{
  "best_metric": 0.02033529244363308,
  "best_model_checkpoint": "./timri-classification/checkpoint-840",
  "epoch": 3.74468085106383,
  "global_step": 880,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 0.00019787234042553193,
      "loss": 0.9777,
      "step": 10
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019574468085106384,
      "loss": 0.4325,
      "step": 20
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019361702127659576,
      "loss": 0.3674,
      "step": 30
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00019148936170212768,
      "loss": 0.5014,
      "step": 40
    },
    {
      "epoch": 0.17,
      "eval_accuracy": 0.8704566635601119,
      "eval_loss": 0.3428270220756531,
      "eval_runtime": 16.978,
      "eval_samples_per_second": 63.199,
      "eval_steps_per_second": 7.951,
      "step": 40
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00018936170212765957,
      "loss": 0.1515,
      "step": 50
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001872340425531915,
      "loss": 0.1543,
      "step": 60
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0001851063829787234,
      "loss": 0.1933,
      "step": 70
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00018297872340425532,
      "loss": 0.1244,
      "step": 80
    },
    {
      "epoch": 0.34,
      "eval_accuracy": 0.9794967381174278,
      "eval_loss": 0.11025915294885635,
      "eval_runtime": 17.4184,
      "eval_samples_per_second": 61.601,
      "eval_steps_per_second": 7.75,
      "step": 80
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00018085106382978726,
      "loss": 0.0805,
      "step": 90
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00017872340425531915,
      "loss": 0.1156,
      "step": 100
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017659574468085107,
      "loss": 0.2888,
      "step": 110
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00017446808510638298,
      "loss": 0.1569,
      "step": 120
    },
    {
      "epoch": 0.51,
      "eval_accuracy": 0.9478098788443616,
      "eval_loss": 0.18117330968379974,
      "eval_runtime": 17.9321,
      "eval_samples_per_second": 59.837,
      "eval_steps_per_second": 7.528,
      "step": 120
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001723404255319149,
      "loss": 0.1843,
      "step": 130
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00017021276595744682,
      "loss": 0.1066,
      "step": 140
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00016808510638297873,
      "loss": 0.1578,
      "step": 150
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00016595744680851065,
      "loss": 0.1581,
      "step": 160
    },
    {
      "epoch": 0.68,
      "eval_accuracy": 0.9561975768872321,
      "eval_loss": 0.15296928584575653,
      "eval_runtime": 19.3736,
      "eval_samples_per_second": 55.385,
      "eval_steps_per_second": 6.968,
      "step": 160
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00016382978723404257,
      "loss": 0.0961,
      "step": 170
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00016170212765957446,
      "loss": 0.0696,
      "step": 180
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00015957446808510637,
      "loss": 0.1529,
      "step": 190
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00015744680851063832,
      "loss": 0.0748,
      "step": 200
    },
    {
      "epoch": 0.85,
      "eval_accuracy": 0.9804287045666356,
      "eval_loss": 0.08471864461898804,
      "eval_runtime": 18.2622,
      "eval_samples_per_second": 58.755,
      "eval_steps_per_second": 7.392,
      "step": 200
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0001553191489361702,
      "loss": 0.2481,
      "step": 210
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00015319148936170213,
      "loss": 0.0898,
      "step": 220
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00015106382978723407,
      "loss": 0.0974,
      "step": 230
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00014893617021276596,
      "loss": 0.0794,
      "step": 240
    },
    {
      "epoch": 1.02,
      "eval_accuracy": 0.9841565703634669,
      "eval_loss": 0.0651417151093483,
      "eval_runtime": 18.0804,
      "eval_samples_per_second": 59.346,
      "eval_steps_per_second": 7.467,
      "step": 240
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00014680851063829788,
      "loss": 0.0721,
      "step": 250
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0001446808510638298,
      "loss": 0.0672,
      "step": 260
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0001425531914893617,
      "loss": 0.0517,
      "step": 270
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00014042553191489363,
      "loss": 0.0345,
      "step": 280
    },
    {
      "epoch": 1.19,
      "eval_accuracy": 0.9869524697110904,
      "eval_loss": 0.059121981263160706,
      "eval_runtime": 17.7664,
      "eval_samples_per_second": 60.395,
      "eval_steps_per_second": 7.599,
      "step": 280
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00013829787234042554,
      "loss": 0.0883,
      "step": 290
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00013617021276595746,
      "loss": 0.0519,
      "step": 300
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00013404255319148938,
      "loss": 0.0269,
      "step": 310
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00013191489361702127,
      "loss": 0.0496,
      "step": 320
    },
    {
      "epoch": 1.36,
      "eval_accuracy": 0.9794967381174278,
      "eval_loss": 0.08983205258846283,
      "eval_runtime": 17.7992,
      "eval_samples_per_second": 60.283,
      "eval_steps_per_second": 7.585,
      "step": 320
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00012978723404255318,
      "loss": 0.0962,
      "step": 330
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00012765957446808513,
      "loss": 0.068,
      "step": 340
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00012553191489361702,
      "loss": 0.0592,
      "step": 350
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00012340425531914893,
      "loss": 0.0588,
      "step": 360
    },
    {
      "epoch": 1.53,
      "eval_accuracy": 0.9683131407269339,
      "eval_loss": 0.09058272838592529,
      "eval_runtime": 18.1011,
      "eval_samples_per_second": 59.278,
      "eval_steps_per_second": 7.458,
      "step": 360
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00012127659574468086,
      "loss": 0.0188,
      "step": 370
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00011914893617021277,
      "loss": 0.0131,
      "step": 380
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00011702127659574468,
      "loss": 0.0488,
      "step": 390
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.00011489361702127661,
      "loss": 0.0125,
      "step": 400
    },
    {
      "epoch": 1.7,
      "eval_accuracy": 0.9925442684063374,
      "eval_loss": 0.02937113679945469,
      "eval_runtime": 18.1506,
      "eval_samples_per_second": 59.117,
      "eval_steps_per_second": 7.438,
      "step": 400
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00011276595744680852,
      "loss": 0.0198,
      "step": 410
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00011063829787234043,
      "loss": 0.0874,
      "step": 420
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00010851063829787234,
      "loss": 0.011,
      "step": 430
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.00010638297872340425,
      "loss": 0.0437,
      "step": 440
    },
    {
      "epoch": 1.87,
      "eval_accuracy": 0.9664492078285182,
      "eval_loss": 0.122951939702034,
      "eval_runtime": 17.8792,
      "eval_samples_per_second": 60.014,
      "eval_steps_per_second": 7.551,
      "step": 440
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00010425531914893618,
      "loss": 0.1603,
      "step": 450
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00010212765957446809,
      "loss": 0.0557,
      "step": 460
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0001,
      "loss": 0.0084,
      "step": 470
    },
    {
      "epoch": 2.04,
      "learning_rate": 9.787234042553192e-05,
      "loss": 0.0223,
      "step": 480
    },
    {
      "epoch": 2.04,
      "eval_accuracy": 0.9888164026095061,
      "eval_loss": 0.04005502909421921,
      "eval_runtime": 17.7107,
      "eval_samples_per_second": 60.585,
      "eval_steps_per_second": 7.623,
      "step": 480
    },
    {
      "epoch": 2.09,
      "learning_rate": 9.574468085106384e-05,
      "loss": 0.0189,
      "step": 490
    },
    {
      "epoch": 2.13,
      "learning_rate": 9.361702127659576e-05,
      "loss": 0.0086,
      "step": 500
    },
    {
      "epoch": 2.17,
      "learning_rate": 9.148936170212766e-05,
      "loss": 0.0064,
      "step": 510
    },
    {
      "epoch": 2.21,
      "learning_rate": 8.936170212765958e-05,
      "loss": 0.0056,
      "step": 520
    },
    {
      "epoch": 2.21,
      "eval_accuracy": 0.994408201304753,
      "eval_loss": 0.025717683136463165,
      "eval_runtime": 17.4332,
      "eval_samples_per_second": 61.549,
      "eval_steps_per_second": 7.744,
      "step": 520
    },
    {
      "epoch": 2.26,
      "learning_rate": 8.723404255319149e-05,
      "loss": 0.0296,
      "step": 530
    },
    {
      "epoch": 2.3,
      "learning_rate": 8.510638297872341e-05,
      "loss": 0.0195,
      "step": 540
    },
    {
      "epoch": 2.34,
      "learning_rate": 8.297872340425533e-05,
      "loss": 0.0059,
      "step": 550
    },
    {
      "epoch": 2.38,
      "learning_rate": 8.085106382978723e-05,
      "loss": 0.0077,
      "step": 560
    },
    {
      "epoch": 2.38,
      "eval_accuracy": 0.9860205032618826,
      "eval_loss": 0.05676997825503349,
      "eval_runtime": 17.6076,
      "eval_samples_per_second": 60.94,
      "eval_steps_per_second": 7.667,
      "step": 560
    },
    {
      "epoch": 2.43,
      "learning_rate": 7.872340425531916e-05,
      "loss": 0.0054,
      "step": 570
    },
    {
      "epoch": 2.47,
      "learning_rate": 7.659574468085106e-05,
      "loss": 0.0357,
      "step": 580
    },
    {
      "epoch": 2.51,
      "learning_rate": 7.446808510638298e-05,
      "loss": 0.005,
      "step": 590
    },
    {
      "epoch": 2.55,
      "learning_rate": 7.23404255319149e-05,
      "loss": 0.0327,
      "step": 600
    },
    {
      "epoch": 2.55,
      "eval_accuracy": 0.9897483690587139,
      "eval_loss": 0.04828759282827377,
      "eval_runtime": 17.4529,
      "eval_samples_per_second": 61.48,
      "eval_steps_per_second": 7.735,
      "step": 600
    },
    {
      "epoch": 2.6,
      "learning_rate": 7.021276595744681e-05,
      "loss": 0.0161,
      "step": 610
    },
    {
      "epoch": 2.64,
      "learning_rate": 6.808510638297873e-05,
      "loss": 0.0066,
      "step": 620
    },
    {
      "epoch": 2.68,
      "learning_rate": 6.595744680851063e-05,
      "loss": 0.0049,
      "step": 630
    },
    {
      "epoch": 2.72,
      "learning_rate": 6.382978723404256e-05,
      "loss": 0.0056,
      "step": 640
    },
    {
      "epoch": 2.72,
      "eval_accuracy": 0.9925442684063374,
      "eval_loss": 0.025615280494093895,
      "eval_runtime": 17.7515,
      "eval_samples_per_second": 60.446,
      "eval_steps_per_second": 7.605,
      "step": 640
    },
    {
      "epoch": 2.77,
      "learning_rate": 6.170212765957447e-05,
      "loss": 0.0046,
      "step": 650
    },
    {
      "epoch": 2.81,
      "learning_rate": 5.9574468085106384e-05,
      "loss": 0.005,
      "step": 660
    },
    {
      "epoch": 2.85,
      "learning_rate": 5.744680851063831e-05,
      "loss": 0.004,
      "step": 670
    },
    {
      "epoch": 2.89,
      "learning_rate": 5.531914893617022e-05,
      "loss": 0.0043,
      "step": 680
    },
    {
      "epoch": 2.89,
      "eval_accuracy": 0.9906803355079217,
      "eval_loss": 0.040840886533260345,
      "eval_runtime": 18.0012,
      "eval_samples_per_second": 59.607,
      "eval_steps_per_second": 7.5,
      "step": 680
    },
    {
      "epoch": 2.94,
      "learning_rate": 5.319148936170213e-05,
      "loss": 0.009,
      "step": 690
    },
    {
      "epoch": 2.98,
      "learning_rate": 5.1063829787234044e-05,
      "loss": 0.0041,
      "step": 700
    },
    {
      "epoch": 3.02,
      "learning_rate": 4.893617021276596e-05,
      "loss": 0.004,
      "step": 710
    },
    {
      "epoch": 3.06,
      "learning_rate": 4.680851063829788e-05,
      "loss": 0.0037,
      "step": 720
    },
    {
      "epoch": 3.06,
      "eval_accuracy": 0.9897483690587139,
      "eval_loss": 0.049264904111623764,
      "eval_runtime": 17.6233,
      "eval_samples_per_second": 60.885,
      "eval_steps_per_second": 7.66,
      "step": 720
    },
    {
      "epoch": 3.11,
      "learning_rate": 4.468085106382979e-05,
      "loss": 0.004,
      "step": 730
    },
    {
      "epoch": 3.15,
      "learning_rate": 4.2553191489361704e-05,
      "loss": 0.0051,
      "step": 740
    },
    {
      "epoch": 3.19,
      "learning_rate": 4.0425531914893614e-05,
      "loss": 0.0039,
      "step": 750
    },
    {
      "epoch": 3.23,
      "learning_rate": 3.829787234042553e-05,
      "loss": 0.0036,
      "step": 760
    },
    {
      "epoch": 3.23,
      "eval_accuracy": 0.9934762348555451,
      "eval_loss": 0.030669014900922775,
      "eval_runtime": 17.7182,
      "eval_samples_per_second": 60.559,
      "eval_steps_per_second": 7.619,
      "step": 760
    },
    {
      "epoch": 3.28,
      "learning_rate": 3.617021276595745e-05,
      "loss": 0.0038,
      "step": 770
    },
    {
      "epoch": 3.32,
      "learning_rate": 3.4042553191489365e-05,
      "loss": 0.0037,
      "step": 780
    },
    {
      "epoch": 3.36,
      "learning_rate": 3.191489361702128e-05,
      "loss": 0.0173,
      "step": 790
    },
    {
      "epoch": 3.4,
      "learning_rate": 2.9787234042553192e-05,
      "loss": 0.0035,
      "step": 800
    },
    {
      "epoch": 3.4,
      "eval_accuracy": 0.9953401677539608,
      "eval_loss": 0.023747418075799942,
      "eval_runtime": 18.3552,
      "eval_samples_per_second": 58.458,
      "eval_steps_per_second": 7.355,
      "step": 800
    },
    {
      "epoch": 3.45,
      "learning_rate": 2.765957446808511e-05,
      "loss": 0.0036,
      "step": 810
    },
    {
      "epoch": 3.49,
      "learning_rate": 2.5531914893617022e-05,
      "loss": 0.0032,
      "step": 820
    },
    {
      "epoch": 3.53,
      "learning_rate": 2.340425531914894e-05,
      "loss": 0.0035,
      "step": 830
    },
    {
      "epoch": 3.57,
      "learning_rate": 2.1276595744680852e-05,
      "loss": 0.0275,
      "step": 840
    },
    {
      "epoch": 3.57,
      "eval_accuracy": 0.9953401677539608,
      "eval_loss": 0.02033529244363308,
      "eval_runtime": 18.0247,
      "eval_samples_per_second": 59.529,
      "eval_steps_per_second": 7.49,
      "step": 840
    },
    {
      "epoch": 3.62,
      "learning_rate": 1.9148936170212766e-05,
      "loss": 0.0035,
      "step": 850
    },
    {
      "epoch": 3.66,
      "learning_rate": 1.7021276595744682e-05,
      "loss": 0.0035,
      "step": 860
    },
    {
      "epoch": 3.7,
      "learning_rate": 1.4893617021276596e-05,
      "loss": 0.0033,
      "step": 870
    },
    {
      "epoch": 3.74,
      "learning_rate": 1.2765957446808511e-05,
      "loss": 0.0033,
      "step": 880
    },
    {
      "epoch": 3.74,
      "eval_accuracy": 0.9953401677539608,
      "eval_loss": 0.020447665825486183,
      "eval_runtime": 17.3813,
      "eval_samples_per_second": 61.733,
      "eval_steps_per_second": 7.767,
      "step": 880
    }
  ],
  "max_steps": 940,
  "num_train_epochs": 4,
  "total_flos": 1.0908742917765612e+18,
  "trial_name": null,
  "trial_params": null
}