INDEX

Explanations

non-roman characters and numbers in a varied format

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

after abbreviations or symbols

np_acts-logits-general · gemini-2.0-flash

Altyn Asyr FK

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

PhysRev

-0.69

 useStyles

-0.56

industan

-0.56

templateUrl

-0.54

 Varian

-0.52

globalData

-0.52

 Wiltshire

-0.50

üğü

-0.50

 Inscrivez

-0.49

Witam

-0.49

POSITIVE LOGITS

 саны

0.77

TagMode

0.66

 vidare

0.58

 videre

0.55

gely

0.54

 мәкал

0.52

 тү

0.52

SerializedSize

0.52

 datter

0.52

Personendaten

0.51

Activations Density 0.007%

non-roman characters and numbers in a varied format

after abbreviations or symbols

Altyn Asyr FK

No Known Activations