INDEX

Explanations

am essentially

lists, description, interests, technical

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

પણે

0.50

യ

0.48

ående

0.48

preneur

0.43

หลาย

0.43

పాల

0.43

ціона

0.43

วล

0.42

ruktur

0.42

timestamps

0.42

POSITIVE LOGITS

 miraculously

0.54

 mirac

0.51

 Pile

0.47

 magically

0.46

 खि

0.46

 absolutamente

0.46

oplanes

0.46

 Feynman

0.45

 Axis

0.44

 simpt

0.44

Activations Density 0.001%