{
  "best_metric": 5.288159547738693,
  "best_model_checkpoint": "/kaggle/working/checkpoint-4000",
  "epoch": 2.873563218390805,
  "eval_steps": 1000,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.017959770114942528,
      "grad_norm": 24.085046768188477,
      "learning_rate": 4.4e-07,
      "loss": 1.4603,
      "step": 25
    },
    {
      "epoch": 0.035919540229885055,
      "grad_norm": 13.078679084777832,
      "learning_rate": 9.400000000000001e-07,
      "loss": 1.3305,
      "step": 50
    },
    {
      "epoch": 0.05387931034482758,
      "grad_norm": 17.343385696411133,
      "learning_rate": 1.44e-06,
      "loss": 1.0971,
      "step": 75
    },
    {
      "epoch": 0.07183908045977011,
      "grad_norm": 10.30687427520752,
      "learning_rate": 1.94e-06,
      "loss": 0.6573,
      "step": 100
    },
    {
      "epoch": 0.08979885057471264,
      "grad_norm": 7.07254695892334,
      "learning_rate": 2.4400000000000004e-06,
      "loss": 0.5107,
      "step": 125
    },
    {
      "epoch": 0.10775862068965517,
      "grad_norm": 7.000039100646973,
      "learning_rate": 2.9400000000000002e-06,
      "loss": 0.4643,
      "step": 150
    },
    {
      "epoch": 0.1257183908045977,
      "grad_norm": 7.2714972496032715,
      "learning_rate": 3.44e-06,
      "loss": 0.4358,
      "step": 175
    },
    {
      "epoch": 0.14367816091954022,
      "grad_norm": 6.5415730476379395,
      "learning_rate": 3.94e-06,
      "loss": 0.4219,
      "step": 200
    },
    {
      "epoch": 0.16163793103448276,
      "grad_norm": 7.611192226409912,
      "learning_rate": 4.440000000000001e-06,
      "loss": 0.4036,
      "step": 225
    },
    {
      "epoch": 0.17959770114942528,
      "grad_norm": 6.730944633483887,
      "learning_rate": 4.94e-06,
      "loss": 0.4024,
      "step": 250
    },
    {
      "epoch": 0.19755747126436782,
      "grad_norm": 6.479249000549316,
      "learning_rate": 5.4400000000000004e-06,
      "loss": 0.3915,
      "step": 275
    },
    {
      "epoch": 0.21551724137931033,
      "grad_norm": 6.75312614440918,
      "learning_rate": 5.94e-06,
      "loss": 0.3638,
      "step": 300
    },
    {
      "epoch": 0.23347701149425287,
      "grad_norm": 5.702291488647461,
      "learning_rate": 6.440000000000001e-06,
      "loss": 0.3708,
      "step": 325
    },
    {
      "epoch": 0.2514367816091954,
      "grad_norm": 5.4078688621521,
      "learning_rate": 6.9400000000000005e-06,
      "loss": 0.3556,
      "step": 350
    },
    {
      "epoch": 0.26939655172413796,
      "grad_norm": 5.735648155212402,
      "learning_rate": 7.440000000000001e-06,
      "loss": 0.3444,
      "step": 375
    },
    {
      "epoch": 0.28735632183908044,
      "grad_norm": 6.246584892272949,
      "learning_rate": 7.94e-06,
      "loss": 0.3443,
      "step": 400
    },
    {
      "epoch": 0.305316091954023,
      "grad_norm": 6.741141319274902,
      "learning_rate": 8.44e-06,
      "loss": 0.3192,
      "step": 425
    },
    {
      "epoch": 0.3232758620689655,
      "grad_norm": 6.674511432647705,
      "learning_rate": 8.94e-06,
      "loss": 0.3249,
      "step": 450
    },
    {
      "epoch": 0.34123563218390807,
      "grad_norm": 6.338778018951416,
      "learning_rate": 9.440000000000001e-06,
      "loss": 0.3331,
      "step": 475
    },
    {
      "epoch": 0.35919540229885055,
      "grad_norm": 6.05772590637207,
      "learning_rate": 9.940000000000001e-06,
      "loss": 0.3067,
      "step": 500
    },
    {
      "epoch": 0.3771551724137931,
      "grad_norm": 6.442326545715332,
      "learning_rate": 9.937142857142858e-06,
      "loss": 0.3275,
      "step": 525
    },
    {
      "epoch": 0.39511494252873564,
      "grad_norm": 5.767732620239258,
      "learning_rate": 9.865714285714285e-06,
      "loss": 0.3218,
      "step": 550
    },
    {
      "epoch": 0.4130747126436782,
      "grad_norm": 5.053823471069336,
      "learning_rate": 9.794285714285714e-06,
      "loss": 0.3078,
      "step": 575
    },
    {
      "epoch": 0.43103448275862066,
      "grad_norm": 5.843025207519531,
      "learning_rate": 9.722857142857143e-06,
      "loss": 0.2923,
      "step": 600
    },
    {
      "epoch": 0.4489942528735632,
      "grad_norm": 6.414527893066406,
      "learning_rate": 9.651428571428572e-06,
      "loss": 0.2823,
      "step": 625
    },
    {
      "epoch": 0.46695402298850575,
      "grad_norm": 5.519237041473389,
      "learning_rate": 9.58e-06,
      "loss": 0.2789,
      "step": 650
    },
    {
      "epoch": 0.4849137931034483,
      "grad_norm": 4.679568290710449,
      "learning_rate": 9.508571428571429e-06,
      "loss": 0.2856,
      "step": 675
    },
    {
      "epoch": 0.5028735632183908,
      "grad_norm": 5.241317272186279,
      "learning_rate": 9.437142857142858e-06,
      "loss": 0.2765,
      "step": 700
    },
    {
      "epoch": 0.5208333333333334,
      "grad_norm": 5.156564235687256,
      "learning_rate": 9.365714285714287e-06,
      "loss": 0.2715,
      "step": 725
    },
    {
      "epoch": 0.5387931034482759,
      "grad_norm": 5.449781894683838,
      "learning_rate": 9.294285714285714e-06,
      "loss": 0.2713,
      "step": 750
    },
    {
      "epoch": 0.5567528735632183,
      "grad_norm": 5.37013053894043,
      "learning_rate": 9.222857142857143e-06,
      "loss": 0.2662,
      "step": 775
    },
    {
      "epoch": 0.5747126436781609,
      "grad_norm": 5.62649393081665,
      "learning_rate": 9.151428571428572e-06,
      "loss": 0.2814,
      "step": 800
    },
    {
      "epoch": 0.5926724137931034,
      "grad_norm": 5.46066427230835,
      "learning_rate": 9.080000000000001e-06,
      "loss": 0.2606,
      "step": 825
    },
    {
      "epoch": 0.610632183908046,
      "grad_norm": 5.137524604797363,
      "learning_rate": 9.00857142857143e-06,
      "loss": 0.2329,
      "step": 850
    },
    {
      "epoch": 0.6285919540229885,
      "grad_norm": 5.287104606628418,
      "learning_rate": 8.937142857142857e-06,
      "loss": 0.2595,
      "step": 875
    },
    {
      "epoch": 0.646551724137931,
      "grad_norm": 6.085468292236328,
      "learning_rate": 8.865714285714287e-06,
      "loss": 0.2558,
      "step": 900
    },
    {
      "epoch": 0.6645114942528736,
      "grad_norm": 6.9088826179504395,
      "learning_rate": 8.794285714285716e-06,
      "loss": 0.2408,
      "step": 925
    },
    {
      "epoch": 0.6824712643678161,
      "grad_norm": 5.21582555770874,
      "learning_rate": 8.722857142857145e-06,
      "loss": 0.2255,
      "step": 950
    },
    {
      "epoch": 0.7004310344827587,
      "grad_norm": 4.872750759124756,
      "learning_rate": 8.651428571428572e-06,
      "loss": 0.2292,
      "step": 975
    },
    {
      "epoch": 0.7183908045977011,
      "grad_norm": 5.2055583000183105,
      "learning_rate": 8.580000000000001e-06,
      "loss": 0.2542,
      "step": 1000
    },
    {
      "epoch": 0.7183908045977011,
      "eval_cer": 9.284704773869347,
      "eval_loss": 0.27573877573013306,
      "eval_runtime": 208.7308,
      "eval_samples_per_second": 2.189,
      "eval_steps_per_second": 0.278,
      "step": 1000
    },
    {
      "epoch": 0.7363505747126436,
      "grad_norm": 5.838357925415039,
      "learning_rate": 8.50857142857143e-06,
      "loss": 0.2147,
      "step": 1025
    },
    {
      "epoch": 0.7543103448275862,
      "grad_norm": 5.988395690917969,
      "learning_rate": 8.437142857142859e-06,
      "loss": 0.2399,
      "step": 1050
    },
    {
      "epoch": 0.7722701149425287,
      "grad_norm": 4.545289993286133,
      "learning_rate": 8.365714285714286e-06,
      "loss": 0.2243,
      "step": 1075
    },
    {
      "epoch": 0.7902298850574713,
      "grad_norm": 4.793825626373291,
      "learning_rate": 8.294285714285715e-06,
      "loss": 0.2275,
      "step": 1100
    },
    {
      "epoch": 0.8081896551724138,
      "grad_norm": 5.085103988647461,
      "learning_rate": 8.222857142857144e-06,
      "loss": 0.2228,
      "step": 1125
    },
    {
      "epoch": 0.8261494252873564,
      "grad_norm": 4.345165729522705,
      "learning_rate": 8.151428571428572e-06,
      "loss": 0.2246,
      "step": 1150
    },
    {
      "epoch": 0.8441091954022989,
      "grad_norm": 5.478984355926514,
      "learning_rate": 8.08e-06,
      "loss": 0.2041,
      "step": 1175
    },
    {
      "epoch": 0.8620689655172413,
      "grad_norm": 5.148238182067871,
      "learning_rate": 8.00857142857143e-06,
      "loss": 0.2123,
      "step": 1200
    },
    {
      "epoch": 0.8800287356321839,
      "grad_norm": 5.2301106452941895,
      "learning_rate": 7.937142857142857e-06,
      "loss": 0.2135,
      "step": 1225
    },
    {
      "epoch": 0.8979885057471264,
      "grad_norm": 5.3838019371032715,
      "learning_rate": 7.865714285714286e-06,
      "loss": 0.1978,
      "step": 1250
    },
    {
      "epoch": 0.915948275862069,
      "grad_norm": 5.178905487060547,
      "learning_rate": 7.794285714285715e-06,
      "loss": 0.2127,
      "step": 1275
    },
    {
      "epoch": 0.9339080459770115,
      "grad_norm": 5.5203351974487305,
      "learning_rate": 7.722857142857142e-06,
      "loss": 0.2128,
      "step": 1300
    },
    {
      "epoch": 0.951867816091954,
      "grad_norm": 5.354878902435303,
      "learning_rate": 7.651428571428571e-06,
      "loss": 0.1924,
      "step": 1325
    },
    {
      "epoch": 0.9698275862068966,
      "grad_norm": 5.155806064605713,
      "learning_rate": 7.58e-06,
      "loss": 0.209,
      "step": 1350
    },
    {
      "epoch": 0.9877873563218391,
      "grad_norm": 4.325410842895508,
      "learning_rate": 7.508571428571429e-06,
      "loss": 0.2019,
      "step": 1375
    },
    {
      "epoch": 1.0057471264367817,
      "grad_norm": 3.931236743927002,
      "learning_rate": 7.4371428571428575e-06,
      "loss": 0.2028,
      "step": 1400
    },
    {
      "epoch": 1.0237068965517242,
      "grad_norm": 4.375692844390869,
      "learning_rate": 7.365714285714286e-06,
      "loss": 0.1572,
      "step": 1425
    },
    {
      "epoch": 1.0416666666666667,
      "grad_norm": 3.4934616088867188,
      "learning_rate": 7.294285714285715e-06,
      "loss": 0.1423,
      "step": 1450
    },
    {
      "epoch": 1.0596264367816093,
      "grad_norm": 4.1985249519348145,
      "learning_rate": 7.222857142857144e-06,
      "loss": 0.1477,
      "step": 1475
    },
    {
      "epoch": 1.0775862068965518,
      "grad_norm": 4.258643627166748,
      "learning_rate": 7.151428571428573e-06,
      "loss": 0.145,
      "step": 1500
    },
    {
      "epoch": 1.0955459770114944,
      "grad_norm": 3.0669021606445312,
      "learning_rate": 7.08e-06,
      "loss": 0.1404,
      "step": 1525
    },
