书生实战营第三期L1-OpenCompass评测

任务

使用 OpenCompass 评测 internlm2-chat-1.8b 模型在 ceval 数据集上的性能

opencompass可以查看网站

实践

步骤一：创建虚拟环境 opencompass

安装依赖包

studio-conda -o internlm-base -t opencompass
source activate opencompass
git clone -b 0.2.4 https://github.com/open-compass/opencompass
cd opencompass
pip install -r requirements.txt 
pip install -e .
pip install protobuf

步骤二：准备数据 CEVEL数据集

export MKL_SERVICE_FORCE_INTEL=1
cp /share/temp/datasets/OpenCompassData-core-20231110.zip /root/opencompass/
unzip OpenCompassData-core-20231110.zip

步骤三：进行评测

python run.py --datasets ceval_gen --hf-path /share/new_models/Shanghai_AI_Laboratory/internlm2-chat-1_8b --tokenizer-path /share/new_models/Shanghai_AI_Laboratory/internlm2-chat-1_8b --tokenizer-kwargs padding_side='left' truncation='left' trust_remote_code=True --model-kwargs trust_remote_code=True device_map='auto' --max-seq-len 2048 --max-out-len 16 --batch-size 4 --num-gpus 1 --debug

输出结果

输出结果：