INDEX

Explanations

Reinforcement Learning

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 همراه

1.36

heny

1.34

&,

1.34

継

1.25

%@

1.24

стной

1.20

docx

1.19

订

1.18

getto

1.17

hea

1.17

POSITIVE LOGITS

GPIO

1.29

ImageFilter

1.25

ᄂ

1.20

ică

1.16

ไตล์

1.16

Aik

1.15

 Cauchy

1.12

 Froome

1.12

-\{

1.11

 preimage

1.11

Activations Density 0.015%