INDEX

Explanations

performance tests and tasks

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

of

-1.27

’

-0.99

利用

-0.94

 this

-0.88

Kad

-0.85

添

-0.84

 coveted

-0.81

通常

-0.81

 Efforts

-0.81

 discussing

-0.80

POSITIVE LOGITS

 performed

1.47

 tests

1.24

ácidos

1.17

 cuarent

1.14

 devised

1.14

 próprios

1.13

ξύ

1.13

 test

1.13

 desafios

1.12

opheles

1.10

Activations Density 0.086%