{
  "best_metric": 1.6119977235794067,
  "best_model_checkpoint": "./results/checkpoint-20",
  "epoch": 6.511627906976744,
  "eval_steps": 10,
  "global_step": 280,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.23,
      "eval_loss": 2.2899038791656494,
      "eval_runtime": 11.3104,
      "eval_samples_per_second": 6.543,
      "eval_steps_per_second": 0.884,
      "step": 10
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0003,
      "loss": 2.4572,
      "step": 20
    },
    {
      "epoch": 0.47,
      "eval_loss": 1.6119977235794067,
      "eval_runtime": 11.7383,
      "eval_samples_per_second": 6.304,
      "eval_steps_per_second": 0.852,
      "step": 20
    },
    {
      "epoch": 0.7,
      "eval_loss": 1.1244831085205078,
      "eval_runtime": 11.8689,
      "eval_samples_per_second": 6.235,
      "eval_steps_per_second": 0.843,
      "step": 30
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0003,
      "loss": 1.1806,
      "step": 40
    },
    {
      "epoch": 0.93,
      "eval_loss": 0.9044250249862671,
      "eval_runtime": 11.9487,
      "eval_samples_per_second": 6.193,
      "eval_steps_per_second": 0.837,
      "step": 40
    },
    {
      "epoch": 1.16,
      "eval_loss": 0.8299169540405273,
      "eval_runtime": 11.9786,
      "eval_samples_per_second": 6.178,
      "eval_steps_per_second": 0.835,
      "step": 50
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0003,
      "loss": 0.7449,
      "step": 60
    },
    {
      "epoch": 1.4,
      "eval_loss": 0.7723114490509033,
      "eval_runtime": 12.0027,
      "eval_samples_per_second": 6.165,
      "eval_steps_per_second": 0.833,
      "step": 60
    },
    {
      "epoch": 1.63,
      "eval_loss": 0.7225701212882996,
      "eval_runtime": 11.9999,
      "eval_samples_per_second": 6.167,
      "eval_steps_per_second": 0.833,
      "step": 70
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0003,
      "loss": 0.7214,
      "step": 80
    },
    {
      "epoch": 1.86,
      "eval_loss": 0.6744211316108704,
      "eval_runtime": 11.9881,
      "eval_samples_per_second": 6.173,
      "eval_steps_per_second": 0.834,
      "step": 80
    },
    {
      "epoch": 2.09,
      "eval_loss": 0.6709991693496704,
      "eval_runtime": 11.9553,
      "eval_samples_per_second": 6.19,
      "eval_steps_per_second": 0.836,
      "step": 90
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0003,
      "loss": 0.6126,
      "step": 100
    },
    {
      "epoch": 2.33,
      "eval_loss": 0.6710302233695984,
      "eval_runtime": 11.9579,
      "eval_samples_per_second": 6.188,
      "eval_steps_per_second": 0.836,
      "step": 100
    },
    {
      "epoch": 2.56,
      "eval_loss": 0.6822958588600159,
      "eval_runtime": 11.9598,
      "eval_samples_per_second": 6.187,
      "eval_steps_per_second": 0.836,
      "step": 110
    },
    {
      "epoch": 2.79,
      "learning_rate": 0.0003,
      "loss": 0.4763,
      "step": 120
    },
    {
      "epoch": 2.79,
      "eval_loss": 0.6459034085273743,
      "eval_runtime": 11.9686,
      "eval_samples_per_second": 6.183,
      "eval_steps_per_second": 0.836,
      "step": 120
    },
    {
      "epoch": 3.02,
      "eval_loss": 0.6284101009368896,
      "eval_runtime": 11.9574,
      "eval_samples_per_second": 6.189,
      "eval_steps_per_second": 0.836,
      "step": 130
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0003,
      "loss": 0.4888,
      "step": 140
    },
    {
      "epoch": 3.26,
      "eval_loss": 0.7302632331848145,
      "eval_runtime": 11.9624,
      "eval_samples_per_second": 6.186,
      "eval_steps_per_second": 0.836,
      "step": 140
    },
    {
      "epoch": 3.49,
      "eval_loss": 0.6239209175109863,
      "eval_runtime": 11.958,
      "eval_samples_per_second": 6.188,
      "eval_steps_per_second": 0.836,
      "step": 150
    },
    {
      "epoch": 3.72,
      "learning_rate": 0.0003,
      "loss": 0.5019,
      "step": 160
    },
    {
      "epoch": 3.72,
      "eval_loss": 0.6370986104011536,
      "eval_runtime": 11.9588,
      "eval_samples_per_second": 6.188,
      "eval_steps_per_second": 0.836,
      "step": 160
    },
    {
      "epoch": 3.95,
      "eval_loss": 0.6650754809379578,
      "eval_runtime": 11.9585,
      "eval_samples_per_second": 6.188,
      "eval_steps_per_second": 0.836,
      "step": 170
    },
    {
      "epoch": 4.19,
      "learning_rate": 0.0003,
      "loss": 0.3947,
      "step": 180
    },
    {
      "epoch": 4.19,
      "eval_loss": 0.6445854902267456,
      "eval_runtime": 11.9509,
      "eval_samples_per_second": 6.192,
      "eval_steps_per_second": 0.837,
      "step": 180
    },
    {
      "epoch": 4.42,
      "eval_loss": 0.6579444408416748,
      "eval_runtime": 11.9603,
      "eval_samples_per_second": 6.187,
      "eval_steps_per_second": 0.836,
      "step": 190
    },
    {
      "epoch": 4.65,
      "learning_rate": 0.0003,
      "loss": 0.3996,
      "step": 200
    },
    {
      "epoch": 4.65,
      "eval_loss": 0.645702600479126,
      "eval_runtime": 11.964,
      "eval_samples_per_second": 6.185,
      "eval_steps_per_second": 0.836,
      "step": 200
    },
    {
      "epoch": 4.88,
      "eval_loss": 0.6404394507408142,
      "eval_runtime": 11.954,
      "eval_samples_per_second": 6.19,
      "eval_steps_per_second": 0.837,
      "step": 210
    },
    {
      "epoch": 5.12,
      "learning_rate": 0.0003,
      "loss": 0.4472,
      "step": 220
    },
    {
      "epoch": 5.12,
      "eval_loss": 0.8005651235580444,
      "eval_runtime": 11.9642,
      "eval_samples_per_second": 6.185,
      "eval_steps_per_second": 0.836,
      "step": 220
    },
    {
      "epoch": 5.35,
      "eval_loss": 0.7091573476791382,
      "eval_runtime": 11.9502,
      "eval_samples_per_second": 6.192,
      "eval_steps_per_second": 0.837,
      "step": 230
    },
    {
      "epoch": 5.58,
      "learning_rate": 0.0003,
      "loss": 0.4052,
      "step": 240
    },
    {
      "epoch": 5.58,
      "eval_loss": 0.6789885759353638,
      "eval_runtime": 11.9644,
      "eval_samples_per_second": 6.185,
      "eval_steps_per_second": 0.836,
      "step": 240
    },
    {
      "epoch": 5.81,
      "eval_loss": 0.6493074297904968,
      "eval_runtime": 11.9517,
      "eval_samples_per_second": 6.192,
      "eval_steps_per_second": 0.837,
      "step": 250
    },
    {
      "epoch": 6.05,
      "learning_rate": 0.0003,
      "loss": 0.4125,
      "step": 260
    },
    {
      "epoch": 6.05,
      "eval_loss": 0.6842274069786072,
      "eval_runtime": 11.9486,
      "eval_samples_per_second": 6.193,
      "eval_steps_per_second": 0.837,
      "step": 260
    },
    {
      "epoch": 6.28,
      "eval_loss": 0.7352823615074158,
      "eval_runtime": 11.9558,
      "eval_samples_per_second": 6.189,
      "eval_steps_per_second": 0.836,
      "step": 270
    },
    {
      "epoch": 6.51,
      "learning_rate": 0.0003,
      "loss": 0.3443,
      "step": 280
    },
    {
      "epoch": 6.51,
      "eval_loss": 0.656093418598175,
      "eval_runtime": 11.9533,
      "eval_samples_per_second": 6.191,
      "eval_steps_per_second": 0.837,
      "step": 280
    }
  ],
  "logging_steps": 20,
  "max_steps": 300,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 20,
  "total_flos": 52877712752640.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}