INDEX

Explanations

Farming

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

 healthy

-0.99

 tend

-0.92

del

-0.85

 Festival

-0.83

Del

-0.74

healthy

-0.65

del

-0.64

 Healthy

-0.63

Del

-0.60

hay

-0.59

POSITIVE LOGITS

 myſelf

1.04

 Monfieur

1.00

 Theſe

0.99

ſelf

0.95

 purpoſe

0.95

 raiſ

0.94

 whoſe

0.90

 itſelf

0.90

 himſelf

0.88

 ſtate

0.88

Activations Density 0.421%