| { | |
| "best_metric": 0.5190737247467041, | |
| "best_model_checkpoint": "/scratch/s3545881/dumped/translation/byt5/2872662/checkpoint-7600", | |
| "epoch": 7.508644821340359, | |
| "global_step": 7600, | |
| "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": 6.3753, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 5e-06, | |
| "loss": 5.7611, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 7.5e-06, | |
| "loss": 4.8915, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 1e-05, | |
| "loss": 3.9166, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 1.25e-05, | |
| "loss": 3.0447, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1.5e-05, | |
| "loss": 2.3818, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.7500000000000002e-05, | |
| "loss": 2.0129, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 2e-05, | |
| "loss": 1.9158, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 2.2499999999999998e-05, | |
| "loss": 1.7987, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.7847, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 2.75e-05, | |
| "loss": 1.7354, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 3e-05, | |
| "loss": 1.6959, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 1.6236, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 3.5000000000000004e-05, | |
| "loss": 1.6548, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 3.75e-05, | |
| "loss": 1.6176, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 4e-05, | |
| "loss": 1.6133, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.25e-05, | |
| "loss": 1.5805, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.4999999999999996e-05, | |
| "loss": 1.563, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 4.75e-05, | |
| "loss": 1.5905, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5598, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 18970, | |
| 10390, | |
| 6480, | |
| 4402 | |
| ], | |
| "eval_loss": 0.8889120221138, | |
| "eval_precisions": [ | |
| 11.51162085077978, | |
| 6.439337597302791, | |
| 4.103343465045593, | |
| 2.8486009370227525 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 1301.2843, | |
| "eval_samples_per_second": 2.661, | |
| "eval_score": 5.425462783898391, | |
| "eval_steps_per_second": 2.661, | |
| "eval_sys_len": 164790, | |
| "eval_totals": [ | |
| 164790, | |
| 161352, | |
| 157920, | |
| 154532 | |
| ], | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 5.25e-05, | |
| "loss": 1.5432, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 5.5e-05, | |
| "loss": 1.4932, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 5.75e-05, | |
| "loss": 1.5247, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 6e-05, | |
| "loss": 1.4728, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 6.25e-05, | |
| "loss": 1.4548, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 1.4672, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 6.75e-05, | |
| "loss": 1.4647, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 7.000000000000001e-05, | |
| "loss": 1.4552, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 7.25e-05, | |
| "loss": 1.4364, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 7.5e-05, | |
| "loss": 1.416, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 7.75e-05, | |
| "loss": 1.4421, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 8e-05, | |
| "loss": 1.4383, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 8.25e-05, | |
| "loss": 1.4207, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 8.5e-05, | |
| "loss": 1.4607, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 8.75e-05, | |
| "loss": 1.4313, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 8.999999999999999e-05, | |
| "loss": 1.3747, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 9.25e-05, | |
| "loss": 1.363, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 9.5e-05, | |
| "loss": 1.3928, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 9.750000000000001e-05, | |
| "loss": 1.3761, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0001, | |
| "loss": 1.3532, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 20469, | |
| 11530, | |
| 7295, | |
| 4985 | |
| ], | |
| "eval_loss": 0.7589057087898254, | |
| "eval_precisions": [ | |
| 30.113870417230626, | |
| 17.87319795380561, | |
| 11.947656326771266, | |
| 8.64805787346252 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 834.6439, | |
| "eval_samples_per_second": 4.149, | |
| "eval_score": 15.35651038926585, | |
| "eval_steps_per_second": 4.149, | |
| "eval_sys_len": 67972, | |
| "eval_totals": [ | |
| 67972, | |
| 64510, | |
| 61058, | |
| 57643 | |
| ], | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0001025, | |
| "loss": 1.3415, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.000105, | |
| "loss": 1.3655, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0001075, | |
| "loss": 1.358, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00011, | |
| "loss": 1.3552, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00011250000000000001, | |
| "loss": 1.3216, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.000115, | |
| "loss": 1.3447, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0001175, | |
| "loss": 1.3075, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00012, | |
| "loss": 1.3461, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0001225, | |
| "loss": 1.3383, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.000125, | |
| "loss": 1.3617, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0001275, | |
| "loss": 1.3188, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00013000000000000002, | |
| "loss": 1.2593, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00013250000000000002, | |
| "loss": 1.2945, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.000135, | |
| "loss": 1.3043, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0001375, | |
| "loss": 1.2945, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.00014000000000000001, | |
| "loss": 1.3191, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.0001425, | |
| "loss": 1.323, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.000145, | |
| "loss": 1.2948, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0001475, | |
| "loss": 1.2496, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00015, | |
| "loss": 1.2418, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 21696, | |
| 12448, | |
| 7897, | |
| 5370 | |
| ], | |
| "eval_loss": 0.6997856497764587, | |
| "eval_precisions": [ | |
| 40.1904303206565, | |
| 24.62755960035612, | |
| 16.76076067577893, | |
| 12.28018020078209 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 616.3616, | |
| "eval_samples_per_second": 5.618, | |
| "eval_score": 21.24519026821098, | |
| "eval_steps_per_second": 5.618, | |
| "eval_sys_len": 53983, | |
| "eval_totals": [ | |
| 53983, | |
| 50545, | |
| 47116, | |
| 43729 | |
| ], | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.0001525, | |
| "loss": 1.2905, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.000155, | |
| "loss": 1.2839, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0001575, | |
| "loss": 1.2958, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00016, | |
| "loss": 1.2598, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.00016250000000000002, | |
| "loss": 1.3098, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.000165, | |
| "loss": 1.2969, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0001675, | |
| "loss": 1.2021, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.00017, | |
| "loss": 1.2976, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.0001725, | |
| "loss": 1.2447, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.000175, | |
| "loss": 1.2866, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.0001775, | |
| "loss": 1.2554, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 1.2714, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.0001825, | |
| "loss": 1.2553, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.000185, | |
| "loss": 1.2123, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.0001875, | |
| "loss": 1.283, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00019, | |
| "loss": 1.219, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00019250000000000002, | |
| "loss": 1.2568, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00019500000000000002, | |
| "loss": 1.2754, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0001975, | |
| "loss": 1.2136, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.0002, | |
| "loss": 1.272, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 21698, | |
| 12261, | |
| 7570, | |
| 5005 | |
| ], | |
| "eval_loss": 0.6696712374687195, | |
| "eval_precisions": [ | |
| 44.981135204610474, | |
| 27.383584589614742, | |
| 18.31909590300801, | |
| 13.20162481536189 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 559.4526, | |
| "eval_samples_per_second": 6.19, | |
| "eval_score": 23.36216210014707, | |
| "eval_steps_per_second": 6.19, | |
| "eval_sys_len": 48238, | |
| "eval_totals": [ | |
| 48238, | |
| 44775, | |
| 41323, | |
| 37912 | |
| ], | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.00020250000000000002, | |
| "loss": 1.2232, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.000205, | |
| "loss": 1.1903, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.0002075, | |
| "loss": 1.2297, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 0.00021, | |
| "loss": 1.2287, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.0002125, | |
| "loss": 1.1917, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.000215, | |
| "loss": 1.1999, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.0002175, | |
| "loss": 1.2152, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.00022, | |
| "loss": 1.1885, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 0.00022250000000000001, | |
| "loss": 1.2142, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.00022500000000000002, | |
| "loss": 1.1942, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.0002275, | |
| "loss": 1.1873, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.00023, | |
| "loss": 1.2147, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.0002325, | |
| "loss": 1.2065, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.000235, | |
| "loss": 1.2102, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.0002375, | |
| "loss": 1.1659, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.00024, | |
| "loss": 1.1554, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.00024249999999999999, | |
| "loss": 1.1555, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.000245, | |
| "loss": 1.1889, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 0.0002475, | |
| "loss": 1.1182, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.00025, | |
| "loss": 1.19, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23510, | |
| 13826, | |
| 8781, | |
| 5898 | |
| ], | |
| "eval_loss": 0.6404776573181152, | |
| "eval_precisions": [ | |
| 55.5647467561627, | |
| 35.58999176276771, | |
| 24.808588783726513, | |
| 18.439317201275557 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 352.6488, | |
| "eval_samples_per_second": 9.82, | |
| "eval_score": 30.84029550072649, | |
| "eval_steps_per_second": 9.82, | |
| "eval_sys_len": 42311, | |
| "eval_totals": [ | |
| 42311, | |
| 38848, | |
| 35395, | |
| 31986 | |
| ], | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0002525, | |
| "loss": 1.1579, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 0.000255, | |
| "loss": 1.1308, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002575, | |
| "loss": 1.1322, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 0.00026000000000000003, | |
| "loss": 1.2081, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 0.00026250000000000004, | |
| "loss": 1.1808, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 0.00026500000000000004, | |
| "loss": 1.1789, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.0002675, | |
| "loss": 1.1565, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.00027, | |
| "loss": 1.1788, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.0002725, | |
| "loss": 1.1433, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.000275, | |
| "loss": 1.1283, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 0.0002775, | |
| "loss": 1.1419, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.00028000000000000003, | |
| "loss": 1.1265, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.0002825, | |
| "loss": 1.1686, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.000285, | |
| "loss": 1.2066, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 0.0002875, | |
| "loss": 1.157, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 0.00029, | |
| "loss": 1.1044, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 0.0002925, | |
| "loss": 1.1021, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.000295, | |
| "loss": 1.1836, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.00029749999999999997, | |
| "loss": 1.1113, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 0.0003, | |
| "loss": 1.1544, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23760, | |
| 14076, | |
| 8957, | |
| 6008 | |
| ], | |
| "eval_loss": 0.6251839995384216, | |
| "eval_precisions": [ | |
| 55.710567656920446, | |
| 35.92099219108865, | |
| 25.06716668532408, | |
| 18.588533770613534 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 361.3566, | |
| "eval_samples_per_second": 9.583, | |
| "eval_score": 31.074852281559178, | |
| "eval_steps_per_second": 9.583, | |
| "eval_sys_len": 42649, | |
| "eval_totals": [ | |
| 42649, | |
| 39186, | |
| 35732, | |
| 32321 | |
| ], | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.0003025, | |
| "loss": 1.1165, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 0.000305, | |
| "loss": 1.1519, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 0.0003075, | |
| "loss": 1.1087, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.00031, | |
| "loss": 1.1868, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.0003125, | |
| "loss": 1.1668, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.000315, | |
| "loss": 1.1627, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.0003175, | |
| "loss": 1.1495, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 0.00032, | |
| "loss": 1.1543, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 0.00032250000000000003, | |
| "loss": 1.12, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.00032500000000000004, | |
| "loss": 1.1203, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.00032750000000000005, | |
| "loss": 1.1677, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 0.00033, | |
| "loss": 1.0841, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.0003325, | |
| "loss": 1.0987, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 0.000335, | |
| "loss": 1.152, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0003375, | |
| "loss": 1.0829, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.00034, | |
| "loss": 1.1071, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 0.00034250000000000003, | |
| "loss": 1.0975, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 0.000345, | |
| "loss": 1.1018, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 0.0003475, | |
| "loss": 1.1136, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.00035, | |
| "loss": 1.1412, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23291, | |
| 13729, | |
| 8658, | |
| 5767 | |
| ], | |
| "eval_loss": 0.6117897033691406, | |
| "eval_precisions": [ | |
| 48.52089496270989, | |
| 30.824670513482566, | |
| 21.07338444687842, | |
| 15.308045549863298 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 514.8274, | |
| "eval_samples_per_second": 6.727, | |
| "eval_score": 26.355442996408566, | |
| "eval_steps_per_second": 6.727, | |
| "eval_sys_len": 48002, | |
| "eval_totals": [ | |
| 48002, | |
| 44539, | |
| 41085, | |
| 37673 | |
| ], | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 0.0003525, | |
| "loss": 1.1045, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.000355, | |
| "loss": 1.1118, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 0.0003575, | |
| "loss": 1.1085, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.00035999999999999997, | |
| "loss": 1.1063, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.0003625, | |
| "loss": 1.1177, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.000365, | |
| "loss": 1.084, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.0003675, | |
| "loss": 1.1397, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.00037, | |
| "loss": 1.0966, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.0003725, | |
| "loss": 1.0949, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 0.000375, | |
| "loss": 1.1, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 0.0003775, | |
| "loss": 1.1104, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 0.00038, | |
| "loss": 1.1082, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 0.00038250000000000003, | |
| "loss": 1.0897, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 0.00038500000000000003, | |
| "loss": 1.1307, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.00038750000000000004, | |
| "loss": 1.0821, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.00039000000000000005, | |
| "loss": 1.0819, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 0.0003925, | |
| "loss": 1.0677, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.000395, | |
| "loss": 1.0852, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 0.0003975, | |
| "loss": 1.0922, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 0.0004, | |
| "loss": 1.0695, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 22253, | |
| 12934, | |
| 8080, | |
| 5242 | |
| ], | |
| "eval_loss": 0.6057727932929993, | |
| "eval_precisions": [ | |
| 41.40632268388441, | |
| 25.723945902943516, | |
| 17.25647651795058, | |
| 12.074724160965609 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 611.9492, | |
| "eval_samples_per_second": 5.659, | |
| "eval_score": 21.70493043603416, | |
| "eval_steps_per_second": 5.659, | |
| "eval_sys_len": 53743, | |
| "eval_totals": [ | |
| 53743, | |
| 50280, | |
| 46823, | |
| 43413 | |
| ], | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 0.0004025, | |
| "loss": 1.0798, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.00040500000000000003, | |
| "loss": 1.119, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.0004075, | |
| "loss": 1.0528, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.00041, | |
| "loss": 1.0703, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 0.0004125, | |
| "loss": 1.0703, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.000415, | |
| "loss": 1.1318, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 0.0004175, | |
| "loss": 1.1081, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 0.00042, | |
| "loss": 1.0828, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.00042249999999999997, | |
| "loss": 1.0449, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.000425, | |
| "loss": 1.0627, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.0004275, | |
| "loss": 1.0338, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 0.00043, | |
| "loss": 1.098, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 0.0004325, | |
| "loss": 1.1104, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 0.000435, | |
| "loss": 1.0415, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 0.0004375, | |
| "loss": 1.079, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 0.00044, | |
| "loss": 1.0831, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.0004425, | |
| "loss": 1.0328, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 0.00044500000000000003, | |
| "loss": 1.0521, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 0.00044750000000000004, | |
| "loss": 1.0577, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.00045000000000000004, | |
| "loss": 1.0597, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23666, | |
| 14098, | |
| 8958, | |
| 5990 | |
| ], | |
| "eval_loss": 0.5892692804336548, | |
| "eval_precisions": [ | |
| 55.72272845000118, | |
| 36.141304347826086, | |
| 25.19547730213197, | |
| 18.63663233875735 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 372.8002, | |
| "eval_samples_per_second": 9.289, | |
| "eval_score": 31.183980903985358, | |
| "eval_steps_per_second": 9.289, | |
| "eval_sys_len": 42471, | |
| "eval_totals": [ | |
| 42471, | |
| 39008, | |
| 35554, | |
| 32141 | |
| ], | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.00045250000000000005, | |
| "loss": 1.0905, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 0.000455, | |
| "loss": 1.0904, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 0.0004575, | |
| "loss": 1.0865, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.00046, | |
| "loss": 1.0519, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 0.0004625, | |
| "loss": 1.0671, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 0.000465, | |
| "loss": 1.0415, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.00046750000000000003, | |
| "loss": 1.0543, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 0.00047, | |
| "loss": 1.0369, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.0004725, | |
| "loss": 1.0153, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 0.000475, | |
| "loss": 1.0551, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0004775, | |
| "loss": 1.0606, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 0.00048, | |
| "loss": 1.0965, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 0.0004825, | |
| "loss": 1.0778, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.00048499999999999997, | |
| "loss": 1.0879, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 0.0004875, | |
| "loss": 1.006, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.00049, | |
| "loss": 1.0478, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 0.0004925, | |
| "loss": 1.0454, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 0.000495, | |
| "loss": 1.0507, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.0004975, | |
| "loss": 0.9912, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 0.0005, | |
| "loss": 1.053, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23393, | |
| 14036, | |
| 9032, | |
| 6018 | |
| ], | |
| "eval_loss": 0.581674337387085, | |
| "eval_precisions": [ | |
| 56.94082710610228, | |
| 37.30994152046784, | |
| 26.435637768541824, | |
| 19.56754999187124 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 351.8793, | |
| "eval_samples_per_second": 9.841, | |
| "eval_score": 32.377524897401834, | |
| "eval_steps_per_second": 9.841, | |
| "eval_sys_len": 41083, | |
| "eval_totals": [ | |
| 41083, | |
| 37620, | |
| 34166, | |
| 30755 | |
| ], | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.0005024999999999999, | |
| "loss": 1.0799, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.000505, | |
| "loss": 1.0574, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 0.0005074999999999999, | |
| "loss": 1.0629, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 0.00051, | |
| "loss": 1.0294, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 0.0005124999999999999, | |
| "loss": 1.0194, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.000515, | |
| "loss": 1.0328, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 0.0005175, | |
| "loss": 1.0219, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 0.0005200000000000001, | |
| "loss": 1.0743, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 0.0005225, | |
| "loss": 1.0261, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 0.0005250000000000001, | |
| "loss": 1.0134, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 0.0005275, | |
| "loss": 1.0279, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 0.0005300000000000001, | |
| "loss": 1.0651, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 0.0005325, | |
| "loss": 1.0665, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.000535, | |
| "loss": 0.9793, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 0.0005375, | |
| "loss": 0.9911, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 0.00054, | |
| "loss": 1.0008, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 0.0005425, | |
| "loss": 1.0268, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.000545, | |
| "loss": 0.9924, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 0.0005475, | |
| "loss": 1.0377, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 0.00055, | |
| "loss": 1.011, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 22978, | |
| 13794, | |
| 8850, | |
| 5936 | |
| ], | |
| "eval_loss": 0.5843345522880554, | |
| "eval_precisions": [ | |
| 40.11592381151905, | |
| 25.631782369555523, | |
| 17.57277312259243, | |
| 12.64350678395706 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 732.1458, | |
| "eval_samples_per_second": 4.73, | |
| "eval_score": 21.862552388934734, | |
| "eval_steps_per_second": 4.73, | |
| "eval_sys_len": 57279, | |
| "eval_totals": [ | |
| 57279, | |
| 53816, | |
| 50362, | |
| 46949 | |
| ], | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 0.0005525, | |
| "loss": 1.0545, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 0.000555, | |
| "loss": 0.9986, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 0.0005575, | |
| "loss": 0.9874, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 0.0005600000000000001, | |
| "loss": 0.9982, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 0.0005625000000000001, | |
| "loss": 1.0326, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 0.000565, | |
| "loss": 1.0462, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 0.0005675, | |
| "loss": 0.9829, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 0.00057, | |
| "loss": 1.005, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.0005725, | |
| "loss": 1.0239, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 0.000575, | |
| "loss": 1.0255, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 0.0005775, | |
| "loss": 0.9746, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 0.00058, | |
| "loss": 1.0039, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 0.0005825, | |
| "loss": 1.0168, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 0.000585, | |
| "loss": 0.9802, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 0.0005875, | |
| "loss": 1.0143, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.00059, | |
| "loss": 0.9687, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 0.0005925, | |
| "loss": 1.0321, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 0.0005949999999999999, | |
| "loss": 1.0312, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 0.0005975, | |
| "loss": 1.0452, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 0.0006, | |
| "loss": 0.9862, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23858, | |
| 14449, | |
| 9220, | |
| 6162 | |
| ], | |
| "eval_loss": 0.5717904567718506, | |
| "eval_precisions": [ | |
| 56.85891325071497, | |
| 37.5327947632283, | |
| 26.309781988357493, | |
| 19.47904153758614 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 372.0168, | |
| "eval_samples_per_second": 9.309, | |
| "eval_score": 32.33877561394095, | |
| "eval_steps_per_second": 9.309, | |
| "eval_sys_len": 41960, | |
| "eval_totals": [ | |
| 41960, | |
| 38497, | |
| 35044, | |
| 31634 | |
| ], | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 0.0006025000000000001, | |
| "loss": 1.0001, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 0.000605, | |
| "loss": 0.9679, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0006075000000000001, | |
| "loss": 1.003, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 0.00061, | |
| "loss": 0.9323, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 0.0006125000000000001, | |
| "loss": 1.0117, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 0.000615, | |
| "loss": 0.9984, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 0.0006175000000000001, | |
| "loss": 1.0382, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 0.00062, | |
| "loss": 1.04, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 0.0006225000000000001, | |
| "loss": 0.9991, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 0.000625, | |
| "loss": 0.9853, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 0.0006274999999999999, | |
| "loss": 0.9927, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 0.00063, | |
| "loss": 0.9613, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 0.0006324999999999999, | |
| "loss": 1.022, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 0.000635, | |
| "loss": 0.9686, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 0.0006374999999999999, | |
| "loss": 1.0023, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 0.00064, | |
| "loss": 1.029, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 0.0006425, | |
| "loss": 0.9776, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 0.0006450000000000001, | |
| "loss": 1.0, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 0.0006475, | |
| "loss": 0.9978, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 0.0006500000000000001, | |
| "loss": 1.0347, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23864, | |
| 14413, | |
| 9231, | |
| 6182 | |
| ], | |
| "eval_loss": 0.5666627883911133, | |
| "eval_precisions": [ | |
| 51.390055343799126, | |
| 33.538883976357795, | |
| 23.357793522267208, | |
| 17.118963225520602 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 469.9629, | |
| "eval_samples_per_second": 7.369, | |
| "eval_score": 28.812718861200807, | |
| "eval_steps_per_second": 7.369, | |
| "eval_sys_len": 46437, | |
| "eval_totals": [ | |
| 46437, | |
| 42974, | |
| 39520, | |
| 36112 | |
| ], | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 0.0006525, | |
| "loss": 0.9897, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 0.0006550000000000001, | |
| "loss": 0.9548, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 0.0006575, | |
| "loss": 0.9939, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 0.00066, | |
| "loss": 0.9751, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 0.0006625, | |
| "loss": 1.0134, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 0.000665, | |
| "loss": 0.9915, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 0.0006675, | |
| "loss": 1.0162, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 0.00067, | |
| "loss": 0.9882, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 0.0006725, | |
| "loss": 1.0007, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 0.000675, | |
| "loss": 1.0213, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 0.0006775, | |
| "loss": 1.0142, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.00068, | |
| "loss": 0.9928, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 0.0006825000000000001, | |
| "loss": 0.9857, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 0.0006850000000000001, | |
| "loss": 1.027, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 0.0006875, | |
| "loss": 0.9968, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 0.00069, | |
| "loss": 1.0254, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 0.0006925, | |
| "loss": 0.9792, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 0.000695, | |
| "loss": 0.9248, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 0.0006975, | |
| "loss": 0.9874, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 0.0007, | |
| "loss": 0.987, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 24280, | |
| 14962, | |
| 9786, | |
| 6610 | |
| ], | |
| "eval_loss": 0.5556033253669739, | |
| "eval_precisions": [ | |
| 53.74416185226996, | |
| 35.868053890780075, | |
| 25.5796324855582, | |
| 18.97135640893175 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 420.0219, | |
| "eval_samples_per_second": 8.245, | |
| "eval_score": 31.09982518936309, | |
| "eval_steps_per_second": 8.245, | |
| "eval_sys_len": 45177, | |
| "eval_totals": [ | |
| 45177, | |
| 41714, | |
| 38257, | |
| 34842 | |
| ], | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 0.0007025, | |
| "loss": 0.9964, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 0.000705, | |
| "loss": 0.9792, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.0007075, | |
| "loss": 0.9914, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 0.00071, | |
| "loss": 0.9503, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 0.0007125, | |
| "loss": 0.9789, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 0.000715, | |
| "loss": 1.0009, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.0007175, | |
| "loss": 0.9373, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 0.0007199999999999999, | |
| "loss": 0.978, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 0.0007225, | |
| "loss": 0.9694, | |
| "step": 2890 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 0.000725, | |
| "loss": 1.0054, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 0.0007275000000000001, | |
| "loss": 1.0342, | |
| "step": 2910 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 0.00073, | |
| "loss": 0.983, | |
| "step": 2920 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 0.0007325000000000001, | |
| "loss": 0.9998, | |
| "step": 2930 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 0.000735, | |
| "loss": 0.9899, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 0.0007375000000000001, | |
| "loss": 0.9702, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 0.00074, | |
| "loss": 0.9673, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 0.0007425000000000001, | |
| "loss": 0.9844, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 0.000745, | |
| "loss": 0.9725, | |
| "step": 2980 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 0.0007475000000000001, | |
| "loss": 0.9842, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 0.00075, | |
| "loss": 0.9833, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23935, | |
| 14729, | |
| 9656, | |
| 6563 | |
| ], | |
| "eval_loss": 0.5605592727661133, | |
| "eval_precisions": [ | |
| 60.27904399728008, | |
| 40.63845050215208, | |
| 29.44890054591479, | |
| 22.339846143372593 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 320.2647, | |
| "eval_samples_per_second": 10.813, | |
| "eval_score": 35.62978402935608, | |
| "eval_steps_per_second": 10.813, | |
| "eval_sys_len": 39707, | |
| "eval_totals": [ | |
| 39707, | |
| 36244, | |
| 32789, | |
| 29378 | |
| ], | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 0.0007524999999999999, | |
| "loss": 0.9558, | |
| "step": 3010 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 0.000755, | |
| "loss": 1.0105, | |
| "step": 3020 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 0.0007574999999999999, | |
| "loss": 0.9627, | |
| "step": 3030 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.00076, | |
| "loss": 1.0326, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 0.0007624999999999999, | |
| "loss": 0.922, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 0.0007650000000000001, | |
| "loss": 0.9251, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 0.0007675, | |
| "loss": 0.9837, | |
| "step": 3070 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 0.0007700000000000001, | |
| "loss": 0.9609, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 0.0007725, | |
| "loss": 0.9575, | |
| "step": 3090 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 0.0007750000000000001, | |
| "loss": 0.9229, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 0.0007775, | |
| "loss": 1.0113, | |
| "step": 3110 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 0.0007800000000000001, | |
| "loss": 0.9542, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 0.0007825, | |
| "loss": 0.9876, | |
| "step": 3130 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 0.000785, | |
| "loss": 0.9544, | |
| "step": 3140 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 0.0007875, | |
| "loss": 0.9329, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 0.00079, | |
