INDEX

Explanations

Various text sources

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

oin

-0.31

own

-0.28

pre

-0.27

Fre

-0.27

Round

-0.27

ive

-0.26

fre

-0.26

Fast

-0.26

æ·±

-0.26

POSITIVE LOGITS

 trusts

0.27

ViewById

0.26

ä¼ļè®®ç²¾ç¥ŀ

0.25

otec

0.25

imet

0.24

Ã¤nde

0.24

 Contracts

0.24

 Locator

0.24

<li

0.24

ç¼©

0.24

Activations Density 0.003%

Various text sources

No Known Activations