INDEX

Explanations

references to ethnic or national identities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_10/width_16k/average_l0_113

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

 graduación

-0.44

 Græ

-0.43

lindung

-0.43

лтамалар

-0.43

</tfoot>

-0.41

 Auflage

-0.40

 Anleitung

-0.39

 milagro

-0.39

 tén

-0.38

 linha

-0.38

POSITIVE LOGITS

 Americans

0.67

 Indians

0.62

外国人

0.60

 foreigners

0.59

Jereo

0.58

 Australians

0.57

therners

0.57

中国人

0.57

 countrymen

0.56

 Canadians

0.56

Activations Density 0.576%

references to ethnic or national identities

No Comments

No Known Activations