INDEX

Explanations

hardware logic design

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

 montagna

-1.21

袈

-1.11

kasse

-1.08

 влияет

-1.08

 digno

-1.02

 краї

-1.02

ינו

-1.02

ssohn

-1.01

 besta

-1.00

 vecchia

-1.00

POSITIVE LOGITS

 nélk

1.15

Елена

1.10

 aerospace

1.09

 diseñados

1.09

äm

1.09

,'

1.09

̿

1.08

 assemblies

1.07

ancang

1.05

ングラス

1.05

Activations Density 0.027%