diff --git "a/trainer_state.json" "b/trainer_state.json" new file mode 100644--- /dev/null +++ "b/trainer_state.json" @@ -0,0 +1,12397 @@ +{ + "best_metric": 0.9958071934678491, + "best_model_checkpoint": "output_dev/v36-deberta_v2_xxlarge-f1-warmup-greater-patience-seqlen-240-6epoch-lr-2e-5/checkpoint-14600", + "epoch": 5.572198921509886, + "eval_steps": 200, + "global_step": 18600, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0, + "learning_rate": 4.990019960079841e-08, + "loss": 1.2142, + "step": 10 + }, + { + "epoch": 0.01, + "learning_rate": 9.980039920159682e-08, + "loss": 1.2, + "step": 20 + }, + { + "epoch": 0.01, + "learning_rate": 1.4970059880239523e-07, + "loss": 1.1833, + "step": 30 + }, + { + "epoch": 0.01, + "learning_rate": 1.9960079840319364e-07, + "loss": 1.1024, + "step": 40 + }, + { + "epoch": 0.01, + "learning_rate": 2.4950099800399204e-07, + "loss": 1.0527, + "step": 50 + }, + { + "epoch": 0.02, + "learning_rate": 2.9940119760479047e-07, + "loss": 0.9926, + "step": 60 + }, + { + "epoch": 0.02, + "learning_rate": 3.4930139720558884e-07, + "loss": 0.9131, + "step": 70 + }, + { + "epoch": 0.02, + "learning_rate": 3.9920159680638727e-07, + "loss": 0.8547, + "step": 80 + }, + { + "epoch": 0.03, + "learning_rate": 4.4910179640718565e-07, + "loss": 0.7847, + "step": 90 + }, + { + "epoch": 0.03, + "learning_rate": 4.990019960079841e-07, + "loss": 0.6674, + "step": 100 + }, + { + "epoch": 0.03, + "learning_rate": 5.489021956087825e-07, + "loss": 0.7092, + "step": 110 + }, + { + "epoch": 0.04, + "learning_rate": 5.988023952095809e-07, + "loss": 0.6892, + "step": 120 + }, + { + "epoch": 0.04, + "learning_rate": 6.487025948103792e-07, + "loss": 0.593, + "step": 130 + }, + { + "epoch": 0.04, + "learning_rate": 6.986027944111777e-07, + "loss": 0.5163, + "step": 140 + }, + { + "epoch": 0.04, + "learning_rate": 7.485029940119761e-07, + "loss": 0.4687, + "step": 150 + }, + { + "epoch": 0.05, + "learning_rate": 7.984031936127745e-07, + "loss": 0.4181, + "step": 160 + }, + { + "epoch": 0.05, + "learning_rate": 8.483033932135728e-07, + "loss": 0.3147, + "step": 170 + }, + { + "epoch": 0.05, + "learning_rate": 8.982035928143713e-07, + "loss": 0.3517, + "step": 180 + }, + { + "epoch": 0.06, + "learning_rate": 9.481037924151697e-07, + "loss": 0.3178, + "step": 190 + }, + { + "epoch": 0.06, + "learning_rate": 9.980039920159682e-07, + "loss": 0.3168, + "step": 200 + }, + { + "epoch": 0.06, + "eval_f1": 0.715117705506412, + "eval_f1_0": 0.9524392895324753, + "eval_f1_1": 0.9676134766619297, + "eval_f1_2": 0.00017229218601095132, + "eval_f1_3": 0.9402457636452318, + "eval_loss": 0.2787778079509735, + "eval_runtime": 153.8618, + "eval_samples_per_second": 77.134, + "eval_steps_per_second": 0.305, + "step": 200 + }, + { + "epoch": 0.06, + "learning_rate": 1.0479041916167664e-06, + "loss": 0.2541, + "step": 210 + }, + { + "epoch": 0.07, + "learning_rate": 1.097804391217565e-06, + "loss": 0.227, + "step": 220 + }, + { + "epoch": 0.07, + "learning_rate": 1.1477045908183634e-06, + "loss": 0.2468, + "step": 230 + }, + { + "epoch": 0.07, + "learning_rate": 1.1976047904191619e-06, + "loss": 0.236, + "step": 240 + }, + { + "epoch": 0.07, + "learning_rate": 1.2475049900199601e-06, + "loss": 0.2776, + "step": 250 + }, + { + "epoch": 0.08, + "learning_rate": 1.2974051896207584e-06, + "loss": 0.2483, + "step": 260 + }, + { + "epoch": 0.08, + "learning_rate": 1.347305389221557e-06, + "loss": 0.2117, + "step": 270 + }, + { + "epoch": 0.08, + "learning_rate": 1.3972055888223554e-06, + "loss": 0.18, + "step": 280 + }, + { + "epoch": 0.09, + "learning_rate": 1.4471057884231539e-06, + "loss": 0.1752, + "step": 290 + }, + { + "epoch": 0.09, + "learning_rate": 1.4970059880239521e-06, + "loss": 0.17, + "step": 300 + }, + { + "epoch": 0.09, + "learning_rate": 1.5469061876247506e-06, + "loss": 0.1736, + "step": 310 + }, + { + "epoch": 0.1, + "learning_rate": 1.596806387225549e-06, + "loss": 0.1771, + "step": 320 + }, + { + "epoch": 0.1, + "learning_rate": 1.6467065868263474e-06, + "loss": 0.2029, + "step": 330 + }, + { + "epoch": 0.1, + "learning_rate": 1.6966067864271456e-06, + "loss": 0.1745, + "step": 340 + }, + { + "epoch": 0.1, + "learning_rate": 1.7465069860279443e-06, + "loss": 0.1408, + "step": 350 + }, + { + "epoch": 0.11, + "learning_rate": 1.7964071856287426e-06, + "loss": 0.1479, + "step": 360 + }, + { + "epoch": 0.11, + "learning_rate": 1.846307385229541e-06, + "loss": 0.1184, + "step": 370 + }, + { + "epoch": 0.11, + "learning_rate": 1.8962075848303393e-06, + "loss": 0.1263, + "step": 380 + }, + { + "epoch": 0.12, + "learning_rate": 1.946107784431138e-06, + "loss": 0.1521, + "step": 390 + }, + { + "epoch": 0.12, + "learning_rate": 1.9960079840319363e-06, + "loss": 0.1625, + "step": 400 + }, + { + "epoch": 0.12, + "eval_f1": 0.9105686190994257, + "eval_f1_0": 0.9668113842193872, + "eval_f1_1": 0.9911796094568739, + "eval_f1_2": 0.693173384704574, + "eval_f1_3": 0.991110098016868, + "eval_loss": 0.13286341726779938, + "eval_runtime": 153.1538, + "eval_samples_per_second": 77.491, + "eval_steps_per_second": 0.307, + "step": 400 + }, + { + "epoch": 0.12, + "learning_rate": 2.0459081836327348e-06, + "loss": 0.0903, + "step": 410 + }, + { + "epoch": 0.13, + "learning_rate": 2.095808383233533e-06, + "loss": 0.0965, + "step": 420 + }, + { + "epoch": 0.13, + "learning_rate": 2.1457085828343317e-06, + "loss": 0.1268, + "step": 430 + }, + { + "epoch": 0.13, + "learning_rate": 2.19560878243513e-06, + "loss": 0.1247, + "step": 440 + }, + { + "epoch": 0.13, + "learning_rate": 2.2455089820359283e-06, + "loss": 0.101, + "step": 450 + }, + { + "epoch": 0.14, + "learning_rate": 2.2954091816367268e-06, + "loss": 0.0945, + "step": 460 + }, + { + "epoch": 0.14, + "learning_rate": 2.3453093812375253e-06, + "loss": 0.166, + "step": 470 + }, + { + "epoch": 0.14, + "learning_rate": 2.3952095808383237e-06, + "loss": 0.1141, + "step": 480 + }, + { + "epoch": 0.15, + "learning_rate": 2.445109780439122e-06, + "loss": 0.1131, + "step": 490 + }, + { + "epoch": 0.15, + "learning_rate": 2.4950099800399203e-06, + "loss": 0.0881, + "step": 500 + }, + { + "epoch": 0.15, + "learning_rate": 2.5449101796407188e-06, + "loss": 0.0874, + "step": 510 + }, + { + "epoch": 0.16, + "learning_rate": 2.594810379241517e-06, + "loss": 0.0625, + "step": 520 + }, + { + "epoch": 0.16, + "learning_rate": 2.6447105788423157e-06, + "loss": 0.093, + "step": 530 + }, + { + "epoch": 0.16, + "learning_rate": 2.694610778443114e-06, + "loss": 0.0936, + "step": 540 + }, + { + "epoch": 0.16, + "learning_rate": 2.7445109780439123e-06, + "loss": 0.0886, + "step": 550 + }, + { + "epoch": 0.17, + "learning_rate": 2.7944111776447107e-06, + "loss": 0.0901, + "step": 560 + }, + { + "epoch": 0.17, + "learning_rate": 2.8443113772455092e-06, + "loss": 0.0808, + "step": 570 + }, + { + "epoch": 0.17, + "learning_rate": 2.8942115768463077e-06, + "loss": 0.0486, + "step": 580 + }, + { + "epoch": 0.18, + "learning_rate": 2.944111776447106e-06, + "loss": 0.0889, + "step": 590 + }, + { + "epoch": 0.18, + "learning_rate": 2.9940119760479042e-06, + "loss": 0.0897, + "step": 600 + }, + { + "epoch": 0.18, + "eval_f1": 0.9655660557043999, + "eval_f1_0": 0.988741287133544, + "eval_f1_1": 0.9930321299437529, + "eval_f1_2": 0.8836948791359373, + "eval_f1_3": 0.9967959266043652, + "eval_loss": 0.055430300533771515, + "eval_runtime": 153.1295, + "eval_samples_per_second": 77.503, + "eval_steps_per_second": 0.307, + "step": 600 + }, + { + "epoch": 0.18, + "learning_rate": 3.0439121756487027e-06, + "loss": 0.0423, + "step": 610 + }, + { + "epoch": 0.19, + "learning_rate": 3.093812375249501e-06, + "loss": 0.0778, + "step": 620 + }, + { + "epoch": 0.19, + "learning_rate": 3.1437125748502993e-06, + "loss": 0.0712, + "step": 630 + }, + { + "epoch": 0.19, + "learning_rate": 3.193612774451098e-06, + "loss": 0.0442, + "step": 640 + }, + { + "epoch": 0.19, + "learning_rate": 3.2435129740518966e-06, + "loss": 0.0853, + "step": 650 + }, + { + "epoch": 0.2, + "learning_rate": 3.2934131736526947e-06, + "loss": 0.0601, + "step": 660 + }, + { + "epoch": 0.2, + "learning_rate": 3.343313373253493e-06, + "loss": 0.1203, + "step": 670 + }, + { + "epoch": 0.2, + "learning_rate": 3.3932135728542912e-06, + "loss": 0.0914, + "step": 680 + }, + { + "epoch": 0.21, + "learning_rate": 3.44311377245509e-06, + "loss": 0.0317, + "step": 690 + }, + { + "epoch": 0.21, + "learning_rate": 3.4930139720558886e-06, + "loss": 0.0712, + "step": 700 + }, + { + "epoch": 0.21, + "learning_rate": 3.5429141716566867e-06, + "loss": 0.0533, + "step": 710 + }, + { + "epoch": 0.22, + "learning_rate": 3.592814371257485e-06, + "loss": 0.0093, + "step": 720 + }, + { + "epoch": 0.22, + "learning_rate": 3.642714570858284e-06, + "loss": 0.0381, + "step": 730 + }, + { + "epoch": 0.22, + "learning_rate": 3.692614770459082e-06, + "loss": 0.0584, + "step": 740 + }, + { + "epoch": 0.22, + "learning_rate": 3.7425149700598806e-06, + "loss": 0.0565, + "step": 750 + }, + { + "epoch": 0.23, + "learning_rate": 3.7924151696606787e-06, + "loss": 0.041, + "step": 760 + }, + { + "epoch": 0.23, + "learning_rate": 3.842315369261477e-06, + "loss": 0.0442, + "step": 770 + }, + { + "epoch": 0.23, + "learning_rate": 3.892215568862276e-06, + "loss": 0.0267, + "step": 780 + }, + { + "epoch": 0.24, + "learning_rate": 3.942115768463074e-06, + "loss": 0.0551, + "step": 790 + }, + { + "epoch": 0.24, + "learning_rate": 3.992015968063873e-06, + "loss": 0.0935, + "step": 800 + }, + { + "epoch": 0.24, + "eval_f1": 0.9565196786765503, + "eval_f1_0": 0.9843092266751751, + "eval_f1_1": 0.9943184551123322, + "eval_f1_2": 0.8496552625559586, + "eval_f1_3": 0.9977957703627355, + "eval_loss": 0.10954605042934418, + "eval_runtime": 153.1467, + "eval_samples_per_second": 77.494, + "eval_steps_per_second": 0.307, + "step": 800 + }, + { + "epoch": 0.24, + "learning_rate": 4.041916167664671e-06, + "loss": 0.0896, + "step": 810 + }, + { + "epoch": 0.25, + "learning_rate": 4.0918163672654696e-06, + "loss": 0.0404, + "step": 820 + }, + { + "epoch": 0.25, + "learning_rate": 4.141716566866268e-06, + "loss": 0.0374, + "step": 830 + }, + { + "epoch": 0.25, + "learning_rate": 4.191616766467066e-06, + "loss": 0.0221, + "step": 840 + }, + { + "epoch": 0.25, + "learning_rate": 4.241516966067864e-06, + "loss": 0.0323, + "step": 850 + }, + { + "epoch": 0.26, + "learning_rate": 4.2914171656686635e-06, + "loss": 0.0164, + "step": 860 + }, + { + "epoch": 0.26, + "learning_rate": 4.341317365269461e-06, + "loss": 0.0569, + "step": 870 + }, + { + "epoch": 0.26, + "learning_rate": 4.39121756487026e-06, + "loss": 0.0829, + "step": 880 + }, + { + "epoch": 0.27, + "learning_rate": 4.441117764471058e-06, + "loss": 0.0326, + "step": 890 + }, + { + "epoch": 0.27, + "learning_rate": 4.4910179640718566e-06, + "loss": 0.0111, + "step": 900 + }, + { + "epoch": 0.27, + "learning_rate": 4.540918163672655e-06, + "loss": 0.0548, + "step": 910 + }, + { + "epoch": 0.28, + "learning_rate": 4.5908183632734535e-06, + "loss": 0.0253, + "step": 920 + }, + { + "epoch": 0.28, + "learning_rate": 4.640718562874252e-06, + "loss": 0.0611, + "step": 930 + }, + { + "epoch": 0.28, + "learning_rate": 4.6906187624750505e-06, + "loss": 0.0753, + "step": 940 + }, + { + "epoch": 0.28, + "learning_rate": 4.740518962075848e-06, + "loss": 0.083, + "step": 950 + }, + { + "epoch": 0.29, + "learning_rate": 4.7904191616766475e-06, + "loss": 0.0605, + "step": 960 + }, + { + "epoch": 0.29, + "learning_rate": 4.840319361277446e-06, + "loss": 0.0613, + "step": 970 + }, + { + "epoch": 0.29, + "learning_rate": 4.890219560878244e-06, + "loss": 0.007, + "step": 980 + }, + { + "epoch": 0.3, + "learning_rate": 4.940119760479042e-06, + "loss": 0.0607, + "step": 990 + }, + { + "epoch": 0.3, + "learning_rate": 4.9900199600798405e-06, + "loss": 0.0206, + "step": 1000 + }, + { + "epoch": 0.3, + "eval_f1": 0.979305315894786, + "eval_f1_0": 0.9936102682168653, + "eval_f1_1": 0.9929366998906228, + "eval_f1_2": 0.9321544346109409, + "eval_f1_3": 0.9985198608607149, + "eval_loss": 0.04428563266992569, + "eval_runtime": 153.0733, + "eval_samples_per_second": 77.531, + "eval_steps_per_second": 0.307, + "step": 1000 + }, + { + "epoch": 0.3, + "learning_rate": 4.997897613791654e-06, + "loss": 0.0793, + "step": 1010 + }, + { + "epoch": 0.31, + "learning_rate": 4.995269631031221e-06, + "loss": 0.0324, + "step": 1020 + }, + { + "epoch": 0.31, + "learning_rate": 4.992641648270788e-06, + "loss": 0.0165, + "step": 1030 + }, + { + "epoch": 0.31, + "learning_rate": 4.990013665510355e-06, + "loss": 0.0414, + "step": 1040 + }, + { + "epoch": 0.31, + "learning_rate": 4.987385682749922e-06, + "loss": 0.0346, + "step": 1050 + }, + { + "epoch": 0.32, + "learning_rate": 4.984757699989489e-06, + "loss": 0.0267, + "step": 1060 + }, + { + "epoch": 0.32, + "learning_rate": 4.982129717229056e-06, + "loss": 0.0739, + "step": 1070 + }, + { + "epoch": 0.32, + "learning_rate": 4.979501734468623e-06, + "loss": 0.1384, + "step": 1080 + }, + { + "epoch": 0.33, + "learning_rate": 4.97687375170819e-06, + "loss": 0.0517, + "step": 1090 + }, + { + "epoch": 0.33, + "learning_rate": 4.974245768947757e-06, + "loss": 0.007, + "step": 1100 + }, + { + "epoch": 0.33, + "learning_rate": 4.971617786187323e-06, + "loss": 0.0662, + "step": 1110 + }, + { + "epoch": 0.34, + "learning_rate": 4.96898980342689e-06, + "loss": 0.0194, + "step": 1120 + }, + { + "epoch": 0.34, + "learning_rate": 4.966361820666457e-06, + "loss": 0.0284, + "step": 1130 + }, + { + "epoch": 0.34, + "learning_rate": 4.963733837906023e-06, + "loss": 0.0335, + "step": 1140 + }, + { + "epoch": 0.34, + "learning_rate": 4.96110585514559e-06, + "loss": 0.0217, + "step": 1150 + }, + { + "epoch": 0.35, + "learning_rate": 4.958477872385157e-06, + "loss": 0.0608, + "step": 1160 + }, + { + "epoch": 0.35, + "learning_rate": 4.955849889624724e-06, + "loss": 0.1114, + "step": 1170 + }, + { + "epoch": 0.35, + "learning_rate": 4.953221906864291e-06, + "loss": 0.0605, + "step": 1180 + }, + { + "epoch": 0.36, + "learning_rate": 4.950593924103858e-06, + "loss": 0.0243, + "step": 1190 + }, + { + "epoch": 0.36, + "learning_rate": 4.947965941343425e-06, + "loss": 0.0096, + "step": 1200 + }, + { + "epoch": 0.36, + "eval_f1": 0.9848431755266841, + "eval_f1_0": 0.9953030664489017, + "eval_f1_1": 0.9963201860243812, + "eval_f1_2": 0.9497379775953829, + "eval_f1_3": 0.9980114720380705, + "eval_loss": 0.035134535282850266, + "eval_runtime": 153.0723, + "eval_samples_per_second": 77.532, + "eval_steps_per_second": 0.307, + "step": 1200 + }, + { + "epoch": 0.36, + "learning_rate": 4.945337958582992e-06, + "loss": 0.0517, + "step": 1210 + }, + { + "epoch": 0.37, + "learning_rate": 4.942709975822559e-06, + "loss": 0.03, + "step": 1220 + }, + { + "epoch": 0.37, + "learning_rate": 4.940081993062125e-06, + "loss": 0.0065, + "step": 1230 + }, + { + "epoch": 0.37, + "learning_rate": 4.937454010301692e-06, + "loss": 0.0273, + "step": 1240 + }, + { + "epoch": 0.37, + "learning_rate": 4.934826027541259e-06, + "loss": 0.0545, + "step": 1250 + }, + { + "epoch": 0.38, + "learning_rate": 4.932198044780826e-06, + "loss": 0.0429, + "step": 1260 + }, + { + "epoch": 0.38, + "learning_rate": 4.929570062020393e-06, + "loss": 0.0661, + "step": 1270 + }, + { + "epoch": 0.38, + "learning_rate": 4.92694207925996e-06, + "loss": 0.0335, + "step": 1280 + }, + { + "epoch": 0.39, + "learning_rate": 4.924314096499527e-06, + "loss": 0.0586, + "step": 1290 + }, + { + "epoch": 0.39, + "learning_rate": 4.9216861137390944e-06, + "loss": 0.0199, + "step": 1300 + }, + { + "epoch": 0.39, + "learning_rate": 4.9190581309786614e-06, + "loss": 0.0438, + "step": 1310 + }, + { + "epoch": 0.4, + "learning_rate": 4.9164301482182284e-06, + "loss": 0.0341, + "step": 1320 + }, + { + "epoch": 0.4, + "learning_rate": 4.913802165457795e-06, + "loss": 0.0414, + "step": 1330 + }, + { + "epoch": 0.4, + "learning_rate": 4.911174182697362e-06, + "loss": 0.0236, + "step": 1340 + }, + { + "epoch": 0.4, + "learning_rate": 4.908546199936929e-06, + "loss": 0.0204, + "step": 1350 + }, + { + "epoch": 0.41, + "learning_rate": 4.905918217176496e-06, + "loss": 0.0594, + "step": 1360 + }, + { + "epoch": 0.41, + "learning_rate": 4.903290234416063e-06, + "loss": 0.0307, + "step": 1370 + }, + { + "epoch": 0.41, + "learning_rate": 4.90066225165563e-06, + "loss": 0.0289, + "step": 1380 + }, + { + "epoch": 0.42, + "learning_rate": 4.898034268895197e-06, + "loss": 0.0286, + "step": 1390 + }, + { + "epoch": 0.42, + "learning_rate": 4.895406286134764e-06, + "loss": 0.0165, + "step": 1400 + }, + { + "epoch": 0.42, + "eval_f1": 0.9806008044565914, + "eval_f1_0": 0.9935918800844773, + "eval_f1_1": 0.9939256908226702, + "eval_f1_2": 0.9369203232630038, + "eval_f1_3": 0.9979653236562142, + "eval_loss": 0.043006036430597305, + "eval_runtime": 153.0999, + "eval_samples_per_second": 77.518, + "eval_steps_per_second": 0.307, + "step": 1400 + }, + { + "epoch": 0.42, + "learning_rate": 4.892778303374331e-06, + "loss": 0.0495, + "step": 1410 + }, + { + "epoch": 0.43, + "learning_rate": 4.890150320613897e-06, + "loss": 0.0302, + "step": 1420 + }, + { + "epoch": 0.43, + "learning_rate": 4.887522337853464e-06, + "loss": 0.0133, + "step": 1430 + }, + { + "epoch": 0.43, + "learning_rate": 4.884894355093031e-06, + "loss": 0.0525, + "step": 1440 + }, + { + "epoch": 0.43, + "learning_rate": 4.882266372332598e-06, + "loss": 0.0161, + "step": 1450 + }, + { + "epoch": 0.44, + "learning_rate": 4.879638389572165e-06, + "loss": 0.0087, + "step": 1460 + }, + { + "epoch": 0.44, + "learning_rate": 4.877010406811732e-06, + "loss": 0.0405, + "step": 1470 + }, + { + "epoch": 0.44, + "learning_rate": 4.874382424051299e-06, + "loss": 0.0454, + "step": 1480 + }, + { + "epoch": 0.45, + "learning_rate": 4.871754441290866e-06, + "loss": 0.0155, + "step": 1490 + }, + { + "epoch": 0.45, + "learning_rate": 4.869126458530433e-06, + "loss": 0.0502, + "step": 1500 + }, + { + "epoch": 0.45, + "learning_rate": 4.86649847577e-06, + "loss": 0.0055, + "step": 1510 + }, + { + "epoch": 0.46, + "learning_rate": 4.863870493009566e-06, + "loss": 0.0089, + "step": 1520 + }, + { + "epoch": 0.46, + "learning_rate": 4.861242510249133e-06, + "loss": 0.0225, + "step": 1530 + }, + { + "epoch": 0.46, + "learning_rate": 4.8586145274887e-06, + "loss": 0.0295, + "step": 1540 + }, + { + "epoch": 0.46, + "learning_rate": 4.855986544728267e-06, + "loss": 0.0668, + "step": 1550 + }, + { + "epoch": 0.47, + "learning_rate": 4.853358561967833e-06, + "loss": 0.0125, + "step": 1560 + }, + { + "epoch": 0.47, + "learning_rate": 4.8507305792074e-06, + "loss": 0.006, + "step": 1570 + }, + { + "epoch": 0.47, + "learning_rate": 4.848102596446967e-06, + "loss": 0.0631, + "step": 1580 + }, + { + "epoch": 0.48, + "learning_rate": 4.845474613686534e-06, + "loss": 0.0364, + "step": 1590 + }, + { + "epoch": 0.48, + "learning_rate": 4.842846630926101e-06, + "loss": 0.0656, + "step": 1600 + }, + { + "epoch": 0.48, + "eval_f1": 0.9573566037869359, + "eval_f1_0": 0.9845514174634512, + "eval_f1_1": 0.9949111421936536, + "eval_f1_2": 0.8514684546002883, + "eval_f1_3": 0.9984954008903504, + "eval_loss": 0.11157545447349548, + "eval_runtime": 153.0673, + "eval_samples_per_second": 77.535, + "eval_steps_per_second": 0.307, + "step": 1600 + }, + { + "epoch": 0.48, + "learning_rate": 4.840218648165668e-06, + "loss": 0.0364, + "step": 1610 + }, + { + "epoch": 0.49, + "learning_rate": 4.837590665405235e-06, + "loss": 0.0274, + "step": 1620 + }, + { + "epoch": 0.49, + "learning_rate": 4.834962682644802e-06, + "loss": 0.0171, + "step": 1630 + }, + { + "epoch": 0.49, + "learning_rate": 4.832334699884369e-06, + "loss": 0.0163, + "step": 1640 + }, + { + "epoch": 0.49, + "learning_rate": 4.8297067171239355e-06, + "loss": 0.0106, + "step": 1650 + }, + { + "epoch": 0.5, + "learning_rate": 4.8270787343635026e-06, + "loss": 0.022, + "step": 1660 + }, + { + "epoch": 0.5, + "learning_rate": 4.8244507516030696e-06, + "loss": 0.0448, + "step": 1670 + }, + { + "epoch": 0.5, + "learning_rate": 4.821822768842637e-06, + "loss": 0.0607, + "step": 1680 + }, + { + "epoch": 0.51, + "learning_rate": 4.819194786082204e-06, + "loss": 0.0159, + "step": 1690 + }, + { + "epoch": 0.51, + "learning_rate": 4.816566803321771e-06, + "loss": 0.009, + "step": 1700 + }, + { + "epoch": 0.51, + "learning_rate": 4.813938820561338e-06, + "loss": 0.041, + "step": 1710 + }, + { + "epoch": 0.52, + "learning_rate": 4.811310837800905e-06, + "loss": 0.0226, + "step": 1720 + }, + { + "epoch": 0.52, + "learning_rate": 4.808682855040472e-06, + "loss": 0.02, + "step": 1730 + }, + { + "epoch": 0.52, + "learning_rate": 4.806054872280038e-06, + "loss": 0.0264, + "step": 1740 + }, + { + "epoch": 0.52, + "learning_rate": 4.803426889519605e-06, + "loss": 0.0642, + "step": 1750 + }, + { + "epoch": 0.53, + "learning_rate": 4.800798906759172e-06, + "loss": 0.0672, + "step": 1760 + }, + { + "epoch": 0.53, + "learning_rate": 4.798170923998739e-06, + "loss": 0.0273, + "step": 1770 + }, + { + "epoch": 0.53, + "learning_rate": 4.795542941238306e-06, + "loss": 0.0089, + "step": 1780 + }, + { + "epoch": 0.54, + "learning_rate": 4.792914958477873e-06, + "loss": 0.0203, + "step": 1790 + }, + { + "epoch": 0.54, + "learning_rate": 4.79028697571744e-06, + "loss": 0.0107, + "step": 1800 + }, + { + "epoch": 0.54, + "eval_f1": 0.9844617361811707, + "eval_f1_0": 0.9950323086176547, + "eval_f1_1": 0.9973182942970593, + "eval_f1_2": 0.9468495385821599, + "eval_f1_3": 0.9986468032278089, + "eval_loss": 0.03598302602767944, + "eval_runtime": 153.2134, + "eval_samples_per_second": 77.461, + "eval_steps_per_second": 0.307, + "step": 1800 + }, + { + "epoch": 0.54, + "learning_rate": 4.787658992957007e-06, + "loss": 0.0107, + "step": 1810 + }, + { + "epoch": 0.55, + "learning_rate": 4.785031010196574e-06, + "loss": 0.0255, + "step": 1820 + }, + { + "epoch": 0.55, + "learning_rate": 4.782403027436141e-06, + "loss": 0.006, + "step": 1830 + }, + { + "epoch": 0.55, + "learning_rate": 4.779775044675707e-06, + "loss": 0.0344, + "step": 1840 + }, + { + "epoch": 0.55, + "learning_rate": 4.777147061915274e-06, + "loss": 0.0458, + "step": 1850 + }, + { + "epoch": 0.56, + "learning_rate": 4.774519079154841e-06, + "loss": 0.0352, + "step": 1860 + }, + { + "epoch": 0.56, + "learning_rate": 4.771891096394408e-06, + "loss": 0.0799, + "step": 1870 + }, + { + "epoch": 0.56, + "learning_rate": 4.769263113633975e-06, + "loss": 0.0626, + "step": 1880 + }, + { + "epoch": 0.57, + "learning_rate": 4.766635130873542e-06, + "loss": 0.0176, + "step": 1890 + }, + { + "epoch": 0.57, + "learning_rate": 4.764007148113109e-06, + "loss": 0.0381, + "step": 1900 + }, + { + "epoch": 0.57, + "learning_rate": 4.761379165352676e-06, + "loss": 0.0297, + "step": 1910 + }, + { + "epoch": 0.58, + "learning_rate": 4.758751182592243e-06, + "loss": 0.0794, + "step": 1920 + }, + { + "epoch": 0.58, + "learning_rate": 4.756123199831809e-06, + "loss": 0.0186, + "step": 1930 + }, + { + "epoch": 0.58, + "learning_rate": 4.753495217071376e-06, + "loss": 0.0342, + "step": 1940 + }, + { + "epoch": 0.58, + "learning_rate": 4.750867234310943e-06, + "loss": 0.0705, + "step": 1950 + }, + { + "epoch": 0.59, + "learning_rate": 4.74823925155051e-06, + "loss": 0.0239, + "step": 1960 + }, + { + "epoch": 0.59, + "learning_rate": 4.745611268790077e-06, + "loss": 0.0408, + "step": 1970 + }, + { + "epoch": 0.59, + "learning_rate": 4.742983286029644e-06, + "loss": 0.0492, + "step": 1980 + }, + { + "epoch": 0.6, + "learning_rate": 4.7403553032692105e-06, + "loss": 0.1006, + "step": 1990 + }, + { + "epoch": 0.6, + "learning_rate": 4.7377273205087775e-06, + "loss": 0.0307, + "step": 2000 + }, + { + "epoch": 0.6, + "eval_f1": 0.9852587720651101, + "eval_f1_0": 0.9952924145851145, + "eval_f1_1": 0.997425129509855, + "eval_f1_2": 0.949825877240174, + "eval_f1_3": 0.9984916669252971, + "eval_loss": 0.03334509953856468, + "eval_runtime": 153.1318, + "eval_samples_per_second": 77.502, + "eval_steps_per_second": 0.307, + "step": 2000 + }, + { + "epoch": 0.6, + "learning_rate": 4.7350993377483445e-06, + "loss": 0.0254, + "step": 2010 + }, + { + "epoch": 0.61, + "learning_rate": 4.7324713549879115e-06, + "loss": 0.0111, + "step": 2020 + }, + { + "epoch": 0.61, + "learning_rate": 4.7298433722274785e-06, + "loss": 0.0662, + "step": 2030 + }, + { + "epoch": 0.61, + "learning_rate": 4.7272153894670456e-06, + "loss": 0.0178, + "step": 2040 + }, + { + "epoch": 0.61, + "learning_rate": 4.7245874067066126e-06, + "loss": 0.0455, + "step": 2050 + }, + { + "epoch": 0.62, + "learning_rate": 4.721959423946179e-06, + "loss": 0.0158, + "step": 2060 + }, + { + "epoch": 0.62, + "learning_rate": 4.719331441185746e-06, + "loss": 0.005, + "step": 2070 + }, + { + "epoch": 0.62, + "learning_rate": 4.716703458425313e-06, + "loss": 0.0226, + "step": 2080 + }, + { + "epoch": 0.63, + "learning_rate": 4.71407547566488e-06, + "loss": 0.0296, + "step": 2090 + }, + { + "epoch": 0.63, + "learning_rate": 4.711447492904447e-06, + "loss": 0.0034, + "step": 2100 + }, + { + "epoch": 0.63, + "learning_rate": 4.708819510144014e-06, + "loss": 0.0272, + "step": 2110 + }, + { + "epoch": 0.64, + "learning_rate": 4.706191527383581e-06, + "loss": 0.0238, + "step": 2120 + }, + { + "epoch": 0.64, + "learning_rate": 4.703563544623148e-06, + "loss": 0.0169, + "step": 2130 + }, + { + "epoch": 0.64, + "learning_rate": 4.700935561862715e-06, + "loss": 0.0296, + "step": 2140 + }, + { + "epoch": 0.64, + "learning_rate": 4.698307579102282e-06, + "loss": 0.0354, + "step": 2150 + }, + { + "epoch": 0.65, + "learning_rate": 4.695679596341848e-06, + "loss": 0.0073, + "step": 2160 + }, + { + "epoch": 0.65, + "learning_rate": 4.693051613581415e-06, + "loss": 0.0033, + "step": 2170 + }, + { + "epoch": 0.65, + "learning_rate": 4.690423630820982e-06, + "loss": 0.0296, + "step": 2180 + }, + { + "epoch": 0.66, + "learning_rate": 4.687795648060549e-06, + "loss": 0.014, + "step": 2190 + }, + { + "epoch": 0.66, + "learning_rate": 4.685167665300116e-06, + "loss": 0.0143, + "step": 2200 + }, + { + "epoch": 0.66, + "eval_f1": 0.988073153460117, + "eval_f1_0": 0.9962023415252138, + "eval_f1_1": 0.9963215311004785, + "eval_f1_2": 0.9614893673366701, + "eval_f1_3": 0.9982793738781053, + "eval_loss": 0.026542503386735916, + "eval_runtime": 153.1413, + "eval_samples_per_second": 77.497, + "eval_steps_per_second": 0.307, + "step": 2200 + }, + { + "epoch": 0.66, + "learning_rate": 4.682539682539683e-06, + "loss": 0.0228, + "step": 2210 + }, + { + "epoch": 0.67, + "learning_rate": 4.67991169977925e-06, + "loss": 0.0045, + "step": 2220 + }, + { + "epoch": 0.67, + "learning_rate": 4.677283717018817e-06, + "loss": 0.0374, + "step": 2230 + }, + { + "epoch": 0.67, + "learning_rate": 4.674655734258384e-06, + "loss": 0.0206, + "step": 2240 + }, + { + "epoch": 0.67, + "learning_rate": 4.67202775149795e-06, + "loss": 0.0119, + "step": 2250 + }, + { + "epoch": 0.68, + "learning_rate": 4.669399768737517e-06, + "loss": 0.0301, + "step": 2260 + }, + { + "epoch": 0.68, + "learning_rate": 4.666771785977084e-06, + "loss": 0.017, + "step": 2270 + }, + { + "epoch": 0.68, + "learning_rate": 4.664143803216651e-06, + "loss": 0.0234, + "step": 2280 + }, + { + "epoch": 0.69, + "learning_rate": 4.661515820456218e-06, + "loss": 0.0046, + "step": 2290 + }, + { + "epoch": 0.69, + "learning_rate": 4.658887837695785e-06, + "loss": 0.0356, + "step": 2300 + }, + { + "epoch": 0.69, + "learning_rate": 4.656259854935352e-06, + "loss": 0.03, + "step": 2310 + }, + { + "epoch": 0.7, + "learning_rate": 4.653631872174919e-06, + "loss": 0.0106, + "step": 2320 + }, + { + "epoch": 0.7, + "learning_rate": 4.651003889414486e-06, + "loss": 0.0374, + "step": 2330 + }, + { + "epoch": 0.7, + "learning_rate": 4.648375906654053e-06, + "loss": 0.0077, + "step": 2340 + }, + { + "epoch": 