INDEX

Explanations

specialized terminology

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

 ними

-1.25

 informou

-1.23

 яке

-1.20

 PUSTAKA

-1.19

 normalt

-1.17

too

-1.16

-1.14

ことが多い

-1.13

立马

-1.12

accompag

-1.12

POSITIVE LOGITS

but

1.72

ínio

1.34

يح

1.27

 viagra

1.24

 löst

1.23

 days

1.21

 días

1.19

୩

1.17

 didn

1.17

 לאחר

1.15

Activations Density 0.174%