{
  "best_metric": 0.33891761302948,
  "best_model_checkpoint": "finetuned-electrical-images/checkpoint-600",
  "epoch": 4.0,
  "eval_steps": 100,
  "global_step": 860,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.046511627906976744,
      "grad_norm": 2.072422742843628,
      "learning_rate": 0.00019767441860465116,
      "loss": 1.5929,
      "step": 10
    },
    {
      "epoch": 0.09302325581395349,
      "grad_norm": 2.2486143112182617,
      "learning_rate": 0.00019534883720930232,
      "loss": 1.2056,
      "step": 20
    },
    {
      "epoch": 0.13953488372093023,
      "grad_norm": 2.1311681270599365,
      "learning_rate": 0.0001930232558139535,
      "loss": 1.0947,
      "step": 30
    },
    {
      "epoch": 0.18604651162790697,
      "grad_norm": 2.3108925819396973,
      "learning_rate": 0.00019069767441860466,
      "loss": 0.965,
      "step": 40
    },
    {
      "epoch": 0.23255813953488372,
      "grad_norm": 2.6261141300201416,
      "learning_rate": 0.00018837209302325584,
      "loss": 0.7767,
      "step": 50
    },
    {
      "epoch": 0.27906976744186046,
      "grad_norm": 4.883328914642334,
      "learning_rate": 0.000186046511627907,
      "loss": 0.8164,
      "step": 60
    },
    {
      "epoch": 0.32558139534883723,
      "grad_norm": 2.2796030044555664,
      "learning_rate": 0.00018372093023255815,
      "loss": 0.8293,
      "step": 70
    },
    {
      "epoch": 0.37209302325581395,
      "grad_norm": 2.7582902908325195,
      "learning_rate": 0.0001813953488372093,
      "loss": 0.7748,
      "step": 80
    },
    {
      "epoch": 0.4186046511627907,
      "grad_norm": 3.0205562114715576,
      "learning_rate": 0.00017906976744186048,
      "loss": 0.8204,
      "step": 90
    },
    {
      "epoch": 0.46511627906976744,
      "grad_norm": 1.5362632274627686,
      "learning_rate": 0.00017674418604651164,
      "loss": 0.7151,
      "step": 100
    },
    {
      "epoch": 0.46511627906976744,
      "eval_accuracy": 0.8201320132013201,
      "eval_loss": 0.5808877348899841,
      "eval_runtime": 7.6253,
      "eval_samples_per_second": 79.472,
      "eval_steps_per_second": 9.967,
      "step": 100
    },
    {
      "epoch": 0.5116279069767442,
      "grad_norm": 3.7735097408294678,
      "learning_rate": 0.0001744186046511628,
      "loss": 0.523,
      "step": 110
    },
    {
      "epoch": 0.5581395348837209,
      "grad_norm": 5.240425109863281,
      "learning_rate": 0.00017209302325581395,
      "loss": 0.5818,
      "step": 120
    },
    {
      "epoch": 0.6046511627906976,
      "grad_norm": 3.7434163093566895,
      "learning_rate": 0.0001697674418604651,
      "loss": 0.7063,
      "step": 130
    },
    {
      "epoch": 0.6511627906976745,
      "grad_norm": 1.7690590620040894,
      "learning_rate": 0.00016744186046511629,
      "loss": 0.7594,
      "step": 140
    },
    {
      "epoch": 0.6976744186046512,
      "grad_norm": 3.7053258419036865,
      "learning_rate": 0.00016511627906976747,
      "loss": 0.5997,
      "step": 150
    },
    {
      "epoch": 0.7441860465116279,
      "grad_norm": 4.870126247406006,
      "learning_rate": 0.00016279069767441862,
      "loss": 0.5614,
      "step": 160
    },
    {
      "epoch": 0.7906976744186046,
      "grad_norm": 2.533661365509033,
      "learning_rate": 0.00016046511627906978,
      "loss": 0.4965,
      "step": 170
    },
    {
      "epoch": 0.8372093023255814,
      "grad_norm": 3.910142660140991,
      "learning_rate": 0.00015813953488372093,
      "loss": 0.5525,
      "step": 180
    },
    {
      "epoch": 0.8837209302325582,
      "grad_norm": 3.5800535678863525,
      "learning_rate": 0.0001558139534883721,
      "loss": 0.5458,
      "step": 190
    },
    {
      "epoch": 0.9302325581395349,
      "grad_norm": 1.315317153930664,
      "learning_rate": 0.00015348837209302327,
      "loss": 0.6882,
      "step": 200
    },
    {
      "epoch": 0.9302325581395349,
      "eval_accuracy": 0.8498349834983498,
      "eval_loss": 0.4638592302799225,
      "eval_runtime": 7.9986,
      "eval_samples_per_second": 75.764,
      "eval_steps_per_second": 9.502,
      "step": 200
    },
    {
      "epoch": 0.9767441860465116,
      "grad_norm": 3.7220730781555176,
      "learning_rate": 0.00015116279069767442,
      "loss": 0.5207,
      "step": 210
    },
    {
      "epoch": 1.0232558139534884,
      "grad_norm": 4.103209018707275,
      "learning_rate": 0.00014883720930232558,
      "loss": 0.5282,
      "step": 220
    },
    {
      "epoch": 1.069767441860465,
      "grad_norm": 2.3725953102111816,
      "learning_rate": 0.00014651162790697673,
      "loss": 0.4849,
      "step": 230
    },
    {
      "epoch": 1.1162790697674418,
      "grad_norm": 2.339578151702881,
      "learning_rate": 0.00014418604651162791,
      "loss": 0.3707,
      "step": 240
    },
    {
      "epoch": 1.1627906976744187,
      "grad_norm": 2.8100476264953613,
      "learning_rate": 0.0001418604651162791,
      "loss": 0.3821,
      "step": 250
    },
    {
      "epoch": 1.2093023255813953,
      "grad_norm": 2.1530966758728027,
      "learning_rate": 0.00013953488372093025,
      "loss": 0.4797,
      "step": 260
    },
    {
      "epoch": 1.255813953488372,
      "grad_norm": 1.164758324623108,
      "learning_rate": 0.0001372093023255814,
      "loss": 0.4341,
      "step": 270
    },
    {
      "epoch": 1.302325581395349,
      "grad_norm": 1.5009866952896118,
      "learning_rate": 0.00013488372093023256,
      "loss": 0.4527,
      "step": 280
    },
    {
      "epoch": 1.3488372093023255,
      "grad_norm": 2.4176268577575684,
      "learning_rate": 0.00013255813953488372,
      "loss": 0.3878,
      "step": 290
    },
    {
      "epoch": 1.3953488372093024,
      "grad_norm": 4.717296123504639,
      "learning_rate": 0.0001302325581395349,
      "loss": 0.3897,
      "step": 300
    },
    {
      "epoch": 1.3953488372093024,
      "eval_accuracy": 0.8465346534653465,
      "eval_loss": 0.47040635347366333,
      "eval_runtime": 8.212,
      "eval_samples_per_second": 73.794,
      "eval_steps_per_second": 9.255,
      "step": 300
    },
    {
      "epoch": 1.441860465116279,
      "grad_norm": 1.026237964630127,
      "learning_rate": 0.00012790697674418605,
      "loss": 0.3683,
      "step": 310
    },
    {
      "epoch": 1.4883720930232558,
      "grad_norm": 2.894584894180298,
      "learning_rate": 0.0001255813953488372,
      "loss": 0.4968,
      "step": 320
    },
    {
      "epoch": 1.5348837209302326,
      "grad_norm": 1.6250619888305664,
      "learning_rate": 0.00012325581395348836,
      "loss": 0.4784,
      "step": 330
    },
    {
      "epoch": 1.5813953488372094,
      "grad_norm": 2.221461296081543,
      "learning_rate": 0.00012093023255813953,
      "loss": 0.5513,
      "step": 340
    },
    {
      "epoch": 1.627906976744186,
      "grad_norm": 6.982600688934326,
      "learning_rate": 0.00011860465116279071,
      "loss": 0.5509,
      "step": 350
    },
    {
      "epoch": 1.6744186046511627,
      "grad_norm": 2.3711423873901367,
      "learning_rate": 0.00011627906976744187,
      "loss": 0.4542,
      "step": 360
    },
    {
      "epoch": 1.7209302325581395,
      "grad_norm": 2.340607166290283,
      "learning_rate": 0.00011395348837209304,
      "loss": 0.3822,
      "step": 370
    },
    {
      "epoch": 1.7674418604651163,
      "grad_norm": 3.709766387939453,
      "learning_rate": 0.00011162790697674419,
      "loss": 0.4252,
      "step": 380
    },
    {
      "epoch": 1.8139534883720931,
      "grad_norm": 3.5805418491363525,
      "learning_rate": 0.00010930232558139534,
      "loss": 0.6467,
      "step": 390
    },
    {
      "epoch": 1.8604651162790697,
      "grad_norm": 2.1463587284088135,
      "learning_rate": 0.00010697674418604651,
      "loss": 0.4909,
      "step": 400
    },
    {
      "epoch": 1.8604651162790697,
      "eval_accuracy": 0.8448844884488449,
      "eval_loss": 0.5023446083068848,
      "eval_runtime": 8.1791,
      "eval_samples_per_second": 74.091,
      "eval_steps_per_second": 9.292,
      "step": 400
    },
    {
      "epoch": 1.9069767441860463,
      "grad_norm": 3.8787500858306885,
      "learning_rate": 0.00010465116279069768,
      "loss": 0.485,
      "step": 410
    },
    {
      "epoch": 1.9534883720930232,
      "grad_norm": 3.9055089950561523,
      "learning_rate": 0.00010232558139534885,
      "loss": 0.4737,
      "step": 420
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.440263032913208,
      "learning_rate": 0.0001,
      "loss": 0.35,
      "step": 430
    },
    {
      "epoch": 2.046511627906977,
      "grad_norm": 1.907047986984253,
      "learning_rate": 9.767441860465116e-05,
      "loss": 0.3646,
      "step": 440
    },
    {
      "epoch": 2.0930232558139537,
      "grad_norm": 2.902924060821533,
      "learning_rate": 9.534883720930233e-05,
      "loss": 0.3787,
      "step": 450
    },
    {
      "epoch": 2.13953488372093,
      "grad_norm": 2.807384729385376,
      "learning_rate": 9.30232558139535e-05,
      "loss": 0.3122,
      "step": 460
    },
    {
      "epoch": 2.186046511627907,
      "grad_norm": 1.7866243124008179,
      "learning_rate": 9.069767441860465e-05,
      "loss": 0.3379,
      "step": 470
    },
    {
      "epoch": 2.2325581395348837,
      "grad_norm": 6.933365821838379,
      "learning_rate": 8.837209302325582e-05,
      "loss": 0.3066,
      "step": 480
    },
    {
      "epoch": 2.2790697674418605,
      "grad_norm": 5.515610694885254,
      "learning_rate": 8.604651162790697e-05,
      "loss": 0.2632,
      "step": 490
    },
    {
      "epoch": 2.3255813953488373,
      "grad_norm": 4.792200088500977,
      "learning_rate": 8.372093023255814e-05,
      "loss": 0.2836,
      "step": 500
    },
    {
      "epoch": 2.3255813953488373,
      "eval_accuracy": 0.8745874587458746,
      "eval_loss": 0.41001710295677185,
      "eval_runtime": 8.2547,
      "eval_samples_per_second": 73.413,
      "eval_steps_per_second": 9.207,
      "step": 500
    },
    {
      "epoch": 2.3720930232558137,
      "grad_norm": 4.973999977111816,
      "learning_rate": 8.139534883720931e-05,
      "loss": 0.3589,
      "step": 510
    },
    {
      "epoch": 2.4186046511627906,
      "grad_norm": 2.9822804927825928,
      "learning_rate": 7.906976744186047e-05,
      "loss": 0.2937,
      "step": 520
    },
    {
      "epoch": 2.4651162790697674,
      "grad_norm": 3.735166549682617,
      "learning_rate": 7.674418604651163e-05,
      "loss": 0.3345,
      "step": 530
    },
    {
      "epoch": 2.511627906976744,
      "grad_norm": 3.042361259460449,
      "learning_rate": 7.441860465116279e-05,
      "loss": 0.3717,
      "step": 540
    },
    {
      "epoch": 2.558139534883721,
      "grad_norm": 2.4927892684936523,
      "learning_rate": 7.209302325581396e-05,
      "loss": 0.249,
      "step": 550
    },
    {
      "epoch": 2.604651162790698,
      "grad_norm": 1.5524264574050903,
      "learning_rate": 6.976744186046513e-05,
      "loss": 0.3304,
      "step": 560
    },
    {
      "epoch": 2.6511627906976747,
      "grad_norm": 0.39165279269218445,
      "learning_rate": 6.744186046511628e-05,
      "loss": 0.285,
      "step": 570
    },
    {
      "epoch": 2.697674418604651,
      "grad_norm": 1.6114171743392944,
      "learning_rate": 6.511627906976745e-05,
      "loss": 0.3908,
      "step": 580
    },
    {
      "epoch": 2.744186046511628,
      "grad_norm": 2.375959634780884,
      "learning_rate": 6.27906976744186e-05,
      "loss": 0.2845,
      "step": 590
    },
    {
      "epoch": 2.7906976744186047,
      "grad_norm": 3.077956199645996,
      "learning_rate": 6.0465116279069765e-05,
      "loss": 0.2669,
      "step": 600
    },
    {
      "epoch": 2.7906976744186047,
      "eval_accuracy": 0.8993399339933993,
      "eval_loss": 0.33891761302948,
      "eval_runtime": 8.0736,
      "eval_samples_per_second": 75.059,
      "eval_steps_per_second": 9.413,
      "step": 600
    },
    {
      "epoch": 2.8372093023255816,
      "grad_norm": 1.2785547971725464,
      "learning_rate": 5.8139534883720933e-05,
      "loss": 0.2499,
      "step": 610
    },
    {
      "epoch": 2.883720930232558,
      "grad_norm": 2.2260217666625977,
      "learning_rate": 5.5813953488372095e-05,
      "loss": 0.2065,
      "step": 620
    },
    {
      "epoch": 2.9302325581395348,
      "grad_norm": 2.7635715007781982,
      "learning_rate": 5.348837209302326e-05,
      "loss": 0.3334,
      "step": 630
    },
    {
      "epoch": 2.9767441860465116,
      "grad_norm": 3.221409797668457,
      "learning_rate": 5.1162790697674425e-05,
      "loss": 0.2453,
      "step": 640
    },
    {
      "epoch": 3.0232558139534884,
      "grad_norm": 0.77796870470047,
      "learning_rate": 4.883720930232558e-05,
      "loss": 0.2191,
      "step": 650
    },
    {
      "epoch": 3.0697674418604652,
      "grad_norm": 1.0451290607452393,
      "learning_rate": 4.651162790697675e-05,
      "loss": 0.2745,
      "step": 660
    },
    {
      "epoch": 3.116279069767442,
      "grad_norm": 4.356563091278076,
      "learning_rate": 4.418604651162791e-05,
      "loss": 0.2399,
      "step": 670
    },
    {
      "epoch": 3.1627906976744184,
      "grad_norm": 2.47353458404541,
      "learning_rate": 4.186046511627907e-05,
      "loss": 0.3016,
      "step": 680
    },
    {
      "epoch": 3.2093023255813953,
      "grad_norm": 1.1897259950637817,
      "learning_rate": 3.953488372093023e-05,
      "loss": 0.1716,
      "step": 690
    },
    {
      "epoch": 3.255813953488372,
      "grad_norm": 2.9624576568603516,
      "learning_rate": 3.7209302325581394e-05,
      "loss": 0.2304,
      "step": 700
    },
    {
      "epoch": 3.255813953488372,
      "eval_accuracy": 0.8927392739273927,
      "eval_loss": 0.36686915159225464,
      "eval_runtime": 8.0401,
      "eval_samples_per_second": 75.372,
      "eval_steps_per_second": 9.453,
      "step": 700
    },
    {
      "epoch": 3.302325581395349,
      "grad_norm": 0.4115903675556183,
      "learning_rate": 3.488372093023256e-05,
      "loss": 0.2835,
      "step": 710
    },
    {
      "epoch": 3.3488372093023258,
      "grad_norm": 3.0008704662323,
      "learning_rate": 3.2558139534883724e-05,
      "loss": 0.1414,
      "step": 720
    },
    {
      "epoch": 3.395348837209302,
      "grad_norm": 3.6043615341186523,
      "learning_rate": 3.0232558139534883e-05,
      "loss": 0.2309,
      "step": 730
    },
    {
      "epoch": 3.441860465116279,
      "grad_norm": 1.3581503629684448,
      "learning_rate": 2.7906976744186048e-05,
      "loss": 0.3342,
      "step": 740
    },
    {
      "epoch": 3.488372093023256,
      "grad_norm": 1.7710747718811035,
      "learning_rate": 2.5581395348837212e-05,
      "loss": 0.2326,
      "step": 750
    },
    {
      "epoch": 3.5348837209302326,
      "grad_norm": 3.192469835281372,
      "learning_rate": 2.3255813953488374e-05,
      "loss": 0.1234,
      "step": 760
    },
    {
      "epoch": 3.5813953488372094,
      "grad_norm": 0.3328302800655365,
      "learning_rate": 2.0930232558139536e-05,
      "loss": 0.228,
      "step": 770
    },
    {
      "epoch": 3.6279069767441863,
      "grad_norm": 2.4526288509368896,
      "learning_rate": 1.8604651162790697e-05,
      "loss": 0.1601,
      "step": 780
    },
    {
      "epoch": 3.6744186046511627,
      "grad_norm": 1.8664888143539429,
      "learning_rate": 1.6279069767441862e-05,
      "loss": 0.2098,
      "step": 790
    },
    {
      "epoch": 3.7209302325581395,
      "grad_norm": 5.262502193450928,
      "learning_rate": 1.3953488372093024e-05,
      "loss": 0.1523,
      "step": 800
    },
    {
      "epoch": 3.7209302325581395,
      "eval_accuracy": 0.8960396039603961,
      "eval_loss": 0.36768776178359985,
      "eval_runtime": 8.208,
      "eval_samples_per_second": 73.83,
      "eval_steps_per_second": 9.259,
      "step": 800
    },
    {
      "epoch": 3.7674418604651163,
      "grad_norm": 6.241558074951172,
      "learning_rate": 1.1627906976744187e-05,
      "loss": 0.1313,
      "step": 810
    },
    {
      "epoch": 3.813953488372093,
      "grad_norm": 2.1938974857330322,
      "learning_rate": 9.302325581395349e-06,
      "loss": 0.1505,
      "step": 820
    },
    {
      "epoch": 3.8604651162790695,
      "grad_norm": 2.1302292346954346,
      "learning_rate": 6.976744186046512e-06,
      "loss": 0.2306,
      "step": 830
    },
    {
      "epoch": 3.9069767441860463,
      "grad_norm": 10.263894081115723,
      "learning_rate": 4.651162790697674e-06,
      "loss": 0.2138,
      "step": 840
    },
    {
      "epoch": 3.953488372093023,
      "grad_norm": 6.005746364593506,
      "learning_rate": 2.325581395348837e-06,
      "loss": 0.1595,
      "step": 850
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.945340156555176,
      "learning_rate": 0.0,
      "loss": 0.2154,
      "step": 860
    },
    {
      "epoch": 4.0,
      "step": 860,
      "total_flos": 1.0638481718004941e+18,
      "train_loss": 0.4333847167880036,
      "train_runtime": 591.3634,
      "train_samples_per_second": 23.214,
      "train_steps_per_second": 1.454
    }
  ],
  "logging_steps": 10,
  "max_steps": 860,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 100,
  "total_flos": 1.0638481718004941e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}