| { | |
| "best_metric": 0.3959246873855591, | |
| "best_model_checkpoint": "/scratch/s3545881/dumped/translation/byt5/2753838/checkpoint-7000", | |
| "epoch": 6.915857072287173, | |
| "global_step": 7000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 2.5e-06, | |
| "loss": 5.8979, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 5e-06, | |
| "loss": 5.3917, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 7.5e-06, | |
| "loss": 4.5382, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 1e-05, | |
| "loss": 3.6672, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 1.25e-05, | |
| "loss": 2.8366, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1.5e-05, | |
| "loss": 2.219, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.7500000000000002e-05, | |
| "loss": 1.802, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 2e-05, | |
| "loss": 1.6535, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 2.2499999999999998e-05, | |
| "loss": 1.5583, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.524, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 2.75e-05, | |
| "loss": 1.4785, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 3e-05, | |
| "loss": 1.4156, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 1.3735, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 3.5000000000000004e-05, | |
| "loss": 1.3937, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 3.75e-05, | |
| "loss": 1.3698, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 4e-05, | |
| "loss": 1.3565, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.25e-05, | |
| "loss": 1.2993, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.4999999999999996e-05, | |
| "loss": 1.3186, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 4.75e-05, | |
| "loss": 1.3235, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 5e-05, | |
| "loss": 1.302, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 20154, | |
| 11402, | |
| 7266, | |
| 4956 | |
| ], | |
| "eval_loss": 0.6817044019699097, | |
| "eval_precisions": [ | |
| 11.08550369901818, | |
| 6.392366386535777, | |
| 4.153542744447938, | |
| 2.8893410366881014 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 839.9367, | |
| "eval_samples_per_second": 4.123, | |
| "eval_score": 5.400185085993893, | |
| "eval_steps_per_second": 4.123, | |
| "eval_sys_len": 181805, | |
| "eval_totals": [ | |
| 181805, | |
| 178369, | |
| 174935, | |
| 171527 | |
| ], | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 5.25e-05, | |
| "loss": 1.2825, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 5.5e-05, | |
| "loss": 1.245, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 5.75e-05, | |
| "loss": 1.2638, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 6e-05, | |
| "loss": 1.2674, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 6.25e-05, | |
| "loss": 1.2135, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 1.2186, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 6.75e-05, | |
| "loss": 1.2385, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 7.000000000000001e-05, | |
| "loss": 1.1997, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 7.25e-05, | |
| "loss": 1.1891, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 7.5e-05, | |
| "loss": 1.1812, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 7.75e-05, | |
| "loss": 1.2191, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 8e-05, | |
| "loss": 1.2003, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 8.25e-05, | |
| "loss": 1.1914, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 8.5e-05, | |
| "loss": 1.2244, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 8.75e-05, | |
| "loss": 1.1732, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 8.999999999999999e-05, | |
| "loss": 1.1308, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 9.25e-05, | |
| "loss": 1.1218, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 9.5e-05, | |
| "loss": 1.1612, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 9.750000000000001e-05, | |
| "loss": 1.1436, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0001, | |
| "loss": 1.1089, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23672, | |
| 14726, | |
| 10093, | |
| 7300 | |
| ], | |
| "eval_loss": 0.5815275311470032, | |
| "eval_precisions": [ | |
| 34.68069209018855, | |
| 22.72741303207087, | |
| 16.455531099698376, | |
| 12.60511456840433 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 489.5104, | |
| "eval_samples_per_second": 7.074, | |
| "eval_score": 20.108242558207973, | |
| "eval_steps_per_second": 7.074, | |
| "eval_sys_len": 68257, | |
| "eval_totals": [ | |
| 68257, | |
| 64794, | |
| 61335, | |
| 57913 | |
| ], | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0001025, | |
| "loss": 1.1308, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.000105, | |
| "loss": 1.1377, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0001075, | |
| "loss": 1.131, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00011, | |
| "loss": 1.1321, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00011250000000000001, | |
| "loss": 1.1045, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.000115, | |
| "loss": 1.115, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0001175, | |
| "loss": 1.0983, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00012, | |
| "loss": 1.0936, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0001225, | |
| "loss": 1.1068, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.000125, | |
| "loss": 1.1375, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0001275, | |
| "loss": 1.1024, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00013000000000000002, | |
| "loss": 1.0603, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00013250000000000002, | |
| "loss": 1.0914, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.000135, | |
| "loss": 1.0847, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0001375, | |
| "loss": 1.0648, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.00014000000000000001, | |
| "loss": 1.1071, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.0001425, | |
| "loss": 1.1093, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.000145, | |
| "loss": 1.1, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0001475, | |
| "loss": 1.0543, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00015, | |
| "loss": 1.0418, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 25499, | |
| 16524, | |
| 11653, | |
| 8563 | |
| ], | |
| "eval_loss": 0.5327444672584534, | |
| "eval_precisions": [ | |
| 44.343773368346, | |
| 30.577350111028867, | |
| 23.035562496293515, | |
| 18.152332902295804 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 410.6954, | |
| "eval_samples_per_second": 8.432, | |
| "eval_score": 27.440431748782675, | |
| "eval_steps_per_second": 8.432, | |
| "eval_sys_len": 57503, | |
| "eval_totals": [ | |
| 57503, | |
| 54040, | |
| 50587, | |
| 47173 | |
| ], | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.0001525, | |
| "loss": 1.0696, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.000155, | |
| "loss": 1.0774, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0001575, | |
| "loss": 1.0745, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00016, | |
| "loss": 1.0698, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.00016250000000000002, | |
| "loss": 1.0809, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.000165, | |
| "loss": 1.0698, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0001675, | |
| "loss": 1.0117, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.00017, | |
| "loss": 1.0663, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.0001725, | |
| "loss": 1.045, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.000175, | |
| "loss": 1.0662, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.0001775, | |
| "loss": 1.0427, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 1.0554, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.0001825, | |
| "loss": 1.0357, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.000185, | |
| "loss": 1.0215, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.0001875, | |
| "loss": 1.0497, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00019, | |
| "loss": 1.0363, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00019250000000000002, | |
| "loss": 1.0369, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00019500000000000002, | |
| "loss": 1.0462, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0001975, | |
| "loss": 1.0045, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0481, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 26102, | |
| 17383, | |
| 12439, | |
| 9213 | |
| ], | |
| "eval_loss": 0.5049412250518799, | |
| "eval_precisions": [ | |
| 45.29316837009145, | |
| 32.09208728722815, | |
| 24.529678564385723, | |
| 19.481096168485156 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 406.3473, | |
| "eval_samples_per_second": 8.522, | |
| "eval_score": 28.86914876707736, | |
| "eval_steps_per_second": 8.522, | |
| "eval_sys_len": 57629, | |
| "eval_totals": [ | |
| 57629, | |
| 54166, | |
| 50710, | |
| 47292 | |
| ], | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.00020250000000000002, | |
| "loss": 1.0076, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.000205, | |
| "loss": 0.9906, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.0002075, | |
| "loss": 1.0277, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 0.00021, | |
| "loss": 1.0217, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.0002125, | |
| "loss": 0.9918, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.000215, | |
| "loss": 0.9979, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.0002175, | |
| "loss": 1.011, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.00022, | |
| "loss": 1.009, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 0.00022250000000000001, | |
| "loss": 1.0135, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.00022500000000000002, | |
| "loss": 0.993, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.0002275, | |
| "loss": 0.9985, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.00023, | |
| "loss": 1.0029, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.0002325, | |
| "loss": 0.9864, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.000235, | |
| "loss": 1.007, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.0002375, | |
| "loss": 0.9633, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.00024, | |
| "loss": 0.9491, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.00024249999999999999, | |
| "loss": 0.951, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.000245, | |
| "loss": 0.9758, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 0.0002475, | |
| "loss": 0.911, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.00025, | |
| "loss": 0.9828, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 27510, | |
| 18772, | |
| 13695, | |
| 10271 | |
| ], | |
| "eval_loss": 0.4844214916229248, | |
| "eval_precisions": [ | |
| 60.232522496879994, | |
| 44.47287372660507, | |
| 35.33828766062858, | |
| 29.06585165690353 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 267.7695, | |
| "eval_samples_per_second": 12.933, | |
| "eval_score": 40.72758885216857, | |
| "eval_steps_per_second": 12.933, | |
| "eval_sys_len": 45673, | |
| "eval_totals": [ | |
| 45673, | |
| 42210, | |
| 38754, | |
| 35337 | |
| ], | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0002525, | |
| "loss": 0.981, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 0.000255, | |
| "loss": 0.9364, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002575, | |
| "loss": 0.9354, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 0.00026000000000000003, | |
| "loss": 0.9955, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 0.00026250000000000004, | |
| "loss": 0.9585, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 0.00026500000000000004, | |
| "loss": 0.9957, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.0002675, | |
| "loss": 0.9416, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.00027, | |
| "loss": 0.9893, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.0002725, | |
| "loss": 0.9634, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.000275, | |
| "loss": 0.9614, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 0.0002775, | |
| "loss": 0.9511, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.00028000000000000003, | |
| "loss": 0.9351, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.0002825, | |
| "loss": 0.9485, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.000285, | |
| "loss": 0.9985, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 0.0002875, | |
| "loss": 0.9533, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 0.00029, | |
| "loss": 0.9334, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 0.0002925, | |
| "loss": 0.8978, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.000295, | |
| "loss": 0.9695, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.00029749999999999997, | |
| "loss": 0.9135, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 0.0003, | |
| "loss": 0.9443, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 27829, | |
| 19245, | |
| 14138, | |
| 10671 | |
| ], | |
| "eval_loss": 0.477074533700943, | |
| "eval_precisions": [ | |
| 66.99971109399075, | |
| 50.54763218028524, | |
| 40.842385024266235, | |
| 34.20521203961919 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 216.6085, | |
| "eval_samples_per_second": 15.987, | |
| "eval_score": 46.63846865322161, | |
| "eval_steps_per_second": 15.987, | |
| "eval_sys_len": 41536, | |
| "eval_totals": [ | |
| 41536, | |
| 38073, | |
| 34616, | |
| 31197 | |
| ], | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.0003025, | |
| "loss": 0.952, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 0.000305, | |
| "loss": 0.9502, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 0.0003075, | |
| "loss": 0.9274, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.00031, | |
| "loss": 0.9873, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.0003125, | |
| "loss": 0.9588, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.000315, | |
| "loss": 0.9835, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.0003175, | |
| "loss": 0.9558, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 0.00032, | |
| "loss": 0.9432, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 0.00032250000000000003, | |
| "loss": 0.9229, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.00032500000000000004, | |
| "loss": 0.9095, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.00032750000000000005, | |
| "loss": 0.9503, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 0.00033, | |
| "loss": 0.8999, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.0003325, | |
| "loss": 0.9129, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 0.000335, | |
| "loss": 0.9508, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0003375, | |
| "loss": 0.9156, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.00034, | |
| "loss": 0.916, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 0.00034250000000000003, | |
| "loss": 0.9156, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 0.000345, | |
| "loss": 0.9125, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 0.0003475, | |
| "loss": 0.9151, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.00035, | |
| "loss": 0.9365, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 27732, | |
| 19213, | |
| 14172, | |
| 10741 | |
| ], | |
| "eval_loss": 0.4637322425842285, | |
| "eval_precisions": [ | |
| 60.56873280041934, | |
| 45.39612031283227, | |
| 36.4618709478234, | |
| 30.29901269393512 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 290.4637, | |
| "eval_samples_per_second": 11.922, | |
| "eval_score": 41.747806624471906, | |
| "eval_steps_per_second": 11.922, | |
| "eval_sys_len": 45786, | |
| "eval_totals": [ | |
| 45786, | |
| 42323, | |
| 38868, | |
| 35450 | |
| ], | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 0.0003525, | |
| "loss": 0.928, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.000355, | |
| "loss": 0.9327, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 0.0003575, | |
| "loss": 0.9028, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.00035999999999999997, | |
| "loss": 0.9209, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.0003625, | |
| "loss": 0.9058, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.000365, | |
| "loss": 0.8938, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.0003675, | |
| "loss": 0.942, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.00037, | |
| "loss": 0.9019, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.0003725, | |
| "loss": 0.9188, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 0.000375, | |
| "loss": 0.9241, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 0.0003775, | |
| "loss": 0.9152, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 0.00038, | |
| "loss": 0.9218, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 0.00038250000000000003, | |
| "loss": 0.9115, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 0.00038500000000000003, | |
| "loss": 0.9215, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.00038750000000000004, | |
| "loss": 0.9093, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.00039000000000000005, | |
| "loss": 0.9076, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 0.0003925, | |
| "loss": 0.8922, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.000395, | |
| "loss": 0.9248, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 0.0003975, | |
| "loss": 0.9009, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 0.0004, | |
| "loss": 0.8859, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 27542, | |
| 19232, | |
| 14275, | |
| 10862 | |
| ], | |
| "eval_loss": 0.45880991220474243, | |
| "eval_precisions": [ | |
| 55.53270425034277, | |
| 41.68816248672317, | |
| 33.448146586063075, | |
| 27.666131784722754 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 328.8521, | |
| "eval_samples_per_second": 10.531, | |
| "eval_score": 38.257846353776635, | |
| "eval_steps_per_second": 10.531, | |
| "eval_sys_len": 49596, | |
| "eval_totals": [ | |
| 49596, | |
| 46133, | |
| 42678, | |
| 39261 | |
| ], | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 0.0004025, | |
| "loss": 0.8769, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.00040500000000000003, | |
| "loss": 0.9273, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.0004075, | |
| "loss": 0.8746, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.00041, | |
| "loss": 0.8925, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 0.0004125, | |
| "loss": 0.8805, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.000415, | |
| "loss": 0.9323, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 0.0004175, | |
| "loss": 0.9355, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 0.00042, | |
| "loss": 0.894, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.00042249999999999997, | |
| "loss": 0.8662, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.000425, | |
| "loss": 0.8803, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.0004275, | |
| "loss": 0.849, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 0.00043, | |
| "loss": 0.8991, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 0.0004325, | |
| "loss": 0.9301, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 0.000435, | |
| "loss": 0.8748, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 0.0004375, | |
| "loss": 0.907, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 0.00044, | |
| "loss": 0.9047, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.0004425, | |
| "loss": 0.8688, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 0.00044500000000000003, | |
| "loss": 0.8672, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 0.00044750000000000004, | |
| "loss": 0.8761, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.00045000000000000004, | |
| "loss": 0.8744, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 28424, | |
| 19992, | |
| 14908, | |
| 11378 | |
| ], | |
| "eval_loss": 0.45066994428634644, | |
| "eval_precisions": [ | |
| 66.64165806996155, | |
| 51.01431524152186, | |
| 41.720538437858565, | |
| 35.20965495899737 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 254.1519, | |
| "eval_samples_per_second": 13.626, | |
| "eval_score": 47.272914915461314, | |
| "eval_steps_per_second": 13.626, | |
| "eval_sys_len": 42652, | |
| "eval_totals": [ | |
| 42652, | |
| 39189, | |
| 35733, | |
| 32315 | |
| ], | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.00045250000000000005, | |
| "loss": 0.9204, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 0.000455, | |
| "loss": 0.9004, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 0.0004575, | |
| "loss": 0.9017, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.00046, | |
| "loss": 0.8787, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 0.0004625, | |
| "loss": 0.9, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 0.000465, | |
| "loss": 0.8753, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.00046750000000000003, | |
| "loss": 0.8835, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 0.00047, | |
| "loss": 0.8641, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.0004725, | |
| "loss": 0.8414, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 0.000475, | |
| "loss": 0.8794, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0004775, | |
| "loss": 0.8829, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 0.00048, | |
| "loss": 0.9059, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 0.0004825, | |
| "loss": 0.904, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.00048499999999999997, | |
| "loss": 0.9051, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 0.0004875, | |
| "loss": 0.8379, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.00049, | |
| "loss": 0.8869, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 0.0004925, | |
| "loss": 0.8636, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 0.000495, | |
| "loss": 0.8725, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.0004975, | |
| "loss": 0.8284, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 0.0005, | |
| "loss": 0.9, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 26501, | |
| 18108, | |
| 13245, | |
| 9944 | |
| ], | |
| "eval_loss": 0.4501844048500061, | |
| "eval_precisions": [ | |
| 47.71687853361662, | |
| 34.77292366778685, | |
| 27.241875771287535, | |
| 22.0 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 416.5193, | |
| "eval_samples_per_second": 8.314, | |
| "eval_score": 31.578626057173885, | |
| "eval_steps_per_second": 8.314, | |
| "eval_sys_len": 55538, | |
| "eval_totals": [ | |
| 55538, | |
| 52075, | |
| 48620, | |
| 45200 | |
| ], | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.0005024999999999999, | |
| "loss": 0.9086, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.000505, | |
| "loss": 0.8709, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 0.0005074999999999999, | |
| "loss": 0.8838, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 0.00051, | |
| "loss": 0.8664, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 0.0005124999999999999, | |
| "loss": 0.8533, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.000515, | |
| "loss": 0.8538, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 0.0005175, | |
| "loss": 0.8728, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 0.0005200000000000001, | |
| "loss": 0.8802, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 0.0005225, | |
| "loss": 0.8765, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 0.0005250000000000001, | |
| "loss": 0.8399, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 0.0005275, | |
| "loss": 0.855, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 0.0005300000000000001, | |
| "loss": 0.8856, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 0.0005325, | |
| "loss": 0.8805, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.000535, | |
| "loss": 0.83, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 0.0005375, | |
| "loss": 0.85, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 0.00054, | |
| "loss": 0.834, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 0.0005425, | |
| "loss": 0.8506, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.000545, | |
| "loss": 0.8277, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 0.0005475, | |
| "loss": 0.8653, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 0.00055, | |
| "loss": 0.8337, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 27954, | |
| 19776, | |
| 14846, | |
| 11400 | |
| ], | |
| "eval_loss": 0.44147148728370667, | |
| "eval_precisions": [ | |
| 54.01217273693363, | |
| 40.95088213368674, | |
| 33.11178517262914, | |
| 27.52891743739586 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 372.4047, | |
| "eval_samples_per_second": 9.299, | |
| "eval_score": 37.6818016846399, | |
| "eval_steps_per_second": 9.299, | |
| "eval_sys_len": 51755, | |
| "eval_totals": [ | |
| 51755, | |
| 48292, | |
| 44836, | |
| 41411 | |
| ], | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 0.0005525, | |
| "loss": 0.862, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 0.000555, | |
| "loss": 0.837, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 0.0005575, | |
| "loss": 0.8238, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 0.0005600000000000001, | |
| "loss": 0.8241, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 0.0005625000000000001, | |
| "loss": 0.8775, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 0.000565, | |
| "loss": 0.8661, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 0.0005675, | |
| "loss": 0.8012, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 0.00057, | |
| "loss": 0.8255, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.0005725, | |
| "loss": 0.8481, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 0.000575, | |
| "loss": 0.8623, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 0.0005775, | |
| "loss": 0.8029, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 0.00058, | |
| "loss": 0.8435, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 0.0005825, | |
| "loss": 0.8466, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 0.000585, | |
| "loss": 0.8272, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 0.0005875, | |
| "loss": 0.8337, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.00059, | |
| "loss": 0.8159, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 0.0005925, | |
| "loss": 0.8517, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 0.0005949999999999999, | |
| "loss": 0.8434, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 0.0005975, | |
| "loss": 0.8524, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 0.0006, | |
| "loss": 0.8227, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 28249, | |
| 20096, | |
| 15121, | |
| 11638 | |
| ], | |
| "eval_loss": 0.43790727853775024, | |
| "eval_precisions": [ | |
| 58.84105063633902, | |
| 45.11291698468998, | |
| 36.800603567864876, | |
| 30.896251460125306 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 312.0886, | |
| "eval_samples_per_second": 11.096, | |
| "eval_score": 41.680749055432095, | |
| "eval_steps_per_second": 11.096, | |
| "eval_sys_len": 48009, | |
| "eval_totals": [ | |
| 48009, | |
| 44546, | |
| 41089, | |
| 37668 | |
| ], | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 0.0006025000000000001, | |
| "loss": 0.863, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 0.000605, | |
| "loss": 0.7905, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0006075000000000001, | |
| "loss": 0.8278, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 0.00061, | |
| "loss": 0.78, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 0.0006125000000000001, | |
| "loss": 0.8673, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 0.000615, | |
| "loss": 0.8446, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 0.0006175000000000001, | |
| "loss": 0.8655, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 0.00062, | |
| "loss": 0.8507, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 0.0006225000000000001, | |
| "loss": 0.8115, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 0.000625, | |
| "loss": 0.8337, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 0.0006274999999999999, | |
| "loss": 0.8307, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 0.00063, | |
| "loss": 0.7888, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 0.0006324999999999999, | |
| "loss": 0.8345, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 0.000635, | |
| "loss": 0.8155, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 0.0006374999999999999, | |
| "loss": 0.8321, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 0.00064, | |
| "loss": 0.8516, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 0.0006425, | |
| "loss": 0.8124, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 0.0006450000000000001, | |
| "loss": 0.8382, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 0.0006475, | |
| "loss": 0.8455, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 0.0006500000000000001, | |
| "loss": 0.8385, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 27823, | |
| 19765, | |
| 14889, | |
| 11503 | |
| ], | |
| "eval_loss": 0.43482282757759094, | |
| "eval_precisions": [ | |
| 56.770046929198124, | |
| 43.39473510878872, | |
| 35.37419814682823, | |
| 29.745804349512554 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 342.9777, | |
| "eval_samples_per_second": 10.097, | |
| "eval_score": 40.12519481921629, | |
| "eval_steps_per_second": 10.097, | |
| "eval_sys_len": 49010, | |
| "eval_totals": [ | |
| 49010, | |
| 45547, | |
| 42090, | |
| 38671 | |
| ], | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 0.0006525, | |
| "loss": 0.8321, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 0.0006550000000000001, | |
| "loss": 0.7944, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 0.0006575, | |
| "loss": 0.8239, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 0.00066, | |
| "loss": 0.8123, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 0.0006625, | |
| "loss": 0.8329, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 0.000665, | |
| "loss": 0.8096, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 0.0006675, | |
| "loss": 0.8491, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 0.00067, | |
| "loss": 0.8353, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 0.0006725, | |
| "loss": 0.8454, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 0.000675, | |
| "loss": 0.8376, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 0.0006775, | |
| "loss": 0.8402, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.00068, | |
| "loss": 0.805, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 0.0006825000000000001, | |
| "loss": 0.8314, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 0.0006850000000000001, | |
| "loss": 0.8562, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 0.0006875, | |
| "loss": 0.8417, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 0.00069, | |
| "loss": 0.8611, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 0.0006925, | |
| "loss": 0.8408, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 0.000695, | |
| "loss": 0.765, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 0.0006975, | |
| "loss": 0.8264, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 0.0007, | |
| "loss": 0.8435, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 28549, | |
| 20299, | |
| 15265, | |
| 11758 | |
| ], | |
| "eval_loss": 0.43037816882133484, | |
| "eval_precisions": [ | |
| 61.14323652874154, | |
| 46.95690393023202, | |
| 38.380308249314865, | |
| 32.33951262445679 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 296.3653, | |
| "eval_samples_per_second": 11.685, | |
| "eval_score": 43.44826631126853, | |
| "eval_steps_per_second": 11.685, | |
| "eval_sys_len": 46692, | |
| "eval_totals": [ | |
| 46692, | |
| 43229, | |
| 39773, | |
| 36358 | |
| ], | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 0.0007025, | |
| "loss": 0.8406, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 0.000705, | |
| "loss": 0.8368, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.0007075, | |
| "loss": 0.8128, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 0.00071, | |
| "loss": 0.813, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 0.0007125, | |
| "loss": 0.8117, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 0.000715, | |
| "loss": 0.8333, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.0007175, | |
| "loss": 0.7817, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 0.0007199999999999999, | |
| "loss": 0.8374, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 0.0007225, | |
| "loss": 0.8163, | |
| "step": 2890 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 0.000725, | |
| "loss": 0.8588, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 0.0007275000000000001, | |
| "loss": 0.8709, | |
| "step": 2910 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 0.00073, | |
| "loss": 0.8248, | |
| "step": 2920 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 0.0007325000000000001, | |
| "loss": 0.8591, | |
| "step": 2930 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 0.000735, | |
| "loss": 0.8216, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 0.0007375000000000001, | |
| "loss": 0.8073, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 0.00074, | |
| "loss": 0.806, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 0.0007425000000000001, | |
| "loss": 0.8289, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 0.000745, | |
| "loss": 0.8071, | |
| "step": 2980 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 0.0007475000000000001, | |
| "loss": 0.8295, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 0.00075, | |
| "loss": 0.8217, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 28370, | |
| 20302, | |
| 15357, | |
| 11894 | |
| ], | |
| "eval_loss": 0.426497220993042, | |
| "eval_precisions": [ | |
| 61.62166858532983, | |
| 47.68414130026306, | |
| 39.25713847490989, | |
| 33.31652661064426 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 307.935, | |
| "eval_samples_per_second": 11.246, | |
| "eval_score": 44.27631255210222, | |
| "eval_steps_per_second": 11.246, | |
| "eval_sys_len": 46039, | |
| "eval_totals": [ | |
| 46039, | |
| 42576, | |
| 39119, | |
| 35700 | |
| ], | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 0.0007524999999999999, | |
| "loss": 0.8048, | |
| "step": 3010 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 0.000755, | |
| "loss": 0.8476, | |
| "step": 3020 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 0.0007574999999999999, | |
| "loss": 0.818, | |
| "step": 3030 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.00076, | |
| "loss": 0.8293, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 0.0007624999999999999, | |
| "loss": 0.7651, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 0.0007650000000000001, | |
| "loss": 0.7878, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 0.0007675, | |
| "loss": 0.81, | |
| "step": 3070 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 0.0007700000000000001, | |
| "loss": 0.7829, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 0.0007725, | |
| "loss": 0.819, | |
| "step": 3090 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 0.0007750000000000001, | |
| "loss": 0.7806, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 0.0007775, | |
| "loss": 0.846, | |
| "step": 3110 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 0.0007800000000000001, | |
| "loss": 0.7805, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 0.0007825, | |
| "loss": 0.8161, | |
| "step": 3130 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 0.000785, | |
| "loss": 0.7853, | |
| "step": 3140 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 0.0007875, | |
| "loss": 0.774, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 0.00079, | |
| "loss": 0.8049, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 0.0007925, | |
| "loss": 0.7877, | |
| "step": 3170 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 0.000795, | |
| "loss": 0.7874, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 0.0007975, | |
| "loss": 0.8034, | |
| "step": 3190 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 0.0008, | |
| "loss": 0.7729, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 29261, | |
| 21059, | |
| 15998, | |
| 12377 | |
| ], | |
| "eval_loss": 0.4273025095462799, | |
| "eval_precisions": [ | |
| 65.0837429658133, | |
| 50.7494698284172, | |
| 42.059047769277285, | |
| 35.75824112327738 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 265.0733, | |
| "eval_samples_per_second": 13.064, | |
| "eval_score": 47.21009820410733, | |
| "eval_steps_per_second": 13.064, | |
| "eval_sys_len": 44959, | |
| "eval_totals": [ | |
| 44959, | |
| 41496, | |
| 38037, | |
| 34613 | |
| ], | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 0.0008025, | |
| "loss": 0.8325, | |
| "step": 3210 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 0.000805, | |
| "loss": 0.8142, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 0.0008075000000000001, | |
| "loss": 0.7777, | |
| "step": 3230 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.0008100000000000001, | |
| "loss": 0.7774, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 0.0008125000000000001, | |
| "loss": 0.8134, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 0.000815, | |
| "loss": 0.7921, | |
| "step": 3260 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 0.0008175, | |
| "loss": 0.7643, | |
| "step": 3270 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 0.00082, | |
| "loss": 0.7803, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 0.0008225, | |
| "loss": 0.7804, | |
| "step": 3290 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 0.000825, | |
| "loss": 0.8059, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 0.0008275, | |
| "loss": 0.795, | |
| "step": 3310 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 0.00083, | |
| "loss": 0.7984, | |
| "step": 3320 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 0.0008325, | |
| "loss": 0.8141, | |
| "step": 3330 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 0.000835, | |
| "loss": 0.8191, | |
| "step": 3340 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 0.0008375, | |
| "loss": 0.7957, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 0.00084, | |
| "loss": 0.8047, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 0.0008425, | |
| "loss": 0.7619, | |
| "step": 3370 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 0.0008449999999999999, | |
| "loss": 0.7813, | |
| "step": 3380 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 0.0008475000000000001, | |
| "loss": 0.8167, | |
| "step": 3390 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 0.00085, | |
| "loss": 0.7897, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 29187, | |
| 21037, | |
| 15999, | |
| 12459 | |
| ], | |
| "eval_loss": 0.42397016286849976, | |
| "eval_precisions": [ | |
| 65.54163298302345, | |
| 51.22355060994911, | |
| 42.53582537952304, | |
| 36.43088979209919 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 240.3753, | |
| "eval_samples_per_second": 14.407, | |
| "eval_score": 47.758743784886654, | |
| "eval_steps_per_second": 14.407, | |
| "eval_sys_len": 44532, | |
| "eval_totals": [ | |
| 44532, | |
| 41069, | |
| 37613, | |
| 34199 | |
| ], | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 0.0008525000000000001, | |
| "loss": 0.7772, | |
| "step": 3410 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 0.000855, | |
| "loss": 0.7755, | |
| "step": 3420 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 0.0008575000000000001, | |
| "loss": 0.7838, | |
| "step": 3430 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 0.00086, | |
| "loss": 0.7617, | |
| "step": 3440 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 0.0008625000000000001, | |
| "loss": 0.8364, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 0.000865, | |
| "loss": 0.8103, | |
| "step": 3460 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 0.0008675000000000001, | |
| "loss": 0.7912, | |
| "step": 3470 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 0.00087, | |
| "loss": 0.777, | |
| "step": 3480 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 0.0008725000000000001, | |
| "loss": 0.8106, | |
| "step": 3490 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 0.000875, | |
| "loss": 0.7864, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 0.0008774999999999999, | |
| "loss": 0.7966, | |
| "step": 3510 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 0.00088, | |
| "loss": 0.773, | |
| "step": 3520 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 0.0008824999999999999, | |
| "loss": 0.7928, | |
| "step": 3530 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 0.000885, | |
| "loss": 0.7953, | |
| "step": 3540 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 0.0008874999999999999, | |
| "loss": 0.7662, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 0.0008900000000000001, | |
| "loss": 0.7974, | |
| "step": 3560 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 0.0008925, | |
| "loss": 0.7905, | |
| "step": 3570 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 0.0008950000000000001, | |
| "loss": 0.7816, | |
| "step": 3580 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 0.0008975, | |
| "loss": 0.7742, | |
| "step": 3590 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 0.0009000000000000001, | |
| "loss": 0.8173, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 28583, | |
| 20488, | |
| 15508, | |
| 11976 | |
| ], | |
| "eval_loss": 0.4235726296901703, | |
| "eval_precisions": [ | |
| 63.69044966353224, | |
| 49.469998792707955, | |
| 40.85891186931893, | |
| 34.680875709486855 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 252.8586, | |
| "eval_samples_per_second": 13.695, | |
| "eval_score": 45.967222888757206, | |
| "eval_steps_per_second": 13.695, | |
| "eval_sys_len": 44878, | |
| "eval_totals": [ | |
| 44878, | |
| 41415, | |
| 37955, | |
| 34532 | |
| ], | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 0.0009025, | |
| "loss": 0.821, | |
| "step": 3610 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 0.0009050000000000001, | |
| "loss": 0.7577, | |
| "step": 3620 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 0.0009075, | |
| "loss": 0.765, | |
| "step": 3630 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 0.00091, | |
| "loss": 0.7477, | |
| "step": 3640 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 0.0009125, | |
| "loss": 0.8144, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 0.000915, | |
| "loss": 0.7645, | |
| "step": 3660 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 0.0009175, | |
| "loss": 0.8026, | |
| "step": 3670 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 0.00092, | |
| "loss": 0.8054, | |
| "step": 3680 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 0.0009225, | |
| "loss": 0.7957, | |
| "step": 3690 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 0.000925, | |
| "loss": 0.7849, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 0.0009275, | |
| "loss": 0.7473, | |
| "step": 3710 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 0.00093, | |
| "loss": 0.7891, | |
| "step": 3720 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 0.0009325000000000001, | |
| "loss": 0.7804, | |
| "step": 3730 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 0.0009350000000000001, | |
| "loss": 0.8034, | |
| "step": 3740 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 0.0009375, | |
| "loss": 0.7809, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 0.00094, | |
| "loss": 0.7682, | |
| "step": 3760 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 0.0009425, | |
