{
  "best_metric": 0.5330188679245284,
  "best_model_checkpoint": "./Validated_cracks_raw_dataset_266_outputs/checkpoint-1280",
  "epoch": 30.0,
  "eval_steps": 500,
  "global_step": 2400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.625,
      "grad_norm": 12.7417573928833,
      "learning_rate": 3.916666666666667e-06,
      "loss": 1.3904,
      "step": 50
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.35377358490566035,
      "eval_loss": 1.347093939781189,
      "eval_runtime": 1.5501,
      "eval_samples_per_second": 136.765,
      "eval_steps_per_second": 17.418,
      "step": 80
    },
    {
      "epoch": 1.25,
      "grad_norm": 13.116891860961914,
      "learning_rate": 8.083333333333334e-06,
      "loss": 1.3764,
      "step": 100
    },
    {
      "epoch": 1.875,
      "grad_norm": Infinity,
      "learning_rate": 9.996791734463078e-06,
      "loss": 1.3264,
      "step": 150
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.33962264150943394,
      "eval_loss": 1.2973852157592773,
      "eval_runtime": 1.5451,
      "eval_samples_per_second": 137.208,
      "eval_steps_per_second": 17.475,
      "step": 160
    },
    {
      "epoch": 2.5,
      "grad_norm": 11.807487487792969,
      "learning_rate": 9.972609476841368e-06,
      "loss": 1.2875,
      "step": 200
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3490566037735849,
      "eval_loss": 1.2867087125778198,
      "eval_runtime": 1.5667,
      "eval_samples_per_second": 135.313,
      "eval_steps_per_second": 17.233,
      "step": 240
    },
    {
      "epoch": 3.125,
      "grad_norm": 13.677336692810059,
      "learning_rate": 9.92483424862726e-06,
      "loss": 1.2754,
      "step": 250
    },
    {
      "epoch": 3.75,
      "grad_norm": 11.059372901916504,
      "learning_rate": 9.8536927234736e-06,
      "loss": 1.2415,
      "step": 300
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.37735849056603776,
      "eval_loss": 1.257009744644165,
      "eval_runtime": 1.5386,
      "eval_samples_per_second": 137.789,
      "eval_steps_per_second": 17.549,
      "step": 320
    },
    {
      "epoch": 4.375,
      "grad_norm": 9.970629692077637,
      "learning_rate": 9.759522438425813e-06,
      "loss": 1.2347,
      "step": 350
    },
    {
      "epoch": 5.0,
      "grad_norm": 36.81943130493164,
      "learning_rate": 9.642770192448537e-06,
      "loss": 1.2209,
      "step": 400
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.4339622641509434,
      "eval_loss": 1.214514136314392,
      "eval_runtime": 1.5466,
      "eval_samples_per_second": 137.075,
      "eval_steps_per_second": 17.458,
      "step": 400
    },
    {
      "epoch": 5.625,
      "grad_norm": 25.951169967651367,
      "learning_rate": 9.50398992654976e-06,
      "loss": 1.1699,
      "step": 450
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.41509433962264153,
      "eval_loss": 1.2138594388961792,
      "eval_runtime": 1.5575,
      "eval_samples_per_second": 136.117,
      "eval_steps_per_second": 17.336,
      "step": 480
    },
    {
      "epoch": 6.25,
      "grad_norm": 33.16694259643555,
      "learning_rate": 9.343840095560373e-06,
      "loss": 1.1707,
      "step": 500
    },
    {
      "epoch": 6.875,
      "grad_norm": 15.973888397216797,
      "learning_rate": 9.163080544038953e-06,
      "loss": 1.1315,
      "step": 550
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.46226415094339623,
      "eval_loss": 1.1762322187423706,
      "eval_runtime": 1.548,
      "eval_samples_per_second": 136.955,
      "eval_steps_per_second": 17.442,
      "step": 560
    },
    {
      "epoch": 7.5,
      "grad_norm": 16.8402042388916,
      "learning_rate": 8.962568901124326e-06,
      "loss": 1.1565,
      "step": 600
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.45754716981132076,
      "eval_loss": 1.1620187759399414,
      "eval_runtime": 1.5613,
      "eval_samples_per_second": 135.788,
      "eval_steps_per_second": 17.294,
      "step": 640
    },
    {
      "epoch": 8.125,
      "grad_norm": 27.61142921447754,
      "learning_rate": 8.743256511440829e-06,
      "loss": 1.1218,
      "step": 650
    },
    {
      "epoch": 8.75,
      "grad_norm": 14.339193344116211,
      "learning_rate": 8.511092281712174e-06,
      "loss": 1.1111,
      "step": 700
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.4811320754716981,
      "eval_loss": 1.1559137105941772,
      "eval_runtime": 1.5476,
      "eval_samples_per_second": 136.986,
      "eval_steps_per_second": 17.446,
      "step": 720
    },
    {
      "epoch": 9.375,
      "grad_norm": 22.12592124938965,
      "learning_rate": 8.257705467351144e-06,
      "loss": 1.0349,
      "step": 750
    },
    {
      "epoch": 10.0,
      "grad_norm": 30.31277847290039,
      "learning_rate": 7.988862191016204e-06,
      "loss": 1.117,
      "step": 800
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.5,
      "eval_loss": 1.125497817993164,
      "eval_runtime": 1.5617,
      "eval_samples_per_second": 135.746,
      "eval_steps_per_second": 17.288,
      "step": 800
    },
    {
      "epoch": 10.625,
      "grad_norm": 53.451171875,
      "learning_rate": 7.705838002605665e-06,
      "loss": 1.0174,
      "step": 850
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.5047169811320755,
      "eval_loss": 1.1186039447784424,
      "eval_runtime": 1.5425,
      "eval_samples_per_second": 137.44,
      "eval_steps_per_second": 17.504,
      "step": 880
    },
    {
      "epoch": 11.25,
      "grad_norm": 18.92786979675293,
      "learning_rate": 7.409975734566998e-06,
      "loss": 1.0209,
      "step": 900
    },
    {
      "epoch": 11.875,
      "grad_norm": 25.6451358795166,
      "learning_rate": 7.102679130713538e-06,
      "loss": 1.0569,
      "step": 950
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.5,
      "eval_loss": 1.1092461347579956,
      "eval_runtime": 1.5482,
      "eval_samples_per_second": 136.93,
      "eval_steps_per_second": 17.439,
      "step": 960
    },
    {
      "epoch": 12.5,
      "grad_norm": 16.849384307861328,
      "learning_rate": 6.785406186042e-06,
      "loss": 1.0089,
      "step": 1000
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.5,
      "eval_loss": 1.1156014204025269,
      "eval_runtime": 1.5482,
      "eval_samples_per_second": 136.935,
      "eval_steps_per_second": 17.44,
      "step": 1040
    },
    {
      "epoch": 13.125,
      "grad_norm": 29.936918258666992,
      "learning_rate": 6.45966222915063e-06,
      "loss": 1.0412,
      "step": 1050
    },
    {
      "epoch": 13.75,
      "grad_norm": 46.736331939697266,
      "learning_rate": 6.126992780079032e-06,
      "loss": 1.0413,
      "step": 1100
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.49528301886792453,
      "eval_loss": 1.108467698097229,
      "eval_runtime": 1.5451,
      "eval_samples_per_second": 137.209,
      "eval_steps_per_second": 17.475,
      "step": 1120
    },
    {
      "epoch": 14.375,
      "grad_norm": 35.94117736816406,
      "learning_rate": 5.788976217456275e-06,
      "loss": 1.019,
      "step": 1150
    },
    {
      "epoch": 15.0,
      "grad_norm": 23.572219848632812,
      "learning_rate": 5.447216289748596e-06,
      "loss": 0.9958,
      "step": 1200
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.5235849056603774,
      "eval_loss": 1.115731954574585,
      "eval_runtime": 1.5428,
      "eval_samples_per_second": 137.41,
      "eval_steps_per_second": 17.5,
      "step": 1200
    },
    {
      "epoch": 15.625,
      "grad_norm": 19.54017448425293,
      "learning_rate": 5.103334506137773e-06,
      "loss": 0.9969,
      "step": 1250
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.5330188679245284,
      "eval_loss": 1.1048997640609741,
      "eval_runtime": 1.5622,
      "eval_samples_per_second": 135.705,
      "eval_steps_per_second": 17.283,
      "step": 1280
    },
    {
      "epoch": 16.25,
      "grad_norm": 33.59697341918945,
      "learning_rate": 4.758962443132227e-06,
      "loss": 0.9203,
      "step": 1300
    },
    {
      "epoch": 16.875,
      "grad_norm": 22.073335647583008,
      "learning_rate": 4.415734003412873e-06,
      "loss": 0.9918,
      "step": 1350
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.5047169811320755,
      "eval_loss": 1.1045137643814087,
      "eval_runtime": 1.5338,
      "eval_samples_per_second": 138.214,
      "eval_steps_per_second": 17.603,
      "step": 1360
    },
    {
      "epoch": 17.5,
      "grad_norm": 39.07722473144531,
      "learning_rate": 4.075277663642208e-06,
      "loss": 0.9798,
      "step": 1400
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.5141509433962265,
      "eval_loss": 1.0932115316390991,
      "eval_runtime": 1.5496,
      "eval_samples_per_second": 136.805,
      "eval_steps_per_second": 17.423,
      "step": 1440
    },
    {
      "epoch": 18.125,
      "grad_norm": 34.9410285949707,
      "learning_rate": 3.739208748017647e-06,
      "loss": 0.9567,
      "step": 1450
    },
    {
      "epoch": 18.75,
      "grad_norm": 13.545350074768066,
      "learning_rate": 3.409121764227809e-06,
      "loss": 0.9232,
      "step": 1500
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.5047169811320755,
      "eval_loss": 1.0961326360702515,
      "eval_runtime": 1.5504,
      "eval_samples_per_second": 136.736,
      "eval_steps_per_second": 17.414,
      "step": 1520
    },
    {
      "epoch": 19.375,
      "grad_norm": 21.65803337097168,
      "learning_rate": 3.0865828381745515e-06,
      "loss": 0.9919,
      "step": 1550
    },
    {
      "epoch": 20.0,
      "grad_norm": 24.88987159729004,
      "learning_rate": 2.7731222833547842e-06,
      "loss": 0.8817,
      "step": 1600
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.49056603773584906,
      "eval_loss": 1.1115567684173584,
      "eval_runtime": 1.544,
      "eval_samples_per_second": 137.306,
      "eval_steps_per_second": 17.487,
      "step": 1600
    },
    {
      "epoch": 20.625,
      "grad_norm": 33.98754119873047,
      "learning_rate": 2.470227340157316e-06,
      "loss": 0.9587,
      "step": 1650
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.5235849056603774,
      "eval_loss": 1.1000138521194458,
      "eval_runtime": 1.5478,
      "eval_samples_per_second": 136.971,
      "eval_steps_per_second": 17.444,
      "step": 1680
    },
    {
      "epoch": 21.25,
      "grad_norm": 18.886362075805664,
      "learning_rate": 2.179335119523745e-06,
      "loss": 0.8872,
      "step": 1700
    },
    {
      "epoch": 21.875,
      "grad_norm": 45.27411651611328,
      "learning_rate": 1.901825784452777e-06,
      "loss": 0.964,
      "step": 1750
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.5094339622641509,
      "eval_loss": 1.097833275794983,
      "eval_runtime": 1.5565,
      "eval_samples_per_second": 136.207,
      "eval_steps_per_second": 17.347,
      "step": 1760
    },
    {
      "epoch": 22.5,
      "grad_norm": 46.455753326416016,
      "learning_rate": 1.6390160016989487e-06,
      "loss": 0.8906,
      "step": 1800
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.5235849056603774,
      "eval_loss": 1.0974699258804321,
      "eval_runtime": 1.5424,
      "eval_samples_per_second": 137.448,
      "eval_steps_per_second": 17.505,
      "step": 1840
    },
    {
      "epoch": 23.125,
      "grad_norm": 40.78911590576172,
      "learning_rate": 1.3921526947346902e-06,
      "loss": 0.945,
      "step": 1850
    },
    {
      "epoch": 23.75,
      "grad_norm": 23.577985763549805,
      "learning_rate": 1.162407127615357e-06,
      "loss": 0.896,
      "step": 1900
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.5141509433962265,
      "eval_loss": 1.099574089050293,
      "eval_runtime": 1.549,
      "eval_samples_per_second": 136.865,
      "eval_steps_per_second": 17.431,
      "step": 1920
    },
    {
      "epoch": 24.375,
      "grad_norm": 33.863609313964844,
      "learning_rate": 9.508693478168346e-07,
      "loss": 0.891,
      "step": 1950
    },
    {
      "epoch": 25.0,
      "grad_norm": 27.020963668823242,
      "learning_rate": 7.585430144121319e-07,
      "loss": 0.9156,
      "step": 2000
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.5235849056603774,
      "eval_loss": 1.0951563119888306,
      "eval_runtime": 1.5411,
      "eval_samples_per_second": 137.565,
      "eval_steps_per_second": 17.52,
      "step": 2000
    },
    {
      "epoch": 25.625,
      "grad_norm": 15.748483657836914,
      "learning_rate": 5.863406361251472e-07,
      "loss": 0.8797,
      "step": 2050
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.5235849056603774,
      "eval_loss": 1.0959974527359009,
      "eval_runtime": 1.5455,
      "eval_samples_per_second": 137.175,
      "eval_steps_per_second": 17.47,
      "step": 2080
    },
    {
      "epoch": 26.25,
      "grad_norm": 25.159223556518555,
      "learning_rate": 4.350792418550509e-07,
      "loss": 0.9148,
      "step": 2100
    },
    {
      "epoch": 26.875,
      "grad_norm": 18.654874801635742,
      "learning_rate": 3.0547650421285216e-07,
      "loss": 0.8781,
      "step": 2150
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.5235849056603774,
      "eval_loss": 1.0947679281234741,
      "eval_runtime": 1.5435,
      "eval_samples_per_second": 137.346,
      "eval_steps_per_second": 17.492,
      "step": 2160
    },
    {
      "epoch": 27.5,
      "grad_norm": 37.69408416748047,
      "learning_rate": 1.9814733446237356e-07,
      "loss": 0.8698,
      "step": 2200
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.5235849056603774,
      "eval_loss": 1.0946481227874756,
      "eval_runtime": 1.5439,
      "eval_samples_per_second": 137.312,
      "eval_steps_per_second": 17.488,
      "step": 2240
    },
    {
      "epoch": 28.125,
      "grad_norm": 18.448745727539062,
      "learning_rate": 1.1360096502120387e-07,
      "loss": 0.915,
      "step": 2250
    },
    {
      "epoch": 28.75,
      "grad_norm": 25.56351661682129,
      "learning_rate": 5.223853336398632e-08,
      "loss": 0.9,
      "step": 2300
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.5283018867924528,
      "eval_loss": 1.0949124097824097,
      "eval_runtime": 1.5384,
      "eval_samples_per_second": 137.806,
      "eval_steps_per_second": 17.551,
      "step": 2320
    },
    {
      "epoch": 29.375,
      "grad_norm": 20.619009017944336,
      "learning_rate": 1.4351178791384702e-08,
      "loss": 0.9145,
      "step": 2350
    },
    {
      "epoch": 30.0,
      "grad_norm": 16.861053466796875,
      "learning_rate": 1.1866109479674593e-10,
      "loss": 0.8853,
      "step": 2400
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.5283018867924528,
      "eval_loss": 1.0948740243911743,
      "eval_runtime": 2.7457,
      "eval_samples_per_second": 77.212,
      "eval_steps_per_second": 9.834,
      "step": 2400
    },
    {
      "epoch": 30.0,
      "step": 2400,
      "total_flos": 1.5090549662800282e+18,
      "train_loss": 1.0357336870829263,
      "train_runtime": 587.3439,
      "train_samples_per_second": 32.485,
      "train_steps_per_second": 4.086
    }
  ],
  "logging_steps": 50,
  "max_steps": 2400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "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": 1.5090549662800282e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}