| { | |
| "best_metric": 0.9836411609498681, | |
| "best_model_checkpoint": "wav2vec2-object/checkpoint-504", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 840, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 3.5714285714285714e-06, | |
| "loss": 2.3951, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 7.142857142857143e-06, | |
| "loss": 2.389, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 1.0714285714285714e-05, | |
| "loss": 2.3731, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 2.3439, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 2.2776, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 2.1428571428571428e-05, | |
| "loss": 2.1571, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.92, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 1.6367, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.933509234828496, | |
| "eval_loss": 1.284836769104004, | |
| "eval_runtime": 11.447, | |
| "eval_samples_per_second": 165.546, | |
| "eval_steps_per_second": 5.242, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 2.9761904761904762e-05, | |
| "loss": 1.377, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 2.9365079365079366e-05, | |
| "loss": 1.1976, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 2.8968253968253967e-05, | |
| "loss": 1.0806, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 0.9608, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 2.8174603174603175e-05, | |
| "loss": 0.8859, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.8128, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 2.738095238095238e-05, | |
| "loss": 0.7591, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 2.6984126984126984e-05, | |
| "loss": 0.7071, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9683377308707124, | |
| "eval_loss": 0.5107668042182922, | |
| "eval_runtime": 11.2841, | |
| "eval_samples_per_second": 167.936, | |
| "eval_steps_per_second": 5.317, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 2.6587301587301588e-05, | |
| "loss": 0.6489, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 2.6190476190476192e-05, | |
| "loss": 0.6104, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 2.5793650793650793e-05, | |
| "loss": 0.5735, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 2.5396825396825397e-05, | |
| "loss": 0.5251, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.4951, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 2.4603174603174605e-05, | |
| "loss": 0.4926, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 2.4206349206349206e-05, | |
| "loss": 0.4804, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.4279, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 2.3412698412698414e-05, | |
| "loss": 0.4246, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.97467018469657, | |
| "eval_loss": 0.30182337760925293, | |
| "eval_runtime": 11.1208, | |
| "eval_samples_per_second": 170.401, | |
| "eval_steps_per_second": 5.395, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 2.301587301587302e-05, | |
| "loss": 0.3991, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 2.261904761904762e-05, | |
| "loss": 0.381, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 2.222222222222222e-05, | |
| "loss": 0.3748, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 2.1825396825396827e-05, | |
| "loss": 0.3619, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 2.1428571428571428e-05, | |
| "loss": 0.3416, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 2.1031746031746032e-05, | |
| "loss": 0.3261, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 2.0634920634920633e-05, | |
| "loss": 0.2987, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 2.023809523809524e-05, | |
| "loss": 0.3173, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9767810026385224, | |
| "eval_loss": 0.2097773551940918, | |
| "eval_runtime": 10.8229, | |
| "eval_samples_per_second": 175.091, | |
| "eval_steps_per_second": 5.544, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 1.984126984126984e-05, | |
| "loss": 0.2915, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.2797, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "learning_rate": 1.9047619047619046e-05, | |
| "loss": 0.2736, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 1.8650793650793654e-05, | |
| "loss": 0.285, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 1.8253968253968254e-05, | |
| "loss": 0.2456, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 1.7896825396825394e-05, | |
| "loss": 0.2508, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 1.7500000000000002e-05, | |
| "loss": 0.2649, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 1.7103174603174603e-05, | |
| "loss": 0.2643, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 1.6706349206349207e-05, | |
| "loss": 0.2585, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9783641160949869, | |
| "eval_loss": 0.1684747189283371, | |
| "eval_runtime": 11.0873, | |
| "eval_samples_per_second": 170.917, | |
| "eval_steps_per_second": 5.412, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 1.6309523809523807e-05, | |
| "loss": 0.2413, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 1.5912698412698415e-05, | |
| "loss": 0.2137, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "learning_rate": 1.5515873015873016e-05, | |
| "loss": 0.2007, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "learning_rate": 1.511904761904762e-05, | |
| "loss": 0.2259, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 1.4722222222222222e-05, | |
| "loss": 0.1921, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "learning_rate": 1.4325396825396825e-05, | |
| "loss": 0.1951, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "learning_rate": 1.3928571428571429e-05, | |
| "loss": 0.2063, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 5.95, | |
| "learning_rate": 1.3531746031746031e-05, | |
| "loss": 0.192, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9836411609498681, | |
| "eval_loss": 0.12848743796348572, | |
| "eval_runtime": 11.2118, | |
| "eval_samples_per_second": 169.018, | |
| "eval_steps_per_second": 5.351, | |
| "step": 504 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 1.3134920634920635e-05, | |
| "loss": 0.2192, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "learning_rate": 1.2738095238095238e-05, | |
| "loss": 0.2063, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "learning_rate": 1.2341269841269842e-05, | |
| "loss": 0.2053, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "learning_rate": 1.1944444444444444e-05, | |
| "loss": 0.2138, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "learning_rate": 1.1547619047619048e-05, | |
| "loss": 0.1777, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 1.1150793650793651e-05, | |
| "loss": 0.1833, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "learning_rate": 1.0753968253968255e-05, | |
| "loss": 0.1803, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "learning_rate": 1.0357142857142857e-05, | |
| "loss": 0.1772, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9810026385224274, | |
| "eval_loss": 0.12096092104911804, | |
| "eval_runtime": 10.7065, | |
| "eval_samples_per_second": 176.995, | |
| "eval_steps_per_second": 5.604, | |
| "step": 588 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 9.960317460317462e-06, | |
| "loss": 0.1887, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 9.563492063492064e-06, | |
| "loss": 0.1695, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "learning_rate": 9.166666666666668e-06, | |
| "loss": 0.1767, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "learning_rate": 8.76984126984127e-06, | |
| "loss": 0.1694, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "learning_rate": 8.373015873015875e-06, | |
| "loss": 0.1722, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 7.62, | |
| "learning_rate": 7.976190476190477e-06, | |
| "loss": 0.1525, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 7.57936507936508e-06, | |
| "loss": 0.1517, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "learning_rate": 7.182539682539683e-06, | |
| "loss": 0.1634, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 7.98, | |
| "learning_rate": 6.785714285714286e-06, | |
| "loss": 0.1435, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9804749340369393, | |
| "eval_loss": 0.1141170933842659, | |
| "eval_runtime": 10.8149, | |
| "eval_samples_per_second": 175.222, | |
| "eval_steps_per_second": 5.548, | |
| "step": 672 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "learning_rate": 6.388888888888889e-06, | |
| "loss": 0.1874, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 8.21, | |
| "learning_rate": 5.992063492063493e-06, | |
| "loss": 0.1268, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 5.595238095238095e-06, | |
| "loss": 0.1529, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 8.45, | |
| "learning_rate": 5.198412698412698e-06, | |
| "loss": 0.1356, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 8.57, | |
| "learning_rate": 4.801587301587302e-06, | |
| "loss": 0.1549, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 8.69, | |
| "learning_rate": 4.404761904761905e-06, | |
| "loss": 0.1775, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 8.81, | |
| "learning_rate": 4.007936507936508e-06, | |
| "loss": 0.1448, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "learning_rate": 3.611111111111111e-06, | |
| "loss": 0.1425, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9836411609498681, | |
| "eval_loss": 0.10937827080488205, | |
| "eval_runtime": 11.2971, | |
| "eval_samples_per_second": 167.742, | |
| "eval_steps_per_second": 5.311, | |
| "step": 756 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 3.2142857142857143e-06, | |
| "loss": 0.1726, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 9.17, | |
| "learning_rate": 2.8174603174603176e-06, | |
| "loss": 0.1471, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 9.29, | |
| "learning_rate": 2.4206349206349204e-06, | |
| "loss": 0.151, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 9.4, | |
| "learning_rate": 2.0238095238095237e-06, | |
| "loss": 0.1586, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 1.626984126984127e-06, | |
| "loss": 0.1112, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 9.64, | |
| "learning_rate": 1.23015873015873e-06, | |
| "loss": 0.1315, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 9.76, | |
| "learning_rate": 8.333333333333333e-07, | |
| "loss": 0.1649, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "learning_rate": 4.365079365079365e-07, | |
| "loss": 0.1458, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 3.968253968253968e-08, | |
| "loss": 0.1398, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9825857519788919, | |
| "eval_loss": 0.1077321395277977, | |
| "eval_runtime": 11.3409, | |
| "eval_samples_per_second": 167.094, | |
| "eval_steps_per_second": 5.291, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 840, | |
| "total_flos": 1.170126007537755e+18, | |
| "train_loss": 0.5051058745100385, | |
| "train_runtime": 1169.0466, | |
| "train_samples_per_second": 91.81, | |
| "train_steps_per_second": 0.719 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 840, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 1.170126007537755e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |