{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 57.0,
  "eval_steps": 500,
  "global_step": 12084,
  "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": 6.3597,
      "step": 212
    },
    {
      "epoch": 1.0,
      "eval_loss": 5.853069305419922,
      "eval_runtime": 3.5937,
      "eval_samples_per_second": 188.106,
      "eval_steps_per_second": 23.652,
      "step": 212
    },
    {
      "epoch": 2.0,
      "learning_rate": 9.900000000000001e-05,
      "loss": 5.5629,
      "step": 424
    },
    {
      "epoch": 2.0,
      "eval_loss": 4.915834903717041,
      "eval_runtime": 3.704,
      "eval_samples_per_second": 182.506,
      "eval_steps_per_second": 22.948,
      "step": 424
    },
    {
      "epoch": 3.0,
      "learning_rate": 9.850000000000001e-05,
      "loss": 4.8228,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_loss": 4.177055835723877,
      "eval_runtime": 3.7104,
      "eval_samples_per_second": 182.193,
      "eval_steps_per_second": 22.909,
      "step": 636
    },
    {
      "epoch": 4.0,
      "learning_rate": 9.8e-05,
      "loss": 4.1035,
      "step": 848
    },
    {
      "epoch": 4.0,
      "eval_loss": 3.462527275085449,
      "eval_runtime": 3.1205,
      "eval_samples_per_second": 216.633,
      "eval_steps_per_second": 27.239,
      "step": 848
    },
    {
      "epoch": 5.0,
      "learning_rate": 9.75e-05,
      "loss": 3.467,
      "step": 1060
    },
    {
      "epoch": 5.0,
      "eval_loss": 2.8398332595825195,
      "eval_runtime": 3.3235,
      "eval_samples_per_second": 203.402,
      "eval_steps_per_second": 25.576,
      "step": 1060
    },
    {
      "epoch": 6.0,
      "learning_rate": 9.7e-05,
      "loss": 2.8887,
      "step": 1272
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.3201420307159424,
      "eval_runtime": 3.4255,
      "eval_samples_per_second": 197.342,
      "eval_steps_per_second": 24.814,
      "step": 1272
    },
    {
      "epoch": 7.0,
      "learning_rate": 9.65e-05,
      "loss": 2.4003,
      "step": 1484
    },
    {
      "epoch": 7.0,
      "eval_loss": 1.8698177337646484,
      "eval_runtime": 3.5218,
      "eval_samples_per_second": 191.945,
      "eval_steps_per_second": 24.135,
      "step": 1484
    },
    {
      "epoch": 8.0,
      "learning_rate": 9.6e-05,
      "loss": 1.9919,
      "step": 1696
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.532776951789856,
      "eval_runtime": 3.5514,
      "eval_samples_per_second": 190.345,
      "eval_steps_per_second": 23.934,
      "step": 1696
    },
    {
      "epoch": 9.0,
      "learning_rate": 9.55e-05,
      "loss": 1.6656,
      "step": 1908
    },
    {
      "epoch": 9.0,
      "eval_loss": 1.2841078042984009,
      "eval_runtime": 3.622,
      "eval_samples_per_second": 186.637,
      "eval_steps_per_second": 23.468,
      "step": 1908
    },
    {
      "epoch": 10.0,
      "learning_rate": 9.5e-05,
      "loss": 1.4036,
      "step": 2120
    },
    {
      "epoch": 10.0,
      "eval_loss": 1.107720971107483,
      "eval_runtime": 3.6471,
      "eval_samples_per_second": 185.353,
      "eval_steps_per_second": 23.306,
      "step": 2120
    },
    {
      "epoch": 11.0,
      "learning_rate": 9.449999999999999e-05,
      "loss": 1.2091,
      "step": 2332
    },
    {
      "epoch": 11.0,
      "eval_loss": 0.9709760546684265,
      "eval_runtime": 3.7402,
      "eval_samples_per_second": 180.741,
      "eval_steps_per_second": 22.726,
      "step": 2332
    },
    {
      "epoch": 12.0,
      "learning_rate": 9.4e-05,
      "loss": 1.0379,
      "step": 2544
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.8528622984886169,
      "eval_runtime": 3.734,
      "eval_samples_per_second": 181.041,
      "eval_steps_per_second": 22.764,
      "step": 2544
    },
    {
      "epoch": 13.0,
      "learning_rate": 9.350000000000001e-05,
      "loss": 0.8994,
      "step": 2756
    },
    {
      "epoch": 13.0,
      "eval_loss": 0.7617917060852051,
      "eval_runtime": 3.7153,
      "eval_samples_per_second": 181.95,
      "eval_steps_per_second": 22.878,
      "step": 2756
    },
    {
      "epoch": 14.0,
      "learning_rate": 9.300000000000001e-05,
      "loss": 0.7828,
      "step": 2968
    },
    {
      "epoch": 14.0,
      "eval_loss": 0.6774648427963257,
      "eval_runtime": 3.6994,
      "eval_samples_per_second": 182.734,
      "eval_steps_per_second": 22.977,
      "step": 2968
    },
    {
      "epoch": 15.0,
      "learning_rate": 9.250000000000001e-05,
      "loss": 0.6788,
      "step": 3180
    },
    {
      "epoch": 15.0,
      "eval_loss": 0.6361714005470276,
      "eval_runtime": 3.6581,
      "eval_samples_per_second": 184.796,
      "eval_steps_per_second": 23.236,
      "step": 3180
    },
    {
      "epoch": 16.0,
      "learning_rate": 9.200000000000001e-05,
      "loss": 0.5819,
      "step": 3392
    },
    {
      "epoch": 16.0,
      "eval_loss": 0.5761357545852661,
      "eval_runtime": 3.7378,
      "eval_samples_per_second": 180.853,
      "eval_steps_per_second": 22.74,
      "step": 3392
    },
    {
      "epoch": 17.0,
      "learning_rate": 9.15e-05,
      "loss": 0.5104,
      "step": 3604
    },
    {
      "epoch": 17.0,
      "eval_loss": 0.5219044089317322,
      "eval_runtime": 3.7453,
      "eval_samples_per_second": 180.495,
      "eval_steps_per_second": 22.695,
      "step": 3604
    },
    {
      "epoch": 18.0,
      "learning_rate": 9.1e-05,
      "loss": 0.4343,
      "step": 3816
    },
    {
      "epoch": 18.0,
      "eval_loss": 0.4844624102115631,
      "eval_runtime": 3.7695,
      "eval_samples_per_second": 179.335,
      "eval_steps_per_second": 22.55,
      "step": 3816
    },
    {
      "epoch": 19.0,
      "learning_rate": 9.05e-05,
      "loss": 0.3784,
      "step": 4028
    },
    {
      "epoch": 19.0,
      "eval_loss": 0.44724151492118835,
      "eval_runtime": 3.7703,
      "eval_samples_per_second": 179.296,
      "eval_steps_per_second": 22.545,
      "step": 4028
    },
    {
      "epoch": 20.0,
      "learning_rate": 9e-05,
      "loss": 0.3224,
      "step": 4240
    },
    {
      "epoch": 20.0,
      "eval_loss": 0.45848363637924194,
      "eval_runtime": 3.6822,
      "eval_samples_per_second": 183.584,
      "eval_steps_per_second": 23.084,
      "step": 4240
    },
    {
      "epoch": 21.0,
      "learning_rate": 8.950000000000001e-05,
      "loss": 0.2768,
      "step": 4452
    },
    {
      "epoch": 21.0,
      "eval_loss": 0.4251560866832733,
      "eval_runtime": 3.9348,
      "eval_samples_per_second": 171.802,
      "eval_steps_per_second": 21.602,
      "step": 4452
    },
    {
      "epoch": 22.0,
      "learning_rate": 8.900000000000001e-05,
      "loss": 0.2269,
      "step": 4664
    },
    {
      "epoch": 22.0,
      "eval_loss": 0.39140933752059937,
      "eval_runtime": 3.6907,
      "eval_samples_per_second": 183.165,
      "eval_steps_per_second": 23.031,
      "step": 4664
    },
    {
      "epoch": 23.0,
      "learning_rate": 8.850000000000001e-05,
      "loss": 0.1868,
      "step": 4876
    },
    {
      "epoch": 23.0,
      "eval_loss": 0.4142400920391083,
      "eval_runtime": 3.8669,
      "eval_samples_per_second": 174.815,
      "eval_steps_per_second": 21.981,
      "step": 4876
    },
    {
      "epoch": 24.0,
      "learning_rate": 8.800000000000001e-05,
      "loss": 0.1539,
      "step": 5088
    },
    {
      "epoch": 24.0,
      "eval_loss": 0.3828276991844177,
      "eval_runtime": 3.7873,
      "eval_samples_per_second": 178.493,
      "eval_steps_per_second": 22.444,
      "step": 5088
    },
    {
      "epoch": 25.0,
      "learning_rate": 8.75e-05,
      "loss": 0.1298,
      "step": 5300
    },
    {
      "epoch": 25.0,
      "eval_loss": 0.40518617630004883,
      "eval_runtime": 3.9459,
      "eval_samples_per_second": 171.319,
      "eval_steps_per_second": 21.542,
      "step": 5300
    },
    {
      "epoch": 26.0,
      "learning_rate": 8.7e-05,
      "loss": 0.1098,
      "step": 5512
    },
    {
      "epoch": 26.0,
      "eval_loss": 0.420820027589798,
      "eval_runtime": 3.7042,
      "eval_samples_per_second": 182.497,
      "eval_steps_per_second": 22.947,
      "step": 5512
    },
    {
      "epoch": 27.0,
      "learning_rate": 8.65e-05,
      "loss": 0.0905,
      "step": 5724
    },
    {
      "epoch": 27.0,
      "eval_loss": 0.39528989791870117,
      "eval_runtime": 3.784,
      "eval_samples_per_second": 178.645,
      "eval_steps_per_second": 22.463,
      "step": 5724
    },
    {
      "epoch": 28.0,
      "learning_rate": 8.6e-05,
      "loss": 0.0731,
      "step": 5936
    },
    {
      "epoch": 28.0,
      "eval_loss": 0.41354966163635254,
      "eval_runtime": 3.7338,
      "eval_samples_per_second": 181.047,
      "eval_steps_per_second": 22.765,
      "step": 5936
    },
    {
      "epoch": 29.0,
      "learning_rate": 8.55e-05,
      "loss": 0.063,
      "step": 6148
    },
    {
      "epoch": 29.0,
      "eval_loss": 0.39981818199157715,
      "eval_runtime": 3.6841,
      "eval_samples_per_second": 183.489,
      "eval_steps_per_second": 23.072,
      "step": 6148
    },
    {
      "epoch": 30.0,
      "learning_rate": 8.5e-05,
      "loss": 0.0543,
      "step": 6360
    },
    {
      "epoch": 30.0,
      "eval_loss": 0.4148443043231964,
      "eval_runtime": 3.9507,
      "eval_samples_per_second": 171.111,
      "eval_steps_per_second": 21.515,
      "step": 6360
    },
    {
      "epoch": 31.0,
      "learning_rate": 8.450000000000001e-05,
      "loss": 0.0452,
      "step": 6572
    },
    {
      "epoch": 31.0,
      "eval_loss": 0.4315831959247589,
      "eval_runtime": 4.0598,
      "eval_samples_per_second": 166.511,
      "eval_steps_per_second": 20.937,
      "step": 6572
    },
    {
      "epoch": 32.0,
      "learning_rate": 8.4e-05,
      "loss": 0.0411,
      "step": 6784
    },
    {
      "epoch": 32.0,
      "eval_loss": 0.4563623070716858,
      "eval_runtime": 3.7904,
      "eval_samples_per_second": 178.346,
      "eval_steps_per_second": 22.425,
      "step": 6784
    },
    {
      "epoch": 33.0,
      "learning_rate": 8.35e-05,
      "loss": 0.0342,
      "step": 6996
    },
    {
      "epoch": 33.0,
      "eval_loss": 0.4423569142818451,
      "eval_runtime": 3.6613,
      "eval_samples_per_second": 184.635,
      "eval_steps_per_second": 23.216,
      "step": 6996
    },
    {
      "epoch": 34.0,
      "learning_rate": 8.3e-05,
      "loss": 0.0285,
      "step": 7208
    },
    {
      "epoch": 34.0,
      "eval_loss": 0.45883628726005554,
      "eval_runtime": 3.8854,
      "eval_samples_per_second": 173.985,
      "eval_steps_per_second": 21.877,
      "step": 7208
    },
    {
      "epoch": 35.0,
      "learning_rate": 8.25e-05,
      "loss": 0.0261,
      "step": 7420
    },
    {
      "epoch": 35.0,
      "eval_loss": 0.4722530245780945,
      "eval_runtime": 3.9505,
      "eval_samples_per_second": 171.117,
      "eval_steps_per_second": 21.516,
      "step": 7420
    },
    {
      "epoch": 36.0,
      "learning_rate": 8.2e-05,
      "loss": 0.0261,
      "step": 7632
    },
    {
      "epoch": 36.0,
      "eval_loss": 0.4807146191596985,
      "eval_runtime": 3.7918,
      "eval_samples_per_second": 178.279,
      "eval_steps_per_second": 22.417,
      "step": 7632
    },
    {
      "epoch": 37.0,
      "learning_rate": 8.15e-05,
      "loss": 0.0199,
      "step": 7844
    },
    {
      "epoch": 37.0,
      "eval_loss": 0.46007126569747925,
      "eval_runtime": 3.8954,
      "eval_samples_per_second": 173.536,
      "eval_steps_per_second": 21.82,
      "step": 7844
    },
    {
      "epoch": 38.0,
      "learning_rate": 8.1e-05,
      "loss": 0.0169,
      "step": 8056
    },
    {
      "epoch": 38.0,
      "eval_loss": 0.4729515314102173,
      "eval_runtime": 3.8647,
      "eval_samples_per_second": 174.918,
      "eval_steps_per_second": 21.994,
      "step": 8056
    },
    {
      "epoch": 39.0,
      "learning_rate": 8.05e-05,
      "loss": 0.0164,
      "step": 8268
    },
    {
      "epoch": 39.0,
      "eval_loss": 0.4991854131221771,
      "eval_runtime": 3.8642,
      "eval_samples_per_second": 174.937,
      "eval_steps_per_second": 21.997,
      "step": 8268
    },
    {
      "epoch": 40.0,
      "learning_rate": 8e-05,
      "loss": 0.0152,
      "step": 8480
    },
    {
      "epoch": 40.0,
      "eval_loss": 0.48734694719314575,
      "eval_runtime": 3.8857,
      "eval_samples_per_second": 173.971,
      "eval_steps_per_second": 21.875,
      "step": 8480
    },
    {
      "epoch": 41.0,
      "learning_rate": 7.950000000000001e-05,
      "loss": 0.0109,
      "step": 8692
    },
    {
      "epoch": 41.0,
      "eval_loss": 0.49513980746269226,
      "eval_runtime": 3.8249,
      "eval_samples_per_second": 176.736,
      "eval_steps_per_second": 22.223,
      "step": 8692
    },
    {
      "epoch": 42.0,
      "learning_rate": 7.900000000000001e-05,
      "loss": 0.0132,
      "step": 8904
    },
    {
      "epoch": 42.0,
      "eval_loss": 0.4910328984260559,
      "eval_runtime": 3.823,
      "eval_samples_per_second": 176.826,
      "eval_steps_per_second": 22.234,
      "step": 8904
    },
    {
      "epoch": 43.0,
      "learning_rate": 7.850000000000001e-05,
      "loss": 0.0086,
      "step": 9116
    },
    {
      "epoch": 43.0,
      "eval_loss": 0.5066259503364563,
      "eval_runtime": 3.8604,
      "eval_samples_per_second": 175.11,
      "eval_steps_per_second": 22.018,
      "step": 9116
    },
    {
      "epoch": 44.0,
      "learning_rate": 7.800000000000001e-05,
      "loss": 0.0084,
      "step": 9328
    },
    {
      "epoch": 44.0,
      "eval_loss": 0.5218311548233032,
      "eval_runtime": 3.8685,
      "eval_samples_per_second": 174.744,
      "eval_steps_per_second": 21.972,
      "step": 9328
    },
    {
      "epoch": 45.0,
      "learning_rate": 7.75e-05,
      "loss": 0.0082,
      "step": 9540
    },
    {
      "epoch": 45.0,
      "eval_loss": 0.5277370810508728,
      "eval_runtime": 3.8269,
      "eval_samples_per_second": 176.644,
      "eval_steps_per_second": 22.211,
      "step": 9540
    },
    {
      "epoch": 46.0,
      "learning_rate": 7.7e-05,
      "loss": 0.0094,
      "step": 9752
    },
    {
      "epoch": 46.0,
      "eval_loss": 0.5367192029953003,
      "eval_runtime": 3.8012,
      "eval_samples_per_second": 177.838,
      "eval_steps_per_second": 22.361,
      "step": 9752
    },
    {
      "epoch": 47.0,
      "learning_rate": 7.65e-05,
      "loss": 0.0117,
      "step": 9964
    },
    {
      "epoch": 47.0,
      "eval_loss": 0.5425814390182495,
      "eval_runtime": 3.8527,
      "eval_samples_per_second": 175.46,
      "eval_steps_per_second": 22.062,
      "step": 9964
    },
    {
      "epoch": 48.0,
      "learning_rate": 7.6e-05,
      "loss": 0.0073,
      "step": 10176
    },
    {
      "epoch": 48.0,
      "eval_loss": 0.5633374452590942,
      "eval_runtime": 3.8494,
      "eval_samples_per_second": 175.614,
      "eval_steps_per_second": 22.082,
      "step": 10176
    },
    {
      "epoch": 49.0,
      "learning_rate": 7.55e-05,
      "loss": 0.0064,
      "step": 10388
    },
    {
      "epoch": 49.0,
      "eval_loss": 0.558996319770813,
      "eval_runtime": 3.8307,
      "eval_samples_per_second": 176.467,
      "eval_steps_per_second": 22.189,
      "step": 10388
    },
    {
      "epoch": 50.0,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.0047,
      "step": 10600
    },
    {
      "epoch": 50.0,
      "eval_loss": 0.5562776923179626,
      "eval_runtime": 3.8234,
      "eval_samples_per_second": 176.804,
      "eval_steps_per_second": 22.231,
      "step": 10600
    },
    {
      "epoch": 51.0,
      "learning_rate": 7.450000000000001e-05,
      "loss": 0.0107,
      "step": 10812
    },
    {
      "epoch": 51.0,
      "eval_loss": 0.5780841708183289,
      "eval_runtime": 3.8575,
      "eval_samples_per_second": 175.242,
      "eval_steps_per_second": 22.035,
      "step": 10812
    },
    {
      "epoch": 52.0,
      "learning_rate": 7.4e-05,
      "loss": 0.0084,
      "step": 11024
    },
    {
      "epoch": 52.0,
      "eval_loss": 0.5766631364822388,
      "eval_runtime": 3.9223,
      "eval_samples_per_second": 172.35,
      "eval_steps_per_second": 21.671,
      "step": 11024
    },
    {
      "epoch": 53.0,
      "learning_rate": 7.35e-05,
      "loss": 0.005,
      "step": 11236
    },
    {
      "epoch": 53.0,
      "eval_loss": 0.5684217810630798,
      "eval_runtime": 3.8567,
      "eval_samples_per_second": 175.281,
      "eval_steps_per_second": 22.04,
      "step": 11236
    },
    {
      "epoch": 54.0,
      "learning_rate": 7.3e-05,
      "loss": 0.0036,
      "step": 11448
    },
    {
      "epoch": 54.0,
      "eval_loss": 0.5706903338432312,
      "eval_runtime": 3.8702,
      "eval_samples_per_second": 174.67,
      "eval_steps_per_second": 21.963,
      "step": 11448
    },
    {
      "epoch": 55.0,
      "learning_rate": 7.25e-05,
      "loss": 0.0049,
      "step": 11660
    },
    {
      "epoch": 55.0,
      "eval_loss": 0.6319368481636047,
      "eval_runtime": 3.8576,
      "eval_samples_per_second": 175.24,
      "eval_steps_per_second": 22.035,
      "step": 11660
    },
    {
      "epoch": 56.0,
      "learning_rate": 7.2e-05,
      "loss": 0.0104,
      "step": 11872
    },
    {
      "epoch": 56.0,
      "eval_loss": 0.5811020731925964,
      "eval_runtime": 3.8577,
      "eval_samples_per_second": 175.232,
      "eval_steps_per_second": 22.034,
      "step": 11872
    },
    {
      "epoch": 57.0,
      "learning_rate": 7.15e-05,
      "loss": 0.0062,
      "step": 12084
    },
    {
      "epoch": 57.0,
      "eval_loss": 0.5707266330718994,
      "eval_runtime": 3.9635,
      "eval_samples_per_second": 170.555,
      "eval_steps_per_second": 21.446,
      "step": 12084
    }
  ],
  "logging_steps": 500,
  "max_steps": 42400,
  "num_train_epochs": 200,
  "save_steps": 500,
  "total_flos": 1.5218906197855332e+16,
  "trial_name": null,
  "trial_params": null
}