INDEX

Explanations

challenges

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_1.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.1.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 neighbour

-0.26

("-

-0.25

éĴ»çŁ³

-0.25

è¯´æĪĳ

-0.25

TLC

-0.25

beg

-0.25

erc

-0.25

ports

-0.25

çĽĽè¡Į

-0.24

[@"

-0.23

POSITIVE LOGITS

aspers

0.35

 tá»¥

0.27

æ°ĳ

0.27

thon

0.26

asher

0.25

tie

0.25

é¦Ĩ

0.25

keiten

0.25

ĥĿ

0.24

ë¯¼

0.24

Activations Density 43.199%

challenges

No Comments

No Known Activations