{
  "best_global_step": 1000,
  "best_metric": 1.2435074923828926,
  "best_model_checkpoint": "/workspace/outputs/checkpoint-1000",
  "epoch": 2.4,
  "eval_steps": 1000,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "grad_norm": 0.2642808258533478,
      "learning_rate": 0.00019849539815926373,
      "loss": 0.18876359939575196,
      "step": 100
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.23475925624370575,
      "learning_rate": 0.00019689475790316127,
      "loss": 0.059173784255981444,
      "step": 200
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.20998287200927734,
      "learning_rate": 0.00019529411764705883,
      "loss": 0.04893703460693359,
      "step": 300
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.23086349666118622,
      "learning_rate": 0.0001936934773909564,
      "loss": 0.044345521926879884,
      "step": 400
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.205961674451828,
      "learning_rate": 0.00019209283713485397,
      "loss": 0.0404533576965332,
      "step": 500
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.1983308494091034,
      "learning_rate": 0.0001904921968787515,
      "loss": 0.037444443702697755,
      "step": 600
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.1748906373977661,
      "learning_rate": 0.00018889155662264907,
      "loss": 0.03432520627975464,
      "step": 700
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.18883433938026428,
      "learning_rate": 0.00018729091636654664,
      "loss": 0.03186330795288086,
      "step": 800
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.19548793137073517,
      "learning_rate": 0.00018569027611044418,
      "loss": 0.031016035079956053,
      "step": 900
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.1149163618683815,
      "learning_rate": 0.00018408963585434175,
      "loss": 0.03159681081771851,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "eval_cer": 1.2435074923828926,
      "eval_loss": 0.028011268004775047,
      "eval_runtime": 115.0642,
      "eval_samples_per_second": 43.454,
      "eval_steps_per_second": 2.72,
      "step": 1000
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.17694653570652008,
      "learning_rate": 0.0001824889955982393,
      "loss": 0.029933228492736816,
      "step": 1100
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.13514423370361328,
      "learning_rate": 0.00018088835534213685,
      "loss": 0.0282124662399292,
      "step": 1200
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.1704186350107193,
      "learning_rate": 0.00017928771508603442,
      "loss": 0.02600616455078125,
      "step": 1300
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.1911018043756485,
      "learning_rate": 0.00017768707482993199,
      "loss": 0.023294129371643067,
      "step": 1400
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.27751773595809937,
      "learning_rate": 0.00017608643457382955,
      "loss": 0.02326359748840332,
      "step": 1500
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.12529133260250092,
      "learning_rate": 0.0001744857943177271,
      "loss": 0.02243619441986084,
      "step": 1600
    },
    {
      "epoch": 1.3599999999999999,
      "grad_norm": 0.10395608842372894,
      "learning_rate": 0.00017288515406162466,
      "loss": 0.022894420623779298,
      "step": 1700
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.11129526793956757,
      "learning_rate": 0.00017128451380552223,
      "loss": 0.022162730693817138,
      "step": 1800
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.11065872758626938,
      "learning_rate": 0.00016968387354941977,
      "loss": 0.023314964771270753,
      "step": 1900
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.08472305536270142,
      "learning_rate": 0.00016808323329331733,
      "loss": 0.02215144395828247,
      "step": 2000
    },
    {
      "epoch": 1.6,
      "eval_cer": 1.3076230467241983,
      "eval_loss": 0.02364002913236618,
      "eval_runtime": 110.8626,
      "eval_samples_per_second": 45.101,
      "eval_steps_per_second": 2.823,
      "step": 2000
    },
    {
      "epoch": 1.6800000000000002,
      "grad_norm": 0.11845964193344116,
      "learning_rate": 0.0001664825930372149,
      "loss": 0.021831510066986085,
      "step": 2100
    },
    {
      "epoch": 1.76,
      "grad_norm": 0.09786409139633179,
      "learning_rate": 0.00016488195278111244,
      "loss": 0.022126307487487794,
      "step": 2200
    },
    {
      "epoch": 1.8399999999999999,
      "grad_norm": 0.10078627616167068,
      "learning_rate": 0.00016328131252501003,
      "loss": 0.02198990821838379,
      "step": 2300
    },
    {
      "epoch": 1.92,
      "grad_norm": 0.12120728194713593,
      "learning_rate": 0.00016168067226890757,
      "loss": 0.022060329914093017,
      "step": 2400
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.11438124626874924,
      "learning_rate": 0.00016008003201280514,
      "loss": 0.021810805797576903,
      "step": 2500
    },
    {
      "epoch": 2.08,
      "grad_norm": 0.06793903559446335,
      "learning_rate": 0.00015847939175670268,
      "loss": 0.019431318044662475,
      "step": 2600
    },
    {
      "epoch": 2.16,
      "grad_norm": 0.06746117025613785,
      "learning_rate": 0.00015687875150060025,
      "loss": 0.01889911890029907,
      "step": 2700
    },
    {
      "epoch": 2.24,
      "grad_norm": 0.13070549070835114,
      "learning_rate": 0.0001552781112444978,
      "loss": 0.019312630891799926,
      "step": 2800
    },
    {
      "epoch": 2.32,
      "grad_norm": 0.04858144372701645,
      "learning_rate": 0.00015367747098839535,
      "loss": 0.018591853380203246,
      "step": 2900
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.05757610872387886,
      "learning_rate": 0.00015207683073229292,
      "loss": 0.01879650592803955,
      "step": 3000
    },
    {
      "epoch": 2.4,
      "eval_cer": 1.2903041497645082,
      "eval_loss": 0.021801782771945,
      "eval_runtime": 108.7599,
      "eval_samples_per_second": 45.973,
      "eval_steps_per_second": 2.878,
      "step": 3000
    }
  ],
  "logging_steps": 100,
  "max_steps": 12500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 1000,
  "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": 1.055415576124416e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}