INDEX

Explanations

code correctness and improvement

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

 参数

1.40

 capp

1.34

ichlet

1.33

닐

1.32

oxi

1.31

อฟ

1.31

备注

1.31

Contato

1.29

 Notre

1.29

Formatting

1.28

POSITIVE LOGITS

 realise

1.66

 realising

1.64

にとって

1.64

 realises

1.63

 számára

1.53

 wanting

1.51

 regering

1.49

 itself

1.48

想要的

1.48

にとっては

1.48

Activations Density 0.717%