Egregoros

@cjd @thecsb @jeff quantum tunneling is the best source for random numbers.

@cjd @thecsb @jeff all software was deterministic until LLM's entered the picture.

@jeff @cjd @thecsb Compilers are deterministic. Unfortunately, they often taken in inputs from all over. These inputs change between builds.

A deterministic system will always give the same output, for a given input.

@jeff @thecsb if it's open source you can remove the back door.

@jeff @thecsb you don't need the raw data, just the weights. It's the data in "compressed" form.

@jeff @thecsb a binary is instructions not data.

@jeff @thecsb not compression, but translation. Translation from computer language to machine code.

@jeff @thecsb gradient descent is used to keep the model stable.

The primary goal of a LLM is word prediction. When the word predicted is not the target word gradient descent adjusts the weights.

Compilation is a deterministic process. AI is a probabilistic process. They are not the same.

@jeff @thecsb yes, compilers are deterministic.

@jeff @thecsb if you give the compiler the same inputs, including flags, you will get the same out.

In the situation you're describing the inputs are being changed on each run. You can think of a compiler as a function. And your source code is not the only input.

Some compilers make controlling inputs impossible, thus you can't get a bit-for-bit binary that's identical for each compiler run.

Compilers are deterministic. For a given set of inputs you will get the same output.

@CSB most Americans do not want this war.