[step1_supervised_finetuning] run_chinese.sh error with deepspeed config

[xf4fresh]

When I run a script bash training_scripts/other_language/run_chinese.sh, I encounter a problem.

Traceback (most recent call last):
  File "xxx/DeepSpeedExamples/applications/DeepSpeed-Chat/training/step1_supervised_finetuning/main.py", line 339, in <module>
    main()
  File "xxx/DeepSpeedExamples/applications/DeepSpeed-Chat/training/step1_supervised_finetuning/main.py", line 284, in main
    model, optimizer, _, lr_scheduler = deepspeed.initialize(
  File "/home/admin/miniconda3/envs/DS_chat/lib/python3.9/site-packages/deepspeed/__init__.py", line 137, in initialize
    assert config is None, "Not sure how to proceed, we were given deepspeed configs in the deepspeed arguments and deepspeed.initialize() function call"
AssertionError: Not sure how to proceed, we were given deepspeed configs in the deepspeed arguments and deepspeed.initialize() function call

The prompt message seems to say that there is a problem with my deepspeed config, but I don't know what my problem is. Below is my deepspeed config json.

{
    "bfloat16": {
        "enabled": "auto"
    },
    "fp16": {
        "enabled": "auto",
        "loss_scale": 0,
        "loss_scale_window": 1000,
        "initial_scale_power": 16,
        "hysteresis": 2,
        "min_loss_scale": 1
    },
    "optimizer": {
        "type": "AdamW",
        "params": {
            "lr": "auto",
            "betas": "auto",
            "eps": "auto",
            "weight_decay": "auto"
        }
    },
    "scheduler": {
        "type": "WarmupLR",
        "params": {
            "warmup_min_lr": "auto",
            "warmup_max_lr": "auto",
            "warmup_num_steps": "auto"
        }
    },
    "zero_optimization": {
        "stage": 1
    },
    "gradient_accumulation_steps": "auto",
    "gradient_clipping": "auto",
    "train_batch_size": "auto",
    "train_micro_batch_size_per_gpu": "auto",
    "steps_per_print": 1000
}

If I remove --deepspeed_config ./ds_config.json \ this line in sh, it seems to work fine.

Can someone help explain? Or kindly provide a correct DeepSpeed ​​Config?