INDEX

Explanations

window state or treatments

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

你需要

1.27

你要

1.15

 brauchen

1.14

 youre

1.12

 ideia

1.09

 אתה

1.09

 smelly

1.08

あなた

1.08

あなたの

1.08

 당신

1.08

POSITIVE LOGITS

 procured

1.51

 obtained

1.43

 acquired

1.31

 published

1.27

 dated

1.26

 constructed

1.23

获得的

1.17

obtained

1.16

 obtenidos

1.16

 purchased

1.16

Activations Density 0.334%