Pin torch to cu121 build + use model.device instead of hardcoded cuda string 8f291e0 shank commited on 13 days ago
Replace unsloth with bitsandbytes+peft: fixes CUDA driver incompatibility on HF A100 c325ad7 shank commited on 13 days ago
Fix Gradio 4.x every= deprecation: use gr.Timer for auto-refresh 5eea2dd shank commited on 13 days ago
Reduce training to 500 steps with tightened curriculum for A10G budget ba8df98 shank commited on 13 days ago
Optimize for A100 80GB: 8 generations, batch 4, lr 2e-5, dense logging 2b1fbf3 shank commited on 13 days ago
Reduce training to 500 steps with tightened curriculum for A10G budget 3152fa9 shank commited on 13 days ago
Fix: Final submission cleanup, unified identity and integrity markers 8807d25 shank commited on about 1 month ago