INDEX

Explanations

calculates mathematical totals

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

 suscit

1.85

 Switching

1.81

 swapped

1.78

 Opened

1.72

 فتحه

1.71

 rife

1.71

 segnal

1.69

 захо

1.68

プト

1.67

jects

1.67

POSITIVE LOGITS

正しい

1.98

 mathematically

1.97

accuracy

1.89

ultimate

1.89

lede

1.85

 automaticamente

1.80

roni

1.79

metrics

1.74

ząc

1.72

出一个

1.71

Activations Density 0.055%