INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_22.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.22.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

DrawerToggle

-0.25

trl

-0.25

 ”↵↵

-0.25

gon

-0.24

çªģå¦Ĥåħ¶

-0.24

laden

-0.24

ç²½åŃĲ

-0.24

Peter

-0.23

ä¿

-0.23

 Affero

-0.23

POSITIVE LOGITS

Bit

0.28

ych

0.27

 Fact

0.26

çĦī

0.25

ritel

0.24

 Fine

0.24

-required

0.24

par

0.24

dy

0.24

bl

0.24

Activations Density 0.820%

No Known Activations

This feature has no known activations.