{
  "best_metric": 0.7540983606557377,
  "best_model_checkpoint": "vit-base-patch16-224-R1-40\\checkpoint-994",
  "epoch": 39.73856209150327,
  "eval_steps": 500,
  "global_step": 1520,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.26,
      "learning_rate": 7.2368421052631575e-06,
      "loss": 1.3841,
      "step": 10
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.4473684210526315e-05,
      "loss": 1.3644,
      "step": 20
    },
    {
      "epoch": 0.78,
      "learning_rate": 2.1710526315789474e-05,
      "loss": 1.3233,
      "step": 30
    },
    {
      "epoch": 0.99,
      "eval_accuracy": 0.5573770491803278,
      "eval_loss": 1.2354576587677002,
      "eval_runtime": 1.2109,
      "eval_samples_per_second": 50.376,
      "eval_steps_per_second": 1.652,
      "step": 38
    },
    {
      "epoch": 1.05,
      "learning_rate": 2.894736842105263e-05,
      "loss": 1.2568,
      "step": 40
    },
    {
      "epoch": 1.31,
      "learning_rate": 3.618421052631579e-05,
      "loss": 1.1698,
      "step": 50
    },
    {
      "epoch": 1.57,
      "learning_rate": 4.342105263157895e-05,
      "loss": 1.0595,
      "step": 60
    },
    {
      "epoch": 1.83,
      "learning_rate": 5.0657894736842104e-05,
      "loss": 0.8643,
      "step": 70
    },
    {
      "epoch": 1.99,
      "eval_accuracy": 0.5901639344262295,
      "eval_loss": 0.9296518564224243,
      "eval_runtime": 0.9419,
      "eval_samples_per_second": 64.761,
      "eval_steps_per_second": 2.123,
      "step": 76
    },
    {
      "epoch": 2.09,
      "learning_rate": 5.484764542936288e-05,
      "loss": 0.7065,
      "step": 80
    },
    {
      "epoch": 2.35,
      "learning_rate": 5.446675900277008e-05,
      "loss": 0.5891,
      "step": 90
    },
    {
      "epoch": 2.61,
      "learning_rate": 5.408587257617729e-05,
      "loss": 0.4685,
      "step": 100
    },
    {
      "epoch": 2.88,
      "learning_rate": 5.370498614958449e-05,
      "loss": 0.4464,
      "step": 110
    },
    {
      "epoch": 2.98,
      "eval_accuracy": 0.639344262295082,
      "eval_loss": 1.1189780235290527,
      "eval_runtime": 0.9517,
      "eval_samples_per_second": 64.093,
      "eval_steps_per_second": 2.101,
      "step": 114
    },
    {
      "epoch": 3.14,
      "learning_rate": 5.332409972299169e-05,
      "loss": 0.3343,
      "step": 120
    },
    {
      "epoch": 3.4,
      "learning_rate": 5.294321329639889e-05,
      "loss": 0.2516,
      "step": 130
    },
    {
      "epoch": 3.66,
      "learning_rate": 5.256232686980609e-05,
      "loss": 0.2748,
      "step": 140
    },
    {
      "epoch": 3.92,
      "learning_rate": 5.21814404432133e-05,
      "loss": 0.3092,
      "step": 150
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7049180327868853,
      "eval_loss": 0.986064612865448,
      "eval_runtime": 0.9785,
      "eval_samples_per_second": 62.339,
      "eval_steps_per_second": 2.044,
      "step": 153
    },
    {
      "epoch": 4.18,
      "learning_rate": 5.18005540166205e-05,
      "loss": 0.1909,
      "step": 160
    },
    {
      "epoch": 4.44,
      "learning_rate": 5.14196675900277e-05,
      "loss": 0.2265,
      "step": 170
    },
    {
      "epoch": 4.71,
      "learning_rate": 5.1038781163434903e-05,
      "loss": 0.1834,
      "step": 180
    },
    {
      "epoch": 4.97,
      "learning_rate": 5.0657894736842104e-05,
      "loss": 0.1628,
      "step": 190
    },
    {
      "epoch": 4.99,
      "eval_accuracy": 0.6721311475409836,
      "eval_loss": 1.1220643520355225,
      "eval_runtime": 0.9649,
