{
  "best_metric": 0.1423337161540985,
  "best_model_checkpoint": "./models/toxicity-scorer-qwen/checkpoint-8816",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 8816,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0,
      "eval_accuracy": 0.5782279248332997,
      "eval_f1": 0.6486677195396867,
      "eval_loss": 2.1836156845092773,
      "eval_precision": 0.8024809024681206,
      "eval_recall": 0.5782279248332997,
      "eval_runtime": 121.6957,
      "eval_samples_per_second": 488.004,
      "eval_steps_per_second": 3.813,
      "step": 0
    },
    {
      "epoch": 0.001134301270417423,
      "grad_norm": 75.0,
      "learning_rate": 1.1342155009451796e-07,
      "loss": 2.1801,
      "step": 10
    },
    {
      "epoch": 0.002268602540834846,
      "grad_norm": 72.5,
      "learning_rate": 2.2684310018903592e-07,
      "loss": 2.2173,
      "step": 20
    },
    {
      "epoch": 0.0034029038112522686,
      "grad_norm": 70.5,
      "learning_rate": 3.402646502835539e-07,
      "loss": 2.1489,
      "step": 30
    },
    {
      "epoch": 0.004537205081669692,
      "grad_norm": 71.5,
      "learning_rate": 4.5368620037807185e-07,
      "loss": 2.1703,
      "step": 40
    },
    {
      "epoch": 0.0056715063520871144,
      "grad_norm": 70.5,
      "learning_rate": 5.671077504725898e-07,
      "loss": 2.2116,
      "step": 50
    },
    {
      "epoch": 0.006805807622504537,
      "grad_norm": 73.5,
      "learning_rate": 6.805293005671078e-07,
      "loss": 2.1757,
      "step": 60
    },
    {
      "epoch": 0.00794010889292196,
      "grad_norm": 71.0,
      "learning_rate": 7.939508506616257e-07,
      "loss": 2.1018,
      "step": 70
    },
    {
      "epoch": 0.009074410163339383,
      "grad_norm": 70.0,
      "learning_rate": 9.073724007561437e-07,
      "loss": 2.1104,
      "step": 80
    },
    {
      "epoch": 0.010208711433756805,
      "grad_norm": 70.5,
      "learning_rate": 1.0207939508506615e-06,
      "loss": 2.0932,
      "step": 90
    },
    {
      "epoch": 0.011343012704174229,
      "grad_norm": 74.5,
      "learning_rate": 1.1342155009451797e-06,
      "loss": 2.0352,
      "step": 100
    },
    {
      "epoch": 0.01247731397459165,
      "grad_norm": 68.5,
      "learning_rate": 1.2476370510396976e-06,
      "loss": 1.946,
      "step": 110
    },
    {
      "epoch": 0.013611615245009074,
      "grad_norm": 74.0,
      "learning_rate": 1.3610586011342156e-06,
      "loss": 1.843,
      "step": 120
    },
    {
      "epoch": 0.014745916515426498,
      "grad_norm": 70.0,
      "learning_rate": 1.4744801512287333e-06,
      "loss": 1.8038,
      "step": 130
    },
    {
      "epoch": 0.01588021778584392,
      "grad_norm": 72.5,
      "learning_rate": 1.5879017013232515e-06,
      "loss": 1.7602,
      "step": 140
    },
    {
      "epoch": 0.01701451905626134,
      "grad_norm": 98.0,
      "learning_rate": 1.7013232514177694e-06,
      "loss": 1.633,
      "step": 150
    },
    {
      "epoch": 0.018148820326678767,
      "grad_norm": 189.0,
      "learning_rate": 1.8147448015122874e-06,
      "loss": 1.527,
      "step": 160
    },
    {
      "epoch": 0.01928312159709619,
      "grad_norm": 290.0,
      "learning_rate": 1.9281663516068055e-06,
      "loss": 1.3534,
      "step": 170
    },
    {
      "epoch": 0.02041742286751361,
      "grad_norm": 201.0,
      "learning_rate": 2.041587901701323e-06,
      "loss": 0.9993,
      "step": 180
    },
    {
      "epoch": 0.021551724137931036,
      "grad_norm": 126.0,
      "learning_rate": 2.1550094517958414e-06,
      "loss": 0.7995,
      "step": 190
    },
    {
      "epoch": 0.022686025408348458,
      "grad_norm": 72.5,
      "learning_rate": 2.2684310018903594e-06,
      "loss": 0.781,
      "step": 200
    },
    {
      "epoch": 0.02382032667876588,
      "grad_norm": 56.5,
      "learning_rate": 2.381852551984877e-06,
      "loss": 0.683,
      "step": 210
    },
    {
      "epoch": 0.0249546279491833,
      "grad_norm": 47.25,
      "learning_rate": 2.4952741020793953e-06,
      "loss": 0.6598,
      "step": 220
    },
    {
      "epoch": 0.026088929219600727,
      "grad_norm": 47.0,
      "learning_rate": 2.6086956521739132e-06,
      "loss": 0.6226,
      "step": 230
    },
    {
      "epoch": 0.02722323049001815,
      "grad_norm": 40.75,
      "learning_rate": 2.722117202268431e-06,
      "loss": 0.5713,
      "step": 240
    },
    {
      "epoch": 0.02835753176043557,
      "grad_norm": 44.0,
      "learning_rate": 2.835538752362949e-06,
      "loss": 0.5529,
      "step": 250
    },
    {
      "epoch": 0.029491833030852996,
      "grad_norm": 54.25,
      "learning_rate": 2.9489603024574667e-06,
      "loss": 0.4694,
      "step": 260
    },
    {
      "epoch": 0.030626134301270418,
      "grad_norm": 34.25,
      "learning_rate": 3.062381852551985e-06,
      "loss": 0.4789,
      "step": 270
    },
    {
      "epoch": 0.03176043557168784,
      "grad_norm": 35.5,
      "learning_rate": 3.175803402646503e-06,
      "loss": 0.4618,
      "step": 280
    },
    {
      "epoch": 0.03289473684210526,
      "grad_norm": 37.75,
      "learning_rate": 3.289224952741021e-06,
      "loss": 0.4601,
      "step": 290
    },
    {
      "epoch": 0.03402903811252268,
      "grad_norm": 33.25,
      "learning_rate": 3.402646502835539e-06,
      "loss": 0.4079,
      "step": 300
    },
    {
      "epoch": 0.03516333938294011,
      "grad_norm": 53.25,
      "learning_rate": 3.516068052930057e-06,
      "loss": 0.4136,
      "step": 310
    },
    {
      "epoch": 0.036297640653357534,
      "grad_norm": 34.0,
      "learning_rate": 3.6294896030245748e-06,
      "loss": 0.4257,
      "step": 320
    },
    {
      "epoch": 0.037431941923774956,
      "grad_norm": 70.0,
      "learning_rate": 3.7429111531190927e-06,
      "loss": 0.3797,
      "step": 330
    },
    {
      "epoch": 0.03856624319419238,
      "grad_norm": 34.0,
      "learning_rate": 3.856332703213611e-06,
      "loss": 0.3806,
      "step": 340
    },
    {
      "epoch": 0.0397005444646098,
      "grad_norm": 31.5,
      "learning_rate": 3.969754253308128e-06,
      "loss": 0.3845,
      "step": 350
    },
    {
      "epoch": 0.04083484573502722,
      "grad_norm": 48.5,
      "learning_rate": 4.083175803402646e-06,
      "loss": 0.3884,
      "step": 360
    },
    {
      "epoch": 0.04196914700544464,
      "grad_norm": 39.5,
      "learning_rate": 4.196597353497165e-06,
      "loss": 0.3725,
      "step": 370
    },
    {
      "epoch": 0.04310344827586207,
      "grad_norm": 38.5,
      "learning_rate": 4.310018903591683e-06,
      "loss": 0.3517,
      "step": 380
    },
    {
      "epoch": 0.044237749546279494,
      "grad_norm": 39.25,
      "learning_rate": 4.4234404536862e-06,
      "loss": 0.3605,
      "step": 390
    },
    {
      "epoch": 0.045372050816696916,
      "grad_norm": 26.625,
      "learning_rate": 4.536862003780719e-06,
      "loss": 0.3566,
      "step": 400
    },
    {
      "epoch": 0.04650635208711434,
      "grad_norm": 24.75,
      "learning_rate": 4.650283553875237e-06,
      "loss": 0.3615,
      "step": 410
    },
    {
      "epoch": 0.04764065335753176,
      "grad_norm": 19.875,
      "learning_rate": 4.763705103969754e-06,
      "loss": 0.3364,
      "step": 420
    },
    {
      "epoch": 0.04877495462794918,
      "grad_norm": 40.75,
      "learning_rate": 4.877126654064273e-06,
      "loss": 0.3319,
      "step": 430
    },
    {
      "epoch": 0.0499092558983666,
      "grad_norm": 40.75,
      "learning_rate": 4.990548204158791e-06,
      "loss": 0.329,
      "step": 440
    },
    {
      "epoch": 0.05104355716878403,
      "grad_norm": 26.875,
      "learning_rate": 5.103969754253308e-06,
      "loss": 0.322,
      "step": 450
    },
    {
      "epoch": 0.052177858439201454,
      "grad_norm": 32.25,
      "learning_rate": 5.2173913043478265e-06,
      "loss": 0.3282,
      "step": 460
    },
    {
      "epoch": 0.053312159709618875,
      "grad_norm": 23.875,
      "learning_rate": 5.3308128544423444e-06,
      "loss": 0.3349,
      "step": 470
    },
    {
      "epoch": 0.0544464609800363,
      "grad_norm": 20.75,
      "learning_rate": 5.444234404536862e-06,
      "loss": 0.3183,
      "step": 480
    },
    {
      "epoch": 0.05558076225045372,
      "grad_norm": 33.75,
      "learning_rate": 5.5576559546313795e-06,
      "loss": 0.3125,
      "step": 490
    },
    {
      "epoch": 0.05671506352087114,
      "grad_norm": 23.875,
      "learning_rate": 5.671077504725898e-06,
      "loss": 0.3047,
      "step": 500
    },
    {
      "epoch": 0.05784936479128856,
      "grad_norm": 25.875,
      "learning_rate": 5.784499054820416e-06,
      "loss": 0.3113,
      "step": 510
    },
    {
      "epoch": 0.05898366606170599,
      "grad_norm": 26.75,
      "learning_rate": 5.897920604914933e-06,
      "loss": 0.2892,
      "step": 520
    },
    {
      "epoch": 0.06011796733212341,
      "grad_norm": 29.5,
      "learning_rate": 6.011342155009452e-06,
      "loss": 0.3104,
      "step": 530
    },
    {
      "epoch": 0.061252268602540835,
      "grad_norm": 20.5,
      "learning_rate": 6.12476370510397e-06,
      "loss": 0.3059,
      "step": 540
    },
    {
      "epoch": 0.06238656987295826,
      "grad_norm": 29.375,
      "learning_rate": 6.238185255198487e-06,
      "loss": 0.3083,
      "step": 550
    },
    {
      "epoch": 0.06352087114337568,
      "grad_norm": 30.625,
      "learning_rate": 6.351606805293006e-06,
      "loss": 0.294,
      "step": 560
    },
    {
      "epoch": 0.06465517241379311,
      "grad_norm": 22.0,
      "learning_rate": 6.465028355387524e-06,
      "loss": 0.3006,
      "step": 570
    },
    {
      "epoch": 0.06578947368421052,
      "grad_norm": 15.4375,
      "learning_rate": 6.578449905482042e-06,
      "loss": 0.2832,
      "step": 580
    },
    {
      "epoch": 0.06692377495462795,
      "grad_norm": 18.25,
      "learning_rate": 6.69187145557656e-06,
      "loss": 0.2919,
      "step": 590
    },
    {
      "epoch": 0.06805807622504537,
      "grad_norm": 17.75,
      "learning_rate": 6.805293005671078e-06,
      "loss": 0.2856,
      "step": 600
    },
    {
      "epoch": 0.0691923774954628,
      "grad_norm": 26.625,
      "learning_rate": 6.918714555765596e-06,
      "loss": 0.2779,
      "step": 610
    },
    {
      "epoch": 0.07032667876588022,
      "grad_norm": 20.5,
      "learning_rate": 7.032136105860114e-06,
      "loss": 0.2941,
      "step": 620
    },
    {
      "epoch": 0.07146098003629764,
      "grad_norm": 23.125,
      "learning_rate": 7.145557655954632e-06,
      "loss": 0.2805,
      "step": 630
    },
    {
      "epoch": 0.07259528130671507,
      "grad_norm": 23.625,
      "learning_rate": 7.2589792060491495e-06,
      "loss": 0.2676,
      "step": 640
    },
    {
      "epoch": 0.07372958257713248,
      "grad_norm": 10.6875,
      "learning_rate": 7.3724007561436675e-06,
      "loss": 0.2927,
      "step": 650
    },
    {
      "epoch": 0.07486388384754991,
      "grad_norm": 20.75,
      "learning_rate": 7.4858223062381854e-06,
      "loss": 0.2588,
      "step": 660
    },
    {
      "epoch": 0.07599818511796733,
      "grad_norm": 20.375,
      "learning_rate": 7.5992438563327025e-06,
      "loss": 0.2629,
      "step": 670
    },
    {
      "epoch": 0.07713248638838476,
      "grad_norm": 22.0,
      "learning_rate": 7.712665406427222e-06,
      "loss": 0.2685,
      "step": 680
    },
    {
      "epoch": 0.07826678765880218,
      "grad_norm": 20.375,
      "learning_rate": 7.826086956521738e-06,
      "loss": 0.2468,
      "step": 690
    },
    {
      "epoch": 0.0794010889292196,
      "grad_norm": 21.875,
      "learning_rate": 7.939508506616256e-06,
      "loss": 0.2488,
      "step": 700
    },
    {
      "epoch": 0.08053539019963703,
      "grad_norm": 26.5,
      "learning_rate": 8.052930056710776e-06,
      "loss": 0.2616,
      "step": 710
    },
    {
      "epoch": 0.08166969147005444,
      "grad_norm": 14.0625,
      "learning_rate": 8.166351606805292e-06,
      "loss": 0.2529,
      "step": 720
    },
    {
      "epoch": 0.08280399274047187,
      "grad_norm": 20.0,
      "learning_rate": 8.279773156899812e-06,
      "loss": 0.2425,
      "step": 730
    },
    {
      "epoch": 0.08393829401088929,
      "grad_norm": 28.375,
      "learning_rate": 8.39319470699433e-06,
      "loss": 0.2323,
      "step": 740
    },
    {
      "epoch": 0.08507259528130671,
      "grad_norm": 12.6875,
      "learning_rate": 8.506616257088846e-06,
      "loss": 0.2394,
      "step": 750
    },
    {
      "epoch": 0.08620689655172414,
      "grad_norm": 25.875,
      "learning_rate": 8.620037807183366e-06,
      "loss": 0.2264,
      "step": 760
    },
    {
      "epoch": 0.08734119782214156,
      "grad_norm": 35.75,
      "learning_rate": 8.733459357277884e-06,
      "loss": 0.2234,
      "step": 770
    },
    {
      "epoch": 0.08847549909255899,
      "grad_norm": 24.625,
      "learning_rate": 8.8468809073724e-06,
      "loss": 0.231,
      "step": 780
    },
    {
      "epoch": 0.0896098003629764,
      "grad_norm": 18.75,
      "learning_rate": 8.96030245746692e-06,
      "loss": 0.2253,
      "step": 790
    },
    {
      "epoch": 0.09074410163339383,
      "grad_norm": 24.125,
      "learning_rate": 9.073724007561438e-06,
      "loss": 0.2098,
      "step": 800
    },
    {
      "epoch": 0.09187840290381125,
      "grad_norm": 26.875,
      "learning_rate": 9.187145557655954e-06,
      "loss": 0.2211,
      "step": 810
    },
    {
      "epoch": 0.09301270417422867,
      "grad_norm": 20.0,
      "learning_rate": 9.300567107750473e-06,
      "loss": 0.2178,
      "step": 820
    },
    {
      "epoch": 0.0941470054446461,
      "grad_norm": 16.5,
      "learning_rate": 9.413988657844991e-06,
      "loss": 0.2264,
      "step": 830
    },
    {
      "epoch": 0.09528130671506352,
      "grad_norm": 11.5625,
      "learning_rate": 9.527410207939508e-06,
      "loss": 0.2138,
      "step": 840
    },
    {
      "epoch": 0.09641560798548095,
      "grad_norm": 21.75,
      "learning_rate": 9.640831758034027e-06,
      "loss": 0.2016,
      "step": 850
    },
    {
      "epoch": 0.09754990925589836,
      "grad_norm": 17.0,
      "learning_rate": 9.754253308128545e-06,
      "loss": 0.2108,
      "step": 860
    },
    {
      "epoch": 0.09868421052631579,
      "grad_norm": 11.4375,
      "learning_rate": 9.867674858223062e-06,
      "loss": 0.2005,
      "step": 870
    },
    {
      "epoch": 0.0998185117967332,
      "grad_norm": 20.25,
      "learning_rate": 9.981096408317581e-06,
      "loss": 0.2117,
      "step": 880
    },
    {
      "epoch": 0.10095281306715063,
      "grad_norm": 18.625,
      "learning_rate": 1.0094517958412099e-05,
      "loss": 0.2037,
      "step": 890
    },
    {
      "epoch": 0.10208711433756806,
      "grad_norm": 23.375,
      "learning_rate": 1.0207939508506615e-05,
      "loss": 0.1903,
      "step": 900
    },
    {
      "epoch": 0.10322141560798548,
      "grad_norm": 14.1875,
      "learning_rate": 1.0321361058601135e-05,
      "loss": 0.2131,
      "step": 910
    },
    {
      "epoch": 0.10435571687840291,
      "grad_norm": 14.1875,
      "learning_rate": 1.0434782608695653e-05,
      "loss": 0.2058,
      "step": 920
    },
    {
      "epoch": 0.10549001814882032,
      "grad_norm": 17.25,
      "learning_rate": 1.0548204158790171e-05,
      "loss": 0.1915,
      "step": 930
    },
    {
      "epoch": 0.10662431941923775,
      "grad_norm": 16.625,
      "learning_rate": 1.0661625708884689e-05,
      "loss": 0.1945,
      "step": 940
    },
    {
      "epoch": 0.10775862068965517,
      "grad_norm": 13.5625,
      "learning_rate": 1.0775047258979205e-05,
      "loss": 0.1945,
      "step": 950
    },
    {
      "epoch": 0.1088929219600726,
      "grad_norm": 21.25,
      "learning_rate": 1.0888468809073725e-05,
      "loss": 0.1886,
      "step": 960
    },
    {
      "epoch": 0.11002722323049002,
      "grad_norm": 12.875,
      "learning_rate": 1.1001890359168243e-05,
      "loss": 0.1917,
      "step": 970
    },
    {
      "epoch": 0.11116152450090744,
      "grad_norm": 16.75,
      "learning_rate": 1.1115311909262759e-05,
      "loss": 0.1886,
      "step": 980
    },
    {
      "epoch": 0.11229582577132487,
      "grad_norm": 13.0625,
      "learning_rate": 1.1228733459357279e-05,
      "loss": 0.2012,
      "step": 990
    },
    {
      "epoch": 0.11343012704174228,
      "grad_norm": 11.25,
      "learning_rate": 1.1342155009451797e-05,
      "loss": 0.1909,
      "step": 1000
    },
    {
      "epoch": 0.11456442831215971,
      "grad_norm": 29.875,
      "learning_rate": 1.1455576559546313e-05,
      "loss": 0.192,
      "step": 1010
    },
    {
      "epoch": 0.11569872958257713,
      "grad_norm": 13.9375,
      "learning_rate": 1.1568998109640832e-05,
      "loss": 0.1837,
      "step": 1020
    },
    {
      "epoch": 0.11683303085299455,
      "grad_norm": 13.1875,
      "learning_rate": 1.168241965973535e-05,
      "loss": 0.1769,
      "step": 1030
    },
    {
      "epoch": 0.11796733212341198,
      "grad_norm": 20.25,
      "learning_rate": 1.1795841209829867e-05,
      "loss": 0.188,
      "step": 1040
    },
    {
      "epoch": 0.1191016333938294,
      "grad_norm": 13.3125,
      "learning_rate": 1.1909262759924386e-05,
      "loss": 0.1758,
      "step": 1050
    },
    {
      "epoch": 0.12023593466424683,
      "grad_norm": 26.375,
      "learning_rate": 1.2022684310018904e-05,
      "loss": 0.196,
      "step": 1060
    },
    {
      "epoch": 0.12137023593466424,
      "grad_norm": 13.6875,
      "learning_rate": 1.213610586011342e-05,
      "loss": 0.184,
      "step": 1070
    },
    {
      "epoch": 0.12250453720508167,
      "grad_norm": 11.375,
      "learning_rate": 1.224952741020794e-05,
      "loss": 0.1791,
      "step": 1080
    },
    {
      "epoch": 0.1236388384754991,
      "grad_norm": 18.0,
      "learning_rate": 1.2362948960302458e-05,
      "loss": 0.1682,
      "step": 1090
    },
    {
      "epoch": 0.12477313974591651,
      "grad_norm": 9.375,
      "learning_rate": 1.2476370510396974e-05,
      "loss": 0.1801,
      "step": 1100
    },
    {
      "epoch": 0.12590744101633394,
      "grad_norm": 14.75,
      "learning_rate": 1.2589792060491494e-05,
      "loss": 0.1845,
      "step": 1110
    },
    {
      "epoch": 0.12704174228675136,
      "grad_norm": 10.75,
      "learning_rate": 1.2703213610586012e-05,
      "loss": 0.1737,
      "step": 1120
    },
    {
      "epoch": 0.12817604355716877,
      "grad_norm": 13.875,
      "learning_rate": 1.281663516068053e-05,
      "loss": 0.181,
      "step": 1130
    },
    {
      "epoch": 0.12931034482758622,
      "grad_norm": 13.5625,
      "learning_rate": 1.2930056710775048e-05,
      "loss": 0.1781,
      "step": 1140
    },
    {
      "epoch": 0.13044464609800363,
      "grad_norm": 12.375,
      "learning_rate": 1.3043478260869566e-05,
      "loss": 0.1908,
      "step": 1150
    },
    {
      "epoch": 0.13157894736842105,
      "grad_norm": 11.8125,
      "learning_rate": 1.3156899810964084e-05,
      "loss": 0.1736,
      "step": 1160
    },
    {
      "epoch": 0.1327132486388385,
      "grad_norm": 14.0,
      "learning_rate": 1.3270321361058602e-05,
      "loss": 0.1767,
      "step": 1170
    },
    {
      "epoch": 0.1338475499092559,
      "grad_norm": 10.25,
      "learning_rate": 1.338374291115312e-05,
      "loss": 0.1693,
      "step": 1180
    },
    {
      "epoch": 0.13498185117967332,
      "grad_norm": 10.125,
      "learning_rate": 1.3497164461247638e-05,
      "loss": 0.166,
      "step": 1190
    },
    {
      "epoch": 0.13611615245009073,
      "grad_norm": 10.625,
      "learning_rate": 1.3610586011342156e-05,
      "loss": 0.1616,
      "step": 1200
    },
    {
      "epoch": 0.13725045372050818,
      "grad_norm": 7.40625,
      "learning_rate": 1.3724007561436673e-05,
      "loss": 0.1792,
      "step": 1210
    },
    {
      "epoch": 0.1383847549909256,
      "grad_norm": 13.375,
      "learning_rate": 1.3837429111531191e-05,
      "loss": 0.1737,
      "step": 1220
    },
    {
      "epoch": 0.139519056261343,
      "grad_norm": 17.125,
      "learning_rate": 1.395085066162571e-05,
      "loss": 0.174,
      "step": 1230
    },
    {
      "epoch": 0.14065335753176045,
      "grad_norm": 22.75,
      "learning_rate": 1.4064272211720227e-05,
      "loss": 0.177,
      "step": 1240
    },
    {
      "epoch": 0.14178765880217786,
      "grad_norm": 15.8125,
      "learning_rate": 1.4177693761814745e-05,
      "loss": 0.1709,
      "step": 1250
    },
    {
      "epoch": 0.14292196007259528,
      "grad_norm": 21.625,
      "learning_rate": 1.4291115311909263e-05,
      "loss": 0.172,
      "step": 1260
    },
    {
      "epoch": 0.1440562613430127,
      "grad_norm": 13.25,
      "learning_rate": 1.4404536862003781e-05,
      "loss": 0.1795,
      "step": 1270
    },
    {
      "epoch": 0.14519056261343014,
      "grad_norm": 13.25,
      "learning_rate": 1.4517958412098299e-05,
      "loss": 0.1685,
      "step": 1280
    },
    {
      "epoch": 0.14632486388384755,
      "grad_norm": 24.625,
      "learning_rate": 1.4631379962192817e-05,
      "loss": 0.1795,
      "step": 1290
    },
    {
      "epoch": 0.14745916515426497,
      "grad_norm": 7.1875,
      "learning_rate": 1.4744801512287335e-05,
      "loss": 0.1653,
      "step": 1300
    },
    {
      "epoch": 0.1485934664246824,
      "grad_norm": 12.125,
      "learning_rate": 1.4858223062381853e-05,
      "loss": 0.1734,
      "step": 1310
    },
    {
      "epoch": 0.14972776769509982,
      "grad_norm": 16.125,
      "learning_rate": 1.4971644612476371e-05,
      "loss": 0.1628,
      "step": 1320
    },
    {
      "epoch": 0.15086206896551724,
      "grad_norm": 13.4375,
      "learning_rate": 1.5085066162570889e-05,
      "loss": 0.1617,
      "step": 1330
    },
    {
      "epoch": 0.15199637023593465,
      "grad_norm": 10.6875,
      "learning_rate": 1.5198487712665405e-05,
      "loss": 0.171,
      "step": 1340
    },
    {
      "epoch": 0.1531306715063521,
      "grad_norm": 21.0,
      "learning_rate": 1.5311909262759923e-05,
      "loss": 0.165,
      "step": 1350
    },
    {
      "epoch": 0.1542649727767695,
      "grad_norm": 14.8125,
      "learning_rate": 1.5425330812854444e-05,
      "loss": 0.1735,
      "step": 1360
    },
    {
      "epoch": 0.15539927404718692,
      "grad_norm": 6.90625,
      "learning_rate": 1.553875236294896e-05,
      "loss": 0.164,
      "step": 1370
    },
    {
      "epoch": 0.15653357531760437,
      "grad_norm": 11.625,
      "learning_rate": 1.5652173913043477e-05,
      "loss": 0.1642,
      "step": 1380
    },
    {
      "epoch": 0.15766787658802178,
      "grad_norm": 13.6875,
      "learning_rate": 1.5765595463137998e-05,
      "loss": 0.1714,
      "step": 1390
    },
    {
      "epoch": 0.1588021778584392,
      "grad_norm": 13.5,
      "learning_rate": 1.5879017013232513e-05,
      "loss": 0.168,
      "step": 1400
    },
    {
      "epoch": 0.1599364791288566,
      "grad_norm": 14.0625,
      "learning_rate": 1.599243856332703e-05,
      "loss": 0.1739,
      "step": 1410
    },
    {
      "epoch": 0.16107078039927405,
      "grad_norm": 8.75,
      "learning_rate": 1.6105860113421552e-05,
      "loss": 0.1753,
      "step": 1420
    },
    {
      "epoch": 0.16220508166969147,
      "grad_norm": 11.8125,
      "learning_rate": 1.621928166351607e-05,
      "loss": 0.1808,
      "step": 1430
    },
    {
      "epoch": 0.16333938294010888,
      "grad_norm": 5.71875,
      "learning_rate": 1.6332703213610585e-05,
      "loss": 0.1583,
      "step": 1440
    },
    {
      "epoch": 0.16447368421052633,
      "grad_norm": 12.0,
      "learning_rate": 1.6446124763705106e-05,
      "loss": 0.1649,
      "step": 1450
    },
    {
      "epoch": 0.16560798548094374,
      "grad_norm": 7.46875,
      "learning_rate": 1.6559546313799624e-05,
      "loss": 0.1713,
      "step": 1460
    },
    {
      "epoch": 0.16674228675136116,
      "grad_norm": 13.1875,
      "learning_rate": 1.667296786389414e-05,
      "loss": 0.1685,
      "step": 1470
    },
    {
      "epoch": 0.16787658802177857,
      "grad_norm": 9.75,
      "learning_rate": 1.678638941398866e-05,
      "loss": 0.1628,
      "step": 1480
    },
    {
      "epoch": 0.16901088929219601,
      "grad_norm": 8.3125,
      "learning_rate": 1.6899810964083178e-05,
      "loss": 0.1636,
      "step": 1490
    },
    {
      "epoch": 0.17014519056261343,
      "grad_norm": 4.40625,
      "learning_rate": 1.7013232514177692e-05,
      "loss": 0.1639,
      "step": 1500
    },
    {
      "epoch": 0.17127949183303084,
      "grad_norm": 9.5,
      "learning_rate": 1.7126654064272214e-05,
      "loss": 0.1772,
      "step": 1510
    },
    {
      "epoch": 0.1724137931034483,
      "grad_norm": 17.5,
      "learning_rate": 1.724007561436673e-05,
      "loss": 0.1622,
      "step": 1520
    },
    {
      "epoch": 0.1735480943738657,
      "grad_norm": 16.625,
      "learning_rate": 1.7353497164461246e-05,
      "loss": 0.163,
      "step": 1530
    },
    {
      "epoch": 0.17468239564428312,
      "grad_norm": 8.125,
      "learning_rate": 1.7466918714555767e-05,
      "loss": 0.1564,
      "step": 1540
    },
    {
      "epoch": 0.17581669691470053,
      "grad_norm": 13.1875,
      "learning_rate": 1.7580340264650285e-05,
      "loss": 0.1662,
      "step": 1550
    },
    {
      "epoch": 0.17695099818511797,
      "grad_norm": 9.125,
      "learning_rate": 1.76937618147448e-05,
      "loss": 0.1593,
      "step": 1560
    },
    {
      "epoch": 0.1780852994555354,
      "grad_norm": 6.5,
      "learning_rate": 1.780718336483932e-05,
      "loss": 0.1633,
      "step": 1570
    },
    {
      "epoch": 0.1792196007259528,
      "grad_norm": 7.0625,
      "learning_rate": 1.792060491493384e-05,
      "loss": 0.1578,
      "step": 1580
    },
    {
      "epoch": 0.18035390199637025,
      "grad_norm": 12.5625,
      "learning_rate": 1.8034026465028354e-05,
      "loss": 0.1624,
      "step": 1590
    },
    {
      "epoch": 0.18148820326678766,
      "grad_norm": 11.375,
      "learning_rate": 1.8147448015122875e-05,
      "loss": 0.1614,
      "step": 1600
    },
    {
      "epoch": 0.18262250453720508,
      "grad_norm": 11.75,
      "learning_rate": 1.8260869565217393e-05,
      "loss": 0.1594,
      "step": 1610
    },
    {
      "epoch": 0.1837568058076225,
      "grad_norm": 5.96875,
      "learning_rate": 1.8374291115311908e-05,
      "loss": 0.1574,
      "step": 1620
    },
    {
      "epoch": 0.18489110707803993,
      "grad_norm": 5.59375,
      "learning_rate": 1.848771266540643e-05,
      "loss": 0.1721,
      "step": 1630
    },
    {
      "epoch": 0.18602540834845735,
      "grad_norm": 8.75,
      "learning_rate": 1.8601134215500947e-05,
      "loss": 0.1525,
      "step": 1640
    },
    {
      "epoch": 0.18715970961887476,
      "grad_norm": 9.75,
      "learning_rate": 1.871455576559546e-05,
      "loss": 0.1621,
      "step": 1650
    },
    {
      "epoch": 0.1882940108892922,
      "grad_norm": 8.375,
      "learning_rate": 1.8827977315689983e-05,
      "loss": 0.1791,
      "step": 1660
    },
    {
      "epoch": 0.18942831215970962,
      "grad_norm": 7.25,
      "learning_rate": 1.89413988657845e-05,
      "loss": 0.1643,
      "step": 1670
    },
    {
      "epoch": 0.19056261343012704,
      "grad_norm": 7.78125,
      "learning_rate": 1.9054820415879015e-05,
      "loss": 0.1617,
      "step": 1680
    },
    {
      "epoch": 0.19169691470054445,
      "grad_norm": 8.9375,
      "learning_rate": 1.9168241965973537e-05,
      "loss": 0.1598,
      "step": 1690
    },
    {
      "epoch": 0.1928312159709619,
      "grad_norm": 7.0625,
      "learning_rate": 1.9281663516068055e-05,
      "loss": 0.1591,
      "step": 1700
    },
    {
      "epoch": 0.1939655172413793,
      "grad_norm": 12.25,
      "learning_rate": 1.939508506616257e-05,
      "loss": 0.1566,
      "step": 1710
    },
    {
      "epoch": 0.19509981851179672,
      "grad_norm": 9.4375,
      "learning_rate": 1.950850661625709e-05,
      "loss": 0.166,
      "step": 1720
    },
    {
      "epoch": 0.19623411978221417,
      "grad_norm": 8.75,
      "learning_rate": 1.962192816635161e-05,
      "loss": 0.1631,
      "step": 1730
    },
    {
      "epoch": 0.19736842105263158,
      "grad_norm": 10.9375,
      "learning_rate": 1.9735349716446123e-05,
      "loss": 0.1651,
      "step": 1740
    },
    {
      "epoch": 0.198502722323049,
      "grad_norm": 8.5,
      "learning_rate": 1.9848771266540644e-05,
      "loss": 0.1647,
      "step": 1750
    },
    {
      "epoch": 0.1996370235934664,
      "grad_norm": 14.375,
      "learning_rate": 1.9962192816635162e-05,
      "loss": 0.1584,
      "step": 1760
    },
    {
      "epoch": 0.20077132486388385,
      "grad_norm": 24.125,
      "learning_rate": 2.0075614366729677e-05,
      "loss": 0.1764,
      "step": 1770
    },
    {
      "epoch": 0.20190562613430127,
      "grad_norm": 15.25,
      "learning_rate": 2.0189035916824198e-05,
      "loss": 0.1751,
      "step": 1780
    },
    {
      "epoch": 0.20303992740471868,
      "grad_norm": 19.125,
      "learning_rate": 2.0302457466918716e-05,
      "loss": 0.179,
      "step": 1790
    },
    {
      "epoch": 0.20417422867513613,
      "grad_norm": 5.5625,
      "learning_rate": 2.041587901701323e-05,
      "loss": 0.1565,
      "step": 1800
    },
    {
      "epoch": 0.20530852994555354,
      "grad_norm": 16.25,
      "learning_rate": 2.0529300567107752e-05,
      "loss": 0.1659,
      "step": 1810
    },
    {
      "epoch": 0.20644283121597096,
      "grad_norm": 14.5,
      "learning_rate": 2.064272211720227e-05,
      "loss": 0.1686,
      "step": 1820
    },
    {
      "epoch": 0.20757713248638837,
      "grad_norm": 8.1875,
      "learning_rate": 2.0756143667296785e-05,
      "loss": 0.164,
      "step": 1830
    },
    {
      "epoch": 0.20871143375680581,
      "grad_norm": 17.25,
      "learning_rate": 2.0869565217391306e-05,
      "loss": 0.1626,
      "step": 1840
    },
    {
      "epoch": 0.20984573502722323,
      "grad_norm": 11.3125,
      "learning_rate": 2.0982986767485824e-05,
      "loss": 0.1633,
      "step": 1850
    },
    {
      "epoch": 0.21098003629764064,
      "grad_norm": 7.25,
      "learning_rate": 2.1096408317580342e-05,
      "loss": 0.1711,
      "step": 1860
    },
    {
      "epoch": 0.2121143375680581,
      "grad_norm": 4.9375,
      "learning_rate": 2.120982986767486e-05,
      "loss": 0.1637,
      "step": 1870
    },
    {
      "epoch": 0.2132486388384755,
      "grad_norm": 10.8125,
      "learning_rate": 2.1323251417769378e-05,
      "loss": 0.1726,
      "step": 1880
    },
    {
      "epoch": 0.21438294010889292,
      "grad_norm": 8.125,
      "learning_rate": 2.1436672967863896e-05,
      "loss": 0.1721,
      "step": 1890
    },
    {
      "epoch": 0.21551724137931033,
      "grad_norm": 8.875,
      "learning_rate": 2.155009451795841e-05,
      "loss": 0.1679,
      "step": 1900
    },
    {
      "epoch": 0.21665154264972777,
      "grad_norm": 5.03125,
      "learning_rate": 2.166351606805293e-05,
      "loss": 0.1545,
      "step": 1910
    },
    {
      "epoch": 0.2177858439201452,
      "grad_norm": 5.3125,
      "learning_rate": 2.177693761814745e-05,
      "loss": 0.1563,
      "step": 1920
    },
    {
      "epoch": 0.2189201451905626,
      "grad_norm": 9.5,
      "learning_rate": 2.1890359168241964e-05,
      "loss": 0.1691,
      "step": 1930
    },
    {
      "epoch": 0.22005444646098005,
      "grad_norm": 6.375,
      "learning_rate": 2.2003780718336485e-05,
      "loss": 0.1596,
      "step": 1940
    },
    {
      "epoch": 0.22118874773139746,
      "grad_norm": 9.875,
      "learning_rate": 2.2117202268431003e-05,
      "loss": 0.1753,
      "step": 1950
    },
    {
      "epoch": 0.22232304900181488,
      "grad_norm": 9.6875,
      "learning_rate": 2.2230623818525518e-05,
      "loss": 0.1733,
      "step": 1960
    },
    {
      "epoch": 0.2234573502722323,
      "grad_norm": 6.8125,
      "learning_rate": 2.234404536862004e-05,
      "loss": 0.1716,
      "step": 1970
    },
    {
      "epoch": 0.22459165154264973,
      "grad_norm": 7.125,
      "learning_rate": 2.2457466918714557e-05,
      "loss": 0.1626,
      "step": 1980
    },
    {
      "epoch": 0.22572595281306715,
      "grad_norm": 6.84375,
      "learning_rate": 2.2570888468809072e-05,
      "loss": 0.1563,
      "step": 1990
    },
    {
      "epoch": 0.22686025408348456,
      "grad_norm": 6.34375,
      "learning_rate": 2.2684310018903593e-05,
      "loss": 0.1603,
      "step": 2000
    },
    {
      "epoch": 0.227994555353902,
      "grad_norm": 9.625,
      "learning_rate": 2.279773156899811e-05,
      "loss": 0.1667,
      "step": 2010
    },
    {
      "epoch": 0.22912885662431942,
      "grad_norm": 17.625,
      "learning_rate": 2.2911153119092626e-05,
      "loss": 0.1583,
      "step": 2020
    },
    {
      "epoch": 0.23026315789473684,
      "grad_norm": 15.0625,
      "learning_rate": 2.3024574669187147e-05,
      "loss": 0.1853,
      "step": 2030
    },
    {
      "epoch": 0.23139745916515425,
      "grad_norm": 6.96875,
      "learning_rate": 2.3137996219281665e-05,
      "loss": 0.1661,
      "step": 2040
    },
    {
      "epoch": 0.2325317604355717,
      "grad_norm": 6.125,
      "learning_rate": 2.325141776937618e-05,
      "loss": 0.1555,
      "step": 2050
    },
    {
      "epoch": 0.2336660617059891,
      "grad_norm": 7.84375,
      "learning_rate": 2.33648393194707e-05,
      "loss": 0.1669,
      "step": 2060
    },
    {
      "epoch": 0.23480036297640652,
      "grad_norm": 7.65625,
      "learning_rate": 2.347826086956522e-05,
      "loss": 0.153,
      "step": 2070
    },
    {
      "epoch": 0.23593466424682397,
      "grad_norm": 5.4375,
      "learning_rate": 2.3591682419659733e-05,
      "loss": 0.1509,
      "step": 2080
    },
    {
      "epoch": 0.23706896551724138,
      "grad_norm": 3.59375,
      "learning_rate": 2.3705103969754255e-05,
      "loss": 0.1562,
      "step": 2090
    },
    {
      "epoch": 0.2382032667876588,
      "grad_norm": 5.625,
      "learning_rate": 2.3818525519848773e-05,
      "loss": 0.1583,
      "step": 2100
    },
    {
      "epoch": 0.23933756805807624,
      "grad_norm": 4.625,
      "learning_rate": 2.3931947069943287e-05,
      "loss": 0.1569,
      "step": 2110
    },
    {
      "epoch": 0.24047186932849365,
      "grad_norm": 7.4375,
      "learning_rate": 2.404536862003781e-05,
      "loss": 0.1492,
      "step": 2120
    },
    {
      "epoch": 0.24160617059891107,
      "grad_norm": 5.0,
      "learning_rate": 2.4158790170132326e-05,
      "loss": 0.15,
      "step": 2130
    },
    {
      "epoch": 0.24274047186932848,
      "grad_norm": 4.75,
      "learning_rate": 2.427221172022684e-05,
      "loss": 0.1653,
      "step": 2140
    },
    {
      "epoch": 0.24387477313974593,
      "grad_norm": 4.90625,
      "learning_rate": 2.4385633270321362e-05,
      "loss": 0.1582,
      "step": 2150
    },
    {
      "epoch": 0.24500907441016334,
      "grad_norm": 3.59375,
      "learning_rate": 2.449905482041588e-05,
      "loss": 0.1594,
      "step": 2160
    },
    {
      "epoch": 0.24614337568058076,
      "grad_norm": 6.9375,
      "learning_rate": 2.4612476370510395e-05,
      "loss": 0.1605,
      "step": 2170
    },
    {
      "epoch": 0.2472776769509982,
      "grad_norm": 5.1875,
      "learning_rate": 2.4725897920604916e-05,
      "loss": 0.1667,
      "step": 2180
    },
    {
      "epoch": 0.2484119782214156,
      "grad_norm": 7.96875,
      "learning_rate": 2.4839319470699434e-05,
      "loss": 0.165,
      "step": 2190
    },
    {
      "epoch": 0.24954627949183303,
      "grad_norm": 6.375,
      "learning_rate": 2.495274102079395e-05,
      "loss": 0.1592,
