INDEX

Explanations

the

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

 للمعارف

-0.65

pleft

-0.56

 تانيه

-0.53

WEBPACK

-0.48

 correct

-0.48

 وصلة

-0.48

 continúas

-0.47

ServletException

-0.47

tamol

-0.47

 '\\;'

-0.46

POSITIVE LOGITS

LogFactory

0.61

ancier

0.58

velocityX

0.54

fection

0.52



0.51

lifies

0.48

genden

0.48

CreateModel

0.48

 introduction

0.47

odes

0.47

Activations Density 0.012%