| { | |
| "best_metric": 34.549659010739205, | |
| "best_model_checkpoint": "./whisper-small-hi/checkpoint-4000", | |
| "epoch": 2.546148949713558, | |
| "eval_steps": 1000, | |
| "global_step": 4000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.015913430935709738, | |
| "grad_norm": 6.266472339630127, | |
| "learning_rate": 1.44e-05, | |
| "loss": 1.7883, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.031826861871419476, | |
| "grad_norm": 11.293601989746094, | |
| "learning_rate": 2.94e-05, | |
| "loss": 1.3114, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.047740292807129214, | |
| "grad_norm": 11.253959655761719, | |
| "learning_rate": 4.4399999999999995e-05, | |
| "loss": 0.7712, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.06365372374283895, | |
| "grad_norm": 6.763392448425293, | |
| "learning_rate": 5.94e-05, | |
| "loss": 0.4958, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07956715467854869, | |
| "grad_norm": 7.226134300231934, | |
| "learning_rate": 7.439999999999999e-05, | |
| "loss": 0.4231, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.09548058561425843, | |
| "grad_norm": 8.527499198913574, | |
| "learning_rate": 8.939999999999999e-05, | |
| "loss": 0.3815, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.11139401654996817, | |
| "grad_norm": 6.013850212097168, | |
| "learning_rate": 0.00010439999999999999, | |
| "loss": 0.3633, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.1273074474856779, | |
| "grad_norm": 5.921957015991211, | |
| "learning_rate": 0.0001194, | |
| "loss": 0.3564, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.14322087842138764, | |
| "grad_norm": 6.438776969909668, | |
| "learning_rate": 0.0001344, | |
| "loss": 0.3549, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.15913430935709738, | |
| "grad_norm": 6.543764114379883, | |
| "learning_rate": 0.0001494, | |
| "loss": 0.3298, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.17504774029280712, | |
| "grad_norm": 5.377121448516846, | |
| "learning_rate": 0.0001644, | |
| "loss": 0.3251, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.19096117122851686, | |
| "grad_norm": 6.325755596160889, | |
| "learning_rate": 0.00017939999999999997, | |
| "loss": 0.3319, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.2068746021642266, | |
| "grad_norm": 3.889002561569214, | |
| "learning_rate": 0.00019439999999999998, | |
| "loss": 0.3311, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.22278803309993633, | |
| "grad_norm": 4.005517482757568, | |
| "learning_rate": 0.00020939999999999997, | |
| "loss": 0.3388, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.23870146403564607, | |
| "grad_norm": 3.7465219497680664, | |
| "learning_rate": 0.00022439999999999998, | |
| "loss": 0.3304, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.2546148949713558, | |
| "grad_norm": 3.612300157546997, | |
| "learning_rate": 0.0002394, | |
| "loss": 0.3253, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.27052832590706555, | |
| "grad_norm": 2.8564534187316895, | |
| "learning_rate": 0.00025439999999999995, | |
| "loss": 0.325, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.2864417568427753, | |
| "grad_norm": 3.6657044887542725, | |
| "learning_rate": 0.0002694, | |
| "loss": 0.3362, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.302355187778485, | |
| "grad_norm": 6.07236385345459, | |
| "learning_rate": 0.0002844, | |
| "loss": 0.3396, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.31826861871419476, | |
| "grad_norm": 3.5293257236480713, | |
| "learning_rate": 0.00029939999999999996, | |
| "loss": 0.3265, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.3341820496499045, | |
| "grad_norm": 2.7845981121063232, | |
| "learning_rate": 0.0002979428571428571, | |
| "loss": 0.3066, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.35009548058561424, | |
| "grad_norm": 3.957970380783081, | |
| "learning_rate": 0.0002958, | |
| "loss": 0.3076, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.366008911521324, | |
| "grad_norm": 2.7481985092163086, | |
| "learning_rate": 0.00029365714285714285, | |
| "loss": 0.3181, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.3819223424570337, | |
| "grad_norm": 5.5784912109375, | |
