{
  "best_global_step": 36,
  "best_metric": 50.0,
  "best_model_checkpoint": "/content/wolof-agri-asr/scripts/../models/whisper-wolof-agri/checkpoint-36",
  "epoch": 14.0,
  "eval_steps": 500,
  "global_step": 42,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 3.1818084716796875,
      "eval_runtime": 6.863,
      "eval_samples_per_second": 0.583,
      "eval_steps_per_second": 0.146,
      "eval_wer": 193.33333333333334,
      "step": 3
    },
    {
      "epoch": 1.8,
      "grad_norm": 5.9897379875183105,
      "learning_rate": 0.0009111111111111111,
      "loss": 5.081877136230469,
      "step": 5
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.4930508136749268,
      "eval_runtime": 5.0686,
      "eval_samples_per_second": 0.789,
      "eval_steps_per_second": 0.197,
      "eval_wer": 60.0,
      "step": 6
    },
    {
      "epoch": 3.0,
      "eval_loss": 3.6675028800964355,
      "eval_runtime": 4.4694,
      "eval_samples_per_second": 0.895,
      "eval_steps_per_second": 0.224,
      "eval_wer": 56.666666666666664,
      "step": 9
    },
    {
      "epoch": 3.4,
      "grad_norm": Infinity,
      "learning_rate": 0.0008,
      "loss": 3.032562828063965,
      "step": 10
    },
    {
      "epoch": 4.0,
      "eval_loss": 3.6675028800964355,
      "eval_runtime": 2.1332,
      "eval_samples_per_second": 1.875,
      "eval_steps_per_second": 0.469,
      "eval_wer": 56.666666666666664,
      "step": 12
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.8512147665023804,
      "learning_rate": 0.000688888888888889,
      "loss": 2.990604019165039,
      "step": 15
    },
    {
      "epoch": 5.0,
      "eval_loss": 2.268052577972412,
      "eval_runtime": 2.4621,
      "eval_samples_per_second": 1.625,
      "eval_steps_per_second": 0.406,
      "eval_wer": 66.66666666666666,
      "step": 15
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.2226436138153076,
      "eval_runtime": 2.5158,
      "eval_samples_per_second": 1.59,
      "eval_steps_per_second": 0.397,
      "eval_wer": 63.33333333333333,
      "step": 18
    },
    {
      "epoch": 6.8,
      "grad_norm": 1.5517808198928833,
      "learning_rate": 0.0005777777777777778,
      "loss": 1.6542428970336913,
      "step": 20
    },
    {
      "epoch": 7.0,
      "eval_loss": 2.1684648990631104,
      "eval_runtime": 2.6024,
      "eval_samples_per_second": 1.537,
      "eval_steps_per_second": 0.384,
      "eval_wer": 66.66666666666666,
      "step": 21
    },
    {
      "epoch": 8.0,
      "eval_loss": 2.1176955699920654,
      "eval_runtime": 2.6084,
      "eval_samples_per_second": 1.534,
      "eval_steps_per_second": 0.383,
      "eval_wer": 56.666666666666664,
      "step": 24
    },
    {
      "epoch": 8.4,
      "grad_norm": 0.7947890758514404,
      "learning_rate": 0.00046666666666666666,
      "loss": 1.241654109954834,
      "step": 25
    },
    {
      "epoch": 9.0,
      "eval_loss": 2.082947254180908,
      "eval_runtime": 2.6456,
      "eval_samples_per_second": 1.512,
      "eval_steps_per_second": 0.378,
      "eval_wer": 56.666666666666664,
      "step": 27
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.344099760055542,
      "learning_rate": 0.00035555555555555557,
      "loss": 1.1321176528930663,
      "step": 30
    },
    {
      "epoch": 10.0,
      "eval_loss": 2.0475993156433105,
      "eval_runtime": 3.1795,
      "eval_samples_per_second": 1.258,
      "eval_steps_per_second": 0.315,
      "eval_wer": 53.333333333333336,
      "step": 30
    },
    {
      "epoch": 11.0,
      "eval_loss": 2.0215885639190674,
      "eval_runtime": 3.7545,
      "eval_samples_per_second": 1.065,
      "eval_steps_per_second": 0.266,
      "eval_wer": 53.333333333333336,
      "step": 33
    },
    {
      "epoch": 11.8,
      "grad_norm": 0.5112788081169128,
      "learning_rate": 0.00024444444444444443,
      "loss": 1.164208221435547,
      "step": 35
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.9927606582641602,
      "eval_runtime": 2.7185,
      "eval_samples_per_second": 1.471,
      "eval_steps_per_second": 0.368,
      "eval_wer": 50.0,
      "step": 36
    },
    {
      "epoch": 13.0,
      "eval_loss": 1.968162178993225,
      "eval_runtime": 4.261,
      "eval_samples_per_second": 0.939,
      "eval_steps_per_second": 0.235,
      "eval_wer": 53.333333333333336,
      "step": 39
    },
    {
      "epoch": 13.4,
      "grad_norm": 0.5217896103858948,
      "learning_rate": 0.00013333333333333334,
      "loss": 0.9501232147216797,
      "step": 40
    },
    {
      "epoch": 14.0,
      "eval_loss": 1.9518566131591797,
      "eval_runtime": 2.434,
      "eval_samples_per_second": 1.643,
      "eval_steps_per_second": 0.411,
      "eval_wer": 50.0,
      "step": 42
    }
  ],
  "logging_steps": 5,
  "max_steps": 45,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "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": 1.623212752896e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}