{ "best_global_step": 6000, "best_metric": 0.85033727, "best_model_checkpoint": "/NHNHOME/data/sanghyeok/qwen-cua/runs/cosmos3_exact_fullmot_8gpu/v4-20260624-133258/checkpoint-6000", "epoch": 1.6143999999999998, "eval_steps": 500, "global_step": 7000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.00023063063063063062, "grad_norm": 42004.87890625, "learning_rate": 5.0000000000000004e-08, "loss": 27.907299041748047, "loss_alignment": 2.697265625, "loss_alignment_w": 1.3486328125, "loss_sft": 0.38892047479748726, "loss_total_v6": 1.744206115603447, "step": 1 }, { "epoch": 0.00046126126126126124, "grad_norm": 39402.1015625, "learning_rate": 1.0000000000000001e-07, "loss": 27.083362579345703, "loss_alignment": 2.703125, "loss_alignment_w": 1.3515625, "loss_sft": 0.3399270214140415, "loss_total_v6": 1.6927101761102676, "step": 2 }, { "epoch": 0.0006918918918918919, "grad_norm": 6210.19970703125, "learning_rate": 1.5000000000000002e-07, "loss": 26.742481231689453, "loss_alignment": 2.67578125, "loss_alignment_w": 1.337890625, "loss_sft": 0.33595574274659157, "loss_total_v6": 1.6714049726724625, "step": 3 }, { "epoch": 0.0009225225225225225, "grad_norm": 6407.359375, "learning_rate": 2.0000000000000002e-07, "loss": 26.819990158081055, "loss_alignment": 2.6640625, "loss_alignment_w": 1.33203125, "loss_sft": 0.3405560739338398, "loss_total_v6": 1.6762493997812271, "step": 4 }, { "epoch": 0.0011531531531531532, "grad_norm": 5712.46142578125, "learning_rate": 2.5000000000000004e-07, "loss": 26.108837127685547, "loss_alignment": 2.53125, "loss_alignment_w": 1.265625, "loss_sft": 0.3662993088364601, "loss_total_v6": 1.6318022906780243, "step": 5 }, { "epoch": 0.0013837837837837837, "grad_norm": 5260.19677734375, "learning_rate": 3.0000000000000004e-07, "loss": 24.735301971435547, "loss_alignment": 2.400390625, "loss_alignment_w": 1.2001953125, "loss_sft": 0.3420378305017948, "loss_total_v6": 1.545956328511238, "step": 6 }, { "epoch": 0.0016144144144144145, "grad_norm": 3917.843505859375, "learning_rate": 3.5000000000000004e-07, "loss": 24.114948272705078, "loss_alignment": 2.29296875, "loss_alignment_w": 1.146484375, "loss_sft": 0.364911250770092, "loss_total_v6": 1.5071841776371002, "step": 7 }, { "epoch": 0.001845045045045045, "grad_norm": 3943.014892578125, "learning_rate": 4.0000000000000003e-07, "loss": 22.030826568603516, "loss_alignment": 2.08984375, "loss_alignment_w": 1.044921875, "loss_sft": 0.32989906892180443, "loss_total_v6": 1.3769266456365585, "step": 8 }, { "epoch": 0.0020756756756756755, "grad_norm": 3444.25244140625, "learning_rate": 4.5000000000000003e-07, "loss": 21.6820011138916, "loss_alignment": 2.05859375, "loss_alignment_w": 1.029296875, "loss_sft": 0.3261638656258583, "loss_total_v6": 1.3551251143217087, "step": 9 }, { "epoch": 0.0023063063063063064, "grad_norm": 2980.263671875, "learning_rate": 5.000000000000001e-07, "loss": 21.306888580322266, "loss_alignment": 2.0068359375, "loss_alignment_w": 1.00341796875, "loss_sft": 0.3288728594779968, "loss_total_v6": 1.331680491566658, "step": 10 }, { "epoch": 0.002536936936936937, "grad_norm": 2632.6884765625, "learning_rate": 5.5e-07, "loss": 18.671615600585938, "loss_alignment": 1.6630859375, "loss_alignment_w": 0.83154296875, "loss_sft": 0.3339071962982416, "loss_total_v6": 1.1669760197401047, "step": 11 }, { "epoch": 0.0027675675675675675, "grad_norm": 2456.209716796875, "learning_rate": 6.000000000000001e-07, "loss": 21.704315185546875, "loss_alignment": 1.9560546875, "loss_alignment_w": 0.97802734375, "loss_sft": 0.37663075514137745, "loss_total_v6": 1.356519654393196, "step": 12 }, { "epoch": 0.0029981981981981984, "grad_norm": 1963.2039794921875, "learning_rate": 6.5e-07, "loss": 18.490877151489258, "loss_alignment": 1.63671875, "loss_alignment_w": 0.818359375, "loss_sft": 0.3408604208379984, "loss_total_v6": 1.1556798219680786, "step": 13 }, { "epoch": 0.003228828828828829, "grad_norm": 1689.7694091796875, "learning_rate": 7.000000000000001e-07, "loss": 18.366527557373047, "loss_alignment": 1.5830078125, "loss_alignment_w": 0.79150390625, "loss_sft": 0.35533586889505386, "loss_total_v6": 1.1479079276323318, "step": 14 }, { "epoch": 0.0034594594594594594, "grad_norm": 1395.5404052734375, "learning_rate": 7.5e-07, "loss": 17.421850204467773, "loss_alignment": 1.5205078125, "loss_alignment_w": 0.76025390625, "loss_sft": 0.32809288799762726, "loss_total_v6": 1.0888655930757523, "step": 15 }, { "epoch": 0.00369009009009009, "grad_norm": 1174.551513671875, "learning_rate": 8.000000000000001e-07, "loss": 16.753416061401367, "loss_alignment": 1.49609375, "loss_alignment_w": 0.748046875, "loss_sft": 0.299255246296525, "loss_total_v6": 1.0470884516835213, "step": 16 }, { "epoch": 0.003920720720720721, "grad_norm": 1010.3545532226562, "learning_rate": 8.500000000000001e-07, "loss": 16.71527862548828, "loss_alignment": 1.392578125, "loss_alignment_w": 0.6962890625, "loss_sft": 0.34869053214788437, "loss_total_v6": 1.0447049289941788, "step": 17 }, { "epoch": 0.004151351351351351, "grad_norm": 971.2501831054688, "learning_rate": 9.000000000000001e-07, "loss": 16.939220428466797, "loss_alignment": 1.453125, "loss_alignment_w": 0.7265625, "loss_sft": 0.3348243907094002, "loss_total_v6": 1.0587013214826584, "step": 18 }, { "epoch": 0.004381981981981982, "grad_norm": 845.4120483398438, "learning_rate": 9.500000000000001e-07, "loss": 15.989473342895508, "loss_alignment": 1.3681640625, "loss_alignment_w": 0.68408203125, "loss_sft": 0.3158704340457916, "loss_total_v6": 0.9993420913815498, "step": 19 }, { "epoch": 0.004612612612612613, "grad_norm": 697.2727661132812, "learning_rate": 1.0000000000000002e-06, "loss": 15.408124923706055, "loss_alignment": 1.306640625, "loss_alignment_w": 0.6533203125, "loss_sft": 0.30797853507101536, "loss_total_v6": 0.963007815182209, "step": 20 }, { "epoch": 0.004843243243243243, "grad_norm": 653.9332885742188, "learning_rate": 1.0500000000000001e-06, "loss": 15.71966552734375, "loss_alignment": 1.3251953125, "loss_alignment_w": 0.66259765625, "loss_sft": 0.3197898827493191, "loss_total_v6": 0.9824790880084038, "step": 21 }, { "epoch": 0.005073873873873874, "grad_norm": 600.2960205078125, "learning_rate": 1.1e-06, "loss": 14.895074844360352, "loss_alignment": 1.2626953125, "loss_alignment_w": 0.63134765625, "loss_sft": 0.301212003454566, "loss_total_v6": 0.9309421926736832, "step": 22 }, { "epoch": 0.005304504504504505, "grad_norm": 588.8832397460938, "learning_rate": 1.1500000000000002e-06, "loss": 15.329338073730469, "loss_alignment": 1.2861328125, "loss_alignment_w": 0.64306640625, "loss_sft": 0.3151087835431099, "loss_total_v6": 0.9580836221575737, "step": 23 }, { "epoch": 0.005535135135135135, "grad_norm": 548.64013671875, "learning_rate": 1.2000000000000002e-06, "loss": 15.263714790344238, "loss_alignment": 1.2919921875, "loss_alignment_w": 0.64599609375, "loss_sft": 0.306612778455019, "loss_total_v6": 0.9539821818470955, "step": 24 }, { "epoch": 0.005765765765765766, "grad_norm": 505.54644775390625, "learning_rate": 1.25e-06, "loss": 15.399370193481445, "loss_alignment": 1.3505859375, "loss_alignment_w": 0.67529296875, "loss_sft": 0.28735072538256645, "loss_total_v6": 0.9624606072902679, "step": 25 }, { "epoch": 0.005996396396396397, "grad_norm": 499.8928527832031, "learning_rate": 1.3e-06, "loss": 14.861554145812988, "loss_alignment": 1.2626953125, "loss_alignment_w": 0.63134765625, "loss_sft": 0.2978046201169491, "loss_total_v6": 0.92884711176157, "step": 26 }, { "epoch": 0.006227027027027027, "grad_norm": 481.2905578613281, "learning_rate": 1.3500000000000002e-06, "loss": 14.165836334228516, "loss_alignment": 1.2177734375, "loss_alignment_w": 0.60888671875, "loss_sft": 0.27425032295286655, "loss_total_v6": 0.8853647857904434, "step": 27 }, { "epoch": 0.006457657657657658, "grad_norm": 450.7752990722656, "learning_rate": 1.4000000000000001e-06, "loss": 14.190394401550293, "loss_alignment": 1.189453125, "loss_alignment_w": 0.5947265625, "loss_sft": 0.2896096520125866, "loss_total_v6": 0.8868996873497963, "step": 28 }, { "epoch": 0.006688288288288288, "grad_norm": 454.749267578125, "learning_rate": 1.45e-06, "loss": 14.550434112548828, "loss_alignment": 1.2021484375, "loss_alignment_w": 0.60107421875, "loss_sft": 0.3060695733875036, "loss_total_v6": 0.90940210968256, "step": 29 }, { "epoch": 0.006918918918918919, "grad_norm": 437.4634704589844, "learning_rate": 1.5e-06, "loss": 14.352893829345703, "loss_alignment": 1.2275390625, "loss_alignment_w": 0.61376953125, "loss_sft": 0.28252343088388443, "loss_total_v6": 0.8970558643341064, "step": 30 }, { "epoch": 0.00714954954954955, "grad_norm": 384.991943359375, "learning_rate": 1.5500000000000002e-06, "loss": 14.228824615478516, "loss_alignment": 1.2216796875, "loss_alignment_w": 0.61083984375, "loss_sft": 0.2779123969376087, "loss_total_v6": 0.8893015682697296, "step": 31 }, { "epoch": 0.00738018018018018, "grad_norm": 404.8675842285156, "learning_rate": 1.6000000000000001e-06, "loss": 14.453071594238281, "loss_alignment": 1.2255859375, "loss_alignment_w": 0.61279296875, "loss_sft": 0.2905239500105381, "loss_total_v6": 0.9033169448375702, "step": 32 }, { "epoch": 0.007610810810810811, "grad_norm": 383.85406494140625, "learning_rate": 1.6500000000000003e-06, "loss": 14.254920959472656, "loss_alignment": 1.205078125, "loss_alignment_w": 0.6025390625, "loss_sft": 0.2863793447613716, "loss_total_v6": 0.8909325674176216, "step": 33 }, { "epoch": 0.007841441441441442, "grad_norm": 370.85400390625, "learning_rate": 1.7000000000000002e-06, "loss": 14.43237018585205, "loss_alignment": 1.24609375, "loss_alignment_w": 0.623046875, "loss_sft": 0.2798612844198942, "loss_total_v6": 0.9020231366157532, "step": 34 }, { "epoch": 0.008072072072072072, "grad_norm": 344.21258544921875, "learning_rate": 1.75e-06, "loss": 14.519047737121582, "loss_alignment": 1.2626953125, "loss_alignment_w": 0.63134765625, "loss_sft": 0.2759707495570183, "loss_total_v6": 0.9074404612183571, "step": 35 }, { "epoch": 0.008302702702702702, "grad_norm": 336.1675109863281, "learning_rate": 1.8000000000000001e-06, "loss": 14.061067581176758, "loss_alignment": 1.2099609375, "loss_alignment_w": 0.60498046875, "loss_sft": 0.2733479682356119, "loss_total_v6": 0.8788167089223862, "step": 36 }, { "epoch": 0.008533333333333334, "grad_norm": 307.4387512207031, "learning_rate": 1.85e-06, "loss": 14.767568588256836, "loss_alignment": 1.2314453125, "loss_alignment_w": 0.61572265625, "loss_sft": 0.30810482054948807, "loss_total_v6": 0.9229730144143105, "step": 37 }, { "epoch": 0.008763963963963964, "grad_norm": 309.9416198730469, "learning_rate": 1.9000000000000002e-06, "loss": 14.85086441040039, "loss_alignment": 1.271484375, "loss_alignment_w": 0.6357421875, "loss_sft": 0.29365747421979904, "loss_total_v6": 0.9281789734959602, "step": 38 }, { "epoch": 0.008994594594594594, "grad_norm": 323.5921936035156, "learning_rate": 1.9500000000000004e-06, "loss": 14.562665939331055, "loss_alignment": 1.22265625, "loss_alignment_w": 0.611328125, "loss_sft": 0.2982892207801342, "loss_total_v6": 0.9101666435599327, "step": 39 }, { "epoch": 0.009225225225225226, "grad_norm": 297.3161315917969, "learning_rate": 2.0000000000000003e-06, "loss": 14.785842895507812, "loss_alignment": 1.2529296875, "loss_alignment_w": 0.62646484375, "loss_sft": 0.29557510651648045, "loss_total_v6": 0.9241151586174965, "step": 40 }, { "epoch": 0.009455855855855856, "grad_norm": 289.91064453125, "learning_rate": 2.05e-06, "loss": 13.859284400939941, "loss_alignment": 1.2119140625, "loss_alignment_w": 0.60595703125, "loss_sft": 0.2606755197048187, "loss_total_v6": 0.8662052974104881, "step": 41 }, { "epoch": 0.009686486486486486, "grad_norm": 261.9313049316406, "learning_rate": 2.1000000000000002e-06, "loss": 14.620847702026367, "loss_alignment": 1.2060546875, "loss_alignment_w": 0.60302734375, "loss_sft": 0.31068412587046623, "loss_total_v6": 0.9138029962778091, "step": 42 }, { "epoch": 0.009917117117117118, "grad_norm": 267.0162048339844, "learning_rate": 2.15e-06, "loss": 14.193038940429688, "loss_alignment": 1.2197265625, "loss_alignment_w": 0.60986328125, "loss_sft": 0.27573682367801666, "loss_total_v6": 0.8870649859309196, "step": 43 }, { "epoch": 0.010147747747747748, "grad_norm": 255.0847625732422, "learning_rate": 2.2e-06, "loss": 14.18613052368164, "loss_alignment": 1.20703125, "loss_alignment_w": 0.603515625, "loss_sft": 0.28372787311673164, "loss_total_v6": 0.8866331502795219, "step": 44 }, { "epoch": 0.010378378378378378, "grad_norm": 269.2639465332031, "learning_rate": 2.25e-06, "loss": 14.068443298339844, "loss_alignment": 1.2138671875, "loss_alignment_w": 0.60693359375, "loss_sft": 0.2732901591807604, "loss_total_v6": 0.8792776837944984, "step": 45 }, { "epoch": 0.01060900900900901, "grad_norm": 267.8614501953125, "learning_rate": 2.3000000000000004e-06, "loss": 14.195666313171387, "loss_alignment": 1.232421875, "loss_alignment_w": 0.6162109375, "loss_sft": 0.27040788903832436, "loss_total_v6": 0.8872291892766953, "step": 46 }, { "epoch": 0.01083963963963964, "grad_norm": 264.6114501953125, "learning_rate": 2.35e-06, "loss": 14.98655891418457, "loss_alignment": 1.234375, "loss_alignment_w": 0.6171875, "loss_sft": 0.3182211723178625, "loss_total_v6": 0.9366598948836327, "step": 47 }, { "epoch": 0.01107027027027027, "grad_norm": 259.51531982421875, "learning_rate": 2.4000000000000003e-06, "loss": 14.442510604858398, "loss_alignment": 1.2275390625, "loss_alignment_w": 0.61376953125, "loss_sft": 0.28833805583417416, "loss_total_v6": 0.9026568979024887, "step": 48 }, { "epoch": 0.011300900900900902, "grad_norm": 243.80714416503906, "learning_rate": 2.4500000000000003e-06, "loss": 13.696667671203613, "loss_alignment": 1.2265625, "loss_alignment_w": 0.61328125, "loss_sft": 0.24248583987355232, "loss_total_v6": 0.8560417667031288, "step": 49 }, { "epoch": 0.011531531531531532, "grad_norm": 243.82412719726562, "learning_rate": 2.5e-06, "loss": 13.829612731933594, "loss_alignment": 1.2041015625, "loss_alignment_w": 0.60205078125, "loss_sft": 0.2616286724805832, "loss_total_v6": 0.8643508180975914, "step": 50 }, { "epoch": 0.011762162162162162, "grad_norm": 235.83316040039062, "learning_rate": 2.55e-06, "loss": 13.486672401428223, "loss_alignment": 1.201171875, "loss_alignment_w": 0.6005859375, "loss_sft": 0.24336870945990086, "loss_total_v6": 0.8429170250892639, "step": 51 }, { "epoch": 0.011992792792792794, "grad_norm": 247.17587280273438, "learning_rate": 2.6e-06, "loss": 14.985547065734863, "loss_alignment": 1.2646484375, "loss_alignment_w": 0.63232421875, "loss_sft": 0.30369265191257, "loss_total_v6": 0.9365966990590096, "step": 52 }, { "epoch": 0.012223423423423424, "grad_norm": 240.42030334472656, "learning_rate": 2.6500000000000005e-06, "loss": 14.617449760437012, "loss_alignment": 1.23828125, "loss_alignment_w": 0.619140625, "loss_sft": 0.2922832425683737, "loss_total_v6": 0.9135906472802162, "step": 53 }, { "epoch": 0.012454054054054054, "grad_norm": 236.35923767089844, "learning_rate": 2.7000000000000004e-06, "loss": 13.872835159301758, "loss_alignment": 1.205078125, "loss_alignment_w": 0.6025390625, "loss_sft": 0.2668019514530897, "loss_total_v6": 0.8670521825551987, "step": 54 }, { "epoch": 0.012684684684684684, "grad_norm": 232.87648010253906, "learning_rate": 2.7500000000000004e-06, "loss": 14.127612113952637, "loss_alignment": 1.22265625, "loss_alignment_w": 0.611328125, "loss_sft": 0.2716171145439148, "loss_total_v6": 0.8829757645726204, "step": 55 }, { "epoch": 0.012915315315315316, "grad_norm": 226.14544677734375, "learning_rate": 2.8000000000000003e-06, "loss": 13.79443359375, "loss_alignment": 1.1826171875, "loss_alignment_w": 0.59130859375, "loss_sft": 0.2709045261144638, "loss_total_v6": 0.8621520698070526, "step": 56 }, { "epoch": 0.013145945945945946, "grad_norm": 220.85328674316406, "learning_rate": 2.85e-06, "loss": 13.899883270263672, "loss_alignment": 1.1708984375, "loss_alignment_w": 0.58544921875, "loss_sft": 0.2831714116036892, "loss_total_v6": 0.8687427043914795, "step": 57 }, { "epoch": 0.013376576576576576, "grad_norm": 217.01284790039062, "learning_rate": 2.9e-06, "loss": 13.438658714294434, "loss_alignment": 1.1904296875, "loss_alignment_w": 0.59521484375, "loss_sft": 0.24454874731600285, "loss_total_v6": 0.8399161919951439, "step": 58 }, { "epoch": 0.013607207207207208, "grad_norm": 218.3524169921875, "learning_rate": 2.95e-06, "loss": 14.624217987060547, "loss_alignment": 1.1982421875, "loss_alignment_w": 0.59912109375, "loss_sft": 0.3158080503344536, "loss_total_v6": 0.9140136316418648, "step": 59 }, { "epoch": 0.013837837837837838, "grad_norm": 232.7371063232422, "learning_rate": 3e-06, "loss": 14.12376880645752, "loss_alignment": 1.2216796875, "loss_alignment_w": 0.61083984375, "loss_sft": 0.2718346621841192, "loss_total_v6": 0.8827355355024338, "step": 60 }, { "epoch": 0.014068468468468468, "grad_norm": 211.91217041015625, "learning_rate": 3.05e-06, "loss": 13.810466766357422, "loss_alignment": 1.21875, "loss_alignment_w": 0.609375, "loss_sft": 0.25457261875271797, "loss_total_v6": 0.8631541952490807, "step": 61 }, { "epoch": 0.0142990990990991, "grad_norm": 198.495361328125, "learning_rate": 3.1000000000000004e-06, "loss": 14.260335922241211, "loss_alignment": 1.1943359375, "loss_alignment_w": 0.59716796875, "loss_sft": 0.29489648155868053, "loss_total_v6": 0.8912709951400757, "step": 62 }, { "epoch": 0.01452972972972973, "grad_norm": 188.2116241455078, "learning_rate": 3.1500000000000003e-06, "loss": 13.509490013122559, "loss_alignment": 1.1962890625, "loss_alignment_w": 0.59814453125, "loss_sft": 0.2463817410171032, "loss_total_v6": 0.8443431556224823, "step": 63 }, { "epoch": 0.01476036036036036, "grad_norm": 191.81105041503906, "learning_rate": 3.2000000000000003e-06, "loss": 13.978200912475586, "loss_alignment": 1.201171875, "loss_alignment_w": 0.6005859375, "loss_sft": 0.27149525098502636, "loss_total_v6": 0.8736375644803047, "step": 64 }, { "epoch": 0.014990990990990992, "grad_norm": 195.96351623535156, "learning_rate": 3.2500000000000002e-06, "loss": 14.43725872039795, "loss_alignment": 1.208984375, "loss_alignment_w": 0.6044921875, "loss_sft": 0.29588338546454906, "loss_total_v6": 0.902328684926033, "step": 65 }, { "epoch": 0.015221621621621622, "grad_norm": 190.3626708984375, "learning_rate": 3.3000000000000006e-06, "loss": 13.658101081848145, "loss_alignment": 1.1669921875, "loss_alignment_w": 0.58349609375, "loss_sft": 0.27034879475831985, "loss_total_v6": 0.8536312729120255, "step": 66 }, { "epoch": 0.015452252252252252, "grad_norm": 177.4860382080078, "learning_rate": 3.3500000000000005e-06, "loss": 13.687713623046875, "loss_alignment": 1.19140625, "loss_alignment_w": 0.595703125, "loss_sft": 0.2602366991341114, "loss_total_v6": 0.8554820790886879, "step": 67 }, { "epoch": 0.015682882882882884, "grad_norm": 181.04669189453125, "learning_rate": 3.4000000000000005e-06, "loss": 13.815826416015625, "loss_alignment": 1.1416015625, "loss_alignment_w": 0.57080078125, "loss_sft": 0.29262734577059746, "loss_total_v6": 0.8634891584515572, "step": 68 }, { "epoch": 0.015913513513513514, "grad_norm": 175.1885223388672, "learning_rate": 3.45e-06, "loss": 13.739995956420898, "loss_alignment": 1.2197265625, "loss_alignment_w": 0.60986328125, "loss_sft": 0.24855076149106026, "loss_total_v6": 0.8587497994303703, "step": 69 }, { "epoch": 0.016144144144144144, "grad_norm": 179.39085388183594, "learning_rate": 3.5e-06, "loss": 13.883562088012695, "loss_alignment": 1.1806640625, "loss_alignment_w": 0.59033203125, "loss_sft": 0.27867237851023674, "loss_total_v6": 0.8677226528525352, "step": 70 }, { "epoch": 0.016374774774774774, "grad_norm": 180.44778442382812, "learning_rate": 3.5500000000000003e-06, "loss": 14.081855773925781, "loss_alignment": 1.17578125, "loss_alignment_w": 0.587890625, "loss_sft": 0.28999752178788185, "loss_total_v6": 0.8801159635186195, "step": 71 }, { "epoch": 0.016605405405405404, "grad_norm": 167.62533569335938, "learning_rate": 3.6000000000000003e-06, "loss": 13.40809154510498, "loss_alignment": 1.173828125, "loss_alignment_w": 0.5869140625, "loss_sft": 0.2503897622227669, "loss_total_v6": 0.8380057290196419, "step": 72 }, { "epoch": 0.016836036036036037, "grad_norm": 166.6565704345703, "learning_rate": 3.65e-06, "loss": 13.494349479675293, "loss_alignment": 1.17578125, "loss_alignment_w": 0.587890625, "loss_sft": 0.2561470456421375, "loss_total_v6": 0.8433968275785446, "step": 73 }, { "epoch": 0.017066666666666667, "grad_norm": 167.3935089111328, "learning_rate": 3.7e-06, "loss": 14.308683395385742, "loss_alignment": 1.212890625, "loss_alignment_w": 0.6064453125, "loss_sft": 0.28693187423050404, "loss_total_v6": 0.8942927122116089, "step": 74 }, { "epoch": 0.017297297297297298, "grad_norm": 151.84249877929688, "learning_rate": 3.7500000000000005e-06, "loss": 13.561010360717773, "loss_alignment": 1.1484375, "loss_alignment_w": 0.57421875, "loss_sft": 0.27224575728178024, "loss_total_v6": 0.8475631326436996, "step": 75 }, { "epoch": 0.017527927927927928, "grad_norm": 162.82933044433594, "learning_rate": 3.8000000000000005e-06, "loss": 13.091299057006836, "loss_alignment": 1.1337890625, "loss_alignment_w": 0.56689453125, "loss_sft": 0.25173887237906456, "loss_total_v6": 0.8182061687111855, "step": 76 }, { "epoch": 0.017758558558558558, "grad_norm": 166.96774291992188, "learning_rate": 3.85e-06, "loss": 14.48149299621582, "loss_alignment": 1.224609375, "loss_alignment_w": 0.6123046875, "loss_sft": 0.29223934933543205, "loss_total_v6": 0.905093364417553, "step": 77 }, { "epoch": 0.017989189189189188, "grad_norm": 149.51303100585938, "learning_rate": 3.900000000000001e-06, "loss": 13.495660781860352, "loss_alignment": 1.1982421875, "loss_alignment_w": 0.59912109375, "loss_sft": 0.24338115006685257, "loss_total_v6": 0.843478798866272, "step": 78 }, { "epoch": 0.01821981981981982, "grad_norm": 151.79576110839844, "learning_rate": 3.95e-06, "loss": 14.008453369140625, "loss_alignment": 1.205078125, "loss_alignment_w": 0.6025390625, "loss_sft": 0.27257731556892395, "loss_total_v6": 0.8755283132195473, "step": 79 }, { "epoch": 0.01845045045045045, "grad_norm": 151.3647003173828, "learning_rate": 4.000000000000001e-06, "loss": 13.735309600830078, "loss_alignment": 1.185546875, "loss_alignment_w": 0.5927734375, "loss_sft": 0.26788063906133175, "loss_total_v6": 0.8584568351507187, "step": 80 }, { "epoch": 0.01868108108108108, "grad_norm": 149.06997680664062, "learning_rate": 4.05e-06, "loss": 14.117412567138672, "loss_alignment": 1.177734375, "loss_alignment_w": 0.5888671875, "loss_sft": 0.2946308050304651, "loss_total_v6": 0.8823382779955864, "step": 81 }, { "epoch": 0.01891171171171171, "grad_norm": 148.46621704101562, "learning_rate": 4.1e-06, "loss": 14.042402267456055, "loss_alignment": 1.1884765625, "loss_alignment_w": 0.59423828125, "loss_sft": 0.2851819097995758, "loss_total_v6": 0.8776501640677452, "step": 82 }, { "epoch": 0.01914234234234234, "grad_norm": 143.84884643554688, "learning_rate": 4.15e-06, "loss": 13.987787246704102, "loss_alignment": 1.16015625, "loss_alignment_w": 0.580078125, "loss_sft": 0.2960202097892761, "loss_total_v6": 0.8742367327213287, "step": 83 }, { "epoch": 0.019372972972972972, "grad_norm": 142.75074768066406, "learning_rate": 4.2000000000000004e-06, "loss": 14.13957405090332, "loss_alignment": 1.1845703125, "loss_alignment_w": 0.59228515625, "loss_sft": 0.29088892228901386, "loss_total_v6": 0.8837234079837799, "step": 84 }, { "epoch": 0.019603603603603605, "grad_norm": 140.70272827148438, "learning_rate": 4.25e-06, "loss": 13.725725173950195, "loss_alignment": 1.1767578125, "loss_alignment_w": 0.58837890625, "loss_sft": 0.2720424458384514, "loss_total_v6": 0.8578578606247902, "step": 85 }, { "epoch": 0.019834234234234235, "grad_norm": 143.4513702392578, "learning_rate": 4.3e-06, "loss": 14.114599227905273, "loss_alignment": 1.1884765625, "loss_alignment_w": 0.59423828125, "loss_sft": 0.28954157046973705, "loss_total_v6": 0.8821624517440796, "step": 86 }, { "epoch": 0.020064864864864865, "grad_norm": 133.37994384765625, "learning_rate": 4.350000000000001e-06, "loss": 13.487284660339355, "loss_alignment": 1.15234375, "loss_alignment_w": 0.576171875, "loss_sft": 0.26615781895816326, "loss_total_v6": 0.8429552838206291, "step": 87 }, { "epoch": 0.020295495495495496, "grad_norm": 127.0411148071289, "learning_rate": 4.4e-06, "loss": 13.314414978027344, "loss_alignment": 1.1396484375, "loss_alignment_w": 0.56982421875, "loss_sft": 0.2619299590587616, "loss_total_v6": 0.8321509212255478, "step": 88 }, { "epoch": 0.020526126126126126, "grad_norm": 117.92483520507812, "learning_rate": 4.450000000000001e-06, "loss": 13.477127075195312, "loss_alignment": 1.1455078125, "loss_alignment_w": 0.57275390625, "loss_sft": 0.2715196833014488, "loss_total_v6": 0.8423204496502876, "step": 89 }, { "epoch": 0.020756756756756756, "grad_norm": 124.33360290527344, "learning_rate": 4.5e-06, "loss": 13.40951156616211, "loss_alignment": 1.1708984375, "loss_alignment_w": 0.58544921875, "loss_sft": 0.2518823593854904, "loss_total_v6": 0.8380945175886154, "step": 90 }, { "epoch": 0.020987387387387386, "grad_norm": 124.25833892822266, "learning_rate": 4.5500000000000005e-06, "loss": 13.833200454711914, "loss_alignment": 1.171875, "loss_alignment_w": 0.5859375, "loss_sft": 0.2773558124899864, "loss_total_v6": 0.8645750507712364, "step": 91 }, { "epoch": 0.02121801801801802, "grad_norm": 121.18716430664062, "learning_rate": 4.600000000000001e-06, "loss": 13.402933120727539, "loss_alignment": 1.1669921875, "loss_alignment_w": 0.58349609375, "loss_sft": 0.25418717600405216, "loss_total_v6": 0.8376833349466324, "step": 92 }, { "epoch": 0.02144864864864865, "grad_norm": 116.82134246826172, "learning_rate": 4.65e-06, "loss": 13.651426315307617, "loss_alignment": 1.1572265625, "loss_alignment_w": 0.57861328125, "loss_sft": 0.2744787894189358, "loss_total_v6": 0.8532141596078873, "step": 93 }, { "epoch": 0.02167927927927928, "grad_norm": 120.26747131347656, "learning_rate": 4.7e-06, "loss": 13.800682067871094, "loss_alignment": 1.138671875, "loss_alignment_w": 0.5693359375, "loss_sft": 0.2911314629018307, "loss_total_v6": 0.8625426217913628, "step": 94 }, { "epoch": 0.02190990990990991, "grad_norm": 112.33533477783203, "learning_rate": 4.75e-06, "loss": 13.906843185424805, "loss_alignment": 1.1357421875, "loss_alignment_w": 0.56787109375, "loss_sft": 0.3004521206021309, "loss_total_v6": 0.8691776990890503, "step": 95 }, { "epoch": 0.02214054054054054, "grad_norm": 118.32575225830078, "learning_rate": 4.800000000000001e-06, "loss": 13.649503707885742, "loss_alignment": 1.150390625, "loss_alignment_w": 0.5751953125, "loss_sft": 0.27521317079663277, "loss_total_v6": 0.8530940189957619, "step": 96 }, { "epoch": 0.02237117117117117, "grad_norm": 108.7224349975586, "learning_rate": 4.85e-06, "loss": 13.355680465698242, "loss_alignment": 1.111328125, "loss_alignment_w": 0.5556640625, "loss_sft": 0.2797220926731825, "loss_total_v6": 0.8347299993038177, "step": 97 }, { "epoch": 0.022601801801801803, "grad_norm": 110.14176940917969, "learning_rate": 4.9000000000000005e-06, "loss": 12.985860824584961, "loss_alignment": 1.1494140625, "loss_alignment_w": 0.57470703125, "loss_sft": 0.23761123977601528, "loss_total_v6": 0.8116163238883018, "step": 98 }, { "epoch": 0.022832432432432433, "grad_norm": 102.81851959228516, "learning_rate": 4.95e-06, "loss": 13.807252883911133, "loss_alignment": 1.1357421875, "loss_alignment_w": 0.56787109375, "loss_sft": 0.29276292584836483, "loss_total_v6": 0.8629533350467682, "step": 99 }, { "epoch": 0.023063063063063063, "grad_norm": 105.96465301513672, "learning_rate": 5e-06, "loss": 14.051258087158203, "loss_alignment": 1.12109375, "loss_alignment_w": 0.560546875, "loss_sft": 0.31605456210672855, "loss_total_v6": 0.8782036378979683, "step": 100 }, { "epoch": 0.023293693693693693, "grad_norm": 99.23560333251953, "learning_rate": 5.050000000000001e-06, "loss": 13.741861343383789, "loss_alignment": 1.140625, "loss_alignment_w": 0.5703125, "loss_sft": 0.28663117811083794, "loss_total_v6": 0.8588663190603256, "step": 101 }, { "epoch": 0.023524324324324324, "grad_norm": 97.75009155273438, "learning_rate": 5.1e-06, "loss": 13.66524887084961, "loss_alignment": 1.1318359375, "loss_alignment_w": 0.56591796875, "loss_sft": 0.28816012665629387, "loss_total_v6": 0.85407804697752, "step": 102 }, { "epoch": 0.023754954954954954, "grad_norm": 98.05671691894531, "learning_rate": 5.150000000000001e-06, "loss": 13.1329345703125, "loss_alignment": 1.134765625, "loss_alignment_w": 0.5673828125, "loss_sft": 0.25385285541415215, "loss_total_v6": 0.8208084255456924, "step": 103 }, { "epoch": 0.023985585585585587, "grad_norm": 96.9218978881836, "learning_rate": 5.2e-06, "loss": 13.656241416931152, "loss_alignment": 1.1259765625, "loss_alignment_w": 0.56298828125, "loss_sft": 0.29157968424260616, "loss_total_v6": 0.8535151109099388, "step": 104 }, { "epoch": 0.024216216216216217, "grad_norm": 95.64869689941406, "learning_rate": 5.2500000000000006e-06, "loss": 13.272480010986328, "loss_alignment": 1.140625, "loss_alignment_w": 0.5703125, "loss_sft": 0.25821044109761715, "loss_total_v6": 0.8295300081372261, "step": 105 }, { "epoch": 0.024446846846846847, "grad_norm": 96.72772216796875, "learning_rate": 5.300000000000001e-06, "loss": 13.157745361328125, "loss_alignment": 1.1474609375, "loss_alignment_w": 0.57373046875, "loss_sft": 0.24865915067493916, "loss_total_v6": 0.8223591074347496, "step": 106 }, { "epoch": 0.024677477477477477, "grad_norm": 95.51960754394531, "learning_rate": 5.3500000000000004e-06, "loss": 13.315793991088867, "loss_alignment": 1.1142578125, "loss_alignment_w": 0.55712890625, "loss_sft": 0.2763746567070484, "loss_total_v6": 0.8322370797395706, "step": 107 }, { "epoch": 0.024908108108108108, "grad_norm": 88.46495056152344, "learning_rate": 5.400000000000001e-06, "loss": 13.481379508972168, "loss_alignment": 1.1220703125, "loss_alignment_w": 0.56103515625, "loss_sft": 0.28251235745847225, "loss_total_v6": 0.8425862416625023, "step": 108 }, { "epoch": 0.025138738738738738, "grad_norm": 87.78988647460938, "learning_rate": 5.450000000000001e-06, "loss": 13.314220428466797, "loss_alignment": 1.134765625, "loss_alignment_w": 0.5673828125, "loss_sft": 0.26484758220613003, "loss_total_v6": 0.8321388363838196, "step": 109 }, { "epoch": 0.025369369369369368, "grad_norm": 92.02143096923828, "learning_rate": 5.500000000000001e-06, "loss": 13.479331016540527, "loss_alignment": 1.1435546875, "loss_alignment_w": 0.57177734375, "loss_sft": 0.2717489767819643, "loss_total_v6": 0.8424581959843636, "step": 110 }, { "epoch": 0.0256, "grad_norm": 89.8924331665039, "learning_rate": 5.550000000000001e-06, "loss": 13.509444236755371, "loss_alignment": 1.1015625, "loss_alignment_w": 0.55078125, "loss_sft": 0.2940320521593094, "loss_total_v6": 0.8443402945995331, "step": 111 }, { "epoch": 0.02583063063063063, "grad_norm": 87.49028778076172, "learning_rate": 5.600000000000001e-06, "loss": 13.377408027648926, "loss_alignment": 1.138671875, "loss_alignment_w": 0.5693359375, "loss_sft": 0.26676733046770096, "loss_total_v6": 0.8360880389809608, "step": 112 }, { "epoch": 0.02606126126126126, "grad_norm": 86.9133071899414, "learning_rate": 5.65e-06, "loss": 13.17496109008789, "loss_alignment": 1.1005859375, "loss_alignment_w": 0.55029296875, "loss_sft": 0.27427130937576294, "loss_total_v6": 0.8234350830316544, "step": 113 }, { "epoch": 0.02629189189189189, "grad_norm": 85.0246810913086, "learning_rate": 5.7e-06, "loss": 12.727971076965332, "loss_alignment": 1.1328125, "loss_alignment_w": 0.56640625, "loss_sft": 0.23028214275836945, "loss_total_v6": 0.7954981848597527, "step": 114 }, { "epoch": 0.02652252252252252, "grad_norm": 83.61204528808594, "learning_rate": 5.75e-06, "loss": 13.73523235321045, "loss_alignment": 1.1171875, "loss_alignment_w": 0.55859375, "loss_sft": 0.2999498061835766, "loss_total_v6": 0.8584520295262337, "step": 115 }, { "epoch": 0.02675315315315315, "grad_norm": 80.1097640991211, "learning_rate": 5.8e-06, "loss": 12.906314849853516, "loss_alignment": 1.1064453125, "loss_alignment_w": 0.55322265625, "loss_sft": 0.2547190170735121, "loss_total_v6": 0.8066446855664253, "step": 116 }, { "epoch": 0.026983783783783785, "grad_norm": 83.9259033203125, "learning_rate": 5.85e-06, "loss": 12.783424377441406, "loss_alignment": 1.0947265625, "loss_alignment_w": 0.54736328125, "loss_sft": 0.2528977580368519, "loss_total_v6": 0.7989640310406685, "step": 117 }, { "epoch": 0.027214414414414415, "grad_norm": 84.3042984008789, "learning_rate": 5.9e-06, "loss": 13.320978164672852, "loss_alignment": 1.13671875, "loss_alignment_w": 0.568359375, "loss_sft": 0.2653156351298094, "loss_total_v6": 0.8325611129403114, "step": 118 }, { "epoch": 0.027445045045045045, "grad_norm": 82.20767974853516, "learning_rate": 5.950000000000001e-06, "loss": 13.106669425964355, "loss_alignment": 1.130859375, "loss_alignment_w": 0.5654296875, "loss_sft": 0.25315733440220356, "loss_total_v6": 0.8191668316721916, "step": 119 }, { "epoch": 0.027675675675675675, "grad_norm": 78.10233306884766, "learning_rate": 6e-06, "loss": 13.150980949401855, "loss_alignment": 1.123046875, "loss_alignment_w": 0.5615234375, "loss_sft": 0.25923801213502884, "loss_total_v6": 0.8219363540410995, "step": 120 }, { "epoch": 0.027906306306306305, "grad_norm": 76.82672119140625, "learning_rate": 6.0500000000000005e-06, "loss": 12.797612190246582, "loss_alignment": 1.1181640625, "loss_alignment_w": 0.55908203125, "loss_sft": 0.24150113947689533, "loss_total_v6": 0.7998507544398308, "step": 121 }, { "epoch": 0.028136936936936936, "grad_norm": 73.5177001953125, "learning_rate": 6.1e-06, "loss": 13.624311447143555, "loss_alignment": 1.1123046875, "loss_alignment_w": 0.55615234375, "loss_sft": 0.29558076336979866, "loss_total_v6": 0.8515194579958916, "step": 122 }, { "epoch": 0.02836756756756757, "grad_norm": 73.36830139160156, "learning_rate": 6.15e-06, "loss": 13.009435653686523, "loss_alignment": 1.1123046875, "loss_alignment_w": 0.55615234375, "loss_sft": 0.25771559216082096, "loss_total_v6": 0.8130897060036659, "step": 123 }, { "epoch": 0.0285981981981982, "grad_norm": 75.23326873779297, "learning_rate": 6.200000000000001e-06, "loss": 13.920466423034668, "loss_alignment": 1.125, "loss_alignment_w": 0.5625, "loss_sft": 0.30699508264660835, "loss_total_v6": 0.8700291514396667, "step": 124 }, { "epoch": 0.02882882882882883, "grad_norm": 72.1257095336914, "learning_rate": 6.25e-06, "loss": 12.943717956542969, "loss_alignment": 1.1171875, "loss_alignment_w": 0.55859375, "loss_sft": 0.25029708072543144, "loss_total_v6": 0.8089823797345161, "step": 125 }, { "epoch": 0.02905945945945946, "grad_norm": 73.22750854492188, "learning_rate": 6.300000000000001e-06, "loss": 13.39422607421875, "loss_alignment": 1.1142578125, "loss_alignment_w": 0.55712890625, "loss_sft": 0.28031538613140583, "loss_total_v6": 0.8371390923857689, "step": 126 }, { "epoch": 0.02929009009009009, "grad_norm": 71.08984375, "learning_rate": 6.35e-06, "loss": 13.257046699523926, "loss_alignment": 1.142578125, "loss_alignment_w": 0.5712890625, "loss_sft": 0.25779518112540245, "loss_total_v6": 0.8285654485225677, "step": 127 }, { "epoch": 0.02952072072072072, "grad_norm": 71.03507995605469, "learning_rate": 6.4000000000000006e-06, "loss": 13.271125793457031, "loss_alignment": 1.1162109375, "loss_alignment_w": 0.55810546875, "loss_sft": 0.2707142271101475, "loss_total_v6": 0.8294453099370003, "step": 128 }, { "epoch": 0.029751351351351353, "grad_norm": 69.75076293945312, "learning_rate": 6.450000000000001e-06, "loss": 12.942028045654297, "loss_alignment": 1.1064453125, "loss_alignment_w": 0.55322265625, "loss_sft": 0.25382305309176445, "loss_total_v6": 0.808876745402813, "step": 129 }, { "epoch": 0.029981981981981983, "grad_norm": 64.99773406982422, "learning_rate": 6.5000000000000004e-06, "loss": 13.286184310913086, "loss_alignment": 1.115234375, "loss_alignment_w": 0.5576171875, "loss_sft": 0.27237267047166824, "loss_total_v6": 0.8303865641355515, "step": 130 }, { "epoch": 0.030212612612612613, "grad_norm": 69.16624450683594, "learning_rate": 6.550000000000001e-06, "loss": 13.426399230957031, "loss_alignment": 1.1298828125, "loss_alignment_w": 0.56494140625, "loss_sft": 0.27396444976329803, "loss_total_v6": 0.8391499742865562, "step": 131 }, { "epoch": 0.030443243243243243, "grad_norm": 64.9174575805664, "learning_rate": 6.600000000000001e-06, "loss": 13.783538818359375, "loss_alignment": 1.1337890625, "loss_alignment_w": 0.56689453125, "loss_sft": 0.29332548193633556, "loss_total_v6": 0.8614711910486221, "step": 132 }, { "epoch": 0.030673873873873873, "grad_norm": 66.26510620117188, "learning_rate": 6.650000000000001e-06, "loss": 12.831928253173828, "loss_alignment": 1.078125, "loss_alignment_w": 0.5390625, "loss_sft": 0.2630093414336443, "loss_total_v6": 0.8019955456256866, "step": 133 }, { "epoch": 0.030904504504504503, "grad_norm": 65.36932373046875, "learning_rate": 6.700000000000001e-06, "loss": 13.280662536621094, "loss_alignment": 1.1123046875, "loss_alignment_w": 0.55615234375, "loss_sft": 0.27565911784768105, "loss_total_v6": 0.8300414457917213, "step": 134 }, { "epoch": 0.031135135135135134, "grad_norm": 61.65110778808594, "learning_rate": 6.750000000000001e-06, "loss": 12.842427253723145, "loss_alignment": 1.099609375, "loss_alignment_w": 0.5498046875, "loss_sft": 0.25116862915456295, "loss_total_v6": 0.8026517480611801, "step": 135 }, { "epoch": 0.03136576576576577, "grad_norm": 67.43919372558594, "learning_rate": 6.800000000000001e-06, "loss": 12.99039077758789, "loss_alignment": 1.1103515625, "loss_alignment_w": 0.55517578125, "loss_sft": 0.2560828011482954, "loss_total_v6": 0.8118994534015656, "step": 136 }, { "epoch": 0.031596396396396394, "grad_norm": 62.54423522949219, "learning_rate": 6.850000000000001e-06, "loss": 13.621335983276367, "loss_alignment": 1.1220703125, "loss_alignment_w": 0.56103515625, "loss_sft": 0.2889861296862364, "loss_total_v6": 0.8513335511088371, "step": 137 }, { "epoch": 0.03182702702702703, "grad_norm": 60.578033447265625, "learning_rate": 6.9e-06, "loss": 12.769184112548828, "loss_alignment": 1.1103515625, "loss_alignment_w": 0.55517578125, "loss_sft": 0.24379848316311836, "loss_total_v6": 0.798074021935463, "step": 138 }, { "epoch": 0.03205765765765766, "grad_norm": 59.837623596191406, "learning_rate": 6.95e-06, "loss": 13.352354049682617, "loss_alignment": 1.08984375, "loss_alignment_w": 0.544921875, "loss_sft": 0.2889593541622162, "loss_total_v6": 0.8345221132040024, "step": 139 }, { "epoch": 0.03228828828828829, "grad_norm": 59.230690002441406, "learning_rate": 7e-06, "loss": 13.216328620910645, "loss_alignment": 1.0859375, "loss_alignment_w": 0.54296875, "loss_sft": 0.2830822691321373, "loss_total_v6": 0.8260205090045929, "step": 140 }, { "epoch": 0.03251891891891892, "grad_norm": 66.27136993408203, "learning_rate": 7.05e-06, "loss": 13.157919883728027, "loss_alignment": 1.109375, "loss_alignment_w": 0.5546875, "loss_sft": 0.26911680586636066, "loss_total_v6": 0.8223699927330017, "step": 141 }, { "epoch": 0.03274954954954955, "grad_norm": 59.802757263183594, "learning_rate": 7.100000000000001e-06, "loss": 13.10572624206543, "loss_alignment": 1.0869140625, "loss_alignment_w": 0.54345703125, "loss_sft": 0.2766121104359627, "loss_total_v6": 0.819107860326767, "step": 142 }, { "epoch": 0.03298018018018018, "grad_norm": 55.83554458618164, "learning_rate": 7.15e-06, "loss": 12.674299240112305, "loss_alignment": 1.07275390625, "loss_alignment_w": 0.536376953125, "loss_sft": 0.2548969890922308, "loss_total_v6": 0.7921437099575996, "step": 143 }, { "epoch": 0.03321081081081081, "grad_norm": 60.96759033203125, "learning_rate": 7.2000000000000005e-06, "loss": 13.645760536193848, "loss_alignment": 1.1142578125, "loss_alignment_w": 0.55712890625, "loss_sft": 0.2972722612321377, "loss_total_v6": 0.8528600484132767, "step": 144 }, { "epoch": 0.03344144144144144, "grad_norm": 60.297576904296875, "learning_rate": 7.25e-06, "loss": 13.138763427734375, "loss_alignment": 1.095703125, "loss_alignment_w": 0.5478515625, "loss_sft": 0.2725582458078861, "loss_total_v6": 0.8211727440357208, "step": 145 }, { "epoch": 0.033672072072072075, "grad_norm": 54.42205810546875, "learning_rate": 7.3e-06, "loss": 13.582732200622559, "loss_alignment": 1.1015625, "loss_alignment_w": 0.55078125, "loss_sft": 0.29902450554072857, "loss_total_v6": 0.8489207550883293, "step": 146 }, { "epoch": 0.0339027027027027, "grad_norm": 55.727134704589844, "learning_rate": 7.350000000000001e-06, "loss": 12.826959609985352, "loss_alignment": 1.107421875, "loss_alignment_w": 0.5537109375, "loss_sft": 0.24873701855540276, "loss_total_v6": 0.801685020327568, "step": 147 }, { "epoch": 0.034133333333333335, "grad_norm": 57.38017272949219, "learning_rate": 7.4e-06, "loss": 13.338040351867676, "loss_alignment": 1.1123046875, "loss_alignment_w": 0.55615234375, "loss_sft": 0.2773072961717844, "loss_total_v6": 0.8336275145411491, "step": 148 }, { "epoch": 0.03436396396396396, "grad_norm": 52.54877471923828, "learning_rate": 7.450000000000001e-06, "loss": 13.3449068069458, "loss_alignment": 1.123046875, "loss_alignment_w": 0.5615234375, "loss_sft": 0.27317408472299576, "loss_total_v6": 0.834056667983532, "step": 149 }, { "epoch": 0.034594594594594595, "grad_norm": 50.48481750488281, "learning_rate": 7.500000000000001e-06, "loss": 13.234891891479492, "loss_alignment": 1.1171875, "loss_alignment_w": 0.55859375, "loss_sft": 0.268419174477458, "loss_total_v6": 0.8271807730197906, "step": 150 }, { "epoch": 0.03482522522522523, "grad_norm": 49.906898498535156, "learning_rate": 7.5500000000000006e-06, "loss": 13.80842399597168, "loss_alignment": 1.0888671875, "loss_alignment_w": 0.54443359375, "loss_sft": 0.3185318298637867, "loss_total_v6": 0.8630264848470688, "step": 151 }, { "epoch": 0.035055855855855855, "grad_norm": 50.628421783447266, "learning_rate": 7.600000000000001e-06, "loss": 12.909797668457031, "loss_alignment": 1.1240234375, "loss_alignment_w": 0.56201171875, "loss_sft": 0.24726151674985886, "loss_total_v6": 0.8068623542785645, "step": 152 }, { "epoch": 0.03528648648648649, "grad_norm": 53.82900619506836, "learning_rate": 7.650000000000001e-06, "loss": 13.25867748260498, "loss_alignment": 1.12890625, "loss_alignment_w": 0.564453125, "loss_sft": 0.26662506349384785, "loss_total_v6": 0.8286673203110695, "step": 153 }, { "epoch": 0.035517117117117115, "grad_norm": 51.980628967285156, "learning_rate": 7.7e-06, "loss": 12.81523609161377, "loss_alignment": 1.09375, "loss_alignment_w": 0.546875, "loss_sft": 0.2542298138141632, "loss_total_v6": 0.8009522259235382, "step": 154 }, { "epoch": 0.03574774774774775, "grad_norm": 52.56926345825195, "learning_rate": 7.75e-06, "loss": 13.733903884887695, "loss_alignment": 1.115234375, "loss_alignment_w": 0.5576171875, "loss_sft": 0.2991649154573679, "loss_total_v6": 0.8583689779043198, "step": 155 }, { "epoch": 0.035978378378378376, "grad_norm": 50.7730827331543, "learning_rate": 7.800000000000002e-06, "loss": 12.890593528747559, "loss_alignment": 1.09765625, "loss_alignment_w": 0.548828125, "loss_sft": 0.2584819495677948, "loss_total_v6": 0.8056621104478836, "step": 156 }, { "epoch": 0.03620900900900901, "grad_norm": 47.27699279785156, "learning_rate": 7.850000000000001e-06, "loss": 12.952641487121582, "loss_alignment": 1.09765625, "loss_alignment_w": 0.548828125, "loss_sft": 0.2608034983277321, "loss_total_v6": 0.8095400854945183, "step": 157 }, { "epoch": 0.03643963963963964, "grad_norm": 49.21956253051758, "learning_rate": 7.9e-06, "loss": 13.268665313720703, "loss_alignment": 1.083984375, "loss_alignment_w": 0.5419921875, "loss_sft": 0.2865517232567072, "loss_total_v6": 0.8292915895581245, "step": 158 }, { "epoch": 0.03667027027027027, "grad_norm": 48.18354034423828, "learning_rate": 7.950000000000002e-06, "loss": 13.265656471252441, "loss_alignment": 1.0791015625, "loss_alignment_w": 0.53955078125, "loss_sft": 0.2899342458695173, "loss_total_v6": 0.8291035443544388, "step": 159 }, { "epoch": 0.0369009009009009, "grad_norm": 51.012447357177734, "learning_rate": 8.000000000000001e-06, "loss": 12.308396339416504, "loss_alignment": 1.09375, "loss_alignment_w": 0.546875, "loss_sft": 0.2218199260532856, "loss_total_v6": 0.7692747563123703, "step": 160 }, { "epoch": 0.03713153153153153, "grad_norm": 53.63130569458008, "learning_rate": 8.050000000000001e-06, "loss": 12.782915115356445, "loss_alignment": 1.0927734375, "loss_alignment_w": 0.54638671875, "loss_sft": 0.25339999608695507, "loss_total_v6": 0.7989322245121002, "step": 161 }, { "epoch": 0.03736216216216216, "grad_norm": 53.36442947387695, "learning_rate": 8.1e-06, "loss": 13.43191909790039, "loss_alignment": 1.09765625, "loss_alignment_w": 0.548828125, "loss_sft": 0.2908804602921009, "loss_total_v6": 0.8394949361681938, "step": 162 }, { "epoch": 0.03759279279279279, "grad_norm": 50.653648376464844, "learning_rate": 8.15e-06, "loss": 12.877388000488281, "loss_alignment": 1.072265625, "loss_alignment_w": 0.5361328125, "loss_sft": 0.2668881379067898, "loss_total_v6": 0.8048367723822594, "step": 163 }, { "epoch": 0.03782342342342342, "grad_norm": 48.06792449951172, "learning_rate": 8.2e-06, "loss": 13.005121231079102, "loss_alignment": 1.0888671875, "loss_alignment_w": 0.54443359375, "loss_sft": 0.2679591942578554, "loss_total_v6": 0.8128200322389603, "step": 164 }, { "epoch": 0.03805405405405406, "grad_norm": 48.830810546875, "learning_rate": 8.25e-06, "loss": 12.914543151855469, "loss_alignment": 1.109375, "loss_alignment_w": 0.5546875, "loss_sft": 0.25256302021443844, "loss_total_v6": 0.8071589693427086, "step": 165 }, { "epoch": 0.03828468468468468, "grad_norm": 43.801578521728516, "learning_rate": 8.3e-06, "loss": 13.15224552154541, "loss_alignment": 1.091796875, "loss_alignment_w": 0.5458984375, "loss_sft": 0.2773223351687193, "loss_total_v6": 0.8220153227448463, "step": 166 }, { "epoch": 0.03851531531531532, "grad_norm": 44.244747161865234, "learning_rate": 8.35e-06, "loss": 13.017324447631836, "loss_alignment": 1.08984375, "loss_alignment_w": 0.544921875, "loss_sft": 0.2679436933249235, "loss_total_v6": 0.8135827407240868, "step": 167 }, { "epoch": 0.038745945945945943, "grad_norm": 44.13385009765625, "learning_rate": 8.400000000000001e-06, "loss": 13.477222442626953, "loss_alignment": 1.1015625, "loss_alignment_w": 0.55078125, "loss_sft": 0.2914993315935135, "loss_total_v6": 0.8423263728618622, "step": 168 }, { "epoch": 0.03897657657657658, "grad_norm": 45.98707962036133, "learning_rate": 8.45e-06, "loss": 13.469715118408203, "loss_alignment": 1.107421875, "loss_alignment_w": 0.5537109375, "loss_sft": 0.2890007384121418, "loss_total_v6": 0.8418572023510933, "step": 169 }, { "epoch": 0.03920720720720721, "grad_norm": 44.11461639404297, "learning_rate": 8.5e-06, "loss": 13.599042892456055, "loss_alignment": 1.083984375, "loss_alignment_w": 0.5419921875, "loss_sft": 0.3072003461420536, "loss_total_v6": 0.8499402031302452, "step": 170 }, { "epoch": 0.03943783783783784, "grad_norm": 45.11695861816406, "learning_rate": 8.550000000000001e-06, "loss": 13.176594734191895, "loss_alignment": 1.08203125, "loss_alignment_w": 0.541015625, "loss_sft": 0.2820332385599613, "loss_total_v6": 0.8235371485352516, "step": 171 }, { "epoch": 0.03966846846846847, "grad_norm": 45.39731979370117, "learning_rate": 8.6e-06, "loss": 13.338546752929688, "loss_alignment": 1.0859375, "loss_alignment_w": 0.54296875, "loss_sft": 0.2900342829525471, "loss_total_v6": 0.8336591795086861, "step": 172 }, { "epoch": 0.0398990990990991, "grad_norm": 45.751182556152344, "learning_rate": 8.65e-06, "loss": 12.86566162109375, "loss_alignment": 1.08984375, "loss_alignment_w": 0.544921875, "loss_sft": 0.25915144197642803, "loss_total_v6": 0.8041038364171982, "step": 173 }, { "epoch": 0.04012972972972973, "grad_norm": 42.74627685546875, "learning_rate": 8.700000000000001e-06, "loss": 12.61233901977539, "loss_alignment": 1.0830078125, "loss_alignment_w": 0.54150390625, "loss_sft": 0.24736235849559307, "loss_total_v6": 0.7882711812853813, "step": 174 }, { "epoch": 0.04036036036036036, "grad_norm": 41.16392517089844, "learning_rate": 8.750000000000001e-06, "loss": 13.168859481811523, "loss_alignment": 1.11328125, "loss_alignment_w": 0.556640625, "loss_sft": 0.26772534288465977, "loss_total_v6": 0.8230537250638008, "step": 175 }, { "epoch": 0.04059099099099099, "grad_norm": 42.52791213989258, "learning_rate": 8.8e-06, "loss": 12.875228881835938, "loss_alignment": 1.0849609375, "loss_alignment_w": 0.54248046875, "loss_sft": 0.26331995241343975, "loss_total_v6": 0.8047018125653267, "step": 176 }, { "epoch": 0.040821621621621625, "grad_norm": 41.527565002441406, "learning_rate": 8.85e-06, "loss": 12.869843482971191, "loss_alignment": 1.064453125, "loss_alignment_w": 0.5322265625, "loss_sft": 0.27355774492025375, "loss_total_v6": 0.8043652325868607, "step": 177 }, { "epoch": 0.04105225225225225, "grad_norm": 43.043827056884766, "learning_rate": 8.900000000000001e-06, "loss": 12.606548309326172, "loss_alignment": 1.0712890625, "loss_alignment_w": 0.53564453125, "loss_sft": 0.25202060863375664, "loss_total_v6": 0.7879092693328857, "step": 178 }, { "epoch": 0.041282882882882885, "grad_norm": 42.63831329345703, "learning_rate": 8.95e-06, "loss": 13.219629287719727, "loss_alignment": 1.099609375, "loss_alignment_w": 0.5498046875, "loss_sft": 0.2764984630048275, "loss_total_v6": 0.8262268751859665, "step": 179 }, { "epoch": 0.04151351351351351, "grad_norm": 44.35379409790039, "learning_rate": 9e-06, "loss": 13.180201530456543, "loss_alignment": 1.0751953125, "loss_alignment_w": 0.53759765625, "loss_sft": 0.2873398754745722, "loss_total_v6": 0.8237626031041145, "step": 180 }, { "epoch": 0.041744144144144145, "grad_norm": 47.52521514892578, "learning_rate": 9.050000000000001e-06, "loss": 13.089662551879883, "loss_alignment": 1.1044921875, "loss_alignment_w": 0.55224609375, "loss_sft": 0.2649117782711983, "loss_total_v6": 0.8181039243936539, "step": 181 }, { "epoch": 0.04197477477477477, "grad_norm": 40.76896286010742, "learning_rate": 9.100000000000001e-06, "loss": 12.886955261230469, "loss_alignment": 1.0556640625, "loss_alignment_w": 0.52783203125, "loss_sft": 0.2784571573138237, "loss_total_v6": 0.8054346889257431, "step": 182 }, { "epoch": 0.042205405405405405, "grad_norm": 38.956302642822266, "learning_rate": 9.15e-06, "loss": 12.926885604858398, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.28545415215194225, "loss_total_v6": 0.8079303503036499, "step": 183 }, { "epoch": 0.04243603603603604, "grad_norm": 37.79641342163086, "learning_rate": 9.200000000000002e-06, "loss": 13.069202423095703, "loss_alignment": 1.068359375, "loss_alignment_w": 0.5341796875, "loss_sft": 0.2817757334560156, "loss_total_v6": 0.8168251514434814, "step": 184 }, { "epoch": 0.042666666666666665, "grad_norm": 41.599178314208984, "learning_rate": 9.250000000000001e-06, "loss": 12.685253143310547, "loss_alignment": 1.0625, "loss_alignment_w": 0.53125, "loss_sft": 0.26177673414349556, "loss_total_v6": 0.7928283214569092, "step": 185 }, { "epoch": 0.0428972972972973, "grad_norm": 56.29331970214844, "learning_rate": 9.3e-06, "loss": 12.879982948303223, "loss_alignment": 1.072265625, "loss_alignment_w": 0.5361328125, "loss_sft": 0.268606748431921, "loss_total_v6": 0.8049989491701126, "step": 186 }, { "epoch": 0.043127927927927925, "grad_norm": 41.38413619995117, "learning_rate": 9.350000000000002e-06, "loss": 12.86756420135498, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.2786795236170292, "loss_total_v6": 0.8042227551341057, "step": 187 }, { "epoch": 0.04335855855855856, "grad_norm": 37.184329986572266, "learning_rate": 9.4e-06, "loss": 12.755568504333496, "loss_alignment": 1.0673828125, "loss_alignment_w": 0.53369140625, "loss_sft": 0.26273819617927074, "loss_total_v6": 0.7972230315208435, "step": 188 }, { "epoch": 0.04358918918918919, "grad_norm": 38.613346099853516, "learning_rate": 9.450000000000001e-06, "loss": 12.525259017944336, "loss_alignment": 1.06640625, "loss_alignment_w": 0.533203125, "loss_sft": 0.2506478801369667, "loss_total_v6": 0.7828286588191986, "step": 189 }, { "epoch": 0.04381981981981982, "grad_norm": 39.6599006652832, "learning_rate": 9.5e-06, "loss": 13.070833206176758, "loss_alignment": 1.0654296875, "loss_alignment_w": 0.53271484375, "loss_sft": 0.28311359509825706, "loss_total_v6": 0.816927045583725, "step": 190 }, { "epoch": 0.04405045045045045, "grad_norm": 39.1950798034668, "learning_rate": 9.55e-06, "loss": 13.242182731628418, "loss_alignment": 1.0849609375, "loss_alignment_w": 0.54248046875, "loss_sft": 0.285476416349411, "loss_total_v6": 0.8276364356279373, "step": 191 }, { "epoch": 0.04428108108108108, "grad_norm": 38.990360260009766, "learning_rate": 9.600000000000001e-06, "loss": 12.815702438354492, "loss_alignment": 1.0595703125, "loss_alignment_w": 0.52978515625, "loss_sft": 0.2713640611618757, "loss_total_v6": 0.8009813949465752, "step": 192 }, { "epoch": 0.04451171171171171, "grad_norm": 37.78145980834961, "learning_rate": 9.65e-06, "loss": 12.611580848693848, "loss_alignment": 1.0634765625, "loss_alignment_w": 0.53173828125, "loss_sft": 0.25654657930135727, "loss_total_v6": 0.7882237955927849, "step": 193 }, { "epoch": 0.04474234234234234, "grad_norm": 37.322025299072266, "learning_rate": 9.7e-06, "loss": 12.771236419677734, "loss_alignment": 1.0693359375, "loss_alignment_w": 0.53466796875, "loss_sft": 0.26290871016681194, "loss_total_v6": 0.7982022985816002, "step": 194 }, { "epoch": 0.04497297297297297, "grad_norm": 39.750526428222656, "learning_rate": 9.75e-06, "loss": 12.976125717163086, "loss_alignment": 1.04931640625, "loss_alignment_w": 0.524658203125, "loss_sft": 0.2869142219424248, "loss_total_v6": 0.8110078498721123, "step": 195 }, { "epoch": 0.045203603603603607, "grad_norm": 37.15281295776367, "learning_rate": 9.800000000000001e-06, "loss": 13.170655250549316, "loss_alignment": 1.0673828125, "loss_alignment_w": 0.53369140625, "loss_sft": 0.2890778165310621, "loss_total_v6": 0.8231659829616547, "step": 196 }, { "epoch": 0.04543423423423423, "grad_norm": 38.61949920654297, "learning_rate": 9.85e-06, "loss": 13.118810653686523, "loss_alignment": 1.06640625, "loss_alignment_w": 0.533203125, "loss_sft": 0.28568489477038383, "loss_total_v6": 0.8199256360530853, "step": 197 }, { "epoch": 0.04566486486486487, "grad_norm": 36.575958251953125, "learning_rate": 9.9e-06, "loss": 12.547130584716797, "loss_alignment": 1.056640625, "loss_alignment_w": 0.5283203125, "loss_sft": 0.25511238537728786, "loss_total_v6": 0.7841956466436386, "step": 198 }, { "epoch": 0.04589549549549549, "grad_norm": 43.46383285522461, "learning_rate": 9.950000000000001e-06, "loss": 12.98065185546875, "loss_alignment": 1.076171875, "loss_alignment_w": 0.5380859375, "loss_sft": 0.273571016266942, "loss_total_v6": 0.8112907558679581, "step": 199 }, { "epoch": 0.04612612612612613, "grad_norm": 36.499000549316406, "learning_rate": 1e-05, "loss": 14.033528327941895, "loss_alignment": 1.087890625, "loss_alignment_w": 0.5439453125, "loss_sft": 0.3321736380457878, "loss_total_v6": 0.877095527946949, "step": 200 }, { "epoch": 0.04635675675675675, "grad_norm": 39.24445724487305, "learning_rate": 9.99999959444427e-06, "loss": 13.304422378540039, "loss_alignment": 1.0693359375, "loss_alignment_w": 0.53466796875, "loss_sft": 0.29617176204919815, "loss_total_v6": 0.8315263986587524, "step": 201 }, { "epoch": 0.04658738738738739, "grad_norm": 35.67110061645508, "learning_rate": 9.999998377777142e-06, "loss": 12.765340805053711, "loss_alignment": 1.03662109375, "loss_alignment_w": 0.518310546875, "loss_sft": 0.27911132015287876, "loss_total_v6": 0.7978338450193405, "step": 202 }, { "epoch": 0.04681801801801802, "grad_norm": 34.882225036621094, "learning_rate": 9.999996349998818e-06, "loss": 13.314420700073242, "loss_alignment": 1.0615234375, "loss_alignment_w": 0.53076171875, "loss_sft": 0.302167821675539, "loss_total_v6": 0.8321513235569, "step": 203 }, { "epoch": 0.04704864864864865, "grad_norm": 31.877431869506836, "learning_rate": 9.999993511109623e-06, "loss": 12.35051155090332, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.2479507140815258, "loss_total_v6": 0.7719069942831993, "step": 204 }, { "epoch": 0.04727927927927928, "grad_norm": 35.33152389526367, "learning_rate": 9.999989861110017e-06, "loss": 13.532279968261719, "loss_alignment": 1.0849609375, "loss_alignment_w": 0.54248046875, "loss_sft": 0.30261570401489735, "loss_total_v6": 0.8457675501704216, "step": 205 }, { "epoch": 0.04750990990990991, "grad_norm": 34.350799560546875, "learning_rate": 9.999985400000596e-06, "loss": 12.891900062561035, "loss_alignment": 1.0546875, "loss_alignment_w": 0.52734375, "loss_sft": 0.27771334536373615, "loss_total_v6": 0.8057437464594841, "step": 206 }, { "epoch": 0.04774054054054054, "grad_norm": 34.251853942871094, "learning_rate": 9.99998012778208e-06, "loss": 12.841141700744629, "loss_alignment": 1.0546875, "loss_alignment_w": 0.52734375, "loss_sft": 0.27493769116699696, "loss_total_v6": 0.8025713786482811, "step": 207 }, { "epoch": 0.047971171171171174, "grad_norm": 38.33256912231445, "learning_rate": 9.999974044455328e-06, "loss": 13.12419319152832, "loss_alignment": 1.060546875, "loss_alignment_w": 0.5302734375, "loss_sft": 0.2901870459318161, "loss_total_v6": 0.82026207447052, "step": 208 }, { "epoch": 0.0482018018018018, "grad_norm": 34.67158889770508, "learning_rate": 9.999967150021323e-06, "loss": 13.387081146240234, "loss_alignment": 1.0634765625, "loss_alignment_w": 0.53173828125, "loss_sft": 0.30594610050320625, "loss_total_v6": 0.8366925418376923, "step": 209 }, { "epoch": 0.048432432432432435, "grad_norm": 34.68839645385742, "learning_rate": 9.999959444481187e-06, "loss": 12.316947937011719, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.24609708972275257, "loss_total_v6": 0.7698092609643936, "step": 210 }, { "epoch": 0.04866306306306306, "grad_norm": 36.85696792602539, "learning_rate": 9.999950927836166e-06, "loss": 13.225957870483398, "loss_alignment": 1.0810546875, "loss_alignment_w": 0.54052734375, "loss_sft": 0.2854083813726902, "loss_total_v6": 0.8266223594546318, "step": 211 }, { "epoch": 0.048893693693693695, "grad_norm": 35.063812255859375, "learning_rate": 9.999941600087643e-06, "loss": 13.298421859741211, "loss_alignment": 1.076171875, "loss_alignment_w": 0.5380859375, "loss_sft": 0.29242455400526524, "loss_total_v6": 0.8311513662338257, "step": 212 }, { "epoch": 0.04912432432432432, "grad_norm": 33.611724853515625, "learning_rate": 9.999931461237135e-06, "loss": 12.990327835083008, "loss_alignment": 1.0830078125, "loss_alignment_w": 0.54150390625, "loss_sft": 0.2696592267602682, "loss_total_v6": 0.8118955418467522, "step": 213 }, { "epoch": 0.049354954954954955, "grad_norm": 32.95344161987305, "learning_rate": 9.999920511286282e-06, "loss": 12.751401901245117, "loss_alignment": 1.087890625, "loss_alignment_w": 0.5439453125, "loss_sft": 0.25292575918138027, "loss_total_v6": 0.7969626113772392, "step": 214 }, { "epoch": 0.04958558558558559, "grad_norm": 35.24388885498047, "learning_rate": 9.999908750236862e-06, "loss": 12.761972427368164, "loss_alignment": 1.060546875, "loss_alignment_w": 0.5302734375, "loss_sft": 0.26800602674484253, "loss_total_v6": 0.7976233139634132, "step": 215 }, { "epoch": 0.049816216216216215, "grad_norm": 32.39956283569336, "learning_rate": 9.999896178090783e-06, "loss": 13.036951065063477, "loss_alignment": 1.064453125, "loss_alignment_w": 0.5322265625, "loss_sft": 0.28206409700214863, "loss_total_v6": 0.8148094788193703, "step": 216 }, { "epoch": 0.05004684684684685, "grad_norm": 32.981082916259766, "learning_rate": 9.999882794850085e-06, "loss": 12.570440292358398, "loss_alignment": 1.060546875, "loss_alignment_w": 0.5302734375, "loss_sft": 0.2554858550429344, "loss_total_v6": 0.7856525182723999, "step": 217 }, { "epoch": 0.050277477477477475, "grad_norm": 32.144203186035156, "learning_rate": 9.999868600516938e-06, "loss": 12.537269592285156, "loss_alignment": 1.05859375, "loss_alignment_w": 0.529296875, "loss_sft": 0.2542062345892191, "loss_total_v6": 0.7835793942213058, "step": 218 }, { "epoch": 0.05050810810810811, "grad_norm": 33.36701965332031, "learning_rate": 9.999853595093645e-06, "loss": 12.951834678649902, "loss_alignment": 1.0703125, "loss_alignment_w": 0.53515625, "loss_sft": 0.2738451324403286, "loss_total_v6": 0.8094896674156189, "step": 219 }, { "epoch": 0.050738738738738735, "grad_norm": 31.848487854003906, "learning_rate": 9.999837778582641e-06, "loss": 12.680641174316406, "loss_alignment": 1.0654296875, "loss_alignment_w": 0.53271484375, "loss_sft": 0.25929114781320095, "loss_total_v6": 0.7925400659441948, "step": 220 }, { "epoch": 0.05096936936936937, "grad_norm": 38.162452697753906, "learning_rate": 9.99982115098649e-06, "loss": 12.656349182128906, "loss_alignment": 1.068359375, "loss_alignment_w": 0.5341796875, "loss_sft": 0.2564149145036936, "loss_total_v6": 0.791021890938282, "step": 221 }, { "epoch": 0.0512, "grad_norm": 33.05305099487305, "learning_rate": 9.99980371230789e-06, "loss": 12.759757995605469, "loss_alignment": 1.0546875, "loss_alignment_w": 0.52734375, "loss_sft": 0.2707209251821041, "loss_total_v6": 0.7974848523736, "step": 222 }, { "epoch": 0.05143063063063063, "grad_norm": 34.71186065673828, "learning_rate": 9.999785462549672e-06, "loss": 13.584798812866211, "loss_alignment": 1.0654296875, "loss_alignment_w": 0.53271484375, "loss_sft": 0.31560257636010647, "loss_total_v6": 0.8490498587489128, "step": 223 }, { "epoch": 0.05166126126126126, "grad_norm": 33.48963165283203, "learning_rate": 9.999766401714795e-06, "loss": 13.192294120788574, "loss_alignment": 1.0634765625, "loss_alignment_w": 0.53173828125, "loss_sft": 0.29352780245244503, "loss_total_v6": 0.8245184049010277, "step": 224 }, { "epoch": 0.05189189189189189, "grad_norm": 35.219173431396484, "learning_rate": 9.99974652980635e-06, "loss": 12.829975128173828, "loss_alignment": 1.068359375, "loss_alignment_w": 0.5341796875, "loss_sft": 0.2678921017795801, "loss_total_v6": 0.8018734529614449, "step": 225 }, { "epoch": 0.05212252252252252, "grad_norm": 32.504066467285156, "learning_rate": 9.999725846827562e-06, "loss": 12.892035484313965, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.27924771048128605, "loss_total_v6": 0.8057522103190422, "step": 226 }, { "epoch": 0.052353153153153156, "grad_norm": 35.384986877441406, "learning_rate": 9.999704352781785e-06, "loss": 12.725044250488281, "loss_alignment": 1.0703125, "loss_alignment_w": 0.53515625, "loss_sft": 0.25896890088915825, "loss_total_v6": 0.7953153178095818, "step": 227 }, { "epoch": 0.05258378378378378, "grad_norm": 29.63697052001953, "learning_rate": 9.999682047672508e-06, "loss": 12.729867935180664, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.27288115210831165, "loss_total_v6": 0.7956167459487915, "step": 228 }, { "epoch": 0.052814414414414416, "grad_norm": 35.225337982177734, "learning_rate": 9.999658931503347e-06, "loss": 12.841331481933594, "loss_alignment": 1.0576171875, "loss_alignment_w": 0.52880859375, "loss_sft": 0.2735457196831703, "loss_total_v6": 0.8025832176208496, "step": 229 }, { "epoch": 0.05304504504504504, "grad_norm": 33.20581817626953, "learning_rate": 9.999635004278054e-06, "loss": 12.403535842895508, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.24929636158049107, "loss_total_v6": 0.7752210199832916, "step": 230 }, { "epoch": 0.05327567567567568, "grad_norm": 53.24517059326172, "learning_rate": 9.99961026600051e-06, "loss": 12.737834930419922, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.27183796651661396, "loss_total_v6": 0.7961146607995033, "step": 231 }, { "epoch": 0.0535063063063063, "grad_norm": 31.463987350463867, "learning_rate": 9.999584716674727e-06, "loss": 12.30654525756836, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.2441346626728773, "loss_total_v6": 0.7691590860486031, "step": 232 }, { "epoch": 0.05373693693693694, "grad_norm": 29.8360595703125, "learning_rate": 9.99955835630485e-06, "loss": 13.121796607971191, "loss_alignment": 1.0634765625, "loss_alignment_w": 0.53173828125, "loss_sft": 0.28826718404889107, "loss_total_v6": 0.8201122656464577, "step": 233 }, { "epoch": 0.05396756756756757, "grad_norm": 32.144710540771484, "learning_rate": 9.999531184895156e-06, "loss": 12.628232955932617, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.26753603108227253, "loss_total_v6": 0.7892645373940468, "step": 234 }, { "epoch": 0.0541981981981982, "grad_norm": 30.990625381469727, "learning_rate": 9.999503202450052e-06, "loss": 12.962984085083008, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.2843990661203861, "loss_total_v6": 0.810186468064785, "step": 235 }, { "epoch": 0.05442882882882883, "grad_norm": 29.518918991088867, "learning_rate": 9.999474408974079e-06, "loss": 13.009281158447266, "loss_alignment": 1.07421875, "loss_alignment_w": 0.537109375, "loss_sft": 0.2767793647944927, "loss_total_v6": 0.8130800426006317, "step": 236 }, { "epoch": 0.05465945945945946, "grad_norm": 30.757413864135742, "learning_rate": 9.999444804471906e-06, "loss": 12.770463943481445, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.27964510954916477, "loss_total_v6": 0.7981540486216545, "step": 237 }, { "epoch": 0.05489009009009009, "grad_norm": 29.442298889160156, "learning_rate": 9.999414388948337e-06, "loss": 13.029792785644531, "loss_alignment": 1.064453125, "loss_alignment_w": 0.5322265625, "loss_sft": 0.2815555967390537, "loss_total_v6": 0.8143620118498802, "step": 238 }, { "epoch": 0.05512072072072072, "grad_norm": 35.66536331176758, "learning_rate": 9.999383162408303e-06, "loss": 13.200323104858398, "loss_alignment": 1.0654296875, "loss_alignment_w": 0.53271484375, "loss_sft": 0.2921985797584057, "loss_total_v6": 0.8250202089548111, "step": 239 }, { "epoch": 0.05535135135135135, "grad_norm": 33.057430267333984, "learning_rate": 9.999351124856873e-06, "loss": 12.950990676879883, "loss_alignment": 1.080078125, "loss_alignment_w": 0.5400390625, "loss_sft": 0.2703897412866354, "loss_total_v6": 0.8094369545578957, "step": 240 }, { "epoch": 0.055581981981981984, "grad_norm": 33.24260330200195, "learning_rate": 9.999318276299246e-06, "loss": 13.350919723510742, "loss_alignment": 1.0615234375, "loss_alignment_w": 0.53076171875, "loss_sft": 0.3037165757268667, "loss_total_v6": 0.834432527422905, "step": 241 }, { "epoch": 0.05581261261261261, "grad_norm": 35.32317352294922, "learning_rate": 9.999284616740746e-06, "loss": 12.927034378051758, "loss_alignment": 1.0732421875, "loss_alignment_w": 0.53662109375, "loss_sft": 0.2729055192321539, "loss_total_v6": 0.8079396858811378, "step": 242 }, { "epoch": 0.056043243243243245, "grad_norm": 32.18104934692383, "learning_rate": 9.999250146186834e-06, "loss": 13.084518432617188, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.29196448624134064, "loss_total_v6": 0.8177823647856712, "step": 243 }, { "epoch": 0.05627387387387387, "grad_norm": 33.46307373046875, "learning_rate": 9.999214864643103e-06, "loss": 12.801172256469727, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.27362983487546444, "loss_total_v6": 0.8000733107328415, "step": 244 }, { "epoch": 0.056504504504504505, "grad_norm": 35.73590850830078, "learning_rate": 9.999178772115279e-06, "loss": 12.982236862182617, "loss_alignment": 1.0595703125, "loss_alignment_w": 0.52978515625, "loss_sft": 0.28096385300159454, "loss_total_v6": 0.811389833688736, "step": 245 }, { "epoch": 0.05673513513513514, "grad_norm": 33.721336364746094, "learning_rate": 9.999141868609212e-06, "loss": 12.636012077331543, "loss_alignment": 1.068359375, "loss_alignment_w": 0.5341796875, "loss_sft": 0.2562576737254858, "loss_total_v6": 0.7897507175803185, "step": 246 }, { "epoch": 0.056965765765765765, "grad_norm": 29.04454803466797, "learning_rate": 9.999104154130894e-06, "loss": 13.163990020751953, "loss_alignment": 1.08203125, "loss_alignment_w": 0.541015625, "loss_sft": 0.28220670856535435, "loss_total_v6": 0.8227493092417717, "step": 247 }, { "epoch": 0.0571963963963964, "grad_norm": 32.37933349609375, "learning_rate": 9.999065628686439e-06, "loss": 13.266929626464844, "loss_alignment": 1.0556640625, "loss_alignment_w": 0.52783203125, "loss_sft": 0.30203769728541374, "loss_total_v6": 0.8291830793023109, "step": 248 }, { "epoch": 0.057427027027027025, "grad_norm": 32.00790786743164, "learning_rate": 9.999026292282096e-06, "loss": 12.84672737121582, "loss_alignment": 1.0634765625, "loss_alignment_w": 0.53173828125, "loss_sft": 0.27150262519717216, "loss_total_v6": 0.8029204905033112, "step": 249 }, { "epoch": 0.05765765765765766, "grad_norm": 30.530569076538086, "learning_rate": 9.998986144924253e-06, "loss": 12.695795059204102, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.26878325268626213, "loss_total_v6": 0.7934872284531593, "step": 250 }, { "epoch": 0.057888288288288285, "grad_norm": 30.405109405517578, "learning_rate": 9.998945186619413e-06, "loss": 12.34244155883789, "loss_alignment": 1.04736328125, "loss_alignment_w": 0.523681640625, "loss_sft": 0.24790403060615063, "loss_total_v6": 0.7714025899767876, "step": 251 }, { "epoch": 0.05811891891891892, "grad_norm": 31.44076156616211, "learning_rate": 9.998903417374228e-06, "loss": 12.861724853515625, "loss_alignment": 1.068359375, "loss_alignment_w": 0.5341796875, "loss_sft": 0.2702579591423273, "loss_total_v6": 0.803857795894146, "step": 252 }, { "epoch": 0.05834954954954955, "grad_norm": 31.23092269897461, "learning_rate": 9.998860837195472e-06, "loss": 13.3262939453125, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.308570833876729, "loss_total_v6": 0.8328933641314507, "step": 253 }, { "epoch": 0.05858018018018018, "grad_norm": 29.7677059173584, "learning_rate": 9.99881744609005e-06, "loss": 13.215841293334961, "loss_alignment": 1.04443359375, "loss_alignment_w": 0.522216796875, "loss_sft": 0.30442944914102554, "loss_total_v6": 0.8259901255369186, "step": 254 }, { "epoch": 0.05881081081081081, "grad_norm": 33.884429931640625, "learning_rate": 9.998773244065003e-06, "loss": 12.970660209655762, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.2862674668431282, "loss_total_v6": 0.8106662929058075, "step": 255 }, { "epoch": 0.05904144144144144, "grad_norm": 36.35212707519531, "learning_rate": 9.998728231127501e-06, "loss": 13.405640602111816, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.3114395570009947, "loss_total_v6": 0.8378525525331497, "step": 256 }, { "epoch": 0.05927207207207207, "grad_norm": 30.161958694458008, "learning_rate": 9.998682407284848e-06, "loss": 12.611045837402344, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.26608044281601906, "loss_total_v6": 0.7881904020905495, "step": 257 }, { "epoch": 0.059502702702702706, "grad_norm": 29.337562561035156, "learning_rate": 9.998635772544475e-06, "loss": 13.30959701538086, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.3118913173675537, "loss_total_v6": 0.8318498283624649, "step": 258 }, { "epoch": 0.05973333333333333, "grad_norm": 28.353355407714844, "learning_rate": 9.99858832691395e-06, "loss": 12.62418270111084, "loss_alignment": 1.03173828125, "loss_alignment_w": 0.515869140625, "loss_sft": 0.2731728497892618, "loss_total_v6": 0.7890114411711693, "step": 259 }, { "epoch": 0.059963963963963966, "grad_norm": 28.200223922729492, "learning_rate": 9.998540070400966e-06, "loss": 12.780755996704102, "loss_alignment": 1.04052734375, "loss_alignment_w": 0.520263671875, "loss_sft": 0.27947965264320374, "loss_total_v6": 0.7987972795963287, "step": 260 }, { "epoch": 0.06019459459459459, "grad_norm": 28.646516799926758, "learning_rate": 9.998491003013353e-06, "loss": 12.74860668182373, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.27129044011235237, "loss_total_v6": 0.7967878952622414, "step": 261 }, { "epoch": 0.060425225225225226, "grad_norm": 29.318601608276367, "learning_rate": 9.998441124759072e-06, "loss": 12.807653427124023, "loss_alignment": 1.03369140625, "loss_alignment_w": 0.516845703125, "loss_sft": 0.2839073408395052, "loss_total_v6": 0.8004783540964127, "step": 262 }, { "epoch": 0.06065585585585585, "grad_norm": 36.07237243652344, "learning_rate": 9.998390435646214e-06, "loss": 12.551568984985352, "loss_alignment": 1.03662109375, "loss_alignment_w": 0.518310546875, "loss_sft": 0.2645755559206009, "loss_total_v6": 0.7844730094075203, "step": 263 }, { "epoch": 0.06088648648648649, "grad_norm": 30.399913787841797, "learning_rate": 9.998338935683002e-06, "loss": 13.357183456420898, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.31532324105501175, "loss_total_v6": 0.8348239660263062, "step": 264 }, { "epoch": 0.06111711711711712, "grad_norm": 29.004905700683594, "learning_rate": 9.998286624877786e-06, "loss": 12.6559419631958, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.26380515471100807, "loss_total_v6": 0.7909963577985764, "step": 265 }, { "epoch": 0.06134774774774775, "grad_norm": 30.646024703979492, "learning_rate": 9.99823350323906e-06, "loss": 12.750419616699219, "loss_alignment": 1.0576171875, "loss_alignment_w": 0.52880859375, "loss_sft": 0.2676653880625963, "loss_total_v6": 0.7969012185931206, "step": 266 }, { "epoch": 0.06157837837837838, "grad_norm": 30.319578170776367, "learning_rate": 9.998179570775435e-06, "loss": 12.901203155517578, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.28770946338772774, "loss_total_v6": 0.8063251972198486, "step": 267 }, { "epoch": 0.06180900900900901, "grad_norm": 28.747060775756836, "learning_rate": 9.998124827495664e-06, "loss": 12.947742462158203, "loss_alignment": 1.056640625, "loss_alignment_w": 0.5283203125, "loss_sft": 0.28077624551951885, "loss_total_v6": 0.8092338964343071, "step": 268 }, { "epoch": 0.06203963963963964, "grad_norm": 29.742773056030273, "learning_rate": 9.998069273408624e-06, "loss": 12.366948127746582, "loss_alignment": 1.04443359375, "loss_alignment_w": 0.522216796875, "loss_sft": 0.2513583414256573, "loss_total_v6": 0.7729342430830002, "step": 269 }, { "epoch": 0.06227027027027027, "grad_norm": 29.887983322143555, "learning_rate": 9.99801290852333e-06, "loss": 12.317023277282715, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.24576609954237938, "loss_total_v6": 0.7698139399290085, "step": 270 }, { "epoch": 0.06250090090090091, "grad_norm": 29.54168701171875, "learning_rate": 9.997955732848925e-06, "loss": 12.794855117797852, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.28609812445938587, "loss_total_v6": 0.7996784374117851, "step": 271 }, { "epoch": 0.06273153153153153, "grad_norm": 37.11092758178711, "learning_rate": 9.997897746394684e-06, "loss": 12.81655502319336, "loss_alignment": 1.0546875, "loss_alignment_w": 0.52734375, "loss_sft": 0.2727449517697096, "loss_total_v6": 0.8010347113013268, "step": 272 }, { "epoch": 0.06296216216216216, "grad_norm": 27.665672302246094, "learning_rate": 9.997838949170013e-06, "loss": 12.4058198928833, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.2530553862452507, "loss_total_v6": 0.7753637358546257, "step": 273 }, { "epoch": 0.06319279279279279, "grad_norm": 27.63239097595215, "learning_rate": 9.99777934118445e-06, "loss": 13.046628952026367, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.29104598611593246, "loss_total_v6": 0.8154142871499062, "step": 274 }, { "epoch": 0.06342342342342343, "grad_norm": 29.918684005737305, "learning_rate": 9.997718922447669e-06, "loss": 12.986869812011719, "loss_alignment": 1.0703125, "loss_alignment_w": 0.53515625, "loss_sft": 0.2785372622311115, "loss_total_v6": 0.8116793483495712, "step": 275 }, { "epoch": 0.06365405405405405, "grad_norm": 27.54312515258789, "learning_rate": 9.997657692969465e-06, "loss": 12.651897430419922, "loss_alignment": 1.0615234375, "loss_alignment_w": 0.53076171875, "loss_sft": 0.25871543027460575, "loss_total_v6": 0.7907436117529869, "step": 276 }, { "epoch": 0.06388468468468468, "grad_norm": 28.56995964050293, "learning_rate": 9.997595652759773e-06, "loss": 13.087834358215332, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.291118873283267, "loss_total_v6": 0.8179896399378777, "step": 277 }, { "epoch": 0.06411531531531532, "grad_norm": 28.35601806640625, "learning_rate": 9.997532801828659e-06, "loss": 12.580802917480469, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.25936841778457165, "loss_total_v6": 0.7863001897931099, "step": 278 }, { "epoch": 0.06434594594594595, "grad_norm": 31.57276725769043, "learning_rate": 9.997469140186318e-06, "loss": 12.98330307006836, "loss_alignment": 1.0595703125, "loss_alignment_w": 0.52978515625, "loss_sft": 0.2831514049321413, "loss_total_v6": 0.8114564642310143, "step": 279 }, { "epoch": 0.06457657657657657, "grad_norm": 28.741912841796875, "learning_rate": 9.997404667843076e-06, "loss": 12.949126243591309, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2931765802204609, "loss_total_v6": 0.8093203976750374, "step": 280 }, { "epoch": 0.0648072072072072, "grad_norm": 26.818317413330078, "learning_rate": 9.997339384809392e-06, "loss": 13.231379508972168, "loss_alignment": 1.0712890625, "loss_alignment_w": 0.53564453125, "loss_sft": 0.28957720287144184, "loss_total_v6": 0.8269612342119217, "step": 281 }, { "epoch": 0.06503783783783784, "grad_norm": 29.923322677612305, "learning_rate": 9.997273291095858e-06, "loss": 12.537126541137695, "loss_alignment": 1.0673828125, "loss_alignment_w": 0.53369140625, "loss_sft": 0.25103859044611454, "loss_total_v6": 0.783570371568203, "step": 282 }, { "epoch": 0.06526846846846847, "grad_norm": 26.492109298706055, "learning_rate": 9.997206386713198e-06, "loss": 12.995796203613281, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.29049343056976795, "loss_total_v6": 0.8122372403740883, "step": 283 }, { "epoch": 0.0654990990990991, "grad_norm": 28.57612419128418, "learning_rate": 9.99713867167226e-06, "loss": 12.968696594238281, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.2853665370494127, "loss_total_v6": 0.8105435222387314, "step": 284 }, { "epoch": 0.06572972972972974, "grad_norm": 28.286470413208008, "learning_rate": 9.997070145984032e-06, "loss": 12.915550231933594, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.28637842275202274, "loss_total_v6": 0.8072219118475914, "step": 285 }, { "epoch": 0.06596036036036036, "grad_norm": 33.249107360839844, "learning_rate": 9.997000809659629e-06, "loss": 13.320640563964844, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.30940767750144005, "loss_total_v6": 0.8325399979948997, "step": 286 }, { "epoch": 0.06619099099099099, "grad_norm": 28.500598907470703, "learning_rate": 9.9969306627103e-06, "loss": 12.60733413696289, "loss_alignment": 1.046875, "loss_alignment_w": 0.5234375, "loss_sft": 0.2634069789201021, "loss_total_v6": 0.7879583537578583, "step": 287 }, { "epoch": 0.06642162162162162, "grad_norm": 29.27279281616211, "learning_rate": 9.996859705147424e-06, "loss": 12.518274307250977, "loss_alignment": 1.046875, "loss_alignment_w": 0.5234375, "loss_sft": 0.2594581823796034, "loss_total_v6": 0.7823921516537666, "step": 288 }, { "epoch": 0.06665225225225226, "grad_norm": 35.1848030090332, "learning_rate": 9.996787936982513e-06, "loss": 12.48763370513916, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.26785812713205814, "loss_total_v6": 0.7804771140217781, "step": 289 }, { "epoch": 0.06688288288288288, "grad_norm": 27.510225296020508, "learning_rate": 9.996715358227208e-06, "loss": 13.165997505187988, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.30473214015364647, "loss_total_v6": 0.8228748515248299, "step": 290 }, { "epoch": 0.06711351351351351, "grad_norm": 27.24018669128418, "learning_rate": 9.996641968893281e-06, "loss": 12.824785232543945, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.27805060520768166, "loss_total_v6": 0.8015491217374802, "step": 291 }, { "epoch": 0.06734414414414415, "grad_norm": 28.01561164855957, "learning_rate": 9.996567768992642e-06, "loss": 13.298786163330078, "loss_alignment": 1.05029296875, "loss_alignment_w": 0.525146484375, "loss_sft": 0.3054935820400715, "loss_total_v6": 0.8311741426587105, "step": 292 }, { "epoch": 0.06757477477477478, "grad_norm": 28.41852569580078, "learning_rate": 9.996492758537326e-06, "loss": 12.937723159790039, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.2884202841669321, "loss_total_v6": 0.8086076453328133, "step": 293 }, { "epoch": 0.0678054054054054, "grad_norm": 28.8185977935791, "learning_rate": 9.996416937539499e-06, "loss": 12.682303428649902, "loss_alignment": 1.0546875, "loss_alignment_w": 0.52734375, "loss_sft": 0.2644915319979191, "loss_total_v6": 0.7926439642906189, "step": 294 }, { "epoch": 0.06803603603603603, "grad_norm": 28.76427459716797, "learning_rate": 9.996340306011464e-06, "loss": 12.992313385009766, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.29705067351460457, "loss_total_v6": 0.812019556760788, "step": 295 }, { "epoch": 0.06826666666666667, "grad_norm": 29.039199829101562, "learning_rate": 9.996262863965651e-06, "loss": 12.661609649658203, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.27711418084800243, "loss_total_v6": 0.7913506180047989, "step": 296 }, { "epoch": 0.0684972972972973, "grad_norm": 27.04611587524414, "learning_rate": 9.996184611414623e-06, "loss": 13.068563461303711, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.2912419196218252, "loss_total_v6": 0.8167851492762566, "step": 297 }, { "epoch": 0.06872792792792792, "grad_norm": 30.949350357055664, "learning_rate": 9.996105548371074e-06, "loss": 12.941619873046875, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.28846551291644573, "loss_total_v6": 0.8088512793183327, "step": 298 }, { "epoch": 0.06895855855855856, "grad_norm": 27.314237594604492, "learning_rate": 9.996025674847831e-06, "loss": 12.688573837280273, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.27243643812835217, "loss_total_v6": 0.7930358275771141, "step": 299 }, { "epoch": 0.06918918918918919, "grad_norm": 31.00360870361328, "learning_rate": 9.995944990857848e-06, "loss": 13.457629203796387, "loss_alignment": 1.05615234375, "loss_alignment_w": 0.528076171875, "loss_sft": 0.31252215057611465, "loss_total_v6": 0.8411018326878548, "step": 300 }, { "epoch": 0.06941981981981982, "grad_norm": 30.850292205810547, "learning_rate": 9.995863496414219e-06, "loss": 12.456998825073242, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.25785625353455544, "loss_total_v6": 0.7785624340176582, "step": 301 }, { "epoch": 0.06965045045045046, "grad_norm": 27.190053939819336, "learning_rate": 9.995781191530158e-06, "loss": 12.752195358276367, "loss_alignment": 1.0556640625, "loss_alignment_w": 0.52783203125, "loss_sft": 0.26823420636355877, "loss_total_v6": 0.7970122620463371, "step": 302 }, { "epoch": 0.06988108108108108, "grad_norm": 28.05088996887207, "learning_rate": 9.995698076219022e-06, "loss": 12.717517852783203, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.27130056358873844, "loss_total_v6": 0.7948448807001114, "step": 303 }, { "epoch": 0.07011171171171171, "grad_norm": 29.68389892578125, "learning_rate": 9.995614150494293e-06, "loss": 12.572183609008789, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2685038149356842, "loss_total_v6": 0.7857614979147911, "step": 304 }, { "epoch": 0.07034234234234234, "grad_norm": 27.66911506652832, "learning_rate": 9.995529414369582e-06, "loss": 12.944347381591797, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.28569102846086025, "loss_total_v6": 0.8090217113494873, "step": 305 }, { "epoch": 0.07057297297297298, "grad_norm": 32.711753845214844, "learning_rate": 9.995443867858642e-06, "loss": 12.799393653869629, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.2754717469215393, "loss_total_v6": 0.7999621108174324, "step": 306 }, { "epoch": 0.0708036036036036, "grad_norm": 29.206090927124023, "learning_rate": 9.995357510975343e-06, "loss": 12.458385467529297, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.25249553099274635, "loss_total_v6": 0.7786491215229034, "step": 307 }, { "epoch": 0.07103423423423423, "grad_norm": 29.795969009399414, "learning_rate": 9.995270343733697e-06, "loss": 12.974613189697266, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.2916415147483349, "loss_total_v6": 0.8109133318066597, "step": 308 }, { "epoch": 0.07126486486486487, "grad_norm": 34.19267654418945, "learning_rate": 9.995182366147846e-06, "loss": 13.260213851928711, "loss_alignment": 1.068359375, "loss_alignment_w": 0.5341796875, "loss_sft": 0.2944311574101448, "loss_total_v6": 0.8287634178996086, "step": 309 }, { "epoch": 0.0714954954954955, "grad_norm": 29.557872772216797, "learning_rate": 9.99509357823206e-06, "loss": 13.121465682983398, "loss_alignment": 1.046875, "loss_alignment_w": 0.5234375, "loss_sft": 0.2967151366174221, "loss_total_v6": 0.8200915977358818, "step": 310 }, { "epoch": 0.07172612612612612, "grad_norm": 26.094026565551758, "learning_rate": 9.995003980000743e-06, "loss": 13.519064903259277, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.32212968170642853, "loss_total_v6": 0.844941571354866, "step": 311 }, { "epoch": 0.07195675675675675, "grad_norm": 25.870399475097656, "learning_rate": 9.994913571468432e-06, "loss": 12.474217414855957, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2683776319026947, "loss_total_v6": 0.7796385884284973, "step": 312 }, { "epoch": 0.07218738738738739, "grad_norm": 27.619983673095703, "learning_rate": 9.99482235264979e-06, "loss": 12.53802490234375, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.268154164776206, "loss_total_v6": 0.7836265712976456, "step": 313 }, { "epoch": 0.07241801801801802, "grad_norm": 28.443084716796875, "learning_rate": 9.994730323559616e-06, "loss": 13.168815612792969, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.3011393826454878, "loss_total_v6": 0.8230509907007217, "step": 314 }, { "epoch": 0.07264864864864864, "grad_norm": 26.589784622192383, "learning_rate": 9.994637484212839e-06, "loss": 12.759736061096191, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.27651794627308846, "loss_total_v6": 0.7974834889173508, "step": 315 }, { "epoch": 0.07287927927927929, "grad_norm": 25.804386138916016, "learning_rate": 9.99454383462452e-06, "loss": 12.768477439880371, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.2949475832283497, "loss_total_v6": 0.7980298474431038, "step": 316 }, { "epoch": 0.07310990990990991, "grad_norm": 30.758575439453125, "learning_rate": 9.994449374809851e-06, "loss": 13.139461517333984, "loss_alignment": 1.0654296875, "loss_alignment_w": 0.53271484375, "loss_sft": 0.28862352669239044, "loss_total_v6": 0.8212163150310516, "step": 317 }, { "epoch": 0.07334054054054054, "grad_norm": 28.832290649414062, "learning_rate": 9.994354104784153e-06, "loss": 12.696924209594727, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2768646255135536, "loss_total_v6": 0.7935577183961868, "step": 318 }, { "epoch": 0.07357117117117117, "grad_norm": 25.769481658935547, "learning_rate": 9.994258024562887e-06, "loss": 12.274060249328613, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.24387438409030437, "loss_total_v6": 0.7671287581324577, "step": 319 }, { "epoch": 0.0738018018018018, "grad_norm": 26.350553512573242, "learning_rate": 9.994161134161635e-06, "loss": 13.026908874511719, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.28967620618641376, "loss_total_v6": 0.8141817972064018, "step": 320 }, { "epoch": 0.07403243243243243, "grad_norm": 30.438613891601562, "learning_rate": 9.994063433596113e-06, "loss": 13.019505500793457, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.2953627686947584, "loss_total_v6": 0.8137191012501717, "step": 321 }, { "epoch": 0.07426306306306306, "grad_norm": 26.311954498291016, "learning_rate": 9.993964922882174e-06, "loss": 12.692558288574219, "loss_alignment": 1.03369140625, "loss_alignment_w": 0.516845703125, "loss_sft": 0.27685118839144707, "loss_total_v6": 0.7932848706841469, "step": 322 }, { "epoch": 0.0744936936936937, "grad_norm": 26.44007110595703, "learning_rate": 9.993865602035798e-06, "loss": 12.4622220993042, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.262043172493577, "loss_total_v6": 0.7788888812065125, "step": 323 }, { "epoch": 0.07472432432432433, "grad_norm": 28.416093826293945, "learning_rate": 9.993765471073094e-06, "loss": 12.338972091674805, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.25571329705417156, "loss_total_v6": 0.7711857333779335, "step": 324 }, { "epoch": 0.07495495495495495, "grad_norm": 27.920427322387695, "learning_rate": 9.993664530010308e-06, "loss": 13.024862289428711, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.29047904163599014, "loss_total_v6": 0.8140538707375526, "step": 325 }, { "epoch": 0.07518558558558558, "grad_norm": 29.480226516723633, "learning_rate": 9.993562778863817e-06, "loss": 12.599124908447266, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.26908893696963787, "loss_total_v6": 0.7874452620744705, "step": 326 }, { "epoch": 0.07541621621621622, "grad_norm": 25.620187759399414, "learning_rate": 9.993460217650122e-06, "loss": 12.142919540405273, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.23804316855967045, "loss_total_v6": 0.7589324861764908, "step": 327 }, { "epoch": 0.07564684684684685, "grad_norm": 28.651870727539062, "learning_rate": 9.993356846385866e-06, "loss": 13.118865013122559, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.29635415971279144, "loss_total_v6": 0.8199290558695793, "step": 328 }, { "epoch": 0.07587747747747747, "grad_norm": 27.005659103393555, "learning_rate": 9.993252665087816e-06, "loss": 12.266449928283691, "loss_alignment": 1.05859375, "loss_alignment_w": 0.529296875, "loss_sft": 0.2365322709083557, "loss_total_v6": 0.7666531205177307, "step": 329 }, { "epoch": 0.07610810810810811, "grad_norm": 27.174407958984375, "learning_rate": 9.993147673772869e-06, "loss": 12.279251098632812, "loss_alignment": 1.02783203125, "loss_alignment_w": 0.513916015625, "loss_sft": 0.2532624900341034, "loss_total_v6": 0.7674531787633896, "step": 330 }, { "epoch": 0.07633873873873874, "grad_norm": 25.61027717590332, "learning_rate": 9.993041872458064e-06, "loss": 12.72581958770752, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.2705528922379017, "loss_total_v6": 0.7953636795282364, "step": 331 }, { "epoch": 0.07656936936936937, "grad_norm": 28.055225372314453, "learning_rate": 9.992935261160559e-06, "loss": 12.83183479309082, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.2811461128294468, "loss_total_v6": 0.8019896224141121, "step": 332 }, { "epoch": 0.0768, "grad_norm": 28.580162048339844, "learning_rate": 9.99282783989765e-06, "loss": 12.116533279418945, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2498675361275673, "loss_total_v6": 0.7572832852602005, "step": 333 }, { "epoch": 0.07703063063063063, "grad_norm": 24.627605438232422, "learning_rate": 9.992719608686763e-06, "loss": 12.505990982055664, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.2655111327767372, "loss_total_v6": 0.7816244140267372, "step": 334 }, { "epoch": 0.07726126126126126, "grad_norm": 28.806821823120117, "learning_rate": 9.992610567545458e-06, "loss": 12.70738410949707, "loss_alignment": 1.03955078125, "loss_alignment_w": 0.519775390625, "loss_sft": 0.2746954970061779, "loss_total_v6": 0.7942114919424057, "step": 335 }, { "epoch": 0.07749189189189189, "grad_norm": 27.055343627929688, "learning_rate": 9.99250071649142e-06, "loss": 12.289006233215332, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.25118665024638176, "loss_total_v6": 0.7680628821253777, "step": 336 }, { "epoch": 0.07772252252252253, "grad_norm": 32.459651947021484, "learning_rate": 9.992390055542472e-06, "loss": 12.789876937866211, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.287495993077755, "loss_total_v6": 0.799367293715477, "step": 337 }, { "epoch": 0.07795315315315315, "grad_norm": 28.091440200805664, "learning_rate": 9.992278584716563e-06, "loss": 12.51415729522705, "loss_alignment": 1.03173828125, "loss_alignment_w": 0.515869140625, "loss_sft": 0.26704393327236176, "loss_total_v6": 0.7821348682045937, "step": 338 }, { "epoch": 0.07818378378378378, "grad_norm": 26.770017623901367, "learning_rate": 9.992166304031781e-06, "loss": 12.908563613891602, "loss_alignment": 1.0595703125, "loss_alignment_w": 0.52978515625, "loss_sft": 0.2766338549554348, "loss_total_v6": 0.8067852407693863, "step": 339 }, { "epoch": 0.07841441441441442, "grad_norm": 27.528425216674805, "learning_rate": 9.992053213506333e-06, "loss": 12.566232681274414, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.2638899050652981, "loss_total_v6": 0.7853895351290703, "step": 340 }, { "epoch": 0.07864504504504505, "grad_norm": 28.959705352783203, "learning_rate": 9.991939313158574e-06, "loss": 12.872640609741211, "loss_alignment": 1.0693359375, "loss_alignment_w": 0.53466796875, "loss_sft": 0.2721456177532673, "loss_total_v6": 0.8045400679111481, "step": 341 }, { "epoch": 0.07887567567567567, "grad_norm": 24.084745407104492, "learning_rate": 9.991824603006972e-06, "loss": 12.56607437133789, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2674506362527609, "loss_total_v6": 0.7853797003626823, "step": 342 }, { "epoch": 0.0791063063063063, "grad_norm": 27.46710205078125, "learning_rate": 9.991709083070143e-06, "loss": 12.952554702758789, "loss_alignment": 1.0634765625, "loss_alignment_w": 0.53173828125, "loss_sft": 0.2766214422881603, "loss_total_v6": 0.8095346689224243, "step": 343 }, { "epoch": 0.07933693693693694, "grad_norm": 25.98877716064453, "learning_rate": 9.991592753366822e-06, "loss": 12.989216804504395, "loss_alignment": 1.0625, "loss_alignment_w": 0.53125, "loss_sft": 0.2813695427030325, "loss_total_v6": 0.8118260577321053, "step": 344 }, { "epoch": 0.07956756756756757, "grad_norm": 28.222457885742188, "learning_rate": 9.991475613915883e-06, "loss": 13.025232315063477, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.29244006983935833, "loss_total_v6": 0.8140770494937897, "step": 345 }, { "epoch": 0.0797981981981982, "grad_norm": 30.32134246826172, "learning_rate": 9.991357664736327e-06, "loss": 12.818130493164062, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.28337190859019756, "loss_total_v6": 0.8011331036686897, "step": 346 }, { "epoch": 0.08002882882882884, "grad_norm": 27.61977767944336, "learning_rate": 9.991238905847289e-06, "loss": 12.817916870117188, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2857237048447132, "loss_total_v6": 0.8011198118329048, "step": 347 }, { "epoch": 0.08025945945945946, "grad_norm": 30.709537506103516, "learning_rate": 9.991119337268032e-06, "loss": 12.494037628173828, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.26325339265167713, "loss_total_v6": 0.7808773219585419, "step": 348 }, { "epoch": 0.08049009009009009, "grad_norm": 30.213470458984375, "learning_rate": 9.990998959017956e-06, "loss": 12.930904388427734, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.2875821776688099, "loss_total_v6": 0.8081815391778946, "step": 349 }, { "epoch": 0.08072072072072072, "grad_norm": 26.426820755004883, "learning_rate": 9.990877771116588e-06, "loss": 12.769051551818848, "loss_alignment": 1.0546875, "loss_alignment_w": 0.52734375, "loss_sft": 0.2701726704835892, "loss_total_v6": 0.798065721988678, "step": 350 }, { "epoch": 0.08095135135135136, "grad_norm": 24.204551696777344, "learning_rate": 9.990755773583588e-06, "loss": 13.04059886932373, "loss_alignment": 1.056640625, "loss_alignment_w": 0.5283203125, "loss_sft": 0.28540488705039024, "loss_total_v6": 0.8150374218821526, "step": 351 }, { "epoch": 0.08118198198198198, "grad_norm": 24.077594757080078, "learning_rate": 9.990632966438743e-06, "loss": 12.424110412597656, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.2589592654258013, "loss_total_v6": 0.7765069007873535, "step": 352 }, { "epoch": 0.08141261261261261, "grad_norm": 25.68401336669922, "learning_rate": 9.99050934970198e-06, "loss": 12.988487243652344, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.2907538451254368, "loss_total_v6": 0.8117804229259491, "step": 353 }, { "epoch": 0.08164324324324325, "grad_norm": 24.25824737548828, "learning_rate": 9.99038492339335e-06, "loss": 12.382295608520508, "loss_alignment": 1.03271484375, "loss_alignment_w": 0.516357421875, "loss_sft": 0.2561932522803545, "loss_total_v6": 0.7738934457302094, "step": 354 }, { "epoch": 0.08187387387387388, "grad_norm": 29.09197235107422, "learning_rate": 9.990259687533038e-06, "loss": 12.530942916870117, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.2639273274689913, "loss_total_v6": 0.7831839099526405, "step": 355 }, { "epoch": 0.0821045045045045, "grad_norm": 28.827463150024414, "learning_rate": 9.990133642141359e-06, "loss": 13.266341209411621, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.3091573305428028, "loss_total_v6": 0.8291463479399681, "step": 356 }, { "epoch": 0.08233513513513513, "grad_norm": 27.91555404663086, "learning_rate": 9.990006787238761e-06, "loss": 12.44978141784668, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.26013651117682457, "loss_total_v6": 0.7781113833189011, "step": 357 }, { "epoch": 0.08256576576576577, "grad_norm": 30.166318893432617, "learning_rate": 9.989879122845825e-06, "loss": 12.479297637939453, "loss_alignment": 1.02978515625, "loss_alignment_w": 0.514892578125, "loss_sft": 0.265383992344141, "loss_total_v6": 0.7799561098217964, "step": 358 }, { "epoch": 0.0827963963963964, "grad_norm": 27.42767906188965, "learning_rate": 9.989750648983258e-06, "loss": 12.149486541748047, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.24599147215485573, "loss_total_v6": 0.7593429014086723, "step": 359 }, { "epoch": 0.08302702702702702, "grad_norm": 26.570077896118164, "learning_rate": 9.989621365671902e-06, "loss": 12.140247344970703, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.2592537850141525, "loss_total_v6": 0.7587654963135719, "step": 360 }, { "epoch": 0.08325765765765766, "grad_norm": 28.698013305664062, "learning_rate": 9.98949127293273e-06, "loss": 12.80367374420166, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2831702493131161, "loss_total_v6": 0.8002296090126038, "step": 361 }, { "epoch": 0.08348828828828829, "grad_norm": 27.846811294555664, "learning_rate": 9.989360370786847e-06, "loss": 12.843279838562012, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.2772838566452265, "loss_total_v6": 0.8027049973607063, "step": 362 }, { "epoch": 0.08371891891891892, "grad_norm": 31.650739669799805, "learning_rate": 9.989228659255486e-06, "loss": 13.265066146850586, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.3115190677344799, "loss_total_v6": 0.8290666490793228, "step": 363 }, { "epoch": 0.08394954954954954, "grad_norm": 26.630075454711914, "learning_rate": 9.989096138360014e-06, "loss": 13.36771297454834, "loss_alignment": 1.0595703125, "loss_alignment_w": 0.52978515625, "loss_sft": 0.30670396983623505, "loss_total_v6": 0.83548204600811, "step": 364 }, { "epoch": 0.08418018018018018, "grad_norm": 25.413103103637695, "learning_rate": 9.988962808121932e-06, "loss": 12.7880859375, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.2753905486315489, "loss_total_v6": 0.799255333840847, "step": 365 }, { "epoch": 0.08441081081081081, "grad_norm": 26.25037956237793, "learning_rate": 9.988828668562866e-06, "loss": 12.583772659301758, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.26754964143037796, "loss_total_v6": 0.7864857912063599, "step": 366 }, { "epoch": 0.08464144144144144, "grad_norm": 33.8400764465332, "learning_rate": 9.988693719704577e-06, "loss": 13.036041259765625, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.28956031054258347, "loss_total_v6": 0.8147525787353516, "step": 367 }, { "epoch": 0.08487207207207208, "grad_norm": 24.07904815673828, "learning_rate": 9.988557961568956e-06, "loss": 12.321378707885742, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.25159251131117344, "loss_total_v6": 0.7700861766934395, "step": 368 }, { "epoch": 0.0851027027027027, "grad_norm": 28.681373596191406, "learning_rate": 9.988421394178027e-06, "loss": 12.657415390014648, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.2724575027823448, "loss_total_v6": 0.7910884618759155, "step": 369 }, { "epoch": 0.08533333333333333, "grad_norm": 30.226078033447266, "learning_rate": 9.988284017553945e-06, "loss": 13.202919006347656, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.30324032716453075, "loss_total_v6": 0.8251824676990509, "step": 370 }, { "epoch": 0.08556396396396396, "grad_norm": 29.26854705810547, "learning_rate": 9.988145831718995e-06, "loss": 12.661002159118652, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.2698588203638792, "loss_total_v6": 0.7913126349449158, "step": 371 }, { "epoch": 0.0857945945945946, "grad_norm": 25.849689483642578, "learning_rate": 9.988006836695593e-06, "loss": 12.96546745300293, "loss_alignment": 1.04638671875, "loss_alignment_w": 0.523193359375, "loss_sft": 0.2857903316617012, "loss_total_v6": 0.8103417456150055, "step": 372 }, { "epoch": 0.08602522522522522, "grad_norm": 27.952510833740234, "learning_rate": 9.987867032506286e-06, "loss": 12.14124870300293, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2505425959825516, "loss_total_v6": 0.7588280662894249, "step": 373 }, { "epoch": 0.08625585585585585, "grad_norm": 29.310665130615234, "learning_rate": 9.987726419173756e-06, "loss": 12.791702270507812, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.2730531617999077, "loss_total_v6": 0.7994813472032547, "step": 374 }, { "epoch": 0.08648648648648649, "grad_norm": 28.38065528869629, "learning_rate": 9.987584996720813e-06, "loss": 13.416007995605469, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.3109278678894043, "loss_total_v6": 0.8385005071759224, "step": 375 }, { "epoch": 0.08671711711711712, "grad_norm": 26.064186096191406, "learning_rate": 9.987442765170398e-06, "loss": 13.14644718170166, "loss_alignment": 1.04833984375, "loss_alignment_w": 0.524169921875, "loss_sft": 0.29762036725878716, "loss_total_v6": 0.8216529786586761, "step": 376 }, { "epoch": 0.08694774774774774, "grad_norm": 26.01885414123535, "learning_rate": 9.987299724545585e-06, "loss": 13.169514656066895, "loss_alignment": 1.05517578125, "loss_alignment_w": 0.527587890625, "loss_sft": 0.29440814070403576, "loss_total_v6": 0.8230946734547615, "step": 377 }, { "epoch": 0.08717837837837838, "grad_norm": 28.56486701965332, "learning_rate": 9.987155874869579e-06, "loss": 12.618545532226562, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2734155859798193, "loss_total_v6": 0.788659118115902, "step": 378 }, { "epoch": 0.08740900900900901, "grad_norm": 28.558500289916992, "learning_rate": 9.987011216165715e-06, "loss": 12.874129295349121, "loss_alignment": 1.05126953125, "loss_alignment_w": 0.525634765625, "loss_sft": 0.2792119514197111, "loss_total_v6": 0.8046330809593201, "step": 379 }, { "epoch": 0.08763963963963964, "grad_norm": 29.600460052490234, "learning_rate": 9.986865748457457e-06, "loss": 12.419700622558594, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.25734091736376286, "loss_total_v6": 0.7762313038110733, "step": 380 }, { "epoch": 0.08787027027027026, "grad_norm": 28.31549835205078, "learning_rate": 9.986719471768406e-06, "loss": 13.217884063720703, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.31191184371709824, "loss_total_v6": 0.8261177837848663, "step": 381 }, { "epoch": 0.0881009009009009, "grad_norm": 29.167112350463867, "learning_rate": 9.98657238612229e-06, "loss": 12.791090965270996, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.28302471339702606, "loss_total_v6": 0.7994431629776955, "step": 382 }, { "epoch": 0.08833153153153153, "grad_norm": 27.890094757080078, "learning_rate": 9.986424491542972e-06, "loss": 12.946578979492188, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.29463479295372963, "loss_total_v6": 0.8091611862182617, "step": 383 }, { "epoch": 0.08856216216216216, "grad_norm": 31.997787475585938, "learning_rate": 9.986275788054442e-06, "loss": 12.577340126037598, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.27114539593458176, "loss_total_v6": 0.7860837578773499, "step": 384 }, { "epoch": 0.0887927927927928, "grad_norm": 28.886192321777344, "learning_rate": 9.986126275680825e-06, "loss": 12.845130920410156, "loss_alignment": 1.0546875, "loss_alignment_w": 0.52734375, "loss_sft": 0.2758889701217413, "loss_total_v6": 0.802820697426796, "step": 385 }, { "epoch": 0.08902342342342343, "grad_norm": 24.481956481933594, "learning_rate": 9.985975954446372e-06, "loss": 12.438318252563477, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.25758895836770535, "loss_total_v6": 0.7773948609828949, "step": 386 }, { "epoch": 0.08925405405405405, "grad_norm": 26.324525833129883, "learning_rate": 9.985824824375469e-06, "loss": 13.024103164672852, "loss_alignment": 1.0556640625, "loss_alignment_w": 0.52783203125, "loss_sft": 0.28528942354023457, "loss_total_v6": 0.8140064477920532, "step": 387 }, { "epoch": 0.08948468468468468, "grad_norm": 28.40669822692871, "learning_rate": 9.985672885492634e-06, "loss": 13.28851318359375, "loss_alignment": 1.0576171875, "loss_alignment_w": 0.52880859375, "loss_sft": 0.30138777010142803, "loss_total_v6": 0.830532044172287, "step": 388 }, { "epoch": 0.08971531531531532, "grad_norm": 26.647748947143555, "learning_rate": 9.985520137822516e-06, "loss": 12.71446418762207, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.27446665801107883, "loss_total_v6": 0.7946540415287018, "step": 389 }, { "epoch": 0.08994594594594595, "grad_norm": 27.44015884399414, "learning_rate": 9.985366581389893e-06, "loss": 13.371804237365723, "loss_alignment": 1.02783203125, "loss_alignment_w": 0.513916015625, "loss_sft": 0.32000596448779106, "loss_total_v6": 0.8357377648353577, "step": 390 }, { "epoch": 0.09017657657657657, "grad_norm": 24.066883087158203, "learning_rate": 9.985212216219673e-06, "loss": 12.679723739624023, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.27612533792853355, "loss_total_v6": 0.7924827262759209, "step": 391 }, { "epoch": 0.09040720720720721, "grad_norm": 27.868654251098633, "learning_rate": 9.985057042336898e-06, "loss": 12.716527938842773, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.27294772304594517, "loss_total_v6": 0.7947830185294151, "step": 392 }, { "epoch": 0.09063783783783784, "grad_norm": 28.37613868713379, "learning_rate": 9.984901059766746e-06, "loss": 13.568058967590332, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.32946423441171646, "loss_total_v6": 0.8480036929249763, "step": 393 }, { "epoch": 0.09086846846846847, "grad_norm": 26.77106285095215, "learning_rate": 9.984744268534515e-06, "loss": 12.696159362792969, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.27873943373560905, "loss_total_v6": 0.7935099452733994, "step": 394 }, { "epoch": 0.09109909909909909, "grad_norm": 24.917787551879883, "learning_rate": 9.984586668665641e-06, "loss": 12.762596130371094, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.2735839206725359, "loss_total_v6": 0.7976622879505157, "step": 395 }, { "epoch": 0.09132972972972973, "grad_norm": 26.535724639892578, "learning_rate": 9.984428260185691e-06, "loss": 12.551704406738281, "loss_alignment": 1.060546875, "loss_alignment_w": 0.5302734375, "loss_sft": 0.25494049303233624, "loss_total_v6": 0.7844815254211426, "step": 396 }, { "epoch": 0.09156036036036036, "grad_norm": 26.879180908203125, "learning_rate": 9.984269043120364e-06, "loss": 12.432920455932617, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.26048649474978447, "loss_total_v6": 0.7770575433969498, "step": 397 }, { "epoch": 0.09179099099099099, "grad_norm": 26.009811401367188, "learning_rate": 9.984109017495485e-06, "loss": 13.032218933105469, "loss_alignment": 1.060546875, "loss_alignment_w": 0.5302734375, "loss_sft": 0.28410294093191624, "loss_total_v6": 0.814513698220253, "step": 398 }, { "epoch": 0.09202162162162163, "grad_norm": 27.232004165649414, "learning_rate": 9.983948183337018e-06, "loss": 12.87534236907959, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.28739018365740776, "loss_total_v6": 0.804708905518055, "step": 399 }, { "epoch": 0.09225225225225225, "grad_norm": 28.40029525756836, "learning_rate": 9.983786540671052e-06, "loss": 12.817012786865234, "loss_alignment": 1.0556640625, "loss_alignment_w": 0.52783203125, "loss_sft": 0.27419257164001465, "loss_total_v6": 0.8010633140802383, "step": 400 }, { "epoch": 0.09248288288288288, "grad_norm": 29.244873046875, "learning_rate": 9.983624089523806e-06, "loss": 12.713988304138184, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.26950833573937416, "loss_total_v6": 0.7946242690086365, "step": 401 }, { "epoch": 0.0927135135135135, "grad_norm": 25.751502990722656, "learning_rate": 9.983460829921637e-06, "loss": 12.905773162841797, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.280228354036808, "loss_total_v6": 0.8066108152270317, "step": 402 }, { "epoch": 0.09294414414414415, "grad_norm": 26.833707809448242, "learning_rate": 9.983296761891029e-06, "loss": 12.675731658935547, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.26797178387641907, "loss_total_v6": 0.7922332361340523, "step": 403 }, { "epoch": 0.09317477477477477, "grad_norm": 26.947769165039062, "learning_rate": 9.983131885458596e-06, "loss": 12.391786575317383, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2619134169071913, "loss_total_v6": 0.7744866758584976, "step": 404 }, { "epoch": 0.0934054054054054, "grad_norm": 24.6778564453125, "learning_rate": 9.982966200651085e-06, "loss": 12.13823127746582, "loss_alignment": 1.03173828125, "loss_alignment_w": 0.515869140625, "loss_sft": 0.24365535378456116, "loss_total_v6": 0.7586394920945168, "step": 405 }, { "epoch": 0.09363603603603604, "grad_norm": 25.216440200805664, "learning_rate": 9.982799707495376e-06, "loss": 13.044157028198242, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.2935160659253597, "loss_total_v6": 0.8152598440647125, "step": 406 }, { "epoch": 0.09386666666666667, "grad_norm": 25.203292846679688, "learning_rate": 9.982632406018475e-06, "loss": 12.067489624023438, "loss_alignment": 1.03271484375, "loss_alignment_w": 0.516357421875, "loss_sft": 0.23793693259358406, "loss_total_v6": 0.7542180567979813, "step": 407 }, { "epoch": 0.0940972972972973, "grad_norm": 64.7546615600586, "learning_rate": 9.982464296247523e-06, "loss": 12.523172378540039, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.26759215258061886, "loss_total_v6": 0.7826983407139778, "step": 408 }, { "epoch": 0.09432792792792793, "grad_norm": 25.339651107788086, "learning_rate": 9.982295378209792e-06, "loss": 11.847034454345703, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.23500757664442062, "loss_total_v6": 0.7404396831989288, "step": 409 }, { "epoch": 0.09455855855855856, "grad_norm": 26.53764533996582, "learning_rate": 9.982125651932681e-06, "loss": 12.475593566894531, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.26135301776230335, "loss_total_v6": 0.7797246128320694, "step": 410 }, { "epoch": 0.09478918918918919, "grad_norm": 28.9260196685791, "learning_rate": 9.981955117443728e-06, "loss": 13.262323379516602, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.30413021333515644, "loss_total_v6": 0.82889524102211, "step": 411 }, { "epoch": 0.09501981981981981, "grad_norm": 26.562707901000977, "learning_rate": 9.981783774770595e-06, "loss": 12.662088394165039, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.28227100521326065, "loss_total_v6": 0.7913805097341537, "step": 412 }, { "epoch": 0.09525045045045046, "grad_norm": 24.736909866333008, "learning_rate": 9.981611623941077e-06, "loss": 12.796920776367188, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.28404521383345127, "loss_total_v6": 0.7998075261712074, "step": 413 }, { "epoch": 0.09548108108108108, "grad_norm": 25.38623046875, "learning_rate": 9.981438664983103e-06, "loss": 12.845602989196777, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2864622585475445, "loss_total_v6": 0.802850216627121, "step": 414 }, { "epoch": 0.09571171171171171, "grad_norm": 25.477598190307617, "learning_rate": 9.98126489792473e-06, "loss": 12.40560245513916, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2644096463918686, "loss_total_v6": 0.7753501757979393, "step": 415 }, { "epoch": 0.09594234234234235, "grad_norm": 26.838645935058594, "learning_rate": 9.981090322794145e-06, "loss": 13.369083404541016, "loss_alignment": 1.0556640625, "loss_alignment_w": 0.52783203125, "loss_sft": 0.3074609972536564, "loss_total_v6": 0.8355677127838135, "step": 416 }, { "epoch": 0.09617297297297298, "grad_norm": 25.650087356567383, "learning_rate": 9.980914939619669e-06, "loss": 12.446285247802734, "loss_alignment": 1.03173828125, "loss_alignment_w": 0.515869140625, "loss_sft": 0.26274083741009235, "loss_total_v6": 0.7778928428888321, "step": 417 }, { "epoch": 0.0964036036036036, "grad_norm": 25.23677635192871, "learning_rate": 9.980738748429755e-06, "loss": 13.263452529907227, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.30446022376418114, "loss_total_v6": 0.8289658352732658, "step": 418 }, { "epoch": 0.09663423423423423, "grad_norm": 24.99830436706543, "learning_rate": 9.980561749252982e-06, "loss": 13.250080108642578, "loss_alignment": 1.046875, "loss_alignment_w": 0.5234375, "loss_sft": 0.30496713519096375, "loss_total_v6": 0.8281299918889999, "step": 419 }, { "epoch": 0.09686486486486487, "grad_norm": 27.430097579956055, "learning_rate": 9.980383942118066e-06, "loss": 13.20534896850586, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.2993790786713362, "loss_total_v6": 0.8253342807292938, "step": 420 }, { "epoch": 0.0970954954954955, "grad_norm": 26.03395652770996, "learning_rate": 9.98020532705385e-06, "loss": 12.827014923095703, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.2785866744816303, "loss_total_v6": 0.8016884475946426, "step": 421 }, { "epoch": 0.09732612612612612, "grad_norm": 26.666982650756836, "learning_rate": 9.980025904089308e-06, "loss": 12.386646270751953, "loss_alignment": 1.01611328125, "loss_alignment_w": 0.508056640625, "loss_sft": 0.2658493835479021, "loss_total_v6": 0.7741654291749, "step": 422 }, { "epoch": 0.09755675675675676, "grad_norm": 23.094533920288086, "learning_rate": 9.979845673253547e-06, "loss": 12.484533309936523, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2616523075848818, "loss_total_v6": 0.7802833169698715, "step": 423 }, { "epoch": 0.09778738738738739, "grad_norm": 24.623455047607422, "learning_rate": 9.979664634575807e-06, "loss": 12.425865173339844, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2630362715572119, "loss_total_v6": 0.7766165882349014, "step": 424 }, { "epoch": 0.09801801801801802, "grad_norm": 26.114046096801758, "learning_rate": 9.979482788085455e-06, "loss": 13.14681339263916, "loss_alignment": 1.02978515625, "loss_alignment_w": 0.514892578125, "loss_sft": 0.3074851632118225, "loss_total_v6": 0.8216758519411087, "step": 425 }, { "epoch": 0.09824864864864864, "grad_norm": 25.341957092285156, "learning_rate": 9.979300133811991e-06, "loss": 13.308479309082031, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.31215720996260643, "loss_total_v6": 0.8317799940705299, "step": 426 }, { "epoch": 0.09847927927927928, "grad_norm": 25.81424331665039, "learning_rate": 9.979116671785044e-06, "loss": 13.01740550994873, "loss_alignment": 1.05859375, "loss_alignment_w": 0.529296875, "loss_sft": 0.2845656629651785, "loss_total_v6": 0.8135878518223763, "step": 427 }, { "epoch": 0.09870990990990991, "grad_norm": 32.62810516357422, "learning_rate": 9.978932402034376e-06, "loss": 13.73437213897705, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.3336637392640114, "loss_total_v6": 0.8583982512354851, "step": 428 }, { "epoch": 0.09894054054054054, "grad_norm": 25.606542587280273, "learning_rate": 9.97874732458988e-06, "loss": 13.019968032836914, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.2913481518626213, "loss_total_v6": 0.8137480169534683, "step": 429 }, { "epoch": 0.09917117117117118, "grad_norm": 24.783437728881836, "learning_rate": 9.97856143948158e-06, "loss": 12.70248031616211, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.2835595887154341, "loss_total_v6": 0.7939050421118736, "step": 430 }, { "epoch": 0.0994018018018018, "grad_norm": 24.40469741821289, "learning_rate": 9.978374746739631e-06, "loss": 12.548602104187012, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.27427787333726883, "loss_total_v6": 0.7842876464128494, "step": 431 }, { "epoch": 0.09963243243243243, "grad_norm": 23.70060157775879, "learning_rate": 9.97818724639432e-06, "loss": 12.215085983276367, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.26062000915408134, "loss_total_v6": 0.7634428963065147, "step": 432 }, { "epoch": 0.09986306306306306, "grad_norm": 25.267507553100586, "learning_rate": 9.977998938476059e-06, "loss": 12.7860746383667, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2840387411415577, "loss_total_v6": 0.7991296872496605, "step": 433 }, { "epoch": 0.1000936936936937, "grad_norm": 24.431865692138672, "learning_rate": 9.9778098230154e-06, "loss": 12.876167297363281, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.2842221427708864, "loss_total_v6": 0.8047604411840439, "step": 434 }, { "epoch": 0.10032432432432432, "grad_norm": 26.13159942626953, "learning_rate": 9.977619900043021e-06, "loss": 13.622238159179688, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.3336286433041096, "loss_total_v6": 0.8513898849487305, "step": 435 }, { "epoch": 0.10055495495495495, "grad_norm": 26.844207763671875, "learning_rate": 9.977429169589731e-06, "loss": 13.220135688781738, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.3043926730751991, "loss_total_v6": 0.8262584805488586, "step": 436 }, { "epoch": 0.10078558558558559, "grad_norm": 37.350826263427734, "learning_rate": 9.977237631686472e-06, "loss": 12.921304702758789, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.28399150259792805, "loss_total_v6": 0.8075815811753273, "step": 437 }, { "epoch": 0.10101621621621622, "grad_norm": 26.511051177978516, "learning_rate": 9.977045286364314e-06, "loss": 12.61196517944336, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2762697022408247, "loss_total_v6": 0.7882478684186935, "step": 438 }, { "epoch": 0.10124684684684684, "grad_norm": 26.451120376586914, "learning_rate": 9.97685213365446e-06, "loss": 12.467689514160156, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.263742933049798, "loss_total_v6": 0.7792306318879128, "step": 439 }, { "epoch": 0.10147747747747747, "grad_norm": 25.49945068359375, "learning_rate": 9.976658173588244e-06, "loss": 12.359931945800781, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.26044133119285107, "loss_total_v6": 0.7724957689642906, "step": 440 }, { "epoch": 0.10170810810810811, "grad_norm": 25.64006996154785, "learning_rate": 9.97646340619713e-06, "loss": 12.925863265991211, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.2820943742990494, "loss_total_v6": 0.8078664913773537, "step": 441 }, { "epoch": 0.10193873873873874, "grad_norm": 28.66044807434082, "learning_rate": 9.976267831512716e-06, "loss": 12.116167068481445, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.23246486112475395, "loss_total_v6": 0.757260374724865, "step": 442 }, { "epoch": 0.10216936936936936, "grad_norm": 25.620702743530273, "learning_rate": 9.976071449566726e-06, "loss": 12.698037147521973, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2827478051185608, "loss_total_v6": 0.7936273217201233, "step": 443 }, { "epoch": 0.1024, "grad_norm": 24.020700454711914, "learning_rate": 9.975874260391019e-06, "loss": 12.959928512573242, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.2935770358890295, "loss_total_v6": 0.8099954947829247, "step": 444 }, { "epoch": 0.10263063063063063, "grad_norm": 23.74622917175293, "learning_rate": 9.975676264017582e-06, "loss": 12.506309509277344, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.26273876056075096, "loss_total_v6": 0.7816444039344788, "step": 445 }, { "epoch": 0.10286126126126126, "grad_norm": 25.433124542236328, "learning_rate": 9.975477460478537e-06, "loss": 12.6243896484375, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2716140877455473, "loss_total_v6": 0.7890243753790855, "step": 446 }, { "epoch": 0.1030918918918919, "grad_norm": 25.27539825439453, "learning_rate": 9.975277849806133e-06, "loss": 12.733757972717285, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.2807078715413809, "loss_total_v6": 0.7958598434925079, "step": 447 }, { "epoch": 0.10332252252252253, "grad_norm": 26.370161056518555, "learning_rate": 9.975077432032749e-06, "loss": 13.046808242797852, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.2934070788323879, "loss_total_v6": 0.8154255300760269, "step": 448 }, { "epoch": 0.10355315315315315, "grad_norm": 24.649459838867188, "learning_rate": 9.9748762071909e-06, "loss": 12.697214126586914, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.2751432452350855, "loss_total_v6": 0.7935758978128433, "step": 449 }, { "epoch": 0.10378378378378378, "grad_norm": 23.58873176574707, "learning_rate": 9.974674175313228e-06, "loss": 12.975317001342773, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.2855972144752741, "loss_total_v6": 0.8109573051333427, "step": 450 }, { "epoch": 0.10401441441441442, "grad_norm": 25.634218215942383, "learning_rate": 9.97447133643251e-06, "loss": 12.543932914733887, "loss_alignment": 1.04248046875, "loss_alignment_w": 0.521240234375, "loss_sft": 0.26251143030822277, "loss_total_v6": 0.7839958071708679, "step": 451 }, { "epoch": 0.10424504504504505, "grad_norm": 25.697792053222656, "learning_rate": 9.974267690581646e-06, "loss": 12.579965591430664, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.26815097965300083, "loss_total_v6": 0.7862478792667389, "step": 452 }, { "epoch": 0.10447567567567567, "grad_norm": 26.819292068481445, "learning_rate": 9.974063237793674e-06, "loss": 12.808547973632812, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.27926351875066757, "loss_total_v6": 0.800534263253212, "step": 453 }, { "epoch": 0.10470630630630631, "grad_norm": 25.386676788330078, "learning_rate": 9.973857978101762e-06, "loss": 12.671823501586914, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.27914105728268623, "loss_total_v6": 0.7919889688491821, "step": 454 }, { "epoch": 0.10493693693693694, "grad_norm": 24.41870880126953, "learning_rate": 9.973651911539209e-06, "loss": 13.011240005493164, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.29959162697196007, "loss_total_v6": 0.8132024630904198, "step": 455 }, { "epoch": 0.10516756756756757, "grad_norm": 25.15342140197754, "learning_rate": 9.973445038139437e-06, "loss": 12.420639038085938, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.26293848641216755, "loss_total_v6": 0.7762899473309517, "step": 456 }, { "epoch": 0.10539819819819819, "grad_norm": 25.641569137573242, "learning_rate": 9.973237357936014e-06, "loss": 12.761083602905273, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2841705195605755, "loss_total_v6": 0.797567717730999, "step": 457 }, { "epoch": 0.10562882882882883, "grad_norm": 25.7365779876709, "learning_rate": 9.973028870962625e-06, "loss": 12.478398323059082, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.26326778531074524, "loss_total_v6": 0.7798998802900314, "step": 458 }, { "epoch": 0.10585945945945946, "grad_norm": 25.91276741027832, "learning_rate": 9.972819577253092e-06, "loss": 13.238592147827148, "loss_alignment": 1.03271484375, "loss_alignment_w": 0.516357421875, "loss_sft": 0.3100170008838177, "loss_total_v6": 0.827412024140358, "step": 459 }, { "epoch": 0.10609009009009009, "grad_norm": 24.56707191467285, "learning_rate": 9.972609476841368e-06, "loss": 13.009156227111816, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.28816989436745644, "loss_total_v6": 0.8130722492933273, "step": 460 }, { "epoch": 0.10632072072072073, "grad_norm": 25.93621253967285, "learning_rate": 9.972398569761534e-06, "loss": 12.85817813873291, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2927718423306942, "loss_total_v6": 0.8036361262202263, "step": 461 }, { "epoch": 0.10655135135135135, "grad_norm": 23.274457931518555, "learning_rate": 9.972186856047808e-06, "loss": 12.038196563720703, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2378914337605238, "loss_total_v6": 0.7523872926831245, "step": 462 }, { "epoch": 0.10678198198198198, "grad_norm": 23.27049446105957, "learning_rate": 9.971974335734532e-06, "loss": 12.750246047973633, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.278640853241086, "loss_total_v6": 0.7968903705477715, "step": 463 }, { "epoch": 0.1070126126126126, "grad_norm": 23.86429786682129, "learning_rate": 9.97176100885618e-06, "loss": 12.72149658203125, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2850227113813162, "loss_total_v6": 0.7950935140252113, "step": 464 }, { "epoch": 0.10724324324324325, "grad_norm": 28.080318450927734, "learning_rate": 9.97154687544736e-06, "loss": 12.794384002685547, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.2782713808119297, "loss_total_v6": 0.7996489778161049, "step": 465 }, { "epoch": 0.10747387387387387, "grad_norm": 24.91501235961914, "learning_rate": 9.97133193554281e-06, "loss": 12.602952003479004, "loss_alignment": 1.02490234375, "loss_alignment_w": 0.512451171875, "loss_sft": 0.2752333376556635, "loss_total_v6": 0.7876845151185989, "step": 466 }, { "epoch": 0.1077045045045045, "grad_norm": 25.558374404907227, "learning_rate": 9.9711161891774e-06, "loss": 13.172500610351562, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.3025140333920717, "loss_total_v6": 0.8232812434434891, "step": 467 }, { "epoch": 0.10793513513513514, "grad_norm": 28.42300033569336, "learning_rate": 9.970899636386121e-06, "loss": 12.942651748657227, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.287278825417161, "loss_total_v6": 0.8089157566428185, "step": 468 }, { "epoch": 0.10816576576576577, "grad_norm": 23.407133102416992, "learning_rate": 9.97068227720411e-06, "loss": 12.150583267211914, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.24952376261353493, "loss_total_v6": 0.7594114542007446, "step": 469 }, { "epoch": 0.1083963963963964, "grad_norm": 28.66659164428711, "learning_rate": 9.970464111666627e-06, "loss": 13.534381866455078, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.3202487640082836, "loss_total_v6": 0.8458988070487976, "step": 470 }, { "epoch": 0.10862702702702702, "grad_norm": 23.014209747314453, "learning_rate": 9.97024513980906e-06, "loss": 12.872102737426758, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.28738607466220856, "loss_total_v6": 0.804506428539753, "step": 471 }, { "epoch": 0.10885765765765766, "grad_norm": 24.248899459838867, "learning_rate": 9.970025361666934e-06, "loss": 12.888838768005371, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2963360957801342, "loss_total_v6": 0.8055524080991745, "step": 472 }, { "epoch": 0.10908828828828829, "grad_norm": 28.174652099609375, "learning_rate": 9.9698047772759e-06, "loss": 13.029820442199707, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.2882712483406067, "loss_total_v6": 0.8143637627363205, "step": 473 }, { "epoch": 0.10931891891891891, "grad_norm": 25.081531524658203, "learning_rate": 9.969583386671742e-06, "loss": 12.955127716064453, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2918885052204132, "loss_total_v6": 0.8096955046057701, "step": 474 }, { "epoch": 0.10954954954954955, "grad_norm": 22.469778060913086, "learning_rate": 9.969361189890373e-06, "loss": 12.303184509277344, "loss_alignment": 1.02099609375, "loss_alignment_w": 0.510498046875, "loss_sft": 0.25965639762580395, "loss_total_v6": 0.7689490169286728, "step": 475 }, { "epoch": 0.10978018018018018, "grad_norm": 24.467241287231445, "learning_rate": 9.969138186967844e-06, "loss": 12.336139678955078, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.26235699839890003, "loss_total_v6": 0.7710087224841118, "step": 476 }, { "epoch": 0.11001081081081081, "grad_norm": 26.22158432006836, "learning_rate": 9.968914377940326e-06, "loss": 12.3512601852417, "loss_alignment": 1.03564453125, "loss_alignment_w": 0.517822265625, "loss_sft": 0.25457407906651497, "loss_total_v6": 0.7719537913799286, "step": 477 }, { "epoch": 0.11024144144144143, "grad_norm": 25.005184173583984, "learning_rate": 9.968689762844127e-06, "loss": 13.022503852844238, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2988308146595955, "loss_total_v6": 0.8139064684510231, "step": 478 }, { "epoch": 0.11047207207207208, "grad_norm": 25.588212966918945, "learning_rate": 9.968464341715684e-06, "loss": 12.509648323059082, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.2595141548663378, "loss_total_v6": 0.7818530276417732, "step": 479 }, { "epoch": 0.1107027027027027, "grad_norm": 25.45505142211914, "learning_rate": 9.968238114591567e-06, "loss": 12.760970115661621, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.2708424925804138, "loss_total_v6": 0.7975606173276901, "step": 480 }, { "epoch": 0.11093333333333333, "grad_norm": 27.34043312072754, "learning_rate": 9.968011081508471e-06, "loss": 12.746341705322266, "loss_alignment": 1.01611328125, "loss_alignment_w": 0.508056640625, "loss_sft": 0.28910846821963787, "loss_total_v6": 0.796646311879158, "step": 481 }, { "epoch": 0.11116396396396397, "grad_norm": 24.632646560668945, "learning_rate": 9.967783242503232e-06, "loss": 13.228341102600098, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.30551582388579845, "loss_total_v6": 0.8267713040113449, "step": 482 }, { "epoch": 0.1113945945945946, "grad_norm": 28.72916030883789, "learning_rate": 9.967554597612807e-06, "loss": 13.038410186767578, "loss_alignment": 1.03271484375, "loss_alignment_w": 0.516357421875, "loss_sft": 0.29836015962064266, "loss_total_v6": 0.814900703728199, "step": 483 }, { "epoch": 0.11162522522522522, "grad_norm": 25.809614181518555, "learning_rate": 9.967325146874287e-06, "loss": 12.830110549926758, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.279329439625144, "loss_total_v6": 0.8018819242715836, "step": 484 }, { "epoch": 0.11185585585585586, "grad_norm": 23.871843338012695, "learning_rate": 9.967094890324894e-06, "loss": 12.810173034667969, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.28450724482536316, "loss_total_v6": 0.8006357923150063, "step": 485 }, { "epoch": 0.11208648648648649, "grad_norm": 24.609878540039062, "learning_rate": 9.966863828001982e-06, "loss": 12.981636047363281, "loss_alignment": 1.03076171875, "loss_alignment_w": 0.515380859375, "loss_sft": 0.2962002754211426, "loss_total_v6": 0.8113522604107857, "step": 486 }, { "epoch": 0.11231711711711712, "grad_norm": 36.355472564697266, "learning_rate": 9.966631959943036e-06, "loss": 12.485045433044434, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.27408977411687374, "loss_total_v6": 0.7803153395652771, "step": 487 }, { "epoch": 0.11254774774774774, "grad_norm": 24.185213088989258, "learning_rate": 9.966399286185666e-06, "loss": 12.907266616821289, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.29031617753207684, "loss_total_v6": 0.8067041337490082, "step": 488 }, { "epoch": 0.11277837837837838, "grad_norm": 26.45436668395996, "learning_rate": 9.966165806767619e-06, "loss": 13.323471069335938, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.3136434778571129, "loss_total_v6": 0.8327169567346573, "step": 489 }, { "epoch": 0.11300900900900901, "grad_norm": 24.941139221191406, "learning_rate": 9.965931521726771e-06, "loss": 12.612738609313965, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2790493071079254, "loss_total_v6": 0.788296140730381, "step": 490 }, { "epoch": 0.11323963963963964, "grad_norm": 26.653406143188477, "learning_rate": 9.965696431101128e-06, "loss": 12.47503662109375, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2646140977740288, "loss_total_v6": 0.7796897664666176, "step": 491 }, { "epoch": 0.11347027027027028, "grad_norm": 24.195295333862305, "learning_rate": 9.965460534928827e-06, "loss": 12.751035690307617, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.2763098683208227, "loss_total_v6": 0.7969397529959679, "step": 492 }, { "epoch": 0.1137009009009009, "grad_norm": 22.938182830810547, "learning_rate": 9.965223833248135e-06, "loss": 12.972322463989258, "loss_alignment": 1.04443359375, "loss_alignment_w": 0.522216796875, "loss_sft": 0.2872410714626312, "loss_total_v6": 0.8107701241970062, "step": 493 }, { "epoch": 0.11393153153153153, "grad_norm": 25.58162498474121, "learning_rate": 9.96498632609745e-06, "loss": 12.76118278503418, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2852905709296465, "loss_total_v6": 0.7975738868117332, "step": 494 }, { "epoch": 0.11416216216216216, "grad_norm": 28.830419540405273, "learning_rate": 9.964748013515301e-06, "loss": 12.562748908996582, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.26301607117056847, "loss_total_v6": 0.7851718217134476, "step": 495 }, { "epoch": 0.1143927927927928, "grad_norm": 24.98674774169922, "learning_rate": 9.96450889554035e-06, "loss": 12.87803840637207, "loss_alignment": 1.03466796875, "loss_alignment_w": 0.517333984375, "loss_sft": 0.2861701287329197, "loss_total_v6": 0.8048774003982544, "step": 496 }, { "epoch": 0.11462342342342342, "grad_norm": 25.032114028930664, "learning_rate": 9.964268972211386e-06, "loss": 13.27956485748291, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.30951080471277237, "loss_total_v6": 0.8299728184938431, "step": 497 }, { "epoch": 0.11485405405405405, "grad_norm": 26.453725814819336, "learning_rate": 9.964028243567327e-06, "loss": 13.283608436584473, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.3043618779629469, "loss_total_v6": 0.8302255198359489, "step": 498 }, { "epoch": 0.11508468468468469, "grad_norm": 27.21307373046875, "learning_rate": 9.963786709647228e-06, "loss": 13.032166481018066, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.3053856622427702, "loss_total_v6": 0.8145104274153709, "step": 499 }, { "epoch": 0.11531531531531532, "grad_norm": 26.143503189086914, "learning_rate": 9.96354437049027e-06, "loss": 12.65481185913086, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.281129565089941, "loss_total_v6": 0.7909257188439369, "step": 500 }, { "epoch": 0.11531531531531532, "eval_loss": 0.9442645907402039, "eval_loss_alignment": 1.0390112193207763, "eval_loss_alignment_w": 0.5195056096603882, "eval_loss_contrast": 0.0, "eval_loss_highlevel": 0.0, "eval_loss_sft": 0.42475900492847785, "eval_loss_total_v6": 0.9442646150991797, "eval_loss_wm_recon": 0.0, "eval_loss_wm_recon_w": 0.0, "eval_runtime": 348.7351, "eval_samples_per_second": 5.004, "eval_steps_per_second": 0.628, "step": 500 }, { "epoch": 0.11554594594594594, "grad_norm": 24.367727279663086, "learning_rate": 9.963301226135767e-06, "loss": 12.849960327148438, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.3003301862627268, "loss_total_v6": 0.8031225502490997, "step": 501 }, { "epoch": 0.11577657657657657, "grad_norm": 26.226634979248047, "learning_rate": 9.96305727662316e-06, "loss": 12.268131256103516, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.24063522182404995, "loss_total_v6": 0.7667582407593727, "step": 502 }, { "epoch": 0.11600720720720721, "grad_norm": 24.092344284057617, "learning_rate": 9.962812521992026e-06, "loss": 12.73026180267334, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2785972896963358, "loss_total_v6": 0.7956413701176643, "step": 503 }, { "epoch": 0.11623783783783784, "grad_norm": 23.22370147705078, "learning_rate": 9.962566962282068e-06, "loss": 12.185972213745117, "loss_alignment": 1.03662109375, "loss_alignment_w": 0.518310546875, "loss_sft": 0.2445944808423519, "loss_total_v6": 0.7616233080625534, "step": 504 }, { "epoch": 0.11646846846846846, "grad_norm": 23.96076011657715, "learning_rate": 9.962320597533119e-06, "loss": 12.908602714538574, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2927190624177456, "loss_total_v6": 0.8067876771092415, "step": 505 }, { "epoch": 0.1166990990990991, "grad_norm": 26.59017562866211, "learning_rate": 9.962073427785149e-06, "loss": 13.494431495666504, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.3166685216128826, "loss_total_v6": 0.8434019461274147, "step": 506 }, { "epoch": 0.11692972972972973, "grad_norm": 27.915149688720703, "learning_rate": 9.961825453078254e-06, "loss": 12.745006561279297, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.2694327440112829, "loss_total_v6": 0.7965628802776337, "step": 507 }, { "epoch": 0.11716036036036036, "grad_norm": 24.36427879333496, "learning_rate": 9.961576673452656e-06, "loss": 12.356354713439941, "loss_alignment": 1.03857421875, "loss_alignment_w": 0.519287109375, "loss_sft": 0.25495349057018757, "loss_total_v6": 0.7722721844911575, "step": 508 }, { "epoch": 0.11739099099099098, "grad_norm": 24.81569480895996, "learning_rate": 9.961327088948718e-06, "loss": 12.52717399597168, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2702530976384878, "loss_total_v6": 0.782948412001133, "step": 509 }, { "epoch": 0.11762162162162162, "grad_norm": 27.2630558013916, "learning_rate": 9.961076699606927e-06, "loss": 12.493364334106445, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.25590234994888306, "loss_total_v6": 0.7808352336287498, "step": 510 }, { "epoch": 0.11785225225225225, "grad_norm": 24.73326873779297, "learning_rate": 9.960825505467902e-06, "loss": 13.154182434082031, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.29758501425385475, "loss_total_v6": 0.8221364095807076, "step": 511 }, { "epoch": 0.11808288288288288, "grad_norm": 25.463603973388672, "learning_rate": 9.960573506572391e-06, "loss": 12.487672805786133, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.26796735264360905, "loss_total_v6": 0.7804795429110527, "step": 512 }, { "epoch": 0.11831351351351352, "grad_norm": 27.720609664916992, "learning_rate": 9.960320702961272e-06, "loss": 12.996042251586914, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.29281291738152504, "loss_total_v6": 0.8122526183724403, "step": 513 }, { "epoch": 0.11854414414414415, "grad_norm": 23.073997497558594, "learning_rate": 9.96006709467556e-06, "loss": 12.270391464233398, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2548145204782486, "loss_total_v6": 0.7668994963169098, "step": 514 }, { "epoch": 0.11877477477477477, "grad_norm": 37.4439697265625, "learning_rate": 9.959812681756394e-06, "loss": 12.747224807739258, "loss_alignment": 1.04248046875, "loss_alignment_w": 0.521240234375, "loss_sft": 0.2759800720959902, "loss_total_v6": 0.7967015281319618, "step": 515 }, { "epoch": 0.11900540540540541, "grad_norm": 25.655426025390625, "learning_rate": 9.959557464245043e-06, "loss": 12.520910263061523, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.27114326879382133, "loss_total_v6": 0.782556876540184, "step": 516 }, { "epoch": 0.11923603603603604, "grad_norm": 24.122783660888672, "learning_rate": 9.959301442182912e-06, "loss": 12.483880996704102, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.2613369096070528, "loss_total_v6": 0.7802425771951675, "step": 517 }, { "epoch": 0.11946666666666667, "grad_norm": 23.750810623168945, "learning_rate": 9.959044615611533e-06, "loss": 12.948976516723633, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.29495254158973694, "loss_total_v6": 0.8093110397458076, "step": 518 }, { "epoch": 0.11969729729729729, "grad_norm": 26.386672973632812, "learning_rate": 9.958786984572568e-06, "loss": 12.599016189575195, "loss_alignment": 1.03466796875, "loss_alignment_w": 0.517333984375, "loss_sft": 0.2717676777392626, "loss_total_v6": 0.7874384820461273, "step": 519 }, { "epoch": 0.11992792792792793, "grad_norm": 27.099491119384766, "learning_rate": 9.958528549107812e-06, "loss": 13.069467544555664, "loss_alignment": 1.02294921875, "loss_alignment_w": 0.511474609375, "loss_sft": 0.3054892234504223, "loss_total_v6": 0.8168417513370514, "step": 520 }, { "epoch": 0.12015855855855856, "grad_norm": 24.893169403076172, "learning_rate": 9.958269309259186e-06, "loss": 12.330387115478516, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2640879452228546, "loss_total_v6": 0.7706492394208908, "step": 521 }, { "epoch": 0.12038918918918919, "grad_norm": 27.23513412475586, "learning_rate": 9.958009265068747e-06, "loss": 12.875131607055664, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.2835469897836447, "loss_total_v6": 0.8046957179903984, "step": 522 }, { "epoch": 0.12061981981981983, "grad_norm": 24.776756286621094, "learning_rate": 9.95774841657868e-06, "loss": 12.95482063293457, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.28469768166542053, "loss_total_v6": 0.8096763342618942, "step": 523 }, { "epoch": 0.12085045045045045, "grad_norm": 22.728933334350586, "learning_rate": 9.9574867638313e-06, "loss": 12.942953109741211, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2942556254565716, "loss_total_v6": 0.8089345842599869, "step": 524 }, { "epoch": 0.12108108108108108, "grad_norm": 24.05179786682129, "learning_rate": 9.957224306869053e-06, "loss": 12.791915893554688, "loss_alignment": 1.03955078125, "loss_alignment_w": 0.519775390625, "loss_sft": 0.2797650769352913, "loss_total_v6": 0.7994947284460068, "step": 525 }, { "epoch": 0.1213117117117117, "grad_norm": 28.712594985961914, "learning_rate": 9.956961045734514e-06, "loss": 12.526187896728516, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2669108249247074, "loss_total_v6": 0.7828867733478546, "step": 526 }, { "epoch": 0.12154234234234235, "grad_norm": 23.803707122802734, "learning_rate": 9.956696980470392e-06, "loss": 13.050599098205566, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.29756551422178745, "loss_total_v6": 0.8156624436378479, "step": 527 }, { "epoch": 0.12177297297297297, "grad_norm": 27.961654663085938, "learning_rate": 9.956432111119522e-06, "loss": 12.640231132507324, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.28887002170085907, "loss_total_v6": 0.7900144308805466, "step": 528 }, { "epoch": 0.1220036036036036, "grad_norm": 26.196334838867188, "learning_rate": 9.956166437724875e-06, "loss": 12.483213424682617, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.28213868476450443, "loss_total_v6": 0.7802008539438248, "step": 529 }, { "epoch": 0.12223423423423424, "grad_norm": 24.495574951171875, "learning_rate": 9.955899960329546e-06, "loss": 12.390857696533203, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.26621939428150654, "loss_total_v6": 0.7744286209344864, "step": 530 }, { "epoch": 0.12246486486486487, "grad_norm": 24.080110549926758, "learning_rate": 9.955632678976765e-06, "loss": 12.432121276855469, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.26049756817519665, "loss_total_v6": 0.7770075500011444, "step": 531 }, { "epoch": 0.1226954954954955, "grad_norm": 21.549753189086914, "learning_rate": 9.95536459370989e-06, "loss": 12.12384033203125, "loss_alignment": 1.01416015625, "loss_alignment_w": 0.507080078125, "loss_sft": 0.25052259862422943, "loss_total_v6": 0.7577399909496307, "step": 532 }, { "epoch": 0.12292612612612612, "grad_norm": 22.790143966674805, "learning_rate": 9.955095704572412e-06, "loss": 12.161748886108398, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2549365255981684, "loss_total_v6": 0.7601092830300331, "step": 533 }, { "epoch": 0.12315675675675676, "grad_norm": 26.23153305053711, "learning_rate": 9.954826011607949e-06, "loss": 12.938619613647461, "loss_alignment": 1.01806640625, "loss_alignment_w": 0.509033203125, "loss_sft": 0.2987455688416958, "loss_total_v6": 0.8086637631058693, "step": 534 }, { "epoch": 0.12338738738738739, "grad_norm": 27.89885711669922, "learning_rate": 9.954555514860254e-06, "loss": 13.03219985961914, "loss_alignment": 1.02685546875, "loss_alignment_w": 0.513427734375, "loss_sft": 0.30081011913716793, "loss_total_v6": 0.8145124986767769, "step": 535 }, { "epoch": 0.12361801801801801, "grad_norm": 23.338523864746094, "learning_rate": 9.954284214373204e-06, "loss": 11.840544700622559, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.23139757104218006, "loss_total_v6": 0.7400340363383293, "step": 536 }, { "epoch": 0.12384864864864865, "grad_norm": 24.284557342529297, "learning_rate": 9.954012110190812e-06, "loss": 12.5368070602417, "loss_alignment": 1.01220703125, "loss_alignment_w": 0.506103515625, "loss_sft": 0.2747766338288784, "loss_total_v6": 0.7835504412651062, "step": 537 }, { "epoch": 0.12407927927927928, "grad_norm": 26.10140037536621, "learning_rate": 9.953739202357219e-06, "loss": 13.316122055053711, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.31113952212035656, "loss_total_v6": 0.8322576880455017, "step": 538 }, { "epoch": 0.12430990990990991, "grad_norm": 23.24429702758789, "learning_rate": 9.953465490916696e-06, "loss": 12.026876449584961, "loss_alignment": 1.02294921875, "loss_alignment_w": 0.511474609375, "loss_sft": 0.2402051743119955, "loss_total_v6": 0.7516797855496407, "step": 539 }, { "epoch": 0.12454054054054053, "grad_norm": 23.460493087768555, "learning_rate": 9.953190975913648e-06, "loss": 12.83413028717041, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.28772884234786034, "loss_total_v6": 0.8021331280469894, "step": 540 }, { "epoch": 0.12477117117117117, "grad_norm": 28.78046226501465, "learning_rate": 9.952915657392603e-06, "loss": 12.419502258300781, "loss_alignment": 1.02880859375, "loss_alignment_w": 0.514404296875, "loss_sft": 0.2625775095075369, "loss_total_v6": 0.7762188836932182, "step": 541 }, { "epoch": 0.12500180180180182, "grad_norm": 24.20979118347168, "learning_rate": 9.952639535398227e-06, "loss": 12.432543754577637, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.27042698860168457, "loss_total_v6": 0.7770339921116829, "step": 542 }, { "epoch": 0.12523243243243243, "grad_norm": 25.16596031188965, "learning_rate": 9.952362609975313e-06, "loss": 12.74860954284668, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2848404571413994, "loss_total_v6": 0.7967880964279175, "step": 543 }, { "epoch": 0.12546306306306307, "grad_norm": 25.63142967224121, "learning_rate": 9.952084881168784e-06, "loss": 12.959830284118652, "loss_alignment": 1.01806640625, "loss_alignment_w": 0.509033203125, "loss_sft": 0.3006967809051275, "loss_total_v6": 0.8099893778562546, "step": 544 }, { "epoch": 0.12569369369369368, "grad_norm": 24.463350296020508, "learning_rate": 9.951806349023693e-06, "loss": 12.692367553710938, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.29104519449174404, "loss_total_v6": 0.7932729944586754, "step": 545 }, { "epoch": 0.12592432432432432, "grad_norm": 25.729496002197266, "learning_rate": 9.951527013585224e-06, "loss": 12.91214370727539, "loss_alignment": 1.02197265625, "loss_alignment_w": 0.510986328125, "loss_sft": 0.29628212936222553, "loss_total_v6": 0.8070090189576149, "step": 546 }, { "epoch": 0.12615495495495496, "grad_norm": 23.19234848022461, "learning_rate": 9.951246874898693e-06, "loss": 13.191242218017578, "loss_alignment": 1.03173828125, "loss_alignment_w": 0.515869140625, "loss_sft": 0.308553047478199, "loss_total_v6": 0.8244526609778404, "step": 547 }, { "epoch": 0.12638558558558557, "grad_norm": 25.43962287902832, "learning_rate": 9.950965933009544e-06, "loss": 12.819999694824219, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.27572198770940304, "loss_total_v6": 0.8012499511241913, "step": 548 }, { "epoch": 0.12661621621621622, "grad_norm": 24.45639991760254, "learning_rate": 9.95068418796335e-06, "loss": 12.720335006713867, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.27782423235476017, "loss_total_v6": 0.7950208857655525, "step": 549 }, { "epoch": 0.12684684684684686, "grad_norm": 22.281585693359375, "learning_rate": 9.950401639805822e-06, "loss": 12.399209022521973, "loss_alignment": 1.02783203125, "loss_alignment_w": 0.513916015625, "loss_sft": 0.26097350753843784, "loss_total_v6": 0.7749505639076233, "step": 550 }, { "epoch": 0.12707747747747747, "grad_norm": 21.927082061767578, "learning_rate": 9.95011828858279e-06, "loss": 12.599535942077637, "loss_alignment": 1.02099609375, "loss_alignment_w": 0.510498046875, "loss_sft": 0.2766830325126648, "loss_total_v6": 0.7874709740281105, "step": 551 }, { "epoch": 0.1273081081081081, "grad_norm": 24.27582359313965, "learning_rate": 9.949834134340219e-06, "loss": 12.769234657287598, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.29613926261663437, "loss_total_v6": 0.7980771660804749, "step": 552 }, { "epoch": 0.12753873873873875, "grad_norm": 26.879119873046875, "learning_rate": 9.94954917712421e-06, "loss": 12.731847763061523, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2777045760303736, "loss_total_v6": 0.7957404851913452, "step": 553 }, { "epoch": 0.12776936936936936, "grad_norm": 26.083024978637695, "learning_rate": 9.94926341698099e-06, "loss": 12.709183692932129, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.27920249477028847, "loss_total_v6": 0.7943239733576775, "step": 554 }, { "epoch": 0.128, "grad_norm": 23.5715274810791, "learning_rate": 9.948976853956908e-06, "loss": 12.455053329467773, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2748245298862457, "loss_total_v6": 0.7784408405423164, "step": 555 }, { "epoch": 0.12823063063063064, "grad_norm": 24.02121925354004, "learning_rate": 9.94868948809846e-06, "loss": 12.642410278320312, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.281071649864316, "loss_total_v6": 0.7901506200432777, "step": 556 }, { "epoch": 0.12846126126126126, "grad_norm": 25.017065048217773, "learning_rate": 9.948401319452256e-06, "loss": 12.64433479309082, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.27853688038885593, "loss_total_v6": 0.7902708947658539, "step": 557 }, { "epoch": 0.1286918918918919, "grad_norm": 25.061304092407227, "learning_rate": 9.948112348065049e-06, "loss": 13.087952613830566, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.29114156402647495, "loss_total_v6": 0.8179970532655716, "step": 558 }, { "epoch": 0.1289225225225225, "grad_norm": 23.611318588256836, "learning_rate": 9.947822573983711e-06, "loss": 12.177079200744629, "loss_alignment": 1.02099609375, "loss_alignment_w": 0.510498046875, "loss_sft": 0.251027163118124, "loss_total_v6": 0.7610674574971199, "step": 559 }, { "epoch": 0.12915315315315315, "grad_norm": 24.822126388549805, "learning_rate": 9.947531997255256e-06, "loss": 12.519229888916016, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2708246409893036, "loss_total_v6": 0.7824518457055092, "step": 560 }, { "epoch": 0.1293837837837838, "grad_norm": 25.345417022705078, "learning_rate": 9.947240617926818e-06, "loss": 12.414539337158203, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.25913933478295803, "loss_total_v6": 0.7759087532758713, "step": 561 }, { "epoch": 0.1296144144144144, "grad_norm": 24.582727432250977, "learning_rate": 9.946948436045664e-06, "loss": 12.559910774230957, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.26393728144466877, "loss_total_v6": 0.7849944084882736, "step": 562 }, { "epoch": 0.12984504504504504, "grad_norm": 27.508033752441406, "learning_rate": 9.946655451659197e-06, "loss": 12.693831443786621, "loss_alignment": 1.02392578125, "loss_alignment_w": 0.511962890625, "loss_sft": 0.28039449639618397, "loss_total_v6": 0.7933644652366638, "step": 563 }, { "epoch": 0.13007567567567568, "grad_norm": 23.069963455200195, "learning_rate": 9.946361664814942e-06, "loss": 12.590835571289062, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.2696695476770401, "loss_total_v6": 0.78692726790905, "step": 564 }, { "epoch": 0.1303063063063063, "grad_norm": 27.444169998168945, "learning_rate": 9.94606707556056e-06, "loss": 13.152751922607422, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.3111521601676941, "loss_total_v6": 0.8220469504594803, "step": 565 }, { "epoch": 0.13053693693693694, "grad_norm": 27.264873504638672, "learning_rate": 9.945771683943837e-06, "loss": 12.742999076843262, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.28636664524674416, "loss_total_v6": 0.7964374423027039, "step": 566 }, { "epoch": 0.13076756756756758, "grad_norm": 24.087949752807617, "learning_rate": 9.945475490012695e-06, "loss": 12.243782043457031, "loss_alignment": 1.02197265625, "loss_alignment_w": 0.510986328125, "loss_sft": 0.255135053768754, "loss_total_v6": 0.7652363628149033, "step": 567 }, { "epoch": 0.1309981981981982, "grad_norm": 30.881187438964844, "learning_rate": 9.945178493815181e-06, "loss": 12.933944702148438, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.3002844490110874, "loss_total_v6": 0.8083715960383415, "step": 568 }, { "epoch": 0.13122882882882883, "grad_norm": 25.42133903503418, "learning_rate": 9.944880695399476e-06, "loss": 12.986190795898438, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.2923955526202917, "loss_total_v6": 0.8116368800401688, "step": 569 }, { "epoch": 0.13145945945945947, "grad_norm": 25.498416900634766, "learning_rate": 9.94458209481389e-06, "loss": 12.87701416015625, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.2819251995533705, "loss_total_v6": 0.804813377559185, "step": 570 }, { "epoch": 0.13169009009009008, "grad_norm": 25.35660171508789, "learning_rate": 9.944282692106862e-06, "loss": 12.413614273071289, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2640253324061632, "loss_total_v6": 0.7758509069681168, "step": 571 }, { "epoch": 0.13192072072072072, "grad_norm": 23.83200454711914, "learning_rate": 9.94398248732696e-06, "loss": 12.527137756347656, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2708153333514929, "loss_total_v6": 0.7829460576176643, "step": 572 }, { "epoch": 0.13215135135135136, "grad_norm": 24.181114196777344, "learning_rate": 9.943681480522886e-06, "loss": 12.517047882080078, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.27139024063944817, "loss_total_v6": 0.7823155000805855, "step": 573 }, { "epoch": 0.13238198198198198, "grad_norm": 24.927688598632812, "learning_rate": 9.943379671743471e-06, "loss": 12.70882511138916, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.27303084544837475, "loss_total_v6": 0.7943015769124031, "step": 574 }, { "epoch": 0.13261261261261262, "grad_norm": 25.325332641601562, "learning_rate": 9.943077061037672e-06, "loss": 13.501618385314941, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.32225996628403664, "loss_total_v6": 0.8438511416316032, "step": 575 }, { "epoch": 0.13284324324324323, "grad_norm": 27.32435417175293, "learning_rate": 9.94277364845458e-06, "loss": 12.961523056030273, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2966826856136322, "loss_total_v6": 0.8100951537489891, "step": 576 }, { "epoch": 0.13307387387387387, "grad_norm": 24.190614700317383, "learning_rate": 9.942469434043418e-06, "loss": 12.751938819885254, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.27476409636437893, "loss_total_v6": 0.796996146440506, "step": 577 }, { "epoch": 0.1333045045045045, "grad_norm": 22.50294303894043, "learning_rate": 9.942164417853534e-06, "loss": 12.70851993560791, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.2840285897254944, "loss_total_v6": 0.7942824959754944, "step": 578 }, { "epoch": 0.13353513513513512, "grad_norm": 26.355436325073242, "learning_rate": 9.94185859993441e-06, "loss": 12.780539512634277, "loss_alignment": 1.0576171875, "loss_alignment_w": 0.52880859375, "loss_sft": 0.2707533333450556, "loss_total_v6": 0.7987837046384811, "step": 579 }, { "epoch": 0.13376576576576577, "grad_norm": 25.781787872314453, "learning_rate": 9.941551980335653e-06, "loss": 13.146215438842773, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.30810393765568733, "loss_total_v6": 0.8216384649276733, "step": 580 }, { "epoch": 0.1339963963963964, "grad_norm": 25.788381576538086, "learning_rate": 9.941244559107007e-06, "loss": 12.348419189453125, "loss_alignment": 1.02685546875, "loss_alignment_w": 0.513427734375, "loss_sft": 0.2591266389936209, "loss_total_v6": 0.7717761546373367, "step": 581 }, { "epoch": 0.13422702702702702, "grad_norm": 25.89301300048828, "learning_rate": 9.940936336298342e-06, "loss": 12.8663911819458, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2868459802120924, "loss_total_v6": 0.804149441421032, "step": 582 }, { "epoch": 0.13445765765765766, "grad_norm": 25.89327621459961, "learning_rate": 9.940627311959656e-06, "loss": 12.864105224609375, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.293569503352046, "loss_total_v6": 0.804006539285183, "step": 583 }, { "epoch": 0.1346882882882883, "grad_norm": 22.89764976501465, "learning_rate": 9.940317486141084e-06, "loss": 12.227923393249512, "loss_alignment": 1.03662109375, "loss_alignment_w": 0.518310546875, "loss_sft": 0.24521753564476967, "loss_total_v6": 0.7642452418804169, "step": 584 }, { "epoch": 0.1349189189189189, "grad_norm": 24.840606689453125, "learning_rate": 9.940006858892883e-06, "loss": 13.188285827636719, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.30459927022457123, "loss_total_v6": 0.8242678567767143, "step": 585 }, { "epoch": 0.13514954954954955, "grad_norm": 25.001466751098633, "learning_rate": 9.939695430265446e-06, "loss": 12.615086555480957, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.2687285467982292, "loss_total_v6": 0.7884429022669792, "step": 586 }, { "epoch": 0.1353801801801802, "grad_norm": 25.73089027404785, "learning_rate": 9.93938320030929e-06, "loss": 12.61333179473877, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.28796699456870556, "loss_total_v6": 0.7883332148194313, "step": 587 }, { "epoch": 0.1356108108108108, "grad_norm": 23.19498062133789, "learning_rate": 9.939070169075072e-06, "loss": 12.659067153930664, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.276787381619215, "loss_total_v6": 0.7911916896700859, "step": 588 }, { "epoch": 0.13584144144144145, "grad_norm": 23.94400405883789, "learning_rate": 9.938756336613568e-06, "loss": 12.258160591125488, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.24574925377964973, "loss_total_v6": 0.7661350145936012, "step": 589 }, { "epoch": 0.13607207207207206, "grad_norm": 23.84123992919922, "learning_rate": 9.938441702975689e-06, "loss": 12.407304763793945, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.26358516700565815, "loss_total_v6": 0.7754565104842186, "step": 590 }, { "epoch": 0.1363027027027027, "grad_norm": 23.549325942993164, "learning_rate": 9.938126268212476e-06, "loss": 12.462172508239746, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.2540292013436556, "loss_total_v6": 0.7788857668638229, "step": 591 }, { "epoch": 0.13653333333333334, "grad_norm": 26.237451553344727, "learning_rate": 9.9378100323751e-06, "loss": 12.755205154418945, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2848102543503046, "loss_total_v6": 0.7972003668546677, "step": 592 }, { "epoch": 0.13676396396396395, "grad_norm": 25.120725631713867, "learning_rate": 9.937492995514862e-06, "loss": 12.411420822143555, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2699611894786358, "loss_total_v6": 0.775713749229908, "step": 593 }, { "epoch": 0.1369945945945946, "grad_norm": 24.85306739807129, "learning_rate": 9.937175157683191e-06, "loss": 12.315450668334961, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.26970041543245316, "loss_total_v6": 0.7697156816720963, "step": 594 }, { "epoch": 0.13722522522522523, "grad_norm": 24.357168197631836, "learning_rate": 9.93685651893165e-06, "loss": 12.384340286254883, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.26970304548740387, "loss_total_v6": 0.7740213051438332, "step": 595 }, { "epoch": 0.13745585585585585, "grad_norm": 32.59584426879883, "learning_rate": 9.936537079311925e-06, "loss": 12.585386276245117, "loss_alignment": 1.03955078125, "loss_alignment_w": 0.519775390625, "loss_sft": 0.2672232072800398, "loss_total_v6": 0.7865866050124168, "step": 596 }, { "epoch": 0.1376864864864865, "grad_norm": 26.867380142211914, "learning_rate": 9.936216838875839e-06, "loss": 12.759641647338867, "loss_alignment": 1.02099609375, "loss_alignment_w": 0.510498046875, "loss_sft": 0.28736103512346745, "loss_total_v6": 0.7974776476621628, "step": 597 }, { "epoch": 0.13791711711711713, "grad_norm": 28.102825164794922, "learning_rate": 9.935895797675342e-06, "loss": 12.953424453735352, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.2943912670016289, "loss_total_v6": 0.8095890507102013, "step": 598 }, { "epoch": 0.13814774774774774, "grad_norm": 23.379560470581055, "learning_rate": 9.935573955762514e-06, "loss": 13.139455795288086, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.3006928935647011, "loss_total_v6": 0.8212159723043442, "step": 599 }, { "epoch": 0.13837837837837838, "grad_norm": 26.562484741210938, "learning_rate": 9.935251313189564e-06, "loss": 12.677003860473633, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2830048892647028, "loss_total_v6": 0.7923127040266991, "step": 600 }, { "epoch": 0.13860900900900902, "grad_norm": 24.05263900756836, "learning_rate": 9.934927870008835e-06, "loss": 13.077047348022461, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.3024991638958454, "loss_total_v6": 0.8173154667019844, "step": 601 }, { "epoch": 0.13883963963963963, "grad_norm": 25.568540573120117, "learning_rate": 9.934603626272791e-06, "loss": 12.652727127075195, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.270150288939476, "loss_total_v6": 0.7907954305410385, "step": 602 }, { "epoch": 0.13907027027027027, "grad_norm": 26.521669387817383, "learning_rate": 9.934278582034037e-06, "loss": 12.762901306152344, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.28051515482366085, "loss_total_v6": 0.7976812869310379, "step": 603 }, { "epoch": 0.13930090090090091, "grad_norm": 25.226123809814453, "learning_rate": 9.9339527373453e-06, "loss": 12.801506042480469, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2807002030313015, "loss_total_v6": 0.8000941276550293, "step": 604 }, { "epoch": 0.13953153153153153, "grad_norm": 24.91452980041504, "learning_rate": 9.933626092259439e-06, "loss": 12.2503023147583, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2525366013869643, "loss_total_v6": 0.7656439170241356, "step": 605 }, { "epoch": 0.13976216216216217, "grad_norm": 27.1458683013916, "learning_rate": 9.933298646829444e-06, "loss": 12.042497634887695, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.24055585451424122, "loss_total_v6": 0.7526560798287392, "step": 606 }, { "epoch": 0.13999279279279278, "grad_norm": 26.242273330688477, "learning_rate": 9.932970401108433e-06, "loss": 13.321020126342773, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.3133987821638584, "loss_total_v6": 0.8325637951493263, "step": 607 }, { "epoch": 0.14022342342342342, "grad_norm": 34.79134750366211, "learning_rate": 9.932641355149655e-06, "loss": 12.188385009765625, "loss_alignment": 1.03955078125, "loss_alignment_w": 0.519775390625, "loss_sft": 0.24373817071318626, "loss_total_v6": 0.761774055659771, "step": 608 }, { "epoch": 0.14045405405405406, "grad_norm": 29.835920333862305, "learning_rate": 9.93231150900649e-06, "loss": 12.684141159057617, "loss_alignment": 1.02783203125, "loss_alignment_w": 0.513916015625, "loss_sft": 0.2799108996987343, "loss_total_v6": 0.7927588224411011, "step": 609 }, { "epoch": 0.14068468468468467, "grad_norm": 24.212839126586914, "learning_rate": 9.931980862732445e-06, "loss": 12.878783226013184, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.2842635363340378, "loss_total_v6": 0.804923951625824, "step": 610 }, { "epoch": 0.14091531531531531, "grad_norm": 24.04381561279297, "learning_rate": 9.931649416381158e-06, "loss": 12.719100952148438, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2772283609956503, "loss_total_v6": 0.7949437871575356, "step": 611 }, { "epoch": 0.14114594594594596, "grad_norm": 24.29496955871582, "learning_rate": 9.931317170006398e-06, "loss": 13.508045196533203, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.3216087967157364, "loss_total_v6": 0.8442528322339058, "step": 612 }, { "epoch": 0.14137657657657657, "grad_norm": 24.3197078704834, "learning_rate": 9.930984123662063e-06, "loss": 13.057924270629883, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.3005868047475815, "loss_total_v6": 0.8161202669143677, "step": 613 }, { "epoch": 0.1416072072072072, "grad_norm": 25.602338790893555, "learning_rate": 9.930650277402178e-06, "loss": 12.29412841796875, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2590598724782467, "loss_total_v6": 0.7683829739689827, "step": 614 }, { "epoch": 0.14183783783783785, "grad_norm": 80.42786407470703, "learning_rate": 9.930315631280904e-06, "loss": 12.958759307861328, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.29649475775659084, "loss_total_v6": 0.8099224865436554, "step": 615 }, { "epoch": 0.14206846846846846, "grad_norm": 23.073598861694336, "learning_rate": 9.929980185352525e-06, "loss": 12.662031173706055, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.278849545866251, "loss_total_v6": 0.7913769856095314, "step": 616 }, { "epoch": 0.1422990990990991, "grad_norm": 23.736671447753906, "learning_rate": 9.92964393967146e-06, "loss": 13.220663070678711, "loss_alignment": 1.02294921875, "loss_alignment_w": 0.511474609375, "loss_sft": 0.31470999494194984, "loss_total_v6": 0.8262913972139359, "step": 617 }, { "epoch": 0.14252972972972974, "grad_norm": 25.609630584716797, "learning_rate": 9.929306894292254e-06, "loss": 12.405035018920898, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.26020853221416473, "loss_total_v6": 0.7753147333860397, "step": 618 }, { "epoch": 0.14276036036036036, "grad_norm": 24.658458709716797, "learning_rate": 9.928969049269583e-06, "loss": 12.700546264648438, "loss_alignment": 1.02392578125, "loss_alignment_w": 0.511962890625, "loss_sft": 0.2814397756010294, "loss_total_v6": 0.7937841042876244, "step": 619 }, { "epoch": 0.142990990990991, "grad_norm": 24.797792434692383, "learning_rate": 9.928630404658255e-06, "loss": 12.47422981262207, "loss_alignment": 1.01513671875, "loss_alignment_w": 0.507568359375, "loss_sft": 0.27110966108739376, "loss_total_v6": 0.7796393409371376, "step": 620 }, { "epoch": 0.1432216216216216, "grad_norm": 24.677888870239258, "learning_rate": 9.928290960513204e-06, "loss": 12.886543273925781, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.28801399655640125, "loss_total_v6": 0.8054089918732643, "step": 621 }, { "epoch": 0.14345225225225225, "grad_norm": 27.8433780670166, "learning_rate": 9.927950716889496e-06, "loss": 12.866753578186035, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.287204310297966, "loss_total_v6": 0.8041720986366272, "step": 622 }, { "epoch": 0.1436828828828829, "grad_norm": 25.594348907470703, "learning_rate": 9.927609673842327e-06, "loss": 12.628081321716309, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.28121371380984783, "loss_total_v6": 0.7892550826072693, "step": 623 }, { "epoch": 0.1439135135135135, "grad_norm": 25.802997589111328, "learning_rate": 9.92726783142702e-06, "loss": 12.578710556030273, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.28073732554912567, "loss_total_v6": 0.7861693948507309, "step": 624 }, { "epoch": 0.14414414414414414, "grad_norm": 22.686880111694336, "learning_rate": 9.92692518969903e-06, "loss": 12.332155227661133, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.26082626916468143, "loss_total_v6": 0.7707596868276596, "step": 625 }, { "epoch": 0.14437477477477478, "grad_norm": 25.577219009399414, "learning_rate": 9.926581748713942e-06, "loss": 12.84030532836914, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.28033286705613136, "loss_total_v6": 0.8025191202759743, "step": 626 }, { "epoch": 0.1446054054054054, "grad_norm": 25.54297637939453, "learning_rate": 9.926237508527468e-06, "loss": 13.148494720458984, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.3054845556616783, "loss_total_v6": 0.8217809498310089, "step": 627 }, { "epoch": 0.14483603603603604, "grad_norm": 26.17448616027832, "learning_rate": 9.925892469195454e-06, "loss": 13.202719688415527, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.31224580109119415, "loss_total_v6": 0.8251699805259705, "step": 628 }, { "epoch": 0.14506666666666668, "grad_norm": 24.702613830566406, "learning_rate": 9.92554663077387e-06, "loss": 13.022109985351562, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.29744813218712807, "loss_total_v6": 0.8138818442821503, "step": 629 }, { "epoch": 0.1452972972972973, "grad_norm": 25.636709213256836, "learning_rate": 9.925199993318822e-06, "loss": 12.453449249267578, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2650349363684654, "loss_total_v6": 0.778340607881546, "step": 630 }, { "epoch": 0.14552792792792793, "grad_norm": 24.314159393310547, "learning_rate": 9.92485255688654e-06, "loss": 12.656688690185547, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.2701689787209034, "loss_total_v6": 0.7910429984331131, "step": 631 }, { "epoch": 0.14575855855855857, "grad_norm": 25.38037109375, "learning_rate": 9.924504321533387e-06, "loss": 12.392719268798828, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2633144538849592, "loss_total_v6": 0.7745449244976044, "step": 632 }, { "epoch": 0.14598918918918918, "grad_norm": 26.19980239868164, "learning_rate": 9.924155287315853e-06, "loss": 12.565692901611328, "loss_alignment": 1.04833984375, "loss_alignment_w": 0.524169921875, "loss_sft": 0.2630474865436554, "loss_total_v6": 0.7853558287024498, "step": 633 }, { "epoch": 0.14621981981981982, "grad_norm": 25.362897872924805, "learning_rate": 9.923805454290563e-06, "loss": 12.188982009887695, "loss_alignment": 1.01513671875, "loss_alignment_w": 0.507568359375, "loss_sft": 0.2540294099599123, "loss_total_v6": 0.7618113979697227, "step": 634 }, { "epoch": 0.14645045045045044, "grad_norm": 26.96984100341797, "learning_rate": 9.923454822514262e-06, "loss": 13.180034637451172, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.30191686749458313, "loss_total_v6": 0.82375218719244, "step": 635 }, { "epoch": 0.14668108108108108, "grad_norm": 28.57961082458496, "learning_rate": 9.923103392043836e-06, "loss": 12.220099449157715, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.25530283339321613, "loss_total_v6": 0.7637562155723572, "step": 636 }, { "epoch": 0.14691171171171172, "grad_norm": 25.511865615844727, "learning_rate": 9.92275116293629e-06, "loss": 12.360071182250977, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.26756058819592, "loss_total_v6": 0.772504448890686, "step": 637 }, { "epoch": 0.14714234234234233, "grad_norm": 26.34673500061035, "learning_rate": 9.922398135248766e-06, "loss": 13.098477363586426, "loss_alignment": 1.0546875, "loss_alignment_w": 0.52734375, "loss_sft": 0.2926538661122322, "loss_total_v6": 0.8186548128724098, "step": 638 }, { "epoch": 0.14737297297297297, "grad_norm": 27.564823150634766, "learning_rate": 9.922044309038534e-06, "loss": 12.17198371887207, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2464819960296154, "loss_total_v6": 0.7607489824295044, "step": 639 }, { "epoch": 0.1476036036036036, "grad_norm": 26.66698455810547, "learning_rate": 9.921689684362989e-06, "loss": 12.12791633605957, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.25361544638872147, "loss_total_v6": 0.757994756102562, "step": 640 }, { "epoch": 0.14783423423423422, "grad_norm": 26.381092071533203, "learning_rate": 9.921334261279663e-06, "loss": 12.611288070678711, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.269009942188859, "loss_total_v6": 0.7882054969668388, "step": 641 }, { "epoch": 0.14806486486486486, "grad_norm": 25.961889266967773, "learning_rate": 9.92097803984621e-06, "loss": 12.409873008728027, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2607396859675646, "loss_total_v6": 0.7756170406937599, "step": 642 }, { "epoch": 0.1482954954954955, "grad_norm": 25.212718963623047, "learning_rate": 9.92062102012042e-06, "loss": 12.753937721252441, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.27939036674797535, "loss_total_v6": 0.7971211150288582, "step": 643 }, { "epoch": 0.14852612612612612, "grad_norm": 24.433774948120117, "learning_rate": 9.920263202160208e-06, "loss": 12.379734992980957, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2654326520860195, "loss_total_v6": 0.7737334221601486, "step": 644 }, { "epoch": 0.14875675675675676, "grad_norm": 24.63347816467285, "learning_rate": 9.91990458602362e-06, "loss": 12.563859939575195, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.27004351653158665, "loss_total_v6": 0.7852412387728691, "step": 645 }, { "epoch": 0.1489873873873874, "grad_norm": 26.40336036682129, "learning_rate": 9.919545171768832e-06, "loss": 12.40711784362793, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.2507866807281971, "loss_total_v6": 0.775444895029068, "step": 646 }, { "epoch": 0.149218018018018, "grad_norm": 25.6297607421875, "learning_rate": 9.919184959454148e-06, "loss": 12.73588752746582, "loss_alignment": 1.00732421875, "loss_alignment_w": 0.503662109375, "loss_sft": 0.29240719601511955, "loss_total_v6": 0.7959930002689362, "step": 647 }, { "epoch": 0.14944864864864865, "grad_norm": 24.551149368286133, "learning_rate": 9.918823949138004e-06, "loss": 12.64642333984375, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.28327562287449837, "loss_total_v6": 0.7904014959931374, "step": 648 }, { "epoch": 0.1496792792792793, "grad_norm": 22.6022891998291, "learning_rate": 9.918462140878965e-06, "loss": 12.472366333007812, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2665834669023752, "loss_total_v6": 0.7795229256153107, "step": 649 }, { "epoch": 0.1499099099099099, "grad_norm": 22.35427474975586, "learning_rate": 9.91809953473572e-06, "loss": 12.745922088623047, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.28310078382492065, "loss_total_v6": 0.7966200858354568, "step": 650 }, { "epoch": 0.15014054054054055, "grad_norm": 24.720354080200195, "learning_rate": 9.917736130767095e-06, "loss": 12.368465423583984, "loss_alignment": 1.02490234375, "loss_alignment_w": 0.512451171875, "loss_sft": 0.262225866317749, "loss_total_v6": 0.7730291187763214, "step": 651 }, { "epoch": 0.15037117117117116, "grad_norm": 24.200786590576172, "learning_rate": 9.91737192903204e-06, "loss": 12.246007919311523, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.2436469253152609, "loss_total_v6": 0.765375442802906, "step": 652 }, { "epoch": 0.1506018018018018, "grad_norm": 24.137710571289062, "learning_rate": 9.917006929589642e-06, "loss": 12.33034610748291, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.252854872494936, "loss_total_v6": 0.7706466391682625, "step": 653 }, { "epoch": 0.15083243243243244, "grad_norm": 29.723697662353516, "learning_rate": 9.916641132499105e-06, "loss": 12.661630630493164, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.27616943418979645, "loss_total_v6": 0.7913519367575645, "step": 654 }, { "epoch": 0.15106306306306305, "grad_norm": 23.08078956604004, "learning_rate": 9.916274537819774e-06, "loss": 12.858598709106445, "loss_alignment": 1.01806640625, "loss_alignment_w": 0.509033203125, "loss_sft": 0.2942935712635517, "loss_total_v6": 0.8036624491214752, "step": 655 }, { "epoch": 0.1512936936936937, "grad_norm": 25.636070251464844, "learning_rate": 9.915907145611117e-06, "loss": 12.70048999786377, "loss_alignment": 1.01806640625, "loss_alignment_w": 0.509033203125, "loss_sft": 0.2853577770292759, "loss_total_v6": 0.7937806099653244, "step": 656 }, { "epoch": 0.15152432432432433, "grad_norm": 24.35089111328125, "learning_rate": 9.915538955932735e-06, "loss": 12.606498718261719, "loss_alignment": 1.02978515625, "loss_alignment_w": 0.514892578125, "loss_sft": 0.27385283447802067, "loss_total_v6": 0.7879061698913574, "step": 657 }, { "epoch": 0.15175495495495495, "grad_norm": 27.134994506835938, "learning_rate": 9.915169968844354e-06, "loss": 12.60207748413086, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.27087569423019886, "loss_total_v6": 0.7876298502087593, "step": 658 }, { "epoch": 0.1519855855855856, "grad_norm": 22.11064910888672, "learning_rate": 9.914800184405836e-06, "loss": 13.124505043029785, "loss_alignment": 1.02783203125, "loss_alignment_w": 0.513916015625, "loss_sft": 0.3062434569001198, "loss_total_v6": 0.8202815800905228, "step": 659 }, { "epoch": 0.15221621621621623, "grad_norm": 23.64436149597168, "learning_rate": 9.914429602677163e-06, "loss": 13.13026237487793, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.3102043494582176, "loss_total_v6": 0.820641353726387, "step": 660 }, { "epoch": 0.15244684684684684, "grad_norm": 23.659337997436523, "learning_rate": 9.914058223718453e-06, "loss": 12.772872924804688, "loss_alignment": 1.056640625, "loss_alignment_w": 0.5283203125, "loss_sft": 0.2696790751069784, "loss_total_v6": 0.7983045652508736, "step": 661 }, { "epoch": 0.15267747747747748, "grad_norm": 26.49545669555664, "learning_rate": 9.913686047589955e-06, "loss": 13.251680374145508, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.3042279165238142, "loss_total_v6": 0.8282299935817719, "step": 662 }, { "epoch": 0.15290810810810812, "grad_norm": 26.87653350830078, "learning_rate": 9.913313074352042e-06, "loss": 12.716268539428711, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2794163729995489, "loss_total_v6": 0.7947667166590691, "step": 663 }, { "epoch": 0.15313873873873873, "grad_norm": 25.33757781982422, "learning_rate": 9.912939304065219e-06, "loss": 12.817451477050781, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2882427964359522, "loss_total_v6": 0.801090694963932, "step": 664 }, { "epoch": 0.15336936936936937, "grad_norm": 25.671043395996094, "learning_rate": 9.91256473679012e-06, "loss": 12.885772705078125, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.28907968662679195, "loss_total_v6": 0.8053608313202858, "step": 665 }, { "epoch": 0.1536, "grad_norm": 25.107229232788086, "learning_rate": 9.912189372587508e-06, "loss": 12.355127334594727, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2680908180773258, "loss_total_v6": 0.7721954509615898, "step": 666 }, { "epoch": 0.15383063063063063, "grad_norm": 24.64372444152832, "learning_rate": 9.911813211518275e-06, "loss": 13.182809829711914, "loss_alignment": 1.0576171875, "loss_alignment_w": 0.52880859375, "loss_sft": 0.2976042404770851, "loss_total_v6": 0.8239256218075752, "step": 667 }, { "epoch": 0.15406126126126127, "grad_norm": 23.380823135375977, "learning_rate": 9.911436253643445e-06, "loss": 13.142879486083984, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.3053624741733074, "loss_total_v6": 0.8214299976825714, "step": 668 }, { "epoch": 0.15429189189189188, "grad_norm": 23.763917922973633, "learning_rate": 9.911058499024164e-06, "loss": 12.579217910766602, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.27706109173595905, "loss_total_v6": 0.7862011268734932, "step": 669 }, { "epoch": 0.15452252252252252, "grad_norm": 24.63629722595215, "learning_rate": 9.910679947721718e-06, "loss": 12.73121166229248, "loss_alignment": 1.04833984375, "loss_alignment_w": 0.524169921875, "loss_sft": 0.27194275334477425, "loss_total_v6": 0.795700691640377, "step": 670 }, { "epoch": 0.15475315315315316, "grad_norm": 23.35222053527832, "learning_rate": 9.910300599797514e-06, "loss": 12.805950164794922, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2834041081368923, "loss_total_v6": 0.8003718703985214, "step": 671 }, { "epoch": 0.15498378378378377, "grad_norm": 26.22959327697754, "learning_rate": 9.909920455313087e-06, "loss": 13.111992835998535, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.3008533287793398, "loss_total_v6": 0.8194995522499084, "step": 672 }, { "epoch": 0.15521441441441441, "grad_norm": 23.45319175720215, "learning_rate": 9.90953951433011e-06, "loss": 12.989176750183105, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.2837474104017019, "loss_total_v6": 0.8118235915899277, "step": 673 }, { "epoch": 0.15544504504504505, "grad_norm": 24.322235107421875, "learning_rate": 9.90915777691038e-06, "loss": 12.360313415527344, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2559942565858364, "loss_total_v6": 0.7725195586681366, "step": 674 }, { "epoch": 0.15567567567567567, "grad_norm": 23.888465881347656, "learning_rate": 9.908775243115822e-06, "loss": 12.709653854370117, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.28756327368319035, "loss_total_v6": 0.7943534329533577, "step": 675 }, { "epoch": 0.1559063063063063, "grad_norm": 23.98606300354004, "learning_rate": 9.908391913008491e-06, "loss": 13.026788711547852, "loss_alignment": 1.0556640625, "loss_alignment_w": 0.52783203125, "loss_sft": 0.2869526781141758, "loss_total_v6": 0.8141743391752243, "step": 676 }, { "epoch": 0.15613693693693695, "grad_norm": 24.908443450927734, "learning_rate": 9.90800778665057e-06, "loss": 12.61132526397705, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.2680662237107754, "loss_total_v6": 0.7882078513503075, "step": 677 }, { "epoch": 0.15636756756756756, "grad_norm": 23.689891815185547, "learning_rate": 9.907622864104376e-06, "loss": 12.503374099731445, "loss_alignment": 1.01708984375, "loss_alignment_w": 0.508544921875, "loss_sft": 0.27422820404171944, "loss_total_v6": 0.7814608290791512, "step": 678 }, { "epoch": 0.1565981981981982, "grad_norm": 25.195524215698242, "learning_rate": 9.907237145432352e-06, "loss": 12.681835174560547, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.27724904753267765, "loss_total_v6": 0.7926146686077118, "step": 679 }, { "epoch": 0.15682882882882884, "grad_norm": 27.656993865966797, "learning_rate": 9.906850630697068e-06, "loss": 13.097196578979492, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.2968157362192869, "loss_total_v6": 0.8185747787356377, "step": 680 }, { "epoch": 0.15705945945945946, "grad_norm": 30.816795349121094, "learning_rate": 9.906463319961225e-06, "loss": 12.6043701171875, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.27751921117305756, "loss_total_v6": 0.7877731472253799, "step": 681 }, { "epoch": 0.1572900900900901, "grad_norm": 27.664409637451172, "learning_rate": 9.906075213287658e-06, "loss": 13.024494171142578, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.2907764557749033, "loss_total_v6": 0.8140308633446693, "step": 682 }, { "epoch": 0.1575207207207207, "grad_norm": 24.397342681884766, "learning_rate": 9.905686310739321e-06, "loss": 12.57359504699707, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2707892246544361, "loss_total_v6": 0.7858496531844139, "step": 683 }, { "epoch": 0.15775135135135135, "grad_norm": 23.545921325683594, "learning_rate": 9.905296612379307e-06, "loss": 12.512341499328613, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.2716758903115988, "loss_total_v6": 0.7820213735103607, "step": 684 }, { "epoch": 0.157981981981982, "grad_norm": 21.402389526367188, "learning_rate": 9.904906118270831e-06, "loss": 12.600812911987305, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.2683552373200655, "loss_total_v6": 0.7875507920980453, "step": 685 }, { "epoch": 0.1582126126126126, "grad_norm": 25.598413467407227, "learning_rate": 9.90451482847724e-06, "loss": 12.463817596435547, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.26908572763204575, "loss_total_v6": 0.7789886444807053, "step": 686 }, { "epoch": 0.15844324324324324, "grad_norm": 24.58620834350586, "learning_rate": 9.90412274306201e-06, "loss": 12.484077453613281, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.26937529630959034, "loss_total_v6": 0.7802548334002495, "step": 687 }, { "epoch": 0.15867387387387388, "grad_norm": 23.421615600585938, "learning_rate": 9.903729862088749e-06, "loss": 12.787694931030273, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2841552961617708, "loss_total_v6": 0.7992309778928757, "step": 688 }, { "epoch": 0.1589045045045045, "grad_norm": 26.85695457458496, "learning_rate": 9.903336185621186e-06, "loss": 12.845093727111816, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2864304017275572, "loss_total_v6": 0.8028183579444885, "step": 689 }, { "epoch": 0.15913513513513514, "grad_norm": 24.27919578552246, "learning_rate": 9.902941713723187e-06, "loss": 12.519546508789062, "loss_alignment": 1.02978515625, "loss_alignment_w": 0.514892578125, "loss_sft": 0.26693821139633656, "loss_total_v6": 0.7824716567993164, "step": 690 }, { "epoch": 0.15936576576576578, "grad_norm": 24.517351150512695, "learning_rate": 9.902546446458745e-06, "loss": 12.715106964111328, "loss_alignment": 1.02587890625, "loss_alignment_w": 0.512939453125, "loss_sft": 0.28256346471607685, "loss_total_v6": 0.7946941703557968, "step": 691 }, { "epoch": 0.1595963963963964, "grad_norm": 23.9361629486084, "learning_rate": 9.902150383891979e-06, "loss": 12.171466827392578, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.24736519902944565, "loss_total_v6": 0.7607166469097137, "step": 692 }, { "epoch": 0.15982702702702703, "grad_norm": 23.45281982421875, "learning_rate": 9.901753526087141e-06, "loss": 12.470564842224121, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.26326651126146317, "loss_total_v6": 0.7794103100895882, "step": 693 }, { "epoch": 0.16005765765765767, "grad_norm": 24.110090255737305, "learning_rate": 9.901355873108611e-06, "loss": 12.199773788452148, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.24403804913163185, "loss_total_v6": 0.7624859064817429, "step": 694 }, { "epoch": 0.16028828828828828, "grad_norm": 23.80487632751465, "learning_rate": 9.900957425020894e-06, "loss": 12.711735725402832, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.27664597891271114, "loss_total_v6": 0.7944835051894188, "step": 695 }, { "epoch": 0.16051891891891892, "grad_norm": 22.770353317260742, "learning_rate": 9.900558181888626e-06, "loss": 11.85948371887207, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.2209540456533432, "loss_total_v6": 0.7412177249789238, "step": 696 }, { "epoch": 0.16074954954954954, "grad_norm": 25.803966522216797, "learning_rate": 9.90015814377658e-06, "loss": 12.534747123718262, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2663318682461977, "loss_total_v6": 0.7834216877818108, "step": 697 }, { "epoch": 0.16098018018018018, "grad_norm": 25.64549446105957, "learning_rate": 9.899757310749644e-06, "loss": 13.11953067779541, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.3005766961723566, "loss_total_v6": 0.8199706599116325, "step": 698 }, { "epoch": 0.16121081081081082, "grad_norm": 27.40300178527832, "learning_rate": 9.899355682872847e-06, "loss": 12.612056732177734, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2826993279159069, "loss_total_v6": 0.7882535308599472, "step": 699 }, { "epoch": 0.16144144144144143, "grad_norm": 25.661256790161133, "learning_rate": 9.89895326021134e-06, "loss": 12.96225643157959, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.2894195709377527, "loss_total_v6": 0.8101409897208214, "step": 700 }, { "epoch": 0.16167207207207207, "grad_norm": 24.793306350708008, "learning_rate": 9.898550042830404e-06, "loss": 12.503229141235352, "loss_alignment": 1.01806640625, "loss_alignment_w": 0.509033203125, "loss_sft": 0.2736087776720524, "loss_total_v6": 0.7814517840743065, "step": 701 }, { "epoch": 0.1619027027027027, "grad_norm": 23.497575759887695, "learning_rate": 9.89814603079545e-06, "loss": 12.104862213134766, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.24152399599552155, "loss_total_v6": 0.7565538957715034, "step": 702 }, { "epoch": 0.16213333333333332, "grad_norm": 23.67327117919922, "learning_rate": 9.89774122417202e-06, "loss": 12.408829689025879, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2580805476754904, "loss_total_v6": 0.7755518779158592, "step": 703 }, { "epoch": 0.16236396396396396, "grad_norm": 25.32233428955078, "learning_rate": 9.89733562302578e-06, "loss": 12.390435218811035, "loss_alignment": 1.01904296875, "loss_alignment_w": 0.509521484375, "loss_sft": 0.2635532021522522, "loss_total_v6": 0.7744021862745285, "step": 704 }, { "epoch": 0.1625945945945946, "grad_norm": 25.682661056518555, "learning_rate": 9.89692922742253e-06, "loss": 12.234058380126953, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.2640030216425657, "loss_total_v6": 0.764628604054451, "step": 705 }, { "epoch": 0.16282522522522522, "grad_norm": 24.020732879638672, "learning_rate": 9.896522037428195e-06, "loss": 13.148879051208496, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.30689707957208157, "loss_total_v6": 0.8218049481511116, "step": 706 }, { "epoch": 0.16305585585585586, "grad_norm": 24.832651138305664, "learning_rate": 9.89611405310883e-06, "loss": 12.551019668579102, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2711178082972765, "loss_total_v6": 0.7844387590885162, "step": 707 }, { "epoch": 0.1632864864864865, "grad_norm": 24.934799194335938, "learning_rate": 9.89570527453062e-06, "loss": 13.548874855041504, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.32210069335997105, "loss_total_v6": 0.8468046858906746, "step": 708 }, { "epoch": 0.1635171171171171, "grad_norm": 23.624908447265625, "learning_rate": 9.895295701759878e-06, "loss": 12.742106437683105, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.27820843271911144, "loss_total_v6": 0.7963816225528717, "step": 709 }, { "epoch": 0.16374774774774775, "grad_norm": 24.615510940551758, "learning_rate": 9.894885334863044e-06, "loss": 12.678679466247559, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.2916697971522808, "loss_total_v6": 0.7924174666404724, "step": 710 }, { "epoch": 0.1639783783783784, "grad_norm": 24.849138259887695, "learning_rate": 9.89447417390669e-06, "loss": 12.69192123413086, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.27943587861955166, "loss_total_v6": 0.7932451143860817, "step": 711 }, { "epoch": 0.164209009009009, "grad_norm": 25.481460571289062, "learning_rate": 9.894062218957517e-06, "loss": 13.060540199279785, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.3001094553619623, "loss_total_v6": 0.8162837699055672, "step": 712 }, { "epoch": 0.16443963963963965, "grad_norm": 27.554046630859375, "learning_rate": 9.89364947008235e-06, "loss": 12.556365966796875, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2719707787036896, "loss_total_v6": 0.7847728803753853, "step": 713 }, { "epoch": 0.16467027027027026, "grad_norm": 24.428707122802734, "learning_rate": 9.89323592734815e-06, "loss": 12.383304595947266, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.25758388079702854, "loss_total_v6": 0.7739565670490265, "step": 714 }, { "epoch": 0.1649009009009009, "grad_norm": 28.150402069091797, "learning_rate": 9.892821590822e-06, "loss": 12.688519477844238, "loss_alignment": 1.02587890625, "loss_alignment_w": 0.512939453125, "loss_sft": 0.280306626111269, "loss_total_v6": 0.7930324897170067, "step": 715 }, { "epoch": 0.16513153153153154, "grad_norm": 22.139755249023438, "learning_rate": 9.892406460571116e-06, "loss": 12.493764877319336, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.26389255560934544, "loss_total_v6": 0.7808603346347809, "step": 716 }, { "epoch": 0.16536216216216215, "grad_norm": 22.481807708740234, "learning_rate": 9.891990536662839e-06, "loss": 12.691936492919922, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.29329181276261806, "loss_total_v6": 0.7932460531592369, "step": 717 }, { "epoch": 0.1655927927927928, "grad_norm": 24.30417823791504, "learning_rate": 9.891573819164643e-06, "loss": 12.875179290771484, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2956654839217663, "loss_total_v6": 0.8046987280249596, "step": 718 }, { "epoch": 0.16582342342342343, "grad_norm": 23.817115783691406, "learning_rate": 9.891156308144131e-06, "loss": 12.651159286499023, "loss_alignment": 1.02490234375, "loss_alignment_w": 0.512451171875, "loss_sft": 0.2786582335829735, "loss_total_v6": 0.7906974256038666, "step": 719 }, { "epoch": 0.16605405405405405, "grad_norm": 27.315670013427734, "learning_rate": 9.890738003669029e-06, "loss": 12.821968078613281, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.28615993820130825, "loss_total_v6": 0.8013729900121689, "step": 720 }, { "epoch": 0.16628468468468469, "grad_norm": 24.666072845458984, "learning_rate": 9.890318905807196e-06, "loss": 12.584797859191895, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.27084854431450367, "loss_total_v6": 0.7865498438477516, "step": 721 }, { "epoch": 0.16651531531531533, "grad_norm": 26.598581314086914, "learning_rate": 9.88989901462662e-06, "loss": 12.73835277557373, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.2739150021225214, "loss_total_v6": 0.7961470633745193, "step": 722 }, { "epoch": 0.16674594594594594, "grad_norm": 24.93560791015625, "learning_rate": 9.889478330195419e-06, "loss": 12.727439880371094, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.2770323306322098, "loss_total_v6": 0.7954649478197098, "step": 723 }, { "epoch": 0.16697657657657658, "grad_norm": 23.574127197265625, "learning_rate": 9.889056852581833e-06, "loss": 13.066350936889648, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.3028072118759155, "loss_total_v6": 0.816646933555603, "step": 724 }, { "epoch": 0.16720720720720722, "grad_norm": 22.91994285583496, "learning_rate": 9.888634581854235e-06, "loss": 12.968406677246094, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.28557727113366127, "loss_total_v6": 0.8105254024267197, "step": 725 }, { "epoch": 0.16743783783783783, "grad_norm": 25.90430450439453, "learning_rate": 9.88821151808113e-06, "loss": 12.688315391540527, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.2753195110708475, "loss_total_v6": 0.7930197343230247, "step": 726 }, { "epoch": 0.16766846846846847, "grad_norm": 25.597047805786133, "learning_rate": 9.887787661331148e-06, "loss": 12.952770233154297, "loss_alignment": 1.02783203125, "loss_alignment_w": 0.513916015625, "loss_sft": 0.29637979343533516, "loss_total_v6": 0.8095481395721436, "step": 727 }, { "epoch": 0.1678990990990991, "grad_norm": 24.90395164489746, "learning_rate": 9.887363011673046e-06, "loss": 12.656736373901367, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2801359910517931, "loss_total_v6": 0.7910460606217384, "step": 728 }, { "epoch": 0.16812972972972973, "grad_norm": 24.685056686401367, "learning_rate": 9.886937569175711e-06, "loss": 12.381047248840332, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.2720454093068838, "loss_total_v6": 0.7738154157996178, "step": 729 }, { "epoch": 0.16836036036036037, "grad_norm": 25.104450225830078, "learning_rate": 9.886511333908165e-06, "loss": 12.776524543762207, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.28714972734451294, "loss_total_v6": 0.7985327914357185, "step": 730 }, { "epoch": 0.16859099099099098, "grad_norm": 23.512847900390625, "learning_rate": 9.886084305939545e-06, "loss": 12.305196762084961, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2591260690242052, "loss_total_v6": 0.7690747976303101, "step": 731 }, { "epoch": 0.16882162162162162, "grad_norm": 27.288726806640625, "learning_rate": 9.885656485339129e-06, "loss": 12.530889511108398, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.275230810046196, "loss_total_v6": 0.7831805944442749, "step": 732 }, { "epoch": 0.16905225225225226, "grad_norm": 23.951526641845703, "learning_rate": 9.88522787217632e-06, "loss": 12.690659523010254, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2843771781772375, "loss_total_v6": 0.7931662201881409, "step": 733 }, { "epoch": 0.16928288288288287, "grad_norm": 23.29867172241211, "learning_rate": 9.884798466520645e-06, "loss": 13.142680168151855, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.31140774488449097, "loss_total_v6": 0.821417473256588, "step": 734 }, { "epoch": 0.16951351351351351, "grad_norm": 22.53425407409668, "learning_rate": 9.884368268441764e-06, "loss": 12.501405715942383, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2725640721619129, "loss_total_v6": 0.7813378721475601, "step": 735 }, { "epoch": 0.16974414414414415, "grad_norm": 25.739639282226562, "learning_rate": 9.883937278009468e-06, "loss": 12.332300186157227, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2612472493201494, "loss_total_v6": 0.7707687616348267, "step": 736 }, { "epoch": 0.16997477477477477, "grad_norm": 23.278682708740234, "learning_rate": 9.883505495293671e-06, "loss": 12.47482967376709, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.26209873892366886, "loss_total_v6": 0.7796768620610237, "step": 737 }, { "epoch": 0.1702054054054054, "grad_norm": 23.390819549560547, "learning_rate": 9.883072920364417e-06, "loss": 13.206645965576172, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.299704372882843, "loss_total_v6": 0.8254154324531555, "step": 738 }, { "epoch": 0.17043603603603605, "grad_norm": 23.876615524291992, "learning_rate": 9.88263955329188e-06, "loss": 12.659538269042969, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2732310798019171, "loss_total_v6": 0.7912211865186691, "step": 739 }, { "epoch": 0.17066666666666666, "grad_norm": 23.94149398803711, "learning_rate": 9.882205394146362e-06, "loss": 12.791393280029297, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.283730274066329, "loss_total_v6": 0.7994620949029922, "step": 740 }, { "epoch": 0.1708972972972973, "grad_norm": 22.329492568969727, "learning_rate": 9.881770442998293e-06, "loss": 12.898660659790039, "loss_alignment": 1.03466796875, "loss_alignment_w": 0.517333984375, "loss_sft": 0.28893911093473434, "loss_total_v6": 0.806166298687458, "step": 741 }, { "epoch": 0.17112792792792791, "grad_norm": 25.857227325439453, "learning_rate": 9.881334699918231e-06, "loss": 12.680828094482422, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.27595018967986107, "loss_total_v6": 0.7925517484545708, "step": 742 }, { "epoch": 0.17135855855855855, "grad_norm": 26.402137756347656, "learning_rate": 9.880898164976865e-06, "loss": 12.216024398803711, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.24204768799245358, "loss_total_v6": 0.7635015472769737, "step": 743 }, { "epoch": 0.1715891891891892, "grad_norm": 22.994632720947266, "learning_rate": 9.88046083824501e-06, "loss": 12.800625801086426, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.2953546531498432, "loss_total_v6": 0.8000390902161598, "step": 744 }, { "epoch": 0.1718198198198198, "grad_norm": 24.855112075805664, "learning_rate": 9.88002271979361e-06, "loss": 12.829227447509766, "loss_alignment": 1.02880859375, "loss_alignment_w": 0.514404296875, "loss_sft": 0.28647635877132416, "loss_total_v6": 0.8018266782164574, "step": 745 }, { "epoch": 0.17205045045045045, "grad_norm": 27.91103172302246, "learning_rate": 9.879583809693737e-06, "loss": 12.68446159362793, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2809228077530861, "loss_total_v6": 0.7927788943052292, "step": 746 }, { "epoch": 0.1722810810810811, "grad_norm": 22.189367294311523, "learning_rate": 9.879144108016595e-06, "loss": 13.01999282836914, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.2901746891438961, "loss_total_v6": 0.8137495145201683, "step": 747 }, { "epoch": 0.1725117117117117, "grad_norm": 23.55867576599121, "learning_rate": 9.878703614833507e-06, "loss": 12.689777374267578, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2807972300797701, "loss_total_v6": 0.7931110635399818, "step": 748 }, { "epoch": 0.17274234234234234, "grad_norm": 24.68488121032715, "learning_rate": 9.878262330215938e-06, "loss": 13.036979675292969, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.30345867201685905, "loss_total_v6": 0.8148112297058105, "step": 749 }, { "epoch": 0.17297297297297298, "grad_norm": 24.618438720703125, "learning_rate": 9.87782025423547e-06, "loss": 12.711692810058594, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.27340843714773655, "loss_total_v6": 0.7944808304309845, "step": 750 }, { "epoch": 0.1732036036036036, "grad_norm": 24.534461975097656, "learning_rate": 9.877377386963819e-06, "loss": 12.377144813537598, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2667202968150377, "loss_total_v6": 0.7735715135931969, "step": 751 }, { "epoch": 0.17343423423423424, "grad_norm": 22.472915649414062, "learning_rate": 9.876933728472826e-06, "loss": 12.66329574584961, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2746103126555681, "loss_total_v6": 0.7914560362696648, "step": 752 }, { "epoch": 0.17366486486486488, "grad_norm": 24.01845932006836, "learning_rate": 9.876489278834465e-06, "loss": 12.930217742919922, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2928493283689022, "loss_total_v6": 0.8081386163830757, "step": 753 }, { "epoch": 0.1738954954954955, "grad_norm": 24.75965690612793, "learning_rate": 9.876044038120835e-06, "loss": 12.801271438598633, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.29275523871183395, "loss_total_v6": 0.8000794276595116, "step": 754 }, { "epoch": 0.17412612612612613, "grad_norm": 22.340593338012695, "learning_rate": 9.875598006404164e-06, "loss": 12.015303611755371, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.24889657646417618, "loss_total_v6": 0.7509565055370331, "step": 755 }, { "epoch": 0.17435675675675677, "grad_norm": 24.294931411743164, "learning_rate": 9.875151183756806e-06, "loss": 12.241830825805664, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.25226653553545475, "loss_total_v6": 0.7651144787669182, "step": 756 }, { "epoch": 0.17458738738738738, "grad_norm": 21.26738929748535, "learning_rate": 9.874703570251247e-06, "loss": 11.928848266601562, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2306604264304042, "loss_total_v6": 0.7455529868602753, "step": 757 }, { "epoch": 0.17481801801801802, "grad_norm": 23.4849796295166, "learning_rate": 9.874255165960102e-06, "loss": 12.750263214111328, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.29255793802440166, "loss_total_v6": 0.7968914285302162, "step": 758 }, { "epoch": 0.17504864864864864, "grad_norm": 23.163942337036133, "learning_rate": 9.87380597095611e-06, "loss": 12.23699951171875, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.248485604301095, "loss_total_v6": 0.7648124620318413, "step": 759 }, { "epoch": 0.17527927927927928, "grad_norm": 25.1905460357666, "learning_rate": 9.873355985312141e-06, "loss": 13.082132339477539, "loss_alignment": 1.0556640625, "loss_alignment_w": 0.52783203125, "loss_sft": 0.2906557582318783, "loss_total_v6": 0.8176332861185074, "step": 760 }, { "epoch": 0.17550990990990992, "grad_norm": 24.54130744934082, "learning_rate": 9.872905209101191e-06, "loss": 12.912846565246582, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.291107464581728, "loss_total_v6": 0.8070529326796532, "step": 761 }, { "epoch": 0.17574054054054053, "grad_norm": 24.862157821655273, "learning_rate": 9.87245364239639e-06, "loss": 12.431596755981445, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2666293941438198, "loss_total_v6": 0.7769748345017433, "step": 762 }, { "epoch": 0.17597117117117117, "grad_norm": 23.288516998291016, "learning_rate": 9.872001285270987e-06, "loss": 12.268001556396484, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2590138800442219, "loss_total_v6": 0.7667500823736191, "step": 763 }, { "epoch": 0.1762018018018018, "grad_norm": 26.815021514892578, "learning_rate": 9.871548137798367e-06, "loss": 12.272871017456055, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.26114930026233196, "loss_total_v6": 0.7670544311404228, "step": 764 }, { "epoch": 0.17643243243243242, "grad_norm": 22.358110427856445, "learning_rate": 9.871094200052043e-06, "loss": 12.690162658691406, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.27610630728304386, "loss_total_v6": 0.7931351214647293, "step": 765 }, { "epoch": 0.17666306306306306, "grad_norm": 22.277019500732422, "learning_rate": 9.87063947210565e-06, "loss": 12.801116943359375, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.28665730357170105, "loss_total_v6": 0.8000698313117027, "step": 766 }, { "epoch": 0.1768936936936937, "grad_norm": 26.713735580444336, "learning_rate": 9.870183954032957e-06, "loss": 12.314277648925781, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.26269964314997196, "loss_total_v6": 0.7696423605084419, "step": 767 }, { "epoch": 0.17712432432432432, "grad_norm": 24.59596061706543, "learning_rate": 9.86972764590786e-06, "loss": 12.499053955078125, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2676105685532093, "loss_total_v6": 0.781190924346447, "step": 768 }, { "epoch": 0.17735495495495496, "grad_norm": 24.57118797302246, "learning_rate": 9.86927054780438e-06, "loss": 12.708545684814453, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.28528139367699623, "loss_total_v6": 0.7942840754985809, "step": 769 }, { "epoch": 0.1775855855855856, "grad_norm": 23.56012725830078, "learning_rate": 9.868812659796669e-06, "loss": 13.22778606414795, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.30717483907938004, "loss_total_v6": 0.826736606657505, "step": 770 }, { "epoch": 0.1778162162162162, "grad_norm": 22.682796478271484, "learning_rate": 9.868353981959007e-06, "loss": 12.4916353225708, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.26833702623844147, "loss_total_v6": 0.7807271853089333, "step": 771 }, { "epoch": 0.17804684684684685, "grad_norm": 24.892833709716797, "learning_rate": 9.867894514365802e-06, "loss": 12.548079490661621, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.2739400193095207, "loss_total_v6": 0.7842549309134483, "step": 772 }, { "epoch": 0.17827747747747746, "grad_norm": 21.71626853942871, "learning_rate": 9.86743425709159e-06, "loss": 12.784870147705078, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.2938206382095814, "loss_total_v6": 0.7990544214844704, "step": 773 }, { "epoch": 0.1785081081081081, "grad_norm": 24.757532119750977, "learning_rate": 9.866973210211036e-06, "loss": 12.74998664855957, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.27050695940852165, "loss_total_v6": 0.796874150633812, "step": 774 }, { "epoch": 0.17873873873873874, "grad_norm": 25.89173126220703, "learning_rate": 9.86651137379893e-06, "loss": 12.959039688110352, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.2864414006471634, "loss_total_v6": 0.8099399656057358, "step": 775 }, { "epoch": 0.17896936936936936, "grad_norm": 23.204212188720703, "learning_rate": 9.866048747930194e-06, "loss": 12.91525650024414, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.2924482598900795, "loss_total_v6": 0.8072035312652588, "step": 776 }, { "epoch": 0.1792, "grad_norm": 22.918310165405273, "learning_rate": 9.865585332679874e-06, "loss": 12.514348030090332, "loss_alignment": 1.01904296875, "loss_alignment_w": 0.509521484375, "loss_sft": 0.2731288019567728, "loss_total_v6": 0.7821467518806458, "step": 777 }, { "epoch": 0.17943063063063064, "grad_norm": 23.311954498291016, "learning_rate": 9.865121128123148e-06, "loss": 12.908238410949707, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.29132301546633244, "loss_total_v6": 0.8067649081349373, "step": 778 }, { "epoch": 0.17966126126126125, "grad_norm": 23.880281448364258, "learning_rate": 9.864656134335321e-06, "loss": 12.876577377319336, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.27785428799688816, "loss_total_v6": 0.8047860562801361, "step": 779 }, { "epoch": 0.1798918918918919, "grad_norm": 24.695812225341797, "learning_rate": 9.864190351391822e-06, "loss": 13.23099422454834, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.302721481770277, "loss_total_v6": 0.8269371464848518, "step": 780 }, { "epoch": 0.18012252252252253, "grad_norm": 26.8072509765625, "learning_rate": 9.863723779368218e-06, "loss": 13.048635482788086, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.2905152812600136, "loss_total_v6": 0.8155396953225136, "step": 781 }, { "epoch": 0.18035315315315315, "grad_norm": 22.871829986572266, "learning_rate": 9.86325641834019e-06, "loss": 13.131160736083984, "loss_alignment": 1.046875, "loss_alignment_w": 0.5234375, "loss_sft": 0.29663441330194473, "loss_total_v6": 0.8206975236535072, "step": 782 }, { "epoch": 0.18058378378378379, "grad_norm": 23.546010971069336, "learning_rate": 9.862788268383557e-06, "loss": 12.805882453918457, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.2796309348195791, "loss_total_v6": 0.8003676533699036, "step": 783 }, { "epoch": 0.18081441441441443, "grad_norm": 22.853708267211914, "learning_rate": 9.862319329574264e-06, "loss": 12.583559036254883, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.26774989254772663, "loss_total_v6": 0.786472424864769, "step": 784 }, { "epoch": 0.18104504504504504, "grad_norm": 26.112226486206055, "learning_rate": 9.861849601988384e-06, "loss": 13.144196510314941, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.29816633462905884, "loss_total_v6": 0.8215122893452644, "step": 785 }, { "epoch": 0.18127567567567568, "grad_norm": 25.142642974853516, "learning_rate": 9.861379085702117e-06, "loss": 12.40846061706543, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2628487255424261, "loss_total_v6": 0.7755288034677505, "step": 786 }, { "epoch": 0.18150630630630632, "grad_norm": 21.545135498046875, "learning_rate": 9.86090778079179e-06, "loss": 12.506804466247559, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.26794239692389965, "loss_total_v6": 0.7816753089427948, "step": 787 }, { "epoch": 0.18173693693693693, "grad_norm": 26.675350189208984, "learning_rate": 9.860435687333857e-06, "loss": 12.471025466918945, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2685290854424238, "loss_total_v6": 0.7794391140341759, "step": 788 }, { "epoch": 0.18196756756756757, "grad_norm": 23.54298973083496, "learning_rate": 9.859962805404908e-06, "loss": 12.253023147583008, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2543851360678673, "loss_total_v6": 0.7658139765262604, "step": 789 }, { "epoch": 0.18219819819819819, "grad_norm": 22.62670135498047, "learning_rate": 9.859489135081653e-06, "loss": 12.930866241455078, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.3053410369902849, "loss_total_v6": 0.8081791624426842, "step": 790 }, { "epoch": 0.18242882882882883, "grad_norm": 24.603057861328125, "learning_rate": 9.859014676440928e-06, "loss": 12.008460998535156, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.2507271971553564, "loss_total_v6": 0.7505288422107697, "step": 791 }, { "epoch": 0.18265945945945947, "grad_norm": 24.812227249145508, "learning_rate": 9.858539429559705e-06, "loss": 12.544007301330566, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2727089710533619, "loss_total_v6": 0.7840004563331604, "step": 792 }, { "epoch": 0.18289009009009008, "grad_norm": 25.129663467407227, "learning_rate": 9.858063394515079e-06, "loss": 12.57248306274414, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.27270341850817204, "loss_total_v6": 0.7857801914215088, "step": 793 }, { "epoch": 0.18312072072072072, "grad_norm": 24.03770637512207, "learning_rate": 9.85758657138427e-06, "loss": 12.771865844726562, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.27968690916895866, "loss_total_v6": 0.7982416152954102, "step": 794 }, { "epoch": 0.18335135135135136, "grad_norm": 22.222652435302734, "learning_rate": 9.857108960244633e-06, "loss": 12.251404762268066, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.25553518906235695, "loss_total_v6": 0.7657127901911736, "step": 795 }, { "epoch": 0.18358198198198197, "grad_norm": 23.445377349853516, "learning_rate": 9.856630561173648e-06, "loss": 12.104195594787598, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.2585416156798601, "loss_total_v6": 0.7565122172236443, "step": 796 }, { "epoch": 0.1838126126126126, "grad_norm": 24.268573760986328, "learning_rate": 9.856151374248919e-06, "loss": 12.762517929077148, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.2802165746688843, "loss_total_v6": 0.797657385468483, "step": 797 }, { "epoch": 0.18404324324324325, "grad_norm": 22.748889923095703, "learning_rate": 9.85567139954818e-06, "loss": 13.074719429016113, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.30258258804678917, "loss_total_v6": 0.8171699494123459, "step": 798 }, { "epoch": 0.18427387387387387, "grad_norm": 23.278162002563477, "learning_rate": 9.855190637149299e-06, "loss": 12.398868560791016, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2651941291987896, "loss_total_v6": 0.7749292477965355, "step": 799 }, { "epoch": 0.1845045045045045, "grad_norm": 24.330097198486328, "learning_rate": 9.854709087130261e-06, "loss": 12.770423889160156, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.2741647008806467, "loss_total_v6": 0.7981515005230904, "step": 800 }, { "epoch": 0.18473513513513515, "grad_norm": 23.45793342590332, "learning_rate": 9.854226749569187e-06, "loss": 12.404296875, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2616577371954918, "loss_total_v6": 0.775268591940403, "step": 801 }, { "epoch": 0.18496576576576576, "grad_norm": 23.697263717651367, "learning_rate": 9.85374362454432e-06, "loss": 13.077534675598145, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.30179719254374504, "loss_total_v6": 0.8173459097743034, "step": 802 }, { "epoch": 0.1851963963963964, "grad_norm": 24.64661407470703, "learning_rate": 9.853259712134038e-06, "loss": 12.65250015258789, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.28344180434942245, "loss_total_v6": 0.7907812744379044, "step": 803 }, { "epoch": 0.185427027027027, "grad_norm": 23.522008895874023, "learning_rate": 9.85277501241684e-06, "loss": 12.738663673400879, "loss_alignment": 1.01708984375, "loss_alignment_w": 0.508544921875, "loss_sft": 0.2894678935408592, "loss_total_v6": 0.7961664944887161, "step": 804 }, { "epoch": 0.18565765765765765, "grad_norm": 22.655078887939453, "learning_rate": 9.852289525471353e-06, "loss": 12.733845710754395, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.281766252592206, "loss_total_v6": 0.795865386724472, "step": 805 }, { "epoch": 0.1858882882882883, "grad_norm": 24.48311996459961, "learning_rate": 9.851803251376337e-06, "loss": 13.228252410888672, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.3051745556294918, "loss_total_v6": 0.8267657458782196, "step": 806 }, { "epoch": 0.1861189189189189, "grad_norm": 25.36807632446289, "learning_rate": 9.851316190210676e-06, "loss": 12.779806137084961, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.27894726768136024, "loss_total_v6": 0.7987379133701324, "step": 807 }, { "epoch": 0.18634954954954955, "grad_norm": 23.582172393798828, "learning_rate": 9.850828342053382e-06, "loss": 12.642708778381348, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2746358644217253, "loss_total_v6": 0.7901693060994148, "step": 808 }, { "epoch": 0.1865801801801802, "grad_norm": 24.952075958251953, "learning_rate": 9.850339706983594e-06, "loss": 12.654132843017578, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2822315841913223, "loss_total_v6": 0.7908833250403404, "step": 809 }, { "epoch": 0.1868108108108108, "grad_norm": 22.736602783203125, "learning_rate": 9.84985028508058e-06, "loss": 12.518464088439941, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.2637272793799639, "loss_total_v6": 0.7824040129780769, "step": 810 }, { "epoch": 0.18704144144144144, "grad_norm": 26.540191650390625, "learning_rate": 9.849360076423736e-06, "loss": 12.565879821777344, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.26937631890177727, "loss_total_v6": 0.7853675484657288, "step": 811 }, { "epoch": 0.18727207207207208, "grad_norm": 23.708396911621094, "learning_rate": 9.848869081092581e-06, "loss": 12.726667404174805, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.28838234208524227, "loss_total_v6": 0.7954166680574417, "step": 812 }, { "epoch": 0.1875027027027027, "grad_norm": 24.052703857421875, "learning_rate": 9.848377299166771e-06, "loss": 13.095020294189453, "loss_alignment": 1.03466796875, "loss_alignment_w": 0.517333984375, "loss_sft": 0.3027832265943289, "loss_total_v6": 0.8184387236833572, "step": 813 }, { "epoch": 0.18773333333333334, "grad_norm": 23.585235595703125, "learning_rate": 9.847884730726081e-06, "loss": 12.79471206665039, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.28399877808988094, "loss_total_v6": 0.7996695339679718, "step": 814 }, { "epoch": 0.18796396396396398, "grad_norm": 25.052701950073242, "learning_rate": 9.847391375850415e-06, "loss": 12.606098175048828, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2717373501509428, "loss_total_v6": 0.7878811433911324, "step": 815 }, { "epoch": 0.1881945945945946, "grad_norm": 21.702341079711914, "learning_rate": 9.84689723461981e-06, "loss": 12.335792541503906, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.27416087314486504, "loss_total_v6": 0.7709870487451553, "step": 816 }, { "epoch": 0.18842522522522523, "grad_norm": 24.946155548095703, "learning_rate": 9.846402307114423e-06, "loss": 12.630730628967285, "loss_alignment": 1.01904296875, "loss_alignment_w": 0.509521484375, "loss_sft": 0.2791057098656893, "loss_total_v6": 0.7894206345081329, "step": 817 }, { "epoch": 0.18865585585585587, "grad_norm": 26.308483123779297, "learning_rate": 9.845906593414545e-06, "loss": 12.980985641479492, "loss_alignment": 1.03759765625, "loss_alignment_w": 0.518798828125, "loss_sft": 0.2923296894878149, "loss_total_v6": 0.8113116174936295, "step": 818 }, { "epoch": 0.18888648648648648, "grad_norm": 21.387008666992188, "learning_rate": 9.845410093600591e-06, "loss": 12.98653793334961, "loss_alignment": 1.03466796875, "loss_alignment_w": 0.517333984375, "loss_sft": 0.2938516177237034, "loss_total_v6": 0.8116586357355118, "step": 819 }, { "epoch": 0.18911711711711712, "grad_norm": 22.98248863220215, "learning_rate": 9.844912807753105e-06, "loss": 12.257267951965332, "loss_alignment": 1.02099609375, "loss_alignment_w": 0.510498046875, "loss_sft": 0.25637470558285713, "loss_total_v6": 0.766079269349575, "step": 820 }, { "epoch": 0.18934774774774774, "grad_norm": 22.083755493164062, "learning_rate": 9.844414735952754e-06, "loss": 13.223897933959961, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.3138287775218487, "loss_total_v6": 0.8264935985207558, "step": 821 }, { "epoch": 0.18957837837837838, "grad_norm": 22.325536727905273, "learning_rate": 9.843915878280342e-06, "loss": 13.130338668823242, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.31117052398622036, "loss_total_v6": 0.8206462040543556, "step": 822 }, { "epoch": 0.18980900900900902, "grad_norm": 24.48786163330078, "learning_rate": 9.843416234816791e-06, "loss": 12.765544891357422, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.289606848731637, "loss_total_v6": 0.7978465929627419, "step": 823 }, { "epoch": 0.19003963963963963, "grad_norm": 23.439231872558594, "learning_rate": 9.842915805643156e-06, "loss": 12.972540855407715, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2957539074122906, "loss_total_v6": 0.810783825814724, "step": 824 }, { "epoch": 0.19027027027027027, "grad_norm": 23.886104583740234, "learning_rate": 9.842414590840618e-06, "loss": 12.257200241088867, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.26692943274974823, "loss_total_v6": 0.76607496291399, "step": 825 }, { "epoch": 0.1905009009009009, "grad_norm": 22.396541595458984, "learning_rate": 9.841912590490484e-06, "loss": 12.463010787963867, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.26604452542960644, "loss_total_v6": 0.7789382115006447, "step": 826 }, { "epoch": 0.19073153153153152, "grad_norm": 23.83783531188965, "learning_rate": 9.84140980467419e-06, "loss": 12.42593002319336, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.2589662689715624, "loss_total_v6": 0.7766206488013268, "step": 827 }, { "epoch": 0.19096216216216216, "grad_norm": 26.750455856323242, "learning_rate": 9.840906233473298e-06, "loss": 13.05970287322998, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.29428926296532154, "loss_total_v6": 0.8162314146757126, "step": 828 }, { "epoch": 0.1911927927927928, "grad_norm": 22.92235565185547, "learning_rate": 9.840401876969501e-06, "loss": 12.694011688232422, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.2742411848157644, "loss_total_v6": 0.793375700712204, "step": 829 }, { "epoch": 0.19142342342342342, "grad_norm": 85.004150390625, "learning_rate": 9.839896735244615e-06, "loss": 12.141441345214844, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.24442055076360703, "loss_total_v6": 0.7588401138782501, "step": 830 }, { "epoch": 0.19165405405405406, "grad_norm": 22.070688247680664, "learning_rate": 9.839390808380586e-06, "loss": 12.498332977294922, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.27429457381367683, "loss_total_v6": 0.7811457961797714, "step": 831 }, { "epoch": 0.1918846846846847, "grad_norm": 24.01607894897461, "learning_rate": 9.838884096459486e-06, "loss": 12.919679641723633, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.2848512288182974, "loss_total_v6": 0.8074799925088882, "step": 832 }, { "epoch": 0.1921153153153153, "grad_norm": 23.066448211669922, "learning_rate": 9.838376599563517e-06, "loss": 12.53470230102539, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.2640707641839981, "loss_total_v6": 0.7834189161658287, "step": 833 }, { "epoch": 0.19234594594594595, "grad_norm": 31.00482940673828, "learning_rate": 9.837868317775002e-06, "loss": 13.262774467468262, "loss_alignment": 1.046875, "loss_alignment_w": 0.5234375, "loss_sft": 0.30499764904379845, "loss_total_v6": 0.8289234191179276, "step": 834 }, { "epoch": 0.19257657657657656, "grad_norm": 24.285654067993164, "learning_rate": 9.8373592511764e-06, "loss": 12.722582817077637, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.2922317199409008, "loss_total_v6": 0.7951614260673523, "step": 835 }, { "epoch": 0.1928072072072072, "grad_norm": 21.2269344329834, "learning_rate": 9.836849399850291e-06, "loss": 12.302351951599121, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2535008601844311, "loss_total_v6": 0.7688969820737839, "step": 836 }, { "epoch": 0.19303783783783784, "grad_norm": 20.61589241027832, "learning_rate": 9.836338763879386e-06, "loss": 12.614731788635254, "loss_alignment": 1.03271484375, "loss_alignment_w": 0.516357421875, "loss_sft": 0.2724142577499151, "loss_total_v6": 0.7884207218885422, "step": 837 }, { "epoch": 0.19326846846846846, "grad_norm": 24.00278663635254, "learning_rate": 9.835827343346518e-06, "loss": 12.824377059936523, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.28219071961939335, "loss_total_v6": 0.8015236034989357, "step": 838 }, { "epoch": 0.1934990990990991, "grad_norm": 22.658174514770508, "learning_rate": 9.835315138334653e-06, "loss": 12.452364921569824, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.25991648249328136, "loss_total_v6": 0.778272807598114, "step": 839 }, { "epoch": 0.19372972972972974, "grad_norm": 25.703563690185547, "learning_rate": 9.834802148926883e-06, "loss": 12.824379920959473, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.28983550146222115, "loss_total_v6": 0.8015237227082253, "step": 840 }, { "epoch": 0.19396036036036035, "grad_norm": 24.417560577392578, "learning_rate": 9.834288375206424e-06, "loss": 12.445411682128906, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.26650096848607063, "loss_total_v6": 0.7778382152318954, "step": 841 }, { "epoch": 0.194190990990991, "grad_norm": 24.789108276367188, "learning_rate": 9.833773817256624e-06, "loss": 12.711368560791016, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.27853039652109146, "loss_total_v6": 0.7944605648517609, "step": 842 }, { "epoch": 0.19442162162162163, "grad_norm": 24.179861068725586, "learning_rate": 9.833258475160955e-06, "loss": 12.843461990356445, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2920047137886286, "loss_total_v6": 0.8027163371443748, "step": 843 }, { "epoch": 0.19465225225225224, "grad_norm": 23.682903289794922, "learning_rate": 9.832742349003014e-06, "loss": 12.835357666015625, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.28698157519102097, "loss_total_v6": 0.8022098541259766, "step": 844 }, { "epoch": 0.19488288288288289, "grad_norm": 24.17443084716797, "learning_rate": 9.832225438866532e-06, "loss": 13.002765655517578, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.29430125653743744, "loss_total_v6": 0.8126728311181068, "step": 845 }, { "epoch": 0.19511351351351353, "grad_norm": 23.436155319213867, "learning_rate": 9.831707744835363e-06, "loss": 13.269872665405273, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.30283197201788425, "loss_total_v6": 0.8293670192360878, "step": 846 }, { "epoch": 0.19534414414414414, "grad_norm": 26.89253044128418, "learning_rate": 9.831189266993486e-06, "loss": 12.870537757873535, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2914538737386465, "loss_total_v6": 0.8044086024165154, "step": 847 }, { "epoch": 0.19557477477477478, "grad_norm": 22.147350311279297, "learning_rate": 9.830670005425012e-06, "loss": 12.608285903930664, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.26866975240409374, "loss_total_v6": 0.7880178615450859, "step": 848 }, { "epoch": 0.1958054054054054, "grad_norm": 33.54662322998047, "learning_rate": 9.830149960214177e-06, "loss": 12.20762825012207, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2509375885128975, "loss_total_v6": 0.7629767805337906, "step": 849 }, { "epoch": 0.19603603603603603, "grad_norm": 23.749765396118164, "learning_rate": 9.829629131445342e-06, "loss": 12.730212211608887, "loss_alignment": 1.0556640625, "loss_alignment_w": 0.52783203125, "loss_sft": 0.268019812181592, "loss_total_v6": 0.7956382557749748, "step": 850 }, { "epoch": 0.19626666666666667, "grad_norm": 22.246925354003906, "learning_rate": 9.829107519202999e-06, "loss": 12.344453811645508, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.26240357756614685, "loss_total_v6": 0.7715283185243607, "step": 851 }, { "epoch": 0.19649729729729729, "grad_norm": 23.584495544433594, "learning_rate": 9.828585123571763e-06, "loss": 13.304180145263672, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.3223407119512558, "loss_total_v6": 0.8315112814307213, "step": 852 }, { "epoch": 0.19672792792792793, "grad_norm": 23.455537796020508, "learning_rate": 9.828061944636381e-06, "loss": 12.59595775604248, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.26438965275883675, "loss_total_v6": 0.7872473523020744, "step": 853 }, { "epoch": 0.19695855855855857, "grad_norm": 22.109041213989258, "learning_rate": 9.827537982481721e-06, "loss": 12.88174819946289, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.28196172416210175, "loss_total_v6": 0.8051092624664307, "step": 854 }, { "epoch": 0.19718918918918918, "grad_norm": 27.144128799438477, "learning_rate": 9.827013237192781e-06, "loss": 12.879188537597656, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.30194331891834736, "loss_total_v6": 0.8049492761492729, "step": 855 }, { "epoch": 0.19741981981981982, "grad_norm": 23.720924377441406, "learning_rate": 9.826487708854693e-06, "loss": 13.080975532531738, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.2917736414819956, "loss_total_v6": 0.8175609856843948, "step": 856 }, { "epoch": 0.19765045045045046, "grad_norm": 24.488256454467773, "learning_rate": 9.825961397552702e-06, "loss": 12.49906063079834, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2711662743240595, "loss_total_v6": 0.7811913192272186, "step": 857 }, { "epoch": 0.19788108108108107, "grad_norm": 24.45775032043457, "learning_rate": 9.82543430337219e-06, "loss": 12.74183177947998, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.27549050375819206, "loss_total_v6": 0.7963645085692406, "step": 858 }, { "epoch": 0.1981117117117117, "grad_norm": 30.167118072509766, "learning_rate": 9.824906426398665e-06, "loss": 12.64076042175293, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.2751396521925926, "loss_total_v6": 0.7900475040078163, "step": 859 }, { "epoch": 0.19834234234234235, "grad_norm": 23.58637809753418, "learning_rate": 9.824377766717758e-06, "loss": 12.819881439208984, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.28767759911715984, "loss_total_v6": 0.8012426421046257, "step": 860 }, { "epoch": 0.19857297297297297, "grad_norm": 23.24453353881836, "learning_rate": 9.82384832441523e-06, "loss": 12.320194244384766, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2585070785135031, "loss_total_v6": 0.7700121849775314, "step": 861 }, { "epoch": 0.1988036036036036, "grad_norm": 25.025495529174805, "learning_rate": 9.82331809957697e-06, "loss": 12.665337562561035, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2789798304438591, "loss_total_v6": 0.7915835976600647, "step": 862 }, { "epoch": 0.19903423423423425, "grad_norm": 25.3750057220459, "learning_rate": 9.822787092288991e-06, "loss": 12.433670043945312, "loss_alignment": 1.02587890625, "loss_alignment_w": 0.512939453125, "loss_sft": 0.26376816630363464, "loss_total_v6": 0.777104340493679, "step": 863 }, { "epoch": 0.19926486486486486, "grad_norm": 24.038957595825195, "learning_rate": 9.822255302637434e-06, "loss": 12.561774253845215, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.267837954685092, "loss_total_v6": 0.7851109206676483, "step": 864 }, { "epoch": 0.1994954954954955, "grad_norm": 36.10112762451172, "learning_rate": 9.821722730708567e-06, "loss": 12.980497360229492, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.28666869550943375, "loss_total_v6": 0.8112811222672462, "step": 865 }, { "epoch": 0.1997261261261261, "grad_norm": 22.575359344482422, "learning_rate": 9.821189376588787e-06, "loss": 12.657843589782715, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2811207100749016, "loss_total_v6": 0.7911152318120003, "step": 866 }, { "epoch": 0.19995675675675675, "grad_norm": 25.06310272216797, "learning_rate": 9.820655240364612e-06, "loss": 12.015800476074219, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.24856137484312057, "loss_total_v6": 0.7509875148534775, "step": 867 }, { "epoch": 0.2001873873873874, "grad_norm": 22.266855239868164, "learning_rate": 9.820120322122695e-06, "loss": 12.340289115905762, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.25909156166017056, "loss_total_v6": 0.7712680995464325, "step": 868 }, { "epoch": 0.200418018018018, "grad_norm": 23.878393173217773, "learning_rate": 9.81958462194981e-06, "loss": 12.598800659179688, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.270991288125515, "loss_total_v6": 0.7874250262975693, "step": 869 }, { "epoch": 0.20064864864864865, "grad_norm": 23.707687377929688, "learning_rate": 9.819048139932858e-06, "loss": 12.85194206237793, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.2873161491006613, "loss_total_v6": 0.8032463192939758, "step": 870 }, { "epoch": 0.2008792792792793, "grad_norm": 23.54559898376465, "learning_rate": 9.818510876158871e-06, "loss": 12.40281867980957, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.26495280861854553, "loss_total_v6": 0.7751762121915817, "step": 871 }, { "epoch": 0.2011099099099099, "grad_norm": 22.475618362426758, "learning_rate": 9.817972830715003e-06, "loss": 13.012706756591797, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.2941291518509388, "loss_total_v6": 0.8132942020893097, "step": 872 }, { "epoch": 0.20134054054054054, "grad_norm": 23.14457130432129, "learning_rate": 9.817434003688537e-06, "loss": 12.374889373779297, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.26439739763736725, "loss_total_v6": 0.7734305784106255, "step": 873 }, { "epoch": 0.20157117117117118, "grad_norm": 22.219589233398438, "learning_rate": 9.816894395166885e-06, "loss": 12.796346664428711, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.28873955458402634, "loss_total_v6": 0.7997716814279556, "step": 874 }, { "epoch": 0.2018018018018018, "grad_norm": 20.445289611816406, "learning_rate": 9.816354005237583e-06, "loss": 12.404266357421875, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2641430664807558, "loss_total_v6": 0.7752666920423508, "step": 875 }, { "epoch": 0.20203243243243243, "grad_norm": 22.716764450073242, "learning_rate": 9.815812833988292e-06, "loss": 13.037355422973633, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.2935944925993681, "loss_total_v6": 0.8148347064852715, "step": 876 }, { "epoch": 0.20226306306306308, "grad_norm": 24.262781143188477, "learning_rate": 9.815270881506804e-06, "loss": 12.580785751342773, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.275526350364089, "loss_total_v6": 0.7862990871071815, "step": 877 }, { "epoch": 0.2024936936936937, "grad_norm": 25.184112548828125, "learning_rate": 9.814728147881034e-06, "loss": 12.534261703491211, "loss_alignment": 1.01904296875, "loss_alignment_w": 0.509521484375, "loss_sft": 0.273473160341382, "loss_total_v6": 0.7833913788199425, "step": 878 }, { "epoch": 0.20272432432432433, "grad_norm": 26.66155433654785, "learning_rate": 9.814184633199027e-06, "loss": 12.85708236694336, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.28652349673211575, "loss_total_v6": 0.8035675957798958, "step": 879 }, { "epoch": 0.20295495495495494, "grad_norm": 22.637008666992188, "learning_rate": 9.813640337548955e-06, "loss": 12.438297271728516, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.26845198310911655, "loss_total_v6": 0.7773936167359352, "step": 880 }, { "epoch": 0.20318558558558558, "grad_norm": 23.498458862304688, "learning_rate": 9.81309526101911e-06, "loss": 12.768396377563477, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2836967520415783, "loss_total_v6": 0.7980247512459755, "step": 881 }, { "epoch": 0.20341621621621622, "grad_norm": 24.4376277923584, "learning_rate": 9.812549403697919e-06, "loss": 12.847145080566406, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.29159399308264256, "loss_total_v6": 0.802946537733078, "step": 882 }, { "epoch": 0.20364684684684684, "grad_norm": 24.316471099853516, "learning_rate": 9.812002765673934e-06, "loss": 12.579879760742188, "loss_alignment": 1.03466796875, "loss_alignment_w": 0.517333984375, "loss_sft": 0.268740676343441, "loss_total_v6": 0.7862425148487091, "step": 883 }, { "epoch": 0.20387747747747748, "grad_norm": 23.906715393066406, "learning_rate": 9.811455347035827e-06, "loss": 12.785493850708008, "loss_alignment": 1.01708984375, "loss_alignment_w": 0.508544921875, "loss_sft": 0.2923337444663048, "loss_total_v6": 0.7990933805704117, "step": 884 }, { "epoch": 0.20410810810810812, "grad_norm": 23.311412811279297, "learning_rate": 9.810907147872407e-06, "loss": 13.171785354614258, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.3129979521036148, "loss_total_v6": 0.8232365995645523, "step": 885 }, { "epoch": 0.20433873873873873, "grad_norm": 26.433364868164062, "learning_rate": 9.810358168272597e-06, "loss": 12.78659439086914, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2903731130063534, "loss_total_v6": 0.7991621568799019, "step": 886 }, { "epoch": 0.20456936936936937, "grad_norm": 24.586984634399414, "learning_rate": 9.80980840832546e-06, "loss": 12.71041488647461, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2814004495739937, "loss_total_v6": 0.7944009155035019, "step": 887 }, { "epoch": 0.2048, "grad_norm": 24.097068786621094, "learning_rate": 9.809257868120176e-06, "loss": 12.578222274780273, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.2662567310035229, "loss_total_v6": 0.7861389368772507, "step": 888 }, { "epoch": 0.20503063063063062, "grad_norm": 24.67955780029297, "learning_rate": 9.808706547746057e-06, "loss": 12.912408828735352, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2879062108695507, "loss_total_v6": 0.8070255070924759, "step": 889 }, { "epoch": 0.20526126126126126, "grad_norm": 23.97585678100586, "learning_rate": 9.808154447292539e-06, "loss": 12.20859432220459, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.25734557770192623, "loss_total_v6": 0.7630371302366257, "step": 890 }, { "epoch": 0.2054918918918919, "grad_norm": 22.874736785888672, "learning_rate": 9.807601566849183e-06, "loss": 13.092277526855469, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.3018946759402752, "loss_total_v6": 0.8182673826813698, "step": 891 }, { "epoch": 0.20572252252252252, "grad_norm": 20.47365951538086, "learning_rate": 9.807047906505683e-06, "loss": 12.028820991516113, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.2509315926581621, "loss_total_v6": 0.7518013119697571, "step": 892 }, { "epoch": 0.20595315315315316, "grad_norm": 24.173322677612305, "learning_rate": 9.806493466351849e-06, "loss": 12.504966735839844, "loss_alignment": 1.02197265625, "loss_alignment_w": 0.510986328125, "loss_sft": 0.2703909892588854, "loss_total_v6": 0.781560406088829, "step": 893 }, { "epoch": 0.2061837837837838, "grad_norm": 22.11171531677246, "learning_rate": 9.805938246477627e-06, "loss": 12.46689224243164, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2651579864323139, "loss_total_v6": 0.7791808173060417, "step": 894 }, { "epoch": 0.2064144144144144, "grad_norm": 23.77330780029297, "learning_rate": 9.805382246973088e-06, "loss": 12.735130310058594, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.2762770652770996, "loss_total_v6": 0.7959456443786621, "step": 895 }, { "epoch": 0.20664504504504505, "grad_norm": 22.555566787719727, "learning_rate": 9.804825467928423e-06, "loss": 12.831475257873535, "loss_alignment": 1.02587890625, "loss_alignment_w": 0.512939453125, "loss_sft": 0.28972965106368065, "loss_total_v6": 0.8019671812653542, "step": 896 }, { "epoch": 0.20687567567567566, "grad_norm": 24.69898223876953, "learning_rate": 9.804267909433956e-06, "loss": 12.327085494995117, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2572287693619728, "loss_total_v6": 0.7704428732395172, "step": 897 }, { "epoch": 0.2071063063063063, "grad_norm": 24.617233276367188, "learning_rate": 9.803709571580135e-06, "loss": 12.611566543579102, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2702632453292608, "loss_total_v6": 0.7882228717207909, "step": 898 }, { "epoch": 0.20733693693693694, "grad_norm": 24.216306686401367, "learning_rate": 9.803150454457538e-06, "loss": 12.861247062683105, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.29293316788971424, "loss_total_v6": 0.8038279339671135, "step": 899 }, { "epoch": 0.20756756756756756, "grad_norm": 23.406410217285156, "learning_rate": 9.802590558156863e-06, "loss": 12.439424514770508, "loss_alignment": 1.02783203125, "loss_alignment_w": 0.513916015625, "loss_sft": 0.2616559434682131, "loss_total_v6": 0.7774640545248985, "step": 900 }, { "epoch": 0.2077981981981982, "grad_norm": 25.050006866455078, "learning_rate": 9.802029882768938e-06, "loss": 13.009140968322754, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.29753781482577324, "loss_total_v6": 0.8130712881684303, "step": 901 }, { "epoch": 0.20802882882882884, "grad_norm": 24.865034103393555, "learning_rate": 9.801468428384716e-06, "loss": 12.552845001220703, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.27681659534573555, "loss_total_v6": 0.7845527902245522, "step": 902 }, { "epoch": 0.20825945945945945, "grad_norm": 25.13810157775879, "learning_rate": 9.80090619509528e-06, "loss": 12.946067810058594, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2960829418152571, "loss_total_v6": 0.8091292157769203, "step": 903 }, { "epoch": 0.2084900900900901, "grad_norm": 23.42059898376465, "learning_rate": 9.800343182991837e-06, "loss": 12.64474868774414, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.27116224728524685, "loss_total_v6": 0.7902967631816864, "step": 904 }, { "epoch": 0.20872072072072073, "grad_norm": 23.095239639282227, "learning_rate": 9.799779392165716e-06, "loss": 12.949701309204102, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2997127082198858, "loss_total_v6": 0.8093562871217728, "step": 905 }, { "epoch": 0.20895135135135134, "grad_norm": 22.288990020751953, "learning_rate": 9.799214822708381e-06, "loss": 12.708976745605469, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.28007457964122295, "loss_total_v6": 0.7943110540509224, "step": 906 }, { "epoch": 0.20918198198198198, "grad_norm": 23.011493682861328, "learning_rate": 9.798649474711415e-06, "loss": 12.621576309204102, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.281188553199172, "loss_total_v6": 0.7888484820723534, "step": 907 }, { "epoch": 0.20941261261261263, "grad_norm": 22.954992294311523, "learning_rate": 9.798083348266531e-06, "loss": 12.20975112915039, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2563192415982485, "loss_total_v6": 0.7631094083189964, "step": 908 }, { "epoch": 0.20964324324324324, "grad_norm": 26.9117488861084, "learning_rate": 9.797516443465566e-06, "loss": 12.87826919555664, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.29334087297320366, "loss_total_v6": 0.804891787469387, "step": 909 }, { "epoch": 0.20987387387387388, "grad_norm": 22.513912200927734, "learning_rate": 9.796948760400488e-06, "loss": 12.328800201416016, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.2543604616075754, "loss_total_v6": 0.7705500200390816, "step": 910 }, { "epoch": 0.2101045045045045, "grad_norm": 24.810108184814453, "learning_rate": 9.796380299163386e-06, "loss": 12.999740600585938, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.29870512150228024, "loss_total_v6": 0.8124837800860405, "step": 911 }, { "epoch": 0.21033513513513513, "grad_norm": 23.534339904785156, "learning_rate": 9.795811059846476e-06, "loss": 13.008622169494629, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.29401116631925106, "loss_total_v6": 0.8130388930439949, "step": 912 }, { "epoch": 0.21056576576576577, "grad_norm": 25.074251174926758, "learning_rate": 9.7952410425421e-06, "loss": 12.850990295410156, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2900490537285805, "loss_total_v6": 0.803186871111393, "step": 913 }, { "epoch": 0.21079639639639638, "grad_norm": 25.146738052368164, "learning_rate": 9.794670247342732e-06, "loss": 12.476019859313965, "loss_alignment": 1.01708984375, "loss_alignment_w": 0.508544921875, "loss_sft": 0.27326624281704426, "loss_total_v6": 0.7797512412071228, "step": 914 }, { "epoch": 0.21102702702702703, "grad_norm": 22.972991943359375, "learning_rate": 9.794098674340966e-06, "loss": 13.000737190246582, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.29356417432427406, "loss_total_v6": 0.8125460743904114, "step": 915 }, { "epoch": 0.21125765765765767, "grad_norm": 24.24443244934082, "learning_rate": 9.793526323629522e-06, "loss": 12.933666229248047, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2983138207346201, "loss_total_v6": 0.8083541095256805, "step": 916 }, { "epoch": 0.21148828828828828, "grad_norm": 24.101886749267578, "learning_rate": 9.79295319530125e-06, "loss": 12.355388641357422, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.27114369347691536, "loss_total_v6": 0.7722118124365807, "step": 917 }, { "epoch": 0.21171891891891892, "grad_norm": 24.54798126220703, "learning_rate": 9.792379289449122e-06, "loss": 12.918851852416992, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.29166590981185436, "loss_total_v6": 0.8074282631278038, "step": 918 }, { "epoch": 0.21194954954954956, "grad_norm": 24.810039520263672, "learning_rate": 9.791804606166241e-06, "loss": 12.743799209594727, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.27668151818215847, "loss_total_v6": 0.7964874729514122, "step": 919 }, { "epoch": 0.21218018018018017, "grad_norm": 24.76127815246582, "learning_rate": 9.791229145545832e-06, "loss": 12.885577201843262, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2876026127487421, "loss_total_v6": 0.805348590016365, "step": 920 }, { "epoch": 0.2124108108108108, "grad_norm": 24.586305618286133, "learning_rate": 9.790652907681248e-06, "loss": 12.390798568725586, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2563737239688635, "loss_total_v6": 0.7744248807430267, "step": 921 }, { "epoch": 0.21264144144144145, "grad_norm": 22.77292251586914, "learning_rate": 9.790075892665966e-06, "loss": 12.670310974121094, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2741941921412945, "loss_total_v6": 0.7918944209814072, "step": 922 }, { "epoch": 0.21287207207207207, "grad_norm": 23.590593338012695, "learning_rate": 9.789498100593593e-06, "loss": 12.538349151611328, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.2644818089902401, "loss_total_v6": 0.7836468368768692, "step": 923 }, { "epoch": 0.2131027027027027, "grad_norm": 23.839313507080078, "learning_rate": 9.78891953155786e-06, "loss": 12.772065162658691, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.28952595591545105, "loss_total_v6": 0.7982540279626846, "step": 924 }, { "epoch": 0.21333333333333335, "grad_norm": 26.022855758666992, "learning_rate": 9.78834018565262e-06, "loss": 13.4238920211792, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.31584561988711357, "loss_total_v6": 0.8389932066202164, "step": 925 }, { "epoch": 0.21356396396396396, "grad_norm": 25.590866088867188, "learning_rate": 9.78776006297186e-06, "loss": 12.270284652709961, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2532819677144289, "loss_total_v6": 0.766892783343792, "step": 926 }, { "epoch": 0.2137945945945946, "grad_norm": 21.347925186157227, "learning_rate": 9.787179163609688e-06, "loss": 12.238226890563965, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.26842925511300564, "loss_total_v6": 0.7648892104625702, "step": 927 }, { "epoch": 0.2140252252252252, "grad_norm": 24.23967933654785, "learning_rate": 9.786597487660336e-06, "loss": 12.74319076538086, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2820298243314028, "loss_total_v6": 0.7964493781328201, "step": 928 }, { "epoch": 0.21425585585585585, "grad_norm": 24.448684692382812, "learning_rate": 9.78601503521817e-06, "loss": 12.433825492858887, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2671653553843498, "loss_total_v6": 0.777114100754261, "step": 929 }, { "epoch": 0.2144864864864865, "grad_norm": 23.131114959716797, "learning_rate": 9.785431806377672e-06, "loss": 12.672477722167969, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2779612150043249, "loss_total_v6": 0.7920298129320145, "step": 930 }, { "epoch": 0.2147171171171171, "grad_norm": 23.230182647705078, "learning_rate": 9.784847801233454e-06, "loss": 12.584029197692871, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.2827786933630705, "loss_total_v6": 0.786501832306385, "step": 931 }, { "epoch": 0.21494774774774775, "grad_norm": 24.895198822021484, "learning_rate": 9.78426301988026e-06, "loss": 12.443059921264648, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2649349085986614, "loss_total_v6": 0.7776912376284599, "step": 932 }, { "epoch": 0.2151783783783784, "grad_norm": 23.084043502807617, "learning_rate": 9.783677462412951e-06, "loss": 12.751327514648438, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.27609923109412193, "loss_total_v6": 0.7969580069184303, "step": 933 }, { "epoch": 0.215409009009009, "grad_norm": 23.080846786499023, "learning_rate": 9.783091128926519e-06, "loss": 12.448031425476074, "loss_alignment": 1.02294921875, "loss_alignment_w": 0.511474609375, "loss_sft": 0.2654592264443636, "loss_total_v6": 0.7780019640922546, "step": 934 }, { "epoch": 0.21563963963963964, "grad_norm": 24.710458755493164, "learning_rate": 9.782504019516079e-06, "loss": 13.027252197265625, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.2950534485280514, "loss_total_v6": 0.8142032623291016, "step": 935 }, { "epoch": 0.21587027027027028, "grad_norm": 24.77289581298828, "learning_rate": 9.781916134276873e-06, "loss": 12.80600357055664, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2843534480780363, "loss_total_v6": 0.8003751784563065, "step": 936 }, { "epoch": 0.2161009009009009, "grad_norm": 22.656431198120117, "learning_rate": 9.78132747330427e-06, "loss": 12.626543045043945, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.27908814884722233, "loss_total_v6": 0.7891589105129242, "step": 937 }, { "epoch": 0.21633153153153153, "grad_norm": 21.381162643432617, "learning_rate": 9.780738036693764e-06, "loss": 12.78838062286377, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.28590708784759045, "loss_total_v6": 0.799273781478405, "step": 938 }, { "epoch": 0.21656216216216218, "grad_norm": 23.362836837768555, "learning_rate": 9.780147824540973e-06, "loss": 12.098993301391602, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.24588735960423946, "loss_total_v6": 0.7561870217323303, "step": 939 }, { "epoch": 0.2167927927927928, "grad_norm": 23.030689239501953, "learning_rate": 9.779556836941646e-06, "loss": 13.011414527893066, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.30387500301003456, "loss_total_v6": 0.8132133856415749, "step": 940 }, { "epoch": 0.21702342342342343, "grad_norm": 26.99878692626953, "learning_rate": 9.778965073991652e-06, "loss": 12.148212432861328, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.2585613578557968, "loss_total_v6": 0.7592632621526718, "step": 941 }, { "epoch": 0.21725405405405404, "grad_norm": 23.704973220825195, "learning_rate": 9.778372535786987e-06, "loss": 12.685983657836914, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2817198671400547, "loss_total_v6": 0.7928740233182907, "step": 942 }, { "epoch": 0.21748468468468468, "grad_norm": 23.361278533935547, "learning_rate": 9.777779222423778e-06, "loss": 12.747486114501953, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.28185587748885155, "loss_total_v6": 0.7967179492115974, "step": 943 }, { "epoch": 0.21771531531531532, "grad_norm": 23.20418930053711, "learning_rate": 9.777185133998268e-06, "loss": 13.148174285888672, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.3042743355035782, "loss_total_v6": 0.8217608854174614, "step": 944 }, { "epoch": 0.21794594594594593, "grad_norm": 25.07304573059082, "learning_rate": 9.776590270606837e-06, "loss": 12.782940864562988, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.28036386892199516, "loss_total_v6": 0.7989338114857674, "step": 945 }, { "epoch": 0.21817657657657658, "grad_norm": 22.76459503173828, "learning_rate": 9.775994632345982e-06, "loss": 12.604528427124023, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.28193895146250725, "loss_total_v6": 0.7877830266952515, "step": 946 }, { "epoch": 0.21840720720720722, "grad_norm": 22.461545944213867, "learning_rate": 9.775398219312328e-06, "loss": 13.161272048950195, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.300469521433115, "loss_total_v6": 0.8225795105099678, "step": 947 }, { "epoch": 0.21863783783783783, "grad_norm": 22.90011978149414, "learning_rate": 9.77480103160263e-06, "loss": 12.669987678527832, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.27821763791143894, "loss_total_v6": 0.7918742373585701, "step": 948 }, { "epoch": 0.21886846846846847, "grad_norm": 23.694366455078125, "learning_rate": 9.774203069313763e-06, "loss": 12.661758422851562, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.27665041387081146, "loss_total_v6": 0.7913598865270615, "step": 949 }, { "epoch": 0.2190990990990991, "grad_norm": 23.701505661010742, "learning_rate": 9.77360433254273e-06, "loss": 12.209529876708984, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.24899648502469063, "loss_total_v6": 0.7630956023931503, "step": 950 }, { "epoch": 0.21932972972972972, "grad_norm": 30.17262077331543, "learning_rate": 9.773004821386659e-06, "loss": 12.15738296508789, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2556708101183176, "loss_total_v6": 0.759836457669735, "step": 951 }, { "epoch": 0.21956036036036036, "grad_norm": 24.73659896850586, "learning_rate": 9.772404535942802e-06, "loss": 11.998815536499023, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.24337994307279587, "loss_total_v6": 0.7499259412288666, "step": 952 }, { "epoch": 0.219790990990991, "grad_norm": 23.66141128540039, "learning_rate": 9.771803476308544e-06, "loss": 12.81752872467041, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.2816864028573036, "loss_total_v6": 0.8010955601930618, "step": 953 }, { "epoch": 0.22002162162162162, "grad_norm": 21.46739959716797, "learning_rate": 9.771201642581384e-06, "loss": 12.514287948608398, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.2631458304822445, "loss_total_v6": 0.7821429818868637, "step": 954 }, { "epoch": 0.22025225225225226, "grad_norm": 23.318098068237305, "learning_rate": 9.77059903485896e-06, "loss": 12.799177169799805, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2924565467983484, "loss_total_v6": 0.7999485731124878, "step": 955 }, { "epoch": 0.22048288288288287, "grad_norm": 26.61558723449707, "learning_rate": 9.769995653239023e-06, "loss": 13.142053604125977, "loss_alignment": 1.02978515625, "loss_alignment_w": 0.514892578125, "loss_sft": 0.3065926060080528, "loss_total_v6": 0.8213783726096153, "step": 956 }, { "epoch": 0.2207135135135135, "grad_norm": 23.6624698638916, "learning_rate": 9.769391497819456e-06, "loss": 13.126968383789062, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.30537512712180614, "loss_total_v6": 0.8204355761408806, "step": 957 }, { "epoch": 0.22094414414414415, "grad_norm": 26.68526268005371, "learning_rate": 9.768786568698267e-06, "loss": 12.213525772094727, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.25017702765762806, "loss_total_v6": 0.7633453533053398, "step": 958 }, { "epoch": 0.22117477477477476, "grad_norm": 28.230464935302734, "learning_rate": 9.768180865973591e-06, "loss": 12.998672485351562, "loss_alignment": 1.0576171875, "loss_alignment_w": 0.52880859375, "loss_sft": 0.2837304659187794, "loss_total_v6": 0.8124170005321503, "step": 959 }, { "epoch": 0.2214054054054054, "grad_norm": 22.9271183013916, "learning_rate": 9.767574389743683e-06, "loss": 12.831197738647461, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.28475323505699635, "loss_total_v6": 0.8019498959183693, "step": 960 }, { "epoch": 0.22163603603603604, "grad_norm": 25.279970169067383, "learning_rate": 9.766967140106928e-06, "loss": 12.913888931274414, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.29565870575606823, "loss_total_v6": 0.8071180284023285, "step": 961 }, { "epoch": 0.22186666666666666, "grad_norm": 26.795703887939453, "learning_rate": 9.766359117161836e-06, "loss": 12.182280540466309, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.25560944713652134, "loss_total_v6": 0.7613925263285637, "step": 962 }, { "epoch": 0.2220972972972973, "grad_norm": 22.71122932434082, "learning_rate": 9.765750321007042e-06, "loss": 12.834102630615234, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.29152660444378853, "loss_total_v6": 0.8021314665675163, "step": 963 }, { "epoch": 0.22232792792792794, "grad_norm": 23.32827377319336, "learning_rate": 9.765140751741307e-06, "loss": 12.391365051269531, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.2705997917801142, "loss_total_v6": 0.7744603082537651, "step": 964 }, { "epoch": 0.22255855855855855, "grad_norm": 28.330554962158203, "learning_rate": 9.764530409463516e-06, "loss": 12.980143547058105, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2921549715101719, "loss_total_v6": 0.8112589567899704, "step": 965 }, { "epoch": 0.2227891891891892, "grad_norm": 27.965320587158203, "learning_rate": 9.763919294272678e-06, "loss": 12.584718704223633, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.27093516662716866, "loss_total_v6": 0.7865448668599129, "step": 966 }, { "epoch": 0.22301981981981983, "grad_norm": 25.499052047729492, "learning_rate": 9.763307406267933e-06, "loss": 12.968740463256836, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.29554699547588825, "loss_total_v6": 0.810546338558197, "step": 967 }, { "epoch": 0.22325045045045044, "grad_norm": 22.79706573486328, "learning_rate": 9.76269474554854e-06, "loss": 12.30155086517334, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.25093308836221695, "loss_total_v6": 0.7688468918204308, "step": 968 }, { "epoch": 0.22348108108108108, "grad_norm": 24.539203643798828, "learning_rate": 9.762081312213889e-06, "loss": 12.3358736038208, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.25721338018774986, "loss_total_v6": 0.7709921002388, "step": 969 }, { "epoch": 0.22371171171171172, "grad_norm": 24.174449920654297, "learning_rate": 9.761467106363488e-06, "loss": 12.360563278198242, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2617777809500694, "loss_total_v6": 0.7725352197885513, "step": 970 }, { "epoch": 0.22394234234234234, "grad_norm": 28.611881256103516, "learning_rate": 9.760852128096981e-06, "loss": 12.70675277709961, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.28120209462940693, "loss_total_v6": 0.7941720560193062, "step": 971 }, { "epoch": 0.22417297297297298, "grad_norm": 24.920175552368164, "learning_rate": 9.760236377514128e-06, "loss": 12.61388111114502, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2715981788933277, "loss_total_v6": 0.7883675694465637, "step": 972 }, { "epoch": 0.2244036036036036, "grad_norm": 23.32066535949707, "learning_rate": 9.759619854714817e-06, "loss": 12.505218505859375, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2711544018238783, "loss_total_v6": 0.7815761417150497, "step": 973 }, { "epoch": 0.22463423423423423, "grad_norm": 21.807153701782227, "learning_rate": 9.759002559799062e-06, "loss": 12.700345039367676, "loss_alignment": 1.01513671875, "loss_alignment_w": 0.507568359375, "loss_sft": 0.2862642202526331, "loss_total_v6": 0.7937715500593185, "step": 974 }, { "epoch": 0.22486486486486487, "grad_norm": 22.84084701538086, "learning_rate": 9.758384492867004e-06, "loss": 12.461706161499023, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2731040846556425, "loss_total_v6": 0.7788566648960114, "step": 975 }, { "epoch": 0.22509549549549548, "grad_norm": 24.331758499145508, "learning_rate": 9.757765654018905e-06, "loss": 12.686637878417969, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2836984973400831, "loss_total_v6": 0.7929148301482201, "step": 976 }, { "epoch": 0.22532612612612613, "grad_norm": 21.90334701538086, "learning_rate": 9.757146043355154e-06, "loss": 12.610679626464844, "loss_alignment": 1.0595703125, "loss_alignment_w": 0.52978515625, "loss_sft": 0.2597403768450022, "loss_total_v6": 0.7881674915552139, "step": 977 }, { "epoch": 0.22555675675675677, "grad_norm": 22.395009994506836, "learning_rate": 9.756525660976266e-06, "loss": 12.360126495361328, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.2697155810892582, "loss_total_v6": 0.7725079283118248, "step": 978 }, { "epoch": 0.22578738738738738, "grad_norm": 23.05516242980957, "learning_rate": 9.755904506982884e-06, "loss": 12.189109802246094, "loss_alignment": 1.02392578125, "loss_alignment_w": 0.511962890625, "loss_sft": 0.25016165897250175, "loss_total_v6": 0.7618193626403809, "step": 979 }, { "epoch": 0.22601801801801802, "grad_norm": 25.938692092895508, "learning_rate": 9.755282581475769e-06, "loss": 13.146878242492676, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.2999513987451792, "loss_total_v6": 0.8216798901557922, "step": 980 }, { "epoch": 0.22624864864864866, "grad_norm": 23.934329986572266, "learning_rate": 9.754659884555813e-06, "loss": 12.523843765258789, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.27422584779560566, "loss_total_v6": 0.7827402576804161, "step": 981 }, { "epoch": 0.22647927927927927, "grad_norm": 24.479318618774414, "learning_rate": 9.75403641632403e-06, "loss": 12.651323318481445, "loss_alignment": 1.04150390625, "loss_alignment_w": 0.520751953125, "loss_sft": 0.26815519481897354, "loss_total_v6": 0.7907076850533485, "step": 982 }, { "epoch": 0.2267099099099099, "grad_norm": 26.56537437438965, "learning_rate": 9.753412176881564e-06, "loss": 12.573041915893555, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.27374546229839325, "loss_total_v6": 0.7858151495456696, "step": 983 }, { "epoch": 0.22694054054054055, "grad_norm": 25.122499465942383, "learning_rate": 9.752787166329677e-06, "loss": 12.474035263061523, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.25902785919606686, "loss_total_v6": 0.7796271815896034, "step": 984 }, { "epoch": 0.22717117117117117, "grad_norm": 23.874195098876953, "learning_rate": 9.75216138476976e-06, "loss": 12.837128639221191, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2872601170092821, "loss_total_v6": 0.8023205175995827, "step": 985 }, { "epoch": 0.2274018018018018, "grad_norm": 21.582439422607422, "learning_rate": 9.751534832303332e-06, "loss": 12.570361137390137, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2693206835538149, "loss_total_v6": 0.785647563636303, "step": 986 }, { "epoch": 0.22763243243243242, "grad_norm": 24.49098014831543, "learning_rate": 9.750907509032028e-06, "loss": 12.993875503540039, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.29974235594272614, "loss_total_v6": 0.8121172115206718, "step": 987 }, { "epoch": 0.22786306306306306, "grad_norm": 24.536741256713867, "learning_rate": 9.750279415057618e-06, "loss": 12.762067794799805, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2911900207400322, "loss_total_v6": 0.797629252076149, "step": 988 }, { "epoch": 0.2280936936936937, "grad_norm": 22.089916229248047, "learning_rate": 9.749650550481994e-06, "loss": 12.661246299743652, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.27628273889422417, "loss_total_v6": 0.7913278937339783, "step": 989 }, { "epoch": 0.2283243243243243, "grad_norm": 23.188623428344727, "learning_rate": 9.749020915407165e-06, "loss": 12.81916618347168, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2923325337469578, "loss_total_v6": 0.8011979013681412, "step": 990 }, { "epoch": 0.22855495495495495, "grad_norm": 22.99237060546875, "learning_rate": 9.748390509935282e-06, "loss": 12.919770240783691, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.28882412426173687, "loss_total_v6": 0.8074856325984001, "step": 991 }, { "epoch": 0.2287855855855856, "grad_norm": 25.423635482788086, "learning_rate": 9.747759334168602e-06, "loss": 12.46037483215332, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.27830044366419315, "loss_total_v6": 0.7787734717130661, "step": 992 }, { "epoch": 0.2290162162162162, "grad_norm": 28.306171417236328, "learning_rate": 9.74712738820952e-06, "loss": 12.621772766113281, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.2882656753063202, "loss_total_v6": 0.7888607680797577, "step": 993 }, { "epoch": 0.22924684684684685, "grad_norm": 22.336688995361328, "learning_rate": 9.746494672160552e-06, "loss": 12.202995300292969, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.25095321238040924, "loss_total_v6": 0.7626871839165688, "step": 994 }, { "epoch": 0.2294774774774775, "grad_norm": 25.456138610839844, "learning_rate": 9.745861186124336e-06, "loss": 13.15328311920166, "loss_alignment": 1.03076171875, "loss_alignment_w": 0.515380859375, "loss_sft": 0.30701977014541626, "loss_total_v6": 0.8220801949501038, "step": 995 }, { "epoch": 0.2297081081081081, "grad_norm": 22.771772384643555, "learning_rate": 9.74522693020364e-06, "loss": 12.638765335083008, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.2715512942522764, "loss_total_v6": 0.7899228483438492, "step": 996 }, { "epoch": 0.22993873873873874, "grad_norm": 24.780941009521484, "learning_rate": 9.744591904501353e-06, "loss": 13.191713333129883, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.3009072579443455, "loss_total_v6": 0.8244820982217789, "step": 997 }, { "epoch": 0.23016936936936938, "grad_norm": 24.808002471923828, "learning_rate": 9.743956109120491e-06, "loss": 13.563459396362305, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.3317249771207571, "loss_total_v6": 0.8477162048220634, "step": 998 }, { "epoch": 0.2304, "grad_norm": 24.584020614624023, "learning_rate": 9.743319544164195e-06, "loss": 13.0745267868042, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.3021432552486658, "loss_total_v6": 0.8171579390764236, "step": 999 }, { "epoch": 0.23063063063063063, "grad_norm": 23.349464416503906, "learning_rate": 9.742682209735727e-06, "loss": 12.45436954498291, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.2784743905067444, "loss_total_v6": 0.7783980891108513, "step": 1000 }, { "epoch": 0.23063063063063063, "eval_loss": 0.9359597563743591, "eval_loss_alignment": 1.034288937642694, "eval_loss_alignment_w": 0.517144468821347, "eval_loss_contrast": 0.0, "eval_loss_highlevel": 0.0, "eval_loss_sft": 0.4188153476594654, "eval_loss_total_v6": 0.9359598157748784, "eval_loss_wm_recon": 0.0, "eval_loss_wm_recon_w": 0.0, "eval_runtime": 253.902, "eval_samples_per_second": 6.873, "eval_steps_per_second": 0.863, "step": 1000 }, { "epoch": 0.23086126126126127, "grad_norm": 22.542177200317383, "learning_rate": 9.742044105938481e-06, "loss": 12.487288475036621, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.2770375423133373, "loss_total_v6": 0.7804555147886276, "step": 1001 }, { "epoch": 0.2310918918918919, "grad_norm": 24.071699142456055, "learning_rate": 9.74140523287597e-06, "loss": 12.898228645324707, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2920554671436548, "loss_total_v6": 0.8061392903327942, "step": 1002 }, { "epoch": 0.23132252252252253, "grad_norm": 25.254854202270508, "learning_rate": 9.740765590651833e-06, "loss": 12.749950408935547, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.28582455962896347, "loss_total_v6": 0.7968719378113747, "step": 1003 }, { "epoch": 0.23155315315315314, "grad_norm": 25.629030227661133, "learning_rate": 9.740125179369833e-06, "loss": 12.645416259765625, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.27747535705566406, "loss_total_v6": 0.7903385013341904, "step": 1004 }, { "epoch": 0.23178378378378378, "grad_norm": 20.90388298034668, "learning_rate": 9.739483999133861e-06, "loss": 11.447298049926758, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.22068992629647255, "loss_total_v6": 0.7154561579227448, "step": 1005 }, { "epoch": 0.23201441441441442, "grad_norm": 20.399381637573242, "learning_rate": 9.73884205004793e-06, "loss": 12.299768447875977, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.26551591232419014, "loss_total_v6": 0.7687355056405067, "step": 1006 }, { "epoch": 0.23224504504504503, "grad_norm": 23.41910743713379, "learning_rate": 9.738199332216178e-06, "loss": 12.58187484741211, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.26997923478484154, "loss_total_v6": 0.7863671481609344, "step": 1007 }, { "epoch": 0.23247567567567567, "grad_norm": 22.213266372680664, "learning_rate": 9.737555845742869e-06, "loss": 12.822225570678711, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2868932690471411, "loss_total_v6": 0.8013891279697418, "step": 1008 }, { "epoch": 0.23270630630630632, "grad_norm": 23.324710845947266, "learning_rate": 9.73691159073239e-06, "loss": 12.255319595336914, "loss_alignment": 1.01513671875, "loss_alignment_w": 0.507568359375, "loss_sft": 0.2588316574692726, "loss_total_v6": 0.7659575045108795, "step": 1009 }, { "epoch": 0.23293693693693693, "grad_norm": 23.14114761352539, "learning_rate": 9.736266567289255e-06, "loss": 12.563039779663086, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.2794069144874811, "loss_total_v6": 0.7851899713277817, "step": 1010 }, { "epoch": 0.23316756756756757, "grad_norm": 23.270511627197266, "learning_rate": 9.735620775518098e-06, "loss": 12.41983699798584, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.26523824594914913, "loss_total_v6": 0.7762398198246956, "step": 1011 }, { "epoch": 0.2333981981981982, "grad_norm": 23.35867691040039, "learning_rate": 9.734974215523684e-06, "loss": 12.210400581359863, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2616241630166769, "loss_total_v6": 0.7631500288844109, "step": 1012 }, { "epoch": 0.23362882882882882, "grad_norm": 23.516233444213867, "learning_rate": 9.734326887410902e-06, "loss": 12.972494125366211, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.2928212434053421, "loss_total_v6": 0.8107808455824852, "step": 1013 }, { "epoch": 0.23385945945945946, "grad_norm": 24.39723014831543, "learning_rate": 9.733678791284755e-06, "loss": 12.668814659118652, "loss_alignment": 1.01318359375, "loss_alignment_w": 0.506591796875, "loss_sft": 0.2831644657999277, "loss_total_v6": 0.7918009236454964, "step": 1014 }, { "epoch": 0.2340900900900901, "grad_norm": 24.1178035736084, "learning_rate": 9.733029927250385e-06, "loss": 12.500619888305664, "loss_alignment": 1.01904296875, "loss_alignment_w": 0.509521484375, "loss_sft": 0.2703634686768055, "loss_total_v6": 0.7812887504696846, "step": 1015 }, { "epoch": 0.23432072072072072, "grad_norm": 23.544580459594727, "learning_rate": 9.73238029541305e-06, "loss": 12.6576509475708, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.27347929030656815, "loss_total_v6": 0.791103184223175, "step": 1016 }, { "epoch": 0.23455135135135136, "grad_norm": 24.548126220703125, "learning_rate": 9.731729895878135e-06, "loss": 12.571098327636719, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.26860381104052067, "loss_total_v6": 0.7856936901807785, "step": 1017 }, { "epoch": 0.23478198198198197, "grad_norm": 26.847015380859375, "learning_rate": 9.73107872875115e-06, "loss": 12.695432662963867, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.2759627141058445, "loss_total_v6": 0.7934645563364029, "step": 1018 }, { "epoch": 0.2350126126126126, "grad_norm": 23.444604873657227, "learning_rate": 9.730426794137727e-06, "loss": 12.709598541259766, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.29409051313996315, "loss_total_v6": 0.7943498864769936, "step": 1019 }, { "epoch": 0.23524324324324325, "grad_norm": 24.543127059936523, "learning_rate": 9.729774092143627e-06, "loss": 12.714371681213379, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2858286164700985, "loss_total_v6": 0.7946482226252556, "step": 1020 }, { "epoch": 0.23547387387387386, "grad_norm": 21.581819534301758, "learning_rate": 9.72912062287473e-06, "loss": 12.56268310546875, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2731437627226114, "loss_total_v6": 0.7851677015423775, "step": 1021 }, { "epoch": 0.2357045045045045, "grad_norm": 23.518329620361328, "learning_rate": 9.728466386437046e-06, "loss": 13.059480667114258, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.3056279346346855, "loss_total_v6": 0.8162175416946411, "step": 1022 }, { "epoch": 0.23593513513513514, "grad_norm": 24.435340881347656, "learning_rate": 9.727811382936704e-06, "loss": 12.688121795654297, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.287987494841218, "loss_total_v6": 0.7930076420307159, "step": 1023 }, { "epoch": 0.23616576576576576, "grad_norm": 22.10631561279297, "learning_rate": 9.727155612479964e-06, "loss": 11.866379737854004, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2302351426333189, "loss_total_v6": 0.741648718714714, "step": 1024 }, { "epoch": 0.2363963963963964, "grad_norm": 21.964418411254883, "learning_rate": 9.726499075173201e-06, "loss": 12.51376724243164, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.26145004853606224, "loss_total_v6": 0.7821104377508163, "step": 1025 }, { "epoch": 0.23662702702702704, "grad_norm": 23.9450740814209, "learning_rate": 9.725841771122927e-06, "loss": 13.114079475402832, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.30705673061311245, "loss_total_v6": 0.8196299746632576, "step": 1026 }, { "epoch": 0.23685765765765765, "grad_norm": 24.355440139770508, "learning_rate": 9.725183700435764e-06, "loss": 12.413655281066895, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.2727864272892475, "loss_total_v6": 0.7758534550666809, "step": 1027 }, { "epoch": 0.2370882882882883, "grad_norm": 23.928258895874023, "learning_rate": 9.72452486321847e-06, "loss": 12.845014572143555, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2965420540422201, "loss_total_v6": 0.802813433110714, "step": 1028 }, { "epoch": 0.23731891891891893, "grad_norm": 21.721492767333984, "learning_rate": 9.723865259577923e-06, "loss": 12.01541519165039, "loss_alignment": 1.01806640625, "loss_alignment_w": 0.509033203125, "loss_sft": 0.24463103711605072, "loss_total_v6": 0.7509634420275688, "step": 1029 }, { "epoch": 0.23754954954954954, "grad_norm": 23.1011905670166, "learning_rate": 9.723204889621125e-06, "loss": 12.492395401000977, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.2767463158816099, "loss_total_v6": 0.7807746529579163, "step": 1030 }, { "epoch": 0.23778018018018018, "grad_norm": 20.72844696044922, "learning_rate": 9.722543753455202e-06, "loss": 12.652212142944336, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.27802214957773685, "loss_total_v6": 0.7907632514834404, "step": 1031 }, { "epoch": 0.23801081081081082, "grad_norm": 25.205036163330078, "learning_rate": 9.721881851187406e-06, "loss": 12.729498863220215, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2813572343438864, "loss_total_v6": 0.795593686401844, "step": 1032 }, { "epoch": 0.23824144144144144, "grad_norm": 22.01713752746582, "learning_rate": 9.721219182925113e-06, "loss": 12.499853134155273, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.27408439852297306, "loss_total_v6": 0.781240776181221, "step": 1033 }, { "epoch": 0.23847207207207208, "grad_norm": 26.043642044067383, "learning_rate": 9.72055574877582e-06, "loss": 12.796199798583984, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.287799596786499, "loss_total_v6": 0.799762487411499, "step": 1034 }, { "epoch": 0.2387027027027027, "grad_norm": 24.378353118896484, "learning_rate": 9.71989154884715e-06, "loss": 13.052568435668945, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.295643949881196, "loss_total_v6": 0.8157855197787285, "step": 1035 }, { "epoch": 0.23893333333333333, "grad_norm": 23.14119529724121, "learning_rate": 9.719226583246856e-06, "loss": 12.687431335449219, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.2696490231901407, "loss_total_v6": 0.7929644286632538, "step": 1036 }, { "epoch": 0.23916396396396397, "grad_norm": 23.517560958862305, "learning_rate": 9.718560852082806e-06, "loss": 12.542579650878906, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2759461235255003, "loss_total_v6": 0.7839112281799316, "step": 1037 }, { "epoch": 0.23939459459459458, "grad_norm": 25.25262451171875, "learning_rate": 9.717894355463e-06, "loss": 12.439157485961914, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.26661362312734127, "loss_total_v6": 0.7774473652243614, "step": 1038 }, { "epoch": 0.23962522522522522, "grad_norm": 22.902822494506836, "learning_rate": 9.717227093495554e-06, "loss": 13.22340202331543, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.31204301677644253, "loss_total_v6": 0.8264626041054726, "step": 1039 }, { "epoch": 0.23985585585585587, "grad_norm": 25.41493797302246, "learning_rate": 9.716559066288716e-06, "loss": 13.239097595214844, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.31256621517241, "loss_total_v6": 0.8274435624480247, "step": 1040 }, { "epoch": 0.24008648648648648, "grad_norm": 21.307510375976562, "learning_rate": 9.715890273950854e-06, "loss": 12.293665885925293, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2517067976295948, "loss_total_v6": 0.7683541029691696, "step": 1041 }, { "epoch": 0.24031711711711712, "grad_norm": 28.09596824645996, "learning_rate": 9.715220716590464e-06, "loss": 13.226947784423828, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.30695459991693497, "loss_total_v6": 0.8266841992735863, "step": 1042 }, { "epoch": 0.24054774774774776, "grad_norm": 21.84491729736328, "learning_rate": 9.714550394316156e-06, "loss": 12.787029266357422, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.29207873716950417, "loss_total_v6": 0.7991892993450165, "step": 1043 }, { "epoch": 0.24077837837837837, "grad_norm": 25.7558650970459, "learning_rate": 9.713879307236678e-06, "loss": 12.68018627166748, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2838293667882681, "loss_total_v6": 0.7925116047263145, "step": 1044 }, { "epoch": 0.241009009009009, "grad_norm": 23.518985748291016, "learning_rate": 9.713207455460893e-06, "loss": 12.947173118591309, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.28814117424190044, "loss_total_v6": 0.8091983124613762, "step": 1045 }, { "epoch": 0.24123963963963965, "grad_norm": 23.472190856933594, "learning_rate": 9.712534839097789e-06, "loss": 12.987951278686523, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.29018630273640156, "loss_total_v6": 0.8117469474673271, "step": 1046 }, { "epoch": 0.24147027027027027, "grad_norm": 25.540904998779297, "learning_rate": 9.711861458256482e-06, "loss": 12.854647636413574, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2922307699918747, "loss_total_v6": 0.803415484726429, "step": 1047 }, { "epoch": 0.2417009009009009, "grad_norm": 22.169803619384766, "learning_rate": 9.711187313046207e-06, "loss": 12.287273406982422, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.2498728707432747, "loss_total_v6": 0.7679545432329178, "step": 1048 }, { "epoch": 0.24193153153153152, "grad_norm": 23.66090965270996, "learning_rate": 9.710512403576326e-06, "loss": 12.519657135009766, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.27387266233563423, "loss_total_v6": 0.7824786305427551, "step": 1049 }, { "epoch": 0.24216216216216216, "grad_norm": 32.39354705810547, "learning_rate": 9.709836729956326e-06, "loss": 