ds_config.json

{
  "train_micro_batch_size_per_gpu": 1,
  "train_batch_size": 16,
  "gradient_clipping": 1.0,
  "zero_optimization": {
    "stage": 1
  },
  "fp16": {
    "enabled": true,
    "loss_scale": 0,
    "loss_scale_window": 500,
    "hysteresis": 2,
    "min_loss_scale": 1,
    "initial_scale_power": 12
  },
  "zero_allow_untested_optimizer": true,
  "steps_per_print": 2000,
  "wall_clock_breakdown": false
}