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

çł´

-0.26

drm

-0.25

ç§°

-0.25

 Taste

-0.25

å£³

-0.25

akening

-0.24

acades

-0.24

tÃ©

-0.24

éĻĨç»Ń

-0.24

å°¤åħ¶

-0.24

POSITIVE LOGITS

inflate

0.28

apyrus

0.27

 infiltration

0.25

TypeID

0.25

 Lund

0.25

odate

0.24

 controller

0.23

èĭĮ

0.23

pike

0.23

 Bands

0.23

Activations Density 0.045%

No Known Activations

This feature has no known activations.