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

Negative Logits

VENT

-0.27

 Ahead

-0.26

ä¹ŁåºĶè¯¥

-0.26

à¸§à¸¥

-0.25

 microbi

-0.25

 INCIDENT

-0.24

ulk

-0.24

.font

-0.24

åºĶå½ĵ

-0.23

SPA

-0.23

POSITIVE LOGITS

ocol

0.35

æĭ¾

0.29

{name

0.25

 immortal

0.25

cess

0.25

 resumes

0.24

 particular

0.24

åħ¬å¸ĥäºĨ

0.24

Bear

0.24

.EXIT

0.23

Activations Density 0.005%

No Known Activations

This feature has no known activations.