| { | |
| "best_metric": 53.673860298812414, | |
| "best_model_checkpoint": "./whisper-medium-te/checkpoint-2000", | |
| "epoch": 2.971768202080238, | |
| "eval_steps": 1000, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03714710252600297, | |
| "grad_norm": 9.450862884521484, | |
| "learning_rate": 4.6000000000000004e-07, | |
| "loss": 1.1494, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.07429420505200594, | |
| "grad_norm": 5.792855262756348, | |
| "learning_rate": 9.600000000000001e-07, | |
| "loss": 0.8665, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.11144130757800892, | |
| "grad_norm": 5.076374530792236, | |
| "learning_rate": 1.46e-06, | |
| "loss": 0.6443, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1485884101040119, | |
| "grad_norm": 4.388276100158691, | |
| "learning_rate": 1.9600000000000003e-06, | |
| "loss": 0.5233, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.18573551263001487, | |
| "grad_norm": 4.2827301025390625, | |
| "learning_rate": 2.46e-06, | |
| "loss": 0.4417, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.22288261515601784, | |
| "grad_norm": 4.33107328414917, | |
| "learning_rate": 2.96e-06, | |
| "loss": 0.3846, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.2600297176820208, | |
| "grad_norm": 5.062885284423828, | |
| "learning_rate": 3.46e-06, | |
| "loss": 0.3456, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.2971768202080238, | |
| "grad_norm": 3.433645725250244, | |
| "learning_rate": 3.96e-06, | |
| "loss": 0.3304, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3343239227340267, | |
| "grad_norm": 4.322815418243408, | |
| "learning_rate": 4.4600000000000005e-06, | |
| "loss": 0.3323, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.37147102526002973, | |
| "grad_norm": 4.137588024139404, | |
| "learning_rate": 4.960000000000001e-06, | |
| "loss": 0.3087, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.4086181277860327, | |
| "grad_norm": 3.188852071762085, | |
| "learning_rate": 5.460000000000001e-06, | |
| "loss": 0.2858, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.4457652303120357, | |
| "grad_norm": 3.801391839981079, | |
| "learning_rate": 5.9600000000000005e-06, | |
| "loss": 0.2804, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.48291233283803864, | |
| "grad_norm": 3.4735641479492188, | |
| "learning_rate": 6.460000000000001e-06, | |
| "loss": 0.2746, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.5200594353640416, | |
| "grad_norm": 4.813545227050781, | |
| "learning_rate": 6.96e-06, | |
| "loss": 0.278, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.5572065378900446, | |
| "grad_norm": 3.2244017124176025, | |
| "learning_rate": 7.4600000000000006e-06, | |
| "loss": 0.2655, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.5943536404160475, | |
| "grad_norm": 3.476280689239502, | |
| "learning_rate": 7.960000000000002e-06, | |
| "loss": 0.2571, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6315007429420505, | |
| "grad_norm": 3.5403590202331543, | |
| "learning_rate": 8.46e-06, | |
| "loss": 0.2522, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.6686478454680534, | |
| "grad_norm": 4.525882244110107, | |
| "learning_rate": 8.96e-06, | |
| "loss": 0.2473, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.7057949479940565, | |
| "grad_norm": 4.107092380523682, | |
| "learning_rate": 9.460000000000001e-06, | |
| "loss": 0.2519, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.7429420505200595, | |
| "grad_norm": 3.5582709312438965, | |
| "learning_rate": 9.960000000000001e-06, | |
| "loss": 0.2508, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.7800891530460624, | |
| "grad_norm": 2.6124589443206787, | |
| "learning_rate": 9.984137931034483e-06, | |
| "loss": 0.2437, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.8172362555720654, | |
| "grad_norm": 3.1363003253936768, | |
| "learning_rate": 9.96689655172414e-06, | |
| "loss": 0.241, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.8543833580980683, | |
| "grad_norm": 2.3347408771514893, | |
| "learning_rate": 9.949655172413793e-06, | |
| "loss": 0.2369, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.8915304606240714, | |
| "grad_norm": 2.609605312347412, | |
| "learning_rate": 9.93241379310345e-06, | |
| "loss": 0.2204, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9286775631500743, | |
| "grad_norm": 3.050915479660034, | |
| "learning_rate": 9.915172413793104e-06, | |
| "loss": 0.2183, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.9658246656760773, | |
| "grad_norm": 2.9425406455993652, | |
| "learning_rate": 9.897931034482759e-06, | |
| "loss": 0.2321, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.0029717682020802, | |
| "grad_norm": 3.074657917022705, | |
| "learning_rate": 9.880689655172414e-06, | |
| "loss": 0.2241, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 1.0401188707280833, | |
| "grad_norm": 2.8157992362976074, | |
| "learning_rate": 9.86344827586207e-06, | |
| "loss": 0.1898, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.0772659732540861, | |
| "grad_norm": 2.4690845012664795, | |
| "learning_rate": 9.846206896551725e-06, | |
| "loss": 0.1857, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 1.1144130757800892, | |
| "grad_norm": 2.4655447006225586, | |
| "learning_rate": 9.82896551724138e-06, | |
| "loss": 0.1743, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.151560178306092, | |
| "grad_norm": 2.103703498840332, | |
| "learning_rate": 9.811724137931035e-06, | |
| "loss": 0.1825, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.188707280832095, | |
| "grad_norm": 2.727170467376709, | |
| "learning_rate": 9.79448275862069e-06, | |
| "loss": 0.1861, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.2258543833580982, | |
| "grad_norm": 2.1180167198181152, | |
| "learning_rate": 9.777241379310347e-06, | |
| "loss": 0.179, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 1.263001485884101, | |
| "grad_norm": 2.076005697250366, | |
| "learning_rate": 9.760000000000001e-06, | |
| "loss": 0.1854, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.300148588410104, | |
| "grad_norm": 1.8460164070129395, | |
| "learning_rate": 9.742758620689656e-06, | |
| "loss": 0.1831, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 1.3372956909361071, | |
| "grad_norm": 4.341026306152344, | |
| "learning_rate": 9.725517241379311e-06, | |
| "loss": 0.177, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.37444279346211, | |
| "grad_norm": 2.86643123626709, | |
| "learning_rate": 9.708275862068966e-06, | |
| "loss": 0.1853, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 1.4115898959881128, | |
| "grad_norm": 2.4118528366088867, | |
| "learning_rate": 9.691034482758621e-06, | |
| "loss": 0.1741, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.4487369985141159, | |
| "grad_norm": 2.506206512451172, | |
| "learning_rate": 9.673793103448277e-06, | |
| "loss": 0.1713, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 1.485884101040119, | |
| "grad_norm": 2.32373309135437, | |
| "learning_rate": 9.65655172413793e-06, | |
| "loss": 0.1662, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.485884101040119, | |
| "eval_loss": 0.2365463376045227, | |
| "eval_runtime": 3136.2768, | |
| "eval_samples_per_second": 0.863, | |
| "eval_steps_per_second": 0.108, | |
| "eval_wer": 55.854935512705914, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.5230312035661218, | |
| "grad_norm": 2.9691216945648193, | |
| "learning_rate": 9.639310344827587e-06, | |
| "loss": 0.1792, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 1.5601783060921248, | |
| "grad_norm": 2.158869981765747, | |
| "learning_rate": 9.622068965517242e-06, | |
| "loss": 0.1654, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.5973254086181279, | |
| "grad_norm": 2.132694721221924, | |
| "learning_rate": 9.604827586206897e-06, | |
| "loss": 0.1805, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 1.6344725111441307, | |
| "grad_norm": 2.4110560417175293, | |
| "learning_rate": 9.587586206896554e-06, | |
| "loss": 0.1688, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.6716196136701336, | |
| "grad_norm": 3.177272319793701, | |
| "learning_rate": 9.570344827586208e-06, | |
| "loss": 0.1746, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 1.7087667161961368, | |
| "grad_norm": 2.058962821960449, | |
| "learning_rate": 9.553103448275863e-06, | |
| "loss": 0.1695, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.7459138187221397, | |
| "grad_norm": 2.6047332286834717, | |
| "learning_rate": 9.535862068965518e-06, | |
| "loss": 0.1767, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 1.7830609212481425, | |
| "grad_norm": 2.676912546157837, | |
| "learning_rate": 9.518620689655173e-06, | |
| "loss": 0.186, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.8202080237741456, | |
| "grad_norm": 2.346376895904541, | |
| "learning_rate": 9.501379310344828e-06, | |
| "loss": 0.1832, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 1.8573551263001487, | |
| "grad_norm": 2.172449827194214, | |
| "learning_rate": 9.484137931034484e-06, | |
| "loss": 0.1791, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.8945022288261515, | |
| "grad_norm": 2.105217933654785, | |
| "learning_rate": 9.46689655172414e-06, | |
| "loss": 0.1771, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 1.9316493313521546, | |
| "grad_norm": 2.7176296710968018, | |
| "learning_rate": 9.449655172413794e-06, | |
| "loss": 0.1711, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.9687964338781576, | |
| "grad_norm": 2.633023262023926, | |
| "learning_rate": 9.432413793103449e-06, | |
| "loss": 0.1717, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 2.0059435364041605, | |
| "grad_norm": 1.6349059343338013, | |
| "learning_rate": 9.415172413793104e-06, | |
| "loss": 0.1649, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.0430906389301633, | |
| "grad_norm": 2.0442230701446533, | |
| "learning_rate": 9.397931034482759e-06, | |
| "loss": 0.1204, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 2.0802377414561666, | |
| "grad_norm": 1.8747773170471191, | |
| "learning_rate": 9.380689655172415e-06, | |
| "loss": 0.1214, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.1173848439821694, | |
| "grad_norm": 2.2875783443450928, | |
| "learning_rate": 9.363448275862069e-06, | |
| "loss": 0.1344, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 2.1545319465081723, | |
| "grad_norm": 1.9838837385177612, | |
| "learning_rate": 9.346206896551725e-06, | |
| "loss": 0.1199, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 2.1916790490341755, | |
| "grad_norm": 2.1067237854003906, | |
| "learning_rate": 9.32896551724138e-06, | |
| "loss": 0.1181, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 2.2288261515601784, | |
| "grad_norm": 2.6312592029571533, | |
| "learning_rate": 9.311724137931035e-06, | |
| "loss": 0.1256, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.265973254086181, | |
| "grad_norm": 1.992762804031372, | |
| "learning_rate": 9.294482758620691e-06, | |
| "loss": 0.1272, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 2.303120356612184, | |
| "grad_norm": 2.416465997695923, | |
| "learning_rate": 9.277241379310346e-06, | |
| "loss": 0.1305, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 2.3402674591381873, | |
| "grad_norm": 1.7433266639709473, | |
| "learning_rate": 9.260000000000001e-06, | |
| "loss": 0.1262, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 2.37741456166419, | |
| "grad_norm": 1.751561164855957, | |
| "learning_rate": 9.242758620689656e-06, | |
| "loss": 0.1216, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.414561664190193, | |
| "grad_norm": 2.123859405517578, | |
| "learning_rate": 9.225517241379311e-06, | |
| "loss": 0.1366, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 2.4517087667161963, | |
| "grad_norm": 1.9784679412841797, | |
| "learning_rate": 9.208275862068966e-06, | |
| "loss": 0.1265, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 2.488855869242199, | |
| "grad_norm": 2.4382882118225098, | |
| "learning_rate": 9.191034482758622e-06, | |
| "loss": 0.1228, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 2.526002971768202, | |
| "grad_norm": 2.2365972995758057, | |
| "learning_rate": 9.173793103448277e-06, | |
| "loss": 0.1326, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.563150074294205, | |
| "grad_norm": 1.7844051122665405, | |
| "learning_rate": 9.156551724137932e-06, | |
| "loss": 0.1234, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 2.600297176820208, | |
| "grad_norm": 1.6666020154953003, | |
| "learning_rate": 9.139310344827587e-06, | |
| "loss": 0.1343, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 2.637444279346211, | |
| "grad_norm": 2.0560336112976074, | |
| "learning_rate": 9.122068965517242e-06, | |
| "loss": 0.1235, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 2.6745913818722142, | |
| "grad_norm": 3.1412951946258545, | |
| "learning_rate": 9.104827586206897e-06, | |
| "loss": 0.1274, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.711738484398217, | |
| "grad_norm": 1.8942031860351562, | |
| "learning_rate": 9.087586206896553e-06, | |
| "loss": 0.1285, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 2.74888558692422, | |
| "grad_norm": 1.8972444534301758, | |
| "learning_rate": 9.070344827586206e-06, | |
| "loss": 0.1335, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 2.7860326894502228, | |
| "grad_norm": 1.9524787664413452, | |
| "learning_rate": 9.053103448275863e-06, | |
| "loss": 0.1209, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 2.8231797919762256, | |
| "grad_norm": 1.6888232231140137, | |
| "learning_rate": 9.035862068965518e-06, | |
| "loss": 0.1361, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.860326894502229, | |
| "grad_norm": 2.2144405841827393, | |
| "learning_rate": 9.018620689655173e-06, | |
| "loss": 0.134, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 2.8974739970282317, | |
| "grad_norm": 2.0730419158935547, | |
| "learning_rate": 9.00137931034483e-06, | |
| "loss": 0.1282, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 2.934621099554235, | |
| "grad_norm": 1.705673098564148, | |
| "learning_rate": 8.984137931034484e-06, | |
| "loss": 0.1206, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 2.971768202080238, | |
| "grad_norm": 1.9598541259765625, | |
| "learning_rate": 8.966896551724139e-06, | |
| "loss": 0.1253, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.971768202080238, | |
| "eval_loss": 0.21802061796188354, | |
| "eval_runtime": 3105.4135, | |
| "eval_samples_per_second": 0.872, | |
| "eval_steps_per_second": 0.109, | |
| "eval_wer": 53.673860298812414, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 15000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 23, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.265527462100992e+19, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |