INDEX

Explanations

explaining reasons or questions

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

ẵ

0.79

Ẵ

0.78

平行

0.78

横浜

0.78

넨

0.78

βάλ

0.74

在新

0.73

 Charleston

0.72

存

0.72

 ලබා

0.71

POSITIVE LOGITS

 reasons

1.87

 explanations

1.85

 explanation

1.82

原因

1.81

理由

1.76

explanation

1.75

理由は

1.71

 Reasons

1.71

 explain

1.69

 причины

1.67

Activations Density 0.263%