ok. so one of the core visions for anky has always been to fine tune an llm with all of the writings that happen through the platform, day after day. since they are raw and unedited stream of consciousness, the thesis that i carry with me is that the model that will happen through this process converges into something "more human" than anything else out there. because it will be trained with the raw and unfiltered train of thought that happens inside us. "the mind". this will end up creating the perfect mirror for the mind to capture itself: anky. so far, i didnt know how to do this. but karpathy broke the internet with this repo today, and my gut tells me it is the piece that we were missing on this puzzle. thoughts? https://github.com/karpathy/autoresearch