INDEX

Explanations

mining practice

text format indicators

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 LaTeX

0.43

 cate

0.42

là

0.42

 tyle

0.41

JSC

0.41

 Latex

0.41

 decor

0.39

às

0.39

 jest

0.39

 při

0.39

POSITIVE LOGITS

 २०२

0.49

Нау

0.49

 Кра

0.46

 Брита

0.45

 துன்ப

0.44

अय

0.44

 университета

0.44

 relentlessly

0.44

新冠

0.43

 катастро

0.43

Activations Density 0.001%