lvj

/

Qwen3-4B-parq-2b-weight-4b-embed-shared-gsm

like 0

• "collections.OrderedDict",
• "torch._utils._rebuild_tensor_v2",
• "torch.device",
• "torch.serialization._get_layout",
• "torchao.quantization.IntxUnpackedToInt8Tensor",
• "torch._utils._rebuild_wrapper_subclass",
• "torch.BFloat16Storage",
• "torch.CharStorage",
• "torch.bfloat16",
• "torch._tensor._rebuild_from_type_v2",
• "torch.int2"

How to fix it?

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

How to fix it?