INDEX

Explanations

keywords related to political events and elections

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-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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 assinatura

-0.53

 berdua

-0.53

这件事情

-0.52

 Komunikasi

-0.50

 heureuse

-0.49

 Kriege

-0.49

 maravilloso

-0.49

 maravillosa

-0.48

GEBURTS

-0.48

 defaultstate

-0.47

POSITIVE LOGITS

 watchdog

0.43

 rival

0.41

 ANGELES

0.40

 authorities

0.38

 Britain

0.38

 antis

0.38

Britain

0.38

 officials

0.38

featureID

0.37

 lawmakers

0.37

Activations Density 0.265%

keywords related to political events and elections

No Comments

No Known Activations