INDEX

Explanations

foreign language phrases and names

understandably causes concern

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

まとめて

0.51

まとめ

0.49

ikult

0.44

intercept

0.43

 এলাকায়

0.43

 whims

0.41

ialog

0.41

 summar

0.41

 bosques

0.41

NCB

0.41

POSITIVE LOGITS

그러

0.46

Young

0.45

 Believe

0.45

That

0.45

Believe

0.44

 দাগ

0.44

టై

0.43

Champion

0.43

冠军

0.42

KAN

0.42

Activations Density 0.001%