INDEX

Explanations

phrases beginning with Well

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

 erwähnten

-2.88

 bestehenden

-2.42

 verwendeten

-2.41

㌧

-2.36

 üblichen

-2.36

钿

-2.23

зации

-2.19

Татьяна

-2.16

 einzige

-2.09

 gewünschten

-2.09

POSITIVE LOGITS

 Well

2.80

2.42

埽

2.27

Well

2.27

敉

2.20

us

2.09

2.03

᠌

2.00

宬

1.98

༷

1.98

Activations Density 0.006%