0.7, + "learning_rate": 4.6457479238936195e-06, + "loss": 0.0295, + "step": 2350 + }, + { + "epoch": 0.71, + "learning_rate": 4.6431199411331865e-06, + "loss": 0.0063, + "step": 2360 + }, + { + "epoch": 0.71, + "learning_rate": 4.6404919583727535e-06, + "loss": 0.0106, + "step": 2370 + }, + { + "epoch": 0.71, + "learning_rate": 4.6378639756123205e-06, + "loss": 0.0349, + "step": 2380 + }, + { + "epoch": 0.72, + "learning_rate": 4.6352359928518875e-06, + "loss": 0.003, + "step": 2390 + }, + { + "epoch": 0.72, + "learning_rate": 4.6326080100914545e-06, + "loss": 0.0121, + "step": 2400 + }, + { + "epoch": 0.72, + "eval_f1": 0.9885120168169891, + "eval_f1_0": 0.9963461094236451, + "eval_f1_1": 0.9977868941982138, + "eval_f1_2": 0.9611902612086393, + "eval_f1_3": 0.9987248024374579, + "eval_loss": 0.024830136448144913, + "eval_runtime": 153.1267, + "eval_samples_per_second": 77.504, + "eval_steps_per_second": 0.307, + "step": 2400 + }, + { + "epoch": 0.72, + "learning_rate": 4.629980027331021e-06, + "loss": 0.0369, + "step": 2410 + }, + { + "epoch": 0.72, + "learning_rate": 4.627352044570588e-06, + "loss": 0.0023, + "step": 2420 + }, + { + "epoch": 0.73, + "learning_rate": 4.624724061810155e-06, + "loss": 0.0391, + "step": 2430 + }, + { + "epoch": 0.73, + "learning_rate": 4.622096079049722e-06, + "loss": 0.0066, + "step": 2440 + }, + { + "epoch": 0.73, + "learning_rate": 4.619468096289289e-06, + "loss": 0.0297, + "step": 2450 + }, + { + "epoch": 0.74, + "learning_rate": 4.616840113528856e-06, + "loss": 0.0055, + "step": 2460 + }, + { + "epoch": 0.74, + "learning_rate": 4.614212130768423e-06, + "loss": 0.002, + "step": 2470 + }, + { + "epoch": 0.74, + "learning_rate": 4.611584148007989e-06, + "loss": 0.0128, + "step": 2480 + }, + { + "epoch": 0.75, + "learning_rate": 4.608956165247556e-06, + "loss": 0.007, + "step": 2490 + }, + { + "epoch": 0.75, + "learning_rate": 4.606328182487123e-06, + "loss": 0.0365, + "step": 2500 + }, + { + "epoch": 0.75, + "learning_rate": 4.60370019972669e-06, + "loss": 0.0178, + "step": 2510 + }, + { + "epoch": 0.75, + "learning_rate": 4.601072216966257e-06, + "loss": 0.0218, + "step": 2520 + }, + { + "epoch": 0.76, + "learning_rate": 4.598444234205824e-06, + "loss": 0.0153, + "step": 2530 + }, + { + "epoch": 0.76, + "learning_rate": 4.595816251445391e-06, + "loss": 0.0184, + "step": 2540 + }, + { + "epoch": 0.76, + "learning_rate": 4.593188268684958e-06, + "loss": 0.0124, + "step": 2550 + }, + { + "epoch": 0.77, + "learning_rate": 4.590560285924525e-06, + "loss": 0.029, + "step": 2560 + }, + { + "epoch": 0.77, + "learning_rate": 4.587932303164091e-06, + "loss": 0.0251, + "step": 2570 + }, + { + "epoch": 0.77, + "learning_rate": 4.585304320403658e-06, + "loss": 0.0308, + "step": 2580 + }, + { + "epoch": 0.78, + "learning_rate": 4.582676337643225e-06, + "loss": 0.017, + "step": 2590 + }, + { + "epoch": 0.78, + "learning_rate": 4.580048354882792e-06, + "loss": 0.0153, + "step": 2600 + }, + { + "epoch": 0.78, + "eval_f1": 0.9824421525927298, + "eval_f1_0": 0.9942432876005016, + "eval_f1_1": 0.9977419540340049, + "eval_f1_2": 0.9390652568676189, + "eval_f1_3": 0.9987181118687942, + "eval_loss": 0.04089111462235451, + "eval_runtime": 153.216, + "eval_samples_per_second": 77.459, + "eval_steps_per_second": 0.307, + "step": 2600 + }, + { + "epoch": 0.78, + "learning_rate": 4.577420372122359e-06, + "loss": 0.0061, + "step": 2610 + }, + { + "epoch": 0.78, + "learning_rate": 4.574792389361926e-06, + "loss": 0.0299, + "step": 2620 + }, + { + "epoch": 0.79, + "learning_rate": 4.572164406601493e-06, + "loss": 0.0334, + "step": 2630 + }, + { + "epoch": 0.79, + "learning_rate": 4.56953642384106e-06, + "loss": 0.0402, + "step": 2640 + }, + { + "epoch": 0.79, + "learning_rate": 4.566908441080627e-06, + "loss": 0.0485, + "step": 2650 + }, + { + "epoch": 0.8, + "learning_rate": 4.564280458320194e-06, + "loss": 0.0308, + "step": 2660 + }, + { + "epoch": 0.8, + "learning_rate": 4.56165247555976e-06, + "loss": 0.0065, + "step": 2670 + }, + { + "epoch": 0.8, + "learning_rate": 4.559024492799327e-06, + "loss": 0.0245, + "step": 2680 + }, + { + "epoch": 0.81, + "learning_rate": 4.5563965100388944e-06, + "loss": 0.0462, + "step": 2690 + }, + { + "epoch": 0.81, + "learning_rate": 4.5537685272784614e-06, + "loss": 0.0054, + "step": 2700 + }, + { + "epoch": 0.81, + "learning_rate": 4.5511405445180285e-06, + "loss": 0.0238, + "step": 2710 + }, + { + "epoch": 0.81, + "learning_rate": 4.5485125617575955e-06, + "loss": 0.0326, + "step": 2720 + }, + { + "epoch": 0.82, + "learning_rate": 4.5458845789971625e-06, + "loss": 0.0269, + "step": 2730 + }, + { + "epoch": 0.82, + "learning_rate": 4.5432565962367295e-06, + "loss": 0.0312, + "step": 2740 + }, + { + "epoch": 0.82, + "learning_rate": 4.5406286134762965e-06, + "loss": 0.0117, + "step": 2750 + }, + { + "epoch": 0.83, + "learning_rate": 4.538000630715863e-06, + "loss": 0.0494, + "step": 2760 + }, + { + "epoch": 0.83, + "learning_rate": 4.53537264795543e-06, + "loss": 0.0224, + "step": 2770 + }, + { + "epoch": 0.83, + "learning_rate": 4.532744665194997e-06, + "loss": 0.0316, + "step": 2780 + }, + { + "epoch": 0.84, + "learning_rate": 4.530116682434564e-06, + "loss": 0.0199, + "step": 2790 + }, + { + "epoch": 0.84, + "learning_rate": 4.527488699674131e-06, + "loss": 0.0149, + "step": 2800 + }, + { + "epoch": 0.84, + "eval_f1": 0.991081764730613, + "eval_f1_0": 0.9971726588118188, + "eval_f1_1": 0.9973864085569214, + "eval_f1_2": 0.9709476248740075, + "eval_f1_3": 0.9988203666797048, + "eval_loss": 0.017994137480854988, + "eval_runtime": 153.2012, + "eval_samples_per_second": 77.467, + "eval_steps_per_second": 0.307, + "step": 2800 + }, + { + "epoch": 0.84, + "learning_rate": 4.524860716913698e-06, + "loss": 0.0417, + "step": 2810 + }, + { + "epoch": 0.84, + "learning_rate": 4.522232734153265e-06, + "loss": 0.0199, + "step": 2820 + }, + { + "epoch": 0.85, + "learning_rate": 4.519604751392832e-06, + "loss": 0.0545, + "step": 2830 + }, + { + "epoch": 0.85, + "learning_rate": 4.516976768632398e-06, + "loss": 0.0061, + "step": 2840 + }, + { + "epoch": 0.85, + "learning_rate": 4.514348785871965e-06, + "loss": 0.0027, + "step": 2850 + }, + { + "epoch": 0.86, + "learning_rate": 4.511720803111532e-06, + "loss": 0.0053, + "step": 2860 + }, + { + "epoch": 0.86, + "learning_rate": 4.509092820351099e-06, + "loss": 0.0269, + "step": 2870 + }, + { + "epoch": 0.86, + "learning_rate": 4.506464837590666e-06, + "loss": 0.0437, + "step": 2880 + }, + { + "epoch": 0.87, + "learning_rate": 4.503836854830232e-06, + "loss": 0.0704, + "step": 2890 + }, + { + "epoch": 0.87, + "learning_rate": 4.501208872069799e-06, + "loss": 0.0695, + "step": 2900 + }, + { + "epoch": 0.87, + "learning_rate": 4.498580889309366e-06, + "loss": 0.0656, + "step": 2910 + }, + { + "epoch": 0.87, + "learning_rate": 4.495952906548933e-06, + "loss": 0.0253, + "step": 2920 + }, + { + "epoch": 0.88, + "learning_rate": 4.4933249237885e-06, + "loss": 0.0422, + "step": 2930 + }, + { + "epoch": 0.88, + "learning_rate": 4.490696941028067e-06, + "loss": 0.0096, + "step": 2940 + }, + { + "epoch": 0.88, + "learning_rate": 4.488068958267634e-06, + "loss": 0.0105, + "step": 2950 + }, + { + "epoch": 0.89, + "learning_rate": 4.485440975507201e-06, + "loss": 0.0183, + "step": 2960 + }, + { + "epoch": 0.89, + "learning_rate": 4.482812992746768e-06, + "loss": 0.004, + "step": 2970 + }, + { + "epoch": 0.89, + "learning_rate": 4.480185009986335e-06, + "loss": 0.0062, + "step": 2980 + }, + { + "epoch": 0.9, + "learning_rate": 4.477557027225901e-06, + "loss": 0.0264, + "step": 2990 + }, + { + "epoch": 0.9, + "learning_rate": 4.474929044465468e-06, + "loss": 0.0079, + "step": 3000 + }, + { + "epoch": 0.9, + "eval_f1": 0.9892999364102113, + "eval_f1_0": 0.9966928439249748, + "eval_f1_1": 0.9975627514849588, + "eval_f1_2": 0.9642531400753838, + "eval_f1_3": 0.9986910101555279, + "eval_loss": 0.025793807581067085, + "eval_runtime": 153.0919, + "eval_samples_per_second": 77.522, + "eval_steps_per_second": 0.307, + "step": 3000 + }, + { + "epoch": 0.9, + "learning_rate": 4.472301061705035e-06, + "loss": 0.0384, + "step": 3010 + }, + { + "epoch": 0.9, + "learning_rate": 4.469673078944602e-06, + "loss": 0.0265, + "step": 3020 + }, + { + "epoch": 0.91, + "learning_rate": 4.467045096184169e-06, + "loss": 0.0047, + "step": 3030 + }, + { + "epoch": 0.91, + "learning_rate": 4.464417113423736e-06, + "loss": 0.0509, + "step": 3040 + }, + { + "epoch": 0.91, + "learning_rate": 4.461789130663303e-06, + "loss": 0.0101, + "step": 3050 + }, + { + "epoch": 0.92, + "learning_rate": 4.45916114790287e-06, + "loss": 0.0111, + "step": 3060 + }, + { + "epoch": 0.92, + "learning_rate": 4.4565331651424374e-06, + "loss": 0.0157, + "step": 3070 + }, + { + "epoch": 0.92, + "learning_rate": 4.453905182382004e-06, + "loss": 0.0131, + "step": 3080 + }, + { + "epoch": 0.93, + "learning_rate": 4.451277199621571e-06, + "loss": 0.0279, + "step": 3090 + }, + { + "epoch": 0.93, + "learning_rate": 4.448649216861138e-06, + "loss": 0.0284, + "step": 3100 + }, + { + "epoch": 0.93, + "learning_rate": 4.446021234100705e-06, + "loss": 0.0179, + "step": 3110 + }, + { + "epoch": 0.93, + "learning_rate": 4.443393251340272e-06, + "loss": 0.0051, + "step": 3120 + }, + { + "epoch": 0.94, + "learning_rate": 4.440765268579839e-06, + "loss": 0.0403, + "step": 3130 + }, + { + "epoch": 0.94, + "learning_rate": 4.438137285819406e-06, + "loss": 0.0046, + "step": 3140 + }, + { + "epoch": 0.94, + "learning_rate": 4.435509303058973e-06, + "loss": 0.0419, + "step": 3150 + }, + { + "epoch": 0.95, + "learning_rate": 4.43288132029854e-06, + "loss": 0.0343, + "step": 3160 + }, + { + "epoch": 0.95, + "learning_rate": 4.430253337538107e-06, + "loss": 0.0065, + "step": 3170 + }, + { + "epoch": 0.95, + "learning_rate": 4.427625354777673e-06, + "loss": 0.009, + "step": 3180 + }, + { + "epoch": 0.96, + "learning_rate": 4.42499737201724e-06, + "loss": 0.0213, + "step": 3190 + }, + { + "epoch": 0.96, + "learning_rate": 4.422369389256807e-06, + "loss": 0.0035, + "step": 3200 + }, + { + "epoch": 0.96, + "eval_f1": 0.9925562042301975, + "eval_f1_0": 0.9976930515391278, + "eval_f1_1": 0.9978202731847592, + "eval_f1_2": 0.9758470449256423, + "eval_f1_3": 0.9988644472712606, + "eval_loss": 0.01738656684756279, + "eval_runtime": 153.1868, + "eval_samples_per_second": 77.474, + "eval_steps_per_second": 0.307, + "step": 3200 + }, + { + "epoch": 0.96, + "learning_rate": 4.419741406496374e-06, + "loss": 0.0339, + "step": 3210 + }, + { + "epoch": 0.96, + "learning_rate": 4.417113423735941e-06, + "loss": 0.0156, + "step": 3220 + }, + { + "epoch": 0.97, + "learning_rate": 4.414485440975508e-06, + "loss": 0.0349, + "step": 3230 + }, + { + "epoch": 0.97, + "learning_rate": 4.411857458215075e-06, + "loss": 0.0213, + "step": 3240 + }, + { + "epoch": 0.97, + "learning_rate": 4.409229475454642e-06, + "loss": 0.0423, + "step": 3250 + }, + { + "epoch": 0.98, + "learning_rate": 4.406601492694209e-06, + "loss": 0.0252, + "step": 3260 + }, + { + "epoch": 0.98, + "learning_rate": 4.403973509933775e-06, + "loss": 0.0048, + "step": 3270 + }, + { + "epoch": 0.98, + "learning_rate": 4.401345527173342e-06, + "loss": 0.0124, + "step": 3280 + }, + { + "epoch": 0.99, + "learning_rate": 4.398717544412909e-06, + "loss": 0.0348, + "step": 3290 + }, + { + "epoch": 0.99, + "learning_rate": 4.396089561652475e-06, + "loss": 0.0441, + "step": 3300 + }, + { + "epoch": 0.99, + "learning_rate": 4.393461578892042e-06, + "loss": 0.0357, + "step": 3310 + }, + { + "epoch": 0.99, + "learning_rate": 4.390833596131609e-06, + "loss": 0.0083, + "step": 3320 + }, + { + "epoch": 1.0, + "learning_rate": 4.388205613371176e-06, + "loss": 0.0088, + "step": 3330 + }, + { + "epoch": 1.0, + "learning_rate": 4.385577630610743e-06, + "loss": 0.0267, + "step": 3340 + }, + { + "epoch": 1.0, + "learning_rate": 4.38294964785031e-06, + "loss": 0.0085, + "step": 3350 + }, + { + "epoch": 1.01, + "learning_rate": 4.380321665089877e-06, + "loss": 0.0111, + "step": 3360 + }, + { + "epoch": 1.01, + "learning_rate": 4.377693682329444e-06, + "loss": 0.0256, + "step": 3370 + }, + { + "epoch": 1.01, + "learning_rate": 4.375065699569011e-06, + "loss": 0.0143, + "step": 3380 + }, + { + "epoch": 1.02, + "learning_rate": 4.372437716808578e-06, + "loss": 0.0104, + "step": 3390 + }, + { + "epoch": 1.02, + "learning_rate": 4.3698097340481445e-06, + "loss": 0.0056, + "step": 3400 + }, + { + "epoch": 1.02, + "eval_f1": 0.9900821647690017, + "eval_f1_0": 0.996891745919871, + "eval_f1_1": 0.9974857484254811, + "eval_f1_2": 0.9671803752422669, + "eval_f1_3": 0.9987707894883877, + "eval_loss": 0.01880444586277008, + "eval_runtime": 153.1416, + "eval_samples_per_second": 77.497, + "eval_steps_per_second": 0.307, + "step": 3400 + }, + { + "epoch": 1.02, + "learning_rate": 4.3671817512877115e-06, + "loss": 0.0173, + "step": 3410 + }, + { + "epoch": 1.02, + "learning_rate": 4.3645537685272785e-06, + "loss": 0.0076, + "step": 3420 + }, + { + "epoch": 1.03, + "learning_rate": 4.3619257857668456e-06, + "loss": 0.0086, + "step": 3430 + }, + { + "epoch": 1.03, + "learning_rate": 4.3592978030064126e-06, + "loss": 0.0024, + "step": 3440 + }, + { + "epoch": 1.03, + "learning_rate": 4.3566698202459796e-06, + "loss": 0.0133, + "step": 3450 + }, + { + "epoch": 1.04, + "learning_rate": 4.354041837485547e-06, + "loss": 0.0051, + "step": 3460 + }, + { + "epoch": 1.04, + "learning_rate": 4.351413854725114e-06, + "loss": 0.0022, + "step": 3470 + }, + { + "epoch": 1.04, + "learning_rate": 4.348785871964681e-06, + "loss": 0.0277, + "step": 3480 + }, + { + "epoch": 1.05, + "learning_rate": 4.346157889204248e-06, + "loss": 0.0177, + "step": 3490 + }, + { + "epoch": 1.05, + "learning_rate": 4.343529906443814e-06, + "loss": 0.0232, + "step": 3500 + }, + { + "epoch": 1.05, + "learning_rate": 4.340901923683381e-06, + "loss": 0.017, + "step": 3510 + }, + { + "epoch": 1.05, + "learning_rate": 4.338273940922948e-06, + "loss": 0.0088, + "step": 3520 + }, + { + "epoch": 1.06, + "learning_rate": 4.335645958162515e-06, + "loss": 0.0061, + "step": 3530 + }, + { + "epoch": 1.06, + "learning_rate": 4.333017975402082e-06, + "loss": 0.019, + "step": 3540 + }, + { + "epoch": 1.06, + "learning_rate": 4.330389992641649e-06, + "loss": 0.0248, + "step": 3550 + }, + { + "epoch": 1.07, + "learning_rate": 4.327762009881216e-06, + "loss": 0.0222, + "step": 3560 + }, + { + "epoch": 1.07, + "learning_rate": 4.325134027120783e-06, + "loss": 0.0079, + "step": 3570 + }, + { + "epoch": 1.07, + "learning_rate": 4.32250604436035e-06, + "loss": 0.0168, + "step": 3580 + }, + { + "epoch": 1.08, + "learning_rate": 4.319878061599916e-06, + "loss": 0.011, + "step": 3590 + }, + { + "epoch": 1.08, + "learning_rate": 4.317250078839483e-06, + "loss": 0.0041, + "step": 3600 + }, + { + "epoch": 1.08, + "eval_f1": 0.9923554131084382, + "eval_f1_0": 0.9975622957389756, + "eval_f1_1": 0.9968584617984346, + "eval_f1_2": 0.9760375374160667, + "eval_f1_3": 0.9989633574802759, + "eval_loss": 0.016382403671741486, + "eval_runtime": 153.1827, + "eval_samples_per_second": 77.476, + "eval_steps_per_second": 0.307, + "step": 3600 + }, + { + "epoch": 1.08, + "learning_rate": 4.31462209607905e-06, + "loss": 0.0037, + "step": 3610 + }, + { + "epoch": 1.08, + "learning_rate": 4.311994113318617e-06, + "loss": 0.0067, + "step": 3620 + }, + { + "epoch": 1.09, + "learning_rate": 4.309366130558184e-06, + "loss": 0.0026, + "step": 3630 + }, + { + "epoch": 1.09, + "learning_rate": 4.306738147797751e-06, + "loss": 0.043, + "step": 3640 + }, + { + "epoch": 1.09, + "learning_rate": 4.304110165037318e-06, + "loss": 0.0474, + "step": 3650 + }, + { + "epoch": 1.1, + "learning_rate": 4.301482182276885e-06, + "loss": 0.0371, + "step": 3660 + }, + { + "epoch": 1.1, + "learning_rate": 4.298854199516452e-06, + "loss": 0.0188, + "step": 3670 + }, + { + "epoch": 1.1, + "learning_rate": 4.296226216756019e-06, + "loss": 0.0068, + "step": 3680 + }, + { + "epoch": 1.11, + "learning_rate": 4.293598233995585e-06, + "loss": 0.0014, + "step": 3690 + }, + { + "epoch": 1.11, + "learning_rate": 4.290970251235152e-06, + "loss": 0.0022, + "step": 3700 + }, + { + "epoch": 1.11, + "learning_rate": 4.288342268474719e-06, + "loss": 0.0011, + "step": 3710 + }, + { + "epoch": 1.11, + "learning_rate": 4.2857142857142855e-06, + "loss": 0.0266, + "step": 3720 + }, + { + "epoch": 1.12, + "learning_rate": 4.2830863029538525e-06, + "loss": 0.0055, + "step": 3730 + }, + { + "epoch": 1.12, + "learning_rate": 4.2804583201934195e-06, + "loss": 0.004, + "step": 3740 + }, + { + "epoch": 1.12, + "learning_rate": 4.2778303374329865e-06, + "loss": 0.0155, + "step": 3750 + }, + { + "epoch": 1.13, + "learning_rate": 4.2752023546725535e-06, + "loss": 0.0435, + "step": 3760 + }, + { + "epoch": 1.13, + "learning_rate": 4.2725743719121205e-06, + "loss": 0.031, + "step": 3770 + }, + { + "epoch": 1.13, + "learning_rate": 4.2699463891516875e-06, + "loss": 0.0036, + "step": 3780 + }, + { + "epoch": 1.14, + "learning_rate": 4.2673184063912545e-06, + "loss": 0.0088, + "step": 3790 + }, + { + "epoch": 1.14, + "learning_rate": 4.2646904236308215e-06, + "loss": 0.0177, + "step": 3800 + }, + { + "epoch": 1.14, + "eval_f1": 0.9917306367952623, + "eval_f1_0": 0.9973764216025945, + "eval_f1_1": 0.9977408367725801, + "eval_f1_2": 0.9728820290446188, + "eval_f1_3": 0.9989232597612554, + "eval_loss": 0.014767622575163841, + "eval_runtime": 153.2056, + "eval_samples_per_second": 77.465, + "eval_steps_per_second": 0.307, + "step": 3800 + }, + { + "epoch": 1.14, + "learning_rate": 4.262062440870388e-06, + "loss": 0.0088, + "step": 3810 + }, + { + "epoch": 1.14, + "learning_rate": 4.259434458109955e-06, + "loss": 0.0242, + "step": 3820 + }, + { + "epoch": 1.15, + "learning_rate": 4.256806475349522e-06, + "loss": 0.0069, + "step": 3830 + }, + { + "epoch": 1.15, + "learning_rate": 4.254178492589089e-06, + "loss": 0.0234, + "step": 3840 + }, + { + "epoch": 1.15, + "learning_rate": 4.251550509828656e-06, + "loss": 0.0048, + "step": 3850 + }, + { + "epoch": 1.16, + "learning_rate": 4.248922527068223e-06, + "loss": 0.002, + "step": 3860 + }, + { + "epoch": 1.16, + "learning_rate": 4.24629454430779e-06, + "loss": 0.0259, + "step": 3870 + }, + { + "epoch": 1.16, + "learning_rate": 4.243666561547357e-06, + "loss": 0.005, + "step": 3880 + }, + { + "epoch": 1.17, + "learning_rate": 4.241038578786924e-06, + "loss": 0.0109, + "step": 3890 + }, + { + "epoch": 1.17, + "learning_rate": 4.238410596026491e-06, + "loss": 0.0076, + "step": 3900 + }, + { + "epoch": 1.17, + "learning_rate": 4.235782613266057e-06, + "loss": 0.0025, + "step": 3910 + }, + { + "epoch": 1.17, + "learning_rate": 4.233154630505624e-06, + "loss": 0.0032, + "step": 3920 + }, + { + "epoch": 1.18, + "learning_rate": 4.230526647745191e-06, + "loss": 0.0047, + "step": 3930 + }, + { + "epoch": 1.18, + "learning_rate": 4.227898664984758e-06, + "loss": 0.0038, + "step": 3940 + }, + { + "epoch": 1.18, + "learning_rate": 4.225270682224325e-06, + "loss": 0.018, + "step": 3950 + }, + { + "epoch": 1.19, + "learning_rate": 4.222642699463892e-06, + "loss": 0.0013, + "step": 3960 + }, + { + "epoch": 1.19, + "learning_rate": 4.220014716703459e-06, + "loss": 0.0257, + "step": 3970 + }, + { + "epoch": 1.19, + "learning_rate": 4.217386733943026e-06, + "loss": 0.0142, + "step": 3980 + }, + { + "epoch": 1.2, + "learning_rate": 4.214758751182593e-06, + "loss": 0.0101, + "step": 3990 + }, + { + "epoch": 1.2, + "learning_rate": 4.212130768422159e-06, + "loss": 0.0183, + "step": 4000 + }, + { + "epoch": 1.2, + "eval_f1": 0.9811645096843833, + "eval_f1_0": 0.9937597363504184, + "eval_f1_1": 0.9977700441964585, + "eval_f1_2": 0.9342515817388692, + "eval_f1_3": 0.9988766764517871, + "eval_loss": 0.042360611259937286, + "eval_runtime": 153.0928, + "eval_samples_per_second": 77.522, + "eval_steps_per_second": 0.307, + "step": 4000 + }, + { + "epoch": 1.2, + "learning_rate": 4.209502785661726e-06, + "loss": 0.028, + "step": 4010 + }, + { + "epoch": 1.2, + "learning_rate": 4.206874802901293e-06, + "loss": 0.0184, + "step": 4020 + }, + { + "epoch": 1.21, + "learning_rate": 4.20424682014086e-06, + "loss": 0.015, + "step": 4030 + }, + { + "epoch": 1.21, + "learning_rate": 4.201618837380427e-06, + "loss": 0.0203, + "step": 4040 + }, + { + "epoch": 1.21, + "learning_rate": 4.198990854619994e-06, + "loss": 0.0323, + "step": 4050 + }, + { + "epoch": 1.22, + "learning_rate": 4.196362871859561e-06, + "loss": 0.0032, + "step": 4060 + }, + { + "epoch": 1.22, + "learning_rate": 4.193734889099128e-06, + "loss": 0.03, + "step": 4070 + }, + { + "epoch": 1.22, + "learning_rate": 4.191106906338695e-06, + "loss": 0.0049, + "step": 4080 + }, + { + "epoch": 1.23, + "learning_rate": 4.188478923578262e-06, + "loss": 0.0201, + "step": 4090 + }, + { + "epoch": 1.23, + "learning_rate": 4.1858509408178285e-06, + "loss": 0.004, + "step": 4100 + }, + { + "epoch": 1.23, + "learning_rate": 4.1832229580573955e-06, + "loss": 0.011, + "step": 4110 + }, + { + "epoch": 1.23, + "learning_rate": 4.1805949752969625e-06, + "loss": 0.0484, + "step": 4120 + }, + { + "epoch": 1.24, + "learning_rate": 4.177966992536529e-06, + "loss": 0.0018, + "step": 4130 + }, + { + "epoch": 1.24, + "learning_rate": 4.175339009776096e-06, + "loss": 0.0382, + "step": 4140 + }, + { + "epoch": 1.24, + "learning_rate": 4.172711027015663e-06, + "loss": 0.0296, + "step": 4150 + }, + { + "epoch": 1.25, + "learning_rate": 4.17008304425523e-06, + "loss": 0.005, + "step": 4160 + }, + { + "epoch": 1.25, + "learning_rate": 4.167455061494797e-06, + "loss": 0.0112, + "step": 4170 + }, + { + "epoch": 1.25, + "learning_rate": 4.164827078734364e-06, + "loss": 0.0304, + "step": 4180 + }, + { + "epoch": 1.26, + "learning_rate": 4.162199095973931e-06, + "loss": 0.0041, + "step": 4190 + }, + { + "epoch": 1.26, + "learning_rate": 4.159571113213498e-06, + "loss": 0.0198, + "step": 4200 + }, + { + "epoch": 1.26, + "eval_f1": 0.9877866699812177, + "eval_f1_0": 0.9960761087184571, + "eval_f1_1": 0.9979039177369216, + "eval_f1_2": 0.9582030901134232, + "eval_f1_3": 0.9989635633560687, + "eval_loss": 0.02484200894832611, + "eval_runtime": 153.1816, + "eval_samples_per_second": 77.477, + "eval_steps_per_second": 0.307, + "step": 4200 + }, + { + "epoch": 1.26, + "learning_rate": 4.156943130453065e-06, + "loss": 0.0019, + "step": 4210 + }, + { + "epoch": 1.26, + "learning_rate": 4.154315147692632e-06, + "loss": 0.0014, + "step": 4220 + }, + { + "epoch": 1.27, + "learning_rate": 4.151687164932198e-06, + "loss": 0.0078, + "step": 4230 + }, + { + "epoch": 1.27, + "learning_rate": 4.149059182171765e-06, + "loss": 0.017, + "step": 4240 + }, + { + "epoch": 1.27, + "learning_rate": 4.146431199411332e-06, + "loss": 0.0196, + "step": 4250 + }, + { + "epoch": 1.28, + "learning_rate": 4.143803216650899e-06, + "loss": 0.0068, + "step": 4260 + }, + { + "epoch": 1.28, + "learning_rate": 4.141175233890466e-06, + "loss": 0.0041, + "step": 4270 + }, + { + "epoch": 1.28, + "learning_rate": 4.138547251130033e-06, + "loss": 0.0166, + "step": 4280 + }, + { + "epoch": 1.29, + "learning_rate": 4.1359192683696e-06, + "loss": 0.0024, + "step": 4290 + }, + { + "epoch": 1.29, + "learning_rate": 4.133291285609167e-06, + "loss": 0.0009, + "step": 4300 + }, + { + "epoch": 1.29, + "learning_rate": 4.130663302848734e-06, + "loss": 0.0029, + "step": 4310 + }, + { + "epoch": 1.29, + "learning_rate": 4.1280353200883e-06, + "loss": 0.0208, + "step": 4320 + }, + { + "epoch": 1.3, + "learning_rate": 4.125407337327867e-06, + "loss": 0.0264, + "step": 4330 + }, + { + "epoch": 1.3, + "learning_rate": 4.122779354567434e-06, + "loss": 0.0229, + "step": 4340 + }, + { + "epoch": 1.3, + "learning_rate": 4.120151371807001e-06, + "loss": 0.0401, + "step": 4350 + }, + { + "epoch": 1.31, + "learning_rate": 4.117523389046568e-06, + "loss": 0.007, + "step": 4360 + }, + { + "epoch": 1.31, + "learning_rate": 4.114895406286135e-06, + "loss": 0.0122, + "step": 4370 + }, + { + "epoch": 1.31, + "learning_rate": 4.112267423525702e-06, + "loss": 0.0035, + "step": 4380 + }, + { + "epoch": 1.32, + "learning_rate": 4.109639440765269e-06, + "loss": 0.0368, + "step": 4390 + }, + { + "epoch": 1.32, + "learning_rate": 4.107011458004836e-06, + "loss": 0.0132, + "step": 4400 + }, + { + "epoch": 1.32, + "eval_f1": 0.9830756707587374, + "eval_f1_0": 0.9944486878150933, + "eval_f1_1": 0.9969642565692834, + "eval_f1_2": 0.9420131319189023, + "eval_f1_3": 0.9988766067316703, + "eval_loss": 0.035511914640665054, + "eval_runtime": 153.2065, + "eval_samples_per_second": 77.464, + "eval_steps_per_second": 0.307, + "step": 4400 + }, + { + "epoch": 1.32, + "learning_rate": 4.104383475244403e-06, + "loss": 0.0035, + "step": 4410 + }, + { + "epoch": 1.32, + "learning_rate": 4.101755492483969e-06, + "loss": 0.0097, + "step": 4420 + }, + { + "epoch": 1.33, + "learning_rate": 4.099127509723536e-06, + "loss": 0.0144, + "step": 4430 + }, + { + "epoch": 1.33, + "learning_rate": 4.096499526963103e-06, + "loss": 0.0128, + "step": 4440 + }, + { + "epoch": 1.33, + "learning_rate": 4.09387154420267e-06, + "loss": 0.0019, + "step": 4450 + }, + { + "epoch": 1.34, + "learning_rate": 4.0912435614422374e-06, + "loss": 0.0019, + "step": 4460 + }, + { + "epoch": 1.34, + "learning_rate": 4.0886155786818044e-06, + "loss": 0.0032, + "step": 4470 + }, + { + "epoch": 1.34, + "learning_rate": 4.0859875959213715e-06, + "loss": 0.006, + "step": 4480 + }, + { + "epoch": 1.35, + "learning_rate": 4.0833596131609385e-06, + "loss": 0.0025, + "step": 4490 + }, + { + "epoch": 1.35, + "learning_rate": 4.0807316304005055e-06, + "loss": 0.0088, + "step": 4500 + }, + { + "epoch": 1.35, + "learning_rate": 4.078103647640072e-06, + "loss": 0.0329, + "step": 4510 + }, + { + "epoch": 1.35, + "learning_rate": 4.075475664879639e-06, + "loss": 0.0125, + "step": 4520 + }, + { + "epoch": 1.36, + "learning_rate": 4.072847682119206e-06, + "loss": 0.0029, + "step": 4530 + }, + { + "epoch": 1.36, + "learning_rate": 4.070219699358773e-06, + "loss": 0.0033, + "step": 4540 + }, + { + "epoch": 1.36, + "learning_rate": 4.067591716598339e-06, + "loss": 0.0233, + "step": 4550 + }, + { + "epoch": 1.37, + "learning_rate": 4.064963733837906e-06, + "loss": 0.0076, + "step": 4560 + }, + { + "epoch": 1.37, + "learning_rate": 4.062335751077473e-06, + "loss": 0.003, + "step": 4570 + }, + { + "epoch": 1.37, + "learning_rate": 4.05970776831704e-06, + "loss": 0.0207, + "step": 4580 + }, + { + "epoch": 1.38, + "learning_rate": 4.057079785556607e-06, + "loss": 0.0096, + "step": 4590 + }, + { + "epoch": 1.38, + "learning_rate": 4.054451802796174e-06, + "loss": 0.0023, + "step": 4600 + }, + { + "epoch": 1.38, + "eval_f1": 0.9931970642159123, + "eval_f1_0": 0.9978763521162372, + "eval_f1_1": 0.997958193777955, + "eval_f1_2": 0.9784022446005611, + "eval_f1_3": 0.9985514663688952, + "eval_loss": 0.014273732900619507, + "eval_runtime": 153.1815, + "eval_samples_per_second": 77.477, + "eval_steps_per_second": 0.307, + "step": 4600 + }, + { + "epoch": 1.38, + "learning_rate": 4.051823820035741e-06, + "loss": 0.018, + "step": 4610 + }, + { + "epoch": 1.38, + "learning_rate": 4.049195837275308e-06, + "loss": 0.0055, + "step": 4620 + }, + { + "epoch": 1.39, + "learning_rate": 4.046567854514875e-06, + "loss": 0.003, + "step": 4630 + }, + { + "epoch": 1.39, + "learning_rate": 4.043939871754441e-06, + "loss": 0.003, + "step": 4640 + }, + { + "epoch": 1.39, + "learning_rate": 4.041311888994008e-06, + "loss": 0.0047, + "step": 4650 + }, + { + "epoch": 1.4, + "learning_rate": 4.038683906233575e-06, + "loss": 0.007, + "step": 4660 + }, + { + "epoch": 1.4, + "learning_rate": 4.036055923473142e-06, + "loss": 0.0058, + "step": 4670 + }, + { + "epoch": 1.4, + "learning_rate": 4.033427940712709e-06, + "loss": 0.0071, + "step": 4680 + }, + { + "epoch": 1.41, + "learning_rate": 4.030799957952276e-06, + "loss": 0.0025, + "step": 4690 + }, + { + "epoch": 1.41, + "learning_rate": 4.028171975191843e-06, + "loss": 0.0286, + "step": 4700 + }, + { + "epoch": 1.41, + "learning_rate": 4.02554399243141e-06, + "loss": 0.0043, + "step": 4710 + }, + { + "epoch": 1.41, + "learning_rate": 4.022916009670977e-06, + "loss": 0.0193, + "step": 4720 + }, + { + "epoch": 1.42, + "learning_rate": 4.020288026910544e-06, + "loss": 0.0266, + "step": 4730 + }, + { + "epoch": 1.42, + "learning_rate": 4.01766004415011e-06, + "loss": 0.0044, + "step": 4740 + }, + { + "epoch": 1.42, + "learning_rate": 4.015032061389677e-06, + "loss": 0.0356, + "step": 4750 + }, + { + "epoch": 1.43, + "learning_rate": 4.012404078629244e-06, + "loss": 0.0149, + "step": 4760 + }, + { + "epoch": 1.43, + "learning_rate": 4.009776095868811e-06, + "loss": 0.018, + "step": 4770 + }, + { + "epoch": 1.43, + "learning_rate": 4.007148113108378e-06, + "loss": 0.0102, + "step": 4780 + }, + { + "epoch": 1.43, + "learning_rate": 4.004520130347945e-06, + "loss": 0.0258, + "step": 4790 + }, + { + "epoch": 1.44, + "learning_rate": 4.001892147587512e-06, + "loss": 0.023, + "step": 4800 + }, + { + "epoch": 1.44, + "eval_f1": 0.9880544825830188, + "eval_f1_0": 0.9961411610030081, + "eval_f1_1": 0.9981075603634558, + "eval_f1_2": 0.9589777020287306, + "eval_f1_3": 0.9989915069368808, + "eval_loss": 0.026987778022885323, + "eval_runtime": 153.188, + "eval_samples_per_second": 77.473, + "eval_steps_per_second": 0.307, + "step": 4800 + }, + { + "epoch": 1.44, + "learning_rate": 3.999264164827079e-06, + "loss": 0.0087, + "step": 4810 + }, + { + "epoch": 1.44, + "learning_rate": 3.996636182066646e-06, + "loss": 0.0044, + "step": 4820 + }, + { + "epoch": 1.45, + "learning_rate": 3.9940081993062126e-06, + "loss": 0.0106, + "step": 4830 + }, + { + "epoch": 1.45, + "learning_rate": 3.99138021654578e-06, + "loss": 0.0066, + "step": 4840 + }, + { + "epoch": 1.45, + "learning_rate": 3.988752233785347e-06, + "loss": 0.0112, + "step": 4850 + }, + { + "epoch": 1.46, + "learning_rate": 3.986124251024914e-06, + "loss": 0.0024, + "step": 4860 + }, + { + "epoch": 1.46, + "learning_rate": 3.983496268264481e-06, + "loss": 0.0477, + "step": 4870 + }, + { + "epoch": 1.46, + "learning_rate": 3.980868285504048e-06, + "loss": 0.0063, + "step": 4880 + }, + { + "epoch": 1.46, + "learning_rate": 3.978240302743615e-06, + "loss": 0.0086, + "step": 4890 + }, + { + "epoch": 1.47, + "learning_rate": 3.975612319983182e-06, + "loss": 0.0021, + "step": 4900 + }, + { + "epoch": 1.47, + "learning_rate": 3.972984337222749e-06, + "loss": 0.0055, + "step": 4910 + }, + { + "epoch": 1.47, + "learning_rate": 3.970356354462316e-06, + "loss": 0.0083, + "step": 4920 + }, + { + "epoch": 1.48, + "learning_rate": 3.967728371701882e-06, + "loss": 0.0447, + "step": 4930 + }, + { + "epoch": 1.48, + "learning_rate": 3.965100388941449e-06, + "loss": 0.004, + "step": 4940 + }, + { + "epoch": 1.48, + "learning_rate": 3.962472406181016e-06, + "loss": 0.0021, + "step": 4950 + }, + { + "epoch": 1.49, + "learning_rate": 3.959844423420583e-06, + "loss": 0.0125, + "step": 4960 + }, + { + "epoch": 1.49, + "learning_rate": 3.957216440660149e-06, + "loss": 0.0045, + "step": 4970 + }, + { + "epoch": 1.49, + "learning_rate": 3.954588457899716e-06, + "loss": 0.0166, + "step": 4980 + }, + { + "epoch": 1.49, + "learning_rate": 3.951960475139283e-06, + "loss": 0.0134, + "step": 4990 + }, + { + "epoch": 1.5, + "learning_rate": 3.94933249237885e-06, + "loss": 0.0103, + "step": 5000 + }, + { + "epoch": 1.5, + "eval_f1": 0.9920083137758474, + "eval_f1_0": 0.9974702907981972, + "eval_f1_1": 0.9983040421287434, + "eval_f1_2": 0.9732429083447697, + "eval_f1_3": 0.9990160138316794, + "eval_loss": 0.0146098081022501, + "eval_runtime": 153.1625, + "eval_samples_per_second": 77.486, + "eval_steps_per_second": 0.307, + "step": 5000 + }, + { + "epoch": 1.5, + "learning_rate": 3.946704509618417e-06, + "loss": 0.0048, + "step": 5010 + }, + { + "epoch": 1.5, + "learning_rate": 3.944076526857984e-06, + "loss": 0.0076, + "step": 5020 + }, + { + "epoch": 1.51, + "learning_rate": 3.941448544097551e-06, + "loss": 0.0082, + "step": 5030 + }, + { + "epoch": 1.51, + "learning_rate": 3.938820561337118e-06, + "loss": 0.0031, + "step": 5040 + }, + { + "epoch": 1.51, + "learning_rate": 3.936192578576685e-06, + "loss": 0.0135, + "step": 5050 + }, + { + "epoch": 1.52, + "learning_rate": 3.933564595816251e-06, + "loss": 0.0029, + "step": 5060 + }, + { + "epoch": 1.52, + "learning_rate": 3.930936613055818e-06, + "loss": 0.0039, + "step": 5070 + }, + { + "epoch": 1.52, + "learning_rate": 3.928308630295385e-06, + "loss": 0.0041, + "step": 5080 + }, + { + "epoch": 1.52, + "learning_rate": 3.925680647534952e-06, + "loss": 0.0029, + "step": 5090 + }, + { + "epoch": 1.53, + "learning_rate": 3.923052664774519e-06, + "loss": 0.0045, + "step": 5100 + }, + { + "epoch": 1.53, + "learning_rate": 3.920424682014086e-06, + "loss": 0.0042, + "step": 5110 + }, + { + "epoch": 1.53, + "learning_rate": 3.917796699253653e-06, + "loss": 0.0025, + "step": 5120 + }, + { + "epoch": 1.54, + "learning_rate": 3.91516871649322e-06, + "loss": 0.0058, + "step": 5130 + }, + { + "epoch": 1.54, + "learning_rate": 3.912540733732787e-06, + "loss": 0.0165, + "step": 5140 + }, + { + "epoch": 1.54, + "learning_rate": 3.9099127509723535e-06, + "loss": 0.0098, + "step": 5150 + }, + { + "epoch": 1.55, + "learning_rate": 3.9072847682119205e-06, + "loss": 0.0109, + "step": 5160 + }, + { + "epoch": 1.55, + "learning_rate": 3.9046567854514875e-06, + "loss": 0.013, + "step": 5170 + }, + { + "epoch": 1.55, + "learning_rate": 3.9020288026910545e-06, + "loss": 0.0276, + "step": 5180 + }, + { + "epoch": 1.55, + "learning_rate": 3.8994008199306215e-06, + "loss": 0.0038, + "step": 5190 + }, + { + "epoch": 1.56, + "learning_rate": 3.8967728371701886e-06, + "loss": 0.0218, + "step": 5200 + }, + { + "epoch": 1.56, + "eval_f1": 0.9919780112294924, + "eval_f1_0": 0.9974857592459158, + "eval_f1_1": 0.9971331899290379, + "eval_f1_2": 0.9742953640344122, + "eval_f1_3": 0.9989977317086037, + "eval_loss": 0.016803942620754242, + "eval_runtime": 153.3435, + "eval_samples_per_second": 77.395, + "eval_steps_per_second": 0.307, + "step": 5200 + }, + { + "epoch": 1.56, + "learning_rate": 3.8941448544097556e-06, + "loss": 0.0109, + "step": 5210 + }, + { + "epoch": 1.56, + "learning_rate": 3.8915168716493226e-06, + "loss": 0.0015, + "step": 5220 + }, + { + "epoch": 1.57, + "learning_rate": 3.88888888888889e-06, + "loss": 0.0072, + "step": 5230 + }, + { + "epoch": 1.57, + "learning_rate": 3.886260906128457e-06, + "loss": 0.0132, + "step": 5240 + }, + { + "epoch": 1.57, + "learning_rate": 3.883632923368023e-06, + "loss": 0.0261, + "step": 5250 + }, + { + "epoch": 1.58, + "learning_rate": 3.88100494060759e-06, + "loss": 0.0166, + "step": 5260 + }, + { + "epoch": 1.58, + "learning_rate": 3.878376957847157e-06, + "loss": 0.0018, + "step": 5270 + }, + { + "epoch": 1.58, + "learning_rate": 3.875748975086724e-06, + "loss": 0.0022, + "step": 5280 + }, + { + "epoch": 1.58, + "learning_rate": 3.873120992326291e-06, + "loss": 0.0232, + "step": 5290 + }, + { + "epoch": 1.59, + "learning_rate": 3.870493009565858e-06, + "loss": 0.01, + "step": 5300 + }, + { + "epoch": 1.59, + "learning_rate": 3.867865026805425e-06, + "loss": 0.0104, + "step": 5310 + }, + { + "epoch": 1.59, + "learning_rate": 3.865237044044992e-06, + "loss": 0.0419, + "step": 5320 + }, + { + "epoch": 1.6, + "learning_rate": 3.862609061284559e-06, + "loss": 0.0037, + "step": 5330 + }, + { + "epoch": 1.6, + "learning_rate": 3.859981078524125e-06, + "loss": 0.0011, + "step": 5340 + }, + { + "epoch": 1.6, + "learning_rate": 3.857353095763692e-06, + "loss": 0.0014, + "step": 5350 + }, + { + "epoch": 1.61, + "learning_rate": 3.854725113003259e-06, + "loss": 0.0666, + "step": 5360 + }, + { + "epoch": 1.61, + "learning_rate": 3.852097130242826e-06, + "loss": 0.0191, + "step": 5370 + }, + { + "epoch": 1.61, + "learning_rate": 3.849469147482393e-06, + "loss": 0.0102, + "step": 5380 + }, + { + "epoch": 1.61, + "learning_rate": 3.84684116472196e-06, + "loss": 0.0556, + "step": 5390 + }, + { + "epoch": 1.62, + "learning_rate": 3.844213181961526e-06, + "loss": 0.0048, + "step": 5400 + }, + { + "epoch": 1.62, + "eval_f1": 0.9823485200918017, + "eval_f1_0": 0.9941904301193779, + "eval_f1_1": 0.9982555420088488, + "eval_f1_2": 0.9379690705502648, + "eval_f1_3": 0.9989790376887151, + "eval_loss": 0.04433099552989006, + "eval_runtime": 153.078, + "eval_samples_per_second": 77.529, + "eval_steps_per_second": 0.307, + "step": 5400 + }, + { + "epoch": 1.62, + "learning_rate": 3.841585199201093e-06, + "loss": 0.0492, + "step": 5410 + }, + { + "epoch": 1.62, + "learning_rate": 3.83895721644066e-06, + "loss": 0.0611, + "step": 5420 + }, + { + "epoch": 1.63, + "learning_rate": 3.836329233680227e-06, + "loss": 0.0226, + "step": 5430 + }, + { + "epoch": 1.63, + "learning_rate": 3.833701250919794e-06, + "loss": 0.0089, + "step": 5440 + }, + { + "epoch": 1.63, + "learning_rate": 3.831073268159361e-06, + "loss": 0.0055, + "step": 5450 + }, + { + "epoch": 1.64, + "learning_rate": 3.828445285398928e-06, + "loss": 0.0252, + "step": 5460 + }, + { + "epoch": 1.64, + "learning_rate": 3.8258173026384944e-06, + "loss": 0.0084, + "step": 5470 + }, + { + "epoch": 1.64, + "learning_rate": 3.8231893198780614e-06, + "loss": 0.0117, + "step": 5480 + }, + { + "epoch": 1.64, + "learning_rate": 3.8205613371176285e-06, + "loss": 0.0025, + "step": 5490 + }, + { + "epoch": 1.65, + "learning_rate": 3.8179333543571955e-06, + "loss": 0.0157, + "step": 5500 + }, + { + "epoch": 1.65, + "learning_rate": 3.8153053715967625e-06, + "loss": 0.0169, + "step": 5510 + }, + { + "epoch": 1.65, + "learning_rate": 3.8126773888363295e-06, + "loss": 0.0169, + "step": 5520 + }, + { + "epoch": 1.66, + "learning_rate": 3.8100494060758965e-06, + "loss": 0.0458, + "step": 5530 + }, + { + "epoch": 1.66, + "learning_rate": 3.807421423315463e-06, + "loss": 0.0331, + "step": 5540 + }, + { + "epoch": 1.66, + "learning_rate": 3.80479344055503e-06, + "loss": 0.0131, + "step": 5550 + }, + { + "epoch": 1.67, + "learning_rate": 3.802165457794597e-06, + "loss": 0.0159, + "step": 5560 + }, + { + "epoch": 1.67, + "learning_rate": 3.799537475034164e-06, + "loss": 0.0162, + "step": 5570 + }, + { + "epoch": 1.67, + "learning_rate": 3.796909492273731e-06, + "loss": 0.0138, + "step": 5580 + }, + { + "epoch": 1.67, + "learning_rate": 3.7942815095132977e-06, + "loss": 0.0168, + "step": 5590 + }, + { + "epoch": 1.68, + "learning_rate": 3.7916535267528647e-06, + "loss": 0.0066, + "step": 5600 + }, + { + "epoch": 1.68, + "eval_f1": 0.990187773769687, + "eval_f1_0": 0.9968801414078322, + "eval_f1_1": 0.997884261630327, + "eval_f1_2": 0.9670851157715061, + "eval_f1_3": 0.9989015762690828, + "eval_loss": 0.02275552973151207, + "eval_runtime": 153.2058, + "eval_samples_per_second": 77.464, + "eval_steps_per_second": 0.307, + "step": 5600 + }, + { + "epoch": 1.68, + "learning_rate": 3.7890255439924317e-06, + "loss": 0.0498, + "step": 5610 + }, + { + "epoch": 1.68, + "learning_rate": 3.7863975612319987e-06, + "loss": 0.0227, + "step": 5620 + }, + { + "epoch": 1.69, + "learning_rate": 3.7837695784715658e-06, + "loss": 0.0153, + "step": 5630 + }, + { + "epoch": 1.69, + "learning_rate": 3.7811415957111323e-06, + "loss": 0.0093, + "step": 5640 + }, + { + "epoch": 1.69, + "learning_rate": 3.7785136129506993e-06, + "loss": 0.0075, + "step": 5650 + }, + { + "epoch": 1.7, + "learning_rate": 3.7758856301902664e-06, + "loss": 0.0146, + "step": 5660 + }, + { + "epoch": 1.7, + "learning_rate": 3.7732576474298334e-06, + "loss": 0.0236, + "step": 5670 + }, + { + "epoch": 1.7, + "learning_rate": 3.7706296646694004e-06, + "loss": 0.003, + "step": 5680 + }, + { + "epoch": 1.7, + "learning_rate": 3.768001681908967e-06, + "loss": 0.0385, + "step": 5690 + }, + { + "epoch": 1.71, + "learning_rate": 3.765373699148534e-06, + "loss": 0.0245, + "step": 5700 + }, + { + "epoch": 1.71, + "learning_rate": 3.762745716388101e-06, + "loss": 0.0259, + "step": 5710 + }, + { + "epoch": 1.71, + "learning_rate": 3.760117733627668e-06, + "loss": 0.0065, + "step": 5720 + }, + { + "epoch": 1.72, + "learning_rate": 3.7574897508672346e-06, + "loss": 0.0078, + "step": 5730 + }, + { + "epoch": 1.72, + "learning_rate": 3.7548617681068016e-06, + "loss": 0.0044, + "step": 5740 + }, + { + "epoch": 1.72, + "learning_rate": 3.7522337853463686e-06, + "loss": 0.0077, + "step": 5750 + }, + { + "epoch": 1.73, + "learning_rate": 3.7496058025859356e-06, + "loss": 0.0041, + "step": 5760 + }, + { + "epoch": 1.73, + "learning_rate": 3.7469778198255026e-06, + "loss": 0.0378, + "step": 5770 + }, + { + "epoch": 1.73, + "learning_rate": 3.744349837065069e-06, + "loss": 0.0052, + "step": 5780 + }, + { + "epoch": 1.73, + "learning_rate": 3.7417218543046362e-06, + "loss": 0.0069, + "step": 5790 + }, + { + "epoch": 1.74, + "learning_rate": 3.7390938715442032e-06, + "loss": 0.0364, + "step": 5800 + }, + { + "epoch": 1.74, + "eval_f1": 0.9930586158488619, + "eval_f1_0": 0.997830070247112, + "eval_f1_1": 0.9983115974664379, + "eval_f1_2": 0.9771260618891283, + "eval_f1_3": 0.9989667337927696, + "eval_loss": 0.012566142715513706, + "eval_runtime": 153.2869, + "eval_samples_per_second": 77.423, + "eval_steps_per_second": 0.307, + "step": 5800 + }, + { + "epoch": 1.74, + "learning_rate": 3.7364658887837702e-06, + "loss": 0.0053, + "step": 5810 + }, + { + "epoch": 1.74, + "learning_rate": 3.7338379060233372e-06, + "loss": 0.0022, + "step": 5820 + }, + { + "epoch": 1.75, + "learning_rate": 3.7312099232629034e-06, + "loss": 0.0055, + "step": 5830 + }, + { + "epoch": 1.75, + "learning_rate": 3.7285819405024704e-06, + "loss": 0.0027, + "step": 5840 + }, + { + "epoch": 1.75, + "learning_rate": 3.7259539577420374e-06, + "loss": 0.0203, + "step": 5850 + }, + { + "epoch": 1.76, + "learning_rate": 3.723325974981604e-06, + "loss": 0.0306, + "step": 5860 + }, + { + "epoch": 1.76, + "learning_rate": 3.720697992221171e-06, + "loss": 0.0047, + "step": 5870 + }, + { + "epoch": 1.76, + "learning_rate": 3.718070009460738e-06, + "loss": 0.0309, + "step": 5880 + }, + { + "epoch": 1.76, + "learning_rate": 3.715442026700305e-06, + "loss": 0.0079, + "step": 5890 + }, + { + "epoch": 1.77, + "learning_rate": 3.712814043939872e-06, + "loss": 0.0045, + "step": 5900 + }, + { + "epoch": 1.77, + "learning_rate": 3.7101860611794386e-06, + "loss": 0.008, + "step": 5910 + }, + { + "epoch": 1.77, + "learning_rate": 3.7075580784190057e-06, + "loss": 0.0049, + "step": 5920 + }, + { + "epoch": 1.78, + "learning_rate": 3.7049300956585727e-06, + "loss": 0.0185, + "step": 5930 + }, + { + "epoch": 1.78, + "learning_rate": 3.7023021128981397e-06, + "loss": 0.0467, + "step": 5940 + }, + { + "epoch": 1.78, + "learning_rate": 3.6996741301377067e-06, + "loss": 0.0153, + "step": 5950 + }, + { + "epoch": 1.79, + "learning_rate": 3.6970461473772733e-06, + "loss": 0.0241, + "step": 5960 + }, + { + "epoch": 1.79, + "learning_rate": 3.6944181646168403e-06, + "loss": 0.0029, + "step": 5970 + }, + { + "epoch": 1.79, + "learning_rate": 3.6917901818564073e-06, + "loss": 0.0029, + "step": 5980 + }, + { + "epoch": 1.79, + "learning_rate": 3.6891621990959743e-06, + "loss": 0.0086, + "step": 5990 + }, + { + "epoch": 1.8, + "learning_rate": 3.686534216335541e-06, + "loss": 0.015, + "step": 6000 + }, + { + "epoch": 1.8, + "eval_f1": 0.9935938444067491, + "eval_f1_0": 0.9979641633032684, + "eval_f1_1": 0.9971702174474558, + "eval_f1_2": 0.9803455304280306, + "eval_f1_3": 0.998895466448242, + "eval_loss": 0.011877385899424553, + "eval_runtime": 153.1522, + "eval_samples_per_second": 77.492, + "eval_steps_per_second": 0.307, + "step": 6000 + }, + { + "epoch": 1.8, + "learning_rate": 3.683906233575108e-06, + "loss": 0.0137, + "step": 6010 + }, + { + "epoch": 1.8, + "learning_rate": 3.681278250814675e-06, + "loss": 0.0158, + "step": 6020 + }, + { + "epoch": 1.81, + "learning_rate": 3.678650268054242e-06, + "loss": 0.0266, + "step": 6030 + }, + { + "epoch": 1.81, + "learning_rate": 3.676022285293809e-06, + "loss": 0.003, + "step": 6040 + }, + { + "epoch": 1.81, + "learning_rate": 3.6733943025333755e-06, + "loss": 0.0027, + "step": 6050 + }, + { + "epoch": 1.82, + "learning_rate": 3.6707663197729425e-06, + "loss": 0.0416, + "step": 6060 + }, + { + "epoch": 1.82, + "learning_rate": 3.6681383370125095e-06, + "loss": 0.0018, + "step": 6070 + }, + { + "epoch": 1.82, + "learning_rate": 3.6655103542520765e-06, + "loss": 0.0065, + "step": 6080 + }, + { + "epoch": 1.82, + "learning_rate": 3.6628823714916436e-06, + "loss": 0.0019, + "step": 6090 + }, + { + "epoch": 1.83, + "learning_rate": 3.66025438873121e-06, + "loss": 0.0046, + "step": 6100 + }, + { + "epoch": 1.83, + "learning_rate": 3.657626405970777e-06, + "loss": 0.0017, + "step": 6110 + }, + { + "epoch": 1.83, + "learning_rate": 3.654998423210344e-06, + "loss": 0.0039, + "step": 6120 + }, + { + "epoch": 1.84, + "learning_rate": 3.652370440449911e-06, + "loss": 0.0018, + "step": 6130 + }, + { + "epoch": 1.84, + "learning_rate": 3.649742457689478e-06, + "loss": 0.0054, + "step": 6140 + }, + { + "epoch": 1.84, + "learning_rate": 3.6471144749290448e-06, + "loss": 0.0075, + "step": 6150 + }, + { + "epoch": 1.85, + "learning_rate": 3.6444864921686118e-06, + "loss": 0.0112, + "step": 6160 + }, + { + "epoch": 1.85, + "learning_rate": 3.641858509408179e-06, + "loss": 0.069, + "step": 6170 + }, + { + "epoch": 1.85, + "learning_rate": 3.639230526647746e-06, + "loss": 0.002, + "step": 6180 + }, + { + "epoch": 1.85, + "learning_rate": 3.6366025438873124e-06, + "loss": 0.008, + "step": 6190 + }, + { + "epoch": 1.86, + "learning_rate": 3.6339745611268794e-06, + "loss": 0.0061, + "step": 6200 + }, + { + "epoch": 1.86, + "eval_f1": 0.9917367001053987, + "eval_f1_0": 0.9973992454549965, + "eval_f1_1": 0.9983099239274149, + "eval_f1_2": 0.9722678167208436, + "eval_f1_3": 0.9989698143183398, + "eval_loss": 0.014565565623342991, + "eval_runtime": 153.2683, + "eval_samples_per_second": 77.433, + "eval_steps_per_second": 0.307, + "step": 6200 + }, + { + "epoch": 1.86, + "learning_rate": 3.6313465783664464e-06, + "loss": 0.0037, + "step": 6210 + }, + { + "epoch": 1.86, + "learning_rate": 3.6287185956060134e-06, + "loss": 0.0423, + "step": 6220 + }, + { + "epoch": 1.87, + "learning_rate": 3.6260906128455804e-06, + "loss": 0.0174, + "step": 6230 + }, + { + "epoch": 1.87, + "learning_rate": 3.623462630085147e-06, + "loss": 0.0018, + "step": 6240 + }, + { + "epoch": 1.87, + "learning_rate": 3.6208346473247136e-06, + "loss": 0.0048, + "step": 6250 + }, + { + "epoch": 1.88, + "learning_rate": 3.6182066645642806e-06, + "loss": 0.0018, + "step": 6260 + }, + { + "epoch": 1.88, + "learning_rate": 3.6155786818038476e-06, + "loss": 0.0023, + "step": 6270 + }, + { + "epoch": 1.88, + "learning_rate": 3.612950699043414e-06, + "loss": 0.0035, + "step": 6280 + }, + { + "epoch": 1.88, + "learning_rate": 3.6103227162829812e-06, + "loss": 0.0043, + "step": 6290 + }, + { + "epoch": 1.89, + "learning_rate": 3.6076947335225482e-06, + "loss": 0.0682, + "step": 6300 + }, + { + "epoch": 1.89, + "learning_rate": 3.6050667507621152e-06, + "loss": 0.0063, + "step": 6310 + }, + { + "epoch": 1.89, + "learning_rate": 3.602438768001682e-06, + "loss": 0.0073, + "step": 6320 + }, + { + "epoch": 1.9, + "learning_rate": 3.599810785241249e-06, + "loss": 0.0092, + "step": 6330 + }, + { + "epoch": 1.9, + "learning_rate": 3.597182802480816e-06, + "loss": 0.0018, + "step": 6340 + }, + { + "epoch": 1.9, + "learning_rate": 3.594554819720383e-06, + "loss": 0.012, + "step": 6350 + }, + { + "epoch": 1.91, + "learning_rate": 3.59192683695995e-06, + "loss": 0.0233, + "step": 6360 + }, + { + "epoch": 1.91, + "learning_rate": 3.5892988541995165e-06, + "loss": 0.0136, + "step": 6370 + }, + { + "epoch": 1.91, + "learning_rate": 3.5866708714390835e-06, + "loss": 0.0258, + "step": 6380 + }, + { + "epoch": 1.91, + "learning_rate": 3.5840428886786505e-06, + "loss": 0.0025, + "step": 6390 + }, + { + "epoch": 1.92, + "learning_rate": 3.5814149059182175e-06, + "loss": 0.0128, + "step": 6400 + }, + { + "epoch": 1.92, + "eval_f1": 0.9914634499892494, + "eval_f1_0": 0.9973011310432432, + "eval_f1_1": 0.9983957680476094, + "eval_f1_2": 0.9711716378258444, + "eval_f1_3": 0.9989852630403009, + "eval_loss": 0.017467861995100975, + "eval_runtime": 153.0977, + "eval_samples_per_second": 77.519, + "eval_steps_per_second": 0.307, + "step": 6400 + }, + { + "epoch": 1.92, + "learning_rate": 3.5787869231577845e-06, + "loss": 0.0242, + "step": 6410 + }, + { + "epoch": 1.92, + "learning_rate": 3.576158940397351e-06, + "loss": 0.0466, + "step": 6420 + }, + { + "epoch": 1.93, + "learning_rate": 3.573530957636918e-06, + "loss": 0.0287, + "step": 6430 + }, + { + "epoch": 1.93, + "learning_rate": 3.570902974876485e-06, + "loss": 0.0118, + "step": 6440 + }, + { + "epoch": 1.93, + "learning_rate": 3.568274992116052e-06, + "loss": 0.0155, + "step": 6450 + }, + { + "epoch": 1.94, + "learning_rate": 3.565647009355619e-06, + "loss": 0.0184, + "step": 6460 + }, + { + "epoch": 1.94, + "learning_rate": 3.5630190265951857e-06, + "loss": 0.0179, + "step": 6470 + }, + { + "epoch": 1.94, + "learning_rate": 3.5603910438347527e-06, + "loss": 0.0101, + "step": 6480 + }, + { + "epoch": 1.94, + "learning_rate": 3.5577630610743197e-06, + "loss": 0.0036, + "step": 6490 + }, + { + "epoch": 1.95, + "learning_rate": 3.5551350783138867e-06, + "loss": 0.0134, + "step": 6500 + }, + { + "epoch": 1.95, + "learning_rate": 3.5525070955534533e-06, + "loss": 0.0485, + "step": 6510 + }, + { + "epoch": 1.95, + "learning_rate": 3.5498791127930203e-06, + "loss": 0.0204, + "step": 6520 + }, + { + "epoch": 1.96, + "learning_rate": 3.5472511300325873e-06, + "loss": 0.019, + "step": 6530 + }, + { + "epoch": 1.96, + "learning_rate": 3.5446231472721544e-06, + "loss": 0.0046, + "step": 6540 + }, + { + "epoch": 1.96, + "learning_rate": 3.5419951645117214e-06, + "loss": 0.0195, + "step": 6550 + }, + { + "epoch": 1.97, + "learning_rate": 3.539367181751288e-06, + "loss": 0.0052, + "step": 6560 + }, + { + "epoch": 1.97, + "learning_rate": 3.536739198990855e-06, + "loss": 0.0051, + "step": 6570 + }, + { + "epoch": 1.97, + "learning_rate": 3.534111216230422e-06, + "loss": 0.0046, + "step": 6580 + }, + { + "epoch": 1.97, + "learning_rate": 3.531483233469989e-06, + "loss": 0.0077, + "step": 6590 + }, + { + "epoch": 1.98, + "learning_rate": 3.528855250709556e-06, + "loss": 0.0131, + "step": 6600 + }, + { + "epoch": 1.98, + "eval_f1": 0.9935970118232959, + "eval_f1_0": 0.9979761564787326, + "eval_f1_1": 0.9972816736632029, + "eval_f1_2": 0.9800643470294387, + "eval_f1_3": 0.9990658701218093, + "eval_loss": 0.013104687444865704, + "eval_runtime": 153.0821, + "eval_samples_per_second": 77.527, + "eval_steps_per_second": 0.307, + "step": 6600 + }, + { + "epoch": 1.98, + "learning_rate": 3.5262272679491226e-06, + "loss": 0.011, + "step": 6610 + }, + { + "epoch": 1.98, + "learning_rate": 3.5235992851886896e-06, + "loss": 0.0285, + "step": 6620 + }, + { + "epoch": 1.99, + "learning_rate": 3.5209713024282566e-06, + "loss": 0.0131, + "step": 6630 + }, + { + "epoch": 1.99, + "learning_rate": 3.5183433196678236e-06, + "loss": 0.0101, + "step": 6640 + }, + { + "epoch": 1.99, + "learning_rate": 3.5157153369073906e-06, + "loss": 0.0116, + "step": 6650 + }, + { + "epoch": 2.0, + "learning_rate": 3.513087354146957e-06, + "loss": 0.0179, + "step": 6660 + }, + { + "epoch": 2.0, + "learning_rate": 3.5104593713865242e-06, + "loss": 0.0096, + "step": 6670 + }, + { + "epoch": 2.0, + "learning_rate": 3.507831388626091e-06, + "loss": 0.0027, + "step": 6680 + }, + { + "epoch": 2.0, + "learning_rate": 3.5052034058656574e-06, + "loss": 0.003, + "step": 6690 + }, + { + "epoch": 2.01, + "learning_rate": 3.5025754231052244e-06, + "loss": 0.0027, + "step": 6700 + }, + { + "epoch": 2.01, + "learning_rate": 3.4999474403447914e-06, + "loss": 0.0021, + "step": 6710 + }, + { + "epoch": 2.01, + "learning_rate": 3.4973194575843584e-06, + "loss": 0.0018, + "step": 6720 + }, + { + "epoch": 2.02, + "learning_rate": 3.4946914748239254e-06, + "loss": 0.0017, + "step": 6730 + }, + { + "epoch": 2.02, + "learning_rate": 3.492063492063492e-06, + "loss": 0.0025, + "step": 6740 + }, + { + "epoch": 2.02, + "learning_rate": 3.489435509303059e-06, + "loss": 0.0028, + "step": 6750 + }, + { + "epoch": 2.03, + "learning_rate": 3.486807526542626e-06, + "loss": 0.0222, + "step": 6760 + }, + { + "epoch": 2.03, + "learning_rate": 3.484179543782193e-06, + "loss": 0.0037, + "step": 6770 + }, + { + "epoch": 2.03, + "learning_rate": 3.4815515610217596e-06, + "loss": 0.0015, + "step": 6780 + }, + { + "epoch": 2.03, + "learning_rate": 3.4789235782613266e-06, + "loss": 0.0218, + "step": 6790 + }, + { + "epoch": 2.04, + "learning_rate": 3.4762955955008937e-06, + "loss": 0.0037, + "step": 6800 + }, + { + "epoch": 2.04, + "eval_f1": 0.9902274199895836, + "eval_f1_0": 0.9969038168318187, + "eval_f1_1": 0.9985555223491728, + "eval_f1_2": 0.9666043873859351, + "eval_f1_3": 0.9988459533914079, + "eval_loss": 0.023619718849658966, + "eval_runtime": 153.1993, + "eval_samples_per_second": 77.468, + "eval_steps_per_second": 0.307, + "step": 6800 + }, + { + "epoch": 2.04, + "learning_rate": 3.4736676127404607e-06, + "loss": 0.0031, + "step": 6810 + }, + { + "epoch": 2.04, + "learning_rate": 3.4710396299800277e-06, + "loss": 0.0013, + "step": 6820 + }, + { + "epoch": 2.05, + "learning_rate": 3.4684116472195943e-06, + "loss": 0.0019, + "step": 6830 + }, + { + "epoch": 2.05, + "learning_rate": 3.4657836644591613e-06, + "loss": 0.0015, + "step": 6840 + }, + { + "epoch": 2.05, + "learning_rate": 3.4631556816987283e-06, + "loss": 0.0188, + "step": 6850 + }, + { + "epoch": 2.06, + "learning_rate": 3.4605276989382953e-06, + "loss": 0.0024, + "step": 6860 + }, + { + "epoch": 2.06, + "learning_rate": 3.4578997161778623e-06, + "loss": 0.002, + "step": 6870 + }, + { + "epoch": 2.06, + "learning_rate": 3.455271733417429e-06, + "loss": 0.0187, + "step": 6880 + }, + { + "epoch": 2.06, + "learning_rate": 3.452643750656996e-06, + "loss": 0.0116, + "step": 6890 + }, + { + "epoch": 2.07, + "learning_rate": 3.450015767896563e-06, + "loss": 0.0061, + "step": 6900 + }, + { + "epoch": 2.07, + "learning_rate": 3.44738778513613e-06, + "loss": 0.0291, + "step": 6910 + }, + { + "epoch": 2.07, + "learning_rate": 3.444759802375697e-06, + "loss": 0.0113, + "step": 6920 + }, + { + "epoch": 2.08, + "learning_rate": 3.4421318196152635e-06, + "loss": 0.0028, + "step": 6930 + }, + { + "epoch": 2.08, + "learning_rate": 3.4395038368548305e-06, + "loss": 0.0015, + "step": 6940 + }, + { + "epoch": 2.08, + "learning_rate": 3.4368758540943975e-06, + "loss": 0.0026, + "step": 6950 + }, + { + "epoch": 2.09, + "learning_rate": 3.4342478713339645e-06, + "loss": 0.0011, + "step": 6960 + }, + { + "epoch": 2.09, + "learning_rate": 3.4316198885735316e-06, + "loss": 0.0015, + "step": 6970 + }, + { + "epoch": 2.09, + "learning_rate": 3.428991905813098e-06, + "loss": 0.0052, + "step": 6980 + }, + { + "epoch": 2.09, + "learning_rate": 3.426363923052665e-06, + "loss": 0.0015, + "step": 6990 + }, + { + "epoch": 2.1, + "learning_rate": 3.423735940292232e-06, + "loss": 0.0226, + "step": 7000 + }, + { + "epoch": 2.1, + "eval_f1": 0.9915189698715074, + "eval_f1_0": 0.9973328123383862, + "eval_f1_1": 0.9985646253940573, + "eval_f1_2": 0.9713880891568768, + "eval_f1_3": 0.9987903525967098, + "eval_loss": 0.018696604296565056, + "eval_runtime": 153.1722, + "eval_samples_per_second": 77.481, + "eval_steps_per_second": 0.307, + "step": 7000 + }, + { + "epoch": 2.1, + "learning_rate": 3.421107957531799e-06, + "loss": 0.0019, + "step": 7010 + }, + { + "epoch": 