INDEX

Explanations

decision and importance criteria

conservative, validation, innovation, notation

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

其他

0.48

风险

0.45

 OTHER

0.45

损伤

0.45

预防

0.44

层面

0.43

퐶

0.43

 المقدم

0.43

笔记本

0.43

区块

0.43

POSITIVE LOGITS

 ем

0.47

(«

0.46

namely

0.46

(=

0.45

 compet

0.44

ைக்

0.44

チャン

0.44

«

0.44

त्वा

0.44

 compétences

0.44

Activations Density 0.000%