INDEX

Explanations

words indicating the novice or beginner status in different contexts

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

references to beginner or novice status in learning contexts

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

authority person or patient

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

 gynhyrchwyd

-0.78

 ligiloj

-0.77

InputBorder

-0.74

ագրություններ

-0.72

lapsingToolbar

-0.71

出版年

-0.69

AndEndTag

-0.68

 تضيفلها

-0.67

OGND

-0.67

-0.65

POSITIVE LOGITS

who

1.19

 whom

0.85

 tarafından

0.83

who

0.79

 számára

0.77

에게

0.76

 kteří

0.72

 którzy

0.68

 alike

0.65

 द्वारा

0.65

Activations Density 0.001%

words indicating the novice or beginner status in different contexts

references to beginner or novice status in learning contexts

authority person or patient

No Comments

No Known Activations