INDEX

Explanations

quantities and approximations

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

 what

-1.43

 very

-1.25

big

-1.20

 although

-1.16

 with

-1.13

 first

-1.10

as

-1.05

 most

-1.04

gallon

-1.02

 huge

-1.02

POSITIVE LOGITS

 dozen

1.41

ようになります

1.23

少々

1.13

だけではなく

1.10

 exuber

1.06

 fifty

1.06

 thousands

1.05

让自己

1.05

1.03

1.02

Activations Density 0.072%