INDEX

Explanations

biggest followed by a key feature

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

此外

1.67

 অনুষ্ঠিত

1.62

 による

1.59

 설립

1.55

 बृहस्पतिवार

1.46

 anticipates

1.46

 memperoleh

1.41

 proclaims

1.40

 heretofore

1.40

 지난

1.39

POSITIVE LOGITS

 shitty

2.31

 fucking

2.15

 crappy

1.94

 fuck

1.91

 freaking

1.91

 fucked

1.89

 shit

1.82

 haha

1.79

 lmao

1.78

 assh

1.77

Activations Density 0.956%