{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 39.0,
  "global_step": 4134,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 1.3832186408159307e-05,
      "loss": 2.0685,
      "step": 106
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6720276077614272,
      "eval_loss": 1.7388017177581787,
      "eval_runtime": 36.8297,
      "eval_samples_per_second": 48.059,
      "eval_steps_per_second": 0.163,
      "step": 106
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.5888124272106204e-05,
      "loss": 1.7284,
      "step": 212
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6931985254177139,
      "eval_loss": 1.5560609102249146,
      "eval_runtime": 37.6763,
      "eval_samples_per_second": 46.979,
      "eval_steps_per_second": 0.159,
      "step": 212
    },
    {
      "epoch": 3.0,
      "learning_rate": 1.7090770826327895e-05,
      "loss": 1.5997,
      "step": 318
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7099605218744038,
      "eval_loss": 1.4386627674102783,
      "eval_runtime": 36.094,
      "eval_samples_per_second": 49.039,
      "eval_steps_per_second": 0.166,
      "step": 318
    },
    {
      "epoch": 4.0,
      "learning_rate": 1.7944062136053104e-05,
      "loss": 1.5195,
      "step": 424
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7235852020285128,
      "eval_loss": 1.3606762886047363,
      "eval_runtime": 37.6742,
      "eval_samples_per_second": 46.982,
      "eval_steps_per_second": 0.159,
      "step": 424
    },
    {
      "epoch": 5.0,
      "learning_rate": 1.860592629580032e-05,
      "loss": 1.4706,
      "step": 530
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7306072855931103,
      "eval_loss": 1.3053652048110962,
      "eval_runtime": 36.8913,
      "eval_samples_per_second": 47.979,
      "eval_steps_per_second": 0.163,
      "step": 530
    },
    {
      "epoch": 6.0,
      "learning_rate": 1.9146708690274792e-05,
      "loss": 1.4153,
      "step": 636
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7296665866066945,
      "eval_loss": 1.3213350772857666,
      "eval_runtime": 37.6502,
      "eval_samples_per_second": 47.012,
      "eval_steps_per_second": 0.159,
      "step": 636
    },
    {
      "epoch": 7.0,
      "learning_rate": 1.9603933689955228e-05,
      "loss": 1.3838,
      "step": 742
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7412067658165602,
      "eval_loss": 1.2423616647720337,
      "eval_runtime": 36.7828,
      "eval_samples_per_second": 48.12,
      "eval_steps_per_second": 0.163,
      "step": 742
    },
    {
      "epoch": 8.0,
      "learning_rate": 2e-05,
      "loss": 1.3512,
      "step": 848
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7385993664885682,
      "eval_loss": 1.2403146028518677,
      "eval_runtime": 37.6141,
      "eval_samples_per_second": 47.057,
      "eval_steps_per_second": 0.16,
      "step": 848
    },
    {
      "epoch": 9.0,
      "learning_rate": 2e-05,
      "loss": 1.3188,
      "step": 954
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.7413660954138545,
      "eval_loss": 1.2292009592056274,
      "eval_runtime": 36.8608,
      "eval_samples_per_second": 48.019,
      "eval_steps_per_second": 0.163,
      "step": 954
    },
    {
      "epoch": 10.0,
      "learning_rate": 2e-05,
      "loss": 1.3098,
      "step": 1060
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.7530054891006436,
      "eval_loss": 1.1540861129760742,
      "eval_runtime": 36.7518,
      "eval_samples_per_second": 48.161,
      "eval_steps_per_second": 0.163,
      "step": 1060
    },
    {
      "epoch": 11.0,
      "learning_rate": 2e-05,
      "loss": 1.2827,
      "step": 1166
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.7532555843965927,
      "eval_loss": 1.1605820655822754,
      "eval_runtime": 37.6556,
      "eval_samples_per_second": 47.005,
      "eval_steps_per_second": 0.159,
      "step": 1166
    },
    {
      "epoch": 12.0,
      "learning_rate": 2e-05,
      "loss": 1.2693,
      "step": 1272
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.7515547286193142,
      "eval_loss": 1.167082667350769,
      "eval_runtime": 36.7311,
      "eval_samples_per_second": 48.188,
      "eval_steps_per_second": 0.163,
      "step": 1272
    },
    {
      "epoch": 13.0,
      "learning_rate": 2e-05,
      "loss": 1.2521,
      "step": 1378
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.7581785009494731,
      "eval_loss": 1.134334921836853,
      "eval_runtime": 37.5802,
      "eval_samples_per_second": 47.099,
