INDEX

Explanations

working fine

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

SECTION

0.51

indeed

0.49

raintree

0.48

睄

0.48

Ⲧ

0.48

 实现

0.47

তথ্য

0.46

াদ

0.46

iciencia

0.46

eningkatan

0.46

POSITIVE LOGITS

 certain

0.45

 traditional

0.44

 certaines

0.44

 kompan

0.44

 smaller

0.44

 certains

0.43

 Naked

0.42

La

0.42

 Corporate

0.42

 popular

0.41

Activations Density 0.004%

No Known Activations