{
  "best_global_step": 945,
  "best_metric": 0.6100375652313232,
  "best_model_checkpoint": "./t5_req_sum_checkpoints/checkpoint-945",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 945,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.372006510113927,
      "grad_norm": 60.87726593017578,
      "learning_rate": 1.76e-05,
      "loss": 73.3444482421875,
      "step": 50
    },
    {
      "epoch": 0.744013020227854,
      "grad_norm": 53.28485107421875,
      "learning_rate": 3.76e-05,
      "loss": 52.544169921875,
      "step": 100
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.1164566278457642,
      "eval_runtime": 29.3037,
      "eval_samples_per_second": 16.312,
      "eval_steps_per_second": 16.312,
      "step": 135
    },
    {
      "epoch": 1.111601953034178,
      "grad_norm": 43.34605407714844,
      "learning_rate": 5.76e-05,
      "loss": 40.7950439453125,
      "step": 150
    },
    {
      "epoch": 1.483608463148105,
      "grad_norm": 49.22634506225586,
      "learning_rate": 7.76e-05,
      "loss": 34.67097412109375,
      "step": 200
    },
    {
      "epoch": 1.855614973262032,
      "grad_norm": 47.0002555847168,
      "learning_rate": 9.76e-05,
      "loss": 32.54794921875,
      "step": 250
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.8499057292938232,
      "eval_runtime": 29.4986,
      "eval_samples_per_second": 16.204,
      "eval_steps_per_second": 16.204,
      "step": 270
    },
    {
      "epoch": 2.223203906068356,
      "grad_norm": 38.79513931274414,
      "learning_rate": 0.0001176,
      "loss": 28.51433837890625,
      "step": 300
    },
    {
      "epoch": 2.595210416182283,
      "grad_norm": 42.362369537353516,
      "learning_rate": 0.00013759999999999998,
      "loss": 27.0824365234375,
      "step": 350
    },
    {
      "epoch": 2.96721692629621,
      "grad_norm": 42.038055419921875,
      "learning_rate": 0.0001576,
      "loss": 25.14374267578125,
      "step": 400
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.7282077670097351,
      "eval_runtime": 29.821,
      "eval_samples_per_second": 16.029,
      "eval_steps_per_second": 16.029,
      "step": 405
    },
    {
      "epoch": 3.3348058591025342,
      "grad_norm": 36.55916976928711,
      "learning_rate": 0.0001776,
      "loss": 22.50442138671875,
      "step": 450
    },
    {
      "epoch": 3.706812369216461,
      "grad_norm": 37.239654541015625,
      "learning_rate": 0.0001976,
      "loss": 22.65784423828125,
      "step": 500
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.6741604804992676,
      "eval_runtime": 29.5681,
      "eval_samples_per_second": 16.166,
      "eval_steps_per_second": 16.166,
      "step": 540
    },
    {
      "epoch": 4.074401302022785,
      "grad_norm": 50.11177444458008,
      "learning_rate": 0.00018482758620689654,
      "loss": 21.0756591796875,
      "step": 550
    },
    {
      "epoch": 4.446407812136712,
      "grad_norm": 46.11113739013672,
      "learning_rate": 0.00016758620689655173,
      "loss": 19.58232177734375,
      "step": 600
    },
    {
      "epoch": 4.818414322250639,
      "grad_norm": 38.068660736083984,
      "learning_rate": 0.0001506896551724138,
      "loss": 19.37177734375,
      "step": 650
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.6635456085205078,
      "eval_runtime": 29.0874,
      "eval_samples_per_second": 16.433,
      "eval_steps_per_second": 16.433,
      "step": 675
    },
    {
      "epoch": 5.1860032550569635,
      "grad_norm": 45.14381790161133,
      "learning_rate": 0.00013379310344827588,
      "loss": 17.655869140625,
      "step": 700
    },
    {
      "epoch": 5.55800976517089,
      "grad_norm": 111.4928970336914,
      "learning_rate": 0.00011689655172413793,
      "loss": 16.826263427734375,
      "step": 750
    },
    {
      "epoch": 5.930016275284817,
      "grad_norm": 41.3922004699707,
      "learning_rate": 0.00010103448275862069,
      "loss": 18.38109130859375,
      "step": 800
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.6126712560653687,
      "eval_runtime": 28.3014,
      "eval_samples_per_second": 16.89,
      "eval_steps_per_second": 16.89,
      "step": 810
    },
    {
      "epoch": 6.2976052080911415,
      "grad_norm": 39.92831802368164,
      "learning_rate": 8.413793103448277e-05,
      "loss": 15.30945556640625,
      "step": 850
    },
    {
      "epoch": 6.6696117182050685,
      "grad_norm": 52.849761962890625,
      "learning_rate": 6.689655172413794e-05,
      "loss": 15.41493896484375,
      "step": 900
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.6100375652313232,
      "eval_runtime": 27.7122,
      "eval_samples_per_second": 17.249,
      "eval_steps_per_second": 17.249,
      "step": 945
    }
  ],
  "logging_steps": 50,
  "max_steps": 1080,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3713190910571520.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}