last-checkpoint/trainer_state.json

{
  "best_metric": 55.83411121482864,
  "best_model_checkpoint": "./linshoufanfork-whisper-small-nan-tw/checkpoint-500",
  "epoch": 0.1607200257152041,
  "eval_steps": 500,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "grad_norm": 48.035465240478516,
      "learning_rate": 2.5e-06,
      "loss": 6.8285,
      "step": 25
    },
    {
      "epoch": 0.02,
      "grad_norm": 142.98660278320312,
      "learning_rate": 5e-06,
      "loss": 3.2493,
      "step": 50
    },
    {
      "epoch": 0.02,
      "grad_norm": 27.529985427856445,
      "learning_rate": 7.500000000000001e-06,
      "loss": 2.2763,
      "step": 75
    },
    {
      "epoch": 0.03,
      "grad_norm": 21.324289321899414,
      "learning_rate": 1e-05,
      "loss": 1.7193,
      "step": 100
    },
    {
      "epoch": 0.04,
      "grad_norm": 13.650047302246094,
      "learning_rate": 9.91697110594487e-06,
      "loss": 1.3177,
      "step": 125
    },
    {
      "epoch": 0.05,
      "grad_norm": 14.939111709594727,
      "learning_rate": 9.83394221188974e-06,
      "loss": 1.2486,
      "step": 150
    },
    {
      "epoch": 0.06,
      "grad_norm": 14.85805606842041,
      "learning_rate": 9.750913317834608e-06,
      "loss": 1.1991,
      "step": 175
    },
    {
      "epoch": 0.06,
      "grad_norm": 15.52128791809082,
      "learning_rate": 9.667884423779476e-06,
      "loss": 1.1126,
      "step": 200
    },
    {
      "epoch": 0.07,
      "grad_norm": 22.642026901245117,
      "learning_rate": 9.584855529724345e-06,
      "loss": 1.0437,
      "step": 225
    },
    {
      "epoch": 0.08,
      "grad_norm": 16.712217330932617,
      "learning_rate": 9.501826635669213e-06,
      "loss": 0.9986,
      "step": 250
    },
    {
      "epoch": 0.09,
      "grad_norm": 13.973222732543945,
      "learning_rate": 9.418797741614083e-06,
      "loss": 0.9188,
      "step": 275
    },
    {
      "epoch": 0.1,
      "grad_norm": 13.404074668884277,
      "learning_rate": 9.335768847558952e-06,
      "loss": 0.9244,
      "step": 300
    },
    {
      "epoch": 0.1,
      "grad_norm": 12.21960163116455,
      "learning_rate": 9.25273995350382e-06,
      "loss": 0.9018,
      "step": 325
    },
    {
      "epoch": 0.11,
      "grad_norm": 14.460400581359863,
      "learning_rate": 9.169711059448689e-06,
      "loss": 0.8398,
      "step": 350
    },
    {
      "epoch": 0.12,
      "grad_norm": 14.076154708862305,
      "learning_rate": 9.086682165393557e-06,
      "loss": 0.8828,
      "step": 375
    },
    {
      "epoch": 0.13,
      "grad_norm": 13.204269409179688,
      "learning_rate": 9.003653271338426e-06,
      "loss": 0.8503,
      "step": 400
    },
    {
      "epoch": 0.14,
      "grad_norm": 10.978958129882812,
      "learning_rate": 8.920624377283296e-06,
      "loss": 0.8198,
      "step": 425
    },
    {
      "epoch": 0.14,
      "grad_norm": 13.963995933532715,
      "learning_rate": 8.837595483228164e-06,
      "loss": 0.8202,
      "step": 450
    },
    {
      "epoch": 0.15,
      "grad_norm": 13.337563514709473,
      "learning_rate": 8.754566589173033e-06,
      "loss": 0.7536,
      "step": 475
    },
    {
      "epoch": 0.16,
      "grad_norm": 12.711252212524414,
      "learning_rate": 8.671537695117903e-06,
      "loss": 0.7938,
      "step": 500
    },
    {
      "epoch": 0.16,
      "eval_cer": 55.83411121482864,
      "eval_loss": 0.7767874002456665,
      "eval_runtime": 1802.202,
      "eval_samples_per_second": 2.461,
      "eval_steps_per_second": 0.308,
      "step": 500
    }
  ],
  "logging_steps": 25,
  "max_steps": 3111,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 2.30868320256e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}