INDEX

Explanations

ultimate achievement to date

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

on

-1.43

 исче

-1.34

-1.33

 lakini

-1.31

when

-1.26

 those

-1.26

for

-1.26

("")){

-1.23

 habt

-1.23

 would

-1.23

POSITIVE LOGITS

ypress

1.51

 lança

1.50

切り

1.45

 preuves

1.38

 autres

1.36

 décontract

1.36

していた

1.34

ників

1.30

gilet

1.30

jercicio

1.30

Activations Density 0.014%