main
Add flash-attention patch for falcon-7b (#3580)
fix typos (#3036)
Changes for orcacode experiment (#3612)
Update Flash Attention forward for Llama 2: (#3595)
Rl training (#2206)
Configure PEFT from config (#3571)
Add LoRA and Prefix-Tuning as Modeling Options for Improved Memory Efficiency + performance (potentially) (#2840)
Add GPTNeoXRewardModel (#2182)
Fix Rope scaling (#3598)