INDEX

Explanations

percentage of quantities

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

ات

3.15

ká

2.34

泂

2.30

 مو

2.27

迩

2.26

ವಾಡ

2.26

壤

2.25

udahan

2.23

 дверей

2.23

 electroweak

2.20

POSITIVE LOGITS

give

2.33

 "*");

2.18

Give

2.04

thood

1.97

ϲ

1.93

zī

1.93

ed

1.90

th

1.89

Cet

1.86

ണ

1.85

Activations Density 0.026%