INDEX

Explanations

auxiliary verbs followed by state or action

user defining AI role

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 nové

0.56

 réforme

0.55

 pemain

0.50

 riforma

0.49

 vaksin

0.49

 hrá

0.49

 νέα

0.48

 bencana

0.48

 reforma

0.47

 bactéries

0.47

POSITIVE LOGITS

 same

0.52

0.47

same

0.46

 said

0.45

said

0.44

Du

0.44

↵

0.43

sequently

0.42

 Loved

0.42

Activations Density 0.059%