{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.2424289485945024,
  "eval_steps": 1000,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.024848578971890044,
      "grad_norm": 1.1403515692366237,
      "learning_rate": 0.0002475,
      "loss": 0.7727,
      "step": 100
    },
    {
      "epoch": 0.04969715794378009,
      "grad_norm": 2.0321728234236067,
      "learning_rate": 0.0004975,
      "loss": 0.6434,
      "step": 200
    },
    {
      "epoch": 0.07454573691567014,
      "grad_norm": 1.5298185029395692,
      "learning_rate": 0.0004896875,
      "loss": 0.6966,
      "step": 300
    },
    {
      "epoch": 0.09939431588756018,
      "grad_norm": 1.3277644876612467,
      "learning_rate": 0.00047927083333333334,
      "loss": 0.6549,
      "step": 400
    },
    {
      "epoch": 0.12424289485945023,
      "grad_norm": 1.5260941221380562,
      "learning_rate": 0.0004688541666666667,
      "loss": 0.6218,
      "step": 500
    },
    {
      "epoch": 0.14909147383134028,
      "grad_norm": 1.461823959618585,
      "learning_rate": 0.0004584375,
      "loss": 0.6,
      "step": 600
    },
    {
      "epoch": 0.17394005280323033,
      "grad_norm": 1.579155099177269,
      "learning_rate": 0.0004480208333333333,
      "loss": 0.5768,
      "step": 700
    },
    {
      "epoch": 0.19878863177512035,
      "grad_norm": 1.0030257236753515,
      "learning_rate": 0.0004376041666666667,
      "loss": 0.5625,
      "step": 800
    },
    {
      "epoch": 0.2236372107470104,
      "grad_norm": 1.1362089000123636,
      "learning_rate": 0.0004271875,
      "loss": 0.5397,
      "step": 900
    },
    {
      "epoch": 0.24848578971890045,
      "grad_norm": 1.2060122481392086,
      "learning_rate": 0.00041677083333333334,
      "loss": 0.5262,
      "step": 1000
    },
    {
      "epoch": 0.24848578971890045,
      "eval_accuracy": 0.8519274864233807,
      "eval_loss": 0.4272424280643463,
      "eval_runtime": 355.1137,
      "eval_samples_per_second": 76.471,
      "eval_steps_per_second": 9.56,
      "step": 1000
    },
    {
      "epoch": 0.2733343686907905,
      "grad_norm": 1.4939232786831804,
      "learning_rate": 0.0004063541666666667,
      "loss": 0.507,
      "step": 1100
    },
    {
      "epoch": 0.29818294766268055,
      "grad_norm": 1.0210090194207913,
      "learning_rate": 0.0003959375,
      "loss": 0.4923,
      "step": 1200
    },
    {
      "epoch": 0.3230315266345706,
      "grad_norm": 1.5265032617099488,
      "learning_rate": 0.0003855208333333333,
      "loss": 0.485,
      "step": 1300
    },
    {
      "epoch": 0.34788010560646065,
      "grad_norm": 1.5974690542066108,
      "learning_rate": 0.0003751041666666667,
      "loss": 0.4734,
      "step": 1400
    },
    {
      "epoch": 0.3727286845783507,
      "grad_norm": 1.3463235139038994,
      "learning_rate": 0.0003646875,
      "loss": 0.4629,
      "step": 1500
    },
    {
      "epoch": 0.3975772635502407,
      "grad_norm": 0.8545527093751057,
      "learning_rate": 0.00035427083333333334,
      "loss": 0.4504,
      "step": 1600
    },
    {
      "epoch": 0.4224258425221308,
      "grad_norm": 1.1247072555863402,
      "learning_rate": 0.0003438541666666667,
      "loss": 0.4366,
      "step": 1700
    },
    {
      "epoch": 0.4472744214940208,
      "grad_norm": 0.9808487633447448,
      "learning_rate": 0.0003334375,
      "loss": 0.4326,
      "step": 1800
    },
    {
      "epoch": 0.4721230004659109,
