{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.6,
  "eval_steps": 5,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "grad_norm": 1.7204986331972942,
      "learning_rate": 0.0001,
      "loss": 2.3815,
      "step": 1
    },
    {
      "epoch": 0.4,
      "grad_norm": 3.2293833170556625,
      "learning_rate": 0.00018660254037844388,
      "loss": 2.0498,
      "step": 5
    },
    {
      "epoch": 0.4,
      "eval_loss": 1.5629652738571167,
      "eval_runtime": 3.0527,
      "eval_samples_per_second": 6.552,
      "eval_steps_per_second": 1.638,
      "step": 5
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.9620857356197902,
      "learning_rate": 0.00011736481776669306,
      "loss": 1.412,
      "step": 10
    },
    {
      "epoch": 0.8,
      "eval_loss": 1.3433921337127686,
      "eval_runtime": 1.8318,
      "eval_samples_per_second": 10.918,
      "eval_steps_per_second": 2.73,
      "step": 10
    },
    {
      "epoch": 1.2,
      "grad_norm": 1.0025754462857623,
      "learning_rate": 3.5721239031346066e-05,
      "loss": 1.1535,
      "step": 15
    },
    {
      "epoch": 1.2,
      "eval_loss": 1.2855589389801025,
      "eval_runtime": 1.8278,
      "eval_samples_per_second": 10.942,
      "eval_steps_per_second": 2.736,
      "step": 15
    },
    {
      "epoch": 1.6,
      "grad_norm": 1.0149567514468123,
      "learning_rate": 0.0,
      "loss": 1.0048,
      "step": 20
    },
    {
      "epoch": 1.6,
      "eval_loss": 1.2704746723175049,
      "eval_runtime": 1.8362,
      "eval_samples_per_second": 10.892,
      "eval_steps_per_second": 2.723,
      "step": 20
    },
    {
      "epoch": 1.6,
      "step": 20,
      "total_flos": 548330668032.0,
      "train_loss": 1.421624231338501,
      "train_runtime": 82.0086,
      "train_samples_per_second": 1.951,
      "train_steps_per_second": 0.244
    }
  ],
  "logging_steps": 5,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 548330668032.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}