3 ms·
You are free to look at every single weight and study how it affects the result. You can see how the model is architected. And you don't need training data to b
by charcircuit 1y ago
You are free to look at every single weight and study how it affects the result. You can see how the model is architected. And you don't need training data to be provided to be able to modify the weights. Software can still be open source even if it isn't friendly to beginners.
- tuckerman 1y agoI think you could say something remarkably similar about just releasing bytecode as well and I think most people would call foul at that. I don't think it's so cut and dry. This isn't entirely about being a beginner or not either. Full fine-tuning without forgetting does really want the training data (or something that is a good replacement). You can do things like LoRa but, depending on your use case, it might not work.