INDEX

Explanations

used to do

formerly, once, or past states

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

 ensuing

0.61

 оказалось

0.57

 thereafter

0.55

 turned

0.54

 decided

0.51

 итоге

0.51

later

0.50

 finally

0.49

Finally

0.48

 later

0.48

POSITIVE LOGITS

曾经

1.48

曾經

1.45

เคย

1.39

 เคย

1.33

曾

1.20

曾在

1.16

 pernah

1.13

 autrefois

1.09

 했던

1.05

 existed

1.02

Activations Density 0.159%