INDEX

Explanations

class names or compound nouns

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

lessons

-1.38

 their

-1.30

its

-1.29

넋

-1.25

 balón

-1.23

المراجع

-1.23

municipi

-1.23

 nomme

-1.22

 jälkeen

-1.21

 powiedzieć

-1.21

POSITIVE LOGITS

1.51

on

1.35

le

1.33

ar

1.30

1.26

1.23

1.21

th

1.20

as

1.20

in

1.18

Activations Density 0.073%