{
  "best_metric": 0.12240613251924515,
  "best_model_checkpoint": "/home1/datahome/villien/project_hub/DinoVdeau/models/DinoVdo-large-2025_01_27_45863-bs32_freeze/checkpoint-22113",
  "epoch": 91.0,
  "eval_steps": 500,
  "global_step": 24843,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2293193717277487,
      "eval_f1_macro": 0.5138028081483725,
      "eval_f1_micro": 0.7433001107419712,
      "eval_loss": 0.16775080561637878,
      "eval_runtime": 464.7451,
      "eval_samples_per_second": 6.165,
      "eval_steps_per_second": 0.194,
      "learning_rate": 0.001,
      "step": 273
    },
    {
      "epoch": 1.8315018315018317,
      "grad_norm": 0.32014167308807373,
      "learning_rate": 0.001,
      "loss": 0.2722,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.243630017452007,
      "eval_f1_macro": 0.5770554152365124,
      "eval_f1_micro": 0.7613556338028169,
      "eval_loss": 0.15366077423095703,
      "eval_runtime": 460.9126,
      "eval_samples_per_second": 6.216,
      "eval_steps_per_second": 0.195,
      "learning_rate": 0.001,
      "step": 546
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.24153577661431064,
      "eval_f1_macro": 0.6194117241441944,
      "eval_f1_micro": 0.7763416930582898,
      "eval_loss": 0.14831368625164032,
      "eval_runtime": 459.029,
      "eval_samples_per_second": 6.241,
      "eval_steps_per_second": 0.196,
      "learning_rate": 0.001,
      "step": 819
    },
    {
      "epoch": 3.663003663003663,
      "grad_norm": 0.2205018699169159,
      "learning_rate": 0.001,
      "loss": 0.169,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.2554973821989529,
      "eval_f1_macro": 0.6275864194841865,
      "eval_f1_micro": 0.7808419243986254,
      "eval_loss": 0.14640773832798004,
      "eval_runtime": 453.7375,
      "eval_samples_per_second": 6.314,
      "eval_steps_per_second": 0.198,
      "learning_rate": 0.001,
      "step": 1092
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.2520069808027923,
      "eval_f1_macro": 0.6420978846376312,
      "eval_f1_micro": 0.7787831782144088,
      "eval_loss": 0.14515382051467896,
      "eval_runtime": 459.4943,
      "eval_samples_per_second": 6.235,
      "eval_steps_per_second": 0.196,
      "learning_rate": 0.001,
      "step": 1365
    },
    {
      "epoch": 5.4945054945054945,
      "grad_norm": 0.19979850947856903,
      "learning_rate": 0.001,
      "loss": 0.1612,
      "step": 1500
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.2579406631762653,
      "eval_f1_macro": 0.614677201484495,
      "eval_f1_micro": 0.7802330599913682,
      "eval_loss": 0.14404040575027466,
      "eval_runtime": 455.8188,
      "eval_samples_per_second": 6.285,
      "eval_steps_per_second": 0.197,
      "learning_rate": 0.001,
      "step": 1638
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.25026178010471206,
      "eval_f1_macro": 0.6141169392172392,
      "eval_f1_micro": 0.7787126995142263,
      "eval_loss": 0.14518576860427856,
      "eval_runtime": 463.2529,
      "eval_samples_per_second": 6.185,
      "eval_steps_per_second": 0.194,
      "learning_rate": 0.001,
      "step": 1911
    },
    {
      "epoch": 7.326007326007326,
      "grad_norm": 0.21145381033420563,
      "learning_rate": 0.001,
      "loss": 0.1594,
      "step": 2000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.2534031413612565,
      "eval_f1_macro": 0.6193231969902891,
      "eval_f1_micro": 0.77764832793959,
      "eval_loss": 0.14463570713996887,
      "eval_runtime": 461.4649,
      "eval_samples_per_second": 6.208,
      "eval_steps_per_second": 0.195,
      "learning_rate": 0.001,
      "step": 2184
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.24537521815008725,
      "eval_f1_macro": 0.6363157565562336,
      "eval_f1_micro": 0.7813486742506164,
      "eval_loss": 0.14786836504936218,
      "eval_runtime": 457.2698,
      "eval_samples_per_second": 6.265,
      "eval_steps_per_second": 0.197,
      "learning_rate": 0.001,
      "step": 2457
    },
    {
      "epoch": 9.157509157509157,
      "grad_norm": 0.1889927238225937,
      "learning_rate": 0.001,
      "loss": 0.1567,
      "step": 2500
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.2607329842931937,
      "eval_f1_macro": 0.6366285266954459,
      "eval_f1_micro": 0.7865724075118186,
      "eval_loss": 0.14251072704792023,
      "eval_runtime": 460.3923,
      "eval_samples_per_second": 6.223,
      "eval_steps_per_second": 0.195,
      "learning_rate": 0.001,
      "step": 2730
    },
    {
      "epoch": 10.989010989010989,
      "grad_norm": 0.1742856353521347,
      "learning_rate": 0.001,
      "loss": 0.1559,
      "step": 3000
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.2617801047120419,
      "eval_f1_macro": 0.6565839087323015,
      "eval_f1_micro": 0.7907930529399456,
      "eval_loss": 0.14535894989967346,
      "eval_runtime": 453.3601,
      "eval_samples_per_second": 6.319,
      "eval_steps_per_second": 0.199,
      "learning_rate": 0.001,
      "step": 3003
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.2600349040139616,
      "eval_f1_macro": 0.6438532757657272,
      "eval_f1_micro": 0.7895364689843217,
      "eval_loss": 0.14042578637599945,
      "eval_runtime": 461.5251,
      "eval_samples_per_second": 6.208,
      "eval_steps_per_second": 0.195,
      "learning_rate": 0.001,
      "step": 3276
    },
    {
      "epoch": 12.820512820512821,
      "grad_norm": 0.16230766475200653,
      "learning_rate": 0.001,
      "loss": 0.1549,
      "step": 3500
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.2530541012216405,
      "eval_f1_macro": 0.6490185335679685,
      "eval_f1_micro": 0.7883035906256719,
      "eval_loss": 0.1413952261209488,
      "eval_runtime": 459.1407,
      "eval_samples_per_second": 6.24,
      "eval_steps_per_second": 0.196,
      "learning_rate": 0.001,
      "step": 3549
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.2649214659685864,
      "eval_f1_macro": 0.6347742331888712,
      "eval_f1_micro": 0.7893909125685383,
      "eval_loss": 0.14056158065795898,
      "eval_runtime": 459.115,
      "eval_samples_per_second": 6.24,
      "eval_steps_per_second": 0.196,
      "learning_rate": 0.001,
      "step": 3822
    },
    {
      "epoch": 14.652014652014651,
      "grad_norm": 0.5134446024894714,
      "learning_rate": 0.001,
      "loss": 0.155,
      "step": 4000
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.2631762652705061,
      "eval_f1_macro": 0.6469635503488436,
      "eval_f1_micro": 0.7906051299796704,
      "eval_loss": 0.13873133063316345,
      "eval_runtime": 461.6498,
      "eval_samples_per_second": 6.206,
      "eval_steps_per_second": 0.195,
      "learning_rate": 0.001,
      "step": 4095
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.2603839441535777,
      "eval_f1_macro": 0.6331356980695925,
      "eval_f1_micro": 0.7857704059362723,
      "eval_loss": 0.14008578658103943,
      "eval_runtime": 459.7885,
      "eval_samples_per_second": 6.231,
      "eval_steps_per_second": 0.196,
      "learning_rate": 0.001,
      "step": 4368
    },
    {
      "epoch": 16.483516483516482,
      "grad_norm": 0.159025177359581,
      "learning_rate": 0.001,
      "loss": 0.1531,
      "step": 4500
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.2520069808027923,
      "eval_f1_macro": 0.665107882553296,
      "eval_f1_micro": 0.7954926273458445,
      "eval_loss": 0.13811993598937988,
      "eval_runtime": 460.3054,
      "eval_samples_per_second": 6.224,
      "eval_steps_per_second": 0.196,
      "learning_rate": 0.001,
      "step": 4641
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.27015706806282724,
      "eval_f1_macro": 0.6498362138588303,
      "eval_f1_micro": 0.791355160264673,
      "eval_loss": 0.13870170712471008,
      "eval_runtime": 463.8864,
      "eval_samples_per_second": 6.176,
      "eval_steps_per_second": 0.194,
      "learning_rate": 0.001,
      "step": 4914
    },
    {
      "epoch": 18.315018315018314,
      "grad_norm": 0.14721287786960602,
      "learning_rate": 0.001,
      "loss": 0.1549,
      "step": 5000
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.2631762652705061,
      "eval_f1_macro": 0.6356042954083428,
      "eval_f1_micro": 0.794037710208935,
      "eval_loss": 0.1373777985572815,
      "eval_runtime": 458.2597,
      "eval_samples_per_second": 6.252,
      "eval_steps_per_second": 0.196,
      "learning_rate": 0.001,
      "step": 5187
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.25514834205933684,
      "eval_f1_macro": 0.6393109338369551,
      "eval_f1_micro": 0.7850155722244155,
      "eval_loss": 0.13864819705486298,
      "eval_runtime": 469.7679,
      "eval_samples_per_second": 6.099,
      "eval_steps_per_second": 0.192,
      "learning_rate": 0.001,
      "step": 5460
    },
    {
      "epoch": 20.146520146520146,
      "grad_norm": 0.29727715253829956,
      "learning_rate": 0.001,
      "loss": 0.1524,
      "step": 5500
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.26457242582897034,
      "eval_f1_macro": 0.6496408648134395,
      "eval_f1_micro": 0.7943311144513635,
      "eval_loss": 0.13566707074642181,
      "eval_runtime": 461.0026,
      "eval_samples_per_second": 6.215,
      "eval_steps_per_second": 0.195,
      "learning_rate": 0.001,
      "step": 5733
    },
    {
      "epoch": 21.978021978021978,
      "grad_norm": 0.16196180880069733,
      "learning_rate": 0.001,
      "loss": 0.1519,
      "step": 6000
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.27225130890052357,
      "eval_f1_macro": 0.6400060646722534,
      "eval_f1_micro": 0.7972140143520473,
      "eval_loss": 0.1371580958366394,
      "eval_runtime": 454.7092,
      "eval_samples_per_second": 6.301,
      "eval_steps_per_second": 0.198,
      "learning_rate": 0.001,
      "step": 6006
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.2603839441535777,
      "eval_f1_macro": 0.6595092645083339,
      "eval_f1_micro": 0.7938437661665805,
      "eval_loss": 0.13614478707313538,
      "eval_runtime": 452.3843,
      "eval_samples_per_second": 6.333,
      "eval_steps_per_second": 0.199,
      "learning_rate": 0.001,
      "step": 6279
    },
    {
      "epoch": 23.80952380952381,
      "grad_norm": 0.19081830978393555,
      "learning_rate": 0.001,
      "loss": 0.1528,
