{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.088803088803089,
  "eval_steps": 200,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15444015444015444,
      "eval_loss": Infinity,
      "eval_runtime": 182.0736,
      "eval_samples_per_second": 38.567,
      "eval_steps_per_second": 4.822,
      "eval_wer": 1.0,
      "step": 200
    },
    {
      "epoch": 0.3088803088803089,
      "eval_loss": Infinity,
      "eval_runtime": 177.405,
      "eval_samples_per_second": 39.582,
      "eval_steps_per_second": 4.949,
      "eval_wer": 0.8236661342890161,
      "step": 400
    },
    {
      "epoch": 0.3861003861003861,
      "grad_norm": 3.3234875202178955,
      "learning_rate": 0.00029699999999999996,
      "loss": 3.6679,
      "step": 500
    },
    {
      "epoch": 0.46332046332046334,
      "eval_loss": Infinity,
      "eval_runtime": 177.3768,
      "eval_samples_per_second": 39.588,
      "eval_steps_per_second": 4.95,
      "eval_wer": 0.7189305701634789,
      "step": 600
    },
    {
      "epoch": 0.6177606177606177,
      "eval_loss": Infinity,
      "eval_runtime": 177.3999,
      "eval_samples_per_second": 39.583,
      "eval_steps_per_second": 4.949,
      "eval_wer": 0.5541142404529106,
      "step": 800
    },
    {
      "epoch": 0.7722007722007722,
      "grad_norm": 3.4127838611602783,
      "learning_rate": 0.0002577428571428571,
      "loss": 0.8341,
      "step": 1000
    },
    {
      "epoch": 0.7722007722007722,
      "eval_loss": Infinity,
      "eval_runtime": 178.223,
      "eval_samples_per_second": 39.4,
      "eval_steps_per_second": 4.926,
      "eval_wer": 0.512381026579928,
      "step": 1000
    },
    {
      "epoch": 0.9266409266409267,
      "eval_loss": Infinity,
      "eval_runtime": 177.4709,
      "eval_samples_per_second": 39.567,
      "eval_steps_per_second": 4.947,
      "eval_wer": 0.489462817965798,
      "step": 1200
    },
    {
      "epoch": 1.0810810810810811,
      "eval_loss": Infinity,
      "eval_runtime": 178.0824,
      "eval_samples_per_second": 39.431,
      "eval_steps_per_second": 4.93,
      "eval_wer": 0.45271577525872253,
      "step": 1400
    },
    {
      "epoch": 1.1583011583011582,
      "grad_norm": 0.5421157479286194,
      "learning_rate": 0.00021488571428571426,
      "loss": 0.6524,
      "step": 1500
    },
    {
      "epoch": 1.2355212355212355,
      "eval_loss": Infinity,
      "eval_runtime": 176.6316,
      "eval_samples_per_second": 39.755,
      "eval_steps_per_second": 4.971,
      "eval_wer": 0.4340436030280602,
      "step": 1600
    },
    {
      "epoch": 1.3899613899613898,
      "eval_loss": Infinity,
      "eval_runtime": 177.018,
      "eval_samples_per_second": 39.668,
      "eval_steps_per_second": 4.96,
      "eval_wer": 0.40710009998311975,
      "step": 1800
    },
    {
      "epoch": 1.5444015444015444,
      "grad_norm": 0.6360776424407959,
      "learning_rate": 0.0001721142857142857,
      "loss": 0.5566,
      "step": 2000
    },
    {
      "epoch": 1.5444015444015444,
      "eval_loss": Infinity,
      "eval_runtime": 176.5522,
      "eval_samples_per_second": 39.773,
      "eval_steps_per_second": 4.973,
      "eval_wer": 0.3941931881630374,
      "step": 2000
    },
    {
      "epoch": 1.698841698841699,
      "eval_loss": Infinity,
      "eval_runtime": 177.2873,
      "eval_samples_per_second": 39.608,
      "eval_steps_per_second": 4.952,
