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

 increasing

-0.30

å¢ŀåĬł

-0.27

cie

-0.26

oint

-0.26

Im

-0.25

ä»ĸå¯¹

-0.25

 Increasing

-0.25

åįĹçľģ

-0.25

aus

-0.25

werk

-0.25

POSITIVE LOGITS

åĴĮå¹³

0.28

 motivating

0.27

 hydration

0.26

 motivation

0.24

-muted

0.24

LERÄ°

0.24

 dehydration

0.23

 fotos

0.23

çİ·

0.23

Fot

0.23

Activations Density 0.005%

No Known Activations

This feature has no known activations.