{
  "best_metric": 0.5094339622641509,
  "best_model_checkpoint": "./Validated_Balanced_Raw_Data_model_vit_outputs/checkpoint-1440",
  "epoch": 25.0,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.625,
      "grad_norm": 10.029894828796387,
      "learning_rate": 1.47e-05,
      "loss": 1.4452,
      "step": 50
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.24056603773584906,
      "eval_loss": 1.4536809921264648,
      "eval_runtime": 1.1349,
      "eval_samples_per_second": 186.798,
      "eval_steps_per_second": 23.79,
      "step": 80
    },
    {
      "epoch": 1.25,
      "grad_norm": 13.045843124389648,
      "learning_rate": 2.97e-05,
      "loss": 1.4021,
      "step": 100
    },
    {
      "epoch": 1.875,
      "grad_norm": 9.291085243225098,
      "learning_rate": 2.9950795096316707e-05,
      "loss": 1.3534,
      "step": 150
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.35377358490566035,
      "eval_loss": 1.4003523588180542,
      "eval_runtime": 1.1414,
      "eval_samples_per_second": 185.736,
      "eval_steps_per_second": 23.655,
      "step": 160
    },
    {
      "epoch": 2.5,
      "grad_norm": 8.364015579223633,
      "learning_rate": 2.979948160693239e-05,
      "loss": 1.2977,
      "step": 200
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.37735849056603776,
      "eval_loss": 1.3416857719421387,
      "eval_runtime": 1.1295,
      "eval_samples_per_second": 187.7,
      "eval_steps_per_second": 23.905,
      "step": 240
    },
    {
      "epoch": 3.125,
      "grad_norm": 10.082795143127441,
      "learning_rate": 2.9547072651695803e-05,
      "loss": 1.2979,
      "step": 250
    },
    {
      "epoch": 3.75,
      "grad_norm": 9.592968940734863,
      "learning_rate": 2.9195292439710517e-05,
      "loss": 1.2604,
      "step": 300
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.37735849056603776,
      "eval_loss": 1.3131976127624512,
      "eval_runtime": 1.1222,
      "eval_samples_per_second": 188.91,
      "eval_steps_per_second": 24.059,
      "step": 320
    },
    {
      "epoch": 4.375,
      "grad_norm": 9.461295127868652,
      "learning_rate": 2.874654398652483e-05,
      "loss": 1.2509,
      "step": 350
    },
    {
      "epoch": 5.0,
      "grad_norm": 20.441808700561523,
      "learning_rate": 2.8203892699098956e-05,
      "loss": 1.2428,
      "step": 400
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.4009433962264151,
      "eval_loss": 1.2442667484283447,
      "eval_runtime": 1.1294,
      "eval_samples_per_second": 187.712,
      "eval_steps_per_second": 23.907,
      "step": 400
    },
    {
      "epoch": 5.625,
      "grad_norm": 7.7924604415893555,
      "learning_rate": 2.757104543596717e-05,
      "loss": 1.213,
      "step": 450
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.419811320754717,
      "eval_loss": 1.2147694826126099,
      "eval_runtime": 1.1426,
      "eval_samples_per_second": 185.546,
      "eval_steps_per_second": 23.631,
      "step": 480
    },
    {
      "epoch": 6.25,
      "grad_norm": 9.47819709777832,
      "learning_rate": 2.6852325185635357e-05,
      "loss": 1.2163,
      "step": 500
    },
    {
      "epoch": 6.875,
      "grad_norm": 9.0564546585083,
      "learning_rate": 2.605264153618602e-05,
      "loss": 1.1426,
      "step": 550
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.4669811320754717,
      "eval_loss": 1.2095811367034912,
      "eval_runtime": 1.125,
      "eval_samples_per_second": 188.45,
      "eval_steps_per_second": 24.001,
      "step": 560
    },
    {
      "epoch": 7.5,
      "grad_norm": 6.277278900146484,
      "learning_rate": 2.517745713781345e-05,
      "loss": 1.1657,
      "step": 600
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.4669811320754717,
      "eval_loss": 1.2065520286560059,
      "eval_runtime": 1.1354,
      "eval_samples_per_second": 186.725,
      "eval_steps_per_second": 23.781,
      "step": 640
    },
    {
      "epoch": 8.125,
      "grad_norm": 9.455373764038086,
      "learning_rate": 2.4252284842435667e-05,
      "loss": 1.1988,
      "step": 650
    },
    {
      "epoch": 8.75,
      "grad_norm": 9.699421882629395,
      "learning_rate": 2.3245704321242494e-05,
      "loss": 1.1249,
      "step": 700
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.4386792452830189,
      "eval_loss": 1.220900297164917,
      "eval_runtime": 1.1295,
      "eval_samples_per_second": 187.686,
      "eval_steps_per_second": 23.903,
      "step": 720
    },
    {
      "epoch": 9.375,
      "grad_norm": 10.024123191833496,
      "learning_rate": 2.218279727850104e-05,
      "loss": 1.0891,
      "step": 750
    },
    {
      "epoch": 10.0,
      "grad_norm": 14.003544807434082,
      "learning_rate": 2.107082444708681e-05,
      "loss": 1.1622,
      "step": 800
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.4811320754716981,
      "eval_loss": 1.1445928812026978,
      "eval_runtime": 1.1414,
      "eval_samples_per_second": 185.734,
      "eval_steps_per_second": 23.655,
      "step": 800
    },
    {
      "epoch": 10.625,
      "grad_norm": 10.41913890838623,
      "learning_rate": 1.9917381728963962e-05,
      "loss": 1.0625,
      "step": 850
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.4669811320754717,
      "eval_loss": 1.1742380857467651,
      "eval_runtime": 1.1293,
      "eval_samples_per_second": 187.73,
      "eval_steps_per_second": 23.909,
      "step": 880
    },
    {
      "epoch": 11.25,
      "grad_norm": 9.851762771606445,
      "learning_rate": 1.8730348307472828e-05,
      "loss": 1.043,
      "step": 900
    },
    {
      "epoch": 11.875,
      "grad_norm": 8.781569480895996,
      "learning_rate": 1.7517832824518216e-05,
      "loss": 1.1157,
      "step": 950
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.44339622641509435,
      "eval_loss": 1.2199500799179077,
      "eval_runtime": 1.1331,
      "eval_samples_per_second": 187.094,
      "eval_steps_per_second": 23.828,
      "step": 960
    },
    {
      "epoch": 12.5,
      "grad_norm": 10.422264099121094,
      "learning_rate": 1.628811799032326e-05,
      "loss": 1.0807,
      "step": 1000
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.4669811320754717,
      "eval_loss": 1.2117412090301514,
      "eval_runtime": 1.1321,
      "eval_samples_per_second": 187.259,
      "eval_steps_per_second": 23.849,
      "step": 1040
    },
    {
      "epoch": 13.125,
      "grad_norm": 8.4778470993042,
      "learning_rate": 1.5049604004119928e-05,
      "loss": 1.0891,
      "step": 1050
    },
    {
      "epoch": 13.75,
      "grad_norm": 11.677724838256836,
      "learning_rate": 1.381075117227066e-05,
      "loss": 1.0629,
      "step": 1100
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.4811320754716981,
      "eval_loss": 1.2296439409255981,
      "eval_runtime": 1.1234,
      "eval_samples_per_second": 188.721,
      "eval_steps_per_second": 24.035,
      "step": 1120
    },
    {
      "epoch": 14.375,
      "grad_norm": 11.892010688781738,
      "learning_rate": 1.2580022115797237e-05,
      "loss": 1.0673,
      "step": 1150
    },
    {
      "epoch": 15.0,
      "grad_norm": 16.539581298828125,
      "learning_rate": 1.1365823962098208e-05,
      "loss": 1.0323,
      "step": 1200
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.49056603773584906,
      "eval_loss": 1.1886956691741943,
      "eval_runtime": 1.1303,
      "eval_samples_per_second": 187.563,
      "eval_steps_per_second": 23.888,
      "step": 1200
    },
    {
      "epoch": 15.625,
      "grad_norm": 6.841868877410889,
      "learning_rate": 1.0176450915744072e-05,
      "loss": 1.0128,
      "step": 1250
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.49528301886792453,
      "eval_loss": 1.2074586153030396,
      "eval_runtime": 1.1321,
      "eval_samples_per_second": 187.267,
      "eval_steps_per_second": 23.85,
      "step": 1280
    },
    {
      "epoch": 16.25,
      "grad_norm": 9.490548133850098,
      "learning_rate": 9.020027600649825e-06,
      "loss": 0.9715,
      "step": 1300
    },
    {
      "epoch": 16.875,
      "grad_norm": 9.569696426391602,
      "learning_rate": 7.904453560655389e-06,
      "loss": 1.0266,
      "step": 1350
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.5,
      "eval_loss": 1.2082122564315796,
      "eval_runtime": 1.1449,
      "eval_samples_per_second": 185.172,
      "eval_steps_per_second": 23.583,
      "step": 1360
    },
    {
      "epoch": 17.5,
      "grad_norm": 9.353178977966309,
      "learning_rate": 6.837349297631114e-06,
      "loss": 1.004,
      "step": 1400
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.5094339622641509,
      "eval_loss": 1.2154083251953125,
      "eval_runtime": 1.1238,
      "eval_samples_per_second": 188.649,
      "eval_steps_per_second": 24.026,
      "step": 1440
    },
    {
      "epoch": 18.125,
      "grad_norm": 9.470197677612305,
      "learning_rate": 5.826004215722937e-06,
      "loss": 1.002,
      "step": 1450
    },
    {
      "epoch": 18.75,
      "grad_norm": 6.417470932006836,
      "learning_rate": 4.87732682733072e-06,
      "loss": 0.9543,
      "step": 1500
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.5047169811320755,
      "eval_loss": 1.2047548294067383,
      "eval_runtime": 1.1355,
      "eval_samples_per_second": 186.702,
      "eval_steps_per_second": 23.778,
      "step": 1520
    },
    {
      "epoch": 19.375,
      "grad_norm": 8.318925857543945,
      "learning_rate": 3.997797560963404e-06,
      "loss": 1.0218,
      "step": 1550
    },
    {
      "epoch": 20.0,
      "grad_norm": 8.360044479370117,
      "learning_rate": 3.193424493341213e-06,
      "loss": 0.9439,
      "step": 1600
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.49056603773584906,
      "eval_loss": 1.2217515707015991,
      "eval_runtime": 1.1305,
      "eval_samples_per_second": 187.525,
      "eval_steps_per_second": 23.883,
      "step": 1600
    },
    {
      "epoch": 20.625,
      "grad_norm": 12.746959686279297,
      "learning_rate": 2.4697023081395654e-06,
      "loss": 0.9891,
      "step": 1650
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.49056603773584906,
      "eval_loss": 1.2135605812072754,
      "eval_runtime": 1.1547,
      "eval_samples_per_second": 183.601,
      "eval_steps_per_second": 23.383,
      "step": 1680
    },
    {
      "epoch": 21.25,
      "grad_norm": 7.136344909667969,
      "learning_rate": 1.8315747617280382e-06,
      "loss": 0.9586,
      "step": 1700
    },
    {
      "epoch": 21.875,
      "grad_norm": 8.792285919189453,
      "learning_rate": 1.283400912301434e-06,
      "loss": 0.9801,
      "step": 1750
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.4858490566037736,
      "eval_loss": 1.216606855392456,
      "eval_runtime": 1.133,
      "eval_samples_per_second": 187.117,
      "eval_steps_per_second": 23.831,
      "step": 1760
    },
    {
      "epoch": 22.5,
      "grad_norm": 8.172861099243164,
      "learning_rate": 8.289253430923128e-07,
      "loss": 0.9632,
      "step": 1800
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.49056603773584906,
      "eval_loss": 1.2148730754852295,
      "eval_runtime": 1.1383,
      "eval_samples_per_second": 186.244,
      "eval_steps_per_second": 23.72,
      "step": 1840
    },
    {
      "epoch": 23.125,
      "grad_norm": 8.669483184814453,
      "learning_rate": 4.7125258307053385e-07,
      "loss": 0.9956,
      "step": 1850
    },
    {
      "epoch": 23.75,
      "grad_norm": 8.101666450500488,
      "learning_rate": 2.1282589986245494e-07,
      "loss": 0.9584,
      "step": 1900
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.49056603773584906,
      "eval_loss": 1.213520884513855,
      "eval_runtime": 1.1315,
      "eval_samples_per_second": 187.37,
      "eval_steps_per_second": 23.863,
      "step": 1920
    },
    {
      "epoch": 24.375,
      "grad_norm": 25.165281295776367,
      "learning_rate": 5.5410609755597306e-08,
      "loss": 0.948,
      "step": 1950
    },
    {
      "epoch": 25.0,
      "grad_norm": 10.21505069732666,
      "learning_rate": 8.201879839297988e-11,
      "loss": 0.9561,
      "step": 2000
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.49056603773584906,
      "eval_loss": 1.2136294841766357,
      "eval_runtime": 2.3217,
      "eval_samples_per_second": 91.313,
      "eval_steps_per_second": 11.63,
      "step": 2000
    },
    {
      "epoch": 25.0,
      "step": 2000,
      "total_flos": 1.2321447211229184e+18,
      "train_loss": 1.1025613689422606,
      "train_runtime": 425.612,
      "train_samples_per_second": 37.358,
      "train_steps_per_second": 4.699
    }
  ],
  "logging_steps": 50,
  "max_steps": 2000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 25,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.2321447211229184e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}