| { | |
| "best_metric": 0.2463931441307068, | |
| "best_model_checkpoint": "./checkpoint-6000", | |
| "epoch": 4.999878498238225, | |
| "global_step": 15430, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 3.7125e-06, | |
| "loss": 7.2804, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 7.4625e-06, | |
| "loss": 2.9939, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.1212499999999998e-05, | |
| "loss": 2.8161, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.49625e-05, | |
| "loss": 1.8175, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.8675e-05, | |
| "loss": 1.3167, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 2.2424999999999996e-05, | |
| "loss": 1.1703, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 2.6174999999999996e-05, | |
| "loss": 1.1128, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 2.9925e-05, | |
| "loss": 1.0763, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 3.3675e-05, | |
| "loss": 1.0522, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 3.7424999999999995e-05, | |
| "loss": 1.0326, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_loss": 0.30916452407836914, | |
| "eval_runtime": 1003.5092, | |
| "eval_samples_per_second": 15.885, | |
| "eval_steps_per_second": 0.994, | |
| "eval_wer": 0.2718170286472251, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.1175e-05, | |
| "loss": 1.0398, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 4.4924999999999994e-05, | |
| "loss": 1.0289, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 4.8675e-05, | |
| "loss": 1.0403, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 5.2424999999999994e-05, | |
| "loss": 1.0439, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 5.6175e-05, | |
| "loss": 1.0455, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 5.9925e-05, | |
| "loss": 1.0567, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 6.36375e-05, | |
| "loss": 1.0687, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 6.738749999999999e-05, | |
| "loss": 1.0601, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 7.11375e-05, | |
| "loss": 1.0657, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 7.48875e-05, | |
| "loss": 1.0828, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_loss": 0.28432372212409973, | |
| "eval_runtime": 1004.4407, | |
| "eval_samples_per_second": 15.871, | |
| "eval_steps_per_second": 0.993, | |
| "eval_wer": 0.26055517279125323, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 7.4463886820551e-05, | |
| "loss": 1.09, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 7.390543559195829e-05, | |
| "loss": 1.0696, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 7.33469843633656e-05, | |
| "loss": 1.0794, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 7.279411764705881e-05, | |
| "loss": 1.0906, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 7.223566641846612e-05, | |
| "loss": 1.0739, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 7.16772151898734e-05, | |
| "loss": 1.0753, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 7.111876396128072e-05, | |
| "loss": 1.0694, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 7.0560312732688e-05, | |
| "loss": 1.0652, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 7.00018615040953e-05, | |
| "loss": 1.0627, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 6.94434102755026e-05, | |
| "loss": 1.0771, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_loss": 0.27740851044654846, | |
| "eval_runtime": 1003.9914, | |
| "eval_samples_per_second": 15.878, | |
| "eval_steps_per_second": 0.993, | |
| "eval_wer": 0.24881682903743005, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 6.88849590469099e-05, | |
| "loss": 1.0548, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 6.832650781831719e-05, | |
| "loss": 1.0594, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 6.776805658972449e-05, | |
| "loss": 1.0605, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 6.720960536113179e-05, | |
| "loss": 1.0501, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 6.665673864482501e-05, | |
| "loss": 1.06, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 6.609828741623231e-05, | |
| "loss": 1.0539, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 6.553983618763961e-05, | |
| "loss": 1.0483, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 6.49813849590469e-05, | |
| "loss": 1.0507, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 6.44229337304542e-05, | |
| "loss": 1.0416, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 6.38644825018615e-05, | |
| "loss": 1.0306, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_loss": 0.25875675678253174, | |
| "eval_runtime": 1004.8682, | |
| "eval_samples_per_second": 15.864, | |
| "eval_steps_per_second": 0.992, | |
| "eval_wer": 0.23513389953832187, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 6.33060312732688e-05, | |
| "loss": 1.0413, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 6.275316455696202e-05, | |
| "loss": 1.0475, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 6.219471332836931e-05, | |
| "loss": 1.0328, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 6.163626209977662e-05, | |
| "loss": 1.036, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 6.10778108711839e-05, | |
| "loss": 1.0248, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 6.051935964259121e-05, | |
| "loss": 1.0184, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 5.996090841399851e-05, | |
| "loss": 1.0268, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 5.94024571854058e-05, | |
| "loss": 1.0177, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 5.88440059568131e-05, | |
| "loss": 1.0155, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 5.82855547282204e-05, | |
| "loss": 1.0052, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "eval_loss": 0.24826321005821228, | |
| "eval_runtime": 1000.3482, | |
| "eval_samples_per_second": 15.935, | |
| "eval_steps_per_second": 0.997, | |
| "eval_wer": 0.22835360553241082, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 5.7727103499627696e-05, | |
| "loss": 1.0076, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 5.717423678332092e-05, | |
| "loss": 1.0107, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 5.661578555472821e-05, | |
| "loss": 1.0135, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 5.605733432613552e-05, | |
| "loss": 1.012, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 5.549888309754281e-05, | |
| "loss": 0.9943, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 5.4940431868950106e-05, | |
| "loss": 1.0049, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 5.4381980640357404e-05, | |
| "loss": 1.0082, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 5.38235294117647e-05, | |
| "loss": 0.9948, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 5.326507818317199e-05, | |
| "loss": 0.9925, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 5.2706626954579296e-05, | |
| "loss": 0.9865, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_loss": 0.2463931441307068, | |
| "eval_runtime": 999.3892, | |
| "eval_samples_per_second": 15.951, | |
| "eval_steps_per_second": 0.998, | |
| "eval_wer": 0.22199184819353135, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 5.2148175725986594e-05, | |
| "loss": 0.987, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 5.158972449739389e-05, | |
| "loss": 0.9789, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 5.103127326880119e-05, | |
| "loss": 0.9722, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 5.047282204020848e-05, | |
| "loss": 0.9747, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 4.9914370811615785e-05, | |
| "loss": 0.9771, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 4.9355919583023076e-05, | |
| "loss": 0.9777, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 4.88030528667163e-05, | |
| "loss": 0.9804, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 4.82446016381236e-05, | |
| "loss": 0.9575, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 4.76861504095309e-05, | |
| "loss": 0.9613, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 4.713328369322412e-05, | |
| "loss": 0.978, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "eval_loss": 0.2513602077960968, | |
| "eval_runtime": 1010.6008, | |
| "eval_samples_per_second": 15.774, | |
| "eval_steps_per_second": 0.987, | |
| "eval_wer": 0.21723340824066503, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 4.658041697691735e-05, | |
| "loss": 0.9837, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 4.602196574832464e-05, | |
| "loss": 1.0137, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 4.546351451973194e-05, | |
| "loss": 1.0971, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 4.4905063291139234e-05, | |
| "loss": 1.1683, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 4.434661206254654e-05, | |
| "loss": 1.2993, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 4.378816083395383e-05, | |
| "loss": 1.3806, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 4.322970960536113e-05, | |
| "loss": 1.4648, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 4.2671258376768425e-05, | |
| "loss": 1.5643, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 4.211280714817572e-05, | |
| "loss": 1.5596, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 4.155435591958301e-05, | |
| "loss": 1.7438, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "eval_loss": 0.7983415722846985, | |
| "eval_runtime": 1011.5068, | |
| "eval_samples_per_second": 15.76, | |
| "eval_steps_per_second": 0.986, | |
| "eval_wer": 0.5072471233652923, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 4.099590469099032e-05, | |
| "loss": 1.7779, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 4.0437453462397615e-05, | |
| "loss": 1.9193, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 3.987900223380491e-05, | |
| "loss": 1.8549, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 3.932055100521221e-05, | |
| "loss": 1.9323, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 3.87620997766195e-05, | |
| "loss": 2.0832, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 3.8203648548026806e-05, | |
| "loss": 2.268, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 3.76451973194341e-05, | |
| "loss": 1.9962, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 3.7086746090841395e-05, | |
| "loss": 2.2105, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 3.65282948622487e-05, | |
| "loss": 2.3688, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 3.596984363365599e-05, | |
| "loss": 2.3309, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "eval_loss": 1.8916891813278198, | |
| "eval_runtime": 1006.0568, | |
| "eval_samples_per_second": 15.845, | |
| "eval_steps_per_second": 0.991, | |
| "eval_wer": 0.9416173544619229, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 3.541139240506329e-05, | |
| "loss": 2.2688, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 3.4852941176470585e-05, | |
| "loss": 2.0991, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 3.430007446016381e-05, | |
| "loss": 2.0663, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 3.374162323157111e-05, | |
| "loss": 2.0995, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 3.3183172002978406e-05, | |
| "loss": 2.2691, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 3.26247207743857e-05, | |
| "loss": 2.1691, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 3.2066269545792995e-05, | |
| "loss": 2.2706, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 3.150781831720029e-05, | |
| "loss": 2.0075, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 3.094936708860759e-05, | |
| "loss": 2.1615, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 3.0390915860014888e-05, | |
| "loss": 2.1834, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "eval_loss": 1.7495689392089844, | |
| "eval_runtime": 1013.0296, | |
| "eval_samples_per_second": 15.736, | |
| "eval_steps_per_second": 0.984, | |
| "eval_wer": 0.9029832005820879, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 2.983246463142219e-05, | |
| "loss": 2.2615, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 2.9274013402829486e-05, | |
| "loss": 2.387, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 2.871556217423678e-05, | |
| "loss": 2.3751, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 2.8157110945644078e-05, | |
| "loss": 2.3223, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 2.7604244229337304e-05, | |
| "loss": 2.2708, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 2.7045793000744598e-05, | |
| "loss": 2.2369, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 2.6487341772151896e-05, | |
| "loss": 2.3768, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 2.5928890543559194e-05, | |
| "loss": 2.2929, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 2.5376023827252416e-05, | |
| "loss": 2.3779, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 2.4817572598659713e-05, | |
| "loss": 2.3047, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "eval_loss": 1.5376594066619873, | |
| "eval_runtime": 1009.9366, | |
| "eval_samples_per_second": 15.784, | |
| "eval_steps_per_second": 0.987, | |
| "eval_wer": 0.874715877993342, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 2.4259121370067014e-05, | |
| "loss": 2.0797, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 2.3700670141474312e-05, | |
| "loss": 2.1414, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 2.3142218912881606e-05, | |
| "loss": 2.1064, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 2.2583767684288904e-05, | |
| "loss": 2.0681, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 2.2025316455696202e-05, | |
| "loss": 2.0853, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 2.14668652271035e-05, | |
| "loss": 2.1272, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 2.0908413998510794e-05, | |
| "loss": 2.1257, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 2.034996276991809e-05, | |
| "loss": 2.0966, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 1.979151154132539e-05, | |
| "loss": 2.2297, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 1.9233060312732687e-05, | |
| "loss": 2.1378, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "eval_loss": 1.3500523567199707, | |
| "eval_runtime": 1013.8007, | |
| "eval_samples_per_second": 15.724, | |
| "eval_steps_per_second": 0.983, | |
| "eval_wer": 0.7923285448445941, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 1.8680193596425912e-05, | |
| "loss": 2.2683, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 1.8127326880119135e-05, | |
| "loss": 2.311, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 1.7568875651526432e-05, | |
| "loss": 2.0594, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 1.701042442293373e-05, | |
| "loss": 1.9798, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 1.6457557706626952e-05, | |
| "loss": 1.8521, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 1.589910647803425e-05, | |
| "loss": 1.8466, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 1.5340655249441547e-05, | |
| "loss": 1.9372, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 1.4782204020848843e-05, | |
| "loss": 1.9591, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "learning_rate": 1.4223752792256143e-05, | |
| "loss": 1.9641, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 1.3670886075949367e-05, | |
| "loss": 1.9812, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "eval_loss": 1.266192078590393, | |
| "eval_runtime": 1008.4278, | |
| "eval_samples_per_second": 15.808, | |
| "eval_steps_per_second": 0.989, | |
| "eval_wer": 0.7696760526197176, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 1.3112434847356663e-05, | |
| "loss": 2.1185, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 1.255398361876396e-05, | |
| "loss": 2.2783, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 1.2001116902457184e-05, | |
| "loss": 2.3792, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 1.1442665673864482e-05, | |
| "loss": 2.4678, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 1.0884214445271778e-05, | |
| "loss": 2.5792, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 1.0325763216679075e-05, | |
| "loss": 2.6183, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 9.767311988086373e-06, | |
| "loss": 2.6677, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 9.20886075949367e-06, | |
| "loss": 2.7043, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 8.650409530900967e-06, | |
| "loss": 2.6737, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "learning_rate": 8.091958302308264e-06, | |
| "loss": 2.6855, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "eval_loss": 2.4120402336120605, | |
| "eval_runtime": 1002.3228, | |
| "eval_samples_per_second": 15.904, | |
| "eval_steps_per_second": 0.995, | |
| "eval_wer": 0.9901740468632286, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 7.533507073715562e-06, | |
| "loss": 2.7223, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "learning_rate": 6.975055845122859e-06, | |
| "loss": 2.7861, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 6.416604616530156e-06, | |
| "loss": 2.7397, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 5.8581533879374525e-06, | |
| "loss": 2.7574, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 5.3052866716306765e-06, | |
| "loss": 2.8119, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 4.746835443037975e-06, | |
| "loss": 2.725, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 4.188384214445272e-06, | |
| "loss": 2.7319, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 3.629932985852568e-06, | |
| "loss": 2.7426, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 3.071481757259866e-06, | |
| "loss": 2.7441, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "learning_rate": 2.513030528667163e-06, | |
| "loss": 2.7482, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "eval_loss": 2.5340933799743652, | |
| "eval_runtime": 1000.6978, | |
| "eval_samples_per_second": 15.93, | |
| "eval_steps_per_second": 0.996, | |
| "eval_wer": 0.9874181438864671, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 1.95457930007446e-06, | |
| "loss": 2.752, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 1.3961280714817571e-06, | |
| "loss": 2.7476, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 8.376768428890544e-07, | |
| "loss": 2.7484, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "learning_rate": 2.7922561429635143e-07, | |
| "loss": 2.7494, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 15430, | |
| "total_flos": 8.828199178194905e+20, | |
| "train_loss": 1.7380384819737829, | |
| "train_runtime": 214447.5226, | |
| "train_samples_per_second": 9.211, | |
| "train_steps_per_second": 0.072 | |
| } | |
| ], | |
| "max_steps": 15430, | |
| "num_train_epochs": 5, | |
| "total_flos": 8.828199178194905e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |