INDEX

Explanations

commas

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

 répondit

-1.13

 afferma

-1.04

noted

-1.04

 fermés

-0.98

 sauvages

-0.98

 scolaires

-0.98

 noted

-0.98

 mondta

-0.98

 katanya

-0.96

 avoient

-0.96

POSITIVE LOGITS

0.36

 look

0.33

 them

0.32

ан

0.30

 couple

0.30

els

0.26

le

0.25

 turn

0.24

 suppose

0.24

reg

0.24

Activations Density 0.002%