{
  "best_metric": 0.6599727869033813,
  "best_model_checkpoint": "/scratch/skscla001/results/xls-r-1b-bem-natbed-combined-model/checkpoint-800",
  "epoch": 2.7534418022528158,
  "eval_steps": 100,
  "global_step": 1100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2503128911138924,
      "grad_norm": 2.3866429328918457,
      "learning_rate": 4.9e-05,
      "loss": 6.1525,
      "step": 100
    },
    {
      "epoch": 0.2503128911138924,
      "eval_loss": 2.9905781745910645,
      "eval_runtime": 96.751,
      "eval_samples_per_second": 14.036,
      "eval_steps_per_second": 1.757,
      "eval_wer": 1.0,
      "step": 100
    },
    {
      "epoch": 0.5006257822277848,
      "grad_norm": 12.731361389160156,
      "learning_rate": 4.958719460825611e-05,
      "loss": 2.8477,
      "step": 200
    },
    {
      "epoch": 0.5006257822277848,
      "eval_loss": 2.3722355365753174,
      "eval_runtime": 95.7895,
      "eval_samples_per_second": 14.177,
      "eval_steps_per_second": 1.775,
      "eval_wer": 1.0,
      "step": 200
    },
    {
      "epoch": 0.7509386733416771,
      "grad_norm": 3.254181385040283,
      "learning_rate": 4.916596461668071e-05,
      "loss": 1.3216,
      "step": 300
    },
    {
      "epoch": 0.7509386733416771,
      "eval_loss": 0.8614773154258728,
      "eval_runtime": 95.7487,
      "eval_samples_per_second": 14.183,
      "eval_steps_per_second": 1.775,
      "eval_wer": 0.8819921491658489,
      "step": 300
    },
    {
      "epoch": 1.0012515644555695,
      "grad_norm": 4.778253555297852,
      "learning_rate": 4.874473462510531e-05,
      "loss": 0.8219,
      "step": 400
    },
    {
      "epoch": 1.0012515644555695,
      "eval_loss": 0.8088704347610474,
      "eval_runtime": 95.7716,
      "eval_samples_per_second": 14.18,
      "eval_steps_per_second": 1.775,
      "eval_wer": 0.8601570166830226,
      "step": 400
    },
    {
      "epoch": 1.2515644555694618,
      "grad_norm": 1.714836597442627,
      "learning_rate": 4.832350463352991e-05,
      "loss": 0.718,
      "step": 500
    },
    {
      "epoch": 1.2515644555694618,
      "eval_loss": 0.7154927253723145,
      "eval_runtime": 96.02,
      "eval_samples_per_second": 14.143,
      "eval_steps_per_second": 1.77,
      "eval_wer": 0.7755152109911678,
      "step": 500
    },
    {
      "epoch": 1.5018773466833542,
      "grad_norm": 3.5530471801757812,
      "learning_rate": 4.7902274641954506e-05,
      "loss": 0.6985,
      "step": 600
    },
    {
      "epoch": 1.5018773466833542,
      "eval_loss": 0.7211880087852478,
      "eval_runtime": 95.9087,
      "eval_samples_per_second": 14.159,
      "eval_steps_per_second": 1.773,
      "eval_wer": 0.6726365718024206,
      "step": 600
    },
    {
      "epoch": 1.7521902377972465,
      "grad_norm": 3.236358642578125,
      "learning_rate": 4.7481044650379106e-05,
      "loss": 0.669,
      "step": 700
    },
    {
      "epoch": 1.7521902377972465,
      "eval_loss": 0.6837976574897766,
      "eval_runtime": 96.2232,
      "eval_samples_per_second": 14.113,
      "eval_steps_per_second": 1.767,
      "eval_wer": 0.7244030094864246,
      "step": 700
    },
    {
      "epoch": 2.002503128911139,
      "grad_norm": 1.7046451568603516,
      "learning_rate": 4.7059814658803705e-05,
      "loss": 0.6842,
      "step": 800
    },
    {
      "epoch": 2.002503128911139,
      "eval_loss": 0.6599727869033813,
      "eval_runtime": 96.1095,
      "eval_samples_per_second": 14.13,
      "eval_steps_per_second": 1.769,
      "eval_wer": 0.7042852469741576,
      "step": 800
    },
    {
      "epoch": 2.252816020025031,
      "grad_norm": 7.314039707183838,
      "learning_rate": 4.663858466722831e-05,
      "loss": 0.5908,
      "step": 900
    },
    {
      "epoch": 2.252816020025031,
      "eval_loss": 0.821216881275177,
      "eval_runtime": 96.0491,
      "eval_samples_per_second": 14.139,
      "eval_steps_per_second": 1.77,
      "eval_wer": 0.862855740922473,
      "step": 900
    },
    {
      "epoch": 2.5031289111389237,
      "grad_norm": 0.7544125318527222,
      "learning_rate": 4.621735467565291e-05,
      "loss": 0.5575,
      "step": 1000
    },
    {
      "epoch": 2.5031289111389237,
      "eval_loss": 0.6882767081260681,
      "eval_runtime": 95.8261,
      "eval_samples_per_second": 14.172,
      "eval_steps_per_second": 1.774,
      "eval_wer": 0.7400228982662741,
      "step": 1000
    },
    {
      "epoch": 2.7534418022528158,
      "grad_norm": 5.456251621246338,
      "learning_rate": 4.579612468407751e-05,
      "loss": 0.583,
      "step": 1100
    },
    {
      "epoch": 2.7534418022528158,
      "eval_loss": 0.7891210913658142,
      "eval_runtime": 96.0541,
      "eval_samples_per_second": 14.138,
      "eval_steps_per_second": 1.77,
      "eval_wer": 0.8344782466470396,
      "step": 1100
    },
    {
      "epoch": 2.7534418022528158,
      "step": 1100,
      "total_flos": 1.1254487729160804e+19,
      "train_loss": 1.4222582556984642,
      "train_runtime": 3256.3096,
      "train_samples_per_second": 58.843,
      "train_steps_per_second": 3.676
    }
  ],
  "logging_steps": 100,
  "max_steps": 11970,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 200,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "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": 1.1254487729160804e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}