checkpoints/checkpoint-400/trainer_state.json

{
  "best_metric": 102.97906602254427,
  "best_model_checkpoint": "./whisper-medium-attempt1-with-silence-text/checkpoint-300",
  "epoch": 12.5,
  "eval_steps": 100,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.78125,
      "grad_norm": 9.995346069335938,
      "learning_rate": 5.000000000000001e-07,
      "loss": 2.8857,
      "step": 25
    },
    {
      "epoch": 1.5625,
      "grad_norm": 10.72589111328125,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 2.824,
      "step": 50
    },
    {
      "epoch": 2.34375,
      "grad_norm": 10.114445686340332,
      "learning_rate": 1.5e-06,
      "loss": 2.6291,
      "step": 75
    },
    {
      "epoch": 3.125,
      "grad_norm": 8.762908935546875,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 2.502,
      "step": 100
    },
    {
      "epoch": 3.125,
      "eval_loss": 2.8539323806762695,
      "eval_runtime": 96.3981,
      "eval_samples_per_second": 1.317,
      "eval_steps_per_second": 0.166,
      "eval_wer": 127.10681696188944,
      "step": 100
    },
    {
      "epoch": 3.90625,
      "grad_norm": 11.801437377929688,
      "learning_rate": 2.5e-06,
      "loss": 2.297,
      "step": 125
    },
    {
      "epoch": 4.6875,
      "grad_norm": 9.479702949523926,
      "learning_rate": 3e-06,
      "loss": 1.9392,
      "step": 150
    },
    {
      "epoch": 5.46875,
      "grad_norm": 7.965484619140625,
      "learning_rate": 3.5e-06,
      "loss": 1.6685,
      "step": 175
    },
    {
      "epoch": 6.25,
      "grad_norm": 9.829634666442871,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.3556,
      "step": 200
    },
    {
      "epoch": 6.25,
      "eval_loss": 2.4153664112091064,
      "eval_runtime": 83.527,
      "eval_samples_per_second": 1.52,
      "eval_steps_per_second": 0.192,
      "eval_wer": 115.96886741814278,
      "step": 200
    },
    {
      "epoch": 7.03125,
      "grad_norm": 7.646000862121582,
      "learning_rate": 4.5e-06,
      "loss": 1.2347,
      "step": 225
    },
    {
      "epoch": 7.8125,
      "grad_norm": 9.588939666748047,
      "learning_rate": 5e-06,
      "loss": 1.0514,
      "step": 250
    },
    {
      "epoch": 8.59375,
      "grad_norm": 11.105179786682129,
      "learning_rate": 5.500000000000001e-06,
      "loss": 0.8558,
      "step": 275
    },
    {
      "epoch": 9.375,
      "grad_norm": 10.725419044494629,
      "learning_rate": 6e-06,
      "loss": 0.7175,
      "step": 300
    },
    {
      "epoch": 9.375,
      "eval_loss": 2.746173143386841,
      "eval_runtime": 72.6785,
      "eval_samples_per_second": 1.747,
      "eval_steps_per_second": 0.22,
      "eval_wer": 102.97906602254427,
      "step": 300
    },
    {
      "epoch": 10.15625,
      "grad_norm": 12.282588958740234,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 0.5836,
      "step": 325
    },
    {
      "epoch": 10.9375,
      "grad_norm": 9.197545051574707,
      "learning_rate": 7e-06,
      "loss": 0.4481,
      "step": 350
    },
    {
      "epoch": 11.71875,
      "grad_norm": 9.019693374633789,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.3159,
      "step": 375
    },
    {
      "epoch": 12.5,
      "grad_norm": 10.85390567779541,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.2532,
      "step": 400
    },
    {
      "epoch": 12.5,
      "eval_loss": 3.102280855178833,
      "eval_runtime": 82.6864,
      "eval_samples_per_second": 1.536,
      "eval_steps_per_second": 0.194,
      "eval_wer": 114.78797638217928,
      "step": 400
    }
  ],
  "logging_steps": 25,
  "max_steps": 2000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 63,
  "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": 6.4706351136768e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}