{
  "best_metric": 80.09197164207703,
  "best_model_checkpoint": "./whisper-small-ha-adam-v5/checkpoint-2000",
  "epoch": 15.923566878980893,
  "eval_steps": 500,
  "global_step": 2500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1592356687898089,
      "grad_norm": 20.107877731323242,
      "learning_rate": 2.5e-05,
      "loss": 3.0102,
      "step": 25
    },
    {
      "epoch": 0.3184713375796178,
      "grad_norm": 21.660022735595703,
      "learning_rate": 5e-05,
      "loss": 2.5653,
      "step": 50
    },
    {
      "epoch": 0.47770700636942676,
      "grad_norm": 14.246131896972656,
      "learning_rate": 5e-05,
      "loss": 2.0386,
      "step": 75
    },
    {
      "epoch": 0.6369426751592356,
      "grad_norm": 13.428327560424805,
      "learning_rate": 5e-05,
      "loss": 1.883,
      "step": 100
    },
    {
      "epoch": 0.7961783439490446,
      "grad_norm": 12.753376007080078,
      "learning_rate": 5e-05,
      "loss": 1.7273,
      "step": 125
    },
    {
      "epoch": 0.9554140127388535,
      "grad_norm": 13.684127807617188,
      "learning_rate": 5e-05,
      "loss": 1.6506,
      "step": 150
    },
    {
      "epoch": 1.1146496815286624,
      "grad_norm": 10.537659645080566,
      "learning_rate": 5e-05,
      "loss": 1.0013,
      "step": 175
    },
    {
      "epoch": 1.2738853503184713,
      "grad_norm": 9.208993911743164,
      "learning_rate": 5e-05,
      "loss": 0.7727,
      "step": 200
    },
    {
      "epoch": 1.4331210191082802,
      "grad_norm": 10.912196159362793,
      "learning_rate": 5e-05,
      "loss": 0.7837,
      "step": 225
    },
    {
      "epoch": 1.5923566878980893,
      "grad_norm": 9.990174293518066,
      "learning_rate": 5e-05,
      "loss": 0.7913,
      "step": 250
    },
    {
      "epoch": 1.7515923566878981,
      "grad_norm": 11.005951881408691,
      "learning_rate": 5e-05,
      "loss": 0.7732,
      "step": 275
    },
    {
      "epoch": 1.910828025477707,
      "grad_norm": 9.276304244995117,
      "learning_rate": 5e-05,
      "loss": 0.8036,
      "step": 300
    },
    {
      "epoch": 2.070063694267516,
      "grad_norm": 6.240780353546143,
      "learning_rate": 5e-05,
      "loss": 0.5545,
      "step": 325
    },
    {
      "epoch": 2.229299363057325,
      "grad_norm": 15.549118995666504,
      "learning_rate": 5e-05,
      "loss": 0.2379,
      "step": 350
    },
    {
      "epoch": 2.388535031847134,
      "grad_norm": 6.316082954406738,
      "learning_rate": 5e-05,
      "loss": 0.2277,
      "step": 375
    },
    {
      "epoch": 2.5477707006369426,
      "grad_norm": 8.334115028381348,
      "learning_rate": 5e-05,
      "loss": 0.248,
      "step": 400
    },
    {
      "epoch": 2.7070063694267517,
      "grad_norm": 8.026554107666016,
      "learning_rate": 5e-05,
      "loss": 0.2683,
      "step": 425
    },
    {
      "epoch": 2.8662420382165603,
      "grad_norm": 7.539698600769043,
      "learning_rate": 5e-05,
      "loss": 0.2704,
      "step": 450
    },
    {
      "epoch": 3.0254777070063694,
      "grad_norm": 3.8269407749176025,
      "learning_rate": 5e-05,
      "loss": 0.2389,
      "step": 475
    },
    {
      "epoch": 3.1847133757961785,
      "grad_norm": 5.021791458129883,
      "learning_rate": 5e-05,
      "loss": 0.0983,
      "step": 500
    },
    {
      "epoch": 3.1847133757961785,
      "eval_loss": 1.8208503723144531,
      "eval_runtime": 234.2259,
      "eval_samples_per_second": 2.818,
