"# Experimental environment: 2 * 4090

2 * 24GB GPU memory

export NCCL_IB_DISABLE=1; export NCCL_P2P_DISABLE=1
export CUDA_LAUNCH_BLOCKING=1
CUDA_VISIBLE_DEVICES=0,1 \
NPROC_PER_NODE=2 \
swift sft \
--model_id_or_path /data/models/Baichuan2-13B-Chat \
--model_type baichuan-13b-chat \
--model_revision master \
--template_type baichuan \
--sft_type lora \
--tuner_backend peft \
--dtype AUTO \
--output_dir output \
--dataset ms-bench-mini \
--train_dataset_sample -1 \
--num_train_epochs 3 \
--max_length 512 \
--check_dataset_strategy warning \
--lora_rank 8 \
--lora_alpha 32 \
--lora_dropout_p 0.05 \
--lora_target_modules W_pack \
--gradient_checkpointing true \
--batch_size 1 \
--weight_decay 0.1 \
--learning_rate 1e-4 \
--gradient_accumulation_steps 16 \
--max_grad_norm 0.5 \
--warmup_ratio 0.03 \
--eval_steps 100 \
--save_steps 100 \
--save_total_limit 2 \
--logging_steps 10 \
Loading checkpoint shards: 83%时显存爆掉了,ModelScope有优化的地方吗？"