opencompass-my-api/configs/eval_qwen_7b_chat_lawbench.py

from mmengine.config import read_base

with read_base():
    from .models.qwen.hf_qwen_7b_chat import models
    from .datasets.lawbench.lawbench_zero_shot_gen_002588 import lawbench_datasets as lawbench_zero_shot_datasets
    from .datasets.lawbench.lawbench_one_shot_gen_002588 import lawbench_datasets as lawbench_one_shot_datasets
    from .summarizers.lawbench import summarizer

datasets = lawbench_zero_shot_datasets + lawbench_one_shot_datasets
for d in datasets:
    d["infer_cfg"]["inferencer"]["save_every"] = 1