INDEX

Explanations

references to "American" and its associations throughout the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

XmlAccessorType

-0.79

rhosis

-0.77

jooq

-0.74

IContainer

-0.73

новниш

-0.72

conazole

-0.71

SourceChecksum

-0.70

 Blom

-0.70

ExtendWith

-0.69

Tikang

-0.69

POSITIVE LOGITS

American

1.88

 America

1.87

 American

1.86

America

1.73

 Americans

1.67

 AMERICAN

1.62

 AMERICA

1.57

AMERICAN

1.55

 american

1.52

 america

1.50

Activations Density 0.148%

references to "American" and its associations throughout the text

No Comments

No Known Activations