INDEX

Explanations

python code documentation

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

相

1.89

いろいろ

1.88

 일반적으로

1.88

 通常

1.83

<0xA5>

1.82

 ponieważ

1.82

 具体

1.80

 最近

1.78

町村

1.78

 다양

1.78

POSITIVE LOGITS

isters

1.51

Off

1.50

better

1.49

:@

1.48

ure

1.47

gl

1.45

:#

1.43

afe

1.42

erede

1.41

bl

1.39

Activations Density 0.004%