{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.3529411764705883,
  "eval_steps": 10,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11764705882352941,
      "grad_norm": 99.34697723388672,
      "learning_rate": 7e-06,
      "loss": 4.6344,
      "step": 10
    },
    {
      "epoch": 0.11764705882352941,
      "eval_loss": 4.356701850891113,
      "eval_runtime": 7.4823,
      "eval_samples_per_second": 727.722,
      "eval_steps_per_second": 11.494,
      "step": 10
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 68.6648941040039,
      "learning_rate": 1.7e-05,
      "loss": 4.1147,
      "step": 20
    },
    {
      "epoch": 0.23529411764705882,
      "eval_loss": 3.802903652191162,
      "eval_runtime": 7.4967,
      "eval_samples_per_second": 726.322,
      "eval_steps_per_second": 11.472,
      "step": 20
    },
    {
      "epoch": 0.35294117647058826,
      "grad_norm": 73.34857940673828,
      "learning_rate": 1.9222222222222225e-05,
      "loss": 3.3201,
      "step": 30
    },
    {
      "epoch": 0.35294117647058826,
      "eval_loss": 2.9514710903167725,
      "eval_runtime": 7.5247,
      "eval_samples_per_second": 723.619,
      "eval_steps_per_second": 11.429,
      "step": 30
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 61.95991134643555,
      "learning_rate": 1.8111111111111112e-05,
      "loss": 2.1842,
      "step": 40
    },
    {
      "epoch": 0.47058823529411764,
      "eval_loss": 2.131948947906494,
      "eval_runtime": 7.7586,
      "eval_samples_per_second": 701.798,
      "eval_steps_per_second": 11.084,
      "step": 40
    },
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 48.887203216552734,
      "learning_rate": 1.7e-05,
      "loss": 1.2624,
      "step": 50
    },
    {
      "epoch": 0.5882352941176471,
      "eval_loss": 1.3057191371917725,
      "eval_runtime": 7.9165,
      "eval_samples_per_second": 687.801,
      "eval_steps_per_second": 10.863,
      "step": 50
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 35.66941833496094,
      "learning_rate": 1.588888888888889e-05,
      "loss": 0.6907,
      "step": 60
    },
    {
      "epoch": 0.7058823529411765,
      "eval_loss": 0.9652644991874695,
      "eval_runtime": 7.9447,
      "eval_samples_per_second": 685.363,
      "eval_steps_per_second": 10.825,
      "step": 60
    },
    {
      "epoch": 0.8235294117647058,
      "grad_norm": 29.34256935119629,
      "learning_rate": 1.477777777777778e-05,
      "loss": 0.3527,
      "step": 70
    },
    {
      "epoch": 0.8235294117647058,
      "eval_loss": 0.8976885080337524,
      "eval_runtime": 8.0315,
      "eval_samples_per_second": 677.953,
      "eval_steps_per_second": 10.708,
      "step": 70
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 32.28174591064453,
      "learning_rate": 1.3666666666666667e-05,
      "loss": 0.2828,
      "step": 80
    },
    {
      "epoch": 0.9411764705882353,
      "eval_loss": 0.8685462474822998,
      "eval_runtime": 8.14,
      "eval_samples_per_second": 668.919,
      "eval_steps_per_second": 10.565,
      "step": 80
    },
    {
      "epoch": 1.0588235294117647,
      "grad_norm": 38.87677764892578,
      "learning_rate": 1.2555555555555557e-05,
      "loss": 0.2758,
      "step": 90
    },
    {
      "epoch": 1.0588235294117647,
      "eval_loss": 0.8556678295135498,
      "eval_runtime": 8.2601,
      "eval_samples_per_second": 659.194,
      "eval_steps_per_second": 10.412,
      "step": 90
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 33.91325378417969,
      "learning_rate": 1.1444444444444444e-05,
      "loss": 0.2576,
      "step": 100
    },
    {
      "epoch": 1.1764705882352942,
      "eval_loss": 0.8554092049598694,
      "eval_runtime": 8.4912,
      "eval_samples_per_second": 641.252,
      "eval_steps_per_second": 10.128,
      "step": 100
    },
    {
      "epoch": 1.2941176470588236,
      "grad_norm": 27.338443756103516,
      "learning_rate": 1.0333333333333335e-05,
      "loss": 0.2877,
      "step": 110
    },
    {
      "epoch": 1.2941176470588236,
      "eval_loss": 0.8585284948348999,
      "eval_runtime": 8.7035,
      "eval_samples_per_second": 625.611,
      "eval_steps_per_second": 9.881,
      "step": 110
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 33.70604705810547,
      "learning_rate": 9.222222222222224e-06,
      "loss": 0.2109,
      "step": 120
    },
    {
      "epoch": 1.4117647058823528,
      "eval_loss": 0.8407207727432251,
      "eval_runtime": 8.9789,
      "eval_samples_per_second": 606.424,
      "eval_steps_per_second": 9.578,
      "step": 120
    },
    {
      "epoch": 1.5294117647058822,
      "grad_norm": 38.09708786010742,
      "learning_rate": 8.111111111111112e-06,
      "loss": 0.206,
      "step": 130
    },
    {
      "epoch": 1.5294117647058822,
      "eval_loss": 0.8386306166648865,
      "eval_runtime": 9.2342,
      "eval_samples_per_second": 589.658,
      "eval_steps_per_second": 9.313,
      "step": 130
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 47.18006896972656,
      "learning_rate": 7e-06,
      "loss": 0.2828,
      "step": 140
    },
    {
      "epoch": 1.6470588235294117,
      "eval_loss": 0.8333184719085693,
      "eval_runtime": 9.1122,
      "eval_samples_per_second": 597.55,
      "eval_steps_per_second": 9.438,
      "step": 140
    },
    {
      "epoch": 1.7647058823529411,
      "grad_norm": 24.672765731811523,
      "learning_rate": 5.88888888888889e-06,
      "loss": 0.2045,
      "step": 150
    },
    {
      "epoch": 1.7647058823529411,
      "eval_loss": 0.8294004201889038,
      "eval_runtime": 8.9028,
      "eval_samples_per_second": 611.608,
      "eval_steps_per_second": 9.66,
      "step": 150
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 21.150001525878906,
      "learning_rate": 4.777777777777778e-06,
      "loss": 0.1932,
      "step": 160
    },
    {
      "epoch": 1.8823529411764706,
      "eval_loss": 0.828009843826294,
      "eval_runtime": 8.7354,
      "eval_samples_per_second": 623.324,
      "eval_steps_per_second": 9.845,
      "step": 160
    },
    {
      "epoch": 2.0,
      "grad_norm": 55.02903366088867,
      "learning_rate": 3.6666666666666666e-06,
      "loss": 0.1763,
      "step": 170
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.824001133441925,
      "eval_runtime": 8.6395,
      "eval_samples_per_second": 630.244,
      "eval_steps_per_second": 9.954,
      "step": 170
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 15.967198371887207,
      "learning_rate": 2.5555555555555557e-06,
      "loss": 0.1603,
      "step": 180
    },
    {
      "epoch": 2.1176470588235294,
      "eval_loss": 0.8176314830780029,
      "eval_runtime": 8.7379,
      "eval_samples_per_second": 623.15,
      "eval_steps_per_second": 9.842,
      "step": 180
    },
    {
      "epoch": 2.235294117647059,
      "grad_norm": 9.89770221710205,
      "learning_rate": 1.4444444444444445e-06,
      "loss": 0.1445,
      "step": 190
    },
    {
      "epoch": 2.235294117647059,
      "eval_loss": 0.8145530223846436,
      "eval_runtime": 8.8692,
      "eval_samples_per_second": 613.923,
      "eval_steps_per_second": 9.696,
      "step": 190
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 31.038183212280273,
      "learning_rate": 3.3333333333333335e-07,
      "loss": 0.1534,
      "step": 200
    },
    {
      "epoch": 2.3529411764705883,
      "eval_loss": 0.8133436441421509,
      "eval_runtime": 9.0044,
      "eval_samples_per_second": 604.702,
      "eval_steps_per_second": 9.551,
      "step": 200
    }
  ],
  "logging_steps": 10,
  "max_steps": 200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 2500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}