{
  "best_metric": 0.2986587882041931,
  "best_model_checkpoint": "./results/checkpoint-180",
  "epoch": 1.510574018126888,
  "eval_steps": 20,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06042296072507553,
      "grad_norm": 0.5974797606468201,
      "learning_rate": 1.97583081570997e-05,
      "loss": 0.5426,
      "step": 20
    },
    {
      "epoch": 0.06042296072507553,
      "eval_accuracy": 0.7450980392156863,
      "eval_loss": 0.7869178652763367,
      "eval_runtime": 15.0039,
      "eval_samples_per_second": 10.197,
      "eval_steps_per_second": 2.599,
      "step": 20
    },
    {
      "epoch": 0.12084592145015106,
      "grad_norm": 13.038370132446289,
      "learning_rate": 1.9516616314199397e-05,
      "loss": 0.5984,
      "step": 40
    },
    {
      "epoch": 0.12084592145015106,
      "eval_accuracy": 0.7908496732026143,
      "eval_loss": 0.39430469274520874,
      "eval_runtime": 15.9435,
      "eval_samples_per_second": 9.596,
      "eval_steps_per_second": 2.446,
      "step": 40
    },
    {
      "epoch": 0.18126888217522658,
      "grad_norm": 0.1351632922887802,
      "learning_rate": 1.9274924471299096e-05,
      "loss": 0.4864,
      "step": 60
    },
    {
      "epoch": 0.18126888217522658,
      "eval_accuracy": 0.7843137254901961,
      "eval_loss": 0.9364686608314514,
      "eval_runtime": 15.2263,
      "eval_samples_per_second": 10.048,
      "eval_steps_per_second": 2.561,
      "step": 60
    },
    {
      "epoch": 0.24169184290030213,
      "grad_norm": 1.4119517803192139,
      "learning_rate": 1.9033232628398792e-05,
      "loss": 0.6039,
      "step": 80
    },
    {
      "epoch": 0.24169184290030213,
      "eval_accuracy": 0.7712418300653595,
      "eval_loss": 0.6580381989479065,
      "eval_runtime": 15.5472,
      "eval_samples_per_second": 9.841,
      "eval_steps_per_second": 2.508,
      "step": 80
    },
    {
      "epoch": 0.3021148036253776,
      "grad_norm": 13.04010009765625,
      "learning_rate": 1.879154078549849e-05,
      "loss": 0.5741,
      "step": 100
    },
    {
      "epoch": 0.3021148036253776,
      "eval_accuracy": 0.8235294117647058,
      "eval_loss": 0.3454345464706421,
      "eval_runtime": 15.5082,
      "eval_samples_per_second": 9.866,
      "eval_steps_per_second": 2.515,
      "step": 100
    },
    {
      "epoch": 0.36253776435045315,
      "grad_norm": 61.882415771484375,
      "learning_rate": 1.854984894259819e-05,
      "loss": 0.4276,
      "step": 120
    },
    {
      "epoch": 0.36253776435045315,
      "eval_accuracy": 0.8169934640522876,
      "eval_loss": 0.5421260595321655,
      "eval_runtime": 15.4627,
      "eval_samples_per_second": 9.895,
      "eval_steps_per_second": 2.522,
      "step": 120
    },
    {
      "epoch": 0.4229607250755287,
      "grad_norm": 8.9187593460083,
      "learning_rate": 1.8308157099697886e-05,
      "loss": 0.4342,
      "step": 140
    },
    {
      "epoch": 0.4229607250755287,
      "eval_accuracy": 0.8562091503267973,
      "eval_loss": 0.4258342981338501,
      "eval_runtime": 15.4753,
      "eval_samples_per_second": 9.887,
      "eval_steps_per_second": 2.52,
      "step": 140
    },
    {
      "epoch": 0.48338368580060426,
      "grad_norm": 40.476078033447266,
      "learning_rate": 1.8066465256797586e-05,
      "loss": 0.4915,
      "step": 160
    },
    {
      "epoch": 0.48338368580060426,
      "eval_accuracy": 0.8300653594771242,
      "eval_loss": 0.5960604548454285,
      "eval_runtime": 15.4631,
      "eval_samples_per_second": 9.895,
      "eval_steps_per_second": 2.522,
      "step": 160
    },
    {
      "epoch": 0.5438066465256798,
      "grad_norm": 3.8627891540527344,
