INDEX

Explanations

medical/scientific texts

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

<bos>

-1.02

 Réponses

-0.55

 AssemblyVersion

-0.55

 čís

-0.53

 Geschäft

-0.53

cuerpo

-0.52

 declarat

-0.51

 numeri

-0.50

Šaltiniai

-0.48

 caoutchouc

-0.46

POSITIVE LOGITS

httphttps

0.64

DotNetBar

0.63

Lucia

0.62

 Lucia

0.60

])));

0.59

<()>

0.58

 Aval

0.58

Ava

0.56

Piv

0.56

Vy

0.55

Activations Density 0.061%