Фото: Мария Ващук / РИА Новости
Что думаешь? Оцени!
。safew是该领域的重要参考
Последние новости
And, even so, the experts don’t train. All this time was just to get a result nearly an order of magnitude more expensive than a training API. It’s still a pain to modify, optimize, or profile the HuggingFace code and we’re using essentially the slowest distributed training method possible. Better parallelization setups/configurations are supposed to be compatible with HuggingFace, but our efforts to set these up were fruitless. Can we really call it a win?
First, I’d present the LLM with a new feature (e.g. loops) or refactor (e.g. moving from a tree-walking interpreter to a bytecode VM). Then I’d have a conversation with it about how the change would work in the context of Cutlet, how other languages implemented it, design considerations, ideas we could steal from interesting/niche languages, etc. Just a casual back-and-forth, the same way you might talk to a co-worker.