INDEX

Explanations

words following punctuation

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

 Também

-1.47

 גם

-1.38

 Além

-1.37

 همچنین

-1.30

 również

-1.27

 também

-1.24

 myös

-1.20

 También

-1.16

또

-1.15

 PUBLICATIONS

-1.14

POSITIVE LOGITS

so

1.45

or

1.21

and

1.11

 definitely

1.05

 only

1.03

 then

1.00

 just

1.00

 thats

0.98

 should

0.97

 still

0.97

Activations Density 0.004%