{
  "best_global_step": 2000,
  "best_metric": 0.6051455140113831,
  "best_model_checkpoint": "./whisper-large-v3-lt1/checkpoint-2000",
  "epoch": 1.179,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "grad_norm": 0.42568153142929077,
      "learning_rate": 9.9e-06,
      "loss": 1.3821,
      "step": 100
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4062618613243103,
      "learning_rate": 9.478947368421053e-06,
      "loss": 1.1435,
      "step": 200
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.5448814630508423,
      "learning_rate": 8.95263157894737e-06,
      "loss": 1.009,
      "step": 300
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.709639310836792,
      "learning_rate": 8.426315789473684e-06,
      "loss": 0.919,
      "step": 400
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.8839156627655029,
      "learning_rate": 7.9e-06,
      "loss": 0.8441,
      "step": 500
    },
    {
      "epoch": 0.25,
      "eval_loss": 0.8364643454551697,
      "eval_runtime": 1166.4094,
      "eval_samples_per_second": 2.503,
      "eval_steps_per_second": 0.313,
      "step": 500
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.2657830715179443,
      "learning_rate": 7.373684210526316e-06,
      "loss": 0.796,
      "step": 600
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.17739737033844,
      "learning_rate": 6.8473684210526325e-06,
      "loss": 0.8109,
      "step": 700
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.271132230758667,
      "learning_rate": 6.321052631578948e-06,
      "loss": 0.7474,
      "step": 800
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.3513257503509521,
      "learning_rate": 5.794736842105264e-06,
      "loss": 0.7134,
      "step": 900
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.3304741382598877,
      "learning_rate": 5.268421052631579e-06,
      "loss": 0.6847,
      "step": 1000
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.6822749972343445,
      "eval_runtime": 1166.2313,
      "eval_samples_per_second": 2.504,
      "eval_steps_per_second": 0.313,
      "step": 1000
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.2514197826385498,
      "learning_rate": 4.7421052631578954e-06,
      "loss": 0.6689,
      "step": 1100
    },
    {
      "epoch": 0.6,
      "grad_norm": 2.271310806274414,
      "learning_rate": 4.215789473684211e-06,
      "loss": 0.6526,
      "step": 1200
    },
    {
      "epoch": 0.65,
      "grad_norm": 2.1460328102111816,
      "learning_rate": 3.6894736842105265e-06,
      "loss": 0.6299,
      "step": 1300
    },
    {
      "epoch": 0.7,
      "grad_norm": 2.67410945892334,
      "learning_rate": 3.1631578947368424e-06,
      "loss": 0.6143,
      "step": 1400
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.5648741722106934,
      "learning_rate": 2.6368421052631584e-06,
      "loss": 0.6632,
      "step": 1500
    },
    {
      "epoch": 0.75,
      "eval_loss": 0.6280742287635803,
      "eval_runtime": 1163.7066,
      "eval_samples_per_second": 2.509,
      "eval_steps_per_second": 0.314,
      "step": 1500
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.4991073608398438,
      "learning_rate": 2.110526315789474e-06,
      "loss": 0.6425,
      "step": 1600
    },
    {
      "epoch": 1.029,
      "grad_norm": 1.7371805906295776,
      "learning_rate": 1.5842105263157894e-06,
      "loss": 0.6223,
      "step": 1700
    },
    {
      "epoch": 1.079,
      "grad_norm": 3.491868495941162,
      "learning_rate": 1.0578947368421054e-06,
      "loss": 0.6013,
      "step": 1800
    },
    {
      "epoch": 1.129,
      "grad_norm": 2.1243815422058105,
      "learning_rate": 5.315789473684211e-07,
      "loss": 0.5967,
      "step": 1900
    },
    {
      "epoch": 1.179,
      "grad_norm": 2.0289883613586426,
      "learning_rate": 5.263157894736842e-09,
      "loss": 0.5986,
      "step": 2000
    },
    {
      "epoch": 1.179,
      "eval_loss": 0.6051455140113831,
      "eval_runtime": 1166.0635,
      "eval_samples_per_second": 2.504,
      "eval_steps_per_second": 0.313,
      "step": 2000
    }
  ],
  "logging_steps": 100,
  "max_steps": 2000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "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": 1.0985211708309504e+20,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}