{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.974958263772955,
  "eval_steps": 10,
  "global_step": 745,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 3e-05,
      "loss": 2.3217,
      "step": 10
    },
    {
      "epoch": 0.07,
      "eval_accuracy": 0.4901315789473684,
      "eval_loss": 0.7262606620788574,
      "eval_runtime": 25.3718,
      "eval_samples_per_second": 11.982,
      "eval_steps_per_second": 2.995,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 3e-05,
      "loss": 0.56,
      "step": 20
    },
    {
      "epoch": 0.13,
      "eval_accuracy": 0.5526315789473685,
      "eval_loss": 0.6898001432418823,
      "eval_runtime": 25.3761,
      "eval_samples_per_second": 11.98,
      "eval_steps_per_second": 2.995,
      "step": 20
    },
    {
      "epoch": 0.2,
      "learning_rate": 3e-05,
      "loss": 0.5281,
      "step": 30
    },
    {
      "epoch": 0.2,
      "eval_accuracy": 0.5888157894736842,
      "eval_loss": 0.6465093493461609,
      "eval_runtime": 25.3947,
      "eval_samples_per_second": 11.971,
      "eval_steps_per_second": 2.993,
      "step": 30
    },
    {
      "epoch": 0.27,
      "learning_rate": 3e-05,
      "loss": 0.994,
      "step": 40
    },
    {
      "epoch": 0.27,
      "eval_accuracy": 0.5986842105263158,
      "eval_loss": 0.7351367473602295,
      "eval_runtime": 25.4621,
      "eval_samples_per_second": 11.939,
      "eval_steps_per_second": 2.985,
      "step": 40
    },
    {
      "epoch": 0.33,
      "learning_rate": 3e-05,
      "loss": 0.4785,
      "step": 50
    },
    {
      "epoch": 0.33,
      "eval_accuracy": 0.6118421052631579,
      "eval_loss": 0.6004362106323242,
      "eval_runtime": 25.4138,
      "eval_samples_per_second": 11.962,
      "eval_steps_per_second": 2.991,
      "step": 50
    },
    {
      "epoch": 0.4,
      "learning_rate": 3e-05,
      "loss": 0.4732,
      "step": 60
    },
    {
      "epoch": 0.4,
      "eval_accuracy": 0.6348684210526315,
      "eval_loss": 0.5782976746559143,
      "eval_runtime": 25.369,
      "eval_samples_per_second": 11.983,
      "eval_steps_per_second": 2.996,
      "step": 60
    },
    {
      "epoch": 0.47,
      "learning_rate": 3e-05,
      "loss": 0.4466,
      "step": 70
    },
    {
      "epoch": 0.47,
      "eval_accuracy": 0.6414473684210527,
      "eval_loss": 0.5713546872138977,
      "eval_runtime": 25.3858,
      "eval_samples_per_second": 11.975,
      "eval_steps_per_second": 2.994,
      "step": 70
    },
    {
      "epoch": 0.53,
      "learning_rate": 3e-05,
      "loss": 0.8737,
      "step": 80
    },
    {
      "epoch": 0.53,
      "eval_accuracy": 0.618421052631579,
      "eval_loss": 0.567269504070282,
      "eval_runtime": 25.3639,
      "eval_samples_per_second": 11.986,
      "eval_steps_per_second": 2.996,
      "step": 80
    },
    {
      "epoch": 0.6,
      "learning_rate": 3e-05,
      "loss": 0.4471,
      "step": 90
    },
    {
      "epoch": 0.6,
      "eval_accuracy": 0.6282894736842105,
      "eval_loss": 0.5630530118942261,
      "eval_runtime": 25.3754,
      "eval_samples_per_second": 11.98,
      "eval_steps_per_second": 2.995,
      "step": 90
    },
    {
      "epoch": 0.67,
      "learning_rate": 3e-05,
      "loss": 0.46,
