INDEX

Explanations

function returns or exceptions

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

 becomes

-1.38

 what

-1.32

 connects

-1.28

the

-1.27

 operates

-1.24

𝟵

-1.23

 receives

-1.22

 takes

-1.21

 demonstrates

-1.19

 informs

-1.19

POSITIVE LOGITS

 '-';

1.41

 utiles

1.38

 tématu

1.30

//});

1.29

 lakini

1.24

 lutter

1.24

 sauvage

1.20

 volant

1.20

 saine

1.18

 gustos

1.17

Activations Density 0.004%