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

arness

-0.28

empt

-0.25

 mastering

-0.24

 increments

-0.24

loon

-0.24

 Tier

-0.24

vag

-0.24

 scrolling

-0.24

 mastery

-0.23

 moderation

-0.23

POSITIVE LOGITS

 ApplicationContext

0.25

 armies

0.25

getStore

0.25

olie

0.25

bury

0.24

%C

0.24

igli

0.24

ien

0.24

å·¥ä¸ļåĮĸ

0.23

iene

0.23

Activations Density 0.002%

No Known Activations

This feature has no known activations.