INDEX

Explanations

learning and generalization

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

一下

1.41

 naquele

1.33

了一下

1.33

 iban

1.29

 iria

1.25

สัก

1.24

 mereka

1.23

 ними

1.23

 antes

1.22

本次

1.20

POSITIVE LOGITS

 meestal

2.03

typically

2.02

 genellikle

1.99

基本的に

1.98

 보통

1.98

usually

1.97

often

1.96

 readily

1.95

 customarily

1.95

 عادة

1.91

Activations Density 0.283%