INDEX

Explanations

works and protects

code definitions and descriptions

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

کرد

0.57

 supporto

0.55

ᱡ

0.54

édrale

0.53

ರ್

0.51

 tratamento

0.51

襟

0.49

락

0.49

Bracelet

0.48

᱒

0.48

POSITIVE LOGITS

ade

0.62

ake

0.56

0.49

ate

0.47

og

0.46

ras

0.46

 पुस्तिका

0.46

ler

0.46

 प्रारंभ

0.46

itsky

0.46

Activations Density 0.001%