| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 99.99602649006623, | |
| "global_step": 18800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.950000000000001e-06, | |
| "loss": 13.7083, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 9.950000000000001e-06, | |
| "loss": 5.8958, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 1.4950000000000001e-05, | |
| "loss": 4.115, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 1.995e-05, | |
| "loss": 3.6068, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 2.495e-05, | |
| "loss": 3.2892, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "eval_loss": 3.241530179977417, | |
| "eval_runtime": 149.1051, | |
| "eval_samples_per_second": 18.403, | |
| "eval_steps_per_second": 2.3, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 2.995e-05, | |
| "loss": 3.2316, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 3.495e-05, | |
| "loss": 3.1529, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 3.995e-05, | |
| "loss": 3.1279, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 4.495e-05, | |
| "loss": 3.0647, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "learning_rate": 4.995e-05, | |
| "loss": 2.9206, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "eval_loss": 2.4381155967712402, | |
| "eval_runtime": 158.7474, | |
| "eval_samples_per_second": 17.285, | |
| "eval_steps_per_second": 2.161, | |
| "eval_wer": 1.0055832685002974, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 5.495e-05, | |
| "loss": 2.2804, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "learning_rate": 5.995000000000001e-05, | |
| "loss": 1.8138, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "learning_rate": 6.49e-05, | |
| "loss": 1.6251, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 7.45, | |
| "learning_rate": 6.99e-05, | |
| "loss": 1.5483, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "learning_rate": 7.49e-05, | |
| "loss": 1.4909, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "eval_loss": 0.5427731275558472, | |
| "eval_runtime": 150.2181, | |
| "eval_samples_per_second": 18.267, | |
| "eval_steps_per_second": 2.283, | |
| "eval_wer": 0.6704956294906411, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 8.51, | |
| "learning_rate": 7.99e-05, | |
| "loss": 1.4577, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 8.49e-05, | |
| "loss": 1.4197, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "learning_rate": 8.985e-05, | |
| "loss": 1.3877, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 10.11, | |
| "learning_rate": 9.485e-05, | |
| "loss": 1.3704, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 10.64, | |
| "learning_rate": 9.985000000000001e-05, | |
| "loss": 1.3395, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 10.64, | |
| "eval_loss": 0.4207160472869873, | |
| "eval_runtime": 153.5378, | |
| "eval_samples_per_second": 17.872, | |
| "eval_steps_per_second": 2.234, | |
| "eval_wer": 0.5995148963434168, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 11.17, | |
| "learning_rate": 9.942261904761904e-05, | |
| "loss": 1.3349, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 11.7, | |
| "learning_rate": 9.882738095238095e-05, | |
| "loss": 1.3064, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 12.23, | |
| "learning_rate": 9.823214285714287e-05, | |
| "loss": 1.3132, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 12.76, | |
| "learning_rate": 9.763690476190477e-05, | |
| "loss": 1.2813, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 13.3, | |
| "learning_rate": 9.704166666666668e-05, | |
| "loss": 1.2718, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 13.3, | |
| "eval_loss": 0.37430423498153687, | |
| "eval_runtime": 148.1352, | |
| "eval_samples_per_second": 18.524, | |
| "eval_steps_per_second": 2.315, | |
| "eval_wer": 0.5648254084481259, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 13.83, | |
| "learning_rate": 9.644642857142857e-05, | |
| "loss": 1.247, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 14.36, | |
| "learning_rate": 9.585119047619047e-05, | |
| "loss": 1.2446, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 14.89, | |
| "learning_rate": 9.525595238095239e-05, | |
| "loss": 1.2404, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 15.42, | |
| "learning_rate": 9.46607142857143e-05, | |
| "loss": 1.234, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 15.95, | |
| "learning_rate": 9.40654761904762e-05, | |
| "loss": 1.1798, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 15.95, | |
| "eval_loss": 0.32250717282295227, | |
| "eval_runtime": 150.8327, | |
| "eval_samples_per_second": 18.192, | |
| "eval_steps_per_second": 2.274, | |
