Show HN: Fine-tune Llama3-8B on 8GB GPU without quantization

3 points| anuarsh | 4 months ago |github.com