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

 domin

-0.28

bin

-0.27

bin

-0.27

æįĲåĬ©

-0.26

ESA

-0.26

çį¾

-0.26

 simult

-0.25

 days

-0.25

 decided

-0.24

BIN

-0.24

POSITIVE LOGITS

çīĻ

0.28

ä¼ģä¸ļæıĲä¾Ľ

0.27

mlin

0.27

 Ð»Ð¸Ð´ÐµÑĢ

0.26

inel

0.26

æ¢ģ

0.24

Law

0.24

Law

0.24

hta

0.23

ixo

0.23

Activations Density 46.364%

No Known Activations

This feature has no known activations.