INDEX

Explanations

rehabilitation

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_7.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.7.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

ily

-0.29

asin

-0.28

è§Ħç¨ĭ

-0.27

neh

-0.27

itsu

-0.26

allon

-0.26

asy

-0.25

å·±

-0.25

antly

-0.25

(dot

-0.25

POSITIVE LOGITS

former

0.26

éĶĻ

0.26

ÙħØ¬

0.26

 behaving

0.26

å¥³ç¥ŀ

0.25

ses

0.25

anity

0.25

åľĥ

0.25

 former

0.25

anical

0.25

Activations Density 0.002%

rehabilitation

No Comments

No Known Activations