2.1, + "learning_rate": 3.4184799747713658e-06, + "loss": 0.0185, + "step": 7020 + }, + { + "epoch": 2.11, + "learning_rate": 3.4158519920109328e-06, + "loss": 0.0036, + "step": 7030 + }, + { + "epoch": 2.11, + "learning_rate": 3.4132240092504998e-06, + "loss": 0.0007, + "step": 7040 + }, + { + "epoch": 2.11, + "learning_rate": 3.4105960264900668e-06, + "loss": 0.0015, + "step": 7050 + }, + { + "epoch": 2.12, + "learning_rate": 3.407968043729634e-06, + "loss": 0.002, + "step": 7060 + }, + { + "epoch": 2.12, + "learning_rate": 3.4053400609692004e-06, + "loss": 0.0016, + "step": 7070 + }, + { + "epoch": 2.12, + "learning_rate": 3.4027120782087674e-06, + "loss": 0.003, + "step": 7080 + }, + { + "epoch": 2.12, + "learning_rate": 3.4000840954483344e-06, + "loss": 0.0011, + "step": 7090 + }, + { + "epoch": 2.13, + "learning_rate": 3.3974561126879014e-06, + "loss": 0.0023, + "step": 7100 + }, + { + "epoch": 2.13, + "learning_rate": 3.3948281299274676e-06, + "loss": 0.003, + "step": 7110 + }, + { + "epoch": 2.13, + "learning_rate": 3.3922001471670346e-06, + "loss": 0.0015, + "step": 7120 + }, + { + "epoch": 2.14, + "learning_rate": 3.3895721644066016e-06, + "loss": 0.0018, + "step": 7130 + }, + { + "epoch": 2.14, + "learning_rate": 3.3869441816461686e-06, + "loss": 0.0016, + "step": 7140 + }, + { + "epoch": 2.14, + "learning_rate": 3.384316198885735e-06, + "loss": 0.0008, + "step": 7150 + }, + { + "epoch": 2.14, + "learning_rate": 3.381688216125302e-06, + "loss": 0.0018, + "step": 7160 + }, + { + "epoch": 2.15, + "learning_rate": 3.379060233364869e-06, + "loss": 0.0104, + "step": 7170 + }, + { + "epoch": 2.15, + "learning_rate": 3.3764322506044362e-06, + "loss": 0.0021, + "step": 7180 + }, + { + "epoch": 2.15, + "learning_rate": 3.3738042678440032e-06, + "loss": 0.0172, + "step": 7190 + }, + { + "epoch": 2.16, + "learning_rate": 3.37117628508357e-06, + "loss": 0.0091, + "step": 7200 + }, + { + "epoch": 2.16, + "eval_f1": 0.9859817799915488, + "eval_f1_0": 0.9954234300747538, + "eval_f1_1": 0.998167859346792, + "eval_f1_2": 0.9512544005545798, + "eval_f1_3": 0.9990814299900695, + "eval_loss": 0.03158416226506233, + "eval_runtime": 153.0778, + "eval_samples_per_second": 77.529, + "eval_steps_per_second": 0.307, + "step": 7200 + }, + { + "epoch": 2.16, + "learning_rate": 3.368548302323137e-06, + "loss": 0.0019, + "step": 7210 + }, + { + "epoch": 2.16, + "learning_rate": 3.365920319562704e-06, + "loss": 0.0065, + "step": 7220 + }, + { + "epoch": 2.17, + "learning_rate": 3.363292336802271e-06, + "loss": 0.0171, + "step": 7230 + }, + { + "epoch": 2.17, + "learning_rate": 3.360664354041838e-06, + "loss": 0.0147, + "step": 7240 + }, + { + "epoch": 2.17, + "learning_rate": 3.3580363712814044e-06, + "loss": 0.009, + "step": 7250 + }, + { + "epoch": 2.17, + "learning_rate": 3.3554083885209715e-06, + "loss": 0.0302, + "step": 7260 + }, + { + "epoch": 2.18, + "learning_rate": 3.3527804057605385e-06, + "loss": 0.0097, + "step": 7270 + }, + { + "epoch": 2.18, + "learning_rate": 3.3501524230001055e-06, + "loss": 0.0026, + "step": 7280 + }, + { + "epoch": 2.18, + "learning_rate": 3.347524440239672e-06, + "loss": 0.007, + "step": 7290 + }, + { + "epoch": 2.19, + "learning_rate": 3.344896457479239e-06, + "loss": 0.0022, + "step": 7300 + }, + { + "epoch": 2.19, + "learning_rate": 3.342268474718806e-06, + "loss": 0.0019, + "step": 7310 + }, + { + "epoch": 2.19, + "learning_rate": 3.339640491958373e-06, + "loss": 0.011, + "step": 7320 + }, + { + "epoch": 2.2, + "learning_rate": 3.33701250919794e-06, + "loss": 0.0097, + "step": 7330 + }, + { + "epoch": 2.2, + "learning_rate": 3.3343845264375067e-06, + "loss": 0.0075, + "step": 7340 + }, + { + "epoch": 2.2, + "learning_rate": 3.3317565436770737e-06, + "loss": 0.0051, + "step": 7350 + }, + { + "epoch": 2.2, + "learning_rate": 3.3291285609166407e-06, + "loss": 0.0039, + "step": 7360 + }, + { + "epoch": 2.21, + "learning_rate": 3.3265005781562077e-06, + "loss": 0.0393, + "step": 7370 + }, + { + "epoch": 2.21, + "learning_rate": 3.3238725953957747e-06, + "loss": 0.0046, + "step": 7380 + }, + { + "epoch": 2.21, + "learning_rate": 3.3212446126353413e-06, + "loss": 0.015, + "step": 7390 + }, + { + "epoch": 2.22, + "learning_rate": 3.3186166298749083e-06, + "loss": 0.0102, + "step": 7400 + }, + { + "epoch": 2.22, + "eval_f1": 0.9922962109513367, + "eval_f1_0": 0.9975603361253395, + "eval_f1_1": 0.998105116186297, + "eval_f1_2": 0.9745372284660181, + "eval_f1_3": 0.9989821630276926, + "eval_loss": 0.017126647755503654, + "eval_runtime": 153.2307, + "eval_samples_per_second": 77.452, + "eval_steps_per_second": 0.307, + "step": 7400 + }, + { + "epoch": 2.22, + "learning_rate": 3.3159886471144753e-06, + "loss": 0.0092, + "step": 7410 + }, + { + "epoch": 2.22, + "learning_rate": 3.3133606643540423e-06, + "loss": 0.007, + "step": 7420 + }, + { + "epoch": 2.23, + "learning_rate": 3.3107326815936094e-06, + "loss": 0.0012, + "step": 7430 + }, + { + "epoch": 2.23, + "learning_rate": 3.308104698833176e-06, + "loss": 0.0028, + "step": 7440 + }, + { + "epoch": 2.23, + "learning_rate": 3.305476716072743e-06, + "loss": 0.0121, + "step": 7450 + }, + { + "epoch": 2.23, + "learning_rate": 3.30284873331231e-06, + "loss": 0.0088, + "step": 7460 + }, + { + "epoch": 2.24, + "learning_rate": 3.300220750551877e-06, + "loss": 0.008, + "step": 7470 + }, + { + "epoch": 2.24, + "learning_rate": 3.297592767791444e-06, + "loss": 0.0092, + "step": 7480 + }, + { + "epoch": 2.24, + "learning_rate": 3.2949647850310106e-06, + "loss": 0.002, + "step": 7490 + }, + { + "epoch": 2.25, + "learning_rate": 3.2923368022705776e-06, + "loss": 0.0126, + "step": 7500 + }, + { + "epoch": 2.25, + "learning_rate": 3.2897088195101446e-06, + "loss": 0.001, + "step": 7510 + }, + { + "epoch": 2.25, + "learning_rate": 3.2870808367497116e-06, + "loss": 0.0058, + "step": 7520 + }, + { + "epoch": 2.26, + "learning_rate": 3.2844528539892778e-06, + "loss": 0.0253, + "step": 7530 + }, + { + "epoch": 2.26, + "learning_rate": 3.2818248712288448e-06, + "loss": 0.018, + "step": 7540 + }, + { + "epoch": 2.26, + "learning_rate": 3.2791968884684118e-06, + "loss": 0.0041, + "step": 7550 + }, + { + "epoch": 2.26, + "learning_rate": 3.276568905707979e-06, + "loss": 0.0022, + "step": 7560 + }, + { + "epoch": 2.27, + "learning_rate": 3.2739409229475454e-06, + "loss": 0.0013, + "step": 7570 + }, + { + "epoch": 2.27, + "learning_rate": 3.2713129401871124e-06, + "loss": 0.0343, + "step": 7580 + }, + { + "epoch": 2.27, + "learning_rate": 3.2686849574266794e-06, + "loss": 0.0043, + "step": 7590 + }, + { + "epoch": 2.28, + "learning_rate": 3.2660569746662464e-06, + "loss": 0.006, + "step": 7600 + }, + { + "epoch": 2.28, + "eval_f1": 0.9939749535471507, + "eval_f1_0": 0.9981190431394863, + "eval_f1_1": 0.998165285358517, + "eval_f1_2": 0.980670645733551, + "eval_f1_3": 0.9989448399570487, + "eval_loss": 0.013017122633755207, + "eval_runtime": 153.0757, + "eval_samples_per_second": 77.53, + "eval_steps_per_second": 0.307, + "step": 7600 + }, + { + "epoch": 2.28, + "learning_rate": 3.263428991905813e-06, + "loss": 0.0015, + "step": 7610 + }, + { + "epoch": 2.28, + "learning_rate": 3.26080100914538e-06, + "loss": 0.0162, + "step": 7620 + }, + { + "epoch": 2.29, + "learning_rate": 3.258173026384947e-06, + "loss": 0.0023, + "step": 7630 + }, + { + "epoch": 2.29, + "learning_rate": 3.255545043624514e-06, + "loss": 0.0076, + "step": 7640 + }, + { + "epoch": 2.29, + "learning_rate": 3.252917060864081e-06, + "loss": 0.0348, + "step": 7650 + }, + { + "epoch": 2.29, + "learning_rate": 3.2502890781036476e-06, + "loss": 0.0012, + "step": 7660 + }, + { + "epoch": 2.3, + "learning_rate": 3.2476610953432146e-06, + "loss": 0.0051, + "step": 7670 + }, + { + "epoch": 2.3, + "learning_rate": 3.2450331125827816e-06, + "loss": 0.0044, + "step": 7680 + }, + { + "epoch": 2.3, + "learning_rate": 3.2424051298223487e-06, + "loss": 0.0166, + "step": 7690 + }, + { + "epoch": 2.31, + "learning_rate": 3.2397771470619157e-06, + "loss": 0.0024, + "step": 7700 + }, + { + "epoch": 2.31, + "learning_rate": 3.2371491643014823e-06, + "loss": 0.0626, + "step": 7710 + }, + { + "epoch": 2.31, + "learning_rate": 3.2345211815410493e-06, + "loss": 0.0298, + "step": 7720 + }, + { + "epoch": 2.32, + "learning_rate": 3.2318931987806163e-06, + "loss": 0.0191, + "step": 7730 + }, + { + "epoch": 2.32, + "learning_rate": 3.2292652160201833e-06, + "loss": 0.0024, + "step": 7740 + }, + { + "epoch": 2.32, + "learning_rate": 3.2266372332597503e-06, + "loss": 0.0124, + "step": 7750 + }, + { + "epoch": 2.32, + "learning_rate": 3.224009250499317e-06, + "loss": 0.0155, + "step": 7760 + }, + { + "epoch": 2.33, + "learning_rate": 3.221381267738884e-06, + "loss": 0.0033, + "step": 7770 + }, + { + "epoch": 2.33, + "learning_rate": 3.218753284978451e-06, + "loss": 0.0032, + "step": 7780 + }, + { + "epoch": 2.33, + "learning_rate": 3.216125302218018e-06, + "loss": 0.0024, + "step": 7790 + }, + { + "epoch": 2.34, + "learning_rate": 3.2134973194575845e-06, + "loss": 0.0014, + "step": 7800 + }, + { + "epoch": 2.34, + "eval_f1": 0.9927733375726729, + "eval_f1_0": 0.9977209199162577, + "eval_f1_1": 0.9981627518065634, + "eval_f1_2": 0.9762276545337066, + "eval_f1_3": 0.9989820240341643, + "eval_loss": 0.01505122147500515, + "eval_runtime": 153.1341, + "eval_samples_per_second": 77.501, + "eval_steps_per_second": 0.307, + "step": 7800 + }, + { + "epoch": 2.34, + "learning_rate": 3.2108693366971515e-06, + "loss": 0.001, + "step": 7810 + }, + { + "epoch": 2.34, + "learning_rate": 3.2082413539367185e-06, + "loss": 0.0034, + "step": 7820 + }, + { + "epoch": 2.35, + "learning_rate": 3.2056133711762855e-06, + "loss": 0.0014, + "step": 7830 + }, + { + "epoch": 2.35, + "learning_rate": 3.2029853884158525e-06, + "loss": 0.0038, + "step": 7840 + }, + { + "epoch": 2.35, + "learning_rate": 3.200357405655419e-06, + "loss": 0.0174, + "step": 7850 + }, + { + "epoch": 2.35, + "learning_rate": 3.197729422894986e-06, + "loss": 0.0022, + "step": 7860 + }, + { + "epoch": 2.36, + "learning_rate": 3.195101440134553e-06, + "loss": 0.0082, + "step": 7870 + }, + { + "epoch": 2.36, + "learning_rate": 3.19247345737412e-06, + "loss": 0.0021, + "step": 7880 + }, + { + "epoch": 2.36, + "learning_rate": 3.189845474613687e-06, + "loss": 0.0056, + "step": 7890 + }, + { + "epoch": 2.37, + "learning_rate": 3.1872174918532537e-06, + "loss": 0.0097, + "step": 7900 + }, + { + "epoch": 2.37, + "learning_rate": 3.1845895090928208e-06, + "loss": 0.0273, + "step": 7910 + }, + { + "epoch": 2.37, + "learning_rate": 3.1819615263323878e-06, + "loss": 0.0028, + "step": 7920 + }, + { + "epoch": 2.38, + "learning_rate": 3.1793335435719548e-06, + "loss": 0.0085, + "step": 7930 + }, + { + "epoch": 2.38, + "learning_rate": 3.176705560811522e-06, + "loss": 0.0046, + "step": 7940 + }, + { + "epoch": 2.38, + "learning_rate": 3.1740775780510884e-06, + "loss": 0.0028, + "step": 7950 + }, + { + "epoch": 2.38, + "learning_rate": 3.171449595290655e-06, + "loss": 0.0014, + "step": 7960 + }, + { + "epoch": 2.39, + "learning_rate": 3.168821612530222e-06, + "loss": 0.0011, + "step": 7970 + }, + { + "epoch": 2.39, + "learning_rate": 3.1661936297697886e-06, + "loss": 0.0012, + "step": 7980 + }, + { + "epoch": 2.39, + "learning_rate": 3.1635656470093556e-06, + "loss": 0.0037, + "step": 7990 + }, + { + "epoch": 2.4, + "learning_rate": 3.1609376642489226e-06, + "loss": 0.001, + "step": 8000 + }, + { + "epoch": 2.4, + "eval_f1": 0.9936892190071861, + "eval_f1_0": 0.998020679637848, + "eval_f1_1": 0.998313736912299, + "eval_f1_2": 0.9794466934757172, + "eval_f1_3": 0.9989757660028803, + "eval_loss": 0.014697457663714886, + "eval_runtime": 153.2338, + "eval_samples_per_second": 77.45, + "eval_steps_per_second": 0.307, + "step": 8000 + }, + { + "epoch": 2.4, + "learning_rate": 3.1583096814884896e-06, + "loss": 0.0013, + "step": 8010 + }, + { + "epoch": 2.4, + "learning_rate": 3.1556816987280566e-06, + "loss": 0.0014, + "step": 8020 + }, + { + "epoch": 2.41, + "learning_rate": 3.153053715967623e-06, + "loss": 0.002, + "step": 8030 + }, + { + "epoch": 2.41, + "learning_rate": 3.15042573320719e-06, + "loss": 0.0013, + "step": 8040 + }, + { + "epoch": 2.41, + "learning_rate": 3.147797750446757e-06, + "loss": 0.0021, + "step": 8050 + }, + { + "epoch": 2.41, + "learning_rate": 3.1451697676863242e-06, + "loss": 0.0083, + "step": 8060 + }, + { + "epoch": 2.42, + "learning_rate": 3.1425417849258912e-06, + "loss": 0.0014, + "step": 8070 + }, + { + "epoch": 2.42, + "learning_rate": 3.139913802165458e-06, + "loss": 0.0013, + "step": 8080 + }, + { + "epoch": 2.42, + "learning_rate": 3.137285819405025e-06, + "loss": 0.0012, + "step": 8090 + }, + { + "epoch": 2.43, + "learning_rate": 3.134657836644592e-06, + "loss": 0.002, + "step": 8100 + }, + { + "epoch": 2.43, + "learning_rate": 3.132029853884159e-06, + "loss": 0.0065, + "step": 8110 + }, + { + "epoch": 2.43, + "learning_rate": 3.1294018711237254e-06, + "loss": 0.0031, + "step": 8120 + }, + { + "epoch": 2.44, + "learning_rate": 3.1267738883632924e-06, + "loss": 0.0019, + "step": 8130 + }, + { + "epoch": 2.44, + "learning_rate": 3.1241459056028594e-06, + "loss": 0.0045, + "step": 8140 + }, + { + "epoch": 2.44, + "learning_rate": 3.1215179228424265e-06, + "loss": 0.0023, + "step": 8150 + }, + { + "epoch": 2.44, + "learning_rate": 3.1188899400819935e-06, + "loss": 0.0101, + "step": 8160 + }, + { + "epoch": 2.45, + "learning_rate": 3.11626195732156e-06, + "loss": 0.0031, + "step": 8170 + }, + { + "epoch": 2.45, + "learning_rate": 3.113633974561127e-06, + "loss": 0.0053, + "step": 8180 + }, + { + "epoch": 2.45, + "learning_rate": 3.111005991800694e-06, + "loss": 0.0093, + "step": 8190 + }, + { + "epoch": 2.46, + "learning_rate": 3.108378009040261e-06, + "loss": 0.0034, + "step": 8200 + }, + { + "epoch": 2.46, + "eval_f1": 0.9929376939918741, + "eval_f1_0": 0.9978118733899454, + "eval_f1_1": 0.9983588957055215, + "eval_f1_2": 0.9764769012588362, + "eval_f1_3": 0.9991031056131934, + "eval_loss": 0.016829997301101685, + "eval_runtime": 153.1299, + "eval_samples_per_second": 77.503, + "eval_steps_per_second": 0.307, + "step": 8200 + }, + { + "epoch": 2.46, + "learning_rate": 3.105750026279828e-06, + "loss": 0.0017, + "step": 8210 + }, + { + "epoch": 2.46, + "learning_rate": 3.1031220435193947e-06, + "loss": 0.0234, + "step": 8220 + }, + { + "epoch": 2.47, + "learning_rate": 3.1004940607589617e-06, + "loss": 0.0025, + "step": 8230 + }, + { + "epoch": 2.47, + "learning_rate": 3.0978660779985287e-06, + "loss": 0.0083, + "step": 8240 + }, + { + "epoch": 2.47, + "learning_rate": 3.0952380952380957e-06, + "loss": 0.0013, + "step": 8250 + }, + { + "epoch": 2.47, + "learning_rate": 3.0926101124776627e-06, + "loss": 0.0027, + "step": 8260 + }, + { + "epoch": 2.48, + "learning_rate": 3.0899821297172293e-06, + "loss": 0.0155, + "step": 8270 + }, + { + "epoch": 2.48, + "learning_rate": 3.0873541469567963e-06, + "loss": 0.0014, + "step": 8280 + }, + { + "epoch": 2.48, + "learning_rate": 3.0847261641963633e-06, + "loss": 0.0027, + "step": 8290 + }, + { + "epoch": 2.49, + "learning_rate": 3.0820981814359303e-06, + "loss": 0.0024, + "step": 8300 + }, + { + "epoch": 2.49, + "learning_rate": 3.079470198675497e-06, + "loss": 0.0021, + "step": 8310 + }, + { + "epoch": 2.49, + "learning_rate": 3.076842215915064e-06, + "loss": 0.0066, + "step": 8320 + }, + { + "epoch": 2.5, + "learning_rate": 3.074214233154631e-06, + "loss": 0.0016, + "step": 8330 + }, + { + "epoch": 2.5, + "learning_rate": 3.071586250394198e-06, + "loss": 0.0016, + "step": 8340 + }, + { + "epoch": 2.5, + "learning_rate": 3.068958267633765e-06, + "loss": 0.0018, + "step": 8350 + }, + { + "epoch": 2.5, + "learning_rate": 3.0663302848733316e-06, + "loss": 0.001, + "step": 8360 + }, + { + "epoch": 2.51, + "learning_rate": 3.0637023021128986e-06, + "loss": 0.0014, + "step": 8370 + }, + { + "epoch": 2.51, + "learning_rate": 3.0610743193524656e-06, + "loss": 0.0018, + "step": 8380 + }, + { + "epoch": 2.51, + "learning_rate": 3.0584463365920317e-06, + "loss": 0.0078, + "step": 8390 + }, + { + "epoch": 2.52, + "learning_rate": 3.0558183538315987e-06, + "loss": 0.0059, + "step": 8400 + }, + { + "epoch": 2.52, + "eval_f1": 0.9934315256501149, + "eval_f1_0": 0.9979495558595631, + "eval_f1_1": 0.998370917231096, + "eval_f1_2": 0.9784080408832903, + "eval_f1_3": 0.9989975886265102, + "eval_loss": 0.016374990344047546, + "eval_runtime": 153.3693, + "eval_samples_per_second": 77.382, + "eval_steps_per_second": 0.306, + "step": 8400 + }, + { + "epoch": 2.52, + "learning_rate": 3.0531903710711658e-06, + "loss": 0.0028, + "step": 8410 + }, + { + "epoch": 2.52, + "learning_rate": 3.0505623883107328e-06, + "loss": 0.001, + "step": 8420 + }, + { + "epoch": 2.53, + "learning_rate": 3.0479344055502998e-06, + "loss": 0.0082, + "step": 8430 + }, + { + "epoch": 2.53, + "learning_rate": 3.0453064227898664e-06, + "loss": 0.0043, + "step": 8440 + }, + { + "epoch": 2.53, + "learning_rate": 3.0426784400294334e-06, + "loss": 0.0016, + "step": 8450 + }, + { + "epoch": 2.53, + "learning_rate": 3.0400504572690004e-06, + "loss": 0.0152, + "step": 8460 + }, + { + "epoch": 2.54, + "learning_rate": 3.0374224745085674e-06, + "loss": 0.0052, + "step": 8470 + }, + { + "epoch": 2.54, + "learning_rate": 3.0347944917481344e-06, + "loss": 0.0163, + "step": 8480 + }, + { + "epoch": 2.54, + "learning_rate": 3.032166508987701e-06, + "loss": 0.0144, + "step": 8490 + }, + { + "epoch": 2.55, + "learning_rate": 3.029538526227268e-06, + "loss": 0.0254, + "step": 8500 + }, + { + "epoch": 2.55, + "learning_rate": 3.026910543466835e-06, + "loss": 0.0029, + "step": 8510 + }, + { + "epoch": 2.55, + "learning_rate": 3.024282560706402e-06, + "loss": 0.0221, + "step": 8520 + }, + { + "epoch": 2.56, + "learning_rate": 3.021654577945969e-06, + "loss": 0.0158, + "step": 8530 + }, + { + "epoch": 2.56, + "learning_rate": 3.0190265951855356e-06, + "loss": 0.0021, + "step": 8540 + }, + { + "epoch": 2.56, + "learning_rate": 3.0163986124251026e-06, + "loss": 0.0319, + "step": 8550 + }, + { + "epoch": 2.56, + "learning_rate": 3.0137706296646696e-06, + "loss": 0.0014, + "step": 8560 + }, + { + "epoch": 2.57, + "learning_rate": 3.0111426469042366e-06, + "loss": 0.0011, + "step": 8570 + }, + { + "epoch": 2.57, + "learning_rate": 3.0085146641438032e-06, + "loss": 0.0016, + "step": 8580 + }, + { + "epoch": 2.57, + "learning_rate": 3.0058866813833702e-06, + "loss": 0.0031, + "step": 8590 + }, + { + "epoch": 2.58, + "learning_rate": 3.0032586986229373e-06, + "loss": 0.0021, + "step": 8600 + }, + { + "epoch": 2.58, + "eval_f1": 0.9944045209341374, + "eval_f1_0": 0.9982423475292743, + "eval_f1_1": 0.9983367696540941, + "eval_f1_2": 0.9819266084672601, + "eval_f1_3": 0.9991123580859212, + "eval_loss": 0.014226074330508709, + "eval_runtime": 153.198, + "eval_samples_per_second": 77.468, + "eval_steps_per_second": 0.307, + "step": 8600 + }, + { + "epoch": 2.58, + "learning_rate": 3.0006307158625043e-06, + "loss": 0.0163, + "step": 8610 + }, + { + "epoch": 2.58, + "learning_rate": 2.9980027331020713e-06, + "loss": 0.003, + "step": 8620 + }, + { + "epoch": 2.59, + "learning_rate": 2.995374750341638e-06, + "loss": 0.0025, + "step": 8630 + }, + { + "epoch": 2.59, + "learning_rate": 2.992746767581205e-06, + "loss": 0.0009, + "step": 8640 + }, + { + "epoch": 2.59, + "learning_rate": 2.990118784820772e-06, + "loss": 0.0018, + "step": 8650 + }, + { + "epoch": 2.59, + "learning_rate": 2.987490802060339e-06, + "loss": 0.0122, + "step": 8660 + }, + { + "epoch": 2.6, + "learning_rate": 2.984862819299906e-06, + "loss": 0.0009, + "step": 8670 + }, + { + "epoch": 2.6, + "learning_rate": 2.9822348365394725e-06, + "loss": 0.0045, + "step": 8680 + }, + { + "epoch": 2.6, + "learning_rate": 2.9796068537790395e-06, + "loss": 0.0016, + "step": 8690 + }, + { + "epoch": 2.61, + "learning_rate": 2.9769788710186065e-06, + "loss": 0.0013, + "step": 8700 + }, + { + "epoch": 2.61, + "learning_rate": 2.9743508882581735e-06, + "loss": 0.0383, + "step": 8710 + }, + { + "epoch": 2.61, + "learning_rate": 2.9717229054977405e-06, + "loss": 0.0053, + "step": 8720 + }, + { + "epoch": 2.62, + "learning_rate": 2.969094922737307e-06, + "loss": 0.0015, + "step": 8730 + }, + { + "epoch": 2.62, + "learning_rate": 2.966466939976874e-06, + "loss": 0.0139, + "step": 8740 + }, + { + "epoch": 2.62, + "learning_rate": 2.963838957216441e-06, + "loss": 0.0029, + "step": 8750 + }, + { + "epoch": 2.62, + "learning_rate": 2.961210974456008e-06, + "loss": 0.0028, + "step": 8760 + }, + { + "epoch": 2.63, + "learning_rate": 2.958582991695575e-06, + "loss": 0.0013, + "step": 8770 + }, + { + "epoch": 2.63, + "learning_rate": 2.9559550089351417e-06, + "loss": 0.003, + "step": 8780 + }, + { + "epoch": 2.63, + "learning_rate": 2.9533270261747088e-06, + "loss": 0.0106, + "step": 8790 + }, + { + "epoch": 2.64, + "learning_rate": 2.9506990434142758e-06, + "loss": 0.0052, + "step": 8800 + }, + { + "epoch": 2.64, + "eval_f1": 0.9918912454619301, + "eval_f1_0": 0.997424502378134, + "eval_f1_1": 0.998214046181853, + "eval_f1_2": 0.972873035606133, + "eval_f1_3": 0.9990533976816002, + "eval_loss": 0.0193399079144001, + "eval_runtime": 153.415, + "eval_samples_per_second": 77.359, + "eval_steps_per_second": 0.306, + "step": 8800 + }, + { + "epoch": 2.64, + "learning_rate": 2.948071060653842e-06, + "loss": 0.0155, + "step": 8810 + }, + { + "epoch": 2.64, + "learning_rate": 2.945443077893409e-06, + "loss": 0.0018, + "step": 8820 + }, + { + "epoch": 2.65, + "learning_rate": 2.942815095132976e-06, + "loss": 0.0025, + "step": 8830 + }, + { + "epoch": 2.65, + "learning_rate": 2.940187112372543e-06, + "loss": 0.0015, + "step": 8840 + }, + { + "epoch": 2.65, + "learning_rate": 2.93755912961211e-06, + "loss": 0.0137, + "step": 8850 + }, + { + "epoch": 2.65, + "learning_rate": 2.9349311468516766e-06, + "loss": 0.0019, + "step": 8860 + }, + { + "epoch": 2.66, + "learning_rate": 2.9323031640912436e-06, + "loss": 0.0017, + "step": 8870 + }, + { + "epoch": 2.66, + "learning_rate": 2.9296751813308106e-06, + "loss": 0.0026, + "step": 8880 + }, + { + "epoch": 2.66, + "learning_rate": 2.9270471985703776e-06, + "loss": 0.0037, + "step": 8890 + }, + { + "epoch": 2.67, + "learning_rate": 2.924419215809944e-06, + "loss": 0.0161, + "step": 8900 + }, + { + "epoch": 2.67, + "learning_rate": 2.921791233049511e-06, + "loss": 0.002, + "step": 8910 + }, + { + "epoch": 2.67, + "learning_rate": 2.919163250289078e-06, + "loss": 0.0052, + "step": 8920 + }, + { + "epoch": 2.68, + "learning_rate": 2.916535267528645e-06, + "loss": 0.0114, + "step": 8930 + }, + { + "epoch": 2.68, + "learning_rate": 2.913907284768212e-06, + "loss": 0.0356, + "step": 8940 + }, + { + "epoch": 2.68, + "learning_rate": 2.911279302007779e-06, + "loss": 0.0127, + "step": 8950 + }, + { + "epoch": 2.68, + "learning_rate": 2.908651319247346e-06, + "loss": 0.0017, + "step": 8960 + }, + { + "epoch": 2.69, + "learning_rate": 2.906023336486913e-06, + "loss": 0.0139, + "step": 8970 + }, + { + "epoch": 2.69, + "learning_rate": 2.90339535372648e-06, + "loss": 0.0015, + "step": 8980 + }, + { + "epoch": 2.69, + "learning_rate": 2.900767370966047e-06, + "loss": 0.0227, + "step": 8990 + }, + { + "epoch": 2.7, + "learning_rate": 2.8981393882056134e-06, + "loss": 0.0014, + "step": 9000 + }, + { + "epoch": 2.7, + "eval_f1": 0.9922554773188612, + "eval_f1_0": 0.9975959459772522, + "eval_f1_1": 0.998426265457006, + "eval_f1_2": 0.9738997655014452, + "eval_f1_3": 0.9990999323397414, + "eval_loss": 0.01743851974606514, + "eval_runtime": 153.1158, + "eval_samples_per_second": 77.51, + "eval_steps_per_second": 0.307, + "step": 9000 + }, + { + "epoch": 2.7, + "learning_rate": 2.8955114054451804e-06, + "loss": 0.0026, + "step": 9010 + }, + { + "epoch": 2.7, + "learning_rate": 2.8928834226847474e-06, + "loss": 0.0019, + "step": 9020 + }, + { + "epoch": 2.71, + "learning_rate": 2.8902554399243145e-06, + "loss": 0.0039, + "step": 9030 + }, + { + "epoch": 2.71, + "learning_rate": 2.8876274571638815e-06, + "loss": 0.0228, + "step": 9040 + }, + { + "epoch": 2.71, + "learning_rate": 2.884999474403448e-06, + "loss": 0.0051, + "step": 9050 + }, + { + "epoch": 2.71, + "learning_rate": 2.882371491643015e-06, + "loss": 0.0477, + "step": 9060 + }, + { + "epoch": 2.72, + "learning_rate": 2.879743508882582e-06, + "loss": 0.0023, + "step": 9070 + }, + { + "epoch": 2.72, + "learning_rate": 2.877115526122149e-06, + "loss": 0.0016, + "step": 9080 + }, + { + "epoch": 2.72, + "learning_rate": 2.8744875433617157e-06, + "loss": 0.0034, + "step": 9090 + }, + { + "epoch": 2.73, + "learning_rate": 2.8718595606012827e-06, + "loss": 0.0017, + "step": 9100 + }, + { + "epoch": 2.73, + "learning_rate": 2.8692315778408497e-06, + "loss": 0.001, + "step": 9110 + }, + { + "epoch": 2.73, + "learning_rate": 2.8666035950804167e-06, + "loss": 0.0065, + "step": 9120 + }, + { + "epoch": 2.74, + "learning_rate": 2.8639756123199837e-06, + "loss": 0.0058, + "step": 9130 + }, + { + "epoch": 2.74, + "learning_rate": 2.8613476295595503e-06, + "loss": 0.0011, + "step": 9140 + }, + { + "epoch": 2.74, + "learning_rate": 2.8587196467991173e-06, + "loss": 0.0039, + "step": 9150 + }, + { + "epoch": 2.74, + "learning_rate": 2.8560916640386843e-06, + "loss": 0.0012, + "step": 9160 + }, + { + "epoch": 2.75, + "learning_rate": 2.8534636812782513e-06, + "loss": 0.0023, + "step": 9170 + }, + { + "epoch": 2.75, + "learning_rate": 2.8508356985178183e-06, + "loss": 0.0017, + "step": 9180 + }, + { + "epoch": 2.75, + "learning_rate": 2.848207715757385e-06, + "loss": 0.0053, + "step": 9190 + }, + { + "epoch": 2.76, + "learning_rate": 2.845579732996952e-06, + "loss": 0.0051, + "step": 9200 + }, + { + "epoch": 2.76, + "eval_f1": 0.9922298711940248, + "eval_f1_0": 0.9975922583433169, + "eval_f1_1": 0.9985417293441998, + "eval_f1_2": 0.9736916764872768, + "eval_f1_3": 0.9990938206013059, + "eval_loss": 0.019912779331207275, + "eval_runtime": 153.1079, + "eval_samples_per_second": 77.514, + "eval_steps_per_second": 0.307, + "step": 9200 + }, + { + "epoch": 2.76, + "learning_rate": 2.842951750236519e-06, + "loss": 0.0239, + "step": 9210 + }, + { + "epoch": 2.76, + "learning_rate": 2.840323767476086e-06, + "loss": 0.0017, + "step": 9220 + }, + { + "epoch": 2.77, + "learning_rate": 2.837695784715653e-06, + "loss": 0.0012, + "step": 9230 + }, + { + "epoch": 2.77, + "learning_rate": 2.835067801955219e-06, + "loss": 0.0028, + "step": 9240 + }, + { + "epoch": 2.77, + "learning_rate": 2.832439819194786e-06, + "loss": 0.0053, + "step": 9250 + }, + { + "epoch": 2.77, + "learning_rate": 2.829811836434353e-06, + "loss": 0.0144, + "step": 9260 + }, + { + "epoch": 2.78, + "learning_rate": 2.8271838536739197e-06, + "loss": 0.0061, + "step": 9270 + }, + { + "epoch": 2.78, + "learning_rate": 2.8245558709134867e-06, + "loss": 0.0116, + "step": 9280 + }, + { + "epoch": 2.78, + "learning_rate": 2.8219278881530538e-06, + "loss": 0.0032, + "step": 9290 + }, + { + "epoch": 2.79, + "learning_rate": 2.8192999053926208e-06, + "loss": 0.0131, + "step": 9300 + }, + { + "epoch": 2.79, + "learning_rate": 2.8166719226321878e-06, + "loss": 0.0158, + "step": 9310 + }, + { + "epoch": 2.79, + "learning_rate": 2.8140439398717544e-06, + "loss": 0.0048, + "step": 9320 + }, + { + "epoch": 2.8, + "learning_rate": 2.8114159571113214e-06, + "loss": 0.0023, + "step": 9330 + }, + { + "epoch": 2.8, + "learning_rate": 2.8087879743508884e-06, + "loss": 0.0054, + "step": 9340 + }, + { + "epoch": 2.8, + "learning_rate": 2.8061599915904554e-06, + "loss": 0.0033, + "step": 9350 + }, + { + "epoch": 2.8, + "learning_rate": 2.8035320088300224e-06, + "loss": 0.0043, + "step": 9360 + }, + { + "epoch": 2.81, + "learning_rate": 2.800904026069589e-06, + "loss": 0.0018, + "step": 9370 + }, + { + "epoch": 2.81, + "learning_rate": 2.798276043309156e-06, + "loss": 0.0026, + "step": 9380 + }, + { + "epoch": 2.81, + "learning_rate": 2.795648060548723e-06, + "loss": 0.0169, + "step": 9390 + }, + { + "epoch": 2.82, + "learning_rate": 2.79302007778829e-06, + "loss": 0.0016, + "step": 9400 + }, + { + "epoch": 2.82, + "eval_f1": 0.9761104838258727, + "eval_f1_0": 0.9918857239064715, + "eval_f1_1": 0.998355256852138, + "eval_f1_2": 0.9151040670242819, + "eval_f1_3": 0.9990968875205993, + "eval_loss": 0.06116287410259247, + "eval_runtime": 153.0787, + "eval_samples_per_second": 77.529, + "eval_steps_per_second": 0.307, + "step": 9400 + }, + { + "epoch": 2.82, + "learning_rate": 2.7903920950278566e-06, + "loss": 0.0257, + "step": 9410 + }, + { + "epoch": 2.82, + "learning_rate": 2.7877641122674236e-06, + "loss": 0.0014, + "step": 9420 + }, + { + "epoch": 2.83, + "learning_rate": 2.7851361295069906e-06, + "loss": 0.0048, + "step": 9430 + }, + { + "epoch": 2.83, + "learning_rate": 2.7825081467465576e-06, + "loss": 0.0025, + "step": 9440 + }, + { + "epoch": 2.83, + "learning_rate": 2.7798801639861246e-06, + "loss": 0.0012, + "step": 9450 + }, + { + "epoch": 2.83, + "learning_rate": 2.7772521812256912e-06, + "loss": 0.0187, + "step": 9460 + }, + { + "epoch": 2.84, + "learning_rate": 2.7746241984652582e-06, + "loss": 0.0066, + "step": 9470 + }, + { + "epoch": 2.84, + "learning_rate": 2.7719962157048252e-06, + "loss": 0.0115, + "step": 9480 + }, + { + "epoch": 2.84, + "learning_rate": 2.7693682329443923e-06, + "loss": 0.0017, + "step": 9490 + }, + { + "epoch": 2.85, + "learning_rate": 2.7667402501839593e-06, + "loss": 0.0021, + "step": 9500 + }, + { + "epoch": 2.85, + "learning_rate": 2.764112267423526e-06, + "loss": 0.0012, + "step": 9510 + }, + { + "epoch": 2.85, + "learning_rate": 2.761484284663093e-06, + "loss": 0.0018, + "step": 9520 + }, + { + "epoch": 2.86, + "learning_rate": 2.75885630190266e-06, + "loss": 0.0013, + "step": 9530 + }, + { + "epoch": 2.86, + "learning_rate": 2.756228319142227e-06, + "loss": 0.0009, + "step": 9540 + }, + { + "epoch": 2.86, + "learning_rate": 2.753600336381794e-06, + "loss": 0.0077, + "step": 9550 + }, + { + "epoch": 2.86, + "learning_rate": 2.7509723536213605e-06, + "loss": 0.0008, + "step": 9560 + }, + { + "epoch": 2.87, + "learning_rate": 2.7483443708609275e-06, + "loss": 0.0066, + "step": 9570 + }, + { + "epoch": 2.87, + "learning_rate": 2.7457163881004945e-06, + "loss": 0.0204, + "step": 9580 + }, + { + "epoch": 2.87, + "learning_rate": 2.7430884053400615e-06, + "loss": 0.0016, + "step": 9590 + }, + { + "epoch": 2.88, + "learning_rate": 2.740460422579628e-06, + "loss": 0.0045, + "step": 9600 + }, + { + "epoch": 2.88, + "eval_f1": 0.9937663379693042, + "eval_f1_0": 0.9980625896221533, + "eval_f1_1": 0.998526551767084, + "eval_f1_2": 0.9793855017063364, + "eval_f1_3": 0.9990907087816429, + "eval_loss": 0.014854556880891323, + "eval_runtime": 153.2727, + "eval_samples_per_second": 77.431, + "eval_steps_per_second": 0.307, + "step": 9600 + }, + { + "epoch": 2.88, + "learning_rate": 2.737832439819195e-06, + "loss": 0.0019, + "step": 9610 + }, + { + "epoch": 2.88, + "learning_rate": 2.735204457058762e-06, + "loss": 0.0123, + "step": 9620 + }, + { + "epoch": 2.88, + "learning_rate": 2.732576474298329e-06, + "loss": 0.0025, + "step": 9630 + }, + { + "epoch": 2.89, + "learning_rate": 2.729948491537896e-06, + "loss": 0.0017, + "step": 9640 + }, + { + "epoch": 2.89, + "learning_rate": 2.7273205087774627e-06, + "loss": 0.0043, + "step": 9650 + }, + { + "epoch": 2.89, + "learning_rate": 2.7246925260170297e-06, + "loss": 0.0029, + "step": 9660 + }, + { + "epoch": 2.9, + "learning_rate": 2.7220645432565963e-06, + "loss": 0.0021, + "step": 9670 + }, + { + "epoch": 2.9, + "learning_rate": 2.719436560496163e-06, + "loss": 0.0033, + "step": 9680 + }, + { + "epoch": 2.9, + "learning_rate": 2.71680857773573e-06, + "loss": 0.0033, + "step": 9690 + }, + { + "epoch": 2.91, + "learning_rate": 2.714180594975297e-06, + "loss": 0.0012, + "step": 9700 + }, + { + "epoch": 2.91, + "learning_rate": 2.711552612214864e-06, + "loss": 0.0099, + "step": 9710 + }, + { + "epoch": 2.91, + "learning_rate": 2.708924629454431e-06, + "loss": 0.0018, + "step": 9720 + }, + { + "epoch": 2.91, + "learning_rate": 2.7062966466939975e-06, + "loss": 0.0022, + "step": 9730 + }, + { + "epoch": 2.92, + "learning_rate": 2.7036686639335645e-06, + "loss": 0.0024, + "step": 9740 + }, + { + "epoch": 2.92, + "learning_rate": 2.7010406811731316e-06, + "loss": 0.0011, + "step": 9750 + }, + { + "epoch": 2.92, + "learning_rate": 2.6984126984126986e-06, + "loss": 0.0017, + "step": 9760 + }, + { + "epoch": 2.93, + "learning_rate": 2.6957847156522656e-06, + "loss": 0.0239, + "step": 9770 + }, + { + "epoch": 2.93, + "learning_rate": 2.693156732891832e-06, + "loss": 0.0018, + "step": 9780 + }, + { + "epoch": 2.93, + "learning_rate": 2.690528750131399e-06, + "loss": 0.0028, + "step": 9790 + }, + { + "epoch": 2.94, + "learning_rate": 2.687900767370966e-06, + "loss": 0.001, + "step": 9800 + }, + { + "epoch": 2.94, + "eval_f1": 0.9929864600519772, + "eval_f1_0": 0.9977926101144773, + "eval_f1_1": 0.9985662422897187, + "eval_f1_2": 0.976521048111115, + "eval_f1_3": 0.9990659396925979, + "eval_loss": 0.016506973654031754, + "eval_runtime": 153.0798, + "eval_samples_per_second": 77.528, + "eval_steps_per_second": 0.307, + "step": 9800 + }, + { + "epoch": 2.94, + "learning_rate": 2.685272784610533e-06, + "loss": 0.0014, + "step": 9810 + }, + { + "epoch": 2.94, + "learning_rate": 2.6826448018501e-06, + "loss": 0.0028, + "step": 9820 + }, + { + "epoch": 2.94, + "learning_rate": 2.680016819089667e-06, + "loss": 0.0031, + "step": 9830 + }, + { + "epoch": 2.95, + "learning_rate": 2.677388836329234e-06, + "loss": 0.0037, + "step": 9840 + }, + { + "epoch": 2.95, + "learning_rate": 2.674760853568801e-06, + "loss": 0.0061, + "step": 9850 + }, + { + "epoch": 2.95, + "learning_rate": 2.672132870808368e-06, + "loss": 0.0019, + "step": 9860 + }, + { + "epoch": 2.96, + "learning_rate": 2.6695048880479344e-06, + "loss": 0.0012, + "step": 9870 + }, + { + "epoch": 2.96, + "learning_rate": 2.6668769052875014e-06, + "loss": 0.0021, + "step": 9880 + }, + { + "epoch": 2.96, + "learning_rate": 2.6642489225270684e-06, + "loss": 0.0013, + "step": 9890 + }, + { + "epoch": 2.97, + "learning_rate": 2.6616209397666354e-06, + "loss": 0.0023, + "step": 9900 + }, + { + "epoch": 2.97, + "learning_rate": 2.6589929570062024e-06, + "loss": 0.0064, + "step": 9910 + }, + { + "epoch": 2.97, + "learning_rate": 2.656364974245769e-06, + "loss": 0.0037, + "step": 9920 + }, + { + "epoch": 2.97, + "learning_rate": 2.653736991485336e-06, + "loss": 0.0036, + "step": 9930 + }, + { + "epoch": 2.98, + "learning_rate": 2.651109008724903e-06, + "loss": 0.0014, + "step": 9940 + }, + { + "epoch": 2.98, + "learning_rate": 2.64848102596447e-06, + "loss": 0.0006, + "step": 9950 + }, + { + "epoch": 2.98, + "learning_rate": 2.645853043204037e-06, + "loss": 0.0017, + "step": 9960 + }, + { + "epoch": 2.99, + "learning_rate": 2.6432250604436037e-06, + "loss": 0.0074, + "step": 9970 + }, + { + "epoch": 2.99, + "learning_rate": 2.6405970776831707e-06, + "loss": 0.0207, + "step": 9980 + }, + { + "epoch": 2.99, + "learning_rate": 2.6379690949227377e-06, + "loss": 0.0014, + "step": 9990 + }, + { + "epoch": 3.0, + "learning_rate": 2.6353411121623047e-06, + "loss": 0.0027, + "step": 10000 + }, + { + "epoch": 3.0, + "eval_f1": 0.9940681360407999, + "eval_f1_0": 0.9981445880450488, + "eval_f1_1": 0.9983542642564972, + "eval_f1_2": 0.9806705305827429, + "eval_f1_3": 0.9991031612789106, + "eval_loss": 0.013135611079633236, + "eval_runtime": 153.0686, + "eval_samples_per_second": 77.534, + "eval_steps_per_second": 0.307, + "step": 10000 + }, + { + "epoch": 3.0, + "learning_rate": 2.6327131294018717e-06, + "loss": 0.0017, + "step": 10010 + }, + { + "epoch": 3.0, + "learning_rate": 2.6300851466414383e-06, + "loss": 0.0085, + "step": 10020 + }, + { + "epoch": 3.0, + "learning_rate": 2.6274571638810053e-06, + "loss": 0.002, + "step": 10030 + }, + { + "epoch": 3.01, + "learning_rate": 2.6248291811205723e-06, + "loss": 0.0156, + "step": 10040 + }, + { + "epoch": 3.01, + "learning_rate": 2.6222011983601393e-06, + "loss": 0.0121, + "step": 10050 + }, + { + "epoch": 3.01, + "learning_rate": 2.6195732155997063e-06, + "loss": 0.0009, + "step": 10060 + }, + { + "epoch": 3.02, + "learning_rate": 2.616945232839273e-06, + "loss": 0.0013, + "step": 10070 + }, + { + "epoch": 3.02, + "learning_rate": 2.61431725007884e-06, + "loss": 0.0011, + "step": 10080 + }, + { + "epoch": 3.02, + "learning_rate": 2.6116892673184065e-06, + "loss": 0.0025, + "step": 10090 + }, + { + "epoch": 3.03, + "learning_rate": 2.609061284557973e-06, + "loss": 0.0015, + "step": 10100 + }, + { + "epoch": 3.03, + "learning_rate": 2.60643330179754e-06, + "loss": 0.0012, + "step": 10110 + }, + { + "epoch": 3.03, + "learning_rate": 2.603805319037107e-06, + "loss": 0.0011, + "step": 10120 + }, + { + "epoch": 3.03, + "learning_rate": 2.601177336276674e-06, + "loss": 0.0017, + "step": 10130 + }, + { + "epoch": 3.04, + "learning_rate": 2.598549353516241e-06, + "loss": 0.001, + "step": 10140 + }, + { + "epoch": 3.04, + "learning_rate": 2.5959213707558077e-06, + "loss": 0.0013, + "step": 10150 + }, + { + "epoch": 3.04, + "learning_rate": 2.5932933879953747e-06, + "loss": 0.0068, + "step": 10160 + }, + { + "epoch": 3.05, + "learning_rate": 2.5906654052349417e-06, + "loss": 0.002, + "step": 10170 + }, + { + "epoch": 3.05, + "learning_rate": 2.5880374224745088e-06, + "loss": 0.001, + "step": 10180 + }, + { + "epoch": 3.05, + "learning_rate": 2.5854094397140753e-06, + "loss": 0.0008, + "step": 10190 + }, + { + "epoch": 3.06, + "learning_rate": 2.5827814569536424e-06, + "loss": 0.002, + "step": 10200 + }, + { + "epoch": 3.06, + "eval_f1": 0.9944674225355606, + "eval_f1_0": 0.9982657574099039, + "eval_f1_1": 0.9985489550367745, + "eval_f1_2": 0.981948854638505, + "eval_f1_3": 0.9991061230570591, + "eval_loss": 0.016152124851942062, + "eval_runtime": 153.2353, + "eval_samples_per_second": 77.45, + "eval_steps_per_second": 0.307, + "step": 10200 + }, + { + "epoch": 3.06, + "learning_rate": 2.5801534741932094e-06, + "loss": 0.0012, + "step": 10210 + }, + { + "epoch": 3.06, + "learning_rate": 2.5775254914327764e-06, + "loss": 0.0009, + "step": 10220 + }, + { + "epoch": 3.06, + "learning_rate": 2.5748975086723434e-06, + "loss": 0.002, + "step": 10230 + }, + { + "epoch": 3.07, + "learning_rate": 2.57226952591191e-06, + "loss": 0.0016, + "step": 10240 + }, + { + "epoch": 3.07, + "learning_rate": 2.569641543151477e-06, + "loss": 0.0014, + "step": 10250 + }, + { + "epoch": 3.07, + "learning_rate": 2.567013560391044e-06, + "loss": 0.0023, + "step": 10260 + }, + { + "epoch": 3.08, + "learning_rate": 2.564385577630611e-06, + "loss": 0.0066, + "step": 10270 + }, + { + "epoch": 3.08, + "learning_rate": 2.561757594870178e-06, + "loss": 0.0016, + "step": 10280 + }, + { + "epoch": 3.08, + "learning_rate": 2.5591296121097446e-06, + "loss": 0.002, + "step": 10290 + }, + { + "epoch": 3.09, + "learning_rate": 2.5565016293493116e-06, + "loss": 0.0027, + "step": 10300 + }, + { + "epoch": 3.09, + "learning_rate": 2.5538736465888786e-06, + "loss": 0.0022, + "step": 10310 + }, + { + "epoch": 3.09, + "learning_rate": 2.5512456638284456e-06, + "loss": 0.0042, + "step": 10320 + }, + { + "epoch": 3.09, + "learning_rate": 2.5486176810680126e-06, + "loss": 0.0026, + "step": 10330 + }, + { + "epoch": 3.1, + "learning_rate": 2.5459896983075792e-06, + "loss": 0.0014, + "step": 10340 + }, + { + "epoch": 3.1, + "learning_rate": 2.5433617155471462e-06, + "loss": 0.0015, + "step": 10350 + }, + { + "epoch": 3.1, + "learning_rate": 2.5407337327867132e-06, + "loss": 0.0011, + "step": 10360 + }, + { + "epoch": 3.11, + "learning_rate": 2.5381057500262803e-06, + "loss": 0.001, + "step": 10370 + }, + { + "epoch": 3.11, + "learning_rate": 2.535477767265847e-06, + "loss": 0.0018, + "step": 10380 + }, + { + "epoch": 3.11, + "learning_rate": 2.532849784505414e-06, + "loss": 0.0026, + "step": 10390 + }, + { + "epoch": 3.12, + "learning_rate": 2.530221801744981e-06, + "loss": 0.0027, + "step": 10400 + }, + { + "epoch": 3.12, + "eval_f1": 0.9943387857877316, + "eval_f1_0": 0.9982231258109571, + "eval_f1_1": 0.9984278236861782, + "eval_f1_2": 0.9815886687876261, + "eval_f1_3": 0.999115524866165, + "eval_loss": 0.01668805629014969, + "eval_runtime": 153.0651, + "eval_samples_per_second": 77.536, + "eval_steps_per_second": 0.307, + "step": 10400 + }, + { + "epoch": 3.12, + "learning_rate": 2.527593818984548e-06, + "loss": 0.0021, + "step": 10410 + }, + { + "epoch": 3.12, + "learning_rate": 2.524965836224115e-06, + "loss": 0.0013, + "step": 10420 + }, + { + "epoch": 3.12, + "learning_rate": 2.5223378534636815e-06, + "loss": 0.0015, + "step": 10430 + }, + { + "epoch": 3.13, + "learning_rate": 2.5197098707032485e-06, + "loss": 0.0012, + "step": 10440 + }, + { + "epoch": 3.13, + "learning_rate": 2.5170818879428155e-06, + "loss": 0.0012, + "step": 10450 + }, + { + "epoch": 3.13, + "learning_rate": 2.5144539051823825e-06, + "loss": 0.0014, + "step": 10460 + }, + { + "epoch": 3.14, + "learning_rate": 2.5118259224219495e-06, + "loss": 0.0014, + "step": 10470 + }, + { + "epoch": 3.14, + "learning_rate": 2.509197939661516e-06, + "loss": 0.0013, + "step": 10480 + }, + { + "epoch": 3.14, + "learning_rate": 2.506569956901083e-06, + "loss": 0.0018, + "step": 10490 + }, + { + "epoch": 3.15, + "learning_rate": 2.50394197414065e-06, + "loss": 0.0014, + "step": 10500 + }, + { + "epoch": 3.15, + "learning_rate": 2.501313991380217e-06, + "loss": 0.0016, + "step": 10510 + }, + { + "epoch": 3.15, + "learning_rate": 2.4986860086197837e-06, + "loss": 0.0026, + "step": 10520 + }, + { + "epoch": 3.15, + "learning_rate": 2.4960580258593507e-06, + "loss": 0.0021, + "step": 10530 + }, + { + "epoch": 3.16, + "learning_rate": 2.4934300430989173e-06, + "loss": 0.0021, + "step": 10540 + }, + { + "epoch": 3.16, + "learning_rate": 2.4908020603384843e-06, + "loss": 0.0015, + "step": 10550 + }, + { + "epoch": 3.16, + "learning_rate": 2.4881740775780513e-06, + "loss": 0.0101, + "step": 10560 + }, + { + "epoch": 3.17, + "learning_rate": 2.4855460948176183e-06, + "loss": 0.001, + "step": 10570 + }, + { + "epoch": 3.17, + "learning_rate": 2.4829181120571853e-06, + "loss": 0.002, + "step": 10580 + }, + { + "epoch": 3.17, + "learning_rate": 2.480290129296752e-06, + "loss": 0.0139, + "step": 10590 + }, + { + "epoch": 3.18, + "learning_rate": 2.477662146536319e-06, + "loss": 0.0016, + "step": 10600 + }, + { + "epoch": 3.18, + "eval_f1": 0.9928868874420482, + "eval_f1_0": 0.9977685703070324, + "eval_f1_1": 0.9987504695321543, + "eval_f1_2": 0.9759378237229549, + "eval_f1_3": 0.9990906862060511, + "eval_loss": 0.018006889149546623, + "eval_runtime": 153.1039, + "eval_samples_per_second": 77.516, + "eval_steps_per_second": 0.307, + "step": 10600 + }, + { + "epoch": 3.18, + "learning_rate": 2.475034163775886e-06, + "loss": 0.013, + "step": 10610 + }, + { + "epoch": 3.18, + "learning_rate": 2.472406181015453e-06, + "loss": 0.0014, + "step": 10620 + }, + { + "epoch": 3.18, + "learning_rate": 2.4697781982550195e-06, + "loss": 0.0014, + "step": 10630 + }, + { + "epoch": 3.19, + "learning_rate": 2.4671502154945866e-06, + "loss": 0.0013, + "step": 10640 + }, + { + "epoch": 3.19, + "learning_rate": 2.4645222327341536e-06, + "loss": 0.0011, + "step": 10650 + }, + { + "epoch": 3.19, + "learning_rate": 2.46189424997372e-06, + "loss": 0.0016, + "step": 10660 + }, + { + "epoch": 3.2, + "learning_rate": 2.459266267213287e-06, + "loss": 0.0025, + "step": 10670 + }, + { + "epoch": 3.2, + "learning_rate": 2.456638284452854e-06, + "loss": 0.0011, + "step": 10680 + }, + { + "epoch": 3.2, + "learning_rate": 2.454010301692421e-06, + "loss": 0.0009, + "step": 10690 + }, + { + "epoch": 3.21, + "learning_rate": 2.4513823189319878e-06, + "loss": 0.0016, + "step": 10700 + }, + { + "epoch": 3.21, + "learning_rate": 2.4487543361715548e-06, + "loss": 0.0007, + "step": 10710 + }, + { + "epoch": 3.21, + "learning_rate": 2.446126353411122e-06, + "loss": 0.0015, + "step": 10720 + }, + { + "epoch": 3.21, + "learning_rate": 2.443498370650689e-06, + "loss": 0.005, + "step": 10730 + }, + { + "epoch": 3.22, + "learning_rate": 2.440870387890256e-06, + "loss": 0.0017, + "step": 10740 + }, + { + "epoch": 3.22, + "learning_rate": 2.4382424051298224e-06, + "loss": 0.0105, + "step": 10750 + }, + { + "epoch": 3.22, + "learning_rate": 2.4356144223693894e-06, + "loss": 0.0013, + "step": 10760 + }, + { + "epoch": 3.23, + "learning_rate": 2.4329864396089564e-06, + "loss": 0.0007, + "step": 10770 + }, + { + "epoch": 3.23, + "learning_rate": 2.4303584568485234e-06, + "loss": 0.0007, + "step": 10780 + }, + { + "epoch": 3.23, + "learning_rate": 2.4277304740880904e-06, + "loss": 0.0015, + "step": 10790 + }, + { + "epoch": 3.24, + "learning_rate": 2.425102491327657e-06, + "loss": 0.0013, + "step": 10800 + }, + { + "epoch": 3.24, + "eval_f1": 0.9937175001119303, + "eval_f1_0": 0.9980423432210683, + "eval_f1_1": 0.9985875841556455, + "eval_f1_2": 0.9791648037199321, + "eval_f1_3": 0.9990752693510749, + "eval_loss": 0.017555907368659973, + "eval_runtime": 153.1273, + "eval_samples_per_second": 77.504, + "eval_steps_per_second": 0.307, + "step": 10800 + }, + { + "epoch": 3.24, + "learning_rate": 2.422474508567224e-06, + "loss": 0.0141, + "step": 10810 + }, + { + "epoch": 3.24, + "learning_rate": 2.419846525806791e-06, + "loss": 0.0086, + "step": 10820 + }, + { + "epoch": 3.24, + "learning_rate": 2.417218543046358e-06, + "loss": 0.0033, + "step": 10830 + }, + { + "epoch": 3.25, + "learning_rate": 2.4145905602859246e-06, + "loss": 0.0012, + "step": 10840 + }, + { + "epoch": 3.25, + "learning_rate": 2.4119625775254917e-06, + "loss": 0.0011, + "step": 10850 + }, + { + "epoch": 3.25, + "learning_rate": 2.4093345947650582e-06, + "loss": 0.0007, + "step": 10860 + }, + { + "epoch": 3.26, + "learning_rate": 2.4067066120046253e-06, + "loss": 0.0012, + "step": 10870 + }, + { + "epoch": 3.26, + "learning_rate": 2.4040786292441923e-06, + "loss": 0.0008, + "step": 10880 + }, + { + "epoch": 3.26, + "learning_rate": 2.4014506464837593e-06, + "loss": 0.0119, + "step": 10890 + }, + { + "epoch": 3.27, + "learning_rate": 2.3988226637233263e-06, + "loss": 0.0018, + "step": 10900 + }, + { + "epoch": 3.27, + "learning_rate": 2.396194680962893e-06, + "loss": 0.0075, + "step": 10910 + }, + { + "epoch": 3.27, + "learning_rate": 2.39356669820246e-06, + "loss": 0.0028, + "step": 10920 + }, + { + "epoch": 3.27, + "learning_rate": 2.390938715442027e-06, + "loss": 0.0009, + "step": 10930 + }, + { + "epoch": 3.28, + "learning_rate": 2.388310732681594e-06, + "loss": 0.0221, + "step": 10940 + }, + { + "epoch": 3.28, + "learning_rate": 2.385682749921161e-06, + "loss": 0.0058, + "step": 10950 + }, + { + "epoch": 3.28, + "learning_rate": 2.3830547671607275e-06, + "loss": 0.0019, + "step": 10960 + }, + { + "epoch": 3.29, + "learning_rate": 2.3804267844002945e-06, + "loss": 0.0016, + "step": 10970 + }, + { + "epoch": 3.29, + "learning_rate": 2.3777988016398615e-06, + "loss": 0.0012, + "step": 10980 + }, + { + "epoch": 3.29, + "learning_rate": 2.3751708188794285e-06, + "loss": 0.0013, + "step": 10990 + }, + { + "epoch": 3.3, + "learning_rate": 2.3725428361189955e-06, + "loss": 0.0011, + "step": 11000 + }, + { + "epoch": 3.3, + "eval_f1": 0.9948161277346173, + "eval_f1_0": 0.9983735715945335, + "eval_f1_1": 0.9984181917535063, + "eval_f1_2": 0.9833696308435398, + "eval_f1_3": 0.9991031167468896, + "eval_loss": 0.013583315536379814, + "eval_runtime": 153.1964, + "eval_samples_per_second": 77.469, + "eval_steps_per_second": 0.307, + "step": 11000 + }, + { + "epoch": 3.3, + "learning_rate": 2.369914853358562e-06, + "loss": 0.0064, + "step": 11010 + }, + { + "epoch": 3.3, + "learning_rate": 2.367286870598129e-06, + "loss": 0.0013, + "step": 11020 + }, + { + "epoch": 3.3, + "learning_rate": 2.364658887837696e-06, + "loss": 0.0023, + "step": 11030 + }, + { + "epoch": 3.31, + "learning_rate": 2.362030905077263e-06, + "loss": 0.0007, + "step": 11040 + }, + { + "epoch": 3.31, + "learning_rate": 2.3594029223168297e-06, + "loss": 0.0015, + "step": 11050 + }, + { + "epoch": 3.31, + "learning_rate": 2.3567749395563967e-06, + "loss": 0.0067, + "step": 11060 + }, + { + "epoch": 3.32, + "learning_rate": 2.3541469567959633e-06, + "loss": 0.0012, + "step": 11070 + }, + { + "epoch": 3.32, + "learning_rate": 2.3515189740355303e-06, + "loss": 0.0023, + "step": 11080 + }, + { + "epoch": 3.32, + "learning_rate": 2.3488909912750974e-06, + "loss": 0.0013, + "step": 11090 + }, + { + "epoch": 3.33, + "learning_rate": 2.3462630085146644e-06, + "loss": 0.0011, + "step": 11100 + }, + { + "epoch": 3.33, + "learning_rate": 2.3436350257542314e-06, + "loss": 0.0182, + "step": 11110 + }, + { + "epoch": 3.33, + "learning_rate": 2.341007042993798e-06, + "loss": 0.001, + "step": 11120 + }, + { + "epoch": 3.33, + "learning_rate": 2.338379060233365e-06, + "loss": 0.0009, + "step": 11130 + }, + { + "epoch": 3.34, + "learning_rate": 2.335751077472932e-06, + "loss": 0.0285, + "step": 11140 + }, + { + "epoch": 3.34, + "learning_rate": 2.333123094712499e-06, + "loss": 0.0043, + "step": 11150 + }, + { + "epoch": 3.34, + "learning_rate": 2.330495111952066e-06, + "loss": 0.0118, + "step": 11160 + }, + { + "epoch": 3.35, + "learning_rate": 2.3278671291916326e-06, + "loss": 0.0015, + "step": 11170 + }, + { + "epoch": 3.35, + "learning_rate": 2.3252391464311996e-06, + "loss": 0.0293, + "step": 11180 + }, + { + "epoch": 3.35, + "learning_rate": 2.3226111636707666e-06, + "loss": 0.001, + "step": 11190 + }, + { + "epoch": 3.36, + "learning_rate": 2.3199831809103336e-06, + "loss": 0.0022, + "step": 11200 + }, + { + "epoch": 3.36, + "eval_f1": 0.9914373770145668, + "eval_f1_0": 0.9972778102568689, + "eval_f1_1": 0.9984624468949863, + "eval_f1_2": 0.970906156427191, + "eval_f1_3": 0.9991030944792206, + "eval_loss": 0.01650741696357727, + "eval_runtime": 153.0993, + "eval_samples_per_second": 77.518, + "eval_steps_per_second": 0.307, + "step": 11200 + }, + { + "epoch": 3.36, + "learning_rate": 2.3173551981499e-06, + "loss": 0.0013, + "step": 11210 + }, + { + "epoch": 3.36, + "learning_rate": 2.3147272153894672e-06, + "loss": 0.0033, + "step": 11220 + }, + { + "epoch": 3.36, + "learning_rate": 2.3120992326290342e-06, + "loss": 0.0012, + "step": 11230 + }, + { + "epoch": 3.37, + "learning_rate": 2.3094712498686012e-06, + "loss": 0.0297, + "step": 11240 + }, + { + "epoch": 3.37, + "learning_rate": 2.3068432671081682e-06, + "loss": 0.0014, + "step": 11250 + }, + { + "epoch": 3.37, + "learning_rate": 2.304215284347735e-06, + "loss": 0.0018, + "step": 11260 + }, + { + "epoch": 3.38, + "learning_rate": 2.301587301587302e-06, + "loss": 0.0016, + "step": 11270 + }, + { + "epoch": 3.38, + "learning_rate": 2.2989593188268684e-06, + "loss": 0.0016, + "step": 11280 + }, + { + "epoch": 3.38, + "learning_rate": 2.2963313360664354e-06, + "loss": 0.0016, + "step": 11290 + }, + { + "epoch": 3.39, + "learning_rate": 2.2937033533060024e-06, + "loss": 0.0018, + "step": 11300 + }, + { + "epoch": 3.39, + "learning_rate": 2.2910753705455695e-06, + "loss": 0.0021, + "step": 11310 + }, + { + "epoch": 3.39, + "learning_rate": 2.288447387785136e-06, + "loss": 0.0009, + "step": 11320 + }, + { + "epoch": 3.39, + "learning_rate": 2.285819405024703e-06, + "loss": 0.0018, + "step": 11330 + }, + { + "epoch": 3.4, + "learning_rate": 2.28319142226427e-06, + "loss": 0.0019, + "step": 11340 + }, + { + "epoch": 3.4, + "learning_rate": 2.280563439503837e-06, + "loss": 0.0028, + "step": 11350 + }, + { + "epoch": 3.4, + "learning_rate": 2.277935456743404e-06, + "loss": 0.0011, + "step": 11360 + }, + { + "epoch": 3.41, + "learning_rate": 2.2753074739829707e-06, + "loss": 0.0013, + "step": 11370 + }, + { + "epoch": 3.41, + "learning_rate": 2.2726794912225377e-06, + "loss": 0.0011, + "step": 11380 + }, + { + "epoch": 3.41, + "learning_rate": 2.2700515084621047e-06, + "loss": 0.0011, + "step": 11390 + }, + { + "epoch": 3.42, + "learning_rate": 2.2674235257016717e-06, + "loss": 0.0154, + "step": 11400 + }, + { + "epoch": 3.42, + "eval_f1": 0.9942353190983868, + "eval_f1_0": 0.99818227671564, + "eval_f1_1": 0.9985109494086917, + "eval_f1_2": 0.9811046238507483, + "eval_f1_3": 0.9991434264184672, + "eval_loss": 0.014985567890107632, + "eval_runtime": 153.1706, + "eval_samples_per_second": 77.482, + "eval_steps_per_second": 0.307, + "step": 11400 + }, + { + "epoch": 3.42, + "learning_rate": 2.2647955429412387e-06, + "loss": 0.0024, + "step": 11410 + }, + { + "epoch": 3.42, + "learning_rate": 2.2621675601808053e-06, + "loss": 0.0044, + "step": 11420 + }, + { + "epoch": 3.42, + "learning_rate": 2.2595395774203723e-06, + "loss": 0.0015, + "step": 11430 + }, + { + "epoch": 3.43, + "learning_rate": 2.2569115946599393e-06, + "loss": 0.0037, + "step": 11440 + }, + { + "epoch": 3.43, + "learning_rate": 2.2542836118995063e-06, + "loss": 0.0011, + "step": 11450 + }, + { + "epoch": 3.43, + "learning_rate": 2.2516556291390733e-06, + "loss": 0.0012, + "step": 11460 + }, + { + "epoch": 3.44, + "learning_rate": 2.24902764637864e-06, + "loss": 0.02, + "step": 11470 + }, + { + "epoch": 3.44, + "learning_rate": 2.2463996636182065e-06, + "loss": 0.0013, + "step": 11480 + }, + { + "epoch": 3.44, + "learning_rate": 2.2437716808577735e-06, + "loss": 0.0015, + "step": 11490 + }, + { + "epoch": 3.45, + "learning_rate": 2.2411436980973405e-06, + "loss": 0.001, + "step": 11500 + }, + { + "epoch": 3.45, + "learning_rate": 2.2385157153369075e-06, + "loss": 0.0103, + "step": 11510 + }, + { + "epoch": 3.45, + "learning_rate": 2.2358877325764746e-06, + "loss": 0.0284, + "step": 11520 + }, + { + "epoch": 3.45, + "learning_rate": 2.233259749816041e-06, + "loss": 0.0025, + "step": 11530 + }, + { + "epoch": 3.46, + "learning_rate": 2.230631767055608e-06, + "loss": 0.0209, + "step": 11540 + }, + { + "epoch": 3.46, + "learning_rate": 2.228003784295175e-06, + "loss": 0.0026, + "step": 11550 + }, + { + "epoch": 3.46, + "learning_rate": 2.225375801534742e-06, + "loss": 0.0016, + "step": 11560 + }, + { + "epoch": 3.47, + "learning_rate": 2.222747818774309e-06, + "loss": 0.0078, + "step": 11570 + }, + { + "epoch": 3.47, + "learning_rate": 2.2201198360138758e-06, + "loss": 0.0029, + "step": 11580 + }, + { + "epoch": 3.47, + "learning_rate": 2.2174918532534428e-06, + "loss": 0.0013, + "step": 11590 + }, + { + "epoch": 3.48, + "learning_rate": 2.2148638704930098e-06, + "loss": 0.002, + "step": 11600 + }, + { + "epoch": 3.48, + "eval_f1": 0.9936369783837441, + "eval_f1_0": 0.9979922839545181, + "eval_f1_1": 0.9986300923681776, + "eval_f1_2": 0.9788378440452727, + "eval_f1_3": 0.999087693167008, + "eval_loss": 0.013910142704844475, + "eval_runtime": 153.1364, + "eval_samples_per_second": 77.5, + "eval_steps_per_second": 0.307, + "step": 11600 + }, + { + "epoch": 3.48, + "learning_rate": 2.212235887732577e-06, + "loss": 0.0008, + "step": 11610 + }, + { + "epoch": 3.48, + "learning_rate": 2.209607904972144e-06, + "loss": 0.0032, + "step": 11620 + }, + { + "epoch": 3.48, + "learning_rate": 2.2069799222117104e-06, + "loss": 0.004, + "step": 11630 + }, + { + "epoch": 3.49, + "learning_rate": 2.2043519394512774e-06, + "loss": 0.0033, + "step": 11640 + }, + { + "epoch": 3.49, + "learning_rate": 2.2017239566908444e-06, + "loss": 0.0008, + "step": 11650 + }, + { + "epoch": 3.49, + "learning_rate": 2.1990959739304114e-06, + "loss": 0.0032, + "step": 11660 + }, + { + "epoch": 3.5, + "learning_rate": 2.196467991169978e-06, + "loss": 0.0011, + "step": 11670 + }, + { + "epoch": 3.5, + "learning_rate": 2.193840008409545e-06, + "loss": 0.0005, + "step": 11680 + }, + { + "epoch": 3.5, + "learning_rate": 2.1912120256491116e-06, + "loss": 0.0127, + "step": 11690 + }, + { + "epoch": 3.51, + "learning_rate": 2.1885840428886786e-06, + "loss": 0.0018, + "step": 11700 + }, + { + "epoch": 3.51, + "learning_rate": 2.1859560601282456e-06, + "loss": 0.0015, + "step": 11710 + }, + { + "epoch": 3.51, + "learning_rate": 2.1833280773678126e-06, + "loss": 0.0021, + "step": 11720 + }, + { + "epoch": 3.51, + "learning_rate": 2.1807000946073796e-06, + "loss": 0.0146, + "step": 11730 + }, + { + "epoch": 3.52, + "learning_rate": 2.1780721118469462e-06, + "loss": 0.0041, + "step": 11740 + }, + { + "epoch": 3.52, + "learning_rate": 2.1754441290865132e-06, + "loss": 0.0011, + "step": 11750 + }, + { + "epoch": 3.52, + "learning_rate": 2.1728161463260803e-06, + "loss": 0.0035, + "step": 11760 + }, + { + "epoch": 3.53, + "learning_rate": 2.1701881635656473e-06, + "loss": 0.0029, + "step": 11770 + }, + { + "epoch": 3.53, + "learning_rate": 2.1675601808052143e-06, + "loss": 0.005, + "step": 11780 + }, + { + "epoch": 3.53, + "learning_rate": 2.164932198044781e-06, + "loss": 0.0064, + "step": 11790 + }, + { + "epoch": 3.54, + "learning_rate": 2.162304215284348e-06, + "loss": 0.0013, + "step": 11800 + }, + { + "epoch": 3.54, + "eval_f1": 0.9910180791472074, + "eval_f1_0": 0.9971343488734166, + "eval_f1_1": 0.9984950182227411, + "eval_f1_2": 0.9694358194453245, + "eval_f1_3": 0.9990071300473475, + "eval_loss": 0.02113513834774494, + "eval_runtime": 153.0763, + "eval_samples_per_second": 77.53, + "eval_steps_per_second": 0.307, + "step": 11800 + }, + { + "epoch": 3.54, + "learning_rate": 2.159676232523915e-06, + "loss": 0.0014, + "step": 11810 + }, + { + "epoch": 3.54, + "learning_rate": 2.157048249763482e-06, + "loss": 0.0009, + "step": 11820 + }, + { + "epoch": 3.54, + "learning_rate": 2.1544202670030485e-06, + "loss": 0.0016, + "step": 11830 + }, + { + "epoch": 3.55, + "learning_rate": 2.1517922842426155e-06, + "loss": 0.0007, + "step": 11840 + }, + { + "epoch": 3.55, + "learning_rate": 2.1491643014821825e-06, + "loss": 0.001, + "step": 11850 + }, + { + "epoch": 3.55, + "learning_rate": 2.1465363187217495e-06, + "loss": 0.0012, + "step": 11860 + }, + { + "epoch": 3.56, + "learning_rate": 2.1439083359613165e-06, + "loss": 0.0021, + "step": 11870 + }, + { + "epoch": 3.56, + "learning_rate": 2.141280353200883e-06, + "loss": 0.0012, + "step": 11880 + }, + { + "epoch": 3.56, + "learning_rate": 2.13865237044045e-06, + "loss": 0.0014, + "step": 11890 + }, + { + "epoch": 3.57, + "learning_rate": 2.136024387680017e-06, + "loss": 0.0007, + "step": 11900 + }, + { + "epoch": 3.57, + "learning_rate": 2.1333964049195837e-06, + "loss": 0.0016, + "step": 11910 + }, + { + "epoch": 3.57, + "learning_rate": 2.1307684221591507e-06, + "loss": 0.0158, + "step": 11920 + }, + { + "epoch": 3.57, + "learning_rate": 2.1281404393987177e-06, + "loss": 0.0011, + "step": 11930 + }, + { + "epoch": 3.58, + "learning_rate": 2.1255124566382847e-06, + "loss": 0.0009, + "step": 11940 + }, + { + "epoch": 3.58, + "learning_rate": 2.1228844738778513e-06, + "loss": 0.0063, + "step": 11950 + }, + { + "epoch": 3.58, + "learning_rate": 2.1202564911174183e-06, + "loss": 0.0009, + "step": 11960 + }, + { + "epoch": 3.59, + "learning_rate": 2.1176285083569853e-06, + "loss": 0.0011, + "step": 11970 + }, + { + "epoch": 3.59, + "learning_rate": 2.1150005255965524e-06, + "loss": 0.0053, + "step": 11980 + }, + { + "epoch": 3.59, + "learning_rate": 2.112372542836119e-06, + "loss": 0.0019, + "step": 11990 + }, + { + "epoch": 3.59, + "learning_rate": 2.109744560075686e-06, + "loss": 0.001, + "step": 12000 + }, + { + "epoch": 3.59, + "eval_f1": 0.9951547260785085, + "eval_f1_0": 0.9984885632168076, + "eval_f1_1": 0.9985354333123745, + "eval_f1_2": 0.9845073845073845, + "eval_f1_3": 0.9990875232774674, + "eval_loss": 0.014154409058392048, + "eval_runtime": 153.1047, + "eval_samples_per_second": 77.516, + "eval_steps_per_second": 0.307, + "step": 12000 + }, + { + "epoch": 3.6, + "learning_rate": 2.107116577315253e-06, + "loss": 0.0035, + "step": 12010 + }, + { + "epoch": 3.6, + "learning_rate": 2.10448859455482e-06, + "loss": 0.0013, + "step": 12020 + }, + { + "epoch": 3.6, + "learning_rate": 2.101860611794387e-06, + "loss": 0.0014, + "step": 12030 + }, + { + "epoch": 3.61, + "learning_rate": 2.0992326290339536e-06, + "loss": 0.0015, + "step": 12040 + }, + { + "epoch": 3.61, + "learning_rate": 2.0966046462735206e-06, + "loss": 0.001, + "step": 12050 + }, + { + "epoch": 3.61, + "learning_rate": 2.0939766635130876e-06, + "loss": 0.0011, + "step": 12060 + }, + { + "epoch": 3.62, + "learning_rate": 2.0913486807526546e-06, + "loss": 0.0017, + "step": 12070 + }, + { + "epoch": 3.62, + "learning_rate": 2.0887206979922216e-06, + "loss": 0.0007, + "step": 12080 + }, + { + "epoch": 3.62, + "learning_rate": 2.086092715231788e-06, + "loss": 0.0005, + "step": 12090 + }, + { + "epoch": 3.62, + "learning_rate": 2.083464732471355e-06, + "loss": 0.0015, + "step": 12100 + }, + { + "epoch": 3.63, + "learning_rate": 2.0808367497109222e-06, + "loss": 0.002, + "step": 12110 + }, + { + "epoch": 3.63, + "learning_rate": 2.078208766950489e-06, + "loss": 0.0021, + "step": 12120 + }, + { + "epoch": 3.63, + "learning_rate": 2.075580784190056e-06, + "loss": 0.0148, + "step": 12130 + }, + { + "epoch": 3.64, + "learning_rate": 2.072952801429623e-06, + "loss": 0.0014, + "step": 12140 + }, + { + "epoch": 3.64, + "learning_rate": 2.0703248186691894e-06, + "loss": 0.0024, + "step": 12150 + }, + { + "epoch": 3.64, + "learning_rate": 2.0676968359087564e-06, + "loss": 0.0016, + "step": 12160 + }, + { + "epoch": 3.65, + "learning_rate": 2.0650688531483234e-06, + "loss": 0.0008, + "step": 12170 + }, + { + "epoch": 3.65, + "learning_rate": 2.0624408703878904e-06, + "loss": 0.0007, + "step": 12180 + }, + { + "epoch": 3.65, + "learning_rate": 2.0598128876274575e-06, + "loss": 0.0024, + "step": 12190 + }, + { + "epoch": 3.65, + "learning_rate": 2.057184904867024e-06, + "loss": 0.0046, + "step": 12200 + }, + { + "epoch": 3.65, + "eval_f1": 0.9947026226164652, + "eval_f1_0": 0.9983469571822656, + "eval_f1_1": 0.9985123555010257, + "eval_f1_2": 0.982838841736585, + "eval_f1_3": 0.9991123360459847, + "eval_loss": 0.01602787896990776, + "eval_runtime": 152.9981, + "eval_samples_per_second": 77.57, + "eval_steps_per_second": 0.307, + "step": 12200 + }, + { + "epoch": 3.66, + "learning_rate": 2.054556922106591e-06, + "loss": 0.0017, + "step": 12210 + }, + { + "epoch": 3.66, + "learning_rate": 2.051928939346158e-06, + "loss": 0.0024, + "step": 12220 + }, + { + "epoch": 3.66, + "learning_rate": 2.049300956585725e-06, + "loss": 0.0006, + "step": 12230 + }, + { + "epoch": 3.67, + "learning_rate": 2.046672973825292e-06, + "loss": 0.0019, + "step": 12240 + }, + { + "epoch": 3.67, + "learning_rate": 2.0440449910648587e-06, + "loss": 0.0192, + "step": 12250 + }, + { + "epoch": 3.67, + "learning_rate": 2.0414170083044257e-06, + "loss": 0.0194, + "step": 12260 + }, + { + "epoch": 3.68, + "learning_rate": 2.0387890255439927e-06, + "loss": 0.0143, + "step": 12270 + }, + { + "epoch": 3.68, + "learning_rate": 2.0361610427835597e-06, + "loss": 0.002, + "step": 12280 + }, + { + "epoch": 3.68, + "learning_rate": 2.0335330600231267e-06, + "loss": 0.0352, + "step": 12290 + }, + { + "epoch": 3.68, + "learning_rate": 2.0309050772626933e-06, + "loss": 0.0009, + "step": 12300 + }, + { + "epoch": 3.69, + "learning_rate": 2.0282770945022603e-06, + "loss": 0.0012, + "step": 12310 + }, + { + "epoch": 3.69, + "learning_rate": 2.0256491117418273e-06, + "loss": 0.0063, + "step": 12320 + }, + { + "epoch": 3.69, + "learning_rate": 2.023021128981394e-06, + "loss": 0.0013, + "step": 12330 + }, + { + "epoch": 3.7, + "learning_rate": 2.020393146220961e-06, + "loss": 0.0011, + "step": 12340 + }, + { + "epoch": 3.7, + "learning_rate": 2.017765163460528e-06, + "loss": 0.0009, + "step": 12350 + }, + { + "epoch": 3.7, + "learning_rate": 2.0151371807000945e-06, + "loss": 0.0014, + "step": 12360 + }, + { + "epoch": 3.71, + "learning_rate": 2.0125091979396615e-06, + "loss": 0.0082, + "step": 12370 + }, + { + "epoch": 3.71, + "learning_rate": 2.0098812151792285e-06, + "loss": 0.0012, + "step": 12380 + }, + { + "epoch": 3.71, + "learning_rate": 2.0072532324187955e-06, + "loss": 0.0027, + "step": 12390 + }, + { + "epoch": 3.71, + "learning_rate": 2.0046252496583625e-06, + "loss": 0.0009, + "step": 12400 + }, + { + "epoch": 3.71, + "eval_f1": 0.9938454734895156, + "eval_f1_0": 0.998058144492058, + "eval_f1_1": 0.9984700978907289, + "eval_f1_2": 0.9797009434999316, + "eval_f1_3": 0.9991527080753438, + "eval_loss": 0.013563607819378376, + "eval_runtime": 153.2157, + "eval_samples_per_second": 77.459, + "eval_steps_per_second": 0.307, + "step": 12400 + }, + { + "epoch": 3.72, + "learning_rate": 2.001997266897929e-06, + "loss": 0.0016, + "step": 12410 + }, + { + "epoch": 3.72, + "learning_rate": 1.999369284137496e-06, + "loss": 0.0025, + "step": 12420 + }, + { + "epoch": 3.72, + "learning_rate": 1.996741301377063e-06, + "loss": 0.0007, + "step": 12430 + }, + { + "epoch": 3.73, + "learning_rate": 1.99411331861663e-06, + "loss": 0.0009, + "step": 12440 + }, + { + "epoch": 3.73, + "learning_rate": 1.991485335856197e-06, + "loss": 0.0067, + "step": 12450 + }, + { + "epoch": 3.73, + "learning_rate": 1.9888573530957638e-06, + "loss": 0.0032, + "step": 12460 + }, + { + "epoch": 3.74, + "learning_rate": 1.9862293703353308e-06, + "loss": 0.0011, + "step": 12470 + }, + { + "epoch": 3.74, + "learning_rate": 1.9836013875748978e-06, + "loss": 0.0014, + "step": 12480 + }, + { + "epoch": 3.74, + "learning_rate": 1.980973404814465e-06, + "loss": 0.0016, + "step": 12490 + }, + { + "epoch": 3.74, + "learning_rate": 1.9783454220540314e-06, + "loss": 0.0009, + "step": 12500 + }, + { + "epoch": 3.75, + "learning_rate": 1.9757174392935984e-06, + "loss": 0.0009, + "step": 12510 + }, + { + "epoch": 3.75, + "learning_rate": 1.9730894565331654e-06, + "loss": 0.0021, + "step": 12520 + }, + { + "epoch": 3.75, + "learning_rate": 1.9704614737727324e-06, + "loss": 0.0012, + "step": 12530 + }, + { + "epoch": 3.76, + "learning_rate": 1.9678334910122994e-06, + "loss": 0.0232, + "step": 12540 + }, + { + "epoch": 3.76, + "learning_rate": 1.965205508251866e-06, + "loss": 0.0014, + "step": 12550 + }, + { + "epoch": 3.76, + "learning_rate": 1.962577525491433e-06, + "loss": 0.0013, + "step": 12560 + }, + { + "epoch": 3.77, + "learning_rate": 1.9599495427309996e-06, + "loss": 0.0026, + "step": 12570 + }, + { + "epoch": 3.77, + "learning_rate": 1.9573215599705666e-06, + "loss": 0.0011, + "step": 12580 + }, + { + "epoch": 3.77, + "learning_rate": 1.9546935772101336e-06, + "loss": 0.001, + "step": 12590 + }, + { + "epoch": 3.77, + "learning_rate": 1.9520655944497006e-06, + "loss": 0.0033, + "step": 12600 + }, + { + "epoch": 3.77, + "eval_f1": 0.9941136426031095, + "eval_f1_0": 0.9981520899343792, + "eval_f1_1": 0.998633683439975, + "eval_f1_2": 0.9805408368353736, + "eval_f1_3": 0.9991279602027104, + "eval_loss": 0.014793042093515396, + "eval_runtime": 153.1084, + "eval_samples_per_second": 77.514, + "eval_steps_per_second": 0.307, + "step": 12600 + }, + { + "epoch": 3.78, + "learning_rate": 1.9494376116892672e-06, + "loss": 0.0013, + "step": 12610 + }, + { + "epoch": 3.78, + "learning_rate": 1.9468096289288342e-06, + "loss": 0.0062, + "step": 12620 + }, + { + "epoch": 3.78, + "learning_rate": 1.9441816461684012e-06, + "loss": 0.001, + "step": 12630 + }, + { + "epoch": 3.79, + "learning_rate": 1.9415536634079682e-06, + "loss": 0.0013, + "step": 12640 + }, + { + "epoch": 3.79, + "learning_rate": 1.9389256806475353e-06, + "loss": 0.001, + "step": 12650 + }, + { + "epoch": 3.79, + "learning_rate": 1.936297697887102e-06, + "loss": 0.0013, + "step": 12660 + }, + { + "epoch": 3.8, + "learning_rate": 1.933669715126669e-06, + "loss": 0.0013, + "step": 12670 + }, + { + "epoch": 3.8, + "learning_rate": 1.931041732366236e-06, + "loss": 0.0009, + "step": 12680 + }, + { + "epoch": 3.8, + "learning_rate": 1.928413749605803e-06, + "loss": 0.0012, + "step": 12690 + }, + { + "epoch": 3.8, + "learning_rate": 1.92578576684537e-06, + "loss": 0.0016, + "step": 12700 + }, + { + "epoch": 3.81, + "learning_rate": 1.9231577840849365e-06, + "loss": 0.0013, + "step": 12710 + }, + { + "epoch": 3.81, + "learning_rate": 1.9205298013245035e-06, + "loss": 0.0018, + "step": 12720 + }, + { + "epoch": 3.81, + "learning_rate": 1.9179018185640705e-06, + "loss": 0.0036, + "step": 12730 + }, + { + "epoch": 3.82, + "learning_rate": 1.9152738358036375e-06, + "loss": 0.0012, + "step": 12740 + }, + { + "epoch": 3.82, + "learning_rate": 1.9126458530432045e-06, + "loss": 0.001, + "step": 12750 + }, + { + "epoch": 3.82, + "learning_rate": 1.910017870282771e-06, + "loss": 0.0017, + "step": 12760 + }, + { + "epoch": 3.83, + "learning_rate": 1.9073898875223377e-06, + "loss": 0.0011, + "step": 12770 + }, + { + "epoch": 3.83, + "learning_rate": 1.904761904761905e-06, + "loss": 0.0017, + "step": 12780 + }, + { + "epoch": 3.83, + "learning_rate": 1.9021339220014717e-06, + "loss": 0.005, + "step": 12790 + }, + { + "epoch": 3.83, + "learning_rate": 1.8995059392410387e-06, + "loss": 0.0008, + "step": 12800 + }, + { + "epoch": 3.83, + "eval_f1": 0.9919555603950767, + "eval_f1_0": 0.9974524093844568, + "eval_f1_1": 0.9984359006796931, + "eval_f1_2": 0.9728773804426739, + "eval_f1_3": 0.9990565510734835, + "eval_loss": 0.01904941350221634, + "eval_runtime": 153.2291, + "eval_samples_per_second": 77.453, + "eval_steps_per_second": 0.307, + "step": 12800 + }, + { + "epoch": 3.84, + "learning_rate": 1.8968779564806055e-06, + "loss": 0.0015, + "step": 12810 + }, + { + "epoch": 3.84, + "learning_rate": 1.8942499737201725e-06, + "loss": 0.0008, + "step": 12820 + }, + { + "epoch": 3.84, + "learning_rate": 1.8916219909597393e-06, + "loss": 0.0012, + "step": 12830 + }, + { + "epoch": 3.85, + "learning_rate": 1.8889940081993063e-06, + "loss": 0.002, + "step": 12840 + }, + { + "epoch": 3.85, + "learning_rate": 1.8863660254388733e-06, + "loss": 0.001, + "step": 12850 + }, + { + "epoch": 3.85, + "learning_rate": 1.8837380426784401e-06, + "loss": 0.0008, + "step": 12860 + }, + { + "epoch": 3.86, + "learning_rate": 1.8811100599180072e-06, + "loss": 0.0014, + "step": 12870 + }, + { + "epoch": 3.86, + "learning_rate": 1.878482077157574e-06, + "loss": 0.0023, + "step": 12880 + }, + { + "epoch": 3.86, + "learning_rate": 1.875854094397141e-06, + "loss": 0.001, + "step": 12890 + }, + { + "epoch": 3.86, + "learning_rate": 1.873226111636708e-06, + "loss": 0.0041, + "step": 12900 + }, + { + "epoch": 3.87, + "learning_rate": 1.8705981288762748e-06, + "loss": 0.0006, + "step": 12910 + }, + { + "epoch": 3.87, + "learning_rate": 1.8679701461158418e-06, + "loss": 0.0024, + "step": 12920 + }, + { + "epoch": 3.87, + "learning_rate": 1.8653421633554086e-06, + "loss": 0.0061, + "step": 12930 + }, + { + "epoch": 3.88, + "learning_rate": 1.8627141805949756e-06, + "loss": 0.0009, + "step": 12940 + }, + { + "epoch": 3.88, + "learning_rate": 1.8600861978345424e-06, + "loss": 0.0292, + "step": 12950 + }, + { + "epoch": 3.88, + "learning_rate": 1.8574582150741094e-06, + "loss": 0.0009, + "step": 12960 + }, + { + "epoch": 3.89, + "learning_rate": 1.854830232313676e-06, + "loss": 0.0015, + "step": 12970 + }, + { + "epoch": 3.89, + "learning_rate": 1.852202249553243e-06, + "loss": 0.0015, + "step": 12980 + }, + { + "epoch": 3.89, + "learning_rate": 1.8495742667928098e-06, + "loss": 0.0031, + "step": 12990 + }, + { + "epoch": 3.89, + "learning_rate": 1.8469462840323768e-06, + "loss": 0.0013, + "step": 13000 + }, + { + "epoch": 3.89, + "eval_f1": 0.993328578780528, + "eval_f1_0": 0.9979199988261016, + "eval_f1_1": 0.9979012876195154, + "eval_f1_2": 0.9783837493533191, + "eval_f1_3": 0.9991092793231764, + "eval_loss": 0.021434780210256577, + "eval_runtime": 153.1857, + "eval_samples_per_second": 77.475, + "eval_steps_per_second": 0.307, + "step": 13000 + }, + { + "epoch": 3.9, + "learning_rate": 1.8443183012719438e-06, + "loss": 0.0198, + "step": 13010 + }, + { + "epoch": 3.9, + "learning_rate": 1.8416903185115106e-06, + "loss": 0.0005, + "step": 13020 + }, + { + "epoch": 3.9, + "learning_rate": 1.8390623357510776e-06, + "loss": 0.0245, + "step": 13030 + }, + { + "epoch": 3.91, + "learning_rate": 1.8364343529906444e-06, + "loss": 0.0008, + "step": 13040 + }, + { + "epoch": 3.91, + "learning_rate": 1.8338063702302114e-06, + "loss": 0.0013, + "step": 13050 + }, + { + "epoch": 3.91, + "learning_rate": 1.8311783874697782e-06, + "loss": 0.0006, + "step": 13060 + }, + { + "epoch": 3.92, + "learning_rate": 1.8285504047093452e-06, + "loss": 0.0016, + "step": 13070 + }, + { + "epoch": 3.92, + "learning_rate": 1.8259224219489122e-06, + "loss": 0.0014, + "step": 13080 + }, + { + "epoch": 3.92, + "learning_rate": 1.823294439188479e-06, + "loss": 0.0021, + "step": 13090 + }, + { + "epoch": 3.92, + "learning_rate": 1.820666456428046e-06, + "loss": 0.0012, + "step": 13100 + }, + { + "epoch": 3.93, + "learning_rate": 1.8180384736676129e-06, + "loss": 0.0046, + "step": 13110 + }, + { + "epoch": 3.93, + "learning_rate": 1.8154104909071799e-06, + "loss": 0.0021, + "step": 13120 + }, + { + "epoch": 3.93, + "learning_rate": 1.8127825081467469e-06, + "loss": 0.0326, + "step": 13130 + }, + { + "epoch": 3.94, + "learning_rate": 1.8101545253863137e-06, + "loss": 0.0019, + "step": 13140 + }, + { + "epoch": 3.94, + "learning_rate": 1.8075265426258807e-06, + "loss": 0.0013, + "step": 13150 + }, + { + "epoch": 3.94, + "learning_rate": 1.8048985598654475e-06, + "loss": 0.0024, + "step": 13160 + }, + { + "epoch": 3.95, + "learning_rate": 1.8022705771050145e-06, + "loss": 0.0014, + "step": 13170 + }, + { + "epoch": 3.95, + "learning_rate": 1.7996425943445813e-06, + "loss": 0.0111, + "step": 13180 + }, + { + "epoch": 3.95, + "learning_rate": 1.797014611584148e-06, + "loss": 0.0015, + "step": 13190 + }, + { + "epoch": 3.95, + "learning_rate": 1.7943866288237149e-06, + "loss": 0.0012, + "step": 13200 + }, + { + "epoch": 3.95, + "eval_f1": 0.9831277397299376, + "eval_f1_0": 0.9944236829249278, + "eval_f1_1": 0.9980741417078757, + "eval_f1_2": 0.9409410688411816, + "eval_f1_3": 0.9990720654457654, + "eval_loss": 0.045238446444272995, + "eval_runtime": 153.1564, + "eval_samples_per_second": 77.489, + "eval_steps_per_second": 0.307, + "step": 13200 + }, + { + "epoch": 3.96, + "learning_rate": 1.791758646063282e-06, + "loss": 0.0021, + "step": 13210 + }, + { + "epoch": 3.96, + "learning_rate": 1.7891306633028487e-06, + "loss": 0.0011, + "step": 13220 + }, + { + "epoch": 3.96, + "learning_rate": 1.7865026805424157e-06, + "loss": 0.001, + "step": 13230 + }, + { + "epoch": 3.97, + "learning_rate": 1.7838746977819827e-06, + "loss": 0.001, + "step": 13240 + }, + { + "epoch": 3.97, + "learning_rate": 1.7812467150215495e-06, + "loss": 0.001, + "step": 13250 + }, + { + "epoch": 3.97, + "learning_rate": 1.7786187322611165e-06, + "loss": 0.0146, + "step": 13260 + }, + { + "epoch": 3.98, + "learning_rate": 1.7759907495006833e-06, + "loss": 0.0006, + "step": 13270 + }, + { + "epoch": 3.98, + "learning_rate": 1.7733627667402503e-06, + "loss": 0.0009, + "step": 13280 + }, + { + "epoch": 3.98, + "learning_rate": 1.7707347839798173e-06, + "loss": 0.0034, + "step": 13290 + }, + { + "epoch": 3.98, + "learning_rate": 1.7681068012193841e-06, + "loss": 0.0012, + "step": 13300 + }, + { + "epoch": 3.99, + "learning_rate": 1.7654788184589511e-06, + "loss": 0.0018, + "step": 13310 + }, + { + "epoch": 3.99, + "learning_rate": 1.762850835698518e-06, + "loss": 0.0009, + "step": 13320 + }, + { + "epoch": 3.99, + "learning_rate": 1.760222852938085e-06, + "loss": 0.0011, + "step": 13330 + }, + { + "epoch": 4.0, + "learning_rate": 1.7575948701776518e-06, + "loss": 0.0022, + "step": 13340 + }, + { + "epoch": 4.0, + "learning_rate": 1.7549668874172188e-06, + "loss": 0.0008, + "step": 13350 + }, + { + "epoch": 4.0, + "learning_rate": 1.7523389046567858e-06, + "loss": 0.0058, + "step": 13360 + }, + { + "epoch": 4.01, + "learning_rate": 1.7497109218963526e-06, + "loss": 0.0009, + "step": 13370 + }, + { + "epoch": 4.01, + "learning_rate": 1.7470829391359196e-06, + "loss": 0.0013, + "step": 13380 + }, + { + "epoch": 4.01, + "learning_rate": 1.7444549563754864e-06, + "loss": 0.0013, + "step": 13390 + }, + { + "epoch": 4.01, + "learning_rate": 1.7418269736150532e-06, + "loss": 0.0006, + "step": 13400 + }, + { + "epoch": 4.01, + "eval_f1": 0.9917532185890476, + "eval_f1_0": 0.9973856942651312, + "eval_f1_1": 0.9988251145657044, + "eval_f1_2": 0.9716896523443804, + "eval_f1_3": 0.9991124131809747, + "eval_loss": 0.02197587490081787, + "eval_runtime": 153.1159, + "eval_samples_per_second": 77.51, + "eval_steps_per_second": 0.307, + "step": 13400 + }, + { + "epoch": 4.02, + "learning_rate": 1.73919899085462e-06, + "loss": 0.0007, + "step": 13410 + }, + { + "epoch": 4.02, + "learning_rate": 1.736571008094187e-06, + "loss": 0.0013, + "step": 13420 + }, + { + "epoch": 4.02, + "learning_rate": 1.7339430253337538e-06, + "loss": 0.0015, + "step": 13430 + }, + { + "epoch": 4.03, + "learning_rate": 1.7313150425733208e-06, + "loss": 0.001, + "step": 13440 + }, + { + "epoch": 4.03, + "learning_rate": 1.7286870598128878e-06, + "loss": 0.0006, + "step": 13450 + }, + { + "epoch": 4.03, + "learning_rate": 1.7260590770524546e-06, + "loss": 0.0008, + "step": 13460 + }, + { + "epoch": 4.04, + "learning_rate": 1.7234310942920216e-06, + "loss": 0.0009, + "step": 13470 + }, + { + "epoch": 4.04, + "learning_rate": 1.7208031115315884e-06, + "loss": 0.0031, + "step": 13480 + }, + { + "epoch": 4.04, + "learning_rate": 1.7181751287711554e-06, + "loss": 0.0007, + "step": 13490 + }, + { + "epoch": 4.04, + "learning_rate": 1.7155471460107222e-06, + "loss": 0.0014, + "step": 13500 + }, + { + "epoch": 4.05, + "learning_rate": 1.7129191632502892e-06, + "loss": 0.0008, + "step": 13510 + }, + { + "epoch": 4.05, + "learning_rate": 1.7102911804898562e-06, + "loss": 0.0011, + "step": 13520 + }, + { + "epoch": 4.05, + "learning_rate": 1.707663197729423e-06, + "loss": 0.0008, + "step": 13530 + }, + { + "epoch": 4.06, + "learning_rate": 1.70503521496899e-06, + "loss": 0.0113, + "step": 13540 + }, + { + "epoch": 4.06, + "learning_rate": 1.7024072322085568e-06, + "loss": 0.0024, + "step": 13550 + }, + { + "epoch": 4.06, + "learning_rate": 1.6997792494481239e-06, + "loss": 0.0042, + "step": 13560 + }, + { + "epoch": 4.07, + "learning_rate": 1.6971512666876907e-06, + "loss": 0.001, + "step": 13570 + }, + { + "epoch": 4.07, + "learning_rate": 1.6945232839272577e-06, + "loss": 0.0012, + "step": 13580 + }, + { + "epoch": 4.07, + "learning_rate": 1.6918953011668247e-06, + "loss": 0.0012, + "step": 13590 + }, + { + "epoch": 4.07, + "learning_rate": 1.6892673184063915e-06, + "loss": 0.0011, + "step": 13600 + }, + { + "epoch": 4.07, + "eval_f1": 0.9945764341532917, + "eval_f1_0": 0.9983035493624414, + "eval_f1_1": 0.9986811939794972, + "eval_f1_2": 0.982208602127454, + "eval_f1_3": 0.999112391143774, + "eval_loss": 0.013722382485866547, + "eval_runtime": 153.2875, + "eval_samples_per_second": 77.423, + "eval_steps_per_second": 0.307, + "step": 13600 + }, + { + "epoch": 4.08, + "learning_rate": 1.686639335645958e-06, + "loss": 0.0021, + "step": 13610 + }, + { + "epoch": 4.08, + "learning_rate": 1.684011352885525e-06, + "loss": 0.0009, + "step": 13620 + }, + { + "epoch": 4.08, + "learning_rate": 1.681383370125092e-06, + "loss": 0.0019, + "step": 13630 + }, + { + "epoch": 4.09, + "learning_rate": 1.6787553873646589e-06, + "loss": 0.0009, + "step": 13640 + }, + { + "epoch": 4.09, + "learning_rate": 1.6761274046042259e-06, + "loss": 0.0013, + "step": 13650 + }, + { + "epoch": 4.09, + "learning_rate": 1.6734994218437927e-06, + "loss": 0.0009, + "step": 13660 + }, + { + "epoch": 4.1, + "learning_rate": 1.6708714390833597e-06, + "loss": 0.0011, + "step": 13670 + }, + { + "epoch": 4.1, + "learning_rate": 1.6682434563229267e-06, + "loss": 0.0012, + "step": 13680 + }, + { + "epoch": 4.1, + "learning_rate": 1.6656154735624935e-06, + "loss": 0.0128, + "step": 13690 + }, + { + "epoch": 4.1, + "learning_rate": 1.6629874908020605e-06, + "loss": 0.001, + "step": 13700 + }, + { + "epoch": 4.11, + "learning_rate": 1.6603595080416273e-06, + "loss": 0.0005, + "step": 13710 + }, + { + "epoch": 4.11, + "learning_rate": 1.6577315252811943e-06, + "loss": 0.0011, + "step": 13720 + }, + { + "epoch": 4.11, + "learning_rate": 1.6551035425207611e-06, + "loss": 0.0018, + "step": 13730 + }, + { + "epoch": 4.12, + "learning_rate": 1.6524755597603281e-06, + "loss": 0.0009, + "step": 13740 + }, + { + "epoch": 4.12, + "learning_rate": 1.6498475769998951e-06, + "loss": 0.0007, + "step": 13750 + }, + { + "epoch": 4.12, + "learning_rate": 1.647219594239462e-06, + "loss": 0.0013, + "step": 13760 + }, + { + "epoch": 4.13, + "learning_rate": 1.644591611479029e-06, + "loss": 0.0011, + "step": 13770 + }, + { + "epoch": 4.13, + "learning_rate": 1.6419636287185958e-06, + "loss": 0.001, + "step": 13780 + }, + { + "epoch": 4.13, + "learning_rate": 1.6393356459581628e-06, + "loss": 0.0009, + "step": 13790 + }, + { + "epoch": 4.13, + "learning_rate": 1.6367076631977298e-06, + "loss": 0.0009, + "step": 13800 + }, + { + "epoch": 4.13, + "eval_f1": 0.9950386146416905, + "eval_f1_0": 0.9984622585356817, + "eval_f1_1": 0.9987485410825407, + "eval_f1_2": 0.9838281999648616, + "eval_f1_3": 0.9991154589836779, + "eval_loss": 0.013640536926686764, + "eval_runtime": 153.1601, + "eval_samples_per_second": 77.488, + "eval_steps_per_second": 0.307, + "step": 13800 + }, + { + "epoch": 4.14, + "learning_rate": 1.6340796804372966e-06, + "loss": 0.0009, + "step": 13810 + }, + { + "epoch": 4.14, + "learning_rate": 1.6314516976768636e-06, + "loss": 0.0015, + "step": 13820 + }, + { + "epoch": 4.14, + "learning_rate": 1.6288237149164302e-06, + "loss": 0.0013, + "step": 13830 + }, + { + "epoch": 4.15, + "learning_rate": 1.6261957321559972e-06, + "loss": 0.0011, + "step": 13840 + }, + { + "epoch": 4.15, + "learning_rate": 1.623567749395564e-06, + "loss": 0.0011, + "step": 13850 + }, + { + "epoch": 4.15, + "learning_rate": 1.620939766635131e-06, + "loss": 0.0017, + "step": 13860 + }, + { + "epoch": 4.16, + "learning_rate": 1.6183117838746978e-06, + "loss": 0.0007, + "step": 13870 + }, + { + "epoch": 4.16, + "learning_rate": 1.6156838011142648e-06, + "loss": 0.001, + "step": 13880 + }, + { + "epoch": 4.16, + "learning_rate": 1.6130558183538316e-06, + "loss": 0.0014, + "step": 13890 + }, + { + "epoch": 4.16, + "learning_rate": 1.6104278355933986e-06, + "loss": 0.0014, + "step": 13900 + }, + { + "epoch": 4.17, + "learning_rate": 1.6077998528329656e-06, + "loss": 0.0013, + "step": 13910 + }, + { + "epoch": 4.17, + "learning_rate": 1.6051718700725324e-06, + "loss": 0.0007, + "step": 13920 + }, + { + "epoch": 4.17, + "learning_rate": 1.6025438873120994e-06, + "loss": 0.0008, + "step": 13930 + }, + { + "epoch": 4.18, + "learning_rate": 1.5999159045516662e-06, + "loss": 0.0008, + "step": 13940 + }, + { + "epoch": 4.18, + "learning_rate": 1.5972879217912332e-06, + "loss": 0.0011, + "step": 13950 + }, + { + "epoch": 4.18, + "learning_rate": 1.5946599390308e-06, + "loss": 0.0032, + "step": 13960 + }, + { + "epoch": 4.19, + "learning_rate": 1.592031956270367e-06, + "loss": 0.0008, + "step": 13970 + }, + { + "epoch": 4.19, + "learning_rate": 1.589403973509934e-06, + "loss": 0.0016, + "step": 13980 + }, + { + "epoch": 4.19, + "learning_rate": 1.5867759907495008e-06, + "loss": 0.0013, + "step": 13990 + }, + { + "epoch": 4.19, + "learning_rate": 1.5841480079890679e-06, + "loss": 0.0011, + "step": 14000 + }, + { + "epoch": 4.19, + "eval_f1": 0.9923091616893841, + "eval_f1_0": 0.9975497273387712, + "eval_f1_1": 0.9982912932666216, + "eval_f1_2": 0.9742708208984919, + "eval_f1_3": 0.9991248052536513, + "eval_loss": 