INDEX

Explanations

asking 'how much'

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

ed

1.44

Statuses

1.43

 başlı

1.36



1.36

 głównie

1.33

鸳

1.32

altra

1.31

beam

1.27

acteur

1.27

 Köz

1.25

POSITIVE LOGITS

 leeway

2.65

)\|_{\

2.36

 verschil

2.36

 différence

2.33

}\|_{

2.22

 difference

2.18

 innervation

2.17

 equates

2.17

}&=

2.15

}<\

2.15

Activations Density 0.110%