INDEX

Explanations

introducing details after "following"

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

rógeno

-1.55

 either

-1.52

 faisant

-1.51

 ayant

-1.49

but

-1.44

();

-1.41

 obligé

-1.38

 sugger

-1.36

ումներ

-1.36

 that

-1.35

POSITIVE LOGITS

1.66

 after

1.59

1.45

 from

1.42

他們

1.22

としている

1.22

 despite

1.20

on

1.19

 unique

1.15

renewable

1.13

Activations Density 0.079%