INDEX

Explanations

what things are made of

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

So

-2.83

看上去

-2.58

-2.56

Of

-2.44

先前

-2.33

 categor

-2.33

what

-2.33

っている

-2.27

 Sekunden

-2.25

黑衣

-2.25

POSITIVE LOGITS

</em>

2.64

↵

2.63

灕

2.44

But

2.41

釗

2.38

 แต่

2.34

崐

2.30

汭

2.27

 белой

2.27

 estekak

2.27

Activations Density 0.008%