INDEX

Explanations

references to the United States

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pography

-0.54

fromnode

-0.53

abstractmethod

-0.52

 CreateTagHelper

-0.51

 carinho

-0.50

 desmotivaciones

-0.49

AnchorTagHelper

-0.48

bcryptjs

-0.48

MessageTagHelper

-0.48

pointment

-0.48

POSITIVE LOGITS

US

0.99

US

0.89

USA

0.80

🇺🇸

0.75

 American

0.66

 США

0.66

USA

0.66

American

0.65

 States

0.63

 America

0.63

Activations Density 0.048%

references to the United States

No Comments

No Known Activations