| "loss": 0.8136, | |
| "step": 3770 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 0.000945, | |
| "loss": 0.7893, | |
| "step": 3780 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 0.0009475, | |
| "loss": 0.7766, | |
| "step": 3790 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 0.00095, | |
| "loss": 0.7888, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 28906, | |
| 20868, | |
| 15901, | |
| 12378 | |
| ], | |
| "eval_loss": 0.42072421312332153, | |
| "eval_precisions": [ | |
| 62.163440860215054, | |
| 48.488509886841555, | |
| 40.17636060437617, | |
| 34.23876963930073 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 262.2659, | |
| "eval_samples_per_second": 13.204, | |
| "eval_score": 45.124838095439365, | |
| "eval_steps_per_second": 13.204, | |
| "eval_sys_len": 46500, | |
| "eval_totals": [ | |
| 46500, | |
| 43037, | |
| 39578, | |
| 36152 | |
| ], | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 0.0009525, | |
| "loss": 0.7763, | |
| "step": 3810 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 0.000955, | |
| "loss": 0.7843, | |
| "step": 3820 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 0.0009575, | |
| "loss": 0.7704, | |
| "step": 3830 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 0.00096, | |
| "loss": 0.7757, | |
| "step": 3840 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 0.0009625, | |
| "loss": 0.7608, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 0.000965, | |
| "loss": 0.7683, | |
| "step": 3860 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 0.0009675, | |
| "loss": 0.8123, | |
| "step": 3870 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 0.0009699999999999999, | |
| "loss": 0.8027, | |
| "step": 3880 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 0.0009725000000000001, | |
| "loss": 0.7979, | |
| "step": 3890 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 0.000975, | |
| "loss": 0.7426, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 0.0009775, | |
| "loss": 0.7936, | |
| "step": 3910 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 0.00098, | |
| "loss": 0.753, | |
| "step": 3920 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 0.0009825, | |
| "loss": 0.7393, | |
| "step": 3930 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 0.000985, | |
| "loss": 0.795, | |
| "step": 3940 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 0.0009875, | |
| "loss": 0.7733, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 0.00099, | |
| "loss": 0.7532, | |
| "step": 3960 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 0.0009925000000000001, | |
| "loss": 0.7695, | |
| "step": 3970 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 0.000995, | |
| "loss": 0.7615, | |
| "step": 3980 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 0.0009975000000000001, | |
| "loss": 0.8031, | |
| "step": 3990 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 0.001, | |
| "loss": 0.8118, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 28407, | |
| 20295, | |
| 15340, | |
| 11840 | |
| ], | |
| "eval_loss": 0.41856399178504944, | |
| "eval_precisions": [ | |
| 61.8942827261635, | |
| 47.82834114957698, | |
| 39.35452423099618, | |
| 33.293029271996176 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 277.9389, | |
| "eval_samples_per_second": 12.46, | |
| "eval_score": 44.378330272467224, | |
| "eval_steps_per_second": 12.46, | |
| "eval_sys_len": 45896, | |
| "eval_totals": [ | |
| 45896, | |
| 42433, | |
| 38979, | |
| 35563 | |
| ], | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 0.001, | |
| "loss": 0.7708, | |
| "step": 4010 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "learning_rate": 0.001, | |
| "loss": 0.7693, | |
| "step": 4020 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 0.001, | |
| "loss": 0.8062, | |
| "step": 4030 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 0.001, | |
| "loss": 0.8227, | |
| "step": 4040 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.001, | |
| "loss": 0.8294, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 0.001, | |
| "loss": 0.795, | |
| "step": 4060 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 0.001, | |
| "loss": 0.7451, | |
| "step": 4070 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 0.001, | |
| "loss": 0.715, | |
| "step": 4080 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 0.001, | |
| "loss": 0.7761, | |
| "step": 4090 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 0.001, | |
| "loss": 0.7503, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 0.001, | |
| "loss": 0.7685, | |
| "step": 4110 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 0.001, | |
| "loss": 0.71, | |
| "step": 4120 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 0.001, | |
| "loss": 0.7266, | |
| "step": 4130 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 0.001, | |
| "loss": 0.7276, | |
| "step": 4140 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 0.001, | |
| "loss": 0.7567, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 0.001, | |
| "loss": 0.7302, | |
| "step": 4160 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 0.001, | |
| "loss": 0.7423, | |
| "step": 4170 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 0.001, | |
| "loss": 0.71, | |
| "step": 4180 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 0.001, | |
| "loss": 0.7361, | |
| "step": 4190 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 0.001, | |
| "loss": 0.7753, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 28723, | |
| 20563, | |
| 15543, | |
| 11998 | |
| ], | |
| "eval_loss": 0.4164888560771942, | |
| "eval_precisions": [ | |
| 60.51023847644729, | |
| 46.72878082036132, | |
| 38.30683918669131, | |
| 32.26656626506024 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 274.3953, | |
| "eval_samples_per_second": 12.62, | |
| "eval_score": 43.23750209983783, | |
| "eval_steps_per_second": 12.62, | |
| "eval_sys_len": 47468, | |
| "eval_totals": [ | |
| 47468, | |
| 44005, | |
| 40575, | |
| 37184 | |
| ], | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 0.001, | |
| "loss": 0.7742, | |
| "step": 4210 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 0.001, | |
| "loss": 0.7177, | |
| "step": 4220 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "learning_rate": 0.001, | |
| "loss": 0.7441, | |
| "step": 4230 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 0.001, | |
| "loss": 0.7241, | |
| "step": 4240 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "learning_rate": 0.001, | |
| "loss": 0.7551, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 0.001, | |
| "loss": 0.7866, | |
| "step": 4260 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 0.001, | |
| "loss": 0.743, | |
| "step": 4270 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "learning_rate": 0.001, | |
| "loss": 0.7721, | |
| "step": 4280 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 0.001, | |
| "loss": 0.7599, | |
| "step": 4290 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 0.001, | |
| "loss": 0.7711, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 0.001, | |
| "loss": 0.7689, | |
| "step": 4310 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 0.001, | |
| "loss": 0.7777, | |
| "step": 4320 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 0.001, | |
| "loss": 0.7573, | |
| "step": 4330 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "learning_rate": 0.001, | |
| "loss": 0.7699, | |
| "step": 4340 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 0.001, | |
| "loss": 0.7455, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 0.001, | |
| "loss": 0.7559, | |
| "step": 4360 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 0.001, | |
| "loss": 0.8071, | |
| "step": 4370 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 0.001, | |
| "loss": 0.75, | |
| "step": 4380 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 0.001, | |
| "loss": 0.755, | |
| "step": 4390 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 0.001, | |
| "loss": 0.7512, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 28272, | |
| 20274, | |
| 15450, | |
| 12013 | |
| ], | |
| "eval_loss": 0.41703328490257263, | |
| "eval_precisions": [ | |
| 62.01904093361997, | |
| 48.1304750373905, | |
| 39.94828700711054, | |
| 34.06493690628101 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 264.2496, | |
| "eval_samples_per_second": 13.105, | |
| "eval_score": 44.89394147732702, | |
| "eval_steps_per_second": 13.105, | |
| "eval_sys_len": 45586, | |
| "eval_totals": [ | |
| 45586, | |
| 42123, | |
| 38675, | |
| 35265 | |
| ], | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 0.001, | |
| "loss": 0.7414, | |
| "step": 4410 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 0.001, | |
| "loss": 0.7468, | |
| "step": 4420 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 0.001, | |
| "loss": 0.7113, | |
| "step": 4430 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 0.001, | |
| "loss": 0.7615, | |
| "step": 4440 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 0.001, | |
| "loss": 0.7114, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 0.001, | |
| "loss": 0.7609, | |
| "step": 4460 | |
| }, | |
| { | |
| "epoch": 4.42, | |
| "learning_rate": 0.001, | |
| "loss": 0.7379, | |
| "step": 4470 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "learning_rate": 0.001, | |
| "loss": 0.7553, | |
| "step": 4480 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 0.001, | |
| "loss": 0.7502, | |
| "step": 4490 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "learning_rate": 0.001, | |
| "loss": 0.7467, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 0.001, | |
| "loss": 0.7537, | |
| "step": 4510 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 0.001, | |
| "loss": 0.7372, | |
| "step": 4520 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 0.001, | |
| "loss": 0.7542, | |
| "step": 4530 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 0.001, | |
| "loss": 0.745, | |
| "step": 4540 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 0.001, | |
| "loss": 0.7757, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "learning_rate": 0.001, | |
| "loss": 0.7365, | |
| "step": 4560 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 0.001, | |
| "loss": 0.7772, | |
| "step": 4570 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 0.001, | |
| "loss": 0.7387, | |
| "step": 4580 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "learning_rate": 0.001, | |
| "loss": 0.7385, | |
| "step": 4590 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "learning_rate": 0.001, | |
| "loss": 0.7509, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 27954, | |
| 19976, | |
| 15088, | |
| 11660 | |
| ], | |
| "eval_loss": 0.41963234543800354, | |
| "eval_precisions": [ | |
| 62.255578816088374, | |
| 48.20579647192258, | |
| 39.72303398888977, | |
| 33.73549749732373 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 281.5455, | |
| "eval_samples_per_second": 12.3, | |
| "eval_score": 44.781822813954804, | |
| "eval_steps_per_second": 12.3, | |
| "eval_sys_len": 44902, | |
| "eval_totals": [ | |
| 44902, | |
| 41439, | |
| 37983, | |
| 34563 | |
| ], | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 0.001, | |
| "loss": 0.804, | |
| "step": 4610 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "learning_rate": 0.001, | |
| "loss": 0.6814, | |
| "step": 4620 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 0.001, | |
| "loss": 0.7433, | |
| "step": 4630 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 0.001, | |
| "loss": 0.7306, | |
| "step": 4640 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 0.001, | |
| "loss": 0.735, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "learning_rate": 0.001, | |
| "loss": 0.7559, | |
| "step": 4660 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "learning_rate": 0.001, | |
| "loss": 0.7659, | |
| "step": 4670 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 0.001, | |
| "loss": 0.7456, | |
| "step": 4680 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 0.001, | |
| "loss": 0.7533, | |
| "step": 4690 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 0.001, | |
| "loss": 0.724, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 0.001, | |
| "loss": 0.7311, | |
| "step": 4710 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "learning_rate": 0.001, | |
| "loss": 0.7224, | |
| "step": 4720 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 0.001, | |
| "loss": 0.7289, | |
| "step": 4730 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 0.001, | |
| "loss": 0.7441, | |
| "step": 4740 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 0.001, | |
| "loss": 0.7435, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 0.001, | |
| "loss": 0.7392, | |
| "step": 4760 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "learning_rate": 0.001, | |
| "loss": 0.7604, | |
| "step": 4770 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 0.001, | |
| "loss": 0.7511, | |
| "step": 4780 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 0.001, | |
| "loss": 0.7351, | |
| "step": 4790 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 0.001, | |
| "loss": 0.728, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 29267, | |
| 21187, | |
| 16147, | |
| 12524 | |
| ], | |
| "eval_loss": 0.4112496078014374, | |
| "eval_precisions": [ | |
| 67.60059130595464, | |
| 53.19223720217921, | |
| 44.385497127463644, | |
| 37.99296201917243 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 244.1633, | |
| "eval_samples_per_second": 14.183, | |
| "eval_score": 49.623312157351805, | |
| "eval_steps_per_second": 14.183, | |
| "eval_sys_len": 43294, | |
| "eval_totals": [ | |
| 43294, | |
| 39831, | |
| 36379, | |
| 32964 | |
| ], | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "learning_rate": 0.001, | |
| "loss": 0.7531, | |
| "step": 4810 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 0.001, | |
| "loss": 0.7829, | |
| "step": 4820 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 0.001, | |
| "loss": 0.7372, | |
| "step": 4830 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 0.001, | |
| "loss": 0.7225, | |
| "step": 4840 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 0.001, | |
| "loss": 0.6885, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 0.001, | |
| "loss": 0.7275, | |
| "step": 4860 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "learning_rate": 0.001, | |
| "loss": 0.7833, | |
| "step": 4870 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 0.001, | |
| "loss": 0.7644, | |
| "step": 4880 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 0.001, | |
| "loss": 0.7405, | |
| "step": 4890 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 0.001, | |
| "loss": 0.7562, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "learning_rate": 0.001, | |
| "loss": 0.7518, | |
| "step": 4910 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "learning_rate": 0.001, | |
| "loss": 0.7408, | |
| "step": 4920 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 0.001, | |
| "loss": 0.7343, | |
| "step": 4930 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 0.001, | |
| "loss": 0.7392, | |
| "step": 4940 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 0.001, | |
| "loss": 0.7592, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 0.001, | |
| "loss": 0.7739, | |
| "step": 4960 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "learning_rate": 0.001, | |
| "loss": 0.7301, | |
| "step": 4970 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "learning_rate": 0.001, | |
| "loss": 0.8052, | |
| "step": 4980 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 0.001, | |
| "loss": 0.7298, | |
| "step": 4990 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "learning_rate": 0.001, | |
| "loss": 0.7329, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 28901, | |
| 20885, | |
| 15902, | |
| 12379 | |
| ], | |
| "eval_loss": 0.4055224657058716, | |
| "eval_precisions": [ | |
| 62.51568245727882, | |
| 48.80699212451216, | |
| 40.40039633139402, | |
| 34.41287668186367 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 294.1071, | |
| "eval_samples_per_second": 11.775, | |
| "eval_score": 45.38312591155178, | |
| "eval_steps_per_second": 11.775, | |
| "eval_sys_len": 46230, | |
| "eval_totals": [ | |
| 46230, | |
| 42791, | |
| 39361, | |
| 35972 | |
| ], | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 0.001, | |
| "loss": 0.7855, | |
| "step": 5010 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 0.001, | |
| "loss": 0.7417, | |
| "step": 5020 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 0.001, | |
| "loss": 0.7425, | |
| "step": 5030 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "learning_rate": 0.001, | |
| "loss": 0.7576, | |
| "step": 5040 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "learning_rate": 0.001, | |
| "loss": 0.778, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.001, | |
| "loss": 0.7713, | |
| "step": 5060 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 0.001, | |
| "loss": 0.6738, | |
| "step": 5070 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 0.001, | |
| "loss": 0.7084, | |
| "step": 5080 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 0.001, | |
| "loss": 0.6656, | |
| "step": 5090 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 0.001, | |
| "loss": 0.715, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 0.001, | |
| "loss": 0.7264, | |
| "step": 5110 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 0.001, | |
| "loss": 0.6993, | |
| "step": 5120 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 0.001, | |
| "loss": 0.7004, | |
| "step": 5130 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 0.001, | |
| "loss": 0.6492, | |
| "step": 5140 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "learning_rate": 0.001, | |
| "loss": 0.7203, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "learning_rate": 0.001, | |
| "loss": 0.7231, | |
| "step": 5160 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 0.001, | |
| "loss": 0.708, | |
| "step": 5170 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 0.001, | |
| "loss": 0.7115, | |
| "step": 5180 | |
| }, | |
| { | |
| "epoch": 5.13, | |
| "learning_rate": 0.001, | |
| "loss": 0.6912, | |
| "step": 5190 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "learning_rate": 0.001, | |
| "loss": 0.7268, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 27727, | |
| 19860, | |
| 15039, | |
| 11611 | |
| ], | |
| "eval_loss": 0.4133273661136627, | |
| "eval_precisions": [ | |
| 65.2108469154966, | |
| 50.85006145022532, | |
| 42.243195415859105, | |
| 36.07917469392828 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 250.1733, | |
| "eval_samples_per_second": 13.842, | |
| "eval_score": 47.41395354619319, | |
| "eval_steps_per_second": 13.842, | |
| "eval_sys_len": 42519, | |
| "eval_totals": [ | |
| 42519, | |
| 39056, | |
| 35601, | |
| 32182 | |
| ], | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 5.15, | |
| "learning_rate": 0.001, | |
| "loss": 0.6696, | |
| "step": 5210 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 0.001, | |
| "loss": 0.7228, | |
| "step": 5220 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "learning_rate": 0.001, | |
| "loss": 0.6866, | |
| "step": 5230 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "learning_rate": 0.001, | |
| "loss": 0.7071, | |
| "step": 5240 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "learning_rate": 0.001, | |
| "loss": 0.6969, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "learning_rate": 0.001, | |
| "loss": 0.7247, | |
| "step": 5260 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "learning_rate": 0.001, | |
| "loss": 0.7311, | |
| "step": 5270 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 0.001, | |
| "loss": 0.7426, | |
| "step": 5280 | |
| }, | |
| { | |
| "epoch": 5.23, | |
| "learning_rate": 0.001, | |
| "loss": 0.7112, | |
| "step": 5290 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 0.001, | |
| "loss": 0.7144, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 5.25, | |
| "learning_rate": 0.001, | |
| "loss": 0.7079, | |
| "step": 5310 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "learning_rate": 0.001, | |
| "loss": 0.6676, | |
| "step": 5320 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "learning_rate": 0.001, | |
| "loss": 0.7082, | |
| "step": 5330 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "learning_rate": 0.001, | |
| "loss": 0.6699, | |
| "step": 5340 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "learning_rate": 0.001, | |
| "loss": 0.7434, | |
| "step": 5350 | |
| }, | |
| { | |
| "epoch": 5.3, | |
| "learning_rate": 0.001, | |
| "loss": 0.7068, | |
| "step": 5360 | |
| }, | |
| { | |
| "epoch": 5.31, | |
| "learning_rate": 0.001, | |
| "loss": 0.7139, | |
| "step": 5370 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "learning_rate": 0.001, | |
| "loss": 0.7299, | |
| "step": 5380 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 0.001, | |
| "loss": 0.7388, | |
| "step": 5390 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "learning_rate": 0.001, | |
| "loss": 0.714, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "eval_bp": 0.9927479429825526, | |
| "eval_counts": [ | |
| 28359, | |
| 20291, | |
| 15288, | |
| 11708 | |
| ], | |
| "eval_loss": 0.41272008419036865, | |
| "eval_precisions": [ | |
| 69.03359298928919, | |
| 53.9410372969668, | |
| 44.75278826732239, | |
| 38.0809887786632 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 222.8345, | |
| "eval_samples_per_second": 15.541, | |
| "eval_score": 49.827146588601636, | |
| "eval_steps_per_second": 15.541, | |
| "eval_sys_len": 41080, | |
| "eval_totals": [ | |
| 41080, | |
| 37617, | |
| 34161, | |
| 30745 | |
| ], | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "learning_rate": 0.001, | |
| "loss": 0.738, | |
| "step": 5410 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "learning_rate": 0.001, | |
| "loss": 0.6744, | |
| "step": 5420 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "learning_rate": 0.001, | |
| "loss": 0.7497, | |
| "step": 5430 | |
| }, | |
| { | |
| "epoch": 5.37, | |
| "learning_rate": 0.001, | |
| "loss": 0.7485, | |
| "step": 5440 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 0.001, | |
| "loss": 0.7213, | |
| "step": 5450 | |
| }, | |
| { | |
| "epoch": 5.39, | |
| "learning_rate": 0.001, | |
| "loss": 0.6867, | |
| "step": 5460 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "learning_rate": 0.001, | |
| "loss": 0.7245, | |
| "step": 5470 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "learning_rate": 0.001, | |
| "loss": 0.7187, | |
| "step": 5480 | |
| }, | |
| { | |
| "epoch": 5.42, | |
| "learning_rate": 0.001, | |
| "loss": 0.7073, | |
| "step": 5490 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 0.001, | |
| "loss": 0.7175, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "learning_rate": 0.001, | |
| "loss": 0.7046, | |
| "step": 5510 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 0.001, | |
| "loss": 0.7133, | |
| "step": 5520 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "learning_rate": 0.001, | |
| "loss": 0.7653, | |
| "step": 5530 | |
| }, | |
| { | |
| "epoch": 5.47, | |
| "learning_rate": 0.001, | |
| "loss": 0.6843, | |
| "step": 5540 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "learning_rate": 0.001, | |
| "loss": 0.7342, | |
| "step": 5550 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "learning_rate": 0.001, | |
| "loss": 0.6872, | |
| "step": 5560 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "learning_rate": 0.001, | |
| "loss": 0.6606, | |
| "step": 5570 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "learning_rate": 0.001, | |
| "loss": 0.7187, | |
| "step": 5580 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "learning_rate": 0.001, | |
| "loss": 0.7269, | |
| "step": 5590 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 0.001, | |
| "loss": 0.7254, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 28433, | |
| 20456, | |
| 15554, | |
| 12067 | |
| ], | |
| "eval_loss": 0.4069550931453705, | |
| "eval_precisions": [ | |
| 61.42230671188784, | |
| 47.76314560567853, | |
| 39.506235554088036, | |
| 33.565130316263804 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 273.5282, | |
| "eval_samples_per_second": 12.66, | |
| "eval_score": 44.4112766348839, | |
| "eval_steps_per_second": 12.66, | |
| "eval_sys_len": 46291, | |
| "eval_totals": [ | |
| 46291, | |
| 42828, | |
| 39371, | |
| 35951 | |
| ], | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 0.001, | |
| "loss": 0.6905, | |
| "step": 5610 | |
| }, | |
| { | |
| "epoch": 5.55, | |
| "learning_rate": 0.001, | |
| "loss": 0.7082, | |
| "step": 5620 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 0.001, | |
| "loss": 0.7296, | |
| "step": 5630 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "learning_rate": 0.001, | |
| "loss": 0.6929, | |
| "step": 5640 | |
| }, | |
| { | |
| "epoch": 5.58, | |
| "learning_rate": 0.001, | |
| "loss": 0.665, | |
| "step": 5650 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "learning_rate": 0.001, | |
| "loss": 0.7081, | |
| "step": 5660 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 0.001, | |
| "loss": 0.7002, | |
| "step": 5670 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 0.001, | |
| "loss": 0.7112, | |
| "step": 5680 | |
| }, | |
| { | |
| "epoch": 5.62, | |
| "learning_rate": 0.001, | |
| "loss": 0.7427, | |
| "step": 5690 | |
| }, | |
| { | |
| "epoch": 5.63, | |
| "learning_rate": 0.001, | |
| "loss": 0.7099, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 5.64, | |
| "learning_rate": 0.001, | |
| "loss": 0.6758, | |
| "step": 5710 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 0.001, | |
| "loss": 0.7245, | |
| "step": 5720 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 0.001, | |
| "loss": 0.7028, | |
| "step": 5730 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "learning_rate": 0.001, | |
| "loss": 0.7325, | |
| "step": 5740 | |
| }, | |
| { | |
| "epoch": 5.68, | |
| "learning_rate": 0.001, | |
| "loss": 0.7156, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 0.001, | |
| "loss": 0.6748, | |
| "step": 5760 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 0.001, | |
| "loss": 0.7264, | |
| "step": 5770 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "learning_rate": 0.001, | |
| "loss": 0.7532, | |
| "step": 5780 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "learning_rate": 0.001, | |
| "loss": 0.7117, | |
| "step": 5790 | |
| }, | |
| { | |
| "epoch": 5.73, | |
| "learning_rate": 0.001, | |
| "loss": 0.7638, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 5.73, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 28770, | |
| 20622, | |
| 15628, | |
| 12103 | |
| ], | |
| "eval_loss": 0.4058038294315338, | |
| "eval_precisions": [ | |
| 59.88759367194005, | |
| 46.26152500168248, | |
| 38.005836575875485, | |
| 32.102596748096865 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 291.6813, | |
| "eval_samples_per_second": 11.873, | |
| "eval_score": 42.878226797603965, | |
| "eval_steps_per_second": 11.873, | |
| "eval_sys_len": 48040, | |
| "eval_totals": [ | |
| 48040, | |
| 44577, | |
| 41120, | |
| 37701 | |
| ], | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "learning_rate": 0.001, | |
| "loss": 0.7201, | |
| "step": 5810 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "learning_rate": 0.001, | |
| "loss": 0.716, | |
| "step": 5820 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 0.001, | |
| "loss": 0.696, | |
| "step": 5830 | |
| }, | |
| { | |
| "epoch": 5.77, | |
| "learning_rate": 0.001, | |
| "loss": 0.7626, | |
| "step": 5840 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "learning_rate": 0.001, | |
| "loss": 0.7009, | |
| "step": 5850 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 0.001, | |
| "loss": 0.7226, | |
| "step": 5860 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "learning_rate": 0.001, | |
| "loss": 0.6715, | |
| "step": 5870 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 0.001, | |
| "loss": 0.7295, | |
| "step": 5880 | |
| }, | |
| { | |
| "epoch": 5.82, | |
| "learning_rate": 0.001, | |
| "loss": 0.6715, | |
| "step": 5890 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "learning_rate": 0.001, | |
| "loss": 0.7223, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "learning_rate": 0.001, | |
| "loss": 0.7099, | |
| "step": 5910 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 0.001, | |
| "loss": 0.6879, | |
| "step": 5920 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 0.001, | |
| "loss": 0.7327, | |
| "step": 5930 | |
| }, | |
| { | |
| "epoch": 5.87, | |
| "learning_rate": 0.001, | |
| "loss": 0.7383, | |
| "step": 5940 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "learning_rate": 0.001, | |
| "loss": 0.6852, | |
| "step": 5950 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "learning_rate": 0.001, | |
| "loss": 0.7159, | |
| "step": 5960 | |
| }, | |
| { | |
| "epoch": 5.9, | |
| "learning_rate": 0.001, | |
| "loss": 0.7374, | |
| "step": 5970 | |
| }, | |
| { | |
| "epoch": 5.91, | |
| "learning_rate": 0.001, | |
| "loss": 0.6971, | |
| "step": 5980 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 0.001, | |
| "loss": 0.7176, | |
| "step": 5990 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 0.001, | |
| "loss": 0.6952, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 29004, | |
| 21041, | |
| 16027, | |
| 12451 | |
| ], | |
| "eval_loss": 0.4028075039386749, | |
| "eval_precisions": [ | |
| 67.53912071535022, | |
| 53.261612454119735, | |
| 44.428119975605696, | |
| 38.096257993452255 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 245.1844, | |
| "eval_samples_per_second": 14.124, | |
| "eval_score": 49.67381229901984, | |
| "eval_steps_per_second": 14.124, | |
| "eval_sys_len": 42944, | |
| "eval_totals": [ | |
| 42944, | |
| 39505, | |
| 36074, | |
| 32683 | |
| ], | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "learning_rate": 0.001, | |
| "loss": 0.688, | |
| "step": 6010 | |
| }, | |
| { | |
| "epoch": 5.95, | |
| "learning_rate": 0.001, | |
| "loss": 0.691, | |
| "step": 6020 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "learning_rate": 0.001, | |
| "loss": 0.7381, | |
| "step": 6030 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "learning_rate": 0.001, | |
| "loss": 0.7044, | |
| "step": 6040 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "learning_rate": 0.001, | |
| "loss": 0.6931, | |
| "step": 6050 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "learning_rate": 0.001, | |
| "loss": 0.6928, | |
| "step": 6060 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 0.001, | |
| "loss": 0.6837, | |
| "step": 6070 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 0.001, | |
| "loss": 0.6665, | |
| "step": 6080 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 0.001, | |
| "loss": 0.6584, | |
| "step": 6090 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 0.001, | |
| "loss": 0.6864, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 0.001, | |
| "loss": 0.6774, | |
| "step": 6110 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 0.001, | |
| "loss": 0.6633, | |
| "step": 6120 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 0.001, | |
| "loss": 0.7094, | |
| "step": 6130 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 0.001, | |
| "loss": 0.6598, | |
| "step": 6140 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 0.001, | |
| "loss": 0.6601, | |
| "step": 6150 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "learning_rate": 0.001, | |
| "loss": 0.6883, | |
| "step": 6160 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "learning_rate": 0.001, | |
| "loss": 0.6636, | |
| "step": 6170 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 0.001, | |
| "loss": 0.6564, | |
| "step": 6180 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 0.001, | |
| "loss": 0.6996, | |
| "step": 6190 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "learning_rate": 0.001, | |
| "loss": 0.6559, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 28941, | |
| 20962, | |
| 15956, | |
| 12381 | |
| ], | |
| "eval_loss": 0.4090527296066284, | |
| "eval_precisions": [ | |
| 65.99097044874134, | |
| 51.8951303443666, | |
| 43.19553859064945, | |
| 36.92954721708525 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 250.6863, | |
| "eval_samples_per_second": 13.814, | |
| "eval_score": 48.34549370674115, | |
| "eval_steps_per_second": 13.814, | |
| "eval_sys_len": 43856, | |
| "eval_totals": [ | |
| 43856, | |
| 40393, | |
| 36939, | |
| 33526 | |
| ], | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 6.14, | |
| "learning_rate": 0.001, | |
| "loss": 0.6669, | |
| "step": 6210 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 0.001, | |
| "loss": 0.6678, | |
| "step": 6220 | |
| }, | |
| { | |
| "epoch": 6.16, | |
| "learning_rate": 0.001, | |
| "loss": 0.6749, | |
| "step": 6230 | |
