INDEX

Explanations

different or other options

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

 所有

1.97

所有

1.93

大多数

1.87

гато

1.87

iren

1.86

}--\

1.82

定义

1.81

 اجمع

1.81

ステム

1.80

itiva

1.76

POSITIVE LOGITS

 someplace

3.14

 somewhere

2.89

 another

2.66

 autre

2.59

 otra

2.51

อื่น

2.45

 innego

2.42

 annan

2.41

 outra

2.39

 outro

2.38

Activations Density 0.513%