INDEX

Explanations

code artifacts

The neuron flags snippets of HTML/XML markup, especially angle-bracketed tags and their closing markers.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

You

-1.70

….

-1.59

for

-1.52

…..

-1.49

If

-1.46

One

-1.45

…”

-1.38

.....

-1.37

.......

-1.37

 first

-1.36

POSITIVE LOGITS

 rése

1.63

 imprimi

1.58

 tehd

1.49

 thier

1.43

 rayures

1.42

 normaux

1.41

 éte

1.41

 coussin

1.38

 kabát

1.37

 enfati

1.34

Activations Density 0.260%