trainer_state.json

{
  "best_metric": 0.24032442271709442,
  "best_model_checkpoint": "/scratch/skscla001/speech/results/xls-r-300m-nyagen-combined-hp-tuning-test-model/checkpoint-600",
  "epoch": 22.225352112676056,
  "eval_steps": 100,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.2253521126760565,
      "eval_loss": 2.900092124938965,
      "eval_runtime": 27.3693,
      "eval_samples_per_second": 12.203,
      "eval_steps_per_second": 1.535,
      "eval_wer": 1.0,
      "step": 100
    },
    {
      "epoch": 4.450704225352113,
      "eval_loss": 0.837561309337616,
      "eval_runtime": 27.5078,
      "eval_samples_per_second": 12.142,
      "eval_steps_per_second": 1.527,
      "eval_wer": 0.7848668009850012,
      "step": 200
    },
    {
      "epoch": 6.676056338028169,
      "eval_loss": 0.30867961049079895,
      "eval_runtime": 27.4066,
      "eval_samples_per_second": 12.187,
      "eval_steps_per_second": 1.532,
      "eval_wer": 0.4329527647190508,
      "step": 300
    },
    {
      "epoch": 8.901408450704226,
      "eval_loss": 0.25156790018081665,
      "eval_runtime": 27.5616,
      "eval_samples_per_second": 12.118,
      "eval_steps_per_second": 1.524,
      "eval_wer": 0.37653906424893663,
      "step": 400
    },
    {
      "epoch": 11.112676056338028,
      "grad_norm": 9.6257905960083,
      "learning_rate": 0.0009213002275800941,
      "loss": 30.7137,
      "step": 500
    },
    {
      "epoch": 11.112676056338028,
      "eval_loss": 0.26118895411491394,
      "eval_runtime": 27.5705,
      "eval_samples_per_second": 12.114,
      "eval_steps_per_second": 1.523,
      "eval_wer": 0.3434072084172823,
      "step": 500
    },
    {
      "epoch": 13.338028169014084,
      "eval_loss": 0.24032442271709442,
      "eval_runtime": 27.4884,
      "eval_samples_per_second": 12.151,
      "eval_steps_per_second": 1.528,
      "eval_wer": 0.32303559435862994,
      "step": 600
    },
    {
      "epoch": 15.56338028169014,
      "eval_loss": 0.24804773926734924,
      "eval_runtime": 27.4749,
      "eval_samples_per_second": 12.157,
      "eval_steps_per_second": 1.529,
      "eval_wer": 0.31564808596373406,
      "step": 700
    },
    {
      "epoch": 17.788732394366196,
      "eval_loss": 0.25463205575942993,
      "eval_runtime": 27.5117,
      "eval_samples_per_second": 12.14,
      "eval_steps_per_second": 1.527,
      "eval_wer": 0.2939332885605552,
      "step": 800
    },
    {
      "epoch": 20.0,
      "eval_loss": 0.2638227641582489,
      "eval_runtime": 27.6123,
      "eval_samples_per_second": 12.096,
      "eval_steps_per_second": 1.521,
      "eval_wer": 0.2919185135437654,
      "step": 900
    },
    {
      "epoch": 22.225352112676056,
      "grad_norm": 2.932446241378784,
      "learning_rate": 0.0003650929645045522,
      "loss": 1.8035,
      "step": 1000
    },
    {
      "epoch": 22.225352112676056,
      "eval_loss": 0.26894354820251465,
      "eval_runtime": 27.579,
      "eval_samples_per_second": 12.111,
      "eval_steps_per_second": 1.523,
      "eval_wer": 0.2675173494515335,
      "step": 1000
    },
    {
      "epoch": 22.225352112676056,
      "step": 1000,
      "total_flos": 1.9575417036880396e+19,
      "train_loss": 16.2585947265625,
      "train_runtime": 7398.0938,
      "train_samples_per_second": 11.516,
      "train_steps_per_second": 0.178
    }
  ],
  "logging_steps": 500,
  "max_steps": 1320,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 400,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 4,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 2
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.9575417036880396e+19,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}