{
  "best_metric": 0.40238592824799724,
  "best_model_checkpoint": "/groups/claytonm/enoriega/kw_pubmed/kw_pubmed_5000_0.000006/checkpoint-435",
  "epoch": 3.795931045193353,
  "global_step": 585,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 5.968831168831169e-06,
      "loss": 4.3584,
      "step": 5
    },
    {
      "epoch": 0.06,
      "learning_rate": 5.937662337662337e-06,
      "loss": 4.0315,
      "step": 10
    },
    {
      "epoch": 0.1,
      "learning_rate": 5.8987012987012994e-06,
      "loss": 3.9357,
      "step": 15
    },
    {
      "epoch": 0.1,
      "eval_accuracy": 0.37917972831765934,
      "eval_loss": 3.8932790756225586,
      "eval_runtime": 16.4606,
      "eval_samples_per_second": 607.512,
      "eval_steps_per_second": 19.015,
      "step": 15
    },
    {
      "epoch": 0.13,
      "learning_rate": 5.85974025974026e-06,
      "loss": 3.8549,
      "step": 20
    },
    {
      "epoch": 0.16,
      "learning_rate": 5.820779220779221e-06,
      "loss": 3.7904,
      "step": 25
    },
    {
      "epoch": 0.19,
      "learning_rate": 5.781818181818182e-06,
      "loss": 3.7574,
      "step": 30
    },
    {
      "epoch": 0.19,
      "eval_accuracy": 0.38153082549634276,
      "eval_loss": 3.8384764194488525,
      "eval_runtime": 16.5354,
      "eval_samples_per_second": 604.762,
      "eval_steps_per_second": 18.929,
      "step": 30
    },
    {
      "epoch": 0.23,
      "learning_rate": 5.742857142857143e-06,
      "loss": 3.7476,
      "step": 35
    },
    {
      "epoch": 0.26,
      "learning_rate": 5.703896103896104e-06,
      "loss": 3.6722,
      "step": 40
    },
    {
      "epoch": 0.29,
      "learning_rate": 5.664935064935065e-06,
      "loss": 3.6652,
      "step": 45
    },
    {
      "epoch": 0.29,
      "eval_accuracy": 0.38283699059561127,
      "eval_loss": 3.812371015548706,
      "eval_runtime": 16.5574,
      "eval_samples_per_second": 603.961,
      "eval_steps_per_second": 18.904,
      "step": 45
    },
    {
      "epoch": 0.32,
      "learning_rate": 5.625974025974026e-06,
      "loss": 3.6491,
      "step": 50
    },
    {
      "epoch": 0.36,
      "learning_rate": 5.587012987012987e-06,
      "loss": 3.6096,
      "step": 55
    },
    {
      "epoch": 0.39,
      "learning_rate": 5.548051948051948e-06,
      "loss": 3.5893,
      "step": 60
    },
    {
      "epoch": 0.39,
      "eval_accuracy": 0.3845785440613027,
      "eval_loss": 3.786245584487915,
      "eval_runtime": 16.5595,
      "eval_samples_per_second": 603.881,
      "eval_steps_per_second": 18.901,
      "step": 60
    },
    {
      "epoch": 0.42,
      "learning_rate": 5.50909090909091e-06,
      "loss": 3.5818,
      "step": 65
    },
    {
      "epoch": 0.45,
      "learning_rate": 5.47012987012987e-06,
      "loss": 3.548,
      "step": 70
    },
    {
      "epoch": 0.49,
      "learning_rate": 5.431168831168831e-06,
      "loss": 3.5135,
      "step": 75
    },
    {
      "epoch": 0.49,
      "eval_accuracy": 0.389063044235458,
      "eval_loss": 3.7630181312561035,
      "eval_runtime": 17.3582,
      "eval_samples_per_second": 576.098,
      "eval_steps_per_second": 18.032,
      "step": 75
    },
    {
      "epoch": 0.52,
      "learning_rate": 5.3922077922077925e-06,
      "loss": 3.51,
      "step": 80
    },
    {
      "epoch": 0.55,
      "learning_rate": 5.353246753246754e-06,
      "loss": 3.502,
      "step": 85
    },
    {
      "epoch": 0.58,
      "learning_rate": 5.314285714285714e-06,
      "loss": 3.4776,
      "step": 90
    },
    {
      "epoch": 0.58,
      "eval_accuracy": 0.38788749564611635,
      "eval_loss": 3.7638540267944336,
      "eval_runtime": 17.4685,
      "eval_samples_per_second": 572.46,
      "eval_steps_per_second": 17.918,
      "step": 90
    },
    {
      "epoch": 0.61,
      "learning_rate": 5.275324675324675e-06,
      "loss": 3.455,
      "step": 95
    },
    {
      "epoch": 0.65,
      "learning_rate": 5.2363636363636365e-06,
      "loss": 3.4606,
      "step": 100
    },
    {
      "epoch": 0.68,
      "learning_rate": 5.197402597402598e-06,
      "loss": 3.4083,
      "step": 105
    },
    {
      "epoch": 0.68,
      "eval_accuracy": 0.3912835249042146,
      "eval_loss": 3.7579634189605713,
      "eval_runtime": 17.4161,
      "eval_samples_per_second": 574.181,
      "eval_steps_per_second": 17.972,
      "step": 105
    },
    {
      "epoch": 0.71,
      "learning_rate": 5.158441558441558e-06,
      "loss": 3.3844,
      "step": 110
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.119480519480519e-06,
      "loss": 3.4324,
      "step": 115
    },
    {
      "epoch": 0.78,
      "learning_rate": 5.0805194805194805e-06,
      "loss": 3.3791,
      "step": 120
    },
    {
      "epoch": 0.78,
      "eval_accuracy": 0.38784395680947403,
      "eval_loss": 3.7581143379211426,
      "eval_runtime": 17.4647,
      "eval_samples_per_second": 572.583,
      "eval_steps_per_second": 17.922,
      "step": 120
    },
    {
      "epoch": 0.81,
      "learning_rate": 5.041558441558442e-06,
      "loss": 3.3814,
      "step": 125
    },
    {
      "epoch": 0.84,
      "learning_rate": 5.002597402597402e-06,
      "loss": 3.354,
      "step": 130
    },
    {
      "epoch": 0.87,
      "learning_rate": 4.963636363636364e-06,
      "loss": 3.3557,
      "step": 135
    },
    {
      "epoch": 0.87,
      "eval_accuracy": 0.39150121908742597,
      "eval_loss": 3.74817156791687,
      "eval_runtime": 17.4451,
      "eval_samples_per_second": 573.227,
      "eval_steps_per_second": 17.942,
      "step": 135
    },
    {
      "epoch": 0.91,
      "learning_rate": 4.924675324675325e-06,
      "loss": 3.3157,
      "step": 140
    },
    {
      "epoch": 0.94,
      "learning_rate": 4.885714285714286e-06,
      "loss": 3.3465,
      "step": 145
    },
    {
      "epoch": 0.97,
      "learning_rate": 4.846753246753247e-06,
      "loss": 3.3077,
      "step": 150
    },
    {
      "epoch": 0.97,
      "eval_accuracy": 0.39080459770114945,
      "eval_loss": 3.749429702758789,
      "eval_runtime": 17.4638,
      "eval_samples_per_second": 572.613,
      "eval_steps_per_second": 17.923,
      "step": 150
    },
    {
      "epoch": 1.01,
      "learning_rate": 4.807792207792208e-06,
      "loss": 3.6416,
      "step": 155
    },
    {
      "epoch": 1.04,
      "learning_rate": 4.768831168831169e-06,
      "loss": 3.2947,
      "step": 160
    },
    {
      "epoch": 1.07,
      "learning_rate": 4.72987012987013e-06,
      "loss": 3.2584,
      "step": 165
    },
    {
      "epoch": 1.07,
      "eval_accuracy": 0.39372169975618254,
      "eval_loss": 3.7410030364990234,
      "eval_runtime": 17.4385,
      "eval_samples_per_second": 573.445,
      "eval_steps_per_second": 17.949,
      "step": 165
    },
    {
      "epoch": 1.1,
      "learning_rate": 4.690909090909092e-06,
      "loss": 3.2643,
      "step": 170
    },
    {
      "epoch": 1.14,
      "learning_rate": 4.651948051948052e-06,
      "loss": 3.2647,
      "step": 175
    },
    {
      "epoch": 1.17,
      "learning_rate": 4.612987012987013e-06,
      "loss": 3.2459,
      "step": 180
    },
    {
      "epoch": 1.17,
      "eval_accuracy": 0.39167537443399514,
      "eval_loss": 3.7585999965667725,
      "eval_runtime": 17.4515,
      "eval_samples_per_second": 573.017,
      "eval_steps_per_second": 17.935,
      "step": 180
    },
    {
      "epoch": 1.2,
      "learning_rate": 4.574025974025974e-06,
      "loss": 3.2303,
      "step": 185
    },
    {
      "epoch": 1.23,
      "learning_rate": 4.535064935064936e-06,
      "loss": 3.2218,
      "step": 190
    },
    {
      "epoch": 1.27,
      "learning_rate": 4.496103896103896e-06,
      "loss": 3.214,
      "step": 195
    },
    {
      "epoch": 1.27,
      "eval_accuracy": 0.3966823406478579,
      "eval_loss": 3.7317585945129395,
      "eval_runtime": 17.4059,
      "eval_samples_per_second": 574.518,
      "eval_steps_per_second": 17.982,
      "step": 195
    },
    {
      "epoch": 1.3,
      "learning_rate": 4.457142857142857e-06,
      "loss": 3.1917,
      "step": 200
    },
    {
      "epoch": 1.33,
      "learning_rate": 4.418181818181818e-06,
      "loss": 3.185,
      "step": 205
    },
    {
      "epoch": 1.36,
      "learning_rate": 4.37922077922078e-06,
      "loss": 3.2013,
      "step": 210
    },
    {
      "epoch": 1.36,
      "eval_accuracy": 0.39707419017763845,
      "eval_loss": 3.7158472537994385,
      "eval_runtime": 17.4332,
      "eval_samples_per_second": 573.618,
      "eval_steps_per_second": 17.954,
      "step": 210
    },
    {
      "epoch": 1.39,
      "learning_rate": 4.34025974025974e-06,
      "loss": 3.1968,
      "step": 215
    },
    {
      "epoch": 1.43,
      "learning_rate": 4.301298701298701e-06,
      "loss": 3.1735,
      "step": 220
    },
    {
      "epoch": 1.46,
      "learning_rate": 4.262337662337662e-06,
      "loss": 3.1887,
      "step": 225
    },
    {
      "epoch": 1.46,
      "eval_accuracy": 0.3923284569836294,
      "eval_loss": 3.7491486072540283,
      "eval_runtime": 17.4492,
      "eval_samples_per_second": 573.091,
      "eval_steps_per_second": 17.938,
      "step": 225
    },
    {
      "epoch": 1.49,
      "learning_rate": 4.2233766233766236e-06,
      "loss": 3.1663,
      "step": 230
    },
    {
      "epoch": 1.52,
      "learning_rate": 4.184415584415584e-06,
      "loss": 3.1689,
      "step": 235
    },
    {
      "epoch": 1.56,
      "learning_rate": 4.145454545454546e-06,
      "loss": 3.1359,
      "step": 240
    },
    {
      "epoch": 1.56,
      "eval_accuracy": 0.39655172413793105,
      "eval_loss": 3.7439537048339844,
      "eval_runtime": 17.4433,
      "eval_samples_per_second": 573.287,
      "eval_steps_per_second": 17.944,
      "step": 240
    },
    {
      "epoch": 1.59,
      "learning_rate": 4.106493506493506e-06,
      "loss": 3.162,
      "step": 245
    },
    {
      "epoch": 1.62,
      "learning_rate": 4.0675324675324675e-06,
      "loss": 3.1473,
      "step": 250
    },
    {
      "epoch": 1.65,
      "learning_rate": 4.028571428571429e-06,
      "loss": 3.1424,
      "step": 255
    },
    {
      "epoch": 1.65,
      "eval_accuracy": 0.3979014280738419,
      "eval_loss": 3.724943161010742,
      "eval_runtime": 17.4625,
      "eval_samples_per_second": 572.655,
      "eval_steps_per_second": 17.924,
      "step": 255
    },
    {
      "epoch": 1.69,
      "learning_rate": 3.98961038961039e-06,
      "loss": 3.1377,
      "step": 260
    },
    {
      "epoch": 1.72,
      "learning_rate": 3.950649350649351e-06,
      "loss": 3.1185,
      "step": 265
    },
    {
      "epoch": 1.75,
      "learning_rate": 3.9116883116883115e-06,
      "loss": 3.124,
      "step": 270
    },
    {
      "epoch": 1.75,
      "eval_accuracy": 0.397466039707419,
      "eval_loss": 3.734914779663086,
      "eval_runtime": 17.4427,
      "eval_samples_per_second": 573.305,
      "eval_steps_per_second": 17.944,
      "step": 270
    },
    {
      "epoch": 1.78,
      "learning_rate": 3.8727272727272735e-06,
      "loss": 3.0861,
      "step": 275
    },
    {
      "epoch": 1.82,
      "learning_rate": 3.833766233766234e-06,
      "loss": 3.08,
      "step": 280
    },
    {
      "epoch": 1.85,
      "learning_rate": 3.794805194805195e-06,
      "loss": 3.0793,
      "step": 285
    },
    {
      "epoch": 1.85,
      "eval_accuracy": 0.3981191222570533,
      "eval_loss": 3.7173280715942383,
      "eval_runtime": 17.4324,
      "eval_samples_per_second": 573.645,
      "eval_steps_per_second": 17.955,
      "step": 285
    },
    {
      "epoch": 1.88,
      "learning_rate": 3.755844155844156e-06,
      "loss": 3.0642,
      "step": 290
    },
    {
      "epoch": 1.91,
      "learning_rate": 3.716883116883117e-06,
      "loss": 3.0848,
      "step": 295
    },
    {
      "epoch": 1.94,
      "learning_rate": 3.677922077922078e-06,
      "loss": 3.0527,
      "step": 300
    },
    {
      "epoch": 1.94,
      "eval_accuracy": 0.39973005921281785,
      "eval_loss": 3.722606658935547,
      "eval_runtime": 17.44,
      "eval_samples_per_second": 573.394,
      "eval_steps_per_second": 17.947,
      "step": 300
    },
    {
      "epoch": 1.98,
      "learning_rate": 3.638961038961039e-06,
      "loss": 3.062,
      "step": 305
    },
    {
      "epoch": 2.01,
      "learning_rate": 3.6e-06,
      "loss": 3.3985,
      "step": 310
    },
    {
      "epoch": 2.05,
      "learning_rate": 3.5610389610389615e-06,
      "loss": 3.049,
      "step": 315
    },
    {
      "epoch": 2.05,
      "eval_accuracy": 0.3986415882967607,
      "eval_loss": 3.7341573238372803,
      "eval_runtime": 17.4676,
      "eval_samples_per_second": 572.489,
      "eval_steps_per_second": 17.919,
      "step": 315
    },
    {
      "epoch": 2.08,
      "learning_rate": 3.522077922077922e-06,
      "loss": 3.0231,
      "step": 320
    },
    {
      "epoch": 2.11,
      "learning_rate": 3.4831168831168834e-06,
      "loss": 3.0292,
      "step": 325
    },
    {
      "epoch": 2.14,
      "learning_rate": 3.4441558441558442e-06,
      "loss": 3.0422,
      "step": 330
    },
    {
      "epoch": 2.14,
      "eval_accuracy": 0.3981191222570533,
      "eval_loss": 3.747906446456909,
      "eval_runtime": 17.4308,
      "eval_samples_per_second": 573.698,
      "eval_steps_per_second": 17.957,
      "step": 330
    },
    {
      "epoch": 2.17,
      "learning_rate": 3.4051948051948054e-06,
      "loss": 3.0296,
      "step": 335
    },
    {
      "epoch": 2.21,
      "learning_rate": 3.366233766233766e-06,
      "loss": 3.042,
      "step": 340
    },
    {
      "epoch": 2.24,
      "learning_rate": 3.3272727272727274e-06,
      "loss": 3.0078,
      "step": 345
    },
    {
      "epoch": 2.24,
      "eval_accuracy": 0.39502786485545105,
      "eval_loss": 3.7774932384490967,
      "eval_runtime": 17.3904,
      "eval_samples_per_second": 575.029,
      "eval_steps_per_second": 17.998,
      "step": 345
    },
    {
      "epoch": 2.27,
      "learning_rate": 3.288311688311688e-06,
      "loss": 3.0348,
      "step": 350
    },
    {
      "epoch": 2.3,
      "learning_rate": 3.2493506493506494e-06,
      "loss": 3.0249,
      "step": 355
    },
    {
      "epoch": 2.34,
      "learning_rate": 3.21038961038961e-06,
      "loss": 2.9674,
      "step": 360
    },
    {
      "epoch": 2.34,
      "eval_accuracy": 0.3997735980494601,
      "eval_loss": 3.741637945175171,
      "eval_runtime": 17.4523,
      "eval_samples_per_second": 572.989,
      "eval_steps_per_second": 17.935,
      "step": 360
    },
    {
      "epoch": 2.37,
      "learning_rate": 3.1714285714285714e-06,
      "loss": 3.0017,
      "step": 365
    },
    {
      "epoch": 2.4,
      "learning_rate": 3.132467532467532e-06,
      "loss": 2.9889,
      "step": 370
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.0935064935064938e-06,
      "loss": 2.9812,
      "step": 375
    },
    {
      "epoch": 2.43,
      "eval_accuracy": 0.3970306513409962,
      "eval_loss": 3.760990619659424,
      "eval_runtime": 17.5073,
      "eval_samples_per_second": 571.19,
      "eval_steps_per_second": 17.878,
      "step": 375
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.054545454545454e-06,
      "loss": 2.9882,
      "step": 380
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.0155844155844158e-06,
      "loss": 2.9912,
      "step": 385
    },
    {
      "epoch": 2.53,
      "learning_rate": 2.9766233766233765e-06,
      "loss": 2.9518,
      "step": 390
    },
    {
      "epoch": 2.53,
      "eval_accuracy": 0.3969435736677116,
      "eval_loss": 3.7642152309417725,
      "eval_runtime": 17.4159,
      "eval_samples_per_second": 574.188,
      "eval_steps_per_second": 17.972,
      "step": 390
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9376623376623377e-06,
      "loss": 2.9598,
      "step": 395
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.898701298701299e-06,
      "loss": 2.9762,
      "step": 400
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.8597402597402597e-06,
      "loss": 2.9826,
      "step": 405
    },
    {
      "epoch": 2.63,
      "eval_accuracy": 0.39946882619296414,
      "eval_loss": 3.749140501022339,
      "eval_runtime": 17.4742,
      "eval_samples_per_second": 572.273,
      "eval_steps_per_second": 17.912,
      "step": 405
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.820779220779221e-06,
      "loss": 2.9705,
      "step": 410
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.7818181818181817e-06,
      "loss": 2.9742,
      "step": 415
    },
    {
      "epoch": 2.72,
      "learning_rate": 2.742857142857143e-06,
      "loss": 2.9689,
      "step": 420
    },
    {
      "epoch": 2.72,
      "eval_accuracy": 0.39568094740508536,
      "eval_loss": 3.760981321334839,
      "eval_runtime": 17.4343,
      "eval_samples_per_second": 573.581,
      "eval_steps_per_second": 17.953,
      "step": 420
    },
    {
      "epoch": 2.76,
      "learning_rate": 2.7038961038961037e-06,
      "loss": 2.9619,
      "step": 425
    },
    {
      "epoch": 2.79,
      "learning_rate": 2.664935064935065e-06,
      "loss": 2.9498,
      "step": 430
    },
    {
      "epoch": 2.82,
      "learning_rate": 2.625974025974026e-06,
      "loss": 2.9517,
      "step": 435
    },
    {
      "epoch": 2.82,
      "eval_accuracy": 0.40238592824799724,
      "eval_loss": 3.7294681072235107,
      "eval_runtime": 17.4607,
      "eval_samples_per_second": 572.714,
      "eval_steps_per_second": 17.926,
      "step": 435
    },
    {
      "epoch": 2.85,
      "learning_rate": 2.587012987012987e-06,
      "loss": 2.9564,
      "step": 440
    },
    {
      "epoch": 2.89,
      "learning_rate": 2.548051948051948e-06,
      "loss": 2.9505,
      "step": 445
    },
    {
      "epoch": 2.92,
      "learning_rate": 2.509090909090909e-06,
      "loss": 2.9421,
      "step": 450
    },
    {
      "epoch": 2.92,
      "eval_accuracy": 0.39855451062347613,
      "eval_loss": 3.736074209213257,
      "eval_runtime": 17.4502,
      "eval_samples_per_second": 573.061,
      "eval_steps_per_second": 17.937,
      "step": 450
    },
    {
      "epoch": 2.95,
      "learning_rate": 2.4701298701298705e-06,
      "loss": 2.9477,
      "step": 455
    },
    {
      "epoch": 2.98,
      "learning_rate": 2.4311688311688313e-06,
      "loss": 2.9389,
      "step": 460
    },
    {
      "epoch": 3.02,
      "learning_rate": 2.3922077922077925e-06,
      "loss": 3.2627,
      "step": 465
    },
    {
      "epoch": 3.02,
      "eval_accuracy": 0.4000348310693138,
      "eval_loss": 3.746295690536499,
      "eval_runtime": 17.4136,
      "eval_samples_per_second": 574.264,
      "eval_steps_per_second": 17.974,
      "step": 465
    },
    {
      "epoch": 3.05,
      "learning_rate": 2.3532467532467533e-06,
      "loss": 2.9154,
      "step": 470
    },
    {
      "epoch": 3.08,
      "learning_rate": 2.3142857142857145e-06,
      "loss": 2.931,
      "step": 475
    },
    {
      "epoch": 3.12,
      "learning_rate": 2.2753246753246757e-06,
      "loss": 2.9189,
      "step": 480
    },
    {
      "epoch": 3.12,
      "eval_accuracy": 0.3969435736677116,
      "eval_loss": 3.7732067108154297,
      "eval_runtime": 17.4617,
      "eval_samples_per_second": 572.681,
      "eval_steps_per_second": 17.925,
      "step": 480
    },
    {
      "epoch": 3.15,
      "learning_rate": 2.2363636363636364e-06,
      "loss": 2.9295,
      "step": 485
    },
    {
      "epoch": 3.18,
      "learning_rate": 2.1974025974025976e-06,
      "loss": 2.9133,
      "step": 490
    },
    {
      "epoch": 3.21,
      "learning_rate": 2.1584415584415584e-06,
      "loss": 2.9063,
      "step": 495
    },
    {
      "epoch": 3.21,
      "eval_accuracy": 0.39990421455938696,
      "eval_loss": 3.763644218444824,
      "eval_runtime": 17.4503,
      "eval_samples_per_second": 573.056,
      "eval_steps_per_second": 17.937,
      "step": 495
    },
    {
      "epoch": 3.25,
      "learning_rate": 2.1194805194805196e-06,
      "loss": 2.9171,
      "step": 500
    },
    {
      "epoch": 3.28,
      "learning_rate": 2.0805194805194804e-06,
      "loss": 2.9087,
      "step": 505
    },
    {
      "epoch": 3.31,
      "learning_rate": 2.0415584415584416e-06,
      "loss": 2.9181,
      "step": 510
    },
    {
      "epoch": 3.31,
      "eval_accuracy": 0.39999129223267155,
      "eval_loss": 3.754517078399658,
      "eval_runtime": 17.4146,
      "eval_samples_per_second": 574.231,
      "eval_steps_per_second": 17.973,
      "step": 510
    },
    {
      "epoch": 3.34,
      "learning_rate": 2.002597402597403e-06,
      "loss": 2.8923,
      "step": 515
    },
    {
      "epoch": 3.38,
      "learning_rate": 1.9636363636363636e-06,
