INDEX

Explanations

Judge names

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

 disponibilités

-1.55

 hés

-1.51

 refroid

-1.50

licopter

-1.50

 prét

-1.45

 négociations

-1.39

짭

-1.39

 SAYS

-1.39

⃢

-1.38

 dépenses

-1.37

POSITIVE LOGITS

 both

1.86

all

1.54

 only

1.45

 each

1.30

 seemingly

1.23

 almost

1.22

 Both

1.16

 allegedly

1.16

1.13

 from

1.11

Activations Density 0.044%