Update performance table for Llama 2 in config hub
rasbt opened this issue · comments
We only have an entry for QLoRA, but I can easily rerun LoRA and Full and add them to the table today.
Already done via #1138
Lightning-AI / litgpt
Pretrain, finetune, deploy 20+ LLMs on your own data. Uses state-of-the-art techniques: flash attention, FSDP, 4-bit, LoRA, and more.