INDEX

Explanations

briefly explain or describe

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

 cannot

1.46

 expects

1.43

 should

1.38

 could

1.34

 doesn

1.32

 will

1.30

 wouldn

1.30

 thinks

1.29

can

1.26

 requires

1.26

POSITIVE LOGITS

 những

2.14

 those

1.78

各种

1.73

 기존

1.73

 tych

1.73

那些

1.72

各種

1.71

 تلك

1.70

各类

1.70

 다양한

1.69

Activations Density 0.135%