INDEX

Explanations

code blocks and section headers

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

 what

-1.73

 those

-1.65

 their

-1.59

its

-1.47

how

-1.46

 many

-1.46

 more

-1.43

 several

-1.41

 then

-1.41

 possibly

-1.41

POSITIVE LOGITS

этому

1.23

 trám

1.21

ricle

1.20

 maïs

1.20

 Dinas

1.20

 FUNCTION

1.16

 nimmer

1.16

 anſ

1.15

 reager

1.15

 geze

1.10

Activations Density 0.000%