INDEX

Explanations

terms related to data privacy and security

oai_token-act-pair · gpt-4o-mini Triggered by @bot

before suffixes or endings

np_acts-logits-general · gemini-2.0-flash

foreign language fragments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

msgSender

-0.59

ControllerAdvice

-0.56

 Wiktionnaire

-0.56

 AssemblyCompany

-0.55

 médicale

-0.51

 שוליים

-0.50

 partenza

-0.49

 régler

-0.49

AndEndTag

-0.49

 intitulée

-0.48

POSITIVE LOGITS

ny

0.59

 мәгълү

0.58

liche

0.56

ية

0.55

 whor

0.55

gridy

0.54

HomeAsUpEnabled

0.52

}>;

0.52

lich

0.52

 NUKAT

0.51

Activations Density 0.045%

terms related to data privacy and security

before suffixes or endings

foreign language fragments

No Comments

No Known Activations