INDEX

Explanations

code, testing, medical, URLs

class definitions and code block endings in programming code.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 cromado

-0.90

ment

-0.88

 kuten

-0.85

 suivants

-0.82

ments

-0.80

 nagu

-0.79

TBA

-0.79

 according

-0.78

 there

-0.78

⮚

-0.78

POSITIVE LOGITS

eraard

0.98

нгредієнти

0.93

ÊNCIAS

0.92

🧷

0.87

fekt

0.87

annin

0.87

 pointe

0.87

orgt

0.87

잘

0.86

やはり

0.86

Activations Density 0.001%