INDEX

Explanations

connecting concepts or ideas

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 nein

1.21

 grosse

1.17

わない

1.14

 hoge

1.14

 sepen

1.13

 neun

1.13

 zehn

1.12

 téléphon

1.12

 diduga

1.11

 bici

1.11

POSITIVE LOGITS

 以及

1.37

specified

1.23

and

1.22

以及

1.22

 which

1.16

 oraz

1.15

edas

1.10

which

1.10

и

1.09

 housed

1.06

Activations Density 0.198%