{
  "best_global_step": 1000,
  "best_metric": 240.63380883232256,
  "best_model_checkpoint": "/workspace/output/whisper-changhua-8_1_1/checkpoint-1000",
  "epoch": 13.157894736842104,
  "eval_steps": 100,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.6578947368421053,
      "grad_norm": 20.157838821411133,
      "learning_rate": 9.800000000000001e-07,
      "loss": 5.3937,
      "step": 50
    },
    {
      "epoch": 1.3157894736842106,
      "grad_norm": 12.150752067565918,
      "learning_rate": 1.98e-06,
      "loss": 2.8495,
      "step": 100
    },
    {
      "epoch": 1.3157894736842106,
      "eval_cer": 260.2801742547194,
      "eval_loss": 2.078307867050171,
      "eval_runtime": 40.147,
      "eval_samples_per_second": 11.284,
      "eval_steps_per_second": 2.84,
      "step": 100
    },
    {
      "epoch": 1.973684210526316,
      "grad_norm": 9.51569652557373,
      "learning_rate": 2.9800000000000003e-06,
      "loss": 1.65,
      "step": 150
    },
    {
      "epoch": 2.6315789473684212,
      "grad_norm": 9.225106239318848,
      "learning_rate": 3.980000000000001e-06,
      "loss": 1.2547,
      "step": 200
    },
    {
      "epoch": 2.6315789473684212,
      "eval_cer": 243.76868540189628,
      "eval_loss": 1.3091058731079102,
      "eval_runtime": 40.2136,
      "eval_samples_per_second": 11.265,
      "eval_steps_per_second": 2.835,
      "step": 200
    },
    {
      "epoch": 3.2894736842105265,
      "grad_norm": 8.73460865020752,
      "learning_rate": 4.980000000000001e-06,
      "loss": 1.0415,
      "step": 250
    },
    {
      "epoch": 3.9473684210526314,
      "grad_norm": 6.480662822723389,
      "learning_rate": 5.98e-06,
      "loss": 0.7814,
      "step": 300
    },
    {
      "epoch": 3.9473684210526314,
      "eval_cer": 251.57598018279663,
      "eval_loss": 0.6244341731071472,
      "eval_runtime": 40.4043,
      "eval_samples_per_second": 11.212,
      "eval_steps_per_second": 2.821,
      "step": 300
    },
    {
      "epoch": 4.605263157894737,
      "grad_norm": 4.831464767456055,
      "learning_rate": 6.98e-06,
      "loss": 0.4204,
      "step": 350
    },
    {
      "epoch": 5.2631578947368425,
      "grad_norm": 4.215816497802734,
      "learning_rate": 7.980000000000002e-06,
      "loss": 0.3407,
      "step": 400
    },
    {
      "epoch": 5.2631578947368425,
      "eval_cer": 290.749124455454,
      "eval_loss": 0.5503445863723755,
      "eval_runtime": 41.2087,
      "eval_samples_per_second": 10.993,
      "eval_steps_per_second": 2.766,
      "step": 400
    },
    {
      "epoch": 5.921052631578947,
      "grad_norm": 4.6707634925842285,
      "learning_rate": 8.98e-06,
      "loss": 0.2667,
      "step": 450
    },
    {
      "epoch": 6.578947368421053,
      "grad_norm": 3.476262331008911,
      "learning_rate": 9.980000000000001e-06,
      "loss": 0.1763,
      "step": 500
    },
    {
      "epoch": 6.578947368421053,
      "eval_cer": 274.81848466729303,
      "eval_loss": 0.5615935921669006,
      "eval_runtime": 40.95,
      "eval_samples_per_second": 11.062,
      "eval_steps_per_second": 2.784,
      "step": 500
    },
    {
      "epoch": 7.2368421052631575,
      "grad_norm": 2.525753974914551,
      "learning_rate": 9.94842105263158e-06,
      "loss": 0.1409,
      "step": 550
    },
    {
      "epoch": 7.894736842105263,
      "grad_norm": 2.9115219116210938,
      "learning_rate": 9.895789473684212e-06,
      "loss": 0.1026,
      "step": 600
    },
    {
      "epoch": 7.894736842105263,
      "eval_cer": 261.57854275219955,
      "eval_loss": 0.586604654788971,
      "eval_runtime": 40.7596,
      "eval_samples_per_second": 11.114,
      "eval_steps_per_second": 2.797,
      "step": 600
    },
    {
      "epoch": 8.552631578947368,
      "grad_norm": 2.416334390640259,
      "learning_rate": 9.843157894736843e-06,
      "loss": 0.0628,
      "step": 650
    },
    {
      "epoch": 9.210526315789474,
      "grad_norm": 1.3677974939346313,
      "learning_rate": 9.790526315789475e-06,
      "loss": 0.0485,
      "step": 700
    },
    {
      "epoch": 9.210526315789474,
      "eval_cer": 269.804390535577,
      "eval_loss": 0.6309823393821716,
      "eval_runtime": 40.7742,
      "eval_samples_per_second": 11.11,
      "eval_steps_per_second": 2.796,
      "step": 700
    },
    {
      "epoch": 9.868421052631579,
      "grad_norm": 2.056694984436035,
      "learning_rate": 9.737894736842107e-06,
      "loss": 0.0351,
      "step": 750
    },
    {
      "epoch": 10.526315789473685,
      "grad_norm": 1.416398525238037,
      "learning_rate": 9.685263157894738e-06,
      "loss": 0.0245,
      "step": 800
    },
    {
      "epoch": 10.526315789473685,
      "eval_cer": 275.9033057145298,
      "eval_loss": 0.6414592862129211,
      "eval_runtime": 40.752,
      "eval_samples_per_second": 11.116,
      "eval_steps_per_second": 2.797,
      "step": 800
    },
    {
      "epoch": 11.18421052631579,
      "grad_norm": 1.9776651859283447,
      "learning_rate": 9.63263157894737e-06,
      "loss": 0.0223,
      "step": 850
    },
    {
      "epoch": 11.842105263157894,
      "grad_norm": 2.234473705291748,
      "learning_rate": 9.58e-06,
      "loss": 0.017,
      "step": 900
    },
    {
      "epoch": 11.842105263157894,
      "eval_cer": 250.7132484838131,
      "eval_loss": 0.6816866397857666,
      "eval_runtime": 40.3529,
      "eval_samples_per_second": 11.226,
      "eval_steps_per_second": 2.825,
      "step": 900
    },
    {
      "epoch": 12.5,
      "grad_norm": 1.4661343097686768,
      "learning_rate": 9.527368421052631e-06,
      "loss": 0.0149,
      "step": 950
    },
    {
      "epoch": 13.157894736842104,
      "grad_norm": 3.28507137298584,
      "learning_rate": 9.474736842105265e-06,
      "loss": 0.0157,
      "step": 1000
    },
    {
      "epoch": 13.157894736842104,
      "eval_cer": 240.63380883232256,
      "eval_loss": 0.7029738426208496,
      "eval_runtime": 40.3273,
      "eval_samples_per_second": 11.233,
      "eval_steps_per_second": 2.827,
      "step": 1000
    }
  ],
  "logging_steps": 50,
  "max_steps": 10000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 132,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 10,
        "early_stopping_threshold": 0.001
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.376956379086848e+19,
  "train_batch_size": 24,
  "trial_name": null,
  "trial_params": null
}