INDEX

Explanations

inquiry or uncertainty

need, will, what, should

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

 現在

0.42

 എന്നാല്‍

0.41

 परंतु

0.40

 しかし

0.39

 પરંતુ

0.39

 ولكن

0.38

 jedoch

0.37

ዓ

0.37

 ระ

0.36

 способом

0.36

POSITIVE LOGITS

 magari

0.38

 darn

0.36

 fudge

0.36

бл

0.36

 уж

0.34

0.33

koj

0.33

cui

0.32

 какую

0.32

лайте

0.32

Activations Density 0.113%