INDEX

Explanations

abstract feelings and 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

No Comments

Negative Logits

囉

1.44

єр

1.36

ИА

1.32

ಲ್ಲೂ

1.29

 собственно

1.29

实在

1.27

 dreaded

1.23

 юбилей

1.23

 رکن

1.23

 вовсе

1.23

POSITIVE LOGITS

°,

1.65

AUC

1.42

 That

1.41

னமாக

1.38

 XNUMX

1.38

′,

1.35

 adaptability

1.33

',

1.33

 SDGs

1.32

 चीजों

1.32

Activations Density 0.083%