INDEX

Explanations

indefinite words like "some"

something, anything, some

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

真的是

0.40

最好的

0.38

வனே

0.37

 ನಿಜ

0.36

 solely

0.35

 정말

0.35

egir

0.35

只想

0.34

ebra

0.33

ිට

0.33

POSITIVE LOGITS

 somewhere

2.58

 somehow

2.41

 Somewhere

2.30

 someplace

2.25

 quelque

2.23

 nějak

2.20

 alguma

2.19

 somew

2.16

 algum

2.13

 algún

2.11

Activations Density 0.639%