| { | |
| "best_metric": 0.6205869842620162, | |
| "best_model_checkpoint": "./whisper-base-full/checkpoint-1720", | |
| "epoch": 9.99883855981417, | |
| "global_step": 2150, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.600000000000001e-07, | |
| "loss": 3.739, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1.9600000000000003e-06, | |
| "loss": 3.3495, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 2.96e-06, | |
| "loss": 2.9472, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 3.96e-06, | |
| "loss": 2.5951, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.960000000000001e-06, | |
| "loss": 2.2619, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 5.9600000000000005e-06, | |
| "loss": 2.1134, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 6.96e-06, | |
| "loss": 2.0152, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 7.960000000000002e-06, | |
| "loss": 1.906, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.8213512897491455, | |
| "eval_runtime": 91.1825, | |
| "eval_samples_per_second": 8.39, | |
| "eval_steps_per_second": 1.053, | |
| "eval_wer": 4.169076988515525, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 8.96e-06, | |
| "loss": 1.7895, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 9.960000000000001e-06, | |
| "loss": 1.656, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1.0960000000000002e-05, | |
| "loss": 1.5357, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 1.196e-05, | |
| "loss": 1.5254, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 1.2960000000000001e-05, | |
| "loss": 1.4317, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 1.396e-05, | |
| "loss": 1.384, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 1.496e-05, | |
| "loss": 1.3297, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 1.5960000000000003e-05, | |
| "loss": 1.2496, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 1.696e-05, | |
| "loss": 1.21, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.241098403930664, | |
| "eval_runtime": 62.8989, | |
| "eval_samples_per_second": 12.162, | |
| "eval_steps_per_second": 1.526, | |
| "eval_wer": 1.553806890684815, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 1.796e-05, | |
| "loss": 1.0152, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 1.896e-05, | |
| "loss": 0.9099, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 1.9960000000000002e-05, | |
| "loss": 0.8608, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 1.970909090909091e-05, | |
| "loss": 0.8557, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 1.940606060606061e-05, | |
| "loss": 0.7938, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 1.9103030303030305e-05, | |
| "loss": 0.7732, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 1.88e-05, | |
| "loss": 0.7453, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 1.8496969696969698e-05, | |
| "loss": 0.7453, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.8530890941619873, | |
| "eval_runtime": 54.5023, | |
| "eval_samples_per_second": 14.036, | |
| "eval_steps_per_second": 1.761, | |
| "eval_wer": 0.8519778817524458, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 1.8193939393939396e-05, | |
| "loss": 0.6502, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 1.789090909090909e-05, | |
| "loss": 0.4396, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 1.758787878787879e-05, | |
| "loss": 0.4515, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 1.7284848484848485e-05, | |
| "loss": 0.4422, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 1.6981818181818183e-05, | |
| "loss": 0.4381, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 1.6678787878787882e-05, | |
| "loss": 0.3936, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 1.6375757575757577e-05, | |
| "loss": 0.4144, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 1.6072727272727272e-05, | |
| "loss": 0.4026, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 1.576969696969697e-05, | |
| "loss": 0.3933, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.6789121031761169, | |
| "eval_runtime": 56.9416, | |
| "eval_samples_per_second": 13.435, | |
| "eval_steps_per_second": 1.686, | |
| "eval_wer": 0.9025946405784773, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 1.546666666666667e-05, | |
| "loss": 0.3063, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 1.5163636363636366e-05, | |
| "loss": 0.2287, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 1.4860606060606062e-05, | |
| "loss": 0.2192, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 4.42, | |
| "learning_rate": 1.4569696969696972e-05, | |
| "loss": 0.2235, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "learning_rate": 1.4266666666666668e-05, | |
| "loss": 0.2162, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 1.3963636363636365e-05, | |
| "loss": 0.2236, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 1.366060606060606e-05, | |
| "loss": 0.2238, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 1.3357575757575759e-05, | |
| "loss": 0.2287, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 1.3054545454545455e-05, | |
| "loss": 0.1939, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.5971026420593262, | |
| "eval_runtime": 53.1774, | |
| "eval_samples_per_second": 14.386, | |
| "eval_steps_per_second": 1.805, | |
| "eval_wer": 0.6712037430880476, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 1.2751515151515152e-05, | |
| "loss": 0.1229, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 5.23, | |
| "learning_rate": 1.244848484848485e-05, | |
| "loss": 0.1179, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "learning_rate": 1.2145454545454547e-05, | |
| "loss": 0.1132, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "learning_rate": 1.1842424242424244e-05, | |
| "loss": 0.1196, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 5.58, | |
| "learning_rate": 1.153939393939394e-05, | |
| "loss": 0.1265, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 1.1236363636363636e-05, | |
| "loss": 0.1058, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 1.0933333333333334e-05, | |
| "loss": 0.1257, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 1.0630303030303031e-05, | |
| "loss": 0.1122, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.5692458748817444, | |
| "eval_runtime": 52.1137, | |
| "eval_samples_per_second": 14.679, | |
| "eval_steps_per_second": 1.842, | |
| "eval_wer": 0.6346235644406636, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 1.0327272727272728e-05, | |
| "loss": 0.0919, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 6.16, | |
| "learning_rate": 1.0024242424242424e-05, | |
| "loss": 0.0605, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "learning_rate": 9.721212121212123e-06, | |
| "loss": 0.0643, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "learning_rate": 9.418181818181818e-06, | |
| "loss": 0.0658, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "learning_rate": 9.115151515151516e-06, | |
| "loss": 0.0633, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "learning_rate": 8.812121212121213e-06, | |
| "loss": 0.0597, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 6.74, | |
| "learning_rate": 8.50909090909091e-06, | |
| "loss": 0.0598, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 6.86, | |
| "learning_rate": 8.206060606060607e-06, | |
| "loss": 0.06, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "learning_rate": 7.903030303030303e-06, | |
| "loss": 0.0472, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.5586017966270447, | |
| "eval_runtime": 52.7179, | |
| "eval_samples_per_second": 14.511, | |
| "eval_steps_per_second": 1.821, | |
| "eval_wer": 0.6875797532964696, | |
| "step": 1505 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "learning_rate": 7.600000000000001e-06, | |
| "loss": 0.0355, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "learning_rate": 7.296969696969698e-06, | |
| "loss": 0.0299, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "learning_rate": 6.993939393939394e-06, | |
| "loss": 0.0309, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 7.44, | |
| "learning_rate": 6.690909090909091e-06, | |
| "loss": 0.0339, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "learning_rate": 6.387878787878789e-06, | |
| "loss": 0.0322, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "learning_rate": 6.0848484848484855e-06, | |
| "loss": 0.0298, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "learning_rate": 5.781818181818181e-06, | |
| "loss": 0.03, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "learning_rate": 5.478787878787879e-06, | |
| "loss": 0.0282, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.5561943650245667, | |
| "eval_runtime": 52.0646, | |
| "eval_samples_per_second": 14.693, | |
| "eval_steps_per_second": 1.844, | |
| "eval_wer": 0.6205869842620162, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 5.1757575757575765e-06, | |
| "loss": 0.0282, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 8.14, | |
| "learning_rate": 4.872727272727273e-06, | |
| "loss": 0.017, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 8.26, | |
| "learning_rate": 4.56969696969697e-06, | |
| "loss": 0.018, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 8.37, | |
| "learning_rate": 4.266666666666668e-06, | |
| "loss": 0.0155, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 3.963636363636364e-06, | |
| "loss": 0.0163, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 8.6, | |
| "learning_rate": 3.660606060606061e-06, | |
| "loss": 0.0178, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 8.72, | |
| "learning_rate": 3.357575757575758e-06, | |
| "loss": 0.0165, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 8.84, | |
| "learning_rate": 3.054545454545455e-06, | |
| "loss": 0.0169, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 8.95, | |
| "learning_rate": 2.7515151515151517e-06, | |
| "loss": 0.0146, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.5539667010307312, | |
| "eval_runtime": 52.9557, | |
| "eval_samples_per_second": 14.446, | |
| "eval_steps_per_second": 1.813, | |
| "eval_wer": 0.672479795831561, | |
| "step": 1935 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "learning_rate": 2.4484848484848485e-06, | |
| "loss": 0.0121, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 9.19, | |
| "learning_rate": 2.1454545454545456e-06, | |
| "loss": 0.0101, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "learning_rate": 1.8424242424242426e-06, | |
| "loss": 0.0097, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 9.42, | |
| "learning_rate": 1.5393939393939395e-06, | |
| "loss": 0.0114, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 9.53, | |
| "learning_rate": 1.2363636363636365e-06, | |
| "loss": 0.0113, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 9.65, | |
| "learning_rate": 9.333333333333334e-07, | |
| "loss": 0.0117, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 9.77, | |
| "learning_rate": 6.303030303030304e-07, | |
| "loss": 0.0105, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "learning_rate": 3.2727272727272733e-07, | |
| "loss": 0.0102, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 2.4242424242424243e-08, | |
| "loss": 0.0102, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.5552583336830139, | |
| "eval_runtime": 52.1272, | |
| "eval_samples_per_second": 14.676, | |
| "eval_steps_per_second": 1.842, | |
| "eval_wer": 0.6288813270948532, | |
| "step": 2150 | |
| } | |
| ], | |
| "max_steps": 2150, | |
| "num_train_epochs": 10, | |
| "total_flos": 4.46411895570432e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |