INDEX

Explanations

foreign language recognition

alternatives and comparisons

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

 COULD

0.48

 WOULD

0.45

ctuation

0.44

اردوش

0.44

 airliner

0.43

adder

0.43

定量

0.42

 messagebox

0.42

 exchanger

0.42

 endereco

0.41

POSITIVE LOGITS

 बीच

0.54

 ቅር

0.50

 ঘুষ

0.49

emer

0.47

même

0.47

 größ

0.47

íe

0.47

 gars

0.47

蚋

0.47

 семь

0.46

Activations Density 0.001%