| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 29.677419354838708, | |
| "global_step": 9200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 7.3425, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_loss": 3.277569532394409, | |
| "eval_runtime": 138.4344, | |
| "eval_samples_per_second": 10.994, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 0.0002897727272727273, | |
| "loss": 2.6649, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "eval_loss": 1.1241918802261353, | |
| "eval_runtime": 115.4749, | |
| "eval_samples_per_second": 13.18, | |
| "eval_wer": 0.8526294539214697, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 0.0002761363636363636, | |
| "loss": 0.7589, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "eval_loss": 0.69745272397995, | |
| "eval_runtime": 137.7134, | |
| "eval_samples_per_second": 11.052, | |
| "eval_wer": 0.6960734833955788, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 0.0002625, | |
| "loss": 0.476, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "eval_loss": 0.6482455730438232, | |
| "eval_runtime": 115.5054, | |
| "eval_samples_per_second": 13.177, | |
| "eval_wer": 0.6553951751286968, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 0.00024886363636363637, | |
| "loss": 0.3479, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "eval_loss": 0.6039410829544067, | |
| "eval_runtime": 115.7857, | |
| "eval_samples_per_second": 13.145, | |
| "eval_wer": 0.5894821843141214, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 0.0002352272727272727, | |
| "loss": 0.2783, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "eval_loss": 0.6183858513832092, | |
| "eval_runtime": 115.9363, | |
| "eval_samples_per_second": 13.128, | |
| "eval_wer": 0.5855455738366812, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 0.00022159090909090908, | |
| "loss": 0.2204, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "eval_loss": 0.5993764996528625, | |
| "eval_runtime": 137.9021, | |
| "eval_samples_per_second": 11.037, | |
| "eval_wer": 0.550721711920864, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "learning_rate": 0.00020795454545454546, | |
| "loss": 0.1881, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "eval_loss": 0.5736687779426575, | |
| "eval_runtime": 138.6384, | |
| "eval_samples_per_second": 10.978, | |
| "eval_wer": 0.537801554456445, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "learning_rate": 0.00019431818181818179, | |
| "loss": 0.1639, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "eval_loss": 0.5788838267326355, | |
| "eval_runtime": 139.0007, | |
| "eval_samples_per_second": 10.95, | |
| "eval_wer": 0.5217522963561119, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 0.00018068181818181817, | |
| "loss": 0.1464, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "eval_loss": 0.5988554358482361, | |
| "eval_runtime": 116.0695, | |
| "eval_samples_per_second": 13.113, | |
| "eval_wer": 0.5138790754012315, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 14.19, | |
| "learning_rate": 0.00016704545454545452, | |
| "loss": 0.1222, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 14.19, | |
| "eval_loss": 0.6150318384170532, | |
| "eval_runtime": 140.7397, | |
| "eval_samples_per_second": 10.814, | |
| "eval_wer": 0.5012617341273847, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 15.48, | |
| "learning_rate": 0.0001534090909090909, | |
| "loss": 0.112, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 15.48, | |
| "eval_loss": 0.6471191644668579, | |
| "eval_runtime": 138.3313, | |
| "eval_samples_per_second": 11.003, | |
| "eval_wer": 0.5148884627031391, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 16.77, | |
| "learning_rate": 0.00013977272727272726, | |
| "loss": 0.105, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 16.77, | |
| "eval_loss": 0.6167843341827393, | |
| "eval_runtime": 116.3058, | |
| "eval_samples_per_second": 13.086, | |
| "eval_wer": 0.49237912587059657, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 18.06, | |
| "learning_rate": 0.00012613636363636364, | |
| "loss": 0.0936, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 18.06, | |
| "eval_loss": 0.6400735378265381, | |
| "eval_runtime": 138.5841, | |
| "eval_samples_per_second": 10.983, | |
| "eval_wer": 0.4856162309478147, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "learning_rate": 0.0001125, | |
| "loss": 0.0849, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "eval_loss": 0.6149299740791321, | |
| "eval_runtime": 116.9081, | |
| "eval_samples_per_second": 13.019, | |
| "eval_wer": 0.47875239729484204, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 20.65, | |
| "learning_rate": 9.886363636363635e-05, | |
| "loss": 0.079, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 20.65, | |
| "eval_loss": 0.6410804986953735, | |
| "eval_runtime": 139.105, | |
| "eval_samples_per_second": 10.941, | |
| "eval_wer": 0.48067023316846674, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "learning_rate": 8.522727272727273e-05, | |
| "loss": 0.0752, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "eval_loss": 0.6123934388160706, | |
| "eval_runtime": 138.8056, | |
| "eval_samples_per_second": 10.965, | |
| "eval_wer": 0.4773392550721712, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 23.23, | |
| "learning_rate": 7.159090909090909e-05, | |
| "loss": 0.0688, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 23.23, | |
| "eval_loss": 0.6324551105499268, | |
| "eval_runtime": 137.9116, | |
| "eval_samples_per_second": 11.036, | |
| "eval_wer": 0.47198950237206017, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 24.52, | |
| "learning_rate": 5.795454545454545e-05, | |
| "loss": 0.0659, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 24.52, | |
| "eval_loss": 0.6281149387359619, | |
| "eval_runtime": 139.9793, | |
| "eval_samples_per_second": 10.873, | |
| "eval_wer": 0.464923791258706, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 25.81, | |
| "learning_rate": 4.431818181818182e-05, | |
| "loss": 0.0582, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 25.81, | |
| "eval_loss": 0.6326279640197754, | |
| "eval_runtime": 139.8038, | |
| "eval_samples_per_second": 10.887, | |
| "eval_wer": 0.4604824871303119, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 27.1, | |
| "learning_rate": 3.068181818181818e-05, | |
| "loss": 0.0551, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 27.1, | |
| "eval_loss": 0.6271815299987793, | |
| "eval_runtime": 139.3604, | |
| "eval_samples_per_second": 10.921, | |
| "eval_wer": 0.46300595538508127, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 28.39, | |
| "learning_rate": 1.7045454545454543e-05, | |
| "loss": 0.0508, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 28.39, | |
| "eval_loss": 0.6332981586456299, | |
| "eval_runtime": 139.0366, | |
| "eval_samples_per_second": 10.947, | |
| "eval_wer": 0.4594730998284042, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 29.68, | |
| "learning_rate": 3.4090909090909087e-06, | |
| "loss": 0.0517, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 29.68, | |
| "eval_loss": 0.6157954931259155, | |
| "eval_runtime": 140.2207, | |
| "eval_samples_per_second": 10.854, | |
| "eval_wer": 0.45119612395276065, | |
| "step": 9200 | |
| } | |
| ], | |
| "max_steps": 9300, | |
| "num_train_epochs": 30, | |
| "total_flos": 1.4085584505165812e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |