INDEX

Explanations

numbers and measurements

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

 bend

-0.98

$.

-0.83

 Bend

-0.83

viders

-0.82

END

-0.81

壌

-0.81

$.

-0.79

猫

-0.79

 bending

-0.77

bend

-0.77

POSITIVE LOGITS

 eight

1.18

ten

1.16

twelve

1.06

 fifteen

1.03

 twelve

1.00

 TWENTY

0.96

 nine

0.95

 douze

0.92

 NINE

0.92

 seven

0.89

Activations Density 0.023%