| "eval_wer": 0.4927005629032996, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 16.49, | |
| "learning_rate": 9.34702380952381e-05, | |
| "loss": 1.1849, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 17.02, | |
| "learning_rate": 9.2875e-05, | |
| "loss": 1.1931, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 17.55, | |
| "learning_rate": 9.227976190476191e-05, | |
| "loss": 1.1628, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 18.08, | |
| "learning_rate": 9.168452380952382e-05, | |
| "loss": 1.1743, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 18.61, | |
| "learning_rate": 9.108928571428572e-05, | |
| "loss": 1.1392, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 18.61, | |
| "eval_loss": 0.3096984922885895, | |
| "eval_runtime": 151.6787, | |
| "eval_samples_per_second": 18.091, | |
| "eval_steps_per_second": 2.261, | |
| "eval_wer": 0.4626790535902247, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 19.15, | |
| "learning_rate": 9.049404761904763e-05, | |
| "loss": 1.1458, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 19.68, | |
| "learning_rate": 8.989880952380953e-05, | |
| "loss": 1.1323, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 20.21, | |
| "learning_rate": 8.930357142857143e-05, | |
| "loss": 1.1293, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 20.74, | |
| "learning_rate": 8.870833333333334e-05, | |
| "loss": 1.1179, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 21.28, | |
| "learning_rate": 8.811309523809524e-05, | |
| "loss": 1.1143, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 21.28, | |
| "eval_loss": 0.29957136511802673, | |
| "eval_runtime": 147.8067, | |
| "eval_samples_per_second": 18.565, | |
| "eval_steps_per_second": 2.321, | |
| "eval_wer": 0.450459933183836, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 21.81, | |
| "learning_rate": 8.751785714285715e-05, | |
| "loss": 1.1121, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 22.34, | |
| "learning_rate": 8.692261904761905e-05, | |
| "loss": 1.1049, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 22.87, | |
| "learning_rate": 8.632738095238096e-05, | |
| "loss": 1.092, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 23.4, | |
| "learning_rate": 8.573214285714286e-05, | |
| "loss": 1.0936, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 23.93, | |
| "learning_rate": 8.513690476190477e-05, | |
| "loss": 1.0923, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 23.93, | |
| "eval_loss": 0.2841183543205261, | |
| "eval_runtime": 145.4218, | |
| "eval_samples_per_second": 18.869, | |
| "eval_steps_per_second": 2.359, | |
| "eval_wer": 0.4229097066495813, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 24.47, | |
| "learning_rate": 8.454166666666667e-05, | |
| "loss": 1.0748, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 8.395238095238095e-05, | |
| "loss": 1.0786, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 25.53, | |
| "learning_rate": 8.335714285714286e-05, | |
| "loss": 1.0685, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 26.06, | |
| "learning_rate": 8.276190476190476e-05, | |
| "loss": 1.0681, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 26.59, | |
| "learning_rate": 8.216666666666667e-05, | |
| "loss": 1.0516, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 26.59, | |
| "eval_loss": 0.2705024182796478, | |
| "eval_runtime": 156.97, | |
| "eval_samples_per_second": 17.481, | |
| "eval_steps_per_second": 2.185, | |
| "eval_wer": 0.4113312891858496, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 27.13, | |
| "learning_rate": 8.157142857142857e-05, | |
| "loss": 1.0631, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 27.66, | |
| "learning_rate": 8.097619047619049e-05, | |
| "loss": 1.0438, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 28.19, | |
| "learning_rate": 8.03809523809524e-05, | |
| "loss": 1.0437, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 28.72, | |
| "learning_rate": 7.978571428571429e-05, | |
| "loss": 1.0334, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 29.25, | |
| "learning_rate": 7.919047619047619e-05, | |
| "loss": 1.051, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 29.25, | |
| "eval_loss": 0.26215311884880066, | |
| "eval_runtime": 149.1886, | |
| "eval_samples_per_second": 18.393, | |
| "eval_steps_per_second": 2.299, | |
| "eval_wer": 0.4078074230012356, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 29.78, | |
| "learning_rate": 7.85952380952381e-05, | |
| "loss": 1.0209, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 30.32, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 1.0393, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 30.85, | |
