INDEX

Explanations

common separators and code snippets

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

 évek

-1.43

Gdzie

-1.41

 světa

-1.40

Może

-1.39

 Meksika

-1.32

Warto

-1.30

Jeśli

-1.29

Wię

-1.29

źródło

-1.25

Leírás

-1.24

POSITIVE LOGITS

 эти

1.22

 however

1.14

↵↵

1.14

 этих

1.13

1.12

to

1.11

1.10

it

1.06

1.05

 estos

1.03

Activations Density 0.001%