How to use finetune result?

[dviettu134]

Hi, which model should I use for inference from the finetuned result?
This is the structure from finetuned result folder:

\---checkpoint-15478
    |   config.json
    |   latest
    |   model.safetensors
    |   rng_state_0.pth
    |   rng_state_1.pth
    |   sentencepiece.bpe.model
    |   special_tokens_map.json
    |   tokenizer.json
    |   tokenizer_config.json
    |   trainer_state.json
    |   training_args.bin
    |   zero_to_fp32.py
    |
    \---global_step15478
            mp_rank_00_model_states.pt
            zero_pp_rank_0_mp_rank_00_optim_states.pt
            zero_pp_rank_1_mp_rank_00_optim_states.pt