INDEX

Explanations

creation followed by agent

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

在

-1.45

</i>

-1.13

-1.12

we

-1.11

 will

-1.10

-1.06

 suggest

-1.05

特别是

-1.05

 help

-1.03

def

-1.02

POSITIVE LOGITS

by

3.73

 bernama

1.63

 tarafından

1.59

 réalisateur

1.38

nze

1.37

beitet

1.33

Kolor

1.29

Irak

1.28

 mężczy

1.24

 بواسطة

1.23

Activations Density 0.074%