INDEX

Explanations

mentions of "America" and "American" in various contexts

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

American and its 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-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

 directiva

-0.44

 InputDecoration

-0.43

 Konflikt

-0.42

arschijnlijk

-0.42

 zanahoria

-0.41

piscina

-0.41

oriasis

-0.40

 kated

-0.40

rzez

-0.40

guitarra

-0.40

POSITIVE LOGITS

 America

1.17

America

1.15

 American

1.10

American

1.09

america

1.06

 america

1.04

 american

0.98

 AMERICA

0.98

Amer

0.96

american

0.94

Activations Density 0.021%

mentions of "America" and "American" in various contexts

American and its contexts

No Comments

No Known Activations