trainer_state.json

{
  "best_metric": 4.021976178493541,
  "best_model_checkpoint": "model_whisper_base_atco2\\checkpoint-9108",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 10120,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 12.400457382202148,
      "learning_rate": 1e-05,
      "loss": 0.2212,
      "step": 1012
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.10344767570495605,
      "eval_runtime": 291.7326,
      "eval_samples_per_second": 6.945,
      "eval_steps_per_second": 0.871,
      "eval_wer": 5.93440697869485,
      "step": 1012
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.802471160888672,
      "learning_rate": 1e-05,
      "loss": 0.0619,
      "step": 2024
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.09189248830080032,
      "eval_runtime": 297.8017,
      "eval_samples_per_second": 6.803,
      "eval_steps_per_second": 0.853,
      "eval_wer": 5.183694011071967,
      "step": 2024
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.2378997802734375,
      "learning_rate": 1e-05,
      "loss": 0.0273,
      "step": 3036
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.08819250017404556,
      "eval_runtime": 287.7608,
      "eval_samples_per_second": 7.041,
      "eval_steps_per_second": 0.883,
      "eval_wer": 4.495889951350445,
      "step": 3036
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.339867353439331,
      "learning_rate": 1e-05,
      "loss": 0.0146,
      "step": 4048
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.0916878879070282,
      "eval_runtime": 284.6865,
      "eval_samples_per_second": 7.117,
      "eval_steps_per_second": 0.892,
      "eval_wer": 4.214896829391042,
      "step": 4048
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.930009841918945,
      "learning_rate": 1e-05,
      "loss": 0.0085,
      "step": 5060
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.09865663200616837,
      "eval_runtime": 286.2104,
      "eval_samples_per_second": 7.079,
      "eval_steps_per_second": 0.887,
      "eval_wer": 4.638483475926858,
      "step": 5060
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.15008990466594696,
      "learning_rate": 1e-05,
      "loss": 0.0059,
      "step": 6072
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.09895044565200806,
      "eval_runtime": 288.0034,
      "eval_samples_per_second": 7.035,
      "eval_steps_per_second": 0.882,
      "eval_wer": 4.999161214561315,
      "step": 6072
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.05994468554854393,
      "learning_rate": 1e-05,
      "loss": 0.004,
      "step": 7084
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.10006546229124069,
      "eval_runtime": 285.4295,
      "eval_samples_per_second": 7.098,
      "eval_steps_per_second": 0.89,
      "eval_wer": 4.172957557456803,
      "step": 7084
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.015984803438186646,
      "learning_rate": 1e-05,
      "loss": 0.0039,
      "step": 8096
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.10367220640182495,
      "eval_runtime": 285.9651,
      "eval_samples_per_second": 7.085,
      "eval_steps_per_second": 0.888,
      "eval_wer": 4.663647039087401,
      "step": 8096
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.4414064586162567,
      "learning_rate": 1e-05,
      "loss": 0.0045,
      "step": 9108
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.10489141196012497,
      "eval_runtime": 284.3163,
      "eval_samples_per_second": 7.126,
      "eval_steps_per_second": 0.893,
      "eval_wer": 4.021976178493541,
      "step": 9108
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.01772579923272133,
      "learning_rate": 1e-05,
      "loss": 0.0024,
      "step": 10120
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.11083633452653885,
      "eval_runtime": 285.82,
      "eval_samples_per_second": 7.088,
      "eval_steps_per_second": 0.889,
      "eval_wer": 4.265223955712129,
      "step": 10120
    }
  ],
  "logging_steps": 500,
  "max_steps": 10120,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 5.2484708892672e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}