sliver.network

- local llms 101 - running a model = inference (using model weights) - inference = predicting the next token based on your input plus all tokens generated so far - together, these make up the "sequence" - tokens ≠ words - they're the chunks representing the text a model sees -...