INDEX

Explanations

advice and conditionals

reviews – and strong sentiment

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

 graças

0.73

 благодаря

0.65

 sayesinde

0.62

 thanks

0.57

 nhờ

0.57

 gracias

0.55

 হওয়ায়

0.55

ďaka

0.55

thanks

0.54

 Whereas

0.54

POSITIVE LOGITS

 lest

1.45

 ASAP

1.45

 asap

1.39

 preferably

1.29

否则

1.28

不然

1.18

preferably

1.16

 otherwise

1.16

以免

1.14

Otherwise

1.10

Activations Density 0.730%