    {
      "epoch": 1.1135057471264367,
      "grad_norm": 4.19215726852417,
      "learning_rate": 7.008571428571429e-06,
      "loss": 0.1427,
      "step": 1550
    },
    {
      "epoch": 1.1314655172413792,
      "grad_norm": 4.806066989898682,
      "learning_rate": 6.937142857142858e-06,
      "loss": 0.1339,
      "step": 1575
    },
    {
      "epoch": 1.1494252873563218,
      "grad_norm": 4.574463367462158,
      "learning_rate": 6.865714285714287e-06,
      "loss": 0.1427,
      "step": 1600
    },
    {
      "epoch": 1.1673850574712643,
      "grad_norm": 4.810680866241455,
      "learning_rate": 6.794285714285714e-06,
      "loss": 0.1313,
      "step": 1625
    },
    {
      "epoch": 1.1853448275862069,
      "grad_norm": 4.358596324920654,
      "learning_rate": 6.722857142857143e-06,
      "loss": 0.1487,
      "step": 1650
    },
    {
      "epoch": 1.2033045977011494,
      "grad_norm": 3.6320159435272217,
      "learning_rate": 6.651428571428572e-06,
      "loss": 0.1373,
      "step": 1675
    },
    {
      "epoch": 1.221264367816092,
      "grad_norm": 4.551269054412842,
      "learning_rate": 6.5800000000000005e-06,
      "loss": 0.1322,
      "step": 1700
    },
    {
      "epoch": 1.2392241379310345,
      "grad_norm": 3.798658847808838,
      "learning_rate": 6.5085714285714295e-06,
      "loss": 0.1324,
      "step": 1725
    },
    {
      "epoch": 1.257183908045977,
      "grad_norm": 3.2752606868743896,
      "learning_rate": 6.437142857142858e-06,
      "loss": 0.1409,
      "step": 1750
    },
    {
      "epoch": 1.2751436781609196,
      "grad_norm": 3.7986721992492676,
      "learning_rate": 6.365714285714286e-06,
      "loss": 0.1316,
      "step": 1775
    },
    {
      "epoch": 1.293103448275862,
      "grad_norm": 3.678849458694458,
      "learning_rate": 6.294285714285715e-06,
      "loss": 0.131,
      "step": 1800
    },
    {
      "epoch": 1.3110632183908046,
      "grad_norm": 4.378350257873535,
      "learning_rate": 6.222857142857144e-06,
      "loss": 0.1213,
      "step": 1825
    },
    {
      "epoch": 1.3290229885057472,
      "grad_norm": 4.888228893280029,
      "learning_rate": 6.151428571428571e-06,
      "loss": 0.1499,
      "step": 1850
    },
    {
      "epoch": 1.3469827586206897,
      "grad_norm": 3.7335093021392822,
      "learning_rate": 6.08e-06,
      "loss": 0.1261,
      "step": 1875
    },
    {
      "epoch": 1.3649425287356323,
      "grad_norm": 4.47926664352417,
      "learning_rate": 6.008571428571429e-06,
      "loss": 0.1361,
      "step": 1900
    },
    {
      "epoch": 1.3829022988505746,
      "grad_norm": 3.6603622436523438,
      "learning_rate": 5.937142857142858e-06,
      "loss": 0.135,
      "step": 1925
    },
    {
      "epoch": 1.4008620689655173,
      "grad_norm": 3.8697991371154785,
      "learning_rate": 5.865714285714286e-06,
      "loss": 0.1157,
      "step": 1950
    },
    {
      "epoch": 1.4188218390804597,
      "grad_norm": 3.898298740386963,
      "learning_rate": 5.794285714285715e-06,
      "loss": 0.1262,
      "step": 1975
    },
    {
      "epoch": 1.4367816091954024,
      "grad_norm": 6.249912261962891,
      "learning_rate": 5.722857142857144e-06,
      "loss": 0.128,
      "step": 2000
    },
    {
      "epoch": 1.4367816091954024,
      "eval_cer": 7.455244974874371,
      "eval_loss": 0.20655690133571625,
      "eval_runtime": 206.1569,
      "eval_samples_per_second": 2.217,
      "eval_steps_per_second": 0.281,
      "step": 2000
    },
    {
      "epoch": 1.4547413793103448,
      "grad_norm": 3.680793285369873,
