{ "best_metric": 0.011560924351215363, "best_model_checkpoint": "/media/metricspace/Data1/LLamaT5/TLAN-T5/checkpoint-4078", "epoch": 14.964143426294822, "global_step": 4695, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.03, "learning_rate": 4.989350372736955e-05, "loss": 39.1826, "step": 10 }, { "epoch": 0.06, "learning_rate": 4.9787007454739086e-05, "loss": 29.1838, "step": 20 }, { "epoch": 0.1, "learning_rate": 4.968051118210863e-05, "loss": 20.9138, "step": 30 }, { "epoch": 0.13, "learning_rate": 4.957401490947817e-05, "loss": 8.3513, "step": 40 }, { "epoch": 0.16, "learning_rate": 4.9467518636847715e-05, "loss": 4.4736, "step": 50 }, { "epoch": 0.19, "learning_rate": 4.936102236421725e-05, "loss": 3.3849, "step": 60 }, { "epoch": 0.22, "learning_rate": 4.925452609158679e-05, "loss": 2.2067, "step": 70 }, { "epoch": 0.25, "learning_rate": 4.9148029818956336e-05, "loss": 0.5818, "step": 80 }, { "epoch": 0.29, "learning_rate": 4.904153354632588e-05, "loss": 0.1404, "step": 90 }, { "epoch": 0.32, "learning_rate": 4.893503727369542e-05, "loss": 0.0717, "step": 100 }, { "epoch": 0.35, "learning_rate": 4.8828541001064965e-05, "loss": 0.0578, "step": 110 }, { "epoch": 0.38, "learning_rate": 4.872204472843451e-05, "loss": 0.0405, "step": 120 }, { "epoch": 0.41, "learning_rate": 4.861554845580405e-05, "loss": 0.0468, "step": 130 }, { "epoch": 0.45, "learning_rate": 4.8509052183173594e-05, "loss": 0.0277, "step": 140 }, { "epoch": 0.48, "learning_rate": 4.840255591054313e-05, "loss": 0.0327, "step": 150 }, { "epoch": 0.51, "learning_rate": 4.829605963791268e-05, "loss": 0.0297, "step": 160 }, { "epoch": 0.54, "learning_rate": 4.818956336528222e-05, "loss": 0.0242, "step": 170 }, { "epoch": 0.57, "learning_rate": 4.8083067092651754e-05, "loss": 0.0453, "step": 180 }, { "epoch": 0.61, "learning_rate": 4.79765708200213e-05, "loss": 0.0239, "step": 190 }, { "epoch": 0.64, "learning_rate": 4.7870074547390844e-05, "loss": 0.0451, "step": 200 }, { "epoch": 0.67, "learning_rate": 4.776357827476038e-05, "loss": 0.0186, "step": 210 }, { "epoch": 0.7, "learning_rate": 4.765708200212993e-05, "loss": 0.0367, "step": 220 }, { "epoch": 0.73, "learning_rate": 4.755058572949947e-05, "loss": 0.0152, "step": 230 }, { "epoch": 0.76, "learning_rate": 4.744408945686901e-05, "loss": 0.0207, "step": 240 }, { "epoch": 0.8, "learning_rate": 4.7337593184238556e-05, "loss": 0.016, "step": 250 }, { "epoch": 0.83, "learning_rate": 4.7231096911608095e-05, "loss": 0.0235, "step": 260 }, { "epoch": 0.86, "learning_rate": 4.712460063897764e-05, "loss": 0.0196, "step": 270 }, { "epoch": 0.89, "learning_rate": 4.7018104366347185e-05, "loss": 0.0216, "step": 280 }, { "epoch": 0.92, "learning_rate": 4.691160809371672e-05, "loss": 0.0278, "step": 290 }, { "epoch": 0.96, "learning_rate": 4.680511182108626e-05, "loss": 0.0297, "step": 300 }, { "epoch": 0.99, "learning_rate": 4.669861554845581e-05, "loss": 0.0215, "step": 310 }, { "epoch": 1.0, "eval_gen_len": 19.0, "eval_loss": 0.015359184704720974, "eval_rouge1": 31.9011, "eval_rouge2": 22.9103, "eval_rougeL": 31.716, "eval_rougeLsum": 31.8299, "eval_runtime": 717.4413, "eval_samples_per_second": 0.389, "eval_steps_per_second": 0.195, "step": 313 }, { "epoch": 1.02, "learning_rate": 4.6592119275825345e-05, "loss": 0.0171, "step": 320 }, { "epoch": 1.05, "learning_rate": 4.648562300319489e-05, "loss": 0.0242, "step": 330 }, { "epoch": 1.08, "learning_rate": 4.637912673056443e-05, "loss": 0.0216, "step": 340 }, { "epoch": 1.12, "learning_rate": 4.6272630457933974e-05, "loss": 0.0135, "step": 350 }, { "epoch": 1.15, "learning_rate": 4.616613418530352e-05, "loss": 0.0183, "step": 360 }, { "epoch": 1.18, "learning_rate": 4.605963791267306e-05, "loss": 0.0234, "step": 370 }, { "epoch": 1.21, "learning_rate": 4.59531416400426e-05, "loss": 0.03, "step": 380 }, { "epoch": 1.24, "learning_rate": 4.584664536741215e-05, "loss": 0.021, "step": 390 }, { "epoch": 1.27, "learning_rate": 4.5740149094781686e-05, "loss": 0.0212, "step": 400 }, { "epoch": 1.31, "learning_rate": 4.563365282215123e-05, "loss": 0.0121, "step": 410 }, { "epoch": 1.34, "learning_rate": 4.552715654952077e-05, "loss": 0.0148, "step": 420 }, { "epoch": 1.37, "learning_rate": 4.542066027689031e-05, "loss": 0.0155, "step": 430 }, { "epoch": 1.4, "learning_rate": 4.531416400425985e-05, "loss": 0.0213, "step": 440 }, { "epoch": 1.43, "learning_rate": 4.520766773162939e-05, "loss": 0.0193, "step": 450 }, { "epoch": 1.47, "learning_rate": 4.5101171458998936e-05, "loss": 0.0191, "step": 460 }, { "epoch": 1.5, "learning_rate": 4.499467518636848e-05, "loss": 0.0314, "step": 470 }, { "epoch": 1.53, "learning_rate": 4.488817891373802e-05, "loss": 0.0185, "step": 480 }, { "epoch": 1.56, "learning_rate": 4.4781682641107565e-05, "loss": 0.0152, "step": 490 }, { "epoch": 1.59, "learning_rate": 4.46751863684771e-05, "loss": 0.0361, "step": 500 }, { "epoch": 1.63, "learning_rate": 4.456869009584665e-05, "loss": 0.0189, "step": 510 }, { "epoch": 1.66, "learning_rate": 4.4462193823216194e-05, "loss": 0.0149, "step": 520 }, { "epoch": 1.69, "learning_rate": 4.435569755058573e-05, "loss": 0.0236, "step": 530 }, { "epoch": 1.72, "learning_rate": 4.424920127795527e-05, "loss": 0.0328, "step": 540 }, { "epoch": 1.75, "learning_rate": 4.4142705005324815e-05, "loss": 0.0182, "step": 550 }, { "epoch": 1.78, "learning_rate": 4.4036208732694354e-05, "loss": 0.015, "step": 560 }, { "epoch": 1.82, "learning_rate": 4.39297124600639e-05, "loss": 0.0123, "step": 570 }, { "epoch": 1.85, "learning_rate": 4.3823216187433444e-05, "loss": 0.0175, "step": 580 }, { "epoch": 1.88, "learning_rate": 4.371671991480298e-05, "loss": 0.0164, "step": 590 }, { "epoch": 1.91, "learning_rate": 4.361022364217253e-05, "loss": 0.0156, "step": 600 }, { "epoch": 1.94, "learning_rate": 4.3503727369542066e-05, "loss": 0.0109, "step": 610 }, { "epoch": 1.98, "learning_rate": 4.339723109691161e-05, "loss": 0.0186, "step": 620 }, { "epoch": 2.0, "eval_gen_len": 17.795698924731184, "eval_loss": 0.013456961140036583, "eval_rouge1": 39.398, "eval_rouge2": 31.4448, "eval_rougeL": 39.0922, "eval_rougeLsum": 39.237, "eval_runtime": 716.8614, "eval_samples_per_second": 0.389, "eval_steps_per_second": 0.195, "step": 627 }, { "epoch": 2.01, "learning_rate": 4.3290734824281156e-05, "loss": 0.0119, "step": 630 }, { "epoch": 2.04, "learning_rate": 4.3184238551650695e-05, "loss": 0.0169, "step": 640 }, { "epoch": 2.07, "learning_rate": 4.307774227902024e-05, "loss": 0.0157, "step": 650 }, { "epoch": 2.1, "learning_rate": 4.297124600638978e-05, "loss": 0.0246, "step": 660 }, { "epoch": 2.14, "learning_rate": 4.2864749733759316e-05, "loss": 0.0389, "step": 670 }, { "epoch": 2.17, "learning_rate": 4.275825346112886e-05, "loss": 0.02, "step": 680 }, { "epoch": 2.2, "learning_rate": 4.265175718849841e-05, "loss": 0.0176, "step": 690 }, { "epoch": 2.23, "learning_rate": 4.2545260915867945e-05, "loss": 0.0147, "step": 700 }, { "epoch": 2.26, "learning_rate": 4.243876464323749e-05, "loss": 0.0153, "step": 710 }, { "epoch": 2.29, "learning_rate": 4.233226837060703e-05, "loss": 0.0134, "step": 720 }, { "epoch": 2.33, "learning_rate": 4.2225772097976574e-05, "loss": 0.0119, "step": 730 }, { "epoch": 2.36, "learning_rate": 4.211927582534612e-05, "loss": 0.0256, "step": 740 }, { "epoch": 2.39, "learning_rate": 4.201277955271566e-05, "loss": 0.0196, "step": 750 }, { "epoch": 2.42, "learning_rate": 4.19062832800852e-05, "loss": 0.0149, "step": 760 }, { "epoch": 2.45, "learning_rate": 4.179978700745474e-05, "loss": 0.0204, "step": 770 }, { "epoch": 2.49, "learning_rate": 4.169329073482428e-05, "loss": 0.0182, "step": 780 }, { "epoch": 2.52, "learning_rate": 4.1586794462193824e-05, "loss": 0.0142, "step": 790 }, { "epoch": 2.55, "learning_rate": 4.148029818956336e-05, "loss": 0.0095, "step": 800 }, { "epoch": 2.58, "learning_rate": 4.137380191693291e-05, "loss": 0.0108, "step": 810 }, { "epoch": 2.61, "learning_rate": 4.126730564430245e-05, "loss": 0.0123, "step": 820 }, { "epoch": 2.65, "learning_rate": 4.116080937167199e-05, "loss": 0.0226, "step": 830 }, { "epoch": 2.68, "learning_rate": 4.1054313099041536e-05, "loss": 0.0224, "step": 840 }, { "epoch": 2.71, "learning_rate": 4.094781682641108e-05, "loss": 0.0147, "step": 850 }, { "epoch": 2.74, "learning_rate": 4.084132055378062e-05, "loss": 0.0158, "step": 860 }, { "epoch": 2.77, "learning_rate": 4.0734824281150165e-05, "loss": 0.0118, "step": 870 }, { "epoch": 2.8, "learning_rate": 4.06283280085197e-05, "loss": 0.0272, "step": 880 }, { "epoch": 2.84, "learning_rate": 4.052183173588925e-05, "loss": 0.0148, "step": 890 }, { "epoch": 2.87, "learning_rate": 4.041533546325879e-05, "loss": 0.0314, "step": 900 }, { "epoch": 2.9, "learning_rate": 4.0308839190628325e-05, "loss": 0.016, "step": 910 }, { "epoch": 2.93, "learning_rate": 4.020234291799787e-05, "loss": 0.0146, "step": 920 }, { "epoch": 2.96, "learning_rate": 4.0095846645367415e-05, "loss": 0.0159, "step": 930 }, { "epoch": 3.0, "learning_rate": 3.9989350372736954e-05, "loss": 0.0166, "step": 940 }, { "epoch": 3.0, "eval_gen_len": 16.892473118279568, "eval_loss": 0.01282462291419506, "eval_rouge1": 44.2185, "eval_rouge2": 36.802, "eval_rougeL": 44.0029, "eval_rougeLsum": 44.0496, "eval_runtime": 709.8988, "eval_samples_per_second": 0.393, "eval_steps_per_second": 0.197, "step": 941 }, { "epoch": 3.03, "learning_rate": 3.98828541001065e-05, "loss": 0.011, "step": 950 }, { "epoch": 3.06, "learning_rate": 3.9776357827476044e-05, "loss": 0.0136, "step": 960 }, { "epoch": 3.09, "learning_rate": 3.966986155484558e-05, "loss": 0.0165, "step": 970 }, { "epoch": 3.12, "learning_rate": 3.956336528221513e-05, "loss": 0.0132, "step": 980 }, { "epoch": 3.16, "learning_rate": 3.9456869009584666e-05, "loss": 0.0184, "step": 990 }, { "epoch": 3.19, "learning_rate": 3.935037273695421e-05, "loss": 0.0303, "step": 1000 }, { "epoch": 3.22, "learning_rate": 3.9243876464323756e-05, "loss": 0.0202, "step": 1010 }, { "epoch": 3.25, "learning_rate": 3.913738019169329e-05, "loss": 0.0159, "step": 1020 }, { "epoch": 3.28, "learning_rate": 3.903088391906283e-05, "loss": 0.0246, "step": 1030 }, { "epoch": 3.31, "learning_rate": 3.892438764643238e-05, "loss": 0.0108, "step": 1040 }, { "epoch": 3.35, "learning_rate": 3.8817891373801916e-05, "loss": 0.0167, "step": 1050 }, { "epoch": 3.38, "learning_rate": 3.871139510117146e-05, "loss": 0.0134, "step": 1060 }, { "epoch": 3.41, "learning_rate": 3.8604898828541e-05, "loss": 0.0122, "step": 1070 }, { "epoch": 3.44, "learning_rate": 3.8498402555910545e-05, "loss": 0.0101, "step": 1080 }, { "epoch": 3.47, "learning_rate": 3.839190628328009e-05, "loss": 0.0118, "step": 1090 }, { "epoch": 3.51, "learning_rate": 3.828541001064963e-05, "loss": 0.0139, "step": 1100 }, { "epoch": 3.54, "learning_rate": 3.8178913738019174e-05, "loss": 0.0175, "step": 1110 }, { "epoch": 3.57, "learning_rate": 3.807241746538872e-05, "loss": 0.0156, "step": 1120 }, { "epoch": 3.6, "learning_rate": 3.796592119275825e-05, "loss": 0.0235, "step": 1130 }, { "epoch": 3.63, "learning_rate": 3.7859424920127795e-05, "loss": 0.0124, "step": 1140 }, { "epoch": 3.67, "learning_rate": 3.775292864749734e-05, "loss": 0.0109, "step": 1150 }, { "epoch": 3.7, "learning_rate": 3.764643237486688e-05, "loss": 0.0189, "step": 1160 }, { "epoch": 3.73, "learning_rate": 3.7539936102236424e-05, "loss": 0.0134, "step": 1170 }, { "epoch": 3.76, "learning_rate": 3.743343982960596e-05, "loss": 0.0314, "step": 1180 }, { "epoch": 3.79, "learning_rate": 3.732694355697551e-05, "loss": 0.0377, "step": 1190 }, { "epoch": 3.82, "learning_rate": 3.722044728434505e-05, "loss": 0.015, "step": 1200 }, { "epoch": 3.86, "learning_rate": 3.711395101171459e-05, "loss": 0.0129, "step": 1210 }, { "epoch": 3.89, "learning_rate": 3.7007454739084136e-05, "loss": 0.0154, "step": 1220 }, { "epoch": 3.92, "learning_rate": 3.6900958466453675e-05, "loss": 0.0117, "step": 1230 }, { "epoch": 3.95, "learning_rate": 3.679446219382322e-05, "loss": 0.0125, "step": 1240 }, { "epoch": 3.98, "learning_rate": 3.668796592119276e-05, "loss": 0.0103, "step": 1250 }, { "epoch": 4.0, "eval_gen_len": 14.734767025089607, "eval_loss": 0.012266965582966805, "eval_rouge1": 54.8551, "eval_rouge2": 49.2036, "eval_rougeL": 54.5865, "eval_rougeLsum": 54.8397, "eval_runtime": 674.9391, "eval_samples_per_second": 0.413, "eval_steps_per_second": 0.207, "step": 1255 }, { "epoch": 4.02, "learning_rate": 3.65814696485623e-05, "loss": 0.0177, "step": 1260 }, { "epoch": 4.05, "learning_rate": 3.647497337593184e-05, "loss": 0.0189, "step": 1270 }, { "epoch": 4.08, "learning_rate": 3.636847710330139e-05, "loss": 0.0113, "step": 1280 }, { "epoch": 4.11, "learning_rate": 3.6261980830670925e-05, "loss": 0.0277, "step": 1290 }, { "epoch": 4.14, "learning_rate": 3.615548455804047e-05, "loss": 0.0236, "step": 1300 }, { "epoch": 4.18, "learning_rate": 3.6048988285410015e-05, "loss": 0.0198, "step": 1310 }, { "epoch": 4.21, "learning_rate": 3.5942492012779554e-05, "loss": 0.0155, "step": 1320 }, { "epoch": 4.24, "learning_rate": 3.58359957401491e-05, "loss": 0.0138, "step": 1330 }, { "epoch": 4.27, "learning_rate": 3.572949946751864e-05, "loss": 0.0154, "step": 1340 }, { "epoch": 4.3, "learning_rate": 3.562300319488818e-05, "loss": 0.0106, "step": 1350 }, { "epoch": 4.33, "learning_rate": 3.551650692225773e-05, "loss": 0.0168, "step": 1360 }, { "epoch": 4.37, "learning_rate": 3.541001064962726e-05, "loss": 0.0114, "step": 1370 }, { "epoch": 4.4, "learning_rate": 3.5303514376996804e-05, "loss": 0.0119, "step": 1380 }, { "epoch": 4.43, "learning_rate": 3.519701810436635e-05, "loss": 0.0168, "step": 1390 }, { "epoch": 4.46, "learning_rate": 3.509052183173589e-05, "loss": 0.012, "step": 1400 }, { "epoch": 4.49, "learning_rate": 3.498402555910543e-05, "loss": 0.011, "step": 1410 }, { "epoch": 4.53, "learning_rate": 3.487752928647498e-05, "loss": 0.014, "step": 1420 }, { "epoch": 4.56, "learning_rate": 3.4771033013844516e-05, "loss": 0.0151, "step": 1430 }, { "epoch": 4.59, "learning_rate": 3.466453674121406e-05, "loss": 0.0148, "step": 1440 }, { "epoch": 4.62, "learning_rate": 3.45580404685836e-05, "loss": 0.0142, "step": 1450 }, { "epoch": 4.65, "learning_rate": 3.4451544195953145e-05, "loss": 0.0183, "step": 1460 }, { "epoch": 4.69, "learning_rate": 3.434504792332269e-05, "loss": 0.0249, "step": 1470 }, { "epoch": 4.72, "learning_rate": 3.423855165069223e-05, "loss": 0.0147, "step": 1480 }, { "epoch": 4.75, "learning_rate": 3.413205537806177e-05, "loss": 0.0205, "step": 1490 }, { "epoch": 4.78, "learning_rate": 3.402555910543131e-05, "loss": 0.0298, "step": 1500 }, { "epoch": 4.81, "learning_rate": 3.391906283280085e-05, "loss": 0.009, "step": 1510 }, { "epoch": 4.84, "learning_rate": 3.3812566560170395e-05, "loss": 0.0167, "step": 1520 }, { "epoch": 4.88, "learning_rate": 3.3706070287539934e-05, "loss": 0.0108, "step": 1530 }, { "epoch": 4.91, "learning_rate": 3.359957401490948e-05, "loss": 0.0126, "step": 1540 }, { "epoch": 4.94, "learning_rate": 3.3493077742279024e-05, "loss": 0.0201, "step": 1550 }, { "epoch": 4.97, "learning_rate": 3.338658146964856e-05, "loss": 0.0132, "step": 1560 }, { "epoch": 5.0, "eval_gen_len": 13.17921146953405, "eval_loss": 0.012017679400742054, "eval_rouge1": 61.0563, "eval_rouge2": 56.3505, "eval_rougeL": 60.775, "eval_rougeLsum": 60.928, "eval_runtime": 625.9017, "eval_samples_per_second": 0.446, "eval_steps_per_second": 0.224, "step": 1568 }, { "epoch": 5.0, "learning_rate": 3.328008519701811e-05, "loss": 0.0088, "step": 1570 }, { "epoch": 5.04, "learning_rate": 3.317358892438765e-05, "loss": 0.012, "step": 1580 }, { "epoch": 5.07, "learning_rate": 3.306709265175719e-05, "loss": 0.0094, "step": 1590 }, { "epoch": 5.1, "learning_rate": 3.2960596379126736e-05, "loss": 0.0263, "step": 1600 }, { "epoch": 5.13, "learning_rate": 3.2854100106496274e-05, "loss": 0.0139, "step": 1610 }, { "epoch": 5.16, "learning_rate": 3.274760383386581e-05, "loss": 0.0111, "step": 1620 }, { "epoch": 5.2, "learning_rate": 3.264110756123536e-05, "loss": 0.0099, "step": 1630 }, { "epoch": 5.23, "learning_rate": 3.2534611288604896e-05, "loss": 0.0284, "step": 1640 }, { "epoch": 5.26, "learning_rate": 3.242811501597444e-05, "loss": 0.023, "step": 1650 }, { "epoch": 5.29, "learning_rate": 3.2321618743343987e-05, "loss": 0.0077, "step": 1660 }, { "epoch": 5.32, "learning_rate": 3.2215122470713525e-05, "loss": 0.0167, "step": 1670 }, { "epoch": 5.35, "learning_rate": 3.210862619808307e-05, "loss": 0.0096, "step": 1680 }, { "epoch": 5.39, "learning_rate": 3.2002129925452615e-05, "loss": 0.0197, "step": 1690 }, { "epoch": 5.42, "learning_rate": 3.1895633652822154e-05, "loss": 0.0105, "step": 1700 }, { "epoch": 5.45, "learning_rate": 3.17891373801917e-05, "loss": 0.0308, "step": 1710 }, { "epoch": 5.48, "learning_rate": 3.168264110756124e-05, "loss": 0.0118, "step": 1720 }, { "epoch": 5.51, "learning_rate": 3.1576144834930775e-05, "loss": 0.0128, "step": 1730 }, { "epoch": 5.55, "learning_rate": 3.146964856230032e-05, "loss": 0.0142, "step": 1740 }, { "epoch": 5.58, "learning_rate": 3.136315228966986e-05, "loss": 0.0086, "step": 1750 }, { "epoch": 5.61, "learning_rate": 3.1256656017039404e-05, "loss": 0.0189, "step": 1760 }, { "epoch": 5.64, "learning_rate": 3.115015974440895e-05, "loss": 0.0137, "step": 1770 }, { "epoch": 5.67, "learning_rate": 3.104366347177849e-05, "loss": 0.0117, "step": 1780 }, { "epoch": 5.71, "learning_rate": 3.093716719914803e-05, "loss": 0.0112, "step": 1790 }, { "epoch": 5.74, "learning_rate": 3.083067092651757e-05, "loss": 0.017, "step": 1800 }, { "epoch": 5.77, "learning_rate": 3.0724174653887116e-05, "loss": 0.0089, "step": 1810 }, { "epoch": 5.8, "learning_rate": 3.061767838125666e-05, "loss": 0.0354, "step": 1820 }, { "epoch": 5.83, "learning_rate": 3.0511182108626203e-05, "loss": 0.0173, "step": 1830 }, { "epoch": 5.86, "learning_rate": 3.0404685835995745e-05, "loss": 0.0132, "step": 1840 }, { "epoch": 5.9, "learning_rate": 3.029818956336528e-05, "loss": 0.0101, "step": 1850 }, { "epoch": 5.93, "learning_rate": 3.0191693290734825e-05, "loss": 0.0172, "step": 1860 }, { "epoch": 5.96, "learning_rate": 3.0085197018104367e-05, "loss": 0.0109, "step": 1870 }, { "epoch": 5.99, "learning_rate": 2.997870074547391e-05, "loss": 0.0179, "step": 1880 }, { "epoch": 6.0, "eval_gen_len": 12.727598566308243, "eval_loss": 0.011885586194694042, "eval_rouge1": 63.4215, "eval_rouge2": 59.2435, "eval_rougeL": 63.1198, "eval_rougeLsum": 63.3059, "eval_runtime": 611.8658, "eval_samples_per_second": 0.456, "eval_steps_per_second": 0.229, "step": 1882 }, { "epoch": 6.02, "learning_rate": 2.987220447284345e-05, "loss": 0.0157, "step": 1890 }, { "epoch": 6.06, "learning_rate": 2.9765708200212995e-05, "loss": 0.0173, "step": 1900 }, { "epoch": 6.09, "learning_rate": 2.9659211927582537e-05, "loss": 0.0132, "step": 1910 }, { "epoch": 6.12, "learning_rate": 2.955271565495208e-05, "loss": 0.0253, "step": 1920 }, { "epoch": 6.15, "learning_rate": 2.944621938232162e-05, "loss": 0.0158, "step": 1930 }, { "epoch": 6.18, "learning_rate": 2.9339723109691166e-05, "loss": 0.0112, "step": 1940 }, { "epoch": 6.22, "learning_rate": 2.9233226837060707e-05, "loss": 0.0097, "step": 1950 }, { "epoch": 6.25, "learning_rate": 2.912673056443025e-05, "loss": 0.0101, "step": 1960 }, { "epoch": 6.28, "learning_rate": 2.9020234291799788e-05, "loss": 0.0197, "step": 1970 }, { "epoch": 6.31, "learning_rate": 2.891373801916933e-05, "loss": 0.0138, "step": 1980 }, { "epoch": 6.34, "learning_rate": 2.880724174653887e-05, "loss": 0.0123, "step": 1990 }, { "epoch": 6.37, "learning_rate": 2.8700745473908413e-05, "loss": 0.0187, "step": 2000 }, { "epoch": 6.41, "learning_rate": 2.8594249201277955e-05, "loss": 0.0164, "step": 2010 }, { "epoch": 6.44, "learning_rate": 2.84877529286475e-05, "loss": 0.0097, "step": 2020 }, { "epoch": 6.47, "learning_rate": 2.838125665601704e-05, "loss": 0.017, "step": 2030 }, { "epoch": 6.5, "learning_rate": 2.8274760383386583e-05, "loss": 0.0072, "step": 2040 }, { "epoch": 6.53, "learning_rate": 2.8168264110756125e-05, "loss": 0.0173, "step": 2050 }, { "epoch": 6.57, "learning_rate": 2.806176783812567e-05, "loss": 0.0196, "step": 2060 }, { "epoch": 6.6, "learning_rate": 2.7955271565495212e-05, "loss": 0.0255, "step": 2070 }, { "epoch": 6.63, "learning_rate": 2.7848775292864754e-05, "loss": 0.019, "step": 2080 }, { "epoch": 6.66, "learning_rate": 2.7742279020234292e-05, "loss": 0.0103, "step": 2090 }, { "epoch": 6.69, "learning_rate": 2.7635782747603834e-05, "loss": 0.0173, "step": 2100 }, { "epoch": 6.73, "learning_rate": 2.7529286474973375e-05, "loss": 0.0113, "step": 2110 }, { "epoch": 6.76, "learning_rate": 2.7422790202342917e-05, "loss": 0.0189, "step": 2120 }, { "epoch": 6.79, "learning_rate": 2.7316293929712462e-05, "loss": 0.0149, "step": 2130 }, { "epoch": 6.82, "learning_rate": 2.7209797657082004e-05, "loss": 0.0128, "step": 2140 }, { "epoch": 6.85, "learning_rate": 2.7103301384451546e-05, "loss": 0.0181, "step": 2150 }, { "epoch": 6.88, "learning_rate": 2.6996805111821088e-05, "loss": 0.0097, "step": 2160 }, { "epoch": 6.92, "learning_rate": 2.6890308839190633e-05, "loss": 0.013, "step": 2170 }, { "epoch": 6.95, "learning_rate": 2.6783812566560174e-05, "loss": 0.014, "step": 2180 }, { "epoch": 6.98, "learning_rate": 2.6677316293929716e-05, "loss": 0.0115, "step": 2190 }, { "epoch": 7.0, "eval_gen_len": 12.92831541218638, "eval_loss": 0.011721653863787651, "eval_rouge1": 62.5874, "eval_rouge2": 58.2782, "eval_rougeL": 62.2116, "eval_rougeLsum": 62.3846, "eval_runtime": 618.8742, "eval_samples_per_second": 0.451, "eval_steps_per_second": 0.226, "step": 2196 }, { "epoch": 7.01, "learning_rate": 2.6570820021299255e-05, "loss": 0.0124, "step": 2200 }, { "epoch": 7.04, "learning_rate": 2.6464323748668796e-05, "loss": 0.0166, "step": 2210 }, { "epoch": 7.08, "learning_rate": 2.6357827476038338e-05, "loss": 0.0099, "step": 2220 }, { "epoch": 7.11, "learning_rate": 2.625133120340788e-05, "loss": 0.0113, "step": 2230 }, { "epoch": 7.14, "learning_rate": 2.6144834930777425e-05, "loss": 0.0174, "step": 2240 }, { "epoch": 7.17, "learning_rate": 2.6038338658146967e-05, "loss": 0.0095, "step": 2250 }, { "epoch": 7.2, "learning_rate": 2.593184238551651e-05, "loss": 0.0068, "step": 2260 }, { "epoch": 7.24, "learning_rate": 2.582534611288605e-05, "loss": 0.0291, "step": 2270 }, { "epoch": 7.27, "learning_rate": 2.5718849840255592e-05, "loss": 0.0171, "step": 2280 }, { "epoch": 7.3, "learning_rate": 2.5612353567625137e-05, "loss": 0.0183, "step": 2290 }, { "epoch": 7.33, "learning_rate": 2.550585729499468e-05, "loss": 0.0105, "step": 2300 }, { "epoch": 7.36, "learning_rate": 2.539936102236422e-05, "loss": 0.0227, "step": 2310 }, { "epoch": 7.39, "learning_rate": 2.529286474973376e-05, "loss": 0.0311, "step": 2320 }, { "epoch": 7.43, "learning_rate": 2.51863684771033e-05, "loss": 0.0133, "step": 2330 }, { "epoch": 7.46, "learning_rate": 2.5079872204472842e-05, "loss": 0.0102, "step": 2340 }, { "epoch": 7.49, "learning_rate": 2.4973375931842384e-05, "loss": 0.0189, "step": 2350 }, { "epoch": 7.52, "learning_rate": 2.486687965921193e-05, "loss": 0.0157, "step": 2360 }, { "epoch": 7.55, "learning_rate": 2.476038338658147e-05, "loss": 0.0114, "step": 2370 }, { "epoch": 7.59, "learning_rate": 2.4653887113951013e-05, "loss": 0.0131, "step": 2380 }, { "epoch": 7.62, "learning_rate": 2.4547390841320554e-05, "loss": 0.0144, "step": 2390 }, { "epoch": 7.65, "learning_rate": 2.44408945686901e-05, "loss": 0.0107, "step": 2400 }, { "epoch": 7.68, "learning_rate": 2.4334398296059638e-05, "loss": 0.0089, "step": 2410 }, { "epoch": 7.71, "learning_rate": 2.422790202342918e-05, "loss": 0.0119, "step": 2420 }, { "epoch": 7.75, "learning_rate": 2.412140575079872e-05, "loss": 0.0136, "step": 2430 }, { "epoch": 7.78, "learning_rate": 2.4014909478168267e-05, "loss": 0.026, "step": 2440 }, { "epoch": 7.81, "learning_rate": 2.390841320553781e-05, "loss": 0.0147, "step": 2450 }, { "epoch": 7.84, "learning_rate": 2.380191693290735e-05, "loss": 0.0114, "step": 2460 }, { "epoch": 7.87, "learning_rate": 2.3695420660276892e-05, "loss": 0.0122, "step": 2470 }, { "epoch": 7.9, "learning_rate": 2.3588924387646434e-05, "loss": 0.0115, "step": 2480 }, { "epoch": 7.94, "learning_rate": 2.3482428115015975e-05, "loss": 0.016, "step": 2490 }, { "epoch": 7.97, "learning_rate": 2.3375931842385517e-05, "loss": 0.0177, "step": 2500 }, { "epoch": 8.0, "learning_rate": 2.326943556975506e-05, "loss": 0.0097, "step": 2510 }, { "epoch": 8.0, "eval_gen_len": 12.827956989247312, "eval_loss": 0.0116655882447958, "eval_rouge1": 63.06, "eval_rouge2": 58.8818, "eval_rougeL": 62.7484, "eval_rougeLsum": 62.8736, "eval_runtime": 618.883, "eval_samples_per_second": 0.451, "eval_steps_per_second": 0.226, "step": 2510 }, { "epoch": 8.03, "learning_rate": 2.3162939297124604e-05, "loss": 0.0122, "step": 2520 }, { "epoch": 8.06, "learning_rate": 2.3056443024494142e-05, "loss": 0.0125, "step": 2530 }, { "epoch": 8.1, "learning_rate": 2.2949946751863684e-05, "loss": 0.013, "step": 2540 }, { "epoch": 8.13, "learning_rate": 2.284345047923323e-05, "loss": 0.0138, "step": 2550 }, { "epoch": 8.16, "learning_rate": 2.273695420660277e-05, "loss": 0.017, "step": 2560 }, { "epoch": 8.19, "learning_rate": 2.2630457933972313e-05, "loss": 0.0199, "step": 2570 }, { "epoch": 8.22, "learning_rate": 2.2523961661341854e-05, "loss": 0.0157, "step": 2580 }, { "epoch": 8.25, "learning_rate": 2.2417465388711396e-05, "loss": 0.0092, "step": 2590 }, { "epoch": 8.29, "learning_rate": 2.2310969116080938e-05, "loss": 0.011, "step": 2600 }, { "epoch": 8.32, "learning_rate": 2.220447284345048e-05, "loss": 0.012, "step": 2610 }, { "epoch": 8.35, "learning_rate": 2.209797657082002e-05, "loss": 0.0155, "step": 2620 }, { "epoch": 8.38, "learning_rate": 2.1991480298189567e-05, "loss": 0.0132, "step": 2630 }, { "epoch": 8.41, "learning_rate": 2.188498402555911e-05, "loss": 0.0119, "step": 2640 }, { "epoch": 8.45, "learning_rate": 2.1778487752928647e-05, "loss": 0.0147, "step": 2650 }, { "epoch": 8.48, "learning_rate": 2.167199148029819e-05, "loss": 0.0152, "step": 2660 }, { "epoch": 8.51, "learning_rate": 2.1565495207667734e-05, "loss": 0.0158, "step": 2670 }, { "epoch": 8.54, "learning_rate": 2.1458998935037275e-05, "loss": 0.017, "step": 2680 }, { "epoch": 8.57, "learning_rate": 2.1352502662406817e-05, "loss": 0.0143, "step": 2690 }, { "epoch": 8.61, "learning_rate": 2.124600638977636e-05, "loss": 0.015, "step": 2700 }, { "epoch": 8.64, "learning_rate": 2.11395101171459e-05, "loss": 0.0113, "step": 2710 }, { "epoch": 8.67, "learning_rate": 2.1033013844515442e-05, "loss": 0.0131, "step": 2720 }, { "epoch": 8.7, "learning_rate": 2.0926517571884984e-05, "loss": 0.0109, "step": 2730 }, { "epoch": 8.73, "learning_rate": 2.0820021299254526e-05, "loss": 0.0132, "step": 2740 }, { "epoch": 8.76, "learning_rate": 2.071352502662407e-05, "loss": 0.0219, "step": 2750 }, { "epoch": 8.8, "learning_rate": 2.0607028753993613e-05, "loss": 0.0108, "step": 2760 }, { "epoch": 8.83, "learning_rate": 2.050053248136315e-05, "loss": 0.0143, "step": 2770 }, { "epoch": 8.86, "learning_rate": 2.0394036208732696e-05, "loss": 0.0114, "step": 2780 }, { "epoch": 8.89, "learning_rate": 2.0287539936102238e-05, "loss": 0.0136, "step": 2790 }, { "epoch": 8.92, "learning_rate": 2.018104366347178e-05, "loss": 0.0114, "step": 2800 }, { "epoch": 8.96, "learning_rate": 2.007454739084132e-05, "loss": 0.0379, "step": 2810 }, { "epoch": 8.99, "learning_rate": 1.9968051118210863e-05, "loss": 0.0172, "step": 2820 }, { "epoch": 9.0, "eval_gen_len": 12.827956989247312, "eval_loss": 0.011623449623584747, "eval_rouge1": 63.4646, "eval_rouge2": 59.358, "eval_rougeL": 63.1834, "eval_rougeLsum": 63.3962, "eval_runtime": 618.8791, "eval_samples_per_second": 0.451, "eval_steps_per_second": 0.226, "step": 2823 }, { "epoch": 9.02, "learning_rate": 1.9861554845580405e-05, "loss": 0.0179, "step": 2830 }, { "epoch": 9.05, "learning_rate": 1.9755058572949947e-05, "loss": 0.0116, "step": 2840 }, { "epoch": 9.08, "learning_rate": 1.964856230031949e-05, "loss": 0.0108, "step": 2850 }, { "epoch": 9.12, "learning_rate": 1.9542066027689034e-05, "loss": 0.0126, "step": 2860 }, { "epoch": 9.15, "learning_rate": 1.9435569755058575e-05, "loss": 0.0092, "step": 2870 }, { "epoch": 9.18, "learning_rate": 1.9329073482428117e-05, "loss": 0.0163, "step": 2880 }, { "epoch": 9.21, "learning_rate": 1.9222577209797655e-05, "loss": 0.0162, "step": 2890 }, { "epoch": 9.24, "learning_rate": 1.91160809371672e-05, "loss": 0.0161, "step": 2900 }, { "epoch": 9.27, "learning_rate": 1.9009584664536742e-05, "loss": 0.0335, "step": 2910 }, { "epoch": 9.31, "learning_rate": 1.8903088391906284e-05, "loss": 0.0087, "step": 2920 }, { "epoch": 9.34, "learning_rate": 1.8796592119275826e-05, "loss": 0.0163, "step": 2930 }, { "epoch": 9.37, "learning_rate": 1.869009584664537e-05, "loss": 0.01, "step": 2940 }, { "epoch": 9.4, "learning_rate": 1.858359957401491e-05, "loss": 0.0215, "step": 2950 }, { "epoch": 9.43, "learning_rate": 1.847710330138445e-05, "loss": 0.0137, "step": 2960 }, { "epoch": 9.47, "learning_rate": 1.8370607028753993e-05, "loss": 0.0208, "step": 2970 }, { "epoch": 9.5, "learning_rate": 1.8264110756123538e-05, "loss": 0.026, "step": 2980 }, { "epoch": 9.53, "learning_rate": 1.815761448349308e-05, "loss": 0.0109, "step": 2990 }, { "epoch": 9.56, "learning_rate": 1.805111821086262e-05, "loss": 0.0122, "step": 3000 }, { "epoch": 9.59, "learning_rate": 1.7944621938232163e-05, "loss": 0.0109, "step": 3010 }, { "epoch": 9.63, "learning_rate": 1.7838125665601705e-05, "loss": 0.0203, "step": 3020 }, { "epoch": 9.66, "learning_rate": 1.7731629392971247e-05, "loss": 0.0081, "step": 3030 }, { "epoch": 9.69, "learning_rate": 1.762513312034079e-05, "loss": 0.0166, "step": 3040 }, { "epoch": 9.72, "learning_rate": 1.7518636847710333e-05, "loss": 0.0131, "step": 3050 }, { "epoch": 9.75, "learning_rate": 1.7412140575079875e-05, "loss": 0.0159, "step": 3060 }, { "epoch": 9.78, "learning_rate": 1.7305644302449414e-05, "loss": 0.0089, "step": 3070 }, { "epoch": 9.82, "learning_rate": 1.7199148029818955e-05, "loss": 0.0112, "step": 3080 }, { "epoch": 9.85, "learning_rate": 1.70926517571885e-05, "loss": 0.0165, "step": 3090 }, { "epoch": 9.88, "learning_rate": 1.6986155484558042e-05, "loss": 0.0094, "step": 3100 }, { "epoch": 9.91, "learning_rate": 1.6879659211927584e-05, "loss": 0.0118, "step": 3110 }, { "epoch": 9.94, "learning_rate": 1.6773162939297126e-05, "loss": 0.0125, "step": 3120 }, { "epoch": 9.98, "learning_rate": 1.6666666666666667e-05, "loss": 0.0122, "step": 3130 }, { "epoch": 10.0, "eval_gen_len": 12.526881720430108, "eval_loss": 0.011583242565393448, "eval_rouge1": 63.9777, "eval_rouge2": 60.058, "eval_rougeL": 63.7162, "eval_rougeLsum": 63.8393, "eval_runtime": 615.3638, "eval_samples_per_second": 0.453, "eval_steps_per_second": 0.228, "step": 3137 }, { "epoch": 10.01, "learning_rate": 1.656017039403621e-05, "loss": 0.0102, "step": 3140 }, { "epoch": 10.04, "learning_rate": 1.645367412140575e-05, "loss": 0.0139, "step": 3150 }, { "epoch": 10.07, "learning_rate": 1.6347177848775293e-05, "loss": 0.0146, "step": 3160 }, { "epoch": 10.1, "learning_rate": 1.6240681576144838e-05, "loss": 0.022, "step": 3170 }, { "epoch": 10.14, "learning_rate": 1.6134185303514376e-05, "loss": 0.0159, "step": 3180 }, { "epoch": 10.17, "learning_rate": 1.6027689030883918e-05, "loss": 0.0124, "step": 3190 }, { "epoch": 10.2, "learning_rate": 1.5921192758253463e-05, "loss": 0.0144, "step": 3200 }, { "epoch": 10.23, "learning_rate": 1.5814696485623005e-05, "loss": 0.0145, "step": 3210 }, { "epoch": 10.26, "learning_rate": 1.5708200212992547e-05, "loss": 0.0134, "step": 3220 }, { "epoch": 10.29, "learning_rate": 1.560170394036209e-05, "loss": 0.0152, "step": 3230 }, { "epoch": 10.33, "learning_rate": 1.549520766773163e-05, "loss": 0.0105, "step": 3240 }, { "epoch": 10.36, "learning_rate": 1.5388711395101172e-05, "loss": 0.0124, "step": 3250 }, { "epoch": 10.39, "learning_rate": 1.5282215122470714e-05, "loss": 0.0179, "step": 3260 }, { "epoch": 10.42, "learning_rate": 1.5175718849840257e-05, "loss": 0.0236, "step": 3270 }, { "epoch": 10.45, "learning_rate": 1.5069222577209799e-05, "loss": 0.0083, "step": 3280 }, { "epoch": 10.49, "learning_rate": 1.4962726304579342e-05, "loss": 0.0232, "step": 3290 }, { "epoch": 10.52, "learning_rate": 1.485623003194888e-05, "loss": 0.0089, "step": 3300 }, { "epoch": 10.55, "learning_rate": 1.4749733759318424e-05, "loss": 0.0116, "step": 3310 }, { "epoch": 10.58, "learning_rate": 1.4643237486687966e-05, "loss": 0.0106, "step": 3320 }, { "epoch": 10.61, "learning_rate": 1.453674121405751e-05, "loss": 0.0327, "step": 3330 }, { "epoch": 10.65, "learning_rate": 1.4430244941427051e-05, "loss": 0.0074, "step": 3340 }, { "epoch": 10.68, "learning_rate": 1.4323748668796594e-05, "loss": 0.0113, "step": 3350 }, { "epoch": 10.71, "learning_rate": 1.4217252396166134e-05, "loss": 0.0139, "step": 3360 }, { "epoch": 10.74, "learning_rate": 1.4110756123535676e-05, "loss": 0.0106, "step": 3370 }, { "epoch": 10.77, "learning_rate": 1.4004259850905218e-05, "loss": 0.011, "step": 3380 }, { "epoch": 10.8, "learning_rate": 1.3897763578274761e-05, "loss": 0.0161, "step": 3390 }, { "epoch": 10.84, "learning_rate": 1.3791267305644303e-05, "loss": 0.0156, "step": 3400 }, { "epoch": 10.87, "learning_rate": 1.3684771033013847e-05, "loss": 0.0076, "step": 3410 }, { "epoch": 10.9, "learning_rate": 1.3578274760383387e-05, "loss": 0.0111, "step": 3420 }, { "epoch": 10.93, "learning_rate": 1.3471778487752928e-05, "loss": 0.0159, "step": 3430 }, { "epoch": 10.96, "learning_rate": 1.3365282215122472e-05, "loss": 0.0138, "step": 3440 }, { "epoch": 11.0, "learning_rate": 1.3258785942492014e-05, "loss": 0.0132, "step": 3450 }, { "epoch": 11.0, "eval_gen_len": 12.577060931899641, "eval_loss": 0.011564880609512329, "eval_rouge1": 63.7793, "eval_rouge2": 59.7984, "eval_rougeL": 63.5653, "eval_rougeLsum": 63.7034, "eval_runtime": 615.3689, "eval_samples_per_second": 0.453, "eval_steps_per_second": 0.228, "step": 3451 }, { "epoch": 11.03, "learning_rate": 1.3152289669861555e-05, "loss": 0.0129, "step": 3460 }, { "epoch": 11.06, "learning_rate": 1.3045793397231099e-05, "loss": 0.0227, "step": 3470 }, { "epoch": 11.09, "learning_rate": 1.2939297124600639e-05, "loss": 0.0112, "step": 3480 }, { "epoch": 11.12, "learning_rate": 1.283280085197018e-05, "loss": 0.0147, "step": 3490 }, { "epoch": 11.16, "learning_rate": 1.2726304579339724e-05, "loss": 0.0113, "step": 3500 }, { "epoch": 11.19, "learning_rate": 1.2619808306709266e-05, "loss": 0.0123, "step": 3510 }, { "epoch": 11.22, "learning_rate": 1.2513312034078809e-05, "loss": 0.0191, "step": 3520 }, { "epoch": 11.25, "learning_rate": 1.240681576144835e-05, "loss": 0.0128, "step": 3530 }, { "epoch": 11.28, "learning_rate": 1.2300319488817893e-05, "loss": 0.0101, "step": 3540 }, { "epoch": 11.31, "learning_rate": 1.2193823216187434e-05, "loss": 0.0129, "step": 3550 }, { "epoch": 11.35, "learning_rate": 1.2087326943556976e-05, "loss": 0.0093, "step": 3560 }, { "epoch": 11.38, "learning_rate": 1.1980830670926518e-05, "loss": 0.0137, "step": 3570 }, { "epoch": 11.41, "learning_rate": 1.1874334398296061e-05, "loss": 0.0114, "step": 3580 }, { "epoch": 11.44, "learning_rate": 1.1767838125665601e-05, "loss": 0.0164, "step": 3590 }, { "epoch": 11.47, "learning_rate": 1.1661341853035145e-05, "loss": 0.0207, "step": 3600 }, { "epoch": 11.51, "learning_rate": 1.1554845580404687e-05, "loss": 0.0109, "step": 3610 }, { "epoch": 11.54, "learning_rate": 1.1448349307774228e-05, "loss": 0.0169, "step": 3620 }, { "epoch": 11.57, "learning_rate": 1.134185303514377e-05, "loss": 0.0102, "step": 3630 }, { "epoch": 11.6, "learning_rate": 1.1235356762513314e-05, "loss": 0.0248, "step": 3640 }, { "epoch": 11.63, "learning_rate": 1.1128860489882854e-05, "loss": 0.0112, "step": 3650 }, { "epoch": 11.67, "learning_rate": 1.1022364217252397e-05, "loss": 0.02, "step": 3660 }, { "epoch": 11.7, "learning_rate": 1.0915867944621939e-05, "loss": 0.0127, "step": 3670 }, { "epoch": 11.73, "learning_rate": 1.080937167199148e-05, "loss": 0.0147, "step": 3680 }, { "epoch": 11.76, "learning_rate": 1.0702875399361024e-05, "loss": 0.014, "step": 3690 }, { "epoch": 11.79, "learning_rate": 1.0596379126730564e-05, "loss": 0.0079, "step": 3700 }, { "epoch": 11.82, "learning_rate": 1.0489882854100107e-05, "loss": 0.0134, "step": 3710 }, { "epoch": 11.86, "learning_rate": 1.038338658146965e-05, "loss": 0.0188, "step": 3720 }, { "epoch": 11.89, "learning_rate": 1.0276890308839191e-05, "loss": 0.0087, "step": 3730 }, { "epoch": 11.92, "learning_rate": 1.0170394036208733e-05, "loss": 0.0104, "step": 3740 }, { "epoch": 11.95, "learning_rate": 1.0063897763578276e-05, "loss": 0.0132, "step": 3750 }, { "epoch": 11.98, "learning_rate": 9.957401490947816e-06, "loss": 0.0242, "step": 3760 }, { "epoch": 12.0, "eval_gen_len": 12.67741935483871, "eval_loss": 0.011580224148929119, "eval_rouge1": 63.3071, "eval_rouge2": 59.216, "eval_rougeL": 62.9553, "eval_rougeLsum": 63.1625, "eval_runtime": 618.8937, "eval_samples_per_second": 0.451, "eval_steps_per_second": 0.226, "step": 3765 }, { "epoch": 12.02, "learning_rate": 9.85090521831736e-06, "loss": 0.0084, "step": 3770 }, { "epoch": 12.05, "learning_rate": 9.744408945686901e-06, "loss": 0.0106, "step": 3780 }, { "epoch": 12.08, "learning_rate": 9.637912673056443e-06, "loss": 0.0106, "step": 3790 }, { "epoch": 12.11, "learning_rate": 9.531416400425985e-06, "loss": 0.0111, "step": 3800 }, { "epoch": 12.14, "learning_rate": 9.424920127795528e-06, "loss": 0.0118, "step": 3810 }, { "epoch": 12.18, "learning_rate": 9.318423855165068e-06, "loss": 0.0114, "step": 3820 }, { "epoch": 12.21, "learning_rate": 9.211927582534612e-06, "loss": 0.0134, "step": 3830 }, { "epoch": 12.24, "learning_rate": 9.105431309904154e-06, "loss": 0.0119, "step": 3840 }, { "epoch": 12.27, "learning_rate": 8.998935037273695e-06, "loss": 0.0097, "step": 3850 }, { "epoch": 12.3, "learning_rate": 8.892438764643237e-06, "loss": 0.0167, "step": 3860 }, { "epoch": 12.33, "learning_rate": 8.78594249201278e-06, "loss": 0.0281, "step": 3870 }, { "epoch": 12.37, "learning_rate": 8.679446219382322e-06, "loss": 0.0084, "step": 3880 }, { "epoch": 12.4, "learning_rate": 8.572949946751864e-06, "loss": 0.0105, "step": 3890 }, { "epoch": 12.43, "learning_rate": 8.466453674121406e-06, "loss": 0.0146, "step": 3900 }, { "epoch": 12.46, "learning_rate": 8.359957401490947e-06, "loss": 0.011, "step": 3910 }, { "epoch": 12.49, "learning_rate": 8.253461128860491e-06, "loss": 0.0077, "step": 3920 }, { "epoch": 12.53, "learning_rate": 8.146964856230033e-06, "loss": 0.0278, "step": 3930 }, { "epoch": 12.56, "learning_rate": 8.040468583599574e-06, "loss": 0.0151, "step": 3940 }, { "epoch": 12.59, "learning_rate": 7.933972310969116e-06, "loss": 0.0124, "step": 3950 }, { "epoch": 12.62, "learning_rate": 7.82747603833866e-06, "loss": 0.0265, "step": 3960 }, { "epoch": 12.65, "learning_rate": 7.7209797657082e-06, "loss": 0.014, "step": 3970 }, { "epoch": 12.69, "learning_rate": 7.614483493077742e-06, "loss": 0.0118, "step": 3980 }, { "epoch": 12.72, "learning_rate": 7.507987220447285e-06, "loss": 0.0163, "step": 3990 }, { "epoch": 12.75, "learning_rate": 7.4014909478168266e-06, "loss": 0.0199, "step": 4000 }, { "epoch": 12.78, "learning_rate": 7.294994675186369e-06, "loss": 0.0082, "step": 4010 }, { "epoch": 12.81, "learning_rate": 7.188498402555911e-06, "loss": 0.0067, "step": 4020 }, { "epoch": 12.84, "learning_rate": 7.082002129925453e-06, "loss": 0.015, "step": 4030 }, { "epoch": 12.88, "learning_rate": 6.975505857294995e-06, "loss": 0.0156, "step": 4040 }, { "epoch": 12.91, "learning_rate": 6.869009584664538e-06, "loss": 0.0218, "step": 4050 }, { "epoch": 12.94, "learning_rate": 6.762513312034079e-06, "loss": 0.0165, "step": 4060 }, { "epoch": 12.97, "learning_rate": 6.656017039403621e-06, "loss": 0.0178, "step": 4070 }, { "epoch": 13.0, "eval_gen_len": 12.67741935483871, "eval_loss": 0.011560924351215363, "eval_rouge1": 63.3607, "eval_rouge2": 59.1699, "eval_rougeL": 63.0136, "eval_rougeLsum": 63.1274, "eval_runtime": 618.877, "eval_samples_per_second": 0.451, "eval_steps_per_second": 0.226, "step": 4078 }, { "epoch": 13.0, "learning_rate": 6.549520766773164e-06, "loss": 0.0092, "step": 4080 }, { "epoch": 13.04, "learning_rate": 6.443024494142705e-06, "loss": 0.0084, "step": 4090 }, { "epoch": 13.07, "learning_rate": 6.336528221512247e-06, "loss": 0.0149, "step": 4100 }, { "epoch": 13.1, "learning_rate": 6.230031948881789e-06, "loss": 0.0138, "step": 4110 }, { "epoch": 13.13, "learning_rate": 6.123535676251332e-06, "loss": 0.0128, "step": 4120 }, { "epoch": 13.16, "learning_rate": 6.0170394036208735e-06, "loss": 0.0124, "step": 4130 }, { "epoch": 13.2, "learning_rate": 5.910543130990415e-06, "loss": 0.0122, "step": 4140 }, { "epoch": 13.23, "learning_rate": 5.804046858359958e-06, "loss": 0.0175, "step": 4150 }, { "epoch": 13.26, "learning_rate": 5.6975505857295e-06, "loss": 0.0142, "step": 4160 }, { "epoch": 13.29, "learning_rate": 5.591054313099041e-06, "loss": 0.009, "step": 4170 }, { "epoch": 13.32, "learning_rate": 5.484558040468584e-06, "loss": 0.0129, "step": 4180 }, { "epoch": 13.35, "learning_rate": 5.378061767838126e-06, "loss": 0.0086, "step": 4190 }, { "epoch": 13.39, "learning_rate": 5.2715654952076674e-06, "loss": 0.0109, "step": 4200 }, { "epoch": 13.42, "learning_rate": 5.16506922257721e-06, "loss": 0.0127, "step": 4210 }, { "epoch": 13.45, "learning_rate": 5.058572949946752e-06, "loss": 0.0235, "step": 4220 }, { "epoch": 13.48, "learning_rate": 4.952076677316294e-06, "loss": 0.0101, "step": 4230 }, { "epoch": 13.51, "learning_rate": 4.845580404685836e-06, "loss": 0.0177, "step": 4240 }, { "epoch": 13.55, "learning_rate": 4.739084132055379e-06, "loss": 0.0242, "step": 4250 }, { "epoch": 13.58, "learning_rate": 4.6325878594249205e-06, "loss": 0.0157, "step": 4260 }, { "epoch": 13.61, "learning_rate": 4.526091586794463e-06, "loss": 0.015, "step": 4270 }, { "epoch": 13.64, "learning_rate": 4.419595314164005e-06, "loss": 0.0139, "step": 4280 }, { "epoch": 13.67, "learning_rate": 4.3130990415335465e-06, "loss": 0.0126, "step": 4290 }, { "epoch": 13.71, "learning_rate": 4.206602768903089e-06, "loss": 0.0137, "step": 4300 }, { "epoch": 13.74, "learning_rate": 4.100106496272631e-06, "loss": 0.0071, "step": 4310 }, { "epoch": 13.77, "learning_rate": 3.993610223642173e-06, "loss": 0.0201, "step": 4320 }, { "epoch": 13.8, "learning_rate": 3.887113951011715e-06, "loss": 0.0188, "step": 4330 }, { "epoch": 13.83, "learning_rate": 3.780617678381257e-06, "loss": 0.0081, "step": 4340 }, { "epoch": 13.86, "learning_rate": 3.6741214057507987e-06, "loss": 0.0154, "step": 4350 }, { "epoch": 13.9, "learning_rate": 3.5676251331203413e-06, "loss": 0.0159, "step": 4360 }, { "epoch": 13.93, "learning_rate": 3.461128860489883e-06, "loss": 0.0128, "step": 4370 }, { "epoch": 13.96, "learning_rate": 3.354632587859425e-06, "loss": 0.0144, "step": 4380 }, { "epoch": 13.99, "learning_rate": 3.2481363152289674e-06, "loss": 0.0225, "step": 4390 }, { "epoch": 14.0, "eval_gen_len": 12.627240143369175, "eval_loss": 0.011567563749849796, "eval_rouge1": 63.5224, "eval_rouge2": 59.4962, "eval_rougeL": 63.22, "eval_rougeLsum": 63.4236, "eval_runtime": 615.3823, "eval_samples_per_second": 0.453, "eval_steps_per_second": 0.228, "step": 4392 }, { "epoch": 14.02, "learning_rate": 3.141640042598509e-06, "loss": 0.0107, "step": 4400 }, { "epoch": 14.06, "learning_rate": 3.0351437699680513e-06, "loss": 0.0207, "step": 4410 }, { "epoch": 14.09, "learning_rate": 2.9286474973375935e-06, "loss": 0.0132, "step": 4420 }, { "epoch": 14.12, "learning_rate": 2.8221512247071357e-06, "loss": 0.0142, "step": 4430 }, { "epoch": 14.15, "learning_rate": 2.7156549520766774e-06, "loss": 0.0139, "step": 4440 }, { "epoch": 14.18, "learning_rate": 2.6091586794462196e-06, "loss": 0.0127, "step": 4450 }, { "epoch": 14.22, "learning_rate": 2.5026624068157617e-06, "loss": 0.0136, "step": 4460 }, { "epoch": 14.25, "learning_rate": 2.3961661341853035e-06, "loss": 0.0086, "step": 4470 }, { "epoch": 14.28, "learning_rate": 2.2896698615548457e-06, "loss": 0.0104, "step": 4480 }, { "epoch": 14.31, "learning_rate": 2.183173588924388e-06, "loss": 0.0205, "step": 4490 }, { "epoch": 14.34, "learning_rate": 2.0766773162939296e-06, "loss": 0.0158, "step": 4500 }, { "epoch": 14.37, "learning_rate": 1.9701810436634718e-06, "loss": 0.0114, "step": 4510 }, { "epoch": 14.41, "learning_rate": 1.8636847710330141e-06, "loss": 0.0296, "step": 4520 }, { "epoch": 14.44, "learning_rate": 1.7571884984025559e-06, "loss": 0.021, "step": 4530 }, { "epoch": 14.47, "learning_rate": 1.650692225772098e-06, "loss": 0.0123, "step": 4540 }, { "epoch": 14.5, "learning_rate": 1.54419595314164e-06, "loss": 0.0178, "step": 4550 }, { "epoch": 14.53, "learning_rate": 1.4376996805111822e-06, "loss": 0.0116, "step": 4560 }, { "epoch": 14.57, "learning_rate": 1.3312034078807244e-06, "loss": 0.0101, "step": 4570 }, { "epoch": 14.6, "learning_rate": 1.2247071352502663e-06, "loss": 0.0166, "step": 4580 }, { "epoch": 14.63, "learning_rate": 1.1182108626198083e-06, "loss": 0.0136, "step": 4590 }, { "epoch": 14.66, "learning_rate": 1.0117145899893504e-06, "loss": 0.0161, "step": 4600 }, { "epoch": 14.69, "learning_rate": 9.052183173588925e-07, "loss": 0.0082, "step": 4610 }, { "epoch": 14.73, "learning_rate": 7.987220447284345e-07, "loss": 0.0159, "step": 4620 }, { "epoch": 14.76, "learning_rate": 6.922257720979766e-07, "loss": 0.0093, "step": 4630 }, { "epoch": 14.79, "learning_rate": 5.857294994675187e-07, "loss": 0.0096, "step": 4640 }, { "epoch": 14.82, "learning_rate": 4.792332268370607e-07, "loss": 0.0135, "step": 4650 }, { "epoch": 14.85, "learning_rate": 3.727369542066028e-07, "loss": 0.0123, "step": 4660 }, { "epoch": 14.88, "learning_rate": 2.6624068157614484e-07, "loss": 0.007, "step": 4670 }, { "epoch": 14.92, "learning_rate": 1.597444089456869e-07, "loss": 0.0184, "step": 4680 }, { "epoch": 14.95, "learning_rate": 5.3248136315228974e-08, "loss": 0.0204, "step": 4690 }, { "epoch": 14.96, "eval_gen_len": 12.577060931899641, "eval_loss": 0.01156473346054554, "eval_rouge1": 63.2308, "eval_rouge2": 59.1657, "eval_rougeL": 62.9973, "eval_rougeLsum": 63.1556, "eval_runtime": 615.4381, "eval_samples_per_second": 0.453, "eval_steps_per_second": 0.227, "step": 4695 } ], "max_steps": 4695, "num_train_epochs": 15, "total_flos": 3.211052520088535e+17, "trial_name": null, "trial_params": null }