      "grad_norm": 0.9117492642967103,
      "learning_rate": 0.0003230208333333333,
      "loss": 0.4259,
      "step": 1900
    },
    {
      "epoch": 0.4969715794378009,
      "grad_norm": 0.7717947637961347,
      "learning_rate": 0.0003126041666666667,
      "loss": 0.413,
      "step": 2000
    },
    {
      "epoch": 0.4969715794378009,
      "eval_accuracy": 0.8710562213369932,
      "eval_loss": 0.3649989068508148,
      "eval_runtime": 354.5714,
      "eval_samples_per_second": 76.588,
      "eval_steps_per_second": 9.575,
      "step": 2000
    },
    {
      "epoch": 0.521820158409691,
      "grad_norm": 0.8113600127860467,
      "learning_rate": 0.0003021875,
      "loss": 0.4058,
      "step": 2100
    },
    {
      "epoch": 0.546668737381581,
      "grad_norm": 0.8903551125095357,
      "learning_rate": 0.00029177083333333334,
      "loss": 0.397,
      "step": 2200
    },
    {
      "epoch": 0.571517316353471,
      "grad_norm": 0.6671784623140349,
      "learning_rate": 0.0002813541666666667,
      "loss": 0.3913,
      "step": 2300
    },
    {
      "epoch": 0.5963658953253611,
      "grad_norm": 0.7312172828583983,
      "learning_rate": 0.0002709375,
      "loss": 0.3818,
      "step": 2400
    },
    {
      "epoch": 0.6212144742972511,
      "grad_norm": 0.7230909781837032,
      "learning_rate": 0.0002605208333333333,
      "loss": 0.3764,
      "step": 2500
    },
    {
      "epoch": 0.6460630532691412,
      "grad_norm": 0.6919092960285033,
      "learning_rate": 0.0002501041666666667,
      "loss": 0.3749,
      "step": 2600
    },
    {
      "epoch": 0.6709116322410312,
      "grad_norm": 0.7449666551062925,
      "learning_rate": 0.0002396875,
      "loss": 0.3679,
      "step": 2700
    },
    {
      "epoch": 0.6957602112129213,
      "grad_norm": 0.6657006999853705,
      "learning_rate": 0.00022927083333333333,
      "loss": 0.3603,
      "step": 2800
    },
    {
      "epoch": 0.7206087901848113,
      "grad_norm": 0.7751989636808677,
      "learning_rate": 0.00021885416666666665,
      "loss": 0.3576,
      "step": 2900
    },
    {
      "epoch": 0.7454573691567014,
      "grad_norm": 0.582330455026636,
      "learning_rate": 0.0002084375,
      "loss": 0.3505,
      "step": 3000
    },
    {
      "epoch": 0.7454573691567014,
      "eval_accuracy": 0.885227854690163,
      "eval_loss": 0.3137596547603607,
      "eval_runtime": 354.0567,
      "eval_samples_per_second": 76.7,
      "eval_steps_per_second": 9.589,
      "step": 3000
    },
    {
      "epoch": 0.7703059481285914,
      "grad_norm": 0.624670881883226,
      "learning_rate": 0.00019802083333333333,
      "loss": 0.3419,
      "step": 3100
    },
    {
      "epoch": 0.7951545271004814,
      "grad_norm": 0.6846077168970752,
      "learning_rate": 0.00018760416666666665,
      "loss": 0.3429,
      "step": 3200
    },
    {
      "epoch": 0.8200031060723715,
      "grad_norm": 0.549136141158353,
      "learning_rate": 0.0001771875,
      "loss": 0.3347,
      "step": 3300
    },
    {
      "epoch": 0.8448516850442616,
      "grad_norm": 0.5196724667426235,
      "learning_rate": 0.00016677083333333333,
      "loss": 0.3347,
      "step": 3400
    },
    {
      "epoch": 0.8697002640161515,
      "grad_norm": 0.559242933165121,
      "learning_rate": 0.00015635416666666665,
      "loss": 0.33,
      "step": 3500
    },
    {
      "epoch": 0.8945488429880416,
      "grad_norm": 0.5987209783232824,
