export HIP_VISIBLE_DEVICES=0,1,2,3
LOAD_MODEL=/public/LLAMA/model/chatflow_7b.bin
SPM_PATH=/public/LLAMA/model/tokenizer.model

python llama_infer.py --test_path ./prompts.txt --prediction_path ./result.txt \
                      --load_model_path $LOAD_MODEL \
                      --config_path config/llama_7b_config.json \
                      --spm_model_path $SPM_PATH \
                      --world_size 4 --seq_length 1024