{
  "best_metric": 1.6099027395248413,
  "best_model_checkpoint": "gemstones_image_detection/checkpoint-4779",
  "epoch": 59.0,
  "eval_steps": 500,
  "global_step": 4779,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.06521739130434782,
      "eval_loss": 4.402697563171387,
      "eval_runtime": 7.0526,
      "eval_samples_per_second": 91.314,
      "eval_steps_per_second": 11.485,
      "step": 81
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.2329192546583851,
      "eval_loss": 4.283246994018555,
      "eval_runtime": 6.9117,
      "eval_samples_per_second": 93.176,
      "eval_steps_per_second": 11.719,
      "step": 162
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.38354037267080743,
      "eval_loss": 4.158541202545166,
      "eval_runtime": 7.0269,
      "eval_samples_per_second": 91.648,
      "eval_steps_per_second": 11.527,
      "step": 243
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5139751552795031,
      "eval_loss": 4.027233123779297,
      "eval_runtime": 6.9926,
      "eval_samples_per_second": 92.098,
      "eval_steps_per_second": 11.584,
      "step": 324
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6024844720496895,
      "eval_loss": 3.892688035964966,
      "eval_runtime": 7.1602,
      "eval_samples_per_second": 89.942,
      "eval_steps_per_second": 11.313,
      "step": 405
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.65527950310559,
      "eval_loss": 3.7535414695739746,
      "eval_runtime": 7.0085,
      "eval_samples_per_second": 91.888,
      "eval_steps_per_second": 11.557,
      "step": 486
    },
    {
      "epoch": 6.17,
      "learning_rate": 9.064449064449066e-06,
      "loss": 4.0521,
      "step": 500
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6816770186335404,
      "eval_loss": 3.6245224475860596,
      "eval_runtime": 6.9342,
      "eval_samples_per_second": 92.873,
      "eval_steps_per_second": 11.681,
      "step": 567
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7065217391304348,
      "eval_loss": 3.5059802532196045,
      "eval_runtime": 6.9825,
      "eval_samples_per_second": 92.231,
      "eval_steps_per_second": 11.6,
      "step": 648
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.7251552795031055,
      "eval_loss": 3.3893799781799316,
      "eval_runtime": 7.0313,
      "eval_samples_per_second": 91.59,
      "eval_steps_per_second": 11.52,
      "step": 729
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.7468944099378882,
      "eval_loss": 3.2792539596557617,
      "eval_runtime": 6.9072,
      "eval_samples_per_second": 93.236,
      "eval_steps_per_second": 11.727,
      "step": 810
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.7593167701863354,
      "eval_loss": 3.185778856277466,
      "eval_runtime": 7.1843,
      "eval_samples_per_second": 89.64,
      "eval_steps_per_second": 11.275,
      "step": 891
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.765527950310559,
      "eval_loss": 3.093205690383911,
      "eval_runtime": 7.203,
      "eval_samples_per_second": 89.407,
      "eval_steps_per_second": 11.245,
      "step": 972
    },
    {
      "epoch": 12.35,
      "learning_rate": 8.024948024948025e-06,
      "loss": 3.1602,
      "step": 1000
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.7639751552795031,
      "eval_loss": 3.006822347640991,
      "eval_runtime": 6.9935,
      "eval_samples_per_second": 92.086,
      "eval_steps_per_second": 11.582,
      "step": 1053
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.7701863354037267,
      "eval_loss": 2.9287173748016357,
      "eval_runtime": 6.8243,
      "eval_samples_per_second": 94.369,
      "eval_steps_per_second": 11.869,
      "step": 1134
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.7763975155279503,
      "eval_loss": 2.8492088317871094,
      "eval_runtime": 6.9345,
      "eval_samples_per_second": 92.869,
      "eval_steps_per_second": 11.681,
      "step": 1215
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.7763975155279503,
      "eval_loss": 2.779766082763672,
      "eval_runtime": 7.4706,
      "eval_samples_per_second": 86.204,
      "eval_steps_per_second": 10.842,
      "step": 1296
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.782608695652174,
      "eval_loss": 2.71138858795166,
      "eval_runtime": 7.6693,
      "eval_samples_per_second": 83.971,
      "eval_steps_per_second": 10.562,
      "step": 1377
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 2.64682674407959,
      "eval_runtime": 7.4341,
      "eval_samples_per_second": 86.628,
      "eval_steps_per_second": 10.896,
      "step": 1458
    },
    {
      "epoch": 18.52,
      "learning_rate": 6.985446985446987e-06,
      "loss": 2.537,
      "step": 1500
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.7841614906832298,
      "eval_loss": 2.586892604827881,
      "eval_runtime": 7.0064,
      "eval_samples_per_second": 91.916,
      "eval_steps_per_second": 11.561,
      "step": 1539
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.7888198757763976,
      "eval_loss": 2.533936023712158,
      "eval_runtime": 6.9165,
      "eval_samples_per_second": 93.111,
      "eval_steps_per_second": 11.711,
      "step": 1620
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.7950310559006211,
      "eval_loss": 2.4751620292663574,
      "eval_runtime": 6.8444,
      "eval_samples_per_second": 94.092,
      "eval_steps_per_second": 11.835,
      "step": 1701
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.8059006211180124,
      "eval_loss": 2.4239892959594727,
      "eval_runtime": 6.7886,
      "eval_samples_per_second": 94.864,
      "eval_steps_per_second": 11.932,
      "step": 1782
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.8027950310559007,
      "eval_loss": 2.3724544048309326,
      "eval_runtime": 6.8741,
      "eval_samples_per_second": 93.685,
      "eval_steps_per_second": 11.783,
      "step": 1863
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.8043478260869565,
      "eval_loss": 2.327075719833374,
      "eval_runtime": 7.4527,
      "eval_samples_per_second": 86.412,
      "eval_steps_per_second": 10.869,
      "step": 1944
    },
    {
      "epoch": 24.69,
      "learning_rate": 5.945945945945947e-06,
      "loss": 2.0899,
      "step": 2000
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.8074534161490683,
      "eval_loss": 2.2822916507720947,
      "eval_runtime": 6.923,
      "eval_samples_per_second": 93.024,
      "eval_steps_per_second": 11.7,
      "step": 2025
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.812111801242236,
      "eval_loss": 2.2414402961730957,
      "eval_runtime": 6.7582,
      "eval_samples_per_second": 95.292,
      "eval_steps_per_second": 11.985,
      "step": 2106
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.8152173913043478,
      "eval_loss": 2.198101043701172,
      "eval_runtime": 6.9757,
      "eval_samples_per_second": 92.32,
      "eval_steps_per_second": 11.612,
      "step": 2187
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.812111801242236,
      "eval_loss": 2.158327579498291,
      "eval_runtime": 6.8707,
      "eval_samples_per_second": 93.732,
      "eval_steps_per_second": 11.789,
      "step": 2268
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.8152173913043478,
      "eval_loss": 2.1208977699279785,
      "eval_runtime": 7.0622,
      "eval_samples_per_second": 91.19,
      "eval_steps_per_second": 11.47,
      "step": 2349
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.8198757763975155,
      "eval_loss": 2.086048126220703,
      "eval_runtime": 7.4205,
      "eval_samples_per_second": 86.787,
      "eval_steps_per_second": 10.916,
      "step": 2430
    },
    {
