INDEX

Explanations

of/in/part

structured data and code

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

Kwa

0.51

Ulu

0.50

 Giri

0.49

 Yatha

0.48

Laut

0.48

 പറയാ

0.47

 அனுப்ப

0.46

 Certaines

0.46

Sau

0.46

 Arrang

0.46

POSITIVE LOGITS

 uterine

0.46

 actuality

0.46

upload

0.46

 mind

0.45

 navegador

0.45

 naveg

0.44

 volv

0.44

 marriage

0.44

 volat

0.44

 upload

0.44

Activations Density 0.002%