INDEX

Explanations

code comments and licensing sections

decorative separator lines made of repeated characters (dashes, equals signs, asterisks) in code comments and documentation.

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

ではでは

-1.42

 potrebbe

-1.42

 LIABLE

-1.42

所以在

-1.36

actualité

-1.35

ではまた

-1.30

 häufigsten

-1.30

首先

-1.25

 arriving

-1.24

↵↵↵↵↵↵↵↵↵↵↵↵

-1.23

POSITIVE LOGITS

and

2.00

 уже

1.54

 estos

1.52

 from

1.50

 después

1.41

//*[@

1.40

ния

1.39

ㄜ

1.38

 которые

1.37

 даже

1.36

Activations Density 0.032%