¦åĽŀ

0.25

çĦķ

0.24

-wife

0.24

tug

0.24

exterity

0.24

ochrome

0.24

pta

0.24

Activations Density 0.016%

No Known Activations

This feature has no known activations.