INDEX

Explanations

appreciated

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 appreciated

-1.35

 appreciate

-1.13

appreciated

-1.06

 appreciates

-1.03

 enjoyed

-1.02

 apprécié

-0.95

 Appreciate

-0.89

 apreci

-0.88

 enjoy

-0.86

Appreciate

-0.83

POSITIVE LOGITS

 Deal

0.52

 deal

0.51

 deals

0.51

the

0.46

 Pole

0.45

 fjer

0.44

 DEAL

0.44

 basically

0.44

toe

0.44

 dealing

0.43

Activations Density 0.025%