INDEX

Explanations

days, generation, greetings, methods

writing and language generation

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

Qaeda

0.47

UISelectors

0.46

なった

0.45

 نکن

0.44

သင့်

0.44

ではない

0.43

分岐

0.42

 মাঝে

0.41

 consultation

0.41

い

0.41

POSITIVE LOGITS

鐲

0.51

呥

0.49

 Amid

0.49

 менедж

0.49

鋈

0.48

 Martel

0.48

ྷ

0.48

 écrire

0.47

🛖

0.47

写

0.46

Activations Density 0.001%