🔧 96. LoRA: Fine-Tune a Billion-Parameter Model on a Laptop
Nachrichtenbereich: 🔧 Programmierung
🔗 Quelle: dev.to
GPT-2 has 117M parameters. LLaMA-2 has 7B. GPT-3 has 175B.
Full fine-tuning means updating every single parameter. For GPT-2 that's manageable. For LLaMA-2 it needs 28GB of GPU memory just to store... [Weiterlesen]