INDEX

Explanations

pursuant to

actions and their context

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

 Hospitality

0.46

 न्याय

0.44

 JUSTICE

0.44

 magist

0.44

 Education

0.44

 influence

0.43

Aś

0.43

 Influence

0.43

 Attractions

0.42

 Missions

0.41

POSITIVE LOGITS

 ње

0.46

botao

0.46

 blancos

0.44

이랑

0.44

bedingungen

0.43

 нему

0.43

 Джон

0.43

lee

0.43

bett

0.43

ন্তী

0.43

Activations Density 0.001%