| }, | |
| { | |
| "epoch": 6.16, | |
| "learning_rate": 0.001, | |
| "loss": 0.6728, | |
| "step": 6240 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "learning_rate": 0.001, | |
| "loss": 0.7081, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 6.18, | |
| "learning_rate": 0.001, | |
| "loss": 0.6408, | |
| "step": 6260 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "learning_rate": 0.001, | |
| "loss": 0.6131, | |
| "step": 6270 | |
| }, | |
| { | |
| "epoch": 6.2, | |
| "learning_rate": 0.001, | |
| "loss": 0.6698, | |
| "step": 6280 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "learning_rate": 0.001, | |
| "loss": 0.6987, | |
| "step": 6290 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "learning_rate": 0.001, | |
| "loss": 0.6764, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "learning_rate": 0.001, | |
| "loss": 0.6549, | |
| "step": 6310 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 0.001, | |
| "loss": 0.6853, | |
| "step": 6320 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 0.001, | |
| "loss": 0.6904, | |
| "step": 6330 | |
| }, | |
| { | |
| "epoch": 6.26, | |
| "learning_rate": 0.001, | |
| "loss": 0.658, | |
| "step": 6340 | |
| }, | |
| { | |
| "epoch": 6.27, | |
| "learning_rate": 0.001, | |
| "loss": 0.6709, | |
| "step": 6350 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "learning_rate": 0.001, | |
| "loss": 0.6653, | |
| "step": 6360 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 0.001, | |
| "loss": 0.6872, | |
| "step": 6370 | |
| }, | |
| { | |
| "epoch": 6.3, | |
| "learning_rate": 0.001, | |
| "loss": 0.6851, | |
| "step": 6380 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "learning_rate": 0.001, | |
| "loss": 0.6811, | |
| "step": 6390 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "learning_rate": 0.001, | |
| "loss": 0.6496, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 28317, | |
| 20318, | |
| 15457, | |
| 12006 | |
| ], | |
| "eval_loss": 0.40866824984550476, | |
| "eval_precisions": [ | |
| 59.82254145980775, | |
| 46.3119985412108, | |
| 38.24854003761259, | |
| 32.45829841304172 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 287.9132, | |
| "eval_samples_per_second": 12.028, | |
| "eval_score": 43.06502943489827, | |
| "eval_steps_per_second": 12.028, | |
| "eval_sys_len": 47335, | |
| "eval_totals": [ | |
| 47335, | |
| 43872, | |
| 40412, | |
| 36989 | |
| ], | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "learning_rate": 0.001, | |
| "loss": 0.6778, | |
| "step": 6410 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "learning_rate": 0.001, | |
| "loss": 0.6646, | |
| "step": 6420 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "learning_rate": 0.001, | |
| "loss": 0.6947, | |
| "step": 6430 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "learning_rate": 0.001, | |
| "loss": 0.6535, | |
| "step": 6440 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "learning_rate": 0.001, | |
| "loss": 0.6883, | |
| "step": 6450 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "learning_rate": 0.001, | |
| "loss": 0.6447, | |
| "step": 6460 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "learning_rate": 0.001, | |
| "loss": 0.7202, | |
| "step": 6470 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.001, | |
| "loss": 0.6652, | |
| "step": 6480 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 0.001, | |
| "loss": 0.626, | |
| "step": 6490 | |
| }, | |
| { | |
| "epoch": 6.42, | |
| "learning_rate": 0.001, | |
| "loss": 0.6858, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "learning_rate": 0.001, | |
| "loss": 0.7343, | |
| "step": 6510 | |
| }, | |
| { | |
| "epoch": 6.44, | |
| "learning_rate": 0.001, | |
| "loss": 0.6565, | |
| "step": 6520 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 0.001, | |
| "loss": 0.658, | |
| "step": 6530 | |
| }, | |
| { | |
| "epoch": 6.46, | |
| "learning_rate": 0.001, | |
| "loss": 0.6337, | |
| "step": 6540 | |
| }, | |
| { | |
| "epoch": 6.47, | |
| "learning_rate": 0.001, | |
| "loss": 0.6634, | |
| "step": 6550 | |
| }, | |
| { | |
| "epoch": 6.48, | |
| "learning_rate": 0.001, | |
| "loss": 0.6761, | |
| "step": 6560 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "learning_rate": 0.001, | |
| "loss": 0.6695, | |
| "step": 6570 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "learning_rate": 0.001, | |
| "loss": 0.6582, | |
| "step": 6580 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "learning_rate": 0.001, | |
| "loss": 0.7001, | |
| "step": 6590 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "learning_rate": 0.001, | |
| "loss": 0.6905, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "eval_bp": 0.999927496828018, | |
| "eval_counts": [ | |
| 28670, | |
| 20768, | |
| 15857, | |
| 12342 | |
| ], | |
| "eval_loss": 0.40128186345100403, | |
| "eval_precisions": [ | |
| 69.29137664346482, | |
| 54.778044470234484, | |
| 46.01700571693897, | |
| 39.75519407311967 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 217.2884, | |
| "eval_samples_per_second": 15.937, | |
| "eval_score": 51.32961397242035, | |
| "eval_steps_per_second": 15.937, | |
| "eval_sys_len": 41376, | |
| "eval_totals": [ | |
| 41376, | |
| 37913, | |
| 34459, | |
| 31045 | |
| ], | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "learning_rate": 0.001, | |
| "loss": 0.6984, | |
| "step": 6610 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "learning_rate": 0.001, | |
| "loss": 0.6716, | |
| "step": 6620 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "learning_rate": 0.001, | |
| "loss": 0.6761, | |
| "step": 6630 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 0.001, | |
| "loss": 0.6962, | |
| "step": 6640 | |
| }, | |
| { | |
| "epoch": 6.57, | |
| "learning_rate": 0.001, | |
| "loss": 0.6691, | |
| "step": 6650 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "learning_rate": 0.001, | |
| "loss": 0.6632, | |
| "step": 6660 | |
| }, | |
| { | |
| "epoch": 6.59, | |
| "learning_rate": 0.001, | |
| "loss": 0.6636, | |
| "step": 6670 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "learning_rate": 0.001, | |
| "loss": 0.6538, | |
| "step": 6680 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "learning_rate": 0.001, | |
| "loss": 0.6698, | |
| "step": 6690 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 0.001, | |
| "loss": 0.6802, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "learning_rate": 0.001, | |
| "loss": 0.6487, | |
| "step": 6710 | |
| }, | |
| { | |
| "epoch": 6.64, | |
| "learning_rate": 0.001, | |
| "loss": 0.672, | |
| "step": 6720 | |
| }, | |
| { | |
| "epoch": 6.65, | |
| "learning_rate": 0.001, | |
| "loss": 0.6524, | |
| "step": 6730 | |
| }, | |
| { | |
| "epoch": 6.66, | |
| "learning_rate": 0.001, | |
| "loss": 0.6616, | |
| "step": 6740 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 0.001, | |
| "loss": 0.6712, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "learning_rate": 0.001, | |
| "loss": 0.6811, | |
| "step": 6760 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "learning_rate": 0.001, | |
| "loss": 0.6924, | |
| "step": 6770 | |
| }, | |
| { | |
| "epoch": 6.7, | |
| "learning_rate": 0.001, | |
| "loss": 0.703, | |
| "step": 6780 | |
| }, | |
| { | |
| "epoch": 6.71, | |
| "learning_rate": 0.001, | |
| "loss": 0.6721, | |
| "step": 6790 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "learning_rate": 0.001, | |
| "loss": 0.7053, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 29107, | |
| 21108, | |
| 16130, | |
| 12557 | |
| ], | |
| "eval_loss": 0.4013018012046814, | |
| "eval_precisions": [ | |
| 66.57441504082706, | |
| 52.43181479457499, | |
| 43.82913972066736, | |
| 37.616080522437244 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 236.2133, | |
| "eval_samples_per_second": 14.66, | |
| "eval_score": 48.97897409756685, | |
| "eval_steps_per_second": 14.66, | |
| "eval_sys_len": 43721, | |
| "eval_totals": [ | |
| 43721, | |
| 40258, | |
| 36802, | |
| 33382 | |
| ], | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 6.73, | |
| "learning_rate": 0.001, | |
| "loss": 0.6696, | |
| "step": 6810 | |
| }, | |
| { | |
| "epoch": 6.74, | |
| "learning_rate": 0.001, | |
| "loss": 0.6624, | |
| "step": 6820 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "learning_rate": 0.001, | |
| "loss": 0.6915, | |
| "step": 6830 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "learning_rate": 0.001, | |
| "loss": 0.6981, | |
| "step": 6840 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "learning_rate": 0.001, | |
| "loss": 0.6435, | |
| "step": 6850 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 0.001, | |
| "loss": 0.6908, | |
| "step": 6860 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "learning_rate": 0.001, | |
| "loss": 0.6837, | |
| "step": 6870 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "learning_rate": 0.001, | |
| "loss": 0.6885, | |
| "step": 6880 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "learning_rate": 0.001, | |
| "loss": 0.6836, | |
| "step": 6890 | |
| }, | |
| { | |
| "epoch": 6.82, | |
| "learning_rate": 0.001, | |
| "loss": 0.6974, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 0.001, | |
| "loss": 0.6849, | |
| "step": 6910 | |
| }, | |
| { | |
| "epoch": 6.84, | |
| "learning_rate": 0.001, | |
| "loss": 0.7251, | |
| "step": 6920 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "learning_rate": 0.001, | |
| "loss": 0.6843, | |
| "step": 6930 | |
| }, | |
| { | |
| "epoch": 6.86, | |
| "learning_rate": 0.001, | |
| "loss": 0.6929, | |
| "step": 6940 | |
| }, | |
| { | |
| "epoch": 6.87, | |
| "learning_rate": 0.001, | |
| "loss": 0.6778, | |
| "step": 6950 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "learning_rate": 0.001, | |
| "loss": 0.6691, | |
| "step": 6960 | |
| }, | |
| { | |
| "epoch": 6.89, | |
| "learning_rate": 0.001, | |
| "loss": 0.6737, | |
| "step": 6970 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "learning_rate": 0.001, | |
| "loss": 0.6999, | |
| "step": 6980 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "learning_rate": 0.001, | |
| "loss": 0.6387, | |
| "step": 6990 | |
| }, | |
| { | |
| "epoch": 6.92, | |
| "learning_rate": 0.001, | |
| "loss": 0.7109, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 6.92, | |
| "eval_bp": 0.990408388303313, | |
| "eval_counts": [ | |
| 28729, | |
| 20906, | |
| 16036, | |
| 12538 | |
| ], | |
| "eval_loss": 0.3959246873855591, | |
| "eval_precisions": [ | |
| 70.09808705836424, | |
| 55.71813117987261, | |
| 47.07609206200094, | |
| 40.91769466744991 | |
| ], | |
| "eval_ref_len": 41379, | |
| "eval_runtime": 205.3727, | |
| "eval_samples_per_second": 16.862, | |
| "eval_score": 51.870194577496974, | |
| "eval_steps_per_second": 16.862, | |
| "eval_sys_len": 40984, | |
| "eval_totals": [ | |
| 40984, | |
| 37521, | |
| 34064, | |
| 30642 | |
| ], | |
| "step": 7000 | |
| } | |
| ], | |
| "max_steps": 253000, | |
| "num_train_epochs": 250, | |
| "total_flos": 9.650740613918746e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |