INDEX

Explanations

hyphenated words or words with a prefix

Prefixes

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

Negative Logits

<bos>

-0.57

val

-0.53

Comma

-0.48

for

-0.47

github

-0.47

console

-0.46

 Walker

-0.46

 Siete

-0.45

Ten

-0.45

ph

-0.45

POSITIVE LOGITS

 GenerationType

0.86

 مرئيه

0.86

 متعلقه

0.80

aarrggbb

0.78

 betweenstory

0.77

 calendriers

0.74

MessageTagHelper

0.72

RenderAtEndOf

0.71

 Extragalactic

0.69

URLException

0.66

Activations Density 1.155%