INDEX

Explanations

forum and q&a responses

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 THEN

-1.09

 WANT

-1.01

 PLUS

-1.00

 ketat

-0.96

nowe

-0.95

-}

-0.94

-0.93

áne

-0.93

 Danke

-0.91

 xxiii

-0.90

POSITIVE LOGITS

 sorry

1.28

you

1.21

we

1.21

 available

1.05

 Sorry

1.00

 unfortunately

0.99

begin

0.95

Sorry

0.94

 generally

0.94

 sappi

0.93

Activations Density 0.065%