Datasets:

---

zx10086

/

1.5B_grpo

like 0

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

How to fix it?

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

How to fix it?

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

How to fix it?

• "torch.bfloat16",
• "torch.device",
• "transformers.integrations.deepspeed.HfTrainerDeepSpeedConfig",
• "accelerate.state.PartialState",
• "transformers.trainer_utils.IntervalStrategy",
• "transformers.trainer_utils.SchedulerType",
• "accelerate.utils.dataclasses.DeepSpeedPlugin",
• "mytrl.trl.trainer.grpo_config.GRPOConfig",
• "transformers.integrations.deepspeed.HfDeepSpeedConfig",
• "accelerate.utils.dataclasses.DistributedType",
• "transformers.trainer_utils.HubStrategy",
• "transformers.trainer_utils.SaveStrategy",
• "transformers.training_args.OptimizerNames",
• "transformers.trainer_pt_utils.AcceleratorConfig"

How to fix it?