INDEX

Explanations

breadth and depth

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

 While

-2.53

 Also

-2.42

 Generally

-2.39

 Aside

-2.38

 Surprisingly

-2.30

 Since

-2.30

ing

-2.28

 faſt

-2.27

 regards

-2.23

ة

-2.22

POSITIVE LOGITS



2.25

 expanse

2.25

fficking

2.19

 Externe

2.14

怎样的

2.08

 andere

2.06

྅

2.03

 baumwolle

1.98

神

1.97

 ornate

1.93

Activations Density 0.003%