INDEX

Explanations

place to do things

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

 there

-1.43

 remplacer

-1.42

 descendre

-1.41

 diminuer

-1.38

 réduire

-1.32

就没有

-1.32

pè

-1.30

additional

-1.29

なんでも

-1.27

logische

-1.27

POSITIVE LOGITS

 both

1.47

 such

1.38

for

1.37

地名

1.36

and

1.35

 that

1.34

ફ

1.30

 before

1.30

colgroup

1.24

isesi

1.22

Activations Density 0.017%