| { | |
| "best_metric": 1.5756924152374268, | |
| "best_model_checkpoint": "/scratch/s3545881/dumped/translation/mt5/3086466/checkpoint-2800", | |
| "epoch": 2.7092404450895016, | |
| "global_step": 2800, | |
| "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": 37.0498, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 5e-06, | |
| "loss": 37.4476, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 7.5e-06, | |
| "loss": 36.636, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 1e-05, | |
| "loss": 35.662, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 1.25e-05, | |
| "loss": 34.4439, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1.5e-05, | |
| "loss": 33.4065, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.7500000000000002e-05, | |
| "loss": 32.8662, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 2e-05, | |
| "loss": 32.7596, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 2.2499999999999998e-05, | |
| "loss": 31.2127, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 2.5e-05, | |
| "loss": 30.3467, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 2.75e-05, | |
| "loss": 28.3883, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 3e-05, | |
| "loss": 26.9506, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 24.2034, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 3.5000000000000004e-05, | |
| "loss": 22.9662, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 3.75e-05, | |
| "loss": 20.9191, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4e-05, | |
| "loss": 18.8087, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 4.25e-05, | |
| "loss": 17.9818, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.4999999999999996e-05, | |
| "loss": 16.0516, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.75e-05, | |
| "loss": 13.4505, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 5e-05, | |
| "loss": 12.6675, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 3907, | |
| 91, | |
| 14, | |
| 6 | |
| ], | |
| "eval_loss": 8.447338104248047, | |
| "eval_precisions": [ | |
| 0.5502692897061474, | |
| 0.012879411906237882, | |
| 0.0019911394295385534, | |
| 0.0008574956553553462 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 1345.1117, | |
| "eval_samples_per_second": 2.575, | |
| "eval_score": 0.010488206448671048, | |
| "eval_steps_per_second": 2.575, | |
| "eval_sys_len": 710016, | |
| "eval_totals": [ | |
| 710016, | |
| 706554, | |
| 703115, | |
| 699712 | |
| ], | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 5.25e-05, | |
| "loss": 11.941, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 5.5e-05, | |
| "loss": 10.961, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 5.75e-05, | |
| "loss": 10.3499, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 6e-05, | |
| "loss": 9.5122, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 6.25e-05, | |
| "loss": 9.1592, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 8.7164, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 6.75e-05, | |
| "loss": 8.4345, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 7.000000000000001e-05, | |
| "loss": 8.0707, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 7.25e-05, | |
| "loss": 7.959, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 7.5e-05, | |
| "loss": 7.661, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 7.75e-05, | |
| "loss": 7.7047, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 8e-05, | |
| "loss": 7.6889, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 8.25e-05, | |
| "loss": 7.7815, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 8.5e-05, | |
| "loss": 8.0411, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 8.75e-05, | |
| "loss": 7.8518, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 8.999999999999999e-05, | |
| "loss": 7.7647, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 9.25e-05, | |
| "loss": 7.4841, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 9.5e-05, | |
| "loss": 7.2992, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 9.750000000000001e-05, | |
| "loss": 7.2381, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0001, | |
| "loss": 7.1215, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 3135, | |
| 982, | |
| 295, | |
| 61 | |
| ], | |
| "eval_loss": 6.286175727844238, | |
| "eval_precisions": [ | |
| 0.9876939944424491, | |
| 0.31213545788873104, | |
| 0.09449404046907482, | |
| 0.019671708213744397 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 1325.4365, | |
| "eval_samples_per_second": 2.613, | |
| "eval_score": 0.15472233223389328, | |
| "eval_steps_per_second": 2.613, | |
| "eval_sys_len": 317406, | |
| "eval_totals": [ | |
| 317406, | |
| 314607, | |
| 312189, | |
| 310090 | |
| ], | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0001025, | |
| "loss": 7.0884, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.000105, | |
| "loss": 7.0528, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0001075, | |
| "loss": 7.0151, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00011, | |
| "loss": 6.7428, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00011250000000000001, | |
| "loss": 6.5513, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.000115, | |
| "loss": 6.4033, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.0001175, | |
| "loss": 6.0151, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00012, | |
