INDEX

Explanations

Names

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ValueStyle

-0.73

Xna

-0.59

AndEndTag

-0.58

foglal

-0.57

 Ferdinand

-0.55

moveToFirst

-0.55

CodedInputStream

-0.53

@"/

-0.53

Ferdinand

-0.52

kháu

-0.50

POSITIVE LOGITS

Traceback

0.56

cshtml

0.54

gratis

0.54

 uska

0.54

greens

0.50

nationality

0.48

médiaire

0.47

 twice

0.46

ﾟ)

0.46

hithe

0.46

Activations Density 0.151%