{
  "best_metric": 0.4415583312511444,
  "best_model_checkpoint": "output_pipe/tf0/origin/checkpoint-200",
  "epoch": 4.0,
  "eval_steps": 200,
  "global_step": 2024,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1976284584980237,
      "grad_norm": 1.5848366022109985,
      "learning_rate": 2.9240121580547114e-05,
      "loss": 0.5377,
      "step": 100
    },
    {
      "epoch": 0.3952569169960474,
      "grad_norm": 1.709352731704712,
      "learning_rate": 2.7720364741641338e-05,
      "loss": 0.4727,
      "step": 200
    },
    {
      "epoch": 0.3952569169960474,
      "eval_accuracy": 0.796,
      "eval_f1": 0.7955288985823337,
      "eval_loss": 0.4415583312511444,
      "eval_matthews_correlation": 0.591370962853321,
      "eval_precision": 0.7960841474628472,
      "eval_recall": 0.7952873521791191,
      "eval_runtime": 0.2556,
      "eval_samples_per_second": 3912.889,
      "eval_steps_per_second": 62.606,
      "step": 200
    },
    {
      "epoch": 0.5928853754940712,
      "grad_norm": 1.3577654361724854,
      "learning_rate": 2.6200607902735562e-05,
      "loss": 0.4454,
      "step": 300
    },
    {
      "epoch": 0.7905138339920948,
      "grad_norm": 2.5865519046783447,
      "learning_rate": 2.4680851063829786e-05,
      "loss": 0.4305,
      "step": 400
    },
    {
      "epoch": 0.7905138339920948,
      "eval_accuracy": 0.796,
      "eval_f1": 0.7928143128747623,
      "eval_loss": 0.4495840072631836,
      "eval_matthews_correlation": 0.5989374522199814,
      "eval_precision": 0.8063936013682609,
      "eval_recall": 0.7927003616180304,
      "eval_runtime": 0.2543,
      "eval_samples_per_second": 3932.409,
      "eval_steps_per_second": 62.919,
      "step": 400
    },
    {
      "epoch": 0.9881422924901185,
      "grad_norm": 0.9050748944282532,
      "learning_rate": 2.3161094224924013e-05,
      "loss": 0.4465,
      "step": 500
    },
    {
      "epoch": 1.1857707509881423,
      "grad_norm": 1.9691131114959717,
      "learning_rate": 2.1641337386018237e-05,
      "loss": 0.3415,
      "step": 600
    },
    {
      "epoch": 1.1857707509881423,
      "eval_accuracy": 0.796,
      "eval_f1": 0.7948767450559262,
      "eval_loss": 0.45666906237602234,
      "eval_matthews_correlation": 0.5923257505144678,
      "eval_precision": 0.7979339496091201,
      "eval_recall": 0.7944023290924309,
      "eval_runtime": 0.2545,
      "eval_samples_per_second": 3928.505,
      "eval_steps_per_second": 62.856,
      "step": 600
    },
    {
      "epoch": 1.383399209486166,
      "grad_norm": 1.9345086812973022,
      "learning_rate": 2.012158054711246e-05,
      "loss": 0.3257,
      "step": 700
    },
    {
      "epoch": 1.5810276679841897,
      "grad_norm": 4.1219024658203125,
      "learning_rate": 1.8601823708206688e-05,
      "loss": 0.3225,
      "step": 800
    },
    {
      "epoch": 1.5810276679841897,
      "eval_accuracy": 0.803,
      "eval_f1": 0.8015663166377074,
      "eval_loss": 0.44956839084625244,
      "eval_matthews_correlation": 0.6073670430930368,
      "eval_precision": 0.8063543296817944,
      "eval_recall": 0.8010359976132408,
      "eval_runtime": 0.2534,
      "eval_samples_per_second": 3946.161,
      "eval_steps_per_second": 63.139,
      "step": 800
    },
    {
      "epoch": 1.7786561264822134,
      "grad_norm": 2.96213436126709,
      "learning_rate": 1.7082066869300912e-05,
      "loss": 0.3249,
      "step": 900
    },
    {
      "epoch": 1.9762845849802373,
      "grad_norm": 2.0221433639526367,
      "learning_rate": 1.5577507598784196e-05,
      "loss": 0.3245,
      "step": 1000
    },
    {
      "epoch": 1.9762845849802373,
      "eval_accuracy": 0.799,
      "eval_f1": 0.7988308167168588,
      "eval_loss": 0.4458507001399994,
      "eval_matthews_correlation": 0.5977016347708501,
      "eval_precision": 0.7987640881147541,
      "eval_recall": 0.7989375718330389,
      "eval_runtime": 0.254,
      "eval_samples_per_second": 3937.334,
      "eval_steps_per_second": 62.997,
      "step": 1000
    },
    {
      "epoch": 2.1739130434782608,
      "grad_norm": 6.168671131134033,
      "learning_rate": 1.405775075987842e-05,
      "loss": 0.1443,
      "step": 1100
    },
    {
      "epoch": 2.3715415019762847,
      "grad_norm": 4.112609386444092,
      "learning_rate": 1.2537993920972645e-05,
      "loss": 0.1103,
      "step": 1200
    },
    {
      "epoch": 2.3715415019762847,
      "eval_accuracy": 0.78,
      "eval_f1": 0.778510705987057,
      "eval_loss": 0.6606701612472534,
      "eval_matthews_correlation": 0.5606591610125751,
      "eval_precision": 0.7825653544908961,
      "eval_recall": 0.7781114968904053,
      "eval_runtime": 0.2532,
      "eval_samples_per_second": 3949.431,
      "eval_steps_per_second": 63.191,
      "step": 1200
    },
    {
      "epoch": 2.5691699604743086,
      "grad_norm": 2.68495512008667,
      "learning_rate": 1.101823708206687e-05,
      "loss": 0.1111,
      "step": 1300
    },
    {
      "epoch": 2.766798418972332,
      "grad_norm": 7.105761528015137,
      "learning_rate": 9.498480243161095e-06,
      "loss": 0.1078,
      "step": 1400
    },
    {
      "epoch": 2.766798418972332,
      "eval_accuracy": 0.799,
      "eval_f1": 0.7989660252582687,
      "eval_loss": 0.6632059812545776,
      "eval_matthews_correlation": 0.6014707475917296,
      "eval_precision": 0.801103974307507,
      "eval_recall": 0.8003672245115353,
      "eval_runtime": 0.2539,
      "eval_samples_per_second": 3938.144,
      "eval_steps_per_second": 63.01,
      "step": 1400
    },
    {
      "epoch": 2.9644268774703555,
      "grad_norm": 5.618806838989258,
      "learning_rate": 7.978723404255319e-06,
      "loss": 0.1095,
      "step": 1500
    },
    {
      "epoch": 3.1620553359683794,
      "grad_norm": 0.052807554602622986,
      "learning_rate": 6.458966565349544e-06,
      "loss": 0.0384,
      "step": 1600
    },
    {
      "epoch": 3.1620553359683794,
      "eval_accuracy": 0.795,
      "eval_f1": 0.7947485669945684,
      "eval_loss": 0.958301842212677,
      "eval_matthews_correlation": 0.589498726885206,
      "eval_precision": 0.7947700219484451,
      "eval_recall": 0.7947287063845806,
      "eval_runtime": 0.253,
      "eval_samples_per_second": 3953.143,
      "eval_steps_per_second": 63.25,
      "step": 1600
    },
    {
      "epoch": 3.3596837944664033,
      "grad_norm": 0.5183179974555969,
      "learning_rate": 4.939209726443769e-06,
      "loss": 0.0242,
      "step": 1700
    },
    {
      "epoch": 3.5573122529644268,
      "grad_norm": 0.16107724606990814,
      "learning_rate": 3.419452887537994e-06,
      "loss": 0.0172,
      "step": 1800
    },
    {
      "epoch": 3.5573122529644268,
      "eval_accuracy": 0.794,
      "eval_f1": 0.7937326775501049,
      "eval_loss": 1.0012874603271484,
      "eval_matthews_correlation": 0.5874717193339695,
      "eval_precision": 0.7937782157435336,
      "eval_recall": 0.79369350969721,
      "eval_runtime": 0.2538,
      "eval_samples_per_second": 3940.36,
      "eval_steps_per_second": 63.046,
      "step": 1800
    },
    {
      "epoch": 3.7549407114624507,
      "grad_norm": 0.045861050486564636,
      "learning_rate": 1.899696048632219e-06,
      "loss": 0.0141,
      "step": 1900
    },
    {
      "epoch": 3.9525691699604746,
      "grad_norm": 0.3307662308216095,
      "learning_rate": 3.7993920972644377e-07,
      "loss": 0.0123,
      "step": 2000
    },
    {
      "epoch": 3.9525691699604746,
      "eval_accuracy": 0.789,
      "eval_f1": 0.7885718580124752,
      "eval_loss": 1.0535693168640137,
      "eval_matthews_correlation": 0.5773352449673653,
      "eval_precision": 0.7889541602465331,
      "eval_recall": 0.788381368862405,
      "eval_runtime": 0.2532,
      "eval_samples_per_second": 3948.915,
      "eval_steps_per_second": 63.183,
      "step": 2000
    },
    {
      "epoch": 4.0,
      "step": 2024,
      "total_flos": 8718835559772720.0,
      "train_loss": 0.2305071633088259,
      "train_runtime": 175.4879,
      "train_samples_per_second": 738.011,
      "train_steps_per_second": 11.534
    }
  ],
  "logging_steps": 100,
  "max_steps": 2024,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8718835559772720.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}