| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 50.0, | |
| "global_step": 2550, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 7.361999999999999e-05, | |
| "loss": 5.3118, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "eval_loss": 2.9093422889709473, | |
| "eval_runtime": 17.6486, | |
| "eval_samples_per_second": 19.322, | |
| "eval_steps_per_second": 2.436, | |
| "eval_wer": 0.9981549815498155, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 7.062e-05, | |
| "loss": 2.2071, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "eval_loss": 1.1736845970153809, | |
| "eval_runtime": 17.7952, | |
| "eval_samples_per_second": 19.163, | |
| "eval_steps_per_second": 2.416, | |
| "eval_wer": 0.777859778597786, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "learning_rate": 6.761999999999999e-05, | |
| "loss": 1.6098, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "eval_loss": 0.9983930587768555, | |
| "eval_runtime": 17.6369, | |
| "eval_samples_per_second": 19.334, | |
| "eval_steps_per_second": 2.438, | |
| "eval_wer": 0.7014760147601476, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "learning_rate": 6.462e-05, | |
| "loss": 1.4333, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "eval_loss": 0.9799550771713257, | |
| "eval_runtime": 17.5884, | |
| "eval_samples_per_second": 19.388, | |
| "eval_steps_per_second": 2.445, | |
| "eval_wer": 0.670479704797048, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "learning_rate": 6.162e-05, | |
| "loss": 1.2859, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "eval_loss": 0.9582107067108154, | |
| "eval_runtime": 17.5894, | |
| "eval_samples_per_second": 19.387, | |
| "eval_steps_per_second": 2.445, | |
| "eval_wer": 0.6487084870848708, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 11.76, | |
| "learning_rate": 5.861999999999999e-05, | |
| "loss": 1.2073, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 11.76, | |
| "eval_loss": 0.8841160535812378, | |
| "eval_runtime": 17.5809, | |
| "eval_samples_per_second": 19.396, | |
| "eval_steps_per_second": 2.446, | |
| "eval_wer": 0.6077490774907749, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 13.73, | |
| "learning_rate": 5.562e-05, | |
| "loss": 1.1417, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 13.73, | |
| "eval_loss": 0.9118468165397644, | |
| "eval_runtime": 17.5423, | |
| "eval_samples_per_second": 19.439, | |
| "eval_steps_per_second": 2.451, | |
| "eval_wer": 0.6343173431734317, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 15.69, | |
| "learning_rate": 5.2619999999999994e-05, | |
| "loss": 1.0988, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 15.69, | |
| "eval_loss": 0.9216808080673218, | |
| "eval_runtime": 17.5542, | |
| "eval_samples_per_second": 19.426, | |
| "eval_steps_per_second": 2.45, | |
| "eval_wer": 0.6195571955719558, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 17.65, | |
| "learning_rate": 4.9619999999999996e-05, | |
| "loss": 1.0279, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 17.65, | |
| "eval_loss": 0.9165470600128174, | |
| "eval_runtime": 17.6001, | |
| "eval_samples_per_second": 19.375, | |
| "eval_steps_per_second": 2.443, | |
| "eval_wer": 0.5867158671586716, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 19.61, | |
| "learning_rate": 4.662e-05, | |
| "loss": 0.9765, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 19.61, | |
| "eval_loss": 0.9305969476699829, | |
| "eval_runtime": 17.6088, | |
| "eval_samples_per_second": 19.365, | |
| "eval_steps_per_second": 2.442, | |
| "eval_wer": 0.5977859778597786, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 21.57, | |
| "learning_rate": 4.362e-05, | |
| "loss": 0.9161, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 21.57, | |
| "eval_loss": 0.9304590225219727, | |
| "eval_runtime": 17.4967, | |
| "eval_samples_per_second": 19.489, | |
| "eval_steps_per_second": 2.458, | |
| "eval_wer": 0.5767527675276752, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 23.53, | |
| "learning_rate": 4.0619999999999994e-05, | |
| "loss": 0.8395, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 23.53, | |
| "eval_loss": 0.982844889163971, | |
| "eval_runtime": 17.7044, | |
| "eval_samples_per_second": 19.261, | |
| "eval_steps_per_second": 2.429, | |
| "eval_wer": 0.5819188191881919, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 25.49, | |
| "learning_rate": 3.762e-05, | |
| "loss": 0.8306, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 25.49, | |
| "eval_loss": 0.9397062063217163, | |
| "eval_runtime": 17.6627, | |
| "eval_samples_per_second": 19.306, | |
| "eval_steps_per_second": 2.435, | |
| "eval_wer": 0.5760147601476014, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 27.45, | |
| "learning_rate": 3.462e-05, | |
| "loss": 0.7819, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 27.45, | |
| "eval_loss": 0.9543871283531189, | |
| "eval_runtime": 17.6517, | |
| "eval_samples_per_second": 19.318, | |
| "eval_steps_per_second": 2.436, | |
| "eval_wer": 0.574169741697417, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 29.41, | |
| "learning_rate": 3.161999999999999e-05, | |
| "loss": 0.7509, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 29.41, | |
| "eval_loss": 0.9278193712234497, | |
| "eval_runtime": 17.4199, | |
| "eval_samples_per_second": 19.575, | |
| "eval_steps_per_second": 2.468, | |
| "eval_wer": 0.5690036900369003, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 31.37, | |
| "learning_rate": 2.8619999999999997e-05, | |
| "loss": 0.7218, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 31.37, | |
| "eval_loss": 0.9002724289894104, | |
| "eval_runtime": 17.4813, | |
| "eval_samples_per_second": 19.507, | |
| "eval_steps_per_second": 2.46, | |
| "eval_wer": 0.5586715867158671, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 2.562e-05, | |
| "loss": 0.6725, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "eval_loss": 0.9659368395805359, | |
| "eval_runtime": 17.6287, | |
| "eval_samples_per_second": 19.343, | |
| "eval_steps_per_second": 2.439, | |
| "eval_wer": 0.5553505535055351, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 35.29, | |
| "learning_rate": 2.2619999999999997e-05, | |
| "loss": 0.6287, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 35.29, | |
| "eval_loss": 0.9521807432174683, | |
| "eval_runtime": 17.6153, | |
| "eval_samples_per_second": 19.358, | |
| "eval_steps_per_second": 2.441, | |
| "eval_wer": 0.5560885608856089, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 37.25, | |
| "learning_rate": 1.962e-05, | |
| "loss": 0.6077, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 37.25, | |
| "eval_loss": 0.9153698086738586, | |
| "eval_runtime": 17.5781, | |
| "eval_samples_per_second": 19.399, | |
| "eval_steps_per_second": 2.446, | |
| "eval_wer": 0.5464944649446495, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 39.22, | |
| "learning_rate": 1.6619999999999997e-05, | |
| "loss": 0.5873, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 39.22, | |
| "eval_loss": 0.9330592751502991, | |
| "eval_runtime": 17.6281, | |
| "eval_samples_per_second": 19.344, | |
| "eval_steps_per_second": 2.439, | |
| "eval_wer": 0.5468634686346864, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 41.18, | |
| "learning_rate": 1.362e-05, | |
| "loss": 0.5621, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 41.18, | |
| "eval_loss": 0.9334675669670105, | |
| "eval_runtime": 17.5009, | |
| "eval_samples_per_second": 19.485, | |
| "eval_steps_per_second": 2.457, | |
| "eval_wer": 0.5490774907749078, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 43.14, | |
| "learning_rate": 1.062e-05, | |
| "loss": 0.5168, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 43.14, | |
| "eval_loss": 0.9632032513618469, | |
| "eval_runtime": 17.6433, | |
| "eval_samples_per_second": 19.327, | |
| "eval_steps_per_second": 2.437, | |
| "eval_wer": 0.5457564575645757, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 45.1, | |
| "learning_rate": 7.619999999999999e-06, | |
| "loss": 0.5114, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 45.1, | |
| "eval_loss": 0.9349461793899536, | |
| "eval_runtime": 17.5705, | |
| "eval_samples_per_second": 19.407, | |
| "eval_steps_per_second": 2.447, | |
| "eval_wer": 0.5387453874538746, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 47.06, | |
| "learning_rate": 4.62e-06, | |
| "loss": 0.4986, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 47.06, | |
| "eval_loss": 0.9364362955093384, | |
| "eval_runtime": 17.6211, | |
| "eval_samples_per_second": 19.352, | |
| "eval_steps_per_second": 2.44, | |
| "eval_wer": 0.5380073800738008, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 49.02, | |
| "learning_rate": 1.62e-06, | |
| "loss": 0.4761, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 49.02, | |
| "eval_loss": 0.9584243297576904, | |
| "eval_runtime": 17.5152, | |
| "eval_samples_per_second": 19.469, | |
| "eval_steps_per_second": 2.455, | |
| "eval_wer": 0.5391143911439115, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 2550, | |
| "total_flos": 1.5012316680494998e+19, | |
| "train_loss": 1.0758988070020488, | |
| "train_runtime": 4106.4423, | |
| "train_samples_per_second": 9.863, | |
| "train_steps_per_second": 0.621 | |
| } | |
| ], | |
| "max_steps": 2550, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.5012316680494998e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |