INDEX

Explanations

pauses and interjections

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 које

1.37

1.31

 которого

1.28

which

1.28

 которой

1.20

\...

1.18

 koje

1.16

 которых

1.15

1.12

aka

1.10

POSITIVE LOGITS

 appunto

1.18

 通过

1.14

 opnieuw

1.11

 respectively

1.09

 Applic

1.07

 必要

1.06

 successively

1.06

 ขั้น

1.06

 bewust

1.05

 broadening

1.04

Activations Density 0.416%