      "eval_steps_per_second": 0.16,
      "step": 1378
    },
    {
      "epoch": 14.0,
      "learning_rate": 2e-05,
      "loss": 1.2421,
      "step": 1484
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.7592175982523275,
      "eval_loss": 1.1171754598617554,
      "eval_runtime": 36.9701,
      "eval_samples_per_second": 47.877,
      "eval_steps_per_second": 0.162,
      "step": 1484
    },
    {
      "epoch": 15.0,
      "learning_rate": 2e-05,
      "loss": 1.2308,
      "step": 1590
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.7617716097984533,
      "eval_loss": 1.1091045141220093,
      "eval_runtime": 37.8672,
      "eval_samples_per_second": 46.742,
      "eval_steps_per_second": 0.158,
      "step": 1590
    },
    {
      "epoch": 16.0,
      "learning_rate": 2e-05,
      "loss": 1.2132,
      "step": 1696
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.7612774124438361,
      "eval_loss": 1.1064728498458862,
      "eval_runtime": 36.7747,
      "eval_samples_per_second": 48.131,
      "eval_steps_per_second": 0.163,
      "step": 1696
    },
    {
      "epoch": 17.0,
      "learning_rate": 2e-05,
      "loss": 1.2055,
      "step": 1802
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.761372646367941,
      "eval_loss": 1.1087865829467773,
      "eval_runtime": 36.8483,
      "eval_samples_per_second": 48.035,
      "eval_steps_per_second": 0.163,
      "step": 1802
    },
    {
      "epoch": 18.0,
      "learning_rate": 2e-05,
      "loss": 1.1931,
      "step": 1908
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.7599678162601031,
      "eval_loss": 1.1089411973953247,
      "eval_runtime": 35.9217,
      "eval_samples_per_second": 49.274,
      "eval_steps_per_second": 0.167,
      "step": 1908
    },
    {
      "epoch": 19.0,
      "learning_rate": 2e-05,
      "loss": 1.1815,
      "step": 2014
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.7659135481142874,
      "eval_loss": 1.0751391649246216,
      "eval_runtime": 36.8871,
      "eval_samples_per_second": 47.984,
      "eval_steps_per_second": 0.163,
      "step": 2014
    },
    {
      "epoch": 20.0,
      "learning_rate": 2e-05,
      "loss": 1.1728,
      "step": 2120
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.768600214617861,
      "eval_loss": 1.069868803024292,
      "eval_runtime": 36.7387,
      "eval_samples_per_second": 48.178,
      "eval_steps_per_second": 0.163,
      "step": 2120
    },
    {
      "epoch": 21.0,
      "learning_rate": 2e-05,
      "loss": 1.164,
      "step": 2226
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.7675022113058035,
      "eval_loss": 1.065330147743225,
      "eval_runtime": 36.8047,
      "eval_samples_per_second": 48.092,
      "eval_steps_per_second": 0.163,
      "step": 2226
    },
    {
      "epoch": 22.0,
      "learning_rate": 2e-05,
      "loss": 1.1524,
      "step": 2332
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.7698299863752147,
      "eval_loss": 1.0548479557037354,
      "eval_runtime": 36.7052,
      "eval_samples_per_second": 48.222,
      "eval_steps_per_second": 0.163,
      "step": 2332
    },
    {
      "epoch": 23.0,
      "learning_rate": 2e-05,
      "loss": 1.1425,
      "step": 2438
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.769697828632807,
      "eval_loss": 1.047703504562378,
      "eval_runtime": 36.8466,
      "eval_samples_per_second": 48.037,
      "eval_steps_per_second": 0.163,
      "step": 2438
    },
    {
      "epoch": 24.0,
      "learning_rate": 2e-05,
      "loss": 1.143,
      "step": 2544
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.7772214047626197,
      "eval_loss": 1.0133404731750488,
      "eval_runtime": 36.8371,
      "eval_samples_per_second": 48.049,
      "eval_steps_per_second": 0.163,
      "step": 2544
    },
    {
      "epoch": 25.0,
      "learning_rate": 2e-05,
      "loss": 1.1308,
      "step": 2650
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.7741821838279008,
      "eval_loss": 1.0260401964187622,
      "eval_runtime": 36.8437,
      "eval_samples_per_second": 48.041,
      "eval_steps_per_second": 0.163,
      "step": 2650
    },
    {
      "epoch": 26.0,
      "learning_rate": 2e-05,
      "loss": 1.1271,
      "step": 2756
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.773661286574258,
      "eval_loss": 1.0230038166046143,
      "eval_runtime": 36.8925,
      "eval_samples_per_second": 47.977,
      "eval_steps_per_second": 0.163,
      "step": 2756
    },
    {
      "epoch": 27.0,
