INDEX

Explanations

markup and code structure

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

 naszych

-1.51

 habituellement

-1.44

 utilisé

-1.43

 τελευτα

-1.43

 uniquement

-1.43

我們的

-1.42

 that

-1.41

 innych

-1.38

 erzielt

-1.37

 last

-1.36

POSITIVE LOGITS

('

1.55

('

1.44

 then

1.40

 include

1.40

and

1.38

 including

1.36

۠

1.36

 zrobić

1.35

 как

1.29

Хотя

1.29

Activations Density 0.047%