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

åıªè¦ģ

-0.26

agan

-0.25

Aggregate

-0.25

 MÃ¡

-0.25

udes

-0.25

Tol

-0.25

 ÑģÐ¾Ð²

-0.25

tek

-0.24

ku

-0.24

Nova

-0.24

POSITIVE LOGITS

_RATE

0.27

 circumcision

0.26

mailto

0.26

 marked

0.25

æ¡¥æ¢ģ

0.24

UPI

0.24

 Gundam

0.24

 bridges

0.24

ponsive

0.24

earch

0.24

Activations Density 0.034%

No Known Activations

This feature has no known activations.