INDEX

Explanations

mathematical and code formatting

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

 expédi

-1.78

 jugado

-1.73

 lyder

-1.70

 menyen

-1.69

-1.67

-1.66

 preguntó

-1.66

 saken

-1.63

 gustado

-1.61

 pierdas

-1.54

POSITIVE LOGITS

as

2.02

,$_

1.64

,:);

1.59

 refer

1.56

силання

1.48

,"\

1.44

So

1.41

 подчер

1.41

use

1.38

ꔹ

1.38

Activations Density 0.002%