0.023798752576112747, + "eval_runtime": 153.0593, + "eval_samples_per_second": 77.539, + "eval_steps_per_second": 0.307, + "step": 14000 + }, + { + "epoch": 4.2, + "learning_rate": 1.5815200252286347e-06, + "loss": 0.0035, + "step": 14010 + }, + { + "epoch": 4.2, + "learning_rate": 1.5788920424682017e-06, + "loss": 0.0067, + "step": 14020 + }, + { + "epoch": 4.2, + "learning_rate": 1.5762640597077687e-06, + "loss": 0.0013, + "step": 14030 + }, + { + "epoch": 4.21, + "learning_rate": 1.5736360769473353e-06, + "loss": 0.0017, + "step": 14040 + }, + { + "epoch": 4.21, + "learning_rate": 1.571008094186902e-06, + "loss": 0.0012, + "step": 14050 + }, + { + "epoch": 4.21, + "learning_rate": 1.568380111426469e-06, + "loss": 0.0004, + "step": 14060 + }, + { + "epoch": 4.22, + "learning_rate": 1.565752128666036e-06, + "loss": 0.0011, + "step": 14070 + }, + { + "epoch": 4.22, + "learning_rate": 1.5631241459056029e-06, + "loss": 0.0017, + "step": 14080 + }, + { + "epoch": 4.22, + "learning_rate": 1.5604961631451699e-06, + "loss": 0.0107, + "step": 14090 + }, + { + "epoch": 4.22, + "learning_rate": 1.5578681803847367e-06, + "loss": 0.0014, + "step": 14100 + }, + { + "epoch": 4.23, + "learning_rate": 1.5552401976243037e-06, + "loss": 0.0012, + "step": 14110 + }, + { + "epoch": 4.23, + "learning_rate": 1.5526122148638705e-06, + "loss": 0.0009, + "step": 14120 + }, + { + "epoch": 4.23, + "learning_rate": 1.5499842321034375e-06, + "loss": 0.0014, + "step": 14130 + }, + { + "epoch": 4.24, + "learning_rate": 1.5473562493430045e-06, + "loss": 0.0011, + "step": 14140 + }, + { + "epoch": 4.24, + "learning_rate": 1.5447282665825713e-06, + "loss": 0.0006, + "step": 14150 + }, + { + "epoch": 4.24, + "learning_rate": 1.5421002838221383e-06, + "loss": 0.0009, + "step": 14160 + }, + { + "epoch": 4.25, + "learning_rate": 1.5394723010617051e-06, + "loss": 0.0009, + "step": 14170 + }, + { + "epoch": 4.25, + "learning_rate": 1.5368443183012721e-06, + "loss": 0.0084, + "step": 14180 + }, + { + "epoch": 4.25, + "learning_rate": 1.5342163355408391e-06, + "loss": 0.0013, + "step": 14190 + }, + { + "epoch": 4.25, + "learning_rate": 1.531588352780406e-06, + "loss": 0.001, + "step": 14200 + }, + { + "epoch": 4.25, + "eval_f1": 0.995143494905595, + "eval_f1_0": 0.998492315814845, + "eval_f1_1": 0.9987099940004178, + "eval_f1_2": 0.9842623517839965, + "eval_f1_3": 0.9991093180231205, + "eval_loss": 0.0133112333714962, + "eval_runtime": 153.1682, + "eval_samples_per_second": 77.483, + "eval_steps_per_second": 0.307, + "step": 14200 + }, + { + "epoch": 4.26, + "learning_rate": 1.528960370019973e-06, + "loss": 0.0019, + "step": 14210 + }, + { + "epoch": 4.26, + "learning_rate": 1.5263323872595397e-06, + "loss": 0.0013, + "step": 14220 + }, + { + "epoch": 4.26, + "learning_rate": 1.5237044044991068e-06, + "loss": 0.0019, + "step": 14230 + }, + { + "epoch": 4.27, + "learning_rate": 1.5210764217386736e-06, + "loss": 0.0014, + "step": 14240 + }, + { + "epoch": 4.27, + "learning_rate": 1.5184484389782404e-06, + "loss": 0.0014, + "step": 14250 + }, + { + "epoch": 4.27, + "learning_rate": 1.5158204562178072e-06, + "loss": 0.0009, + "step": 14260 + }, + { + "epoch": 4.28, + "learning_rate": 1.5131924734573742e-06, + "loss": 0.0009, + "step": 14270 + }, + { + "epoch": 4.28, + "learning_rate": 1.510564490696941e-06, + "loss": 0.0211, + "step": 14280 + }, + { + "epoch": 4.28, + "learning_rate": 1.507936507936508e-06, + "loss": 0.002, + "step": 14290 + }, + { + "epoch": 4.28, + "learning_rate": 1.505308525176075e-06, + "loss": 0.0013, + "step": 14300 + }, + { + "epoch": 4.29, + "learning_rate": 1.5026805424156418e-06, + "loss": 0.0009, + "step": 14310 + }, + { + "epoch": 4.29, + "learning_rate": 1.5000525596552088e-06, + "loss": 0.0013, + "step": 14320 + }, + { + "epoch": 4.29, + "learning_rate": 1.4974245768947756e-06, + "loss": 0.0092, + "step": 14330 + }, + { + "epoch": 4.3, + "learning_rate": 1.4947965941343426e-06, + "loss": 0.0009, + "step": 14340 + }, + { + "epoch": 4.3, + "learning_rate": 1.4921686113739094e-06, + "loss": 0.0019, + "step": 14350 + }, + { + "epoch": 4.3, + "learning_rate": 1.4895406286134764e-06, + "loss": 0.0051, + "step": 14360 + }, + { + "epoch": 4.3, + "learning_rate": 1.4869126458530434e-06, + "loss": 0.0014, + "step": 14370 + }, + { + "epoch": 4.31, + "learning_rate": 1.4842846630926102e-06, + "loss": 0.002, + "step": 14380 + }, + { + "epoch": 4.31, + "learning_rate": 1.4816566803321772e-06, + "loss": 0.0009, + "step": 14390 + }, + { + "epoch": 4.31, + "learning_rate": 1.479028697571744e-06, + "loss": 0.0007, + "step": 14400 + }, + { + "epoch": 4.31, + "eval_f1": 0.9958044718258009, + "eval_f1_0": 0.9986962178379164, + "eval_f1_1": 0.99874291452202, + "eval_f1_2": 0.9866477372408563, + "eval_f1_3": 0.9991310177024108, + "eval_loss": 0.011024106293916702, + "eval_runtime": 153.2376, + "eval_samples_per_second": 77.448, + "eval_steps_per_second": 0.307, + "step": 14400 + }, + { + "epoch": 4.32, + "learning_rate": 1.476400714811311e-06, + "loss": 0.0009, + "step": 14410 + }, + { + "epoch": 4.32, + "learning_rate": 1.473772732050878e-06, + "loss": 0.0044, + "step": 14420 + }, + { + "epoch": 4.32, + "learning_rate": 1.4711447492904448e-06, + "loss": 0.0007, + "step": 14430 + }, + { + "epoch": 4.33, + "learning_rate": 1.4685167665300119e-06, + "loss": 0.001, + "step": 14440 + }, + { + "epoch": 4.33, + "learning_rate": 1.4658887837695787e-06, + "loss": 0.0008, + "step": 14450 + }, + { + "epoch": 4.33, + "learning_rate": 1.4632608010091457e-06, + "loss": 0.0007, + "step": 14460 + }, + { + "epoch": 4.33, + "learning_rate": 1.4606328182487122e-06, + "loss": 0.0017, + "step": 14470 + }, + { + "epoch": 4.34, + "learning_rate": 1.4580048354882793e-06, + "loss": 0.0021, + "step": 14480 + }, + { + "epoch": 4.34, + "learning_rate": 1.455376852727846e-06, + "loss": 0.0266, + "step": 14490 + }, + { + "epoch": 4.34, + "learning_rate": 1.452748869967413e-06, + "loss": 0.0137, + "step": 14500 + }, + { + "epoch": 4.35, + "learning_rate": 1.4501208872069799e-06, + "loss": 0.0016, + "step": 14510 + }, + { + "epoch": 4.35, + "learning_rate": 1.4474929044465469e-06, + "loss": 0.0016, + "step": 14520 + }, + { + "epoch": 4.35, + "learning_rate": 1.4448649216861139e-06, + "loss": 0.0046, + "step": 14530 + }, + { + "epoch": 4.36, + "learning_rate": 1.4422369389256807e-06, + "loss": 0.0009, + "step": 14540 + }, + { + "epoch": 4.36, + "learning_rate": 1.4396089561652477e-06, + "loss": 0.0006, + "step": 14550 + }, + { + "epoch": 4.36, + "learning_rate": 1.4369809734048145e-06, + "loss": 0.0011, + "step": 14560 + }, + { + "epoch": 4.36, + "learning_rate": 1.4343529906443815e-06, + "loss": 0.001, + "step": 14570 + }, + { + "epoch": 4.37, + "learning_rate": 1.4317250078839485e-06, + "loss": 0.0014, + "step": 14580 + }, + { + "epoch": 4.37, + "learning_rate": 1.4290970251235153e-06, + "loss": 0.0009, + "step": 14590 + }, + { + "epoch": 4.37, + "learning_rate": 1.4264690423630823e-06, + "loss": 0.0007, + "step": 14600 + }, + { + "epoch": 4.37, + "eval_f1": 0.9958071934678491, + "eval_f1_0": 0.998687253085116, + "eval_f1_1": 0.99839320159376, + "eval_f1_2": 0.9870297264647511, + "eval_f1_3": 0.9991185927277693, + "eval_loss": 0.012305574491620064, + "eval_runtime": 153.2375, + "eval_samples_per_second": 77.448, + "eval_steps_per_second": 0.307, + "step": 14600 + }, + { + "epoch": 4.38, + "learning_rate": 1.4238410596026491e-06, + "loss": 0.0022, + "step": 14610 + }, + { + "epoch": 4.38, + "learning_rate": 1.4212130768422161e-06, + "loss": 0.0007, + "step": 14620 + }, + { + "epoch": 4.38, + "learning_rate": 1.418585094081783e-06, + "loss": 0.0011, + "step": 14630 + }, + { + "epoch": 4.39, + "learning_rate": 1.41595711132135e-06, + "loss": 0.0009, + "step": 14640 + }, + { + "epoch": 4.39, + "learning_rate": 1.413329128560917e-06, + "loss": 0.0046, + "step": 14650 + }, + { + "epoch": 4.39, + "learning_rate": 1.4107011458004837e-06, + "loss": 0.001, + "step": 14660 + }, + { + "epoch": 4.39, + "learning_rate": 1.4080731630400508e-06, + "loss": 0.0008, + "step": 14670 + }, + { + "epoch": 4.4, + "learning_rate": 1.4054451802796173e-06, + "loss": 0.0037, + "step": 14680 + }, + { + "epoch": 4.4, + "learning_rate": 1.4028171975191844e-06, + "loss": 0.0026, + "step": 14690 + }, + { + "epoch": 4.4, + "learning_rate": 1.4001892147587512e-06, + "loss": 0.0008, + "step": 14700 + }, + { + "epoch": 4.41, + "learning_rate": 1.3975612319983182e-06, + "loss": 0.0011, + "step": 14710 + }, + { + "epoch": 4.41, + "learning_rate": 1.394933249237885e-06, + "loss": 0.0019, + "step": 14720 + }, + { + "epoch": 4.41, + "learning_rate": 1.392305266477452e-06, + "loss": 0.0012, + "step": 14730 + }, + { + "epoch": 4.42, + "learning_rate": 1.389677283717019e-06, + "loss": 0.002, + "step": 14740 + }, + { + "epoch": 4.42, + "learning_rate": 1.3870493009565858e-06, + "loss": 0.0008, + "step": 14750 + }, + { + "epoch": 4.42, + "learning_rate": 1.3844213181961528e-06, + "loss": 0.0012, + "step": 14760 + }, + { + "epoch": 4.42, + "learning_rate": 1.3817933354357196e-06, + "loss": 0.0005, + "step": 14770 + }, + { + "epoch": 4.43, + "learning_rate": 1.3791653526752866e-06, + "loss": 0.0007, + "step": 14780 + }, + { + "epoch": 4.43, + "learning_rate": 1.3765373699148534e-06, + "loss": 0.0005, + "step": 14790 + }, + { + "epoch": 4.43, + "learning_rate": 1.3739093871544204e-06, + "loss": 0.001, + "step": 14800 + }, + { + "epoch": 4.43, + "eval_f1": 0.9956210234698608, + "eval_f1_0": 0.9986354102207614, + "eval_f1_1": 0.9986101872317422, + "eval_f1_2": 0.9861167592664909, + "eval_f1_3": 0.999121737160449, + "eval_loss": 0.012135359458625317, + "eval_runtime": 153.1317, + "eval_samples_per_second": 77.502, + "eval_steps_per_second": 0.307, + "step": 14800 + }, + { + "epoch": 4.44, + "learning_rate": 1.3712814043939874e-06, + "loss": 0.0017, + "step": 14810 + }, + { + "epoch": 4.44, + "learning_rate": 1.3686534216335542e-06, + "loss": 0.0044, + "step": 14820 + }, + { + "epoch": 4.44, + "learning_rate": 1.3660254388731212e-06, + "loss": 0.0014, + "step": 14830 + }, + { + "epoch": 4.45, + "learning_rate": 1.363397456112688e-06, + "loss": 0.0016, + "step": 14840 + }, + { + "epoch": 4.45, + "learning_rate": 1.360769473352255e-06, + "loss": 0.014, + "step": 14850 + }, + { + "epoch": 4.45, + "learning_rate": 1.3581414905918218e-06, + "loss": 0.0025, + "step": 14860 + }, + { + "epoch": 4.45, + "learning_rate": 1.3555135078313888e-06, + "loss": 0.0012, + "step": 14870 + }, + { + "epoch": 4.46, + "learning_rate": 1.3528855250709559e-06, + "loss": 0.0008, + "step": 14880 + }, + { + "epoch": 4.46, + "learning_rate": 1.3502575423105224e-06, + "loss": 0.001, + "step": 14890 + }, + { + "epoch": 4.46, + "learning_rate": 1.3476295595500892e-06, + "loss": 0.0009, + "step": 14900 + }, + { + "epoch": 4.47, + "learning_rate": 1.3450015767896562e-06, + "loss": 0.0012, + "step": 14910 + }, + { + "epoch": 4.47, + "learning_rate": 1.3423735940292233e-06, + "loss": 0.0008, + "step": 14920 + }, + { + "epoch": 4.47, + "learning_rate": 1.33974561126879e-06, + "loss": 0.001, + "step": 14930 + }, + { + "epoch": 4.48, + "learning_rate": 1.337117628508357e-06, + "loss": 0.0012, + "step": 14940 + }, + { + "epoch": 4.48, + "learning_rate": 1.3344896457479239e-06, + "loss": 0.0008, + "step": 14950 + }, + { + "epoch": 4.48, + "learning_rate": 1.3318616629874909e-06, + "loss": 0.0014, + "step": 14960 + }, + { + "epoch": 4.48, + "learning_rate": 1.3292336802270579e-06, + "loss": 0.0007, + "step": 14970 + }, + { + "epoch": 4.49, + "learning_rate": 1.3266056974666247e-06, + "loss": 0.0012, + "step": 14980 + }, + { + "epoch": 4.49, + "learning_rate": 1.3239777147061917e-06, + "loss": 0.0014, + "step": 14990 + }, + { + "epoch": 4.49, + "learning_rate": 1.3213497319457585e-06, + "loss": 0.0008, + "step": 15000 + }, + { + "epoch": 4.49, + "eval_f1": 0.9952924326390867, + "eval_f1_0": 0.9985425928445921, + "eval_f1_1": 0.998687057840373, + "eval_f1_2": 0.9848215254421179, + "eval_f1_3": 0.999118554429264, + "eval_loss": 0.013778653927147388, + "eval_runtime": 153.1609, + "eval_samples_per_second": 77.487, + "eval_steps_per_second": 0.307, + "step": 15000 + }, + { + "epoch": 4.5, + "learning_rate": 1.3187217491853255e-06, + "loss": 0.0008, + "step": 15010 + }, + { + "epoch": 4.5, + "learning_rate": 1.3160937664248923e-06, + "loss": 0.0012, + "step": 15020 + }, + { + "epoch": 4.5, + "learning_rate": 1.3134657836644593e-06, + "loss": 0.001, + "step": 15030 + }, + { + "epoch": 4.51, + "learning_rate": 1.3108378009040263e-06, + "loss": 0.0006, + "step": 15040 + }, + { + "epoch": 4.51, + "learning_rate": 1.3082098181435931e-06, + "loss": 0.007, + "step": 15050 + }, + { + "epoch": 4.51, + "learning_rate": 1.3055818353831601e-06, + "loss": 0.001, + "step": 15060 + }, + { + "epoch": 4.51, + "learning_rate": 1.302953852622727e-06, + "loss": 0.0013, + "step": 15070 + }, + { + "epoch": 4.52, + "learning_rate": 1.300325869862294e-06, + "loss": 0.0009, + "step": 15080 + }, + { + "epoch": 4.52, + "learning_rate": 1.297697887101861e-06, + "loss": 0.0017, + "step": 15090 + }, + { + "epoch": 4.52, + "learning_rate": 1.2950699043414277e-06, + "loss": 0.0005, + "step": 15100 + }, + { + "epoch": 4.53, + "learning_rate": 1.2924419215809943e-06, + "loss": 0.0018, + "step": 15110 + }, + { + "epoch": 4.53, + "learning_rate": 1.2898139388205613e-06, + "loss": 0.0007, + "step": 15120 + }, + { + "epoch": 4.53, + "learning_rate": 1.2871859560601283e-06, + "loss": 0.0011, + "step": 15130 + }, + { + "epoch": 4.54, + "learning_rate": 1.2845579732996951e-06, + "loss": 0.0012, + "step": 15140 + }, + { + "epoch": 4.54, + "learning_rate": 1.2819299905392622e-06, + "loss": 0.0015, + "step": 15150 + }, + { + "epoch": 4.54, + "learning_rate": 1.279302007778829e-06, + "loss": 0.0014, + "step": 15160 + }, + { + "epoch": 4.54, + "learning_rate": 1.276674025018396e-06, + "loss": 0.0009, + "step": 15170 + }, + { + "epoch": 4.55, + "learning_rate": 1.2740460422579628e-06, + "loss": 0.0008, + "step": 15180 + }, + { + "epoch": 4.55, + "learning_rate": 1.2714180594975298e-06, + "loss": 0.0008, + "step": 15190 + }, + { + "epoch": 4.55, + "learning_rate": 1.2687900767370968e-06, + "loss": 0.0009, + "step": 15200 + }, + { + "epoch": 4.55, + "eval_f1": 0.9950761598172075, + "eval_f1_0": 0.9984721078783185, + "eval_f1_1": 0.9987006814714878, + "eval_f1_2": 0.9840070826936979, + "eval_f1_3": 0.9991247672253258, + "eval_loss": 0.014415456913411617, + "eval_runtime": 153.1152, + "eval_samples_per_second": 77.51, + "eval_steps_per_second": 0.307, + "step": 15200 + }, + { + "epoch": 4.56, + "learning_rate": 1.2661620939766636e-06, + "loss": 0.002, + "step": 15210 + }, + { + "epoch": 4.56, + "learning_rate": 1.2635341112162306e-06, + "loss": 0.0107, + "step": 15220 + }, + { + "epoch": 4.56, + "learning_rate": 1.2609061284557974e-06, + "loss": 0.0007, + "step": 15230 + }, + { + "epoch": 4.57, + "learning_rate": 1.2582781456953644e-06, + "loss": 0.0016, + "step": 15240 + }, + { + "epoch": 4.57, + "learning_rate": 1.2556501629349312e-06, + "loss": 0.001, + "step": 15250 + }, + { + "epoch": 4.57, + "learning_rate": 1.2530221801744982e-06, + "loss": 0.001, + "step": 15260 + }, + { + "epoch": 4.57, + "learning_rate": 1.2503941974140652e-06, + "loss": 0.0015, + "step": 15270 + }, + { + "epoch": 4.58, + "learning_rate": 1.247766214653632e-06, + "loss": 0.0006, + "step": 15280 + }, + { + "epoch": 4.58, + "learning_rate": 1.2451382318931988e-06, + "loss": 0.0016, + "step": 15290 + }, + { + "epoch": 4.58, + "learning_rate": 1.2425102491327658e-06, + "loss": 0.0006, + "step": 15300 + }, + { + "epoch": 4.59, + "learning_rate": 1.2398822663723326e-06, + "loss": 0.0013, + "step": 15310 + }, + { + "epoch": 4.59, + "learning_rate": 1.2372542836118996e-06, + "loss": 0.0011, + "step": 15320 + }, + { + "epoch": 4.59, + "learning_rate": 1.2346263008514664e-06, + "loss": 0.0018, + "step": 15330 + }, + { + "epoch": 4.6, + "learning_rate": 1.2319983180910334e-06, + "loss": 0.0008, + "step": 15340 + }, + { + "epoch": 4.6, + "learning_rate": 1.2293703353306005e-06, + "loss": 0.0011, + "step": 15350 + }, + { + "epoch": 4.6, + "learning_rate": 1.2267423525701673e-06, + "loss": 0.0136, + "step": 15360 + }, + { + "epoch": 4.6, + "learning_rate": 1.224114369809734e-06, + "loss": 0.0014, + "step": 15370 + }, + { + "epoch": 4.61, + "learning_rate": 1.221486387049301e-06, + "loss": 0.0022, + "step": 15380 + }, + { + "epoch": 4.61, + "learning_rate": 1.2188584042888679e-06, + "loss": 0.001, + "step": 15390 + }, + { + "epoch": 4.61, + "learning_rate": 1.2162304215284349e-06, + "loss": 0.0011, + "step": 15400 + }, + { + "epoch": 4.61, + "eval_f1": 0.995032009945664, + "eval_f1_0": 0.9984474306158969, + "eval_f1_1": 0.9984507657972037, + "eval_f1_2": 0.9841639094741274, + "eval_f1_3": 0.9990659338954281, + "eval_loss": 0.01461731269955635, + "eval_runtime": 153.208, + "eval_samples_per_second": 77.463, + "eval_steps_per_second": 0.307, + "step": 15400 + }, + { + "epoch": 4.62, + "learning_rate": 1.2136024387680017e-06, + "loss": 0.0032, + "step": 15410 + }, + { + "epoch": 4.62, + "learning_rate": 1.2109744560075687e-06, + "loss": 0.001, + "step": 15420 + }, + { + "epoch": 4.62, + "learning_rate": 1.2083464732471357e-06, + "loss": 0.0009, + "step": 15430 + }, + { + "epoch": 4.63, + "learning_rate": 1.2057184904867025e-06, + "loss": 0.0012, + "step": 15440 + }, + { + "epoch": 4.63, + "learning_rate": 1.2030905077262695e-06, + "loss": 0.0012, + "step": 15450 + }, + { + "epoch": 4.63, + "learning_rate": 1.2004625249658363e-06, + "loss": 0.001, + "step": 15460 + }, + { + "epoch": 4.63, + "learning_rate": 1.1978345422054033e-06, + "loss": 0.0013, + "step": 15470 + }, + { + "epoch": 4.64, + "learning_rate": 1.19520655944497e-06, + "loss": 0.001, + "step": 15480 + }, + { + "epoch": 4.64, + "learning_rate": 1.192578576684537e-06, + "loss": 0.0009, + "step": 15490 + }, + { + "epoch": 4.64, + "learning_rate": 1.189950593924104e-06, + "loss": 0.0006, + "step": 15500 + }, + { + "epoch": 4.65, + "learning_rate": 1.187322611163671e-06, + "loss": 0.0012, + "step": 15510 + }, + { + "epoch": 4.65, + "learning_rate": 1.1846946284032377e-06, + "loss": 0.001, + "step": 15520 + }, + { + "epoch": 4.65, + "learning_rate": 1.1820666456428047e-06, + "loss": 0.0009, + "step": 15530 + }, + { + "epoch": 4.66, + "learning_rate": 1.1794386628823715e-06, + "loss": 0.0009, + "step": 15540 + }, + { + "epoch": 4.66, + "learning_rate": 1.1768106801219385e-06, + "loss": 0.0075, + "step": 15550 + }, + { + "epoch": 4.66, + "learning_rate": 1.1741826973615055e-06, + "loss": 0.0007, + "step": 15560 + }, + { + "epoch": 4.66, + "learning_rate": 1.1715547146010723e-06, + "loss": 0.0013, + "step": 15570 + }, + { + "epoch": 4.67, + "learning_rate": 1.1689267318406394e-06, + "loss": 0.0011, + "step": 15580 + }, + { + "epoch": 4.67, + "learning_rate": 1.1662987490802062e-06, + "loss": 0.0009, + "step": 15590 + }, + { + "epoch": 4.67, + "learning_rate": 1.163670766319773e-06, + "loss": 0.0016, + "step": 15600 + }, + { + "epoch": 4.67, + "eval_f1": 0.994661222330338, + "eval_f1_0": 0.9983344600018881, + "eval_f1_1": 0.9987661558858579, + "eval_f1_2": 0.9824319428977768, + "eval_f1_3": 0.9991123305358296, + "eval_loss": 0.015577760525047779, + "eval_runtime": 153.1847, + "eval_samples_per_second": 77.475, + "eval_steps_per_second": 0.307, + "step": 15600 + }, + { + "epoch": 4.68, + "learning_rate": 1.16104278355934e-06, + "loss": 0.001, + "step": 15610 + }, + { + "epoch": 4.68, + "learning_rate": 1.1584148007989068e-06, + "loss": 0.0016, + "step": 15620 + }, + { + "epoch": 4.68, + "learning_rate": 1.1557868180384738e-06, + "loss": 0.0011, + "step": 15630 + }, + { + "epoch": 4.69, + "learning_rate": 1.1531588352780408e-06, + "loss": 0.0008, + "step": 15640 + }, + { + "epoch": 4.69, + "learning_rate": 1.1505308525176076e-06, + "loss": 0.001, + "step": 15650 + }, + { + "epoch": 4.69, + "learning_rate": 1.1479028697571746e-06, + "loss": 0.0015, + "step": 15660 + }, + { + "epoch": 4.69, + "learning_rate": 1.1452748869967414e-06, + "loss": 0.004, + "step": 15670 + }, + { + "epoch": 4.7, + "learning_rate": 1.1426469042363084e-06, + "loss": 0.0009, + "step": 15680 + }, + { + "epoch": 4.7, + "learning_rate": 1.1400189214758752e-06, + "loss": 0.0009, + "step": 15690 + }, + { + "epoch": 4.7, + "learning_rate": 1.137390938715442e-06, + "loss": 0.0008, + "step": 15700 + }, + { + "epoch": 4.71, + "learning_rate": 1.134762955955009e-06, + "loss": 0.0013, + "step": 15710 + }, + { + "epoch": 4.71, + "learning_rate": 1.1321349731945758e-06, + "loss": 0.0009, + "step": 15720 + }, + { + "epoch": 4.71, + "learning_rate": 1.1295069904341428e-06, + "loss": 0.0024, + "step": 15730 + }, + { + "epoch": 4.72, + "learning_rate": 1.1268790076737098e-06, + "loss": 0.0011, + "step": 15740 + }, + { + "epoch": 4.72, + "learning_rate": 1.1242510249132766e-06, + "loss": 0.0012, + "step": 15750 + }, + { + "epoch": 4.72, + "learning_rate": 1.1216230421528436e-06, + "loss": 0.0008, + "step": 15760 + }, + { + "epoch": 4.72, + "learning_rate": 1.1189950593924104e-06, + "loss": 0.0009, + "step": 15770 + }, + { + "epoch": 4.73, + "learning_rate": 1.1163670766319774e-06, + "loss": 0.0008, + "step": 15780 + }, + { + "epoch": 4.73, + "learning_rate": 1.1137390938715445e-06, + "loss": 0.0017, + "step": 15790 + }, + { + "epoch": 4.73, + "learning_rate": 1.111111111111111e-06, + "loss": 0.0009, + "step": 15800 + }, + { + "epoch": 4.73, + "eval_f1": 0.9949544347555679, + "eval_f1_0": 0.9984277834502506, + "eval_f1_1": 0.9987446888840763, + "eval_f1_2": 0.9835453008271705, + "eval_f1_3": 0.999099965860774, + "eval_loss": 0.015532366931438446, + "eval_runtime": 153.1944, + "eval_samples_per_second": 77.47, + "eval_steps_per_second": 0.307, + "step": 15800 + }, + { + "epoch": 4.74, + "learning_rate": 1.108483128350678e-06, + "loss": 0.0013, + "step": 15810 + }, + { + "epoch": 4.74, + "learning_rate": 1.105855145590245e-06, + "loss": 0.0008, + "step": 15820 + }, + { + "epoch": 4.74, + "learning_rate": 1.1032271628298119e-06, + "loss": 0.0006, + "step": 15830 + }, + { + "epoch": 4.75, + "learning_rate": 1.1005991800693789e-06, + "loss": 0.001, + "step": 15840 + }, + { + "epoch": 4.75, + "learning_rate": 1.0979711973089457e-06, + "loss": 0.0144, + "step": 15850 + }, + { + "epoch": 4.75, + "learning_rate": 1.0953432145485127e-06, + "loss": 0.0015, + "step": 15860 + }, + { + "epoch": 4.75, + "learning_rate": 1.0927152317880797e-06, + "loss": 0.0011, + "step": 15870 + }, + { + "epoch": 4.76, + "learning_rate": 1.0900872490276465e-06, + "loss": 0.0011, + "step": 15880 + }, + { + "epoch": 4.76, + "learning_rate": 1.0874592662672135e-06, + "loss": 0.0014, + "step": 15890 + }, + { + "epoch": 4.76, + "learning_rate": 1.0848312835067803e-06, + "loss": 0.0007, + "step": 15900 + }, + { + "epoch": 4.77, + "learning_rate": 1.082203300746347e-06, + "loss": 0.0026, + "step": 15910 + }, + { + "epoch": 4.77, + "learning_rate": 1.079575317985914e-06, + "loss": 0.001, + "step": 15920 + }, + { + "epoch": 4.77, + "learning_rate": 1.076947335225481e-06, + "loss": 0.0013, + "step": 15930 + }, + { + "epoch": 4.78, + "learning_rate": 1.074319352465048e-06, + "loss": 0.0015, + "step": 15940 + }, + { + "epoch": 4.78, + "learning_rate": 1.071691369704615e-06, + "loss": 0.0009, + "step": 15950 + }, + { + "epoch": 4.78, + "learning_rate": 1.0690633869441817e-06, + "loss": 0.0007, + "step": 15960 + }, + { + "epoch": 4.78, + "learning_rate": 1.0664354041837487e-06, + "loss": 0.0007, + "step": 15970 + }, + { + "epoch": 4.79, + "learning_rate": 1.0638074214233155e-06, + "loss": 0.0007, + "step": 15980 + }, + { + "epoch": 4.79, + "learning_rate": 1.0611794386628825e-06, + "loss": 0.003, + "step": 15990 + }, + { + "epoch": 4.79, + "learning_rate": 1.0585514559024493e-06, + "loss": 0.001, + "step": 16000 + }, + { + "epoch": 4.79, + "eval_f1": 0.9920330423874137, + "eval_f1_0": 0.9974704429001112, + "eval_f1_1": 0.9985448368549305, + "eval_f1_2": 0.9730106668717138, + "eval_f1_3": 0.9991062229228993, + "eval_loss": 0.020844409242272377, + "eval_runtime": 153.1204, + "eval_samples_per_second": 77.508, + "eval_steps_per_second": 0.307, + "step": 16000 + }, + { + "epoch": 4.8, + "learning_rate": 1.0559234731420161e-06, + "loss": 0.0015, + "step": 16010 + }, + { + "epoch": 4.8, + "learning_rate": 1.0532954903815831e-06, + "loss": 0.0017, + "step": 16020 + }, + { + "epoch": 4.8, + "learning_rate": 1.0506675076211502e-06, + "loss": 0.0006, + "step": 16030 + }, + { + "epoch": 4.81, + "learning_rate": 1.048039524860717e-06, + "loss": 0.0007, + "step": 16040 + }, + { + "epoch": 4.81, + "learning_rate": 1.045411542100284e-06, + "loss": 0.0009, + "step": 16050 + }, + { + "epoch": 4.81, + "learning_rate": 1.0427835593398508e-06, + "loss": 0.0057, + "step": 16060 + }, + { + "epoch": 4.81, + "learning_rate": 1.0401555765794178e-06, + "loss": 0.0009, + "step": 16070 + }, + { + "epoch": 4.82, + "learning_rate": 1.0375275938189846e-06, + "loss": 0.0009, + "step": 16080 + }, + { + "epoch": 4.82, + "learning_rate": 1.0348996110585516e-06, + "loss": 0.0008, + "step": 16090 + }, + { + "epoch": 4.82, + "learning_rate": 1.0322716282981186e-06, + "loss": 0.0014, + "step": 16100 + }, + { + "epoch": 4.83, + "learning_rate": 1.0296436455376854e-06, + "loss": 0.0007, + "step": 16110 + }, + { + "epoch": 4.83, + "learning_rate": 1.0270156627772522e-06, + "loss": 0.0015, + "step": 16120 + }, + { + "epoch": 4.83, + "learning_rate": 1.0243876800168192e-06, + "loss": 0.0007, + "step": 16130 + }, + { + "epoch": 4.84, + "learning_rate": 1.021759697256386e-06, + "loss": 0.0017, + "step": 16140 + }, + { + "epoch": 4.84, + "learning_rate": 1.019131714495953e-06, + "loss": 0.0015, + "step": 16150 + }, + { + "epoch": 4.84, + "learning_rate": 1.0165037317355198e-06, + "loss": 0.0012, + "step": 16160 + }, + { + "epoch": 4.84, + "learning_rate": 1.0138757489750868e-06, + "loss": 0.0007, + "step": 16170 + }, + { + "epoch": 4.85, + "learning_rate": 1.0112477662146538e-06, + "loss": 0.0007, + "step": 16180 + }, + { + "epoch": 4.85, + "learning_rate": 1.0086197834542206e-06, + "loss": 0.0034, + "step": 16190 + }, + { + "epoch": 4.85, + "learning_rate": 1.0059918006937876e-06, + "loss": 0.0016, + "step": 16200 + }, + { + "epoch": 4.85, + "eval_f1": 0.9944965177719658, + "eval_f1_0": 0.9982769405207954, + "eval_f1_1": 0.9986871836543824, + "eval_f1_2": 0.9819282275617314, + "eval_f1_3": 0.999093719350954, + "eval_loss": 0.014544418081641197, + "eval_runtime": 153.3038, + "eval_samples_per_second": 77.415, + "eval_steps_per_second": 0.307, + "step": 16200 + }, + { + "epoch": 4.86, + "learning_rate": 1.0033638179333544e-06, + "loss": 0.0022, + "step": 16210 + }, + { + "epoch": 4.86, + "learning_rate": 1.0007358351729214e-06, + "loss": 0.0012, + "step": 16220 + }, + { + "epoch": 4.86, + "learning_rate": 9.981078524124882e-07, + "loss": 0.0009, + "step": 16230 + }, + { + "epoch": 4.87, + "learning_rate": 9.95479869652055e-07, + "loss": 0.0006, + "step": 16240 + }, + { + "epoch": 4.87, + "learning_rate": 9.92851886891622e-07, + "loss": 0.0089, + "step": 16250 + }, + { + "epoch": 4.87, + "learning_rate": 9.90223904131189e-07, + "loss": 0.006, + "step": 16260 + }, + { + "epoch": 4.87, + "learning_rate": 9.875959213707559e-07, + "loss": 0.0012, + "step": 16270 + }, + { + "epoch": 4.88, + "learning_rate": 9.849679386103229e-07, + "loss": 0.0026, + "step": 16280 + }, + { + "epoch": 4.88, + "learning_rate": 9.823399558498897e-07, + "loss": 0.0008, + "step": 16290 + }, + { + "epoch": 4.88, + "learning_rate": 9.797119730894567e-07, + "loss": 0.0007, + "step": 16300 + }, + { + "epoch": 4.89, + "learning_rate": 9.770839903290235e-07, + "loss": 0.0007, + "step": 16310 + }, + { + "epoch": 4.89, + "learning_rate": 9.744560075685905e-07, + "loss": 0.008, + "step": 16320 + }, + { + "epoch": 