      "step": 2200
    },
    {
      "epoch": 0.25068058076225047,
      "grad_norm": 9.5,
      "learning_rate": 2.506616257088847e-05,
      "loss": 0.156,
      "step": 2210
    },
    {
      "epoch": 0.2518148820326679,
      "grad_norm": 7.53125,
      "learning_rate": 2.5179584120982988e-05,
      "loss": 0.1588,
      "step": 2220
    },
    {
      "epoch": 0.2529491833030853,
      "grad_norm": 7.125,
      "learning_rate": 2.5293005671077503e-05,
      "loss": 0.173,
      "step": 2230
    },
    {
      "epoch": 0.2540834845735027,
      "grad_norm": 5.5,
      "learning_rate": 2.5406427221172024e-05,
      "loss": 0.1667,
      "step": 2240
    },
    {
      "epoch": 0.25521778584392013,
      "grad_norm": 12.125,
      "learning_rate": 2.5519848771266542e-05,
      "loss": 0.1624,
      "step": 2250
    },
    {
      "epoch": 0.25635208711433755,
      "grad_norm": 7.46875,
      "learning_rate": 2.563327032136106e-05,
      "loss": 0.152,
      "step": 2260
    },
    {
      "epoch": 0.257486388384755,
      "grad_norm": 6.9375,
      "learning_rate": 2.5746691871455578e-05,
      "loss": 0.1643,
      "step": 2270
    },
    {
      "epoch": 0.25862068965517243,
      "grad_norm": 7.0,
      "learning_rate": 2.5860113421550096e-05,
      "loss": 0.1637,
      "step": 2280
    },
    {
      "epoch": 0.25975499092558985,
      "grad_norm": 2.78125,
      "learning_rate": 2.5973534971644614e-05,
      "loss": 0.1664,
      "step": 2290
    },
    {
      "epoch": 0.26088929219600726,
      "grad_norm": 7.5,
      "learning_rate": 2.608695652173913e-05,
      "loss": 0.1615,
      "step": 2300
    },
    {
      "epoch": 0.2620235934664247,
      "grad_norm": 4.78125,
      "learning_rate": 2.620037807183365e-05,
      "loss": 0.1518,
      "step": 2310
    },
    {
      "epoch": 0.2631578947368421,
      "grad_norm": 6.59375,
      "learning_rate": 2.6313799621928167e-05,
      "loss": 0.1572,
      "step": 2320
    },
    {
      "epoch": 0.2642921960072595,
      "grad_norm": 3.953125,
      "learning_rate": 2.6427221172022685e-05,
      "loss": 0.1546,
      "step": 2330
    },
    {
      "epoch": 0.265426497277677,
      "grad_norm": 4.96875,
      "learning_rate": 2.6540642722117203e-05,
      "loss": 0.1575,
      "step": 2340
    },
    {
      "epoch": 0.2665607985480944,
      "grad_norm": 5.59375,
      "learning_rate": 2.665406427221172e-05,
      "loss": 0.1622,
      "step": 2350
    },
    {
      "epoch": 0.2676950998185118,
      "grad_norm": 7.53125,
      "learning_rate": 2.676748582230624e-05,
      "loss": 0.1588,
      "step": 2360
    },
    {
      "epoch": 0.2688294010889292,
      "grad_norm": 3.53125,
      "learning_rate": 2.6880907372400757e-05,
      "loss": 0.1574,
      "step": 2370
    },
    {
      "epoch": 0.26996370235934664,
      "grad_norm": 6.5625,
      "learning_rate": 2.6994328922495275e-05,
      "loss": 0.1571,
      "step": 2380
    },
    {
      "epoch": 0.27109800362976405,
      "grad_norm": 7.09375,
      "learning_rate": 2.7107750472589793e-05,
      "loss": 0.1558,
      "step": 2390
    },
    {
      "epoch": 0.27223230490018147,
      "grad_norm": 3.859375,
      "learning_rate": 2.722117202268431e-05,
      "loss": 0.1715,
      "step": 2400
    },
    {
      "epoch": 0.27336660617059894,
      "grad_norm": 7.25,
      "learning_rate": 2.733459357277883e-05,
      "loss": 0.165,
      "step": 2410
    },
    {
      "epoch": 0.27450090744101635,
      "grad_norm": 7.75,
      "learning_rate": 2.7448015122873347e-05,
      "loss": 0.1572,
      "step": 2420
    },
    {
      "epoch": 0.27563520871143377,
      "grad_norm": 5.28125,
      "learning_rate": 2.7561436672967865e-05,
      "loss": 0.1647,
      "step": 2430
    },
    {
      "epoch": 0.2767695099818512,
      "grad_norm": 7.25,
      "learning_rate": 2.7674858223062383e-05,
      "loss": 0.1687,
      "step": 2440
    },
    {
      "epoch": 0.2779038112522686,
      "grad_norm": 2.84375,
      "learning_rate": 2.77882797731569e-05,
      "loss": 0.1533,
      "step": 2450
    },
    {
      "epoch": 0.279038112522686,
      "grad_norm": 3.46875,
      "learning_rate": 2.790170132325142e-05,
      "loss": 0.16,
      "step": 2460
    },
    {
      "epoch": 0.2801724137931034,
      "grad_norm": 5.1875,
      "learning_rate": 2.8015122873345937e-05,
      "loss": 0.1684,
      "step": 2470
    },
    {
      "epoch": 0.2813067150635209,
      "grad_norm": 10.0625,
      "learning_rate": 2.8128544423440455e-05,
      "loss": 0.1679,
      "step": 2480
    },
    {
      "epoch": 0.2824410163339383,
      "grad_norm": 7.9375,
      "learning_rate": 2.8241965973534973e-05,
      "loss": 0.1674,
      "step": 2490
    },
    {
      "epoch": 0.2835753176043557,
      "grad_norm": 5.03125,
      "learning_rate": 2.835538752362949e-05,
      "loss": 0.1676,
      "step": 2500
    },
    {
      "epoch": 0.28470961887477314,
      "grad_norm": 4.75,
      "learning_rate": 2.846880907372401e-05,
      "loss": 0.1547,
      "step": 2510
    },
    {
      "epoch": 0.28584392014519056,
      "grad_norm": 4.03125,
      "learning_rate": 2.8582230623818526e-05,
      "loss": 0.1579,
      "step": 2520
    },
    {
      "epoch": 0.28697822141560797,
      "grad_norm": 7.53125,
      "learning_rate": 2.8695652173913044e-05,
      "loss": 0.1559,
      "step": 2530
    },
    {
      "epoch": 0.2881125226860254,
      "grad_norm": 6.90625,
      "learning_rate": 2.8809073724007562e-05,
      "loss": 0.1673,
      "step": 2540
    },
    {
      "epoch": 0.28924682395644286,
      "grad_norm": 6.53125,
      "learning_rate": 2.892249527410208e-05,
      "loss": 0.1645,
      "step": 2550
    },
    {
      "epoch": 0.29038112522686027,
      "grad_norm": 6.46875,
      "learning_rate": 2.9035916824196598e-05,
      "loss": 0.1581,
      "step": 2560
    },
    {
      "epoch": 0.2915154264972777,
      "grad_norm": 4.46875,
      "learning_rate": 2.9149338374291116e-05,
      "loss": 0.1565,
      "step": 2570
    },
    {
      "epoch": 0.2926497277676951,
      "grad_norm": 5.125,
      "learning_rate": 2.9262759924385634e-05,
      "loss": 0.1707,
      "step": 2580
    },
    {
      "epoch": 0.2937840290381125,
      "grad_norm": 6.34375,
      "learning_rate": 2.9376181474480152e-05,
      "loss": 0.1638,
      "step": 2590
    },
    {
      "epoch": 0.29491833030852993,
      "grad_norm": 3.25,
      "learning_rate": 2.948960302457467e-05,
      "loss": 0.1658,
      "step": 2600
    },
    {
      "epoch": 0.29605263157894735,
      "grad_norm": 4.21875,
      "learning_rate": 2.9603024574669188e-05,
      "loss": 0.1556,
      "step": 2610
    },
    {
      "epoch": 0.2971869328493648,
      "grad_norm": 4.875,
      "learning_rate": 2.9716446124763706e-05,
      "loss": 0.1717,
      "step": 2620
    },
    {
      "epoch": 0.29832123411978223,
      "grad_norm": 5.09375,
      "learning_rate": 2.9829867674858224e-05,
      "loss": 0.1761,
      "step": 2630
    },
    {
      "epoch": 0.29945553539019965,
      "grad_norm": 4.25,
      "learning_rate": 2.9943289224952742e-05,
      "loss": 0.1669,
      "step": 2640
    },
    {
      "epoch": 0.30058983666061706,
      "grad_norm": 6.78125,
      "learning_rate": 2.9999996733838852e-05,
      "loss": 0.1562,
      "step": 2650
    },
    {
      "epoch": 0.3017241379310345,
      "grad_norm": 7.15625,
      "learning_rate": 2.99999706045582e-05,
      "loss": 0.156,
      "step": 2660
    },
    {
      "epoch": 0.3028584392014519,
      "grad_norm": 3.46875,
      "learning_rate": 2.9999918346042412e-05,
      "loss": 0.1568,
      "step": 2670
    },
    {
      "epoch": 0.3039927404718693,
      "grad_norm": 6.46875,
      "learning_rate": 2.999983995838252e-05,
      "loss": 0.152,
      "step": 2680
    },
    {
      "epoch": 0.3051270417422868,
      "grad_norm": 7.875,
      "learning_rate": 2.9999735441715067e-05,
      "loss": 0.159,
      "step": 2690
    },
    {
      "epoch": 0.3062613430127042,
      "grad_norm": 3.171875,
      "learning_rate": 2.999960479622213e-05,
      "loss": 0.1666,
      "step": 2700
    },
    {
      "epoch": 0.3073956442831216,
      "grad_norm": 3.34375,
      "learning_rate": 2.9999448022131267e-05,
      "loss": 0.166,
      "step": 2710
    },
    {
      "epoch": 0.308529945553539,
      "grad_norm": 4.65625,
      "learning_rate": 2.9999265119715588e-05,
      "loss": 0.1708,
      "step": 2720
    },
    {
      "epoch": 0.30966424682395643,
      "grad_norm": 5.125,
      "learning_rate": 2.999905608929369e-05,
      "loss": 0.1618,
      "step": 2730
    },
    {
      "epoch": 0.31079854809437385,
      "grad_norm": 3.875,
      "learning_rate": 2.9998820931229702e-05,
      "loss": 0.1446,
      "step": 2740
    },
    {
      "epoch": 0.31193284936479126,
      "grad_norm": 5.5625,
      "learning_rate": 2.9998559645933254e-05,
      "loss": 0.1689,
      "step": 2750
    },
    {
      "epoch": 0.31306715063520874,
      "grad_norm": 3.90625,
      "learning_rate": 2.999827223385949e-05,
      "loss": 0.1722,
      "step": 2760
    },
    {
      "epoch": 0.31420145190562615,
      "grad_norm": 2.90625,
      "learning_rate": 2.9997958695509075e-05,
      "loss": 0.1568,
      "step": 2770
    },
    {
      "epoch": 0.31533575317604357,
      "grad_norm": 5.8125,
      "learning_rate": 2.9997619031428166e-05,
      "loss": 0.1615,
      "step": 2780
    },
    {
      "epoch": 0.316470054446461,
      "grad_norm": 3.40625,
      "learning_rate": 2.9997253242208454e-05,
      "loss": 0.1606,
      "step": 2790
    },
    {
      "epoch": 0.3176043557168784,
      "grad_norm": 6.40625,
      "learning_rate": 2.9996861328487123e-05,
      "loss": 0.1632,
      "step": 2800
    },
    {
      "epoch": 0.3187386569872958,
      "grad_norm": 3.671875,
      "learning_rate": 2.9996443290946863e-05,
      "loss": 0.1694,
      "step": 2810
    },
    {
      "epoch": 0.3198729582577132,
      "grad_norm": 6.65625,
      "learning_rate": 2.9995999130315883e-05,
      "loss": 0.1624,
      "step": 2820
    },
    {
      "epoch": 0.3210072595281307,
      "grad_norm": 3.65625,
      "learning_rate": 2.9995528847367885e-05,
      "loss": 0.1694,
      "step": 2830
    },
    {
      "epoch": 0.3221415607985481,
      "grad_norm": 3.5625,
      "learning_rate": 2.9995032442922084e-05,
      "loss": 0.1625,
      "step": 2840
    },
    {
      "epoch": 0.3232758620689655,
      "grad_norm": 3.0625,
      "learning_rate": 2.9994509917843184e-05,
      "loss": 0.1598,
      "step": 2850
    },
    {
      "epoch": 0.32441016333938294,
      "grad_norm": 6.125,
      "learning_rate": 2.9993961273041404e-05,
      "loss": 0.1481,
      "step": 2860
    },
    {
      "epoch": 0.32554446460980035,
      "grad_norm": 4.1875,
      "learning_rate": 2.9993386509472464e-05,
      "loss": 0.1578,
      "step": 2870
    },
    {
      "epoch": 0.32667876588021777,
      "grad_norm": 3.5,
      "learning_rate": 2.999278562813756e-05,
      "loss": 0.158,
      "step": 2880
    },
    {
      "epoch": 0.3278130671506352,
      "grad_norm": 3.03125,
      "learning_rate": 2.999215863008341e-05,
      "loss": 0.1612,
      "step": 2890
    },
    {
      "epoch": 0.32894736842105265,
      "grad_norm": 3.609375,
      "learning_rate": 2.9991505516402214e-05,
      "loss": 0.1589,
      "step": 2900
    },
    {
      "epoch": 0.33008166969147007,
      "grad_norm": 4.46875,
      "learning_rate": 2.9990826288231663e-05,
      "loss": 0.164,
      "step": 2910
    },
    {
      "epoch": 0.3312159709618875,
      "grad_norm": 5.3125,
      "learning_rate": 2.9990120946754937e-05,
      "loss": 0.1638,
      "step": 2920
    },
    {
      "epoch": 0.3323502722323049,
      "grad_norm": 6.125,
      "learning_rate": 2.998938949320071e-05,
      "loss": 0.1558,
      "step": 2930
    },
    {
      "epoch": 0.3334845735027223,
      "grad_norm": 4.15625,
      "learning_rate": 2.998863192884314e-05,
      "loss": 0.1564,
      "step": 2940
    },
    {
      "epoch": 0.33461887477313973,
      "grad_norm": 2.796875,
      "learning_rate": 2.998784825500187e-05,
      "loss": 0.1573,
      "step": 2950
    },
    {
      "epoch": 0.33575317604355714,
      "grad_norm": 5.375,
      "learning_rate": 2.998703847304202e-05,
      "loss": 0.1619,
      "step": 2960
    },
    {
      "epoch": 0.3368874773139746,
      "grad_norm": 3.359375,
      "learning_rate": 2.998620258437419e-05,
      "loss": 0.1536,
      "step": 2970
    },
    {
      "epoch": 0.33802177858439203,
      "grad_norm": 4.03125,
      "learning_rate": 2.998534059045446e-05,
      "loss": 0.1555,
      "step": 2980
    },
    {
      "epoch": 0.33915607985480944,
      "grad_norm": 2.515625,
      "learning_rate": 2.998445249278438e-05,
      "loss": 0.1657,
      "step": 2990
    },
    {
      "epoch": 0.34029038112522686,
      "grad_norm": 3.515625,
      "learning_rate": 2.9983538292910982e-05,
      "loss": 0.156,
      "step": 3000
    },
    {
      "epoch": 0.3414246823956443,
      "grad_norm": 5.0625,
      "learning_rate": 2.9982597992426747e-05,
      "loss": 0.1722,
      "step": 3010
    },
    {
      "epoch": 0.3425589836660617,
      "grad_norm": 2.234375,
      "learning_rate": 2.998163159296964e-05,
      "loss": 0.1544,
      "step": 3020
    },
    {
      "epoch": 0.3436932849364791,
      "grad_norm": 3.609375,
      "learning_rate": 2.9980639096223087e-05,
      "loss": 0.1416,
      "step": 3030
    },
    {
      "epoch": 0.3448275862068966,
      "grad_norm": 4.03125,
      "learning_rate": 2.9979620503915965e-05,
      "loss": 0.1604,
      "step": 3040
    },
    {
      "epoch": 0.345961887477314,
      "grad_norm": 3.53125,
      "learning_rate": 2.997857581782261e-05,
      "loss": 0.1603,
      "step": 3050
    },
    {
      "epoch": 0.3470961887477314,
      "grad_norm": 4.625,
      "learning_rate": 2.997750503976282e-05,
      "loss": 0.1574,
      "step": 3060
    },
    {
      "epoch": 0.3482304900181488,
      "grad_norm": 4.1875,
      "learning_rate": 2.997640817160184e-05,
      "loss": 0.1489,
      "step": 3070
    },
    {
      "epoch": 0.34936479128856623,
      "grad_norm": 2.328125,
      "learning_rate": 2.9975285215250364e-05,
      "loss": 0.1513,
      "step": 3080
    },
    {
      "epoch": 0.35049909255898365,
      "grad_norm": 2.5625,
      "learning_rate": 2.9974136172664526e-05,
      "loss": 0.168,
      "step": 3090
    },
    {
      "epoch": 0.35163339382940106,
      "grad_norm": 5.46875,
      "learning_rate": 2.9972961045845898e-05,
      "loss": 0.1561,
      "step": 3100
    },
    {
      "epoch": 0.35276769509981853,
      "grad_norm": 3.09375,
      "learning_rate": 2.9971759836841502e-05,
      "loss": 0.1654,
      "step": 3110
    },
    {
      "epoch": 0.35390199637023595,
      "grad_norm": 4.0625,
      "learning_rate": 2.9970532547743787e-05,
      "loss": 0.1514,
      "step": 3120
    },
    {
      "epoch": 0.35503629764065336,
      "grad_norm": 3.03125,
      "learning_rate": 2.996927918069063e-05,
      "loss": 0.1551,
      "step": 3130
    },
    {
      "epoch": 0.3561705989110708,
      "grad_norm": 2.71875,
      "learning_rate": 2.9967999737865338e-05,
      "loss": 0.1599,
      "step": 3140
    },
    {
      "epoch": 0.3573049001814882,
      "grad_norm": 2.90625,
      "learning_rate": 2.9966694221496635e-05,
      "loss": 0.1585,
      "step": 3150
    },
    {
      "epoch": 0.3584392014519056,
      "grad_norm": 2.8125,
      "learning_rate": 2.9965362633858673e-05,
      "loss": 0.1658,
      "step": 3160
    },
    {
      "epoch": 0.359573502722323,
      "grad_norm": 2.40625,
      "learning_rate": 2.9964004977271013e-05,
      "loss": 0.1498,
      "step": 3170
    },
    {
      "epoch": 0.3607078039927405,
      "grad_norm": 6.8125,
      "learning_rate": 2.996262125409863e-05,
      "loss": 0.1654,
      "step": 3180
    },
    {
      "epoch": 0.3618421052631579,
      "grad_norm": 3.75,
      "learning_rate": 2.9961211466751902e-05,
      "loss": 0.1632,
      "step": 3190
    },
    {
      "epoch": 0.3629764065335753,
      "grad_norm": 2.40625,
      "learning_rate": 2.995977561768661e-05,
      "loss": 0.142,
      "step": 3200
    },
    {
      "epoch": 0.36411070780399274,
      "grad_norm": 3.140625,
      "learning_rate": 2.995831370940394e-05,
      "loss": 0.1525,
      "step": 3210
    },
    {
      "epoch": 0.36524500907441015,
      "grad_norm": 4.5,
      "learning_rate": 2.995682574445046e-05,
      "loss": 0.1604,
      "step": 3220
    },
    {
      "epoch": 0.36637931034482757,
      "grad_norm": 4.6875,
      "learning_rate": 2.9955311725418136e-05,
      "loss": 0.1599,
      "step": 3230
    },
    {
      "epoch": 0.367513611615245,
      "grad_norm": 3.578125,
      "learning_rate": 2.995377165494432e-05,
      "loss": 0.1513,
      "step": 3240
    },
    {
      "epoch": 0.36864791288566245,
      "grad_norm": 6.34375,
      "learning_rate": 2.9952205535711742e-05,
      "loss": 0.1603,
      "step": 3250
    },
    {
      "epoch": 0.36978221415607987,
      "grad_norm": 4.75,
      "learning_rate": 2.9950613370448504e-05,
      "loss": 0.1651,
      "step": 3260
    },
    {
      "epoch": 0.3709165154264973,
      "grad_norm": 3.65625,
      "learning_rate": 2.9948995161928085e-05,
      "loss": 0.1591,
      "step": 3270
    },
    {
      "epoch": 0.3720508166969147,
      "grad_norm": 3.125,
      "learning_rate": 2.9947350912969326e-05,
      "loss": 0.1494,
      "step": 3280
    },
    {
      "epoch": 0.3731851179673321,
      "grad_norm": 3.140625,
      "learning_rate": 2.9945680626436432e-05,
      "loss": 0.1415,
      "step": 3290
    },
    {
      "epoch": 0.37431941923774953,
      "grad_norm": 3.0625,
      "learning_rate": 2.9943984305238957e-05,
      "loss": 0.1593,
      "step": 3300
    },
    {
      "epoch": 0.37545372050816694,
      "grad_norm": 5.40625,
      "learning_rate": 2.9942261952331814e-05,
      "loss": 0.157,
      "step": 3310
    },
    {
      "epoch": 0.3765880217785844,
      "grad_norm": 3.609375,
      "learning_rate": 2.9940513570715263e-05,
      "loss": 0.1597,
      "step": 3320
    },
    {
      "epoch": 0.37772232304900183,
      "grad_norm": 5.96875,
      "learning_rate": 2.99387391634349e-05,
      "loss": 0.1571,
      "step": 3330
    },
    {
      "epoch": 0.37885662431941924,
      "grad_norm": 3.046875,
      "learning_rate": 2.993693873358166e-05,
      "loss": 0.1503,
      "step": 3340
    },
    {
      "epoch": 0.37999092558983666,
      "grad_norm": 4.125,
      "learning_rate": 2.9935112284291806e-05,
      "loss": 0.1536,
      "step": 3350
    },
    {
      "epoch": 0.3811252268602541,
      "grad_norm": 2.3125,
      "learning_rate": 2.9933259818746917e-05,
      "loss": 0.1621,
      "step": 3360
    },
    {
      "epoch": 0.3822595281306715,
      "grad_norm": 3.4375,
      "learning_rate": 2.9931381340173914e-05,
      "loss": 0.1562,
      "step": 3370
    },
    {
      "epoch": 0.3833938294010889,
      "grad_norm": 2.984375,
      "learning_rate": 2.9929476851845005e-05,
      "loss": 0.1522,
      "step": 3380
    },
    {
      "epoch": 0.3845281306715064,
      "grad_norm": 3.765625,
      "learning_rate": 2.9927546357077727e-05,
      "loss": 0.1589,
      "step": 3390
    },
    {
      "epoch": 0.3856624319419238,
      "grad_norm": 3.6875,
      "learning_rate": 2.9925589859234904e-05,
      "loss": 0.153,
      "step": 3400
    },
    {
      "epoch": 0.3867967332123412,
      "grad_norm": 4.40625,
      "learning_rate": 2.9923607361724663e-05,
      "loss": 0.1477,
      "step": 3410
    },
    {
      "epoch": 0.3879310344827586,
      "grad_norm": 5.59375,
      "learning_rate": 2.9921598868000427e-05,
      "loss": 0.159,
      "step": 3420
    },
    {
      "epoch": 0.38906533575317603,
      "grad_norm": 2.78125,
      "learning_rate": 2.9919564381560884e-05,
      "loss": 0.1523,
      "step": 3430
    },
    {
      "epoch": 0.39019963702359345,
      "grad_norm": 2.453125,
      "learning_rate": 2.9917503905950022e-05,
      "loss": 0.1512,
      "step": 3440
    },
    {
      "epoch": 0.39133393829401086,
      "grad_norm": 2.9375,
      "learning_rate": 2.991541744475709e-05,
      "loss": 0.1487,
      "step": 3450
    },
    {
      "epoch": 0.39246823956442833,
      "grad_norm": 3.0,
      "learning_rate": 2.9913305001616607e-05,
      "loss": 0.1545,
      "step": 3460
    },
    {
      "epoch": 0.39360254083484575,
      "grad_norm": 2.96875,
      "learning_rate": 2.9911166580208346e-05,
      "loss": 0.1679,
      "step": 3470
    },
    {
      "epoch": 0.39473684210526316,
      "grad_norm": 3.03125,
      "learning_rate": 2.9909002184257327e-05,
      "loss": 0.1575,
      "step": 3480
    },
    {
      "epoch": 0.3958711433756806,
      "grad_norm": 3.421875,
      "learning_rate": 2.9906811817533835e-05,
      "loss": 0.1557,
      "step": 3490
    },
    {
      "epoch": 0.397005444646098,
      "grad_norm": 3.203125,
      "learning_rate": 2.9904595483853386e-05,
      "loss": 0.1452,
      "step": 3500
    },
    {
      "epoch": 0.3981397459165154,
      "grad_norm": 2.96875,
      "learning_rate": 2.9902353187076715e-05,
      "loss": 0.1481,
      "step": 3510
    },
    {
      "epoch": 0.3992740471869328,
      "grad_norm": 4.375,
      "learning_rate": 2.9900084931109812e-05,
      "loss": 0.1578,
      "step": 3520
    },
    {
      "epoch": 0.4004083484573503,
      "grad_norm": 2.734375,
      "learning_rate": 2.989779071990386e-05,
      "loss": 0.1624,
      "step": 3530
    },
    {
      "epoch": 0.4015426497277677,
      "grad_norm": 3.1875,
      "learning_rate": 2.989547055745527e-05,
      "loss": 0.1498,
      "step": 3540
    },
    {
      "epoch": 0.4026769509981851,
      "grad_norm": 3.515625,
      "learning_rate": 2.9893124447805654e-05,
      "loss": 0.1411,
      "step": 3550
    },
    {
      "epoch": 0.40381125226860254,
      "grad_norm": 3.15625,
      "learning_rate": 2.9890752395041822e-05,
      "loss": 0.1577,
      "step": 3560
    },
    {
      "epoch": 0.40494555353901995,
      "grad_norm": 6.78125,
      "learning_rate": 2.9888354403295782e-05,
      "loss": 0.1748,
      "step": 3570
    },
    {
      "epoch": 0.40607985480943737,
      "grad_norm": 3.53125,
      "learning_rate": 2.9885930476744722e-05,
      "loss": 0.1546,
      "step": 3580
    },
    {
      "epoch": 0.4072141560798548,
      "grad_norm": 4.25,
      "learning_rate": 2.9883480619610992e-05,
      "loss": 0.1571,
      "step": 3590
    },
    {
      "epoch": 0.40834845735027225,
      "grad_norm": 2.125,
      "learning_rate": 2.9881004836162145e-05,
      "loss": 0.1588,
      "step": 3600
    },
    {
      "epoch": 0.40948275862068967,
      "grad_norm": 3.0,
      "learning_rate": 2.987850313071087e-05,
      "loss": 0.1535,
      "step": 3610
    },
    {
      "epoch": 0.4106170598911071,
      "grad_norm": 2.765625,
      "learning_rate": 2.9875975507615015e-05,
      "loss": 0.1507,
      "step": 3620
    },
    {
      "epoch": 0.4117513611615245,
      "grad_norm": 2.890625,
      "learning_rate": 2.9873421971277588e-05,
      "loss": 0.1567,
      "step": 3630
    },
    {
      "epoch": 0.4128856624319419,
      "grad_norm": 2.28125,
      "learning_rate": 2.9870842526146716e-05,
      "loss": 0.1465,
      "step": 3640
    },
    {
      "epoch": 0.41401996370235933,
      "grad_norm": 4.53125,
      "learning_rate": 2.9868237176715685e-05,
      "loss": 0.1575,
      "step": 3650
    },
    {
      "epoch": 0.41515426497277674,
      "grad_norm": 3.9375,
      "learning_rate": 2.9865605927522878e-05,
      "loss": 0.1605,
      "step": 3660
    },
    {
      "epoch": 0.4162885662431942,
      "grad_norm": 2.59375,
      "learning_rate": 2.9862948783151812e-05,
      "loss": 0.1392,
      "step": 3670
    },
    {
      "epoch": 0.41742286751361163,
      "grad_norm": 4.34375,
      "learning_rate": 2.98602657482311e-05,
      "loss": 0.1652,
      "step": 3680
    },
    {
      "epoch": 0.41855716878402904,
      "grad_norm": 3.390625,
      "learning_rate": 2.9857556827434468e-05,
      "loss": 0.1489,
      "step": 3690
    },
    {
      "epoch": 0.41969147005444646,
      "grad_norm": 3.890625,
      "learning_rate": 2.985482202548072e-05,
      "loss": 0.1605,
      "step": 3700
    },
    {
      "epoch": 0.4208257713248639,
      "grad_norm": 3.578125,
      "learning_rate": 2.985206134713376e-05,
      "loss": 0.1498,
      "step": 3710
    },
    {
      "epoch": 0.4219600725952813,
      "grad_norm": 2.546875,
      "learning_rate": 2.984927479720255e-05,
      "loss": 0.1589,
      "step": 3720
    },
    {
      "epoch": 0.4230943738656987,
      "grad_norm": 5.6875,
      "learning_rate": 2.9846462380541135e-05,
      "loss": 0.1394,
      "step": 3730
    },
    {
      "epoch": 0.4242286751361162,
      "grad_norm": 3.125,
      "learning_rate": 2.9843624102048597e-05,
      "loss": 0.148,
      "step": 3740
    },
    {
      "epoch": 0.4253629764065336,
      "grad_norm": 2.40625,
      "learning_rate": 2.98407599666691e-05,
      "loss": 0.149,
      "step": 3750
    },
    {
      "epoch": 0.426497277676951,
      "grad_norm": 3.015625,
      "learning_rate": 2.983786997939182e-05,
      "loss": 0.1548,
      "step": 3760
    },
    {
      "epoch": 0.4276315789473684,
      "grad_norm": 2.09375,
      "learning_rate": 2.983495414525098e-05,
      "loss": 0.146,
      "step": 3770
    },
    {
      "epoch": 0.42876588021778583,
      "grad_norm": 2.859375,
      "learning_rate": 2.9832012469325833e-05,
      "loss": 0.1502,
      "step": 3780
    },
    {
      "epoch": 0.42990018148820325,
      "grad_norm": 2.734375,
      "learning_rate": 2.982904495674062e-05,
      "loss": 0.156,
      "step": 3790
    },
    {
      "epoch": 0.43103448275862066,
      "grad_norm": 2.890625,
      "learning_rate": 2.982605161266462e-05,
      "loss": 0.1453,
      "step": 3800
    },
    {
      "epoch": 0.43216878402903813,
      "grad_norm": 3.015625,
      "learning_rate": 2.9823032442312097e-05,
      "loss": 0.1481,
      "step": 3810
    },
    {
      "epoch": 0.43330308529945555,
      "grad_norm": 2.828125,
      "learning_rate": 2.9819987450942293e-05,
      "loss": 0.1392,
      "step": 3820
    },
    {
      "epoch": 0.43443738656987296,
      "grad_norm": 3.46875,
      "learning_rate": 2.9816916643859443e-05,
      "loss": 0.1522,
      "step": 3830
    },
    {
      "epoch": 0.4355716878402904,
      "grad_norm": 2.796875,
      "learning_rate": 2.9813820026412754e-05,
      "loss": 0.1488,
      "step": 3840
    },
    {
      "epoch": 0.4367059891107078,
      "grad_norm": 3.0,
      "learning_rate": 2.9810697603996373e-05,
      "loss": 0.1413,
      "step": 3850
    },
    {
      "epoch": 0.4378402903811252,
      "grad_norm": 3.859375,
      "learning_rate": 2.9807549382049423e-05,
      "loss": 0.1568,
      "step": 3860
    },
    {
      "epoch": 0.4389745916515426,
      "grad_norm": 1.7109375,
      "learning_rate": 2.9804375366055948e-05,
      "loss": 0.1505,
      "step": 3870
    },
    {
      "epoch": 0.4401088929219601,
      "grad_norm": 2.171875,
      "learning_rate": 2.9801175561544938e-05,
      "loss": 0.1591,
      "step": 3880
    },
    {
      "epoch": 0.4412431941923775,
      "grad_norm": 3.75,
      "learning_rate": 2.9797949974090294e-05,
      "loss": 0.151,
      "step": 3890
    },
    {
      "epoch": 0.4423774954627949,
      "grad_norm": 2.578125,
      "learning_rate": 2.979469860931085e-05,
      "loss": 0.1593,
      "step": 3900
    },
    {
      "epoch": 0.44351179673321234,
      "grad_norm": 4.21875,
      "learning_rate": 2.979142147287031e-05,
      "loss": 0.1549,
      "step": 3910
    },
    {
      "epoch": 0.44464609800362975,
      "grad_norm": 2.28125,
      "learning_rate": 2.9788118570477303e-05,
      "loss": 0.145,
      "step": 3920
    },
    {
      "epoch": 0.44578039927404717,
      "grad_norm": 3.359375,
      "learning_rate": 2.9784789907885325e-05,
      "loss": 0.167,
      "step": 3930
    },
    {
      "epoch": 0.4469147005444646,
      "grad_norm": 2.296875,
      "learning_rate": 2.9781435490892745e-05,
      "loss": 0.1485,
      "step": 3940
    },
    {
      "epoch": 0.44804900181488205,
      "grad_norm": 2.125,
      "learning_rate": 2.97780553253428e-05,
      "loss": 0.1489,
      "step": 3950
    },
    {
      "epoch": 0.44918330308529947,
      "grad_norm": 2.453125,
      "learning_rate": 2.9774649417123576e-05,
      "loss": 0.1462,
      "step": 3960
    },
    {
      "epoch": 0.4503176043557169,
      "grad_norm": 5.03125,
      "learning_rate": 2.977121777216801e-05,
      "loss": 0.1532,
      "step": 3970
    },
    {
      "epoch": 0.4514519056261343,
      "grad_norm": 3.640625,
      "learning_rate": 2.9767760396453853e-05,
      "loss": 0.1583,
      "step": 3980
    },
    {
      "epoch": 0.4525862068965517,
      "grad_norm": 2.171875,
      "learning_rate": 2.9764277296003693e-05,
      "loss": 0.1598,
      "step": 3990
    },
    {
      "epoch": 0.4537205081669691,
      "grad_norm": 3.375,
      "learning_rate": 2.976076847688493e-05,
      "loss": 0.146,
      "step": 4000
    },
    {
      "epoch": 0.45485480943738654,
      "grad_norm": 2.28125,
      "learning_rate": 2.975723394520976e-05,
      "loss": 0.1567,
      "step": 4010
    },
    {
      "epoch": 0.455989110707804,
      "grad_norm": 3.84375,
      "learning_rate": 2.9753673707135162e-05,
      "loss": 0.1357,
      "step": 4020
    },
    {
      "epoch": 0.45712341197822143,
      "grad_norm": 3.390625,
      "learning_rate": 2.97500877688629e-05,
      "loss": 0.1553,
      "step": 4030
    },
    {
      "epoch": 0.45825771324863884,
      "grad_norm": 2.75,
      "learning_rate": 2.9746476136639518e-05,
      "loss": 0.1568,
      "step": 4040
    },
    {
      "epoch": 0.45939201451905626,
      "grad_norm": 2.046875,
      "learning_rate": 2.9742838816756303e-05,
      "loss": 0.1618,
      "step": 4050
    },
    {
      "epoch": 0.4605263157894737,
      "grad_norm": 2.796875,
      "learning_rate": 2.973917581554929e-05,
      "loss": 0.1559,
      "step": 4060
    },
    {
      "epoch": 0.4616606170598911,
      "grad_norm": 2.90625,
      "learning_rate": 2.973548713939926e-05,
      "loss": 0.1518,
      "step": 4070
    },
    {
      "epoch": 0.4627949183303085,
      "grad_norm": 2.78125,
      "learning_rate": 2.9731772794731704e-05,
      "loss": 0.1401,
      "step": 4080
    },
    {
      "epoch": 0.463929219600726,
      "grad_norm": 1.84375,
      "learning_rate": 2.9728032788016836e-05,
      "loss": 0.1428,
      "step": 4090
    },
    {
      "epoch": 0.4650635208711434,
      "grad_norm": 2.875,
      "learning_rate": 2.972426712576957e-05,
      "loss": 0.1546,
      "step": 4100
    },
    {
      "epoch": 0.4661978221415608,
      "grad_norm": 3.3125,
      "learning_rate": 2.972047581454951e-05,
      "loss": 0.1524,
      "step": 4110
    },
    {
      "epoch": 0.4673321234119782,
      "grad_norm": 3.09375,
      "learning_rate": 2.9716658860960942e-05,
      "loss": 0.1619,
      "step": 4120
    },
    {
      "epoch": 0.46846642468239563,
      "grad_norm": 2.984375,
      "learning_rate": 2.9712816271652816e-05,
      "loss": 0.1502,
      "step": 4130
    },
    {
      "epoch": 0.46960072595281305,
      "grad_norm": 2.390625,
      "learning_rate": 2.970894805331874e-05,
      "loss": 0.1528,
      "step": 4140
    },
    {
      "epoch": 0.4707350272232305,
      "grad_norm": 2.46875,
      "learning_rate": 2.9705054212696965e-05,
      "loss": 0.1445,
      "step": 4150
    },
    {
      "epoch": 0.47186932849364793,
      "grad_norm": 3.234375,
      "learning_rate": 2.9701134756570376e-05,
      "loss": 0.1408,
      "step": 4160
    },
    {
      "epoch": 0.47300362976406535,
      "grad_norm": 2.203125,
      "learning_rate": 2.9697189691766483e-05,
      "loss": 0.1539,
      "step": 4170
    },
    {
      "epoch": 0.47413793103448276,
      "grad_norm": 3.59375,
      "learning_rate": 2.9693219025157398e-05,
      "loss": 0.1573,
      "step": 4180
    },
    {
      "epoch": 0.4752722323049002,
      "grad_norm": 3.265625,
      "learning_rate": 2.9689222763659834e-05,
      "loss": 0.1519,
      "step": 4190
    },
    {
      "epoch": 0.4764065335753176,
      "grad_norm": 2.484375,
      "learning_rate": 2.9685200914235086e-05,
      "loss": 0.1489,
      "step": 4200
    },
    {
      "epoch": 0.477540834845735,
      "grad_norm": 2.90625,
      "learning_rate": 2.9681153483889026e-05,
      "loss": 0.162,
      "step": 4210
    },
    {
      "epoch": 0.4786751361161525,
      "grad_norm": 2.546875,
      "learning_rate": 2.967708047967209e-05,
      "loss": 0.1507,
      "step": 4220
    },
    {
      "epoch": 0.4798094373865699,
      "grad_norm": 2.46875,
      "learning_rate": 2.9672981908679253e-05,
      "loss": 0.1469,
      "step": 4230
    },
    {
      "epoch": 0.4809437386569873,
      "grad_norm": 2.265625,
      "learning_rate": 2.9668857778050027e-05,
      "loss": 0.1506,
      "step": 4240
    },
    {
      "epoch": 0.4820780399274047,
      "grad_norm": 3.703125,
      "learning_rate": 2.966470809496846e-05,
      "loss": 0.1553,
      "step": 4250
    },
    {
      "epoch": 0.48321234119782214,
      "grad_norm": 1.8203125,
      "learning_rate": 2.9660532866663094e-05,
      "loss": 0.1488,
      "step": 4260
    },
    {
      "epoch": 0.48434664246823955,
      "grad_norm": 3.21875,
      "learning_rate": 2.965633210040698e-05,
      "loss": 0.1396,
      "step": 4270
    },
    {
      "epoch": 0.48548094373865697,
      "grad_norm": 2.1875,
      "learning_rate": 2.965210580351766e-05,
      "loss": 0.1578,
      "step": 4280
    },
    {
      "epoch": 0.48661524500907444,
      "grad_norm": 3.1875,
      "learning_rate": 2.9647853983357133e-05,
      "loss": 0.1541,
      "step": 4290
    },
    {
      "epoch": 0.48774954627949185,
      "grad_norm": 2.375,
      "learning_rate": 2.964357664733187e-05,
      "loss": 0.1564,
      "step": 4300
    },
    {
      "epoch": 0.48888384754990927,
      "grad_norm": 3.046875,
      "learning_rate": 2.9639273802892792e-05,
      "loss": 0.149,
      "step": 4310
    },
    {
      "epoch": 0.4900181488203267,
      "grad_norm": 4.34375,
      "learning_rate": 2.9634945457535244e-05,
      "loss": 0.1523,
      "step": 4320
    },
    {
      "epoch": 0.4911524500907441,
      "grad_norm": 4.0,
      "learning_rate": 2.9630591618798997e-05,
      "loss": 0.1524,
      "step": 4330
    },
    {
      "epoch": 0.4922867513611615,
      "grad_norm": 2.40625,
      "learning_rate": 2.9626212294268235e-05,
      "loss": 0.1539,
      "step": 4340
    },
    {
      "epoch": 0.4934210526315789,
      "grad_norm": 2.96875,
      "learning_rate": 2.9621807491571533e-05,
      "loss": 0.1565,
      "step": 4350
    },
    {
      "epoch": 0.4945553539019964,
      "grad_norm": 2.234375,
      "learning_rate": 2.9617377218381844e-05,
      "loss": 0.148,
      "step": 4360
    },
    {
      "epoch": 0.4956896551724138,
      "grad_norm": 2.453125,
      "learning_rate": 2.9612921482416495e-05,
      "loss": 0.1564,
      "step": 4370
    },
    {
      "epoch": 0.4968239564428312,
      "grad_norm": 2.78125,
      "learning_rate": 2.9608440291437168e-05,
      "loss": 0.1529,
      "step": 4380
    },
    {
      "epoch": 0.49795825771324864,
      "grad_norm": 2.296875,
      "learning_rate": 2.9603933653249878e-05,
      "loss": 0.1502,
      "step": 4390
    },
    {
      "epoch": 0.49909255898366606,
      "grad_norm": 3.21875,
      "learning_rate": 2.959940157570498e-05,
      "loss": 0.1499,
      "step": 4400
    },
    {
      "epoch": 0.5002268602540835,
      "grad_norm": 2.3125,
      "learning_rate": 2.9594844066697136e-05,
      "loss": 0.1458,
      "step": 4410
    },
    {
      "epoch": 0.5013611615245009,
      "grad_norm": 2.171875,
      "learning_rate": 2.959026113416531e-05,
      "loss": 0.147,
      "step": 4420
    },
    {
      "epoch": 0.5024954627949183,
      "grad_norm": 3.15625,
      "learning_rate": 2.9585652786092754e-05,
      "loss": 0.1517,
      "step": 4430
    },
    {
      "epoch": 0.5036297640653358,
      "grad_norm": 2.859375,
      "learning_rate": 2.958101903050699e-05,
      "loss": 0.1434,
      "step": 4440
    },
    {
      "epoch": 0.5047640653357531,
      "grad_norm": 3.234375,
      "learning_rate": 2.9576359875479798e-05,
      "loss": 0.1426,
      "step": 4450
    },
    {
      "epoch": 0.5058983666061706,
      "grad_norm": 2.59375,
      "learning_rate": 2.9571675329127203e-05,
      "loss": 0.1481,
      "step": 4460
    },
    {
      "epoch": 0.5070326678765881,
      "grad_norm": 2.6875,
      "learning_rate": 2.9566965399609468e-05,
      "loss": 0.151,
      "step": 4470
    },
    {
      "epoch": 0.5081669691470054,
      "grad_norm": 3.234375,
      "learning_rate": 2.9562230095131065e-05,
      "loss": 0.1503,
      "step": 4480
    },
    {
      "epoch": 0.5093012704174229,
      "grad_norm": 2.609375,
      "learning_rate": 2.9557469423940662e-05,
      "loss": 0.1531,
      "step": 4490
    },
    {
      "epoch": 0.5104355716878403,
      "grad_norm": 3.234375,
      "learning_rate": 2.955268339433113e-05,
      "loss": 0.1481,
      "step": 4500
    },
    {
      "epoch": 0.5115698729582577,
      "grad_norm": 3.671875,
      "learning_rate": 2.95478720146395e-05,
      "loss": 0.1454,
      "step": 4510
    },
    {
      "epoch": 0.5127041742286751,
      "grad_norm": 3.09375,
      "learning_rate": 2.9543035293246968e-05,
      "loss": 0.1524,
      "step": 4520
    },
    {
      "epoch": 0.5138384754990926,
      "grad_norm": 3.09375,
      "learning_rate": 2.9538173238578872e-05,
      "loss": 0.1529,
      "step": 4530
    },
    {
      "epoch": 0.51497277676951,
      "grad_norm": 3.03125,
      "learning_rate": 2.953328585910468e-05,
      "loss": 0.1509,
      "step": 4540
    },
    {
      "epoch": 0.5161070780399274,
      "grad_norm": 2.984375,
      "learning_rate": 2.9528373163337977e-05,
      "loss": 0.1467,
      "step": 4550
    },
    {
      "epoch": 0.5172413793103449,
      "grad_norm": 2.234375,
      "learning_rate": 2.9523435159836437e-05,
      "loss": 0.1541,
      "step": 4560
    },
    {
      "epoch": 0.5183756805807622,
      "grad_norm": 1.7890625,
      "learning_rate": 2.9518471857201838e-05,
      "loss": 0.1388,
      "step": 4570
    },
    {
      "epoch": 0.5195099818511797,
      "grad_norm": 3.3125,
      "learning_rate": 2.9513483264080016e-05,
      "loss": 0.1515,
      "step": 4580
    },
    {
      "epoch": 0.520644283121597,
      "grad_norm": 2.953125,
      "learning_rate": 2.9508469389160854e-05,
      "loss": 0.1513,
      "step": 4590
    },
    {
      "epoch": 0.5217785843920145,
      "grad_norm": 2.84375,
      "learning_rate": 2.950343024117829e-05,
      "loss": 0.1505,
      "step": 4600
    },
    {
      "epoch": 0.522912885662432,
      "grad_norm": 4.15625,
      "learning_rate": 2.949836582891028e-05,
      "loss": 0.1563,
      "step": 4610
    },
    {
      "epoch": 0.5240471869328494,
      "grad_norm": 2.5,
      "learning_rate": 2.9493276161178788e-05,
      "loss": 0.1562,
      "step": 4620
    },
    {
      "epoch": 0.5251814882032668,
      "grad_norm": 2.90625,
      "learning_rate": 2.9488161246849768e-05,
      "loss": 0.1507,
      "step": 4630
    },
    {
      "epoch": 0.5263157894736842,
      "grad_norm": 2.5,
      "learning_rate": 2.948302109483317e-05,
      "loss": 0.1632,
      "step": 4640
    },
    {
      "epoch": 0.5274500907441017,
      "grad_norm": 3.140625,
      "learning_rate": 2.9477855714082884e-05,
      "loss": 0.1526,
      "step": 4650
    },
    {
      "epoch": 0.528584392014519,
      "grad_norm": 2.375,
      "learning_rate": 2.9472665113596758e-05,
      "loss": 0.1442,
      "step": 4660
    },
    {
      "epoch": 0.5297186932849365,
      "grad_norm": 2.96875,
      "learning_rate": 2.946744930241658e-05,
      "loss": 0.1521,
      "step": 4670
    },
    {
      "epoch": 0.530852994555354,
      "grad_norm": 2.875,
      "learning_rate": 2.946220828962803e-05,
      "loss": 0.1487,
      "step": 4680
    },
    {
      "epoch": 0.5319872958257713,
      "grad_norm": 2.703125,
      "learning_rate": 2.9456942084360715e-05,
      "loss": 0.1534,
      "step": 4690
    },
    {
      "epoch": 0.5331215970961888,
      "grad_norm": 2.40625,
      "learning_rate": 2.9451650695788114e-05,
      "loss": 0.1461,
      "step": 4700
    },
    {
      "epoch": 0.5342558983666061,
      "grad_norm": 3.921875,
      "learning_rate": 2.944633413312756e-05,
      "loss": 0.1524,
      "step": 4710
    },
    {
      "epoch": 0.5353901996370236,
      "grad_norm": 2.921875,
      "learning_rate": 2.9440992405640273e-05,
      "loss": 0.1463,
      "step": 4720
    },
    {
      "epoch": 0.536524500907441,
      "grad_norm": 2.515625,
      "learning_rate": 2.9435625522631277e-05,
      "loss": 0.1513,
      "step": 4730
    },
    {
      "epoch": 0.5376588021778584,
      "grad_norm": 3.265625,
      "learning_rate": 2.943023349344942e-05,
      "loss": 0.1538,
      "step": 4740
    },
    {
      "epoch": 0.5387931034482759,
      "grad_norm": 2.890625,
      "learning_rate": 2.9424816327487373e-05,
      "loss": 0.156,
      "step": 4750
    },
    {
      "epoch": 0.5399274047186933,
      "grad_norm": 2.625,
      "learning_rate": 2.9419374034181578e-05,
      "loss": 0.1546,
      "step": 4760
    },
    {
      "epoch": 0.5410617059891107,
      "grad_norm": 3.046875,
      "learning_rate": 2.9413906623012244e-05,
      "loss": 0.1485,
      "step": 4770
    },
    {
      "epoch": 0.5421960072595281,
      "grad_norm": 3.984375,
      "learning_rate": 2.940841410350335e-05,
      "loss": 0.1475,
      "step": 4780
    },
    {
      "epoch": 0.5433303085299456,
      "grad_norm": 2.828125,
      "learning_rate": 2.94028964852226e-05,
      "loss": 0.154,
      "step": 4790
    },
    {
      "epoch": 0.5444646098003629,
      "grad_norm": 2.828125,
      "learning_rate": 2.9397353777781423e-05,
      "loss": 0.1464,
      "step": 4800
    },
    {
      "epoch": 0.5455989110707804,
      "grad_norm": 2.703125,
      "learning_rate": 2.939178599083495e-05,
      "loss": 0.1555,
      "step": 4810
    },
    {
      "epoch": 0.5467332123411979,
      "grad_norm": 4.09375,
      "learning_rate": 2.9386193134082003e-05,
      "loss": 0.1574,
      "step": 4820
    },
    {
      "epoch": 0.5478675136116152,
      "grad_norm": 2.0625,
      "learning_rate": 2.9380575217265072e-05,
      "loss": 0.1519,
      "step": 4830
    },
    {
      "epoch": 0.5490018148820327,
      "grad_norm": 3.125,
      "learning_rate": 2.93749322501703e-05,
      "loss": 0.148,
      "step": 4840
    },
    {
      "epoch": 0.5501361161524501,
      "grad_norm": 3.078125,
      "learning_rate": 2.9369264242627467e-05,
      "loss": 0.1508,
      "step": 4850
    },
    {
      "epoch": 0.5512704174228675,
      "grad_norm": 2.078125,
      "learning_rate": 2.936357120450997e-05,
      "loss": 0.1455,
      "step": 4860
    },
    {
      "epoch": 0.5524047186932849,
      "grad_norm": 3.0,
      "learning_rate": 2.935785314573482e-05,
      "loss": 0.1408,
      "step": 4870
    },
    {
      "epoch": 0.5535390199637024,
      "grad_norm": 2.5625,
      "learning_rate": 2.9352110076262595e-05,
      "loss": 0.1588,
      "step": 4880
    },
    {
      "epoch": 0.5546733212341198,
      "grad_norm": 2.34375,
      "learning_rate": 2.934634200609745e-05,
      "loss": 0.146,
      "step": 4890
    },
    {
      "epoch": 0.5558076225045372,
      "grad_norm": 2.015625,
      "learning_rate": 2.9340548945287088e-05,
      "loss": 0.1488,
      "step": 4900
    },
    {
      "epoch": 0.5569419237749547,
      "grad_norm": 2.3125,
      "learning_rate": 2.9334730903922747e-05,
      "loss": 0.1502,
      "step": 4910
    },
    {
      "epoch": 0.558076225045372,
      "grad_norm": 3.53125,
      "learning_rate": 2.932888789213918e-05,
      "loss": 0.1345,
      "step": 4920
    },
    {
      "epoch": 0.5592105263157895,
      "grad_norm": 2.21875,
      "learning_rate": 2.9323019920114634e-05,
      "loss": 0.1617,
      "step": 4930
    },
    {
      "epoch": 0.5603448275862069,
      "grad_norm": 2.390625,
      "learning_rate": 2.9317126998070836e-05,
      "loss": 0.1445,
      "step": 4940
    },
    {
      "epoch": 0.5614791288566243,
      "grad_norm": 2.296875,
      "learning_rate": 2.9311209136272977e-05,
      "loss": 0.1419,
      "step": 4950
    },
    {
      "epoch": 0.5626134301270418,
      "grad_norm": 2.46875,
      "learning_rate": 2.930526634502969e-05,
      "loss": 0.1555,
      "step": 4960
    },
    {
      "epoch": 0.5637477313974592,
      "grad_norm": 3.375,
      "learning_rate": 2.9299298634693033e-05,
      "loss": 0.1436,
      "step": 4970
    },
    {
      "epoch": 0.5648820326678766,
      "grad_norm": 2.484375,
      "learning_rate": 2.9293306015658474e-05,
      "loss": 0.1554,
      "step": 4980
    },
    {
      "epoch": 0.566016333938294,
      "grad_norm": 2.390625,
      "learning_rate": 2.9287288498364876e-05,
      "loss": 0.1545,
      "step": 4990
    },
    {
      "epoch": 0.5671506352087115,
      "grad_norm": 1.6484375,
      "learning_rate": 2.928124609329446e-05,
      "loss": 0.158,
      "step": 5000
    },
    {
      "epoch": 0.5682849364791288,
      "grad_norm": 1.7265625,
      "learning_rate": 2.9275178810972817e-05,
      "loss": 0.145,
      "step": 5010
    },
    {
      "epoch": 0.5694192377495463,
      "grad_norm": 2.25,
      "learning_rate": 2.926908666196885e-05,
      "loss": 0.1479,
      "step": 5020
    },
    {
      "epoch": 0.5705535390199638,
      "grad_norm": 2.328125,
      "learning_rate": 2.926296965689481e-05,
      "loss": 0.1463,
      "step": 5030
    },
    {
      "epoch": 0.5716878402903811,
      "grad_norm": 1.7890625,
      "learning_rate": 2.9256827806406216e-05,
      "loss": 0.151,
      "step": 5040
    },
    {
      "epoch": 0.5728221415607986,
      "grad_norm": 2.796875,
      "learning_rate": 2.9250661121201884e-05,
      "loss": 0.1503,
      "step": 5050
    },
    {
      "epoch": 0.5739564428312159,
      "grad_norm": 3.390625,
      "learning_rate": 2.9244469612023886e-05,
      "loss": 0.1518,
      "step": 5060
    },
    {
      "epoch": 0.5750907441016334,
      "grad_norm": 2.796875,
      "learning_rate": 2.9238253289657535e-05,
      "loss": 0.1434,
      "step": 5070
    },
    {
      "epoch": 0.5762250453720508,
      "grad_norm": 2.1875,
      "learning_rate": 2.9232012164931376e-05,
      "loss": 0.1548,
      "step": 5080
    },
    {
      "epoch": 0.5773593466424682,
      "grad_norm": 2.515625,
      "learning_rate": 2.9225746248717137e-05,
      "loss": 0.1472,
      "step": 5090
    },
    {
      "epoch": 0.5784936479128857,
      "grad_norm": 3.046875,
      "learning_rate": 2.921945555192976e-05,
      "loss": 0.1474,
      "step": 5100
    },
    {
      "epoch": 0.5796279491833031,
      "grad_norm": 2.4375,
      "learning_rate": 2.9213140085527332e-05,
      "loss": 0.1377,
      "step": 5110
    },
    {
      "epoch": 0.5807622504537205,
      "grad_norm": 2.296875,
      "learning_rate": 2.920679986051109e-05,
      "loss": 0.1351,
      "step": 5120
    },
    {
      "epoch": 0.5818965517241379,
      "grad_norm": 3.09375,
      "learning_rate": 2.920043488792542e-05,
      "loss": 0.1317,
      "step": 5130
    },
    {
      "epoch": 0.5830308529945554,
      "grad_norm": 2.25,
      "learning_rate": 2.919404517885779e-05,
      "loss": 0.1446,
      "step": 5140
    },
    {
      "epoch": 0.5841651542649727,
      "grad_norm": 2.125,
      "learning_rate": 2.9187630744438775e-05,
      "loss": 0.1412,
      "step": 5150
    },
    {
      "epoch": 0.5852994555353902,
      "grad_norm": 2.703125,
      "learning_rate": 2.918119159584201e-05,
      "loss": 0.1517,
      "step": 5160
    },
    {
      "epoch": 0.5864337568058077,
      "grad_norm": 2.328125,
      "learning_rate": 2.9174727744284182e-05,
      "loss": 0.156,
      "step": 5170
    },
    {
      "epoch": 0.587568058076225,
      "grad_norm": 2.828125,
      "learning_rate": 2.916823920102502e-05,
      "loss": 0.1451,
      "step": 5180
    },
    {
      "epoch": 0.5887023593466425,
      "grad_norm": 1.875,
      "learning_rate": 2.916172597736726e-05,
      "loss": 0.1433,
      "step": 5190
    },
    {
      "epoch": 0.5898366606170599,
      "grad_norm": 2.4375,
      "learning_rate": 2.9155188084656615e-05,
      "loss": 0.1564,
      "step": 5200
    },
    {
      "epoch": 0.5909709618874773,
      "grad_norm": 2.40625,
      "learning_rate": 2.91486255342818e-05,
      "loss": 0.1464,
      "step": 5210
    },
    {
      "epoch": 0.5921052631578947,
      "grad_norm": 1.703125,
      "learning_rate": 2.9142038337674452e-05,
      "loss": 0.1528,
      "step": 5220
    },
    {
      "epoch": 0.5932395644283122,
      "grad_norm": 2.6875,
      "learning_rate": 2.9135426506309157e-05,
      "loss": 0.1532,
      "step": 5230
    },
    {
      "epoch": 0.5943738656987296,
      "grad_norm": 2.484375,
      "learning_rate": 2.9128790051703414e-05,
      "loss": 0.1393,
      "step": 5240
    },
    {
      "epoch": 0.595508166969147,
      "grad_norm": 2.53125,
      "learning_rate": 2.9122128985417612e-05,
      "loss": 0.151,
      "step": 5250
    },
    {
      "epoch": 0.5966424682395645,
      "grad_norm": 3.203125,
      "learning_rate": 2.9115443319055004e-05,
      "loss": 0.1571,
      "step": 5260
    },
    {
      "epoch": 0.5977767695099818,
      "grad_norm": 1.953125,
      "learning_rate": 2.910873306426171e-05,
      "loss": 0.142,
      "step": 5270
    },
    {
      "epoch": 0.5989110707803993,
      "grad_norm": 2.015625,
      "learning_rate": 2.910199823272667e-05,
      "loss": 0.1471,
      "step": 5280
    },
    {
      "epoch": 0.6000453720508166,
      "grad_norm": 2.28125,
      "learning_rate": 2.9095238836181648e-05,
      "loss": 0.1522,
      "step": 5290
    },
    {
      "epoch": 0.6011796733212341,
      "grad_norm": 2.3125,
      "learning_rate": 2.9088454886401184e-05,
      "loss": 0.14,
      "step": 5300
    },
    {
      "epoch": 0.6023139745916516,
      "grad_norm": 2.3125,
      "learning_rate": 2.90816463952026e-05,
      "loss": 0.1561,
      "step": 5310
    },
    {
      "epoch": 0.603448275862069,
      "grad_norm": 2.28125,
      "learning_rate": 2.9074813374445955e-05,
      "loss": 0.1443,
      "step": 5320
    },
    {
      "epoch": 0.6045825771324864,
      "grad_norm": 2.578125,
      "learning_rate": 2.9067955836034055e-05,
      "loss": 0.1519,
      "step": 5330
    },
    {
      "epoch": 0.6057168784029038,
      "grad_norm": 2.34375,
      "learning_rate": 2.9061073791912405e-05,
      "loss": 0.1456,
      "step": 5340
    },
    {
      "epoch": 0.6068511796733213,
      "grad_norm": 3.71875,
      "learning_rate": 2.9054167254069193e-05,
      "loss": 0.1507,
      "step": 5350
    },
    {
      "epoch": 0.6079854809437386,
      "grad_norm": 3.421875,
      "learning_rate": 2.9047236234535285e-05,
      "loss": 0.1453,
      "step": 5360
    },
    {
      "epoch": 0.6091197822141561,
      "grad_norm": 3.359375,
      "learning_rate": 2.9040280745384188e-05,
      "loss": 0.1458,
      "step": 5370
    },
    {
      "epoch": 0.6102540834845736,
      "grad_norm": 2.515625,
      "learning_rate": 2.9033300798732024e-05,
      "loss": 0.1494,
      "step": 5380
    },
    {
      "epoch": 0.6113883847549909,
      "grad_norm": 1.7265625,
      "learning_rate": 2.9026296406737536e-05,
      "loss": 0.1492,
      "step": 5390
    },
    {
      "epoch": 0.6125226860254084,
      "grad_norm": 2.515625,
      "learning_rate": 2.9019267581602042e-05,
      "loss": 0.1536,
      "step": 5400
    },
    {
      "epoch": 0.6136569872958257,
      "grad_norm": 2.5625,
      "learning_rate": 2.9012214335569417e-05,
      "loss": 0.1515,
      "step": 5410
    },
    {
      "epoch": 0.6147912885662432,
      "grad_norm": 3.171875,
      "learning_rate": 2.900513668092608e-05,
      "loss": 0.1413,
      "step": 5420
    },
    {
      "epoch": 0.6159255898366606,
      "grad_norm": 3.078125,
      "learning_rate": 2.8998034630000974e-05,
      "loss": 0.1442,
      "step": 5430
    },
    {
      "epoch": 0.617059891107078,
      "grad_norm": 2.515625,
      "learning_rate": 2.899090819516553e-05,
      "loss": 0.1499,
      "step": 5440
    },
    {
      "epoch": 0.6181941923774955,
      "grad_norm": 3.28125,
      "learning_rate": 2.8983757388833657e-05,
      "loss": 0.1562,
      "step": 5450
    },
    {
      "epoch": 0.6193284936479129,
      "grad_norm": 2.6875,
      "learning_rate": 2.8976582223461722e-05,
      "loss": 0.1541,
      "step": 5460
    },
    {
      "epoch": 0.6204627949183303,
      "grad_norm": 2.890625,
      "learning_rate": 2.896938271154852e-05,
      "loss": 0.1499,
      "step": 5470
    },
    {
      "epoch": 0.6215970961887477,
      "grad_norm": 4.84375,
      "learning_rate": 2.8962158865635258e-05,
      "loss": 0.1545,
      "step": 5480
    },
    {
      "epoch": 0.6227313974591652,
      "grad_norm": 2.3125,
      "learning_rate": 2.8954910698305532e-05,
      "loss": 0.1469,
      "step": 5490
    },
    {
      "epoch": 0.6238656987295825,
      "grad_norm": 3.453125,
      "learning_rate": 2.8947638222185312e-05,
      "loss": 0.1492,
      "step": 5500
    },
    {
      "epoch": 0.625,
      "grad_norm": 2.375,
      "learning_rate": 2.8940341449942892e-05,
      "loss": 0.1431,
      "step": 5510
    },
    {
      "epoch": 0.6261343012704175,
      "grad_norm": 1.8828125,
      "learning_rate": 2.893302039428891e-05,
      "loss": 0.1462,
      "step": 5520
    },
    {
      "epoch": 0.6272686025408348,
