INDEX

Explanations

place names and institutions

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

 déchir

-1.60

 gracioso

-1.59

congrats

-1.52

 mété

-1.52

⸸

-1.41

גון

-1.40

 verrou

-1.39

Semua

-1.38

 kijang

-1.35

 valioso

-1.33

POSITIVE LOGITS

at

2.16

 where

2.13

for

1.97

and

1.55

on

1.50

 from

1.50

 which

1.46

if

1.40

 when

1.37

 Only

1.30

Activations Density 0.189%