{
  "best_metric": 20.214938516576623,
  "best_model_checkpoint": "./checkpoint-5000",
  "epoch": 1.0,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 4.973833272194737e-06,
      "loss": 0.4248,
      "step": 25
    },
    {
      "epoch": 0.01,
      "learning_rate": 6.195318418690893e-06,
      "loss": 0.3353,
      "step": 50
    },
    {
      "epoch": 0.01,
      "learning_rate": 6.881634451095711e-06,
      "loss": 0.3227,
      "step": 75
    },
    {
      "epoch": 0.02,
      "learning_rate": 7.344547104469332e-06,
      "loss": 0.2826,
      "step": 100
    },
    {
      "epoch": 0.03,
      "learning_rate": 7.716963756434345e-06,
      "loss": 0.3717,
      "step": 125
    },
    {
      "epoch": 0.03,
      "learning_rate": 8.019180844200955e-06,
      "loss": 0.3568,
      "step": 150
    },
    {
      "epoch": 0.04,
      "learning_rate": 8.27351214279797e-06,
      "loss": 0.3498,
      "step": 175
    },
    {
      "epoch": 0.04,
      "learning_rate": 8.49307723936858e-06,
      "loss": 0.3809,
      "step": 200
    },
    {
      "epoch": 0.04,
      "learning_rate": 8.686247975778677e-06,
      "loss": 0.3468,
      "step": 225
    },
    {
      "epoch": 0.05,
      "learning_rate": 8.858694625217149e-06,
      "loss": 0.3089,
      "step": 250
    },
    {
      "epoch": 0.06,
      "learning_rate": 9.014436199608479e-06,
      "loss": 0.3316,
      "step": 275
    },
    {
      "epoch": 0.06,
      "learning_rate": 9.156425255148058e-06,
      "loss": 0.3564,
      "step": 300
    },
    {
      "epoch": 0.07,
      "learning_rate": 9.28689473531776e-06,
      "loss": 0.2979,
      "step": 325
    },
    {
      "epoch": 0.07,
      "learning_rate": 9.407574351377137e-06,
      "loss": 0.321,
      "step": 350
    },
    {
      "epoch": 0.07,
      "learning_rate": 9.519831289296397e-06,
      "loss": 0.2998,
      "step": 375
    },
    {
      "epoch": 0.08,
      "learning_rate": 9.624764935335318e-06,
      "loss": 0.2896,
      "step": 400
    },
    {
      "epoch": 0.09,
      "learning_rate": 9.723272550712454e-06,
      "loss": 0.2802,
      "step": 425
    },
    {
      "epoch": 0.09,
      "learning_rate": 9.816095971633122e-06,
      "loss": 0.3408,
      "step": 450
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.900435550016748e-06,
      "loss": 0.3171,
      "step": 475
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.98382788472848e-06,
      "loss": 0.3302,
      "step": 500
    },
    {
      "epoch": 0.1,
      "learning_rate": 1e-05,
      "loss": 0.277,
      "step": 525
    },
    {
      "epoch": 0.11,
      "learning_rate": 1e-05,
      "loss": 0.2987,
      "step": 550
    },
    {
      "epoch": 0.12,
      "learning_rate": 1e-05,
      "loss": 0.2941,
      "step": 575
    },
    {
      "epoch": 0.12,
      "learning_rate": 1e-05,
      "loss": 0.2529,
      "step": 600
    },
    {
      "epoch": 0.12,
      "learning_rate": 1e-05,
      "loss": 0.339,
      "step": 625
    },
    {
      "epoch": 0.13,
      "learning_rate": 1e-05,
      "loss": 0.2919,
      "step": 650
    },
    {
      "epoch": 0.14,
      "learning_rate": 1e-05,
      "loss": 0.3202,
      "step": 675
    },
    {
      "epoch": 0.14,
      "learning_rate": 1e-05,
      "loss": 0.3154,
      "step": 700
    },
    {
      "epoch": 0.14,
      "learning_rate": 1e-05,
      "loss": 0.3017,
      "step": 725
    },
    {
      "epoch": 0.15,
      "learning_rate": 1e-05,
      "loss": 0.3514,
      "step": 750
    },
    {
      "epoch": 0.15,
      "learning_rate": 1e-05,
      "loss": 0.3026,
      "step": 775
    },
    {
      "epoch": 0.16,
      "learning_rate": 1e-05,
      "loss": 0.302,
      "step": 800
    },
    {
      "epoch": 0.17,
      "learning_rate": 1e-05,
      "loss": 0.2944,
      "step": 825
    },
    {
      "epoch": 0.17,
      "learning_rate": 1e-05,
      "loss": 0.3156,
      "step": 850
    },
    {
      "epoch": 0.17,
      "learning_rate": 1e-05,
      "loss": 0.3407,
      "step": 875
    },
    {
      "epoch": 0.18,
      "learning_rate": 1e-05,
      "loss": 0.302,
      "step": 900
    },
    {
      "epoch": 0.18,
      "learning_rate": 1e-05,
      "loss": 0.2982,
      "step": 925
    },
    {
      "epoch": 0.19,
      "learning_rate": 1e-05,
      "loss": 0.294,
      "step": 950
    },
    {
      "epoch": 0.2,
      "learning_rate": 1e-05,
      "loss": 0.2874,
      "step": 975
    },
    {
      "epoch": 0.2,
      "learning_rate": 1e-05,
      "loss": 0.2767,
      "step": 1000
    },
    {
      "epoch": 0.2,
      "eval_loss": 1.0654296875,
      "eval_runtime": 2194.1241,
      "eval_samples_per_second": 2.577,
      "eval_steps_per_second": 0.161,
      "eval_wer": 25.397216227568325,
      "step": 1000
    },
    {
      "epoch": 0.2,
      "learning_rate": 1e-05,
      "loss": 0.2915,
      "step": 1025
    },
    {
      "epoch": 0.21,
      "learning_rate": 1e-05,
      "loss": 0.2937,
      "step": 1050
    },
    {
      "epoch": 0.21,
      "learning_rate": 1e-05,
      "loss": 0.2843,
      "step": 1075
    },
    {
      "epoch": 0.22,
      "learning_rate": 1e-05,
      "loss": 0.2789,
      "step": 1100
    },
    {
      "epoch": 0.23,
      "learning_rate": 1e-05,
      "loss": 0.28,
      "step": 1125
    },
    {
      "epoch": 0.23,
      "learning_rate": 1e-05,
      "loss": 0.2801,
      "step": 1150
    },
    {
      "epoch": 0.23,
      "learning_rate": 1e-05,
      "loss": 0.3144,
      "step": 1175
    },
    {
      "epoch": 0.24,
      "learning_rate": 1e-05,
      "loss": 0.3067,
      "step": 1200
    },
    {
      "epoch": 0.24,
      "learning_rate": 1e-05,
      "loss": 0.3055,
      "step": 1225
    },
    {
      "epoch": 0.25,
      "learning_rate": 1e-05,
      "loss": 0.345,
      "step": 1250
    },
    {
      "epoch": 0.26,
      "learning_rate": 1e-05,
      "loss": 0.2893,
      "step": 1275
    },
    {
      "epoch": 0.26,
      "learning_rate": 1e-05,
      "loss": 0.2749,
      "step": 1300
    },
    {
      "epoch": 0.27,
      "learning_rate": 1e-05,
      "loss": 0.3303,
      "step": 1325
    },
    {
      "epoch": 0.27,
      "learning_rate": 1e-05,
      "loss": 0.3312,
      "step": 1350
    },
    {
      "epoch": 0.28,
      "learning_rate": 1e-05,
      "loss": 0.2868,
      "step": 1375
    },
    {
      "epoch": 0.28,
      "learning_rate": 1e-05,
      "loss": 0.2923,
      "step": 1400
    },
    {
      "epoch": 0.28,
      "learning_rate": 1e-05,
      "loss": 0.2963,
      "step": 1425
    },
    {
      "epoch": 0.29,
      "learning_rate": 1e-05,
      "loss": 0.321,
      "step": 1450
    },
    {
      "epoch": 0.29,
      "learning_rate": 1e-05,
      "loss": 0.2929,
      "step": 1475
    },
    {
      "epoch": 0.3,
      "learning_rate": 1e-05,
      "loss": 0.2385,
      "step": 1500
    },
    {
      "epoch": 0.3,
      "learning_rate": 1e-05,
      "loss": 0.2804,
      "step": 1525
    },
    {
      "epoch": 0.31,
      "learning_rate": 1e-05,
      "loss": 0.2676,
      "step": 1550
    },
    {
      "epoch": 0.32,
      "learning_rate": 1e-05,
      "loss": 0.2758,
      "step": 1575
    },
    {
      "epoch": 0.32,
      "learning_rate": 1e-05,
      "loss": 0.2623,
      "step": 1600
    },
    {
      "epoch": 0.33,
      "learning_rate": 1e-05,
      "loss": 0.3237,
      "step": 1625
    },
    {
      "epoch": 0.33,
      "learning_rate": 1e-05,
      "loss": 0.2559,
      "step": 1650
    },
    {
      "epoch": 0.34,
      "learning_rate": 1e-05,
      "loss": 0.2841,
      "step": 1675
    },
    {
      "epoch": 0.34,
      "learning_rate": 1e-05,
      "loss": 0.2503,
      "step": 1700
    },
    {
      "epoch": 0.34,
      "learning_rate": 1e-05,
      "loss": 0.3199,
      "step": 1725
    },
    {
      "epoch": 0.35,
      "learning_rate": 1e-05,
      "loss": 0.2686,
      "step": 1750
    },
    {
      "epoch": 0.35,
      "learning_rate": 1e-05,
      "loss": 0.2463,
      "step": 1775
    },
    {
      "epoch": 0.36,
      "learning_rate": 1e-05,
      "loss": 0.2853,
      "step": 1800
    },
    {
      "epoch": 0.36,
      "learning_rate": 1e-05,
      "loss": 0.272,
      "step": 1825
    },
    {
      "epoch": 0.37,
      "learning_rate": 1e-05,
      "loss": 0.2859,
      "step": 1850
    },
    {
      "epoch": 0.38,
      "learning_rate": 1e-05,
      "loss": 0.2862,
      "step": 1875
    },
    {
      "epoch": 0.38,
      "learning_rate": 1e-05,
      "loss": 0.3104,
      "step": 1900
    },
    {
      "epoch": 0.39,
      "learning_rate": 1e-05,
      "loss": 0.2786,
      "step": 1925
    },
    {
      "epoch": 0.39,
      "learning_rate": 1e-05,
      "loss": 0.2828,
      "step": 1950
    },
    {
      "epoch": 0.4,
      "learning_rate": 1e-05,
      "loss": 0.2732,
      "step": 1975
    },
    {
      "epoch": 0.4,
      "learning_rate": 1e-05,
      "loss": 0.2528,
      "step": 2000
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.93701171875,
      "eval_runtime": 2189.9196,
      "eval_samples_per_second": 2.582,
      "eval_steps_per_second": 0.162,
      "eval_wer": 22.131135669075793,
      "step": 2000
    },
    {
      "epoch": 0.41,
      "learning_rate": 1e-05,
      "loss": 0.2352,
      "step": 2025
    },
    {
      "epoch": 0.41,
      "learning_rate": 1e-05,
      "loss": 0.2332,
      "step": 2050
    },
    {
      "epoch": 0.41,
      "learning_rate": 1e-05,
      "loss": 0.2303,
      "step": 2075
    },
    {
      "epoch": 0.42,
      "learning_rate": 1e-05,
      "loss": 0.2777,
      "step": 2100
    },
    {
      "epoch": 0.42,
      "learning_rate": 1e-05,
      "loss": 0.306,
      "step": 2125
    },
    {
      "epoch": 0.43,
      "learning_rate": 1e-05,
      "loss": 0.2996,
      "step": 2150
    },
    {
      "epoch": 0.43,
      "learning_rate": 1e-05,
      "loss": 0.2405,
      "step": 2175
    },
    {
      "epoch": 0.44,
      "learning_rate": 1e-05,
      "loss": 0.2519,
      "step": 2200
    },
    {
      "epoch": 0.45,
      "learning_rate": 1e-05,
      "loss": 0.3032,
      "step": 2225
    },
    {
      "epoch": 0.45,
      "learning_rate": 1e-05,
      "loss": 0.2881,
      "step": 2250
    },
    {
      "epoch": 0.46,
      "learning_rate": 1e-05,
      "loss": 0.2801,
      "step": 2275
    },
    {
      "epoch": 0.46,
      "learning_rate": 1e-05,
      "loss": 0.2945,
      "step": 2300
    },
    {
      "epoch": 0.47,
      "learning_rate": 1e-05,
      "loss": 0.2836,
      "step": 2325
    },
    {
      "epoch": 0.47,
      "learning_rate": 1e-05,
      "loss": 0.2651,
      "step": 2350
    },
    {
      "epoch": 0.47,
      "learning_rate": 1e-05,
      "loss": 0.2837,
      "step": 2375
    },
    {
      "epoch": 0.48,
      "learning_rate": 1e-05,
      "loss": 0.2498,
      "step": 2400
    },
    {
      "epoch": 0.48,
      "learning_rate": 1e-05,
      "loss": 0.2304,
      "step": 2425
    },
    {
      "epoch": 0.49,
      "learning_rate": 1e-05,
      "loss": 0.2475,
      "step": 2450
    },
    {
      "epoch": 0.49,
      "learning_rate": 1e-05,
      "loss": 0.2568,
      "step": 2475
    },
    {
      "epoch": 0.5,
      "learning_rate": 1e-05,
      "loss": 0.2431,
      "step": 2500
    },
    {
      "epoch": 0.51,
      "learning_rate": 1e-05,
      "loss": 0.2973,
      "step": 2525
    },
    {
      "epoch": 0.51,
      "learning_rate": 1e-05,
      "loss": 0.2509,
      "step": 2550
    },
    {
      "epoch": 0.52,
      "learning_rate": 1e-05,
      "loss": 0.2662,
      "step": 2575
    },
    {
      "epoch": 0.52,
      "learning_rate": 1e-05,
      "loss": 0.2174,
      "step": 2600
    },
    {
      "epoch": 0.53,
      "learning_rate": 1e-05,
      "loss": 0.2375,
      "step": 2625
    },
    {
      "epoch": 0.53,
      "learning_rate": 1e-05,
      "loss": 0.3,
      "step": 2650
    },
    {
      "epoch": 0.54,
      "learning_rate": 1e-05,
      "loss": 0.2586,
      "step": 2675
    },
    {
      "epoch": 0.54,
      "learning_rate": 1e-05,
      "loss": 0.2233,
      "step": 2700
    },
    {
      "epoch": 0.55,
      "learning_rate": 1e-05,
      "loss": 0.2458,
      "step": 2725
    },
    {
      "epoch": 0.55,
      "learning_rate": 1e-05,
      "loss": 0.2033,
      "step": 2750
    },
    {
      "epoch": 0.56,
      "learning_rate": 1e-05,
      "loss": 0.2896,
      "step": 2775
    },
    {
      "epoch": 0.56,
      "learning_rate": 1e-05,
      "loss": 0.269,
      "step": 2800
    },
    {
      "epoch": 0.56,
      "learning_rate": 1e-05,
      "loss": 0.2408,
      "step": 2825
    },
    {
      "epoch": 0.57,
      "learning_rate": 1e-05,
      "loss": 0.2474,
      "step": 2850
    },
    {
      "epoch": 0.57,
      "learning_rate": 1e-05,
      "loss": 0.2357,
      "step": 2875
    },
    {
      "epoch": 0.58,
      "learning_rate": 1e-05,
      "loss": 0.3061,
      "step": 2900
    },
    {
      "epoch": 0.58,
      "learning_rate": 1e-05,
      "loss": 0.3008,
      "step": 2925
    },
    {
      "epoch": 0.59,
      "learning_rate": 1e-05,
      "loss": 0.2849,
      "step": 2950
    },
    {
      "epoch": 0.59,
      "learning_rate": 1e-05,
      "loss": 0.2692,
      "step": 2975
    },
    {
      "epoch": 0.6,
      "learning_rate": 1e-05,
      "loss": 0.3038,
      "step": 3000
    },
    {
      "epoch": 0.6,
      "eval_loss": 0.99658203125,
      "eval_runtime": 2194.0772,
      "eval_samples_per_second": 2.577,
      "eval_steps_per_second": 0.161,
      "eval_wer": 20.575583332126357,
      "step": 3000
    },
    {
      "epoch": 0.6,
      "learning_rate": 1e-05,
      "loss": 0.2707,
      "step": 3025
    },
    {
      "epoch": 0.61,
      "learning_rate": 1e-05,
      "loss": 0.2503,
      "step": 3050
    },
    {
      "epoch": 0.61,
      "learning_rate": 1e-05,
      "loss": 0.2505,
      "step": 3075
    },
    {
      "epoch": 0.62,
      "learning_rate": 1e-05,
      "loss": 0.2751,
      "step": 3100
    },
    {
      "epoch": 0.62,
      "learning_rate": 1e-05,
      "loss": 0.2501,
      "step": 3125
    },
    {
      "epoch": 0.63,
      "learning_rate": 1e-05,
      "loss": 0.2805,
      "step": 3150
    },
    {
      "epoch": 0.64,
      "learning_rate": 1e-05,
      "loss": 0.2512,
      "step": 3175
    },
    {
      "epoch": 0.64,
      "learning_rate": 1e-05,
      "loss": 0.2928,
      "step": 3200
    },
    {
      "epoch": 0.65,
      "learning_rate": 1e-05,
      "loss": 0.2776,
      "step": 3225
    },
    {
      "epoch": 0.65,
      "learning_rate": 1e-05,
      "loss": 0.2497,
      "step": 3250
    },
    {
      "epoch": 0.66,
      "learning_rate": 1e-05,
      "loss": 0.2434,
      "step": 3275
    },
    {
      "epoch": 0.66,
      "learning_rate": 1e-05,
      "loss": 0.2577,
      "step": 3300
    },
    {
      "epoch": 0.67,
      "learning_rate": 1e-05,
      "loss": 0.241,
      "step": 3325
    },
    {
      "epoch": 0.67,
      "learning_rate": 1e-05,
      "loss": 0.2555,
      "step": 3350
    },
    {
      "epoch": 0.68,
      "learning_rate": 1e-05,
      "loss": 0.2963,
      "step": 3375
    },
    {
      "epoch": 0.68,
      "learning_rate": 1e-05,
      "loss": 0.2876,
      "step": 3400
    },
    {
      "epoch": 0.69,
      "learning_rate": 1e-05,
      "loss": 0.2444,
      "step": 3425
    },
    {
      "epoch": 0.69,
      "learning_rate": 1e-05,
      "loss": 0.2383,
      "step": 3450
    },
    {
      "epoch": 0.69,
      "learning_rate": 1e-05,
      "loss": 0.246,
      "step": 3475
    },
    {
      "epoch": 0.7,
      "learning_rate": 1e-05,
      "loss": 0.2611,
      "step": 3500
    },
    {
      "epoch": 0.7,
      "learning_rate": 1e-05,
      "loss": 0.2633,
      "step": 3525
    },
    {
      "epoch": 0.71,
      "learning_rate": 1e-05,
      "loss": 0.261,
      "step": 3550
    },
    {
      "epoch": 0.71,
      "learning_rate": 1e-05,
      "loss": 0.236,
      "step": 3575
    },
    {
      "epoch": 0.72,
      "learning_rate": 1e-05,
      "loss": 0.2509,
      "step": 3600
    },
    {
      "epoch": 0.72,
      "learning_rate": 1e-05,
      "loss": 0.2725,
      "step": 3625
    },
    {
      "epoch": 0.73,
      "learning_rate": 1e-05,
      "loss": 0.2433,
      "step": 3650
    },
    {
      "epoch": 0.73,
      "learning_rate": 1e-05,
      "loss": 0.2287,
      "step": 3675
    },
    {
      "epoch": 0.74,
      "learning_rate": 1e-05,
      "loss": 0.2596,
      "step": 3700
    },
    {
      "epoch": 0.74,
      "learning_rate": 1e-05,
      "loss": 0.247,
      "step": 3725
    },
    {
      "epoch": 0.75,
      "learning_rate": 1e-05,
      "loss": 0.2559,
      "step": 3750
    },
    {
      "epoch": 0.76,
      "learning_rate": 1e-05,
      "loss": 0.2356,
      "step": 3775
    },
    {
      "epoch": 0.76,
      "learning_rate": 1e-05,
      "loss": 0.2166,
      "step": 3800
    },
    {
      "epoch": 0.77,
      "learning_rate": 1e-05,
      "loss": 0.2397,
      "step": 3825
    },
    {
      "epoch": 0.77,
      "learning_rate": 1e-05,
      "loss": 0.2638,
      "step": 3850
    },
    {
      "epoch": 0.78,
      "learning_rate": 1e-05,
      "loss": 0.2146,
      "step": 3875
    },
    {
      "epoch": 0.78,
      "learning_rate": 1e-05,
      "loss": 0.2261,
      "step": 3900
    },
    {
      "epoch": 0.79,
      "learning_rate": 1e-05,
      "loss": 0.2542,
      "step": 3925
    },
    {
      "epoch": 0.79,
      "learning_rate": 1e-05,
      "loss": 0.2548,
      "step": 3950
    },
    {
      "epoch": 0.8,
      "learning_rate": 1e-05,
      "loss": 0.2075,
      "step": 3975
    },
    {
      "epoch": 0.8,
      "learning_rate": 1e-05,
      "loss": 0.2718,
      "step": 4000
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.8720703125,
      "eval_runtime": 2119.3543,
      "eval_samples_per_second": 2.668,
      "eval_steps_per_second": 0.167,
      "eval_wer": 24.929391828280924,
      "step": 4000
    },
    {
      "epoch": 0.81,
      "learning_rate": 1e-05,
      "loss": 0.2896,
      "step": 4025
    },
    {
      "epoch": 0.81,
      "learning_rate": 1e-05,
      "loss": 0.22,
      "step": 4050
    },
    {
      "epoch": 0.81,
      "learning_rate": 1e-05,
      "loss": 0.2534,
      "step": 4075
    },
    {
      "epoch": 0.82,
      "learning_rate": 1e-05,
      "loss": 0.251,
      "step": 4100
    },
    {
      "epoch": 0.82,
      "learning_rate": 1e-05,
      "loss": 0.2616,
      "step": 4125
    },
    {
      "epoch": 0.83,
      "learning_rate": 1e-05,
      "loss": 0.267,
      "step": 4150
    },
    {
      "epoch": 0.83,
      "learning_rate": 1e-05,
      "loss": 0.2211,
      "step": 4175
    },
    {
      "epoch": 0.84,
      "learning_rate": 1e-05,
      "loss": 0.2264,
      "step": 4200
    },
    {
      "epoch": 0.84,
      "learning_rate": 1e-05,
      "loss": 0.2683,
      "step": 4225
    },
    {
      "epoch": 0.85,
      "learning_rate": 1e-05,
      "loss": 0.2711,
      "step": 4250
    },
    {
      "epoch": 0.85,
      "learning_rate": 1e-05,
      "loss": 0.2582,
      "step": 4275
    },
    {
      "epoch": 0.86,
      "learning_rate": 1e-05,
      "loss": 0.2348,
      "step": 4300
    },
    {
      "epoch": 0.86,
      "learning_rate": 1e-05,
      "loss": 0.2457,
      "step": 4325
    },
    {
      "epoch": 0.87,
      "learning_rate": 1e-05,
      "loss": 0.2883,
      "step": 4350
    },
    {
      "epoch": 0.88,
      "learning_rate": 1e-05,
      "loss": 0.2408,
      "step": 4375
    },
    {
      "epoch": 0.88,
      "learning_rate": 1e-05,
      "loss": 0.2509,
      "step": 4400
    },
    {
      "epoch": 0.89,
      "learning_rate": 1e-05,
      "loss": 0.2546,
      "step": 4425
    },
    {
      "epoch": 0.89,
      "learning_rate": 1e-05,
      "loss": 0.2427,
      "step": 4450
    },
    {
      "epoch": 0.9,
      "learning_rate": 1e-05,
      "loss": 0.2382,
      "step": 4475
    },
    {
      "epoch": 0.9,
      "learning_rate": 1e-05,
      "loss": 0.2371,
      "step": 4500
    },
    {
      "epoch": 0.91,
      "learning_rate": 1e-05,
      "loss": 0.2422,
      "step": 4525
    },
    {
      "epoch": 0.91,
      "learning_rate": 1e-05,
      "loss": 0.2686,
      "step": 4550
    },
    {
      "epoch": 0.92,
      "learning_rate": 1e-05,
      "loss": 0.2689,
      "step": 4575
    },
    {
      "epoch": 0.92,
      "learning_rate": 1e-05,
      "loss": 0.2693,
      "step": 4600
    },
    {
      "epoch": 0.93,
      "learning_rate": 1e-05,
      "loss": 0.2359,
      "step": 4625
    },
    {
      "epoch": 0.93,
      "learning_rate": 1e-05,
      "loss": 0.2611,
      "step": 4650
    },
    {
      "epoch": 0.94,
      "learning_rate": 1e-05,
      "loss": 0.2249,
      "step": 4675
    },
    {
      "epoch": 0.94,
      "learning_rate": 1e-05,
      "loss": 0.2564,
      "step": 4700
    },
    {
      "epoch": 0.94,
      "learning_rate": 1e-05,
      "loss": 0.2328,
      "step": 4725
    },
    {
      "epoch": 0.95,
      "learning_rate": 1e-05,
      "loss": 0.2576,
      "step": 4750
    },
    {
      "epoch": 0.95,
      "learning_rate": 1e-05,
      "loss": 0.2797,
      "step": 4775
    },
    {
      "epoch": 0.96,
      "learning_rate": 1e-05,
      "loss": 0.2583,
      "step": 4800
    },
    {
      "epoch": 0.96,
      "learning_rate": 1e-05,
      "loss": 0.2488,
      "step": 4825
    },
    {
      "epoch": 0.97,
      "learning_rate": 1e-05,
      "loss": 0.2767,
      "step": 4850
    },
    {
      "epoch": 0.97,
      "learning_rate": 1e-05,
      "loss": 0.2527,
      "step": 4875
    },
    {
      "epoch": 0.98,
      "learning_rate": 1e-05,
      "loss": 0.2304,
      "step": 4900
    },
    {
      "epoch": 0.98,
      "learning_rate": 1e-05,
      "loss": 0.23,
      "step": 4925
    },
    {
      "epoch": 0.99,
      "learning_rate": 1e-05,
      "loss": 0.2285,
      "step": 4950
    },
    {
      "epoch": 0.99,
      "learning_rate": 1e-05,
      "loss": 0.2367,
      "step": 4975
    },
    {
      "epoch": 1.0,
      "learning_rate": 1e-05,
      "loss": 0.2269,
      "step": 5000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.8486328125,
      "eval_runtime": 2156.4186,
      "eval_samples_per_second": 2.622,
      "eval_steps_per_second": 0.164,
      "eval_wer": 20.214938516576623,
      "step": 5000
    },
    {
      "epoch": 1.0,
      "step": 5000,
      "total_flos": 3.397073665165099e+20,
      "train_loss": 0.27437216796875,
      "train_runtime": 51922.2504,
      "train_samples_per_second": 3.082,
      "train_steps_per_second": 0.096
    }
  ],
  "max_steps": 5000,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 3.397073665165099e+20,
  "trial_name": null,
  "trial_params": null
}