Kunwar Satyam Singh a28991ee08 fix: pin padding_free=False to keep max_length enforced (trl 1.6+) ... Newer Unsloth versions auto-enable SFTConfig's padding_free when it's left unset, which conflicts with an explicit max_length + packing=False and raises a ValueError at SFTTrainer init. Verified end-to-end on a GTX 1650 (4GB VRAM): 60 steps complete cleanly, peak reserved memory 1.543 GB (42% of total), train_loss 1.662 -> 1.264.		2026-07-29 14:22:44 +05:30
..
4gb_vram_qlora_guide.md	fix: update examples to trl 1.x API (SFTConfig + processing_class)	2026-06-29 15:18:01 +05:30
train_4gb_vram.py	fix: pin padding_free=False to keep max_length enforced (trl 1.6+)	2026-07-29 14:22:44 +05:30