| "learning_rate": 7.740476190476192e-05, | |
| "loss": 1.0175, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 31.38, | |
| "learning_rate": 7.680952380952381e-05, | |
| "loss": 1.0208, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 31.91, | |
| "learning_rate": 7.621428571428571e-05, | |
| "loss": 1.021, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 31.91, | |
| "eval_loss": 0.26111042499542236, | |
| "eval_runtime": 144.8158, | |
| "eval_samples_per_second": 18.948, | |
| "eval_steps_per_second": 2.369, | |
| "eval_wer": 0.40085121962381587, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 32.45, | |
| "learning_rate": 7.561904761904762e-05, | |
| "loss": 1.0091, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 32.97, | |
| "learning_rate": 7.502380952380953e-05, | |
| "loss": 1.0007, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 33.51, | |
| "learning_rate": 7.442857142857144e-05, | |
| "loss": 1.0141, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 34.04, | |
| "learning_rate": 7.383333333333333e-05, | |
| "loss": 1.0072, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 34.57, | |
| "learning_rate": 7.323809523809523e-05, | |
| "loss": 0.9886, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 34.57, | |
| "eval_loss": 0.24984091520309448, | |
| "eval_runtime": 146.0157, | |
| "eval_samples_per_second": 18.792, | |
| "eval_steps_per_second": 2.349, | |
| "eval_wer": 0.3920644364102329, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 35.11, | |
| "learning_rate": 7.264285714285715e-05, | |
| "loss": 0.9955, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 35.64, | |
| "learning_rate": 7.205357142857144e-05, | |
| "loss": 0.9811, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 36.17, | |
| "learning_rate": 7.145833333333334e-05, | |
| "loss": 0.987, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 36.7, | |
| "learning_rate": 7.086309523809524e-05, | |
| "loss": 0.9776, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 37.23, | |
| "learning_rate": 7.026785714285714e-05, | |
| "loss": 0.984, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 37.23, | |
| "eval_loss": 0.25214260816574097, | |
| "eval_runtime": 151.9558, | |
| "eval_samples_per_second": 18.058, | |
| "eval_steps_per_second": 2.257, | |
| "eval_wer": 0.38446753009015605, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 37.76, | |
| "learning_rate": 6.967261904761905e-05, | |
| "loss": 0.98, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 38.3, | |
| "learning_rate": 6.907738095238096e-05, | |
| "loss": 0.9762, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 38.83, | |
| "learning_rate": 6.848214285714286e-05, | |
| "loss": 0.9601, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 39.36, | |
| "learning_rate": 6.788690476190477e-05, | |
| "loss": 0.9633, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 39.89, | |
| "learning_rate": 6.729166666666667e-05, | |
| "loss": 0.9631, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 39.89, | |
| "eval_loss": 0.24125833809375763, | |
| "eval_runtime": 185.0035, | |
| "eval_samples_per_second": 14.832, | |
| "eval_steps_per_second": 1.854, | |
| "eval_wer": 0.3790673195734749, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 40.42, | |
| "learning_rate": 6.669642857142858e-05, | |
| "loss": 0.9653, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 40.95, | |
| "learning_rate": 6.610119047619048e-05, | |
| "loss": 0.9482, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 41.49, | |
| "learning_rate": 6.550595238095238e-05, | |
| "loss": 0.9547, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 42.02, | |
| "learning_rate": 6.491071428571429e-05, | |
| "loss": 0.9635, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 42.55, | |
| "learning_rate": 6.432142857142857e-05, | |
| "loss": 0.9353, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 42.55, | |
| "eval_loss": 0.23913756012916565, | |
| "eval_runtime": 145.0257, | |
| "eval_samples_per_second": 18.921, | |
| "eval_steps_per_second": 2.365, | |
| "eval_wer": 0.3611734016749806, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 43.08, | |
| "learning_rate": 6.372619047619049e-05, | |
| "loss": 0.9484, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 43.61, | |
| "learning_rate": 6.313095238095238e-05, | |
| "loss": 0.933, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 44.15, | |
| "learning_rate": 6.253571428571429e-05, | |
| "loss": 0.9315, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 44.68, | |
| "learning_rate": 6.194047619047619e-05, | |
| "loss": 0.9337, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 45.21, | |
| "learning_rate": 6.13452380952381e-05, | |
| "loss": 0.922, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 45.21, | |
| "eval_loss": 0.2362910658121109, | |
| "eval_runtime": 152.2446, | |
| "eval_samples_per_second": 18.024, | |
| "eval_steps_per_second": 2.253, | |
| "eval_wer": 0.3570545970436136, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 45.74, | |
| "learning_rate": 6.0750000000000006e-05, | |
| "loss": 0.9274, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 46.28, | |
| "learning_rate": 6.0154761904761904e-05, | |
| "loss": 0.9243, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 46.81, | |
| "learning_rate": 5.955952380952381e-05, | |
| "loss": 0.9148, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 47.34, | |
| "learning_rate": 5.896428571428572e-05, | |
| "loss": 0.9309, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 47.87, | |
| "learning_rate": 5.8369047619047624e-05, | |
| "loss": 0.9116, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 47.87, | |
| "eval_loss": 0.2284734547138214, | |
| "eval_runtime": 145.6764, | |
| "eval_samples_per_second": 18.836, | |
| "eval_steps_per_second": 2.355, | |
| "eval_wer": 0.366756670175278, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 48.4, | |
| "learning_rate": 5.777380952380953e-05, | |
| "loss": 0.908, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 48.93, | |
| "learning_rate": 5.7178571428571426e-05, | |
| "loss": 0.9045, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 49.47, | |
| "learning_rate": 5.658333333333333e-05, | |
| "loss": 0.9069, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 5.598809523809524e-05, | |
| "loss": 0.886, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 50.53, | |
| "learning_rate": 5.539285714285715e-05, | |
| "loss": 0.8951, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 50.53, | |
| "eval_loss": 0.22562462091445923, | |
| "eval_runtime": 146.467, | |
| "eval_samples_per_second": 18.735, | |
| "eval_steps_per_second": 2.342, | |
| "eval_wer": 0.3729348771223285, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 51.06, | |
| "learning_rate": 5.479761904761905e-05, | |
| "loss": 0.9011, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 51.59, | |
| "learning_rate": 5.420238095238096e-05, | |
| "loss": 0.8858, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 52.13, | |
| "learning_rate": 5.360714285714285e-05, | |
| "loss": 0.8955, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 52.66, | |
| "learning_rate": 5.3011904761904765e-05, | |
| "loss": 0.8747, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 53.19, | |
| "learning_rate": 5.241666666666667e-05, | |
| "loss": 0.8865, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 53.19, | |
| "eval_loss": 0.22283457219600677, | |
| "eval_runtime": 147.1995, | |
| "eval_samples_per_second": 18.641, | |
| "eval_steps_per_second": 2.33, | |
| "eval_wer": 0.3663447897121413, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 53.72, | |
| "learning_rate": 5.1821428571428574e-05, | |
| "loss": 0.8671, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 54.25, | |
| "learning_rate": 5.1226190476190485e-05, | |
| "loss": 0.8771, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 54.78, | |
| "learning_rate": 5.063690476190477e-05, | |
| "loss": 0.8777, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 55.32, | |
| "learning_rate": 5.0041666666666666e-05, | |
| "loss": 0.8805, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 55.85, | |
| "learning_rate": 4.944642857142857e-05, | |
| "loss": 0.8792, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 55.85, | |
| "eval_loss": 0.22211238741874695, | |
| "eval_runtime": 149.6455, | |
| "eval_samples_per_second": 18.337, | |
| "eval_steps_per_second": 2.292, | |
| "eval_wer": 0.3656125577776761, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 56.38, | |
| "learning_rate": 4.885119047619048e-05, | |
| "loss": 0.8805, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 56.91, | |
| "learning_rate": 4.8255952380952386e-05, | |
| "loss": 0.8653, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 57.45, | |
| "learning_rate": 4.7660714285714284e-05, | |
| "loss": 0.8749, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 57.97, | |
| "learning_rate": 4.7065476190476195e-05, | |
| "loss": 0.8713, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 58.51, | |
| "learning_rate": 4.64702380952381e-05, | |
| "loss": 0.8682, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 58.51, | |
| "eval_loss": 0.22277939319610596, | |
| "eval_runtime": 153.8771, | |
| "eval_samples_per_second": 17.832, | |
| "eval_steps_per_second": 2.229, | |
| "eval_wer": 0.3322960047595076, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 59.04, | |
| "learning_rate": 4.5875000000000004e-05, | |
| "loss": 0.8556, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 59.57, | |
| "learning_rate": 4.528571428571429e-05, | |
| "loss": 0.8508, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 60.11, | |
| "learning_rate": 4.469047619047619e-05, | |
| "loss": 0.8594, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 60.64, | |
| "learning_rate": 4.4095238095238096e-05, | |
| "loss": 0.8441, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 61.17, | |
| "learning_rate": 4.35e-05, | |
| "loss": 0.8492, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 61.17, | |
| "eval_loss": 0.2166604995727539, | |
| "eval_runtime": 147.2191, | |
| "eval_samples_per_second": 18.639, | |
| "eval_steps_per_second": 2.33, | |
| "eval_wer": 0.3446066541577045, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 61.7, | |
| "learning_rate": 4.290476190476191e-05, | |
| "loss": 0.8353, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 62.23, | |
| "learning_rate": 4.230952380952381e-05, | |
| "loss": 0.8407, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 62.76, | |
| "learning_rate": 4.1714285714285714e-05, | |
| "loss": 0.8335, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 63.3, | |
| "learning_rate": 4.1119047619047625e-05, | |
| "loss": 0.8596, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 63.83, | |
| "learning_rate": 4.052380952380952e-05, | |
| "loss": 0.8365, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 63.83, | |
| "eval_loss": 0.21556589007377625, | |
| "eval_runtime": 144.6076, | |
| "eval_samples_per_second": 18.975, | |
| "eval_steps_per_second": 2.372, | |
| "eval_wer": 0.33211294677589126, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 64.36, | |
| "learning_rate": 3.9928571428571434e-05, | |
| "loss": 0.8412, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 64.89, | |
| "learning_rate": 3.933333333333333e-05, | |
| "loss": 0.8301, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 65.42, | |
| "learning_rate": 3.873809523809524e-05, | |
| "loss": 0.8283, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 65.95, | |
| "learning_rate": 3.814285714285715e-05, | |
| "loss": 0.8257, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 66.49, | |
| "learning_rate": 3.7547619047619045e-05, | |
| "loss": 0.8298, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 66.49, | |
| "eval_loss": 0.21417230367660522, | |
| "eval_runtime": 151.8253, | |
| "eval_samples_per_second": 18.073, | |
| "eval_steps_per_second": 2.259, | |
| "eval_wer": 0.3400302045672967, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 67.02, | |
| "learning_rate": 3.6952380952380956e-05, | |
| "loss": 0.834, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 67.55, | |
| "learning_rate": 3.6357142857142854e-05, | |
| "loss": 0.8155, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 68.08, | |
| "learning_rate": 3.5761904761904765e-05, | |
| "loss": 0.8157, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 68.61, | |
| "learning_rate": 3.516666666666667e-05, | |
| "loss": 0.8027, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 69.15, | |
| "learning_rate": 3.4571428571428574e-05, | |
| "loss": 0.808, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 69.15, | |
| "eval_loss": 0.20793649554252625, | |
| "eval_runtime": 147.4919, | |
| "eval_samples_per_second": 18.604, | |
| "eval_steps_per_second": 2.326, | |
| "eval_wer": 0.31476820282824586, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 69.68, | |
| "learning_rate": 3.397619047619048e-05, | |
| "loss": 0.807, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 70.21, | |
| "learning_rate": 3.338095238095238e-05, | |
| "loss": 0.8164, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 70.74, | |
| "learning_rate": 3.278571428571429e-05, | |
| "loss": 0.7979, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 71.28, | |
| "learning_rate": 3.219047619047619e-05, | |
| "loss": 0.815, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 71.81, | |
| "learning_rate": 3.15952380952381e-05, | |
| "loss": 0.7999, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 71.81, | |
| "eval_loss": 0.21165262162685394, | |
| "eval_runtime": 145.5259, | |
| "eval_samples_per_second": 18.856, | |
| "eval_steps_per_second": 2.357, | |
| "eval_wer": 0.32254816713193907, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 72.34, | |
| "learning_rate": 3.1e-05, | |
| "loss": 0.8143, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 72.87, | |
| "learning_rate": 3.040476190476191e-05, | |
