INDEX

Explanations

by rival, by existing

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

’?

-1.43

шка

-1.41

anii

-1.33

ってみました

-1.30

)";

-1.28

 Cina

-1.27

 doet

-1.27

"!

-1.26

imine

-1.25

 dificil

-1.23

POSITIVE LOGITS

at

1.43

 Such

1.39

 after

1.34

它的

1.30

 такі

1.27

 такими

1.27

 similar

1.27

 where

1.22

such

1.21

“

1.19

Activations Density 0.268%