INDEX

Explanations

date and time formats

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

</b>

-1.69

 cuando

-1.59

when

-1.58

-1.43

not

-1.40

ню

-1.40

 także

-1.39

 acest

-1.38

not

-1.38

ئيس

-1.34

POSITIVE LOGITS

 butik

1.34

 Lider

1.33

◆

1.30

 féminin

1.26

 nettsted

1.25

 évé

1.24

cembre

1.23



1.22

 日常

1.22

 लेकिन

1.21

Activations Density 0.018%