INDEX

Explanations

specialising in various areas

showcasing work

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

0.53

 sprawl

0.50

 considere

0.49

 encroach

0.47

 upkeep

0.46

 waging

0.46

 barbar

0.44

 canoe

0.43

 definir

0.43

 maintenance

0.43

POSITIVE LOGITS

0.57

 Effect

0.55

亮度

0.52

 эффект

0.50

 میر

0.50

alui

0.49

Data

0.49

ról

0.49

یک

0.48

Activations Density 0.000%