INDEX

Explanations

scored by opponents

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 internationaux

-1.02

 münchen

-0.99

 réelle

-0.95

/\/

-0.94

Gä

-0.92

 yüzde

-0.89

 berguna

-0.88

Pager

-0.87

が終わ

-0.85

 françaises

-0.85

POSITIVE LOGITS

 their

1.44

 themselves

1.26

 GOALS

1.24

 HAPPENED

1.05

：……

1.04

 TÉCNICA

1.04

 PLAINT

1.01

 Their

1.00

ayakan

0.99

 بوابة

0.99

Activations Density 0.006%