{
  "best_metric": 44.119327197715016,
  "best_model_checkpoint": "all_lang_models/sanskrit_models/whisper-medium-sa_alldata_multigpu/checkpoint-750",
  "epoch": 4.62962962962963,
  "global_step": 750,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06,
      "learning_rate": 5e-06,
      "loss": 2.2524,
      "step": 10
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.51838730514401e-06,
      "loss": 0.8538,
      "step": 20
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.95635910224439e-06,
      "loss": 0.5239,
      "step": 30
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.894014962593517e-06,
      "loss": 0.3782,
      "step": 40
    },
    {
      "epoch": 0.31,
      "learning_rate": 9.831670822942643e-06,
      "loss": 0.2879,
      "step": 50
    },
    {
      "epoch": 0.31,
      "eval_loss": 0.313720703125,
      "eval_runtime": 266.2905,
      "eval_samples_per_second": 6.084,
      "eval_steps_per_second": 0.049,
      "eval_wer": 62.215169787369085,
      "step": 50
    },
    {
      "epoch": 0.37,
      "learning_rate": 9.76932668329177e-06,
      "loss": 0.2618,
      "step": 60
    },
    {
      "epoch": 0.43,
      "learning_rate": 9.706982543640898e-06,
      "loss": 0.2358,
      "step": 70
    },
    {
      "epoch": 0.49,
      "learning_rate": 9.644638403990026e-06,
      "loss": 0.23,
      "step": 80
    },
    {
      "epoch": 0.56,
      "learning_rate": 9.582294264339152e-06,
      "loss": 0.2194,
      "step": 90
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.51995012468828e-06,
      "loss": 0.2092,
      "step": 100
    },
    {
      "epoch": 0.62,
      "eval_loss": 0.2529296875,
      "eval_runtime": 377.5429,
      "eval_samples_per_second": 4.291,
      "eval_steps_per_second": 0.034,
      "eval_wer": 52.7642018406855,
      "step": 100
    },
    {
      "epoch": 0.68,
      "learning_rate": 9.457605985037407e-06,
      "loss": 0.2067,
      "step": 110
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.395261845386533e-06,
      "loss": 0.1968,
      "step": 120
    },
    {
      "epoch": 0.8,
      "learning_rate": 9.332917705735661e-06,
      "loss": 0.1987,
      "step": 130
    },
    {
      "epoch": 0.86,
      "learning_rate": 9.270573566084789e-06,
      "loss": 0.1911,
      "step": 140
    },
    {
      "epoch": 0.93,
      "learning_rate": 9.208229426433916e-06,
      "loss": 0.1798,
      "step": 150
    },
    {
      "epoch": 0.93,
      "eval_loss": 0.2269287109375,
      "eval_runtime": 266.0913,
      "eval_samples_per_second": 6.088,
      "eval_steps_per_second": 0.049,
      "eval_wer": 52.92922881624881,
      "step": 150
    },
    {
      "epoch": 0.99,
      "learning_rate": 9.145885286783042e-06,
      "loss": 0.1803,
      "step": 160
    },
    {
      "epoch": 1.05,
      "learning_rate": 9.08354114713217e-06,
      "loss": 0.1587,
      "step": 170
    },
    {
      "epoch": 1.11,
      "learning_rate": 9.021197007481298e-06,
      "loss": 0.1569,
      "step": 180
    },
    {
      "epoch": 1.17,
      "learning_rate": 8.958852867830424e-06,
      "loss": 0.1586,
      "step": 190
    },
    {
      "epoch": 1.23,
      "learning_rate": 8.896508728179551e-06,
      "loss": 0.1591,
      "step": 200
    },
    {
      "epoch": 1.23,
      "eval_loss": 0.213623046875,
      "eval_runtime": 176.3797,
      "eval_samples_per_second": 9.185,
      "eval_steps_per_second": 0.074,
      "eval_wer": 51.36781973976515,
      "step": 200
    },
    {
      "epoch": 1.3,
      "learning_rate": 8.834164588528679e-06,
      "loss": 0.1522,
      "step": 210
    },
    {
      "epoch": 1.36,
      "learning_rate": 8.771820448877807e-06,
      "loss": 0.1551,
      "step": 220
    },
    {
      "epoch": 1.42,
      "learning_rate": 8.709476309226933e-06,
      "loss": 0.1435,
      "step": 230
    },
    {
      "epoch": 1.48,
      "learning_rate": 8.64713216957606e-06,
      "loss": 0.1535,
      "step": 240
    },
    {
      "epoch": 1.54,
      "learning_rate": 8.584788029925188e-06,
      "loss": 0.1506,
      "step": 250
    },
    {
      "epoch": 1.54,
      "eval_loss": 0.2042236328125,
      "eval_runtime": 300.2364,
      "eval_samples_per_second": 5.396,
      "eval_steps_per_second": 0.043,
      "eval_wer": 48.08632180260235,
      "step": 250
    },
    {
      "epoch": 1.6,
      "learning_rate": 8.522443890274314e-06,
      "loss": 0.1525,
      "step": 260
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.460099750623442e-06,
      "loss": 0.1518,
      "step": 270
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.39775561097257e-06,
      "loss": 0.147,
      "step": 280
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.335411471321697e-06,
      "loss": 0.1448,
      "step": 290
    },
    {
      "epoch": 1.85,
      "learning_rate": 8.273067331670823e-06,
      "loss": 0.1441,
      "step": 300
    },
    {
      "epoch": 1.85,
      "eval_loss": 0.192626953125,
      "eval_runtime": 177.3251,
      "eval_samples_per_second": 9.136,
      "eval_steps_per_second": 0.073,
      "eval_wer": 46.9438273563948,
      "step": 300
    },
    {
      "epoch": 1.91,
      "learning_rate": 8.21072319201995e-06,
      "loss": 0.1449,
      "step": 310
    },
    {
      "epoch": 1.98,
      "learning_rate": 8.148379052369078e-06,
      "loss": 0.1411,
      "step": 320
    },
    {
      "epoch": 2.04,
      "learning_rate": 8.086034912718204e-06,
      "loss": 0.1237,
      "step": 330
    },
    {
      "epoch": 2.1,
      "learning_rate": 8.023690773067332e-06,
      "loss": 0.1177,
      "step": 340
    },
    {
      "epoch": 2.16,
      "learning_rate": 7.96134663341646e-06,
      "loss": 0.1179,
      "step": 350
    },
    {
      "epoch": 2.16,
      "eval_loss": 0.19287109375,
      "eval_runtime": 298.9529,
      "eval_samples_per_second": 5.419,
      "eval_steps_per_second": 0.043,
      "eval_wer": 50.38400507775309,
      "step": 350
    },
    {
      "epoch": 2.22,
      "learning_rate": 7.899002493765587e-06,
      "loss": 0.1197,
      "step": 360
    },
    {
      "epoch": 2.28,
      "learning_rate": 7.836658354114713e-06,
      "loss": 0.1145,
      "step": 370
    },
    {
      "epoch": 2.35,
      "learning_rate": 7.77431421446384e-06,
      "loss": 0.1151,
      "step": 380
    },
    {
      "epoch": 2.41,
      "learning_rate": 7.711970074812968e-06,
      "loss": 0.1121,
      "step": 390
    },
    {
      "epoch": 2.47,
      "learning_rate": 7.649625935162094e-06,
      "loss": 0.1174,
      "step": 400
    },
    {
      "epoch": 2.47,
      "eval_loss": 0.1871337890625,
      "eval_runtime": 179.1586,
      "eval_samples_per_second": 9.042,
      "eval_steps_per_second": 0.073,
      "eval_wer": 48.61948587749921,
      "step": 400
    },
    {
      "epoch": 2.53,
      "learning_rate": 7.587281795511222e-06,
      "loss": 0.1161,
      "step": 410
    },
    {
      "epoch": 2.59,
      "learning_rate": 7.52493765586035e-06,
      "loss": 0.1164,
      "step": 420
    },
    {
      "epoch": 2.65,
      "learning_rate": 7.4625935162094764e-06,
      "loss": 0.1115,
      "step": 430
    },
    {
      "epoch": 2.72,
      "learning_rate": 7.400249376558603e-06,
      "loss": 0.1171,
      "step": 440
    },
    {
      "epoch": 2.78,
      "learning_rate": 7.337905236907731e-06,
      "loss": 0.1128,
      "step": 450
    },
    {
      "epoch": 2.78,
      "eval_loss": 0.18505859375,
      "eval_runtime": 173.6435,
      "eval_samples_per_second": 9.329,
      "eval_steps_per_second": 0.075,
      "eval_wer": 47.07711837511901,
      "step": 450
    },
    {
      "epoch": 2.84,
      "learning_rate": 7.275561097256858e-06,
      "loss": 0.1128,
      "step": 460
    },
    {
      "epoch": 2.9,
      "learning_rate": 7.213216957605985e-06,
      "loss": 0.1183,
      "step": 470
    },
    {
      "epoch": 2.96,
      "learning_rate": 7.150872817955112e-06,
      "loss": 0.1137,
      "step": 480
    },
    {
      "epoch": 3.02,
      "learning_rate": 7.08852867830424e-06,
      "loss": 0.1017,
      "step": 490
    },
    {
      "epoch": 3.09,
      "learning_rate": 7.026184538653367e-06,
      "loss": 0.0865,
      "step": 500
    },
    {
      "epoch": 3.09,
      "eval_loss": 0.1888427734375,
      "eval_runtime": 803.7461,
      "eval_samples_per_second": 2.016,
      "eval_steps_per_second": 0.016,
      "eval_wer": 44.804823865439545,
      "step": 500
    },
    {
      "epoch": 3.15,
      "learning_rate": 6.9638403990024935e-06,
      "loss": 0.0876,
      "step": 510
    },
    {
      "epoch": 3.21,
      "learning_rate": 6.901496259351621e-06,
      "loss": 0.0876,
      "step": 520
    },
    {
      "epoch": 3.27,
      "learning_rate": 6.839152119700748e-06,
      "loss": 0.0845,
      "step": 530
    },
    {
      "epoch": 3.33,
      "learning_rate": 6.776807980049876e-06,
      "loss": 0.0912,
      "step": 540
    },
    {
      "epoch": 3.4,
      "learning_rate": 6.7144638403990024e-06,
      "loss": 0.0934,
      "step": 550
    },
    {
      "epoch": 3.4,
      "eval_loss": 0.193603515625,
      "eval_runtime": 374.8954,
      "eval_samples_per_second": 4.321,
      "eval_steps_per_second": 0.035,
      "eval_wer": 50.733100602983185,
      "step": 550
    },
    {
      "epoch": 3.46,
      "learning_rate": 6.65211970074813e-06,
      "loss": 0.0909,
      "step": 560
    },
    {
      "epoch": 3.52,
      "learning_rate": 6.589775561097257e-06,
      "loss": 0.0956,
      "step": 570
    },
    {
      "epoch": 3.58,
      "learning_rate": 6.527431421446384e-06,
      "loss": 0.0942,
      "step": 580
    },
    {
      "epoch": 3.64,
      "learning_rate": 6.465087281795511e-06,
      "loss": 0.0867,
      "step": 590
    },
    {
      "epoch": 3.7,
      "learning_rate": 6.402743142144638e-06,
      "loss": 0.09,
      "step": 600
    },
    {
      "epoch": 3.7,
      "eval_loss": 0.191650390625,
      "eval_runtime": 362.3419,
      "eval_samples_per_second": 4.471,
      "eval_steps_per_second": 0.036,
      "eval_wer": 50.01586797841955,
      "step": 600
    },
    {
      "epoch": 3.77,
      "learning_rate": 6.340399002493766e-06,
      "loss": 0.0812,
      "step": 610
    },
    {
      "epoch": 3.83,
      "learning_rate": 6.278054862842893e-06,
      "loss": 0.0882,
      "step": 620
    },
    {
      "epoch": 3.89,
      "learning_rate": 6.21571072319202e-06,
      "loss": 0.0898,
      "step": 630
    },
    {
      "epoch": 3.95,
      "learning_rate": 6.153366583541147e-06,
      "loss": 0.0932,
      "step": 640
    },
    {
      "epoch": 4.01,
      "learning_rate": 6.091022443890274e-06,
      "loss": 0.0826,
      "step": 650
    },
    {
      "epoch": 4.01,
      "eval_loss": 0.1953125,
      "eval_runtime": 282.5656,
      "eval_samples_per_second": 5.733,
      "eval_steps_per_second": 0.046,
      "eval_wer": 48.44811171056808,
      "step": 650
    },
    {
      "epoch": 4.07,
      "learning_rate": 6.028678304239402e-06,
      "loss": 0.0659,
      "step": 660
    },
    {
      "epoch": 4.14,
      "learning_rate": 5.9663341645885284e-06,
      "loss": 0.0664,
      "step": 670
    },
    {
      "epoch": 4.2,
      "learning_rate": 5.903990024937656e-06,
      "loss": 0.0696,
      "step": 680
    },
    {
      "epoch": 4.26,
      "learning_rate": 5.841645885286783e-06,
      "loss": 0.068,
      "step": 690
    },
    {
      "epoch": 4.32,
      "learning_rate": 5.779301745635911e-06,
      "loss": 0.0691,
      "step": 700
    },
    {
      "epoch": 4.32,
      "eval_loss": 0.206298828125,
      "eval_runtime": 347.7771,
      "eval_samples_per_second": 4.658,
      "eval_steps_per_second": 0.037,
      "eval_wer": 46.702634084417646,
      "step": 700
    },
    {
      "epoch": 4.38,
      "learning_rate": 5.716957605985037e-06,
      "loss": 0.0673,
      "step": 710
    },
    {
      "epoch": 4.44,
      "learning_rate": 5.654613466334164e-06,
      "loss": 0.0688,
      "step": 720
    },
    {
      "epoch": 4.51,
      "learning_rate": 5.592269326683292e-06,
      "loss": 0.0665,
      "step": 730
    },
    {
      "epoch": 4.57,
      "learning_rate": 5.529925187032419e-06,
      "loss": 0.065,
      "step": 740
    },
    {
      "epoch": 4.63,
      "learning_rate": 5.467581047381546e-06,
      "loss": 0.0662,
      "step": 750
    },
    {
      "epoch": 4.63,
      "eval_loss": 0.203369140625,
      "eval_runtime": 168.1018,
      "eval_samples_per_second": 9.637,
      "eval_steps_per_second": 0.077,
      "eval_wer": 44.119327197715016,
      "step": 750
    }
  ],
  "max_steps": 1620,
  "num_train_epochs": 10,
  "total_flos": 1.9577651816212044e+20,
  "trial_name": null,
  "trial_params": null
}