INDEX

Explanations

code definitions and structure

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Labeled

1.56

FieldContext

1.45

Luke

1.41

Clic

1.38

 الماء

1.37

 الجرس

1.37

Function

1.37

Aprend

1.35

Elegant

1.35

Lemma

1.33

POSITIVE LOGITS

 industrial

0.95

 economic

0.94

√

0.94

 largest

0.93

 significant

0.92

 standard

0.92

otho

0.91

sab

0.89

 penal

0.88

 king

0.88

Activations Density 0.001%