INDEX

Explanations

forced actions and conditions

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

 procedentes

-1.36

が良い

-1.34

-1.32

なもの

-1.31

感じです

-1.22

 friv

-1.22

 inga

-1.21

亵

-1.20

esis

-1.20

POSITIVE LOGITS

for

2.13

by

1.85

 have

1.45

1.43

 through

1.39

 memilih

1.38

on

1.37

ländischen

1.35

เพื่อ

1.35

 become

1.34

Activations Density 0.069%