baseline-policy-error-gen-qwen-260515-273

Training Info

• Source checkpoint: checkpoint-273 (best by eval_loss, load_best_model_at_end=True)
• Epoch: 3.0
• Best eval_loss: 0.4269
• Training script: finetuning/baseline_model/train_qwen.py
• Base model: Qwen/Qwen2.5-7B-Instruct