Logarithmic magnitude adjustment for extended context

Read the equation

For scale>1, the adjustment adds 0.1mscaleln(scale) to one. The code returns one for scale≤1. One parameter specifies the context extension, while the other controls adjustment strength; the returned value should not be confused with the input parameter also named mscale.

Terms & definitions

How it works

Read the definition from its right-hand expression to the quantity on the left. Its surrounding source records the intended input, state, or parameter context. Where the source leaves an operation unspecified, this page preserves that gap.

Assumptions & limits

Source recovery establishes that the expression appears in this record. Its mathematical validity, implementation behavior, and interpretation must be assessed separately. See the linked source and research notes for the checks available for this version.

Original context

are interpolated by the factor of 40.

  THE MAGNITUDE CORRECTION
      yarn_get_mscale(scale, mscale) = 0.1 * mscale * ln(scale) + 1.0   for scale>1
      With scale=40, mscale=1.0:  0.1 * ln(40) + 1.0 = 1.3688879

      cos/sin caches are multiplied by _mscale, computed as
          yarn_get_mscale(factor, mscale) / yarn_get_mscale(factor, mscale_all_dim)

yarn_get_mscale(scale, mscale) = 0.1 * mscale * ln(scale) + 1.0   for scale>1

Related builds