{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 60.0,
  "global_step": 18720,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4657039711191336,
      "eval_loss": 0.07504455745220184,
      "eval_runtime": 4.3586,
      "eval_samples_per_second": 63.552,
      "eval_steps_per_second": 8.03,
      "step": 312
    },
    {
      "best_epoch": 0,
      "best_eval_accuracy": 0.4657039711191336,
      "epoch": 1.0,
      "step": 312
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0019465811965811966,
      "loss": 0.1248,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5342960288808665,
      "eval_loss": 0.07155032455921173,
      "eval_runtime": 4.4541,
      "eval_samples_per_second": 62.19,
      "eval_steps_per_second": 7.858,
      "step": 624
    },
    {
      "best_epoch": 1,
      "best_eval_accuracy": 0.5342960288808665,
      "epoch": 2.0,
      "step": 624
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.14003151655197144,
      "eval_runtime": 4.4833,
      "eval_samples_per_second": 61.785,
      "eval_steps_per_second": 7.807,
      "step": 936
    },
    {
      "best_epoch": 1,
      "best_eval_accuracy": 0.5342960288808665,
      "epoch": 3.0,
      "step": 936
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0018931623931623931,
      "loss": 0.0956,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.12427954375743866,
      "eval_runtime": 4.5014,
      "eval_samples_per_second": 61.536,
      "eval_steps_per_second": 7.775,
      "step": 1248
    },
    {
      "best_epoch": 1,
      "best_eval_accuracy": 0.5342960288808665,
      "epoch": 4.0,
      "step": 1248
    },
    {
      "epoch": 4.81,
      "learning_rate": 0.0018397435897435897,
      "loss": 0.0927,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5667870036101083,
      "eval_loss": 0.07349628210067749,
      "eval_runtime": 4.5133,
      "eval_samples_per_second": 61.374,
      "eval_steps_per_second": 7.755,
      "step": 1560
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 5.0,
      "step": 1560
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.11670589447021484,
      "eval_runtime": 4.5097,
      "eval_samples_per_second": 61.423,
      "eval_steps_per_second": 7.761,
      "step": 1872
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 6.0,
      "step": 1872
    },
    {
      "epoch": 6.41,
      "learning_rate": 0.0017863247863247865,
      "loss": 0.0924,
      "step": 2000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.08031181991100311,
      "eval_runtime": 4.494,
      "eval_samples_per_second": 61.637,
      "eval_steps_per_second": 7.788,
      "step": 2184
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 7.0,
      "step": 2184
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.4981949458483754,
      "eval_loss": 0.07143595069646835,
      "eval_runtime": 4.4931,
      "eval_samples_per_second": 61.651,
      "eval_steps_per_second": 7.79,
      "step": 2496
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 8.0,
      "step": 2496
    },
    {
      "epoch": 8.01,
      "learning_rate": 0.001732905982905983,
      "loss": 0.0913,
      "step": 2500
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07295498251914978,
      "eval_runtime": 4.4955,
      "eval_samples_per_second": 61.618,
      "eval_steps_per_second": 7.786,
      "step": 2808
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 9.0,
      "step": 2808
    },
    {
      "epoch": 9.62,
      "learning_rate": 0.0016794871794871796,
      "loss": 0.0885,
      "step": 3000
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.5342960288808665,
      "eval_loss": 0.07082638889551163,
      "eval_runtime": 4.4971,
      "eval_samples_per_second": 61.595,
      "eval_steps_per_second": 7.783,
      "step": 3120
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 10.0,
      "step": 3120
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.49097472924187724,
      "eval_loss": 0.07016535848379135,
      "eval_runtime": 4.5004,
      "eval_samples_per_second": 61.55,
      "eval_steps_per_second": 7.777,
      "step": 3432
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 11.0,
      "step": 3432
    },
    {
      "epoch": 11.22,
      "learning_rate": 0.0016260683760683761,
      "loss": 0.0817,
      "step": 3500
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.5306859205776173,
      "eval_loss": 0.07026981562376022,
      "eval_runtime": 4.4966,
      "eval_samples_per_second": 61.602,
      "eval_steps_per_second": 7.784,
      "step": 3744
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 12.0,
      "step": 3744
    },
    {
      "epoch": 12.82,
      "learning_rate": 0.0015726495726495727,
      "loss": 0.0789,
      "step": 4000
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07234571874141693,
      "eval_runtime": 4.4954,
      "eval_samples_per_second": 61.619,
      "eval_steps_per_second": 7.786,
      "step": 4056
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 13.0,
      "step": 4056
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.48736462093862815,
      "eval_loss": 0.07004748284816742,
      "eval_runtime": 4.4985,
      "eval_samples_per_second": 61.576,
      "eval_steps_per_second": 7.78,
      "step": 4368
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 14.0,
      "step": 4368
    },
    {
      "epoch": 14.42,
      "learning_rate": 0.0015192307692307692,
      "loss": 0.0785,
      "step": 4500
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.47653429602888087,
      "eval_loss": 0.06995787471532822,
      "eval_runtime": 4.4997,
      "eval_samples_per_second": 61.56,
      "eval_steps_per_second": 7.778,
      "step": 4680
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 15.0,
      "step": 4680
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.48014440433212996,
      "eval_loss": 0.0701388567686081,
      "eval_runtime": 4.4975,
      "eval_samples_per_second": 61.59,
      "eval_steps_per_second": 7.782,
      "step": 4992
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 16.0,
      "step": 4992
    },
    {
      "epoch": 16.03,
      "learning_rate": 0.0014658119658119658,
      "loss": 0.0788,
      "step": 5000
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.4548736462093863,
      "eval_loss": 0.07330285012722015,
      "eval_runtime": 4.5002,
      "eval_samples_per_second": 61.553,
      "eval_steps_per_second": 7.777,
      "step": 5304
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 17.0,
      "step": 5304
    },
    {
      "epoch": 17.63,
      "learning_rate": 0.0014123931623931626,
      "loss": 0.0926,
      "step": 5500
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.5306859205776173,
      "eval_loss": 0.0858425423502922,
      "eval_runtime": 4.5001,
      "eval_samples_per_second": 61.555,
      "eval_steps_per_second": 7.778,
      "step": 5616
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 18.0,
      "step": 5616
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.4981949458483754,
      "eval_loss": 0.07387208193540573,
      "eval_runtime": 4.5041,
      "eval_samples_per_second": 61.5,
      "eval_steps_per_second": 7.771,
      "step": 5928
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 19.0,
      "step": 5928
    },
    {
      "epoch": 19.23,
      "learning_rate": 0.0013589743589743591,
      "loss": 0.0845,
      "step": 6000
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.5234657039711191,
      "eval_loss": 0.09436681866645813,
      "eval_runtime": 4.4999,
      "eval_samples_per_second": 61.556,
      "eval_steps_per_second": 7.778,
      "step": 6240
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 20.0,
      "step": 6240
    },
    {
      "epoch": 20.83,
      "learning_rate": 0.0013055555555555557,
      "loss": 0.0826,
      "step": 6500
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.4620938628158845,
      "eval_loss": 0.07167434692382812,
      "eval_runtime": 4.5025,
      "eval_samples_per_second": 61.521,
      "eval_steps_per_second": 7.773,
      "step": 6552
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 21.0,
      "step": 6552
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.0710165873169899,
      "eval_runtime": 4.506,
      "eval_samples_per_second": 61.473,
      "eval_steps_per_second": 7.767,
      "step": 6864
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 22.0,
      "step": 6864
    },
    {
      "epoch": 22.44,
      "learning_rate": 0.0012521367521367522,
      "loss": 0.0818,
      "step": 7000
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.49097472924187724,
      "eval_loss": 0.07119692116975784,
      "eval_runtime": 4.5019,
      "eval_samples_per_second": 61.529,
      "eval_steps_per_second": 7.774,
      "step": 7176
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 23.0,
      "step": 7176
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.48375451263537905,
      "eval_loss": 0.0713801309466362,
      "eval_runtime": 4.5015,
      "eval_samples_per_second": 61.535,
      "eval_steps_per_second": 7.775,
      "step": 7488
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 24.0,
      "step": 7488
    },
    {
      "epoch": 24.04,
      "learning_rate": 0.0011987179487179488,
      "loss": 0.0809,
      "step": 7500
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.5126353790613718,
      "eval_loss": 0.07450491935014725,
      "eval_runtime": 4.5009,
      "eval_samples_per_second": 61.543,
      "eval_steps_per_second": 7.776,
      "step": 7800
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 25.0,
      "step": 7800
    },
    {
      "epoch": 25.64,
      "learning_rate": 0.0011452991452991453,
      "loss": 0.0805,
      "step": 8000
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07136508822441101,
      "eval_runtime": 4.5023,
      "eval_samples_per_second": 61.524,
      "eval_steps_per_second": 7.774,
      "step": 8112
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 26.0,
      "step": 8112
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.0737999975681305,
      "eval_runtime": 4.504,
      "eval_samples_per_second": 61.501,
      "eval_steps_per_second": 7.771,
      "step": 8424
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 27.0,
      "step": 8424
    },
    {
      "epoch": 27.24,
      "learning_rate": 0.001091880341880342,
      "loss": 0.0805,
      "step": 8500
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.47653429602888087,
      "eval_loss": 0.07094266265630722,
      "eval_runtime": 4.4965,
      "eval_samples_per_second": 61.603,
      "eval_steps_per_second": 7.784,
      "step": 8736
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 28.0,
      "step": 8736
    },
    {
      "epoch": 28.85,
      "learning_rate": 0.0010384615384615387,
      "loss": 0.0809,
      "step": 9000
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07367567718029022,
      "eval_runtime": 4.4933,
      "eval_samples_per_second": 61.647,
      "eval_steps_per_second": 7.789,
      "step": 9048
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 29.0,
      "step": 9048
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.5595667870036101,
      "eval_loss": 0.07294340431690216,
      "eval_runtime": 4.4987,
      "eval_samples_per_second": 61.573,
      "eval_steps_per_second": 7.78,
      "step": 9360
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 30.0,
      "step": 9360
    },
    {
      "epoch": 30.45,
      "learning_rate": 0.0009850427350427352,
      "loss": 0.0797,
      "step": 9500
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.5595667870036101,
      "eval_loss": 0.07358906418085098,
      "eval_runtime": 4.4995,
      "eval_samples_per_second": 61.563,
      "eval_steps_per_second": 7.779,
      "step": 9672
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 31.0,
      "step": 9672
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.4693140794223827,
      "eval_loss": 0.07046783715486526,
      "eval_runtime": 4.4971,
      "eval_samples_per_second": 61.596,
      "eval_steps_per_second": 7.783,
      "step": 9984
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 32.0,
      "step": 9984
    },
    {
      "epoch": 32.05,
      "learning_rate": 0.0009316239316239317,
      "loss": 0.08,
      "step": 10000
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.4657039711191336,
      "eval_loss": 0.07109773904085159,
      "eval_runtime": 4.5029,
      "eval_samples_per_second": 61.516,
      "eval_steps_per_second": 7.773,
      "step": 10296
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 33.0,
      "step": 10296
    },
    {
      "epoch": 33.65,
      "learning_rate": 0.0008782051282051282,
      "loss": 0.0798,
      "step": 10500
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.51985559566787,
      "eval_loss": 0.07305511832237244,
      "eval_runtime": 4.5017,
      "eval_samples_per_second": 61.532,
      "eval_steps_per_second": 7.775,
      "step": 10608
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 34.0,
      "step": 10608
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07443650811910629,
      "eval_runtime": 4.505,
      "eval_samples_per_second": 61.488,
      "eval_steps_per_second": 7.769,
      "step": 10920
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 35.0,
      "step": 10920
    },
    {
      "epoch": 35.26,
      "learning_rate": 0.0008247863247863248,
      "loss": 0.0795,
      "step": 11000
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.072080597281456,
      "eval_runtime": 4.5002,
      "eval_samples_per_second": 61.553,
      "eval_steps_per_second": 7.778,
      "step": 11232
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 36.0,
      "step": 11232
    },
    {
      "epoch": 36.86,
      "learning_rate": 0.0007713675213675214,
      "loss": 0.0796,
      "step": 11500
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.47653429602888087,
      "eval_loss": 0.07079458236694336,
      "eval_runtime": 4.506,
      "eval_samples_per_second": 61.473,
      "eval_steps_per_second": 7.767,
      "step": 11544
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 37.0,
      "step": 11544
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07135061919689178,
      "eval_runtime": 4.5013,
      "eval_samples_per_second": 61.538,
      "eval_steps_per_second": 7.776,
      "step": 11856
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 38.0,
      "step": 11856
    },
    {
      "epoch": 38.46,
      "learning_rate": 0.000717948717948718,
      "loss": 0.0792,
      "step": 12000
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07073673605918884,
      "eval_runtime": 4.4981,
      "eval_samples_per_second": 61.582,
      "eval_steps_per_second": 7.781,
      "step": 12168
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 39.0,
      "step": 12168
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.4693140794223827,
      "eval_loss": 0.0705353394150734,
      "eval_runtime": 4.5016,
      "eval_samples_per_second": 61.534,
      "eval_steps_per_second": 7.775,
      "step": 12480
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 40.0,
      "step": 12480
    },
    {
      "epoch": 40.06,
      "learning_rate": 0.0006645299145299145,
      "loss": 0.0785,
      "step": 12500
    },
    {
      "epoch": 41.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07056449353694916,
      "eval_runtime": 4.521,
      "eval_samples_per_second": 61.27,
      "eval_steps_per_second": 7.742,
      "step": 12792
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 41.0,
      "step": 12792
    },
    {
      "epoch": 41.67,
      "learning_rate": 0.0006111111111111112,
      "loss": 0.0782,
      "step": 13000
    },
    {
      "epoch": 42.0,
      "eval_accuracy": 0.47653429602888087,
      "eval_loss": 0.0707937479019165,
      "eval_runtime": 4.4933,
      "eval_samples_per_second": 61.647,
      "eval_steps_per_second": 7.789,
      "step": 13104
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 42.0,
      "step": 13104
    },
    {
      "epoch": 43.0,
      "eval_accuracy": 0.47653429602888087,
      "eval_loss": 0.07090503722429276,
      "eval_runtime": 4.4923,
      "eval_samples_per_second": 61.66,
      "eval_steps_per_second": 7.791,
      "step": 13416
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 43.0,
      "step": 13416
    },
    {
      "epoch": 43.27,
      "learning_rate": 0.0005576923076923078,
      "loss": 0.0779,
      "step": 13500
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07051144540309906,
      "eval_runtime": 4.4928,
      "eval_samples_per_second": 61.654,
      "eval_steps_per_second": 7.79,
      "step": 13728
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 44.0,
      "step": 13728
    },
    {
      "epoch": 44.87,
      "learning_rate": 0.0005042735042735043,
      "loss": 0.078,
      "step": 14000
    },
    {
      "epoch": 45.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07053770124912262,
      "eval_runtime": 4.4949,
      "eval_samples_per_second": 61.625,
      "eval_steps_per_second": 7.787,
      "step": 14040
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 45.0,
      "step": 14040
    },
    {
      "epoch": 46.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07043004035949707,
      "eval_runtime": 4.4904,
      "eval_samples_per_second": 61.687,
      "eval_steps_per_second": 7.794,
      "step": 14352
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 46.0,
      "step": 14352
    },
    {
      "epoch": 46.47,
      "learning_rate": 0.00045085470085470087,
      "loss": 0.0776,
      "step": 14500
    },
    {
      "epoch": 47.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07083167135715485,
      "eval_runtime": 4.4905,
      "eval_samples_per_second": 61.686,
      "eval_steps_per_second": 7.794,
      "step": 14664
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 47.0,
      "step": 14664
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07037856429815292,
      "eval_runtime": 4.4905,
      "eval_samples_per_second": 61.686,
      "eval_steps_per_second": 7.794,
      "step": 14976
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 48.0,
      "step": 14976
    },
    {
      "epoch": 48.08,
      "learning_rate": 0.0003974358974358974,
      "loss": 0.0778,
      "step": 15000
    },
    {
      "epoch": 49.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07035572826862335,
      "eval_runtime": 4.487,
      "eval_samples_per_second": 61.733,
      "eval_steps_per_second": 7.8,
      "step": 15288
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 49.0,
      "step": 15288
    },
    {
      "epoch": 49.68,
      "learning_rate": 0.00034401709401709403,
      "loss": 0.0778,
      "step": 15500
    },
    {
      "epoch": 50.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07087621092796326,
      "eval_runtime": 4.4865,
      "eval_samples_per_second": 61.741,
      "eval_steps_per_second": 7.801,
      "step": 15600
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 50.0,
      "step": 15600
    },
    {
      "epoch": 51.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.0709025114774704,
      "eval_runtime": 4.4859,
      "eval_samples_per_second": 61.748,
      "eval_steps_per_second": 7.802,
      "step": 15912
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 51.0,
      "step": 15912
    },
    {
      "epoch": 51.28,
      "learning_rate": 0.00029059829059829064,
      "loss": 0.078,
      "step": 16000
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07118164002895355,
      "eval_runtime": 4.4885,
      "eval_samples_per_second": 61.714,
      "eval_steps_per_second": 7.798,
      "step": 16224
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 52.0,
      "step": 16224
    },
    {
      "epoch": 52.88,
      "learning_rate": 0.00023717948717948717,
      "loss": 0.0776,
      "step": 16500
    },
    {
      "epoch": 53.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.0703905001282692,
      "eval_runtime": 4.4924,
      "eval_samples_per_second": 61.66,
      "eval_steps_per_second": 7.791,
      "step": 16536
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 53.0,
      "step": 16536
    },
    {
      "epoch": 54.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07079659402370453,
      "eval_runtime": 4.4847,
      "eval_samples_per_second": 61.765,
      "eval_steps_per_second": 7.804,
      "step": 16848
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 54.0,
      "step": 16848
    },
    {
      "epoch": 54.49,
      "learning_rate": 0.00018376068376068378,
      "loss": 0.0772,
      "step": 17000
    },
    {
      "epoch": 55.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07165035605430603,
      "eval_runtime": 4.4905,
      "eval_samples_per_second": 61.685,
      "eval_steps_per_second": 7.794,
      "step": 17160
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 55.0,
      "step": 17160
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07045500725507736,
      "eval_runtime": 4.488,
      "eval_samples_per_second": 61.719,
      "eval_steps_per_second": 7.798,
      "step": 17472
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 56.0,
      "step": 17472
    },
    {
      "epoch": 56.09,
      "learning_rate": 0.00013034188034188036,
      "loss": 0.0772,
      "step": 17500
    },
    {
      "epoch": 57.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.070336252450943,
      "eval_runtime": 4.4882,
      "eval_samples_per_second": 61.718,
      "eval_steps_per_second": 7.798,
      "step": 17784
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 57.0,
      "step": 17784
    },
    {
      "epoch": 57.69,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.0774,
      "step": 18000
    },
    {
      "epoch": 58.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07103137671947479,
      "eval_runtime": 4.4808,
      "eval_samples_per_second": 61.82,
      "eval_steps_per_second": 7.811,
      "step": 18096
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 58.0,
      "step": 18096
    },
    {
      "epoch": 59.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07040625810623169,
      "eval_runtime": 4.466,
      "eval_samples_per_second": 62.024,
      "eval_steps_per_second": 7.837,
      "step": 18408
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 59.0,
      "step": 18408
    },
    {
      "epoch": 59.29,
      "learning_rate": 2.3504273504273504e-05,
      "loss": 0.0774,
      "step": 18500
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.07038838416337967,
      "eval_runtime": 4.4631,
      "eval_samples_per_second": 62.065,
      "eval_steps_per_second": 7.842,
      "step": 18720
    },
    {
      "best_epoch": 4,
      "best_eval_accuracy": 0.5667870036101083,
      "epoch": 60.0,
      "step": 18720
    },
    {
      "epoch": 60.0,
      "step": 18720,
      "total_flos": 6.96152728406016e+16,
      "train_loss": 0.08253046080597445,
      "train_runtime": 4087.2245,
      "train_samples_per_second": 36.553,
      "train_steps_per_second": 4.58
    }
  ],
  "max_steps": 18720,
  "num_train_epochs": 60,
  "total_flos": 6.96152728406016e+16,
  "trial_name": null,
  "trial_params": null
}