INDEX

Explanations

digital high/low states

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

ječ

-0.83

加藤

-0.75

 aeruginosa

-0.75

setPixel

-0.73

caling

-0.73

 срока

-0.71

<0x9A>

-0.71

№

-0.71

 секунд

-0.70

 evaporate

-0.70

POSITIVE LOGITS

don

1.03

 High

0.96

 floating

0.94

Dont

0.93

High

0.92

 pulled

0.92

 Dont

0.89

 đừng

0.89

 Undead

0.88

 Pulled

0.88

Activations Density 0.029%