INDEX

Explanations

code explanation comments

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

বিক

1.65

 얘기

1.56

ട്ടില്‍

1.53

商

1.52

ക്ട

1.49

想法

1.48

 얘기를

1.43

 convinced

1.42

提起

1.41

씨

1.41

POSITIVE LOGITS

 abond

1.29

Eksterne

1.21

 пропор

1.18

یثیت

1.16

 optimal

1.15

or

1.15

‍या

1.13

<bos>

1.12

冋

1.12

and

1.11

Activations Density 0.312%