INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_20.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.20.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aus

-0.28

orning

-0.24

PEED

-0.23

igte

-0.23

 Pett

-0.23

æ´¾

-0.23

imal

-0.23

 hardly

-0.22

peed

-0.22

igm

-0.22

POSITIVE LOGITS

squeeze

0.28

ä½ľèĢħæľ¬äºº

0.27

åĴ§

0.26

din

0.25

éĻį

0.25

ç»¼åĲĪæĢ§

0.25

#%

0.25

 insect

0.25

éĢĨ

0.24

 descon

0.24

Activations Density 19.216%

No Known Activations

This feature has no known activations.