INDEX

Explanations

mathematical variables and expressions

This neuron detects the appearance of differential‐operator tokens (the “d…,” “dv,” “du,” etc.) used in integral expressions.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

-1.72

 omdat

-1.68

点点头

-1.66

嵛

-1.52

duur

-1.52

📙

-1.52

ㅔ

-1.48

 dann

-1.47

尪

-1.45

点了点头

-1.45

POSITIVE LOGITS

of

1.65

 includes

1.58

 there

1.56

 including

1.55

 from

1.55

who

1.54

They

1.54

 istimewa

1.43

):

1.43

 especially

1.41

Activations Density 0.006%