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

ancements

-0.26

 cart

-0.26

cat

-0.26

zin

-0.26

è¡¥æ°´

-0.25

umn

-0.25

graded

-0.25

Ð»ÑĮÐ½

-0.25

atum

-0.24

respons

-0.24

POSITIVE LOGITS

æľĢå¤ļ

0.29

åĨ²

0.26

 longest

0.25

-ground

0.25

ICAST

0.24

 Ð¾Ð±Ñĭ

0.24

ENSOR

0.24

ISC

0.24

åĮį

0.24

 silence

0.24

Activations Density 29.026%

No Known Activations

This feature has no known activations.