4.89, + "learning_rate": 9.718280248081573e-07, + "loss": 0.0058, + "step": 16330 + }, + { + "epoch": 4.9, + "learning_rate": 9.692000420477243e-07, + "loss": 0.0007, + "step": 16340 + }, + { + "epoch": 4.9, + "learning_rate": 9.66572059287291e-07, + "loss": 0.0012, + "step": 16350 + }, + { + "epoch": 4.9, + "learning_rate": 9.63944076526858e-07, + "loss": 0.0015, + "step": 16360 + }, + { + "epoch": 4.9, + "learning_rate": 9.61316093766425e-07, + "loss": 0.0021, + "step": 16370 + }, + { + "epoch": 4.91, + "learning_rate": 9.58688111005992e-07, + "loss": 0.0009, + "step": 16380 + }, + { + "epoch": 4.91, + "learning_rate": 9.560601282455587e-07, + "loss": 0.0006, + "step": 16390 + }, + { + "epoch": 4.91, + "learning_rate": 9.534321454851257e-07, + "loss": 0.0008, + "step": 16400 + }, + { + "epoch": 4.91, + "eval_f1": 0.9952574357684303, + "eval_f1_0": 0.9985215018271703, + "eval_f1_1": 0.9986774058753848, + "eval_f1_2": 0.9847153928596434, + "eval_f1_3": 0.9991154425115225, + "eval_loss": 0.013253765180706978, + "eval_runtime": 153.1517, + "eval_samples_per_second": 77.492, + "eval_steps_per_second": 0.307, + "step": 16400 + }, + { + "epoch": 4.92, + "learning_rate": 9.508041627246926e-07, + "loss": 0.0017, + "step": 16410 + }, + { + "epoch": 4.92, + "learning_rate": 9.481761799642595e-07, + "loss": 0.0031, + "step": 16420 + }, + { + "epoch": 4.92, + "learning_rate": 9.455481972038265e-07, + "loss": 0.001, + "step": 16430 + }, + { + "epoch": 4.93, + "learning_rate": 9.429202144433932e-07, + "loss": 0.0019, + "step": 16440 + }, + { + "epoch": 4.93, + "learning_rate": 9.402922316829602e-07, + "loss": 0.0008, + "step": 16450 + }, + { + "epoch": 4.93, + "learning_rate": 9.376642489225271e-07, + "loss": 0.0011, + "step": 16460 + }, + { + "epoch": 4.93, + "learning_rate": 9.35036266162094e-07, + "loss": 0.0013, + "step": 16470 + }, + { + "epoch": 4.94, + "learning_rate": 9.324082834016609e-07, + "loss": 0.001, + "step": 16480 + }, + { + "epoch": 4.94, + "learning_rate": 9.297803006412279e-07, + "loss": 0.0008, + "step": 16490 + }, + { + "epoch": 4.94, + "learning_rate": 9.271523178807948e-07, + "loss": 0.0009, + "step": 16500 + }, + { + "epoch": 4.95, + "learning_rate": 9.245243351203618e-07, + "loss": 0.0009, + "step": 16510 + }, + { + "epoch": 4.95, + "learning_rate": 9.218963523599287e-07, + "loss": 0.0014, + "step": 16520 + }, + { + "epoch": 4.95, + "learning_rate": 9.192683695994956e-07, + "loss": 0.0061, + "step": 16530 + }, + { + "epoch": 4.96, + "learning_rate": 9.166403868390625e-07, + "loss": 0.0017, + "step": 16540 + }, + { + "epoch": 4.96, + "learning_rate": 9.140124040786293e-07, + "loss": 0.0009, + "step": 16550 + }, + { + "epoch": 4.96, + "learning_rate": 9.113844213181962e-07, + "loss": 0.0021, + "step": 16560 + }, + { + "epoch": 4.96, + "learning_rate": 9.087564385577631e-07, + "loss": 0.0032, + "step": 16570 + }, + { + "epoch": 4.97, + "learning_rate": 9.0612845579733e-07, + "loss": 0.0005, + "step": 16580 + }, + { + "epoch": 4.97, + "learning_rate": 9.03500473036897e-07, + "loss": 0.0009, + "step": 16590 + }, + { + "epoch": 4.97, + "learning_rate": 9.008724902764639e-07, + "loss": 0.0105, + "step": 16600 + }, + { + "epoch": 4.97, + "eval_f1": 0.9956387975196138, + "eval_f1_0": 0.998639219592666, + "eval_f1_1": 0.9987463915628702, + "eval_f1_2": 0.986063261702769, + "eval_f1_3": 0.9991063172201501, + "eval_loss": 0.012624002993106842, + "eval_runtime": 153.1353, + "eval_samples_per_second": 77.5, + "eval_steps_per_second": 0.307, + "step": 16600 + }, + { + "epoch": 4.98, + "learning_rate": 8.982445075160308e-07, + "loss": 0.0014, + "step": 16610 + }, + { + "epoch": 4.98, + "learning_rate": 8.956165247555977e-07, + "loss": 0.0014, + "step": 16620 + }, + { + "epoch": 4.98, + "learning_rate": 8.929885419951646e-07, + "loss": 0.0011, + "step": 16630 + }, + { + "epoch": 4.99, + "learning_rate": 8.903605592347315e-07, + "loss": 0.0006, + "step": 16640 + }, + { + "epoch": 4.99, + "learning_rate": 8.877325764742983e-07, + "loss": 0.001, + "step": 16650 + }, + { + "epoch": 4.99, + "learning_rate": 8.851045937138652e-07, + "loss": 0.0011, + "step": 16660 + }, + { + "epoch": 4.99, + "learning_rate": 8.824766109534321e-07, + "loss": 0.0008, + "step": 16670 + }, + { + "epoch": 5.0, + "learning_rate": 8.798486281929991e-07, + "loss": 0.0019, + "step": 16680 + }, + { + "epoch": 5.0, + "learning_rate": 8.77220645432566e-07, + "loss": 0.0013, + "step": 16690 + }, + { + "epoch": 5.0, + "learning_rate": 8.745926626721329e-07, + "loss": 0.0009, + "step": 16700 + }, + { + "epoch": 5.01, + "learning_rate": 8.719646799116998e-07, + "loss": 0.0009, + "step": 16710 + }, + { + "epoch": 5.01, + "learning_rate": 8.693366971512668e-07, + "loss": 0.0017, + "step": 16720 + }, + { + "epoch": 5.01, + "learning_rate": 8.667087143908337e-07, + "loss": 0.0006, + "step": 16730 + }, + { + "epoch": 5.01, + "learning_rate": 8.640807316304007e-07, + "loss": 0.0008, + "step": 16740 + }, + { + "epoch": 5.02, + "learning_rate": 8.614527488699676e-07, + "loss": 0.0008, + "step": 16750 + }, + { + "epoch": 5.02, + "learning_rate": 8.588247661095344e-07, + "loss": 0.0009, + "step": 16760 + }, + { + "epoch": 5.02, + "learning_rate": 8.561967833491013e-07, + "loss": 0.0009, + "step": 16770 + }, + { + "epoch": 5.03, + "learning_rate": 8.535688005886682e-07, + "loss": 0.0011, + "step": 16780 + }, + { + "epoch": 5.03, + "learning_rate": 8.509408178282351e-07, + "loss": 0.0017, + "step": 16790 + }, + { + "epoch": 5.03, + "learning_rate": 8.48312835067802e-07, + "loss": 0.0007, + "step": 16800 + }, + { + "epoch": 5.03, + "eval_f1": 0.9952139453310066, + "eval_f1_0": 0.9985211697171175, + "eval_f1_1": 0.998777295254449, + "eval_f1_2": 0.9844325273962575, + "eval_f1_3": 0.9991247889562022, + "eval_loss": 0.01218812633305788, + "eval_runtime": 153.1536, + "eval_samples_per_second": 77.491, + "eval_steps_per_second": 0.307, + "step": 16800 + }, + { + "epoch": 5.04, + "learning_rate": 8.456848523073689e-07, + "loss": 0.0012, + "step": 16810 + }, + { + "epoch": 5.04, + "learning_rate": 8.430568695469359e-07, + "loss": 0.001, + "step": 16820 + }, + { + "epoch": 5.04, + "learning_rate": 8.404288867865028e-07, + "loss": 0.0009, + "step": 16830 + }, + { + "epoch": 5.04, + "learning_rate": 8.378009040260697e-07, + "loss": 0.0007, + "step": 16840 + }, + { + "epoch": 5.05, + "learning_rate": 8.351729212656366e-07, + "loss": 0.0016, + "step": 16850 + }, + { + "epoch": 5.05, + "learning_rate": 8.325449385052035e-07, + "loss": 0.0006, + "step": 16860 + }, + { + "epoch": 5.05, + "learning_rate": 8.299169557447703e-07, + "loss": 0.0014, + "step": 16870 + }, + { + "epoch": 5.06, + "learning_rate": 8.272889729843372e-07, + "loss": 0.0007, + "step": 16880 + }, + { + "epoch": 5.06, + "learning_rate": 8.246609902239041e-07, + "loss": 0.0008, + "step": 16890 + }, + { + "epoch": 5.06, + "learning_rate": 8.220330074634711e-07, + "loss": 0.0015, + "step": 16900 + }, + { + "epoch": 5.07, + "learning_rate": 8.19405024703038e-07, + "loss": 0.0009, + "step": 16910 + }, + { + "epoch": 5.07, + "learning_rate": 8.167770419426049e-07, + "loss": 0.001, + "step": 16920 + }, + { + "epoch": 5.07, + "learning_rate": 8.141490591821718e-07, + "loss": 0.0082, + "step": 16930 + }, + { + "epoch": 5.07, + "learning_rate": 8.115210764217388e-07, + "loss": 0.0005, + "step": 16940 + }, + { + "epoch": 5.08, + "learning_rate": 8.088930936613057e-07, + "loss": 0.001, + "step": 16950 + }, + { + "epoch": 5.08, + "learning_rate": 8.062651109008727e-07, + "loss": 0.0007, + "step": 16960 + }, + { + "epoch": 5.08, + "learning_rate": 8.036371281404394e-07, + "loss": 0.0012, + "step": 16970 + }, + { + "epoch": 5.09, + "learning_rate": 8.010091453800064e-07, + "loss": 0.0009, + "step": 16980 + }, + { + "epoch": 5.09, + "learning_rate": 7.983811626195733e-07, + "loss": 0.0012, + "step": 16990 + }, + { + "epoch": 5.09, + "learning_rate": 7.957531798591402e-07, + "loss": 0.0009, + "step": 17000 + }, + { + "epoch": 5.09, + "eval_f1": 0.9945871771903947, + "eval_f1_0": 0.9982992029336855, + "eval_f1_1": 0.998671439006014, + "eval_f1_2": 0.9822439160523908, + "eval_f1_3": 0.9991341507694886, + "eval_loss": 0.013804233632981777, + "eval_runtime": 153.0847, + "eval_samples_per_second": 77.526, + "eval_steps_per_second": 0.307, + "step": 17000 + }, + { + "epoch": 5.1, + "learning_rate": 7.931251970987071e-07, + "loss": 0.0012, + "step": 17010 + }, + { + "epoch": 5.1, + "learning_rate": 7.90497214338274e-07, + "loss": 0.001, + "step": 17020 + }, + { + "epoch": 5.1, + "learning_rate": 7.878692315778409e-07, + "loss": 0.0011, + "step": 17030 + }, + { + "epoch": 5.1, + "learning_rate": 7.852412488174079e-07, + "loss": 0.0051, + "step": 17040 + }, + { + "epoch": 5.11, + "learning_rate": 7.826132660569748e-07, + "loss": 0.0009, + "step": 17050 + }, + { + "epoch": 5.11, + "learning_rate": 7.799852832965417e-07, + "loss": 0.0008, + "step": 17060 + }, + { + "epoch": 5.11, + "learning_rate": 7.773573005361086e-07, + "loss": 0.0182, + "step": 17070 + }, + { + "epoch": 5.12, + "learning_rate": 7.747293177756754e-07, + "loss": 0.001, + "step": 17080 + }, + { + "epoch": 5.12, + "learning_rate": 7.721013350152423e-07, + "loss": 0.0019, + "step": 17090 + }, + { + "epoch": 5.12, + "learning_rate": 7.694733522548092e-07, + "loss": 0.0009, + "step": 17100 + }, + { + "epoch": 5.13, + "learning_rate": 7.668453694943761e-07, + "loss": 0.0015, + "step": 17110 + }, + { + "epoch": 5.13, + "learning_rate": 7.64217386733943e-07, + "loss": 0.0018, + "step": 17120 + }, + { + "epoch": 5.13, + "learning_rate": 7.6158940397351e-07, + "loss": 0.0007, + "step": 17130 + }, + { + "epoch": 5.13, + "learning_rate": 7.589614212130769e-07, + "loss": 0.0008, + "step": 17140 + }, + { + "epoch": 5.14, + "learning_rate": 7.563334384526438e-07, + "loss": 0.0014, + "step": 17150 + }, + { + "epoch": 5.14, + "learning_rate": 7.537054556922108e-07, + "loss": 0.0014, + "step": 17160 + }, + { + "epoch": 5.14, + "learning_rate": 7.510774729317777e-07, + "loss": 0.0008, + "step": 17170 + }, + { + "epoch": 5.15, + "learning_rate": 7.484494901713446e-07, + "loss": 0.0011, + "step": 17180 + }, + { + "epoch": 5.15, + "learning_rate": 7.458215074109114e-07, + "loss": 0.0006, + "step": 17190 + }, + { + "epoch": 5.15, + "learning_rate": 7.431935246504783e-07, + "loss": 0.0012, + "step": 17200 + }, + { + "epoch": 5.15, + "eval_f1": 0.9954414657636388, + "eval_f1_0": 0.998593031247741, + "eval_f1_1": 0.9987407110669838, + "eval_f1_2": 0.9853259366509948, + "eval_f1_3": 0.9991061840888353, + "eval_loss": 0.012961289845407009, + "eval_runtime": 153.1424, + "eval_samples_per_second": 77.496, + "eval_steps_per_second": 0.307, + "step": 17200 + }, + { + "epoch": 5.16, + "learning_rate": 7.405655418900453e-07, + "loss": 0.0012, + "step": 17210 + }, + { + "epoch": 5.16, + "learning_rate": 7.379375591296122e-07, + "loss": 0.0008, + "step": 17220 + }, + { + "epoch": 5.16, + "learning_rate": 7.353095763691791e-07, + "loss": 0.0032, + "step": 17230 + }, + { + "epoch": 5.16, + "learning_rate": 7.32681593608746e-07, + "loss": 0.0016, + "step": 17240 + }, + { + "epoch": 5.17, + "learning_rate": 7.300536108483129e-07, + "loss": 0.001, + "step": 17250 + }, + { + "epoch": 5.17, + "learning_rate": 7.274256280878798e-07, + "loss": 0.0143, + "step": 17260 + }, + { + "epoch": 5.17, + "learning_rate": 7.247976453274468e-07, + "loss": 0.0011, + "step": 17270 + }, + { + "epoch": 5.18, + "learning_rate": 7.221696625670137e-07, + "loss": 0.0108, + "step": 17280 + }, + { + "epoch": 5.18, + "learning_rate": 7.195416798065805e-07, + "loss": 0.0023, + "step": 17290 + }, + { + "epoch": 5.18, + "learning_rate": 7.169136970461474e-07, + "loss": 0.0007, + "step": 17300 + }, + { + "epoch": 5.19, + "learning_rate": 7.142857142857143e-07, + "loss": 0.001, + "step": 17310 + }, + { + "epoch": 5.19, + "learning_rate": 7.116577315252812e-07, + "loss": 0.001, + "step": 17320 + }, + { + "epoch": 5.19, + "learning_rate": 7.090297487648481e-07, + "loss": 0.0013, + "step": 17330 + }, + { + "epoch": 5.19, + "learning_rate": 7.06401766004415e-07, + "loss": 0.0014, + "step": 17340 + }, + { + "epoch": 5.2, + "learning_rate": 7.03773783243982e-07, + "loss": 0.0013, + "step": 17350 + }, + { + "epoch": 5.2, + "learning_rate": 7.011458004835489e-07, + "loss": 0.0009, + "step": 17360 + }, + { + "epoch": 5.2, + "learning_rate": 6.985178177231158e-07, + "loss": 0.0011, + "step": 17370 + }, + { + "epoch": 5.21, + "learning_rate": 6.958898349626827e-07, + "loss": 0.0007, + "step": 17380 + }, + { + "epoch": 5.21, + "learning_rate": 6.932618522022497e-07, + "loss": 0.0011, + "step": 17390 + }, + { + "epoch": 5.21, + "learning_rate": 6.906338694418165e-07, + "loss": 0.0009, + "step": 17400 + }, + { + "epoch": 5.21, + "eval_f1": 0.9947430710226786, + "eval_f1_0": 0.9983573068836467, + "eval_f1_1": 0.9987637923401124, + "eval_f1_2": 0.982729464060428, + "eval_f1_3": 0.9991217208065272, + "eval_loss": 0.015339343808591366, + "eval_runtime": 153.1884, + "eval_samples_per_second": 77.473, + "eval_steps_per_second": 0.307, + "step": 17400 + }, + { + "epoch": 5.22, + "learning_rate": 6.880058866813834e-07, + "loss": 0.0009, + "step": 17410 + }, + { + "epoch": 5.22, + "learning_rate": 6.853779039209503e-07, + "loss": 0.0017, + "step": 17420 + }, + { + "epoch": 5.22, + "learning_rate": 6.827499211605173e-07, + "loss": 0.0009, + "step": 17430 + }, + { + "epoch": 5.22, + "learning_rate": 6.801219384000842e-07, + "loss": 0.002, + "step": 17440 + }, + { + "epoch": 5.23, + "learning_rate": 6.774939556396511e-07, + "loss": 0.0007, + "step": 17450 + }, + { + "epoch": 5.23, + "learning_rate": 6.74865972879218e-07, + "loss": 0.0007, + "step": 17460 + }, + { + "epoch": 5.23, + "learning_rate": 6.722379901187849e-07, + "loss": 0.002, + "step": 17470 + }, + { + "epoch": 5.24, + "learning_rate": 6.696100073583518e-07, + "loss": 0.0005, + "step": 17480 + }, + { + "epoch": 5.24, + "learning_rate": 6.669820245979187e-07, + "loss": 0.0013, + "step": 17490 + }, + { + "epoch": 5.24, + "learning_rate": 6.643540418374857e-07, + "loss": 0.0005, + "step": 17500 + }, + { + "epoch": 5.25, + "learning_rate": 6.617260590770525e-07, + "loss": 0.0006, + "step": 17510 + }, + { + "epoch": 5.25, + "learning_rate": 6.590980763166194e-07, + "loss": 0.001, + "step": 17520 + }, + { + "epoch": 5.25, + "learning_rate": 6.564700935561863e-07, + "loss": 0.0013, + "step": 17530 + }, + { + "epoch": 5.25, + "learning_rate": 6.538421107957532e-07, + "loss": 0.001, + "step": 17540 + }, + { + "epoch": 5.26, + "learning_rate": 6.512141280353201e-07, + "loss": 0.0008, + "step": 17550 + }, + { + "epoch": 5.26, + "learning_rate": 6.48586145274887e-07, + "loss": 0.001, + "step": 17560 + }, + { + "epoch": 5.26, + "learning_rate": 6.459581625144539e-07, + "loss": 0.0013, + "step": 17570 + }, + { + "epoch": 5.27, + "learning_rate": 6.433301797540209e-07, + "loss": 0.0008, + "step": 17580 + }, + { + "epoch": 5.27, + "learning_rate": 6.407021969935878e-07, + "loss": 0.0012, + "step": 17590 + }, + { + "epoch": 5.27, + "learning_rate": 6.380742142331547e-07, + "loss": 0.0008, + "step": 17600 + }, + { + "epoch": 5.27, + "eval_f1": 0.9949769082175013, + "eval_f1_0": 0.9984244829968855, + "eval_f1_1": 0.9985524501937416, + "eval_f1_2": 0.9837841617872118, + "eval_f1_3": 0.9991465378921658, + "eval_loss": 0.014890719205141068, + "eval_runtime": 153.1661, + "eval_samples_per_second": 77.485, + "eval_steps_per_second": 0.307, + "step": 17600 + }, + { + "epoch": 5.28, + "learning_rate": 6.354462314727215e-07, + "loss": 0.0007, + "step": 17610 + }, + { + "epoch": 5.28, + "learning_rate": 6.328182487122885e-07, + "loss": 0.0006, + "step": 17620 + }, + { + "epoch": 5.28, + "learning_rate": 6.301902659518554e-07, + "loss": 0.0007, + "step": 17630 + }, + { + "epoch": 5.28, + "learning_rate": 6.275622831914223e-07, + "loss": 0.0011, + "step": 17640 + }, + { + "epoch": 5.29, + "learning_rate": 6.249343004309892e-07, + "loss": 0.0007, + "step": 17650 + }, + { + "epoch": 5.29, + "learning_rate": 6.223063176705562e-07, + "loss": 0.0009, + "step": 17660 + }, + { + "epoch": 5.29, + "learning_rate": 6.196783349101231e-07, + "loss": 0.0014, + "step": 17670 + }, + { + "epoch": 5.3, + "learning_rate": 6.1705035214969e-07, + "loss": 0.0007, + "step": 17680 + }, + { + "epoch": 5.3, + "learning_rate": 6.144223693892568e-07, + "loss": 0.0009, + "step": 17690 + }, + { + "epoch": 5.3, + "learning_rate": 6.117943866288238e-07, + "loss": 0.0007, + "step": 17700 + }, + { + "epoch": 5.31, + "learning_rate": 6.091664038683907e-07, + "loss": 0.0011, + "step": 17710 + }, + { + "epoch": 5.31, + "learning_rate": 6.065384211079576e-07, + "loss": 0.0011, + "step": 17720 + }, + { + "epoch": 5.31, + "learning_rate": 6.039104383475245e-07, + "loss": 0.001, + "step": 17730 + }, + { + "epoch": 5.31, + "learning_rate": 6.012824555870914e-07, + "loss": 0.0016, + "step": 17740 + }, + { + "epoch": 5.32, + "learning_rate": 5.986544728266583e-07, + "loss": 0.0007, + "step": 17750 + }, + { + "epoch": 5.32, + "learning_rate": 5.960264900662252e-07, + "loss": 0.0011, + "step": 17760 + }, + { + "epoch": 5.32, + "learning_rate": 5.933985073057921e-07, + "loss": 0.0008, + "step": 17770 + }, + { + "epoch": 5.33, + "learning_rate": 5.90770524545359e-07, + "loss": 0.0006, + "step": 17780 + }, + { + "epoch": 5.33, + "learning_rate": 5.881425417849259e-07, + "loss": 0.0011, + "step": 17790 + }, + { + "epoch": 5.33, + "learning_rate": 5.855145590244928e-07, + "loss": 0.001, + "step": 17800 + }, + { + "epoch": 5.33, + "eval_f1": 0.9954665894094483, + "eval_f1_0": 0.998584781766185, + "eval_f1_1": 0.9986139574295141, + "eval_f1_2": 0.9855117779137499, + "eval_f1_3": 0.9991558405283445, + "eval_loss": 0.013635880313813686, + "eval_runtime": 153.1797, + "eval_samples_per_second": 77.478, + "eval_steps_per_second": 0.307, + "step": 17800 + }, + { + "epoch": 5.34, + "learning_rate": 5.828865762640597e-07, + "loss": 0.0013, + "step": 17810 + }, + { + "epoch": 5.34, + "learning_rate": 5.802585935036266e-07, + "loss": 0.0011, + "step": 17820 + }, + { + "epoch": 5.34, + "learning_rate": 5.776306107431935e-07, + "loss": 0.0011, + "step": 17830 + }, + { + "epoch": 5.34, + "learning_rate": 5.750026279827606e-07, + "loss": 0.0115, + "step": 17840 + }, + { + "epoch": 5.35, + "learning_rate": 5.723746452223274e-07, + "loss": 0.0011, + "step": 17850 + }, + { + "epoch": 5.35, + "learning_rate": 5.697466624618943e-07, + "loss": 0.0005, + "step": 17860 + }, + { + "epoch": 5.35, + "learning_rate": 5.671186797014612e-07, + "loss": 0.0007, + "step": 17870 + }, + { + "epoch": 5.36, + "learning_rate": 5.644906969410282e-07, + "loss": 0.0008, + "step": 17880 + }, + { + "epoch": 5.36, + "learning_rate": 5.618627141805951e-07, + "loss": 0.0018, + "step": 17890 + }, + { + "epoch": 5.36, + "learning_rate": 5.59234731420162e-07, + "loss": 0.0006, + "step": 17900 + }, + { + "epoch": 5.37, + "learning_rate": 5.566067486597288e-07, + "loss": 0.0008, + "step": 17910 + }, + { + "epoch": 5.37, + "learning_rate": 5.539787658992958e-07, + "loss": 0.0021, + "step": 17920 + }, + { + "epoch": 5.37, + "learning_rate": 5.513507831388627e-07, + "loss": 0.0008, + "step": 17930 + }, + { + "epoch": 5.37, + "learning_rate": 5.487228003784296e-07, + "loss": 0.0008, + "step": 17940 + }, + { + "epoch": 5.38, + "learning_rate": 5.460948176179965e-07, + "loss": 0.0017, + "step": 17950 + }, + { + "epoch": 5.38, + "learning_rate": 5.434668348575634e-07, + "loss": 0.0006, + "step": 17960 + }, + { + "epoch": 5.38, + "learning_rate": 5.408388520971303e-07, + "loss": 0.0007, + "step": 17970 + }, + { + "epoch": 5.39, + "learning_rate": 5.382108693366972e-07, + "loss": 0.0009, + "step": 17980 + }, + { + "epoch": 5.39, + "learning_rate": 5.355828865762641e-07, + "loss": 0.0014, + "step": 17990 + }, + { + "epoch": 5.39, + "learning_rate": 5.32954903815831e-07, + "loss": 0.0009, + "step": 18000 + }, + { + "epoch": 5.39, + "eval_f1": 0.9945377485487029, + "eval_f1_0": 0.9982842337181957, + "eval_f1_1": 0.9987349633139361, + "eval_f1_2": 0.981985264567984, + "eval_f1_3": 0.9991465325946961, + "eval_loss": 0.015714893117547035, + "eval_runtime": 153.1648, + "eval_samples_per_second": 77.485, + "eval_steps_per_second": 0.307, + "step": 18000 + }, + { + "epoch": 5.4, + "learning_rate": 5.303269210553979e-07, + "loss": 0.0007, + "step": 18010 + }, + { + "epoch": 5.4, + "learning_rate": 5.276989382949648e-07, + "loss": 0.0009, + "step": 18020 + }, + { + "epoch": 5.4, + "learning_rate": 5.250709555345317e-07, + "loss": 0.001, + "step": 18030 + }, + { + "epoch": 5.4, + "learning_rate": 5.224429727740986e-07, + "loss": 0.0008, + "step": 18040 + }, + { + "epoch": 5.41, + "learning_rate": 5.198149900136655e-07, + "loss": 0.0006, + "step": 18050 + }, + { + "epoch": 5.41, + "learning_rate": 5.171870072532324e-07, + "loss": 0.0009, + "step": 18060 + }, + { + "epoch": 5.41, + "learning_rate": 5.145590244927994e-07, + "loss": 0.0006, + "step": 18070 + }, + { + "epoch": 5.42, + "learning_rate": 5.119310417323663e-07, + "loss": 0.0009, + "step": 18080 + }, + { + "epoch": 5.42, + "learning_rate": 5.093030589719332e-07, + "loss": 0.0009, + "step": 18090 + }, + { + "epoch": 5.42, + "learning_rate": 5.066750762115001e-07, + "loss": 0.0007, + "step": 18100 + }, + { + "epoch": 5.43, + "learning_rate": 5.040470934510671e-07, + "loss": 0.0008, + "step": 18110 + }, + { + "epoch": 5.43, + "learning_rate": 5.014191106906339e-07, + "loss": 0.0008, + "step": 18120 + }, + { + "epoch": 5.43, + "learning_rate": 4.987911279302008e-07, + "loss": 0.0008, + "step": 18130 + }, + { + "epoch": 5.43, + "learning_rate": 4.961631451697677e-07, + "loss": 0.0011, + "step": 18140 + }, + { + "epoch": 5.44, + "learning_rate": 4.935351624093347e-07, + "loss": 0.0023, + "step": 18150 + }, + { + "epoch": 5.44, + "learning_rate": 4.909071796489016e-07, + "loss": 0.0032, + "step": 18160 + }, + { + "epoch": 5.44, + "learning_rate": 4.882791968884684e-07, + "loss": 0.0009, + "step": 18170 + }, + { + "epoch": 5.45, + "learning_rate": 4.856512141280353e-07, + "loss": 0.0017, + "step": 18180 + }, + { + "epoch": 5.45, + "learning_rate": 4.830232313676023e-07, + "loss": 0.0101, + "step": 18190 + }, + { + "epoch": 5.45, + "learning_rate": 4.803952486071692e-07, + "loss": 0.0007, + "step": 18200 + }, + { + "epoch": 5.45, + "eval_f1": 0.9907806831919377, + "eval_f1_0": 0.9970540101211753, + "eval_f1_1": 0.9987080499605131, + "eval_f1_2": 0.9682172462675949, + "eval_f1_3": 0.9991434264184672, + "eval_loss": 0.02452138625085354, + "eval_runtime": 153.1312, + "eval_samples_per_second": 77.502, + "eval_steps_per_second": 0.307, + "step": 18200 + }, + { + "epoch": 5.46, + "learning_rate": 4.777672658467361e-07, + "loss": 0.0009, + "step": 18210 + }, + { + "epoch": 5.46, + "learning_rate": 4.75139283086303e-07, + "loss": 0.0008, + "step": 18220 + }, + { + "epoch": 5.46, + "learning_rate": 4.7251130032586987e-07, + "loss": 0.0011, + "step": 18230 + }, + { + "epoch": 5.46, + "learning_rate": 4.698833175654368e-07, + "loss": 0.0014, + "step": 18240 + }, + { + "epoch": 5.47, + "learning_rate": 4.6725533480500373e-07, + "loss": 0.0015, + "step": 18250 + }, + { + "epoch": 5.47, + "learning_rate": 4.6462735204457064e-07, + "loss": 0.0008, + "step": 18260 + }, + { + "epoch": 5.47, + "learning_rate": 4.6199936928413754e-07, + "loss": 0.0008, + "step": 18270 + }, + { + "epoch": 5.48, + "learning_rate": 4.593713865237044e-07, + "loss": 0.001, + "step": 18280 + }, + { + "epoch": 5.48, + "learning_rate": 4.5674340376327135e-07, + "loss": 0.0005, + "step": 18290 + }, + { + "epoch": 5.48, + "learning_rate": 4.5411542100283825e-07, + "loss": 0.0005, + "step": 18300 + }, + { + "epoch": 5.49, + "learning_rate": 4.5148743824240516e-07, + "loss": 0.0007, + "step": 18310 + }, + { + "epoch": 5.49, + "learning_rate": 4.488594554819721e-07, + "loss": 0.0009, + "step": 18320 + }, + { + "epoch": 5.49, + "learning_rate": 4.4623147272153897e-07, + "loss": 0.0009, + "step": 18330 + }, + { + "epoch": 5.49, + "learning_rate": 4.4360348996110587e-07, + "loss": 0.0009, + "step": 18340 + }, + { + "epoch": 5.5, + "learning_rate": 4.409755072006728e-07, + "loss": 0.001, + "step": 18350 + }, + { + "epoch": 5.5, + "learning_rate": 4.3834752444023973e-07, + "loss": 0.0007, + "step": 18360 + }, + { + "epoch": 5.5, + "learning_rate": 4.3571954167980664e-07, + "loss": 0.0017, + "step": 18370 + }, + { + "epoch": 5.51, + "learning_rate": 4.3309155891937354e-07, + "loss": 0.001, + "step": 18380 + }, + { + "epoch": 5.51, + "learning_rate": 4.304635761589404e-07, + "loss": 0.0008, + "step": 18390 + }, + { + "epoch": 5.51, + "learning_rate": 4.2783559339850735e-07, + "loss": 0.0008, + "step": 18400 + }, + { + "epoch": 5.51, + "eval_f1": 0.993857841784884, + "eval_f1_0": 0.9980656120138868, + "eval_f1_1": 0.998681128248776, + "eval_f1_2": 0.9795411845083052, + "eval_f1_3": 0.9991434423685681, + "eval_loss": 0.01854228787124157, + "eval_runtime": 153.1554, + "eval_samples_per_second": 77.49, + "eval_steps_per_second": 0.307, + "step": 18400 + }, + { + "epoch": 5.52, + "learning_rate": 4.2520761063807425e-07, + "loss": 0.0016, + "step": 18410 + }, + { + "epoch": 5.52, + "learning_rate": 4.2257962787764116e-07, + "loss": 0.0008, + "step": 18420 + }, + { + "epoch": 5.52, + "learning_rate": 4.199516451172081e-07, + "loss": 0.0006, + "step": 18430 + }, + { + "epoch": 5.52, + "learning_rate": 4.1732366235677497e-07, + "loss": 0.002, + "step": 18440 + }, + { + "epoch": 5.53, + "learning_rate": 4.1469567959634187e-07, + "loss": 0.0009, + "step": 18450 + }, + { + "epoch": 5.53, + "learning_rate": 4.120676968359088e-07, + "loss": 0.0007, + "step": 18460 + }, + { + "epoch": 5.53, + "learning_rate": 4.0943971407547573e-07, + "loss": 0.0011, + "step": 18470 + }, + { + "epoch": 5.54, + "learning_rate": 4.0681173131504264e-07, + "loss": 0.001, + "step": 18480 + }, + { + "epoch": 5.54, + "learning_rate": 4.041837485546095e-07, + "loss": 0.0006, + "step": 18490 + }, + { + "epoch": 5.54, + "learning_rate": 4.015557657941764e-07, + "loss": 0.0013, + "step": 18500 + }, + { + "epoch": 5.55, + "learning_rate": 3.9892778303374335e-07, + "loss": 0.0008, + "step": 18510 + }, + { + "epoch": 5.55, + "learning_rate": 3.9629980027331025e-07, + "loss": 0.001, + "step": 18520 + }, + { + "epoch": 5.55, + "learning_rate": 3.9367181751287716e-07, + "loss": 0.0007, + "step": 18530 + }, + { + "epoch": 5.55, + "learning_rate": 3.910438347524441e-07, + "loss": 0.0006, + "step": 18540 + }, + { + "epoch": 5.56, + "learning_rate": 3.8841585199201097e-07, + "loss": 0.0007, + "step": 18550 + }, + { + "epoch": 5.56, + "learning_rate": 3.8578786923157787e-07, + "loss": 0.001, + "step": 18560 + }, + { + "epoch": 5.56, + "learning_rate": 3.8315988647114477e-07, + "loss": 0.0008, + "step": 18570 + }, + { + "epoch": 5.57, + "learning_rate": 3.8053190371071173e-07, + "loss": 0.0008, + "step": 18580 + }, + { + "epoch": 5.57, + "learning_rate": 3.7790392095027863e-07, + "loss": 0.0008, + "step": 18590 + }, + { + "epoch": 5.57, + "learning_rate": 3.752759381898455e-07, + "loss": 0.0012, + "step": 18600 + }, + { + "epoch": 5.57, + "eval_f1": 0.994690157199797, + "eval_f1_0": 0.998336381654215, + "eval_f1_1": 0.9987022422263926, + "eval_f1_2": 0.9825940663663268, + "eval_f1_3": 0.9991279385522539, + "eval_loss": 0.016709227114915848, + "eval_runtime": 153.1387, + "eval_samples_per_second": 77.498, + "eval_steps_per_second": 0.307, + "step": 18600 + }, + { + "epoch": 5.57, + "step": 18600, + "total_flos": 1.611826247266992e+18, + "train_loss": 0.020591222444869658, + "train_runtime": 40906.9498, + "train_samples_per_second": 15.665, + "train_steps_per_second": 0.49 + } + ], + "logging_steps": 10, + "max_steps": 20028, + "num_train_epochs": 6, + "save_steps": 200, + "total_flos": 1.611826247266992e+18, + "trial_name": null, + "trial_params": null +}