| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 49.999556933983165, | |
| "global_step": 28200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 3.7125e-06, | |
| "loss": 9.6392, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 7.4625e-06, | |
| "loss": 5.3472, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 1.1212499999999998e-05, | |
| "loss": 3.7052, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.49625e-05, | |
| "loss": 3.35, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.8712499999999997e-05, | |
| "loss": 3.1484, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "eval_loss": 3.084352731704712, | |
| "eval_runtime": 31.0524, | |
| "eval_samples_per_second": 5.732, | |
| "eval_steps_per_second": 1.932, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 2.2462499999999997e-05, | |
| "loss": 3.0531, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 2.6212499999999997e-05, | |
| "loss": 2.9911, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 2.99625e-05, | |
| "loss": 2.9576, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 3.37125e-05, | |
| "loss": 2.9406, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 3.7462499999999996e-05, | |
| "loss": 2.6539, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "eval_loss": 1.7271525859832764, | |
| "eval_runtime": 30.9027, | |
| "eval_samples_per_second": 5.76, | |
| "eval_steps_per_second": 1.942, | |
| "eval_wer": 0.9358414511252939, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 4.12125e-05, | |
| "loss": 1.5429, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 4.4962499999999995e-05, | |
| "loss": 1.0875, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 4.871249999999999e-05, | |
| "loss": 0.9728, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 5.2424999999999994e-05, | |
| "loss": 0.9117, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 5.6175e-05, | |
| "loss": 0.8732, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "eval_loss": 0.19752363860607147, | |
| "eval_runtime": 31.0609, | |
| "eval_samples_per_second": 5.731, | |
| "eval_steps_per_second": 1.932, | |
| "eval_wer": 0.16090023513604298, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 5.9925e-05, | |
| "loss": 0.8518, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 6.367499999999999e-05, | |
| "loss": 0.8373, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 6.7425e-05, | |
| "loss": 0.7958, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 7.1175e-05, | |
| "loss": 0.8058, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 7.492499999999999e-05, | |
| "loss": 0.8075, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "eval_loss": 0.1483038365840912, | |
| "eval_runtime": 31.0609, | |
| "eval_samples_per_second": 5.731, | |
| "eval_steps_per_second": 1.932, | |
| "eval_wer": 0.1467920725562647, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 7.471946564885497e-05, | |
| "loss": 0.785, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 7.443320610687022e-05, | |
| "loss": 0.7656, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 7.414694656488549e-05, | |
| "loss": 0.7681, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 7.386068702290076e-05, | |
| "loss": 0.7535, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "learning_rate": 7.357442748091603e-05, | |
| "loss": 0.7358, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "eval_loss": 0.1331482231616974, | |
| "eval_runtime": 30.9431, | |
| "eval_samples_per_second": 5.752, | |
| "eval_steps_per_second": 1.939, | |
| "eval_wer": 0.14007389989922742, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "learning_rate": 7.328816793893128e-05, | |
| "loss": 0.7478, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 7.300190839694656e-05, | |
| "loss": 0.746, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 7.271564885496182e-05, | |
| "loss": 0.7545, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "learning_rate": 7.242938931297709e-05, | |
| "loss": 0.7268, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "learning_rate": 7.214312977099236e-05, | |
| "loss": 0.7079, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "eval_loss": 0.12731623649597168, | |
| "eval_runtime": 30.9985, | |
| "eval_samples_per_second": 5.742, | |
| "eval_steps_per_second": 1.936, | |
| "eval_wer": 0.1363789049378569, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "learning_rate": 7.185687022900763e-05, | |
| "loss": 0.7344, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "learning_rate": 7.15706106870229e-05, | |
| "loss": 0.7334, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 7.128435114503815e-05, | |
| "loss": 0.7274, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 7.099809160305343e-05, | |
| "loss": 0.7271, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "learning_rate": 7.071469465648854e-05, | |
| "loss": 0.7032, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "eval_loss": 0.11333052814006805, | |
| "eval_runtime": 31.1935, | |
| "eval_samples_per_second": 5.706, | |
| "eval_steps_per_second": 1.923, | |
| "eval_wer": 0.12395028552233793, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "learning_rate": 7.04284351145038e-05, | |
| "loss": 0.6852, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 7.014217557251907e-05, | |
| "loss": 0.699, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 6.74, | |
| "learning_rate": 6.985591603053435e-05, | |
| "loss": 0.6894, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "learning_rate": 6.956965648854962e-05, | |
| "loss": 0.7014, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "learning_rate": 6.928339694656487e-05, | |
| "loss": 0.7129, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "eval_loss": 0.11242285370826721, | |
| "eval_runtime": 31.0672, | |
| "eval_samples_per_second": 5.73, | |
| "eval_steps_per_second": 1.931, | |
| "eval_wer": 0.1289889150151159, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "learning_rate": 6.899713740458015e-05, | |
| "loss": 0.6902, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 7.45, | |
| "learning_rate": 6.871087786259541e-05, | |
| "loss": 0.6898, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 7.62, | |
| "learning_rate": 6.842461832061069e-05, | |
| "loss": 0.6785, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "learning_rate": 6.813835877862594e-05, | |
| "loss": 0.7024, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 7.98, | |
| "learning_rate": 6.785209923664122e-05, | |
| "loss": 0.6771, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 7.98, | |
| "eval_loss": 0.11212227493524551, | |
| "eval_runtime": 31.0691, | |
| "eval_samples_per_second": 5.729, | |
| "eval_steps_per_second": 1.931, | |
| "eval_wer": 0.12999664091367147, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 8.16, | |
| "learning_rate": 6.756583969465648e-05, | |
| "loss": 0.6555, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 6.727958015267174e-05, | |
| "loss": 0.6897, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 8.51, | |
| "learning_rate": 6.699618320610687e-05, | |
| "loss": 0.6809, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 8.69, | |
| "learning_rate": 6.670992366412213e-05, | |
| "loss": 0.6821, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 8.86, | |
| "learning_rate": 6.642366412213741e-05, | |
| "loss": 0.6859, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 8.86, | |
| "eval_loss": 0.109534852206707, | |
| "eval_runtime": 31.3176, | |
| "eval_samples_per_second": 5.684, | |
| "eval_steps_per_second": 1.916, | |
| "eval_wer": 0.13134027544507895, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 6.613740458015266e-05, | |
| "loss": 0.6849, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 9.22, | |
| "learning_rate": 6.585114503816793e-05, | |
| "loss": 0.6615, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 9.4, | |
| "learning_rate": 6.55648854961832e-05, | |
| "loss": 0.6587, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "learning_rate": 6.527862595419846e-05, | |
| "loss": 0.6741, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 9.75, | |
| "learning_rate": 6.499236641221373e-05, | |
| "loss": 0.6496, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 9.75, | |
| "eval_loss": 0.1090548112988472, | |
| "eval_runtime": 31.0434, | |
| "eval_samples_per_second": 5.734, | |
| "eval_steps_per_second": 1.933, | |
| "eval_wer": 0.12495801142089352, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 9.93, | |
| "learning_rate": 6.4706106870229e-05, | |
| "loss": 0.6672, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 10.11, | |
| "learning_rate": 6.441984732824428e-05, | |
| "loss": 0.6511, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "learning_rate": 6.413645038167938e-05, | |
| "loss": 0.6393, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 10.46, | |
| "learning_rate": 6.385019083969465e-05, | |
| "loss": 0.6531, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 10.64, | |
| "learning_rate": 6.356393129770992e-05, | |
| "loss": 0.6431, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 10.64, | |
| "eval_loss": 0.11021342128515244, | |
| "eval_runtime": 30.9854, | |
| "eval_samples_per_second": 5.745, | |
| "eval_steps_per_second": 1.936, | |
| "eval_wer": 0.12932482364796774, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 10.82, | |
| "learning_rate": 6.32776717557252e-05, | |
| "loss": 0.6476, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 10.99, | |
| "learning_rate": 6.299141221374044e-05, | |
| "loss": 0.6504, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 11.17, | |
| "learning_rate": 6.270515267175572e-05, | |
| "loss": 0.6366, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 11.35, | |
| "learning_rate": 6.241889312977098e-05, | |
| "loss": 0.6435, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 11.52, | |
| "learning_rate": 6.213263358778625e-05, | |
| "loss": 0.6422, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 11.52, | |
| "eval_loss": 0.11072035878896713, | |
| "eval_runtime": 31.126, | |
| "eval_samples_per_second": 5.719, | |
| "eval_steps_per_second": 1.928, | |
| "eval_wer": 0.11790393013100436, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 11.7, | |
| "learning_rate": 6.184637404580152e-05, | |
| "loss": 0.636, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 11.88, | |
| "learning_rate": 6.156011450381679e-05, | |
| "loss": 0.647, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 6.127385496183207e-05, | |
| "loss": 0.644, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 12.23, | |
| "learning_rate": 6.098759541984732e-05, | |
| "loss": 0.611, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "learning_rate": 6.070133587786259e-05, | |
| "loss": 0.6334, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "eval_loss": 0.10494451969861984, | |
| "eval_runtime": 31.1005, | |
| "eval_samples_per_second": 5.723, | |
| "eval_steps_per_second": 1.929, | |
| "eval_wer": 0.12361437688948607, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 12.59, | |
| "learning_rate": 6.0415076335877855e-05, | |
| "loss": 0.6305, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 12.77, | |
| "learning_rate": 6.0128816793893126e-05, | |
| "loss": 0.6292, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 12.94, | |
| "learning_rate": 5.984255725190839e-05, | |
| "loss": 0.6341, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 13.12, | |
| "learning_rate": 5.955629770992366e-05, | |
| "loss": 0.625, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 13.3, | |
| "learning_rate": 5.927003816793893e-05, | |
| "loss": 0.599, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 13.3, | |
| "eval_loss": 0.10921534150838852, | |
| "eval_runtime": 31.0896, | |
| "eval_samples_per_second": 5.725, | |
| "eval_steps_per_second": 1.93, | |
| "eval_wer": 0.11521666106818945, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 13.47, | |
| "learning_rate": 5.898377862595419e-05, | |
| "loss": 0.6144, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 13.65, | |
| "learning_rate": 5.869751908396946e-05, | |
| "loss": 0.6314, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 13.83, | |
| "learning_rate": 5.8411259541984726e-05, | |
| "loss": 0.6185, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 14.01, | |
| "learning_rate": 5.8124999999999997e-05, | |
| "loss": 0.6122, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 14.18, | |
| "learning_rate": 5.783874045801526e-05, | |
| "loss": 0.6205, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 14.18, | |
| "eval_loss": 0.1046827957034111, | |
| "eval_runtime": 31.2229, | |
| "eval_samples_per_second": 5.701, | |
| "eval_steps_per_second": 1.922, | |
| "eval_wer": 0.12193483372522673, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 14.36, | |
| "learning_rate": 5.755248091603053e-05, | |
| "loss": 0.612, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 14.54, | |
| "learning_rate": 5.7266221374045794e-05, | |
| "loss": 0.6196, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 14.72, | |
| "learning_rate": 5.6979961832061064e-05, | |
| "loss": 0.6114, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 14.89, | |
| "learning_rate": 5.6693702290076334e-05, | |
| "loss": 0.6005, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 15.07, | |
| "learning_rate": 5.64074427480916e-05, | |
| "loss": 0.5944, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 15.07, | |
| "eval_loss": 0.10684490948915482, | |
| "eval_runtime": 31.2028, | |
| "eval_samples_per_second": 5.705, | |
| "eval_steps_per_second": 1.923, | |
| "eval_wer": 0.12025529056096741, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 15.25, | |
| "learning_rate": 5.612118320610687e-05, | |
| "loss": 0.6108, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 15.43, | |
| "learning_rate": 5.583492366412213e-05, | |
| "loss": 0.5995, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 15.6, | |
| "learning_rate": 5.55486641221374e-05, | |
| "loss": 0.6117, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 15.78, | |
| "learning_rate": 5.5262404580152665e-05, | |
| "loss": 0.6006, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 15.96, | |
| "learning_rate": 5.4976145038167935e-05, | |
| "loss": 0.6102, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 15.96, | |
| "eval_loss": 0.10560546070337296, | |
| "eval_runtime": 31.2837, | |
| "eval_samples_per_second": 5.69, | |
| "eval_steps_per_second": 1.918, | |
| "eval_wer": 0.11588847833389318, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "learning_rate": 5.46898854961832e-05, | |
| "loss": 0.5968, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 16.31, | |
| "learning_rate": 5.440362595419847e-05, | |
| "loss": 0.5912, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 16.49, | |
| "learning_rate": 5.411736641221374e-05, | |
| "loss": 0.5889, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "learning_rate": 5.3831106870229e-05, | |
| "loss": 0.6014, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 16.84, | |
| "learning_rate": 5.354484732824427e-05, | |
| "loss": 0.5983, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 16.84, | |
| "eval_loss": 0.1061149537563324, | |
| "eval_runtime": 31.4987, | |
| "eval_samples_per_second": 5.651, | |
| "eval_steps_per_second": 1.905, | |
| "eval_wer": 0.11521666106818945, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 17.02, | |
| "learning_rate": 5.3258587786259536e-05, | |
| "loss": 0.5992, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 17.2, | |
| "learning_rate": 5.2972328244274806e-05, | |
| "loss": 0.5887, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 17.38, | |
| "learning_rate": 5.268606870229007e-05, | |
| "loss": 0.6012, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 17.55, | |
| "learning_rate": 5.239980916030534e-05, | |
| "loss": 0.5985, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 17.73, | |
| "learning_rate": 5.211354961832061e-05, | |
| "loss": 0.5882, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 17.73, | |
| "eval_loss": 0.10430345684289932, | |
| "eval_runtime": 30.988, | |
| "eval_samples_per_second": 5.744, | |
| "eval_steps_per_second": 1.936, | |
| "eval_wer": 0.11353711790393013, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 17.91, | |
| "learning_rate": 5.182729007633587e-05, | |
| "loss": 0.5906, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 18.09, | |
| "learning_rate": 5.1541030534351143e-05, | |
| "loss": 0.5843, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 18.26, | |
| "learning_rate": 5.1257633587786254e-05, | |
| "loss": 0.5976, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 18.44, | |
| "learning_rate": 5.0971374045801525e-05, | |
| "loss": 0.59, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "learning_rate": 5.068511450381679e-05, | |
| "loss": 0.5876, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "eval_loss": 0.10231117904186249, | |
| "eval_runtime": 31.0041, | |
| "eval_samples_per_second": 5.741, | |
| "eval_steps_per_second": 1.935, | |
| "eval_wer": 0.11588847833389318, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 18.79, | |
| "learning_rate": 5.039885496183206e-05, | |
| "loss": 0.5845, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "learning_rate": 5.011259541984732e-05, | |
| "loss": 0.5812, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 19.15, | |
| "learning_rate": 4.982633587786259e-05, | |
| "loss": 0.5695, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 19.33, | |
| "learning_rate": 4.954007633587786e-05, | |
| "loss": 0.5665, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 19.5, | |
| "learning_rate": 4.9253816793893125e-05, | |
| "loss": 0.5717, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 19.5, | |
| "eval_loss": 0.10367337614297867, | |
| "eval_runtime": 31.2366, | |
| "eval_samples_per_second": 5.698, | |
| "eval_steps_per_second": 1.921, | |
| "eval_wer": 0.1232784682566342, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 19.68, | |
| "learning_rate": 4.8967557251908396e-05, | |
| "loss": 0.5657, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 19.86, | |
| "learning_rate": 4.868129770992366e-05, | |
| "loss": 0.5733, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 20.04, | |
| "learning_rate": 4.839503816793893e-05, | |
| "loss": 0.5816, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 20.21, | |
| "learning_rate": 4.810877862595419e-05, | |
| "loss": 0.5694, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 20.39, | |
| "learning_rate": 4.782251908396946e-05, | |
| "loss": 0.5537, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 20.39, | |
| "eval_loss": 0.10704291611909866, | |
| "eval_runtime": 31.0837, | |
| "eval_samples_per_second": 5.726, | |
| "eval_steps_per_second": 1.93, | |
| "eval_wer": 0.11924756466241182, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 20.57, | |
| "learning_rate": 4.7536259541984726e-05, | |
| "loss": 0.5726, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 20.74, | |
| "learning_rate": 4.7249999999999997e-05, | |
| "loss": 0.5742, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 20.92, | |
| "learning_rate": 4.696374045801527e-05, | |
| "loss": 0.563, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 21.1, | |
| "learning_rate": 4.667748091603053e-05, | |
| "loss": 0.5655, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 21.28, | |
| "learning_rate": 4.63912213740458e-05, | |
| "loss": 0.5636, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 21.28, | |
| "eval_loss": 0.10363561660051346, | |
| "eval_runtime": 31.1987, | |
| "eval_samples_per_second": 5.705, | |
| "eval_steps_per_second": 1.923, | |
| "eval_wer": 0.11689620423244877, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 21.45, | |
| "learning_rate": 4.6104961832061064e-05, | |
| "loss": 0.5612, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 21.63, | |
| "learning_rate": 4.5818702290076334e-05, | |
| "loss": 0.5732, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 21.81, | |
| "learning_rate": 4.55324427480916e-05, | |
| "loss": 0.5726, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 21.99, | |
| "learning_rate": 4.524618320610687e-05, | |
| "loss": 0.5547, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 22.16, | |
| "learning_rate": 4.495992366412213e-05, | |
| "loss": 0.5536, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 22.16, | |
| "eval_loss": 0.10082551091909409, | |
| "eval_runtime": 31.121, | |
| "eval_samples_per_second": 5.72, | |
| "eval_steps_per_second": 1.928, | |
| "eval_wer": 0.11823983876385623, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 22.34, | |
| "learning_rate": 4.46736641221374e-05, | |
| "loss": 0.5509, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 22.52, | |
| "learning_rate": 4.439026717557252e-05, | |
| "loss": 0.5645, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 22.69, | |
| "learning_rate": 4.410400763358778e-05, | |
| "loss": 0.5527, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 22.87, | |
| "learning_rate": 4.381774809160305e-05, | |
| "loss": 0.547, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 23.05, | |
| "learning_rate": 4.3534351145038163e-05, | |
| "loss": 0.5656, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 23.05, | |
| "eval_loss": 0.10101909935474396, | |
| "eval_runtime": 30.9319, | |
| "eval_samples_per_second": 5.755, | |
| "eval_steps_per_second": 1.94, | |
| "eval_wer": 0.11723211286530064, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 23.23, | |
| "learning_rate": 4.3248091603053434e-05, | |
| "loss": 0.5483, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 23.4, | |
| "learning_rate": 4.29618320610687e-05, | |
| "loss": 0.5501, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 23.58, | |
| "learning_rate": 4.267557251908397e-05, | |
| "loss": 0.5429, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 23.76, | |
| "learning_rate": 4.238931297709923e-05, | |
| "loss": 0.5455, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 23.94, | |
| "learning_rate": 4.21030534351145e-05, | |
| "loss": 0.5504, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 23.94, | |
| "eval_loss": 0.10192937403917313, | |
| "eval_runtime": 31.0113, | |
| "eval_samples_per_second": 5.74, | |
| "eval_steps_per_second": 1.935, | |
| "eval_wer": 0.11051394020826336, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 24.11, | |
| "learning_rate": 4.181679389312977e-05, | |
| "loss": 0.5601, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 24.29, | |
| "learning_rate": 4.1530534351145035e-05, | |
| "loss": 0.5419, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 24.47, | |
| "learning_rate": 4.1244274809160305e-05, | |
| "loss": 0.5389, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 24.65, | |
| "learning_rate": 4.095801526717557e-05, | |
| "loss": 0.5572, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 24.82, | |
| "learning_rate": 4.067175572519084e-05, | |
| "loss": 0.5476, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 24.82, | |
| "eval_loss": 0.10260963439941406, | |
| "eval_runtime": 31.1565, | |
| "eval_samples_per_second": 5.713, | |
| "eval_steps_per_second": 1.926, | |
| "eval_wer": 0.11656029559959691, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 4.03854961832061e-05, | |
| "loss": 0.5554, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 25.18, | |
| "learning_rate": 4.009923664122137e-05, | |
| "loss": 0.537, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 25.35, | |
| "learning_rate": 3.9812977099236635e-05, | |
| "loss": 0.5346, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 25.53, | |
| "learning_rate": 3.9526717557251906e-05, | |
| "loss": 0.5245, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 25.71, | |
| "learning_rate": 3.9240458015267176e-05, | |
| "loss": 0.5375, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 25.71, | |
| "eval_loss": 0.1107296496629715, | |
| "eval_runtime": 30.9057, | |
| "eval_samples_per_second": 5.759, | |
| "eval_steps_per_second": 1.941, | |
| "eval_wer": 0.11891165602955996, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 25.89, | |
| "learning_rate": 3.895419847328244e-05, | |
| "loss": 0.5437, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 26.06, | |
| "learning_rate": 3.866793893129771e-05, | |
| "loss": 0.5284, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 26.24, | |
| "learning_rate": 3.838167938931297e-05, | |
| "loss": 0.5276, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 26.42, | |
| "learning_rate": 3.809541984732824e-05, | |
| "loss": 0.5441, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 26.6, | |
| "learning_rate": 3.7809160305343507e-05, | |
| "loss": 0.5318, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 26.6, | |
| "eval_loss": 0.10514429956674576, | |
| "eval_runtime": 31.3716, | |
| "eval_samples_per_second": 5.674, | |
| "eval_steps_per_second": 1.913, | |
| "eval_wer": 0.11420893516963386, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 26.77, | |
| "learning_rate": 3.752290076335878e-05, | |
| "loss": 0.5343, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 26.95, | |
| "learning_rate": 3.723664122137404e-05, | |
| "loss": 0.5394, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 27.13, | |
| "learning_rate": 3.695038167938931e-05, | |
| "loss": 0.5352, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 27.3, | |
| "learning_rate": 3.6664122137404574e-05, | |
| "loss": 0.5398, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 27.48, | |
| "learning_rate": 3.6377862595419844e-05, | |
| "loss": 0.5278, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 27.48, | |
| "eval_loss": 0.10493182390928268, | |
| "eval_runtime": 31.3849, | |
| "eval_samples_per_second": 5.672, | |
| "eval_steps_per_second": 1.912, | |
| "eval_wer": 0.11656029559959691, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 27.66, | |
| "learning_rate": 3.609160305343511e-05, | |
| "loss": 0.538, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 27.84, | |
| "learning_rate": 3.580534351145038e-05, | |
| "loss": 0.5143, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 28.01, | |
| "learning_rate": 3.551908396946565e-05, | |
| "loss": 0.5229, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 28.19, | |
| "learning_rate": 3.523282442748091e-05, | |
| "loss": 0.522, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 28.37, | |
| "learning_rate": 3.494656488549618e-05, | |
| "loss": 0.5204, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 28.37, | |
| "eval_loss": 0.10806475579738617, | |
| "eval_runtime": 31.4114, | |
| "eval_samples_per_second": 5.667, | |
| "eval_steps_per_second": 1.91, | |
| "eval_wer": 0.11823983876385623, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 28.55, | |
| "learning_rate": 3.4660305343511445e-05, | |
| "loss": 0.5322, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 28.72, | |
| "learning_rate": 3.4374045801526715e-05, | |
| "loss": 0.5151, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 28.9, | |
| "learning_rate": 3.408778625954198e-05, | |
| "loss": 0.5393, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 29.08, | |
| "learning_rate": 3.380152671755725e-05, | |
| "loss": 0.513, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 29.26, | |
| "learning_rate": 3.351526717557251e-05, | |
| "loss": 0.512, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 29.26, | |
| "eval_loss": 0.10623880475759506, | |
| "eval_runtime": 31.0174, | |
| "eval_samples_per_second": 5.739, | |
| "eval_steps_per_second": 1.934, | |
| "eval_wer": 0.11555256970104132, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 29.43, | |
| "learning_rate": 3.322900763358778e-05, | |
| "loss": 0.5258, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 29.61, | |
| "learning_rate": 3.294274809160305e-05, | |
| "loss": 0.5366, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 29.79, | |
| "learning_rate": 3.2656488549618316e-05, | |
| "loss": 0.515, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 29.96, | |
| "learning_rate": 3.2370229007633586e-05, | |
| "loss": 0.5246, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 30.14, | |
| "learning_rate": 3.208396946564885e-05, | |
| "loss": 0.5082, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 30.14, | |
| "eval_loss": 0.10452543944120407, | |
| "eval_runtime": 30.9393, | |
| "eval_samples_per_second": 5.753, | |
| "eval_steps_per_second": 1.939, | |
| "eval_wer": 0.11353711790393013, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 30.32, | |
| "learning_rate": 3.179770992366412e-05, | |
| "loss": 0.5029, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 30.5, | |
| "learning_rate": 3.151431297709923e-05, | |
| "loss": 0.5219, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 30.67, | |
| "learning_rate": 3.12280534351145e-05, | |
| "loss": 0.5235, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 30.85, | |
| "learning_rate": 3.0941793893129764e-05, | |
| "loss": 0.5141, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 31.03, | |
| "learning_rate": 3.0655534351145035e-05, | |
| "loss": 0.5193, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 31.03, | |
| "eval_loss": 0.10911107808351517, | |
| "eval_runtime": 30.9608, | |
| "eval_samples_per_second": 5.749, | |
| "eval_steps_per_second": 1.938, | |
| "eval_wer": 0.11454484380248572, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 31.21, | |
| "learning_rate": 3.0369274809160305e-05, | |
| "loss": 0.5038, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 31.38, | |
| "learning_rate": 3.008301526717557e-05, | |
| "loss": 0.5086, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 31.56, | |
| "learning_rate": 2.979675572519084e-05, | |
| "loss": 0.5001, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 31.74, | |
| "learning_rate": 2.9510496183206105e-05, | |
| "loss": 0.5206, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 31.91, | |
| "learning_rate": 2.9224236641221372e-05, | |
| "loss": 0.5129, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 31.91, | |
| "eval_loss": 0.10398419201374054, | |
| "eval_runtime": 30.9924, | |
| "eval_samples_per_second": 5.743, | |
| "eval_steps_per_second": 1.936, | |
| "eval_wer": 0.10883439704400404, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 32.09, | |