      "grad_norm": 2.3125,
      "learning_rate": 2.89256750679763e-05,
      "loss": 0.1319,
      "step": 5530
    },
    {
      "epoch": 0.6284029038112523,
      "grad_norm": 2.375,
      "learning_rate": 2.8918305483800263e-05,
      "loss": 0.1538,
      "step": 5540
    },
    {
      "epoch": 0.6295372050816697,
      "grad_norm": 2.359375,
      "learning_rate": 2.8910911654598268e-05,
      "loss": 0.1581,
      "step": 5550
    },
    {
      "epoch": 0.6306715063520871,
      "grad_norm": 2.296875,
      "learning_rate": 2.8903493593250014e-05,
      "loss": 0.1477,
      "step": 5560
    },
    {
      "epoch": 0.6318058076225045,
      "grad_norm": 1.890625,
      "learning_rate": 2.889605131267741e-05,
      "loss": 0.1467,
      "step": 5570
    },
    {
      "epoch": 0.632940108892922,
      "grad_norm": 1.6484375,
      "learning_rate": 2.8888584825844556e-05,
      "loss": 0.1407,
      "step": 5580
    },
    {
      "epoch": 0.6340744101633394,
      "grad_norm": 3.4375,
      "learning_rate": 2.8881094145757715e-05,
      "loss": 0.1553,
      "step": 5590
    },
    {
      "epoch": 0.6352087114337568,
      "grad_norm": 2.75,
      "learning_rate": 2.8873579285465292e-05,
      "loss": 0.1457,
      "step": 5600
    },
    {
      "epoch": 0.6363430127041743,
      "grad_norm": 2.015625,
      "learning_rate": 2.8866040258057824e-05,
      "loss": 0.1486,
      "step": 5610
    },
    {
      "epoch": 0.6374773139745916,
      "grad_norm": 1.71875,
      "learning_rate": 2.8858477076667932e-05,
      "loss": 0.1411,
      "step": 5620
    },
    {
      "epoch": 0.6386116152450091,
      "grad_norm": 2.15625,
      "learning_rate": 2.8850889754470324e-05,
      "loss": 0.1581,
      "step": 5630
    },
    {
      "epoch": 0.6397459165154264,
      "grad_norm": 2.796875,
      "learning_rate": 2.8843278304681748e-05,
      "loss": 0.1584,
      "step": 5640
    },
    {
      "epoch": 0.6408802177858439,
      "grad_norm": 2.90625,
      "learning_rate": 2.883564274056099e-05,
      "loss": 0.1508,
      "step": 5650
    },
    {
      "epoch": 0.6420145190562614,
      "grad_norm": 2.203125,
      "learning_rate": 2.8827983075408842e-05,
      "loss": 0.1457,
      "step": 5660
    },
    {
      "epoch": 0.6431488203266787,
      "grad_norm": 3.625,
      "learning_rate": 2.882029932256807e-05,
      "loss": 0.1391,
      "step": 5670
    },
    {
      "epoch": 0.6442831215970962,
      "grad_norm": 3.28125,
      "learning_rate": 2.8812591495423413e-05,
      "loss": 0.1492,
      "step": 5680
    },
    {
      "epoch": 0.6454174228675136,
      "grad_norm": 2.828125,
      "learning_rate": 2.8804859607401533e-05,
      "loss": 0.1622,
      "step": 5690
    },
    {
      "epoch": 0.646551724137931,
      "grad_norm": 2.078125,
      "learning_rate": 2.8797103671971014e-05,
      "loss": 0.1436,
      "step": 5700
    },
    {
      "epoch": 0.6476860254083484,
      "grad_norm": 2.40625,
      "learning_rate": 2.8789323702642333e-05,
      "loss": 0.1476,
      "step": 5710
    },
    {
      "epoch": 0.6488203266787659,
      "grad_norm": 3.96875,
      "learning_rate": 2.878151971296781e-05,
      "loss": 0.1434,
      "step": 5720
    },
    {
      "epoch": 0.6499546279491834,
      "grad_norm": 3.515625,
      "learning_rate": 2.877369171654164e-05,
      "loss": 0.148,
      "step": 5730
    },
    {
      "epoch": 0.6510889292196007,
      "grad_norm": 2.59375,
      "learning_rate": 2.876583972699981e-05,
      "loss": 0.1433,
      "step": 5740
    },
    {
      "epoch": 0.6522232304900182,
      "grad_norm": 2.34375,
      "learning_rate": 2.8757963758020122e-05,
      "loss": 0.1501,
      "step": 5750
    },
    {
      "epoch": 0.6533575317604355,
      "grad_norm": 2.328125,
      "learning_rate": 2.8750063823322127e-05,
      "loss": 0.1441,
      "step": 5760
    },
    {
      "epoch": 0.654491833030853,
      "grad_norm": 2.75,
      "learning_rate": 2.874213993666714e-05,
      "loss": 0.1493,
      "step": 5770
    },
    {
      "epoch": 0.6556261343012704,
      "grad_norm": 3.09375,
      "learning_rate": 2.87341921118582e-05,
      "loss": 0.1527,
      "step": 5780
    },
    {
      "epoch": 0.6567604355716878,
      "grad_norm": 2.21875,
      "learning_rate": 2.8726220362740038e-05,
      "loss": 0.1307,
      "step": 5790
    },
    {
      "epoch": 0.6578947368421053,
      "grad_norm": 2.265625,
      "learning_rate": 2.8718224703199056e-05,
      "loss": 0.1534,
      "step": 5800
    },
    {
      "epoch": 0.6590290381125227,
      "grad_norm": 2.375,
      "learning_rate": 2.871020514716332e-05,
      "loss": 0.1371,
      "step": 5810
    },
    {
      "epoch": 0.6601633393829401,
      "grad_norm": 3.0,
      "learning_rate": 2.8702161708602506e-05,
      "loss": 0.1483,
      "step": 5820
    },
    {
      "epoch": 0.6612976406533575,
      "grad_norm": 2.765625,
      "learning_rate": 2.8694094401527912e-05,
      "loss": 0.1431,
      "step": 5830
    },
    {
      "epoch": 0.662431941923775,
      "grad_norm": 3.328125,
      "learning_rate": 2.8686003239992394e-05,
      "loss": 0.1519,
      "step": 5840
    },
    {
      "epoch": 0.6635662431941923,
      "grad_norm": 2.4375,
      "learning_rate": 2.8677888238090373e-05,
      "loss": 0.1468,
      "step": 5850
    },
    {
      "epoch": 0.6647005444646098,
      "grad_norm": 3.40625,
      "learning_rate": 2.8669749409957802e-05,
      "loss": 0.1508,
      "step": 5860
    },
    {
      "epoch": 0.6658348457350273,
      "grad_norm": 2.90625,
      "learning_rate": 2.8661586769772126e-05,
      "loss": 0.1492,
      "step": 5870
    },
    {
      "epoch": 0.6669691470054446,
      "grad_norm": 2.65625,
      "learning_rate": 2.8653400331752274e-05,
      "loss": 0.1418,
      "step": 5880
    },
    {
      "epoch": 0.6681034482758621,
      "grad_norm": 2.640625,
      "learning_rate": 2.8645190110158638e-05,
      "loss": 0.1381,
      "step": 5890
    },
    {
      "epoch": 0.6692377495462795,
      "grad_norm": 3.140625,
      "learning_rate": 2.863695611929303e-05,
      "loss": 0.1357,
      "step": 5900
    },
    {
      "epoch": 0.6703720508166969,
      "grad_norm": 2.984375,
      "learning_rate": 2.862869837349867e-05,
      "loss": 0.1538,
      "step": 5910
    },
    {
      "epoch": 0.6715063520871143,
      "grad_norm": 2.859375,
      "learning_rate": 2.862041688716016e-05,
      "loss": 0.1522,
      "step": 5920
    },
    {
      "epoch": 0.6726406533575318,
      "grad_norm": 2.984375,
      "learning_rate": 2.861211167470345e-05,
      "loss": 0.1523,
      "step": 5930
    },
    {
      "epoch": 0.6737749546279492,
      "grad_norm": 2.90625,
      "learning_rate": 2.8603782750595833e-05,
      "loss": 0.1451,
      "step": 5940
    },
    {
      "epoch": 0.6749092558983666,
      "grad_norm": 3.125,
      "learning_rate": 2.8595430129345898e-05,
      "loss": 0.1399,
      "step": 5950
    },
    {
      "epoch": 0.6760435571687841,
      "grad_norm": 2.703125,
      "learning_rate": 2.8587053825503505e-05,
      "loss": 0.1362,
      "step": 5960
    },
    {
      "epoch": 0.6771778584392014,
      "grad_norm": 2.046875,
      "learning_rate": 2.8578653853659787e-05,
      "loss": 0.1525,
      "step": 5970
    },
    {
      "epoch": 0.6783121597096189,
      "grad_norm": 2.328125,
      "learning_rate": 2.8570230228447085e-05,
      "loss": 0.1468,
      "step": 5980
    },
    {
      "epoch": 0.6794464609800362,
      "grad_norm": 1.984375,
      "learning_rate": 2.8561782964538964e-05,
      "loss": 0.1399,
      "step": 5990
    },
    {
      "epoch": 0.6805807622504537,
      "grad_norm": 2.296875,
      "learning_rate": 2.855331207665015e-05,
      "loss": 0.1454,
      "step": 6000
    },
    {
      "epoch": 0.6817150635208712,
      "grad_norm": 3.015625,
      "learning_rate": 2.8544817579536522e-05,
      "loss": 0.1481,
      "step": 6010
    },
    {
      "epoch": 0.6828493647912885,
      "grad_norm": 3.40625,
      "learning_rate": 2.85362994879951e-05,
      "loss": 0.1342,
      "step": 6020
    },
    {
      "epoch": 0.683983666061706,
      "grad_norm": 2.0625,
      "learning_rate": 2.8527757816863992e-05,
      "loss": 0.145,
      "step": 6030
    },
    {
      "epoch": 0.6851179673321234,
      "grad_norm": 2.4375,
      "learning_rate": 2.8519192581022377e-05,
      "loss": 0.1463,
      "step": 6040
    },
    {
      "epoch": 0.6862522686025408,
      "grad_norm": 2.296875,
      "learning_rate": 2.851060379539049e-05,
      "loss": 0.1587,
      "step": 6050
    },
    {
      "epoch": 0.6873865698729582,
      "grad_norm": 2.40625,
      "learning_rate": 2.850199147492959e-05,
      "loss": 0.1414,
      "step": 6060
    },
    {
      "epoch": 0.6885208711433757,
      "grad_norm": 2.546875,
      "learning_rate": 2.849335563464193e-05,
      "loss": 0.148,
      "step": 6070
    },
    {
      "epoch": 0.6896551724137931,
      "grad_norm": 1.90625,
      "learning_rate": 2.848469628957073e-05,
      "loss": 0.1536,
      "step": 6080
    },
    {
      "epoch": 0.6907894736842105,
      "grad_norm": 2.21875,
      "learning_rate": 2.8476013454800157e-05,
      "loss": 0.1517,
      "step": 6090
    },
    {
      "epoch": 0.691923774954628,
      "grad_norm": 2.1875,
      "learning_rate": 2.8467307145455296e-05,
      "loss": 0.1478,
      "step": 6100
    },
    {
      "epoch": 0.6930580762250453,
      "grad_norm": 1.859375,
      "learning_rate": 2.8458577376702127e-05,
      "loss": 0.1489,
      "step": 6110
    },
    {
      "epoch": 0.6941923774954628,
      "grad_norm": 2.875,
      "learning_rate": 2.8449824163747487e-05,
      "loss": 0.1412,
      "step": 6120
    },
    {
      "epoch": 0.6953266787658802,
      "grad_norm": 2.90625,
      "learning_rate": 2.844104752183906e-05,
      "loss": 0.1533,
      "step": 6130
    },
    {
      "epoch": 0.6964609800362976,
      "grad_norm": 4.28125,
      "learning_rate": 2.843224746626534e-05,
      "loss": 0.1423,
      "step": 6140
    },
    {
      "epoch": 0.6975952813067151,
      "grad_norm": 2.484375,
      "learning_rate": 2.84234240123556e-05,
      "loss": 0.1389,
      "step": 6150
    },
    {
      "epoch": 0.6987295825771325,
      "grad_norm": 2.671875,
      "learning_rate": 2.841457717547988e-05,
      "loss": 0.1415,
      "step": 6160
    },
    {
      "epoch": 0.6998638838475499,
      "grad_norm": 3.5625,
      "learning_rate": 2.8405706971048953e-05,
      "loss": 0.1395,
      "step": 6170
    },
    {
      "epoch": 0.7009981851179673,
      "grad_norm": 1.921875,
      "learning_rate": 2.8396813414514283e-05,
      "loss": 0.1342,
      "step": 6180
    },
    {
      "epoch": 0.7021324863883848,
      "grad_norm": 3.203125,
      "learning_rate": 2.8387896521368033e-05,
      "loss": 0.1454,
      "step": 6190
    },
    {
      "epoch": 0.7032667876588021,
      "grad_norm": 2.265625,
      "learning_rate": 2.8378956307143e-05,
      "loss": 0.1388,
      "step": 6200
    },
    {
      "epoch": 0.7044010889292196,
      "grad_norm": 2.5625,
      "learning_rate": 2.836999278741261e-05,
      "loss": 0.1429,
      "step": 6210
    },
    {
      "epoch": 0.7055353901996371,
      "grad_norm": 2.90625,
      "learning_rate": 2.8361005977790896e-05,
      "loss": 0.1515,
      "step": 6220
    },
    {
      "epoch": 0.7066696914700544,
      "grad_norm": 2.1875,
      "learning_rate": 2.835199589393245e-05,
      "loss": 0.1443,
      "step": 6230
    },
    {
      "epoch": 0.7078039927404719,
      "grad_norm": 3.0,
      "learning_rate": 2.8342962551532404e-05,
      "loss": 0.1458,
      "step": 6240
    },
    {
      "epoch": 0.7089382940108893,
      "grad_norm": 1.984375,
      "learning_rate": 2.8333905966326415e-05,
      "loss": 0.1483,
      "step": 6250
    },
    {
      "epoch": 0.7100725952813067,
      "grad_norm": 2.609375,
      "learning_rate": 2.8324826154090622e-05,
      "loss": 0.1413,
      "step": 6260
    },
    {
      "epoch": 0.7112068965517241,
      "grad_norm": 3.46875,
      "learning_rate": 2.8315723130641625e-05,
      "loss": 0.1392,
      "step": 6270
    },
    {
      "epoch": 0.7123411978221416,
      "grad_norm": 3.5,
      "learning_rate": 2.8306596911836467e-05,
      "loss": 0.1477,
      "step": 6280
    },
    {
      "epoch": 0.713475499092559,
      "grad_norm": 2.703125,
      "learning_rate": 2.8297447513572572e-05,
      "loss": 0.152,
      "step": 6290
    },
    {
      "epoch": 0.7146098003629764,
      "grad_norm": 2.40625,
      "learning_rate": 2.8288274951787765e-05,
      "loss": 0.151,
      "step": 6300
    },
    {
      "epoch": 0.7157441016333939,
      "grad_norm": 2.546875,
      "learning_rate": 2.8279079242460217e-05,
      "loss": 0.1421,
      "step": 6310
    },
    {
      "epoch": 0.7168784029038112,
      "grad_norm": 2.234375,
      "learning_rate": 2.8269860401608406e-05,
      "loss": 0.1412,
      "step": 6320
    },
    {
      "epoch": 0.7180127041742287,
      "grad_norm": 2.0625,
      "learning_rate": 2.8260618445291117e-05,
      "loss": 0.1459,
      "step": 6330
    },
    {
      "epoch": 0.719147005444646,
      "grad_norm": 2.671875,
      "learning_rate": 2.8251353389607404e-05,
      "loss": 0.1455,
      "step": 6340
    },
    {
      "epoch": 0.7202813067150635,
      "grad_norm": 2.09375,
      "learning_rate": 2.8242065250696548e-05,
      "loss": 0.156,
      "step": 6350
    },
    {
      "epoch": 0.721415607985481,
      "grad_norm": 3.28125,
      "learning_rate": 2.8232754044738043e-05,
      "loss": 0.1465,
      "step": 6360
    },
    {
      "epoch": 0.7225499092558983,
      "grad_norm": 2.3125,
      "learning_rate": 2.8223419787951573e-05,
      "loss": 0.1458,
      "step": 6370
    },
    {
      "epoch": 0.7236842105263158,
      "grad_norm": 1.9921875,
      "learning_rate": 2.8214062496596966e-05,
      "loss": 0.1545,
      "step": 6380
    },
    {
      "epoch": 0.7248185117967332,
      "grad_norm": 2.078125,
      "learning_rate": 2.8204682186974177e-05,
      "loss": 0.1422,
      "step": 6390
    },
    {
      "epoch": 0.7259528130671506,
      "grad_norm": 3.234375,
      "learning_rate": 2.819527887542326e-05,
      "loss": 0.1538,
      "step": 6400
    },
    {
      "epoch": 0.727087114337568,
      "grad_norm": 2.296875,
      "learning_rate": 2.8185852578324338e-05,
      "loss": 0.1389,
      "step": 6410
    },
    {
      "epoch": 0.7282214156079855,
      "grad_norm": 2.484375,
      "learning_rate": 2.817640331209757e-05,
      "loss": 0.1457,
      "step": 6420
    },
    {
      "epoch": 0.729355716878403,
      "grad_norm": 1.9921875,
      "learning_rate": 2.816693109320313e-05,
      "loss": 0.1479,
      "step": 6430
    },
    {
      "epoch": 0.7304900181488203,
      "grad_norm": 1.8828125,
      "learning_rate": 2.8157435938141168e-05,
      "loss": 0.1373,
      "step": 6440
    },
    {
      "epoch": 0.7316243194192378,
      "grad_norm": 2.984375,
      "learning_rate": 2.8147917863451802e-05,
      "loss": 0.1523,
      "step": 6450
    },
    {
      "epoch": 0.7327586206896551,
      "grad_norm": 2.90625,
      "learning_rate": 2.8138376885715057e-05,
      "loss": 0.1473,
      "step": 6460
    },
    {
      "epoch": 0.7338929219600726,
      "grad_norm": 3.34375,
      "learning_rate": 2.8128813021550862e-05,
      "loss": 0.1509,
      "step": 6470
    },
    {
      "epoch": 0.73502722323049,
      "grad_norm": 3.21875,
      "learning_rate": 2.8119226287619017e-05,
      "loss": 0.1533,
      "step": 6480
    },
    {
      "epoch": 0.7361615245009074,
      "grad_norm": 2.46875,
      "learning_rate": 2.810961670061916e-05,
      "loss": 0.1383,
      "step": 6490
    },
    {
      "epoch": 0.7372958257713249,
      "grad_norm": 2.015625,
      "learning_rate": 2.809998427729072e-05,
      "loss": 0.1466,
      "step": 6500
    },
    {
      "epoch": 0.7384301270417423,
      "grad_norm": 3.25,
      "learning_rate": 2.809032903441294e-05,
      "loss": 0.1509,
      "step": 6510
    },
    {
      "epoch": 0.7395644283121597,
      "grad_norm": 1.984375,
      "learning_rate": 2.8080650988804782e-05,
      "loss": 0.1404,
      "step": 6520
    },
    {
      "epoch": 0.7406987295825771,
      "grad_norm": 2.59375,
      "learning_rate": 2.807095015732494e-05,
      "loss": 0.1653,
      "step": 6530
    },
    {
      "epoch": 0.7418330308529946,
      "grad_norm": 1.8125,
      "learning_rate": 2.8061226556871805e-05,
      "loss": 0.1437,
      "step": 6540
    },
    {
      "epoch": 0.7429673321234119,
      "grad_norm": 2.390625,
      "learning_rate": 2.805148020438342e-05,
      "loss": 0.1458,
      "step": 6550
    },
    {
      "epoch": 0.7441016333938294,
      "grad_norm": 2.015625,
      "learning_rate": 2.8041711116837478e-05,
      "loss": 0.1417,
      "step": 6560
    },
    {
      "epoch": 0.7452359346642469,
      "grad_norm": 2.328125,
      "learning_rate": 2.8031919311251256e-05,
      "loss": 0.1418,
      "step": 6570
    },
    {
      "epoch": 0.7463702359346642,
      "grad_norm": 2.4375,
      "learning_rate": 2.8022104804681617e-05,
      "loss": 0.1489,
      "step": 6580
    },
    {
      "epoch": 0.7475045372050817,
      "grad_norm": 2.875,
      "learning_rate": 2.8012267614224965e-05,
      "loss": 0.1462,
      "step": 6590
    },
    {
      "epoch": 0.7486388384754991,
      "grad_norm": 2.265625,
      "learning_rate": 2.800240775701721e-05,
      "loss": 0.1604,
      "step": 6600
    },
    {
      "epoch": 0.7497731397459165,
      "grad_norm": 2.515625,
      "learning_rate": 2.799252525023377e-05,
      "loss": 0.1476,
      "step": 6610
    },
    {
      "epoch": 0.7509074410163339,
      "grad_norm": 3.359375,
      "learning_rate": 2.7982620111089486e-05,
      "loss": 0.1526,
      "step": 6620
    },
    {
      "epoch": 0.7520417422867514,
      "grad_norm": 2.546875,
      "learning_rate": 2.7972692356838652e-05,
      "loss": 0.1477,
      "step": 6630
    },
    {
      "epoch": 0.7531760435571688,
      "grad_norm": 1.8984375,
      "learning_rate": 2.7962742004774935e-05,
      "loss": 0.1442,
      "step": 6640
    },
    {
      "epoch": 0.7543103448275862,
      "grad_norm": 2.78125,
      "learning_rate": 2.795276907223137e-05,
      "loss": 0.1446,
      "step": 6650
    },
    {
      "epoch": 0.7554446460980037,
      "grad_norm": 3.078125,
      "learning_rate": 2.7942773576580347e-05,
      "loss": 0.1422,
      "step": 6660
    },
    {
      "epoch": 0.756578947368421,
      "grad_norm": 2.9375,
      "learning_rate": 2.793275553523353e-05,
      "loss": 0.1489,
      "step": 6670
    },
    {
      "epoch": 0.7577132486388385,
      "grad_norm": 2.890625,
      "learning_rate": 2.792271496564188e-05,
      "loss": 0.1477,
      "step": 6680
    },
    {
      "epoch": 0.7588475499092558,
      "grad_norm": 2.5625,
      "learning_rate": 2.791265188529558e-05,
      "loss": 0.1515,
      "step": 6690
    },
    {
      "epoch": 0.7599818511796733,
      "grad_norm": 2.65625,
      "learning_rate": 2.790256631172405e-05,
      "loss": 0.1277,
      "step": 6700
    },
    {
      "epoch": 0.7611161524500908,
      "grad_norm": 3.1875,
      "learning_rate": 2.789245826249587e-05,
      "loss": 0.1442,
      "step": 6710
    },
    {
      "epoch": 0.7622504537205081,
      "grad_norm": 3.140625,
      "learning_rate": 2.7882327755218787e-05,
      "loss": 0.1472,
      "step": 6720
    },
    {
      "epoch": 0.7633847549909256,
      "grad_norm": 2.03125,
      "learning_rate": 2.787217480753966e-05,
      "loss": 0.1364,
      "step": 6730
    },
    {
      "epoch": 0.764519056261343,
      "grad_norm": 2.328125,
      "learning_rate": 2.7861999437144433e-05,
      "loss": 0.143,
      "step": 6740
    },
    {
      "epoch": 0.7656533575317604,
      "grad_norm": 2.640625,
      "learning_rate": 2.785180166175813e-05,
      "loss": 0.1431,
      "step": 6750
    },
    {
      "epoch": 0.7667876588021778,
      "grad_norm": 1.8046875,
      "learning_rate": 2.784158149914479e-05,
      "loss": 0.1351,
      "step": 6760
    },
    {
      "epoch": 0.7679219600725953,
      "grad_norm": 3.171875,
      "learning_rate": 2.7831338967107436e-05,
      "loss": 0.1435,
      "step": 6770
    },
    {
      "epoch": 0.7690562613430127,
      "grad_norm": 1.78125,
      "learning_rate": 2.782107408348809e-05,
      "loss": 0.1498,
      "step": 6780
    },
    {
      "epoch": 0.7701905626134301,
      "grad_norm": 2.34375,
      "learning_rate": 2.7810786866167672e-05,
      "loss": 0.1455,
      "step": 6790
    },
    {
      "epoch": 0.7713248638838476,
      "grad_norm": 2.578125,
      "learning_rate": 2.7800477333066043e-05,
      "loss": 0.1488,
      "step": 6800
    },
    {
      "epoch": 0.7724591651542649,
      "grad_norm": 3.625,
      "learning_rate": 2.7790145502141907e-05,
      "loss": 0.1441,
      "step": 6810
    },
    {
      "epoch": 0.7735934664246824,
      "grad_norm": 3.125,
      "learning_rate": 2.777979139139283e-05,
      "loss": 0.1442,
      "step": 6820
    },
    {
      "epoch": 0.7747277676950998,
      "grad_norm": 1.921875,
      "learning_rate": 2.7769415018855173e-05,
      "loss": 0.1453,
      "step": 6830
    },
    {
      "epoch": 0.7758620689655172,
      "grad_norm": 2.875,
      "learning_rate": 2.7759016402604084e-05,
      "loss": 0.1354,
      "step": 6840
    },
    {
      "epoch": 0.7769963702359347,
      "grad_norm": 2.578125,
      "learning_rate": 2.7748595560753463e-05,
      "loss": 0.1459,
      "step": 6850
    },
    {
      "epoch": 0.7781306715063521,
      "grad_norm": 2.046875,
      "learning_rate": 2.7738152511455922e-05,
      "loss": 0.1478,
      "step": 6860
    },
    {
      "epoch": 0.7792649727767695,
      "grad_norm": 1.9765625,
      "learning_rate": 2.7727687272902747e-05,
      "loss": 0.1401,
      "step": 6870
    },
    {
      "epoch": 0.7803992740471869,
      "grad_norm": 2.703125,
      "learning_rate": 2.771719986332389e-05,
      "loss": 0.1398,
      "step": 6880
    },
    {
      "epoch": 0.7815335753176044,
      "grad_norm": 2.21875,
      "learning_rate": 2.7706690300987926e-05,
      "loss": 0.1389,
      "step": 6890
    },
    {
      "epoch": 0.7826678765880217,
      "grad_norm": 3.03125,
      "learning_rate": 2.7696158604202e-05,
      "loss": 0.1462,
      "step": 6900
    },
    {
      "epoch": 0.7838021778584392,
      "grad_norm": 2.09375,
      "learning_rate": 2.7685604791311832e-05,
      "loss": 0.1429,
      "step": 6910
    },
    {
      "epoch": 0.7849364791288567,
      "grad_norm": 3.0625,
      "learning_rate": 2.7675028880701675e-05,
      "loss": 0.1522,
      "step": 6920
    },
    {
      "epoch": 0.786070780399274,
      "grad_norm": 2.4375,
      "learning_rate": 2.766443089079424e-05,
      "loss": 0.1394,
      "step": 6930
    },
    {
      "epoch": 0.7872050816696915,
      "grad_norm": 2.890625,
      "learning_rate": 2.7653810840050736e-05,
      "loss": 0.1481,
      "step": 6940
    },
    {
      "epoch": 0.7883393829401089,
      "grad_norm": 3.40625,
      "learning_rate": 2.764316874697078e-05,
      "loss": 0.1453,
      "step": 6950
    },
    {
      "epoch": 0.7894736842105263,
      "grad_norm": 2.140625,
      "learning_rate": 2.7632504630092395e-05,
      "loss": 0.1376,
      "step": 6960
    },
    {
      "epoch": 0.7906079854809437,
      "grad_norm": 2.25,
      "learning_rate": 2.7621818507991964e-05,
      "loss": 0.1395,
      "step": 6970
    },
    {
      "epoch": 0.7917422867513612,
      "grad_norm": 2.546875,
      "learning_rate": 2.76111103992842e-05,
      "loss": 0.1401,
      "step": 6980
    },
    {
      "epoch": 0.7928765880217786,
      "grad_norm": 1.9375,
      "learning_rate": 2.7600380322622127e-05,
      "loss": 0.1417,
      "step": 6990
    },
    {
      "epoch": 0.794010889292196,
      "grad_norm": 3.078125,
      "learning_rate": 2.758962829669702e-05,
      "loss": 0.1417,
      "step": 7000
    },
    {
      "epoch": 0.7951451905626135,
      "grad_norm": 2.59375,
      "learning_rate": 2.7578854340238402e-05,
      "loss": 0.1468,
      "step": 7010
    },
    {
      "epoch": 0.7962794918330308,
      "grad_norm": 2.4375,
      "learning_rate": 2.7568058472013985e-05,
      "loss": 0.1462,
      "step": 7020
    },
    {
      "epoch": 0.7974137931034483,
      "grad_norm": 3.0,
      "learning_rate": 2.755724071082967e-05,
      "loss": 0.1498,
      "step": 7030
    },
    {
      "epoch": 0.7985480943738656,
      "grad_norm": 2.34375,
      "learning_rate": 2.754640107552947e-05,
      "loss": 0.1345,
      "step": 7040
    },
    {
      "epoch": 0.7996823956442831,
      "grad_norm": 1.6171875,
      "learning_rate": 2.7535539584995523e-05,
      "loss": 0.1368,
      "step": 7050
    },
    {
      "epoch": 0.8008166969147006,
      "grad_norm": 2.28125,
      "learning_rate": 2.7524656258148023e-05,
      "loss": 0.1401,
      "step": 7060
    },
    {
      "epoch": 0.801950998185118,
      "grad_norm": 2.171875,
      "learning_rate": 2.7513751113945217e-05,
      "loss": 0.1326,
      "step": 7070
    },
    {
      "epoch": 0.8030852994555354,
      "grad_norm": 1.8984375,
      "learning_rate": 2.750282417138333e-05,
      "loss": 0.1371,
      "step": 7080
    },
    {
      "epoch": 0.8042196007259528,
      "grad_norm": 1.859375,
      "learning_rate": 2.7491875449496592e-05,
      "loss": 0.1307,
      "step": 7090
    },
    {
      "epoch": 0.8053539019963702,
      "grad_norm": 2.265625,
      "learning_rate": 2.7480904967357146e-05,
      "loss": 0.1364,
      "step": 7100
    },
    {
      "epoch": 0.8064882032667876,
      "grad_norm": 2.78125,
      "learning_rate": 2.7469912744075055e-05,
      "loss": 0.1497,
      "step": 7110
    },
    {
      "epoch": 0.8076225045372051,
      "grad_norm": 2.375,
      "learning_rate": 2.745889879879825e-05,
      "loss": 0.1511,
      "step": 7120
    },
    {
      "epoch": 0.8087568058076225,
      "grad_norm": 2.53125,
      "learning_rate": 2.744786315071249e-05,
      "loss": 0.1467,
      "step": 7130
    },
    {
      "epoch": 0.8098911070780399,
      "grad_norm": 2.953125,
      "learning_rate": 2.7436805819041366e-05,
      "loss": 0.1407,
      "step": 7140
    },
    {
      "epoch": 0.8110254083484574,
      "grad_norm": 2.625,
      "learning_rate": 2.742572682304621e-05,
      "loss": 0.1465,
      "step": 7150
    },
    {
      "epoch": 0.8121597096188747,
      "grad_norm": 2.890625,
      "learning_rate": 2.741462618202611e-05,
      "loss": 0.1394,
      "step": 7160
    },
    {
      "epoch": 0.8132940108892922,
      "grad_norm": 2.703125,
      "learning_rate": 2.7403503915317854e-05,
      "loss": 0.148,
      "step": 7170
    },
    {
      "epoch": 0.8144283121597096,
      "grad_norm": 2.5,
      "learning_rate": 2.7392360042295902e-05,
      "loss": 0.1367,
      "step": 7180
    },
    {
      "epoch": 0.815562613430127,
      "grad_norm": 1.9453125,
      "learning_rate": 2.738119458237235e-05,
      "loss": 0.1427,
      "step": 7190
    },
    {
      "epoch": 0.8166969147005445,
      "grad_norm": 3.65625,
      "learning_rate": 2.7370007554996898e-05,
      "loss": 0.1427,
      "step": 7200
    },
    {
      "epoch": 0.8178312159709619,
      "grad_norm": 1.984375,
      "learning_rate": 2.735879897965682e-05,
      "loss": 0.1425,
      "step": 7210
    },
    {
      "epoch": 0.8189655172413793,
      "grad_norm": 2.1875,
      "learning_rate": 2.734756887587691e-05,
      "loss": 0.1248,
      "step": 7220
    },
    {
      "epoch": 0.8200998185117967,
      "grad_norm": 2.171875,
      "learning_rate": 2.7336317263219487e-05,
      "loss": 0.1459,
      "step": 7230
    },
    {
      "epoch": 0.8212341197822142,
      "grad_norm": 2.765625,
      "learning_rate": 2.732504416128432e-05,
      "loss": 0.146,
      "step": 7240
    },
    {
      "epoch": 0.8223684210526315,
      "grad_norm": 1.921875,
      "learning_rate": 2.7313749589708607e-05,
      "loss": 0.1421,
      "step": 7250
    },
    {
      "epoch": 0.823502722323049,
      "grad_norm": 2.78125,
      "learning_rate": 2.7302433568166972e-05,
      "loss": 0.1401,
      "step": 7260
    },
    {
      "epoch": 0.8246370235934665,
      "grad_norm": 2.609375,
      "learning_rate": 2.7291096116371372e-05,
      "loss": 0.1538,
      "step": 7270
    },
    {
      "epoch": 0.8257713248638838,
      "grad_norm": 1.8984375,
      "learning_rate": 2.7279737254071114e-05,
      "loss": 0.1331,
      "step": 7280
    },
    {
      "epoch": 0.8269056261343013,
      "grad_norm": 2.578125,
      "learning_rate": 2.7268357001052795e-05,
      "loss": 0.1491,
      "step": 7290
    },
    {
      "epoch": 0.8280399274047187,
      "grad_norm": 1.8359375,
      "learning_rate": 2.7256955377140275e-05,
      "loss": 0.142,
      "step": 7300
    },
    {
      "epoch": 0.8291742286751361,
      "grad_norm": 2.609375,
      "learning_rate": 2.724553240219464e-05,
      "loss": 0.1454,
      "step": 7310
    },
    {
      "epoch": 0.8303085299455535,
      "grad_norm": 2.71875,
      "learning_rate": 2.723408809611417e-05,
      "loss": 0.1432,
      "step": 7320
    },
    {
      "epoch": 0.831442831215971,
      "grad_norm": 3.078125,
      "learning_rate": 2.72226224788343e-05,
      "loss": 0.1461,
      "step": 7330
    },
    {
      "epoch": 0.8325771324863884,
      "grad_norm": 4.03125,
      "learning_rate": 2.7211135570327592e-05,
      "loss": 0.143,
      "step": 7340
    },
    {
      "epoch": 0.8337114337568058,
      "grad_norm": 2.40625,
      "learning_rate": 2.719962739060369e-05,
      "loss": 0.1635,
      "step": 7350
    },
    {
      "epoch": 0.8348457350272233,
      "grad_norm": 1.7265625,
      "learning_rate": 2.71880979597093e-05,
      "loss": 0.1393,
      "step": 7360
    },
    {
      "epoch": 0.8359800362976406,
      "grad_norm": 2.328125,
      "learning_rate": 2.717654729772814e-05,
      "loss": 0.1359,
      "step": 7370
    },
    {
      "epoch": 0.8371143375680581,
      "grad_norm": 2.640625,
      "learning_rate": 2.7164975424780914e-05,
      "loss": 0.1436,
      "step": 7380
    },
    {
      "epoch": 0.8382486388384754,
      "grad_norm": 1.9765625,
      "learning_rate": 2.715338236102527e-05,
      "loss": 0.1415,
      "step": 7390
    },
    {
      "epoch": 0.8393829401088929,
      "grad_norm": 2.609375,
      "learning_rate": 2.7141768126655777e-05,
      "loss": 0.1401,
      "step": 7400
    },
    {
      "epoch": 0.8405172413793104,
      "grad_norm": 2.96875,
      "learning_rate": 2.7130132741903887e-05,
      "loss": 0.1418,
      "step": 7410
    },
    {
      "epoch": 0.8416515426497277,
      "grad_norm": 1.921875,
      "learning_rate": 2.711847622703787e-05,
      "loss": 0.1589,
      "step": 7420
    },
    {
      "epoch": 0.8427858439201452,
      "grad_norm": 2.234375,
      "learning_rate": 2.7106798602362833e-05,
      "loss": 0.1465,
      "step": 7430
    },
    {
      "epoch": 0.8439201451905626,
      "grad_norm": 1.9140625,
      "learning_rate": 2.709509988822064e-05,
      "loss": 0.1378,
      "step": 7440
    },
    {
      "epoch": 0.84505444646098,
      "grad_norm": 1.875,
      "learning_rate": 2.7083380104989887e-05,
      "loss": 0.138,
      "step": 7450
    },
    {
      "epoch": 0.8461887477313974,
      "grad_norm": 2.40625,
      "learning_rate": 2.7071639273085892e-05,
      "loss": 0.1449,
      "step": 7460
    },
    {
      "epoch": 0.8473230490018149,
      "grad_norm": 3.109375,
      "learning_rate": 2.7059877412960616e-05,
      "loss": 0.1375,
      "step": 7470
    },
    {
      "epoch": 0.8484573502722323,
      "grad_norm": 1.96875,
      "learning_rate": 2.704809454510266e-05,
      "loss": 0.1462,
      "step": 7480
    },
    {
      "epoch": 0.8495916515426497,
      "grad_norm": 3.046875,
      "learning_rate": 2.7036290690037223e-05,
      "loss": 0.1371,
      "step": 7490
    },
    {
      "epoch": 0.8507259528130672,
      "grad_norm": 2.09375,
      "learning_rate": 2.7024465868326057e-05,
      "loss": 0.1327,
      "step": 7500
    },
    {
      "epoch": 0.8518602540834845,
      "grad_norm": 2.90625,
      "learning_rate": 2.7012620100567437e-05,
      "loss": 0.1525,
      "step": 7510
    },
    {
      "epoch": 0.852994555353902,
      "grad_norm": 2.34375,
      "learning_rate": 2.7000753407396134e-05,
      "loss": 0.1401,
      "step": 7520
    },
    {
      "epoch": 0.8541288566243194,
      "grad_norm": 3.203125,
      "learning_rate": 2.6988865809483353e-05,
      "loss": 0.1451,
      "step": 7530
    },
    {
      "epoch": 0.8552631578947368,
      "grad_norm": 2.078125,
      "learning_rate": 2.697695732753673e-05,
      "loss": 0.1294,
      "step": 7540
    },
    {
      "epoch": 0.8563974591651543,
      "grad_norm": 2.03125,
      "learning_rate": 2.6965027982300274e-05,
      "loss": 0.139,
      "step": 7550
    },
    {
      "epoch": 0.8575317604355717,
      "grad_norm": 2.40625,
      "learning_rate": 2.6953077794554335e-05,
      "loss": 0.1349,
      "step": 7560
    },
    {
      "epoch": 0.8586660617059891,
      "grad_norm": 2.125,
      "learning_rate": 2.6941106785115576e-05,
      "loss": 0.1454,
      "step": 7570
    },
    {
      "epoch": 0.8598003629764065,
      "grad_norm": 3.046875,
      "learning_rate": 2.6929114974836918e-05,
      "loss": 0.1443,
      "step": 7580
    },
    {
      "epoch": 0.860934664246824,
      "grad_norm": 2.484375,
      "learning_rate": 2.6917102384607526e-05,
      "loss": 0.1397,
      "step": 7590
    },
    {
      "epoch": 0.8620689655172413,
      "grad_norm": 2.484375,
      "learning_rate": 2.690506903535277e-05,
      "loss": 0.1517,
      "step": 7600
    },
    {
      "epoch": 0.8632032667876588,
      "grad_norm": 2.0625,
      "learning_rate": 2.6893014948034162e-05,
      "loss": 0.1355,
      "step": 7610
    },
    {
      "epoch": 0.8643375680580763,
      "grad_norm": 3.0625,
      "learning_rate": 2.6880940143649346e-05,
      "loss": 0.1473,
      "step": 7620
    },
    {
      "epoch": 0.8654718693284936,
      "grad_norm": 2.703125,
      "learning_rate": 2.6868844643232063e-05,
      "loss": 0.1478,
      "step": 7630
    },
    {
      "epoch": 0.8666061705989111,
      "grad_norm": 1.984375,
      "learning_rate": 2.6856728467852102e-05,
      "loss": 0.1491,
      "step": 7640
    },
    {
      "epoch": 0.8677404718693285,
      "grad_norm": 1.640625,
      "learning_rate": 2.684459163861526e-05,
      "loss": 0.1451,
      "step": 7650
    },
    {
      "epoch": 0.8688747731397459,
      "grad_norm": 2.328125,
      "learning_rate": 2.683243417666331e-05,
      "loss": 0.1391,
      "step": 7660
    },
    {
      "epoch": 0.8700090744101633,
      "grad_norm": 2.421875,
      "learning_rate": 2.6820256103173988e-05,
      "loss": 0.1413,
      "step": 7670
    },
    {
      "epoch": 0.8711433756805808,
      "grad_norm": 2.828125,
      "learning_rate": 2.6808057439360903e-05,
      "loss": 0.1417,
      "step": 7680
    },
    {
      "epoch": 0.8722776769509982,
      "grad_norm": 2.625,
      "learning_rate": 2.6795838206473557e-05,
      "loss": 0.1477,
      "step": 7690
    },
    {
      "epoch": 0.8734119782214156,
      "grad_norm": 1.953125,
      "learning_rate": 2.678359842579727e-05,
      "loss": 0.1438,
      "step": 7700
    },
    {
      "epoch": 0.8745462794918331,
      "grad_norm": 2.75,
      "learning_rate": 2.677133811865315e-05,
      "loss": 0.1361,
      "step": 7710
    },
    {
      "epoch": 0.8756805807622504,
      "grad_norm": 2.203125,
      "learning_rate": 2.6759057306398077e-05,
      "loss": 0.1394,
      "step": 7720
    },
    {
      "epoch": 0.8768148820326679,
      "grad_norm": 2.09375,
      "learning_rate": 2.6746756010424644e-05,
      "loss": 0.1382,
      "step": 7730
    },
    {
      "epoch": 0.8779491833030852,
      "grad_norm": 2.796875,