      "learning_rate": 2e-05,
      "loss": 1.1202,
      "step": 2862
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.7754215166511026,
      "eval_loss": 1.0241199731826782,
      "eval_runtime": 36.7992,
      "eval_samples_per_second": 48.099,
      "eval_steps_per_second": 0.163,
      "step": 2862
    },
    {
      "epoch": 28.0,
      "learning_rate": 2e-05,
      "loss": 1.1168,
      "step": 2968
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.7775761513243871,
      "eval_loss": 1.0062930583953857,
      "eval_runtime": 38.045,
      "eval_samples_per_second": 46.524,
      "eval_steps_per_second": 0.158,
      "step": 2968
    },
    {
      "epoch": 29.0,
      "learning_rate": 2e-05,
      "loss": 1.1019,
      "step": 3074
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.7796242613030875,
      "eval_loss": 0.9990780353546143,
      "eval_runtime": 37.599,
      "eval_samples_per_second": 47.076,
      "eval_steps_per_second": 0.16,
      "step": 3074
    },
    {
      "epoch": 30.0,
      "learning_rate": 2e-05,
      "loss": 1.1071,
      "step": 3180
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.7805559249900252,
      "eval_loss": 0.991283655166626,
      "eval_runtime": 36.7892,
      "eval_samples_per_second": 48.112,
      "eval_steps_per_second": 0.163,
      "step": 3180
    },
    {
      "epoch": 31.0,
      "learning_rate": 2e-05,
      "loss": 1.0963,
      "step": 3286
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.7864944315242504,
      "eval_loss": 0.9553370475769043,
      "eval_runtime": 36.7173,
      "eval_samples_per_second": 48.206,
      "eval_steps_per_second": 0.163,
      "step": 3286
    },
    {
      "epoch": 32.0,
      "learning_rate": 2e-05,
      "loss": 1.089,
      "step": 3392
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.783273055389144,
      "eval_loss": 0.9850459098815918,
      "eval_runtime": 36.7845,
      "eval_samples_per_second": 48.118,
      "eval_steps_per_second": 0.163,
      "step": 3392
    },
    {
      "epoch": 33.0,
      "learning_rate": 2e-05,
      "loss": 1.0807,
      "step": 3498
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.7774915196015286,
      "eval_loss": 0.9987505674362183,
      "eval_runtime": 36.8264,
      "eval_samples_per_second": 48.063,
      "eval_steps_per_second": 0.163,
      "step": 3498
    },
    {
      "epoch": 34.0,
      "learning_rate": 2e-05,
      "loss": 1.0735,
      "step": 3604
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.7789523242042387,
      "eval_loss": 0.9969209432601929,
      "eval_runtime": 37.6346,
      "eval_samples_per_second": 47.031,
      "eval_steps_per_second": 0.159,
      "step": 3604
    },
    {
      "epoch": 35.0,
      "learning_rate": 2e-05,
      "loss": 1.0766,
      "step": 3710
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.7880373776195184,
      "eval_loss": 0.9506202936172485,
      "eval_runtime": 36.9744,
      "eval_samples_per_second": 47.871,
      "eval_steps_per_second": 0.162,
      "step": 3710
    },
    {
      "epoch": 36.0,
      "learning_rate": 2e-05,
      "loss": 1.0698,
      "step": 3816
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.7880023166711217,
      "eval_loss": 0.9565942883491516,
      "eval_runtime": 38.7958,
      "eval_samples_per_second": 45.623,
      "eval_steps_per_second": 0.155,
      "step": 3816
    },
    {
      "epoch": 37.0,
      "learning_rate": 2e-05,
      "loss": 1.0608,
      "step": 3922
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.7855791137596568,
      "eval_loss": 0.9620457291603088,
      "eval_runtime": 37.7795,
      "eval_samples_per_second": 46.851,
      "eval_steps_per_second": 0.159,
      "step": 3922
    },
    {
      "epoch": 38.0,
      "learning_rate": 2e-05,
      "loss": 1.0543,
      "step": 4028
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.7822399039183305,
      "eval_loss": 0.9812787175178528,
      "eval_runtime": 36.8483,
      "eval_samples_per_second": 48.035,
      "eval_steps_per_second": 0.163,
      "step": 4028
    },
    {
      "epoch": 39.0,
      "learning_rate": 2e-05,
      "loss": 1.0521,
      "step": 4134
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.7884438270952546,
      "eval_loss": 0.9478756785392761,
      "eval_runtime": 36.7861,
      "eval_samples_per_second": 48.116,
      "eval_steps_per_second": 0.163,
      "step": 4134
    }
  ],
  "max_steps": 4240,
  "num_train_epochs": 40,
  "total_flos": 498219970723840.0,
  "trial_name": null,
  "trial_params": null
}