| { | |
| "best_metric": 0.14731654524803162, | |
| "best_model_checkpoint": "./xls-r-1b-bem-sv-male/checkpoint-3500", | |
| "epoch": 4.997397188964081, | |
| "eval_steps": 500, | |
| "global_step": 4800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.20822488287350338, | |
| "grad_norm": 6.451992988586426, | |
| "learning_rate": 1.9600000000000002e-05, | |
| "loss": 4.6771, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.41644976574700676, | |
| "grad_norm": 3.6971700191497803, | |
| "learning_rate": 3.960000000000001e-05, | |
| "loss": 0.9808, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.5205622071837585, | |
| "eval_loss": 0.2580466568470001, | |
| "eval_runtime": 117.6846, | |
| "eval_samples_per_second": 8.922, | |
| "eval_steps_per_second": 2.235, | |
| "eval_wer": 0.8361904761904762, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.6246746486205101, | |
| "grad_norm": 3.8861448764801025, | |
| "learning_rate": 4.8883720930232564e-05, | |
| "loss": 0.5099, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.8328995314940135, | |
| "grad_norm": 6.251056671142578, | |
| "learning_rate": 4.655813953488372e-05, | |
| "loss": 0.4325, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.041124414367517, | |
| "grad_norm": 2.006500005722046, | |
| "learning_rate": 4.423255813953489e-05, | |
| "loss": 0.3877, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.041124414367517, | |
| "eval_loss": 0.20405621826648712, | |
| "eval_runtime": 117.4432, | |
| "eval_samples_per_second": 8.94, | |
| "eval_steps_per_second": 2.239, | |
| "eval_wer": 0.7790476190476191, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.2493492972410203, | |
| "grad_norm": 1.86452317237854, | |
| "learning_rate": 4.190697674418605e-05, | |
| "loss": 0.3249, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.4575741801145237, | |
| "grad_norm": 2.239264965057373, | |
| "learning_rate": 3.958139534883721e-05, | |
| "loss": 0.298, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.5616866215512752, | |
| "eval_loss": 0.1842915266752243, | |
| "eval_runtime": 118.4618, | |
| "eval_samples_per_second": 8.864, | |
| "eval_steps_per_second": 2.22, | |
| "eval_wer": 0.7619047619047619, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.665799062988027, | |
| "grad_norm": 0.9172901511192322, | |
| "learning_rate": 3.725581395348837e-05, | |
| "loss": 0.2978, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.8740239458615304, | |
| "grad_norm": 2.832592487335205, | |
| "learning_rate": 3.4930232558139534e-05, | |
| "loss": 0.2734, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.082248828735034, | |
| "grad_norm": 1.0995967388153076, | |
| "learning_rate": 3.26046511627907e-05, | |
| "loss": 0.2433, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.082248828735034, | |
| "eval_loss": 0.16212375462055206, | |
| "eval_runtime": 117.6082, | |
| "eval_samples_per_second": 8.928, | |
| "eval_steps_per_second": 2.236, | |
| "eval_wer": 0.700952380952381, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.2904737116085374, | |
| "grad_norm": 1.1086174249649048, | |
| "learning_rate": 3.0279069767441864e-05, | |
| "loss": 0.203, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.4986985944820406, | |
| "grad_norm": 1.091976523399353, | |
| "learning_rate": 2.7953488372093022e-05, | |
| "loss": 0.188, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.6028110359187924, | |
| "eval_loss": 0.1653192788362503, | |
| "eval_runtime": 118.9756, | |
| "eval_samples_per_second": 8.825, | |
| "eval_steps_per_second": 2.211, | |
| "eval_wer": 0.6838095238095238, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.706923477355544, | |
| "grad_norm": 0.8310242295265198, | |
| "learning_rate": 2.5627906976744187e-05, | |
| "loss": 0.1958, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.9151483602290473, | |
| "grad_norm": 1.4682559967041016, | |
| "learning_rate": 2.3302325581395352e-05, | |
| "loss": 0.2035, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.123373243102551, | |
| "grad_norm": 0.5501867532730103, | |
| "learning_rate": 2.0976744186046513e-05, | |
| "loss": 0.1521, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.123373243102551, | |
| "eval_loss": 0.14906759560108185, | |
| "eval_runtime": 118.0235, | |
| "eval_samples_per_second": 8.897, | |
| "eval_steps_per_second": 2.228, | |
| "eval_wer": 0.659047619047619, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.331598125976054, | |
| "grad_norm": 0.9272790551185608, | |
| "learning_rate": 1.8651162790697675e-05, | |
| "loss": 0.1381, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 3.5398230088495577, | |
| "grad_norm": 0.9619794487953186, | |
| "learning_rate": 1.6325581395348837e-05, | |
| "loss": 0.1316, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 3.643935450286309, | |
| "eval_loss": 0.14731654524803162, | |
| "eval_runtime": 117.6012, | |
| "eval_samples_per_second": 8.928, | |
| "eval_steps_per_second": 2.236, | |
| "eval_wer": 0.6228571428571429, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.748047891723061, | |
| "grad_norm": 0.7800536155700684, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 0.127, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 3.956272774596564, | |
| "grad_norm": 4.820044994354248, | |
| "learning_rate": 1.1674418604651163e-05, | |
| "loss": 0.128, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 4.164497657470068, | |
| "grad_norm": 0.6071628928184509, | |
| "learning_rate": 9.348837209302326e-06, | |
| "loss": 0.0925, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.164497657470068, | |
| "eval_loss": 0.15515577793121338, | |
| "eval_runtime": 118.2807, | |
| "eval_samples_per_second": 8.877, | |
| "eval_steps_per_second": 2.224, | |
| "eval_wer": 0.6447619047619048, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.372722540343571, | |
| "grad_norm": 0.38652822375297546, | |
| "learning_rate": 7.023255813953489e-06, | |
| "loss": 0.0777, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 4.580947423217075, | |
| "grad_norm": 0.60927414894104, | |
| "learning_rate": 4.697674418604651e-06, | |
| "loss": 0.0767, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 4.685059864653827, | |
| "eval_loss": 0.15233619511127472, | |
| "eval_runtime": 118.0689, | |
| "eval_samples_per_second": 8.893, | |
| "eval_steps_per_second": 2.228, | |
| "eval_wer": 0.6228571428571429, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.7891723060905775, | |
| "grad_norm": 0.5293630361557007, | |
| "learning_rate": 2.372093023255814e-06, | |
| "loss": 0.0731, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 4.997397188964081, | |
| "grad_norm": 0.3965218961238861, | |
| "learning_rate": 4.651162790697675e-08, | |
| "loss": 0.0804, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 4.997397188964081, | |
| "step": 4800, | |
| "total_flos": 2.172568340790461e+19, | |
| "train_loss": 0.42886508484681446, | |
| "train_runtime": 12219.0167, | |
| "train_samples_per_second": 3.144, | |
| "train_steps_per_second": 0.393 | |
| } | |
| ], | |
| "logging_steps": 200, | |
| "max_steps": 4800, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.172568340790461e+19, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |