{
  "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": 0.9621952419095638,
      "learning_rate": 0.0001,
      "loss": 1.3382,
      "step": 1
    },
    {
      "epoch": 0.4,
      "grad_norm": 2.3491127966382805,
      "learning_rate": 0.00018660254037844388,
      "loss": 1.5884,
      "step": 5
    },
    {
      "epoch": 0.4,
      "eval_loss": 1.322584867477417,
      "eval_runtime": 3.0601,
      "eval_samples_per_second": 6.536,
      "eval_steps_per_second": 1.634,
      "step": 5
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.3901558326910664,
      "learning_rate": 0.00011736481776669306,
      "loss": 1.061,
      "step": 10
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.9965066909790039,
      "eval_runtime": 1.8319,
      "eval_samples_per_second": 10.917,
      "eval_steps_per_second": 2.729,
      "step": 10
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.7027836902663052,
      "learning_rate": 3.5721239031346066e-05,
      "loss": 0.9222,
      "step": 15
    },
    {
      "epoch": 1.2,
      "eval_loss": 0.9271305799484253,
      "eval_runtime": 1.8367,
      "eval_samples_per_second": 10.889,
      "eval_steps_per_second": 2.722,
      "step": 15
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.594962428089565,
      "learning_rate": 0.0,
      "loss": 0.817,
      "step": 20
    },
    {
      "epoch": 1.6,
      "eval_loss": 0.9120187759399414,
      "eval_runtime": 1.8236,
      "eval_samples_per_second": 10.967,
      "eval_steps_per_second": 2.742,
      "step": 20
    },
    {
      "epoch": 1.6,
      "step": 20,
      "total_flos": 1558264479744.0,
      "train_loss": 1.084645891189575,
      "train_runtime": 82.262,
      "train_samples_per_second": 1.945,
      "train_steps_per_second": 0.243
    }
  ],
  "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": 1558264479744.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}