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

Not in Any Lists

No Comments

Negative Logits

åħ¬æĬ¥

-0.28

 Setter

-0.27

èĤĮèĤī

-0.26

åĲįç§°

-0.25

æ¬¾

-0.25

Ã³r

-0.23

ç¬¦åı·

-0.23

å¹³åĿ¦

-0.23

_letters

-0.23

setter

-0.23

POSITIVE LOGITS

 incarceration

0.25

agal

0.25

 stay

0.25

 meal

0.24

æĹ¶åĢĻ

0.23

æľªæĪĲ

0.23

enos

0.22

gou

0.22

.substr

0.22

ç¼

0.22

Activations Density 0.001%

No Known Activations

This feature has no known activations.