| "loss": 5.5711, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0001225, | |
| "loss": 5.0243, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.000125, | |
| "loss": 4.8831, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0001275, | |
| "loss": 4.636, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.00013000000000000002, | |
| "loss": 4.5649, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00013250000000000002, | |
| "loss": 4.1967, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.000135, | |
| "loss": 4.4032, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0001375, | |
| "loss": 4.3111, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00014000000000000001, | |
| "loss": 4.2533, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.0001425, | |
| "loss": 4.2514, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.000145, | |
| "loss": 4.1326, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.0001475, | |
| "loss": 4.0604, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00015, | |
| "loss": 4.0115, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 22228, | |
| 12576, | |
| 7955, | |
| 5383 | |
| ], | |
| "eval_loss": 2.3380367755889893, | |
| "eval_precisions": [ | |
| 45.45975130890052, | |
| 27.68032047190368, | |
| 18.94273127753304, | |
| 13.947402513278922 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 572.0306, | |
| "eval_samples_per_second": 6.054, | |
| "eval_score": 24.01228700570531, | |
| "eval_steps_per_second": 6.054, | |
| "eval_sys_len": 48896, | |
| "eval_totals": [ | |
| 48896, | |
| 45433, | |
| 41995, | |
| 38595 | |
| ], | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.0001525, | |
| "loss": 3.9441, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.000155, | |
| "loss": 3.9152, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.0001575, | |
| "loss": 3.9774, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00016, | |
| "loss": 3.7911, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00016250000000000002, | |
| "loss": 3.8047, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.000165, | |
| "loss": 3.8829, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.0001675, | |
| "loss": 3.6344, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00017, | |
| "loss": 3.8138, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0001725, | |
| "loss": 3.7081, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.000175, | |
| "loss": 3.7708, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.0001775, | |
| "loss": 3.7171, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 3.5893, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0001825, | |
| "loss": 3.5494, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.000185, | |
| "loss": 3.6416, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0001875, | |
| "loss": 3.6174, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.00019, | |
| "loss": 3.6616, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00019250000000000002, | |
| "loss": 3.456, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00019500000000000002, | |
| "loss": 3.6063, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.0001975, | |
| "loss": 3.6423, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.0002, | |
| "loss": 3.6089, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 22303, | |
| 12687, | |
| 7909, | |
| 5220 | |
| ], | |
| "eval_loss": 2.027892589569092, | |
| "eval_precisions": [ | |
| 51.82164598726707, | |
| 32.05811749842072, | |
| 21.88494424306151, | |
| 15.94282572842221 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 511.1135, | |
| "eval_samples_per_second": 6.775, | |
| "eval_score": 27.592432621043134, | |
| "eval_steps_per_second": 6.775, | |
| "eval_sys_len": 43038, | |
| "eval_totals": [ | |
| 43038, | |
| 39575, | |
| 36139, | |
| 32742 | |
| ], | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00020250000000000002, | |
| "loss": 3.4158, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.000205, | |
| "loss": 3.409, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002075, | |
| "loss": 3.5841, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.00021, | |
| "loss": 3.4777, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.0002125, | |
| "loss": 3.4107, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 0.000215, | |
| "loss": 3.3874, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.0002175, | |
| "loss": 3.4342, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.00022, | |
| "loss": 3.4097, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.00022250000000000001, | |
| "loss": 3.3245, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.00022500000000000002, | |
| "loss": 3.3971, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 0.0002275, | |
| "loss": 3.2777, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.00023, | |
| "loss": 3.2984, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.0002325, | |
| "loss": 3.2537, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.000235, | |
| "loss": 3.3175, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.0002375, | |
| "loss": 3.287, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.00024, | |
| "loss": 3.2052, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.00024249999999999999, | |
| "loss": 3.2984, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.000245, | |
| "loss": 3.137, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.0002475, | |
