trainer_state.json

{
  "best_metric": 0.3383650779724121,
  "best_model_checkpoint": "/scratch/skscla001/speech/results/xls-r-300m-bemgen-combined-hp-tuning-test-model/checkpoint-500",
  "epoch": 22.228163992869874,
  "eval_steps": 100,
  "global_step": 800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.7985739750445635,
      "eval_loss": 2.9111883640289307,
      "eval_runtime": 39.716,
      "eval_samples_per_second": 13.622,
      "eval_steps_per_second": 3.424,
      "eval_wer": 1.0,
      "step": 100
    },
    {
      "epoch": 5.570409982174688,
      "eval_loss": 1.6968202590942383,
      "eval_runtime": 39.8919,
      "eval_samples_per_second": 13.562,
      "eval_steps_per_second": 3.409,
      "eval_wer": 0.9970951343500363,
      "step": 200
    },
    {
      "epoch": 8.342245989304812,
      "eval_loss": 0.3893207609653473,
      "eval_runtime": 40.2019,
      "eval_samples_per_second": 13.457,
      "eval_steps_per_second": 3.383,
      "eval_wer": 0.615831517792302,
      "step": 300
    },
    {
      "epoch": 11.114081996434937,
      "eval_loss": 0.34530752897262573,
      "eval_runtime": 40.0609,
      "eval_samples_per_second": 13.504,
      "eval_steps_per_second": 3.395,
      "eval_wer": 0.5106753812636166,
      "step": 400
    },
    {
      "epoch": 13.9126559714795,
      "grad_norm": 7.798405170440674,
      "learning_rate": 0.0005819322340474784,
      "loss": 31.2339,
      "step": 500
    },
    {
      "epoch": 13.9126559714795,
      "eval_loss": 0.3383650779724121,
      "eval_runtime": 39.984,
      "eval_samples_per_second": 13.53,
      "eval_steps_per_second": 3.401,
      "eval_wer": 0.49092229484386346,
      "step": 500
    },
    {
      "epoch": 16.684491978609625,
      "eval_loss": 0.3592090606689453,
      "eval_runtime": 39.9246,
      "eval_samples_per_second": 13.551,
      "eval_steps_per_second": 3.406,
      "eval_wer": 0.46187363834422657,
      "step": 600
    },
    {
      "epoch": 19.45632798573975,
      "eval_loss": 0.367597371339798,
      "eval_runtime": 39.9675,
      "eval_samples_per_second": 13.536,
      "eval_steps_per_second": 3.403,
      "eval_wer": 0.4335511982570806,
      "step": 700
    },
    {
      "epoch": 22.228163992869874,
      "eval_loss": 0.3925014138221741,
      "eval_runtime": 39.8278,
      "eval_samples_per_second": 13.583,
      "eval_steps_per_second": 3.415,
      "eval_wer": 0.42280319535221494,
      "step": 800
    },
    {
      "epoch": 22.228163992869874,
      "step": 800,
      "total_flos": 3.4912812658192863e+19,
      "train_loss": 20.112410697937012,
      "train_runtime": 9256.0708,
      "train_samples_per_second": 14.527,
      "train_steps_per_second": 0.113
    }
  ],
  "logging_steps": 500,
  "max_steps": 1050,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 400,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 3
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.4912812658192863e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}