| "learning_rate": 2.893797709923664e-05, | |
| "loss": 0.5105, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 32.27, | |
| "learning_rate": 2.8651717557251906e-05, | |
| "loss": 0.5062, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 32.45, | |
| "learning_rate": 2.8365458015267172e-05, | |
| "loss": 0.5021, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 32.62, | |
| "learning_rate": 2.807919847328244e-05, | |
| "loss": 0.5122, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 32.8, | |
| "learning_rate": 2.779293893129771e-05, | |
| "loss": 0.5126, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 32.8, | |
| "eval_loss": 0.10847991704940796, | |
| "eval_runtime": 30.8721, | |
| "eval_samples_per_second": 5.766, | |
| "eval_steps_per_second": 1.944, | |
| "eval_wer": 0.11689620423244877, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 32.98, | |
| "learning_rate": 2.7506679389312976e-05, | |
| "loss": 0.5179, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 33.16, | |
| "learning_rate": 2.7220419847328243e-05, | |
| "loss": 0.5049, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 2.693416030534351e-05, | |
| "loss": 0.5173, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 33.51, | |
| "learning_rate": 2.6647900763358777e-05, | |
| "loss": 0.5115, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 33.69, | |
| "learning_rate": 2.6361641221374043e-05, | |
| "loss": 0.496, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 33.69, | |
| "eval_loss": 0.10698471963405609, | |
| "eval_runtime": 31.005, | |
| "eval_samples_per_second": 5.741, | |
| "eval_steps_per_second": 1.935, | |
| "eval_wer": 0.11656029559959691, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 33.86, | |
| "learning_rate": 2.607538167938931e-05, | |
| "loss": 0.4938, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 34.04, | |
| "learning_rate": 2.5789122137404577e-05, | |
| "loss": 0.5128, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 34.22, | |
| "learning_rate": 2.550572519083969e-05, | |
| "loss": 0.497, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 34.4, | |
| "learning_rate": 2.521946564885496e-05, | |
| "loss": 0.4879, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 34.57, | |
| "learning_rate": 2.493320610687023e-05, | |
| "loss": 0.5017, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 34.57, | |
| "eval_loss": 0.11190272867679596, | |
| "eval_runtime": 30.8238, | |
| "eval_samples_per_second": 5.775, | |
| "eval_steps_per_second": 1.947, | |
| "eval_wer": 0.11622438696674504, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 34.75, | |
| "learning_rate": 2.4646946564885495e-05, | |
| "loss": 0.4994, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 34.93, | |
| "learning_rate": 2.436354961832061e-05, | |
| "loss": 0.4973, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 35.11, | |
| "learning_rate": 2.4077290076335876e-05, | |
| "loss": 0.5012, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 35.28, | |
| "learning_rate": 2.3791030534351143e-05, | |
| "loss": 0.4889, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 35.46, | |
| "learning_rate": 2.350477099236641e-05, | |
| "loss": 0.4808, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 35.46, | |
| "eval_loss": 0.11005562543869019, | |
| "eval_runtime": 31.0411, | |
| "eval_samples_per_second": 5.734, | |
| "eval_steps_per_second": 1.933, | |
| "eval_wer": 0.113873026536782, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 35.64, | |
| "learning_rate": 2.3218511450381677e-05, | |
| "loss": 0.4984, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 35.82, | |
| "learning_rate": 2.2932251908396944e-05, | |
| "loss": 0.4929, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 35.99, | |
| "learning_rate": 2.2645992366412214e-05, | |
| "loss": 0.4861, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 36.17, | |
| "learning_rate": 2.235973282442748e-05, | |
| "loss": 0.506, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 36.35, | |
| "learning_rate": 2.2073473282442747e-05, | |
| "loss": 0.4939, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 36.35, | |
| "eval_loss": 0.10813739150762558, | |
| "eval_runtime": 31.0724, | |
| "eval_samples_per_second": 5.729, | |
| "eval_steps_per_second": 1.931, | |
| "eval_wer": 0.11252939200537454, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 36.52, | |
| "learning_rate": 2.1787213740458014e-05, | |
| "loss": 0.4789, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 36.7, | |
| "learning_rate": 2.150095419847328e-05, | |
| "loss": 0.4904, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 36.88, | |
| "learning_rate": 2.1214694656488548e-05, | |
| "loss": 0.4916, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 37.06, | |
| "learning_rate": 2.0928435114503815e-05, | |
| "loss": 0.487, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 37.23, | |
| "learning_rate": 2.064217557251908e-05, | |
| "loss": 0.4738, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 37.23, | |
| "eval_loss": 0.10911141335964203, | |
| "eval_runtime": 31.0688, | |
| "eval_samples_per_second": 5.729, | |
| "eval_steps_per_second": 1.931, | |
| "eval_wer": 0.10984212294255963, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 37.41, | |
| "learning_rate": 2.0355916030534352e-05, | |
| "loss": 0.4757, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 37.59, | |
| "learning_rate": 2.006965648854962e-05, | |
| "loss": 0.4752, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 37.77, | |
| "learning_rate": 1.9783396946564885e-05, | |
| "loss": 0.4939, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 37.94, | |
| "learning_rate": 1.9497137404580152e-05, | |
| "loss": 0.4673, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 38.12, | |
| "learning_rate": 1.921087786259542e-05, | |
| "loss": 0.4978, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 38.12, | |
| "eval_loss": 0.10570317506790161, | |
| "eval_runtime": 31.0697, | |
| "eval_samples_per_second": 5.729, | |
| "eval_steps_per_second": 1.931, | |
| "eval_wer": 0.1091703056768559, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 38.3, | |
| "learning_rate": 1.8924618320610686e-05, | |
| "loss": 0.4829, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 38.47, | |
| "learning_rate": 1.8638358778625953e-05, | |
| "loss": 0.4857, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 38.65, | |
| "learning_rate": 1.835209923664122e-05, | |
| "loss": 0.4618, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 38.83, | |
| "learning_rate": 1.8065839694656486e-05, | |
| "loss": 0.4831, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 39.01, | |
| "learning_rate": 1.77824427480916e-05, | |
| "loss": 0.4972, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 39.01, | |
| "eval_loss": 0.10742757469415665, | |
| "eval_runtime": 31.1659, | |
| "eval_samples_per_second": 5.711, | |
| "eval_steps_per_second": 1.925, | |
| "eval_wer": 0.11051394020826336, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 39.18, | |
| "learning_rate": 1.7496183206106867e-05, | |
| "loss": 0.4974, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 39.36, | |
| "learning_rate": 1.7209923664122138e-05, | |
| "loss": 0.4778, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 39.54, | |
| "learning_rate": 1.6923664122137404e-05, | |
| "loss": 0.4863, | |
| "step": 22300 | |
| }, | |
| { | |
| "epoch": 39.72, | |
| "learning_rate": 1.663740458015267e-05, | |
| "loss": 0.4815, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 39.89, | |
| "learning_rate": 1.6351145038167938e-05, | |
| "loss": 0.4773, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 39.89, | |
| "eval_loss": 0.10622948408126831, | |
| "eval_runtime": 31.5999, | |
| "eval_samples_per_second": 5.633, | |
| "eval_steps_per_second": 1.899, | |
| "eval_wer": 0.11084984884111522, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 40.07, | |
| "learning_rate": 1.6064885496183205e-05, | |
| "loss": 0.4808, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 40.25, | |
| "learning_rate": 1.577862595419847e-05, | |
| "loss": 0.4804, | |
| "step": 22700 | |
| }, | |
| { | |
| "epoch": 40.43, | |
| "learning_rate": 1.549236641221374e-05, | |
| "loss": 0.4796, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 40.6, | |
| "learning_rate": 1.5206106870229005e-05, | |
| "loss": 0.4866, | |
| "step": 22900 | |
| }, | |
| { | |
| "epoch": 40.78, | |
| "learning_rate": 1.4919847328244272e-05, | |
| "loss": 0.4741, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 40.78, | |
| "eval_loss": 0.10567349940538406, | |
| "eval_runtime": 31.2551, | |
| "eval_samples_per_second": 5.695, | |
| "eval_steps_per_second": 1.92, | |
| "eval_wer": 0.10849848841115217, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 40.96, | |
| "learning_rate": 1.4633587786259542e-05, | |
| "loss": 0.4976, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 41.13, | |
| "learning_rate": 1.4347328244274809e-05, | |
| "loss": 0.4811, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 41.31, | |
| "learning_rate": 1.4061068702290076e-05, | |
| "loss": 0.4904, | |
| "step": 23300 | |
| }, | |
| { | |
| "epoch": 41.49, | |
| "learning_rate": 1.3774809160305341e-05, | |
| "loss": 0.4658, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 41.67, | |
| "learning_rate": 1.3488549618320608e-05, | |
| "loss": 0.4776, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 41.67, | |
| "eval_loss": 0.10771991312503815, | |
| "eval_runtime": 31.4029, | |
| "eval_samples_per_second": 5.668, | |
| "eval_steps_per_second": 1.911, | |
| "eval_wer": 0.10849848841115217, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 41.84, | |
| "learning_rate": 1.3202290076335878e-05, | |
| "loss": 0.4745, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 42.02, | |
| "learning_rate": 1.2916030534351145e-05, | |
| "loss": 0.4746, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 42.2, | |
| "learning_rate": 1.262977099236641e-05, | |
| "loss": 0.4858, | |
| "step": 23800 | |
| }, | |
| { | |
| "epoch": 42.38, | |
| "learning_rate": 1.2343511450381677e-05, | |
| "loss": 0.4687, | |
| "step": 23900 | |
| }, | |
| { | |
| "epoch": 42.55, | |
| "learning_rate": 1.2057251908396947e-05, | |
| "loss": 0.4637, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 42.55, | |
| "eval_loss": 0.10607700049877167, | |
| "eval_runtime": 31.045, | |
| "eval_samples_per_second": 5.734, | |
| "eval_steps_per_second": 1.933, | |
| "eval_wer": 0.10950621430970776, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 42.73, | |
| "learning_rate": 1.1770992366412214e-05, | |
| "loss": 0.4462, | |
| "step": 24100 | |
| }, | |
| { | |
| "epoch": 42.91, | |
| "learning_rate": 1.1484732824427479e-05, | |
| "loss": 0.4815, | |
| "step": 24200 | |
| }, | |
| { | |
| "epoch": 43.09, | |
| "learning_rate": 1.1198473282442746e-05, | |
| "loss": 0.461, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 43.26, | |
| "learning_rate": 1.0912213740458016e-05, | |
| "loss": 0.4594, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 43.44, | |
| "learning_rate": 1.0625954198473283e-05, | |
| "loss": 0.4853, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 43.44, | |
| "eval_loss": 0.10806521028280258, | |
| "eval_runtime": 31.4451, | |
| "eval_samples_per_second": 5.661, | |
| "eval_steps_per_second": 1.908, | |
| "eval_wer": 0.10749076251259658, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 43.62, | |
| "learning_rate": 1.033969465648855e-05, | |
| "loss": 0.4747, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 43.79, | |
| "learning_rate": 1.0053435114503815e-05, | |
| "loss": 0.4649, | |
| "step": 24700 | |
| }, | |
| { | |
| "epoch": 43.97, | |
| "learning_rate": 9.767175572519081e-06, | |
| "loss": 0.471, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 44.15, | |
| "learning_rate": 9.480916030534352e-06, | |
| "loss": 0.453, | |
| "step": 24900 | |
| }, | |
| { | |
| "epoch": 44.33, | |
| "learning_rate": 9.194656488549618e-06, | |
| "loss": 0.4602, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 44.33, | |
| "eval_loss": 0.1075613871216774, | |
| "eval_runtime": 31.217, | |
| "eval_samples_per_second": 5.702, | |
| "eval_steps_per_second": 1.922, | |
| "eval_wer": 0.10849848841115217, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 44.5, | |
| "learning_rate": 8.908396946564884e-06, | |
| "loss": 0.4579, | |
| "step": 25100 | |
| }, | |
| { | |
| "epoch": 44.68, | |
| "learning_rate": 8.622137404580152e-06, | |
| "loss": 0.4752, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 44.86, | |
| "learning_rate": 8.335877862595419e-06, | |
| "loss": 0.4708, | |
| "step": 25300 | |
| }, | |
| { | |
| "epoch": 45.04, | |
| "learning_rate": 8.049618320610687e-06, | |
| "loss": 0.4613, | |
| "step": 25400 | |
| }, | |
| { | |
| "epoch": 45.21, | |
| "learning_rate": 7.763358778625954e-06, | |
| "loss": 0.4667, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 45.21, | |
| "eval_loss": 0.1077902689576149, | |
| "eval_runtime": 31.0572, | |
| "eval_samples_per_second": 5.731, | |
| "eval_steps_per_second": 1.932, | |
| "eval_wer": 0.10782667114544844, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 45.39, | |
| "learning_rate": 7.477099236641221e-06, | |
| "loss": 0.4709, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 45.57, | |
| "learning_rate": 7.190839694656488e-06, | |
| "loss": 0.4523, | |
| "step": 25700 | |
| }, | |
| { | |
| "epoch": 45.74, | |
| "learning_rate": 6.9045801526717555e-06, | |
| "loss": 0.4537, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 45.92, | |
| "learning_rate": 6.62118320610687e-06, | |
| "loss": 0.4642, | |
| "step": 25900 | |
| }, | |
| { | |
| "epoch": 46.1, | |
| "learning_rate": 6.334923664122137e-06, | |
| "loss": 0.4484, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 46.1, | |
| "eval_loss": 0.10564640909433365, | |
| "eval_runtime": 31.5659, | |
| "eval_samples_per_second": 5.639, | |
| "eval_steps_per_second": 1.901, | |
| "eval_wer": 0.10816257977830031, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 46.28, | |
| "learning_rate": 6.048664122137404e-06, | |
| "loss": 0.4505, | |
| "step": 26100 | |
| }, | |
| { | |
| "epoch": 46.45, | |
| "learning_rate": 5.762404580152671e-06, | |
| "loss": 0.4758, | |
| "step": 26200 | |
| }, | |
| { | |
| "epoch": 46.63, | |
| "learning_rate": 5.476145038167939e-06, | |
| "loss": 0.4528, | |
| "step": 26300 | |
| }, | |
| { | |
| "epoch": 46.81, | |
| "learning_rate": 5.1898854961832056e-06, | |
| "loss": 0.4579, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 46.99, | |
| "learning_rate": 4.903625954198473e-06, | |
| "loss": 0.4601, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 46.99, | |
| "eval_loss": 0.10663167387247086, | |
| "eval_runtime": 31.3245, | |
| "eval_samples_per_second": 5.682, | |
| "eval_steps_per_second": 1.915, | |
| "eval_wer": 0.10782667114544844, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 47.16, | |
| "learning_rate": 4.61736641221374e-06, | |
| "loss": 0.4634, | |
| "step": 26600 | |
| }, | |
| { | |
| "epoch": 47.34, | |
| "learning_rate": 4.331106870229007e-06, | |
| "loss": 0.4603, | |
| "step": 26700 | |
| }, | |
| { | |
| "epoch": 47.52, | |
| "learning_rate": 4.0448473282442745e-06, | |
| "loss": 0.4653, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 47.69, | |
| "learning_rate": 3.7585877862595418e-06, | |
| "loss": 0.4619, | |
| "step": 26900 | |
| }, | |
| { | |
| "epoch": 47.87, | |
| "learning_rate": 3.472328244274809e-06, | |
| "loss": 0.4691, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 47.87, | |
| "eval_loss": 0.10679604858160019, | |
| "eval_runtime": 31.1673, | |
| "eval_samples_per_second": 5.711, | |
| "eval_steps_per_second": 1.925, | |
| "eval_wer": 0.10849848841115217, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 48.05, | |
| "learning_rate": 3.1860687022900762e-06, | |
| "loss": 0.4582, | |
| "step": 27100 | |
| }, | |
| { | |
| "epoch": 48.23, | |
| "learning_rate": 2.8998091603053435e-06, | |
| "loss": 0.4439, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 48.4, | |
| "learning_rate": 2.6135496183206107e-06, | |
| "loss": 0.4617, | |
| "step": 27300 | |
| }, | |
| { | |
| "epoch": 48.58, | |
| "learning_rate": 2.3272900763358775e-06, | |
| "loss": 0.4538, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 48.76, | |
| "learning_rate": 2.0410305343511447e-06, | |
| "loss": 0.4457, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 48.76, | |
| "eval_loss": 0.10663049668073654, | |
| "eval_runtime": 31.2143, | |
| "eval_samples_per_second": 5.703, | |
| "eval_steps_per_second": 1.922, | |
| "eval_wer": 0.10782667114544844, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 48.94, | |
| "learning_rate": 1.7547709923664122e-06, | |
| "loss": 0.4766, | |
| "step": 27600 | |
| }, | |
| { | |
| "epoch": 49.11, | |
| "learning_rate": 1.4685114503816792e-06, | |
| "loss": 0.4576, | |
| "step": 27700 | |
| }, | |
| { | |
| "epoch": 49.29, | |
| "learning_rate": 1.185114503816794e-06, | |
| "loss": 0.4616, | |
| "step": 27800 | |
| }, | |
| { | |
| "epoch": 49.47, | |
| "learning_rate": 8.98854961832061e-07, | |
| "loss": 0.4588, | |
| "step": 27900 | |
| }, | |
| { | |
| "epoch": 49.65, | |
| "learning_rate": 6.125954198473282e-07, | |
| "loss": 0.475, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 49.65, | |
| "eval_loss": 0.1059938296675682, | |
| "eval_runtime": 32.0006, | |
| "eval_samples_per_second": 5.562, | |
| "eval_steps_per_second": 1.875, | |
| "eval_wer": 0.10816257977830031, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 49.82, | |
| "learning_rate": 3.263358778625954e-07, | |
| "loss": 0.4622, | |
| "step": 28100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 4.0076335877862596e-08, | |
| "loss": 0.4567, | |
| "step": 28200 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 28200, | |
| "total_flos": 6.992186583697227e+19, | |
| "train_loss": 0.6889870901987062, | |
| "train_runtime": 132713.9902, | |
| "train_samples_per_second": 2.551, | |
| "train_steps_per_second": 0.212 | |
| } | |
| ], | |
| "max_steps": 28200, | |
| "num_train_epochs": 50, | |
| "total_flos": 6.992186583697227e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |