set -x
export CUDA_VISIBLE_DEVICES="0,1,2,3,4,5,6,7"
MODEL_NAME_OR_PATH="/volume/ailab4sci/ztgao/em/checkpoints/qwen25_math_7b_1shot_pi1/step_10"
OUTPUT_DIR="/volume/ailab4sci/ztgao/em/checkpoints/qwen25_math_7b_1shot_pi1/step_10/temp03/eval"
mkdir -p $OUTPUT_DIR
PROMPT_TYPE="qwen25-math-cot"
MAX_TOKENS_PER_CALL="3072"
SPLIT="test"
NUM_TEST_SAMPLE=-1
DATA_NAMES="aime25x8,aime24x8,amc23x8,minerva_math,olympiadbench,math500"
IFS=',' read -ra DATASETS <<< "$DATA_NAMES"
ALL_EXIST=true

TOKENIZERS_PARALLELISM=false \
python3 -u math_eval.py \
    --model_name_or_path ${MODEL_NAME_OR_PATH} \
    --data_name ${DATA_NAMES} \
    --output_dir ${OUTPUT_DIR} \
    --split ${SPLIT} \
    --prompt_type ${PROMPT_TYPE} \
    --num_test_sample ${NUM_TEST_SAMPLE} \
    --seed 0 \
    --temperature 0.3 \
    --n_sampling 1 \
    --top_p 1 \
    --start 0 \
    --end -1 \
    --use_vllm \
    --save_outputs \
    --max_tokens_per_call ${MAX_TOKENS_PER_CALL} \
    --overwrite