{
  "best_metric": 18.926838201629888,
  "best_model_checkpoint": "./checkpoint-1000",
  "epoch": 9.7799511002445,
  "eval_steps": 1000,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.24,
      "grad_norm": 39.86157989501953,
      "learning_rate": 5.000000000000001e-07,
      "loss": 2.0555,
      "step": 25
    },
    {
      "epoch": 0.49,
      "grad_norm": Infinity,
      "learning_rate": 9.800000000000001e-07,
      "loss": 1.5219,
      "step": 50
    },
    {
      "epoch": 0.73,
      "grad_norm": 6.197519779205322,
      "learning_rate": 1.48e-06,
      "loss": 1.0167,
      "step": 75
    },
    {
      "epoch": 0.98,
      "grad_norm": 5.485505104064941,
      "learning_rate": 1.98e-06,
      "loss": 0.7299,
      "step": 100
    },
    {
      "epoch": 1.22,
      "grad_norm": 5.534335613250732,
      "learning_rate": 2.4800000000000004e-06,
      "loss": 0.6317,
      "step": 125
    },
    {
      "epoch": 1.47,
      "grad_norm": 4.898209095001221,
      "learning_rate": 2.9800000000000003e-06,
      "loss": 0.5503,
      "step": 150
    },
    {
      "epoch": 1.71,
      "grad_norm": 5.0602946281433105,
      "learning_rate": 3.48e-06,
      "loss": 0.4998,
      "step": 175
    },
    {
      "epoch": 1.96,
      "grad_norm": 4.8069305419921875,
      "learning_rate": 3.980000000000001e-06,
      "loss": 0.4457,
      "step": 200
    },
    {
      "epoch": 2.2,
      "grad_norm": 4.571457386016846,
      "learning_rate": 4.48e-06,
      "loss": 0.3778,
      "step": 225
    },
    {
      "epoch": 2.44,
      "grad_norm": 4.707397937774658,
      "learning_rate": 4.980000000000001e-06,
      "loss": 0.333,
      "step": 250
    },
    {
      "epoch": 2.69,
      "grad_norm": 4.317914009094238,
      "learning_rate": 5.480000000000001e-06,
      "loss": 0.2859,
      "step": 275
    },
    {
      "epoch": 2.93,
      "grad_norm": 2.6433582305908203,
      "learning_rate": 5.98e-06,
      "loss": 0.2229,
      "step": 300
    },
    {
      "epoch": 3.18,
      "grad_norm": 2.5579586029052734,
      "learning_rate": 6.480000000000001e-06,
      "loss": 0.1767,
      "step": 325
    },
    {
      "epoch": 3.42,
      "grad_norm": 2.2651941776275635,
      "learning_rate": 6.98e-06,
      "loss": 0.1474,
      "step": 350
    },
    {
      "epoch": 3.67,
      "grad_norm": 2.528773546218872,
      "learning_rate": 7.48e-06,
      "loss": 0.1493,
      "step": 375
    },
    {
      "epoch": 3.91,
      "grad_norm": 2.422232151031494,
      "learning_rate": 7.980000000000002e-06,
      "loss": 0.142,
      "step": 400
    },
    {
      "epoch": 4.16,
      "grad_norm": 2.209630012512207,
      "learning_rate": 8.48e-06,
      "loss": 0.1126,
      "step": 425
    },
    {
      "epoch": 4.4,
      "grad_norm": 1.9945831298828125,
      "learning_rate": 8.98e-06,
      "loss": 0.0909,
      "step": 450
    },
    {
      "epoch": 4.65,
      "grad_norm": 1.8972020149230957,
      "learning_rate": 9.48e-06,
      "loss": 0.0892,
      "step": 475
    },
    {
      "epoch": 4.89,
      "grad_norm": 2.345607042312622,
      "learning_rate": 9.980000000000001e-06,
      "loss": 0.0909,
      "step": 500
    },
    {
      "epoch": 5.13,
      "grad_norm": 1.3913277387619019,
      "learning_rate": 9.946666666666667e-06,
      "loss": 0.0713,
      "step": 525
    },
    {
      "epoch": 5.38,
      "grad_norm": 2.5306901931762695,
      "learning_rate": 9.891111111111113e-06,
      "loss": 0.0535,
      "step": 550
    },
    {
      "epoch": 5.62,
      "grad_norm": 1.3689407110214233,
      "learning_rate": 9.835555555555556e-06,
      "loss": 0.051,
      "step": 575
    },
    {
      "epoch": 5.87,
      "grad_norm": 2.091756820678711,
      "learning_rate": 9.780000000000001e-06,
      "loss": 0.0573,
      "step": 600
    },
    {
      "epoch": 6.11,
      "grad_norm": 1.5179609060287476,
      "learning_rate": 9.724444444444445e-06,
      "loss": 0.0454,
      "step": 625
    },
    {
      "epoch": 6.36,
      "grad_norm": 1.3439680337905884,
      "learning_rate": 9.66888888888889e-06,
      "loss": 0.029,
      "step": 650
    },
    {
      "epoch": 6.6,
      "grad_norm": 1.279569387435913,
      "learning_rate": 9.613333333333335e-06,
      "loss": 0.032,
      "step": 675
    },
    {
      "epoch": 6.85,
      "grad_norm": 1.6171939373016357,
      "learning_rate": 9.557777777777777e-06,
      "loss": 0.0314,
      "step": 700
    },
    {
      "epoch": 7.09,
      "grad_norm": 0.8949939608573914,
      "learning_rate": 9.502222222222223e-06,
      "loss": 0.0245,
      "step": 725
    },
    {
      "epoch": 7.33,
      "grad_norm": 1.3610639572143555,
      "learning_rate": 9.446666666666667e-06,
      "loss": 0.0165,
      "step": 750
    },
    {
      "epoch": 7.58,
      "grad_norm": 1.5446442365646362,
      "learning_rate": 9.391111111111111e-06,
      "loss": 0.0169,
      "step": 775
    },
    {
      "epoch": 7.82,
      "grad_norm": 1.2825671434402466,
      "learning_rate": 9.335555555555557e-06,
      "loss": 0.0193,
      "step": 800
    },
    {
      "epoch": 8.07,
      "grad_norm": 0.7063258290290833,
      "learning_rate": 9.280000000000001e-06,
      "loss": 0.0154,
      "step": 825
    },
    {
      "epoch": 8.31,
      "grad_norm": 0.72287917137146,
      "learning_rate": 9.224444444444445e-06,
      "loss": 0.0102,
      "step": 850
    },
    {
      "epoch": 8.56,
      "grad_norm": 1.2877657413482666,
      "learning_rate": 9.168888888888889e-06,
      "loss": 0.0093,
      "step": 875
    },
    {
      "epoch": 8.8,
      "grad_norm": 1.6262348890304565,
      "learning_rate": 9.113333333333335e-06,
      "loss": 0.0117,
      "step": 900
    },
    {
      "epoch": 9.05,
      "grad_norm": 0.49374520778656006,
      "learning_rate": 9.057777777777779e-06,
      "loss": 0.0121,
      "step": 925
    },
    {
      "epoch": 9.29,
      "grad_norm": 0.8290354013442993,
      "learning_rate": 9.002222222222223e-06,
      "loss": 0.0063,
      "step": 950
    },
    {
      "epoch": 9.54,
      "grad_norm": 0.9974227547645569,
      "learning_rate": 8.946666666666669e-06,
      "loss": 0.006,
      "step": 975
    },
    {
      "epoch": 9.78,
      "grad_norm": 0.8566415309906006,
      "learning_rate": 8.891111111111111e-06,
      "loss": 0.007,
      "step": 1000
    },
    {
      "epoch": 9.78,
      "eval_loss": 0.41129249334335327,
      "eval_runtime": 1450.197,
      "eval_samples_per_second": 1.996,
      "eval_steps_per_second": 0.499,
      "eval_wer": 18.926838201629888,
      "step": 1000
    }
  ],
  "logging_steps": 25,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 1000,
  "total_flos": 1.845907654606848e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}