| "loss": 3.2062, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.00025, | |
| "loss": 3.3031, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 22475, | |
| 12870, | |
| 8033, | |
| 5319 | |
| ], | |
| "eval_loss": 1.880346655845642, | |
| "eval_precisions": [ | |
| 52.89355392906733, | |
| 32.97632468996618, | |
| 22.57031271950774, | |
| 16.520172686896295 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 493.392, | |
| "eval_samples_per_second": 7.019, | |
| "eval_score": 28.39810035019441, | |
| "eval_steps_per_second": 7.019, | |
| "eval_sys_len": 42491, | |
| "eval_totals": [ | |
| 42491, | |
| 39028, | |
| 35591, | |
| 32197 | |
| ], | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 0.0002525, | |
| "loss": 3.3379, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.000255, | |
| "loss": 3.3144, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0002575, | |
| "loss": 3.3039, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 0.00026000000000000003, | |
| "loss": 3.2099, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.00026250000000000004, | |
| "loss": 3.203, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 0.00026500000000000004, | |
| "loss": 3.027, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 0.0002675, | |
| "loss": 3.0334, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 0.00027, | |
| "loss": 3.1445, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 0.0002725, | |
| "loss": 3.0927, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.000275, | |
| "loss": 3.1355, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.0002775, | |
| "loss": 2.9411, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.00028000000000000003, | |
| "loss": 3.0683, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0002825, | |
| "loss": 3.0704, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 0.000285, | |
| "loss": 3.1367, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.0002875, | |
| "loss": 3.1829, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.00029, | |
| "loss": 3.101, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.0002925, | |
| "loss": 3.0338, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 0.000295, | |
| "loss": 3.0992, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 0.00029749999999999997, | |
| "loss": 3.1358, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 0.0003, | |
| "loss": 3.0265, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 22069, | |
| 12741, | |
| 8000, | |
| 5303 | |
| ], | |
| "eval_loss": 1.8022379875183105, | |
| "eval_precisions": [ | |
| 49.50426200089726, | |
| 30.98718291704161, | |
| 21.2286055459732, | |
| 15.46019066499519 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 683.7176, | |
| "eval_samples_per_second": 5.065, | |
| "eval_score": 26.637302937486467, | |
| "eval_steps_per_second": 5.065, | |
| "eval_sys_len": 44580, | |
| "eval_totals": [ | |
| 44580, | |
| 41117, | |
| 37685, | |
| 34301 | |
| ], | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.0003025, | |
| "loss": 2.9454, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.000305, | |
| "loss": 3.0222, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 0.0003075, | |
| "loss": 3.08, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.00031, | |
| "loss": 2.9951, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 0.0003125, | |
| "loss": 2.9722, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 0.000315, | |
| "loss": 2.9865, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.0003175, | |
| "loss": 2.9643, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.00032, | |
| "loss": 2.9591, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.00032250000000000003, | |
| "loss": 2.9174, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 0.00032500000000000004, | |
| "loss": 2.9484, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 0.00032750000000000005, | |
| "loss": 2.9186, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.00033, | |
| "loss": 2.9222, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.0003325, | |
| "loss": 2.9231, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 0.000335, | |
| "loss": 2.8752, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.0003375, | |
| "loss": 2.8569, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 0.00034, | |
| "loss": 3.0137, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.00034250000000000003, | |
| "loss": 2.9199, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.000345, | |
| "loss": 2.9726, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.0003475, | |
| "loss": 2.906, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 0.00035, | |
| "loss": 2.9055, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23204, | |
| 13680, | |
| 8622, | |
| 5764 | |
| ], | |
| "eval_loss": 1.7418313026428223, | |
| "eval_precisions": [ | |
| 54.13147949423786, | |
| 34.697035026758314, | |
| 23.951330629479415, | |
| 17.671224477282482 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 544.287, | |
| "eval_samples_per_second": 6.362, | |
| "eval_score": 29.859644843289843, | |
| "eval_steps_per_second": 6.362, | |
| "eval_sys_len": 42866, | |
| "eval_totals": [ | |
| 42866, | |
| 39427, | |
| 35998, | |
| 32618 | |
| ], | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 0.0003525, | |
| "loss": 2.8119, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 0.000355, | |
| "loss": 3.0053, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0003575, | |
| "loss": 2.8639, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 0.00035999999999999997, | |
| "loss": 2.9254, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.0003625, | |
| "loss": 2.9477, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 0.000365, | |
| "loss": 3.0299, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.0003675, | |
| "loss": 2.8582, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.00037, | |
| "loss": 2.9433, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.0003725, | |
| "loss": 2.856, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.000375, | |
| "loss": 2.7648, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.0003775, | |
| "loss": 2.842, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.00038, | |
| "loss": 2.7609, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 0.00038250000000000003, | |
| "loss": 2.9978, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 0.00038500000000000003, | |
| "loss": 2.8184, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 0.00038750000000000004, | |
| "loss": 2.8703, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 0.00039000000000000005, | |
| "loss": 2.7835, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 0.0003925, | |
| "loss": 2.8729, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.000395, | |
| "loss": 2.8479, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.0003975, | |
| "loss": 2.8868, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 0.0004, | |
| "loss": 2.829, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23892, | |
| 14346, | |
| 9260, | |
| 6235 | |
| ], | |
| "eval_loss": 1.7061405181884766, | |
| "eval_precisions": [ | |
| 57.79110831599826, | |
| 37.873227909923706, | |
| 26.900618772332454, | |
| 20.09993552546744 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 422.3766, | |
| "eval_samples_per_second": 8.199, | |
| "eval_score": 32.98280367671718, | |
| "eval_steps_per_second": 8.199, | |
| "eval_sys_len": 41342, | |
| "eval_totals": [ | |
| 41342, | |
| 37879, | |
| 34423, | |
| 31020 | |
| ], | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.0004025, | |
| "loss": 2.7557, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 0.00040500000000000003, | |
| "loss": 2.8196, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 0.0004075, | |
| "loss": 2.6888, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 0.00041, | |
| "loss": 2.8251, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0004125, | |
| "loss": 2.8098, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.000415, | |
| "loss": 2.8685, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.0004175, | |
| "loss": 2.7419, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 0.00042, | |
| "loss": 2.8122, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.00042249999999999997, | |
| "loss": 2.8728, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.000425, | |
| "loss": 2.8772, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 0.0004275, | |
| "loss": 2.7859, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 0.00043, | |
| "loss": 2.8488, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.0004325, | |
| "loss": 2.6726, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.000435, | |
| "loss": 2.8862, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.0004375, | |
| "loss": 2.7678, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 0.00044, | |
| "loss": 2.8049, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 0.0004425, | |
| "loss": 2.7327, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 0.00044500000000000003, | |
| "loss": 2.791, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 0.00044750000000000004, | |
| "loss": 2.7355, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 0.00045000000000000004, | |
| "loss": 2.7519, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 22400, | |
| 13234, | |
| 8438, | |
| 5686 | |
| ], | |
| "eval_loss": 1.6596581935882568, | |
| "eval_precisions": [ | |
| 51.272660684856255, | |
| 32.89993784959602, | |
| 22.93246364995244, | |
| 17.015291618038724 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 755.4171, | |
| "eval_samples_per_second": 4.584, | |
| "eval_score": 28.483461117844286, | |
| "eval_steps_per_second": 4.584, | |
| "eval_sys_len": 43688, | |
| "eval_totals": [ | |
| 43688, | |
| 40225, | |
| 36795, | |
| 33417 | |
| ], | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.00045250000000000005, | |
| "loss": 2.8522, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 0.000455, | |
| "loss": 2.8933, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 0.0004575, | |
| "loss": 2.7223, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.00046, | |
| "loss": 2.7376, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.0004625, | |
| "loss": 2.6918, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 0.000465, | |
| "loss": 2.8248, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 0.00046750000000000003, | |
| "loss": 2.8406, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.00047, | |
| "loss": 2.6332, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 0.0004725, | |
| "loss": 2.8686, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 0.000475, | |
| "loss": 2.5939, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.0004775, | |
| "loss": 2.7392, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 0.00048, | |
| "loss": 2.7951, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.0004825, | |
| "loss": 2.7636, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 0.00048499999999999997, | |
| "loss": 2.8033, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0004875, | |
| "loss": 2.7528, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 0.00049, | |
| "loss": 2.71, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 0.0004925, | |
| "loss": 2.7453, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.000495, | |
| "loss": 2.8172, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 0.0004975, | |
| "loss": 2.8219, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.0005, | |
| "loss": 2.6677, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 22981, | |
| 13820, | |
| 8891, | |
| 6002 | |
| ], | |
| "eval_loss": 1.6349472999572754, | |
| "eval_precisions": [ | |
| 57.158135601651495, | |
| 37.612606482867484, | |
| 26.690081652257444, | |
| 20.05211813443806 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 477.7754, | |
| "eval_samples_per_second": 7.248, | |
| "eval_score": 32.751440814753295, | |
| "eval_steps_per_second": 7.248, | |
| "eval_sys_len": 40206, | |
| "eval_totals": [ | |
| 40206, | |
| 36743, | |
| 33312, | |
| 29932 | |
| ], | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.0005024999999999999, | |
| "loss": 2.7718, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 0.000505, | |
| "loss": 2.6684, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 0.0005074999999999999, | |
| "loss": 2.8251, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.00051, | |
| "loss": 2.6859, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 0.0005124999999999999, | |
| "loss": 2.7663, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.000515, | |
| "loss": 2.5372, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0005175, | |
| "loss": 2.714, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 0.0005200000000000001, | |
| "loss": 2.4456, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 0.0005225, | |
| "loss": 2.4832, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 0.0005250000000000001, | |
| "loss": 2.498, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0005275, | |
| "loss": 2.5938, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 0.0005300000000000001, | |
| "loss": 2.5316, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 0.0005325, | |
| "loss": 2.47, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 0.000535, | |
| "loss": 2.3812, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 0.0005375, | |
| "loss": 2.528, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 0.00054, | |
| "loss": 2.4435, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 0.0005425, | |
| "loss": 2.4662, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.000545, | |
| "loss": 2.4676, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 0.0005475, | |
| "loss": 2.5348, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 0.00055, | |
| "loss": 2.4479, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 24046, | |
| 14704, | |
| 9579, | |
| 6510 | |
| ], | |
| "eval_loss": 1.636635661125183, | |
| "eval_precisions": [ | |
| 61.04905047222504, | |
| 40.929714683368125, | |
| 29.480195734465884, | |
| 22.361144505890838 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 347.8604, | |
| "eval_samples_per_second": 9.955, | |
| "eval_score": 35.824942780304305, | |
| "eval_steps_per_second": 9.955, | |
| "eval_sys_len": 39388, | |
| "eval_totals": [ | |
| 39388, | |
| 35925, | |
| 32493, | |
| 29113 | |
| ], | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 0.0005525, | |
| "loss": 2.4267, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.000555, | |
| "loss": 2.3329, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 0.0005575, | |
| "loss": 2.531, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 0.0005600000000000001, | |
| "loss": 2.5151, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 0.0005625000000000001, | |
| "loss": 2.5409, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 0.000565, | |
| "loss": 2.4443, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 0.0005675, | |
| "loss": 2.4513, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 0.00057, | |
| "loss": 2.385, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 0.0005725, | |
| "loss": 2.4345, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 0.000575, | |
| "loss": 2.5575, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 0.0005775, | |
| "loss": 2.4997, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 0.00058, | |
| "loss": 2.4726, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 0.0005825, | |
| "loss": 2.4253, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.000585, | |
| "loss": 2.4573, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 0.0005875, | |
| "loss": 2.4581, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 0.00059, | |
| "loss": 2.5404, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 0.0005925, | |
| "loss": 2.5728, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 0.0005949999999999999, | |
| "loss": 2.4639, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 0.0005975, | |
| "loss": 2.6251, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 0.0006, | |
