INDEX

Explanations

Further

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

to

-0.74

 Cytometry

-0.52

mıştır

-0.52

ToCheck

-0.51

éfonos

-0.51

 aveug

-0.51

 ragazze

-0.50

 elér

-0.49

StateToProps

-0.48

 fédé

-0.48

POSITIVE LOGITS

URLException

0.66

 demonstrating

0.66

astify

0.61

 BoxDecoration

0.60

Placed

0.59

 comp

0.59

 illustrating

0.58

HostException

0.57

 developing

0.55

itse

0.55

Activations Density 0.002%