INDEX

Explanations

Sequence of events

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

AndEndTag

-0.76

quelize

-0.59

RUnlock

-0.58

RectangleBorder

-0.57

geschäft

-0.56

 dima

-0.56

 Forman

-0.56

choon

-0.55

 مشين

-0.54

 intptr

-0.53

POSITIVE LOGITS

He

0.54

He

0.53

SEB

0.49

(';

0.49

Afterwards

0.48

seeking

0.47

 Himself

0.46

findIndex

0.45

tas

0.44

 Hän

0.43

Activations Density 0.006%