INDEX

Explanations

what it could be

research and translation tasks

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

 amaze

0.48

 acrob

0.45

 aided

0.44

 stacked

0.42

ユ

0.42

cij

0.41

 большин

0.40

契約

0.40

 illuminate

0.40

 stacking

0.40

POSITIVE LOGITS

agm

0.44

 articolo

0.42

กล

0.42

mutable

0.40

이에요

0.40

 মুছে

0.40

 artikel

0.40

OUBLE

0.40

igts

0.40

 Dubrovnik

0.40

Activations Density 0.000%