INDEX

Explanations

group labels and categories

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

 borboleta

-1.21

obviously

-1.13

 místa

-1.12

 květ

-1.10

 読む

-1.10

 článek

-1.06

 hudeb

-1.06

 москва

-1.06

 手順

-1.05

Notable

-1.05

POSITIVE LOGITS

 group

1.27

of

1.23

 then

1.16

for

1.12

ष्ट

1.04

 conspira

1.00

 membership

0.98

group

0.94

 jPanel

0.94

Activations Density 0.040%