| "loss": 0.929, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 0.0007925, | |
| "loss": 0.9549, | |
| "step": 3170 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 0.000795, | |
| "loss": 0.9595, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 0.0007975, | |
| "loss": 0.9594, | |
| "step": 3190 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 0.0008, | |
| "loss": 0.9311, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23463, | |
| 14060, | |
| 8878, | |
| 5926 | |
| ], | |
| "eval_loss": 0.5624625086784363, | |
| "eval_precisions": [ | |
| 53.87724172770902, | |
| 35.074589632290575, | |
| 24.23961120515481, | |
| 17.846709832856497 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 375.5697, | |
| "eval_samples_per_second": 9.221, | |
| "eval_score": 30.06910081052927, | |
| "eval_steps_per_second": 9.221, | |
| "eval_sys_len": 43549, | |
| "eval_totals": [ | |
| 43549, | |
| 40086, | |
| 36626, | |
| 33205 | |
| ], | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 0.0008025, | |
| "loss": 1.0039, | |
| "step": 3210 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 0.000805, | |
| "loss": 0.9571, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 0.0008075000000000001, | |
| "loss": 0.9164, | |
| "step": 3230 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.0008100000000000001, | |
| "loss": 0.9176, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 0.0008125000000000001, | |
| "loss": 0.9473, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 0.000815, | |
| "loss": 0.9515, | |
| "step": 3260 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 0.0008175, | |
| "loss": 0.9084, | |
| "step": 3270 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 0.00082, | |
| "loss": 0.9383, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 0.0008225, | |
| "loss": 0.9123, | |
| "step": 3290 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 0.000825, | |
| "loss": 0.9527, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 0.0008275, | |
| "loss": 0.926, | |
| "step": 3310 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 0.00083, | |
| "loss": 0.9563, | |
| "step": 3320 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 0.0008325, | |
| "loss": 0.9396, | |
| "step": 3330 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 0.000835, | |
| "loss": 0.9786, | |
| "step": 3340 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 0.0008375, | |
| "loss": 0.9732, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 0.00084, | |
| "loss": 0.966, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 0.0008425, | |
| "loss": 0.919, | |
| "step": 3370 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 0.0008449999999999999, | |
| "loss": 0.9251, | |
| "step": 3380 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 0.0008475000000000001, | |
| "loss": 0.9586, | |
| "step": 3390 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 0.00085, | |
| "loss": 0.9546, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 24677, | |
| 15294, | |
| 10029, | |
| 6853 | |
| ], | |
| "eval_loss": 0.5574368834495544, | |
| "eval_precisions": [ | |
| 58.379465341850015, | |
| 39.41041564666168, | |
| 28.37057991513437, | |
| 21.456526503647577 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 346.8216, | |
| "eval_samples_per_second": 9.985, | |
| "eval_score": 34.40128694748695, | |
| "eval_steps_per_second": 9.985, | |
| "eval_sys_len": 42270, | |
| "eval_totals": [ | |
| 42270, | |
| 38807, | |
| 35350, | |
| 31939 | |
| ], | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 0.0008525000000000001, | |
| "loss": 0.9117, | |
| "step": 3410 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 0.000855, | |
| "loss": 0.9558, | |
| "step": 3420 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 0.0008575000000000001, | |
| "loss": 0.938, | |
| "step": 3430 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 0.00086, | |
| "loss": 0.9186, | |
| "step": 3440 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 0.0008625000000000001, | |
| "loss": 0.9847, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 0.000865, | |
| "loss": 0.9426, | |
| "step": 3460 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 0.0008675000000000001, | |
| "loss": 0.933, | |
| "step": 3470 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 0.00087, | |
| "loss": 0.9508, | |
| "step": 3480 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 0.0008725000000000001, | |
| "loss": 0.9779, | |
| "step": 3490 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 0.000875, | |
| "loss": 0.9269, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 0.0008774999999999999, | |
| "loss": 0.9694, | |
| "step": 3510 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 0.00088, | |
| "loss": 0.9086, | |
| "step": 3520 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 0.0008824999999999999, | |
| "loss": 0.9309, | |
| "step": 3530 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 0.000885, | |
| "loss": 0.9549, | |
| "step": 3540 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 0.0008874999999999999, | |
| "loss": 0.9123, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 0.0008900000000000001, | |
| "loss": 0.9475, | |
| "step": 3560 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 0.0008925, | |
| "loss": 0.9416, | |
| "step": 3570 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 0.0008950000000000001, | |
| "loss": 0.9449, | |
| "step": 3580 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 0.0008975, | |
| "loss": 0.9189, | |
| "step": 3590 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 0.0009000000000000001, | |
| "loss": 0.9664, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23278, | |
| 14055, | |
| 9060, | |
| 6074 | |
| ], | |
| "eval_loss": 0.5639054775238037, | |
| "eval_precisions": [ | |
| 44.65032416465262, | |
| 28.87756569620513, | |
| 20.037598142209443, | |
| 14.5311004784689 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 598.5697, | |
| "eval_samples_per_second": 5.785, | |
| "eval_score": 24.75325645086999, | |
| "eval_steps_per_second": 5.785, | |
| "eval_sys_len": 52134, | |
| "eval_totals": [ | |
| 52134, | |
| 48671, | |
| 45215, | |
| 41800 | |
| ], | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 0.0009025, | |
| "loss": 0.9641, | |
| "step": 3610 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 0.0009050000000000001, | |
| "loss": 0.9034, | |
| "step": 3620 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 0.0009075, | |
| "loss": 0.9138, | |
| "step": 3630 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 0.00091, | |
| "loss": 0.9151, | |
| "step": 3640 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 0.0009125, | |
| "loss": 0.9672, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 0.000915, | |
| "loss": 0.8992, | |
| "step": 3660 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 0.0009175, | |
| "loss": 0.9544, | |
| "step": 3670 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 0.00092, | |
| "loss": 0.9457, | |
| "step": 3680 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 0.0009225, | |
| "loss": 0.9213, | |
| "step": 3690 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 0.000925, | |
| "loss": 0.9314, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 0.0009275, | |
| "loss": 0.8843, | |
| "step": 3710 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 0.00093, | |
| "loss": 0.9525, | |
| "step": 3720 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 0.0009325000000000001, | |
| "loss": 0.9207, | |
| "step": 3730 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 0.0009350000000000001, | |
| "loss": 0.9335, | |
| "step": 3740 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 0.0009375, | |
| "loss": 0.9347, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 0.00094, | |
| "loss": 0.8988, | |
| "step": 3760 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 0.0009425, | |
| "loss": 0.9692, | |
| "step": 3770 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 0.000945, | |
| "loss": 0.9483, | |
| "step": 3780 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 0.0009475, | |
| "loss": 0.9303, | |
| "step": 3790 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 0.00095, | |
| "loss": 0.946, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 24172, | |
| 14980, | |
| 9782, | |
| 6671 | |
| ], | |
| "eval_loss": 0.5469653010368347, | |
| "eval_precisions": [ | |
| 56.98792908336477, | |
| 38.45660154545221, | |
| 27.558034708136127, | |
| 20.794887780548628 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 348.1102, | |
| "eval_samples_per_second": 9.948, | |
| "eval_score": 33.47646952015742, | |
| "eval_steps_per_second": 9.948, | |
| "eval_sys_len": 42416, | |
| "eval_totals": [ | |
| 42416, | |
| 38953, | |
| 35496, | |
| 32080 | |
| ], | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 0.0009525, | |
| "loss": 0.9247, | |
| "step": 3810 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 0.000955, | |
| "loss": 0.927, | |
| "step": 3820 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 0.0009575, | |
| "loss": 0.9174, | |
| "step": 3830 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 0.00096, | |
| "loss": 0.9457, | |
| "step": 3840 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 0.0009625, | |
| "loss": 0.9072, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 0.000965, | |
| "loss": 0.9264, | |
| "step": 3860 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 0.0009675, | |
| "loss": 0.9585, | |
| "step": 3870 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 0.0009699999999999999, | |
| "loss": 0.9321, | |
| "step": 3880 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 0.0009725000000000001, | |
| "loss": 0.9433, | |
| "step": 3890 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 0.000975, | |
| "loss": 0.8985, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 0.0009775, | |
| "loss": 0.9225, | |
| "step": 3910 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 0.00098, | |
| "loss": 0.9036, | |
| "step": 3920 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 0.0009825, | |
| "loss": 0.8769, | |
| "step": 3930 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 0.000985, | |
| "loss": 0.9365, | |
| "step": 3940 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 0.0009875, | |
| "loss": 0.9244, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 0.00099, | |
| "loss": 0.887, | |
| "step": 3960 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 0.0009925000000000001, | |
| "loss": 0.9069, | |
| "step": 3970 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 0.000995, | |
| "loss": 0.9175, | |
| "step": 3980 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 0.0009975000000000001, | |
| "loss": 0.9568, | |
| "step": 3990 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 0.001, | |
| "loss": 0.9456, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23116, | |
| 14094, | |
| 9177, | |
| 6228 | |
| ], | |
| "eval_loss": 0.5421901345252991, | |
| "eval_precisions": [ | |
| 53.0512014320795, | |
| 35.13836948391922, | |
| 25.036830905221805, | |
| 18.735334817399675 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 404.6207, | |
| "eval_samples_per_second": 8.559, | |
| "eval_score": 30.579431762487346, | |
| "eval_steps_per_second": 8.559, | |
| "eval_sys_len": 43573, | |
| "eval_totals": [ | |
| 43573, | |
| 40110, | |
| 36654, | |
| 33242 | |
| ], | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 0.001, | |
| "loss": 0.9335, | |
| "step": 4010 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "learning_rate": 0.001, | |
| "loss": 0.9347, | |
| "step": 4020 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 0.001, | |
| "loss": 0.9165, | |
| "step": 4030 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 0.001, | |
| "loss": 0.9636, | |
| "step": 4040 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.001, | |
| "loss": 0.9575, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 0.001, | |
| "loss": 0.9575, | |
| "step": 4060 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 0.001, | |
| "loss": 0.9079, | |
| "step": 4070 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 0.001, | |
| "loss": 0.8527, | |
| "step": 4080 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 0.001, | |
| "loss": 0.8988, | |
| "step": 4090 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 0.001, | |
| "loss": 0.8654, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 0.001, | |
| "loss": 0.9086, | |
| "step": 4110 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 0.001, | |
| "loss": 0.8412, | |
| "step": 4120 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 0.001, | |
| "loss": 0.8634, | |
| "step": 4130 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 0.001, | |
| "loss": 0.8624, | |
| "step": 4140 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 0.001, | |
| "loss": 0.8997, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 0.001, | |
| "loss": 0.8843, | |
| "step": 4160 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 0.001, | |
| "loss": 0.869, | |
| "step": 4170 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 0.001, | |
| "loss": 0.8582, | |
| "step": 4180 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 0.001, | |
| "loss": 0.8732, | |
| "step": 4190 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 0.001, | |
| "loss": 0.9132, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23011, | |
| 14101, | |
| 9241, | |
| 6287 | |
| ], | |
| "eval_loss": 0.5446494817733765, | |
| "eval_precisions": [ | |
| 46.28024376018181, | |
| 30.48337584850188, | |
| 21.58960820503236, | |
| 15.961308994897053 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 565.396, | |
| "eval_samples_per_second": 6.125, | |
| "eval_score": 26.405414619893257, | |
| "eval_steps_per_second": 6.125, | |
| "eval_sys_len": 49721, | |
| "eval_totals": [ | |
| 49721, | |
| 46258, | |
| 42803, | |
| 39389 | |
| ], | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 0.001, | |
| "loss": 0.9235, | |
| "step": 4210 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 0.001, | |
| "loss": 0.8471, | |
| "step": 4220 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "learning_rate": 0.001, | |
| "loss": 0.8852, | |
| "step": 4230 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 0.001, | |
| "loss": 0.8511, | |
| "step": 4240 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "learning_rate": 0.001, | |
| "loss": 0.889, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 0.001, | |
| "loss": 0.931, | |
| "step": 4260 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 0.001, | |
| "loss": 0.8874, | |
| "step": 4270 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "learning_rate": 0.001, | |
| "loss": 0.9147, | |
| "step": 4280 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 0.001, | |
| "loss": 0.8831, | |
| "step": 4290 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 0.001, | |
| "loss": 0.89, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 0.001, | |
| "loss": 0.8954, | |
| "step": 4310 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 0.001, | |
| "loss": 0.9295, | |
| "step": 4320 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 0.001, | |
| "loss": 0.9186, | |
| "step": 4330 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "learning_rate": 0.001, | |
| "loss": 0.9342, | |
| "step": 4340 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 0.001, | |
| "loss": 0.8757, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 0.001, | |
| "loss": 0.8796, | |
| "step": 4360 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 0.001, | |
| "loss": 0.9383, | |
| "step": 4370 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 0.001, | |
| "loss": 0.8797, | |
| "step": 4380 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 0.001, | |
| "loss": 0.9009, | |
| "step": 4390 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 0.001, | |
| "loss": 0.8949, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23669, | |
| 14687, | |
| 9709, | |
| 6652 | |
| ], | |
| "eval_loss": 0.5430779457092285, | |
| "eval_precisions": [ | |
| 55.20080227622557, | |
| 37.26246352911328, | |
| 26.99944382647386, | |
| 20.434996313590563 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 372.9478, | |
| "eval_samples_per_second": 9.285, | |
| "eval_score": 32.638980427379494, | |
| "eval_steps_per_second": 9.285, | |
| "eval_sys_len": 42878, | |
| "eval_totals": [ | |
| 42878, | |
| 39415, | |
| 35960, | |
| 32552 | |
| ], | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 0.001, | |
| "loss": 0.8768, | |
| "step": 4410 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 0.001, | |
| "loss": 0.8992, | |
| "step": 4420 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 0.001, | |
| "loss": 0.8467, | |
| "step": 4430 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 0.001, | |
| "loss": 0.8987, | |
| "step": 4440 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 0.001, | |
| "loss": 0.8622, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 0.001, | |
| "loss": 0.8811, | |
| "step": 4460 | |
| }, | |
| { | |
| "epoch": 4.42, | |
| "learning_rate": 0.001, | |
| "loss": 0.8575, | |
| "step": 4470 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "learning_rate": 0.001, | |
| "loss": 0.8882, | |
| "step": 4480 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 0.001, | |
| "loss": 0.8656, | |
| "step": 4490 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "learning_rate": 0.001, | |
| "loss": 0.8873, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 0.001, | |
| "loss": 0.8901, | |
| "step": 4510 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 0.001, | |
| "loss": 0.8735, | |
| "step": 4520 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 0.001, | |
| "loss": 0.8884, | |
| "step": 4530 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 0.001, | |
| "loss": 0.8678, | |
| "step": 4540 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 0.001, | |
| "loss": 0.9001, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "learning_rate": 0.001, | |
| "loss": 0.8658, | |
| "step": 4560 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 0.001, | |
| "loss": 0.8967, | |
| "step": 4570 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 0.001, | |
| "loss": 0.8948, | |
| "step": 4580 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "learning_rate": 0.001, | |
| "loss": 0.8629, | |
| "step": 4590 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "learning_rate": 0.001, | |
| "loss": 0.889, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 24079, | |
| 14897, | |
| 9754, | |
| 6690 | |
| ], | |
| "eval_loss": 0.5361272096633911, | |
| "eval_precisions": [ | |
| 53.15335202313415, | |
| 35.606386538553465, | |
| 25.41295398884894, | |
| 19.1323247633483 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 436.9295, | |
| "eval_samples_per_second": 7.926, | |
| "eval_score": 30.972079373426904, | |
| "eval_steps_per_second": 7.926, | |
| "eval_sys_len": 45301, | |
| "eval_totals": [ | |
| 45301, | |
| 41838, | |
| 38382, | |
| 34967 | |
| ], | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 0.001, | |
| "loss": 0.9193, | |
| "step": 4610 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "learning_rate": 0.001, | |
| "loss": 0.8286, | |
| "step": 4620 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 0.001, | |
| "loss": 0.8695, | |
| "step": 4630 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 0.001, | |
| "loss": 0.8635, | |
| "step": 4640 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 0.001, | |
| "loss": 0.8585, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "learning_rate": 0.001, | |
| "loss": 0.8848, | |
| "step": 4660 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "learning_rate": 0.001, | |
| "loss": 0.91, | |
| "step": 4670 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 0.001, | |
| "loss": 0.8693, | |
| "step": 4680 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 0.001, | |
| "loss": 0.8727, | |
| "step": 4690 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 0.001, | |
| "loss": 0.8518, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 0.001, | |
| "loss": 0.8639, | |
| "step": 4710 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "learning_rate": 0.001, | |
| "loss": 0.8717, | |
| "step": 4720 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 0.001, | |
| "loss": 0.8594, | |
| "step": 4730 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 0.001, | |
| "loss": 0.8901, | |
| "step": 4740 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 0.001, | |
| "loss": 0.8933, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 0.001, | |
| "loss": 0.8565, | |
| "step": 4760 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "learning_rate": 0.001, | |
| "loss": 0.8815, | |
| "step": 4770 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 0.001, | |
| "loss": 0.8678, | |
| "step": 4780 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 0.001, | |
| "loss": 0.8687, | |
| "step": 4790 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 0.001, | |
| "loss": 0.8657, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 25059, | |
| 15944, | |
| 10770, | |
| 7500 | |
| ], | |
| "eval_loss": 0.532075822353363, | |
| "eval_precisions": [ | |
| 60.45013750180924, | |
| 41.96783448711537, | |
| 31.18485059068798, | |
| 24.09793400379141 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 325.3938, | |
| "eval_samples_per_second": 10.642, | |
| "eval_score": 37.15859914069821, | |
| "eval_steps_per_second": 10.642, | |
| "eval_sys_len": 41454, | |
| "eval_totals": [ | |
| 41454, | |
| 37991, | |
| 34536, | |
| 31123 | |
| ], | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "learning_rate": 0.001, | |
| "loss": 0.8662, | |
| "step": 4810 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 0.001, | |
| "loss": 0.9208, | |
| "step": 4820 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 0.001, | |
| "loss": 0.866, | |
| "step": 4830 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 0.001, | |
| "loss": 0.8501, | |
| "step": 4840 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 0.001, | |
| "loss": 0.8426, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 0.001, | |
| "loss": 0.8651, | |
| "step": 4860 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "learning_rate": 0.001, | |
| "loss": 0.926, | |
| "step": 4870 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 0.001, | |
| "loss": 0.8888, | |
| "step": 4880 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 0.001, | |
| "loss": 0.8945, | |
| "step": 4890 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 0.001, | |
| "loss": 0.8741, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "learning_rate": 0.001, | |
| "loss": 0.8563, | |
| "step": 4910 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "learning_rate": 0.001, | |
| "loss": 0.8722, | |
| "step": 4920 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 0.001, | |
| "loss": 0.8682, | |
| "step": 4930 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 0.001, | |
| "loss": 0.8606, | |
| "step": 4940 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 0.001, | |
| "loss": 0.8976, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 0.001, | |
| "loss": 0.8993, | |
| "step": 4960 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "learning_rate": 0.001, | |
| "loss": 0.856, | |
| "step": 4970 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "learning_rate": 0.001, | |
| "loss": 0.9346, | |
| "step": 4980 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 0.001, | |
| "loss": 0.8674, | |
| "step": 4990 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "learning_rate": 0.001, | |
| "loss": 0.8872, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 24630, | |
| 15582, | |
| 10441, | |
| 7231 | |
| ], | |
| "eval_loss": 0.528194785118103, | |
| "eval_precisions": [ | |
| 54.96908965117058, | |
| 37.68866099071207, | |
| 27.554628945423836, | |
| 20.9673210195146 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 404.3358, | |
| "eval_samples_per_second": 8.565, | |
| "eval_score": 33.07628238949202, | |
| "eval_steps_per_second": 8.565, | |
| "eval_sys_len": 44807, | |
| "eval_totals": [ | |
| 44807, | |
| 41344, | |
| 37892, | |
| 34487 | |
| ], | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 0.001, | |
| "loss": 0.8873, | |
| "step": 5010 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 0.001, | |
| "loss": 0.8444, | |
| "step": 5020 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 0.001, | |
| "loss": 0.8671, | |
| "step": 5030 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "learning_rate": 0.001, | |
| "loss": 0.877, | |
| "step": 5040 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "learning_rate": 0.001, | |
| "loss": 0.9055, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.001, | |
| "loss": 0.9148, | |
| "step": 5060 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 0.001, | |
| "loss": 0.7979, | |
| "step": 5070 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 0.001, | |
| "loss": 0.8536, | |
| "step": 5080 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 0.001, | |
| "loss": 0.7932, | |
| "step": 5090 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 0.001, | |
| "loss": 0.8313, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 0.001, | |
| "loss": 0.8412, | |
| "step": 5110 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 0.001, | |
| "loss": 0.8166, | |
| "step": 5120 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 0.001, | |
| "loss": 0.812, | |
| "step": 5130 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 0.001, | |
| "loss": 0.7519, | |
| "step": 5140 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "learning_rate": 0.001, | |
| "loss": 0.8299, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "learning_rate": 0.001, | |
| "loss": 0.8467, | |
| "step": 5160 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 0.001, | |
| "loss": 0.8133, | |
| "step": 5170 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 0.001, | |
| "loss": 0.826, | |
| "step": 5180 | |
| }, | |
| { | |
| "epoch": 5.13, | |
| "learning_rate": 0.001, | |
| "loss": 0.8168, | |
| "step": 5190 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "learning_rate": 0.001, | |
| "loss": 0.8453, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23509, | |
| 14648, | |
| 9751, | |
| 6804 | |
| ], | |
| "eval_loss": 0.5316904187202454, | |
| "eval_precisions": [ | |
| 56.44553290595212, | |
| 38.359608233383966, | |
| 28.075782442198612, | |
| 21.72691276025035 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 351.3653, | |
| "eval_samples_per_second": 9.856, | |
| "eval_score": 33.90067982700938, | |
| "eval_steps_per_second": 9.856, | |
| "eval_sys_len": 41649, | |
| "eval_totals": [ | |
| 41649, | |
| 38186, | |
| 34731, | |
| 31316 | |
| ], | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 5.15, | |
| "learning_rate": 0.001, | |
| "loss": 0.7617, | |
| "step": 5210 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 0.001, | |
| "loss": 0.8376, | |
| "step": 5220 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "learning_rate": 0.001, | |
| "loss": 0.8217, | |
| "step": 5230 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "learning_rate": 0.001, | |
| "loss": 0.8061, | |
| "step": 5240 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "learning_rate": 0.001, | |
| "loss": 0.8295, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "learning_rate": 0.001, | |
| "loss": 0.8573, | |
| "step": 5260 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "learning_rate": 0.001, | |
| "loss": 0.8613, | |
| "step": 5270 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 0.001, | |
| "loss": 0.8865, | |
| "step": 5280 | |
| }, | |
| { | |
| "epoch": 5.23, | |
| "learning_rate": 0.001, | |
| "loss": 0.8281, | |
| "step": 5290 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 0.001, | |
| "loss": 0.8184, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 5.25, | |
| "learning_rate": 0.001, | |
| "loss": 0.8123, | |
| "step": 5310 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "learning_rate": 0.001, | |
| "loss": 0.8018, | |
| "step": 5320 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "learning_rate": 0.001, | |
| "loss": 0.8409, | |
| "step": 5330 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "learning_rate": 0.001, | |
| "loss": 0.7875, | |
| "step": 5340 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "learning_rate": 0.001, | |
| "loss": 0.8604, | |
| "step": 5350 | |
| }, | |
| { | |
| "epoch": 5.3, | |
| "learning_rate": 0.001, | |
| "loss": 0.8453, | |
| "step": 5360 | |
| }, | |
| { | |
| "epoch": 5.31, | |
| "learning_rate": 0.001, | |
| "loss": 0.8294, | |
| "step": 5370 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "learning_rate": 0.001, | |
| "loss": 0.8425, | |
| "step": 5380 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 0.001, | |
| "loss": 0.8778, | |
| "step": 5390 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "learning_rate": 0.001, | |
| "loss": 0.82, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23012, | |
| 14249, | |
| 9439, | |
| 6467 | |
| ], | |
| "eval_loss": 0.5350695848464966, | |
| "eval_precisions": [ | |
| 45.35635446231473, | |
| 30.1419414887991, | |
| 21.54137569035556, | |
| 16.00425658285488 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 567.2966, | |
| "eval_samples_per_second": 6.104, | |
| "eval_score": 26.20170640056699, | |
| "eval_steps_per_second": 6.104, | |
| "eval_sys_len": 50736, | |
| "eval_totals": [ | |
| 50736, | |
| 47273, | |
| 43818, | |
| 40408 | |
| ], | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "learning_rate": 0.001, | |
| "loss": 0.8829, | |
| "step": 5410 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "learning_rate": 0.001, | |
| "loss": 0.787, | |
| "step": 5420 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "learning_rate": 0.001, | |
| "loss": 0.8567, | |
| "step": 5430 | |
| }, | |
| { | |
| "epoch": 5.37, | |
| "learning_rate": 0.001, | |
| "loss": 0.8524, | |
| "step": 5440 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 0.001, | |
| "loss": 0.8345, | |
| "step": 5450 | |
| }, | |
| { | |
| "epoch": 5.39, | |
| "learning_rate": 0.001, | |
| "loss": 0.8319, | |
| "step": 5460 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "learning_rate": 0.001, | |
| "loss": 0.8393, | |
| "step": 5470 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "learning_rate": 0.001, | |
| "loss": 0.867, | |
| "step": 5480 | |
| }, | |
| { | |
| "epoch": 5.42, | |
| "learning_rate": 0.001, | |
| "loss": 0.8375, | |
| "step": 5490 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 0.001, | |
| "loss": 0.838, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "learning_rate": 0.001, | |
| "loss": 0.8331, | |
| "step": 5510 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 0.001, | |
| "loss": 0.8475, | |
| "step": 5520 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "learning_rate": 0.001, | |
| "loss": 0.867, | |
| "step": 5530 | |
| }, | |
| { | |
| "epoch": 5.47, | |
| "learning_rate": 0.001, | |
| "loss": 0.8039, | |
| "step": 5540 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "learning_rate": 0.001, | |
| "loss": 0.8637, | |
| "step": 5550 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "learning_rate": 0.001, | |
| "loss": 0.8048, | |
| "step": 5560 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "learning_rate": 0.001, | |
| "loss": 0.7821, | |
| "step": 5570 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "learning_rate": 0.001, | |
| "loss": 0.8345, | |
| "step": 5580 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "learning_rate": 0.001, | |
| "loss": 0.836, | |
| "step": 5590 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 0.001, | |
| "loss": 0.8249, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 24069, | |
| 15065, | |
| 10055, | |
| 6996 | |
| ], | |
| "eval_loss": 0.5313393473625183, | |
| "eval_precisions": [ | |
| 57.161517087420144, | |
| 38.98405962115723, | |
| 28.57427036858109, | |
| 22.013152512507475 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 360.3262, | |
| "eval_samples_per_second": 9.611, | |
| "eval_score": 34.4082076392569, | |
| "eval_steps_per_second": 9.611, | |
| "eval_sys_len": 42107, | |
| "eval_totals": [ | |
| 42107, | |
| 38644, | |
| 35189, | |
| 31781 | |
| ], | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 0.001, | |
| "loss": 0.7977, | |
| "step": 5610 | |
| }, | |
| { | |
| "epoch": 5.55, | |
| "learning_rate": 0.001, | |
| "loss": 0.8259, | |
| "step": 5620 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 0.001, | |
| "loss": 0.861, | |
| "step": 5630 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "learning_rate": 0.001, | |
| "loss": 0.83, | |
| "step": 5640 | |
| }, | |
| { | |
| "epoch": 5.58, | |
| "learning_rate": 0.001, | |
| "loss": 0.7905, | |
| "step": 5650 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "learning_rate": 0.001, | |
| "loss": 0.8341, | |
| "step": 5660 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 0.001, | |
| "loss": 0.8224, | |
| "step": 5670 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 0.001, | |
| "loss": 0.8408, | |
| "step": 5680 | |
| }, | |
| { | |
| "epoch": 5.62, | |
| "learning_rate": 0.001, | |
| "loss": 0.8591, | |
| "step": 5690 | |
| }, | |
| { | |
| "epoch": 5.63, | |
| "learning_rate": 0.001, | |
| "loss": 0.8173, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 5.64, | |
| "learning_rate": 0.001, | |
| "loss": 0.791, | |
| "step": 5710 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 0.001, | |
| "loss": 0.8623, | |
| "step": 5720 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 0.001, | |
| "loss": 0.8324, | |
| "step": 5730 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "learning_rate": 0.001, | |
| "loss": 0.868, | |
| "step": 5740 | |
| }, | |
| { | |
| "epoch": 5.68, | |
| "learning_rate": 0.001, | |
| "loss": 0.8237, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 0.001, | |
| "loss": 0.8018, | |
| "step": 5760 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 0.001, | |
| "loss": 0.8745, | |
| "step": 5770 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "learning_rate": 0.001, | |
| "loss": 0.8817, | |
| "step": 5780 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "learning_rate": 0.001, | |
| "loss": 0.8403, | |
| "step": 5790 | |
| }, | |
| { | |
| "epoch": 5.73, | |
| "learning_rate": 0.001, | |
| "loss": 0.8985, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 5.73, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23804, | |
| 14873, | |
| 9870, | |
| 6815 | |
| ], | |
| "eval_loss": 0.5328965783119202, | |
| "eval_precisions": [ | |
| 56.88069010012187, | |
| 38.74589694159329, | |
| 28.2581310123683, | |
| 21.6253093863045 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 352.1325, | |
| "eval_samples_per_second": 9.834, | |
| "eval_score": 34.066230834274144, | |
| "eval_steps_per_second": 9.834, | |
| "eval_sys_len": 41849, | |
| "eval_totals": [ | |
| 41849, | |
| 38386, | |
| 34928, | |
| 31514 | |
| ], | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "learning_rate": 0.001, | |
| "loss": 0.8409, | |
| "step": 5810 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "learning_rate": 0.001, | |
| "loss": 0.8561, | |
| "step": 5820 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 0.001, | |
| "loss": 0.8231, | |
| "step": 5830 | |
| }, | |
| { | |
| "epoch": 5.77, | |
| "learning_rate": 0.001, | |
| "loss": 0.9071, | |
| "step": 5840 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "learning_rate": 0.001, | |
| "loss": 0.8143, | |
| "step": 5850 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 0.001, | |
| "loss": 0.8396, | |
| "step": 5860 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "learning_rate": 0.001, | |
| "loss": 0.7873, | |
| "step": 5870 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 0.001, | |
| "loss": 0.8594, | |
| "step": 5880 | |
| }, | |
| { | |
| "epoch": 5.82, | |
| "learning_rate": 0.001, | |
| "loss": 0.7969, | |
| "step": 5890 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "learning_rate": 0.001, | |
| "loss": 0.8291, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "learning_rate": 0.001, | |
| "loss": 0.8373, | |
| "step": 5910 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 0.001, | |
| "loss": 0.8049, | |
| "step": 5920 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 0.001, | |
| "loss": 0.8478, | |
| "step": 5930 | |
| }, | |
| { | |
| "epoch": 5.87, | |
| "learning_rate": 0.001, | |
| "loss": 0.8803, | |
| "step": 5940 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "learning_rate": 0.001, | |
| "loss": 0.8188, | |
| "step": 5950 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "learning_rate": 0.001, | |
| "loss": 0.8392, | |
| "step": 5960 | |
| }, | |
| { | |
| "epoch": 5.9, | |
| "learning_rate": 0.001, | |
| "loss": 0.8316, | |
| "step": 5970 | |
| }, | |
| { | |
| "epoch": 5.91, | |
| "learning_rate": 0.001, | |
| "loss": 0.8086, | |
| "step": 5980 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 0.001, | |
| "loss": 0.8363, | |
| "step": 5990 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 0.001, | |
| "loss": 0.8258, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 24039, | |
| 15198, | |
| 10262, | |
| 7183 | |
| ], | |
| "eval_loss": 0.5226385593414307, | |
| "eval_precisions": [ | |
| 47.00074297110233, | |
| 31.87299456829478, | |
| 23.20249615628109, | |
| 17.5993531631303 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 521.1715, | |
| "eval_samples_per_second": 6.645, | |
| "eval_score": 27.966605219472754, | |
| "eval_steps_per_second": 6.645, | |
| "eval_sys_len": 51146, | |
| "eval_totals": [ | |
| 51146, | |
| 47683, | |
| 44228, | |
| 40814 | |
| ], | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "learning_rate": 0.001, | |
| "loss": 0.8284, | |
| "step": 6010 | |
| }, | |
| { | |
| "epoch": 5.95, | |
| "learning_rate": 0.001, | |
| "loss": 0.808, | |
| "step": 6020 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "learning_rate": 0.001, | |
| "loss": 0.8575, | |
| "step": 6030 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "learning_rate": 0.001, | |
| "loss": 0.8036, | |
| "step": 6040 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "learning_rate": 0.001, | |
| "loss": 0.823, | |
| "step": 6050 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "learning_rate": 0.001, | |
| "loss": 0.8291, | |
| "step": 6060 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 0.001, | |
| "loss": 0.8116, | |
| "step": 6070 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 0.001, | |
| "loss": 0.7767, | |
| "step": 6080 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 0.001, | |
| "loss": 0.7698, | |
| "step": 6090 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 0.001, | |
| "loss": 0.8037, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 0.001, | |
| "loss": 0.7878, | |
| "step": 6110 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 0.001, | |
| "loss": 0.7725, | |
| "step": 6120 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 0.001, | |
| "loss": 0.813, | |
| "step": 6130 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 0.001, | |
| "loss": 0.7674, | |
| "step": 6140 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 0.001, | |
| "loss": 0.7548, | |
| "step": 6150 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "learning_rate": 0.001, | |
| "loss": 0.7822, | |
| "step": 6160 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "learning_rate": 0.001, | |
| "loss": 0.767, | |
| "step": 6170 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 0.001, | |
| "loss": 0.7692, | |
| "step": 6180 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 0.001, | |
| "loss": 0.7928, | |
| "step": 6190 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "learning_rate": 0.001, | |
| "loss": 0.7583, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23731, | |
| 14954, | |
| 10071, | |
| 7011 | |
| ], | |
| "eval_loss": 0.5309858322143555, | |
| "eval_precisions": [ | |
| 55.32346427322532, | |
| 37.92351389734226, | |
| 27.992884342774552, | |
| 21.529249193919853 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 380.8194, | |
| "eval_samples_per_second": 9.094, | |
| "eval_score": 33.53309336901789, | |
| "eval_steps_per_second": 9.094, | |
| "eval_sys_len": 42895, | |
| "eval_totals": [ | |
| 42895, | |
| 39432, | |
| 35977, | |
| 32565 | |
| ], | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 6.14, | |
| "learning_rate": 0.001, | |
| "loss": 0.7889, | |
| "step": 6210 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 0.001, | |
| "loss": 0.7824, | |
| "step": 6220 | |
| }, | |
| { | |
| "epoch": 6.16, | |
| "learning_rate": 0.001, | |
| "loss": 0.7889, | |
| "step": 6230 | |
| }, | |
| { | |
| "epoch": 6.16, | |
| "learning_rate": 0.001, | |
| "loss": 0.7915, | |
| "step": 6240 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "learning_rate": 0.001, | |
| "loss": 0.8305, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 6.18, | |
| "learning_rate": 0.001, | |
| "loss": 0.7535, | |
| "step": 6260 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "learning_rate": 0.001, | |
| "loss": 0.7537, | |
| "step": 6270 | |
| }, | |
| { | |
| "epoch": 6.2, | |
| "learning_rate": 0.001, | |
| "loss": 0.7631, | |
| "step": 6280 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "learning_rate": 0.001, | |
| "loss": 0.8173, | |
| "step": 6290 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "learning_rate": 0.001, | |
| "loss": 0.7826, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "learning_rate": 0.001, | |
| "loss": 0.7687, | |
| "step": 6310 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 0.001, | |
| "loss": 0.7918, | |
| "step": 6320 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 0.001, | |
| "loss": 0.8001, | |
| "step": 6330 | |
| }, | |
| { | |
| "epoch": 6.26, | |
| "learning_rate": 0.001, | |
| "loss": 0.7871, | |
| "step": 6340 | |
| }, | |
| { | |
| "epoch": 6.27, | |
| "learning_rate": 0.001, | |
| "loss": 0.785, | |
| "step": 6350 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "learning_rate": 0.001, | |
| "loss": 0.7807, | |
| "step": 6360 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 0.001, | |
| "loss": 0.7969, | |
| "step": 6370 | |
| }, | |
| { | |
| "epoch": 6.3, | |
| "learning_rate": 0.001, | |
| "loss": 0.8125, | |
| "step": 6380 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "learning_rate": 0.001, | |
| "loss": 0.7843, | |
| "step": 6390 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "learning_rate": 0.001, | |
| "loss": 0.7758, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 24184, | |
| 15282, | |
| 10297, | |
| 7193 | |
| ], | |
| "eval_loss": 0.5329195261001587, | |
| "eval_precisions": [ | |
| 57.96323370802675, | |
| 39.94249869315212, | |
| 29.58482976583824, | |
| 22.912021405364083 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 342.9579, | |
| "eval_samples_per_second": 10.097, | |
| "eval_score": 35.39403526330641, | |
| "eval_steps_per_second": 10.097, | |
| "eval_sys_len": 41723, | |
| "eval_totals": [ | |
| 41723, | |
| 38260, | |
| 34805, | |
| 31394 | |
| ], | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "learning_rate": 0.001, | |
| "loss": 0.8018, | |
| "step": 6410 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "learning_rate": 0.001, | |
| "loss": 0.7649, | |
| "step": 6420 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "learning_rate": 0.001, | |
| "loss": 0.7967, | |
| "step": 6430 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "learning_rate": 0.001, | |
| "loss": 0.7843, | |
| "step": 6440 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "learning_rate": 0.001, | |
| "loss": 0.7841, | |
| "step": 6450 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "learning_rate": 0.001, | |
| "loss": 0.7514, | |
| "step": 6460 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "learning_rate": 0.001, | |
| "loss": 0.8351, | |
| "step": 6470 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.001, | |
| "loss": 0.8028, | |
| "step": 6480 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 0.001, | |
| "loss": 0.7477, | |
| "step": 6490 | |
| }, | |
| { | |
| "epoch": 6.42, | |
| "learning_rate": 0.001, | |
| "loss": 0.8054, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "learning_rate": 0.001, | |
| "loss": 0.8242, | |
| "step": 6510 | |
| }, | |
| { | |
| "epoch": 6.44, | |
| "learning_rate": 0.001, | |
| "loss": 0.7628, | |
| "step": 6520 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 0.001, | |
| "loss": 0.7794, | |
| "step": 6530 | |
| }, | |
| { | |
| "epoch": 6.46, | |
| "learning_rate": 0.001, | |
| "loss": 0.7423, | |
| "step": 6540 | |
| }, | |
| { | |
| "epoch": 6.47, | |
| "learning_rate": 0.001, | |
| "loss": 0.7758, | |
| "step": 6550 | |
| }, | |
| { | |
| "epoch": 6.48, | |
| "learning_rate": 0.001, | |
| "loss": 0.7897, | |
| "step": 6560 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "learning_rate": 0.001, | |
| "loss": 0.7728, | |
| "step": 6570 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "learning_rate": 0.001, | |
| "loss": 0.7827, | |
| "step": 6580 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "learning_rate": 0.001, | |
| "loss": 0.8261, | |
| "step": 6590 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "learning_rate": 0.001, | |
| "loss": 0.8049, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23829, | |
| 15082, | |
| 10178, | |
| 7117 | |
| ], | |
| "eval_loss": 0.5237361788749695, | |
| "eval_precisions": [ | |
| 59.904972597918444, | |
| 41.53104777640094, | |
| 30.973828362751064, | |
| 24.170487349295296 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 306.2297, | |
| "eval_samples_per_second": 11.309, | |
| "eval_score": 36.942732402749876, | |
| "eval_steps_per_second": 11.309, | |
| "eval_sys_len": 39778, | |
| "eval_totals": [ | |
| 39778, | |
| 36315, | |
| 32860, | |
| 29445 | |
| ], | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "learning_rate": 0.001, | |
| "loss": 0.8156, | |
| "step": 6610 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "learning_rate": 0.001, | |
| "loss": 0.8011, | |
| "step": 6620 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "learning_rate": 0.001, | |
| "loss": 0.7951, | |
| "step": 6630 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 0.001, | |
| "loss": 0.8282, | |
| "step": 6640 | |
| }, | |
| { | |
| "epoch": 6.57, | |
| "learning_rate": 0.001, | |
| "loss": 0.7806, | |
| "step": 6650 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "learning_rate": 0.001, | |
| "loss": 0.7797, | |
| "step": 6660 | |
| }, | |
| { | |
| "epoch": 6.59, | |
| "learning_rate": 0.001, | |
| "loss": 0.7625, | |
| "step": 6670 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "learning_rate": 0.001, | |
| "loss": 0.7496, | |
| "step": 6680 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "learning_rate": 0.001, | |
| "loss": 0.7774, | |
| "step": 6690 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 0.001, | |
| "loss": 0.7788, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "learning_rate": 0.001, | |
| "loss": 0.759, | |
| "step": 6710 | |
| }, | |
| { | |
| "epoch": 6.64, | |
| "learning_rate": 0.001, | |
| "loss": 0.8003, | |
| "step": 6720 | |
| }, | |
| { | |
| "epoch": 6.65, | |
| "learning_rate": 0.001, | |
| "loss": 0.7566, | |
| "step": 6730 | |
| }, | |
| { | |
| "epoch": 6.66, | |
| "learning_rate": 0.001, | |
| "loss": 0.7698, | |
| "step": 6740 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 0.001, | |
| "loss": 0.8026, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "learning_rate": 0.001, | |
| "loss": 0.8094, | |
| "step": 6760 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "learning_rate": 0.001, | |
| "loss": 0.8155, | |
| "step": 6770 | |
| }, | |
| { | |
| "epoch": 6.7, | |
| "learning_rate": 0.001, | |
| "loss": 0.8205, | |
| "step": 6780 | |
| }, | |
