checkpoint-300/trainer_state.json

{
  "best_metric": 27.73044151820294,
  "best_model_checkpoint": "./whisper-small-clean_v6/checkpoint-200",
  "epoch": 1.002,
  "eval_steps": 100,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "grad_norm": 24.222782135009766,
      "learning_rate": 4.2000000000000004e-06,
      "loss": 1.5789,
      "step": 25
    },
    {
      "epoch": 0.1,
      "grad_norm": 3.7403714656829834,
      "learning_rate": 9.200000000000002e-06,
      "loss": 0.2766,
      "step": 50
    },
    {
      "epoch": 0.15,
      "grad_norm": 3.9098994731903076,
      "learning_rate": 9.533333333333334e-06,
      "loss": 0.1533,
      "step": 75
    },
    {
      "epoch": 0.2,
      "grad_norm": 3.8178775310516357,
      "learning_rate": 8.977777777777778e-06,
      "loss": 0.1354,
      "step": 100
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.4426867365837097,
      "eval_runtime": 177.5025,
      "eval_samples_per_second": 1.792,
      "eval_steps_per_second": 0.056,
      "eval_wer": 34.77924089852827,
      "step": 100
    },
    {
      "epoch": 0.25,
      "grad_norm": 3.6577646732330322,
      "learning_rate": 8.422222222222223e-06,
      "loss": 0.1327,
      "step": 125
    },
    {
      "epoch": 0.3,
      "grad_norm": 3.8943512439727783,
      "learning_rate": 7.866666666666667e-06,
      "loss": 0.1225,
      "step": 150
    },
    {
      "epoch": 0.35,
      "grad_norm": 3.568549394607544,
      "learning_rate": 7.3111111111111125e-06,
      "loss": 0.1349,
      "step": 175
    },
    {
      "epoch": 0.4,
      "grad_norm": 2.832570791244507,
      "learning_rate": 6.755555555555556e-06,
      "loss": 0.1115,
      "step": 200
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.3909415304660797,
      "eval_runtime": 151.5663,
      "eval_samples_per_second": 2.098,
      "eval_steps_per_second": 0.066,
      "eval_wer": 27.73044151820294,
      "step": 200
    },
    {
      "epoch": 0.45,
      "grad_norm": 3.3633787631988525,
      "learning_rate": 6.200000000000001e-06,
      "loss": 0.1162,
      "step": 225
    },
    {
      "epoch": 0.5,
      "grad_norm": 3.3805878162384033,
      "learning_rate": 5.6444444444444445e-06,
      "loss": 0.1205,
      "step": 250
    },
    {
      "epoch": 0.55,
      "grad_norm": 3.420900821685791,
      "learning_rate": 5.088888888888889e-06,
      "loss": 0.1055,
      "step": 275
    },
    {
      "epoch": 1.002,
      "grad_norm": 1.5796862840652466,
      "learning_rate": 4.533333333333334e-06,
      "loss": 0.1105,
      "step": 300
    },
    {
      "epoch": 1.002,
      "eval_loss": 0.37623485922813416,
      "eval_runtime": 149.2618,
      "eval_samples_per_second": 2.13,
      "eval_steps_per_second": 0.067,
      "eval_wer": 27.885360185902403,
      "step": 300
    }
  ],
  "logging_steps": 25,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.52958485553152e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}