trainer_state.json

{
  "best_metric": 10.136499555844708,
  "best_model_checkpoint": "./whisper-large-v3-ta/checkpoint-500",
  "epoch": 0.014250292130988685,
  "eval_steps": 500,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0007125146065494343,
      "grad_norm": 3.9452075958251953,
      "learning_rate": 5e-06,
      "loss": 0.5658,
      "step": 25
    },
    {
      "epoch": 0.0014250292130988686,
      "grad_norm": 6.297654151916504,
      "learning_rate": 1e-05,
      "loss": 0.2991,
      "step": 50
    },
    {
      "epoch": 0.002137543819648303,
      "grad_norm": 5.006739139556885,
      "learning_rate": 1e-05,
      "loss": 0.3457,
      "step": 75
    },
    {
      "epoch": 0.002850058426197737,
      "grad_norm": 3.2383642196655273,
      "learning_rate": 1e-05,
      "loss": 0.209,
      "step": 100
    },
    {
      "epoch": 0.003562573032747171,
      "grad_norm": 3.5915024280548096,
      "learning_rate": 1e-05,
      "loss": 0.2085,
      "step": 125
    },
    {
      "epoch": 0.004275087639296606,
      "grad_norm": 3.4095346927642822,
      "learning_rate": 1e-05,
      "loss": 0.1911,
      "step": 150
    },
    {
      "epoch": 0.0049876022458460395,
      "grad_norm": 2.908203363418579,
      "learning_rate": 1e-05,
      "loss": 0.1881,
      "step": 175
    },
    {
      "epoch": 0.005700116852395474,
      "grad_norm": 2.7387442588806152,
      "learning_rate": 1e-05,
      "loss": 0.1907,
      "step": 200
    },
    {
      "epoch": 0.006412631458944908,
      "grad_norm": 3.590416669845581,
      "learning_rate": 1e-05,
      "loss": 0.1678,
      "step": 225
    },
    {
      "epoch": 0.007125146065494342,
      "grad_norm": 2.9250645637512207,
      "learning_rate": 1e-05,
      "loss": 0.1907,
      "step": 250
    },
    {
      "epoch": 0.007837660672043776,
      "grad_norm": 3.180162191390991,
      "learning_rate": 1e-05,
      "loss": 0.178,
      "step": 275
    },
    {
      "epoch": 0.008550175278593211,
      "grad_norm": 3.205806255340576,
      "learning_rate": 1e-05,
      "loss": 0.1698,
      "step": 300
    },
    {
      "epoch": 0.009262689885142646,
      "grad_norm": 3.011608123779297,
      "learning_rate": 1e-05,
      "loss": 0.177,
      "step": 325
    },
    {
      "epoch": 0.009975204491692079,
      "grad_norm": 2.0001132488250732,
      "learning_rate": 1e-05,
      "loss": 0.1689,
      "step": 350
    },
    {
      "epoch": 0.010687719098241514,
      "grad_norm": 2.6594409942626953,
      "learning_rate": 1e-05,
      "loss": 0.1726,
      "step": 375
    },
    {
      "epoch": 0.011400233704790949,
      "grad_norm": 2.706954002380371,
      "learning_rate": 1e-05,
      "loss": 0.1781,
      "step": 400
    },
    {
      "epoch": 0.012112748311340382,
      "grad_norm": 2.7729318141937256,
      "learning_rate": 1e-05,
      "loss": 0.1596,
      "step": 425
    },
    {
      "epoch": 0.012825262917889817,
      "grad_norm": 2.474515914916992,
      "learning_rate": 1e-05,
      "loss": 0.1608,
      "step": 450
    },
    {
      "epoch": 0.013537777524439252,
      "grad_norm": 2.822539806365967,
      "learning_rate": 1e-05,
      "loss": 0.1775,
      "step": 475
    },
    {
      "epoch": 0.014250292130988685,
      "grad_norm": 1.9778774976730347,
      "learning_rate": 1e-05,
      "loss": 0.1489,
      "step": 500
    },
    {
      "epoch": 0.014250292130988685,
      "eval_loss": 0.15865246951580048,
      "eval_runtime": 75329.7882,
      "eval_samples_per_second": 0.414,
      "eval_steps_per_second": 0.052,
      "eval_wer": 10.136499555844708,
      "eval_wer_ortho": 36.29684933552337,
      "step": 500
    }
  ],
  "logging_steps": 25,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 1.358999322624e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}