INDEX

Explanations

progress over time

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

 when

-2.25

 first

-1.55

 know

-1.48

 want

-1.40

 When

-1.36

by

-1.36

 knew

-1.36

 encantado

-1.34

 because

-1.33

at

-1.32

POSITIVE LOGITS

 várias

1.48

 różnych

1.41

 anything

1.39

いずれ

1.39

 AMAZING

1.34

 včetně

1.34

 Dinas

1.34

 entweder

1.31

 innych

1.30

 รวม

1.28

Activations Density 0.128%