| "loss": 0.7952, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 73.4, | |
| "learning_rate": 2.980952380952381e-05, | |
| "loss": 0.8075, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 73.93, | |
| "learning_rate": 2.9214285714285715e-05, | |
| "loss": 0.8021, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 74.47, | |
| "learning_rate": 2.8619047619047623e-05, | |
| "loss": 0.7871, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 74.47, | |
| "eval_loss": 0.2087966501712799, | |
| "eval_runtime": 154.8982, | |
| "eval_samples_per_second": 17.715, | |
| "eval_steps_per_second": 2.214, | |
| "eval_wer": 0.31742254359068234, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "learning_rate": 2.8023809523809524e-05, | |
| "loss": 0.7898, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 75.53, | |
| "learning_rate": 2.742857142857143e-05, | |
| "loss": 0.7972, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 76.06, | |
| "learning_rate": 2.6833333333333333e-05, | |
| "loss": 0.801, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 76.59, | |
| "learning_rate": 2.623809523809524e-05, | |
| "loss": 0.786, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 77.13, | |
| "learning_rate": 2.5642857142857145e-05, | |
| "loss": 0.7858, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 77.13, | |
| "eval_loss": 0.2059505730867386, | |
| "eval_runtime": 147.0151, | |
| "eval_samples_per_second": 18.665, | |
| "eval_steps_per_second": 2.333, | |
| "eval_wer": 0.3008100315775022, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 77.66, | |
| "learning_rate": 2.5047619047619046e-05, | |
| "loss": 0.7753, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 78.19, | |
| "learning_rate": 2.4452380952380954e-05, | |
| "loss": 0.7794, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 78.72, | |
| "learning_rate": 2.385714285714286e-05, | |
| "loss": 0.7775, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 79.25, | |
| "learning_rate": 2.3261904761904763e-05, | |
| "loss": 0.7753, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 79.78, | |
| "learning_rate": 2.2666666666666668e-05, | |
| "loss": 0.7764, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 79.78, | |
| "eval_loss": 0.2128456085920334, | |
| "eval_runtime": 150.2974, | |
| "eval_samples_per_second": 18.257, | |
| "eval_steps_per_second": 2.282, | |
| "eval_wer": 0.3145851448446295, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 80.32, | |
| "learning_rate": 2.2071428571428572e-05, | |
| "loss": 0.7847, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 80.85, | |
| "learning_rate": 2.1476190476190477e-05, | |
| "loss": 0.774, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 81.38, | |
| "learning_rate": 2.0886904761904763e-05, | |
| "loss": 0.7739, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 81.91, | |
| "learning_rate": 2.0291666666666667e-05, | |
| "loss": 0.7579, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 82.45, | |
| "learning_rate": 1.9696428571428572e-05, | |
| "loss": 0.7684, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 82.45, | |
| "eval_loss": 0.20856936275959015, | |
| "eval_runtime": 149.2381, | |
| "eval_samples_per_second": 18.387, | |
| "eval_steps_per_second": 2.298, | |
| "eval_wer": 0.31005445975012585, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 82.97, | |
| "learning_rate": 1.9101190476190476e-05, | |
| "loss": 0.7584, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 83.51, | |
| "learning_rate": 1.850595238095238e-05, | |
| "loss": 0.7653, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 84.04, | |
| "learning_rate": 1.7910714285714285e-05, | |
| "loss": 0.7628, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 84.57, | |
| "learning_rate": 1.731547619047619e-05, | |
| "loss": 0.76, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 85.11, | |
| "learning_rate": 1.6720238095238098e-05, | |
| "loss": 0.7717, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 85.11, | |
| "eval_loss": 0.20475880801677704, | |
| "eval_runtime": 151.8607, | |
| "eval_samples_per_second": 18.069, | |
| "eval_steps_per_second": 2.259, | |
| "eval_wer": 0.3068509450368404, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 85.64, | |
| "learning_rate": 1.6125000000000002e-05, | |
| "loss": 0.7459, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 86.17, | |
| "learning_rate": 1.5529761904761907e-05, | |
| "loss": 0.7561, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 86.7, | |
| "learning_rate": 1.493452380952381e-05, | |
| "loss": 0.7457, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 87.23, | |
| "learning_rate": 1.4339285714285716e-05, | |