      "epoch": 30.86,
      "learning_rate": 4.906444906444907e-06,
      "loss": 1.7517,
      "step": 2500
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.8152173913043478,
      "eval_loss": 2.054029941558838,
      "eval_runtime": 7.2868,
      "eval_samples_per_second": 88.38,
      "eval_steps_per_second": 11.116,
      "step": 2511
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.8214285714285714,
      "eval_loss": 2.0214383602142334,
      "eval_runtime": 6.9044,
      "eval_samples_per_second": 93.273,
      "eval_steps_per_second": 11.732,
      "step": 2592
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.8245341614906833,
      "eval_loss": 1.9886717796325684,
      "eval_runtime": 7.0403,
      "eval_samples_per_second": 91.474,
      "eval_steps_per_second": 11.505,
      "step": 2673
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.8167701863354038,
      "eval_loss": 1.9592698812484741,
      "eval_runtime": 7.0078,
      "eval_samples_per_second": 91.898,
      "eval_steps_per_second": 11.559,
      "step": 2754
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.8136645962732919,
      "eval_loss": 1.9315446615219116,
      "eval_runtime": 6.9305,
      "eval_samples_per_second": 92.922,
      "eval_steps_per_second": 11.687,
      "step": 2835
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.8136645962732919,
      "eval_loss": 1.903202772140503,
      "eval_runtime": 7.1188,
      "eval_samples_per_second": 90.464,
      "eval_steps_per_second": 11.378,
      "step": 2916
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.8214285714285714,
      "eval_loss": 1.8818532228469849,
      "eval_runtime": 7.3502,
      "eval_samples_per_second": 87.617,
      "eval_steps_per_second": 11.02,
      "step": 2997
    },
    {
      "epoch": 37.04,
      "learning_rate": 3.866943866943867e-06,
      "loss": 1.4983,
      "step": 3000
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.8198757763975155,
      "eval_loss": 1.85394287109375,
      "eval_runtime": 7.066,
      "eval_samples_per_second": 91.141,
      "eval_steps_per_second": 11.463,
      "step": 3078
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.8152173913043478,
      "eval_loss": 1.8299955129623413,
      "eval_runtime": 6.9274,
      "eval_samples_per_second": 92.965,
      "eval_steps_per_second": 11.693,
      "step": 3159
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.8198757763975155,
      "eval_loss": 1.811854362487793,
      "eval_runtime": 6.9595,
      "eval_samples_per_second": 92.536,
      "eval_steps_per_second": 11.639,
      "step": 3240
    },
    {
      "epoch": 41.0,
      "eval_accuracy": 0.8136645962732919,
      "eval_loss": 1.7893236875534058,
      "eval_runtime": 6.921,
      "eval_samples_per_second": 93.05,
      "eval_steps_per_second": 11.704,
      "step": 3321
    },
    {
      "epoch": 42.0,
      "eval_accuracy": 0.8214285714285714,
      "eval_loss": 1.7720342874526978,
      "eval_runtime": 6.8638,
      "eval_samples_per_second": 93.826,
      "eval_steps_per_second": 11.801,
      "step": 3402
    },
    {
      "epoch": 43.0,
      "eval_accuracy": 0.8167701863354038,
      "eval_loss": 1.7533122301101685,
      "eval_runtime": 7.194,
      "eval_samples_per_second": 89.519,
      "eval_steps_per_second": 11.259,
      "step": 3483
    },
    {
      "epoch": 43.21,
      "learning_rate": 2.827442827442828e-06,
      "loss": 1.3099,
      "step": 3500
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.8214285714285714,
      "eval_loss": 1.7392221689224243,
      "eval_runtime": 6.9012,
      "eval_samples_per_second": 93.317,
      "eval_steps_per_second": 11.737,
      "step": 3564
    },
    {
      "epoch": 45.0,
      "eval_accuracy": 0.8167701863354038,
      "eval_loss": 1.71939218044281,
      "eval_runtime": 7.3602,
      "eval_samples_per_second": 87.497,
      "eval_steps_per_second": 11.005,
      "step": 3645
    },
    {
      "epoch": 46.0,
      "eval_accuracy": 0.8183229813664596,
      "eval_loss": 1.7061350345611572,
      "eval_runtime": 7.1127,
      "eval_samples_per_second": 90.542,
      "eval_steps_per_second": 11.388,
      "step": 3726
    },
    {
      "epoch": 47.0,
      "eval_accuracy": 0.8198757763975155,
      "eval_loss": 1.6922202110290527,
      "eval_runtime": 7.1567,
      "eval_samples_per_second": 89.986,
      "eval_steps_per_second": 11.318,
      "step": 3807
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.8136645962732919,
      "eval_loss": 1.6803785562515259,
      "eval_runtime": 6.8048,
      "eval_samples_per_second": 94.64,
      "eval_steps_per_second": 11.903,
      "step": 3888
    },
    {
      "epoch": 49.0,
      "eval_accuracy": 0.8152173913043478,
      "eval_loss": 1.6694267988204956,
      "eval_runtime": 7.2998,
      "eval_samples_per_second": 88.221,
      "eval_steps_per_second": 11.096,
      "step": 3969
    },
    {
      "epoch": 49.38,
      "learning_rate": 1.7879417879417881e-06,
      "loss": 1.1777,
      "step": 4000
    },
    {
      "epoch": 50.0,
      "eval_accuracy": 0.8167701863354038,
      "eval_loss": 1.6600834131240845,
      "eval_runtime": 6.8099,
      "eval_samples_per_second": 94.568,
      "eval_steps_per_second": 11.894,
      "step": 4050
    },
    {
      "epoch": 51.0,
      "eval_accuracy": 0.8214285714285714,
      "eval_loss": 1.6486986875534058,
      "eval_runtime": 6.809,
      "eval_samples_per_second": 94.581,
      "eval_steps_per_second": 11.896,
      "step": 4131
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.8152173913043478,
      "eval_loss": 1.641870379447937,
      "eval_runtime": 6.9774,
      "eval_samples_per_second": 92.297,
      "eval_steps_per_second": 11.609,
      "step": 4212
    },
    {
      "epoch": 53.0,
      "eval_accuracy": 0.8167701863354038,
      "eval_loss": 1.6351778507232666,
      "eval_runtime": 7.0388,
      "eval_samples_per_second": 91.493,
      "eval_steps_per_second": 11.508,
      "step": 4293
    },
    {
      "epoch": 54.0,
      "eval_accuracy": 0.812111801242236,
      "eval_loss": 1.6281042098999023,
      "eval_runtime": 7.0002,
      "eval_samples_per_second": 91.997,
      "eval_steps_per_second": 11.571,
      "step": 4374
    },
    {
      "epoch": 55.0,
      "eval_accuracy": 0.8167701863354038,
      "eval_loss": 1.6229327917099,
      "eval_runtime": 7.1102,
      "eval_samples_per_second": 90.574,
      "eval_steps_per_second": 11.392,
      "step": 4455
    },
    {
      "epoch": 55.56,
      "learning_rate": 7.484407484407485e-07,
      "loss": 1.0992,
      "step": 4500
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.8136645962732919,
      "eval_loss": 1.6176931858062744,
      "eval_runtime": 6.9832,
      "eval_samples_per_second": 92.221,
      "eval_steps_per_second": 11.599,
      "step": 4536
    },
    {
      "epoch": 57.0,
      "eval_accuracy": 0.8183229813664596,
      "eval_loss": 1.6147724390029907,
      "eval_runtime": 6.8902,
      "eval_samples_per_second": 93.465,
      "eval_steps_per_second": 11.756,
      "step": 4617
    },
    {
      "epoch": 58.0,
      "eval_accuracy": 0.8167701863354038,
      "eval_loss": 1.6121727228164673,
      "eval_runtime": 6.9785,
      "eval_samples_per_second": 92.283,
      "eval_steps_per_second": 11.607,
      "step": 4698
    },
    {
      "epoch": 59.0,
      "eval_accuracy": 0.8167701863354038,
      "eval_loss": 1.6099027395248413,
      "eval_runtime": 7.0345,
      "eval_samples_per_second": 91.548,
      "eval_steps_per_second": 11.515,
      "step": 4779
    }
  ],
  "logging_steps": 500,
  "max_steps": 4860,
  "num_train_epochs": 60,
  "save_steps": 500,
  "total_flos": 1.1781939502129306e+19,
  "trial_name": null,
  "trial_params": null
}