{
  "best_metric": 9.288054935192335,
  "best_model_checkpoint": "./checkpoint-900",
  "epoch": 13.01525,
  "eval_steps": 100,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 4.974489795918367e-07,
      "loss": 0.1595,
      "step": 25
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.846938775510204e-07,
      "loss": 0.1463,
      "step": 50
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.719387755102041e-07,
      "loss": 0.1424,
      "step": 75
    },
    {
      "epoch": 1.02,
      "learning_rate": 4.5918367346938775e-07,
      "loss": 0.1368,
      "step": 100
    },
    {
      "epoch": 1.02,
      "eval_loss": 0.16647553443908691,
      "eval_runtime": 1871.163,
      "eval_samples_per_second": 1.66,
      "eval_steps_per_second": 0.104,
      "eval_wer": 9.69505958965582,
      "step": 100
    },
    {
      "epoch": 1.05,
      "learning_rate": 4.464285714285714e-07,
      "loss": 0.1338,
      "step": 125
    },
    {
      "epoch": 1.07,
      "learning_rate": 4.336734693877551e-07,
      "loss": 0.1281,
      "step": 150
    },
    {
      "epoch": 2.02,
      "learning_rate": 4.2091836734693876e-07,
      "loss": 0.1275,
      "step": 175
    },
    {
      "epoch": 2.05,
      "learning_rate": 4.0816326530612243e-07,
      "loss": 0.1196,
      "step": 200
    },
    {
      "epoch": 2.05,
      "eval_loss": 0.15860922634601593,
      "eval_runtime": 1883.9234,
      "eval_samples_per_second": 1.649,
      "eval_steps_per_second": 0.104,
      "eval_wer": 9.601135438625786,
      "step": 200
    },
    {
      "epoch": 2.07,
      "learning_rate": 3.9540816326530615e-07,
      "loss": 0.1189,
      "step": 225
    },
    {
      "epoch": 3.02,
      "learning_rate": 3.8265306122448977e-07,
      "loss": 0.1189,
      "step": 250
    },
    {
      "epoch": 3.05,
      "learning_rate": 3.6989795918367343e-07,
      "loss": 0.1136,
      "step": 275
    },
    {
      "epoch": 3.07,
      "learning_rate": 3.5714285714285716e-07,
      "loss": 0.107,
      "step": 300
    },
    {
      "epoch": 3.07,
      "eval_loss": 0.15376655757427216,
      "eval_runtime": 1823.9097,
      "eval_samples_per_second": 1.703,
      "eval_steps_per_second": 0.107,
      "eval_wer": 9.400763916428376,
      "step": 300
    },
    {
      "epoch": 4.02,
      "learning_rate": 3.443877551020408e-07,
      "loss": 0.1128,
      "step": 325
    },
    {
      "epoch": 4.05,
      "learning_rate": 3.3163265306122444e-07,
      "loss": 0.1085,
      "step": 350
    },
    {
      "epoch": 4.07,
      "learning_rate": 3.1887755102040816e-07,
      "loss": 0.1035,
      "step": 375
    },
    {
      "epoch": 5.02,
      "learning_rate": 3.0612244897959183e-07,
      "loss": 0.1051,
      "step": 400
    },
    {
      "epoch": 5.02,
      "eval_loss": 0.15036460757255554,
      "eval_runtime": 1824.8717,
      "eval_samples_per_second": 1.703,
      "eval_steps_per_second": 0.107,
      "eval_wer": 9.323537392248127,
      "step": 400
    },
    {
      "epoch": 5.05,
      "learning_rate": 2.933673469387755e-07,
      "loss": 0.102,
      "step": 425
    },
    {
      "epoch": 5.07,
      "learning_rate": 2.8061224489795917e-07,
      "loss": 0.101,
      "step": 450
    },
    {
      "epoch": 6.02,
      "learning_rate": 2.6785714285714284e-07,
      "loss": 0.0995,
      "step": 475
    },
    {
      "epoch": 6.05,
      "learning_rate": 2.551020408163265e-07,
      "loss": 0.0988,
      "step": 500
    },
    {
      "epoch": 6.05,
      "eval_loss": 0.14860908687114716,
      "eval_runtime": 1843.7431,
      "eval_samples_per_second": 1.685,
      "eval_steps_per_second": 0.106,
      "eval_wer": 9.446682390265284,
      "step": 500
    },
    {
      "epoch": 6.07,
      "learning_rate": 2.423469387755102e-07,
      "loss": 0.0955,
      "step": 525
    },
    {
      "epoch": 7.02,
      "learning_rate": 2.2959183673469388e-07,
      "loss": 0.0952,
      "step": 550
    },
    {
      "epoch": 7.04,
      "learning_rate": 2.1683673469387754e-07,
      "loss": 0.0973,
      "step": 575
    },
    {
      "epoch": 7.07,
      "learning_rate": 2.0408163265306121e-07,
      "loss": 0.0939,
      "step": 600
    },
    {
      "epoch": 7.07,
      "eval_loss": 0.14743566513061523,
      "eval_runtime": 1839.7788,
      "eval_samples_per_second": 1.689,
      "eval_steps_per_second": 0.106,
      "eval_wer": 9.442507983552838,
      "step": 600
    },
    {
      "epoch": 8.02,
      "learning_rate": 1.9132653061224488e-07,
      "loss": 0.0923,
      "step": 625
    },
    {
      "epoch": 8.04,
      "learning_rate": 1.7857142857142858e-07,
      "loss": 0.0933,
      "step": 650
    },
    {
      "epoch": 8.07,
      "learning_rate": 1.6581632653061222e-07,
      "loss": 0.0904,
      "step": 675
    },
    {
      "epoch": 9.02,
      "learning_rate": 1.5306122448979592e-07,
      "loss": 0.0901,
      "step": 700
    },
    {
      "epoch": 9.02,
      "eval_loss": 0.14642581343650818,
      "eval_runtime": 1823.9218,
      "eval_samples_per_second": 1.703,
      "eval_steps_per_second": 0.107,
      "eval_wer": 9.300578155329674,
      "step": 700
    },
    {
      "epoch": 9.04,
      "learning_rate": 1.4030612244897959e-07,
      "loss": 0.0894,
      "step": 725
    },
    {
      "epoch": 9.07,
      "learning_rate": 1.2755102040816326e-07,
      "loss": 0.0898,
      "step": 750
    },
    {
      "epoch": 10.02,
      "learning_rate": 1.1479591836734694e-07,
      "loss": 0.0891,
      "step": 775
    },
    {
      "epoch": 10.04,
      "learning_rate": 1.0204081632653061e-07,
      "loss": 0.0859,
      "step": 800
    },
    {
      "epoch": 10.04,
      "eval_loss": 0.14592841267585754,
      "eval_runtime": 1840.3966,
      "eval_samples_per_second": 1.688,
      "eval_steps_per_second": 0.106,
      "eval_wer": 9.436246373484169,
      "step": 800
    },
    {
      "epoch": 10.07,
      "learning_rate": 8.928571428571429e-08,
      "loss": 0.0884,
      "step": 825
    },
    {
      "epoch": 11.02,
      "learning_rate": 7.653061224489796e-08,
      "loss": 0.0881,
      "step": 850
    },
    {
      "epoch": 11.04,
      "learning_rate": 6.377551020408163e-08,
      "loss": 0.0877,
      "step": 875
    },
    {
      "epoch": 11.07,
      "learning_rate": 5.1020408163265303e-08,
      "loss": 0.0859,
      "step": 900
    },
    {
      "epoch": 11.07,
      "eval_loss": 0.14575552940368652,
      "eval_runtime": 1821.0167,
      "eval_samples_per_second": 1.706,
      "eval_steps_per_second": 0.107,
      "eval_wer": 9.288054935192335,
      "step": 900
    },
    {
      "epoch": 12.02,
      "learning_rate": 3.826530612244898e-08,
      "loss": 0.0864,
      "step": 925
    },
    {
      "epoch": 12.04,
      "learning_rate": 2.5510204081632652e-08,
      "loss": 0.0869,
      "step": 950
    },
    {
      "epoch": 12.07,
      "learning_rate": 1.2755102040816326e-08,
      "loss": 0.0876,
      "step": 975
    },
    {
      "epoch": 13.02,
      "learning_rate": 0.0,
      "loss": 0.0839,
      "step": 1000
    },
    {
      "epoch": 13.02,
      "eval_loss": 0.14563478529453278,
      "eval_runtime": 1822.7536,
      "eval_samples_per_second": 1.705,
      "eval_steps_per_second": 0.107,
      "eval_wer": 9.290142138548559,
      "step": 1000
    },
    {
      "epoch": 13.02,
      "step": 1000,
      "total_flos": 1.846196240007168e+19,
      "train_loss": 0.10451480412483215,
      "train_runtime": 27555.4596,
      "train_samples_per_second": 2.323,
      "train_steps_per_second": 0.036
    }
  ],
  "logging_steps": 25,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 100,
  "total_flos": 1.846196240007168e+19,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}