LCM-Lab

/

full_streaming_64k_qwen3-4b_MLP8.0_wfrozen

like 0

Follow

Long-Context Model Laboratory 24

• "training.script_arguments.TrainingArguments",
• "transformers.trainer_pt_utils.AcceleratorConfig",
• "accelerate.utils.dataclasses.DistributedType",
• "torch.device",
• "transformers.trainer_utils.SaveStrategy",
• "transformers.trainer_utils.IntervalStrategy",
• "__builtin__.getattr",
• "transformers.trainer_utils.FSDPOption",
• "transformers.trainer_utils.HubStrategy",
• "accelerate.state.PartialState",
• "transformers.training_args.OptimizerNames",
• "transformers.trainer_utils.SchedulerType"

How to fix it?