      "step": 100
    },
    {
      "epoch": 0.67,
      "eval_accuracy": 0.6348684210526315,
      "eval_loss": 0.5503756999969482,
      "eval_runtime": 25.3686,
      "eval_samples_per_second": 11.983,
      "eval_steps_per_second": 2.996,
      "step": 100
    },
    {
      "epoch": 0.73,
      "learning_rate": 3e-05,
      "loss": 0.3294,
      "step": 110
    },
    {
      "epoch": 0.73,
      "eval_accuracy": 0.625,
      "eval_loss": 0.6009898781776428,
      "eval_runtime": 25.3647,
      "eval_samples_per_second": 11.985,
      "eval_steps_per_second": 2.996,
      "step": 110
    },
    {
      "epoch": 0.8,
      "learning_rate": 3e-05,
      "loss": 0.6526,
      "step": 120
    },
    {
      "epoch": 0.8,
      "eval_accuracy": 0.6282894736842105,
      "eval_loss": 0.5730816721916199,
      "eval_runtime": 25.3832,
      "eval_samples_per_second": 11.976,
      "eval_steps_per_second": 2.994,
      "step": 120
    },
    {
      "epoch": 0.87,
      "learning_rate": 3e-05,
      "loss": 0.3712,
      "step": 130
    },
    {
      "epoch": 0.87,
      "eval_accuracy": 0.6447368421052632,
      "eval_loss": 0.5378755331039429,
      "eval_runtime": 25.3825,
      "eval_samples_per_second": 11.977,
      "eval_steps_per_second": 2.994,
      "step": 130
    },
    {
      "epoch": 0.93,
      "learning_rate": 3e-05,
      "loss": 0.3341,
      "step": 140
    },
    {
      "epoch": 0.93,
      "eval_accuracy": 0.6282894736842105,
      "eval_loss": 0.5408769249916077,
      "eval_runtime": 25.3787,
      "eval_samples_per_second": 11.979,
      "eval_steps_per_second": 2.995,
      "step": 140
    },
    {
      "epoch": 1.0,
      "learning_rate": 3e-05,
      "loss": 0.552,
      "step": 150
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6381578947368421,
      "eval_loss": 0.5310616493225098,
      "eval_runtime": 25.4512,
      "eval_samples_per_second": 11.944,
      "eval_steps_per_second": 2.986,
      "step": 150
    },
    {
      "epoch": 1.07,
      "learning_rate": 3e-05,
      "loss": 0.4681,
      "step": 160
    },
    {
      "epoch": 1.07,
      "eval_accuracy": 0.6414473684210527,
      "eval_loss": 0.5371212959289551,
      "eval_runtime": 25.3633,
      "eval_samples_per_second": 11.986,
      "eval_steps_per_second": 2.996,
      "step": 160
    },
    {
      "epoch": 1.14,
      "learning_rate": 3e-05,
      "loss": 0.3119,
      "step": 170
    },
    {
      "epoch": 1.14,
      "eval_accuracy": 0.6282894736842105,
      "eval_loss": 0.6171860694885254,
      "eval_runtime": 25.3622,
      "eval_samples_per_second": 11.986,
      "eval_steps_per_second": 2.997,
      "step": 170
    },
    {
      "epoch": 1.2,
      "learning_rate": 3e-05,
      "loss": 0.3082,
      "step": 180
    },
    {
      "epoch": 1.2,
      "eval_accuracy": 0.6513157894736842,
      "eval_loss": 0.5360857844352722,
      "eval_runtime": 25.3868,
      "eval_samples_per_second": 11.975,
      "eval_steps_per_second": 2.994,
      "step": 180
    },
    {
      "epoch": 1.27,
      "learning_rate": 3e-05,
      "loss": 0.5217,
      "step": 190
    },
    {
      "epoch": 1.27,
      "eval_accuracy": 0.625,
      "eval_loss": 0.5467653870582581,
      "eval_runtime": 25.3846,
      "eval_samples_per_second": 11.976,
      "eval_steps_per_second": 2.994,
      "step": 190
    },
    {
      "epoch": 1.34,
      "learning_rate": 3e-05,
      "loss": 0.3888,
      "step": 200
    },
    {
      "epoch": 1.34,
      "eval_accuracy": 0.631578947368421,
      "eval_loss": 0.5891463756561279,
      "eval_runtime": 25.374,
      "eval_samples_per_second": 11.981,
      "eval_steps_per_second": 2.995,
      "step": 200
    },
    {
      "epoch": 1.4,
      "learning_rate": 3e-05,
      "loss": 0.2841,
      "step": 210
    },
    {
      "epoch": 1.4,
      "eval_accuracy": 0.6282894736842105,
      "eval_loss": 0.5428625345230103,
      "eval_runtime": 25.3734,
      "eval_samples_per_second": 11.981,
      "eval_steps_per_second": 2.995,
      "step": 210
    },
    {
      "epoch": 1.47,
      "learning_rate": 3e-05,
      "loss": 0.2728,
      "step": 220
    },
    {
      "epoch": 1.47,
      "eval_accuracy": 0.6381578947368421,
      "eval_loss": 0.5246651768684387,
      "eval_runtime": 25.3858,
      "eval_samples_per_second": 11.975,
      "eval_steps_per_second": 2.994,
      "step": 220
    },
    {
      "epoch": 1.54,
      "learning_rate": 3e-05,
      "loss": 0.5563,
      "step": 230
    },
    {
      "epoch": 1.54,
      "eval_accuracy": 0.6513157894736842,
      "eval_loss": 0.5003584027290344,
      "eval_runtime": 25.3763,
      "eval_samples_per_second": 11.98,
      "eval_steps_per_second": 2.995,
      "step": 230
    },
    {
      "epoch": 1.6,
      "learning_rate": 3e-05,
      "loss": 0.2862,
      "step": 240
    },
    {
      "epoch": 1.6,
      "eval_accuracy": 0.6546052631578947,
      "eval_loss": 0.4741169810295105,
      "eval_runtime": 25.3852,
      "eval_samples_per_second": 11.975,
      "eval_steps_per_second": 2.994,
      "step": 240
    },
    {
      "epoch": 1.67,
      "learning_rate": 3e-05,
      "loss": 0.2289,
      "step": 250
    },
    {
      "epoch": 1.67,
      "eval_accuracy": 0.6513157894736842,
      "eval_loss": 0.5441343188285828,
      "eval_runtime": 25.3897,
      "eval_samples_per_second": 11.973,
      "eval_steps_per_second": 2.993,
      "step": 250
    },
    {
      "epoch": 1.74,
      "learning_rate": 3e-05,
      "loss": 0.2481,
      "step": 260
    },
    {
      "epoch": 1.74,
      "eval_accuracy": 0.6513157894736842,
      "eval_loss": 0.5170696377754211,
      "eval_runtime": 25.3747,
      "eval_samples_per_second": 11.98,
      "eval_steps_per_second": 2.995,
      "step": 260
    },
    {
      "epoch": 1.8,
      "learning_rate": 3e-05,
      "loss": 0.329,
      "step": 270
    },
    {
      "epoch": 1.8,
      "eval_accuracy": 0.6546052631578947,
      "eval_loss": 0.5371391177177429,
      "eval_runtime": 25.3708,
      "eval_samples_per_second": 11.982,
      "eval_steps_per_second": 2.996,
      "step": 270
    },
    {
      "epoch": 1.87,
      "learning_rate": 3e-05,
      "loss": 0.1741,
      "step": 280
    },
    {
      "epoch": 1.87,
      "eval_accuracy": 0.6677631578947368,
      "eval_loss": 0.5411613583564758,
      "eval_runtime": 25.3792,
      "eval_samples_per_second": 11.978,
      "eval_steps_per_second": 2.995,
      "step": 280
    },
    {
      "epoch": 1.94,
      "learning_rate": 3e-05,
      "loss": 0.2888,
      "step": 290
    },
    {
      "epoch": 1.94,
      "eval_accuracy": 0.6710526315789473,
      "eval_loss": 0.5130823850631714,
      "eval_runtime": 25.3665,
      "eval_samples_per_second": 11.984,
      "eval_steps_per_second": 2.996,
      "step": 290
    },
    {
      "epoch": 2.0,
      "learning_rate": 3e-05,
      "loss": 0.4157,
      "step": 300
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6447368421052632,
      "eval_loss": 0.45547178387641907,
      "eval_runtime": 25.356,
      "eval_samples_per_second": 11.989,
      "eval_steps_per_second": 2.997,
      "step": 300
    },
    {
      "epoch": 2.07,
      "learning_rate": 3e-05,
      "loss": 0.1982,
      "step": 310
    },
    {
      "epoch": 2.07,
      "eval_accuracy": 0.6611842105263158,
      "eval_loss": 0.5670450329780579,
      "eval_runtime": 25.3778,
      "eval_samples_per_second": 11.979,
      "eval_steps_per_second": 2.995,
      "step": 310
    },
    {
      "epoch": 2.14,
      "learning_rate": 3e-05,
      "loss": 0.106,
      "step": 320
    },
    {
      "epoch": 2.14,
      "eval_accuracy": 0.6677631578947368,
      "eval_loss": 0.7942893505096436,
      "eval_runtime": 25.3984,
      "eval_samples_per_second": 11.969,
      "eval_steps_per_second": 2.992,
      "step": 320
    },
    {
      "epoch": 2.2,
      "learning_rate": 3e-05,
      "loss": 0.1718,
      "step": 330
    },
    {
      "epoch": 2.2,
      "eval_accuracy": 0.6644736842105263,
      "eval_loss": 0.7496399879455566,
      "eval_runtime": 25.3753,
      "eval_samples_per_second": 11.98,
      "eval_steps_per_second": 2.995,
      "step": 330
    },
    {
      "epoch": 2.27,
      "learning_rate": 3e-05,
      "loss": 0.214,
      "step": 340
    },
    {
      "epoch": 2.27,
      "eval_accuracy": 0.6842105263157895,
      "eval_loss": 0.626396656036377,
      "eval_runtime": 25.3727,
      "eval_samples_per_second": 11.981,
      "eval_steps_per_second": 2.995,
      "step": 340
    },
    {
      "epoch": 2.34,
      "learning_rate": 3e-05,
      "loss": 0.1571,
      "step": 350
    },
    {
      "epoch": 2.34,
      "eval_accuracy": 0.631578947368421,
      "eval_loss": 0.6138848066329956,
      "eval_runtime": 25.3708,
      "eval_samples_per_second": 11.982,
      "eval_steps_per_second": 2.996,
      "step": 350
    },
    {
      "epoch": 2.4,
      "learning_rate": 3e-05,
      "loss": 0.1432,
      "step": 360
    },
    {
      "epoch": 2.4,
      "eval_accuracy": 0.6842105263157895,
      "eval_loss": 0.6198970675468445,
      "eval_runtime": 25.3819,
      "eval_samples_per_second": 11.977,
      "eval_steps_per_second": 2.994,
      "step": 360
    },
    {
      "epoch": 2.47,
      "learning_rate": 3e-05,
      "loss": 0.1038,
      "step": 370
    },
    {
      "epoch": 2.47,
      "eval_accuracy": 0.6973684210526315,
      "eval_loss": 0.636822521686554,
      "eval_runtime": 25.4286,
      "eval_samples_per_second": 11.955,
      "eval_steps_per_second": 2.989,
      "step": 370
    },
    {
      "epoch": 2.54,
      "learning_rate": 3e-05,
      "loss": 0.1728,
      "step": 380
    },
    {
      "epoch": 2.54,
      "eval_accuracy": 0.6677631578947368,
      "eval_loss": 0.7889474630355835,
      "eval_runtime": 25.3888,
      "eval_samples_per_second": 11.974,
      "eval_steps_per_second": 2.993,
      "step": 380
    },
    {
      "epoch": 2.6,
      "learning_rate": 3e-05,
      "loss": 0.14,
      "step": 390
    },
    {
      "epoch": 2.6,
      "eval_accuracy": 0.6546052631578947,
      "eval_loss": 0.795179545879364,
      "eval_runtime": 25.3862,
      "eval_samples_per_second": 11.975,
      "eval_steps_per_second": 2.994,
      "step": 390
    },
    {
      "epoch": 2.67,
      "learning_rate": 3e-05,
      "loss": 0.1522,
      "step": 400
    },
    {
      "epoch": 2.67,
      "eval_accuracy": 0.6578947368421053,
      "eval_loss": 0.7745038866996765,
      "eval_runtime": 25.3703,
      "eval_samples_per_second": 11.983,
      "eval_steps_per_second": 2.996,
      "step": 400
    },
    {
      "epoch": 2.74,
      "learning_rate": 3e-05,
      "loss": 0.1345,
      "step": 410
    },
    {
      "epoch": 2.74,
      "eval_accuracy": 0.6513157894736842,
      "eval_loss": 0.7230806946754456,
      "eval_runtime": 25.3678,
      "eval_samples_per_second": 11.984,
      "eval_steps_per_second": 2.996,
      "step": 410
    },
    {
      "epoch": 2.8,
      "learning_rate": 3e-05,
      "loss": 0.1587,
      "step": 420
    },
    {
      "epoch": 2.8,
      "eval_accuracy": 0.6480263157894737,
      "eval_loss": 0.7153956890106201,
      "eval_runtime": 25.3506,
      "eval_samples_per_second": 11.992,
      "eval_steps_per_second": 2.998,
      "step": 420
    },
    {
      "epoch": 2.87,
      "learning_rate": 3e-05,
      "loss": 0.1391,
      "step": 430
    },
    {
      "epoch": 2.87,
      "eval_accuracy": 0.6513157894736842,
      "eval_loss": 0.6923157572746277,
      "eval_runtime": 25.3652,
      "eval_samples_per_second": 11.985,
      "eval_steps_per_second": 2.996,
      "step": 430
    },
    {
      "epoch": 2.94,
      "learning_rate": 3e-05,
      "loss": 0.129,
      "step": 440
    },
    {
      "epoch": 2.94,
      "eval_accuracy": 0.6710526315789473,
      "eval_loss": 0.6483842730522156,
      "eval_runtime": 25.3639,
      "eval_samples_per_second": 11.986,
      "eval_steps_per_second": 2.996,
      "step": 440
    },
    {
      "epoch": 3.01,
      "learning_rate": 3e-05,
      "loss": 0.2092,
      "step": 450
    },
    {
      "epoch": 3.01,
      "eval_accuracy": 0.6743421052631579,
      "eval_loss": 0.5821840167045593,
      "eval_runtime": 25.3814,
      "eval_samples_per_second": 11.977,
      "eval_steps_per_second": 2.994,
      "step": 450
    },
    {
      "epoch": 3.07,
      "learning_rate": 3e-05,
      "loss": 0.015,
      "step": 460
    },
    {
      "epoch": 3.07,
      "eval_accuracy": 0.6578947368421053,
      "eval_loss": 1.1217145919799805,
      "eval_runtime": 25.3698,
      "eval_samples_per_second": 11.983,
      "eval_steps_per_second": 2.996,
      "step": 460
    },
    {
      "epoch": 3.14,
      "learning_rate": 3e-05,
      "loss": 0.051,
      "step": 470
    },
    {
      "epoch": 3.14,
      "eval_accuracy": 0.6480263157894737,
      "eval_loss": 1.5789867639541626,
      "eval_runtime": 25.3704,
      "eval_samples_per_second": 11.982,
      "eval_steps_per_second": 2.996,
      "step": 470
    },