      "step": 6500
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.2649214659685864,
      "eval_f1_macro": 0.6417599876035757,
      "eval_f1_micro": 0.7953625165167725,
      "eval_loss": 0.1362675279378891,
      "eval_runtime": 458.9922,
      "eval_samples_per_second": 6.242,
      "eval_steps_per_second": 0.196,
      "learning_rate": 0.001,
      "step": 6552
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.2732984293193717,
      "eval_f1_macro": 0.6499701126111167,
      "eval_f1_micro": 0.7963302752293578,
      "eval_loss": 0.1358671337366104,
      "eval_runtime": 456.4797,
      "eval_samples_per_second": 6.276,
      "eval_steps_per_second": 0.197,
      "learning_rate": 0.001,
      "step": 6825
    },
    {
      "epoch": 25.641025641025642,
      "grad_norm": 0.1495918333530426,
      "learning_rate": 0.001,
      "loss": 0.1524,
      "step": 7000
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.26910994764397905,
      "eval_f1_macro": 0.6554940295679257,
      "eval_f1_micro": 0.7983850403739906,
      "eval_loss": 0.13484793901443481,
      "eval_runtime": 468.4399,
      "eval_samples_per_second": 6.116,
      "eval_steps_per_second": 0.192,
      "learning_rate": 0.001,
      "step": 7098
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.268760907504363,
      "eval_f1_macro": 0.6535438446852289,
      "eval_f1_micro": 0.7944040481353914,
      "eval_loss": 0.13669784367084503,
      "eval_runtime": 466.0343,
      "eval_samples_per_second": 6.148,
      "eval_steps_per_second": 0.193,
      "learning_rate": 0.001,
      "step": 7371
    },
    {
      "epoch": 27.47252747252747,
      "grad_norm": 0.16602398455142975,
      "learning_rate": 0.001,
      "loss": 0.1521,
      "step": 7500
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.26771378708551485,
      "eval_f1_macro": 0.6398313564596081,
      "eval_f1_micro": 0.7922701137002292,
      "eval_loss": 0.13569706678390503,
      "eval_runtime": 460.9998,
      "eval_samples_per_second": 6.215,
      "eval_steps_per_second": 0.195,
      "learning_rate": 0.001,
      "step": 7644
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.2638743455497382,
      "eval_f1_macro": 0.6636571549498853,
      "eval_f1_micro": 0.7924252834077636,
      "eval_loss": 0.14052562415599823,
      "eval_runtime": 467.3084,
      "eval_samples_per_second": 6.131,
      "eval_steps_per_second": 0.193,
      "learning_rate": 0.001,
      "step": 7917
    },
    {
      "epoch": 29.304029304029303,
      "grad_norm": 0.1555439680814743,
      "learning_rate": 0.001,
      "loss": 0.1525,
      "step": 8000
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.27225130890052357,
      "eval_f1_macro": 0.6405329741259155,
      "eval_f1_micro": 0.7875164835164835,
      "eval_loss": 0.13725879788398743,
      "eval_runtime": 459.9825,
      "eval_samples_per_second": 6.228,
      "eval_steps_per_second": 0.196,
      "learning_rate": 0.001,
      "step": 8190
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.2719022687609075,
      "eval_f1_macro": 0.6562409583519914,
      "eval_f1_micro": 0.7986043969902056,
      "eval_loss": 0.13544805347919464,
      "eval_runtime": 461.8222,
      "eval_samples_per_second": 6.204,
      "eval_steps_per_second": 0.195,
      "learning_rate": 0.001,
      "step": 8463
    },
    {
      "epoch": 31.135531135531135,
      "grad_norm": 0.20161172747612,
      "learning_rate": 0.001,
      "loss": 0.1522,
      "step": 8500
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.2649214659685864,
      "eval_f1_macro": 0.6463119642593244,
      "eval_f1_micro": 0.793030759303076,
      "eval_loss": 0.13693773746490479,
      "eval_runtime": 462.8659,
      "eval_samples_per_second": 6.19,
      "eval_steps_per_second": 0.194,
      "learning_rate": 0.001,
      "step": 8736
    },
    {
      "epoch": 32.967032967032964,
      "grad_norm": 0.5906699299812317,
      "learning_rate": 0.0001,
      "loss": 0.1486,
      "step": 9000
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.27469458987783596,
      "eval_f1_macro": 0.6660418237263234,
      "eval_f1_micro": 0.8014840718154292,
      "eval_loss": 0.13195939362049103,
      "eval_runtime": 465.1048,
      "eval_samples_per_second": 6.16,
      "eval_steps_per_second": 0.194,
      "learning_rate": 0.0001,
      "step": 9009
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.28342059336823733,
      "eval_f1_macro": 0.6755989146580753,
      "eval_f1_micro": 0.8042847388187094,
      "eval_loss": 0.1299898624420166,
      "eval_runtime": 462.0703,
      "eval_samples_per_second": 6.2,
      "eval_steps_per_second": 0.195,
      "learning_rate": 0.0001,
      "step": 9282
    },
    {
      "epoch": 34.798534798534796,
      "grad_norm": 0.1343473345041275,
      "learning_rate": 0.0001,
      "loss": 0.1419,
      "step": 9500
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.28097731239092494,
      "eval_f1_macro": 0.6772201041047545,
      "eval_f1_micro": 0.8066335550130482,
      "eval_loss": 0.12946264445781708,
      "eval_runtime": 463.7568,
      "eval_samples_per_second": 6.178,
      "eval_steps_per_second": 0.194,
      "learning_rate": 0.0001,
      "step": 9555
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.28167539267015707,
      "eval_f1_macro": 0.6794992142877868,
      "eval_f1_micro": 0.8046182685753238,
      "eval_loss": 0.13086125254631042,
      "eval_runtime": 457.8267,
      "eval_samples_per_second": 6.258,
      "eval_steps_per_second": 0.197,
      "learning_rate": 0.0001,
      "step": 9828
    },
    {
      "epoch": 36.63003663003663,
      "grad_norm": 0.16144825518131256,
      "learning_rate": 0.0001,
      "loss": 0.1396,
      "step": 10000
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.2823734729493892,
      "eval_f1_macro": 0.6791512030316924,
      "eval_f1_micro": 0.805367401183229,
      "eval_loss": 0.1278763711452484,
      "eval_runtime": 471.6542,
      "eval_samples_per_second": 6.074,
      "eval_steps_per_second": 0.191,
      "learning_rate": 0.0001,
      "step": 10101
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.2848167539267016,
      "eval_f1_macro": 0.6814340581384846,
      "eval_f1_micro": 0.8077836187892149,
      "eval_loss": 0.12904110550880432,
      "eval_runtime": 470.311,
      "eval_samples_per_second": 6.092,
      "eval_steps_per_second": 0.191,
      "learning_rate": 0.0001,
      "step": 10374
    },
    {
      "epoch": 38.46153846153846,
      "grad_norm": 0.16166280210018158,
      "learning_rate": 0.0001,
      "loss": 0.1366,
      "step": 10500
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.293891797556719,
      "eval_f1_macro": 0.6832531367067615,
      "eval_f1_micro": 0.8115661433562018,
      "eval_loss": 0.12716704607009888,
      "eval_runtime": 462.4995,
      "eval_samples_per_second": 6.195,
      "eval_steps_per_second": 0.195,
      "learning_rate": 0.0001,
      "step": 10647
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.2907504363001745,
      "eval_f1_macro": 0.6878954476405744,
      "eval_f1_micro": 0.8115966316142277,
      "eval_loss": 0.1293308585882187,
      "eval_runtime": 477.5548,
      "eval_samples_per_second": 5.999,
      "eval_steps_per_second": 0.188,
      "learning_rate": 0.0001,
      "step": 10920
    },
    {
      "epoch": 40.29304029304029,
      "grad_norm": 0.23697003722190857,
      "learning_rate": 0.0001,
      "loss": 0.1361,
      "step": 11000
    },
    {
      "epoch": 41.0,
      "eval_accuracy": 0.2917975567190227,
      "eval_f1_macro": 0.6863712244346338,
      "eval_f1_micro": 0.8088835942818243,
      "eval_loss": 0.12695887684822083,
      "eval_runtime": 460.0604,
      "eval_samples_per_second": 6.227,
      "eval_steps_per_second": 0.196,
      "learning_rate": 0.0001,
      "step": 11193
    },
    {
      "epoch": 42.0,
      "eval_accuracy": 0.2914485165794066,
      "eval_f1_macro": 0.6837014102587237,
      "eval_f1_micro": 0.8109225964379168,
      "eval_loss": 0.12624548375606537,
      "eval_runtime": 468.1922,
      "eval_samples_per_second": 6.119,
      "eval_steps_per_second": 0.192,
      "learning_rate": 0.0001,
      "step": 11466
    },
    {
      "epoch": 42.124542124542124,
      "grad_norm": 0.23925267159938812,
      "learning_rate": 0.0001,
      "loss": 0.135,
      "step": 11500
    },
    {
      "epoch": 43.0,
      "eval_accuracy": 0.2949389179755672,
      "eval_f1_macro": 0.6983715804658985,
      "eval_f1_micro": 0.8122734552811262,
      "eval_loss": 0.1261172592639923,
      "eval_runtime": 464.3762,
      "eval_samples_per_second": 6.17,
      "eval_steps_per_second": 0.194,
      "learning_rate": 0.0001,
      "step": 11739
    },
    {
      "epoch": 43.956043956043956,
      "grad_norm": 0.7063220143318176,
      "learning_rate": 0.0001,
      "loss": 0.134,
      "step": 12000
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.29354275741710295,
      "eval_f1_macro": 0.6833912042783733,
      "eval_f1_micro": 0.8106054209237524,
      "eval_loss": 0.12830273807048798,
      "eval_runtime": 456.3209,
      "eval_samples_per_second": 6.278,
      "eval_steps_per_second": 0.197,
      "learning_rate": 0.0001,
      "step": 12012
    },
    {
      "epoch": 45.0,
      "eval_accuracy": 0.2931937172774869,
      "eval_f1_macro": 0.7009703620992983,
      "eval_f1_micro": 0.8113048605846179,
      "eval_loss": 0.12624593079090118,
      "eval_runtime": 455.5361,
      "eval_samples_per_second": 6.289,
      "eval_steps_per_second": 0.198,
      "learning_rate": 0.0001,
      "step": 12285
    },
    {
      "epoch": 45.78754578754579,
      "grad_norm": 0.22236338257789612,
      "learning_rate": 0.0001,
      "loss": 0.1331,
      "step": 12500
    },
    {
      "epoch": 46.0,
      "eval_accuracy": 0.29598603839441534,
      "eval_f1_macro": 0.6963625779503319,
      "eval_f1_micro": 0.8147029204431017,
      "eval_loss": 0.12462077289819717,
      "eval_runtime": 465.158,
      "eval_samples_per_second": 6.159,
      "eval_steps_per_second": 0.193,
      "learning_rate": 0.0001,
      "step": 12558
    },
    {
      "epoch": 47.0,
      "eval_accuracy": 0.2987783595113438,
      "eval_f1_macro": 0.6923480831102572,
      "eval_f1_micro": 0.8126030873334743,
      "eval_loss": 0.12529432773590088,
      "eval_runtime": 457.756,
      "eval_samples_per_second": 6.259,
      "eval_steps_per_second": 0.197,
      "learning_rate": 0.0001,
      "step": 12831
    },
    {
      "epoch": 47.61904761904762,
      "grad_norm": 0.20239435136318207,
      "learning_rate": 0.0001,
      "loss": 0.1325,
      "step": 13000
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.29773123909249566,
      "eval_f1_macro": 0.6954224588663396,
      "eval_f1_micro": 0.8133141600406023,
      "eval_loss": 0.12631145119667053,
      "eval_runtime": 454.7373,
      "eval_samples_per_second": 6.3,
      "eval_steps_per_second": 0.198,
      "learning_rate": 0.0001,
      "step": 13104
    },
    {
      "epoch": 49.0,
      "eval_accuracy": 0.3036649214659686,
      "eval_f1_macro": 0.6951680615776381,
      "eval_f1_micro": 0.8157960928984657,
      "eval_loss": 0.1252526491880417,
      "eval_runtime": 455.0418,
      "eval_samples_per_second": 6.296,
      "eval_steps_per_second": 0.198,
      "learning_rate": 0.0001,
      "step": 13377
    },
    {
      "epoch": 49.45054945054945,
      "grad_norm": 0.2540508806705475,
      "learning_rate": 0.0001,
      "loss": 0.1312,
      "step": 13500
    },
    {
      "epoch": 50.0,
      "eval_accuracy": 0.30052356020942406,
      "eval_f1_macro": 0.7008159807881353,
      "eval_f1_micro": 0.8136319704164391,
      "eval_loss": 0.12632089853286743,
      "eval_runtime": 450.3202,
      "eval_samples_per_second": 6.362,
      "eval_steps_per_second": 0.2,
      "learning_rate": 0.0001,
      "step": 13650
    },
    {
      "epoch": 51.0,
      "eval_accuracy": 0.3008726003490401,
      "eval_f1_macro": 0.7019225754370848,
      "eval_f1_micro": 0.8157828600150792,
      "eval_loss": 0.1246422603726387,
      "eval_runtime": 455.9576,
      "eval_samples_per_second": 6.283,
      "eval_steps_per_second": 0.197,
      "learning_rate": 0.0001,
      "step": 13923
    },
    {
      "epoch": 51.282051282051285,
      "grad_norm": 0.2041085660457611,
      "learning_rate": 0.0001,
      "loss": 0.1301,
      "step": 14000
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.29109947643979056,
      "eval_f1_macro": 0.6948529573216252,
      "eval_f1_micro": 0.8091726695366779,
      "eval_loss": 0.12534211575984955,
      "eval_runtime": 453.0007,
      "eval_samples_per_second": 6.324,
      "eval_steps_per_second": 0.199,
      "learning_rate": 0.0001,
      "step": 14196
    },
    {
      "epoch": 53.0,
      "eval_accuracy": 0.3022687609075044,
      "eval_f1_macro": 0.7018961268425457,
      "eval_f1_micro": 0.8154071593050032,
      "eval_loss": 0.12436465919017792,
      "eval_runtime": 452.0514,
      "eval_samples_per_second": 6.338,
      "eval_steps_per_second": 0.199,
      "learning_rate": 1e-05,
      "step": 14469
    },
    {
      "epoch": 53.11355311355312,
      "grad_norm": 0.19839347898960114,
      "learning_rate": 1e-05,
      "loss": 0.1306,
      "step": 14500
    },
    {
      "epoch": 54.0,
      "eval_accuracy": 0.3008726003490401,
      "eval_f1_macro": 0.7040300990277949,
      "eval_f1_micro": 0.8154115879111705,
      "eval_loss": 0.12488020956516266,
      "eval_runtime": 453.4122,
      "eval_samples_per_second": 6.319,
      "eval_steps_per_second": 0.198,
      "learning_rate": 1e-05,
      "step": 14742
    },
    {
      "epoch": 54.94505494505494,
      "grad_norm": 0.2790488600730896,
      "learning_rate": 1e-05,
      "loss": 0.1282,
      "step": 15000
    },
    {
      "epoch": 55.0,
      "eval_accuracy": 0.30052356020942406,
      "eval_f1_macro": 0.6998211272872638,
      "eval_f1_micro": 0.814350990015231,
      "eval_loss": 0.12366042286157608,
      "eval_runtime": 451.3538,
      "eval_samples_per_second": 6.348,
      "eval_steps_per_second": 0.199,
      "learning_rate": 1e-05,
      "step": 15015
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.3033158813263525,
      "eval_f1_macro": 0.7003615661638819,
      "eval_f1_micro": 0.8167849686847599,
      "eval_loss": 0.12352865934371948,
      "eval_runtime": 453.6557,
      "eval_samples_per_second": 6.315,
      "eval_steps_per_second": 0.198,
      "learning_rate": 1e-05,
      "step": 15288
    },
    {
      "epoch": 56.776556776556774,
      "grad_norm": 0.29984405636787415,
      "learning_rate": 1e-05,
      "loss": 0.1281,
      "step": 15500
    },
    {
      "epoch": 57.0,
      "eval_accuracy": 0.3029668411867365,
      "eval_f1_macro": 0.7002357451773985,
      "eval_f1_micro": 0.815746583643278,
      "eval_loss": 0.1239086389541626,
      "eval_runtime": 461.4901,
      "eval_samples_per_second": 6.208,
      "eval_steps_per_second": 0.195,
      "learning_rate": 1e-05,
      "step": 15561
    },
    {
      "epoch": 58.0,
      "eval_accuracy": 0.30261780104712044,
      "eval_f1_macro": 0.699458636808175,
      "eval_f1_micro": 0.8157211095281991,
      "eval_loss": 0.12343526631593704,
      "eval_runtime": 461.5197,
      "eval_samples_per_second": 6.208,
      "eval_steps_per_second": 0.195,
      "learning_rate": 1e-05,
      "step": 15834
    },
    {
      "epoch": 58.608058608058606,
      "grad_norm": 0.22550125420093536,
      "learning_rate": 1e-05,
      "loss": 0.129,
      "step": 16000
    },
    {
      "epoch": 59.0,
      "eval_accuracy": 0.30471204188481676,
      "eval_f1_macro": 0.7012031830256923,
      "eval_f1_micro": 0.8149766777324873,
      "eval_loss": 0.12345146387815475,
      "eval_runtime": 458.3652,
      "eval_samples_per_second": 6.25,
      "eval_steps_per_second": 0.196,
      "learning_rate": 1e-05,
      "step": 16107
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.29808027923211167,
      "eval_f1_macro": 0.6932117075627285,
      "eval_f1_micro": 0.8127865511971473,
      "eval_loss": 0.1239377036690712,
      "eval_runtime": 461.8366,
      "eval_samples_per_second": 6.203,
      "eval_steps_per_second": 0.195,
      "learning_rate": 1e-05,
      "step": 16380
    },
    {
      "epoch": 60.43956043956044,
      "grad_norm": 0.21306775510311127,
      "learning_rate": 1e-05,
      "loss": 0.1284,
      "step": 16500
    },
    {
      "epoch": 61.0,
      "eval_accuracy": 0.3008726003490401,
      "eval_f1_macro": 0.707636320445804,
      "eval_f1_micro": 0.8174437097179208,
      "eval_loss": 0.12398885935544968,
      "eval_runtime": 462.2279,
      "eval_samples_per_second": 6.198,
      "eval_steps_per_second": 0.195,
      "learning_rate": 1e-05,
      "step": 16653
    },
    {
      "epoch": 62.0,
      "eval_accuracy": 0.3019197207678883,
      "eval_f1_macro": 0.7031525654438611,
      "eval_f1_micro": 0.8151862464183381,
      "eval_loss": 0.12334412336349487,
      "eval_runtime": 460.3594,
      "eval_samples_per_second": 6.223,
      "eval_steps_per_second": 0.195,
      "learning_rate": 1e-05,
      "step": 16926
    },
    {
      "epoch": 62.27106227106227,
      "grad_norm": 0.24677294492721558,
      "learning_rate": 1e-05,
      "loss": 0.127,
      "step": 17000
    },
    {
      "epoch": 63.0,
      "eval_accuracy": 0.3022687609075044,
      "eval_f1_macro": 0.7023118424286922,
      "eval_f1_micro": 0.8157861555919809,
      "eval_loss": 0.12325507402420044,
      "eval_runtime": 461.2281,
      "eval_samples_per_second": 6.212,
      "eval_steps_per_second": 0.195,
      "learning_rate": 1e-05,
      "step": 17199
    },
    {
      "epoch": 64.0,
      "eval_accuracy": 0.30471204188481676,
      "eval_f1_macro": 0.6998879503661606,
      "eval_f1_micro": 0.8152265453932963,
      "eval_loss": 0.12346883863210678,
      "eval_runtime": 466.8309,
      "eval_samples_per_second": 6.137,
      "eval_steps_per_second": 0.193,
      "learning_rate": 1e-05,
      "step": 17472
    },
    {
      "epoch": 64.1025641025641,
      "grad_norm": 0.1753869205713272,
      "learning_rate": 1e-05,
      "loss": 0.1279,
      "step": 17500
    },
    {
      "epoch": 65.0,
      "eval_accuracy": 0.29773123909249566,
      "eval_f1_macro": 0.7001319611176023,
      "eval_f1_micro": 0.814535597814578,
      "eval_loss": 0.12324482202529907,
      "eval_runtime": 462.9556,
      "eval_samples_per_second": 6.188,
      "eval_steps_per_second": 0.194,
      "learning_rate": 1e-05,
      "step": 17745
    },
    {
      "epoch": 65.93406593406593,
      "grad_norm": 0.23069432377815247,
      "learning_rate": 1e-05,
      "loss": 0.1273,
      "step": 18000
    },
    {
      "epoch": 66.0,
      "eval_accuracy": 0.3012216404886562,
      "eval_f1_macro": 0.7004348905275414,
      "eval_f1_micro": 0.814516812060642,
      "eval_loss": 0.12292143702507019,
      "eval_runtime": 464.3912,
      "eval_samples_per_second": 6.169,
      "eval_steps_per_second": 0.194,
      "learning_rate": 1e-05,
      "step": 18018
    },
    {
      "epoch": 67.0,
      "eval_accuracy": 0.30157068062827225,
      "eval_f1_macro": 0.699284425146712,
      "eval_f1_micro": 0.8159166351527375,
      "eval_loss": 0.12375594675540924,
      "eval_runtime": 464.0044,
      "eval_samples_per_second": 6.175,
      "eval_steps_per_second": 0.194,
      "learning_rate": 1e-05,
      "step": 18291
    },
    {
      "epoch": 67.76556776556777,
      "grad_norm": 0.4648727476596832,
      "learning_rate": 1e-05,
      "loss": 0.1272,
      "step": 18500
    },
    {
      "epoch": 68.0,
      "eval_accuracy": 0.299825479930192,
      "eval_f1_macro": 0.7038988669971363,
      "eval_f1_micro": 0.8175921663807173,
      "eval_loss": 0.1228519007563591,
      "eval_runtime": 462.9465,
      "eval_samples_per_second": 6.189,
      "eval_steps_per_second": 0.194,
      "learning_rate": 1e-05,
      "step": 18564
    },
    {
      "epoch": 69.0,
      "eval_accuracy": 0.3057591623036649,
      "eval_f1_macro": 0.7006054696492156,
      "eval_f1_micro": 0.8157472332516685,
      "eval_loss": 0.12302352488040924,
      "eval_runtime": 460.1932,
      "eval_samples_per_second": 6.226,
      "eval_steps_per_second": 0.196,
      "learning_rate": 1e-05,
      "step": 18837
    },
    {
      "epoch": 69.59706959706959,
      "grad_norm": 0.19116108119487762,
      "learning_rate": 1e-05,
      "loss": 0.1274,
      "step": 19000
    },
    {
      "epoch": 70.0,
      "eval_accuracy": 0.30401396160558464,
      "eval_f1_macro": 0.7009163688044155,
      "eval_f1_micro": 0.8169984042999916,
      "eval_loss": 0.12284138053655624,
      "eval_runtime": 460.5489,
      "eval_samples_per_second": 6.221,
      "eval_steps_per_second": 0.195,
      "learning_rate": 1e-05,
      "step": 19110
    },
    {
      "epoch": 71.0,
      "eval_accuracy": 0.3019197207678883,
      "eval_f1_macro": 0.7043489270867269,
      "eval_f1_micro": 0.8157594696169925,
      "eval_loss": 0.12295401096343994,
      "eval_runtime": 459.6032,
      "eval_samples_per_second": 6.234,
      "eval_steps_per_second": 0.196,
      "learning_rate": 1e-05,
      "step": 19383
    },
    {
      "epoch": 71.42857142857143,
      "grad_norm": 0.23645249009132385,
      "learning_rate": 1e-05,
      "loss": 0.1272,
      "step": 19500
    },
    {
      "epoch": 72.0,
      "eval_accuracy": 0.30157068062827225,
      "eval_f1_macro": 0.7024968584049214,
      "eval_f1_micro": 0.8171375869875074,
      "eval_loss": 0.123215451836586,
      "eval_runtime": 456.5626,
      "eval_samples_per_second": 6.275,
      "eval_steps_per_second": 0.197,
      "learning_rate": 1e-05,
      "step": 19656
    },
    {
      "epoch": 73.0,
      "eval_accuracy": 0.3054101221640489,
      "eval_f1_macro": 0.7048999311445782,
      "eval_f1_micro": 0.8173905730984317,
      "eval_loss": 0.12291014939546585,
      "eval_runtime": 459.2654,
      "eval_samples_per_second": 6.238,
      "eval_steps_per_second": 0.196,
      "learning_rate": 1e-05,
      "step": 19929
    },
    {
      "epoch": 73.26007326007326,
      "grad_norm": 0.18364407122135162,
      "learning_rate": 1e-05,
      "loss": 0.1267,
      "step": 20000
    },
    {
      "epoch": 74.0,
      "eval_accuracy": 0.3008726003490401,
      "eval_f1_macro": 0.6942481843546244,
      "eval_f1_micro": 0.8141268691688323,
      "eval_loss": 0.12304174154996872,
      "eval_runtime": 452.0481,
      "eval_samples_per_second": 6.338,
      "eval_steps_per_second": 0.199,
      "learning_rate": 1e-05,
      "step": 20202
    },
    {
      "epoch": 75.0,
      "eval_accuracy": 0.3033158813263525,
      "eval_f1_macro": 0.7001404945828892,
      "eval_f1_micro": 0.8161419818297063,
      "eval_loss": 0.1232200339436531,
      "eval_runtime": 455.6916,
      "eval_samples_per_second": 6.287,
      "eval_steps_per_second": 0.198,
      "learning_rate": 1.0000000000000002e-06,
      "step": 20475
    },
    {
      "epoch": 75.0915750915751,
      "grad_norm": 0.23693855106830597,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.1269,
      "step": 20500
    },
    {
      "epoch": 76.0,
      "eval_accuracy": 0.3057591623036649,
      "eval_f1_macro": 0.7020186938135398,
      "eval_f1_micro": 0.8171137406830337,
      "eval_loss": 0.12267689406871796,
      "eval_runtime": 450.1022,
      "eval_samples_per_second": 6.365,
      "eval_steps_per_second": 0.2,
      "learning_rate": 1.0000000000000002e-06,
      "step": 20748
    },
    {
      "epoch": 76.92307692307692,
      "grad_norm": 0.2620922923088074,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.1261,
      "step": 21000
    },
    {
      "epoch": 77.0,
      "eval_accuracy": 0.3078534031413613,
      "eval_f1_macro": 0.7059725105446049,
      "eval_f1_micro": 0.8190563444800733,
      "eval_loss": 0.12284990400075912,
      "eval_runtime": 450.7915,
      "eval_samples_per_second": 6.355,
      "eval_steps_per_second": 0.2,
      "learning_rate": 1.0000000000000002e-06,
      "step": 21021
    },
    {
      "epoch": 78.0,
      "eval_accuracy": 0.3019197207678883,
      "eval_f1_macro": 0.70715796895925,
      "eval_f1_micro": 0.8165993945509586,
      "eval_loss": 0.123690165579319,
      "eval_runtime": 449.2258,
      "eval_samples_per_second": 6.378,
      "eval_steps_per_second": 0.2,
      "learning_rate": 1.0000000000000002e-06,
      "step": 21294
    },
    {
      "epoch": 78.75457875457876,
      "grad_norm": 0.2725893557071686,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.1268,
      "step": 21500
    },
    {
      "epoch": 79.0,
      "eval_accuracy": 0.30471204188481676,
      "eval_f1_macro": 0.6992489783675215,
      "eval_f1_micro": 0.8156292286874154,
      "eval_loss": 0.12329532951116562,
      "eval_runtime": 448.681,
      "eval_samples_per_second": 6.385,
      "eval_steps_per_second": 0.201,
      "learning_rate": 1.0000000000000002e-06,
      "step": 21567
    },
    {
      "epoch": 80.0,
      "eval_accuracy": 0.30261780104712044,
      "eval_f1_macro": 0.6994382162058129,
      "eval_f1_micro": 0.8171630910227036,
      "eval_loss": 0.12325812131166458,
      "eval_runtime": 447.9437,
      "eval_samples_per_second": 6.396,
      "eval_steps_per_second": 0.201,
      "learning_rate": 1.0000000000000002e-06,
      "step": 21840
    },
    {
      "epoch": 80.58608058608058,
      "grad_norm": 0.23704515397548676,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.1271,
      "step": 22000
    },
    {
      "epoch": 81.0,
      "eval_accuracy": 0.3054101221640489,
      "eval_f1_macro": 0.7037260965073638,
      "eval_f1_micro": 0.8175692748252337,
      "eval_loss": 0.12240613251924515,
      "eval_runtime": 448.5942,
      "eval_samples_per_second": 6.387,
      "eval_steps_per_second": 0.201,
      "learning_rate": 1.0000000000000002e-06,
      "step": 22113
    },
    {
      "epoch": 82.0,
      "eval_accuracy": 0.30017452006980805,
      "eval_f1_macro": 0.6972115267564822,
      "eval_f1_micro": 0.8150502097368756,
      "eval_loss": 0.12270382046699524,
      "eval_runtime": 454.1003,
      "eval_samples_per_second": 6.309,
      "eval_steps_per_second": 0.198,
      "learning_rate": 1.0000000000000002e-06,
      "step": 22386
    },
    {
      "epoch": 82.41758241758242,
      "grad_norm": 0.24043193459510803,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.1263,
      "step": 22500
    },
    {
      "epoch": 83.0,
      "eval_accuracy": 0.2994764397905759,
      "eval_f1_macro": 0.6939342317359001,
      "eval_f1_micro": 0.8146007733820593,
      "eval_loss": 0.12315402179956436,
      "eval_runtime": 453.2474,
      "eval_samples_per_second": 6.321,
      "eval_steps_per_second": 0.199,
      "learning_rate": 1.0000000000000002e-06,
      "step": 22659
    },
    {
      "epoch": 84.0,
      "eval_accuracy": 0.30261780104712044,
      "eval_f1_macro": 0.7017348072852434,
      "eval_f1_micro": 0.817694818782958,
      "eval_loss": 0.1225922629237175,
      "eval_runtime": 454.9911,
      "eval_samples_per_second": 6.297,
      "eval_steps_per_second": 0.198,
      "learning_rate": 1.0000000000000002e-06,
      "step": 22932
    },
    {
      "epoch": 84.24908424908425,
      "grad_norm": 0.20054659247398376,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.1265,
      "step": 23000
    },
    {
      "epoch": 85.0,
      "eval_accuracy": 0.30715532286212915,
      "eval_f1_macro": 0.7059249231157801,
      "eval_f1_micro": 0.8180560794668679,
      "eval_loss": 0.1230376735329628,
      "eval_runtime": 454.3422,
      "eval_samples_per_second": 6.306,
      "eval_steps_per_second": 0.198,
      "learning_rate": 1.0000000000000002e-06,
      "step": 23205
    },
    {
      "epoch": 86.0,
      "eval_accuracy": 0.3033158813263525,
      "eval_f1_macro": 0.7058661509975368,
      "eval_f1_micro": 0.8168233713901948,
      "eval_loss": 0.12335028499364853,
      "eval_runtime": 455.4692,
      "eval_samples_per_second": 6.29,
      "eval_steps_per_second": 0.198,
      "learning_rate": 1.0000000000000002e-06,
      "step": 23478
    },
    {
      "epoch": 86.08058608058609,
      "grad_norm": 0.2747795581817627,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.1265,
      "step": 23500
    },
    {
      "epoch": 87.0,
      "eval_accuracy": 0.30052356020942406,
      "eval_f1_macro": 0.6943550775207357,
      "eval_f1_micro": 0.8143178633603668,
      "eval_loss": 0.12355341017246246,
      "eval_runtime": 467.692,
      "eval_samples_per_second": 6.126,
      "eval_steps_per_second": 0.192,
      "learning_rate": 1.0000000000000002e-06,
      "step": 23751
    },
    {
      "epoch": 87.91208791208791,
      "grad_norm": 0.2173408567905426,
      "learning_rate": 1.0000000000000002e-07,
      "loss": 0.1266,
      "step": 24000
    },
    {
      "epoch": 88.0,
      "eval_accuracy": 0.3078534031413613,
      "eval_f1_macro": 0.702889046791746,
      "eval_f1_micro": 0.8186198516780492,
      "eval_loss": 0.12264065444469452,
      "eval_runtime": 469.0124,
      "eval_samples_per_second": 6.109,
      "eval_steps_per_second": 0.192,
      "learning_rate": 1.0000000000000002e-07,
      "step": 24024
    },
    {
      "epoch": 89.0,
      "eval_accuracy": 0.3075043630017452,
      "eval_f1_macro": 0.7083793564773274,
      "eval_f1_micro": 0.817912934805249,
      "eval_loss": 0.1229795441031456,
      "eval_runtime": 470.229,
      "eval_samples_per_second": 6.093,
      "eval_steps_per_second": 0.191,
      "learning_rate": 1.0000000000000002e-07,
      "step": 24297
    },
    {
      "epoch": 89.74358974358974,
      "grad_norm": 0.22348648309707642,
      "learning_rate": 1.0000000000000002e-07,
      "loss": 0.1263,
      "step": 24500
    },
    {
      "epoch": 90.0,
      "eval_accuracy": 0.3036649214659686,
      "eval_f1_macro": 0.710478296923057,
      "eval_f1_micro": 0.8195029312195524,
      "eval_loss": 0.12317965924739838,
      "eval_runtime": 465.171,
      "eval_samples_per_second": 6.159,
      "eval_steps_per_second": 0.193,
      "learning_rate": 1.0000000000000002e-07,
      "step": 24570
    },
    {
      "epoch": 91.0,
      "eval_accuracy": 0.3019197207678883,
      "eval_f1_macro": 0.6971981356235558,
      "eval_f1_micro": 0.8155397390272835,
      "eval_loss": 0.12267619371414185,
      "eval_runtime": 458.3455,
      "eval_samples_per_second": 6.251,
      "eval_steps_per_second": 0.196,
      "learning_rate": 1.0000000000000002e-07,
      "step": 24843
    },
    {
      "epoch": 91.0,
      "learning_rate": 1.0000000000000002e-07,
      "step": 24843,
      "total_flos": 1.1760993126572918e+21,
      "train_loss": 0.1412498749900842,
      "train_runtime": 170277.6523,
      "train_samples_per_second": 7.678,
      "train_steps_per_second": 0.24
    }
  ],
  "logging_steps": 500,
  "max_steps": 40950,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 150,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 10,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 10
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.1760993126572918e+21,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}