INDEX

Explanations

time units followed by ago or later

time ago

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

一旦

0.39

પણ

0.39

最後まで

0.38

すぐに

0.37

 commandes

0.37

 점점

0.36

সক

0.35

ku

0.35

 nächsten

0.35

 parib

0.34

POSITIVE LOGITS

ago

2.19

Ago

1.62

Ago

1.54

ago

1.51

 geleden

1.42

AGO

1.41

AGO

1.29

 later

1.13

 назад

1.06

 earlier

0.92

Activations Density 0.128%