INDEX

Explanations

Licenses, injury prevention, code

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

è®ºè¯ģ

-0.29

å¾Ĳ

-0.26

èĢĥ

-0.25

 thrott

-0.25

åİ»ä¹°

-0.25

ç»ĨåĪĨ

-0.25

erox

-0.24

 Hamp

-0.24

å¸Ĥåľºä¸Ĭ

-0.24

 hire

-0.24

POSITIVE LOGITS

åĳ¨å²ģ

0.33

åľ°åĿª

0.30

emplo

0.28

avings

0.28

 potentially

0.27

grams

0.26

renched

0.26

çģ¯åħ·

0.26

 drones

0.26

éĩįå¿ĥ

0.26

Activations Density 0.002%

Licenses, injury prevention, code

No Comments

No Known Activations