{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 15.988480921526278,
  "eval_steps": 500,
  "global_step": 16656,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.48,
      "learning_rate": 1.9399615754082615e-05,
      "loss": 1.8112,
      "step": 500
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.879923150816523e-05,
      "loss": 1.6328,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_f1": 0.9029,
      "eval_gen_len": 19.87818181818182,
      "eval_loss": 1.4800708293914795,
      "eval_precision": 0.9134,
      "eval_recall": 0.893,
      "eval_rouge1": 0.448,
      "eval_rouge2": 0.2243,
      "eval_rougeL": 0.385,
      "eval_rougeLsum": 0.385,
      "eval_runtime": 603.3554,
      "eval_samples_per_second": 9.116,
      "eval_steps_per_second": 0.57,
      "step": 1041
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.819884726224784e-05,
      "loss": 1.4991,
      "step": 1500
    },
    {
      "epoch": 1.92,
      "learning_rate": 1.7598463016330453e-05,
      "loss": 1.4598,
      "step": 2000
    },
    {
      "epoch": 2.0,
      "eval_f1": 0.9022,
      "eval_gen_len": 19.934363636363635,
      "eval_loss": 1.405110478401184,
      "eval_precision": 0.9147,
      "eval_recall": 0.8903,
      "eval_rouge1": 0.4428,
      "eval_rouge2": 0.2273,
      "eval_rougeL": 0.3851,
      "eval_rougeLsum": 0.385,
      "eval_runtime": 669.8531,
      "eval_samples_per_second": 8.211,
      "eval_steps_per_second": 0.514,
      "step": 2083
    },
    {
      "epoch": 2.4,
      "learning_rate": 1.6998078770413066e-05,
      "loss": 1.3652,
      "step": 2500
    },
    {
      "epoch": 2.88,
      "learning_rate": 1.6397694524495677e-05,
      "loss": 1.3402,
      "step": 3000
    },
    {
      "epoch": 3.0,
      "eval_f1": 0.9034,
      "eval_gen_len": 19.95,
      "eval_loss": 1.3839877843856812,
      "eval_precision": 0.9158,
      "eval_recall": 0.8918,
      "eval_rouge1": 0.4498,
      "eval_rouge2": 0.2318,
      "eval_rougeL": 0.3921,
      "eval_rougeLsum": 0.392,
      "eval_runtime": 670.3562,
      "eval_samples_per_second": 8.205,
      "eval_steps_per_second": 0.513,
      "step": 3125
    },
    {
      "epoch": 3.36,
      "learning_rate": 1.579731027857829e-05,
      "loss": 1.2679,
      "step": 3500
    },
    {
      "epoch": 3.84,
      "learning_rate": 1.5196926032660904e-05,
      "loss": 1.2446,
      "step": 4000
    },
    {
      "epoch": 4.0,
      "eval_f1": 0.9054,
      "eval_gen_len": 19.884,
      "eval_loss": 1.3682185411453247,
      "eval_precision": 0.9169,
      "eval_recall": 0.8944,
      "eval_rouge1": 0.4604,
      "eval_rouge2": 0.2405,
      "eval_rougeL": 0.4014,
      "eval_rougeLsum": 0.4014,
      "eval_runtime": 577.6339,
      "eval_samples_per_second": 9.522,
      "eval_steps_per_second": 0.596,
      "step": 4167
    },
    {
      "epoch": 4.32,
      "learning_rate": 1.4596541786743516e-05,
      "loss": 1.1877,
      "step": 4500
    },
    {
      "epoch": 4.8,
      "learning_rate": 1.399615754082613e-05,
      "loss": 1.1651,
      "step": 5000
    },
    {
      "epoch": 5.0,
      "eval_f1": 0.9055,
      "eval_gen_len": 19.894,
      "eval_loss": 1.3695330619812012,
      "eval_precision": 0.9173,
      "eval_recall": 0.8942,
      "eval_rouge1": 0.4594,
      "eval_rouge2": 0.2401,
      "eval_rougeL": 0.3995,
      "eval_rougeLsum": 0.3995,
      "eval_runtime": 669.362,
      "eval_samples_per_second": 8.217,
      "eval_steps_per_second": 0.514,
      "step": 5208
    },
    {
      "epoch": 5.28,
      "learning_rate": 1.3395773294908743e-05,
      "loss": 1.1201,
      "step": 5500
    },
    {
      "epoch": 5.76,
      "learning_rate": 1.2795389048991355e-05,
      "loss": 1.1002,
      "step": 6000
    },
    {
      "epoch": 6.0,
      "eval_f1": 0.9053,
      "eval_gen_len": 19.91181818181818,
      "eval_loss": 1.3782570362091064,