      "eval_wer": 0.39070027138275354,
      "step": 2200
    },
    {
      "epoch": 1.8532818532818531,
      "eval_loss": Infinity,
      "eval_runtime": 177.1067,
      "eval_samples_per_second": 39.648,
      "eval_steps_per_second": 4.957,
      "eval_wer": 0.37781932920416034,
      "step": 2400
    },
    {
      "epoch": 1.9305019305019306,
      "grad_norm": 3.162639617919922,
      "learning_rate": 0.0001295142857142857,
      "loss": 0.5324,
      "step": 2500
    },
    {
      "epoch": 2.0077220077220077,
      "eval_loss": Infinity,
      "eval_runtime": 177.9344,
      "eval_samples_per_second": 39.464,
      "eval_steps_per_second": 4.934,
      "eval_wer": 0.3624453014426136,
      "step": 2600
    },
    {
      "epoch": 2.1621621621621623,
      "eval_loss": Infinity,
      "eval_runtime": 178.1586,
      "eval_samples_per_second": 39.414,
      "eval_steps_per_second": 4.928,
      "eval_wer": 0.35340786620440706,
      "step": 2800
    },
    {
      "epoch": 2.3166023166023164,
      "grad_norm": 0.8761777877807617,
      "learning_rate": 8.674285714285714e-05,
      "loss": 0.4444,
      "step": 3000
    },
    {
      "epoch": 2.3166023166023164,
      "eval_loss": Infinity,
      "eval_runtime": 177.4921,
      "eval_samples_per_second": 39.562,
      "eval_steps_per_second": 4.947,
      "eval_wer": 0.34959032890550945,
      "step": 3000
    },
    {
      "epoch": 2.471042471042471,
      "eval_loss": Infinity,
      "eval_runtime": 177.8631,
      "eval_samples_per_second": 39.48,
      "eval_steps_per_second": 4.936,
      "eval_wer": 0.3415916793268669,
      "step": 3200
    },
    {
      "epoch": 2.6254826254826256,
      "eval_loss": Infinity,
      "eval_runtime": 177.4214,
      "eval_samples_per_second": 39.578,
      "eval_steps_per_second": 4.949,
      "eval_wer": 0.3406697570540039,
      "step": 3400
    },
    {
      "epoch": 2.7027027027027026,
      "grad_norm": 3.2265968322753906,
      "learning_rate": 4.405714285714285e-05,
      "loss": 0.4254,
      "step": 3500
    },
    {
      "epoch": 2.7799227799227797,
      "eval_loss": Infinity,
      "eval_runtime": 178.3037,
      "eval_samples_per_second": 39.382,
      "eval_steps_per_second": 4.924,
      "eval_wer": 0.3346448002285328,
      "step": 3600
    },
    {
      "epoch": 2.9343629343629343,
      "eval_loss": Infinity,
      "eval_runtime": 177.742,
      "eval_samples_per_second": 39.507,
      "eval_steps_per_second": 4.94,
      "eval_wer": 0.33778712684871387,
      "step": 3800
    },
    {
      "epoch": 3.088803088803089,
      "grad_norm": 3.3560454845428467,
      "learning_rate": 1.2857142857142856e-06,
      "loss": 0.4273,
      "step": 4000
    },
    {
      "epoch": 3.088803088803089,
      "eval_loss": Infinity,
      "eval_runtime": 177.2694,
      "eval_samples_per_second": 39.612,
      "eval_steps_per_second": 4.953,
      "eval_wer": 0.33521613234129305,
      "step": 4000
    },
    {
      "epoch": 3.088803088803089,
      "step": 4000,
      "total_flos": 1.6887033849357677e+19,
      "train_loss": 0.9425676002502441,
      "train_runtime": 7915.8947,
      "train_samples_per_second": 16.17,
      "train_steps_per_second": 0.505
    }
  ],
  "logging_steps": 500,
  "max_steps": 4000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.6887033849357677e+19,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}