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

omon

-0.28

_ub

-0.27

Ð½ÐµÑĢ

-0.27

ãģ¾ãģĦ

-0.26

ä¸Ģäºĭ

-0.26

ä¼ªè£ħ

-0.26

SIN

-0.25

beer

-0.25

éĢīåĩº

-0.24

_then

-0.23

POSITIVE LOGITS

dcc

0.29

è½®

0.27

dfa

0.25

_tokenize

0.25

iap

0.24

KHR

0.24

EEK

0.23

æĻļå¹´

0.23

.untracked

0.23

kÃ¼

0.23

Activations Density 0.002%

No Known Activations

This feature has no known activations.