      "eval_samples_per_second": 63.216,
      "eval_steps_per_second": 2.073,
      "step": 191
    },
    {
      "epoch": 5.23,
      "learning_rate": 5.027700831024931e-05,
      "loss": 0.1763,
      "step": 200
    },
    {
      "epoch": 5.49,
      "learning_rate": 4.989612188365651e-05,
      "loss": 0.13,
      "step": 210
    },
    {
      "epoch": 5.75,
      "learning_rate": 4.9515235457063714e-05,
      "loss": 0.121,
      "step": 220
    },
    {
      "epoch": 5.99,
      "eval_accuracy": 0.6885245901639344,
      "eval_loss": 1.170951247215271,
      "eval_runtime": 0.977,
      "eval_samples_per_second": 62.435,
      "eval_steps_per_second": 2.047,
      "step": 229
    },
    {
      "epoch": 6.01,
      "learning_rate": 4.9134349030470915e-05,
      "loss": 0.1121,
      "step": 230
    },
    {
      "epoch": 6.27,
      "learning_rate": 4.8753462603878116e-05,
      "loss": 0.1008,
      "step": 240
    },
    {
      "epoch": 6.54,
      "learning_rate": 4.837257617728532e-05,
      "loss": 0.12,
      "step": 250
    },
    {
      "epoch": 6.8,
      "learning_rate": 4.7991689750692524e-05,
      "loss": 0.1138,
      "step": 260
    },
    {
      "epoch": 6.98,
      "eval_accuracy": 0.7213114754098361,
      "eval_loss": 1.199324607849121,
      "eval_runtime": 0.9471,
      "eval_samples_per_second": 64.409,
      "eval_steps_per_second": 2.112,
      "step": 267
    },
    {
      "epoch": 7.06,
      "learning_rate": 4.7610803324099725e-05,
      "loss": 0.1016,
      "step": 270
    },
    {
      "epoch": 7.32,
      "learning_rate": 4.7229916897506926e-05,
      "loss": 0.0953,
      "step": 280
    },
    {
      "epoch": 7.58,
      "learning_rate": 4.684903047091413e-05,
      "loss": 0.1073,
      "step": 290
    },
    {
      "epoch": 7.84,
      "learning_rate": 4.6468144044321335e-05,
      "loss": 0.1124,
      "step": 300
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6885245901639344,
      "eval_loss": 1.2635865211486816,
      "eval_runtime": 0.9388,
      "eval_samples_per_second": 64.974,
      "eval_steps_per_second": 2.13,
      "step": 306
    },
    {
      "epoch": 8.1,
      "learning_rate": 4.6087257617728535e-05,
      "loss": 0.0796,
      "step": 310
    },
    {
      "epoch": 8.37,
      "learning_rate": 4.570637119113573e-05,
      "loss": 0.0944,
      "step": 320
    },
    {
      "epoch": 8.63,
      "learning_rate": 4.532548476454294e-05,
      "loss": 0.0701,
      "step": 330
    },
    {
      "epoch": 8.89,
      "learning_rate": 4.494459833795014e-05,
      "loss": 0.0748,
      "step": 340
    },
    {
      "epoch": 8.99,
      "eval_accuracy": 0.7049180327868853,
      "eval_loss": 1.388106346130371,
      "eval_runtime": 0.9485,
      "eval_samples_per_second": 64.314,
      "eval_steps_per_second": 2.109,
      "step": 344
    },
    {
      "epoch": 9.15,
      "learning_rate": 4.4563711911357346e-05,
      "loss": 0.0708,
      "step": 350
    },
    {
      "epoch": 9.41,
      "learning_rate": 4.418282548476455e-05,
      "loss": 0.1,
      "step": 360
    },
    {
      "epoch": 9.67,
      "learning_rate": 4.380193905817174e-05,
      "loss": 0.0566,
      "step": 370
    },
    {
      "epoch": 9.93,
      "learning_rate": 4.342105263157895e-05,
      "loss": 0.0877,
      "step": 380
    },
    {
      "epoch": 9.99,
      "eval_accuracy": 0.7213114754098361,
      "eval_loss": 1.2892450094223022,
      "eval_runtime": 0.9419,
      "eval_samples_per_second": 64.76,
      "eval_steps_per_second": 2.123,
      "step": 382
    },
    {
      "epoch": 10.2,
      "learning_rate": 4.304016620498615e-05,
      "loss": 0.0929,
      "step": 390
    },
    {
      "epoch": 10.46,
      "learning_rate": 4.265927977839336e-05,
      "loss": 0.1008,
      "step": 400
    },
    {
      "epoch": 10.72,
      "learning_rate": 4.227839335180056e-05,
      "loss": 0.0762,
      "step": 410
    },
    {
      "epoch": 10.98,
      "learning_rate": 4.189750692520776e-05,
      "loss": 0.0642,
      "step": 420
    },
    {
      "epoch": 10.98,
      "eval_accuracy": 0.7049180327868853,
      "eval_loss": 1.3759475946426392,
      "eval_runtime": 0.9604,
      "eval_samples_per_second": 63.518,
      "eval_steps_per_second": 2.083,
      "step": 420
    },
    {
      "epoch": 11.24,
      "learning_rate": 4.151662049861496e-05,
      "loss": 0.0588,
      "step": 430
    },
    {
      "epoch": 11.5,
      "learning_rate": 4.113573407202216e-05,
      "loss": 0.0582,
      "step": 440
    },
    {
      "epoch": 11.76,
      "learning_rate": 4.075484764542937e-05,
      "loss": 0.0675,
      "step": 450
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.7213114754098361,
      "eval_loss": 1.4282864332199097,
      "eval_runtime": 0.9099,
      "eval_samples_per_second": 67.038,
      "eval_steps_per_second": 2.198,
      "step": 459
    },
    {
      "epoch": 12.03,
      "learning_rate": 4.037396121883656e-05,
      "loss": 0.066,
      "step": 460
    },
    {
      "epoch": 12.29,
      "learning_rate": 3.999307479224377e-05,
      "loss": 0.0565,
      "step": 470
    },
    {
      "epoch": 12.55,
      "learning_rate": 3.961218836565097e-05,
      "loss": 0.0619,
      "step": 480
    },
    {
      "epoch": 12.81,
      "learning_rate": 3.923130193905817e-05,
      "loss": 0.0694,
      "step": 490
    },
    {
      "epoch": 12.99,
      "eval_accuracy": 0.7213114754098361,
      "eval_loss": 1.361587643623352,
      "eval_runtime": 0.9412,
      "eval_samples_per_second": 64.813,
      "eval_steps_per_second": 2.125,
      "step": 497
    },
    {
      "epoch": 13.07,
      "learning_rate": 3.885041551246538e-05,
      "loss": 0.074,
      "step": 500
    },
    {
      "epoch": 13.33,
      "learning_rate": 3.8469529085872574e-05,
      "loss": 0.0706,
      "step": 510
    },
    {
      "epoch": 13.59,
      "learning_rate": 3.808864265927978e-05,
      "loss": 0.0495,
      "step": 520
    },
    {
      "epoch": 13.86,
      "learning_rate": 3.770775623268698e-05,
      "loss": 0.0689,
      "step": 530
    },
    {
      "epoch": 13.99,
      "eval_accuracy": 0.7213114754098361,
      "eval_loss": 1.3863976001739502,
      "eval_runtime": 1.0042,
      "eval_samples_per_second": 60.748,
      "eval_steps_per_second": 1.992,
      "step": 535
    },
    {
      "epoch": 14.12,
      "learning_rate": 3.732686980609418e-05,
      "loss": 0.049,
      "step": 540
    },
    {
      "epoch": 14.38,
      "learning_rate": 3.694598337950139e-05,
      "loss": 0.0489,
      "step": 550
    },
    {
      "epoch": 14.64,
      "learning_rate": 3.6565096952908585e-05,
      "loss": 0.0458,
      "step": 560
    },
    {
      "epoch": 14.9,
      "learning_rate": 3.618421052631579e-05,
      "loss": 0.0378,
      "step": 570
    },
    {
      "epoch": 14.98,
      "eval_accuracy": 0.7213114754098361,
      "eval_loss": 1.4321750402450562,
      "eval_runtime": 0.9754,
      "eval_samples_per_second": 62.537,
      "eval_steps_per_second": 2.05,
      "step": 573
    },
    {
      "epoch": 15.16,
      "learning_rate": 3.5803324099722994e-05,
      "loss": 0.0311,
      "step": 580
    },
    {
      "epoch": 15.42,
      "learning_rate": 3.5422437673130195e-05,
      "loss": 0.0626,
      "step": 590
    },
    {
      "epoch": 15.69,
      "learning_rate": 3.5041551246537395e-05,
      "loss": 0.0515,
      "step": 600
    },
    {
      "epoch": 15.95,
      "learning_rate": 3.4660664819944596e-05,
      "loss": 0.0472,
      "step": 610
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.7213114754098361,
      "eval_loss": 1.600350260734558,
      "eval_runtime": 0.907,
      "eval_samples_per_second": 67.252,
      "eval_steps_per_second": 2.205,
      "step": 612
    },
    {
      "epoch": 16.21,
      "learning_rate": 3.4279778393351804e-05,
      "loss": 0.0378,
      "step": 620
    },
    {
      "epoch": 16.47,
      "learning_rate": 3.3898891966759005e-05,
      "loss": 0.0345,
      "step": 630
    },
    {
      "epoch": 16.73,
      "learning_rate": 3.3518005540166206e-05,
      "loss": 0.0591,
      "step": 640
    },
    {
      "epoch": 16.99,
      "learning_rate": 3.313711911357341e-05,
      "loss": 0.044,
      "step": 650
    },
    {
      "epoch": 16.99,
      "eval_accuracy": 0.7049180327868853,
      "eval_loss": 1.5810414552688599,
      "eval_runtime": 0.9357,
      "eval_samples_per_second": 65.194,
      "eval_steps_per_second": 2.138,
      "step": 650
    },
    {
      "epoch": 17.25,
      "learning_rate": 3.275623268698061e-05,
      "loss": 0.0495,
      "step": 660
    },
    {
      "epoch": 17.52,
      "learning_rate": 3.2375346260387815e-05,
      "loss": 0.0409,
      "step": 670
    },
    {
      "epoch": 17.78,
      "learning_rate": 3.1994459833795016e-05,
      "loss": 0.0386,
      "step": 680
    },
    {
      "epoch": 17.99,
      "eval_accuracy": 0.6885245901639344,
      "eval_loss": 1.640351414680481,
      "eval_runtime": 0.9225,
      "eval_samples_per_second": 66.127,
      "eval_steps_per_second": 2.168,
      "step": 688
    },
    {
      "epoch": 18.04,
      "learning_rate": 3.161357340720222e-05,
      "loss": 0.0323,
      "step": 690
    },
    {
      "epoch": 18.3,
      "learning_rate": 3.123268698060942e-05,
      "loss": 0.0399,
      "step": 700
    },
    {
      "epoch": 18.56,
      "learning_rate": 3.085180055401662e-05,
      "loss": 0.0475,
      "step": 710
    },
    {
      "epoch": 18.82,
      "learning_rate": 3.0470914127423827e-05,
      "loss": 0.0341,
      "step": 720
    },
    {
      "epoch": 18.98,
      "eval_accuracy": 0.7377049180327869,
      "eval_loss": 1.5698273181915283,
      "eval_runtime": 0.9437,
      "eval_samples_per_second": 64.641,
      "eval_steps_per_second": 2.119,
      "step": 726
    },
    {
      "epoch": 19.08,
      "learning_rate": 3.0090027700831024e-05,
      "loss": 0.0457,
      "step": 730
    },
    {
      "epoch": 19.35,
      "learning_rate": 2.9709141274238225e-05,
      "loss": 0.0337,
      "step": 740
    },
    {
      "epoch": 19.61,
      "learning_rate": 2.9328254847645433e-05,
      "loss": 0.0291,
      "step": 750
    },
    {
      "epoch": 19.87,
      "learning_rate": 2.894736842105263e-05,
      "loss": 0.0328,
      "step": 760
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.6885245901639344,
      "eval_loss": 1.6720421314239502,
      "eval_runtime": 0.925,
      "eval_samples_per_second": 65.945,
      "eval_steps_per_second": 2.162,
      "step": 765
    },
    {
      "epoch": 20.13,
      "learning_rate": 2.8566481994459838e-05,
      "loss": 0.0226,
      "step": 770
    },
    {
      "epoch": 20.39,
      "learning_rate": 2.8185595567867035e-05,
      "loss": 0.027,
      "step": 780
    },
    {
      "epoch": 20.65,
      "learning_rate": 2.7804709141274236e-05,
      "loss": 0.0241,
      "step": 790
    },
    {
      "epoch": 20.92,
      "learning_rate": 2.742382271468144e-05,
      "loss": 0.0444,
      "step": 800
    },
    {
      "epoch": 20.99,
      "eval_accuracy": 0.7213114754098361,
      "eval_loss": 1.6269422769546509,
      "eval_runtime": 0.9772,
      "eval_samples_per_second": 62.423,
      "eval_steps_per_second": 2.047,
      "step": 803
    },
    {
      "epoch": 21.18,
      "learning_rate": 2.7042936288088645e-05,
      "loss": 0.0312,
      "step": 810
    },
    {
      "epoch": 21.44,
      "learning_rate": 2.6662049861495846e-05,
      "loss": 0.0254,
      "step": 820
    },
    {
      "epoch": 21.7,
      "learning_rate": 2.6281163434903047e-05,
      "loss": 0.0239,
      "step": 830
    },
    {
      "epoch": 21.96,
      "learning_rate": 2.590027700831025e-05,
      "loss": 0.0342,
      "step": 840
    },
    {
      "epoch": 21.99,
      "eval_accuracy": 0.7377049180327869,
      "eval_loss": 1.6344714164733887,
      "eval_runtime": 0.938,
      "eval_samples_per_second": 65.032,
      "eval_steps_per_second": 2.132,
      "step": 841
    },
    {
      "epoch": 22.22,
      "learning_rate": 2.5519390581717452e-05,
      "loss": 0.0203,
      "step": 850
    },
    {
      "epoch": 22.48,
      "learning_rate": 2.5138504155124656e-05,
      "loss": 0.0249,
      "step": 860
    },
    {
      "epoch": 22.75,
      "learning_rate": 2.4757617728531857e-05,
      "loss": 0.0324,
      "step": 870
    },
    {
      "epoch": 22.98,
      "eval_accuracy": 0.7049180327868853,
      "eval_loss": 1.7916051149368286,
      "eval_runtime": 0.9514,
      "eval_samples_per_second": 64.117,
      "eval_steps_per_second": 2.102,
      "step": 879
    },
    {
      "epoch": 23.01,
      "learning_rate": 2.4376731301939058e-05,
      "loss": 0.0287,
      "step": 880
    },
    {
      "epoch": 23.27,
      "learning_rate": 2.3995844875346262e-05,
      "loss": 0.0389,
      "step": 890
    },
    {
      "epoch": 23.53,
      "learning_rate": 2.3614958448753463e-05,
      "loss": 0.0333,
      "step": 900
    },
    {
      "epoch": 23.79,
      "learning_rate": 2.3234072022160667e-05,
      "loss": 0.023,
      "step": 910
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.6885245901639344,
      "eval_loss": 1.8753165006637573,
      "eval_runtime": 0.9604,
      "eval_samples_per_second": 63.517,
      "eval_steps_per_second": 2.083,
      "step": 918
    },
    {
      "epoch": 24.05,
      "learning_rate": 2.2853185595567865e-05,
      "loss": 0.0264,
      "step": 920
    },
    {
      "epoch": 24.31,
      "learning_rate": 2.247229916897507e-05,
      "loss": 0.0305,
      "step": 930
    },
    {
      "epoch": 24.58,
      "learning_rate": 2.2091412742382273e-05,
      "loss": 0.0265,
      "step": 940
    },
    {
      "epoch": 24.84,
      "learning_rate": 2.1710526315789474e-05,
      "loss": 0.048,
      "step": 950
    },
    {
      "epoch": 24.99,
      "eval_accuracy": 0.7377049180327869,
      "eval_loss": 1.767918586730957,
      "eval_runtime": 0.9362,
      "eval_samples_per_second": 65.159,
      "eval_steps_per_second": 2.136,
      "step": 956
    },
    {
      "epoch": 25.1,
      "learning_rate": 2.132963988919668e-05,
      "loss": 0.0378,
      "step": 960
    },
    {
      "epoch": 25.36,
      "learning_rate": 2.094875346260388e-05,
      "loss": 0.0326,
      "step": 970
    },
    {
      "epoch": 25.62,
      "learning_rate": 2.056786703601108e-05,
      "loss": 0.0228,
      "step": 980
    },
    {
      "epoch": 25.88,
      "learning_rate": 2.018698060941828e-05,
      "loss": 0.0202,
      "step": 990
    },
    {
      "epoch": 25.99,
      "eval_accuracy": 0.7540983606557377,
      "eval_loss": 1.7211973667144775,
      "eval_runtime": 0.9572,
      "eval_samples_per_second": 63.731,
      "eval_steps_per_second": 2.09,
      "step": 994
    },
    {
      "epoch": 26.14,
      "learning_rate": 1.9806094182825486e-05,
      "loss": 0.023,
      "step": 1000
    },
    {
      "epoch": 26.41,
      "learning_rate": 1.942520775623269e-05,
      "loss": 0.0152,
      "step": 1010
    },
    {
      "epoch": 26.67,
      "learning_rate": 1.904432132963989e-05,
      "loss": 0.0267,
      "step": 1020
    },
    {
      "epoch": 26.93,
      "learning_rate": 1.866343490304709e-05,
      "loss": 0.0336,
      "step": 1030
    },
    {
      "epoch": 26.98,
      "eval_accuracy": 0.7377049180327869,
      "eval_loss": 1.7305350303649902,
      "eval_runtime": 0.9424,
      "eval_samples_per_second": 64.731,
      "eval_steps_per_second": 2.122,
      "step": 1032
    },
    {
      "epoch": 27.19,
      "learning_rate": 1.8282548476454293e-05,
      "loss": 0.0205,
      "step": 1040
    },
    {
      "epoch": 27.45,
      "learning_rate": 1.7901662049861497e-05,
      "loss": 0.0316,
      "step": 1050
    },
    {
      "epoch": 27.71,
      "learning_rate": 1.7520775623268698e-05,
      "loss": 0.023,
      "step": 1060
    },
    {
      "epoch": 27.97,
      "learning_rate": 1.7139889196675902e-05,
      "loss": 0.0163,
      "step": 1070
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.7049180327868853,
      "eval_loss": 1.7576478719711304,
      "eval_runtime": 1.0009,
      "eval_samples_per_second": 60.943,
      "eval_steps_per_second": 1.998,
      "step": 1071
    },
    {
      "epoch": 28.24,
      "learning_rate": 1.6759002770083103e-05,
      "loss": 0.0183,
      "step": 1080
    },
    {
      "epoch": 28.5,
      "learning_rate": 1.6378116343490304e-05,
      "loss": 0.0432,
      "step": 1090
    },
    {
      "epoch": 28.76,
      "learning_rate": 1.5997229916897508e-05,
      "loss": 0.0186,
      "step": 1100
    },
    {
      "epoch": 28.99,
      "eval_accuracy": 0.7377049180327869,
      "eval_loss": 1.7540286779403687,
      "eval_runtime": 0.9541,
      "eval_samples_per_second": 63.935,
      "eval_steps_per_second": 2.096,
      "step": 1109
    },
    {
      "epoch": 29.02,
      "learning_rate": 1.561634349030471e-05,
      "loss": 0.0181,
      "step": 1110
    },
    {
      "epoch": 29.28,
      "learning_rate": 1.5235457063711913e-05,
      "loss": 0.0186,
      "step": 1120
    },
    {
      "epoch": 29.54,
      "learning_rate": 1.4854570637119112e-05,
      "loss": 0.0112,
      "step": 1130
    },
    {
      "epoch": 29.8,
      "learning_rate": 1.4473684210526315e-05,
      "loss": 0.0189,
      "step": 1140
    },
    {
      "epoch": 29.99,
      "eval_accuracy": 0.7540983606557377,
      "eval_loss": 1.6593692302703857,
      "eval_runtime": 0.9808,
      "eval_samples_per_second": 62.194,
      "eval_steps_per_second": 2.039,
      "step": 1147
    },
    {
      "epoch": 30.07,
      "learning_rate": 1.4092797783933518e-05,
      "loss": 0.0205,
      "step": 1150
    },
    {
      "epoch": 30.33,
      "learning_rate": 1.371191135734072e-05,
      "loss": 0.0135,
      "step": 1160
    },
    {
      "epoch": 30.59,
      "learning_rate": 1.3331024930747923e-05,
      "loss": 0.0116,
      "step": 1170
    },
    {
      "epoch": 30.85,
      "learning_rate": 1.2950138504155125e-05,
      "loss": 0.039,
      "step": 1180
    },
    {
      "epoch": 30.98,
      "eval_accuracy": 0.7213114754098361,
      "eval_loss": 1.7422717809677124,
      "eval_runtime": 0.9576,
      "eval_samples_per_second": 63.702,
      "eval_steps_per_second": 2.089,
      "step": 1185
    },
    {
      "epoch": 31.11,
      "learning_rate": 1.2569252077562328e-05,
      "loss": 0.0209,
      "step": 1190
    },
    {
      "epoch": 31.37,
      "learning_rate": 1.2188365650969529e-05,
      "loss": 0.0217,
      "step": 1200
    },
    {
      "epoch": 31.63,
      "learning_rate": 1.1807479224376732e-05,
      "loss": 0.0242,
      "step": 1210
    },
    {
      "epoch": 31.9,
      "learning_rate": 1.1426592797783932e-05,
      "loss": 0.0194,
      "step": 1220
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.7377049180327869,
      "eval_loss": 1.7148489952087402,
      "eval_runtime": 1.0439,
      "eval_samples_per_second": 58.435,
      "eval_steps_per_second": 1.916,
      "step": 1224
    },
    {
      "epoch": 32.16,
      "learning_rate": 1.1045706371191137e-05,
      "loss": 0.0153,
      "step": 1230
    },
    {
      "epoch": 32.42,
      "learning_rate": 1.066481994459834e-05,
      "loss": 0.0214,
      "step": 1240
    },
    {
      "epoch": 32.68,
      "learning_rate": 1.028393351800554e-05,
      "loss": 0.017,
      "step": 1250
    },
    {
      "epoch": 32.94,
      "learning_rate": 9.903047091412743e-06,
      "loss": 0.0205,
      "step": 1260
    },
    {
      "epoch": 32.99,
      "eval_accuracy": 0.7377049180327869,
      "eval_loss": 1.6965419054031372,
      "eval_runtime": 0.9377,
      "eval_samples_per_second": 65.055,
      "eval_steps_per_second": 2.133,
      "step": 1262
    },
    {
      "epoch": 33.2,
      "learning_rate": 9.522160664819945e-06,
      "loss": 0.0069,
      "step": 1270
    },
    {
      "epoch": 33.46,
      "learning_rate": 9.141274238227146e-06,
      "loss": 0.0153,
      "step": 1280
    },
    {
      "epoch": 33.73,
      "learning_rate": 8.760387811634349e-06,
      "loss": 0.024,
      "step": 1290
    },
    {
      "epoch": 33.99,
      "learning_rate": 8.379501385041551e-06,
      "loss": 0.0186,
      "step": 1300
    },
    {
      "epoch": 33.99,
      "eval_accuracy": 0.7540983606557377,
      "eval_loss": 1.7553398609161377,
      "eval_runtime": 0.9513,
      "eval_samples_per_second": 64.125,
      "eval_steps_per_second": 2.102,
      "step": 1300
    },
    {
      "epoch": 34.25,
      "learning_rate": 7.998614958448754e-06,
      "loss": 0.0094,
      "step": 1310
    },
    {
      "epoch": 34.51,
      "learning_rate": 7.617728531855957e-06,
      "loss": 0.0176,
      "step": 1320
    },
    {
      "epoch": 34.77,
      "learning_rate": 7.2368421052631575e-06,
      "loss": 0.0177,
      "step": 1330
    },
    {
      "epoch": 34.98,
      "eval_accuracy": 0.7377049180327869,
      "eval_loss": 1.7475531101226807,
      "eval_runtime": 0.9569,
      "eval_samples_per_second": 63.746,
      "eval_steps_per_second": 2.09,
      "step": 1338
    },
    {
      "epoch": 35.03,
      "learning_rate": 6.85595567867036e-06,
      "loss": 0.0166,
      "step": 1340
    },
    {
      "epoch": 35.29,
      "learning_rate": 6.475069252077563e-06,
      "loss": 0.0109,
      "step": 1350
    },
    {
      "epoch": 35.56,
      "learning_rate": 6.0941828254847645e-06,
      "loss": 0.0134,
      "step": 1360
    },
    {
      "epoch": 35.82,
      "learning_rate": 5.713296398891966e-06,
      "loss": 0.0132,
      "step": 1370
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.7540983606557377,
      "eval_loss": 1.750593662261963,
      "eval_runtime": 0.9217,
      "eval_samples_per_second": 66.183,
      "eval_steps_per_second": 2.17,
      "step": 1377
    },
    {
      "epoch": 36.08,
      "learning_rate": 5.33240997229917e-06,
      "loss": 0.0117,
      "step": 1380
    },
    {
      "epoch": 36.34,
      "learning_rate": 4.951523545706371e-06,
      "loss": 0.0091,
      "step": 1390
    },
    {
      "epoch": 36.6,
      "learning_rate": 4.570637119113573e-06,
      "loss": 0.0144,
      "step": 1400
    },
    {
      "epoch": 36.86,
      "learning_rate": 4.189750692520776e-06,
      "loss": 0.0068,
      "step": 1410
    },
    {
      "epoch": 36.99,
      "eval_accuracy": 0.7377049180327869,
      "eval_loss": 1.6917140483856201,
      "eval_runtime": 0.9809,
      "eval_samples_per_second": 62.188,
      "eval_steps_per_second": 2.039,
      "step": 1415
    },
    {
      "epoch": 37.12,
      "learning_rate": 3.8088642659279783e-06,
      "loss": 0.0165,
      "step": 1420
    },
    {
      "epoch": 37.39,
      "learning_rate": 3.42797783933518e-06,
      "loss": 0.0159,
      "step": 1430
    },
    {
      "epoch": 37.65,
      "learning_rate": 3.0470914127423822e-06,
      "loss": 0.0155,
      "step": 1440
    },
    {
      "epoch": 37.91,
      "learning_rate": 2.666204986149585e-06,
      "loss": 0.0121,
      "step": 1450
    },
    {
      "epoch": 37.99,
      "eval_accuracy": 0.7540983606557377,
      "eval_loss": 1.727645993232727,
      "eval_runtime": 1.0046,
      "eval_samples_per_second": 60.719,
      "eval_steps_per_second": 1.991,
      "step": 1453
    },
    {
      "epoch": 38.17,
      "learning_rate": 2.2853185595567866e-06,
      "loss": 0.0115,
      "step": 1460
    },
    {
      "epoch": 38.43,
      "learning_rate": 1.9044321329639892e-06,
      "loss": 0.0094,
      "step": 1470
    },
    {
      "epoch": 38.69,
      "learning_rate": 1.5235457063711911e-06,
      "loss": 0.0095,
      "step": 1480
    },
    {
      "epoch": 38.95,
      "learning_rate": 1.1426592797783933e-06,
      "loss": 0.0129,
      "step": 1490
    },
    {
      "epoch": 38.98,
      "eval_accuracy": 0.7540983606557377,
      "eval_loss": 1.7218044996261597,
      "eval_runtime": 0.9303,
      "eval_samples_per_second": 65.571,
      "eval_steps_per_second": 2.15,
      "step": 1491
    },
    {
      "epoch": 39.22,
      "learning_rate": 7.617728531855956e-07,
      "loss": 0.0077,
      "step": 1500
    },
    {
      "epoch": 39.48,
      "learning_rate": 3.808864265927978e-07,
      "loss": 0.0115,
      "step": 1510
    },
    {
      "epoch": 39.74,
      "learning_rate": 0.0,
      "loss": 0.0067,
      "step": 1520
    },
    {
      "epoch": 39.74,
      "eval_accuracy": 0.7540983606557377,
      "eval_loss": 1.7220282554626465,
      "eval_runtime": 0.9466,
      "eval_samples_per_second": 64.441,
      "eval_steps_per_second": 2.113,
      "step": 1520
    },
    {
      "epoch": 39.74,
      "step": 1520,
      "total_flos": 1.5028755889037378e+19,
      "train_loss": 0.11905086408124159,
      "train_runtime": 3040.3234,
      "train_samples_per_second": 64.204,
      "train_steps_per_second": 0.5
    }
  ],
  "logging_steps": 10,
  "max_steps": 1520,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 40,
  "save_steps": 500,
  "total_flos": 1.5028755889037378e+19,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}