INDEX

Explanations

starts task-oriented phrases

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

 évidemment

1.61

 belki

1.55

もちろん

1.51

おそらく

1.49

 doubtless

1.42

或許

1.39

 sicherlich

1.39

 zelfs

1.38

 вполне

1.37

bleed

1.37

POSITIVE LOGITS

_:

1.49

(_)

1.44

 नरेश

1.41

 _____

1.40

 ()=>

1.36

 """"

1.36

_.

1.35

ராட்சி

1.34

__

1.33

 मंत्रिपरिषद

1.31

Activations Density 0.244%