| "loss": 2.4606, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 22981, | |
| 13834, | |
| 8940, | |
| 6036 | |
| ], | |
| "eval_loss": 1.6156156063079834, | |
| "eval_precisions": [ | |
| 52.36283266496537, | |
| 34.22139764996908, | |
| 24.18438565167992, | |
| 17.986233201227687 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 623.2248, | |
| "eval_samples_per_second": 5.557, | |
| "eval_score": 29.713173488652338, | |
| "eval_steps_per_second": 5.557, | |
| "eval_sys_len": 43888, | |
| "eval_totals": [ | |
| 43888, | |
| 40425, | |
| 36966, | |
| 33559 | |
| ], | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.0006025000000000001, | |
| "loss": 2.5344, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 0.000605, | |
| "loss": 2.5511, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 0.0006075000000000001, | |
| "loss": 2.4595, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 0.00061, | |
| "loss": 2.4675, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 0.0006125000000000001, | |
| "loss": 2.4186, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 0.000615, | |
| "loss": 2.3482, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 0.0006175000000000001, | |
| "loss": 2.4653, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.00062, | |
| "loss": 2.5461, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 0.0006225000000000001, | |
| "loss": 2.4725, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 0.000625, | |
| "loss": 2.4625, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 0.0006274999999999999, | |
| "loss": 2.576, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 0.00063, | |
| "loss": 2.5898, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 0.0006324999999999999, | |
| "loss": 2.4081, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 0.000635, | |
| "loss": 2.3564, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 0.0006374999999999999, | |
| "loss": 2.3773, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 0.00064, | |
| "loss": 2.4186, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 0.0006425, | |
| "loss": 2.4812, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 0.0006450000000000001, | |
| "loss": 2.5489, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 0.0006475, | |
| "loss": 2.4855, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 0.0006500000000000001, | |
| "loss": 2.4259, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 24252, | |
| 14861, | |
| 9670, | |
| 6560 | |
| ], | |
| "eval_loss": 1.5984951257705688, | |
| "eval_precisions": [ | |
| 60.03416095254598, | |
| 40.23663832782802, | |
| 28.86136397552604, | |
| 21.777379411081235 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 371.9049, | |
| "eval_samples_per_second": 9.312, | |
| "eval_score": 35.10230625697626, | |
| "eval_steps_per_second": 9.312, | |
| "eval_sys_len": 40397, | |
| "eval_totals": [ | |
| 40397, | |
| 36934, | |
| 33505, | |
| 30123 | |
| ], | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 0.0006525, | |
| "loss": 2.4609, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 0.0006550000000000001, | |
| "loss": 2.4513, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 0.0006575, | |
| "loss": 2.4666, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 0.00066, | |
| "loss": 2.5221, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 0.0006625, | |
| "loss": 2.4292, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 0.000665, | |
| "loss": 2.4118, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 0.0006675, | |
| "loss": 2.5148, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 0.00067, | |
| "loss": 2.5603, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 0.0006725, | |
| "loss": 2.4776, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 0.000675, | |
| "loss": 2.4421, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 0.0006775, | |
| "loss": 2.5626, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 0.00068, | |
| "loss": 2.4886, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 0.0006825000000000001, | |
| "loss": 2.5124, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 0.0006850000000000001, | |
| "loss": 2.4808, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 0.0006875, | |
| "loss": 2.4922, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 0.00069, | |
| "loss": 2.5108, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 0.0006925, | |
| "loss": 2.4229, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.000695, | |
| "loss": 2.3996, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 0.0006975, | |
| "loss": 2.56, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 0.0007, | |
| "loss": 2.4794, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 24440, | |
| 15095, | |
| 9863, | |
| 6691 | |
| ], | |
| "eval_loss": 1.5756924152374268, | |
| "eval_precisions": [ | |
| 61.18412817624233, | |
| 41.3765692670358, | |
| 29.839954013251443, | |
| 22.549118727462677 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 376.0541, | |
| "eval_samples_per_second": 9.209, | |
| "eval_score": 36.12686234680497, | |
| "eval_steps_per_second": 9.209, | |
| "eval_sys_len": 39945, | |
| "eval_totals": [ | |
| 39945, | |
| 36482, | |
| 33053, | |
| 29673 | |
| ], | |
| "step": 2800 | |
| } | |
| ], | |
| "max_steps": 258250, | |
| "num_train_epochs": 250, | |
| "total_flos": 4.09482327561216e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |