INDEX

Explanations

function calls with 'w'

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

-2.17

*/

-1.70

 necessarie

-1.52

 łatwo

-1.50

 takiej

-1.50

 emphas

-1.48

 jardim

-1.45

旁边

-1.44

 oczywiście

-1.42

 prefeito

-1.41

POSITIVE LOGITS

</h2>

1.81

 världen

1.75

двига

1.52

isierung

1.48

嬗

1.44

mancy

1.42

[]{"

1.41

urably

1.41

of

1.40

之际

1.37

Activations Density 0.026%