| { | |
| "epoch": 6.71, | |
| "learning_rate": 0.001, | |
| "loss": 0.8219, | |
| "step": 6790 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "learning_rate": 0.001, | |
| "loss": 0.7997, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 24844, | |
| 15901, | |
| 10792, | |
| 7563 | |
| ], | |
| "eval_loss": 0.5219168066978455, | |
| "eval_precisions": [ | |
| 58.95306345213801, | |
| 41.1101631376199, | |
| 30.64168086314594, | |
| 23.78301886792453 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 345.1356, | |
| "eval_samples_per_second": 10.034, | |
| "eval_score": 36.455141296935956, | |
| "eval_steps_per_second": 10.034, | |
| "eval_sys_len": 42142, | |
| "eval_totals": [ | |
| 42142, | |
| 38679, | |
| 35220, | |
| 31800 | |
| ], | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 6.73, | |
| "learning_rate": 0.001, | |
| "loss": 0.7818, | |
| "step": 6810 | |
| }, | |
| { | |
| "epoch": 6.74, | |
| "learning_rate": 0.001, | |
| "loss": 0.7812, | |
| "step": 6820 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "learning_rate": 0.001, | |
| "loss": 0.8151, | |
| "step": 6830 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "learning_rate": 0.001, | |
| "loss": 0.8026, | |
| "step": 6840 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "learning_rate": 0.001, | |
| "loss": 0.7498, | |
| "step": 6850 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 0.001, | |
| "loss": 0.7914, | |
| "step": 6860 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "learning_rate": 0.001, | |
| "loss": 0.798, | |
| "step": 6870 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "learning_rate": 0.001, | |
| "loss": 0.8053, | |
| "step": 6880 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "learning_rate": 0.001, | |
| "loss": 0.8059, | |
| "step": 6890 | |
| }, | |
| { | |
| "epoch": 6.82, | |
| "learning_rate": 0.001, | |
| "loss": 0.7825, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 0.001, | |
| "loss": 0.7939, | |
| "step": 6910 | |
| }, | |
| { | |
| "epoch": 6.84, | |
| "learning_rate": 0.001, | |
| "loss": 0.8363, | |
| "step": 6920 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "learning_rate": 0.001, | |
| "loss": 0.8025, | |
| "step": 6930 | |
| }, | |
| { | |
| "epoch": 6.86, | |
| "learning_rate": 0.001, | |
| "loss": 0.8241, | |
| "step": 6940 | |
| }, | |
| { | |
| "epoch": 6.87, | |
| "learning_rate": 0.001, | |
| "loss": 0.7841, | |
| "step": 6950 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "learning_rate": 0.001, | |
| "loss": 0.7926, | |
| "step": 6960 | |
| }, | |
| { | |
| "epoch": 6.89, | |
| "learning_rate": 0.001, | |
| "loss": 0.7734, | |
| "step": 6970 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "learning_rate": 0.001, | |
| "loss": 0.8046, | |
| "step": 6980 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "learning_rate": 0.001, | |
| "loss": 0.7715, | |
| "step": 6990 | |
| }, | |
| { | |
| "epoch": 6.92, | |
| "learning_rate": 0.001, | |
| "loss": 0.8206, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 6.92, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23850, | |
| 15135, | |
| 10232, | |
| 7207 | |
| ], | |
| "eval_loss": 0.5204957723617554, | |
| "eval_precisions": [ | |
| 54.366409081583804, | |
| 37.45730832054645, | |
| 27.692224417440254, | |
| 21.495466475781438 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 400.5458, | |
| "eval_samples_per_second": 8.646, | |
| "eval_score": 33.18126834458568, | |
| "eval_steps_per_second": 8.646, | |
| "eval_sys_len": 43869, | |
| "eval_totals": [ | |
| 43869, | |
| 40406, | |
| 36949, | |
| 33528 | |
| ], | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "learning_rate": 0.001, | |
| "loss": 0.8061, | |
| "step": 7010 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "learning_rate": 0.001, | |
| "loss": 0.7957, | |
| "step": 7020 | |
| }, | |
| { | |
| "epoch": 6.95, | |
| "learning_rate": 0.001, | |
| "loss": 0.759, | |
| "step": 7030 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "learning_rate": 0.001, | |
| "loss": 0.8067, | |
| "step": 7040 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "learning_rate": 0.001, | |
| "loss": 0.8093, | |
| "step": 7050 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "learning_rate": 0.001, | |
| "loss": 0.8292, | |
| "step": 7060 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "learning_rate": 0.001, | |
| "loss": 0.817, | |
| "step": 7070 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "learning_rate": 0.001, | |
| "loss": 0.795, | |
| "step": 7080 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 0.001, | |
| "loss": 0.7871, | |
| "step": 7090 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 0.001, | |
| "loss": 0.7257, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 0.001, | |
| "loss": 0.7181, | |
| "step": 7110 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 0.001, | |
| "loss": 0.7531, | |
| "step": 7120 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 0.001, | |
| "loss": 0.7277, | |
| "step": 7130 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 0.001, | |
| "loss": 0.7455, | |
| "step": 7140 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 0.001, | |
| "loss": 0.7343, | |
| "step": 7150 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "learning_rate": 0.001, | |
| "loss": 0.7322, | |
| "step": 7160 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 0.001, | |
| "loss": 0.7532, | |
| "step": 7170 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "learning_rate": 0.001, | |
| "loss": 0.729, | |
| "step": 7180 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 0.001, | |
| "loss": 0.7284, | |
| "step": 7190 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "learning_rate": 0.001, | |
| "loss": 0.762, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 23056, | |
| 14413, | |
| 9691, | |
| 6788 | |
| ], | |
| "eval_loss": 0.5327216982841492, | |
| "eval_precisions": [ | |
| 53.40993328391401, | |
| 36.30021407883138, | |
| 26.735268152725666, | |
| 20.676840598251545 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 416.5042, | |
| "eval_samples_per_second": 8.314, | |
| "eval_score": 32.17547644018389, | |
| "eval_steps_per_second": 8.314, | |
| "eval_sys_len": 43168, | |
| "eval_totals": [ | |
| 43168, | |
| 39705, | |
| 36248, | |
| 32829 | |
| ], | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "learning_rate": 0.001, | |
| "loss": 0.7273, | |
| "step": 7210 | |
| }, | |
| { | |
| "epoch": 7.13, | |
| "learning_rate": 0.001, | |
| "loss": 0.7223, | |
| "step": 7220 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 0.001, | |
| "loss": 0.7728, | |
| "step": 7230 | |
| }, | |
| { | |
| "epoch": 7.15, | |
| "learning_rate": 0.001, | |
| "loss": 0.739, | |
| "step": 7240 | |
| }, | |
| { | |
| "epoch": 7.16, | |
| "learning_rate": 0.001, | |
| "loss": 0.7332, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "learning_rate": 0.001, | |
| "loss": 0.7259, | |
| "step": 7260 | |
| }, | |
| { | |
| "epoch": 7.18, | |
| "learning_rate": 0.001, | |
| "loss": 0.7399, | |
| "step": 7270 | |
| }, | |
| { | |
| "epoch": 7.19, | |
| "learning_rate": 0.001, | |
| "loss": 0.7036, | |
| "step": 7280 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 0.001, | |
| "loss": 0.7782, | |
| "step": 7290 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "learning_rate": 0.001, | |
| "loss": 0.7463, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 7.22, | |
| "learning_rate": 0.001, | |
| "loss": 0.7454, | |
| "step": 7310 | |
| }, | |
| { | |
| "epoch": 7.23, | |
| "learning_rate": 0.001, | |
| "loss": 0.742, | |
| "step": 7320 | |
| }, | |
| { | |
| "epoch": 7.24, | |
| "learning_rate": 0.001, | |
| "loss": 0.7275, | |
| "step": 7330 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "learning_rate": 0.001, | |
| "loss": 0.7719, | |
| "step": 7340 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "learning_rate": 0.001, | |
| "loss": 0.7818, | |
| "step": 7350 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "learning_rate": 0.001, | |
| "loss": 0.777, | |
| "step": 7360 | |
| }, | |
| { | |
| "epoch": 7.28, | |
| "learning_rate": 0.001, | |
| "loss": 0.7185, | |
| "step": 7370 | |
| }, | |
| { | |
| "epoch": 7.29, | |
| "learning_rate": 0.001, | |
| "loss": 0.758, | |
| "step": 7380 | |
| }, | |
| { | |
| "epoch": 7.3, | |
| "learning_rate": 0.001, | |
| "loss": 0.7341, | |
| "step": 7390 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "learning_rate": 0.001, | |
| "loss": 0.7819, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 24477, | |
| 15690, | |
| 10775, | |
| 7636 | |
| ], | |
| "eval_loss": 0.5353978276252747, | |
| "eval_precisions": [ | |
| 54.04027023446814, | |
| 37.50806817910162, | |
| 28.078907593683223, | |
| 21.844604645840484 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 408.1893, | |
| "eval_samples_per_second": 8.484, | |
| "eval_score": 33.391940001171015, | |
| "eval_steps_per_second": 8.484, | |
| "eval_sys_len": 45294, | |
| "eval_totals": [ | |
| 45294, | |
| 41831, | |
| 38374, | |
| 34956 | |
| ], | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 7.32, | |
| "learning_rate": 0.001, | |
| "loss": 0.7884, | |
| "step": 7410 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "learning_rate": 0.001, | |
| "loss": 0.7573, | |
| "step": 7420 | |
| }, | |
| { | |
| "epoch": 7.34, | |
| "learning_rate": 0.001, | |
| "loss": 0.7994, | |
| "step": 7430 | |
| }, | |
| { | |
| "epoch": 7.35, | |
| "learning_rate": 0.001, | |
| "loss": 0.7525, | |
| "step": 7440 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "learning_rate": 0.001, | |
| "loss": 0.773, | |
| "step": 7450 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 0.001, | |
| "loss": 0.755, | |
| "step": 7460 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "learning_rate": 0.001, | |
| "loss": 0.7779, | |
| "step": 7470 | |
| }, | |
| { | |
| "epoch": 7.39, | |
| "learning_rate": 0.001, | |
| "loss": 0.7441, | |
| "step": 7480 | |
| }, | |
| { | |
| "epoch": 7.4, | |
| "learning_rate": 0.001, | |
| "loss": 0.7377, | |
| "step": 7490 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "learning_rate": 0.001, | |
| "loss": 0.7757, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 0.001, | |
| "loss": 0.7742, | |
| "step": 7510 | |
| }, | |
| { | |
| "epoch": 7.43, | |
| "learning_rate": 0.001, | |
| "loss": 0.7297, | |
| "step": 7520 | |
| }, | |
| { | |
| "epoch": 7.44, | |
| "learning_rate": 0.001, | |
| "loss": 0.7489, | |
| "step": 7530 | |
| }, | |
| { | |
| "epoch": 7.45, | |
| "learning_rate": 0.001, | |
| "loss": 0.7525, | |
| "step": 7540 | |
| }, | |
| { | |
| "epoch": 7.46, | |
| "learning_rate": 0.001, | |
| "loss": 0.7681, | |
| "step": 7550 | |
| }, | |
| { | |
| "epoch": 7.47, | |
| "learning_rate": 0.001, | |
| "loss": 0.7535, | |
| "step": 7560 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 0.001, | |
| "loss": 0.7314, | |
| "step": 7570 | |
| }, | |
| { | |
| "epoch": 7.49, | |
| "learning_rate": 0.001, | |
| "loss": 0.7592, | |
| "step": 7580 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "learning_rate": 0.001, | |
| "loss": 0.7432, | |
| "step": 7590 | |
| }, | |
| { | |
| "epoch": 7.51, | |
| "learning_rate": 0.001, | |
| "loss": 0.7491, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 7.51, | |
| "eval_bp": 1.0, | |
| "eval_counts": [ | |
| 24779, | |
| 15986, | |
| 10939, | |
| 7748 | |
| ], | |
| "eval_loss": 0.5190737247467041, | |
| "eval_precisions": [ | |
| 60.838714429521964, | |
| 42.89701067997638, | |
| 32.35337612019757, | |
| 25.48851898151194 | |
| ], | |
| "eval_ref_len": 39265, | |
| "eval_runtime": 314.2121, | |
| "eval_samples_per_second": 11.021, | |
| "eval_score": 38.30167936745805, | |
| "eval_steps_per_second": 11.021, | |
| "eval_sys_len": 40729, | |
| "eval_totals": [ | |
| 40729, | |
| 37266, | |
| 33811, | |
| 30398 | |
| ], | |
| "step": 7600 | |
| } | |
| ], | |
| "max_steps": 253000, | |
| "num_train_epochs": 250, | |
| "total_flos": 1.1343638633720448e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |