INDEX

Explanations

Language, word usage

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

 OFDb

-0.86

 myſelf

-0.74

abestanden

-0.71

 Monfieur

-0.69

neſs

-0.69

 ſeveral

-0.68

 Majefty

-0.67

 ſche

-0.63

parsedMessage

-0.63

localctx

-0.62

POSITIVE LOGITS

ArgsConstructor

0.57

<eos>

0.55

richTextPanel

0.54

0.51

hup

0.47

for

0.46

gatsby

0.46

 rather

0.45

0.44

 poor

0.44

Activations Density 0.009%