INDEX

Explanations

warranties or conditions

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 工事

-0.84

 いる

-0.84

 moreno

-0.84

久了

-0.83

LEBR

-0.82

ѐ

-0.82

 Privacidad

-0.81

 alene

-0.80

 gòn

-0.79

⌀

-0.79

POSITIVE LOGITS

asantry

1.07

 MEANS

0.99

teness

0.99

ходу

0.95

わかります

0.93

 SHALL

0.93

 REMEMBER

0.93

 simplu

0.90

 INCLUDING

0.89

haling

0.86

Activations Density 1.036%