INDEX

Explanations

network protocol headers

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

but

-1.55

chnung

-1.30

pædia

-1.28

 イエロー

-1.26

</

-1.23

anInt

-1.22

"}";

-1.18

 évolution

-1.17

of

-1.16

 LIABLE

-1.16

POSITIVE LOGITS

Each

1.48

 that

1.43

 Instyt

1.37

 Pozna

1.32

 Projektu

1.30

İL

1.29

 ekst

1.29

靦

1.28

 kadang

1.27

 каждой

1.23

Activations Density 0.081%