INDEX

Explanations

types and formats

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

 nuovamente

1.48

 tanggal

1.35

 lágrimas

1.33

 derrota

1.31

 notificación

1.29

 komment

1.28

幸せ

1.25

 agrade

1.25

 abril

1.25

 victoire

1.25

POSITIVE LOGITS

 subtypes

2.15

types

2.15

examples

2.11

 types

2.05

Types

2.00

 varieties

1.98

类型的

1.96

常见的

1.92

TYPES

1.92

 examples

1.88

Activations Density 3.510%