INDEX

Explanations

elements related to code syntax and structure

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

technical jargon or abbreviations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

enumi

-0.48

Clik

-0.43

ok

-0.42

DataMember

-0.39

 exchanges

-0.38

 dulces

-0.38

 eşofman

-0.38

 kita

-0.37

 paramInt

-0.37

raad

-0.36

POSITIVE LOGITS

 autorytatywna

0.83

 kasarigan

0.73

SourceChecksum

0.67

GEBURTSDATUM

0.66

 disambiguazione

0.66

 Италијани

0.66

Personensuche

0.61

 ModelExpression

0.59

 Taktlose

0.59

 Administrativna

0.56

Activations Density 0.007%

elements related to code syntax and structure

technical jargon or abbreviations

No Comments

No Known Activations