{
  "zero_optimization": {
    "stage": 2,
    "offload_optimizer": {
        "device": "cpu"
    },
    "contiguous_gradients": true,
    "overlap_comm": true
  },  
  "fp16": {
    "enabled": "auto"
  },
  "train_micro_batch_size_per_gpu": "auto",
  "gradient_accumulation_steps": "auto"
}