{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "global_step": 8450,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 0.0002,
      "loss": 1.1569,
      "step": 845
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 14.5711,
      "eval_loss": 0.14331915974617004,
      "eval_per": 0.1751,
      "eval_runtime": 93.1351,
      "eval_samples_per_second": 58.485,
      "eval_steps_per_second": 1.836,
      "step": 845
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.00017777777777777779,
      "loss": 0.1409,
      "step": 1690
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 14.5844,
      "eval_loss": 0.10491351783275604,
      "eval_per": 0.1383,
      "eval_runtime": 92.5515,
      "eval_samples_per_second": 58.854,
      "eval_steps_per_second": 1.848,
      "step": 1690
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.00015555555555555556,
      "loss": 0.1073,
      "step": 2535
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 14.6058,
      "eval_loss": 0.09005340933799744,
      "eval_per": 0.1281,
      "eval_runtime": 91.8497,
      "eval_samples_per_second": 59.303,
      "eval_steps_per_second": 1.862,
      "step": 2535
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.00013333333333333334,
      "loss": 0.0927,
      "step": 3380
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 14.5798,
      "eval_loss": 0.08551742881536484,
      "eval_per": 0.1237,
      "eval_runtime": 92.071,
      "eval_samples_per_second": 59.161,
      "eval_steps_per_second": 1.857,
      "step": 3380
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.00011111111111111112,
      "loss": 0.0833,
      "step": 4225
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 14.6139,
      "eval_loss": 0.08394311368465424,
      "eval_per": 0.1207,
      "eval_runtime": 92.0384,
      "eval_samples_per_second": 59.182,
      "eval_steps_per_second": 1.858,
      "step": 4225
    },
    {
      "epoch": 6.0,
      "learning_rate": 8.888888888888889e-05,
      "loss": 0.0762,
      "step": 5070
    },
    {
      "epoch": 6.0,
      "eval_gen_len": 14.5913,
      "eval_loss": 0.08085508644580841,
      "eval_per": 0.118,
      "eval_runtime": 91.0725,
      "eval_samples_per_second": 59.809,
      "eval_steps_per_second": 1.878,
      "step": 5070
    },
    {
      "epoch": 7.0,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.0707,
      "step": 5915
    },
    {
      "epoch": 7.0,
      "eval_gen_len": 14.6055,
      "eval_loss": 0.08100058883428574,
      "eval_per": 0.1171,
      "eval_runtime": 90.5081,
      "eval_samples_per_second": 60.182,
      "eval_steps_per_second": 1.889,
      "step": 5915
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.0667,
      "step": 6760
    },
    {
      "epoch": 8.0,
      "eval_gen_len": 14.6025,
      "eval_loss": 0.07957806438207626,
      "eval_per": 0.1166,
      "eval_runtime": 91.0929,
      "eval_samples_per_second": 59.796,
      "eval_steps_per_second": 1.877,
      "step": 6760
    },
    {
      "epoch": 9.0,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.0638,
      "step": 7605
    },
    {
      "epoch": 9.0,
      "eval_gen_len": 14.6115,
      "eval_loss": 0.07928925007581711,
      "eval_per": 0.115,
      "eval_runtime": 90.6785,
      "eval_samples_per_second": 60.069,
      "eval_steps_per_second": 1.886,
      "step": 7605
    },
    {
      "epoch": 10.0,
      "learning_rate": 0.0,
      "loss": 0.0609,
      "step": 8450
    },
    {
      "epoch": 10.0,
      "eval_gen_len": 14.6046,
      "eval_loss": 0.0802011638879776,
      "eval_per": 0.1151,
      "eval_runtime": 90.409,
      "eval_samples_per_second": 60.248,
      "eval_steps_per_second": 1.891,
      "step": 8450
    },
    {
      "epoch": 10.0,
      "step": 8450,
      "total_flos": 1.2418278660440064e+17,
      "train_loss": 0.19193305201784394,
      "train_runtime": 3656.7872,
      "train_samples_per_second": 295.702,
      "train_steps_per_second": 2.311
    }
  ],
  "max_steps": 8450,
  "num_train_epochs": 10,
  "total_flos": 1.2418278660440064e+17,
  "trial_name": null,
  "trial_params": null
}