INDEX

Explanations

instructional steps or modifications

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

とされる

0.91

itudes

0.90

であり

0.90

ledged

0.87

らしい

0.85

的女

0.85

と共に

0.85

 নেতৃ

0.84

ildiği

0.83

recent

0.83

POSITIVE LOGITS

 Alternatively

2.55

If

2.46

 Remember

2.40

Use

2.39

 Если

2.34

Alternatively

2.24

 Assuming

2.23

 如果

2.20

 Nếu

2.15

 Once

2.12

Activations Density 3.787%