INDEX

Explanations

danger

np_max-act · gemini-2.0-flash

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

Not in Any Lists

No Comments

Negative Logits

eam

-0.30

ters

-0.29

uates

-0.28

CAPE

-0.26

lename

-0.26

_stdio

-0.26

æ®ļ

-0.25

çĤİ

-0.25

McM

-0.25

 reopening

-0.25

POSITIVE LOGITS

ously

0.29

forth

0.27

frame

0.26

 competing

0.25

è¢Ń

0.25

 mutually

0.24

/star

0.24

 parallel

0.24

Dual

0.23

 Household

0.23

Activations Density 0.101%

danger

No Comments

No Known Activations