INDEX

Explanations

scientific/medical topics

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

ſelf

-0.95

 myſelf

-0.95

AndEndTag

-0.94

 itſelf

-0.93

 EconPapers

-0.91

 houſe

-0.91

ſelves

-0.91

 purpoſe

-0.89

 Jefus

-0.87

 Anſ

-0.86

POSITIVE LOGITS

Cap

0.51

Cap

0.51

Mu

0.50

Min

0.49

Cal

0.49

Qu

0.49

Mu

0.48

Mel

0.48

Ra

0.48

Lo

0.48

Activations Density 0.016%