      "loss": 2.8953,
      "step": 520
    },
    {
      "epoch": 3.41,
      "learning_rate": 1.924675324675325e-06,
      "loss": 2.9137,
      "step": 525
    },
    {
      "epoch": 3.41,
      "eval_accuracy": 0.4009056078021595,
      "eval_loss": 3.7548601627349854,
      "eval_runtime": 17.4295,
      "eval_samples_per_second": 573.739,
      "eval_steps_per_second": 17.958,
      "step": 525
    },
    {
      "epoch": 3.44,
      "learning_rate": 1.8857142857142858e-06,
      "loss": 2.8988,
      "step": 530
    },
    {
      "epoch": 3.47,
      "learning_rate": 1.8467532467532468e-06,
      "loss": 2.9048,
      "step": 535
    },
    {
      "epoch": 3.5,
      "learning_rate": 1.8077922077922078e-06,
      "loss": 2.8904,
      "step": 540
    },
    {
      "epoch": 3.5,
      "eval_accuracy": 0.40238592824799724,
      "eval_loss": 3.7537996768951416,
      "eval_runtime": 17.315,
      "eval_samples_per_second": 577.533,
      "eval_steps_per_second": 18.077,
      "step": 540
    },
    {
      "epoch": 3.54,
      "learning_rate": 1.7688311688311688e-06,
      "loss": 2.9169,
      "step": 545
    },
    {
      "epoch": 3.57,
      "learning_rate": 1.7298701298701297e-06,
      "loss": 2.9047,
      "step": 550
    },
    {
      "epoch": 3.6,
      "learning_rate": 1.690909090909091e-06,
      "loss": 2.8881,
      "step": 555
    },
    {
      "epoch": 3.6,
      "eval_accuracy": 0.3987286659700453,
      "eval_loss": 3.7595129013061523,
      "eval_runtime": 17.237,
      "eval_samples_per_second": 580.149,
      "eval_steps_per_second": 18.159,
      "step": 555
    },
    {
      "epoch": 3.63,
      "learning_rate": 1.651948051948052e-06,
      "loss": 2.8934,
      "step": 560
    },
    {
      "epoch": 3.67,
      "learning_rate": 1.612987012987013e-06,
      "loss": 2.8792,
      "step": 565
    },
    {
      "epoch": 3.7,
      "learning_rate": 1.574025974025974e-06,
      "loss": 2.8876,
      "step": 570
    },
    {
      "epoch": 3.7,
      "eval_accuracy": 0.39803204458376873,
      "eval_loss": 3.7674362659454346,
      "eval_runtime": 16.7838,
      "eval_samples_per_second": 595.813,
      "eval_steps_per_second": 18.649,
      "step": 570
    },
    {
      "epoch": 3.73,
      "learning_rate": 1.535064935064935e-06,
      "loss": 2.8712,
      "step": 575
    },
    {
      "epoch": 3.76,
      "learning_rate": 1.4961038961038961e-06,
      "loss": 2.8557,
      "step": 580
    },
    {
      "epoch": 3.8,
      "learning_rate": 1.4571428571428571e-06,
      "loss": 2.8805,
      "step": 585
    },
    {
      "epoch": 3.8,
      "eval_accuracy": 0.4021246952281435,
      "eval_loss": 3.73473858833313,
      "eval_runtime": 17.3707,
      "eval_samples_per_second": 575.682,
      "eval_steps_per_second": 18.019,
      "step": 585
    },
    {
      "epoch": 3.8,
      "step": 585,
      "total_flos": 3.8216290317512755e+17,
      "train_loss": 3.178084816891923,
      "train_runtime": 19361.8142,
      "train_samples_per_second": 319.258,
      "train_steps_per_second": 0.04
    },
    {
      "epoch": 3.8,
      "eval_accuracy": 0.39707419017763845,
      "eval_loss": 3.7512481212615967,
      "eval_runtime": 17.386,
      "eval_samples_per_second": 575.175,
      "eval_steps_per_second": 18.003,
      "step": 585
    }
  ],
  "max_steps": 770,
  "num_train_epochs": 5,
  "total_flos": 3.8216290317512755e+17,
  "trial_name": null,
  "trial_params": null
}