INDEX

Explanations

do you want?

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

 changed

1.37

 newly

1.32

 designated

1.30

 noted

1.24

 changing

1.20

 created

1.19

 selected

1.19

 described

1.16

 preceding

1.16

 contacted

1.15

POSITIVE LOGITS

Α

1.90

 στο

1.85

 την

1.83

 του

1.80

 της

1.79

Φ

1.77

Γ

1.77

Μ

1.76

 από

1.76

 Κα

1.72

Activations Density 0.027%