Viewing profile — halfeatenscone

It only takes a basic understanding of neural nets to recognize this as crankery. Giving an LLM a tricky question does not force it to do any more computation than a simple prompt.…

I don't think this follows. Calling it a derivative work already feels like a stretch, but even granting that framing, the use is clearly transformative and therefore likely to be …

Such logs would not be used for training the base model, but rather for fine-tuning the model for instruction following. Instruction tuning requires far less data than is needed fo…

"Classify the following strings according to whether they represent Python programs which terminate..."

I'm not sure how easy it is to grok these by playing with them. If you're confused, the "About" section has some explanation of how the matrices are constructed and how to interpre…

It's guaranteed to not exist in the training set (i.e. to not be a real GitHub repo that existed before 2015). But the model can certainly repeat itself.