INDEX

Explanations

mathematical instructions

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

 this

-1.38

同上

-1.31

 этом

-1.27

selben

-1.22

 same

-1.20

selbe

-1.16

 desselben

-1.11

囡

-1.10

 demselben

-1.09

一部分

-1.09

POSITIVE LOGITS

by

1.44

1.34

>+</

1.17

我已经

1.16

 любого

1.14

他们

1.13

for

1.10

 resourceCulture

1.09

ivir

1.06

или

1.04

Activations Density 0.068%