INDEX

Explanations

the word "hydrogen"

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

__*/

-0.96

 myſelf

-0.91

таратура

-0.86

 kasarigan

-0.81

Билгалдахарш

-0.80

 abstrait

-0.80

 romantique

-0.80

 imprimée

-0.80

authorities

-0.78

MeasureSpec

-0.77

POSITIVE LOGITS

 hydrogen

1.05

 Hydrogen

0.93

Hydrogen

0.92

hydrogen

0.77

 mots

0.68

 Watson

0.66

észetes

0.65

naire

0.64

sian

0.62

Watson

0.62

Activations Density 0.003%