INDEX

Explanations

specific names and compounds

code comments and documentation blocks in programming files.

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

ニ

-0.76

 apparently

-0.76

 percor

-0.75

۰

-0.75

VALIDATE

-0.72

着

-0.72

 уточ

-0.72

]);

-0.72

hatic

-0.71

tedly

-0.70

POSITIVE LOGITS

 dates

0.86

 Josie

0.84

𓂃

0.83

 matchups

0.82

 reagieren

0.81

утбо

0.80

σαι

0.80

ₜ

0.79

ські

0.79

 Sita

0.79

Activations Density 0.008%