| "learning_rate": 0.0002915142857142857, | |
| "loss": 0.325, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.39783577339274345, | |
| "grad_norm": 2.688514232635498, | |
| "learning_rate": 0.0002893714285714285, | |
| "loss": 0.2907, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.4137492043284532, | |
| "grad_norm": 1.8559094667434692, | |
| "learning_rate": 0.0002872285714285714, | |
| "loss": 0.2837, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.42966263526416293, | |
| "grad_norm": 2.075263023376465, | |
| "learning_rate": 0.00028508571428571426, | |
| "loss": 0.2884, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.44557606619987267, | |
| "grad_norm": 2.253748893737793, | |
| "learning_rate": 0.00028294285714285713, | |
| "loss": 0.2652, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.4614894971355824, | |
| "grad_norm": 2.5322704315185547, | |
| "learning_rate": 0.0002808, | |
| "loss": 0.2763, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.47740292807129214, | |
| "grad_norm": 2.8500661849975586, | |
| "learning_rate": 0.0002786571428571428, | |
| "loss": 0.2698, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.49331635900700194, | |
| "grad_norm": 2.4770638942718506, | |
| "learning_rate": 0.0002765142857142857, | |
| "loss": 0.2643, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.5092297899427116, | |
| "grad_norm": 2.8346035480499268, | |
| "learning_rate": 0.00027437142857142854, | |
| "loss": 0.2682, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.5251432208784214, | |
| "grad_norm": 2.551896095275879, | |
| "learning_rate": 0.0002722285714285714, | |
| "loss": 0.2484, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.5410566518141311, | |
| "grad_norm": 2.1376454830169678, | |
| "learning_rate": 0.0002700857142857143, | |
| "loss": 0.2433, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.5569700827498408, | |
| "grad_norm": 2.4351694583892822, | |
| "learning_rate": 0.0002679428571428571, | |
| "loss": 0.248, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.5728835136855506, | |
| "grad_norm": 2.142422914505005, | |
| "learning_rate": 0.00026579999999999996, | |
| "loss": 0.2592, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.5887969446212603, | |
| "grad_norm": 2.287863254547119, | |
| "learning_rate": 0.0002636571428571428, | |
| "loss": 0.2361, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.60471037555697, | |
| "grad_norm": 1.8011589050292969, | |
| "learning_rate": 0.0002615142857142857, | |
| "loss": 0.2349, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.6206238064926798, | |
| "grad_norm": 2.0155577659606934, | |
| "learning_rate": 0.00025937142857142856, | |
| "loss": 0.2343, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 0.6365372374283895, | |
| "grad_norm": 1.7602595090866089, | |
| "learning_rate": 0.0002572285714285714, | |
| "loss": 0.2369, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6365372374283895, | |
| "eval_loss": 0.24331499636173248, | |
| "eval_runtime": 3161.0794, | |
| "eval_samples_per_second": 2.642, | |
| "eval_steps_per_second": 0.331, | |
| "eval_wer": 62.18807974706697, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6524506683640993, | |
| "grad_norm": 2.084760904312134, | |
| "learning_rate": 0.0002550857142857143, | |
| "loss": 0.2352, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 0.668364099299809, | |
| "grad_norm": 1.6412831544876099, | |
| "learning_rate": 0.00025294285714285716, | |
| "loss": 0.2242, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.6842775302355187, | |
| "grad_norm": 1.3746275901794434, | |
| "learning_rate": 0.00025079999999999997, | |
| "loss": 0.2206, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 0.7001909611712285, | |
| "grad_norm": 2.017609119415283, | |
| "learning_rate": 0.00024865714285714284, | |
| "loss": 0.2223, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.7161043921069382, | |
| "grad_norm": 2.0669291019439697, | |
| "learning_rate": 0.0002465142857142857, | |
| "loss": 0.2233, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 0.732017823042648, | |
| "grad_norm": 1.699652075767517, | |
| "learning_rate": 0.00024437142857142857, | |
| "loss": 0.1984, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.7479312539783577, | |
| "grad_norm": 1.5082184076309204, | |
| "learning_rate": 0.00024222857142857138, | |
| "loss": 0.2143, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 0.7638446849140674, | |
| "grad_norm": 1.363891363143921, | |
| "learning_rate": 0.00024008571428571425, | |
| "loss": 0.2067, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.7797581158497772, | |
| "grad_norm": 1.3396174907684326, | |
| "learning_rate": 0.00023794285714285712, | |
| "loss": 0.2083, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 0.7956715467854869, | |
| "grad_norm": 1.6803642511367798, | |
| "learning_rate": 0.00023579999999999999, | |
| "loss": 0.2072, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.8115849777211966, | |
| "grad_norm": 1.8154999017715454, | |
| "learning_rate": 0.00023365714285714283, | |
| "loss": 0.2137, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 0.8274984086569064, | |
| "grad_norm": 1.9345273971557617, | |
| "learning_rate": 0.0002315142857142857, | |
| "loss": 0.1854, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.8434118395926161, | |
| "grad_norm": 1.8204659223556519, | |
| "learning_rate": 0.00022937142857142856, | |
| "loss": 0.1944, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 0.8593252705283259, | |
| "grad_norm": 1.7036362886428833, | |
| "learning_rate": 0.00022722857142857143, | |
| "loss": 0.1925, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.8752387014640356, | |
| "grad_norm": 1.5732781887054443, | |
| "learning_rate": 0.0002250857142857143, | |
| "loss": 0.2006, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 0.8911521323997453, | |
| "grad_norm": 1.4482449293136597, | |
| "learning_rate": 0.0002229428571428571, | |
| "loss": 0.1814, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.9070655633354551, | |
| "grad_norm": 1.474438190460205, | |
| "learning_rate": 0.00022079999999999997, | |
| "loss": 0.1876, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 0.9229789942711648, | |
| "grad_norm": 1.6966629028320312, | |
| "learning_rate": 0.00021865714285714284, | |
| "loss": 0.186, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.9388924252068745, | |
| "grad_norm": 1.6632884740829468, | |
| "learning_rate": 0.0002165142857142857, | |
| "loss": 0.1846, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 0.9548058561425843, | |
| "grad_norm": 1.4217106103897095, | |
| "learning_rate": 0.00021437142857142855, | |
| "loss": 0.1714, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.9707192870782941, | |
| "grad_norm": 1.280887484550476, | |
| "learning_rate": 0.00021222857142857141, | |
| "loss": 0.1794, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 0.9866327180140039, | |
| "grad_norm": 1.7453186511993408, | |
| "learning_rate": 0.00021008571428571428, | |
| "loss": 0.1788, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.0025461489497136, | |
| "grad_norm": 1.3471609354019165, | |
| "learning_rate": 0.00020794285714285712, | |
| "loss": 0.1631, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 1.0184595798854232, | |
| "grad_norm": 1.6721386909484863, | |
| "learning_rate": 0.0002058, | |
| "loss": 0.1388, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.034373010821133, | |
| "grad_norm": 1.3706496953964233, | |
| "learning_rate": 0.00020365714285714283, | |
| "loss": 0.139, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 1.0502864417568427, | |
| "grad_norm": 1.8108116388320923, | |
| "learning_rate": 0.0002015142857142857, | |
| "loss": 0.1486, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 1.0661998726925526, | |
| "grad_norm": 1.5080534219741821, | |
| "learning_rate": 0.00019937142857142856, | |
| "loss": 0.1431, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 1.0821133036282622, | |
| "grad_norm": 1.377472996711731, | |
| "learning_rate": 0.00019722857142857143, | |
| "loss": 0.1493, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.098026734563972, | |
| "grad_norm": 1.4746475219726562, | |
| "learning_rate": 0.00019508571428571427, | |
| "loss": 0.1386, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 1.1139401654996817, | |
| "grad_norm": 1.4167110919952393, | |
| "learning_rate": 0.0001929428571428571, | |
| "loss": 0.144, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.1298535964353915, | |
| "grad_norm": 1.4052375555038452, | |
| "learning_rate": 0.00019079999999999998, | |
| "loss": 0.1479, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 1.1457670273711011, | |
| "grad_norm": 1.1903959512710571, | |
| "learning_rate": 0.00018865714285714284, | |
| "loss": 0.1319, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.161680458306811, | |
| "grad_norm": 1.5119489431381226, | |
| "learning_rate": 0.00018651428571428568, | |
| "loss": 0.1408, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 1.1775938892425206, | |
| "grad_norm": 1.2824383974075317, | |
| "learning_rate": 0.00018437142857142855, | |
| "loss": 0.1379, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.1935073201782305, | |
| "grad_norm": 2.1692168712615967, | |
| "learning_rate": 0.00018222857142857142, | |
| "loss": 0.1411, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 1.20942075111394, | |
| "grad_norm": 1.2325959205627441, | |
| "learning_rate": 0.00018008571428571428, | |
| "loss": 0.1333, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.22533418204965, | |
| "grad_norm": 0.9051541090011597, | |
| "learning_rate": 0.00017794285714285715, | |
| "loss": 0.1523, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 1.2412476129853596, | |
| "grad_norm": 1.3505010604858398, | |
| "learning_rate": 0.00017579999999999996, | |
| "loss": 0.1363, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 1.2571610439210694, | |
| "grad_norm": 1.3619211912155151, | |
| "learning_rate": 0.00017365714285714283, | |
| "loss": 0.1308, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 1.273074474856779, | |
| "grad_norm": 1.2368829250335693, | |
| "learning_rate": 0.0001715142857142857, | |
| "loss": 0.1242, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.273074474856779, | |
| "eval_loss": 0.17340172827243805, | |
| "eval_runtime": 3339.9389, | |
| "eval_samples_per_second": 2.501, | |
| "eval_steps_per_second": 0.313, | |
| "eval_wer": 49.436910454391054, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.288987905792489, | |
| "grad_norm": 1.3111966848373413, | |
| "learning_rate": 0.00016937142857142856, | |
| "loss": 0.1318, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 1.3049013367281985, | |
| "grad_norm": 1.3677690029144287, | |
| "learning_rate": 0.0001672285714285714, | |
| "loss": 0.1368, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 1.3208147676639084, | |
| "grad_norm": 1.1350477933883667, | |
| "learning_rate": 0.00016508571428571427, | |
| "loss": 0.1218, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 1.336728198599618, | |
| "grad_norm": 1.4157230854034424, | |
| "learning_rate": 0.00016294285714285714, | |
| "loss": 0.1245, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.3526416295353278, | |
| "grad_norm": 1.6313213109970093, | |
| "learning_rate": 0.0001608, | |
| "loss": 0.1249, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 1.3685550604710375, | |
| "grad_norm": 1.1400282382965088, | |
| "learning_rate": 0.00015865714285714282, | |
| "loss": 0.1258, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 1.3844684914067473, | |
| "grad_norm": 1.0388585329055786, | |
| "learning_rate": 0.00015651428571428569, | |
| "loss": 0.1278, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 1.400381922342457, | |
| "grad_norm": 1.1972434520721436, | |
| "learning_rate": 0.00015437142857142855, | |
| "loss": 0.1294, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.4162953532781668, | |
| "grad_norm": 1.5634195804595947, | |
| "learning_rate": 0.00015222857142857142, | |
| "loss": 0.122, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 1.4322087842138764, | |
| "grad_norm": 1.0193285942077637, | |
| "learning_rate": 0.00015008571428571429, | |
| "loss": 0.116, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.4481222151495863, | |
| "grad_norm": 1.2780238389968872, | |
| "learning_rate": 0.00014794285714285713, | |
| "loss": 0.124, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 1.464035646085296, | |
| "grad_norm": 1.1720143556594849, | |
| "learning_rate": 0.0001458, | |
| "loss": 0.1267, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.4799490770210058, | |
| "grad_norm": 1.1181762218475342, | |
| "learning_rate": 0.00014365714285714286, | |
| "loss": 0.1131, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 1.4958625079567156, | |
| "grad_norm": 1.2259148359298706, | |
| "learning_rate": 0.0001415142857142857, | |
| "loss": 0.1121, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 1.5117759388924252, | |
| "grad_norm": 1.2877577543258667, | |
| "learning_rate": 0.00013937142857142857, | |
| "loss": 0.1125, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 1.5276893698281349, | |
| "grad_norm": 0.9741705656051636, | |
| "learning_rate": 0.0001372285714285714, | |
| "loss": 0.1122, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.5436028007638447, | |
| "grad_norm": 1.1237064599990845, | |
| "learning_rate": 0.00013508571428571427, | |
| "loss": 0.1135, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 1.5595162316995546, | |
| "grad_norm": 0.9983360767364502, | |
| "learning_rate": 0.00013294285714285711, | |
| "loss": 0.1053, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 1.5754296626352642, | |
| "grad_norm": 0.941571831703186, | |
| "learning_rate": 0.00013079999999999998, | |
| "loss": 0.1104, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 1.5913430935709738, | |
| "grad_norm": 0.8732393383979797, | |
| "learning_rate": 0.00012865714285714285, | |
| "loss": 0.1073, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.6072565245066837, | |
| "grad_norm": 1.1299751996994019, | |
| "learning_rate": 0.0001265142857142857, | |
| "loss": 0.1121, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 1.6231699554423935, | |
| "grad_norm": 0.8921777606010437, | |
| "learning_rate": 0.00012437142857142855, | |
| "loss": 0.1097, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 1.6390833863781031, | |
| "grad_norm": 0.8513890504837036, | |
| "learning_rate": 0.00012222857142857142, | |
| "loss": 0.1082, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 1.6549968173138128, | |
| "grad_norm": 1.0893051624298096, | |
| "learning_rate": 0.00012008571428571428, | |
| "loss": 0.1029, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.6709102482495226, | |
| "grad_norm": 0.8744276165962219, | |
| "learning_rate": 0.00011794285714285713, | |
| "loss": 0.1086, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 1.6868236791852325, | |
| "grad_norm": 1.1604117155075073, | |
| "learning_rate": 0.0001158, | |
| "loss": 0.1008, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 1.702737110120942, | |
| "grad_norm": 0.9912447929382324, | |
| "learning_rate": 0.00011365714285714284, | |
| "loss": 0.1022, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 1.7186505410566517, | |
| "grad_norm": 1.088443636894226, | |
| "learning_rate": 0.0001115142857142857, | |
| "loss": 0.1029, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.7345639719923616, | |
| "grad_norm": 1.1775622367858887, | |
| "learning_rate": 0.00010937142857142856, | |
| "loss": 0.1041, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 1.7504774029280714, | |
| "grad_norm": 1.1188244819641113, | |
| "learning_rate": 0.00010722857142857142, | |
| "loss": 0.1068, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 1.766390833863781, | |
| "grad_norm": 0.9570010900497437, | |
| "learning_rate": 0.00010508571428571429, | |
| "loss": 0.098, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 1.7823042647994907, | |
| "grad_norm": 1.4812625646591187, | |
| "learning_rate": 0.00010294285714285713, | |
| "loss": 0.0949, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.7982176957352005, | |
| "grad_norm": 0.8738514184951782, | |
| "learning_rate": 0.0001008, | |
| "loss": 0.1003, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 1.8141311266709104, | |
| "grad_norm": 0.9198557734489441, | |
| "learning_rate": 9.865714285714285e-05, | |
| "loss": 0.0913, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 1.83004455760662, | |
| "grad_norm": 0.8232097625732422, | |
| "learning_rate": 9.65142857142857e-05, | |
| "loss": 0.1017, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 1.8459579885423296, | |
| "grad_norm": 0.7927132844924927, | |
| "learning_rate": 9.437142857142856e-05, | |
| "loss": 0.1006, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.8618714194780395, | |
| "grad_norm": 0.8228179812431335, | |
| "learning_rate": 9.222857142857142e-05, | |
| "loss": 0.0954, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 1.8777848504137493, | |
| "grad_norm": 1.0207316875457764, | |
| "learning_rate": 9.008571428571428e-05, | |
| "loss": 0.1015, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 1.893698281349459, | |
| "grad_norm": 1.3909655809402466, | |
| "learning_rate": 8.794285714285713e-05, | |
| "loss": 0.0966, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 1.9096117122851686, | |
| "grad_norm": 0.8367329835891724, | |
| "learning_rate": 8.579999999999998e-05, | |
| "loss": 0.1022, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.9096117122851686, | |
| "eval_loss": 0.11972030252218246, | |
| "eval_runtime": 3680.6627, | |
| "eval_samples_per_second": 2.269, | |
| "eval_steps_per_second": 0.284, | |
| "eval_wer": 39.05306890334718, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.9255251432208784, | |
| "grad_norm": 0.8776394724845886, | |
| "learning_rate": 8.365714285714285e-05, | |
| "loss": 0.0951, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 1.9414385741565883, | |
| "grad_norm": 0.7883875966072083, | |
| "learning_rate": 8.151428571428572e-05, | |
| "loss": 0.0887, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 1.957352005092298, | |
| "grad_norm": 0.9402434229850769, | |
| "learning_rate": 7.937142857142856e-05, | |
| "loss": 0.0989, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 1.9732654360280075, | |
| "grad_norm": 0.9308575391769409, | |
| "learning_rate": 7.722857142857143e-05, | |
| "loss": 0.0913, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 1.9891788669637174, | |
| "grad_norm": 0.8554713129997253, | |
| "learning_rate": 7.508571428571428e-05, | |
| "loss": 0.0841, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 2.0050922978994272, | |
| "grad_norm": 0.6500595211982727, | |
| "learning_rate": 7.294285714285713e-05, | |
| "loss": 0.0778, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 2.021005728835137, | |
| "grad_norm": 0.8015878200531006, | |
| "learning_rate": 7.079999999999999e-05, | |
| "loss": 0.0485, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 2.0369191597708465, | |
| "grad_norm": 0.9871166944503784, | |
| "learning_rate": 6.865714285714285e-05, | |
| "loss": 0.0538, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.0528325907065565, | |
| "grad_norm": 0.7539300918579102, | |
| "learning_rate": 6.65142857142857e-05, | |
| "loss": 0.0567, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 2.068746021642266, | |
| "grad_norm": 0.7191395163536072, | |
| "learning_rate": 6.437142857142857e-05, | |
| "loss": 0.0474, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 2.084659452577976, | |
| "grad_norm": 0.8346642851829529, | |
| "learning_rate": 6.222857142857143e-05, | |
| "loss": 0.0565, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 2.1005728835136854, | |
| "grad_norm": 0.890232264995575, | |
| "learning_rate": 6.008571428571428e-05, | |
| "loss": 0.0513, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.1164863144493955, | |
| "grad_norm": 0.7644281983375549, | |
| "learning_rate": 5.794285714285714e-05, | |
| "loss": 0.0494, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 2.132399745385105, | |
| "grad_norm": 1.0637160539627075, | |
| "learning_rate": 5.5799999999999994e-05, | |
| "loss": 0.0481, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 2.1483131763208148, | |
| "grad_norm": 0.7097823619842529, | |
| "learning_rate": 5.3657142857142855e-05, | |
| "loss": 0.0466, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 2.1642266072565244, | |
| "grad_norm": 1.0402812957763672, | |
| "learning_rate": 5.151428571428571e-05, | |
| "loss": 0.053, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.1801400381922345, | |
| "grad_norm": 1.0634698867797852, | |
| "learning_rate": 4.937142857142856e-05, | |
| "loss": 0.0486, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 2.196053469127944, | |
| "grad_norm": 0.5193982720375061, | |
| "learning_rate": 4.722857142857142e-05, | |
| "loss": 0.0468, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 2.2119669000636537, | |
| "grad_norm": 0.6877008676528931, | |
| "learning_rate": 4.5085714285714275e-05, | |
| "loss": 0.0489, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 2.2278803309993633, | |
| "grad_norm": 0.7828670740127563, | |
| "learning_rate": 4.294285714285714e-05, | |
| "loss": 0.0461, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.2437937619350734, | |
| "grad_norm": 0.714133620262146, | |
| "learning_rate": 4.08e-05, | |
| "loss": 0.047, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 2.259707192870783, | |
| "grad_norm": 0.6755720376968384, | |
| "learning_rate": 3.8657142857142856e-05, | |
| "loss": 0.0456, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 2.2756206238064927, | |
| "grad_norm": 0.908721387386322, | |
| "learning_rate": 3.651428571428571e-05, | |
| "loss": 0.0427, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 2.2915340547422023, | |
| "grad_norm": 0.703593373298645, | |
| "learning_rate": 3.437142857142857e-05, | |
| "loss": 0.049, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.307447485677912, | |
| "grad_norm": 0.46566998958587646, | |
| "learning_rate": 3.222857142857142e-05, | |
| "loss": 0.0436, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 2.323360916613622, | |
| "grad_norm": 0.7065137028694153, | |
| "learning_rate": 3.0085714285714283e-05, | |
| "loss": 0.0429, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 2.3392743475493316, | |
| "grad_norm": 0.7245275974273682, | |
| "learning_rate": 2.794285714285714e-05, | |
| "loss": 0.0436, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 2.3551877784850412, | |
| "grad_norm": 0.5561261177062988, | |
| "learning_rate": 2.5799999999999997e-05, | |
| "loss": 0.0466, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.3711012094207513, | |
| "grad_norm": 0.6039161086082458, | |
| "learning_rate": 2.3657142857142857e-05, | |
| "loss": 0.0416, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 2.387014640356461, | |
| "grad_norm": 0.5791841149330139, | |
| "learning_rate": 2.1514285714285714e-05, | |
| "loss": 0.042, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 2.4029280712921706, | |
| "grad_norm": 0.5699833631515503, | |
| "learning_rate": 1.937142857142857e-05, | |
| "loss": 0.0433, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 2.41884150222788, | |
| "grad_norm": 0.8795793056488037, | |
| "learning_rate": 1.7228571428571428e-05, | |
| "loss": 0.0428, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.4347549331635903, | |
| "grad_norm": 0.7127372622489929, | |
| "learning_rate": 1.5085714285714285e-05, | |
| "loss": 0.0386, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 2.4506683640993, | |
| "grad_norm": 0.7226797342300415, | |
| "learning_rate": 1.2942857142857141e-05, | |
| "loss": 0.0443, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 2.4665817950350095, | |
| "grad_norm": 0.738706648349762, | |
| "learning_rate": 1.0799999999999998e-05, | |
| "loss": 0.0426, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 2.482495225970719, | |
| "grad_norm": 0.9381580352783203, | |
| "learning_rate": 8.657142857142855e-06, | |
| "loss": 0.0436, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.498408656906429, | |
| "grad_norm": 0.863370418548584, | |
| "learning_rate": 6.514285714285714e-06, | |
| "loss": 0.0436, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 2.514322087842139, | |
| "grad_norm": 0.7624787092208862, | |
| "learning_rate": 4.371428571428571e-06, | |
| "loss": 0.0427, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 2.5302355187778485, | |
| "grad_norm": 0.6790493726730347, | |
| "learning_rate": 2.228571428571428e-06, | |
| "loss": 0.0423, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 2.546148949713558, | |
| "grad_norm": 0.7205957174301147, | |
| "learning_rate": 8.571428571428572e-08, | |
| "loss": 0.046, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.546148949713558, | |
| "eval_loss": 0.10670246928930283, | |
| "eval_runtime": 3687.1388, | |
| "eval_samples_per_second": 2.265, | |
| "eval_steps_per_second": 0.283, | |
| "eval_wer": 34.549659010739205, | |
| "step": 4000 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 4000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 1000, | |
| "total_flos": 4.15026143428608e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |