{
  "best_metric": 0.2479187548160553,
  "best_model_checkpoint": "./mms-1b-bem-genbed-all/checkpoint-3600",
  "epoch": 5.0,
  "eval_steps": 200,
  "global_step": 3640,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.27472527472527475,
      "eval_loss": 0.6311588883399963,
      "eval_runtime": 129.4588,
      "eval_samples_per_second": 14.978,
      "eval_steps_per_second": 1.877,
      "eval_wer": 0.6736216216216216,
      "step": 200
    },
    {
      "epoch": 0.5494505494505495,
      "eval_loss": 0.3198663890361786,
      "eval_runtime": 127.9369,
      "eval_samples_per_second": 15.156,
      "eval_steps_per_second": 1.899,
      "eval_wer": 0.49194594594594593,
      "step": 400
    },
    {
      "epoch": 0.6868131868131868,
      "grad_norm": 1.675441861152649,
      "learning_rate": 0.00029699999999999996,
      "loss": 2.9901,
      "step": 500
    },
    {
      "epoch": 0.8241758241758241,
      "eval_loss": 0.3013916313648224,
      "eval_runtime": 128.0261,
      "eval_samples_per_second": 15.145,
      "eval_steps_per_second": 1.898,
      "eval_wer": 0.4612972972972973,
      "step": 600
    },
    {
      "epoch": 1.098901098901099,
      "eval_loss": 0.28253573179244995,
      "eval_runtime": 127.3129,
      "eval_samples_per_second": 15.23,
      "eval_steps_per_second": 1.909,
      "eval_wer": 0.4432972972972973,
      "step": 800
    },
    {
      "epoch": 1.3736263736263736,
      "grad_norm": 0.6800592541694641,
      "learning_rate": 0.00025280254777070065,
      "loss": 0.3968,
      "step": 1000
    },
    {
      "epoch": 1.3736263736263736,
      "eval_loss": 0.2782987058162689,
      "eval_runtime": 127.2721,
      "eval_samples_per_second": 15.235,
      "eval_steps_per_second": 1.909,
      "eval_wer": 0.4541081081081081,
      "step": 1000
    },
    {
      "epoch": 1.6483516483516483,
      "eval_loss": 0.27315396070480347,
      "eval_runtime": 128.0724,
      "eval_samples_per_second": 15.14,
      "eval_steps_per_second": 1.897,
      "eval_wer": 0.4294054054054054,
      "step": 1200
    },
    {
      "epoch": 1.9230769230769231,
      "eval_loss": 0.26492610573768616,
      "eval_runtime": 127.9148,
      "eval_samples_per_second": 15.159,
      "eval_steps_per_second": 1.9,
      "eval_wer": 0.4243783783783784,
      "step": 1400
    },
    {
      "epoch": 2.0604395604395602,
      "grad_norm": 0.9411349296569824,
      "learning_rate": 0.00020503184713375794,
      "loss": 0.3766,
      "step": 1500
    },
    {
      "epoch": 2.197802197802198,
      "eval_loss": 0.26205211877822876,
      "eval_runtime": 128.7617,
      "eval_samples_per_second": 15.059,
      "eval_steps_per_second": 1.887,
      "eval_wer": 0.42043243243243245,
      "step": 1600
    },
    {
      "epoch": 2.4725274725274726,
      "eval_loss": 0.262787789106369,
      "eval_runtime": 128.5209,
      "eval_samples_per_second": 15.087,
      "eval_steps_per_second": 1.891,
      "eval_wer": 0.4170810810810811,
      "step": 1800
    },
    {
      "epoch": 2.7472527472527473,
      "grad_norm": 1.4841364622116089,
      "learning_rate": 0.00015726114649681526,
      "loss": 0.3537,
      "step": 2000
    },
    {
      "epoch": 2.7472527472527473,
      "eval_loss": 0.2579393982887268,
      "eval_runtime": 128.5847,
      "eval_samples_per_second": 15.08,
      "eval_steps_per_second": 1.89,
      "eval_wer": 0.4187027027027027,
      "step": 2000
    },
    {
      "epoch": 3.021978021978022,
      "eval_loss": 0.25568732619285583,
      "eval_runtime": 128.193,
      "eval_samples_per_second": 15.126,
      "eval_steps_per_second": 1.896,
      "eval_wer": 0.40335135135135136,
      "step": 2200
    },
    {
      "epoch": 3.2967032967032965,
      "eval_loss": 0.252371609210968,
      "eval_runtime": 128.8222,
      "eval_samples_per_second": 15.052,
      "eval_steps_per_second": 1.886,
      "eval_wer": 0.4090810810810811,
      "step": 2400
    },
    {
      "epoch": 3.4340659340659343,
      "grad_norm": 2.9654958248138428,
      "learning_rate": 0.00010949044585987259,
      "loss": 0.3529,
      "step": 2500
    },
    {
      "epoch": 3.571428571428571,
      "eval_loss": 0.25349992513656616,
      "eval_runtime": 128.7325,
      "eval_samples_per_second": 15.062,
      "eval_steps_per_second": 1.888,
      "eval_wer": 0.4061081081081081,
      "step": 2600
    },
    {
      "epoch": 3.8461538461538463,
      "eval_loss": 0.24949324131011963,
      "eval_runtime": 129.6607,
      "eval_samples_per_second": 14.954,
      "eval_steps_per_second": 1.874,
      "eval_wer": 0.4034054054054054,
      "step": 2800
    },
    {
      "epoch": 4.1208791208791204,
      "grad_norm": 0.9809963703155518,
      "learning_rate": 6.171974522292994e-05,
      "loss": 0.3393,
      "step": 3000
    },
    {
      "epoch": 4.1208791208791204,
      "eval_loss": 0.24937787652015686,
      "eval_runtime": 130.0963,
      "eval_samples_per_second": 14.904,
      "eval_steps_per_second": 1.868,
      "eval_wer": 0.4064864864864865,
      "step": 3000
    },
    {
      "epoch": 4.395604395604396,
      "eval_loss": 0.24878770112991333,
      "eval_runtime": 129.7286,
      "eval_samples_per_second": 14.947,
      "eval_steps_per_second": 1.873,
      "eval_wer": 0.40664864864864864,
      "step": 3200
    },
    {
      "epoch": 4.670329670329671,
      "eval_loss": 0.2481740117073059,
      "eval_runtime": 129.3443,
      "eval_samples_per_second": 14.991,
      "eval_steps_per_second": 1.879,
      "eval_wer": 0.40275675675675676,
      "step": 3400
    },
    {
      "epoch": 4.8076923076923075,
      "grad_norm": 0.8750921487808228,
      "learning_rate": 1.394904458598726e-05,
      "loss": 0.3332,
      "step": 3500
    },
    {
      "epoch": 4.945054945054945,
      "eval_loss": 0.2479187548160553,
      "eval_runtime": 129.4727,
      "eval_samples_per_second": 14.976,
      "eval_steps_per_second": 1.877,
      "eval_wer": 0.40616216216216217,
      "step": 3600
    },
    {
      "epoch": 5.0,
      "step": 3640,
      "total_flos": 1.737871213031041e+19,
      "train_loss": 0.719452987922417,
      "train_runtime": 6513.7768,
      "train_samples_per_second": 4.465,
      "train_steps_per_second": 0.559
    }
  ],
  "logging_steps": 500,
  "max_steps": 3640,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.737871213031041e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}