INDEX

Explanations

conditions, exceptions, clarifications

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

gilet

-1.29

pantalon

-1.25

embang

-1.09

mayacak

-1.08

 fortfarande

-1.08

 kabát

-1.06

washingtonpost

-1.05

 cuillère

-1.04

 progressivement

-1.03

pantal

-1.02

POSITIVE LOGITS

 cualquier

1.22

any

1.19

enciales

1.13

偶尔

1.12

 السياس

1.11

any

1.09

と思いますが

1.08

each

1.07

 qualquer

1.05

meet

1.03

Activations Density 0.023%