INDEX

Explanations

training or trained models

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

「

-1.69

-1.54

 unſ

-1.52

但他

-1.46

ada

-1.43

 должен

-1.43

цією

-1.42

«

-1.41

…

-1.40

...

-1.38

POSITIVE LOGITS

 décé

1.59

 acogedor

1.54

 遊戲

1.53

 mußten

1.52

 biß

1.52

 Olímpicos

1.49

 lentejuelas

1.47

鈍

1.43

ưới

1.41

 inolvid

1.41

Activations Density 0.013%