INDEX

Explanations

for loops

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

 aisladas

-1.41

 cubiertas

-1.38

 protegida

-1.32

 cercanas

-1.28

 gewähr

-1.27

 absoluta

-1.26

 esconder

-1.26

 ofrecerte

-1.25

 soportar

-1.24

 aislada

-1.23

POSITIVE LOGITS

for

10.56

 для

4.56

for

4.50

 برای

4.47

สำหรับ

4.13

 για

4.00

对于

3.28

 für

3.17

For

3.11

 pentru

3.06

Activations Density 0.083%