INDEX

Explanations

Latin and foreign phrases

The neuron detects the preposition “in” when it appears in Latin experimental phrases such as “in vivo” or “in vitro.”

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

↑↑↑</

-1.53

anwal

-1.50

 varför

-1.49

そして

-1.48

尤其是

-1.42

 behöver

-1.41

 särskilt

-1.41

сында

-1.38

)、

-1.38

でお

-1.36

POSITIVE LOGITS

and

1.89

縠

1.54

诓

1.51

 ぬいぐるみ

1.50

 ſta

1.47

gira

1.45

 simpel

1.44

 hvil

1.43

 voren

1.42

1.41

Activations Density 0.072%