| { | |
| "best_metric": 1.4220008850097656, | |
| "best_model_checkpoint": "/scratch/s3545881/dumped/translation/mt5/2811562/checkpoint-4600", | |
| "epoch": 14.9755832881172, | |
| "global_step": 4600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.5e-06, | |
| "loss": 7.1735, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 5e-06, | |
| "loss": 7.1895, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 7.5e-06, | |
| "loss": 7.1285, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1e-05, | |
| "loss": 7.0295, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.25e-05, | |
| "loss": 6.871, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.5e-05, | |
| "loss": 6.7699, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1.7500000000000002e-05, | |
| "loss": 6.6541, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 2e-05, | |
| "loss": 6.4698, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 2.2499999999999998e-05, | |
| "loss": 6.4136, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 2.5e-05, | |
| "loss": 6.2236, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 2.75e-05, | |
| "loss": 6.1051, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 3e-05, | |
| "loss": 5.9834, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 5.8334, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 3.5000000000000004e-05, | |
| "loss": 5.7289, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 3.75e-05, | |
| "loss": 5.6266, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 4e-05, | |
| "loss": 5.488, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.25e-05, | |
| "loss": 5.4496, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 4.4999999999999996e-05, | |
| "loss": 5.3379, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 4.75e-05, | |
| "loss": 5.2659, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 5e-05, | |
| "loss": 5.1961, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 12902, | |
| 6326, | |
| 3612, | |
| 2252 | |
| ], | |
| "eval_loss": 2.750916004180908, | |
| "eval_precisions": [ | |
| 4.406812080307677, | |
| 2.1854789674423745, | |
| 1.2622265088534077, | |
| 0.7959228393098232 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 262.2904, | |
| "eval_samples_per_second": 13.203, | |
| "eval_score": 1.7636802703289813, | |
| "eval_steps_per_second": 13.203, | |
| "eval_sys_len": 292774, | |
| "eval_totals": [ | |
| 292774, | |
| 289456, | |
| 286161, | |
| 282942 | |
| ], | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 5.25e-05, | |
| "loss": 5.17, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 5.5e-05, | |
| "loss": 5.0736, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 5.75e-05, | |
| "loss": 4.9931, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 6e-05, | |
| "loss": 4.9457, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 6.25e-05, | |
| "loss": 4.9108, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 4.8353, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 6.75e-05, | |
| "loss": 4.7755, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 7.000000000000001e-05, | |
| "loss": 4.7669, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 7.25e-05, | |
| "loss": 4.6519, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 7.5e-05, | |
| "loss": 4.5968, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 7.75e-05, | |
| "loss": 4.5827, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 8e-05, | |
| "loss": 4.5738, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 8.25e-05, | |
| "loss": 4.5119, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 8.5e-05, | |
| "loss": 4.4241, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 8.75e-05, | |
| "loss": 4.4606, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 8.999999999999999e-05, | |
| "loss": 4.3099, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 9.25e-05, | |
| "loss": 4.4264, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 9.5e-05, | |
| "loss": 4.2987, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 9.750000000000001e-05, | |
| "loss": 4.3585, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 0.0001, | |
| "loss": 4.2855, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 21482, | |
| 12393, | |
| 7972, | |
| 5427 | |
| ], | |
| "eval_loss": 2.258103609085083, | |
| "eval_precisions": [ | |
| 30.616840545008838, | |
| 18.57993133536229, | |
| 12.606144942203388, | |
| 9.071761697006169 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 225.927, | |
| "eval_samples_per_second": 15.328, | |
| "eval_score": 15.970540029112566, | |
| "eval_steps_per_second": 15.328, | |
| "eval_sys_len": 70164, | |
| "eval_totals": [ | |
| 70164, | |
| 66701, | |
| 63239, | |
| 59823 | |
| ], | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0001025, | |
| "loss": 4.2679, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 0.000105, | |
| "loss": 4.1738, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.0001075, | |
| "loss": 4.1978, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.00011, | |
| "loss": 4.2151, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.00011250000000000001, | |
| "loss": 4.139, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 0.000115, | |
| "loss": 4.1825, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.0001175, | |
| "loss": 4.1172, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.00012, | |
| "loss": 4.2029, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0001225, | |
| "loss": 4.1657, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 0.000125, | |
| "loss": 4.0543, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 0.0001275, | |
| "loss": 4.0173, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00013000000000000002, | |
| "loss": 4.0702, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 0.00013250000000000002, | |
| "loss": 4.0172, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 0.000135, | |
| "loss": 3.998, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.0001375, | |
| "loss": 4.0506, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.00014000000000000001, | |
| "loss": 3.9215, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 0.0001425, | |
| "loss": 3.986, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.000145, | |
| "loss": 3.9276, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.0001475, | |
| "loss": 3.8933, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 0.00015, | |
| "loss": 3.9534, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 22456, | |
| 13442, | |
| 8970, | |
| 6310 | |
| ], | |
| "eval_loss": 2.0556671619415283, | |
| "eval_precisions": [ | |
| 34.72666821309828, | |
| 21.963334531551258, | |
| 15.534619514391604, | |
| 11.61485081083071 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 214.8511, | |
| "eval_samples_per_second": 16.118, | |
| "eval_score": 19.2605601155173, | |
| "eval_steps_per_second": 16.118, | |
| "eval_sys_len": 64665, | |
| "eval_totals": [ | |
| 64665, | |
| 61202, | |
| 57742, | |
| 54327 | |
| ], | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.0001525, | |
| "loss": 3.8561, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 0.000155, | |
| "loss": 3.8084, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 0.0001575, | |
| "loss": 3.8783, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 0.00016, | |
| "loss": 3.8229, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 0.00016250000000000002, | |
| "loss": 3.7835, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.000165, | |
| "loss": 3.7847, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 0.0001675, | |
| "loss": 3.8852, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 0.00017, | |
| "loss": 3.7897, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 0.0001725, | |
| "loss": 3.7977, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 0.000175, | |
| "loss": 3.7706, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 0.0001775, | |
| "loss": 3.7135, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 3.7296, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 0.0001825, | |
| "loss": 3.7966, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 0.000185, | |
| "loss": 3.6942, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 0.0001875, | |
| "loss": 3.7341, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 0.00019, | |
| "loss": 3.7491, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 0.00019250000000000002, | |
| "loss": 3.6826, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 0.00019500000000000002, | |
| "loss": 3.6763, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 0.0001975, | |
| "loss": 3.5826, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 0.0002, | |
| "loss": 3.6549, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 24601, | |
| 15366, | |
| 10549, | |
| 7581 | |
| ], | |
| "eval_loss": 1.962100863456726, | |
| "eval_precisions": [ | |
| 53.96731380936712, | |
| 36.479749299653385, | |
| 27.26614800072372, | |
| 21.474095685919043 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 160.9582, | |
| "eval_samples_per_second": 21.515, | |
| "eval_score": 32.766520790805245, | |
| "eval_steps_per_second": 21.515, | |
| "eval_sys_len": 45585, | |
| "eval_totals": [ | |
| 45585, | |
| 42122, | |
| 38689, | |
| 35303 | |
| ], | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 0.00020250000000000002, | |
| "loss": 3.75, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 0.000205, | |
| "loss": 3.5627, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 0.0002075, | |
| "loss": 3.6651, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 0.00021, | |
| "loss": 3.6312, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 0.0002125, | |
| "loss": 3.6042, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.000215, | |
| "loss": 3.5747, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 0.0002175, | |
| "loss": 3.5851, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 0.00022, | |
| "loss": 3.5618, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 0.00022250000000000001, | |
| "loss": 3.5375, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 0.00022500000000000002, | |
| "loss": 3.5951, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 0.0002275, | |
| "loss": 3.582, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.00023, | |
| "loss": 3.5011, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 0.0002325, | |
| "loss": 3.4628, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 0.000235, | |
| "loss": 3.5, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 0.0002375, | |
| "loss": 3.5281, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 0.00024, | |
| "loss": 3.4292, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 0.00024249999999999999, | |
| "loss": 3.5526, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 0.000245, | |
| "loss": 3.4211, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 0.0002475, | |
| "loss": 3.4732, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 0.00025, | |
| "loss": 3.5164, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 24977, | |
| 15865, | |
| 10957, | |
| 7898 | |
| ], | |
| "eval_loss": 1.8600678443908691, | |
| "eval_precisions": [ | |
| 52.542230262742706, | |
| 35.99627898534283, | |
| 26.975060932076122, | |
| 21.22547702230583 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 165.696, | |
| "eval_samples_per_second": 20.9, | |
| "eval_score": 32.258673250001976, | |
| "eval_steps_per_second": 20.9, | |
| "eval_sys_len": 47537, | |
| "eval_totals": [ | |
| 47537, | |
| 44074, | |
| 40619, | |
| 37210 | |
| ], | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 0.0002525, | |
| "loss": 3.4451, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 0.000255, | |
| "loss": 3.4723, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 0.0002575, | |
| "loss": 3.4181, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 0.00026000000000000003, | |
| "loss": 3.432, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 0.00026250000000000004, | |
| "loss": 3.364, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 0.00026500000000000004, | |
| "loss": 3.3962, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 0.0002675, | |
| "loss": 3.4944, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 0.00027, | |
| "loss": 3.497, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 0.0002725, | |
| "loss": 3.4347, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 0.000275, | |
| "loss": 3.4106, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 0.0002775, | |
| "loss": 3.4384, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 0.00028000000000000003, | |
| "loss": 3.3933, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 0.0002825, | |
| "loss": 3.3448, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 0.000285, | |
| "loss": 3.3148, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 0.0002875, | |
| "loss": 3.2697, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 0.00029, | |
| "loss": 3.4034, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 0.0002925, | |
| "loss": 3.4009, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 0.000295, | |
| "loss": 3.3068, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 0.00029749999999999997, | |
| "loss": 3.4363, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 0.0003, | |
| "loss": 3.4333, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 24869, | |
| 15901, | |
| 11072, | |
| 8028 | |
| ], | |
| "eval_loss": 1.7562536001205444, | |
| "eval_precisions": [ | |
| 47.29745150247242, | |
| 32.37371989331596, | |
| 24.24773334501336, | |
| 18.999834331290085 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 214.2902, | |
| "eval_samples_per_second": 16.16, | |
| "eval_score": 28.980963954064624, | |
| "eval_steps_per_second": 16.16, | |
| "eval_sys_len": 52580, | |
| "eval_totals": [ | |
| 52580, | |
| 49117, | |
| 45662, | |
| 42253 | |
| ], | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 0.0003025, | |
| "loss": 3.3664, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "learning_rate": 0.000305, | |
| "loss": 3.3555, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0003075, | |
| "loss": 3.3637, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 0.00031, | |
| "loss": 3.2577, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 0.0003125, | |
| "loss": 3.2929, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 0.000315, | |
| "loss": 3.2327, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 0.0003175, | |
| "loss": 3.3175, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 0.00032, | |
| "loss": 3.2598, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "learning_rate": 0.00032250000000000003, | |
| "loss": 3.2131, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "learning_rate": 0.00032500000000000004, | |
| "loss": 3.2873, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 0.00032750000000000005, | |
| "loss": 3.2196, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 0.00033, | |
| "loss": 3.3356, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 0.0003325, | |
| "loss": 3.2612, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 0.000335, | |
| "loss": 3.1025, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 0.0003375, | |
| "loss": 3.2298, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "learning_rate": 0.00034, | |
| "loss": 3.1861, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 0.00034250000000000003, | |
| "loss": 3.2453, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 0.000345, | |
| "loss": 3.2116, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "learning_rate": 0.0003475, | |
| "loss": 3.2456, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "learning_rate": 0.00035, | |
| "loss": 3.2148, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 26323, | |
| 17238, | |
| 12161, | |
| 8884 | |
| ], | |
| "eval_loss": 1.7239768505096436, | |
| "eval_precisions": [ | |
| 59.99954412837345, | |
| 42.65881363062684, | |
| 32.90848081398495, | |
| 26.484617219174815 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 129.8828, | |
| "eval_samples_per_second": 26.662, | |
| "eval_score": 38.646891207773166, | |
| "eval_steps_per_second": 26.662, | |
| "eval_sys_len": 43872, | |
| "eval_totals": [ | |
| 43872, | |
| 40409, | |
| 36954, | |
| 33544 | |
| ], | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 0.0003525, | |
| "loss": 3.2301, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 0.000355, | |
| "loss": 3.1593, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "learning_rate": 0.0003575, | |
| "loss": 3.2608, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 0.00035999999999999997, | |
| "loss": 3.1374, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 0.0003625, | |
| "loss": 3.2217, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "learning_rate": 0.000365, | |
| "loss": 3.2159, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 0.0003675, | |
| "loss": 3.2236, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 0.00037, | |
| "loss": 3.2447, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "learning_rate": 0.0003725, | |
| "loss": 3.2312, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 0.000375, | |
| "loss": 3.2011, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "learning_rate": 0.0003775, | |
| "loss": 3.2031, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 0.00038, | |
| "loss": 3.1596, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "learning_rate": 0.00038250000000000003, | |
| "loss": 3.1573, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 0.00038500000000000003, | |
| "loss": 3.1637, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 0.00038750000000000004, | |
| "loss": 3.1586, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 0.00039000000000000005, | |
| "loss": 3.0617, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 0.0003925, | |
| "loss": 3.107, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "learning_rate": 0.000395, | |
| "loss": 3.0754, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "learning_rate": 0.0003975, | |
| "loss": 3.1476, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "learning_rate": 0.0004, | |
| "loss": 3.1108, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 26002, | |
| 17132, | |
| 12142, | |
| 8953 | |
| ], | |
| "eval_loss": 1.6534518003463745, | |
| "eval_precisions": [ | |
| 60.742401943607355, | |
| 43.54412362749085, | |
| 33.831150738367235, | |
| 27.56295794593929 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 115.9758, | |
| "eval_samples_per_second": 29.86, | |
| "eval_score": 39.62926586495389, | |
| "eval_steps_per_second": 29.86, | |
| "eval_sys_len": 42807, | |
| "eval_totals": [ | |
| 42807, | |
| 39344, | |
| 35890, | |
| 32482 | |
| ], | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 0.0004025, | |
| "loss": 3.0389, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "learning_rate": 0.00040500000000000003, | |
| "loss": 3.1482, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 5.31, | |
| "learning_rate": 0.0004075, | |
| "loss": 3.1614, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "learning_rate": 0.00041, | |
| "loss": 3.103, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 5.37, | |
| "learning_rate": 0.0004125, | |
| "loss": 3.144, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "learning_rate": 0.000415, | |
| "loss": 3.0297, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "learning_rate": 0.0004175, | |
| "loss": 3.0478, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 5.47, | |
| "learning_rate": 0.00042, | |
| "loss": 3.1079, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "learning_rate": 0.00042249999999999997, | |
| "loss": 3.0638, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 0.000425, | |
| "loss": 3.0575, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "learning_rate": 0.0004275, | |
| "loss": 3.1212, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 0.00043, | |
| "loss": 3.1422, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 5.63, | |
| "learning_rate": 0.0004325, | |
| "loss": 3.015, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 0.000435, | |
| "loss": 3.0151, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 0.0004375, | |
| "loss": 2.9851, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 5.73, | |
| "learning_rate": 0.00044, | |
| "loss": 3.0168, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 0.0004425, | |
| "loss": 2.9517, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 0.00044500000000000003, | |
| "loss": 3.0005, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "learning_rate": 0.00044750000000000004, | |
| "loss": 3.0775, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 0.00045000000000000004, | |
| "loss": 3.0713, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 25470, | |
| 16777, | |
| 11904, | |
| 8755 | |
| ], | |
| "eval_loss": 1.617377758026123, | |
| "eval_precisions": [ | |
| 61.505397116708124, | |
| 44.2104985770001, | |
| 34.51135012901168, | |
| 28.164709667041983 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 121.9732, | |
| "eval_samples_per_second": 28.391, | |
| "eval_score": 40.32055180650292, | |
| "eval_steps_per_second": 28.391, | |
| "eval_sys_len": 41411, | |
| "eval_totals": [ | |
| 41411, | |
| 37948, | |
| 34493, | |
| 31085 | |
| ], | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "learning_rate": 0.00045250000000000005, | |
| "loss": 3.1174, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 0.000455, | |
| "loss": 2.9954, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "learning_rate": 0.0004575, | |
| "loss": 3.0373, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "learning_rate": 0.00046, | |
| "loss": 3.0608, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 0.0004625, | |
| "loss": 3.0306, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 0.000465, | |
| "loss": 2.9395, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "learning_rate": 0.00046750000000000003, | |
| "loss": 3.0109, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 0.00047, | |
| "loss": 2.9562, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 0.0004725, | |
| "loss": 2.9876, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "learning_rate": 0.000475, | |
| "loss": 2.9542, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "learning_rate": 0.0004775, | |
| "loss": 2.8958, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 0.00048, | |
| "loss": 3.0231, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "learning_rate": 0.0004825, | |
| "loss": 2.9398, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "learning_rate": 0.00048499999999999997, | |
| "loss": 2.9273, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "learning_rate": 0.0004875, | |
| "loss": 2.9431, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "learning_rate": 0.00049, | |
| "loss": 2.976, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 0.0004925, | |
| "loss": 2.8617, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 0.000495, | |
| "loss": 2.9501, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 6.48, | |
| "learning_rate": 0.0004975, | |
| "loss": 2.9997, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "learning_rate": 0.0005, | |
| "loss": 2.8715, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 26989, | |
| 18197, | |
| 13091, | |
| 9738 | |
| ], | |
| "eval_loss": 1.6082161664962769, | |
| "eval_precisions": [ | |
| 62.49623711937015, | |
| 45.81088565530437, | |
| 36.09717090387691, | |
| 29.638422205989773 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 116.2487, | |
| "eval_samples_per_second": 29.79, | |
| "eval_score": 41.834800362190585, | |
| "eval_steps_per_second": 29.79, | |
| "eval_sys_len": 43185, | |
| "eval_totals": [ | |
| 43185, | |
| 39722, | |
| 36266, | |
| 32856 | |
| ], | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "learning_rate": 0.0005024999999999999, | |
| "loss": 2.9772, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "learning_rate": 0.000505, | |
| "loss": 2.9945, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "learning_rate": 0.0005074999999999999, | |
| "loss": 2.8798, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 6.64, | |
| "learning_rate": 0.00051, | |
| "loss": 2.9645, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 0.0005124999999999999, | |
| "loss": 2.8499, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 6.71, | |
| "learning_rate": 0.000515, | |
| "loss": 2.9474, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 6.74, | |
| "learning_rate": 0.0005175, | |
| "loss": 2.9326, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "learning_rate": 0.0005200000000000001, | |
| "loss": 2.8708, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "learning_rate": 0.0005225, | |
| "loss": 2.9616, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 6.84, | |
| "learning_rate": 0.0005250000000000001, | |
| "loss": 2.9599, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 6.87, | |
| "learning_rate": 0.0005275, | |
| "loss": 2.8761, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "learning_rate": 0.0005300000000000001, | |
| "loss": 3.0114, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "learning_rate": 0.0005325, | |
| "loss": 2.9201, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "learning_rate": 0.000535, | |
| "loss": 2.9802, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 0.0005375, | |
| "loss": 2.9193, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 0.00054, | |
| "loss": 2.8004, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 0.0005425, | |
| "loss": 2.8722, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 0.000545, | |
| "loss": 2.8149, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 7.13, | |
| "learning_rate": 0.0005475, | |
| "loss": 2.9112, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 7.16, | |
| "learning_rate": 0.00055, | |
| "loss": 2.8271, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 7.16, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 26323, | |
| 17758, | |
| 12855, | |
| 9588 | |
| ], | |
| "eval_loss": 1.547306776046753, | |
| "eval_precisions": [ | |
| 57.49011728219801, | |
| 41.9572819204234, | |
| 33.07177772060715, | |
| 27.03739213806328 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 141.8578, | |
| "eval_samples_per_second": 24.412, | |
| "eval_score": 38.32266653815659, | |
| "eval_steps_per_second": 24.412, | |
| "eval_sys_len": 45787, | |
| "eval_totals": [ | |
| 45787, | |
| 42324, | |
| 38870, | |
| 35462 | |
| ], | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 7.19, | |
| "learning_rate": 0.0005525, | |
| "loss": 2.797, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 7.23, | |
| "learning_rate": 0.000555, | |
| "loss": 2.8883, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "learning_rate": 0.0005575, | |
| "loss": 2.81, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 7.29, | |
| "learning_rate": 0.0005600000000000001, | |
| "loss": 2.8967, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "learning_rate": 0.0005625000000000001, | |
| "loss": 2.8093, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "learning_rate": 0.000565, | |
| "loss": 2.8463, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 7.39, | |
| "learning_rate": 0.0005675, | |
| "loss": 2.84, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 0.00057, | |
| "loss": 2.9007, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 7.46, | |
| "learning_rate": 0.0005725, | |
| "loss": 2.7673, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 7.49, | |
| "learning_rate": 0.000575, | |
| "loss": 2.8213, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "learning_rate": 0.0005775, | |
| "loss": 2.8536, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 7.55, | |
| "learning_rate": 0.00058, | |
| "loss": 2.7698, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 0.0005825, | |
| "loss": 2.8104, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 7.62, | |
| "learning_rate": 0.000585, | |
| "loss": 2.8719, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "learning_rate": 0.0005875, | |
| "loss": 2.8486, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 7.68, | |
| "learning_rate": 0.00059, | |
| "loss": 2.8941, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "learning_rate": 0.0005925, | |
| "loss": 2.7646, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 7.75, | |
| "learning_rate": 0.0005949999999999999, | |
| "loss": 2.8403, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 0.0005975, | |
| "loss": 2.8481, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 7.81, | |
| "learning_rate": 0.0006, | |
| "loss": 2.8294, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 7.81, | |
| "eval_bp": 0.9479912131785159, | |
| "eval_counts": [ | |
| 27654, | |
| 18964, | |
| 13797, | |
| 10304 | |
| ], | |
| "eval_loss": 1.5549582242965698, | |
| "eval_precisions": [ | |
| 70.40044805376645, | |
| 52.945446423585906, | |
| 42.60042609689073, | |
| 35.52858423556996 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 69.016, | |
| "eval_samples_per_second": 50.177, | |
| "eval_score": 46.201202847952445, | |
| "eval_steps_per_second": 50.177, | |
| "eval_sys_len": 39281, | |
| "eval_totals": [ | |
| 39281, | |
| 35818, | |
| 32387, | |
| 29002 | |
| ], | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 7.85, | |
| "learning_rate": 0.0006025000000000001, | |
| "loss": 2.8759, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 7.88, | |
| "learning_rate": 0.000605, | |
| "loss": 2.7903, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "learning_rate": 0.0006075000000000001, | |
| "loss": 2.8623, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 7.94, | |
| "learning_rate": 0.00061, | |
| "loss": 2.7207, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 7.98, | |
| "learning_rate": 0.0006125000000000001, | |
| "loss": 2.889, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 0.000615, | |
| "loss": 2.8509, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 0.0006175000000000001, | |
| "loss": 2.7259, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 0.00062, | |
| "loss": 2.7449, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 8.11, | |
| "learning_rate": 0.0006225000000000001, | |
| "loss": 2.8018, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 8.14, | |
| "learning_rate": 0.000625, | |
| "loss": 2.7617, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.17, | |
| "learning_rate": 0.0006274999999999999, | |
| "loss": 2.7699, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "learning_rate": 0.00063, | |
| "loss": 2.6918, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 8.24, | |
| "learning_rate": 0.0006324999999999999, | |
| "loss": 2.75, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 8.27, | |
| "learning_rate": 0.000635, | |
| "loss": 2.7171, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 8.3, | |
| "learning_rate": 0.0006374999999999999, | |
| "loss": 2.8189, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 0.00064, | |
| "loss": 2.8228, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 8.37, | |
| "learning_rate": 0.0006425, | |
| "loss": 2.6978, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "learning_rate": 0.0006450000000000001, | |
| "loss": 2.714, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 8.43, | |
| "learning_rate": 0.0006475, | |
| "loss": 2.6573, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 8.46, | |
| "learning_rate": 0.0006500000000000001, | |
| "loss": 2.7587, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 8.46, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 26118, | |
| 17728, | |
| 12867, | |
| 9601 | |
| ], | |
| "eval_loss": 1.5125658512115479, | |
| "eval_precisions": [ | |
| 53.19239934013564, | |
| 38.84482229720847, | |
| 30.503532312360722, | |
| 24.765270326042096 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 170.3639, | |
| "eval_samples_per_second": 20.327, | |
| "eval_score": 35.34630181468264, | |
| "eval_steps_per_second": 20.327, | |
| "eval_sys_len": 49101, | |
| "eval_totals": [ | |
| 49101, | |
| 45638, | |
| 42182, | |
| 38768 | |
| ], | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "learning_rate": 0.0006525, | |
| "loss": 2.75, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 8.53, | |
| "learning_rate": 0.0006550000000000001, | |
| "loss": 2.7836, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "learning_rate": 0.0006575, | |
| "loss": 2.7503, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 8.59, | |
| "learning_rate": 0.00066, | |
| "loss": 2.698, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 8.63, | |
| "learning_rate": 0.0006625, | |
| "loss": 2.7258, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "learning_rate": 0.000665, | |
| "loss": 2.5896, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 8.69, | |
| "learning_rate": 0.0006675, | |
| "loss": 2.6823, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 8.72, | |
| "learning_rate": 0.00067, | |
| "loss": 2.7439, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "learning_rate": 0.0006725, | |
| "loss": 2.7195, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 8.79, | |
| "learning_rate": 0.000675, | |
| "loss": 2.7273, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "learning_rate": 0.0006775, | |
| "loss": 2.7541, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 8.86, | |
| "learning_rate": 0.00068, | |
| "loss": 2.7274, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 0.0006825000000000001, | |
| "loss": 2.7677, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 8.92, | |
| "learning_rate": 0.0006850000000000001, | |
| "loss": 2.6633, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 8.95, | |
| "learning_rate": 0.0006875, | |
| "loss": 2.7939, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "learning_rate": 0.00069, | |
| "loss": 2.79, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 0.0006925, | |
| "loss": 2.7182, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 0.000695, | |
| "loss": 2.6684, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "learning_rate": 0.0006975, | |
| "loss": 2.5166, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 9.12, | |
| "learning_rate": 0.0007, | |
| "loss": 2.6355, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 9.12, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 27093, | |
| 18616, | |
| 13629, | |
| 10263 | |
| ], | |
| "eval_loss": 1.4999642372131348, | |
| "eval_precisions": [ | |
| 64.89341317365269, | |
| 48.62224776033641, | |
| 39.12893686658436, | |
| 32.66598765039149 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 102.7522, | |
| "eval_samples_per_second": 33.702, | |
| "eval_score": 44.8133460943785, | |
| "eval_steps_per_second": 33.702, | |
| "eval_sys_len": 41750, | |
| "eval_totals": [ | |
| 41750, | |
| 38287, | |
| 34831, | |
| 31418 | |
| ], | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 9.15, | |
| "learning_rate": 0.0007025, | |
| "loss": 2.5984, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 9.18, | |
| "learning_rate": 0.000705, | |
| "loss": 2.6082, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 9.21, | |
| "learning_rate": 0.0007075, | |
| "loss": 2.5937, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "learning_rate": 0.00071, | |
| "loss": 2.6601, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 9.28, | |
| "learning_rate": 0.0007125, | |
| "loss": 2.7424, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 9.31, | |
| "learning_rate": 0.000715, | |
| "loss": 2.6533, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 9.34, | |
| "learning_rate": 0.0007175, | |
| "loss": 2.6721, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "learning_rate": 0.0007199999999999999, | |
| "loss": 2.6039, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "learning_rate": 0.0007225, | |
| "loss": 2.714, | |
| "step": 2890 | |
| }, | |
| { | |
| "epoch": 9.44, | |
| "learning_rate": 0.000725, | |
| "loss": 2.6207, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "learning_rate": 0.0007275000000000001, | |
| "loss": 2.6425, | |
| "step": 2910 | |
| }, | |
| { | |
| "epoch": 9.51, | |
| "learning_rate": 0.00073, | |
| "loss": 2.6648, | |
| "step": 2920 | |
| }, | |
| { | |
| "epoch": 9.54, | |
| "learning_rate": 0.0007325000000000001, | |
| "loss": 2.66, | |
| "step": 2930 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "learning_rate": 0.000735, | |
| "loss": 2.6729, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 0.0007375000000000001, | |
| "loss": 2.6945, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 9.64, | |
| "learning_rate": 0.00074, | |
| "loss": 2.7379, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "learning_rate": 0.0007425000000000001, | |
| "loss": 2.6595, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 9.7, | |
| "learning_rate": 0.000745, | |
| "loss": 2.5923, | |
| "step": 2980 | |
| }, | |
| { | |
| "epoch": 9.73, | |
| "learning_rate": 0.0007475000000000001, | |
| "loss": 2.6219, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 9.77, | |
| "learning_rate": 0.00075, | |
| "loss": 2.6973, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.77, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 26621, | |
| 18360, | |
| 13471, | |
| 10153 | |
| ], | |
| "eval_loss": 1.4800695180892944, | |
| "eval_precisions": [ | |
| 62.9993373722075, | |
| 47.328126208336556, | |
| 38.12151569176783, | |
| 31.803658689387294 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 122.6573, | |
| "eval_samples_per_second": 28.233, | |
| "eval_score": 43.60393001478689, | |
| "eval_steps_per_second": 28.233, | |
| "eval_sys_len": 42256, | |
| "eval_totals": [ | |
| 42256, | |
| 38793, | |
| 35337, | |
| 31924 | |
| ], | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "learning_rate": 0.0007524999999999999, | |
| "loss": 2.6431, | |
| "step": 3010 | |
| }, | |
| { | |
| "epoch": 9.83, | |
| "learning_rate": 0.000755, | |
| "loss": 2.7345, | |
| "step": 3020 | |
| }, | |
| { | |
| "epoch": 9.86, | |
| "learning_rate": 0.0007574999999999999, | |
| "loss": 2.659, | |
| "step": 3030 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "learning_rate": 0.00076, | |
| "loss": 2.6094, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 9.93, | |
| "learning_rate": 0.0007624999999999999, | |
| "loss": 2.6922, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 9.96, | |
| "learning_rate": 0.0007650000000000001, | |
| "loss": 2.5544, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "learning_rate": 0.0007675, | |
| "loss": 2.6085, | |
| "step": 3070 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 0.0007700000000000001, | |
| "loss": 2.6645, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 0.0007725, | |
| "loss": 2.5658, | |
| "step": 3090 | |
| }, | |
| { | |
| "epoch": 10.09, | |
| "learning_rate": 0.0007750000000000001, | |
| "loss": 2.471, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 10.12, | |
| "learning_rate": 0.0007775, | |
| "loss": 2.6254, | |
| "step": 3110 | |
| }, | |
| { | |
| "epoch": 10.16, | |
| "learning_rate": 0.0007800000000000001, | |
| "loss": 2.6021, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 10.19, | |
| "learning_rate": 0.0007825, | |
| "loss": 2.5589, | |
| "step": 3130 | |
| }, | |
| { | |
| "epoch": 10.22, | |
| "learning_rate": 0.000785, | |
| "loss": 2.6455, | |
| "step": 3140 | |
| }, | |
| { | |
| "epoch": 10.26, | |
| "learning_rate": 0.0007875, | |
| "loss": 2.5829, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 10.29, | |
| "learning_rate": 0.00079, | |
| "loss": 2.5544, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "learning_rate": 0.0007925, | |
| "loss": 2.6645, | |
| "step": 3170 | |
| }, | |
| { | |
| "epoch": 10.35, | |
| "learning_rate": 0.000795, | |
| "loss": 2.5514, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 10.39, | |
| "learning_rate": 0.0007975, | |
| "loss": 2.6226, | |
| "step": 3190 | |
| }, | |
| { | |
| "epoch": 10.42, | |
| "learning_rate": 0.0008, | |
| "loss": 2.6525, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 10.42, | |
| "eval_bp": 0.9967563932434949, | |
| "eval_counts": [ | |
| 27078, | |
| 18724, | |
| 13712, | |
| 10292 | |
| ], | |
| "eval_loss": 1.4687154293060303, | |
| "eval_precisions": [ | |
| 65.65159413262214, | |
| 49.55799057752369, | |
| 39.94639631766008, | |
| 33.29235944879343 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 100.2708, | |
| "eval_samples_per_second": 34.536, | |
| "eval_score": 45.46051134991977, | |
| "eval_steps_per_second": 34.536, | |
| "eval_sys_len": 41245, | |
| "eval_totals": [ | |
| 41245, | |
| 37782, | |
| 34326, | |
| 30914 | |
| ], | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 10.45, | |
| "learning_rate": 0.0008025, | |
| "loss": 2.4619, | |
| "step": 3210 | |
| }, | |
| { | |
| "epoch": 10.48, | |
| "learning_rate": 0.000805, | |
| "loss": 2.591, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 10.52, | |
| "learning_rate": 0.0008075000000000001, | |
| "loss": 2.5628, | |
| "step": 3230 | |
| }, | |
| { | |
| "epoch": 10.55, | |
| "learning_rate": 0.0008100000000000001, | |
| "loss": 2.5603, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 10.58, | |
| "learning_rate": 0.0008125000000000001, | |
| "loss": 2.5817, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 10.61, | |
| "learning_rate": 0.000815, | |
| "loss": 2.5175, | |
| "step": 3260 | |
| }, | |
| { | |
| "epoch": 10.65, | |
| "learning_rate": 0.0008175, | |
| "loss": 2.703, | |
| "step": 3270 | |
| }, | |
| { | |
| "epoch": 10.68, | |
| "learning_rate": 0.00082, | |
| "loss": 2.5288, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 10.71, | |
| "learning_rate": 0.0008225, | |
| "loss": 2.5492, | |
| "step": 3290 | |
| }, | |
| { | |
| "epoch": 10.74, | |
| "learning_rate": 0.000825, | |
| "loss": 2.5673, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 10.78, | |
| "learning_rate": 0.0008275, | |
| "loss": 2.554, | |
| "step": 3310 | |
| }, | |
| { | |
| "epoch": 10.81, | |
| "learning_rate": 0.00083, | |
| "loss": 2.5119, | |
| "step": 3320 | |
| }, | |
| { | |
| "epoch": 10.84, | |
| "learning_rate": 0.0008325, | |
| "loss": 2.5016, | |
| "step": 3330 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "learning_rate": 0.000835, | |
| "loss": 2.4994, | |
| "step": 3340 | |
| }, | |
| { | |
| "epoch": 10.91, | |
| "learning_rate": 0.0008375, | |
| "loss": 2.5911, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 10.94, | |
| "learning_rate": 0.00084, | |
| "loss": 2.5726, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "learning_rate": 0.0008425, | |
| "loss": 2.51, | |
| "step": 3370 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 0.0008449999999999999, | |
| "loss": 2.5334, | |
| "step": 3380 | |
| }, | |
| { | |
| "epoch": 11.04, | |
| "learning_rate": 0.0008475000000000001, | |
| "loss": 2.4746, | |
| "step": 3390 | |
| }, | |
| { | |
| "epoch": 11.07, | |
| "learning_rate": 0.00085, | |
| "loss": 2.4877, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 11.07, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 28177, | |
| 19707, | |
| 14557, | |
| 11007 | |
| ], | |
| "eval_loss": 1.4748998880386353, | |
| "eval_precisions": [ | |
| 66.65956943458718, | |
| 50.782075398768264, | |
| 41.14936680235188, | |
| 34.40870299165338 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 101.4694, | |
| "eval_samples_per_second": 34.129, | |
| "eval_score": 46.7897906835126, | |
| "eval_steps_per_second": 34.129, | |
| "eval_sys_len": 42270, | |
| "eval_totals": [ | |
| 42270, | |
| 38807, | |
| 35376, | |
| 31989 | |
| ], | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 11.1, | |
| "learning_rate": 0.0008525000000000001, | |
| "loss": 2.4992, | |
| "step": 3410 | |
| }, | |
| { | |
| "epoch": 11.13, | |
| "learning_rate": 0.000855, | |
| "loss": 2.4663, | |
| "step": 3420 | |
| }, | |
| { | |
| "epoch": 11.17, | |
| "learning_rate": 0.0008575000000000001, | |
| "loss": 2.4946, | |
| "step": 3430 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "learning_rate": 0.00086, | |
| "loss": 2.3708, | |
| "step": 3440 | |
| }, | |
| { | |
| "epoch": 11.23, | |
| "learning_rate": 0.0008625000000000001, | |
| "loss": 2.4729, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 11.26, | |
| "learning_rate": 0.000865, | |
| "loss": 2.4967, | |
| "step": 3460 | |
| }, | |
| { | |
| "epoch": 11.3, | |
| "learning_rate": 0.0008675000000000001, | |
| "loss": 2.509, | |
| "step": 3470 | |
| }, | |
| { | |
| "epoch": 11.33, | |
| "learning_rate": 0.00087, | |
| "loss": 2.5677, | |
| "step": 3480 | |
| }, | |
| { | |
| "epoch": 11.36, | |
| "learning_rate": 0.0008725000000000001, | |
| "loss": 2.5343, | |
| "step": 3490 | |
| }, | |
| { | |
| "epoch": 11.39, | |
| "learning_rate": 0.000875, | |
| "loss": 2.5012, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 11.43, | |
| "learning_rate": 0.0008774999999999999, | |
| "loss": 2.4763, | |
| "step": 3510 | |
| }, | |
| { | |
| "epoch": 11.46, | |
| "learning_rate": 0.00088, | |
| "loss": 2.4803, | |
| "step": 3520 | |
| }, | |
| { | |
| "epoch": 11.49, | |
| "learning_rate": 0.0008824999999999999, | |
| "loss": 2.5076, | |
| "step": 3530 | |
| }, | |
| { | |
| "epoch": 11.52, | |
| "learning_rate": 0.000885, | |
| "loss": 2.5069, | |
| "step": 3540 | |
| }, | |
| { | |
| "epoch": 11.56, | |
| "learning_rate": 0.0008874999999999999, | |
| "loss": 2.583, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 11.59, | |
| "learning_rate": 0.0008900000000000001, | |
| "loss": 2.4486, | |
| "step": 3560 | |
| }, | |
| { | |
| "epoch": 11.62, | |
| "learning_rate": 0.0008925, | |
| "loss": 2.4851, | |
| "step": 3570 | |
| }, | |
| { | |
| "epoch": 11.65, | |
| "learning_rate": 0.0008950000000000001, | |
| "loss": 2.4667, | |
| "step": 3580 | |
| }, | |
| { | |
