INDEX

Explanations

course codes

The neuron primarily activates on numeric tokens (course numbers, section numbers, measurements, decimals, etc.).

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

But

-1.60

我来

-1.52

anguardia

-1.52

My

-1.50

its

-1.45

What

-1.44

 când

-1.40

自身的

-1.38

When

-1.38

 любого

-1.32

POSITIVE LOGITS

 egyéb

1.45

phany

1.39

рования

1.33

ické

1.33

or

1.32

ktionen

1.30

inches

1.30

and

1.28

vollen

1.28

ului

1.27

Activations Density 0.021%