    {
      "epoch": 3.21,
      "learning_rate": 3e-05,
      "loss": 0.0999,
      "step": 480
    },
    {
      "epoch": 3.21,
      "eval_accuracy": 0.6677631578947368,
      "eval_loss": 1.5168237686157227,
      "eval_runtime": 25.3794,
      "eval_samples_per_second": 11.978,
      "eval_steps_per_second": 2.995,
      "step": 480
    },
    {
      "epoch": 3.27,
      "learning_rate": 3e-05,
      "loss": 0.1776,
      "step": 490
    },
    {
      "epoch": 3.27,
      "eval_accuracy": 0.6875,
      "eval_loss": 1.2342281341552734,
      "eval_runtime": 25.3944,
      "eval_samples_per_second": 11.971,
      "eval_steps_per_second": 2.993,
      "step": 490
    },
    {
      "epoch": 3.34,
      "learning_rate": 3e-05,
      "loss": 0.0612,
      "step": 500
    },
    {
      "epoch": 3.34,
      "eval_accuracy": 0.6973684210526315,
      "eval_loss": 1.0370548963546753,
      "eval_runtime": 25.385,
      "eval_samples_per_second": 11.976,
      "eval_steps_per_second": 2.994,
      "step": 500
    },
    {
      "epoch": 3.41,
      "learning_rate": 3e-05,
      "loss": 0.0858,
      "step": 510
    },
    {
      "epoch": 3.41,
      "eval_accuracy": 0.6776315789473685,
      "eval_loss": 1.0277096033096313,
      "eval_runtime": 25.386,
      "eval_samples_per_second": 11.975,
      "eval_steps_per_second": 2.994,
      "step": 510
    },
    {
      "epoch": 3.47,
      "learning_rate": 3e-05,
      "loss": 0.0316,
      "step": 520
    },
    {
      "epoch": 3.47,
      "eval_accuracy": 0.680921052631579,
      "eval_loss": 1.0386649370193481,
      "eval_runtime": 25.3706,
      "eval_samples_per_second": 11.982,
      "eval_steps_per_second": 2.996,
      "step": 520
    },
    {
      "epoch": 3.54,
      "learning_rate": 3e-05,
      "loss": 0.1899,
      "step": 530
    },
    {
      "epoch": 3.54,
      "eval_accuracy": 0.6907894736842105,
      "eval_loss": 0.8184946775436401,
      "eval_runtime": 25.3771,
      "eval_samples_per_second": 11.979,
      "eval_steps_per_second": 2.995,
      "step": 530
    },
    {
      "epoch": 3.61,
      "learning_rate": 3e-05,
      "loss": 0.1517,
      "step": 540
    },
    {
      "epoch": 3.61,
      "eval_accuracy": 0.6842105263157895,
      "eval_loss": 0.7053664922714233,
      "eval_runtime": 25.3758,
      "eval_samples_per_second": 11.98,
      "eval_steps_per_second": 2.995,
      "step": 540
    },
    {
      "epoch": 3.67,
      "learning_rate": 3e-05,
      "loss": 0.0324,
      "step": 550
    },
    {
      "epoch": 3.67,
      "eval_accuracy": 0.6842105263157895,
      "eval_loss": 0.8504552245140076,
      "eval_runtime": 25.3714,
      "eval_samples_per_second": 11.982,
      "eval_steps_per_second": 2.995,
      "step": 550
    },
    {
      "epoch": 3.74,
      "learning_rate": 3e-05,
      "loss": 0.0646,
      "step": 560
    },
    {
      "epoch": 3.74,
      "eval_accuracy": 0.6611842105263158,
      "eval_loss": 1.0056827068328857,
      "eval_runtime": 25.3614,
      "eval_samples_per_second": 11.987,
      "eval_steps_per_second": 2.997,
      "step": 560
    },
    {
      "epoch": 3.81,
      "learning_rate": 3e-05,
      "loss": 0.1038,
      "step": 570
    },
    {
      "epoch": 3.81,
      "eval_accuracy": 0.6644736842105263,
