Updated README with new information

README.md

@@ -520,6 +520,7 @@ use_cpu: false
 ```bash
 deepspeed --num_gpus 8 src/train_bash.py \
     --deepspeed ds_config.json \ 
+        --ddp_timeout 180000000 \ # If the training data is too large, it is recommended to add the ddp_timeout command line option to prevent NCCL errors.
     ... # arguments (same as above)
 ```
 

README_zh.md

@@ -519,7 +519,9 @@ use_cpu: false
 ```bash
 deepspeed --num_gpus 8 src/train_bash.py \
     --deepspeed ds_config.json \
+    --ddp_timeout 180000000 \  # 如训练数据过大，建议加上ddp_timeout命令行，防止nccl报错
     ... # 参数同上
+
 ```
 
 <details><summary>使用 DeepSpeed ZeRO-2 进行全参数训练的 ds_config.json 示例</summary>