| { | |
| "best_metric": 29.213729575611364, | |
| "best_model_checkpoint": "./whisper-small-vietnamese\\checkpoint-1000", | |
| "epoch": 5.747126436781609, | |
| "eval_steps": 1000, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 118.77095794677734, | |
| "learning_rate": 4.2000000000000006e-07, | |
| "loss": 5.0409, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 29.974971771240234, | |
| "learning_rate": 9.200000000000001e-07, | |
| "loss": 4.0132, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 25.654932022094727, | |
| "learning_rate": 1.42e-06, | |
| "loss": 2.9199, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 16.862260818481445, | |
| "learning_rate": 1.9200000000000003e-06, | |
| "loss": 1.7066, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 13.715407371520996, | |
| "learning_rate": 2.42e-06, | |
| "loss": 1.2867, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 16.711305618286133, | |
| "learning_rate": 2.92e-06, | |
| "loss": 1.1578, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 14.38102912902832, | |
| "learning_rate": 3.4200000000000007e-06, | |
| "loss": 1.0035, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 15.285762786865234, | |
| "learning_rate": 3.920000000000001e-06, | |
| "loss": 0.8052, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "grad_norm": 8.91514778137207, | |
| "learning_rate": 4.42e-06, | |
| "loss": 0.286, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 7.789880275726318, | |
| "learning_rate": 4.92e-06, | |
| "loss": 0.2305, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 8.093725204467773, | |
| "learning_rate": 5.420000000000001e-06, | |
| "loss": 0.2184, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 4.7622785568237305, | |
| "learning_rate": 5.92e-06, | |
| "loss": 0.2219, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 7.469940662384033, | |
| "learning_rate": 6.42e-06, | |
| "loss": 0.2067, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 4.024804592132568, | |
| "learning_rate": 6.92e-06, | |
| "loss": 0.2128, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 4.569382667541504, | |
| "learning_rate": 7.420000000000001e-06, | |
| "loss": 0.1294, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 5.233477592468262, | |
| "learning_rate": 7.92e-06, | |
| "loss": 0.1217, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "grad_norm": 6.401954174041748, | |
| "learning_rate": 8.42e-06, | |
| "loss": 0.1272, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "grad_norm": 4.058222770690918, | |
| "learning_rate": 8.920000000000001e-06, | |
| "loss": 0.1188, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "grad_norm": 4.228504657745361, | |
| "learning_rate": 9.42e-06, | |
| "loss": 0.1294, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "grad_norm": 5.160878658294678, | |
| "learning_rate": 9.920000000000002e-06, | |
| "loss": 0.1241, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "grad_norm": 3.9591662883758545, | |
| "learning_rate": 9.940000000000001e-06, | |
| "loss": 0.1264, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "grad_norm": 4.170319080352783, | |
| "learning_rate": 9.86857142857143e-06, | |
| "loss": 0.0793, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "grad_norm": 3.4333674907684326, | |
| "learning_rate": 9.797142857142858e-06, | |
| "loss": 0.0643, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "grad_norm": 4.089048862457275, | |
| "learning_rate": 9.725714285714287e-06, | |
| "loss": 0.0668, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "grad_norm": 3.6274781227111816, | |
| "learning_rate": 9.654285714285716e-06, | |
| "loss": 0.069, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "grad_norm": 3.2134461402893066, | |
| "learning_rate": 9.582857142857143e-06, | |
| "loss": 0.0759, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "grad_norm": 3.4892189502716064, | |
| "learning_rate": 9.511428571428572e-06, | |
| "loss": 0.0698, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "grad_norm": 2.3172402381896973, | |
| "learning_rate": 9.440000000000001e-06, | |
| "loss": 0.0651, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "grad_norm": 2.6511449813842773, | |
| "learning_rate": 9.368571428571428e-06, | |
| "loss": 0.0373, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "grad_norm": 3.1017539501190186, | |
| "learning_rate": 9.297142857142857e-06, | |
| "loss": 0.0415, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "grad_norm": 3.3337509632110596, | |
| "learning_rate": 9.225714285714286e-06, | |
| "loss": 0.0388, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "grad_norm": 2.3944547176361084, | |
| "learning_rate": 9.154285714285715e-06, | |
| "loss": 0.0411, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "grad_norm": 3.1757237911224365, | |
| "learning_rate": 9.082857142857143e-06, | |
| "loss": 0.0381, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "grad_norm": 4.307478427886963, | |
| "learning_rate": 9.011428571428572e-06, | |
| "loss": 0.0434, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "grad_norm": 1.4969714879989624, | |
| "learning_rate": 8.94e-06, | |
| "loss": 0.0384, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "grad_norm": 1.416281819343567, | |
| "learning_rate": 8.86857142857143e-06, | |
| "loss": 0.0207, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "grad_norm": 4.1161322593688965, | |
| "learning_rate": 8.797142857142857e-06, | |
| "loss": 0.0222, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "grad_norm": 0.9343141317367554, | |
| "learning_rate": 8.725714285714286e-06, | |
| "loss": 0.0213, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "grad_norm": 3.1798384189605713, | |
| "learning_rate": 8.654285714285715e-06, | |
| "loss": 0.0239, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "grad_norm": 1.7415159940719604, | |
| "learning_rate": 8.582857142857144e-06, | |
| "loss": 0.0243, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "eval_loss": 0.6310169696807861, | |
| "eval_runtime": 482.8872, | |
| "eval_samples_per_second": 2.562, | |
| "eval_steps_per_second": 0.321, | |
| "eval_wer": 29.213729575611364, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 4000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 23, | |
| "save_steps": 1000, | |
| "total_flos": 4.6014942081024e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |