INDEX

Explanations

a followed by specific words

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

and

1.85

 because

1.66

But

1.62

 Because

1.59

 Apparently

1.58

 Despite

1.57

To

1.55

If

1.52

And

1.48

POSITIVE LOGITS

素敵な

2.36

 meravigli

2.33

 emocionante

2.29

 prachtige

2.22

 hossz

2.18

 vaš

2.16

 nagyobb

2.13

婄

2.13

大きく

2.11

 yüksek

2.11

Activations Density 0.034%