Jeremy Howard sur X : "FSDP/QDoRA with Llama3 : I believe that this combination is likely to create better task-specific models than anything else at any cost."
Tags:
Au sujet de ce document
Infos sur le fichier
Documents with similar tags (experimental)
Answer.AI - You can now train a 70b language model at home
Tags:
2024-03-09 A propos
The End of Finetuning — with Jeremy Howard of Fast.ai
Tags:
2023-10-20 A propos