INDEX

Explanations

each element

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

那些

1.66

futbol

1.50

 causas

1.48

 często

1.48

所有人

1.47

大多數

1.45

 pocas

1.44

 Semua

1.44

 Porque

1.44

 kebanyakan

1.44

POSITIVE LOGITS

 subsequent

2.21

 succeeding

2.20

 uniquely

2.13

 unique

2.08

new

2.07

other

2.06

 individually

2.04

 uniqu

2.02

 nuovo

2.02

 successive

2.00

Activations Density 0.081%