INDEX

Explanations

new location and humbled

common issues and monitoring

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

突き

0.46

𒉰

0.42

٦

0.41

めっちゃ

0.40

쭌

0.40

 گاه

0.40

 هذه

0.40

 леп

0.40

 લગાવ

0.40

XO

0.39

POSITIVE LOGITS

镑

0.45

 হত্যাকাণ্ড

0.44

 włos

0.44

 couplers

0.43

 chwili

0.43

 hilft

0.42

 morals

0.41

 behalf

0.40

 auxiliary

0.40

zyć

0.40

Activations Density 0.001%