INDEX

Explanations

list/section beginnings

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

);

-3.11

).

-2.95

or

-2.89

ちゃんの

-2.69

 Since

-2.63

In

-2.59

 wholly

-2.47

 utterly

-2.47

 amidst

-2.45

Get

-2.41

POSITIVE LOGITS

栳

3.88

轳

3.03

卝

3.02

킁

2.81

 одну

2.81

其妙

2.80

 больших

2.78

釹

2.75

amp

2.73

 prendre

2.73

Activations Density 0.002%