INDEX

Explanations

and words that continue a list

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

 ու

-1.20

 decorativos

-1.01

 sowie

-1.01

に於

-0.99

Parabéns

-0.97

-0.96

よくある

-0.94

 getDescription

-0.94

什么意思

-0.92

百貨

-0.92

POSITIVE LOGITS

and

3.63

And

1.68

And

1.64

 และ

1.48

 και

1.42

และ

1.41

而

1.36

 और

1.31

mathrm

1.30

и

1.22

Activations Density 0.035%