INDEX

Explanations

use or test ID

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

 envisaged

1.16

()]

1.15

 afforded

1.11

夰

1.09

muir

1.09

ultz

1.09

 pioneered

1.08

$(\

1.08

 modulates

1.08

 precluded

1.07

POSITIVE LOGITS

Thinking

1.35

กว่า

1.07

°

1.06

Warning

1.05

Charm

1.05

Properties

1.05

 ว่า

1.04

 pensamento

1.03

ว่า

1.03

Sign

1.01

Activations Density 0.000%