INDEX

Explanations

common types or versions

most common

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

 исходя

0.43

umma

0.43

স্থানের

0.43

Mobs

0.42

 appellee

0.42

 મહત્વ

0.42

σταν

0.41

зін

0.41

مكان

0.41

股票

0.39

POSITIVE LOGITS

 type

0.62

 version

0.61

type

0.56

 VERSION

0.56

 Version

0.53

形で

0.52

 types

0.51

 타입

0.50

的版本

0.50

 tipo

0.50

Activations Density 0.033%