add npu examples

Former-commit-id: 0f21e68e2dbd84c820d66d5c6d980004efc51d51

examples/lora_multi_gpu/ds_zero3.sh

@@ -1,9 +1,15 @@
 #!/bin/bash
 
 NPROC_PER_NODE=4
+NNODES=1
+RANK=0
+MASTER_ADDR=127.0.0.1
+MASTER_PORT=29500
 
-CUDA_VISIBLE_DEVICES=0,1,2,3 python -m torch.distributed.run \
+CUDA_VISIBLE_DEVICES=0,1,2,3 torchrun \
     --nproc_per_node $NPROC_PER_NODE \
-    --nnodes 1 \
-    --standalone \
+    --nnodes $NNODES \
+    --node_rank $RANK \
+    --master_addr $MASTER_ADDR \
+    --master_port $MASTER_PORT \
     src/train.py examples/lora_multi_gpu/llama3_lora_sft_ds.yaml