INDEX

Explanations

references to America and American identity

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

"American" or "America"

np_acts-logits-general · gemini-2.0-flash

American culture and contexts

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

 Regno

-0.45

 hints

-0.45

 hint

-0.45

knitted

-0.43

piscina

-0.42

guitarra

-0.42

GetAxis

-0.41

 Grazing

-0.40

Zü

-0.40

 Psychiatry

-0.40

POSITIVE LOGITS

america

0.99

America

0.98

 America

0.96

American

0.94

 American

0.93

american

0.91

 america

0.90

 american

0.88

 AMERICA

0.84

 Ameri

0.84

Activations Density 0.077%

references to America and American identity

"American" or "America"

American culture and contexts

No Comments

No Known Activations