INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_30.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.30.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uring

-0.29

URED

-0.29

ä½ıäºĨ

-0.28

usting

-0.26

åĳ±

-0.25

emale

-0.25

:message

-0.25

åķ¥

-0.25

èĮ¹

-0.24

ook

-0.24

POSITIVE LOGITS

affe

0.28

heim

0.27

 Campo

0.26

éľ²

0.25

SPD

0.25

 GLFW

0.25

ijk

0.25

arcy

0.25

 æĶ

0.24

 field

0.24

Activations Density 0.905%

No Known Activations

This feature has no known activations.