INDEX

Explanations

comparisons and qualities

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

 infatti

-1.34

 şeyler

-1.30

 этого

-1.27

 döner

-1.23

 inoltre

-1.19

이며

-1.17

 habido

-1.15

 therefore

-1.14

傒

-1.13

それでは

-1.12

POSITIVE LOGITS

its

1.52

 both

1.45

if

1.24

any

1.23

 only

1.13

挽回

1.13

 before

1.10

 either

1.10

 every

1.09

 time

1.06

Activations Density 0.065%