INDEX

Explanations

numbers and specific nouns

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

𝐞

3.34

𝐥

3.20

𝐫

2.99

𝐤

2.99

𝐮

2.95

 ràng

2.92

𝐨

2.92

্লেখ

2.76

lé

2.72

𝐚

2.68

POSITIVE LOGITS

ar

2.70

ما

2.57

[{'

2.42

effectively

2.32

ின்

2.32

StackNavigator

2.28

za

2.26

Des

2.19

 dites

2.17

[{

2.11

Activations Density 0.013%