| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 29.512554112554113, |
| "global_step": 8500, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.17, |
| "learning_rate": 8.333333333333334e-06, |
| "loss": 8.4908, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.35, |
| "learning_rate": 1.6666666666666667e-05, |
| "loss": 8.1905, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.35, |
| "eval_loss": 771.380126953125, |
| "eval_runtime": 237.9275, |
| "eval_samples_per_second": 7.061, |
| "eval_wer": 1.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.52, |
| "learning_rate": 2.5e-05, |
| "loss": 6.0833, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.69, |
| "learning_rate": 3.3333333333333335e-05, |
| "loss": 4.2899, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.69, |
| "eval_loss": 315.0731506347656, |
| "eval_runtime": 181.0423, |
| "eval_samples_per_second": 9.28, |
| "eval_wer": 1.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.87, |
| "learning_rate": 4.1666666666666665e-05, |
| "loss": 3.5966, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.04, |
| "learning_rate": 5e-05, |
| "loss": 3.2533, |
| "step": 300 |
| }, |
| { |
| "epoch": 1.04, |
| "eval_loss": 284.96343994140625, |
| "eval_runtime": 179.8422, |
| "eval_samples_per_second": 9.342, |
| "eval_wer": 0.9928532160527762, |
| "step": 300 |
| }, |
| { |
| "epoch": 1.21, |
| "learning_rate": 5.833333333333333e-05, |
| "loss": 3.0363, |
| "step": 350 |
| }, |
| { |
| "epoch": 1.39, |
| "learning_rate": 6.666666666666667e-05, |
| "loss": 3.0043, |
| "step": 400 |
| }, |
| { |
| "epoch": 1.39, |
| "eval_loss": 280.4071044921875, |
| "eval_runtime": 180.8047, |
| "eval_samples_per_second": 9.292, |
| "eval_wer": 1.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 1.56, |
| "learning_rate": 7.5e-05, |
| "loss": 2.9767, |
| "step": 450 |
| }, |
| { |
| "epoch": 1.73, |
| "learning_rate": 8.333333333333333e-05, |
| "loss": 2.9848, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.73, |
| "eval_loss": 279.59625244140625, |
| "eval_runtime": 179.9553, |
| "eval_samples_per_second": 9.336, |
| "eval_wer": 1.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.91, |
| "learning_rate": 9.166666666666667e-05, |
| "loss": 2.967, |
| "step": 550 |
| }, |
| { |
| "epoch": 2.08, |
| "learning_rate": 0.0001, |
| "loss": 3.0088, |
| "step": 600 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_loss": 276.5469665527344, |
| "eval_runtime": 181.9549, |
| "eval_samples_per_second": 9.233, |
| "eval_wer": 1.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 2.26, |
| "learning_rate": 0.00010833333333333334, |
| "loss": 2.945, |
| "step": 650 |
| }, |
| { |
| "epoch": 2.43, |
| "learning_rate": 0.00011666666666666667, |
| "loss": 2.9085, |
| "step": 700 |
| }, |
| { |
| "epoch": 2.43, |
| "eval_loss": 261.8099365234375, |
| "eval_runtime": 182.2193, |
| "eval_samples_per_second": 9.22, |
| "eval_wer": 1.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 2.6, |
| "learning_rate": 0.000125, |
| "loss": 2.6998, |
| "step": 750 |
| }, |
| { |
| "epoch": 2.78, |
| "learning_rate": 0.00013333333333333334, |
| "loss": 2.0025, |
| "step": 800 |
| }, |
| { |
| "epoch": 2.78, |
| "eval_loss": 113.09307098388672, |
| "eval_runtime": 182.7306, |
| "eval_samples_per_second": 9.194, |
| "eval_wer": 0.919392523364486, |
| "step": 800 |
| }, |
| { |
| "epoch": 2.95, |
| "learning_rate": 0.00014166666666666668, |
| "loss": 1.3056, |
| "step": 850 |
| }, |
| { |
| "epoch": 3.12, |
| "learning_rate": 0.00015, |
| "loss": 0.9594, |
| "step": 900 |
| }, |
| { |
| "epoch": 3.12, |
| "eval_loss": 48.241703033447266, |
| "eval_runtime": 182.3931, |
| "eval_samples_per_second": 9.211, |
| "eval_wer": 0.44193238042880706, |
| "step": 900 |
| }, |
| { |
| "epoch": 3.3, |
| "learning_rate": 0.00015833333333333332, |
| "loss": 0.7557, |
| "step": 950 |
| }, |
| { |
| "epoch": 3.47, |
| "learning_rate": 0.00016666666666666666, |
| "loss": 0.6721, |
| "step": 1000 |
| }, |
| { |
| "epoch": 3.47, |
| "eval_loss": 36.373294830322266, |
| "eval_runtime": 182.5274, |
| "eval_samples_per_second": 9.204, |
| "eval_wer": 0.3460005497526113, |
| "step": 1000 |
| }, |
| { |
| "epoch": 3.64, |
| "learning_rate": 0.000175, |
| "loss": 0.5788, |
| "step": 1050 |
| }, |
| { |
| "epoch": 3.82, |
| "learning_rate": 0.00018333333333333334, |
| "loss": 0.5235, |
| "step": 1100 |
| }, |
| { |
| "epoch": 3.82, |
| "eval_loss": 29.015974044799805, |
| "eval_runtime": 182.315, |
| "eval_samples_per_second": 9.215, |
| "eval_wer": 0.2875893347993403, |
| "step": 1100 |
| }, |
| { |
| "epoch": 3.99, |
| "learning_rate": 0.00019166666666666667, |
| "loss": 0.5232, |
| "step": 1150 |
| }, |
| { |
| "epoch": 4.17, |
| "learning_rate": 0.0002, |
| "loss": 0.4547, |
| "step": 1200 |
| }, |
| { |
| "epoch": 4.17, |
| "eval_loss": 26.849903106689453, |
| "eval_runtime": 181.8627, |
| "eval_samples_per_second": 9.238, |
| "eval_wer": 0.26580538757559097, |
| "step": 1200 |
| }, |
| { |
| "epoch": 4.34, |
| "learning_rate": 0.00020833333333333335, |
| "loss": 0.4091, |
| "step": 1250 |
| }, |
| { |
| "epoch": 4.51, |
| "learning_rate": 0.00021666666666666668, |
| "loss": 0.4003, |
| "step": 1300 |
| }, |
| { |
| "epoch": 4.51, |
| "eval_loss": 24.326810836791992, |
| "eval_runtime": 180.904, |
| "eval_samples_per_second": 9.287, |
| "eval_wer": 0.23735568993952721, |
| "step": 1300 |
| }, |
| { |
| "epoch": 4.69, |
| "learning_rate": 0.00022500000000000002, |
| "loss": 0.3822, |
| "step": 1350 |
| }, |
| { |
| "epoch": 4.86, |
| "learning_rate": 0.00023333333333333333, |
| "loss": 0.3476, |
| "step": 1400 |
| }, |
| { |
| "epoch": 4.86, |
| "eval_loss": 22.954877853393555, |
| "eval_runtime": 182.9264, |
| "eval_samples_per_second": 9.184, |
| "eval_wer": 0.22161902144035184, |
| "step": 1400 |
| }, |
| { |
| "epoch": 5.03, |
| "learning_rate": 0.00024166666666666667, |
| "loss": 0.3594, |
| "step": 1450 |
| }, |
| { |
| "epoch": 5.21, |
| "learning_rate": 0.00025, |
| "loss": 0.3125, |
| "step": 1500 |
| }, |
| { |
| "epoch": 5.21, |
| "eval_loss": 22.298479080200195, |
| "eval_runtime": 181.7168, |
| "eval_samples_per_second": 9.245, |
| "eval_wer": 0.20794392523364486, |
| "step": 1500 |
| }, |
| { |
| "epoch": 5.38, |
| "learning_rate": 0.00025833333333333334, |
| "loss": 0.2878, |
| "step": 1550 |
| }, |
| { |
| "epoch": 5.55, |
| "learning_rate": 0.0002666666666666667, |
| "loss": 0.2762, |
| "step": 1600 |
| }, |
| { |
| "epoch": 5.55, |
| "eval_loss": 23.098875045776367, |
| "eval_runtime": 183.2329, |
| "eval_samples_per_second": 9.169, |
| "eval_wer": 0.21247938427707533, |
| "step": 1600 |
| }, |
| { |
| "epoch": 5.73, |
| "learning_rate": 0.000275, |
| "loss": 0.2703, |
| "step": 1650 |
| }, |
| { |
| "epoch": 5.9, |
| "learning_rate": 0.00028333333333333335, |
| "loss": 0.2827, |
| "step": 1700 |
| }, |
| { |
| "epoch": 5.9, |
| "eval_loss": 21.197721481323242, |
| "eval_runtime": 182.7196, |
| "eval_samples_per_second": 9.194, |
| "eval_wer": 0.19529961517317207, |
| "step": 1700 |
| }, |
| { |
| "epoch": 6.08, |
| "learning_rate": 0.0002916666666666667, |
| "loss": 0.2624, |
| "step": 1750 |
| }, |
| { |
| "epoch": 6.25, |
| "learning_rate": 0.0003, |
| "loss": 0.2306, |
| "step": 1800 |
| }, |
| { |
| "epoch": 6.25, |
| "eval_loss": 22.732194900512695, |
| "eval_runtime": 181.3273, |
| "eval_samples_per_second": 9.265, |
| "eval_wer": 0.18973336998350743, |
| "step": 1800 |
| }, |
| { |
| "epoch": 6.42, |
| "learning_rate": 0.00030833333333333337, |
| "loss": 0.2384, |
| "step": 1850 |
| }, |
| { |
| "epoch": 6.6, |
| "learning_rate": 0.00031666666666666665, |
| "loss": 0.2362, |
| "step": 1900 |
| }, |
| { |
| "epoch": 6.6, |
| "eval_loss": 22.076433181762695, |
| "eval_runtime": 181.988, |
| "eval_samples_per_second": 9.231, |
| "eval_wer": 0.18629741616272677, |
| "step": 1900 |
| }, |
| { |
| "epoch": 6.77, |
| "learning_rate": 0.00032500000000000004, |
| "loss": 0.2269, |
| "step": 1950 |
| }, |
| { |
| "epoch": 6.94, |
| "learning_rate": 0.0003333333333333333, |
| "loss": 0.2342, |
| "step": 2000 |
| }, |
| { |
| "epoch": 6.94, |
| "eval_loss": 22.727157592773438, |
| "eval_runtime": 182.2314, |
| "eval_samples_per_second": 9.219, |
| "eval_wer": 0.17983782297965914, |
| "step": 2000 |
| }, |
| { |
| "epoch": 7.12, |
| "learning_rate": 0.00034166666666666666, |
| "loss": 0.2271, |
| "step": 2050 |
| }, |
| { |
| "epoch": 7.29, |
| "learning_rate": 0.00035, |
| "loss": 0.2165, |
| "step": 2100 |
| }, |
| { |
| "epoch": 7.29, |
| "eval_loss": 21.797693252563477, |
| "eval_runtime": 182.6296, |
| "eval_samples_per_second": 9.199, |
| "eval_wer": 0.17468389224848818, |
| "step": 2100 |
| }, |
| { |
| "epoch": 7.46, |
| "learning_rate": 0.00035833333333333333, |
| "loss": 0.2037, |
| "step": 2150 |
| }, |
| { |
| "epoch": 7.64, |
| "learning_rate": 0.00036666666666666667, |
| "loss": 0.2134, |
| "step": 2200 |
| }, |
| { |
| "epoch": 7.64, |
| "eval_loss": 22.025192260742188, |
| "eval_runtime": 182.2586, |
| "eval_samples_per_second": 9.218, |
| "eval_wer": 0.18148708081363388, |
| "step": 2200 |
| }, |
| { |
| "epoch": 7.81, |
| "learning_rate": 0.000375, |
| "loss": 0.2096, |
| "step": 2250 |
| }, |
| { |
| "epoch": 7.98, |
| "learning_rate": 0.00038333333333333334, |
| "loss": 0.2033, |
| "step": 2300 |
| }, |
| { |
| "epoch": 7.98, |
| "eval_loss": 21.954696655273438, |
| "eval_runtime": 182.9541, |
| "eval_samples_per_second": 9.183, |
| "eval_wer": 0.17990654205607476, |
| "step": 2300 |
| }, |
| { |
| "epoch": 8.16, |
| "learning_rate": 0.0003916666666666667, |
| "loss": 0.2029, |
| "step": 2350 |
| }, |
| { |
| "epoch": 8.33, |
| "learning_rate": 0.0004, |
| "loss": 0.1948, |
| "step": 2400 |
| }, |
| { |
| "epoch": 8.33, |
| "eval_loss": 20.202104568481445, |
| "eval_runtime": 181.6845, |
| "eval_samples_per_second": 9.247, |
| "eval_wer": 0.17303463441451347, |
| "step": 2400 |
| }, |
| { |
| "epoch": 8.51, |
| "learning_rate": 0.00040833333333333336, |
| "loss": 0.1828, |
| "step": 2450 |
| }, |
| { |
| "epoch": 8.68, |
| "learning_rate": 0.0004166666666666667, |
| "loss": 0.1876, |
| "step": 2500 |
| }, |
| { |
| "epoch": 8.68, |
| "eval_loss": 23.213821411132812, |
| "eval_runtime": 182.8529, |
| "eval_samples_per_second": 9.188, |
| "eval_wer": 0.1705607476635514, |
| "step": 2500 |
| }, |
| { |
| "epoch": 8.85, |
| "learning_rate": 0.000425, |
| "loss": 0.1721, |
| "step": 2550 |
| }, |
| { |
| "epoch": 9.03, |
| "learning_rate": 0.00043333333333333337, |
| "loss": 0.1774, |
| "step": 2600 |
| }, |
| { |
| "epoch": 9.03, |
| "eval_loss": 24.49812126159668, |
| "eval_runtime": 181.9478, |
| "eval_samples_per_second": 9.233, |
| "eval_wer": 0.174890049477735, |
| "step": 2600 |
| }, |
| { |
| "epoch": 9.2, |
| "learning_rate": 0.00044166666666666665, |
| "loss": 0.1613, |
| "step": 2650 |
| }, |
| { |
| "epoch": 9.37, |
| "learning_rate": 0.00045000000000000004, |
| "loss": 0.183, |
| "step": 2700 |
| }, |
| { |
| "epoch": 9.37, |
| "eval_loss": 25.343666076660156, |
| "eval_runtime": 181.6008, |
| "eval_samples_per_second": 9.251, |
| "eval_wer": 0.1805937328202309, |
| "step": 2700 |
| }, |
| { |
| "epoch": 9.55, |
| "learning_rate": 0.0004583333333333333, |
| "loss": 0.1738, |
| "step": 2750 |
| }, |
| { |
| "epoch": 9.72, |
| "learning_rate": 0.00046666666666666666, |
| "loss": 0.1886, |
| "step": 2800 |
| }, |
| { |
| "epoch": 9.72, |
| "eval_loss": 24.328418731689453, |
| "eval_runtime": 182.0168, |
| "eval_samples_per_second": 9.23, |
| "eval_wer": 0.18519791094007695, |
| "step": 2800 |
| }, |
| { |
| "epoch": 9.89, |
| "learning_rate": 0.000475, |
| "loss": 0.1855, |
| "step": 2850 |
| }, |
| { |
| "epoch": 10.07, |
| "learning_rate": 0.00048333333333333334, |
| "loss": 0.1784, |
| "step": 2900 |
| }, |
| { |
| "epoch": 10.07, |
| "eval_loss": 24.794740676879883, |
| "eval_runtime": 182.2346, |
| "eval_samples_per_second": 9.219, |
| "eval_wer": 0.1759208356239692, |
| "step": 2900 |
| }, |
| { |
| "epoch": 10.24, |
| "learning_rate": 0.0004916666666666666, |
| "loss": 0.1639, |
| "step": 2950 |
| }, |
| { |
| "epoch": 10.42, |
| "learning_rate": 0.0005, |
| "loss": 0.1951, |
| "step": 3000 |
| }, |
| { |
| "epoch": 10.42, |
| "eval_loss": 23.346879959106445, |
| "eval_runtime": 183.0341, |
| "eval_samples_per_second": 9.179, |
| "eval_wer": 0.1854040681693238, |
| "step": 3000 |
| }, |
| { |
| "epoch": 10.59, |
| "learning_rate": 0.0004955673758865248, |
| "loss": 0.1841, |
| "step": 3050 |
| }, |
| { |
| "epoch": 10.76, |
| "learning_rate": 0.0004911347517730497, |
| "loss": 0.1638, |
| "step": 3100 |
| }, |
| { |
| "epoch": 10.76, |
| "eval_loss": 26.021047592163086, |
| "eval_runtime": 182.7843, |
| "eval_samples_per_second": 9.191, |
| "eval_wer": 0.18945849367784498, |
| "step": 3100 |
| }, |
| { |
| "epoch": 10.94, |
| "learning_rate": 0.0004867021276595745, |
| "loss": 0.1699, |
| "step": 3150 |
| }, |
| { |
| "epoch": 11.11, |
| "learning_rate": 0.00048226950354609925, |
| "loss": 0.178, |
| "step": 3200 |
| }, |
| { |
| "epoch": 11.11, |
| "eval_loss": 25.51027488708496, |
| "eval_runtime": 182.9857, |
| "eval_samples_per_second": 9.181, |
| "eval_wer": 0.18499175371083013, |
| "step": 3200 |
| }, |
| { |
| "epoch": 11.28, |
| "learning_rate": 0.0004778368794326241, |
| "loss": 0.1443, |
| "step": 3250 |
| }, |
| { |
| "epoch": 11.46, |
| "learning_rate": 0.00047340425531914893, |
| "loss": 0.1448, |
| "step": 3300 |
| }, |
| { |
| "epoch": 11.46, |
| "eval_loss": 23.841064453125, |
| "eval_runtime": 182.7518, |
| "eval_samples_per_second": 9.193, |
| "eval_wer": 0.17702034084661902, |
| "step": 3300 |
| }, |
| { |
| "epoch": 11.63, |
| "learning_rate": 0.00046897163120567377, |
| "loss": 0.1415, |
| "step": 3350 |
| }, |
| { |
| "epoch": 11.8, |
| "learning_rate": 0.0004645390070921986, |
| "loss": 0.1451, |
| "step": 3400 |
| }, |
| { |
| "epoch": 11.8, |
| "eval_loss": 24.386722564697266, |
| "eval_runtime": 182.0009, |
| "eval_samples_per_second": 9.231, |
| "eval_wer": 0.17440901594282573, |
| "step": 3400 |
| }, |
| { |
| "epoch": 11.98, |
| "learning_rate": 0.00046010638297872344, |
| "loss": 0.1613, |
| "step": 3450 |
| }, |
| { |
| "epoch": 12.15, |
| "learning_rate": 0.0004556737588652483, |
| "loss": 0.1479, |
| "step": 3500 |
| }, |
| { |
| "epoch": 12.15, |
| "eval_loss": 25.48659324645996, |
| "eval_runtime": 183.184, |
| "eval_samples_per_second": 9.171, |
| "eval_wer": 0.18196811434854315, |
| "step": 3500 |
| }, |
| { |
| "epoch": 12.33, |
| "learning_rate": 0.00045124113475177307, |
| "loss": 0.1249, |
| "step": 3550 |
| }, |
| { |
| "epoch": 12.5, |
| "learning_rate": 0.00044680851063829785, |
| "loss": 0.1255, |
| "step": 3600 |
| }, |
| { |
| "epoch": 12.5, |
| "eval_loss": 26.741592407226562, |
| "eval_runtime": 182.0975, |
| "eval_samples_per_second": 9.226, |
| "eval_wer": 0.17447773501924135, |
| "step": 3600 |
| }, |
| { |
| "epoch": 12.67, |
| "learning_rate": 0.0004423758865248227, |
| "loss": 0.1236, |
| "step": 3650 |
| }, |
| { |
| "epoch": 12.85, |
| "learning_rate": 0.0004379432624113475, |
| "loss": 0.1307, |
| "step": 3700 |
| }, |
| { |
| "epoch": 12.85, |
| "eval_loss": 25.000507354736328, |
| "eval_runtime": 182.3463, |
| "eval_samples_per_second": 9.213, |
| "eval_wer": 0.1726910390324354, |
| "step": 3700 |
| }, |
| { |
| "epoch": 13.02, |
| "learning_rate": 0.00043351063829787236, |
| "loss": 0.1504, |
| "step": 3750 |
| }, |
| { |
| "epoch": 13.19, |
| "learning_rate": 0.0004290780141843972, |
| "loss": 0.1185, |
| "step": 3800 |
| }, |
| { |
| "epoch": 13.19, |
| "eval_loss": 24.76068115234375, |
| "eval_runtime": 182.3064, |
| "eval_samples_per_second": 9.215, |
| "eval_wer": 0.1630703683342496, |
| "step": 3800 |
| }, |
| { |
| "epoch": 13.37, |
| "learning_rate": 0.000424645390070922, |
| "loss": 0.1151, |
| "step": 3850 |
| }, |
| { |
| "epoch": 13.54, |
| "learning_rate": 0.0004202127659574468, |
| "loss": 0.1296, |
| "step": 3900 |
| }, |
| { |
| "epoch": 13.54, |
| "eval_loss": 24.10735511779785, |
| "eval_runtime": 183.409, |
| "eval_samples_per_second": 9.16, |
| "eval_wer": 0.16197086311159978, |
| "step": 3900 |
| }, |
| { |
| "epoch": 13.71, |
| "learning_rate": 0.0004157801418439716, |
| "loss": 0.1099, |
| "step": 3950 |
| }, |
| { |
| "epoch": 13.89, |
| "learning_rate": 0.00041134751773049644, |
| "loss": 0.1141, |
| "step": 4000 |
| }, |
| { |
| "epoch": 13.89, |
| "eval_loss": 27.41958236694336, |
| "eval_runtime": 182.5069, |
| "eval_samples_per_second": 9.205, |
| "eval_wer": 0.17647058823529413, |
| "step": 4000 |
| }, |
| { |
| "epoch": 14.06, |
| "learning_rate": 0.0004069148936170213, |
| "loss": 0.1016, |
| "step": 4050 |
| }, |
| { |
| "epoch": 14.24, |
| "learning_rate": 0.0004024822695035461, |
| "loss": 0.1112, |
| "step": 4100 |
| }, |
| { |
| "epoch": 14.24, |
| "eval_loss": 26.094676971435547, |
| "eval_runtime": 182.6788, |
| "eval_samples_per_second": 9.196, |
| "eval_wer": 0.16760582737768004, |
| "step": 4100 |
| }, |
| { |
| "epoch": 14.41, |
| "learning_rate": 0.00039804964539007096, |
| "loss": 0.0982, |
| "step": 4150 |
| }, |
| { |
| "epoch": 14.58, |
| "learning_rate": 0.00039361702127659574, |
| "loss": 0.1021, |
| "step": 4200 |
| }, |
| { |
| "epoch": 14.58, |
| "eval_loss": 27.608373641967773, |
| "eval_runtime": 182.4822, |
| "eval_samples_per_second": 9.206, |
| "eval_wer": 0.1689802089059923, |
| "step": 4200 |
| }, |
| { |
| "epoch": 14.75, |
| "learning_rate": 0.0003891843971631206, |
| "loss": 0.1064, |
| "step": 4250 |
| }, |
| { |
| "epoch": 14.93, |
| "learning_rate": 0.00038475177304964536, |
| "loss": 0.112, |
| "step": 4300 |
| }, |
| { |
| "epoch": 14.93, |
| "eval_loss": 27.2677059173584, |
| "eval_runtime": 182.8419, |
| "eval_samples_per_second": 9.188, |
| "eval_wer": 0.17550852116547552, |
| "step": 4300 |
| }, |
| { |
| "epoch": 15.1, |
| "learning_rate": 0.0003803191489361702, |
| "loss": 0.1037, |
| "step": 4350 |
| }, |
| { |
| "epoch": 15.28, |
| "learning_rate": 0.00037588652482269504, |
| "loss": 0.1002, |
| "step": 4400 |
| }, |
| { |
| "epoch": 15.28, |
| "eval_loss": 25.244844436645508, |
| "eval_runtime": 182.1487, |
| "eval_samples_per_second": 9.223, |
| "eval_wer": 0.17021715228147333, |
| "step": 4400 |
| }, |
| { |
| "epoch": 15.45, |
| "learning_rate": 0.0003714539007092199, |
| "loss": 0.0889, |
| "step": 4450 |
| }, |
| { |
| "epoch": 15.62, |
| "learning_rate": 0.0003670212765957447, |
| "loss": 0.0893, |
| "step": 4500 |
| }, |
| { |
| "epoch": 15.62, |
| "eval_loss": 31.379478454589844, |
| "eval_runtime": 181.9808, |
| "eval_samples_per_second": 9.232, |
| "eval_wer": 0.17145409565695438, |
| "step": 4500 |
| }, |
| { |
| "epoch": 15.8, |
| "learning_rate": 0.0003625886524822695, |
| "loss": 0.0925, |
| "step": 4550 |
| }, |
| { |
| "epoch": 15.97, |
| "learning_rate": 0.00035815602836879434, |
| "loss": 0.0911, |
| "step": 4600 |
| }, |
| { |
| "epoch": 15.97, |
| "eval_loss": 29.340713500976562, |
| "eval_runtime": 182.5822, |
| "eval_samples_per_second": 9.201, |
| "eval_wer": 0.1657504123144585, |
| "step": 4600 |
| }, |
| { |
| "epoch": 16.15, |
| "learning_rate": 0.0003537234042553192, |
| "loss": 0.0945, |
| "step": 4650 |
| }, |
| { |
| "epoch": 16.32, |
| "learning_rate": 0.00034929078014184396, |
| "loss": 0.0798, |
| "step": 4700 |
| }, |
| { |
| "epoch": 16.32, |
| "eval_loss": 27.59283447265625, |
| "eval_runtime": 183.4998, |
| "eval_samples_per_second": 9.155, |
| "eval_wer": 0.16877405167674547, |
| "step": 4700 |
| }, |
| { |
| "epoch": 16.49, |
| "learning_rate": 0.0003448581560283688, |
| "loss": 0.0868, |
| "step": 4750 |
| }, |
| { |
| "epoch": 16.66, |
| "learning_rate": 0.00034042553191489364, |
| "loss": 0.0893, |
| "step": 4800 |
| }, |
| { |
| "epoch": 16.66, |
| "eval_loss": 26.155086517333984, |
| "eval_runtime": 183.3578, |
| "eval_samples_per_second": 9.162, |
| "eval_wer": 0.16877405167674547, |
| "step": 4800 |
| }, |
| { |
| "epoch": 16.84, |
| "learning_rate": 0.0003359929078014184, |
| "loss": 0.0844, |
| "step": 4850 |
| }, |
| { |
| "epoch": 17.01, |
| "learning_rate": 0.00033156028368794326, |
| "loss": 0.088, |
| "step": 4900 |
| }, |
| { |
| "epoch": 17.01, |
| "eval_loss": 27.282428741455078, |
| "eval_runtime": 183.8154, |
| "eval_samples_per_second": 9.14, |
| "eval_wer": 0.16870533260032985, |
| "step": 4900 |
| }, |
| { |
| "epoch": 17.19, |
| "learning_rate": 0.0003271276595744681, |
| "loss": 0.0844, |
| "step": 4950 |
| }, |
| { |
| "epoch": 17.36, |
| "learning_rate": 0.00032269503546099293, |
| "loss": 0.0647, |
| "step": 5000 |
| }, |
| { |
| "epoch": 17.36, |
| "eval_loss": 27.251529693603516, |
| "eval_runtime": 183.0422, |
| "eval_samples_per_second": 9.178, |
| "eval_wer": 0.1663001649257834, |
| "step": 5000 |
| }, |
| { |
| "epoch": 17.53, |
| "learning_rate": 0.0003182624113475177, |
| "loss": 0.0811, |
| "step": 5050 |
| }, |
| { |
| "epoch": 17.71, |
| "learning_rate": 0.00031382978723404256, |
| "loss": 0.0727, |
| "step": 5100 |
| }, |
| { |
| "epoch": 17.71, |
| "eval_loss": 24.84876823425293, |
| "eval_runtime": 182.8979, |
| "eval_samples_per_second": 9.185, |
| "eval_wer": 0.165131940626718, |
| "step": 5100 |
| }, |
| { |
| "epoch": 17.88, |
| "learning_rate": 0.0003093971631205674, |
| "loss": 0.0744, |
| "step": 5150 |
| }, |
| { |
| "epoch": 18.06, |
| "learning_rate": 0.0003049645390070922, |
| "loss": 0.0801, |
| "step": 5200 |
| }, |
| { |
| "epoch": 18.06, |
| "eval_loss": 26.26492691040039, |
| "eval_runtime": 182.8513, |
| "eval_samples_per_second": 9.188, |
| "eval_wer": 0.1648570643210555, |
| "step": 5200 |
| }, |
| { |
| "epoch": 18.23, |
| "learning_rate": 0.000300531914893617, |
| "loss": 0.0669, |
| "step": 5250 |
| }, |
| { |
| "epoch": 18.4, |
| "learning_rate": 0.00029609929078014185, |
| "loss": 0.0613, |
| "step": 5300 |
| }, |
| { |
| "epoch": 18.4, |
| "eval_loss": 25.977203369140625, |
| "eval_runtime": 182.5431, |
| "eval_samples_per_second": 9.203, |
| "eval_wer": 0.16114623419461244, |
| "step": 5300 |
| }, |
| { |
| "epoch": 18.57, |
| "learning_rate": 0.0002916666666666667, |
| "loss": 0.0641, |
| "step": 5350 |
| }, |
| { |
| "epoch": 18.75, |
| "learning_rate": 0.00028723404255319153, |
| "loss": 0.0588, |
| "step": 5400 |
| }, |
| { |
| "epoch": 18.75, |
| "eval_loss": 26.410762786865234, |
| "eval_runtime": 182.6733, |
| "eval_samples_per_second": 9.197, |
| "eval_wer": 0.1667811984606927, |
| "step": 5400 |
| }, |
| { |
| "epoch": 18.92, |
| "learning_rate": 0.0002828014184397163, |
| "loss": 0.0667, |
| "step": 5450 |
| }, |
| { |
| "epoch": 19.1, |
| "learning_rate": 0.00027836879432624115, |
| "loss": 0.0768, |
| "step": 5500 |
| }, |
| { |
| "epoch": 19.1, |
| "eval_loss": 26.26220703125, |
| "eval_runtime": 182.9701, |
| "eval_samples_per_second": 9.182, |
| "eval_wer": 0.15874106652006598, |
| "step": 5500 |
| }, |
| { |
| "epoch": 19.27, |
| "learning_rate": 0.00027393617021276593, |
| "loss": 0.0746, |
| "step": 5550 |
| }, |
| { |
| "epoch": 19.44, |
| "learning_rate": 0.00026950354609929077, |
| "loss": 0.0616, |
| "step": 5600 |
| }, |
| { |
| "epoch": 19.44, |
| "eval_loss": 23.946300506591797, |
| "eval_runtime": 183.2503, |
| "eval_samples_per_second": 9.168, |
| "eval_wer": 0.15681693238042882, |
| "step": 5600 |
| }, |
| { |
| "epoch": 19.62, |
| "learning_rate": 0.0002650709219858156, |
| "loss": 0.061, |
| "step": 5650 |
| }, |
| { |
| "epoch": 19.79, |
| "learning_rate": 0.00026063829787234045, |
| "loss": 0.066, |
| "step": 5700 |
| }, |
| { |
| "epoch": 19.79, |
| "eval_loss": 27.404529571533203, |
| "eval_runtime": 183.1378, |
| "eval_samples_per_second": 9.173, |
| "eval_wer": 0.16004672897196262, |
| "step": 5700 |
| }, |
| { |
| "epoch": 19.96, |
| "learning_rate": 0.0002562056737588653, |
| "loss": 0.0606, |
| "step": 5750 |
| }, |
| { |
| "epoch": 20.14, |
| "learning_rate": 0.00025177304964539007, |
| "loss": 0.0499, |
| "step": 5800 |
| }, |
| { |
| "epoch": 20.14, |
| "eval_loss": 26.04939079284668, |
| "eval_runtime": 182.3343, |
| "eval_samples_per_second": 9.214, |
| "eval_wer": 0.15805387575590985, |
| "step": 5800 |
| }, |
| { |
| "epoch": 20.31, |
| "learning_rate": 0.0002473404255319149, |
| "loss": 0.0635, |
| "step": 5850 |
| }, |
| { |
| "epoch": 20.48, |
| "learning_rate": 0.00024290780141843972, |
| "loss": 0.0576, |
| "step": 5900 |
| }, |
| { |
| "epoch": 20.48, |
| "eval_loss": 25.570714950561523, |
| "eval_runtime": 182.4858, |
| "eval_samples_per_second": 9.206, |
| "eval_wer": 0.1499450247388675, |
| "step": 5900 |
| }, |
| { |
| "epoch": 20.66, |
| "learning_rate": 0.00023847517730496453, |
| "loss": 0.0535, |
| "step": 5950 |
| }, |
| { |
| "epoch": 20.83, |
| "learning_rate": 0.00023404255319148937, |
| "loss": 0.0538, |
| "step": 6000 |
| }, |
| { |
| "epoch": 20.83, |
| "eval_loss": 26.44669532775879, |
| "eval_runtime": 182.7081, |
| "eval_samples_per_second": 9.195, |
| "eval_wer": 0.15276250687190765, |
| "step": 6000 |
| }, |
| { |
| "epoch": 21.01, |
| "learning_rate": 0.0002296099290780142, |
| "loss": 0.06, |
| "step": 6050 |
| }, |
| { |
| "epoch": 21.18, |
| "learning_rate": 0.00022517730496453902, |
| "loss": 0.0547, |
| "step": 6100 |
| }, |
| { |
| "epoch": 21.18, |
| "eval_loss": 26.02959442138672, |
| "eval_runtime": 183.3837, |
| "eval_samples_per_second": 9.161, |
| "eval_wer": 0.15083837273227046, |
| "step": 6100 |
| }, |
| { |
| "epoch": 21.35, |
| "learning_rate": 0.00022074468085106383, |
| "loss": 0.0506, |
| "step": 6150 |
| }, |
| { |
| "epoch": 21.53, |
| "learning_rate": 0.00021631205673758867, |
| "loss": 0.0578, |
| "step": 6200 |
| }, |
| { |
| "epoch": 21.53, |
| "eval_loss": 25.79215431213379, |
| "eval_runtime": 182.8615, |
| "eval_samples_per_second": 9.187, |
| "eval_wer": 0.1515255634964266, |
| "step": 6200 |
| }, |
| { |
| "epoch": 21.7, |
| "learning_rate": 0.00021187943262411348, |
| "loss": 0.0504, |
| "step": 6250 |
| }, |
| { |
| "epoch": 21.87, |
| "learning_rate": 0.0002074468085106383, |
| "loss": 0.0425, |
| "step": 6300 |
| }, |
| { |
| "epoch": 21.87, |
| "eval_loss": 26.764015197753906, |
| "eval_runtime": 182.7224, |
| "eval_samples_per_second": 9.194, |
| "eval_wer": 0.15592358438702583, |
| "step": 6300 |
| }, |
| { |
| "epoch": 22.05, |
| "learning_rate": 0.00020301418439716313, |
| "loss": 0.0509, |
| "step": 6350 |
| }, |
| { |
| "epoch": 22.22, |
| "learning_rate": 0.00019858156028368796, |
| "loss": 0.0473, |
| "step": 6400 |
| }, |
| { |
| "epoch": 22.22, |
| "eval_loss": 26.02473258972168, |
| "eval_runtime": 183.8649, |
| "eval_samples_per_second": 9.137, |
| "eval_wer": 0.15028862012094557, |
| "step": 6400 |
| }, |
| { |
| "epoch": 22.39, |
| "learning_rate": 0.00019414893617021275, |
| "loss": 0.039, |
| "step": 6450 |
| }, |
| { |
| "epoch": 22.57, |
| "learning_rate": 0.00018971631205673758, |
| "loss": 0.0519, |
| "step": 6500 |
| }, |
| { |
| "epoch": 22.57, |
| "eval_loss": 26.268949508666992, |
| "eval_runtime": 183.1205, |
| "eval_samples_per_second": 9.174, |
| "eval_wer": 0.15166300164925783, |
| "step": 6500 |
| }, |
| { |
| "epoch": 22.74, |
| "learning_rate": 0.00018528368794326242, |
| "loss": 0.0533, |
| "step": 6550 |
| }, |
| { |
| "epoch": 22.91, |
| "learning_rate": 0.00018085106382978726, |
| "loss": 0.0412, |
| "step": 6600 |
| }, |
| { |
| "epoch": 22.91, |
| "eval_loss": 26.845340728759766, |
| "eval_runtime": 183.0902, |
| "eval_samples_per_second": 9.176, |
| "eval_wer": 0.15063221550302364, |
| "step": 6600 |
| }, |
| { |
| "epoch": 23.09, |
| "learning_rate": 0.00017641843971631204, |
| "loss": 0.0535, |
| "step": 6650 |
| }, |
| { |
| "epoch": 23.26, |
| "learning_rate": 0.00017198581560283688, |
| "loss": 0.0449, |
| "step": 6700 |
| }, |
| { |
| "epoch": 23.26, |
| "eval_loss": 24.5848445892334, |
| "eval_runtime": 183.5855, |
| "eval_samples_per_second": 9.151, |
| "eval_wer": 0.14898295766904893, |
| "step": 6700 |
| }, |
| { |
| "epoch": 23.44, |
| "learning_rate": 0.00016755319148936172, |
| "loss": 0.0366, |
| "step": 6750 |
| }, |
| { |
| "epoch": 23.61, |
| "learning_rate": 0.00016312056737588653, |
| "loss": 0.0419, |
| "step": 6800 |
| }, |
| { |
| "epoch": 23.61, |
| "eval_loss": 26.725446701049805, |
| "eval_runtime": 182.6943, |
| "eval_samples_per_second": 9.196, |
| "eval_wer": 0.15063221550302364, |
| "step": 6800 |
| }, |
| { |
| "epoch": 23.78, |
| "learning_rate": 0.00015868794326241134, |
| "loss": 0.0404, |
| "step": 6850 |
| }, |
| { |
| "epoch": 23.96, |
| "learning_rate": 0.00015425531914893618, |
| "loss": 0.0417, |
| "step": 6900 |
| }, |
| { |
| "epoch": 23.96, |
| "eval_loss": 26.662160873413086, |
| "eval_runtime": 182.9944, |
| "eval_samples_per_second": 9.181, |
| "eval_wer": 0.14444749862561848, |
| "step": 6900 |
| }, |
| { |
| "epoch": 24.13, |
| "learning_rate": 0.000149822695035461, |
| "loss": 0.0341, |
| "step": 6950 |
| }, |
| { |
| "epoch": 24.3, |
| "learning_rate": 0.0001453900709219858, |
| "loss": 0.0383, |
| "step": 7000 |
| }, |
| { |
| "epoch": 24.3, |
| "eval_loss": 25.917531967163086, |
| "eval_runtime": 182.9271, |
| "eval_samples_per_second": 9.184, |
| "eval_wer": 0.14699010445299615, |
| "step": 7000 |
| }, |
| { |
| "epoch": 24.48, |
| "learning_rate": 0.00014095744680851064, |
| "loss": 0.0361, |
| "step": 7050 |
| }, |
| { |
| "epoch": 24.65, |
| "learning_rate": 0.00013652482269503548, |
| "loss": 0.037, |
| "step": 7100 |
| }, |
| { |
| "epoch": 24.65, |
| "eval_loss": 26.451175689697266, |
| "eval_runtime": 182.8419, |
| "eval_samples_per_second": 9.188, |
| "eval_wer": 0.14314183617372184, |
| "step": 7100 |
| }, |
| { |
| "epoch": 24.82, |
| "learning_rate": 0.0001320921985815603, |
| "loss": 0.0358, |
| "step": 7150 |
| }, |
| { |
| "epoch": 25.0, |
| "learning_rate": 0.0001276595744680851, |
| "loss": 0.0421, |
| "step": 7200 |
| }, |
| { |
| "epoch": 25.0, |
| "eval_loss": 24.637380599975586, |
| "eval_runtime": 182.468, |
| "eval_samples_per_second": 9.207, |
| "eval_wer": 0.14307311709730622, |
| "step": 7200 |
| }, |
| { |
| "epoch": 25.17, |
| "learning_rate": 0.00012322695035460994, |
| "loss": 0.034, |
| "step": 7250 |
| }, |
| { |
| "epoch": 25.35, |
| "learning_rate": 0.00011879432624113476, |
| "loss": 0.0366, |
| "step": 7300 |
| }, |
| { |
| "epoch": 25.35, |
| "eval_loss": 25.25677490234375, |
| "eval_runtime": 183.0484, |
| "eval_samples_per_second": 9.178, |
| "eval_wer": 0.14369158878504673, |
| "step": 7300 |
| }, |
| { |
| "epoch": 25.52, |
| "learning_rate": 0.00011436170212765957, |
| "loss": 0.0358, |
| "step": 7350 |
| }, |
| { |
| "epoch": 25.69, |
| "learning_rate": 0.00010992907801418441, |
| "loss": 0.0411, |
| "step": 7400 |
| }, |
| { |
| "epoch": 25.69, |
| "eval_loss": 23.710235595703125, |
| "eval_runtime": 183.8494, |
| "eval_samples_per_second": 9.138, |
| "eval_wer": 0.1435541506322155, |
| "step": 7400 |
| }, |
| { |
| "epoch": 25.87, |
| "learning_rate": 0.00010549645390070922, |
| "loss": 0.0324, |
| "step": 7450 |
| }, |
| { |
| "epoch": 26.04, |
| "learning_rate": 0.00010106382978723403, |
| "loss": 0.0395, |
| "step": 7500 |
| }, |
| { |
| "epoch": 26.04, |
| "eval_loss": 25.433012008666992, |
| "eval_runtime": 183.0989, |
| "eval_samples_per_second": 9.175, |
| "eval_wer": 0.14108026388125344, |
| "step": 7500 |
| }, |
| { |
| "epoch": 26.21, |
| "learning_rate": 9.663120567375887e-05, |
| "loss": 0.0373, |
| "step": 7550 |
| }, |
| { |
| "epoch": 26.39, |
| "learning_rate": 9.219858156028368e-05, |
| "loss": 0.0378, |
| "step": 7600 |
| }, |
| { |
| "epoch": 26.39, |
| "eval_loss": 24.688125610351562, |
| "eval_runtime": 187.9819, |
| "eval_samples_per_second": 8.937, |
| "eval_wer": 0.13963716327652556, |
| "step": 7600 |
| }, |
| { |
| "epoch": 26.56, |
| "learning_rate": 8.776595744680852e-05, |
| "loss": 0.0305, |
| "step": 7650 |
| }, |
| { |
| "epoch": 26.73, |
| "learning_rate": 8.333333333333333e-05, |
| "loss": 0.0295, |
| "step": 7700 |
| }, |
| { |
| "epoch": 26.73, |
| "eval_loss": 24.53731918334961, |
| "eval_runtime": 184.8939, |
| "eval_samples_per_second": 9.086, |
| "eval_wer": 0.14114898295766906, |
| "step": 7700 |
| }, |
| { |
| "epoch": 26.91, |
| "learning_rate": 7.890070921985815e-05, |
| "loss": 0.0391, |
| "step": 7750 |
| }, |
| { |
| "epoch": 27.08, |
| "learning_rate": 7.446808510638298e-05, |
| "loss": 0.0295, |
| "step": 7800 |
| }, |
| { |
| "epoch": 27.08, |
| "eval_loss": 24.03723907470703, |
| "eval_runtime": 183.5479, |
| "eval_samples_per_second": 9.153, |
| "eval_wer": 0.14183617372182517, |
| "step": 7800 |
| }, |
| { |
| "epoch": 27.26, |
| "learning_rate": 7.00354609929078e-05, |
| "loss": 0.0367, |
| "step": 7850 |
| }, |
| { |
| "epoch": 27.43, |
| "learning_rate": 6.560283687943263e-05, |
| "loss": 0.0357, |
| "step": 7900 |
| }, |
| { |
| "epoch": 27.43, |
| "eval_loss": 23.551311492919922, |
| "eval_runtime": 186.1508, |
| "eval_samples_per_second": 9.025, |
| "eval_wer": 0.1392935678944475, |
| "step": 7900 |
| }, |
| { |
| "epoch": 27.6, |
| "learning_rate": 6.117021276595745e-05, |
| "loss": 0.0278, |
| "step": 7950 |
| }, |
| { |
| "epoch": 27.78, |
| "learning_rate": 5.673758865248227e-05, |
| "loss": 0.0292, |
| "step": 8000 |
| }, |
| { |
| "epoch": 27.78, |
| "eval_loss": 26.10912322998047, |
| "eval_runtime": 183.6318, |
| "eval_samples_per_second": 9.149, |
| "eval_wer": 0.1403243540406817, |
| "step": 8000 |
| }, |
| { |
| "epoch": 27.95, |
| "learning_rate": 5.2304964539007095e-05, |
| "loss": 0.0293, |
| "step": 8050 |
| }, |
| { |
| "epoch": 28.12, |
| "learning_rate": 4.787234042553191e-05, |
| "loss": 0.0247, |
| "step": 8100 |
| }, |
| { |
| "epoch": 28.12, |
| "eval_loss": 25.034645080566406, |
| "eval_runtime": 183.9776, |
| "eval_samples_per_second": 9.132, |
| "eval_wer": 0.13688840021990104, |
| "step": 8100 |
| }, |
| { |
| "epoch": 28.3, |
| "learning_rate": 4.343971631205674e-05, |
| "loss": 0.0334, |
| "step": 8150 |
| }, |
| { |
| "epoch": 28.47, |
| "learning_rate": 3.900709219858156e-05, |
| "loss": 0.0238, |
| "step": 8200 |
| }, |
| { |
| "epoch": 28.47, |
| "eval_loss": 24.014360427856445, |
| "eval_runtime": 183.9625, |
| "eval_samples_per_second": 9.132, |
| "eval_wer": 0.1355827377680044, |
| "step": 8200 |
| }, |
| { |
| "epoch": 28.64, |
| "learning_rate": 3.4574468085106386e-05, |
| "loss": 0.0342, |
| "step": 8250 |
| }, |
| { |
| "epoch": 28.82, |
| "learning_rate": 3.0141843971631203e-05, |
| "loss": 0.0278, |
| "step": 8300 |
| }, |
| { |
| "epoch": 28.82, |
| "eval_loss": 24.382017135620117, |
| "eval_runtime": 183.4893, |
| "eval_samples_per_second": 9.156, |
| "eval_wer": 0.13585761407366684, |
| "step": 8300 |
| }, |
| { |
| "epoch": 28.99, |
| "learning_rate": 2.5709219858156028e-05, |
| "loss": 0.0209, |
| "step": 8350 |
| }, |
| { |
| "epoch": 29.17, |
| "learning_rate": 2.1276595744680852e-05, |
| "loss": 0.0289, |
| "step": 8400 |
| }, |
| { |
| "epoch": 29.17, |
| "eval_loss": 25.17869758605957, |
| "eval_runtime": 184.0892, |
| "eval_samples_per_second": 9.126, |
| "eval_wer": 0.13551401869158877, |
| "step": 8400 |
| }, |
| { |
| "epoch": 29.34, |
| "learning_rate": 1.6843971631205673e-05, |
| "loss": 0.0306, |
| "step": 8450 |
| }, |
| { |
| "epoch": 29.51, |
| "learning_rate": 1.2411347517730498e-05, |
| "loss": 0.0238, |
| "step": 8500 |
| }, |
| { |
| "epoch": 29.51, |
| "eval_loss": 25.424150466918945, |
| "eval_runtime": 182.9499, |
| "eval_samples_per_second": 9.183, |
| "eval_wer": 0.1351704233095107, |
| "step": 8500 |
| } |
| ], |
| "max_steps": 8640, |
| "num_train_epochs": 30, |
| "total_flos": 1.2862009633623437e+19, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
