INDEX

Explanations

X

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_19.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.19.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 hood

-0.30

WER

-0.30

å»ĵ

-0.29

erÃŃa

-0.26

mek

-0.26

 Mash

-0.25

wij

-0.24

åĽŁ

-0.24

_ds

-0.24

 Ward

-0.24

POSITIVE LOGITS

ä¾§éĿ¢

0.28

ayers

0.28

 agony

0.27

ä¾§

0.27

ippers

0.26

gtest

0.25

éĴ§

0.25

 sides

0.24

ryptography

0.24

/*@

0.23

Activations Density 0.376%

X

No Comments

No Known Activations