INDEX

Explanations

names of places and entities

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

 then

-1.73

 übri

-1.52

 there

-1.41

 people

-1.35

-1.25

 Provinsi

-1.23

Kraj

-1.21

 estábamos

-1.20

 silang

-1.20

 masakan

-1.20

POSITIVE LOGITS

нить

1.30

🈷

1.27

婓

1.23

↵↵

1.23

 através

1.21

通过

1.20

が必要です

1.19

bibitem

1.18

用

1.17

ntax

1.16

Activations Density 0.024%