| "epoch": 11.69, | |
| "learning_rate": 0.0008975, | |
| "loss": 2.5137, | |
| "step": 3590 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "learning_rate": 0.0009000000000000001, | |
| "loss": 2.4139, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 26811, | |
| 18539, | |
| 13636, | |
| 10289 | |
| ], | |
| "eval_loss": 1.4529476165771484, | |
| "eval_precisions": [ | |
| 64.42938504794175, | |
| 48.595019659239846, | |
| 39.303625987202395, | |
| 32.89322250639386 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 110.5018, | |
| "eval_samples_per_second": 31.339, | |
| "eval_score": 44.8542601307095, | |
| "eval_steps_per_second": 31.339, | |
| "eval_sys_len": 41613, | |
| "eval_totals": [ | |
| 41613, | |
| 38150, | |
| 34694, | |
| 31280 | |
| ], | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 11.75, | |
| "learning_rate": 0.0009025, | |
| "loss": 2.5072, | |
| "step": 3610 | |
| }, | |
| { | |
| "epoch": 11.79, | |
| "learning_rate": 0.0009050000000000001, | |
| "loss": 2.5271, | |
| "step": 3620 | |
| }, | |
| { | |
| "epoch": 11.82, | |
| "learning_rate": 0.0009075, | |
| "loss": 2.5521, | |
| "step": 3630 | |
| }, | |
| { | |
| "epoch": 11.85, | |
| "learning_rate": 0.00091, | |
| "loss": 2.3746, | |
| "step": 3640 | |
| }, | |
| { | |
| "epoch": 11.88, | |
| "learning_rate": 0.0009125, | |
| "loss": 2.5233, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 11.92, | |
| "learning_rate": 0.000915, | |
| "loss": 2.4914, | |
| "step": 3660 | |
| }, | |
| { | |
| "epoch": 11.95, | |
| "learning_rate": 0.0009175, | |
| "loss": 2.5744, | |
| "step": 3670 | |
| }, | |
| { | |
| "epoch": 11.98, | |
| "learning_rate": 0.00092, | |
| "loss": 2.5242, | |
| "step": 3680 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 0.0009225, | |
| "loss": 2.4918, | |
| "step": 3690 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 0.000925, | |
| "loss": 2.4723, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 12.08, | |
| "learning_rate": 0.0009275, | |
| "loss": 2.3919, | |
| "step": 3710 | |
| }, | |
| { | |
| "epoch": 12.11, | |
| "learning_rate": 0.00093, | |
| "loss": 2.4419, | |
| "step": 3720 | |
| }, | |
| { | |
| "epoch": 12.14, | |
| "learning_rate": 0.0009325000000000001, | |
| "loss": 2.3503, | |
| "step": 3730 | |
| }, | |
| { | |
| "epoch": 12.18, | |
| "learning_rate": 0.0009350000000000001, | |
| "loss": 2.388, | |
| "step": 3740 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "learning_rate": 0.0009375, | |
| "loss": 2.4935, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 12.24, | |
| "learning_rate": 0.00094, | |
| "loss": 2.3981, | |
| "step": 3760 | |
| }, | |
| { | |
| "epoch": 12.27, | |
| "learning_rate": 0.0009425, | |
| "loss": 2.3666, | |
| "step": 3770 | |
| }, | |
| { | |
| "epoch": 12.31, | |
| "learning_rate": 0.000945, | |
| "loss": 2.4323, | |
| "step": 3780 | |
| }, | |
| { | |
| "epoch": 12.34, | |
| "learning_rate": 0.0009475, | |
| "loss": 2.4228, | |
| "step": 3790 | |
| }, | |
| { | |
| "epoch": 12.37, | |
| "learning_rate": 0.00095, | |
| "loss": 2.4972, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 12.37, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 26521, | |
| 18238, | |
| 13350, | |
| 10018 | |
| ], | |
| "eval_loss": 1.4506553411483765, | |
| "eval_precisions": [ | |
| 58.50005514503143, | |
| 43.556553305311425, | |
| 34.72312534137904, | |
| 28.567354853427627 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 144.0478, | |
| "eval_samples_per_second": 24.041, | |
| "eval_score": 39.87260907407475, | |
| "eval_steps_per_second": 24.041, | |
| "eval_sys_len": 45335, | |
| "eval_totals": [ | |
| 45335, | |
| 41872, | |
| 38447, | |
| 35068 | |
| ], | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 12.4, | |
| "learning_rate": 0.0009525, | |
| "loss": 2.4917, | |
| "step": 3810 | |
| }, | |
| { | |
| "epoch": 12.44, | |
| "learning_rate": 0.000955, | |
| "loss": 2.432, | |
| "step": 3820 | |
| }, | |
| { | |
| "epoch": 12.47, | |
| "learning_rate": 0.0009575, | |
| "loss": 2.4027, | |
| "step": 3830 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "learning_rate": 0.00096, | |
| "loss": 2.4329, | |
| "step": 3840 | |
| }, | |
| { | |
| "epoch": 12.53, | |
| "learning_rate": 0.0009625, | |
| "loss": 2.4021, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 12.57, | |
| "learning_rate": 0.000965, | |
| "loss": 2.5415, | |
| "step": 3860 | |
| }, | |
| { | |
| "epoch": 12.6, | |
| "learning_rate": 0.0009675, | |
| "loss": 2.3346, | |
| "step": 3870 | |
| }, | |
| { | |
| "epoch": 12.63, | |
| "learning_rate": 0.0009699999999999999, | |
| "loss": 2.4081, | |
| "step": 3880 | |
| }, | |
| { | |
| "epoch": 12.66, | |
| "learning_rate": 0.0009725000000000001, | |
| "loss": 2.419, | |
| "step": 3890 | |
| }, | |
| { | |
| "epoch": 12.7, | |
| "learning_rate": 0.000975, | |
| "loss": 2.4059, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 12.73, | |
| "learning_rate": 0.0009775, | |
| "loss": 2.3385, | |
| "step": 3910 | |
| }, | |
| { | |
| "epoch": 12.76, | |
| "learning_rate": 0.00098, | |
| "loss": 2.4485, | |
| "step": 3920 | |
| }, | |
| { | |
| "epoch": 12.79, | |
| "learning_rate": 0.0009825, | |
| "loss": 2.3762, | |
| "step": 3930 | |
| }, | |
| { | |
| "epoch": 12.83, | |
| "learning_rate": 0.000985, | |
| "loss": 2.4831, | |
| "step": 3940 | |
| }, | |
| { | |
| "epoch": 12.86, | |
| "learning_rate": 0.0009875, | |
| "loss": 2.4976, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 12.89, | |
| "learning_rate": 0.00099, | |
| "loss": 2.4762, | |
| "step": 3960 | |
| }, | |
| { | |
| "epoch": 12.92, | |
| "learning_rate": 0.0009925000000000001, | |
| "loss": 2.4236, | |
| "step": 3970 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "learning_rate": 0.000995, | |
| "loss": 2.4216, | |
| "step": 3980 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "learning_rate": 0.0009975000000000001, | |
| "loss": 2.3395, | |
| "step": 3990 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 0.001, | |
| "loss": 2.3016, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "eval_bp": 0.9913107345232208, | |
| "eval_counts": [ | |
| 28315, | |
| 19934, | |
| 14780, | |
| 11227 | |
| ], | |
| "eval_loss": 1.4572792053222656, | |
| "eval_precisions": [ | |
| 69.02562102337826, | |
| 53.07524362319612, | |
| 43.30881706566648, | |
| 36.52244632400781 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 79.9221, | |
| "eval_samples_per_second": 43.33, | |
| "eval_score": 48.63729731945582, | |
| "eval_steps_per_second": 43.33, | |
| "eval_sys_len": 41021, | |
| "eval_totals": [ | |
| 41021, | |
| 37558, | |
| 34127, | |
| 30740 | |
| ], | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 13.05, | |
| "learning_rate": 0.001, | |
| "loss": 2.4317, | |
| "step": 4010 | |
| }, | |
| { | |
| "epoch": 13.09, | |
| "learning_rate": 0.001, | |
| "loss": 2.3301, | |
| "step": 4020 | |
| }, | |
| { | |
| "epoch": 13.12, | |
| "learning_rate": 0.001, | |
| "loss": 2.3254, | |
| "step": 4030 | |
| }, | |
| { | |
| "epoch": 13.15, | |
| "learning_rate": 0.001, | |
| "loss": 2.3389, | |
| "step": 4040 | |
| }, | |
| { | |
| "epoch": 13.19, | |
| "learning_rate": 0.001, | |
| "loss": 2.4051, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 13.22, | |
| "learning_rate": 0.001, | |
| "loss": 2.3279, | |
| "step": 4060 | |
| }, | |
| { | |
| "epoch": 13.25, | |
| "learning_rate": 0.001, | |
| "loss": 2.3066, | |
| "step": 4070 | |
| }, | |
| { | |
| "epoch": 13.28, | |
| "learning_rate": 0.001, | |
| "loss": 2.3249, | |
| "step": 4080 | |
| }, | |
| { | |
| "epoch": 13.32, | |
| "learning_rate": 0.001, | |
| "loss": 2.3425, | |
| "step": 4090 | |
| }, | |
| { | |
| "epoch": 13.35, | |
| "learning_rate": 0.001, | |
| "loss": 2.2547, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 13.38, | |
| "learning_rate": 0.001, | |
| "loss": 2.3656, | |
| "step": 4110 | |
| }, | |
| { | |
| "epoch": 13.41, | |
| "learning_rate": 0.001, | |
| "loss": 2.4137, | |
| "step": 4120 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "learning_rate": 0.001, | |
| "loss": 2.3343, | |
| "step": 4130 | |
| }, | |
| { | |
| "epoch": 13.48, | |
| "learning_rate": 0.001, | |
| "loss": 2.3188, | |
| "step": 4140 | |
| }, | |
| { | |
| "epoch": 13.51, | |
| "learning_rate": 0.001, | |
| "loss": 2.4557, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 13.54, | |
| "learning_rate": 0.001, | |
| "loss": 2.2675, | |
| "step": 4160 | |
| }, | |
| { | |
| "epoch": 13.58, | |
| "learning_rate": 0.001, | |
| "loss": 2.3623, | |
| "step": 4170 | |
| }, | |
| { | |
| "epoch": 13.61, | |
| "learning_rate": 0.001, | |
| "loss": 2.396, | |
| "step": 4180 | |
| }, | |
| { | |
| "epoch": 13.64, | |
| "learning_rate": 0.001, | |
| "loss": 2.3394, | |
| "step": 4190 | |
| }, | |
| { | |
| "epoch": 13.67, | |
| "learning_rate": 0.001, | |
| "loss": 2.385, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 13.67, | |
| "eval_bp": 0.9989119002923912, | |
| "eval_counts": [ | |
| 26679, | |
| 18540, | |
| 13665, | |
| 10316 | |
| ], | |
| "eval_loss": 1.4272891283035278, | |
| "eval_precisions": [ | |
| 64.54492669473073, | |
| 48.955665284782555, | |
| 39.701908829425605, | |
| 33.265615426784045 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 99.3005, | |
| "eval_samples_per_second": 34.874, | |
| "eval_score": 45.14869395785754, | |
| "eval_steps_per_second": 34.874, | |
| "eval_sys_len": 41334, | |
| "eval_totals": [ | |
| 41334, | |
| 37871, | |
| 34419, | |
| 31011 | |
| ], | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 13.71, | |
| "learning_rate": 0.001, | |
| "loss": 2.2847, | |
| "step": 4210 | |
| }, | |
| { | |
| "epoch": 13.74, | |
| "learning_rate": 0.001, | |
| "loss": 2.3304, | |
| "step": 4220 | |
| }, | |
| { | |
| "epoch": 13.77, | |
| "learning_rate": 0.001, | |
| "loss": 2.3681, | |
| "step": 4230 | |
| }, | |
| { | |
| "epoch": 13.8, | |
| "learning_rate": 0.001, | |
| "loss": 2.3145, | |
| "step": 4240 | |
| }, | |
| { | |
| "epoch": 13.84, | |
| "learning_rate": 0.001, | |
| "loss": 2.3572, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 13.87, | |
| "learning_rate": 0.001, | |
| "loss": 2.3935, | |
| "step": 4260 | |
| }, | |
| { | |
| "epoch": 13.9, | |
| "learning_rate": 0.001, | |
| "loss": 2.2827, | |
| "step": 4270 | |
| }, | |
| { | |
| "epoch": 13.93, | |
| "learning_rate": 0.001, | |
| "loss": 2.3749, | |
| "step": 4280 | |
| }, | |
| { | |
| "epoch": 13.97, | |
| "learning_rate": 0.001, | |
| "loss": 2.3992, | |
| "step": 4290 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 0.001, | |
| "loss": 2.398, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "learning_rate": 0.001, | |
| "loss": 2.3011, | |
| "step": 4310 | |
| }, | |
| { | |
| "epoch": 14.06, | |
| "learning_rate": 0.001, | |
| "loss": 2.2635, | |
| "step": 4320 | |
| }, | |
| { | |
| "epoch": 14.1, | |
| "learning_rate": 0.001, | |
| "loss": 2.2014, | |
| "step": 4330 | |
| }, | |
| { | |
| "epoch": 14.13, | |
| "learning_rate": 0.001, | |
| "loss": 2.2523, | |
| "step": 4340 | |
| }, | |
| { | |
| "epoch": 14.16, | |
| "learning_rate": 0.001, | |
| "loss": 2.2941, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 14.19, | |
| "learning_rate": 0.001, | |
| "loss": 2.2856, | |
| "step": 4360 | |
| }, | |
| { | |
| "epoch": 14.23, | |
| "learning_rate": 0.001, | |
| "loss": 2.2582, | |
| "step": 4370 | |
| }, | |
| { | |
| "epoch": 14.26, | |
| "learning_rate": 0.001, | |
| "loss": 2.2663, | |
| "step": 4380 | |
| }, | |
| { | |
| "epoch": 14.29, | |
| "learning_rate": 0.001, | |
| "loss": 2.2206, | |
| "step": 4390 | |
| }, | |
| { | |
| "epoch": 14.32, | |
| "learning_rate": 0.001, | |
| "loss": 2.2694, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 14.32, | |
| "eval_bp": 0.9634105401802233, | |
| "eval_counts": [ | |
| 27903, | |
| 19593, | |
| 14525, | |
| 11034 | |
| ], | |
| "eval_loss": 1.4508273601531982, | |
| "eval_precisions": [ | |
| 69.94635515892911, | |
| 53.78407312855143, | |
| 44.05119340066115, | |
| 37.32873236577692 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 78.5863, | |
| "eval_samples_per_second": 44.066, | |
| "eval_score": 48.04700463705004, | |
| "eval_steps_per_second": 44.066, | |
| "eval_sys_len": 39892, | |
| "eval_totals": [ | |
| 39892, | |
| 36429, | |
| 32973, | |
| 29559 | |
| ], | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 14.36, | |
| "learning_rate": 0.001, | |
| "loss": 2.3494, | |
| "step": 4410 | |
| }, | |
| { | |
| "epoch": 14.39, | |
| "learning_rate": 0.001, | |
| "loss": 2.3291, | |
| "step": 4420 | |
| }, | |
| { | |
| "epoch": 14.42, | |
| "learning_rate": 0.001, | |
| "loss": 2.2417, | |
| "step": 4430 | |
| }, | |
| { | |
| "epoch": 14.45, | |
| "learning_rate": 0.001, | |
| "loss": 2.2454, | |
| "step": 4440 | |
| }, | |
| { | |
| "epoch": 14.49, | |
| "learning_rate": 0.001, | |
| "loss": 2.2897, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 14.52, | |
| "learning_rate": 0.001, | |
| "loss": 2.2596, | |
| "step": 4460 | |
| }, | |
| { | |
| "epoch": 14.55, | |
| "learning_rate": 0.001, | |
| "loss": 2.2098, | |
| "step": 4470 | |
| }, | |
| { | |
| "epoch": 14.58, | |
| "learning_rate": 0.001, | |
| "loss": 2.2473, | |
| "step": 4480 | |
| }, | |
| { | |
| "epoch": 14.62, | |
| "learning_rate": 0.001, | |
| "loss": 2.3361, | |
| "step": 4490 | |
| }, | |
| { | |
| "epoch": 14.65, | |
| "learning_rate": 0.001, | |
| "loss": 2.2572, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 14.68, | |
| "learning_rate": 0.001, | |
| "loss": 2.2152, | |
| "step": 4510 | |
| }, | |
| { | |
| "epoch": 14.72, | |
| "learning_rate": 0.001, | |
| "loss": 2.3067, | |
| "step": 4520 | |
| }, | |
| { | |
| "epoch": 14.75, | |
| "learning_rate": 0.001, | |
| "loss": 2.2733, | |
| "step": 4530 | |
| }, | |
| { | |
| "epoch": 14.78, | |
| "learning_rate": 0.001, | |
| "loss": 2.2986, | |
| "step": 4540 | |
| }, | |
| { | |
| "epoch": 14.81, | |
| "learning_rate": 0.001, | |
| "loss": 2.2666, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 14.85, | |
| "learning_rate": 0.001, | |
| "loss": 2.3507, | |
| "step": 4560 | |
| }, | |
| { | |
| "epoch": 14.88, | |
| "learning_rate": 0.001, | |
| "loss": 2.3074, | |
| "step": 4570 | |
| }, | |
| { | |
| "epoch": 14.91, | |
| "learning_rate": 0.001, | |
| "loss": 2.3392, | |
| "step": 4580 | |
| }, | |
| { | |
| "epoch": 14.94, | |
| "learning_rate": 0.001, | |
| "loss": 2.2536, | |
| "step": 4590 | |
| }, | |
| { | |
| "epoch": 14.98, | |
| "learning_rate": 0.001, | |
| "loss": 2.2633, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 14.98, | |
| "eval_bp": 0.9646873396517708, | |
| "eval_counts": [ | |
| 27585, | |
| 19365, | |
| 14361, | |
| 10897 | |
| ], | |
| "eval_loss": 1.4220008850097656, | |
| "eval_precisions": [ | |
| 69.06091179931403, | |
| 53.08388157894737, | |
| 43.48523845571537, | |
| 36.79802789315503 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 85.4344, | |
| "eval_samples_per_second": 40.534, | |
| "eval_score": 47.47630639176475, | |
| "eval_steps_per_second": 40.534, | |
| "eval_sys_len": 39943, | |
| "eval_totals": [ | |
| 39943, | |
| 36480, | |
| 33025, | |
| 29613 | |
| ], | |
| "step": 4600 | |
| } | |
| ], | |
| "max_steps": 76750, | |
| "num_train_epochs": 250, | |
| "total_flos": 4.434135803203584e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |