INDEX

Explanations

have/has/had + been

past states or completed actions

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 venne

0.53

 দিলো

0.48

 উঠল

0.45

ようになりました

0.45

 можа

0.43

 উঠলো

0.42

 пришли

0.42

 کہنے

0.41

 கட்டப்பட்டது

0.39

 alcanzó

0.39

POSITIVE LOGITS

 been

2.25

been

2.17

 BEEN

2.03

 Been

1.94

Been

1.89

 olnud

1.88

 ollut

1.81

 vært

1.77

 varit

1.76

 été

1.73

Activations Density 0.275%