Model save

Fix: add assistant_only_loss=False to prevent all labels being masked to -100

Training in progress, step 500

Optimal config: high-quality Hebrew data, constant LR, packing disabled, lora_dropout=0.1

Switch to packing=True, batch=4, grad_acc=4, step-based checkpoints for speed

Disable eval to fix OOM on A10G

Fix OOM: reduce max_length to 2048, disable packing, increase grad accum, remove liger

Add training script with QLoRA

Add entrypoint script with fixed args

Fix typo in exclude_modules and add entrypoint script

Upload training script

Upload training script

initial commit