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

Negative Logits

æģ¤

-0.28

duÄŁ

-0.27

 remot

-0.26

bohydr

-0.26

ç³Ĭ

-0.26

-worthy

-0.25

cal

-0.24

è¿ŀçº¿

-0.24

æĿĲæĸĻ

-0.24

-end

-0.23

POSITIVE LOGITS

 nonexistent

0.28

ÑĩÐ¸ÑĤÑĭÐ²Ð°

0.27

Unavailable

0.26

usi

0.25

-Identifier

0.24

 unbe

0.24

Ð¶Ð¸

0.24

Ð½Ð°ÑĤ

0.24

urge

0.24

ÐĽÐĺ

0.24

Activations Density 0.197%

No Known Activations

This feature has no known activations.