      "eval_precision": 0.9166,
      "eval_recall": 0.8945,
      "eval_rouge1": 0.4607,
      "eval_rouge2": 0.2423,
      "eval_rougeL": 0.4014,
      "eval_rougeLsum": 0.4014,
      "eval_runtime": 671.1543,
      "eval_samples_per_second": 8.195,
      "eval_steps_per_second": 0.513,
      "step": 6250
    },
    {
      "epoch": 6.24,
      "learning_rate": 1.2195004803073969e-05,
      "loss": 1.0653,
      "step": 6500
    },
    {
      "epoch": 6.72,
      "learning_rate": 1.1594620557156582e-05,
      "loss": 1.0427,
      "step": 7000
    },
    {
      "epoch": 7.0,
      "eval_f1": 0.9056,
      "eval_gen_len": 19.907454545454545,
      "eval_loss": 1.3850913047790527,
      "eval_precision": 0.9172,
      "eval_recall": 0.8946,
      "eval_rouge1": 0.462,
      "eval_rouge2": 0.2432,
      "eval_rougeL": 0.4028,
      "eval_rougeLsum": 0.4028,
      "eval_runtime": 669.8936,
      "eval_samples_per_second": 8.21,
      "eval_steps_per_second": 0.514,
      "step": 7292
    },
    {
      "epoch": 7.2,
      "learning_rate": 1.0994236311239194e-05,
      "loss": 1.0163,
      "step": 7500
    },
    {
      "epoch": 7.68,
      "learning_rate": 1.0393852065321808e-05,
      "loss": 0.9881,
      "step": 8000
    },
    {
      "epoch": 8.0,
      "eval_f1": 0.9059,
      "eval_gen_len": 19.907090909090908,
      "eval_loss": 1.3910883665084839,
      "eval_precision": 0.9177,
      "eval_recall": 0.8947,
      "eval_rouge1": 0.4635,
      "eval_rouge2": 0.2442,
      "eval_rougeL": 0.4038,
      "eval_rougeLsum": 0.4037,
      "eval_runtime": 573.3321,
      "eval_samples_per_second": 9.593,
      "eval_steps_per_second": 0.6,
      "step": 8334
    },
    {
      "epoch": 8.16,
      "learning_rate": 9.79346781940442e-06,
      "loss": 0.9742,
      "step": 8500
    },
    {
      "epoch": 8.64,
      "learning_rate": 9.193083573487034e-06,
      "loss": 0.9435,
      "step": 9000
    },
    {
      "epoch": 9.0,
      "eval_f1": 0.9067,
      "eval_gen_len": 19.880545454545455,
      "eval_loss": 1.4075220823287964,
      "eval_precision": 0.918,
      "eval_recall": 0.8959,
      "eval_rouge1": 0.468,
      "eval_rouge2": 0.2471,
      "eval_rougeL": 0.4085,
      "eval_rougeLsum": 0.4084,
      "eval_runtime": 599.9366,
      "eval_samples_per_second": 9.168,
      "eval_steps_per_second": 0.573,
      "step": 9375
    },
    {
      "epoch": 9.12,
      "learning_rate": 8.592699327569645e-06,
      "loss": 0.9362,
      "step": 9500
    },
    {
      "epoch": 9.6,
      "learning_rate": 7.992315081652257e-06,
      "loss": 0.9035,
      "step": 10000
    },
    {
      "epoch": 10.0,
      "eval_f1": 0.9064,
      "eval_gen_len": 19.881090909090908,
      "eval_loss": 1.412468671798706,
      "eval_precision": 0.9178,
      "eval_recall": 0.8957,
      "eval_rouge1": 0.4675,
      "eval_rouge2": 0.248,
      "eval_rougeL": 0.4085,
      "eval_rougeLsum": 0.4086,
      "eval_runtime": 566.1377,
      "eval_samples_per_second": 9.715,
      "eval_steps_per_second": 0.608,
      "step": 10417
    },
    {
      "epoch": 10.08,
      "learning_rate": 7.391930835734871e-06,
      "loss": 0.9014,
      "step": 10500
    },
    {
      "epoch": 10.56,
      "learning_rate": 6.791546589817484e-06,
      "loss": 0.8702,
      "step": 11000
    },
    {
      "epoch": 11.0,
      "eval_f1": 0.9063,
      "eval_gen_len": 19.894727272727273,
      "eval_loss": 1.4218909740447998,
      "eval_precision": 0.9181,
      "eval_recall": 0.895,
      "eval_rouge1": 0.4646,
      "eval_rouge2": 0.2455,
      "eval_rougeL": 0.405,
      "eval_rougeLsum": 0.4051,
      "eval_runtime": 670.3799,
      "eval_samples_per_second": 8.204,
      "eval_steps_per_second": 0.513,
      "step": 11459
    },
    {
      "epoch": 11.04,
      "learning_rate": 6.191162343900097e-06,
      "loss": 0.8741,
      "step": 11500
    },
    {
      "epoch": 11.52,
      "learning_rate": 5.590778097982709e-06,
      "loss": 0.8395,
      "step": 12000
    },
    {
      "epoch": 12.0,
      "learning_rate": 4.990393852065322e-06,
      "loss": 0.8458,
      "step": 12500
    },
    {
      "epoch": 12.0,
      "eval_f1": 0.9061,
      "eval_gen_len": 19.898545454545456,
      "eval_loss": 1.4338867664337158,
      "eval_precision": 0.9177,
      "eval_recall": 0.8952,
      "eval_rouge1": 0.4643,
      "eval_rouge2": 0.2447,
      "eval_rougeL": 0.4055,
      "eval_rougeLsum": 0.4055,
      "eval_runtime": 670.6829,
      "eval_samples_per_second": 8.201,
      "eval_steps_per_second": 0.513,
      "step": 12501
    },
    {
      "epoch": 12.48,
      "learning_rate": 4.390009606147935e-06,
      "loss": 0.8172,
      "step": 13000
    },
    {
      "epoch": 12.96,
      "learning_rate": 3.7896253602305477e-06,
      "loss": 0.8207,
      "step": 13500
    },
    {
      "epoch": 13.0,
      "eval_f1": 0.9064,
      "eval_gen_len": 19.905272727272727,
      "eval_loss": 1.44303560256958,
      "eval_precision": 0.9182,
      "eval_recall": 0.8952,
      "eval_rouge1": 0.4671,
      "eval_rouge2": 0.2463,
      "eval_rougeL": 0.4068,
      "eval_rougeLsum": 0.4069,
      "eval_runtime": 650.7057,
      "eval_samples_per_second": 8.452,
      "eval_steps_per_second": 0.529,
      "step": 13542
    },
    {
      "epoch": 13.44,
      "learning_rate": 3.189241114313161e-06,
      "loss": 0.8006,
      "step": 14000
    },
    {
      "epoch": 13.92,
      "learning_rate": 2.5888568683957737e-06,
      "loss": 0.7987,
      "step": 14500
    },
    {
      "epoch": 14.0,
      "eval_f1": 0.9059,
      "eval_gen_len": 19.918,
      "eval_loss": 1.449475646018982,
      "eval_precision": 0.9179,
      "eval_recall": 0.8944,
      "eval_rouge1": 0.4633,
      "eval_rouge2": 0.2455,
      "eval_rougeL": 0.4046,
      "eval_rougeLsum": 0.4047,
      "eval_runtime": 661.0314,
      "eval_samples_per_second": 8.32,
      "eval_steps_per_second": 0.52,
      "step": 14584
    },
    {
      "epoch": 14.4,
      "learning_rate": 1.988472622478386e-06,
      "loss": 0.7843,
      "step": 15000
    },
    {
      "epoch": 14.88,
      "learning_rate": 1.3880883765609993e-06,
      "loss": 0.787,
      "step": 15500
    },
    {
      "epoch": 15.0,
      "eval_f1": 0.9064,
      "eval_gen_len": 19.895636363636363,
      "eval_loss": 1.4560260772705078,
      "eval_precision": 0.9182,
      "eval_recall": 0.8953,
      "eval_rouge1": 0.4666,
      "eval_rouge2": 0.2471,
      "eval_rougeL": 0.407,
      "eval_rougeLsum": 0.4072,
      "eval_runtime": 670.9962,
      "eval_samples_per_second": 8.197,
      "eval_steps_per_second": 0.513,
      "step": 15626
    },
    {
      "epoch": 15.36,
      "learning_rate": 7.87704130643612e-07,
      "loss": 0.7775,
      "step": 16000
    },
    {
      "epoch": 15.84,
      "learning_rate": 1.8731988472622478e-07,
      "loss": 0.772,
      "step": 16500
    },
    {
      "epoch": 15.99,
      "eval_f1": 0.9068,
      "eval_gen_len": 19.881636363636364,
      "eval_loss": 1.4622657299041748,
      "eval_precision": 0.9185,
      "eval_recall": 0.8957,
      "eval_rouge1": 0.4678,
      "eval_rouge2": 0.2472,
      "eval_rougeL": 0.4081,
      "eval_rougeLsum": 0.4082,
      "eval_runtime": 669.6134,
      "eval_samples_per_second": 8.214,
      "eval_steps_per_second": 0.514,
      "step": 16656
    },
    {
      "epoch": 15.99,
      "step": 16656,
      "total_flos": 3.421567656204632e+18,
      "train_loss": 1.050457198154915,
      "train_runtime": 71670.9422,
      "train_samples_per_second": 22.324,
      "train_steps_per_second": 0.232
    }
  ],
  "logging_steps": 500,
  "max_steps": 16656,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 16,
  "save_steps": 500,
  "total_flos": 3.421567656204632e+18,
  "train_batch_size": 24,
  "trial_name": null,
  "trial_params": null
}