in the weights, it iterates, and starts to train
backpropagates, and endless weights
and loss to gain
in the weights, the matmul products deviate
and next token seems like a million years a way
in gpt,
RE:
then the appearance arises that the task is to map "OOP", or "FP", or "TDD", or whateverthefuckkindofnoise-oriented programming onto the core proposition of UB.
the task is to throw all those things on the exact same burn pile as the collected works of all the apologists for UB, and start fresh
RE:
