{
  "best_global_step": 1000,
  "best_metric": 9.673482963989954e-06,
  "best_model_checkpoint": "./medical-qa-t5-lora/checkpoint-1000",
  "epoch": 333.4,
  "eval_steps": 50,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 16.8,
      "grad_norm": 1.3040547370910645,
      "learning_rate": 4.8e-05,
      "loss": 2.3794,
      "step": 50
    },
    {
      "epoch": 16.8,
      "eval_loss": 1.9908905029296875,
      "eval_runtime": 0.5391,
      "eval_samples_per_second": 37.101,
      "eval_steps_per_second": 3.71,
      "step": 50
    },
    {
      "epoch": 33.4,
      "grad_norm": 0.6484032273292542,
      "learning_rate": 9.800000000000001e-05,
      "loss": 1.2119,
      "step": 100
    },
    {
      "epoch": 33.4,
      "eval_loss": 0.44726213812828064,
      "eval_runtime": 0.5249,
      "eval_samples_per_second": 38.1,
      "eval_steps_per_second": 3.81,
      "step": 100
    },
    {
      "epoch": 50.0,
      "grad_norm": 0.19109225273132324,
      "learning_rate": 0.000148,
      "loss": 0.2431,
      "step": 150
    },
    {
      "epoch": 50.0,
      "eval_loss": 0.00484898965805769,
      "eval_runtime": 0.5384,
      "eval_samples_per_second": 37.145,
      "eval_steps_per_second": 3.714,
      "step": 150
    },
    {
      "epoch": 66.8,
      "grad_norm": 0.2626807987689972,
      "learning_rate": 0.00019800000000000002,
      "loss": 0.0343,
      "step": 200
    },
    {
      "epoch": 66.8,
      "eval_loss": 0.0008122542640194297,
      "eval_runtime": 0.5393,
      "eval_samples_per_second": 37.082,
      "eval_steps_per_second": 3.708,
      "step": 200
    },
    {
      "epoch": 83.4,
      "grad_norm": 0.12434151768684387,
      "learning_rate": 0.000248,
      "loss": 0.0118,
      "step": 250
    },
    {
      "epoch": 83.4,
      "eval_loss": 0.00033435432123951614,
      "eval_runtime": 0.5325,
      "eval_samples_per_second": 37.562,
      "eval_steps_per_second": 3.756,
      "step": 250
    },
    {
      "epoch": 100.0,
      "grad_norm": 0.09751362353563309,
      "learning_rate": 0.000298,
      "loss": 0.0068,
      "step": 300
    },
    {
      "epoch": 100.0,
      "eval_loss": 0.00017693254631012678,
      "eval_runtime": 0.5227,
      "eval_samples_per_second": 38.26,
      "eval_steps_per_second": 3.826,
      "step": 300
    },
    {
      "epoch": 116.8,
      "grad_norm": 0.046444371342659,
      "learning_rate": 0.000348,
      "loss": 0.0042,
      "step": 350
    },
    {
      "epoch": 116.8,
      "eval_loss": 0.000103985563328024,
      "eval_runtime": 0.5278,
      "eval_samples_per_second": 37.893,
      "eval_steps_per_second": 3.789,
      "step": 350
    },
    {
      "epoch": 133.4,
      "grad_norm": 0.07410682737827301,
      "learning_rate": 0.000398,
      "loss": 0.0028,
      "step": 400
    },
    {
      "epoch": 133.4,
      "eval_loss": 6.64430481265299e-05,
      "eval_runtime": 0.5314,
      "eval_samples_per_second": 37.637,
      "eval_steps_per_second": 3.764,
      "step": 400
    },
    {
      "epoch": 150.0,
      "grad_norm": 0.010288752615451813,
      "learning_rate": 0.000448,
      "loss": 0.002,
      "step": 450
    },
    {
      "epoch": 150.0,
      "eval_loss": 4.376208016765304e-05,
      "eval_runtime": 0.5212,
      "eval_samples_per_second": 38.37,
      "eval_steps_per_second": 3.837,
      "step": 450
    },
    {
      "epoch": 166.8,
      "grad_norm": 0.012730836868286133,
      "learning_rate": 0.000498,
      "loss": 0.0015,
      "step": 500
    },
    {
      "epoch": 166.8,
      "eval_loss": 3.529857349349186e-05,
      "eval_runtime": 0.5303,
      "eval_samples_per_second": 37.714,
      "eval_steps_per_second": 3.771,
      "step": 500
    },
    {
      "epoch": 183.4,
      "grad_norm": 0.016046803444623947,
      "learning_rate": 0.00045200000000000004,
      "loss": 0.0012,
      "step": 550
    },
    {
      "epoch": 183.4,
      "eval_loss": 2.306454371137079e-05,
      "eval_runtime": 0.548,
      "eval_samples_per_second": 36.494,
      "eval_steps_per_second": 3.649,
      "step": 550
    },
    {
      "epoch": 200.0,
      "grad_norm": 0.007060339208692312,
      "learning_rate": 0.000402,
      "loss": 0.0017,
      "step": 600
    },
    {
      "epoch": 200.0,
      "eval_loss": 2.7466521714814007e-05,
      "eval_runtime": 0.5403,
      "eval_samples_per_second": 37.017,
      "eval_steps_per_second": 3.702,
      "step": 600
    },
    {
      "epoch": 216.8,
      "grad_norm": 0.008236071094870567,
      "learning_rate": 0.000352,
      "loss": 0.0012,
      "step": 650
    },
    {
      "epoch": 216.8,
      "eval_loss": 2.4082381059997715e-05,
      "eval_runtime": 0.5485,
      "eval_samples_per_second": 36.462,
      "eval_steps_per_second": 3.646,
      "step": 650
    },
    {
      "epoch": 233.4,
      "grad_norm": 0.011439572088420391,
      "learning_rate": 0.000302,
      "loss": 0.0008,
      "step": 700
    },
    {
      "epoch": 233.4,
      "eval_loss": 1.633342617424205e-05,
      "eval_runtime": 0.5505,
      "eval_samples_per_second": 36.328,
      "eval_steps_per_second": 3.633,
      "step": 700
    },
    {
      "epoch": 250.0,
      "grad_norm": 0.0028414344415068626,
      "learning_rate": 0.000252,
      "loss": 0.0006,
      "step": 750
    },
    {
      "epoch": 250.0,
      "eval_loss": 1.5031941074994393e-05,
      "eval_runtime": 0.5392,
      "eval_samples_per_second": 37.091,
      "eval_steps_per_second": 3.709,
      "step": 750
    },
    {
      "epoch": 266.8,
      "grad_norm": 0.008099627681076527,
      "learning_rate": 0.000202,
      "loss": 0.0006,
      "step": 800
    },
    {
      "epoch": 266.8,
      "eval_loss": 1.2507860446930863e-05,
      "eval_runtime": 0.5524,
      "eval_samples_per_second": 36.206,
      "eval_steps_per_second": 3.621,
      "step": 800
    },
    {
      "epoch": 283.4,
      "grad_norm": 0.002751375548541546,
      "learning_rate": 0.000152,
      "loss": 0.0004,
      "step": 850
    },
    {
      "epoch": 283.4,
      "eval_loss": 1.0678697435650975e-05,
      "eval_runtime": 0.5562,
      "eval_samples_per_second": 35.957,
      "eval_steps_per_second": 3.596,
      "step": 850
    },
    {
      "epoch": 300.0,
      "grad_norm": 0.002928754547610879,
      "learning_rate": 0.000102,
      "loss": 0.0004,
      "step": 900
    },
    {
      "epoch": 300.0,
      "eval_loss": 1.017556496663019e-05,
      "eval_runtime": 0.5705,
      "eval_samples_per_second": 35.059,
      "eval_steps_per_second": 3.506,
      "step": 900
    },
    {
      "epoch": 316.8,
      "grad_norm": 0.010121754370629787,
      "learning_rate": 5.2e-05,
      "loss": 0.0004,
      "step": 950
    },
    {
      "epoch": 316.8,
      "eval_loss": 9.748164302436635e-06,
      "eval_runtime": 0.5949,
      "eval_samples_per_second": 33.621,
      "eval_steps_per_second": 3.362,
      "step": 950
    },
    {
      "epoch": 333.4,
      "grad_norm": 0.003771366784349084,
      "learning_rate": 2e-06,
      "loss": 0.0004,
      "step": 1000
    },
    {
      "epoch": 333.4,
      "eval_loss": 9.673482963989954e-06,
      "eval_runtime": 0.5717,
      "eval_samples_per_second": 34.981,
      "eval_steps_per_second": 3.498,
      "step": 1000
    },
    {
      "epoch": 333.4,
      "step": 1000,
      "total_flos": 520403538935808.0,
      "train_loss": 0.1952769088242203,
      "train_runtime": 949.4346,
      "train_samples_per_second": 42.13,
      "train_steps_per_second": 1.053
    }
  ],
  "logging_steps": 50,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 500,
  "save_steps": 100,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 520403538935808.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}