      "eval_steps_per_second": 0.179,
      "eval_wer": 83.21517532094272,
      "eval_wer_ortho": 85.52734375,
      "step": 500
    },
    {
      "epoch": 3.343949044585987,
      "grad_norm": 4.637480735778809,
      "learning_rate": 5e-05,
      "loss": 0.1165,
      "step": 525
    },
    {
      "epoch": 3.5031847133757963,
      "grad_norm": 4.222353458404541,
      "learning_rate": 5e-05,
      "loss": 0.1142,
      "step": 550
    },
    {
      "epoch": 3.662420382165605,
      "grad_norm": 7.10737943649292,
      "learning_rate": 5e-05,
      "loss": 0.1279,
      "step": 575
    },
    {
      "epoch": 3.821656050955414,
      "grad_norm": 5.412852764129639,
      "learning_rate": 5e-05,
      "loss": 0.1173,
      "step": 600
    },
    {
      "epoch": 3.980891719745223,
      "grad_norm": 6.706114768981934,
      "learning_rate": 5e-05,
      "loss": 0.1257,
      "step": 625
    },
    {
      "epoch": 4.140127388535032,
      "grad_norm": 2.5801165103912354,
      "learning_rate": 5e-05,
      "loss": 0.0762,
      "step": 650
    },
    {
      "epoch": 4.2993630573248405,
      "grad_norm": 4.530727863311768,
      "learning_rate": 5e-05,
      "loss": 0.0808,
      "step": 675
    },
    {
      "epoch": 4.45859872611465,
      "grad_norm": 3.3920371532440186,
      "learning_rate": 5e-05,
      "loss": 0.0668,
      "step": 700
    },
    {
      "epoch": 4.617834394904459,
      "grad_norm": 5.061642169952393,
      "learning_rate": 5e-05,
      "loss": 0.0865,
      "step": 725
    },
    {
      "epoch": 4.777070063694268,
      "grad_norm": 4.725163459777832,
      "learning_rate": 5e-05,
      "loss": 0.0859,
      "step": 750
    },
    {
      "epoch": 4.936305732484076,
      "grad_norm": 6.57818603515625,
      "learning_rate": 5e-05,
      "loss": 0.0999,
      "step": 775
    },
    {
      "epoch": 5.095541401273885,
      "grad_norm": 2.8120875358581543,
      "learning_rate": 5e-05,
      "loss": 0.0699,
      "step": 800
    },
    {
      "epoch": 5.254777070063694,
      "grad_norm": 4.141892433166504,
      "learning_rate": 5e-05,
      "loss": 0.0558,
      "step": 825
    },
    {
      "epoch": 5.414012738853503,
      "grad_norm": 3.9606029987335205,
      "learning_rate": 5e-05,
      "loss": 0.0733,
      "step": 850
    },
    {
      "epoch": 5.573248407643312,
      "grad_norm": 5.46377420425415,
      "learning_rate": 5e-05,
      "loss": 0.0645,
      "step": 875
    },
    {
      "epoch": 5.732484076433121,
      "grad_norm": 4.627284526824951,
      "learning_rate": 5e-05,
      "loss": 0.0665,
      "step": 900
    },
    {
      "epoch": 5.89171974522293,
      "grad_norm": 5.00495719909668,
      "learning_rate": 5e-05,
      "loss": 0.0675,
      "step": 925
    },
    {
      "epoch": 6.050955414012739,
      "grad_norm": 2.5908775329589844,
      "learning_rate": 5e-05,
      "loss": 0.0596,
      "step": 950
    },
    {
      "epoch": 6.210191082802548,
      "grad_norm": 5.327170372009277,
      "learning_rate": 5e-05,
      "loss": 0.0487,
      "step": 975
    },
    {
      "epoch": 6.369426751592357,
      "grad_norm": 2.1913540363311768,
      "learning_rate": 5e-05,
      "loss": 0.0514,
      "step": 1000
    },
    {
      "epoch": 6.369426751592357,
      "eval_loss": 1.975398063659668,
      "eval_runtime": 246.0885,
      "eval_samples_per_second": 2.682,
      "eval_steps_per_second": 0.171,
      "eval_wer": 88.50354474037172,
      "eval_wer_ortho": 89.62890625,
      "step": 1000
    },
    {
      "epoch": 6.528662420382165,
      "grad_norm": 4.72745943069458,
      "learning_rate": 5e-05,
      "loss": 0.0627,
      "step": 1025
    },
    {
      "epoch": 6.687898089171974,
      "grad_norm": 4.439643383026123,
      "learning_rate": 5e-05,
      "loss": 0.0579,
      "step": 1050
    },
    {
      "epoch": 6.8471337579617835,
      "grad_norm": 4.479223251342773,
      "learning_rate": 5e-05,
      "loss": 0.0551,
      "step": 1075
    },
    {
      "epoch": 7.006369426751593,
      "grad_norm": 5.471689224243164,
      "learning_rate": 5e-05,
      "loss": 0.0637,
      "step": 1100
    },
    {
      "epoch": 7.165605095541402,
      "grad_norm": 1.6073918342590332,
      "learning_rate": 5e-05,
      "loss": 0.0521,
      "step": 1125
    },
    {
      "epoch": 7.32484076433121,
      "grad_norm": 2.2465476989746094,
      "learning_rate": 5e-05,
      "loss": 0.0414,
      "step": 1150
    },
    {
      "epoch": 7.484076433121019,
      "grad_norm": 2.386169910430908,
      "learning_rate": 5e-05,
      "loss": 0.0617,
      "step": 1175
    },
    {
      "epoch": 7.643312101910828,
      "grad_norm": 2.7703945636749268,
      "learning_rate": 5e-05,
      "loss": 0.0511,
      "step": 1200
    },
    {
      "epoch": 7.802547770700637,
      "grad_norm": 3.860624074935913,
      "learning_rate": 5e-05,
      "loss": 0.0543,
      "step": 1225
    },
    {
      "epoch": 7.961783439490446,
      "grad_norm": 7.631902694702148,
      "learning_rate": 5e-05,
      "loss": 0.0536,
      "step": 1250
    },
    {
      "epoch": 8.121019108280255,
      "grad_norm": 1.4489890336990356,
      "learning_rate": 5e-05,
      "loss": 0.031,
      "step": 1275
    },
    {
      "epoch": 8.280254777070065,
      "grad_norm": 1.750957727432251,
      "learning_rate": 5e-05,
      "loss": 0.0309,
      "step": 1300
    },
    {
      "epoch": 8.439490445859873,
      "grad_norm": 2.7531442642211914,
      "learning_rate": 5e-05,
      "loss": 0.0414,
      "step": 1325
    },
    {
      "epoch": 8.598726114649681,
      "grad_norm": 2.628211736679077,
      "learning_rate": 5e-05,
      "loss": 0.0349,
      "step": 1350
    },
    {
      "epoch": 8.757961783439491,
      "grad_norm": 3.8813140392303467,
      "learning_rate": 5e-05,
      "loss": 0.0465,
      "step": 1375
    },
    {
      "epoch": 8.9171974522293,
      "grad_norm": 1.4133429527282715,
      "learning_rate": 5e-05,
      "loss": 0.0395,
      "step": 1400
    },
    {
      "epoch": 9.07643312101911,
      "grad_norm": 2.883612871170044,
      "learning_rate": 5e-05,
      "loss": 0.0421,
      "step": 1425
    },
    {
      "epoch": 9.235668789808917,
      "grad_norm": 4.271716594696045,
      "learning_rate": 5e-05,
      "loss": 0.0289,
      "step": 1450
    },
    {
      "epoch": 9.394904458598726,
      "grad_norm": 2.3349599838256836,
      "learning_rate": 5e-05,
      "loss": 0.0311,
      "step": 1475
    },
    {
      "epoch": 9.554140127388536,
      "grad_norm": 1.0621289014816284,
      "learning_rate": 5e-05,
      "loss": 0.0393,
      "step": 1500
    },
    {
      "epoch": 9.554140127388536,
      "eval_loss": 2.148721694946289,
      "eval_runtime": 240.3237,
      "eval_samples_per_second": 2.746,
      "eval_steps_per_second": 0.175,
      "eval_wer": 88.82927763939452,
      "eval_wer_ortho": 89.5703125,
      "step": 1500
    },
    {
      "epoch": 9.713375796178344,
      "grad_norm": 1.477952003479004,
      "learning_rate": 5e-05,
      "loss": 0.0428,
      "step": 1525
    },
    {
      "epoch": 9.872611464968152,
      "grad_norm": 2.882534980773926,
      "learning_rate": 5e-05,
      "loss": 0.0382,
      "step": 1550
    },
    {
      "epoch": 10.031847133757962,
      "grad_norm": 2.2467527389526367,
      "learning_rate": 5e-05,
      "loss": 0.0386,
      "step": 1575
    },
    {
      "epoch": 10.19108280254777,
      "grad_norm": 2.468972682952881,
      "learning_rate": 5e-05,
      "loss": 0.0332,
      "step": 1600
    },
    {
      "epoch": 10.35031847133758,
      "grad_norm": 2.6261913776397705,
      "learning_rate": 5e-05,
      "loss": 0.0291,
      "step": 1625
    },
    {
      "epoch": 10.509554140127388,
      "grad_norm": 2.022120714187622,
      "learning_rate": 5e-05,
      "loss": 0.0335,
      "step": 1650
    },
    {
      "epoch": 10.668789808917197,
      "grad_norm": 1.5274375677108765,
      "learning_rate": 5e-05,
      "loss": 0.0327,
      "step": 1675
    },
    {
      "epoch": 10.828025477707007,
      "grad_norm": 4.571503639221191,
      "learning_rate": 5e-05,
      "loss": 0.0369,
      "step": 1700
    },
    {
      "epoch": 10.987261146496815,
      "grad_norm": 6.347606182098389,
      "learning_rate": 5e-05,
      "loss": 0.0393,
      "step": 1725
    },
    {
      "epoch": 11.146496815286625,
      "grad_norm": 2.409583568572998,
      "learning_rate": 5e-05,
      "loss": 0.0231,
      "step": 1750
    },
    {
      "epoch": 11.305732484076433,
      "grad_norm": 2.1078784465789795,
      "learning_rate": 5e-05,
      "loss": 0.032,
      "step": 1775
    },
    {
      "epoch": 11.464968152866241,
      "grad_norm": 0.528281569480896,
      "learning_rate": 5e-05,
      "loss": 0.034,
      "step": 1800
    },
    {
      "epoch": 11.624203821656051,
      "grad_norm": 2.900444984436035,
      "learning_rate": 5e-05,
      "loss": 0.0433,
      "step": 1825
    },
    {
      "epoch": 11.78343949044586,
      "grad_norm": 2.4435691833496094,
      "learning_rate": 5e-05,
      "loss": 0.0232,
      "step": 1850
    },
    {
      "epoch": 11.94267515923567,
      "grad_norm": 3.484557628631592,
      "learning_rate": 5e-05,
      "loss": 0.0328,
      "step": 1875
    },
    {
      "epoch": 12.101910828025478,
      "grad_norm": 5.183062553405762,
      "learning_rate": 5e-05,
      "loss": 0.0261,
      "step": 1900
    },
    {
      "epoch": 12.261146496815286,
      "grad_norm": 6.2796630859375,
      "learning_rate": 5e-05,
      "loss": 0.0363,
      "step": 1925
    },
    {
      "epoch": 12.420382165605096,
      "grad_norm": 2.45373797416687,
      "learning_rate": 5e-05,
      "loss": 0.0323,
      "step": 1950
    },
    {
      "epoch": 12.579617834394904,
      "grad_norm": 5.370733261108398,
      "learning_rate": 5e-05,
      "loss": 0.0317,
      "step": 1975
    },
    {
      "epoch": 12.738853503184714,
      "grad_norm": 2.645756959915161,
      "learning_rate": 5e-05,
      "loss": 0.0313,
      "step": 2000
    },
    {
      "epoch": 12.738853503184714,
      "eval_loss": 2.1685144901275635,
      "eval_runtime": 237.2073,
      "eval_samples_per_second": 2.782,
      "eval_steps_per_second": 0.177,
      "eval_wer": 80.09197164207703,
      "eval_wer_ortho": 81.9921875,
      "step": 2000
    },
    {
      "epoch": 12.898089171974522,
      "grad_norm": 2.6092240810394287,
      "learning_rate": 5e-05,
      "loss": 0.0373,
      "step": 2025
    },
    {
      "epoch": 13.05732484076433,
      "grad_norm": 3.123652935028076,
      "learning_rate": 5e-05,
      "loss": 0.0293,
      "step": 2050
    },
    {
      "epoch": 13.21656050955414,
      "grad_norm": 2.5134246349334717,
      "learning_rate": 5e-05,
      "loss": 0.0268,
      "step": 2075
    },
    {
      "epoch": 13.375796178343949,
      "grad_norm": 1.4545310735702515,
      "learning_rate": 5e-05,
      "loss": 0.0294,
      "step": 2100
    },
    {
      "epoch": 13.535031847133759,
      "grad_norm": 2.473706007003784,
      "learning_rate": 5e-05,
      "loss": 0.0307,
      "step": 2125
    },
    {
      "epoch": 13.694267515923567,
      "grad_norm": 2.8176300525665283,
      "learning_rate": 5e-05,
      "loss": 0.0279,
      "step": 2150
    },
    {
      "epoch": 13.853503184713375,
      "grad_norm": 38.75226974487305,
      "learning_rate": 5e-05,
      "loss": 0.0454,
      "step": 2175
    },
    {
      "epoch": 14.012738853503185,
      "grad_norm": 0.736247181892395,
      "learning_rate": 5e-05,
      "loss": 0.0247,
      "step": 2200
    },
    {
      "epoch": 14.171974522292993,
      "grad_norm": 2.7903378009796143,
      "learning_rate": 5e-05,
      "loss": 0.0289,
      "step": 2225
    },
    {
      "epoch": 14.331210191082803,
      "grad_norm": 2.184035301208496,
      "learning_rate": 5e-05,
      "loss": 0.0216,
      "step": 2250
    },
    {
      "epoch": 14.490445859872612,
      "grad_norm": 2.464597702026367,
      "learning_rate": 5e-05,
      "loss": 0.0275,
      "step": 2275
    },
    {
      "epoch": 14.64968152866242,
      "grad_norm": 4.4987335205078125,
      "learning_rate": 5e-05,
      "loss": 0.0374,
      "step": 2300
    },
    {
      "epoch": 14.80891719745223,
      "grad_norm": 2.5459258556365967,
      "learning_rate": 5e-05,
      "loss": 0.031,
      "step": 2325
    },
    {
      "epoch": 14.968152866242038,
      "grad_norm": 2.8609278202056885,
      "learning_rate": 5e-05,
      "loss": 0.0338,
      "step": 2350
    },
    {
      "epoch": 15.127388535031848,
      "grad_norm": 0.5692533850669861,
      "learning_rate": 5e-05,
      "loss": 0.0154,
      "step": 2375
    },
    {
      "epoch": 15.286624203821656,
      "grad_norm": 1.785417914390564,
      "learning_rate": 5e-05,
      "loss": 0.0285,
      "step": 2400
    },
    {
      "epoch": 15.445859872611464,
      "grad_norm": 3.1533737182617188,
      "learning_rate": 5e-05,
      "loss": 0.027,
      "step": 2425
    },
    {
      "epoch": 15.605095541401274,
      "grad_norm": 0.5182532072067261,
      "learning_rate": 5e-05,
      "loss": 0.0293,
      "step": 2450
    },
    {
      "epoch": 15.764331210191083,
      "grad_norm": 2.0394535064697266,
      "learning_rate": 5e-05,
      "loss": 0.0316,
      "step": 2475
    },
    {
      "epoch": 15.923566878980893,
      "grad_norm": 0.6453192234039307,
      "learning_rate": 5e-05,
      "loss": 0.0213,
      "step": 2500
    },
    {
      "epoch": 15.923566878980893,
      "eval_loss": 2.2915444374084473,
      "eval_runtime": 239.3393,
      "eval_samples_per_second": 2.758,
      "eval_steps_per_second": 0.175,
      "eval_wer": 80.80091971642076,
      "eval_wer_ortho": 82.578125,
      "step": 2500
    }
  ],
  "logging_steps": 25,
  "max_steps": 2826,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 18,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.15261008887808e+19,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}