INDEX

Explanations

as possible

heads up, sick, along with

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

 Анаста

0.51

kl

0.47

 Через

0.47

клад

0.45

햐

0.45

UID

0.45

circum

0.43

ORT

0.43

kehr

0.43

Ｌ

0.42

POSITIVE LOGITS

 músicas

0.54

 princes

0.49

 perguntas

0.46

 notamment

0.46

 preguntar

0.46

 pregunt

0.45

 choir

0.45

 tentar

0.45

 nought

0.45

 tragedies

0.45

Activations Density 0.000%