INDEX

Explanations

various relationships or comparisons

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

ざ

1.43

就是

1.40

 그러면은

1.37

 보면은

1.37

 이런

1.37

많

1.33

이제

1.31

 아래

1.30

有不少

1.30

 문제는

1.28

POSITIVE LOGITS

 consistent

1.42

 tailored

1.34

ostar

1.33

 drawn

1.29

 affirmed

1.28

 promising

1.28

 suitable

1.26

 greater

1.24

 adequate

1.24

 beginning

1.24

Activations Density 0.003%