zhouyang.xie 97fe68c387 更换unsloth grpo的训练数据集并验证 hai 3 meses
..
train.jsonl 97fe68c387 更换unsloth grpo的训练数据集并验证 hai 3 meses
train_windturbine_old.jsonl 97fe68c387 更换unsloth grpo的训练数据集并验证 hai 3 meses
unified_chip2.jsonl 7270ddb56d Initialize commit hai 3 meses