INDEX

Explanations

the

np_max-act-logits · gemini-2.0-flash

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

Negative Logits

å°¾

-0.27

Mud

-0.27

ampo

-0.26

orce

-0.26

orno

-0.25

-meta

-0.25

OTS

-0.24

-power

-0.24

 tail

-0.24

 propTypes

-0.24

POSITIVE LOGITS

åĴĮç¤¾ä¼ļ

0.28

åįĩ

0.27

Chr

0.26

ony

0.25

æ³ķå¸Ī

0.25

 exchanges

0.25

èĲ½

0.25

CHR

0.24

 descend

0.24

Prof

0.23

Activations Density 0.003%

the

No Known Activations