INDEX

Explanations

asking for type or kind

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

 Кроме

1.09

 Meir

1.06

 Phelps

1.01

 খান

0.98

 Harold

0.96

Кроме

0.96

 குறித்து

0.95

Gow

0.95

 பிரச்சினை

0.95

 negó

0.94

POSITIVE LOGITS

 supervision

0.95

 surveillance

0.94

 Supervision

0.84

 intervening

0.83

 supervising

0.83

 bystand

0.82

 tightly

0.81

 assisting

0.80

 assistive

0.79

 telescopes

0.78

Activations Density 0.221%