INDEX

Explanations

restricted scope and limited views

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Without

-1.64

 without

-1.43

 undeniably

-1.31

iyle

-1.31

Almost

-1.30

Surprisingly

-1.30

Unlike

-1.30

 Definitely

-1.27

 Surprisingly

-1.27

🧾

-1.26

POSITIVE LOGITS

 view

1.59

ſſen

1.56

 productividad

1.56

 mögliche

1.56

 personalización

1.54

 nowych

1.51

 quirky

1.47

 overly

1.46

 walls

1.43

再度

1.41

Activations Density 0.060%