INDEX

Explanations

measurement result or report

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

 chré

-1.43

 blauen

-1.36

 Kontrol

-1.27

kredit

-1.26

 этого

-1.23

 blaue

-1.19

moderne

-1.18

kontrol

-1.16

 Banyak

-1.15

alcool

-1.15

POSITIVE LOGITS

 such

1.44

 other

1.41

其他

1.37

或

1.30

 autres

1.23

他

1.22

other

1.21

("")){

1.20

 سایر

1.18

 provide

1.15

Activations Density 0.001%