INDEX

Explanations

programming imports and modules

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

 όμως

-1.33

 malgré

-1.25

љено

-1.20

 referirse

-1.17

 selben

-1.16

onav

-1.15

的同时

-1.14

 répond

-1.14

we

-1.13

isn

-1.13

POSITIVE LOGITS

to

1.41

んでいた

1.26

嫑

1.24

됐

1.23

 that

1.20

おわりに

1.19

 mieste

1.18

 nepiecieš

1.17

erweise

1.12

 novin

1.12

Activations Density 0.021%