INDEX

Explanations

code-related structures and syntax elements

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

code keywords and structures

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

 betweenstory

-0.45

MemberId

-0.43

 arşivlendi

-0.41

verwijspagina

-0.40

 Bands

-0.39

entious

-0.39

 commerciales

-0.38

oredCriteria

-0.38

irvana

-0.38

 convierten

-0.38

POSITIVE LOGITS

AnimationsModule

0.63

ніципалі

0.54

 useRouter

0.53

FieldNumber

0.48

styleable

0.44

OptionsMenu

0.43

yarnpkg

0.43

initComponents

0.42

${

0.42

 lateinit

0.41

Activations Density 0.297%

code-related structures and syntax elements

code keywords and structures

No Comments

No Known Activations