INDEX

Explanations

mathematical symbols and notation

This neuron detects LaTeX‐style names of mathematical function spaces or algebraic/cohomological objects—i.e. capital letters like C, H, L, Ext followed by parentheses or superscripts.

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

-1.73

use

-1.69

-1.66

</h2>

-1.62

-1.56

-1.53

</h3>

-1.53

 antidesliz

-1.45

Its

-1.40

</strong>

-1.38

POSITIVE LOGITS

퐼

1.80

鶿

1.75

 musíte

1.70

簓

1.59

 môžete

1.51

 budete

1.50

نے

1.44

 chacune

1.40

会う

1.39

榦

1.38

Activations Density 0.022%