INDEX

Explanations

legal contexts and Q&A format

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

期刊论文

-0.88

 dinámica

-0.87

不吃

-0.86

Useless

-0.82

eniably

-0.82

凄く

-0.81

ڦ

-0.78

 nemus

-0.78

έρα

-0.77

 estratégico

-0.77

POSITIVE LOGITS

Manager

0.85

Planet

0.83

CHARGE

0.83

 charge

0.83

 במק

0.79

似

0.79

Family

0.77

pump

0.75

Hold

0.74

О

0.74

Activations Density 0.008%