Trained BLOOM model saved differently comparing to OPT model

[alibabadoufu]

Hi, I following the script to train the bloom model for my own dataset. However, I found that it saved the model differently as compared to other models such as OPT. The screenshot below shows the saved files.

When I load the model and tokenizer using the following script:

  tokenizer = AutoTokenizer.from_pretrained(model_path,fast_tokenizer=True)
  model = AutoModelForCausalLM.from_pretrained(
              model_path,
          )

I always have this error:

Exception has occurred: OSError
Can't load tokenizer for 'xxx/DeepSpeed/DeepSpeedExamples/applications/DeepSpeed-Chat/output/actor-models/bloom-1b-v1'. If you were trying to load it from 'https://huggingface.co/models', make sure you don't have a local directory with the same name. Otherwise, make sure 'xxx/DeepSpeed/DeepSpeedExamples/applications/DeepSpeed-Chat/output/actor-models/bloom-1b-v1' is the correct path to a directory containing all relevant files for a BloomTokenizerFast tokenizer.