      "learning_rate": 5.651428571428572e-06,
      "loss": 0.1249,
      "step": 2025
    },
    {
      "epoch": 1.4727011494252873,
      "grad_norm": 4.394272804260254,
      "learning_rate": 5.580000000000001e-06,
      "loss": 0.1327,
      "step": 2050
    },
    {
      "epoch": 1.4906609195402298,
      "grad_norm": 3.971735954284668,
      "learning_rate": 5.508571428571429e-06,
      "loss": 0.1256,
      "step": 2075
    },
    {
      "epoch": 1.5086206896551724,
      "grad_norm": 3.7754571437835693,
      "learning_rate": 5.437142857142857e-06,
      "loss": 0.12,
      "step": 2100
    },
    {
      "epoch": 1.526580459770115,
      "grad_norm": 5.009246826171875,
      "learning_rate": 5.365714285714286e-06,
      "loss": 0.1258,
      "step": 2125
    },
    {
      "epoch": 1.5445402298850575,
      "grad_norm": 3.481464147567749,
      "learning_rate": 5.294285714285715e-06,
      "loss": 0.1174,
      "step": 2150
    },
    {
      "epoch": 1.5625,
      "grad_norm": 3.6977555751800537,
      "learning_rate": 5.2228571428571425e-06,
      "loss": 0.1248,
      "step": 2175
    },
    {
      "epoch": 1.5804597701149425,
      "grad_norm": 3.6558139324188232,
      "learning_rate": 5.1514285714285715e-06,
      "loss": 0.1145,
      "step": 2200
    },
    {
      "epoch": 1.598419540229885,
      "grad_norm": 3.2170028686523438,
      "learning_rate": 5.0800000000000005e-06,
      "loss": 0.1181,
      "step": 2225
    },
    {
      "epoch": 1.6163793103448276,
      "grad_norm": 3.2904043197631836,
      "learning_rate": 5.0085714285714295e-06,
      "loss": 0.1261,
      "step": 2250
    },
    {
      "epoch": 1.6343390804597702,
      "grad_norm": 6.5361223220825195,
      "learning_rate": 4.937142857142858e-06,
      "loss": 0.1257,
      "step": 2275
    },
    {
      "epoch": 1.6522988505747125,
      "grad_norm": 3.8789713382720947,
      "learning_rate": 4.865714285714287e-06,
      "loss": 0.1166,
      "step": 2300
    },
    {
      "epoch": 1.6702586206896552,
      "grad_norm": 3.6503453254699707,
      "learning_rate": 4.794285714285715e-06,
      "loss": 0.1166,
      "step": 2325
    },
    {
      "epoch": 1.6882183908045976,
      "grad_norm": 2.955310344696045,
      "learning_rate": 4.722857142857144e-06,
      "loss": 0.1148,
      "step": 2350
    },
    {
      "epoch": 1.7061781609195403,
      "grad_norm": 3.2080459594726562,
      "learning_rate": 4.651428571428572e-06,
      "loss": 0.1153,
      "step": 2375
    },
    {
      "epoch": 1.7241379310344827,
      "grad_norm": 2.9274492263793945,
      "learning_rate": 4.58e-06,
      "loss": 0.1263,
      "step": 2400
    },
    {
      "epoch": 1.7420977011494254,
      "grad_norm": 3.9885170459747314,
      "learning_rate": 4.508571428571429e-06,
      "loss": 0.1041,
      "step": 2425
    },
    {
      "epoch": 1.7600574712643677,
      "grad_norm": 5.113827228546143,
      "learning_rate": 4.437142857142857e-06,
      "loss": 0.1179,
      "step": 2450
    },
    {
      "epoch": 1.7780172413793105,
      "grad_norm": 5.254714012145996,
      "learning_rate": 4.3657142857142855e-06,
      "loss": 0.1183,
      "step": 2475
    },
    {
      "epoch": 1.7959770114942528,
      "grad_norm": 5.9119696617126465,
      "learning_rate": 4.2942857142857146e-06,
      "loss": 0.1183,
      "step": 2500
    },
    {
      "epoch": 1.8139367816091954,
      "grad_norm": 3.5279624462127686,
      "learning_rate": 4.222857142857143e-06,
      "loss": 0.1174,
      "step": 2525
    },
    {
      "epoch": 1.831896551724138,
      "grad_norm": 3.6319098472595215,
      "learning_rate": 4.151428571428572e-06,
      "loss": 0.1084,
      "step": 2550
    },
    {
      "epoch": 1.8498563218390804,
      "grad_norm": 4.165521621704102,
      "learning_rate": 4.08e-06,
      "loss": 0.1086,
      "step": 2575
    },
    {
      "epoch": 1.867816091954023,
      "grad_norm": 3.5935449600219727,
      "learning_rate": 4.008571428571429e-06,
      "loss": 0.1013,
      "step": 2600
    },
    {
      "epoch": 1.8857758620689655,
      "grad_norm": 4.520440101623535,
      "learning_rate": 3.937142857142858e-06,
      "loss": 0.1085,
      "step": 2625
    },
    {
      "epoch": 1.903735632183908,
      "grad_norm": 4.8552327156066895,
      "learning_rate": 3.865714285714286e-06,
      "loss": 0.1075,
      "step": 2650
    },
    {
      "epoch": 1.9216954022988506,
      "grad_norm": 5.094956874847412,
      "learning_rate": 3.7942857142857147e-06,
      "loss": 0.1153,
      "step": 2675
    },
    {
      "epoch": 1.9396551724137931,
      "grad_norm": 4.34927225112915,
      "learning_rate": 3.722857142857143e-06,
      "loss": 0.1092,
      "step": 2700
    },
    {
      "epoch": 1.9576149425287355,
      "grad_norm": 2.9122841358184814,
      "learning_rate": 3.651428571428572e-06,
      "loss": 0.1094,
      "step": 2725
    },
    {
      "epoch": 1.9755747126436782,
      "grad_norm": 4.0797529220581055,
      "learning_rate": 3.58e-06,
      "loss": 0.1031,
      "step": 2750
    },
    {
      "epoch": 1.9935344827586206,
      "grad_norm": 3.478084087371826,
      "learning_rate": 3.508571428571429e-06,
      "loss": 0.1035,
      "step": 2775
    },
    {
      "epoch": 2.0114942528735633,
      "grad_norm": 2.5905420780181885,
      "learning_rate": 3.437142857142857e-06,
      "loss": 0.086,
      "step": 2800
    },
    {
      "epoch": 2.0294540229885056,
      "grad_norm": 2.674304246902466,
      "learning_rate": 3.3657142857142862e-06,
      "loss": 0.0741,
      "step": 2825
    },
    {
      "epoch": 2.0474137931034484,
      "grad_norm": 3.0569241046905518,
      "learning_rate": 3.2942857142857144e-06,
      "loss": 0.0722,
      "step": 2850
    },
    {
      "epoch": 2.0653735632183907,
      "grad_norm": 2.8325390815734863,
      "learning_rate": 3.222857142857143e-06,
      "loss": 0.0759,
      "step": 2875
    },
    {
      "epoch": 2.0833333333333335,
      "grad_norm": 3.281291961669922,
      "learning_rate": 3.151428571428572e-06,
      "loss": 0.0803,
      "step": 2900
    },
    {
      "epoch": 2.101293103448276,
      "grad_norm": 2.3541665077209473,
      "learning_rate": 3.08e-06,
      "loss": 0.0672,
      "step": 2925
    },
    {
      "epoch": 2.1192528735632186,
      "grad_norm": 3.7438085079193115,
      "learning_rate": 3.008571428571429e-06,
      "loss": 0.0737,
      "step": 2950
    },
    {
      "epoch": 2.137212643678161,
      "grad_norm": 2.571523427963257,
      "learning_rate": 2.9371428571428573e-06,
      "loss": 0.065,
      "step": 2975
    },
    {
      "epoch": 2.1551724137931036,
      "grad_norm": 2.559622287750244,
      "learning_rate": 2.865714285714286e-06,
      "loss": 0.0725,
      "step": 3000
    },
    {
      "epoch": 2.1551724137931036,
      "eval_cer": 5.755339195979899,
      "eval_loss": 0.17403484880924225,
      "eval_runtime": 203.7679,
      "eval_samples_per_second": 2.243,
      "eval_steps_per_second": 0.285,
      "step": 3000
    },
    {
      "epoch": 2.173132183908046,
      "grad_norm": 2.8300039768218994,
      "learning_rate": 2.7942857142857145e-06,
      "loss": 0.0691,
      "step": 3025
    },
    {
      "epoch": 2.1910919540229887,
      "grad_norm": 2.468639850616455,
      "learning_rate": 2.722857142857143e-06,
      "loss": 0.0834,
      "step": 3050
    },
    {
      "epoch": 2.209051724137931,
      "grad_norm": 3.254262685775757,
      "learning_rate": 2.6514285714285713e-06,
      "loss": 0.0727,
      "step": 3075
    },
    {
      "epoch": 2.2270114942528734,
      "grad_norm": 3.118223190307617,
      "learning_rate": 2.5800000000000003e-06,
      "loss": 0.0783,
      "step": 3100
    },
    {
      "epoch": 2.244971264367816,
      "grad_norm": 2.0613930225372314,
      "learning_rate": 2.5085714285714285e-06,
      "loss": 0.0677,
      "step": 3125
    },
    {
      "epoch": 2.2629310344827585,
      "grad_norm": 4.234766483306885,
      "learning_rate": 2.4371428571428575e-06,
      "loss": 0.0695,
      "step": 3150
    },
    {
      "epoch": 2.280890804597701,
      "grad_norm": 3.091860294342041,
      "learning_rate": 2.365714285714286e-06,
      "loss": 0.0674,
      "step": 3175
    },
    {
      "epoch": 2.2988505747126435,
      "grad_norm": 2.968728542327881,
      "learning_rate": 2.2942857142857146e-06,
      "loss": 0.0793,
      "step": 3200
    },
    {
      "epoch": 2.3168103448275863,
      "grad_norm": 2.635185956954956,
      "learning_rate": 2.222857142857143e-06,
      "loss": 0.0669,
      "step": 3225
    },
    {
      "epoch": 2.3347701149425286,
      "grad_norm": 4.698062896728516,
      "learning_rate": 2.1514285714285714e-06,
      "loss": 0.0792,
      "step": 3250
    },
    {
      "epoch": 2.3527298850574714,
      "grad_norm": 2.380126953125,
      "learning_rate": 2.08e-06,
      "loss": 0.0686,
      "step": 3275
    },
    {
      "epoch": 2.3706896551724137,
      "grad_norm": 2.4431111812591553,
      "learning_rate": 2.0085714285714286e-06,
      "loss": 0.0683,
      "step": 3300
    },
    {
      "epoch": 2.3886494252873565,
      "grad_norm": 2.8130743503570557,
      "learning_rate": 1.9371428571428576e-06,
      "loss": 0.0726,
      "step": 3325
    },
    {
      "epoch": 2.406609195402299,
      "grad_norm": 2.115403890609741,
      "learning_rate": 1.865714285714286e-06,
      "loss": 0.0715,
      "step": 3350
    },
    {
      "epoch": 2.4245689655172415,
      "grad_norm": 3.2712090015411377,
      "learning_rate": 1.7942857142857146e-06,
      "loss": 0.0677,
      "step": 3375
    },
    {
      "epoch": 2.442528735632184,
      "grad_norm": 2.903533697128296,
      "learning_rate": 1.7228571428571432e-06,
      "loss": 0.0722,
      "step": 3400
    },
    {
      "epoch": 2.4604885057471266,
      "grad_norm": 2.4043657779693604,
      "learning_rate": 1.6514285714285715e-06,
      "loss": 0.0723,
      "step": 3425
    },
    {
      "epoch": 2.478448275862069,
      "grad_norm": 2.5334341526031494,
      "learning_rate": 1.5800000000000001e-06,
      "loss": 0.0729,
      "step": 3450
    },
    {
      "epoch": 2.4964080459770113,
      "grad_norm": 2.8699309825897217,
      "learning_rate": 1.5085714285714287e-06,
      "loss": 0.0733,
      "step": 3475
    },
    {
      "epoch": 2.514367816091954,
      "grad_norm": 3.6407063007354736,
      "learning_rate": 1.4371428571428573e-06,
      "loss": 0.0912,
      "step": 3500
    },
    {
      "epoch": 2.532327586206897,
      "grad_norm": 2.9028170108795166,
      "learning_rate": 1.3657142857142857e-06,
      "loss": 0.0704,
      "step": 3525
    },
    {
      "epoch": 2.550287356321839,
      "grad_norm": 2.9066832065582275,
      "learning_rate": 1.2942857142857143e-06,
      "loss": 0.0666,
      "step": 3550
    },
    {
      "epoch": 2.5682471264367814,
      "grad_norm": 3.7357940673828125,
      "learning_rate": 1.222857142857143e-06,
      "loss": 0.075,
      "step": 3575
    },
    {
      "epoch": 2.586206896551724,
      "grad_norm": 2.8801915645599365,
      "learning_rate": 1.1514285714285714e-06,
      "loss": 0.0792,
      "step": 3600
    },
    {
      "epoch": 2.6041666666666665,
      "grad_norm": 3.622131824493408,
      "learning_rate": 1.08e-06,
      "loss": 0.07,
      "step": 3625
    },
    {
      "epoch": 2.6221264367816093,
      "grad_norm": 3.064757823944092,
      "learning_rate": 1.0085714285714286e-06,
      "loss": 0.066,
      "step": 3650
    },
    {
      "epoch": 2.6400862068965516,
      "grad_norm": 2.172606945037842,
      "learning_rate": 9.371428571428571e-07,
      "loss": 0.0579,
      "step": 3675
    },
    {
      "epoch": 2.6580459770114944,
      "grad_norm": 4.104800701141357,
      "learning_rate": 8.657142857142858e-07,
      "loss": 0.078,
      "step": 3700
    },
    {
      "epoch": 2.6760057471264367,
      "grad_norm": 3.803704261779785,
      "learning_rate": 7.942857142857144e-07,
      "loss": 0.0768,
      "step": 3725
    },
    {
      "epoch": 2.6939655172413794,
      "grad_norm": 1.876494288444519,
      "learning_rate": 7.228571428571429e-07,
      "loss": 0.066,
      "step": 3750
    },
    {
      "epoch": 2.7119252873563218,
      "grad_norm": 4.310028076171875,
      "learning_rate": 6.514285714285715e-07,
      "loss": 0.0665,
      "step": 3775
    },
    {
      "epoch": 2.7298850574712645,
      "grad_norm": 3.0695154666900635,
      "learning_rate": 5.800000000000001e-07,
      "loss": 0.0695,
      "step": 3800
    },
    {
      "epoch": 2.747844827586207,
      "grad_norm": 3.2505669593811035,
      "learning_rate": 5.085714285714286e-07,
      "loss": 0.065,
      "step": 3825
    },
    {
      "epoch": 2.765804597701149,
      "grad_norm": 2.551088809967041,
      "learning_rate": 4.371428571428572e-07,
      "loss": 0.0649,
      "step": 3850
    },
    {
      "epoch": 2.783764367816092,
      "grad_norm": 3.110604763031006,
      "learning_rate": 3.657142857142858e-07,
      "loss": 0.0704,
      "step": 3875
    },
    {
      "epoch": 2.8017241379310347,
      "grad_norm": 2.836848020553589,
      "learning_rate": 2.942857142857143e-07,
      "loss": 0.0698,
      "step": 3900
    },
    {
      "epoch": 2.819683908045977,
      "grad_norm": 2.818726062774658,
      "learning_rate": 2.228571428571429e-07,
      "loss": 0.0722,
      "step": 3925
    },
    {
      "epoch": 2.8376436781609193,
      "grad_norm": 2.192012071609497,
      "learning_rate": 1.5142857142857144e-07,
      "loss": 0.0657,
      "step": 3950
    },
    {
      "epoch": 2.855603448275862,
      "grad_norm": 3.1882786750793457,
      "learning_rate": 8e-08,
      "loss": 0.0751,
      "step": 3975
    },
    {
      "epoch": 2.873563218390805,
      "grad_norm": 3.386892557144165,
      "learning_rate": 8.571428571428572e-09,
      "loss": 0.0673,
      "step": 4000
    },
    {
      "epoch": 2.873563218390805,
      "eval_cer": 5.288159547738693,
      "eval_loss": 0.16544735431671143,
      "eval_runtime": 204.9568,
      "eval_samples_per_second": 2.23,
      "eval_steps_per_second": 0.283,
      "step": 4000
    }
  ],
  "logging_steps": 25,
  "max_steps": 4000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.14987227430912e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}