ReaLHF
PublicSuper-Efficient RLHF Training of LLMs with Parameter Reallocation
deepspeeddistributed-computingdistributed-systemslarge-language-modelslarge-scale-machine-learningllmllm-frameworkllm-trainingmegatron-lmreinforcement-learning
Creat:2024-06-18T11:01:31
Update:2025-03-26T23:53:59
307
Stars
0
Stars Increase