| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.909466234149218, | |
| "global_step": 6700, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4, | |
| "eval_cer": 0.06875412712023102, | |
| "eval_loss": 0.2844763696193695, | |
| "eval_runtime": 65.9715, | |
| "eval_samples_per_second": 7.579, | |
| "eval_steps_per_second": 0.955, | |
| "eval_wer": 0.25628282600101654, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0002394, | |
| "loss": 3.0984, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_cer": 0.04085233040615791, | |
| "eval_loss": 0.15484967827796936, | |
| "eval_runtime": 65.7396, | |
| "eval_samples_per_second": 7.606, | |
| "eval_steps_per_second": 0.958, | |
| "eval_wer": 0.1411306263059807, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.0002857074569789675, | |
| "loss": 0.2022, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "eval_cer": 0.03730423898366892, | |
| "eval_loss": 0.14234833419322968, | |
| "eval_runtime": 65.8678, | |
| "eval_samples_per_second": 7.591, | |
| "eval_steps_per_second": 0.956, | |
| "eval_wer": 0.11916191336759474, | |
| "step": 1005 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.0002665869980879541, | |
| "loss": 0.1458, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "eval_cer": 0.03469244946433676, | |
| "eval_loss": 0.12395954132080078, | |
| "eval_runtime": 66.8459, | |
| "eval_samples_per_second": 7.48, | |
| "eval_steps_per_second": 0.942, | |
| "eval_wer": 0.11165075958660417, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0002474665391969407, | |
| "loss": 0.1354, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_cer": 0.03423908222701871, | |
| "eval_loss": 0.12080004811286926, | |
| "eval_runtime": 65.8681, | |
| "eval_samples_per_second": 7.591, | |
| "eval_steps_per_second": 0.956, | |
| "eval_wer": 0.10814931947817247, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 0.00022834608030592734, | |
| "loss": 0.1169, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "eval_cer": 0.03338162679991721, | |
| "eval_loss": 0.11872279644012451, | |
| "eval_runtime": 66.8663, | |
| "eval_samples_per_second": 7.478, | |
| "eval_steps_per_second": 0.942, | |
| "eval_wer": 0.10735867171175241, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_cer": 0.03327321289534116, | |
| "eval_loss": 0.12243428081274033, | |
| "eval_runtime": 67.2919, | |
| "eval_samples_per_second": 7.43, | |
| "eval_steps_per_second": 0.936, | |
| "eval_wer": 0.10199356186818773, | |
| "step": 2345 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 0.00020922562141491394, | |
| "loss": 0.1107, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "eval_cer": 0.0313808974700137, | |
| "eval_loss": 0.11623063683509827, | |
| "eval_runtime": 66.9398, | |
| "eval_samples_per_second": 7.469, | |
| "eval_steps_per_second": 0.941, | |
| "eval_wer": 0.09724967526966737, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 0.00019010516252390057, | |
| "loss": 0.0976, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "eval_cer": 0.03366744527561771, | |
| "eval_loss": 0.11647358536720276, | |
| "eval_runtime": 67.2006, | |
| "eval_samples_per_second": 7.44, | |
| "eval_steps_per_second": 0.937, | |
| "eval_wer": 0.10052521601626475, | |
| "step": 3015 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 0.0001709847036328872, | |
| "loss": 0.0914, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "eval_cer": 0.03200181346894927, | |
| "eval_loss": 0.1139749139547348, | |
| "eval_runtime": 67.5106, | |
| "eval_samples_per_second": 7.406, | |
| "eval_steps_per_second": 0.933, | |
| "eval_wer": 0.09871802112159034, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 0.00015186424474187378, | |
| "loss": 0.0852, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "eval_cer": 0.03156815785064506, | |
| "eval_loss": 0.12490493804216385, | |
| "eval_runtime": 67.225, | |
| "eval_samples_per_second": 7.438, | |
| "eval_steps_per_second": 0.937, | |
| "eval_wer": 0.09561190489636867, | |
| "step": 3685 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 0.0001327437858508604, | |
| "loss": 0.0777, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "eval_cer": 0.03226792032563595, | |
| "eval_loss": 0.11327209323644638, | |
| "eval_runtime": 67.468, | |
| "eval_samples_per_second": 7.411, | |
| "eval_steps_per_second": 0.934, | |
| "eval_wer": 0.09532953069407579, | |
| "step": 4020 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "eval_cer": 0.03157801366015198, | |
| "eval_loss": 0.1231861561536789, | |
| "eval_runtime": 67.6103, | |
| "eval_samples_per_second": 7.395, | |
| "eval_steps_per_second": 0.932, | |
| "eval_wer": 0.09521658101315864, | |
| "step": 4355 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "learning_rate": 0.00011362332695984703, | |
| "loss": 0.0714, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "eval_cer": 0.032208785468594466, | |
| "eval_loss": 0.13348866999149323, | |
| "eval_runtime": 67.4714, | |
| "eval_samples_per_second": 7.411, | |
| "eval_steps_per_second": 0.934, | |
| "eval_wer": 0.09555543005591009, | |
| "step": 4690 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "learning_rate": 9.450286806883364e-05, | |
| "loss": 0.0631, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "eval_cer": 0.03205109251648384, | |
| "eval_loss": 0.12228710949420929, | |
| "eval_runtime": 67.5997, | |
| "eval_samples_per_second": 7.396, | |
| "eval_steps_per_second": 0.932, | |
| "eval_wer": 0.09521658101315864, | |
| "step": 5025 | |
| }, | |
| { | |
| "epoch": 6.14, | |
| "learning_rate": 7.543021032504778e-05, | |
| "loss": 0.0621, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "eval_cer": 0.032080659945004586, | |
| "eval_loss": 0.1337049901485443, | |
| "eval_runtime": 67.7409, | |
| "eval_samples_per_second": 7.381, | |
| "eval_steps_per_second": 0.93, | |
| "eval_wer": 0.09504715649178291, | |
| "step": 5360 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "learning_rate": 5.6309751434034414e-05, | |
| "loss": 0.0538, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "eval_cer": 0.03255373880133645, | |
| "eval_loss": 0.13896578550338745, | |
| "eval_runtime": 67.8406, | |
| "eval_samples_per_second": 7.37, | |
| "eval_steps_per_second": 0.929, | |
| "eval_wer": 0.09431298356582142, | |
| "step": 5695 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 3.718929254302103e-05, | |
| "loss": 0.0513, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "eval_cer": 0.03282970146753004, | |
| "eval_loss": 0.1389976292848587, | |
| "eval_runtime": 68.1201, | |
| "eval_samples_per_second": 7.34, | |
| "eval_steps_per_second": 0.925, | |
| "eval_wer": 0.09549895521545151, | |
| "step": 6030 | |
| }, | |
| { | |
| "epoch": 7.51, | |
| "eval_cer": 0.03255373880133645, | |
| "eval_loss": 0.1410442739725113, | |
| "eval_runtime": 67.6146, | |
| "eval_samples_per_second": 7.395, | |
| "eval_steps_per_second": 0.932, | |
| "eval_wer": 0.09470830744903146, | |
| "step": 6365 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "learning_rate": 1.8068833652007646e-05, | |
| "loss": 0.0442, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "eval_cer": 0.03214965061155298, | |
| "eval_loss": 0.1414366066455841, | |
| "eval_runtime": 68.0581, | |
| "eval_samples_per_second": 7.347, | |
| "eval_steps_per_second": 0.926, | |
| "eval_wer": 0.09453888292765573, | |
| "step": 6700 | |
| } | |
| ], | |
| "max_steps": 6776, | |
| "num_train_epochs": 8, | |
| "total_flos": 2.1224029808181802e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |