INDEX

Explanations

code and instructions

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

 cardia

0.89

磬

0.86

rectomy

0.86

اكم

0.83

lene

0.83

モリ

0.80

 trab

0.78

Ruf

0.78

 fecal

0.77

渭

0.77

POSITIVE LOGITS

 Sections

1.05

 Section

0.96

Sections

0.94

 Schedule

0.93

 Chapter

0.92

 Enemy

0.87

 Prevention

0.87

第二百

0.86

 Shops

0.85

Section

0.85

Activations Density 0.062%