INDEX

Explanations

pollution

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

BeginContext

-0.88

 defaultstate

-0.79

Personendaten

-0.77

 VizieR

-0.76

CloseOperation

-0.74

 aDecoder

-0.72

 DBNull

-0.72

rungsseite

-0.71

 ویکی‌پدی

-0.70

StoreMessageInfo

-0.68

POSITIVE LOGITS

Coordenadas

0.46

ьаж

0.46

nodoc

0.43

 agenda

0.41

gas

0.38

ینده

0.38

áctenos

0.38

これで

0.37

 gian

0.37

 relevant

0.37

Activations Density 0.023%