      "learning_rate": 0.0001459375,
      "loss": 0.3236,
      "step": 3600
    },
    {
      "epoch": 0.9193974219599317,
      "grad_norm": 0.5980851789962784,
      "learning_rate": 0.00013552083333333333,
      "loss": 0.3242,
      "step": 3700
    },
    {
      "epoch": 0.9442460009318218,
      "grad_norm": 0.8472335802810602,
      "learning_rate": 0.00012510416666666665,
      "loss": 0.3135,
      "step": 3800
    },
    {
      "epoch": 0.9690945799037117,
      "grad_norm": 0.5414234864761139,
      "learning_rate": 0.0001146875,
      "loss": 0.3151,
      "step": 3900
    },
    {
      "epoch": 0.9939431588756018,
      "grad_norm": 0.5498938130901921,
      "learning_rate": 0.00010427083333333333,
      "loss": 0.3111,
      "step": 4000
    },
    {
      "epoch": 0.9939431588756018,
      "eval_accuracy": 0.894954167881237,
      "eval_loss": 0.2828981876373291,
      "eval_runtime": 353.5701,
      "eval_samples_per_second": 76.805,
      "eval_steps_per_second": 9.602,
      "step": 4000
    },
    {
      "epoch": 1.0187917378474918,
      "grad_norm": 0.49437309189757744,
      "learning_rate": 9.385416666666667e-05,
      "loss": 0.3059,
      "step": 4100
    },
    {
      "epoch": 1.043640316819382,
      "grad_norm": 0.5262199037896036,
      "learning_rate": 8.34375e-05,
      "loss": 0.3021,
      "step": 4200
    },
    {
      "epoch": 1.068488895791272,
      "grad_norm": 0.508417592763766,
      "learning_rate": 7.302083333333333e-05,
      "loss": 0.2961,
      "step": 4300
    },
    {
      "epoch": 1.093337474763162,
      "grad_norm": 0.4111721436877036,
      "learning_rate": 6.260416666666667e-05,
      "loss": 0.2966,
      "step": 4400
    },
    {
      "epoch": 1.118186053735052,
      "grad_norm": 0.44511878991978343,
      "learning_rate": 5.21875e-05,
      "loss": 0.293,
      "step": 4500
    },
    {
      "epoch": 1.143034632706942,
      "grad_norm": 0.44581104227849994,
      "learning_rate": 4.177083333333334e-05,
      "loss": 0.2903,
      "step": 4600
    },
    {
      "epoch": 1.167883211678832,
      "grad_norm": 0.40843300918294273,
      "learning_rate": 3.135416666666667e-05,
      "loss": 0.2872,
      "step": 4700
    },
    {
      "epoch": 1.1927317906507222,
      "grad_norm": 0.45173761297005016,
      "learning_rate": 2.09375e-05,
      "loss": 0.2829,
      "step": 4800
    },
    {
      "epoch": 1.2175803696226122,
      "grad_norm": 0.3729668197663827,
      "learning_rate": 1.0520833333333334e-05,
      "loss": 0.2833,
      "step": 4900
    },
    {
      "epoch": 1.2424289485945024,
      "grad_norm": 0.39171821421954367,
      "learning_rate": 1.0416666666666668e-07,
      "loss": 0.2817,
      "step": 5000
    },
    {
      "epoch": 1.2424289485945024,
      "eval_accuracy": 0.9025104707867149,
      "eval_loss": 0.2596372365951538,
      "eval_runtime": 353.5151,
      "eval_samples_per_second": 76.817,
      "eval_steps_per_second": 9.604,
      "step": 5000
    },
    {
      "epoch": 1.2424289485945024,
      "step": 5000,
      "total_flos": 643481070469120.0,
      "train_loss": 0.4145617248535156,
      "train_runtime": 25261.0456,
      "train_samples_per_second": 25.335,
      "train_steps_per_second": 0.198
    }
  ],
  "logging_steps": 100,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 250,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 643481070469120.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}