| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.3166023166023164, | |
| "eval_steps": 200, | |
| "global_step": 6000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07722007722007722, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 199.2963, | |
| "eval_samples_per_second": 35.234, | |
| "eval_steps_per_second": 4.406, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15444015444015444, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 187.0234, | |
| "eval_samples_per_second": 37.546, | |
| "eval_steps_per_second": 4.695, | |
| "eval_wer": 0.9937672860426162, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.19305019305019305, | |
| "grad_norm": 4.210392951965332, | |
| "learning_rate": 0.00029699999999999996, | |
| "loss": 3.9317, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.23166023166023167, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 180.1705, | |
| "eval_samples_per_second": 38.974, | |
| "eval_steps_per_second": 4.873, | |
| "eval_wer": 0.7576253359822368, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.3088803088803089, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 179.0087, | |
| "eval_samples_per_second": 39.227, | |
| "eval_steps_per_second": 4.905, | |
| "eval_wer": 0.6941685170036228, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.3861003861003861, | |
| "grad_norm": 3.149083137512207, | |
| "learning_rate": 0.0002730545454545454, | |
| "loss": 0.9699, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3861003861003861, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 179.6063, | |
| "eval_samples_per_second": 39.097, | |
| "eval_steps_per_second": 4.888, | |
| "eval_wer": 0.5762923142846013, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.46332046332046334, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 180.175, | |
| "eval_samples_per_second": 38.973, | |
| "eval_steps_per_second": 4.873, | |
| "eval_wer": 0.5518548816433589, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.5405405405405406, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 180.7382, | |
| "eval_samples_per_second": 38.852, | |
| "eval_steps_per_second": 4.858, | |
| "eval_wer": 0.5173931673873242, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.5791505791505791, | |
| "grad_norm": 4.50093412399292, | |
| "learning_rate": 0.0002458363636363636, | |
| "loss": 0.8031, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.6177606177606177, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 180.2438, | |
| "eval_samples_per_second": 38.958, | |
| "eval_steps_per_second": 4.871, | |
| "eval_wer": 0.5338059808084349, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.694980694980695, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 180.7932, | |
| "eval_samples_per_second": 38.84, | |
| "eval_steps_per_second": 4.856, | |
| "eval_wer": 0.47772454001272513, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.7722007722007722, | |
| "grad_norm": 3.7705626487731934, | |
| "learning_rate": 0.0002186181818181818, | |
| "loss": 0.7169, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.7722007722007722, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 179.9877, | |
| "eval_samples_per_second": 39.014, | |
| "eval_steps_per_second": 4.878, | |
| "eval_wer": 0.45044343162842637, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.8494208494208494, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 182.9186, | |
| "eval_samples_per_second": 38.389, | |
| "eval_steps_per_second": 4.8, | |
| "eval_wer": 0.4499500084401335, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.9266409266409267, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 180.2372, | |
| "eval_samples_per_second": 38.96, | |
| "eval_steps_per_second": 4.871, | |
| "eval_wer": 0.4431719320114786, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.9652509652509652, | |
| "grad_norm": 3.5582635402679443, | |
| "learning_rate": 0.00019134545454545454, | |
| "loss": 0.6687, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0038610038610039, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 180.3998, | |
| "eval_samples_per_second": 38.925, | |
| "eval_steps_per_second": 4.867, | |
| "eval_wer": 0.4175658655032267, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 180.1365, | |
| "eval_samples_per_second": 38.982, | |
| "eval_steps_per_second": 4.874, | |
| "eval_wer": 0.40537311882409466, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.1583011583011582, | |
| "grad_norm": 0.5179678201675415, | |
| "learning_rate": 0.0001641272727272727, | |
| "loss": 0.5609, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.1583011583011582, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 180.8398, | |
| "eval_samples_per_second": 38.83, | |
| "eval_steps_per_second": 4.855, | |
| "eval_wer": 0.4009193253087141, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.2355212355212355, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 181.6098, | |
| "eval_samples_per_second": 38.665, | |
| "eval_steps_per_second": 4.835, | |
| "eval_wer": 0.4022567618454027, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.3127413127413128, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 182.1259, | |
| "eval_samples_per_second": 38.556, | |
| "eval_steps_per_second": 4.821, | |
| "eval_wer": 0.39188189007050755, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.3513513513513513, | |
| "grad_norm": 0.5364826321601868, | |
| "learning_rate": 0.00013696363636363636, | |
| "loss": 0.5324, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.3899613899613898, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 181.3225, | |
| "eval_samples_per_second": 38.727, | |
| "eval_steps_per_second": 4.842, | |
| "eval_wer": 0.3794813862594627, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.4671814671814671, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 181.5845, | |
| "eval_samples_per_second": 38.671, | |
| "eval_steps_per_second": 4.835, | |
| "eval_wer": 0.37523534987599494, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.5444015444015444, | |
| "grad_norm": 1.197091817855835, | |
| "learning_rate": 0.00010974545454545454, | |
| "loss": 0.5196, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.5444015444015444, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 181.3235, | |
| "eval_samples_per_second": 38.726, | |
| "eval_steps_per_second": 4.842, | |
| "eval_wer": 0.36617194499629935, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.6216216216216215, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 181.9653, | |
| "eval_samples_per_second": 38.59, | |
| "eval_steps_per_second": 4.825, | |
| "eval_wer": 0.37034007245529976, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 1.698841698841699, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 181.5241, | |
| "eval_samples_per_second": 38.684, | |
| "eval_steps_per_second": 4.837, | |
| "eval_wer": 0.3613935309623051, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 1.7374517374517375, | |
| "grad_norm": 0.7301501631736755, | |
| "learning_rate": 8.247272727272728e-05, | |
| "loss": 0.4967, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.776061776061776, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 181.6896, | |
| "eval_samples_per_second": 38.648, | |
| "eval_steps_per_second": 4.832, | |
| "eval_wer": 0.3530313064028151, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 1.8532818532818531, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 184.0459, | |
| "eval_samples_per_second": 38.154, | |
| "eval_steps_per_second": 4.771, | |
| "eval_wer": 0.34805812005765263, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 1.9305019305019306, | |
| "grad_norm": 1.2918003797531128, | |
| "learning_rate": 5.519999999999999e-05, | |
| "loss": 0.4735, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.9305019305019306, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 182.1737, | |
| "eval_samples_per_second": 38.546, | |
| "eval_steps_per_second": 4.82, | |
| "eval_wer": 0.35057717528209525, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.0077220077220077, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 182.5384, | |
| "eval_samples_per_second": 38.469, | |
| "eval_steps_per_second": 4.81, | |
| "eval_wer": 0.3432277667406801, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 2.0849420849420848, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 182.9254, | |
| "eval_samples_per_second": 38.387, | |
| "eval_steps_per_second": 4.8, | |
| "eval_wer": 0.33689117421733994, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 2.1235521235521237, | |
| "grad_norm": 0.6086732745170593, | |
| "learning_rate": 2.7927272727272724e-05, | |
| "loss": 0.4244, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.1621621621621623, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 182.4452, | |
| "eval_samples_per_second": 38.488, | |
| "eval_steps_per_second": 4.812, | |
| "eval_wer": 0.32959370495890306, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 2.2393822393822393, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 182.1862, | |
| "eval_samples_per_second": 38.543, | |
| "eval_steps_per_second": 4.819, | |
| "eval_wer": 0.32954176567592486, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 2.3166023166023164, | |
| "grad_norm": 1.0216798782348633, | |
| "learning_rate": 6.545454545454546e-07, | |
| "loss": 0.3674, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.3166023166023164, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 182.1846, | |
| "eval_samples_per_second": 38.543, | |
| "eval_steps_per_second": 4.819, | |
| "eval_wer": 0.3289055094594419, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.3166023166023164, | |
| "step": 6000, | |
| "total_flos": 1.261526897313927e+19, | |
| "train_loss": 0.8721037826538086, | |
| "train_runtime": 9008.7975, | |
| "train_samples_per_second": 10.656, | |
| "train_steps_per_second": 0.666 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 1.261526897313927e+19, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |