thanks! honestly most of that stability is from the data and tokenizer, not the dynamics model itself. the hardest part was to build a diverse enough but small dataset, that would contain round trips to teach memory and mix non trivial and random behavior on all stages of crafting tree, like, the world model isn't afraid of lava yet :)
thanks! honestly most of that stability is from the data and tokenizer, not the dynamics model itself. the hardest part was to build a diverse enough but small dataset, that would contain round trips to teach memory and mix non trivial and random behavior on all stages of crafting tree, like, the world model isn't afraid of lava yet :)
pretty sweet looks stable enough even tho obv its not perfectly persistent