INDEX

Explanations

right

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

SourceChecksum

-0.74

 HasFactory

-0.65

TargetException

-0.63

Халык

-0.61

ositol

-0.59

 незавершена

-0.59

 ModelExpression

-0.59

!*\

-0.58

sal

-0.58

 برانيه

-0.58

POSITIVE LOGITS

ly

0.81

theless

0.72

 démocr

0.69

 réguli

0.67

 présidenti

0.67

withstanding

0.66

 automatiques

0.65

 preuve

0.61

 nationaux

0.60

 vägen

0.59

Activations Density 0.033%