RWKV-LM: RNN with transformer-level performance, without using attentiongithub.com4 points·vletal··0 commentsOpen articleSaveView on HN