      "learning_rate": 1.782477341389728e-05,
      "loss": 0.4127,
      "step": 180
    },
    {
      "epoch": 0.5438066465256798,
      "eval_accuracy": 0.869281045751634,
      "eval_loss": 0.2986587882041931,
      "eval_runtime": 15.5147,
      "eval_samples_per_second": 9.862,
      "eval_steps_per_second": 2.514,
      "step": 180
    },
    {
      "epoch": 0.6042296072507553,
      "grad_norm": 16.525339126586914,
      "learning_rate": 1.758308157099698e-05,
      "loss": 0.3166,
      "step": 200
    },
    {
      "epoch": 0.6042296072507553,
      "eval_accuracy": 0.869281045751634,
      "eval_loss": 0.33075031638145447,
      "eval_runtime": 15.5417,
      "eval_samples_per_second": 9.845,
      "eval_steps_per_second": 2.509,
      "step": 200
    },
    {
      "epoch": 0.6646525679758308,
      "grad_norm": 1.5080480575561523,
      "learning_rate": 1.7341389728096677e-05,
      "loss": 0.4018,
      "step": 220
    },
    {
      "epoch": 0.6646525679758308,
      "eval_accuracy": 0.803921568627451,
      "eval_loss": 0.5285586714744568,
      "eval_runtime": 15.4908,
      "eval_samples_per_second": 9.877,
      "eval_steps_per_second": 2.518,
      "step": 220
    },
    {
      "epoch": 0.7250755287009063,
      "grad_norm": 21.941341400146484,
      "learning_rate": 1.7099697885196376e-05,
      "loss": 0.3007,
      "step": 240
    },
    {
      "epoch": 0.7250755287009063,
      "eval_accuracy": 0.8627450980392157,
      "eval_loss": 0.584548830986023,
      "eval_runtime": 15.5027,
      "eval_samples_per_second": 9.869,
      "eval_steps_per_second": 2.516,
      "step": 240
    },
    {
      "epoch": 0.7854984894259819,
      "grad_norm": 7.442047119140625,
      "learning_rate": 1.6858006042296072e-05,
      "loss": 0.4893,
      "step": 260
    },
    {
      "epoch": 0.7854984894259819,
      "eval_accuracy": 0.8627450980392157,
      "eval_loss": 0.36624589562416077,
      "eval_runtime": 15.471,
      "eval_samples_per_second": 9.889,
      "eval_steps_per_second": 2.521,
      "step": 260
    },
    {
      "epoch": 0.8459214501510574,
      "grad_norm": 27.062305450439453,
      "learning_rate": 1.661631419939577e-05,
      "loss": 0.274,
      "step": 280
    },
    {
      "epoch": 0.8459214501510574,
      "eval_accuracy": 0.869281045751634,
      "eval_loss": 0.34829556941986084,
      "eval_runtime": 15.4719,
      "eval_samples_per_second": 9.889,
      "eval_steps_per_second": 2.521,
      "step": 280
    },
    {
      "epoch": 0.9063444108761329,
      "grad_norm": 9.517237663269043,
      "learning_rate": 1.637462235649547e-05,
      "loss": 0.5741,
      "step": 300
    },
    {
      "epoch": 0.9063444108761329,
      "eval_accuracy": 0.8823529411764706,
      "eval_loss": 0.32800009846687317,
      "eval_runtime": 15.4828,
      "eval_samples_per_second": 9.882,
      "eval_steps_per_second": 2.519,
      "step": 300
    },
    {
      "epoch": 0.9667673716012085,
      "grad_norm": 40.94089889526367,
      "learning_rate": 1.6132930513595166e-05,
      "loss": 0.3752,
      "step": 320
    },
    {
      "epoch": 0.9667673716012085,
      "eval_accuracy": 0.8888888888888888,
      "eval_loss": 0.5250552296638489,
      "eval_runtime": 15.485,
      "eval_samples_per_second": 9.881,
      "eval_steps_per_second": 2.519,
      "step": 320
    },
    {
      "epoch": 1.027190332326284,
      "grad_norm": 0.2767094373703003,
      "learning_rate": 1.5891238670694865e-05,
      "loss": 0.2711,
      "step": 340
    },
    {
      "epoch": 1.027190332326284,
      "eval_accuracy": 0.8562091503267973,
      "eval_loss": 0.6096686720848083,
      "eval_runtime": 15.4954,
      "eval_samples_per_second": 9.874,
      "eval_steps_per_second": 2.517,
      "step": 340
    },
    {
      "epoch": 1.0876132930513596,
      "grad_norm": 1.7755597829818726,
      "learning_rate": 1.5649546827794565e-05,
      "loss": 0.2369,
      "step": 360
    },
    {
      "epoch": 1.0876132930513596,
      "eval_accuracy": 0.869281045751634,
      "eval_loss": 0.5457373857498169,
      "eval_runtime": 15.479,
      "eval_samples_per_second": 9.884,
      "eval_steps_per_second": 2.52,
      "step": 360
    },
    {
      "epoch": 1.148036253776435,
      "grad_norm": 197.30111694335938,
      "learning_rate": 1.540785498489426e-05,
      "loss": 0.3756,
      "step": 380
    },
    {
      "epoch": 1.148036253776435,
      "eval_accuracy": 0.8758169934640523,
      "eval_loss": 0.6890403628349304,
      "eval_runtime": 15.448,
      "eval_samples_per_second": 9.904,
      "eval_steps_per_second": 2.525,
      "step": 380
    },
    {
      "epoch": 1.2084592145015105,
      "grad_norm": 0.16098028421401978,
      "learning_rate": 1.516616314199396e-05,
      "loss": 0.6575,
      "step": 400
    },
    {
      "epoch": 1.2084592145015105,
      "eval_accuracy": 0.869281045751634,
      "eval_loss": 0.47085481882095337,
      "eval_runtime": 15.4824,
      "eval_samples_per_second": 9.882,
      "eval_steps_per_second": 2.519,
      "step": 400
    },
    {
      "epoch": 1.2688821752265862,
      "grad_norm": 0.3975774943828583,
      "learning_rate": 1.4924471299093657e-05,
      "loss": 0.3268,
      "step": 420
    },
    {
      "epoch": 1.2688821752265862,
      "eval_accuracy": 0.8496732026143791,
      "eval_loss": 0.5218892693519592,
      "eval_runtime": 15.4867,
      "eval_samples_per_second": 9.879,
      "eval_steps_per_second": 2.518,
      "step": 420
    },
    {
      "epoch": 1.3293051359516617,
      "grad_norm": 8.90912914276123,
      "learning_rate": 1.4682779456193355e-05,
      "loss": 0.3994,
      "step": 440
    },
    {
      "epoch": 1.3293051359516617,
      "eval_accuracy": 0.869281045751634,
      "eval_loss": 0.42816421389579773,
      "eval_runtime": 15.4691,
      "eval_samples_per_second": 9.891,
      "eval_steps_per_second": 2.521,
      "step": 440
    },
    {
      "epoch": 1.3897280966767371,
      "grad_norm": 93.57767486572266,
      "learning_rate": 1.4441087613293053e-05,
      "loss": 0.0879,
      "step": 460
    },
    {
      "epoch": 1.3897280966767371,
      "eval_accuracy": 0.8758169934640523,
      "eval_loss": 0.6293966174125671,
      "eval_runtime": 15.4702,
      "eval_samples_per_second": 9.89,
      "eval_steps_per_second": 2.521,
      "step": 460
    },
    {
      "epoch": 1.4501510574018126,
      "grad_norm": 0.061734456568956375,
      "learning_rate": 1.419939577039275e-05,
      "loss": 0.2566,
      "step": 480
    },
    {
      "epoch": 1.4501510574018126,
      "eval_accuracy": 0.8627450980392157,
      "eval_loss": 0.7143150568008423,
      "eval_runtime": 15.449,
      "eval_samples_per_second": 9.904,
      "eval_steps_per_second": 2.524,
      "step": 480
    },
    {
      "epoch": 1.510574018126888,
      "grad_norm": 0.11357846111059189,
      "learning_rate": 1.3957703927492448e-05,
      "loss": 0.2897,
      "step": 500
    },
    {
      "epoch": 1.510574018126888,
      "eval_accuracy": 0.869281045751634,
      "eval_loss": 0.6120204329490662,
      "eval_runtime": 15.463,
      "eval_samples_per_second": 9.895,
      "eval_steps_per_second": 2.522,
      "step": 500
    }
  ],
  "logging_steps": 20,
  "max_steps": 1655,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1604904119249676.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}