INDEX

Explanations

Is followed by There

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

EDI

-0.73

ară

-0.71

ᄂ

-0.71

بة

-0.70

▬▬

-0.70

Referencie

-0.69

 preuve

-0.68

$(\

-0.68

 perawatan

-0.68

xml

-0.68

POSITIVE LOGITS

 capture

1.42

 capturing

1.22

 Capture

1.21

Capture

1.17

capture

1.17

 captures

1.12

 sniff

1.09

captures

1.06

 packet

1.06

Capt

1.04

Activations Density 0.048%