      "eval_loss": 1.0026819705963135,
      "eval_runtime": 25.376,
      "eval_samples_per_second": 11.98,
      "eval_steps_per_second": 2.995,
      "step": 570
    },
    {
      "epoch": 3.87,
      "learning_rate": 3e-05,
      "loss": 0.0844,
      "step": 580
    },
    {
      "epoch": 3.87,
      "eval_accuracy": 0.6513157894736842,
      "eval_loss": 0.9926165342330933,
      "eval_runtime": 25.3651,
      "eval_samples_per_second": 11.985,
      "eval_steps_per_second": 2.996,
      "step": 580
    },
    {
      "epoch": 3.94,
      "learning_rate": 3e-05,
      "loss": 0.0986,
      "step": 590
    },
    {
      "epoch": 3.94,
      "eval_accuracy": 0.6578947368421053,
      "eval_loss": 0.9245979189872742,
      "eval_runtime": 25.3574,
      "eval_samples_per_second": 11.989,
      "eval_steps_per_second": 2.997,
      "step": 590
    },
    {
      "epoch": 4.01,
      "learning_rate": 3e-05,
      "loss": 0.0627,
      "step": 600
    },
    {
      "epoch": 4.01,
      "eval_accuracy": 0.6546052631578947,
      "eval_loss": 0.8538947105407715,
      "eval_runtime": 25.3724,
      "eval_samples_per_second": 11.982,
      "eval_steps_per_second": 2.995,
      "step": 600
    },
    {
      "epoch": 4.07,
      "learning_rate": 3e-05,
      "loss": 0.0513,
      "step": 610
    },
    {
      "epoch": 4.07,
      "eval_accuracy": 0.6513157894736842,
      "eval_loss": 0.924721896648407,
      "eval_runtime": 25.4077,
      "eval_samples_per_second": 11.965,
      "eval_steps_per_second": 2.991,
      "step": 610
    },
    {
      "epoch": 4.14,
      "learning_rate": 3e-05,
      "loss": 0.0484,
      "step": 620
    },
    {
      "epoch": 4.14,
      "eval_accuracy": 0.6546052631578947,
      "eval_loss": 1.112806797027588,
      "eval_runtime": 25.3865,
      "eval_samples_per_second": 11.975,
      "eval_steps_per_second": 2.994,
      "step": 620
    },
    {
      "epoch": 4.21,
      "learning_rate": 3e-05,
      "loss": 0.0244,
      "step": 630
    },
    {
      "epoch": 4.21,
      "eval_accuracy": 0.6480263157894737,
      "eval_loss": 1.2701855897903442,
      "eval_runtime": 25.3744,
      "eval_samples_per_second": 11.981,
      "eval_steps_per_second": 2.995,
      "step": 630
    },
    {
      "epoch": 4.27,
      "learning_rate": 3e-05,
      "loss": 0.0672,
      "step": 640
    },
    {
      "epoch": 4.27,
      "eval_accuracy": 0.6414473684210527,
      "eval_loss": 1.716863751411438,
      "eval_runtime": 25.3891,
      "eval_samples_per_second": 11.974,
      "eval_steps_per_second": 2.993,
      "step": 640
    },
    {
      "epoch": 4.34,
      "learning_rate": 3e-05,
      "loss": 0.0824,
      "step": 650
    },
    {
      "epoch": 4.34,
      "eval_accuracy": 0.6414473684210527,
      "eval_loss": 1.662705659866333,
      "eval_runtime": 25.3803,
      "eval_samples_per_second": 11.978,
      "eval_steps_per_second": 2.994,
      "step": 650
    },
    {
      "epoch": 4.41,
      "learning_rate": 3e-05,
      "loss": 0.0068,
      "step": 660
    },
    {
      "epoch": 4.41,
      "eval_accuracy": 0.6348684210526315,
      "eval_loss": 1.342494010925293,
      "eval_runtime": 25.376,
      "eval_samples_per_second": 11.98,
      "eval_steps_per_second": 2.995,
      "step": 660
    },
    {
      "epoch": 4.47,
      "learning_rate": 3e-05,
      "loss": 0.044,
      "step": 670
    },
    {
      "epoch": 4.47,
      "eval_accuracy": 0.6611842105263158,
      "eval_loss": 1.2208458185195923,
      "eval_runtime": 25.3753,
      "eval_samples_per_second": 11.98,
      "eval_steps_per_second": 2.995,
      "step": 670
    },
    {
      "epoch": 4.54,
      "learning_rate": 3e-05,
      "loss": 0.0378,
      "step": 680
    },
    {
      "epoch": 4.54,
      "eval_accuracy": 0.6447368421052632,
      "eval_loss": 1.289115309715271,
      "eval_runtime": 25.3905,
      "eval_samples_per_second": 11.973,
      "eval_steps_per_second": 2.993,
      "step": 680
    },
    {
      "epoch": 4.61,
      "learning_rate": 3e-05,
      "loss": 0.0411,
      "step": 690
    },
    {
      "epoch": 4.61,
      "eval_accuracy": 0.6611842105263158,
      "eval_loss": 1.3528344631195068,
      "eval_runtime": 25.3678,
      "eval_samples_per_second": 11.984,
      "eval_steps_per_second": 2.996,
      "step": 690
    },
    {
      "epoch": 4.67,
      "learning_rate": 3e-05,
      "loss": 0.0215,
      "step": 700
    },
    {
      "epoch": 4.67,
      "eval_accuracy": 0.6677631578947368,
      "eval_loss": 1.2606314420700073,
      "eval_runtime": 25.3852,
      "eval_samples_per_second": 11.975,
      "eval_steps_per_second": 2.994,
      "step": 700
    },
    {
      "epoch": 4.74,
      "learning_rate": 3e-05,
      "loss": 0.0438,
      "step": 710
    },
    {
      "epoch": 4.74,
      "eval_accuracy": 0.6546052631578947,
      "eval_loss": 1.2514981031417847,
      "eval_runtime": 25.3679,
      "eval_samples_per_second": 11.984,
      "eval_steps_per_second": 2.996,
      "step": 710
    },
    {
      "epoch": 4.81,
      "learning_rate": 3e-05,
      "loss": 0.0936,
      "step": 720
    },
    {
      "epoch": 4.81,
      "eval_accuracy": 0.6644736842105263,
      "eval_loss": 1.0857858657836914,
      "eval_runtime": 25.3812,
      "eval_samples_per_second": 11.977,
      "eval_steps_per_second": 2.994,
      "step": 720
    },
    {
      "epoch": 4.87,
      "learning_rate": 3e-05,
      "loss": 0.0305,
      "step": 730
    },
    {
      "epoch": 4.87,
      "eval_accuracy": 0.6578947368421053,
      "eval_loss": 0.9838737845420837,
      "eval_runtime": 25.3673,
      "eval_samples_per_second": 11.984,
      "eval_steps_per_second": 2.996,
      "step": 730
    },
    {
      "epoch": 4.94,
      "learning_rate": 3e-05,
      "loss": 0.0282,
      "step": 740
    },
    {
      "epoch": 4.94,
      "eval_accuracy": 0.680921052631579,
      "eval_loss": 1.0233386754989624,
      "eval_runtime": 25.3685,
      "eval_samples_per_second": 11.983,
      "eval_steps_per_second": 2.996,
      "step": 740
    },
    {
      "epoch": 4.97,
      "step": 745,
      "total_flos": 2.959225892752589e+17,
      "train_loss": 0.25739487704614666,
      "train_runtime": 4901.8806,
      "train_samples_per_second": 2.444,
      "train_steps_per_second": 0.152
    }
  ],
  "logging_steps": 10,
  "max_steps": 745,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 150,
  "total_flos": 2.959225892752589e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}