LifelongAlignment

/

Qwen2.5-0.5B-Instruct_CPPO_REWARD_0

like 0

Follow

Lifelong Alignment of Agents 6

• "numpy.ndarray",
• "collections.OrderedDict",
• "torch._utils._rebuild_tensor_v2",
• "torch.ByteStorage",
• "_codecs.encode",
• "numpy._core.multiarray._reconstruct",
• "numpy.dtype"

How to fix it?

• "accelerate.utils.dataclasses.DistributedType",
• "transformers.trainer_utils.HubStrategy",
• "transformers.trainer_pt_utils.AcceleratorConfig",
• "transformers.trainer_utils.SchedulerType",
• "transformers.training_args.OptimizerNames",
• "transformers.trainer_utils.IntervalStrategy",
• "accelerate.state.PartialState",
• "trl.trainer.reward_config.RewardConfig",
• "transformers.trainer_utils.SaveStrategy",
• "torch.device"

How to fix it?