INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_18.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.18.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

å¯¹æĪĳæĿ¥è¯´

-0.29

à¸£à¸Ńà¸¢

-0.27

åĲ¾

-0.27

å¹¶æĹł

-0.27

heets

-0.26

å²¿

-0.25

ainers

-0.25

.sheet

-0.25

sheet

-0.25

lift

-0.24

POSITIVE LOGITS

ç´ł

0.27

bak

0.27

 reversal

0.27

spo

0.25

-HT

0.24

ingu

0.24

soc

0.24

oca

0.24

è°ĥåº¦

0.23

æİĴåºı

0.23

Activations Density 1.127%

No Known Activations

This feature has no known activations.