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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

æ¾Ī

-0.27

DUCT

-0.25

COPE

-0.25

éĻĲæľŁ

-0.25

èĢĲå¿ĥ

-0.24

ç«ĭéĿ¢

-0.24

æĪĲäº¤

-0.24

stant

-0.23

ducted

-0.23

åĪĨè§£

-0.23

POSITIVE LOGITS

æĿŀ

0.28

.angular

0.27

 sing

0.26

 positivity

0.26

cers

0.26

appen

0.25

ableView

0.25

 Bien

0.24

 orphan

0.24

iph

0.24

Activations Density 0.040%

No Known Activations

This feature has no known activations.