INDEX

Explanations

Code/Symbols

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

 betweenstory

-0.86

 مشين

-0.78

脚注の使い方

-0.76

parsedMessage

-0.73



-0.73

AddTagHelper

-0.72

ArrowToggle

-0.69

transQ

-0.69

 Мексичка

-0.68

 استنادى

-0.67

POSITIVE LOGITS

<eos>

0.50

↵

0.48

MergeFrom

0.47

larım

0.46

iana

0.46

 lois

0.44

toHexString

0.42

khu

0.41

 wealth

0.41

dega

0.41

Activations Density 0.001%