INDEX

Explanations

providing

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

特色

1.00

 bền

0.99

特色的

0.94

 übrigens

0.89

formatics

0.88

 versatile

0.86

 соответ

0.86

เหมาะ

0.86

 좋습니다

0.86

quiries

0.85

POSITIVE LOGITS

 mistakenly

2.07

 botched

2.05

 needlessly

2.04

 inability

2.04

 carelessly

2.03

 plötzlich

1.99

 unnecessarily

1.97

 negligently

1.96

 inadvertently

1.94

 falsely

1.93

Activations Density 4.114%