INDEX

Explanations

if needed or optional

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

 सारी

1.29

 огром

1.23

 dietro

1.21

world

1.19

两个

1.16

 Uncle

1.14

 намного

1.13

 הרבה

1.12

 enorme

1.11

 весь

1.11

POSITIVE LOGITS

 intervene

1.83

ください

1.59

 emergencies

1.53

imetre

1.52

 dependant

1.51

 opportune

1.50

aided

1.47

 contraindicated

1.47

 summoned

1.46

 depending

1.45

Activations Density 0.228%