| "loss": 0.7599, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 87.76, | |
| "learning_rate": 1.374404761904762e-05, | |
| "loss": 0.7435, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 87.76, | |
| "eval_loss": 0.20274706184864044, | |
| "eval_runtime": 146.6164, | |
| "eval_samples_per_second": 18.716, | |
| "eval_steps_per_second": 2.339, | |
| "eval_wer": 0.3054780101597181, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 88.3, | |
| "learning_rate": 1.3148809523809525e-05, | |
| "loss": 0.7519, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 88.83, | |
| "learning_rate": 1.255357142857143e-05, | |
| "loss": 0.7483, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 89.36, | |
| "learning_rate": 1.1958333333333334e-05, | |
| "loss": 0.7483, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 89.89, | |
| "learning_rate": 1.1363095238095238e-05, | |
| "loss": 0.7353, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 90.42, | |
| "learning_rate": 1.0767857142857143e-05, | |
| "loss": 0.7378, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 90.42, | |
| "eval_loss": 0.20591045916080475, | |
| "eval_runtime": 153.244, | |
| "eval_samples_per_second": 17.906, | |
| "eval_steps_per_second": 2.238, | |
| "eval_wer": 0.29925403871676354, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 90.95, | |
| "learning_rate": 1.0172619047619047e-05, | |
| "loss": 0.7383, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 91.49, | |
| "learning_rate": 9.577380952380953e-06, | |
| "loss": 0.7466, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 92.02, | |
| "learning_rate": 8.982142857142856e-06, | |
| "loss": 0.7377, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 92.55, | |
| "learning_rate": 8.386904761904762e-06, | |
| "loss": 0.7369, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 93.08, | |
| "learning_rate": 7.791666666666667e-06, | |
| "loss": 0.7406, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 93.08, | |
| "eval_loss": 0.2040216028690338, | |
| "eval_runtime": 148.029, | |
| "eval_samples_per_second": 18.537, | |
| "eval_steps_per_second": 2.317, | |
| "eval_wer": 0.2966454624502311, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 93.61, | |
| "learning_rate": 7.2023809523809524e-06, | |
| "loss": 0.7348, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 94.15, | |
| "learning_rate": 6.613095238095239e-06, | |
| "loss": 0.7356, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 94.68, | |
| "learning_rate": 6.017857142857143e-06, | |
| "loss": 0.7305, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 95.21, | |
| "learning_rate": 5.422619047619048e-06, | |
| "loss": 0.741, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 95.74, | |
| "learning_rate": 4.827380952380952e-06, | |
| "loss": 0.7361, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 95.74, | |
| "eval_loss": 0.20563913881778717, | |
| "eval_runtime": 145.4837, | |
| "eval_samples_per_second": 18.861, | |
| "eval_steps_per_second": 2.358, | |
| "eval_wer": 0.30003203514713284, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 96.28, | |
| "learning_rate": 4.232142857142858e-06, | |
| "loss": 0.7359, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 96.81, | |
| "learning_rate": 3.636904761904762e-06, | |
| "loss": 0.7246, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 97.34, | |
| "learning_rate": 3.041666666666667e-06, | |
| "loss": 0.7353, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 97.87, | |
| "learning_rate": 2.4464285714285715e-06, | |
| "loss": 0.7305, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 98.4, | |
| "learning_rate": 1.8511904761904762e-06, | |
| "loss": 0.7379, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 98.4, | |
| "eval_loss": 0.20308499038219452, | |
| "eval_runtime": 146.2002, | |
| "eval_samples_per_second": 18.769, | |
| "eval_steps_per_second": 2.346, | |
| "eval_wer": 0.29756075236831264, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 98.93, | |
| "learning_rate": 1.2559523809523812e-06, | |
| "loss": 0.724, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 99.47, | |
| "learning_rate": 6.607142857142858e-07, | |
| "loss": 0.7339, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 6.547619047619047e-08, | |
| "loss": 0.7277, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 18800, | |
| "total_flos": 1.0789860816574084e+20, | |
| "train_loss": 1.1418190615227881, | |
| "train_runtime": 52190.9896, | |
| "train_samples_per_second": 11.561, | |
| "train_steps_per_second": 0.36 | |
| } | |
| ], | |
| "max_steps": 18800, | |
| "num_train_epochs": 100, | |
| "total_flos": 1.0789860816574084e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |