{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 35.0,
  "eval_steps": 500,
  "global_step": 7420,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 9.95e-05,
      "loss": 6.3714,
      "step": 212
    },
    {
      "epoch": 1.0,
      "eval_loss": 5.850927352905273,
      "eval_runtime": 3.8282,
      "eval_samples_per_second": 176.584,
      "eval_steps_per_second": 22.204,
      "step": 212
    },
    {
      "epoch": 2.0,
      "learning_rate": 9.900000000000001e-05,
      "loss": 5.5438,
      "step": 424
    },
    {
      "epoch": 2.0,
      "eval_loss": 4.862037658691406,
      "eval_runtime": 3.7425,
      "eval_samples_per_second": 180.626,
      "eval_steps_per_second": 22.712,
      "step": 424
    },
    {
      "epoch": 3.0,
      "learning_rate": 9.850000000000001e-05,
      "loss": 4.6993,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_loss": 4.018372058868408,
      "eval_runtime": 4.0424,
      "eval_samples_per_second": 167.229,
      "eval_steps_per_second": 21.027,
      "step": 636
    },
    {
      "epoch": 4.0,
      "learning_rate": 9.8e-05,
      "loss": 3.967,
      "step": 848
    },
    {
      "epoch": 4.0,
      "eval_loss": 3.290557384490967,
      "eval_runtime": 4.3933,
      "eval_samples_per_second": 153.872,
      "eval_steps_per_second": 19.348,
      "step": 848
    },
    {
      "epoch": 5.0,
      "learning_rate": 9.75e-05,
      "loss": 3.305,
      "step": 1060
    },
    {
      "epoch": 5.0,
      "eval_loss": 2.6497254371643066,
      "eval_runtime": 4.3562,
      "eval_samples_per_second": 155.182,
      "eval_steps_per_second": 19.513,
      "step": 1060
    },
    {
      "epoch": 6.0,
      "learning_rate": 9.7e-05,
      "loss": 2.7185,
      "step": 1272
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.1158342361450195,
      "eval_runtime": 4.5634,
      "eval_samples_per_second": 148.135,
      "eval_steps_per_second": 18.626,
      "step": 1272
    },
    {
      "epoch": 7.0,
      "learning_rate": 9.65e-05,
      "loss": 2.2181,
      "step": 1484
    },
    {
      "epoch": 7.0,
      "eval_loss": 1.6882902383804321,
      "eval_runtime": 4.3232,
      "eval_samples_per_second": 156.367,
      "eval_steps_per_second": 19.662,
      "step": 1484
    },
    {
      "epoch": 8.0,
      "learning_rate": 9.6e-05,
      "loss": 1.8106,
      "step": 1696
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.3730388879776,
      "eval_runtime": 4.3486,
      "eval_samples_per_second": 155.451,
      "eval_steps_per_second": 19.546,
      "step": 1696
    },
    {
      "epoch": 9.0,
      "learning_rate": 9.55e-05,
      "loss": 1.497,
      "step": 1908
    },
    {
      "epoch": 9.0,
      "eval_loss": 1.142008900642395,
      "eval_runtime": 4.2743,
      "eval_samples_per_second": 158.154,
      "eval_steps_per_second": 19.886,
      "step": 1908
    },
    {
      "epoch": 10.0,
      "learning_rate": 9.5e-05,
      "loss": 1.2526,
      "step": 2120
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.9753935933113098,
      "eval_runtime": 4.2134,
      "eval_samples_per_second": 160.44,
      "eval_steps_per_second": 20.174,
      "step": 2120
    },
    {
      "epoch": 11.0,
      "learning_rate": 9.449999999999999e-05,
      "loss": 1.0576,
      "step": 2332
    },
    {
      "epoch": 11.0,
      "eval_loss": 0.8463678956031799,
      "eval_runtime": 4.432,
      "eval_samples_per_second": 152.527,
      "eval_steps_per_second": 19.179,
      "step": 2332
    },
    {
      "epoch": 12.0,
      "learning_rate": 9.4e-05,
      "loss": 0.8965,
      "step": 2544
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.7349700927734375,
      "eval_runtime": 4.5659,
      "eval_samples_per_second": 148.054,
      "eval_steps_per_second": 18.616,
      "step": 2544
    },
    {
      "epoch": 13.0,
      "learning_rate": 9.350000000000001e-05,
      "loss": 0.7613,
      "step": 2756
    },
    {
      "epoch": 13.0,
      "eval_loss": 0.6494072675704956,
      "eval_runtime": 4.499,
      "eval_samples_per_second": 150.254,
      "eval_steps_per_second": 18.893,
      "step": 2756
    },
    {
      "epoch": 14.0,
      "learning_rate": 9.300000000000001e-05,
      "loss": 0.645,
      "step": 2968
    },
    {
      "epoch": 14.0,
      "eval_loss": 0.5844011902809143,
      "eval_runtime": 4.7455,
      "eval_samples_per_second": 142.449,
      "eval_steps_per_second": 17.912,
      "step": 2968
    },
    {
      "epoch": 15.0,
      "learning_rate": 9.250000000000001e-05,
      "loss": 0.548,
      "step": 3180
    },
    {
      "epoch": 15.0,
      "eval_loss": 0.5139679312705994,
      "eval_runtime": 4.7986,
      "eval_samples_per_second": 140.876,
      "eval_steps_per_second": 17.714,
      "step": 3180
    },
    {
      "epoch": 16.0,
      "learning_rate": 9.200000000000001e-05,
      "loss": 0.4564,
      "step": 3392
    },
    {
      "epoch": 16.0,
      "eval_loss": 0.47241923213005066,
      "eval_runtime": 4.6502,
      "eval_samples_per_second": 145.371,
      "eval_steps_per_second": 18.279,
      "step": 3392
    },
    {
      "epoch": 17.0,
      "learning_rate": 9.15e-05,
      "loss": 0.3785,
      "step": 3604
    },
    {
      "epoch": 17.0,
      "eval_loss": 0.4442097246646881,
      "eval_runtime": 4.6864,
      "eval_samples_per_second": 144.247,
      "eval_steps_per_second": 18.138,
      "step": 3604
    },
    {
      "epoch": 18.0,
      "learning_rate": 9.1e-05,
      "loss": 0.3144,
      "step": 3816
    },
    {
      "epoch": 18.0,
      "eval_loss": 0.407890647649765,
      "eval_runtime": 4.6905,
      "eval_samples_per_second": 144.121,
      "eval_steps_per_second": 18.122,
      "step": 3816
    },
    {
      "epoch": 19.0,
      "learning_rate": 9.05e-05,
      "loss": 0.2609,
      "step": 4028
    },
    {
      "epoch": 19.0,
      "eval_loss": 0.3970014452934265,
      "eval_runtime": 4.6686,
      "eval_samples_per_second": 144.799,
      "eval_steps_per_second": 18.207,
      "step": 4028
    },
    {
      "epoch": 20.0,
      "learning_rate": 9e-05,
      "loss": 0.2106,
      "step": 4240
    },
    {
      "epoch": 20.0,
      "eval_loss": 0.37661823630332947,
      "eval_runtime": 4.7062,
      "eval_samples_per_second": 143.641,
      "eval_steps_per_second": 18.061,
      "step": 4240
    },
    {
      "epoch": 21.0,
      "learning_rate": 8.950000000000001e-05,
      "loss": 0.1727,
      "step": 4452
    },
    {
      "epoch": 21.0,
      "eval_loss": 0.37585482001304626,
      "eval_runtime": 4.8302,
      "eval_samples_per_second": 139.953,
      "eval_steps_per_second": 17.598,
      "step": 4452
    },
    {
      "epoch": 22.0,
      "learning_rate": 8.900000000000001e-05,
      "loss": 0.1427,
      "step": 4664
    },
    {
      "epoch": 22.0,
      "eval_loss": 0.35890331864356995,
      "eval_runtime": 4.823,
      "eval_samples_per_second": 140.16,
      "eval_steps_per_second": 17.624,
      "step": 4664
    },
    {
      "epoch": 23.0,
      "learning_rate": 8.850000000000001e-05,
      "loss": 0.1115,
      "step": 4876
    },
    {
      "epoch": 23.0,
      "eval_loss": 0.35352450609207153,
      "eval_runtime": 4.9556,
      "eval_samples_per_second": 136.411,
      "eval_steps_per_second": 17.152,
      "step": 4876
    },
    {
      "epoch": 24.0,
      "learning_rate": 8.800000000000001e-05,
      "loss": 0.0949,
      "step": 5088
    },
    {
      "epoch": 24.0,
      "eval_loss": 0.3508472442626953,
      "eval_runtime": 4.4886,
      "eval_samples_per_second": 150.605,
      "eval_steps_per_second": 18.937,
      "step": 5088
    },
    {
      "epoch": 25.0,
      "learning_rate": 8.75e-05,
      "loss": 0.0779,
      "step": 5300
    },
    {
      "epoch": 25.0,
      "eval_loss": 0.35696595907211304,
      "eval_runtime": 4.5496,
      "eval_samples_per_second": 148.586,
      "eval_steps_per_second": 18.683,
      "step": 5300
    },
    {
      "epoch": 26.0,
      "learning_rate": 8.7e-05,
      "loss": 0.0652,
      "step": 5512
    },
    {
      "epoch": 26.0,
      "eval_loss": 0.3560032546520233,
      "eval_runtime": 5.7794,
      "eval_samples_per_second": 116.966,
      "eval_steps_per_second": 14.707,
      "step": 5512
    },
    {
      "epoch": 27.0,
      "learning_rate": 8.65e-05,
      "loss": 0.0549,
      "step": 5724
    },
    {
      "epoch": 27.0,
      "eval_loss": 0.355691134929657,
      "eval_runtime": 5.1711,
      "eval_samples_per_second": 130.726,
      "eval_steps_per_second": 16.437,
      "step": 5724
    },
    {
      "epoch": 28.0,
      "learning_rate": 8.6e-05,
      "loss": 0.0431,
      "step": 5936
    },
    {
      "epoch": 28.0,
      "eval_loss": 0.3634819984436035,
      "eval_runtime": 4.8044,
      "eval_samples_per_second": 140.705,
      "eval_steps_per_second": 17.692,
      "step": 5936
    },
    {
      "epoch": 29.0,
      "learning_rate": 8.55e-05,
      "loss": 0.0389,
      "step": 6148
    },
    {
      "epoch": 29.0,
      "eval_loss": 0.36610594391822815,
      "eval_runtime": 4.5861,
      "eval_samples_per_second": 147.403,
      "eval_steps_per_second": 18.534,
      "step": 6148
    },
    {
      "epoch": 30.0,
      "learning_rate": 8.5e-05,
      "loss": 0.0327,
      "step": 6360
    },
    {
      "epoch": 30.0,
      "eval_loss": 0.3725152611732483,
      "eval_runtime": 5.6392,
      "eval_samples_per_second": 119.874,
      "eval_steps_per_second": 15.073,
      "step": 6360
    },
    {
      "epoch": 31.0,
      "learning_rate": 8.450000000000001e-05,
      "loss": 0.0293,
      "step": 6572
    },
    {
      "epoch": 31.0,
      "eval_loss": 0.38449323177337646,
      "eval_runtime": 5.8725,
      "eval_samples_per_second": 115.113,
      "eval_steps_per_second": 14.474,
      "step": 6572
    },
    {
      "epoch": 32.0,
      "learning_rate": 8.4e-05,
      "loss": 0.0279,
      "step": 6784
    },
    {
      "epoch": 32.0,
      "eval_loss": 0.3865319490432739,
      "eval_runtime": 4.6302,
      "eval_samples_per_second": 145.997,
      "eval_steps_per_second": 18.358,
      "step": 6784
    },
    {
      "epoch": 33.0,
      "learning_rate": 8.35e-05,
      "loss": 0.0219,
      "step": 6996
    },
    {
      "epoch": 33.0,
      "eval_loss": 0.39288753271102905,
      "eval_runtime": 4.7491,
      "eval_samples_per_second": 142.342,
      "eval_steps_per_second": 17.898,
      "step": 6996
    },
    {
      "epoch": 34.0,
      "learning_rate": 8.3e-05,
      "loss": 0.02,
      "step": 7208
    },
    {
      "epoch": 34.0,
      "eval_loss": 0.39614447951316833,
      "eval_runtime": 4.6865,
      "eval_samples_per_second": 144.245,
      "eval_steps_per_second": 18.137,
      "step": 7208
    },
    {
      "epoch": 35.0,
      "learning_rate": 8.25e-05,
      "loss": 0.0166,
      "step": 7420
    },
    {
      "epoch": 35.0,
      "eval_loss": 0.3927687704563141,
      "eval_runtime": 4.4645,
      "eval_samples_per_second": 151.418,
      "eval_steps_per_second": 19.039,
      "step": 7420
    }
  ],
  "logging_steps": 500,
  "max_steps": 42400,
  "num_train_epochs": 200,
  "save_steps": 500,
  "total_flos": 9983766957232464.0,
  "trial_name": null,
  "trial_params": null
}