INDEX

Explanations

Scientific/technical writing

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

eters

-0.29

ilo

-0.28

-0.26

[T

-0.25

obe

-0.25

ÅĻ

-0.24

éĿ¢åĲĳ

-0.24

éŃı

-0.24

kay

-0.24

uvw

-0.24

POSITIVE LOGITS

allon

0.28

Lah

0.27

éĢ¢

0.25

Dun

0.25

æĽ´èĥ½

0.25

å°Ĭ

0.24

ItemImage

0.24

 hust

0.24

alia

0.24

imed

0.24

Activations Density 0.014%

Scientific/technical writing

No Known Activations