      "learning_rate": 2.673443425216111e-05,
      "loss": 0.1511,
      "step": 7740
    },
    {
      "epoch": 0.8790834845735027,
      "grad_norm": 2.640625,
      "learning_rate": 2.6722092053071406e-05,
      "loss": 0.14,
      "step": 7750
    },
    {
      "epoch": 0.8802177858439202,
      "grad_norm": 2.265625,
      "learning_rate": 2.6709729434655054e-05,
      "loss": 0.1501,
      "step": 7760
    },
    {
      "epoch": 0.8813520871143375,
      "grad_norm": 2.328125,
      "learning_rate": 2.669734641844714e-05,
      "loss": 0.1454,
      "step": 7770
    },
    {
      "epoch": 0.882486388384755,
      "grad_norm": 2.78125,
      "learning_rate": 2.668494302601829e-05,
      "loss": 0.1441,
      "step": 7780
    },
    {
      "epoch": 0.8836206896551724,
      "grad_norm": 2.96875,
      "learning_rate": 2.667251927897463e-05,
      "loss": 0.1409,
      "step": 7790
    },
    {
      "epoch": 0.8847549909255898,
      "grad_norm": 2.9375,
      "learning_rate": 2.666007519895773e-05,
      "loss": 0.14,
      "step": 7800
    },
    {
      "epoch": 0.8858892921960072,
      "grad_norm": 2.640625,
      "learning_rate": 2.6647610807644587e-05,
      "loss": 0.1403,
      "step": 7810
    },
    {
      "epoch": 0.8870235934664247,
      "grad_norm": 1.859375,
      "learning_rate": 2.6635126126747577e-05,
      "loss": 0.1498,
      "step": 7820
    },
    {
      "epoch": 0.8881578947368421,
      "grad_norm": 3.046875,
      "learning_rate": 2.6622621178014422e-05,
      "loss": 0.139,
      "step": 7830
    },
    {
      "epoch": 0.8892921960072595,
      "grad_norm": 3.25,
      "learning_rate": 2.6610095983228145e-05,
      "loss": 0.1463,
      "step": 7840
    },
    {
      "epoch": 0.890426497277677,
      "grad_norm": 1.71875,
      "learning_rate": 2.6597550564207048e-05,
      "loss": 0.1241,
      "step": 7850
    },
    {
      "epoch": 0.8915607985480943,
      "grad_norm": 2.4375,
      "learning_rate": 2.6584984942804642e-05,
      "loss": 0.1412,
      "step": 7860
    },
    {
      "epoch": 0.8926950998185118,
      "grad_norm": 1.9765625,
      "learning_rate": 2.657239914090965e-05,
      "loss": 0.1374,
      "step": 7870
    },
    {
      "epoch": 0.8938294010889292,
      "grad_norm": 2.078125,
      "learning_rate": 2.6559793180445936e-05,
      "loss": 0.145,
      "step": 7880
    },
    {
      "epoch": 0.8949637023593466,
      "grad_norm": 2.203125,
      "learning_rate": 2.6547167083372486e-05,
      "loss": 0.1402,
      "step": 7890
    },
    {
      "epoch": 0.8960980036297641,
      "grad_norm": 2.09375,
      "learning_rate": 2.6534520871683354e-05,
      "loss": 0.1426,
      "step": 7900
    },
    {
      "epoch": 0.8972323049001815,
      "grad_norm": 3.96875,
      "learning_rate": 2.6521854567407647e-05,
      "loss": 0.1439,
      "step": 7910
    },
    {
      "epoch": 0.8983666061705989,
      "grad_norm": 1.953125,
      "learning_rate": 2.6509168192609464e-05,
      "loss": 0.1447,
      "step": 7920
    },
    {
      "epoch": 0.8995009074410163,
      "grad_norm": 2.625,
      "learning_rate": 2.6496461769387862e-05,
      "loss": 0.144,
      "step": 7930
    },
    {
      "epoch": 0.9006352087114338,
      "grad_norm": 2.109375,
      "learning_rate": 2.6483735319876827e-05,
      "loss": 0.1482,
      "step": 7940
    },
    {
      "epoch": 0.9017695099818511,
      "grad_norm": 2.078125,
      "learning_rate": 2.6470988866245228e-05,
      "loss": 0.1417,
      "step": 7950
    },
    {
      "epoch": 0.9029038112522686,
      "grad_norm": 3.09375,
      "learning_rate": 2.6458222430696782e-05,
      "loss": 0.1501,
      "step": 7960
    },
    {
      "epoch": 0.9040381125226861,
      "grad_norm": 2.359375,
      "learning_rate": 2.6445436035470014e-05,
      "loss": 0.1418,
      "step": 7970
    },
    {
      "epoch": 0.9051724137931034,
      "grad_norm": 2.265625,
      "learning_rate": 2.6432629702838213e-05,
      "loss": 0.1517,
      "step": 7980
    },
    {
      "epoch": 0.9063067150635209,
      "grad_norm": 2.578125,
      "learning_rate": 2.6419803455109406e-05,
      "loss": 0.1488,
      "step": 7990
    },
    {
      "epoch": 0.9074410163339383,
      "grad_norm": 2.390625,
      "learning_rate": 2.6406957314626298e-05,
      "loss": 0.1348,
      "step": 8000
    },
    {
      "epoch": 0.9085753176043557,
      "grad_norm": 2.34375,
      "learning_rate": 2.639409130376626e-05,
      "loss": 0.1528,
      "step": 8010
    },
    {
      "epoch": 0.9097096188747731,
      "grad_norm": 2.734375,
      "learning_rate": 2.6381205444941272e-05,
      "loss": 0.1453,
      "step": 8020
    },
    {
      "epoch": 0.9108439201451906,
      "grad_norm": 3.140625,
      "learning_rate": 2.6368299760597884e-05,
      "loss": 0.1448,
      "step": 8030
    },
    {
      "epoch": 0.911978221415608,
      "grad_norm": 1.8984375,
      "learning_rate": 2.6355374273217186e-05,
      "loss": 0.1332,
      "step": 8040
    },
    {
      "epoch": 0.9131125226860254,
      "grad_norm": 2.984375,
      "learning_rate": 2.634242900531476e-05,
      "loss": 0.1447,
      "step": 8050
    },
    {
      "epoch": 0.9142468239564429,
      "grad_norm": 1.8828125,
      "learning_rate": 2.6329463979440648e-05,
      "loss": 0.1339,
      "step": 8060
    },
    {
      "epoch": 0.9153811252268602,
      "grad_norm": 3.0625,
      "learning_rate": 2.6316479218179313e-05,
      "loss": 0.1477,
      "step": 8070
    },
    {
      "epoch": 0.9165154264972777,
      "grad_norm": 2.5625,
      "learning_rate": 2.630347474414958e-05,
      "loss": 0.1368,
      "step": 8080
    },
    {
      "epoch": 0.917649727767695,
      "grad_norm": 2.328125,
      "learning_rate": 2.6290450580004633e-05,
      "loss": 0.1334,
      "step": 8090
    },
    {
      "epoch": 0.9187840290381125,
      "grad_norm": 3.09375,
      "learning_rate": 2.6277406748431946e-05,
      "loss": 0.1437,
      "step": 8100
    },
    {
      "epoch": 0.91991833030853,
      "grad_norm": 2.53125,
      "learning_rate": 2.6264343272153247e-05,
      "loss": 0.1466,
      "step": 8110
    },
    {
      "epoch": 0.9210526315789473,
      "grad_norm": 2.953125,
      "learning_rate": 2.62512601739245e-05,
      "loss": 0.141,
      "step": 8120
    },
    {
      "epoch": 0.9221869328493648,
      "grad_norm": 3.0625,
      "learning_rate": 2.6238157476535828e-05,
      "loss": 0.1484,
      "step": 8130
    },
    {
      "epoch": 0.9233212341197822,
      "grad_norm": 2.34375,
      "learning_rate": 2.6225035202811515e-05,
      "loss": 0.1469,
      "step": 8140
    },
    {
      "epoch": 0.9244555353901996,
      "grad_norm": 2.53125,
      "learning_rate": 2.6211893375609933e-05,
      "loss": 0.1369,
      "step": 8150
    },
    {
      "epoch": 0.925589836660617,
      "grad_norm": 2.09375,
      "learning_rate": 2.6198732017823523e-05,
      "loss": 0.1377,
      "step": 8160
    },
    {
      "epoch": 0.9267241379310345,
      "grad_norm": 3.171875,
      "learning_rate": 2.6185551152378745e-05,
      "loss": 0.139,
      "step": 8170
    },
    {
      "epoch": 0.927858439201452,
      "grad_norm": 2.453125,
      "learning_rate": 2.617235080223603e-05,
      "loss": 0.143,
      "step": 8180
    },
    {
      "epoch": 0.9289927404718693,
      "grad_norm": 1.8046875,
      "learning_rate": 2.6159130990389774e-05,
      "loss": 0.1486,
      "step": 8190
    },
    {
      "epoch": 0.9301270417422868,
      "grad_norm": 1.7109375,
      "learning_rate": 2.614589173986825e-05,
      "loss": 0.1434,
      "step": 8200
    },
    {
      "epoch": 0.9312613430127041,
      "grad_norm": 2.296875,
      "learning_rate": 2.6132633073733606e-05,
      "loss": 0.1388,
      "step": 8210
    },
    {
      "epoch": 0.9323956442831216,
      "grad_norm": 1.84375,
      "learning_rate": 2.6119355015081807e-05,
      "loss": 0.1378,
      "step": 8220
    },
    {
      "epoch": 0.933529945553539,
      "grad_norm": 2.078125,
      "learning_rate": 2.6106057587042604e-05,
      "loss": 0.1457,
      "step": 8230
    },
    {
      "epoch": 0.9346642468239564,
      "grad_norm": 2.140625,
      "learning_rate": 2.609274081277947e-05,
      "loss": 0.1434,
      "step": 8240
    },
    {
      "epoch": 0.9357985480943739,
      "grad_norm": 2.109375,
      "learning_rate": 2.6079404715489616e-05,
      "loss": 0.1452,
      "step": 8250
    },
    {
      "epoch": 0.9369328493647913,
      "grad_norm": 2.453125,
      "learning_rate": 2.6066049318403864e-05,
      "loss": 0.1486,
      "step": 8260
    },
    {
      "epoch": 0.9380671506352087,
      "grad_norm": 2.859375,
      "learning_rate": 2.6052674644786695e-05,
      "loss": 0.133,
      "step": 8270
    },
    {
      "epoch": 0.9392014519056261,
      "grad_norm": 2.125,
      "learning_rate": 2.603928071793615e-05,
      "loss": 0.1482,
      "step": 8280
    },
    {
      "epoch": 0.9403357531760436,
      "grad_norm": 2.4375,
      "learning_rate": 2.602586756118381e-05,
      "loss": 0.1464,
      "step": 8290
    },
    {
      "epoch": 0.941470054446461,
      "grad_norm": 2.640625,
      "learning_rate": 2.601243519789476e-05,
      "loss": 0.1483,
      "step": 8300
    },
    {
      "epoch": 0.9426043557168784,
      "grad_norm": 2.53125,
      "learning_rate": 2.599898365146754e-05,
      "loss": 0.1439,
      "step": 8310
    },
    {
      "epoch": 0.9437386569872959,
      "grad_norm": 2.359375,
      "learning_rate": 2.59855129453341e-05,
      "loss": 0.1422,
      "step": 8320
    },
    {
      "epoch": 0.9448729582577132,
      "grad_norm": 2.578125,
      "learning_rate": 2.5972023102959766e-05,
      "loss": 0.1491,
      "step": 8330
    },
    {
      "epoch": 0.9460072595281307,
      "grad_norm": 3.46875,
      "learning_rate": 2.5958514147843208e-05,
      "loss": 0.1433,
      "step": 8340
    },
    {
      "epoch": 0.947141560798548,
      "grad_norm": 2.265625,
      "learning_rate": 2.5944986103516376e-05,
      "loss": 0.1456,
      "step": 8350
    },
    {
      "epoch": 0.9482758620689655,
      "grad_norm": 1.796875,
      "learning_rate": 2.593143899354449e-05,
      "loss": 0.1367,
      "step": 8360
    },
    {
      "epoch": 0.949410163339383,
      "grad_norm": 2.296875,
      "learning_rate": 2.591787284152596e-05,
      "loss": 0.1474,
      "step": 8370
    },
    {
      "epoch": 0.9505444646098004,
      "grad_norm": 2.828125,
      "learning_rate": 2.5904287671092383e-05,
      "loss": 0.1442,
      "step": 8380
    },
    {
      "epoch": 0.9516787658802178,
      "grad_norm": 2.0625,
      "learning_rate": 2.589068350590848e-05,
      "loss": 0.1487,
      "step": 8390
    },
    {
      "epoch": 0.9528130671506352,
      "grad_norm": 2.6875,
      "learning_rate": 2.5877060369672057e-05,
      "loss": 0.1431,
      "step": 8400
    },
    {
      "epoch": 0.9539473684210527,
      "grad_norm": 2.0625,
      "learning_rate": 2.586341828611397e-05,
      "loss": 0.1399,
      "step": 8410
    },
    {
      "epoch": 0.95508166969147,
      "grad_norm": 2.671875,
      "learning_rate": 2.584975727899808e-05,
      "loss": 0.1495,
      "step": 8420
    },
    {
      "epoch": 0.9562159709618875,
      "grad_norm": 2.0625,
      "learning_rate": 2.5836077372121213e-05,
      "loss": 0.1373,
      "step": 8430
    },
    {
      "epoch": 0.957350272232305,
      "grad_norm": 3.484375,
      "learning_rate": 2.5822378589313112e-05,
      "loss": 0.1366,
      "step": 8440
    },
    {
      "epoch": 0.9584845735027223,
      "grad_norm": 2.75,
      "learning_rate": 2.5808660954436413e-05,
      "loss": 0.1381,
      "step": 8450
    },
    {
      "epoch": 0.9596188747731398,
      "grad_norm": 2.203125,
      "learning_rate": 2.579492449138657e-05,
      "loss": 0.1449,
      "step": 8460
    },
    {
      "epoch": 0.9607531760435571,
      "grad_norm": 2.1875,
      "learning_rate": 2.5781169224091858e-05,
      "loss": 0.1442,
      "step": 8470
    },
    {
      "epoch": 0.9618874773139746,
      "grad_norm": 1.53125,
      "learning_rate": 2.57673951765133e-05,
      "loss": 0.1361,
      "step": 8480
    },
    {
      "epoch": 0.963021778584392,
      "grad_norm": 2.421875,
      "learning_rate": 2.575360237264462e-05,
      "loss": 0.1525,
      "step": 8490
    },
    {
      "epoch": 0.9641560798548094,
      "grad_norm": 2.359375,
      "learning_rate": 2.573979083651224e-05,
      "loss": 0.1443,
      "step": 8500
    },
    {
      "epoch": 0.9652903811252269,
      "grad_norm": 2.6875,
      "learning_rate": 2.572596059217519e-05,
      "loss": 0.139,
      "step": 8510
    },
    {
      "epoch": 0.9664246823956443,
      "grad_norm": 1.9765625,
      "learning_rate": 2.57121116637251e-05,
      "loss": 0.1433,
      "step": 8520
    },
    {
      "epoch": 0.9675589836660617,
      "grad_norm": 2.234375,
      "learning_rate": 2.569824407528614e-05,
      "loss": 0.1398,
      "step": 8530
    },
    {
      "epoch": 0.9686932849364791,
      "grad_norm": 1.6484375,
      "learning_rate": 2.5684357851014998e-05,
      "loss": 0.1436,
      "step": 8540
    },
    {
      "epoch": 0.9698275862068966,
      "grad_norm": 2.5,
      "learning_rate": 2.567045301510081e-05,
      "loss": 0.1479,
      "step": 8550
    },
    {
      "epoch": 0.9709618874773139,
      "grad_norm": 3.296875,
      "learning_rate": 2.5656529591765142e-05,
      "loss": 0.1372,
      "step": 8560
    },
    {
      "epoch": 0.9720961887477314,
      "grad_norm": 2.125,
      "learning_rate": 2.5642587605261933e-05,
      "loss": 0.1278,
      "step": 8570
    },
    {
      "epoch": 0.9732304900181489,
      "grad_norm": 2.4375,
      "learning_rate": 2.562862707987746e-05,
      "loss": 0.1412,
      "step": 8580
    },
    {
      "epoch": 0.9743647912885662,
      "grad_norm": 2.328125,
      "learning_rate": 2.5614648039930294e-05,
      "loss": 0.1403,
      "step": 8590
    },
    {
      "epoch": 0.9754990925589837,
      "grad_norm": 1.921875,
      "learning_rate": 2.560065050977126e-05,
      "loss": 0.1418,
      "step": 8600
    },
    {
      "epoch": 0.9766333938294011,
      "grad_norm": 3.296875,
      "learning_rate": 2.5586634513783382e-05,
      "loss": 0.1471,
      "step": 8610
    },
    {
      "epoch": 0.9777676950998185,
      "grad_norm": 2.8125,
      "learning_rate": 2.557260007638187e-05,
      "loss": 0.1391,
      "step": 8620
    },
    {
      "epoch": 0.9789019963702359,
      "grad_norm": 3.0625,
      "learning_rate": 2.555854722201404e-05,
      "loss": 0.1447,
      "step": 8630
    },
    {
      "epoch": 0.9800362976406534,
      "grad_norm": 1.7578125,
      "learning_rate": 2.5544475975159292e-05,
      "loss": 0.1356,
      "step": 8640
    },
    {
      "epoch": 0.9811705989110708,
      "grad_norm": 1.734375,
      "learning_rate": 2.553038636032907e-05,
      "loss": 0.1309,
      "step": 8650
    },
    {
      "epoch": 0.9823049001814882,
      "grad_norm": 2.5,
      "learning_rate": 2.5516278402066823e-05,
      "loss": 0.1456,
      "step": 8660
    },
    {
      "epoch": 0.9834392014519057,
      "grad_norm": 2.25,
      "learning_rate": 2.5502152124947927e-05,
      "loss": 0.1422,
      "step": 8670
    },
    {
      "epoch": 0.984573502722323,
      "grad_norm": 2.59375,
      "learning_rate": 2.5488007553579697e-05,
      "loss": 0.1515,
      "step": 8680
    },
    {
      "epoch": 0.9857078039927405,
      "grad_norm": 2.0625,
      "learning_rate": 2.54738447126013e-05,
      "loss": 0.1359,
      "step": 8690
    },
    {
      "epoch": 0.9868421052631579,
      "grad_norm": 1.796875,
      "learning_rate": 2.5459663626683728e-05,
      "loss": 0.1488,
      "step": 8700
    },
    {
      "epoch": 0.9879764065335753,
      "grad_norm": 2.265625,
      "learning_rate": 2.5445464320529756e-05,
      "loss": 0.1399,
      "step": 8710
    },
    {
      "epoch": 0.9891107078039928,
      "grad_norm": 2.015625,
      "learning_rate": 2.543124681887391e-05,
      "loss": 0.1339,
      "step": 8720
    },
    {
      "epoch": 0.9902450090744102,
      "grad_norm": 1.9609375,
      "learning_rate": 2.541701114648239e-05,
      "loss": 0.1406,
      "step": 8730
    },
    {
      "epoch": 0.9913793103448276,
      "grad_norm": 3.421875,
      "learning_rate": 2.540275732815307e-05,
      "loss": 0.1445,
      "step": 8740
    },
    {
      "epoch": 0.992513611615245,
      "grad_norm": 1.9921875,
      "learning_rate": 2.5388485388715415e-05,
      "loss": 0.1467,
      "step": 8750
    },
    {
      "epoch": 0.9936479128856625,
      "grad_norm": 1.90625,
      "learning_rate": 2.5374195353030473e-05,
      "loss": 0.1421,
      "step": 8760
    },
    {
      "epoch": 0.9947822141560798,
      "grad_norm": 2.375,
      "learning_rate": 2.53598872459908e-05,
      "loss": 0.1482,
      "step": 8770
    },
    {
      "epoch": 0.9959165154264973,
      "grad_norm": 2.234375,
      "learning_rate": 2.534556109252044e-05,
      "loss": 0.1485,
      "step": 8780
    },
    {
      "epoch": 0.9970508166969148,
      "grad_norm": 2.109375,
      "learning_rate": 2.5331216917574864e-05,
      "loss": 0.1352,
      "step": 8790
    },
    {
      "epoch": 0.9981851179673321,
      "grad_norm": 2.21875,
      "learning_rate": 2.5316854746140955e-05,
      "loss": 0.1428,
      "step": 8800
    },
    {
      "epoch": 0.9993194192377496,
      "grad_norm": 1.9921875,
      "learning_rate": 2.5302474603236922e-05,
      "loss": 0.1413,
      "step": 8810
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9400585976965043,
      "eval_f1": 0.9377027870372914,
      "eval_loss": 0.1423337161540985,
      "eval_precision": 0.9371459150383395,
      "eval_recall": 0.9400585976965043,
      "eval_runtime": 122.6099,
      "eval_samples_per_second": 484.366,
      "eval_steps_per_second": 3.784,
      "step": 8816
    }
  ],
  "logging_steps": 10,
  "max_steps": 26448,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.962809651809747e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}