{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 29.997830802603037,
  "global_step": 6900,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.43,
      "eval_loss": 3.6933705806732178,
      "eval_runtime": 193.3371,
      "eval_samples_per_second": 25.339,
      "eval_steps_per_second": 0.797,
      "eval_wer": 1.0,
      "step": 100
    },
    {
      "epoch": 0.87,
      "eval_loss": 3.0763044357299805,
      "eval_runtime": 193.2997,
      "eval_samples_per_second": 25.344,
      "eval_steps_per_second": 0.797,
      "eval_wer": 1.0,
      "step": 200
    },
    {
      "epoch": 1.3,
      "eval_loss": 2.9737112522125244,
      "eval_runtime": 192.6243,
      "eval_samples_per_second": 25.433,
      "eval_steps_per_second": 0.799,
      "eval_wer": 1.0,
      "step": 300
    },
    {
      "epoch": 1.74,
      "eval_loss": 2.5733721256256104,
      "eval_runtime": 192.0804,
      "eval_samples_per_second": 25.505,
      "eval_steps_per_second": 0.802,
      "eval_wer": 1.0,
      "step": 400
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0002982,
      "loss": 5.0957,
      "step": 500
    },
    {
      "epoch": 2.17,
      "eval_loss": 1.1900373697280884,
      "eval_runtime": 192.7964,
      "eval_samples_per_second": 25.41,
      "eval_steps_per_second": 0.799,
      "eval_wer": 0.9011191003946586,
      "step": 500
    },
    {
      "epoch": 2.61,
      "eval_loss": 0.9725514054298401,
      "eval_runtime": 192.2119,
      "eval_samples_per_second": 25.487,
      "eval_steps_per_second": 0.801,
      "eval_wer": 0.7572309023084824,
      "step": 600
    },
    {
      "epoch": 3.04,
      "eval_loss": 0.8960436582565308,
      "eval_runtime": 193.7378,
      "eval_samples_per_second": 25.287,
      "eval_steps_per_second": 0.795,
      "eval_wer": 0.6209114991620263,
      "step": 700
    },
    {
      "epoch": 3.48,
      "eval_loss": 0.7850536704063416,
      "eval_runtime": 193.8724,
      "eval_samples_per_second": 25.269,
      "eval_steps_per_second": 0.794,
      "eval_wer": 0.5515488998215927,
      "step": 800
    },
    {
      "epoch": 3.91,
      "eval_loss": 0.7271264791488647,
      "eval_runtime": 195.9681,
      "eval_samples_per_second": 24.999,
      "eval_steps_per_second": 0.786,
      "eval_wer": 0.511542412283073,
      "step": 900
    },
    {
      "epoch": 4.35,
      "learning_rate": 0.00024945762711864404,
      "loss": 1.0312,
      "step": 1000
    },
    {
      "epoch": 4.35,
      "eval_loss": 0.7053087949752808,
      "eval_runtime": 192.6577,
      "eval_samples_per_second": 25.429,
      "eval_steps_per_second": 0.799,
      "eval_wer": 0.49548575444666704,
      "step": 1000
    },
    {
      "epoch": 4.78,
      "eval_loss": 0.6822740435600281,
      "eval_runtime": 193.7373,
      "eval_samples_per_second": 25.287,
      "eval_steps_per_second": 0.795,
      "eval_wer": 0.47369843758447316,
      "step": 1100
    },
    {
      "epoch": 5.22,
      "eval_loss": 0.6767629384994507,
      "eval_runtime": 192.3102,
      "eval_samples_per_second": 25.474,
      "eval_steps_per_second": 0.801,
      "eval_wer": 0.45945288425150027,
      "step": 1200
    },
    {
      "epoch": 5.65,
      "eval_loss": 0.6635140776634216,
      "eval_runtime": 192.3892,
      "eval_samples_per_second": 25.464,
      "eval_steps_per_second": 0.8,
      "eval_wer": 0.4488295399253933,
      "step": 1300
    },
    {
      "epoch": 6.09,
      "eval_loss": 0.6602210402488708,
      "eval_runtime": 191.9311,
      "eval_samples_per_second": 25.525,
      "eval_steps_per_second": 0.802,
      "eval_wer": 0.4389630750932584,
      "step": 1400
    },
    {
      "epoch": 6.52,
      "learning_rate": 0.0001986101694915254,
      "loss": 0.6815,
      "step": 1500
    },
    {
      "epoch": 6.52,
      "eval_loss": 0.6463663578033447,
      "eval_runtime": 193.1979,
      "eval_samples_per_second": 25.357,
      "eval_steps_per_second": 0.797,
      "eval_wer": 0.4310158404065524,
      "step": 1500
    },
    {
      "epoch": 6.95,
      "eval_loss": 0.6455022096633911,
      "eval_runtime": 198.417,
      "eval_samples_per_second": 24.69,
      "eval_steps_per_second": 0.776,
      "eval_wer": 0.43942260907174135,
      "step": 1600
    },
    {
      "epoch": 7.39,
      "eval_loss": 0.6629974842071533,
      "eval_runtime": 197.9313,
      "eval_samples_per_second": 24.751,
      "eval_steps_per_second": 0.778,
      "eval_wer": 0.4311780288695464,
      "step": 1700
    },
    {
      "epoch": 7.82,
      "eval_loss": 0.6521152853965759,
      "eval_runtime": 202.8016,
      "eval_samples_per_second": 24.157,
      "eval_steps_per_second": 0.759,
      "eval_wer": 0.4125804184462345,
      "step": 1800
    },
    {
      "epoch": 8.26,
      "eval_loss": 0.6281991600990295,
      "eval_runtime": 195.8017,
      "eval_samples_per_second": 25.02,
      "eval_steps_per_second": 0.787,
      "eval_wer": 0.42839379358814944,
      "step": 1900
    },
    {
      "epoch": 8.69,
      "learning_rate": 0.00014776271186440676,
      "loss": 0.544,
      "step": 2000
    },
    {
      "epoch": 8.69,
      "eval_loss": 0.6247670650482178,
      "eval_runtime": 197.0538,
      "eval_samples_per_second": 24.861,
      "eval_steps_per_second": 0.782,
      "eval_wer": 0.4178245120830405,
      "step": 2000
    },
    {
      "epoch": 9.13,
      "eval_loss": 0.6510103940963745,
      "eval_runtime": 199.9527,
      "eval_samples_per_second": 24.501,
      "eval_steps_per_second": 0.77,
      "eval_wer": 0.4103638427853165,
      "step": 2100
    },
    {
      "epoch": 9.56,
      "eval_loss": 0.6527135968208313,
      "eval_runtime": 202.4561,
      "eval_samples_per_second": 24.198,
      "eval_steps_per_second": 0.761,
      "eval_wer": 0.4012542574471536,
      "step": 2200
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.6510817408561707,
      "eval_runtime": 202.1587,
      "eval_samples_per_second": 24.233,
      "eval_steps_per_second": 0.762,
      "eval_wer": 0.40641725685246255,
      "step": 2300
    },
    {
      "epoch": 10.43,
      "eval_loss": 0.6734431982040405,
      "eval_runtime": 202.8997,
      "eval_samples_per_second": 24.145,
      "eval_steps_per_second": 0.759,
      "eval_wer": 0.4060928799264746,
      "step": 2400
    },
    {
      "epoch": 10.87,
      "learning_rate": 9.691525423728813e-05,
      "loss": 0.4478,
      "step": 2500
    },
    {
      "epoch": 10.87,
      "eval_loss": 0.6756427884101868,
      "eval_runtime": 201.7844,
      "eval_samples_per_second": 24.278,
      "eval_steps_per_second": 0.763,
      "eval_wer": 0.41452668000216253,
      "step": 2500
    },
    {
      "epoch": 11.3,
      "eval_loss": 0.6727204918861389,
      "eval_runtime": 202.413,
      "eval_samples_per_second": 24.203,
      "eval_steps_per_second": 0.761,
      "eval_wer": 0.3989565875547386,
      "step": 2600
    },
    {
      "epoch": 11.74,
      "eval_loss": 0.6618625521659851,
      "eval_runtime": 205.4327,
      "eval_samples_per_second": 23.847,
      "eval_steps_per_second": 0.75,
      "eval_wer": 0.4007406606476726,
      "step": 2700
    },
    {
      "epoch": 12.17,
      "eval_loss": 0.6613638997077942,
      "eval_runtime": 202.1353,
      "eval_samples_per_second": 24.236,
      "eval_steps_per_second": 0.762,
      "eval_wer": 0.40193004270962857,
      "step": 2800
    },
    {
      "epoch": 12.61,
      "eval_loss": 0.6695073246955872,
      "eval_runtime": 202.6466,
      "eval_samples_per_second": 24.175,
      "eval_steps_per_second": 0.76,
      "eval_wer": 0.40036222090068657,
      "step": 2900
    },
    {
      "epoch": 13.04,
      "learning_rate": 4.606779661016948e-05,
      "loss": 0.3919,
      "step": 3000
    },
    {
      "epoch": 13.04,
      "eval_loss": 0.6778153777122498,
      "eval_runtime": 202.4696,
      "eval_samples_per_second": 24.196,
      "eval_steps_per_second": 0.761,
      "eval_wer": 0.3966048548413256,
      "step": 3000
    },
    {
      "epoch": 13.48,
      "eval_loss": 0.6872451305389404,
      "eval_runtime": 201.7642,
      "eval_samples_per_second": 24.281,
      "eval_steps_per_second": 0.763,
      "eval_wer": 0.3971184516408066,
      "step": 3100
    },
    {
      "epoch": 13.91,
      "eval_loss": 0.688228964805603,
      "eval_runtime": 204.8269,
      "eval_samples_per_second": 23.918,
      "eval_steps_per_second": 0.752,
      "eval_wer": 0.39452343623290265,
      "step": 3200
    },
    {
      "epoch": 14.35,
      "eval_loss": 0.7177483439445496,
      "eval_runtime": 201.1511,
      "eval_samples_per_second": 24.355,
      "eval_steps_per_second": 0.766,
      "eval_wer": 0.4009839433421636,
      "step": 3300
    },
    {
      "epoch": 14.78,
      "eval_loss": 0.6887663006782532,
      "eval_runtime": 201.3869,
      "eval_samples_per_second": 24.326,
      "eval_steps_per_second": 0.765,
      "eval_wer": 0.4042547440125426,
      "step": 3400
    },
    {
      "epoch": 15.22,
      "learning_rate": 0.000159515625,
      "loss": 0.3767,
      "step": 3500
    },
    {
      "epoch": 15.22,
      "eval_loss": 0.7124109268188477,
      "eval_runtime": 201.401,
      "eval_samples_per_second": 24.325,
      "eval_steps_per_second": 0.765,
      "eval_wer": 0.4202303076174515,
      "step": 3500
    },
    {
      "epoch": 15.65,
      "eval_loss": 0.7275710701942444,
      "eval_runtime": 201.0133,
      "eval_samples_per_second": 24.372,
      "eval_steps_per_second": 0.766,
      "eval_wer": 0.41201275882575555,
      "step": 3600
    },
    {
      "epoch": 16.09,
      "eval_loss": 0.7265337109565735,
      "eval_runtime": 201.5776,
      "eval_samples_per_second": 24.303,
      "eval_steps_per_second": 0.764,
      "eval_wer": 0.40336270746607555,
      "step": 3700
    },
    {
      "epoch": 16.52,
      "eval_loss": 0.7392393946647644,
      "eval_runtime": 200.4049,
      "eval_samples_per_second": 24.446,
      "eval_steps_per_second": 0.768,
      "eval_wer": 0.40771476455641453,
      "step": 3800
    },
    {
      "epoch": 16.95,
      "eval_loss": 0.7403403520584106,
      "eval_runtime": 200.8623,
      "eval_samples_per_second": 24.39,
      "eval_steps_per_second": 0.767,
      "eval_wer": 0.39652376060982863,
      "step": 3900
    },
    {
      "epoch": 17.39,
      "learning_rate": 0.000136078125,
      "loss": 0.3603,
      "step": 4000
    },
    {
      "epoch": 17.39,
      "eval_loss": 0.7444584369659424,
      "eval_runtime": 203.8048,
      "eval_samples_per_second": 24.038,
      "eval_steps_per_second": 0.756,
      "eval_wer": 0.4016326971941396,
      "step": 4000
    },
    {
      "epoch": 17.82,
      "eval_loss": 0.757925271987915,
      "eval_runtime": 201.3248,
      "eval_samples_per_second": 24.334,
      "eval_steps_per_second": 0.765,
      "eval_wer": 0.4012272260366546,
      "step": 4100
    },
    {
      "epoch": 18.26,
      "eval_loss": 0.7225211262702942,
      "eval_runtime": 200.7191,
      "eval_samples_per_second": 24.407,
      "eval_steps_per_second": 0.767,
      "eval_wer": 0.3962804779153376,
      "step": 4200
    },
    {
      "epoch": 18.69,
      "eval_loss": 0.7354857325553894,
      "eval_runtime": 200.5207,
      "eval_samples_per_second": 24.431,
      "eval_steps_per_second": 0.768,
      "eval_wer": 0.3951181272638806,
      "step": 4300
    },
    {
      "epoch": 19.13,
      "eval_loss": 0.7482349276542664,
      "eval_runtime": 200.7847,
      "eval_samples_per_second": 24.399,
      "eval_steps_per_second": 0.767,
      "eval_wer": 0.39252311185597666,
      "step": 4400
    },
    {
      "epoch": 19.56,
      "learning_rate": 0.000112640625,
      "loss": 0.3153,
      "step": 4500
    },
    {
      "epoch": 19.56,
      "eval_loss": 0.772259533405304,
      "eval_runtime": 200.8729,
      "eval_samples_per_second": 24.389,
      "eval_steps_per_second": 0.767,
      "eval_wer": 0.3971725144618046,
      "step": 4500
    },
    {
      "epoch": 20.0,
      "eval_loss": 0.7469175457954407,
      "eval_runtime": 200.5892,
      "eval_samples_per_second": 24.423,
      "eval_steps_per_second": 0.768,
      "eval_wer": 0.3897929393955777,
      "step": 4600
    },
    {
      "epoch": 20.43,
      "eval_loss": 0.7799847722053528,
      "eval_runtime": 200.7406,
      "eval_samples_per_second": 24.405,
      "eval_steps_per_second": 0.767,
      "eval_wer": 0.39444234200140565,
      "step": 4700
    },
    {
      "epoch": 20.87,
      "eval_loss": 0.7827086448669434,
      "eval_runtime": 200.8886,
      "eval_samples_per_second": 24.387,
      "eval_steps_per_second": 0.767,
      "eval_wer": 0.3897388765745797,
      "step": 4800
    },
    {
      "epoch": 21.3,
      "eval_loss": 0.7935096621513367,
      "eval_runtime": 204.4398,
      "eval_samples_per_second": 23.963,
      "eval_steps_per_second": 0.753,
      "eval_wer": 0.3913607612045196,
      "step": 4900
    },
    {
      "epoch": 21.74,
      "learning_rate": 8.9203125e-05,
      "loss": 0.286,
      "step": 5000
    },
    {
      "epoch": 21.74,
      "eval_loss": 0.7984351515769958,
      "eval_runtime": 200.4244,
      "eval_samples_per_second": 24.443,
      "eval_steps_per_second": 0.768,
      "eval_wer": 0.37497972644212574,
      "step": 5000
    },
    {
      "epoch": 22.17,
      "eval_loss": 0.7945317625999451,
      "eval_runtime": 200.6266,
      "eval_samples_per_second": 24.418,
      "eval_steps_per_second": 0.768,
      "eval_wer": 0.3830080553603287,
      "step": 5100
    },
    {
      "epoch": 22.61,
      "eval_loss": 0.8010774254798889,
      "eval_runtime": 200.4942,
      "eval_samples_per_second": 24.435,
      "eval_steps_per_second": 0.768,
      "eval_wer": 0.37746661620803373,
      "step": 5200
    },
    {
      "epoch": 23.04,
      "eval_loss": 0.7977942228317261,
      "eval_runtime": 201.0949,
      "eval_samples_per_second": 24.362,
      "eval_steps_per_second": 0.766,
      "eval_wer": 0.3823863329188517,
      "step": 5300
    },
    {
      "epoch": 23.48,
      "eval_loss": 0.8160513043403625,
      "eval_runtime": 201.474,
      "eval_samples_per_second": 24.316,
      "eval_steps_per_second": 0.764,
      "eval_wer": 0.3833324322863167,
      "step": 5400
    },
    {
      "epoch": 23.91,
      "learning_rate": 6.576562499999999e-05,
      "loss": 0.2615,
      "step": 5500
    },
    {
      "epoch": 23.91,
      "eval_loss": 0.7822675704956055,
      "eval_runtime": 200.6337,
      "eval_samples_per_second": 24.418,
      "eval_steps_per_second": 0.768,
      "eval_wer": 0.3857922906417257,
      "step": 5500
    },
    {
      "epoch": 24.35,
      "eval_loss": 0.8311657309532166,
      "eval_runtime": 201.9809,
      "eval_samples_per_second": 24.255,
      "eval_steps_per_second": 0.762,
      "eval_wer": 0.3863329188517057,
      "step": 5600
    },
    {
      "epoch": 24.78,
      "eval_loss": 0.8427298069000244,
      "eval_runtime": 201.3389,
      "eval_samples_per_second": 24.332,
      "eval_steps_per_second": 0.765,
      "eval_wer": 0.3818997675298697,
      "step": 5700
    },
    {
      "epoch": 25.22,
      "eval_loss": 0.8432016968727112,
      "eval_runtime": 205.0165,
      "eval_samples_per_second": 23.896,
      "eval_steps_per_second": 0.751,
      "eval_wer": 0.3801697572579337,
      "step": 5800
    },
    {
      "epoch": 25.65,
      "eval_loss": 0.8285790085792542,
      "eval_runtime": 200.9997,
      "eval_samples_per_second": 24.373,
      "eval_steps_per_second": 0.766,
      "eval_wer": 0.37943990917446074,
      "step": 5900
    },
    {
      "epoch": 26.09,
      "learning_rate": 4.2328124999999994e-05,
      "loss": 0.2408,
      "step": 6000
    },
    {
      "epoch": 26.09,
      "eval_loss": 0.8224411606788635,
      "eval_runtime": 200.5192,
      "eval_samples_per_second": 24.432,
      "eval_steps_per_second": 0.768,
      "eval_wer": 0.3824403957398497,
      "step": 6000
    },
    {
      "epoch": 26.52,
      "eval_loss": 0.8228213787078857,
      "eval_runtime": 201.4468,
      "eval_samples_per_second": 24.319,
      "eval_steps_per_second": 0.764,
      "eval_wer": 0.3823052386873547,
      "step": 6100
    },
    {
      "epoch": 26.95,
      "eval_loss": 0.832374095916748,
      "eval_runtime": 200.428,
      "eval_samples_per_second": 24.443,
      "eval_steps_per_second": 0.768,
      "eval_wer": 0.3794669405849597,
      "step": 6200
    },
    {
      "epoch": 27.39,
      "eval_loss": 0.8564098477363586,
      "eval_runtime": 199.9963,
      "eval_samples_per_second": 24.495,
      "eval_steps_per_second": 0.77,
      "eval_wer": 0.37441206682164674,
      "step": 6300
    },
    {
      "epoch": 27.82,
      "eval_loss": 0.8629336953163147,
      "eval_runtime": 200.1836,
      "eval_samples_per_second": 24.473,
      "eval_steps_per_second": 0.769,
      "eval_wer": 0.37738552197653674,
      "step": 6400
    },
    {
      "epoch": 28.26,
      "learning_rate": 1.8890625e-05,
      "loss": 0.2254,
      "step": 6500
    },
    {
      "epoch": 28.26,
      "eval_loss": 0.8544773459434509,
      "eval_runtime": 200.1232,
      "eval_samples_per_second": 24.48,
      "eval_steps_per_second": 0.77,
      "eval_wer": 0.3778450559550197,
      "step": 6500
    },
    {
      "epoch": 28.69,
      "eval_loss": 0.8491827845573425,
      "eval_runtime": 201.2475,
      "eval_samples_per_second": 24.343,
      "eval_steps_per_second": 0.765,
      "eval_wer": 0.3767097367140617,
      "step": 6600
    },
    {
      "epoch": 29.13,
      "eval_loss": 0.8510707020759583,
      "eval_runtime": 203.9748,
      "eval_samples_per_second": 24.018,
      "eval_steps_per_second": 0.755,
      "eval_wer": 0.37508785208412176,
      "step": 6700
    },
    {
      "epoch": 29.56,
      "eval_loss": 0.8490829467773438,
      "eval_runtime": 200.922,
      "eval_samples_per_second": 24.383,
      "eval_steps_per_second": 0.766,
      "eval_wer": 0.3752770719576147,
      "step": 6800
    },
    {
      "epoch": 30.0,
      "eval_loss": 0.8535701632499695,
      "eval_runtime": 201.4279,
      "eval_samples_per_second": 24.321,
      "eval_steps_per_second": 0.765,
      "eval_wer": 0.3736551873276748,
      "step": 6900
    },
    {
      "epoch": 30.0,
      "step": 6900,
      "total_flos": 5.975239123849969e+19,
      "train_loss": 0.15137778351272363,
      "train_runtime": 18309.2993,
      "train_samples_per_second": 24.135,
      "train_steps_per_second": 0.377
    }
  ],
  "max_steps": 6900,
  "num_train_epochs": 30,
  "total_flos": 5.975239123849969e+19,
  "trial_name": null,
  "trial_params": null
}