12.405813217163086, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.269641250371933, "loss_total_v6": 0.7753632962703705, "step": 1050 }, { "epoch": 0.2423927927927928, "grad_norm": 21.16622543334961, "learning_rate": 9.709160292295814e-06, "loss": 12.470780372619629, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.26666740514338017, "loss_total_v6": 0.7794237583875656, "step": 1051 }, { "epoch": 0.2426234234234234, "grad_norm": 22.974027633666992, "learning_rate": 9.708483090704524e-06, "loss": 12.607147216796875, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2723827753216028, "loss_total_v6": 0.7879467085003853, "step": 1052 }, { "epoch": 0.24285405405405405, "grad_norm": 23.86160659790039, "learning_rate": 9.707805125292313e-06, "loss": 12.770502090454102, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.28773465380072594, "loss_total_v6": 0.7981563806533813, "step": 1053 }, { "epoch": 0.2430846846846847, "grad_norm": 24.19799041748047, "learning_rate": 9.707126396169165e-06, "loss": 12.810291290283203, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2881156709045172, "loss_total_v6": 0.8006431460380554, "step": 1054 }, { "epoch": 0.2433153153153153, "grad_norm": 25.496429443359375, "learning_rate": 9.706446903445179e-06, "loss": 12.317127227783203, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.26829461194574833, "loss_total_v6": 0.769820474088192, "step": 1055 }, { "epoch": 0.24354594594594595, "grad_norm": 23.249876022338867, "learning_rate": 9.70576664723059e-06, "loss": 12.150760650634766, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.26117726042866707, "loss_total_v6": 0.7594225034117699, "step": 1056 }, { "epoch": 0.2437765765765766, "grad_norm": 24.769763946533203, "learning_rate": 9.705085627635746e-06, "loss": 12.843718528747559, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.2860087789595127, "loss_total_v6": 0.8027324080467224, "step": 1057 }, { "epoch": 0.2440072072072072, "grad_norm": 22.553253173828125, "learning_rate": 9.704403844771128e-06, "loss": 12.418383598327637, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.273173488676548, "loss_total_v6": 0.7761489599943161, "step": 1058 }, { "epoch": 0.24423783783783784, "grad_norm": 22.987247467041016, "learning_rate": 9.703721298747332e-06, "loss": 12.965853691101074, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.29625146463513374, "loss_total_v6": 0.8103658556938171, "step": 1059 }, { "epoch": 0.24446846846846848, "grad_norm": 27.094568252563477, "learning_rate": 9.703037989675088e-06, "loss": 12.887202262878418, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.29711884818971157, "loss_total_v6": 0.8054501265287399, "step": 1060 }, { "epoch": 0.2446990990990991, "grad_norm": 21.26520538330078, "learning_rate": 9.702353917665238e-06, "loss": 12.414787292480469, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2630306128412485, "loss_total_v6": 0.7759242653846741, "step": 1061 }, { "epoch": 0.24492972972972973, "grad_norm": 23.411996841430664, "learning_rate": 9.701669082828754e-06, "loss": 12.69228458404541, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2814727444201708, "loss_total_v6": 0.7932677865028381, "step": 1062 }, { "epoch": 0.24516036036036035, "grad_norm": 22.597999572753906, "learning_rate": 9.700983485276735e-06, "loss": 12.773681640625, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.28906249813735485, "loss_total_v6": 0.7983551099896431, "step": 1063 }, { "epoch": 0.245390990990991, "grad_norm": 22.7351016998291, "learning_rate": 9.7002971251204e-06, "loss": 12.674641609191895, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.28163648396730423, "loss_total_v6": 0.7921650856733322, "step": 1064 }, { "epoch": 0.24562162162162163, "grad_norm": 25.207950592041016, "learning_rate": 9.699610002471088e-06, "loss": 12.327203750610352, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2591282017529011, "loss_total_v6": 0.7704502120614052, "step": 1065 }, { "epoch": 0.24585225225225224, "grad_norm": 24.08560562133789, "learning_rate": 9.69892211744027e-06, "loss": 12.858226776123047, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.29094386473298073, "loss_total_v6": 0.8036391660571098, "step": 1066 }, { "epoch": 0.24608288288288288, "grad_norm": 26.72699546813965, "learning_rate": 9.698233470139534e-06, "loss": 12.999317169189453, "loss_alignment": 1.01708984375, "loss_alignment_w": 0.508544921875, "loss_sft": 0.30401916801929474, "loss_total_v6": 0.8124572932720184, "step": 1067 }, { "epoch": 0.24631351351351352, "grad_norm": 26.671133041381836, "learning_rate": 9.697544060680594e-06, "loss": 12.467048645019531, "loss_alignment": 0.9892578125, "loss_alignment_w": 0.49462890625, "loss_sft": 0.2846226133406162, "loss_total_v6": 0.7791904881596565, "step": 1068 }, { "epoch": 0.24654414414414413, "grad_norm": 22.280960083007812, "learning_rate": 9.696853889175288e-06, "loss": 12.181533813476562, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.25776006281375885, "loss_total_v6": 0.7613458782434464, "step": 1069 }, { "epoch": 0.24677477477477477, "grad_norm": 22.314363479614258, "learning_rate": 9.696162955735577e-06, "loss": 12.106840133666992, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.246438916772604, "loss_total_v6": 0.75667754560709, "step": 1070 }, { "epoch": 0.24700540540540541, "grad_norm": 22.273122787475586, "learning_rate": 9.695471260473546e-06, "loss": 13.122560501098633, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.30334484949707985, "loss_total_v6": 0.8201600462198257, "step": 1071 }, { "epoch": 0.24723603603603603, "grad_norm": 25.70305824279785, "learning_rate": 9.694778803501404e-06, "loss": 11.923201560974121, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.24507803283631802, "loss_total_v6": 0.7452000975608826, "step": 1072 }, { "epoch": 0.24746666666666667, "grad_norm": 25.214195251464844, "learning_rate": 9.694085584931483e-06, "loss": 12.771313667297363, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2832534611225128, "loss_total_v6": 0.7982071116566658, "step": 1073 }, { "epoch": 0.2476972972972973, "grad_norm": 22.883501052856445, "learning_rate": 9.693391604876236e-06, "loss": 12.633014678955078, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.276364604011178, "loss_total_v6": 0.7895634695887566, "step": 1074 }, { "epoch": 0.24792792792792792, "grad_norm": 23.78745460510254, "learning_rate": 9.692696863448246e-06, "loss": 12.625741004943848, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.2668004557490349, "loss_total_v6": 0.7891088128089905, "step": 1075 }, { "epoch": 0.24815855855855856, "grad_norm": 21.20071029663086, "learning_rate": 9.692001360760212e-06, "loss": 12.681442260742188, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.27754492685198784, "loss_total_v6": 0.7925901040434837, "step": 1076 }, { "epoch": 0.2483891891891892, "grad_norm": 21.150041580200195, "learning_rate": 9.69130509692496e-06, "loss": 12.739497184753418, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.279952697455883, "loss_total_v6": 0.796218566596508, "step": 1077 }, { "epoch": 0.24861981981981982, "grad_norm": 24.902557373046875, "learning_rate": 9.690608072055443e-06, "loss": 12.703495979309082, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.28957393020391464, "loss_total_v6": 0.7939684987068176, "step": 1078 }, { "epoch": 0.24885045045045046, "grad_norm": 25.65896224975586, "learning_rate": 9.689910286264731e-06, "loss": 13.165438652038574, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.3015996888279915, "loss_total_v6": 0.8228399232029915, "step": 1079 }, { "epoch": 0.24908108108108107, "grad_norm": 19.464406967163086, "learning_rate": 9.689211739666023e-06, "loss": 12.111993789672852, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.24045910872519016, "loss_total_v6": 0.7569996267557144, "step": 1080 }, { "epoch": 0.2493117117117117, "grad_norm": 27.616714477539062, "learning_rate": 9.688512432372637e-06, "loss": 12.920912742614746, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.29724208638072014, "loss_total_v6": 0.8075570166110992, "step": 1081 }, { "epoch": 0.24954234234234235, "grad_norm": 23.828371047973633, "learning_rate": 9.687812364498015e-06, "loss": 12.410472869873047, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2719009555876255, "loss_total_v6": 0.7756546065211296, "step": 1082 }, { "epoch": 0.24977297297297296, "grad_norm": 22.523862838745117, "learning_rate": 9.687111536155726e-06, "loss": 12.627577781677246, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2696007899940014, "loss_total_v6": 0.7892235964536667, "step": 1083 }, { "epoch": 0.25000360360360363, "grad_norm": 28.991422653198242, "learning_rate": 9.68640994745946e-06, "loss": 12.164283752441406, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.25584274530410767, "loss_total_v6": 0.7602677643299103, "step": 1084 }, { "epoch": 0.25023423423423424, "grad_norm": 35.2864875793457, "learning_rate": 9.685707598523027e-06, "loss": 12.7783203125, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2861022725701332, "loss_total_v6": 0.798644982278347, "step": 1085 }, { "epoch": 0.25046486486486486, "grad_norm": 25.258121490478516, "learning_rate": 9.685004489460366e-06, "loss": 12.573537826538086, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2712129224091768, "loss_total_v6": 0.7858461067080498, "step": 1086 }, { "epoch": 0.25069549549549547, "grad_norm": 22.333642959594727, "learning_rate": 9.684300620385538e-06, "loss": 11.866546630859375, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.23479275219142437, "loss_total_v6": 0.7416591793298721, "step": 1087 }, { "epoch": 0.25092612612612614, "grad_norm": 20.431243896484375, "learning_rate": 9.683595991412725e-06, "loss": 12.316461563110352, "loss_alignment": 1.02880859375, "loss_alignment_w": 0.514404296875, "loss_sft": 0.2566562704741955, "loss_total_v6": 0.769778847694397, "step": 1088 }, { "epoch": 0.25115675675675675, "grad_norm": 23.474624633789062, "learning_rate": 9.682890602656233e-06, "loss": 12.452032089233398, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.26509894244372845, "loss_total_v6": 0.7782519906759262, "step": 1089 }, { "epoch": 0.25138738738738736, "grad_norm": 23.264171600341797, "learning_rate": 9.682184454230495e-06, "loss": 12.823043823242188, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2841062508523464, "loss_total_v6": 0.8014402762055397, "step": 1090 }, { "epoch": 0.25161801801801803, "grad_norm": 25.824480056762695, "learning_rate": 9.68147754625006e-06, "loss": 12.404045104980469, "loss_alignment": 0.990234375, "loss_alignment_w": 0.4951171875, "loss_sft": 0.27934211678802967, "loss_total_v6": 0.7752527967095375, "step": 1091 }, { "epoch": 0.25184864864864864, "grad_norm": 24.009593963623047, "learning_rate": 9.680769878829606e-06, "loss": 12.124979019165039, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.25321828573942184, "loss_total_v6": 0.7578111737966537, "step": 1092 }, { "epoch": 0.25207927927927926, "grad_norm": 26.123138427734375, "learning_rate": 9.680061452083933e-06, "loss": 12.38564395904541, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.26458125188946724, "loss_total_v6": 0.7741027474403381, "step": 1093 }, { "epoch": 0.2523099099099099, "grad_norm": 22.45441246032715, "learning_rate": 9.67935226612796e-06, "loss": 11.859773635864258, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.23200426623225212, "loss_total_v6": 0.7412358224391937, "step": 1094 }, { "epoch": 0.25254054054054054, "grad_norm": 21.16131591796875, "learning_rate": 9.678642321076738e-06, "loss": 12.812429428100586, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2823289856314659, "loss_total_v6": 0.8007768839597702, "step": 1095 }, { "epoch": 0.25277117117117115, "grad_norm": 24.10404396057129, "learning_rate": 9.677931617045432e-06, "loss": 12.762038230895996, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2812089715152979, "loss_total_v6": 0.7976274266839027, "step": 1096 }, { "epoch": 0.2530018018018018, "grad_norm": 23.591150283813477, "learning_rate": 9.677220154149338e-06, "loss": 12.554746627807617, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2712134141474962, "loss_total_v6": 0.7846716716885567, "step": 1097 }, { "epoch": 0.25323243243243243, "grad_norm": 21.51446533203125, "learning_rate": 9.676507932503868e-06, "loss": 12.353499412536621, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.26714989356696606, "loss_total_v6": 0.7720937281847, "step": 1098 }, { "epoch": 0.25346306306306304, "grad_norm": 23.32235336303711, "learning_rate": 9.675794952224558e-06, "loss": 12.827682495117188, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2872190196067095, "loss_total_v6": 0.801730141043663, "step": 1099 }, { "epoch": 0.2536936936936937, "grad_norm": 22.809057235717773, "learning_rate": 9.675081213427076e-06, "loss": 12.545978546142578, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.275517700240016, "loss_total_v6": 0.7841236367821693, "step": 1100 }, { "epoch": 0.2539243243243243, "grad_norm": 23.835668563842773, "learning_rate": 9.6743667162272e-06, "loss": 13.050821304321289, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.3107477892190218, "loss_total_v6": 0.815676361322403, "step": 1101 }, { "epoch": 0.25415495495495494, "grad_norm": 24.39108657836914, "learning_rate": 9.67365146074084e-06, "loss": 13.004148483276367, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.30558761954307556, "loss_total_v6": 0.8127592578530312, "step": 1102 }, { "epoch": 0.2543855855855856, "grad_norm": 23.65277099609375, "learning_rate": 9.672935447084028e-06, "loss": 12.627264022827148, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2780040577054024, "loss_total_v6": 0.7892039939761162, "step": 1103 }, { "epoch": 0.2546162162162162, "grad_norm": 24.11916732788086, "learning_rate": 9.672218675372914e-06, "loss": 12.334083557128906, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.25940564274787903, "loss_total_v6": 0.7708802297711372, "step": 1104 }, { "epoch": 0.25484684684684683, "grad_norm": 21.62306785583496, "learning_rate": 9.671501145723777e-06, "loss": 12.436820030212402, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.26150839030742645, "loss_total_v6": 0.7773012518882751, "step": 1105 }, { "epoch": 0.2550774774774775, "grad_norm": 22.692779541015625, "learning_rate": 9.670782858253015e-06, "loss": 12.229000091552734, "loss_alignment": 1.02880859375, "loss_alignment_w": 0.514404296875, "loss_sft": 0.2513730227947235, "loss_total_v6": 0.7643124610185623, "step": 1106 }, { "epoch": 0.2553081081081081, "grad_norm": 22.90222930908203, "learning_rate": 9.670063813077151e-06, "loss": 12.699310302734375, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2823696658015251, "loss_total_v6": 0.7937069237232208, "step": 1107 }, { "epoch": 0.2555387387387387, "grad_norm": 21.012948989868164, "learning_rate": 9.669344010312829e-06, "loss": 12.127869606018066, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.25208667293190956, "loss_total_v6": 0.7579918429255486, "step": 1108 }, { "epoch": 0.2557693693693694, "grad_norm": 20.361215591430664, "learning_rate": 9.668623450076818e-06, "loss": 12.987752914428711, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.3016637358814478, "loss_total_v6": 0.8117345571517944, "step": 1109 }, { "epoch": 0.256, "grad_norm": 23.305294036865234, "learning_rate": 9.667902132486009e-06, "loss": 12.33319091796875, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.25503157638013363, "loss_total_v6": 0.7708244025707245, "step": 1110 }, { "epoch": 0.2562306306306306, "grad_norm": 23.323036193847656, "learning_rate": 9.667180057657416e-06, "loss": 12.734735488891602, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.27989922650158405, "loss_total_v6": 0.7959209457039833, "step": 1111 }, { "epoch": 0.2564612612612613, "grad_norm": 24.469621658325195, "learning_rate": 9.666457225708175e-06, "loss": 12.42912483215332, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.27218159288167953, "loss_total_v6": 0.7768202647566795, "step": 1112 }, { "epoch": 0.2566918918918919, "grad_norm": 20.970325469970703, "learning_rate": 9.665733636755546e-06, "loss": 12.118587493896484, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.24464010633528233, "loss_total_v6": 0.7574117332696915, "step": 1113 }, { "epoch": 0.2569225225225225, "grad_norm": 23.696699142456055, "learning_rate": 9.66500929091691e-06, "loss": 12.374366760253906, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.25723880156874657, "loss_total_v6": 0.7733978927135468, "step": 1114 }, { "epoch": 0.2571531531531532, "grad_norm": 22.787275314331055, "learning_rate": 9.66428418830977e-06, "loss": 12.741899490356445, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.27720371074974537, "loss_total_v6": 0.7963687255978584, "step": 1115 }, { "epoch": 0.2573837837837838, "grad_norm": 21.322113037109375, "learning_rate": 9.663558329051761e-06, "loss": 12.716476440429688, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.27797991037368774, "loss_total_v6": 0.7947798147797585, "step": 1116 }, { "epoch": 0.2576144144144144, "grad_norm": 24.8878116607666, "learning_rate": 9.662831713260629e-06, "loss": 12.320792198181152, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.26155034825205803, "loss_total_v6": 0.7700494974851608, "step": 1117 }, { "epoch": 0.257845045045045, "grad_norm": 23.598709106445312, "learning_rate": 9.662104341054246e-06, "loss": 12.743583679199219, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2882189489901066, "loss_total_v6": 0.7964739799499512, "step": 1118 }, { "epoch": 0.2580756756756757, "grad_norm": 25.581884384155273, "learning_rate": 9.66137621255061e-06, "loss": 13.14660358428955, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.29784375987946987, "loss_total_v6": 0.8216627314686775, "step": 1119 }, { "epoch": 0.2583063063063063, "grad_norm": 23.371074676513672, "learning_rate": 9.66064732786784e-06, "loss": 13.039287567138672, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.30203133076429367, "loss_total_v6": 0.8149555176496506, "step": 1120 }, { "epoch": 0.2585369369369369, "grad_norm": 19.74445152282715, "learning_rate": 9.659917687124176e-06, "loss": 12.323590278625488, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.25945171900093555, "loss_total_v6": 0.7702243998646736, "step": 1121 }, { "epoch": 0.2587675675675676, "grad_norm": 23.084110260009766, "learning_rate": 9.659187290437982e-06, "loss": 12.771614074707031, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.28105972707271576, "loss_total_v6": 0.7982258796691895, "step": 1122 }, { "epoch": 0.2589981981981982, "grad_norm": 24.293790817260742, "learning_rate": 9.658456137927745e-06, "loss": 12.149514198303223, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.25200515054166317, "loss_total_v6": 0.7593446299433708, "step": 1123 }, { "epoch": 0.2592288288288288, "grad_norm": 23.402883529663086, "learning_rate": 9.657724229712076e-06, "loss": 12.242809295654297, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2522819023579359, "loss_total_v6": 0.7651755586266518, "step": 1124 }, { "epoch": 0.2594594594594595, "grad_norm": 22.882713317871094, "learning_rate": 9.656991565909703e-06, "loss": 12.848730087280273, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2919982671737671, "loss_total_v6": 0.8030456677079201, "step": 1125 }, { "epoch": 0.2596900900900901, "grad_norm": 22.17045021057129, "learning_rate": 9.656258146639485e-06, "loss": 12.517449378967285, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2740245349705219, "loss_total_v6": 0.7823405638337135, "step": 1126 }, { "epoch": 0.2599207207207207, "grad_norm": 25.139514923095703, "learning_rate": 9.655523972020396e-06, "loss": 12.620325088500977, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2795998156070709, "loss_total_v6": 0.7887703254818916, "step": 1127 }, { "epoch": 0.26015135135135137, "grad_norm": 23.16448211669922, "learning_rate": 9.654789042171535e-06, "loss": 12.641230583190918, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.27582517452538013, "loss_total_v6": 0.7900769039988518, "step": 1128 }, { "epoch": 0.260381981981982, "grad_norm": 22.414552688598633, "learning_rate": 9.654053357212125e-06, "loss": 12.032176971435547, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.24776912294328213, "loss_total_v6": 0.7520110756158829, "step": 1129 }, { "epoch": 0.2606126126126126, "grad_norm": 21.978240966796875, "learning_rate": 9.653316917261511e-06, "loss": 12.51583194732666, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.2763343323022127, "loss_total_v6": 0.7822394743561745, "step": 1130 }, { "epoch": 0.26084324324324326, "grad_norm": 21.72887420654297, "learning_rate": 9.65257972243916e-06, "loss": 12.830194473266602, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.28916131891310215, "loss_total_v6": 0.801887147128582, "step": 1131 }, { "epoch": 0.2610738738738739, "grad_norm": 24.298343658447266, "learning_rate": 9.65184177286466e-06, "loss": 12.532620429992676, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2700746897608042, "loss_total_v6": 0.7832887694239616, "step": 1132 }, { "epoch": 0.2613045045045045, "grad_norm": 24.201873779296875, "learning_rate": 9.651103068657726e-06, "loss": 12.791471481323242, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.27912698313593864, "loss_total_v6": 0.799466960132122, "step": 1133 }, { "epoch": 0.26153513513513515, "grad_norm": 23.312280654907227, "learning_rate": 9.650363609938189e-06, "loss": 13.15722370147705, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.31790145114064217, "loss_total_v6": 0.8223264887928963, "step": 1134 }, { "epoch": 0.26176576576576577, "grad_norm": 22.5322265625, "learning_rate": 9.649623396826006e-06, "loss": 12.853593826293945, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2846270576119423, "loss_total_v6": 0.8033496141433716, "step": 1135 }, { "epoch": 0.2619963963963964, "grad_norm": 23.717878341674805, "learning_rate": 9.648882429441258e-06, "loss": 12.883007049560547, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2872893922030926, "loss_total_v6": 0.8051879405975342, "step": 1136 }, { "epoch": 0.26222702702702705, "grad_norm": 27.021011352539062, "learning_rate": 9.648140707904146e-06, "loss": 12.544909477233887, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2752372957766056, "loss_total_v6": 0.7840568497776985, "step": 1137 }, { "epoch": 0.26245765765765766, "grad_norm": 22.00205421447754, "learning_rate": 9.64739823233499e-06, "loss": 12.134845733642578, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.23859138041734695, "loss_total_v6": 0.7584278210997581, "step": 1138 }, { "epoch": 0.2626882882882883, "grad_norm": 24.964441299438477, "learning_rate": 9.646655002854242e-06, "loss": 12.32586669921875, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.26314923353493214, "loss_total_v6": 0.7703666463494301, "step": 1139 }, { "epoch": 0.26291891891891894, "grad_norm": 23.232242584228516, "learning_rate": 9.645911019582467e-06, "loss": 13.36491584777832, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.31301411241292953, "loss_total_v6": 0.8353071957826614, "step": 1140 }, { "epoch": 0.26314954954954956, "grad_norm": 22.248319625854492, "learning_rate": 9.645166282640355e-06, "loss": 12.62210750579834, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.28235095739364624, "loss_total_v6": 0.7888817191123962, "step": 1141 }, { "epoch": 0.26338018018018017, "grad_norm": 24.897193908691406, "learning_rate": 9.644420792148722e-06, "loss": 13.044946670532227, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.29767001792788506, "loss_total_v6": 0.8153091669082642, "step": 1142 }, { "epoch": 0.26361081081081084, "grad_norm": 24.4261474609375, "learning_rate": 9.643674548228502e-06, "loss": 12.829292297363281, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.28225376829504967, "loss_total_v6": 0.8018307834863663, "step": 1143 }, { "epoch": 0.26384144144144145, "grad_norm": 23.95362663269043, "learning_rate": 9.642927551000749e-06, "loss": 12.633620262145996, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2729081716388464, "loss_total_v6": 0.7896013036370277, "step": 1144 }, { "epoch": 0.26407207207207206, "grad_norm": 23.994850158691406, "learning_rate": 9.642179800586648e-06, "loss": 12.798249244689941, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.29067428410053253, "loss_total_v6": 0.7998905926942825, "step": 1145 }, { "epoch": 0.26430270270270273, "grad_norm": 25.00999641418457, "learning_rate": 9.641431297107496e-06, "loss": 12.83055305480957, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2892142105847597, "loss_total_v6": 0.8019095361232758, "step": 1146 }, { "epoch": 0.26453333333333334, "grad_norm": 24.53915786743164, "learning_rate": 9.640682040684722e-06, "loss": 12.676427841186523, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.28725660778582096, "loss_total_v6": 0.7922767624258995, "step": 1147 }, { "epoch": 0.26476396396396396, "grad_norm": 25.707748413085938, "learning_rate": 9.639932031439866e-06, "loss": 13.321281433105469, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.31922866217792034, "loss_total_v6": 0.8325801193714142, "step": 1148 }, { "epoch": 0.26499459459459457, "grad_norm": 24.236526489257812, "learning_rate": 9.6391812694946e-06, "loss": 12.454511642456055, "loss_alignment": 1.02099609375, "loss_alignment_w": 0.510498046875, "loss_sft": 0.26868718303740025, "loss_total_v6": 0.778407022356987, "step": 1149 }, { "epoch": 0.26522522522522524, "grad_norm": 21.314550399780273, "learning_rate": 9.638429754970715e-06, "loss": 12.5938720703125, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2769393529742956, "loss_total_v6": 0.7871169745922089, "step": 1150 }, { "epoch": 0.26545585585585585, "grad_norm": 24.158525466918945, "learning_rate": 9.637677487990122e-06, "loss": 12.920478820800781, "loss_alignment": 1.0576171875, "loss_alignment_w": 0.52880859375, "loss_sft": 0.2794689629226923, "loss_total_v6": 0.8075298815965652, "step": 1151 }, { "epoch": 0.26568648648648646, "grad_norm": 21.361587524414062, "learning_rate": 9.636924468674856e-06, "loss": 12.754653930664062, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2833566777408123, "loss_total_v6": 0.7971658557653427, "step": 1152 }, { "epoch": 0.26591711711711713, "grad_norm": 25.378496170043945, "learning_rate": 9.636170697147071e-06, "loss": 12.325769424438477, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.25426261499524117, "loss_total_v6": 0.7703606337308884, "step": 1153 }, { "epoch": 0.26614774774774774, "grad_norm": 24.147592544555664, "learning_rate": 9.63541617352905e-06, "loss": 12.20349407196045, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2524797059595585, "loss_total_v6": 0.7627183720469475, "step": 1154 }, { "epoch": 0.26637837837837836, "grad_norm": 24.88386344909668, "learning_rate": 9.634660897943192e-06, "loss": 13.318374633789062, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.3154917247593403, "loss_total_v6": 0.8323984518647194, "step": 1155 }, { "epoch": 0.266609009009009, "grad_norm": 22.533039093017578, "learning_rate": 9.633904870512015e-06, "loss": 12.324976921081543, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2566544711589813, "loss_total_v6": 0.7703110799193382, "step": 1156 }, { "epoch": 0.26683963963963964, "grad_norm": 24.741456985473633, "learning_rate": 9.63314809135817e-06, "loss": 12.708182334899902, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.27785818465054035, "loss_total_v6": 0.7942613735795021, "step": 1157 }, { "epoch": 0.26707027027027025, "grad_norm": 25.210878372192383, "learning_rate": 9.632390560604421e-06, "loss": 13.13252067565918, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.30454716086387634, "loss_total_v6": 0.8207825273275375, "step": 1158 }, { "epoch": 0.2673009009009009, "grad_norm": 22.560070037841797, "learning_rate": 9.631632278373655e-06, "loss": 13.144708633422852, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.3006245531141758, "loss_total_v6": 0.8215443417429924, "step": 1159 }, { "epoch": 0.26753153153153153, "grad_norm": 24.677942276000977, "learning_rate": 9.630873244788884e-06, "loss": 12.604652404785156, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.27048730850219727, "loss_total_v6": 0.7877907678484917, "step": 1160 }, { "epoch": 0.26776216216216214, "grad_norm": 25.51542091369629, "learning_rate": 9.630113459973238e-06, "loss": 12.996732711791992, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.30253018252551556, "loss_total_v6": 0.8122958168387413, "step": 1161 }, { "epoch": 0.2679927927927928, "grad_norm": 25.02305030822754, "learning_rate": 9.629352924049975e-06, "loss": 13.577466011047363, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.3319442570209503, "loss_total_v6": 0.8485916033387184, "step": 1162 }, { "epoch": 0.2682234234234234, "grad_norm": 20.48662567138672, "learning_rate": 9.628591637142467e-06, "loss": 12.324910163879395, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.26423393562436104, "loss_total_v6": 0.770306907594204, "step": 1163 }, { "epoch": 0.26845405405405404, "grad_norm": 21.69611930847168, "learning_rate": 9.627829599374214e-06, "loss": 12.51229476928711, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.26827022060751915, "loss_total_v6": 0.7820184007287025, "step": 1164 }, { "epoch": 0.2686846846846847, "grad_norm": 24.243104934692383, "learning_rate": 9.627066810868833e-06, "loss": 13.452155113220215, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.32748454064130783, "loss_total_v6": 0.8407596871256828, "step": 1165 }, { "epoch": 0.2689153153153153, "grad_norm": 22.379047393798828, "learning_rate": 9.62630327175007e-06, "loss": 12.372241973876953, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2643844820559025, "loss_total_v6": 0.773265115916729, "step": 1166 }, { "epoch": 0.26914594594594593, "grad_norm": 22.32411003112793, "learning_rate": 9.625538982141782e-06, "loss": 12.436528205871582, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2673953250050545, "loss_total_v6": 0.7772830203175545, "step": 1167 }, { "epoch": 0.2693765765765766, "grad_norm": 22.533845901489258, "learning_rate": 9.624773942167958e-06, "loss": 12.337639808654785, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.27593953534960747, "loss_total_v6": 0.7711025103926659, "step": 1168 }, { "epoch": 0.2696072072072072, "grad_norm": 21.74884796142578, "learning_rate": 9.624008151952705e-06, "loss": 12.635259628295898, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.28279151394963264, "loss_total_v6": 0.7897037491202354, "step": 1169 }, { "epoch": 0.2698378378378378, "grad_norm": 26.72321891784668, "learning_rate": 9.623241611620248e-06, "loss": 12.708406448364258, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2864629216492176, "loss_total_v6": 0.7942754477262497, "step": 1170 }, { "epoch": 0.2700684684684685, "grad_norm": 23.942047119140625, "learning_rate": 9.622474321294939e-06, "loss": 12.619261741638184, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.26819605380296707, "loss_total_v6": 0.7887038588523865, "step": 1171 }, { "epoch": 0.2702990990990991, "grad_norm": 28.22985076904297, "learning_rate": 9.62170628110125e-06, "loss": 13.170461654663086, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.30467547103762627, "loss_total_v6": 0.8231538981199265, "step": 1172 }, { "epoch": 0.2705297297297297, "grad_norm": 24.07362174987793, "learning_rate": 9.620937491163772e-06, "loss": 12.829639434814453, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2956574410200119, "loss_total_v6": 0.8018525093793869, "step": 1173 }, { "epoch": 0.2707603603603604, "grad_norm": 22.896493911743164, "learning_rate": 9.620167951607221e-06, "loss": 12.111995697021484, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.23844502121210098, "loss_total_v6": 0.756999708712101, "step": 1174 }, { "epoch": 0.270990990990991, "grad_norm": 24.623870849609375, "learning_rate": 9.619397662556434e-06, "loss": 13.235554695129395, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.31251268461346626, "loss_total_v6": 0.8272221833467484, "step": 1175 }, { "epoch": 0.2712216216216216, "grad_norm": 25.47265625, "learning_rate": 9.61862662413637e-06, "loss": 12.841730117797852, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.29217116720974445, "loss_total_v6": 0.8026081845164299, "step": 1176 }, { "epoch": 0.2714522522522523, "grad_norm": 22.379608154296875, "learning_rate": 9.617854836472107e-06, "loss": 13.05029296875, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.2913360595703125, "loss_total_v6": 0.8156433254480362, "step": 1177 }, { "epoch": 0.2716828828828829, "grad_norm": 23.090167999267578, "learning_rate": 9.617082299688846e-06, "loss": 12.352386474609375, "loss_alignment": 1.02099609375, "loss_alignment_w": 0.510498046875, "loss_sft": 0.2630367577075958, "loss_total_v6": 0.7720241770148277, "step": 1178 }, { "epoch": 0.2719135135135135, "grad_norm": 22.482284545898438, "learning_rate": 9.61630901391191e-06, "loss": 12.437015533447266, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.27124048955738544, "loss_total_v6": 0.7773134857416153, "step": 1179 }, { "epoch": 0.2721441441441441, "grad_norm": 25.267473220825195, "learning_rate": 9.615534979266745e-06, "loss": 12.862543106079102, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2838131245225668, "loss_total_v6": 0.8039089366793633, "step": 1180 }, { "epoch": 0.2723747747747748, "grad_norm": 22.45997428894043, "learning_rate": 9.614760195878914e-06, "loss": 12.399130821228027, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2694677598774433, "loss_total_v6": 0.7749456837773323, "step": 1181 }, { "epoch": 0.2726054054054054, "grad_norm": 24.901548385620117, "learning_rate": 9.613984663874106e-06, "loss": 13.011419296264648, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.2930568177253008, "loss_total_v6": 0.8132136985659599, "step": 1182 }, { "epoch": 0.272836036036036, "grad_norm": 22.099538803100586, "learning_rate": 9.613208383378128e-06, "loss": 12.543184280395508, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2798444274812937, "loss_total_v6": 0.7839490324258804, "step": 1183 }, { "epoch": 0.2730666666666667, "grad_norm": 24.870922088623047, "learning_rate": 9.612431354516912e-06, "loss": 12.800883293151855, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.28777191787958145, "loss_total_v6": 0.800055243074894, "step": 1184 }, { "epoch": 0.2732972972972973, "grad_norm": 23.262130737304688, "learning_rate": 9.611653577416508e-06, "loss": 12.803354263305664, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.28461513482034206, "loss_total_v6": 0.8002096265554428, "step": 1185 }, { "epoch": 0.2735279279279279, "grad_norm": 22.134187698364258, "learning_rate": 9.610875052203087e-06, "loss": 12.622334480285645, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2778332456946373, "loss_total_v6": 0.7888959050178528, "step": 1186 }, { "epoch": 0.2737585585585586, "grad_norm": 22.364990234375, "learning_rate": 9.610095779002948e-06, "loss": 11.99643325805664, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2367765549570322, "loss_total_v6": 0.7497770264744759, "step": 1187 }, { "epoch": 0.2739891891891892, "grad_norm": 22.3172550201416, "learning_rate": 9.609315757942504e-06, "loss": 12.565763473510742, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.27231390401721, "loss_total_v6": 0.7853601798415184, "step": 1188 }, { "epoch": 0.2742198198198198, "grad_norm": 23.523176193237305, "learning_rate": 9.60853498914829e-06, "loss": 12.724100112915039, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.28367482125759125, "loss_total_v6": 0.7952562570571899, "step": 1189 }, { "epoch": 0.27445045045045047, "grad_norm": 22.35941505432129, "learning_rate": 9.607753472746967e-06, "loss": 12.297576904296875, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.26502794586122036, "loss_total_v6": 0.7685985267162323, "step": 1190 }, { "epoch": 0.2746810810810811, "grad_norm": 23.033855438232422, "learning_rate": 9.60697120886531e-06, "loss": 12.464855194091797, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2638098578900099, "loss_total_v6": 0.7790534347295761, "step": 1191 }, { "epoch": 0.2749117117117117, "grad_norm": 21.344724655151367, "learning_rate": 9.606188197630224e-06, "loss": 12.707330703735352, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.29666479863226414, "loss_total_v6": 0.7942081466317177, "step": 1192 }, { "epoch": 0.27514234234234236, "grad_norm": 22.888879776000977, "learning_rate": 9.60540443916873e-06, "loss": 12.594267845153809, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.28354061767458916, "loss_total_v6": 0.7871417179703712, "step": 1193 }, { "epoch": 0.275372972972973, "grad_norm": 24.192598342895508, "learning_rate": 9.604619933607971e-06, "loss": 12.795145034790039, "loss_alignment": 1.02294921875, "loss_alignment_w": 0.511474609375, "loss_sft": 0.2886187005788088, "loss_total_v6": 0.7996965944766998, "step": 1194 }, { "epoch": 0.2756036036036036, "grad_norm": 23.280414581298828, "learning_rate": 9.60383468107521e-06, "loss": 12.250431060791016, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.25538276694715023, "loss_total_v6": 0.7656519487500191, "step": 1195 }, { "epoch": 0.27583423423423425, "grad_norm": 21.438404083251953, "learning_rate": 9.603048681697834e-06, "loss": 11.942338943481445, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.24035366624593735, "loss_total_v6": 0.7463961541652679, "step": 1196 }, { "epoch": 0.27606486486486487, "grad_norm": 22.04668617248535, "learning_rate": 9.602261935603348e-06, "loss": 12.830268859863281, "loss_alignment": 1.01416015625, "loss_alignment_w": 0.507080078125, "loss_sft": 0.2942166347056627, "loss_total_v6": 0.8018917739391327, "step": 1197 }, { "epoch": 0.2762954954954955, "grad_norm": 22.741605758666992, "learning_rate": 9.601474442919383e-06, "loss": 12.362622261047363, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2617538720369339, "loss_total_v6": 0.7726639062166214, "step": 1198 }, { "epoch": 0.27652612612612615, "grad_norm": 24.359724044799805, "learning_rate": 9.600686203773684e-06, "loss": 13.454967498779297, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.3361442759633064, "loss_total_v6": 0.840935505926609, "step": 1199 }, { "epoch": 0.27675675675675676, "grad_norm": 22.654001235961914, "learning_rate": 9.599897218294122e-06, "loss": 12.661117553710938, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.28131006099283695, "loss_total_v6": 0.7913198247551918, "step": 1200 }, { "epoch": 0.2769873873873874, "grad_norm": 23.184101104736328, "learning_rate": 9.59910748660869e-06, "loss": 12.88853645324707, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.28691774792969227, "loss_total_v6": 0.8055334761738777, "step": 1201 }, { "epoch": 0.27721801801801804, "grad_norm": 23.34012794494629, "learning_rate": 9.598317008845496e-06, "loss": 12.668179512023926, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2780435159802437, "loss_total_v6": 0.7917611822485924, "step": 1202 }, { "epoch": 0.27744864864864865, "grad_norm": 23.147594451904297, "learning_rate": 9.597525785132778e-06, "loss": 13.131240844726562, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.3074579518288374, "loss_total_v6": 0.820702575147152, "step": 1203 }, { "epoch": 0.27767927927927927, "grad_norm": 22.0787410736084, "learning_rate": 9.596733815598888e-06, "loss": 12.817591667175293, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.28339933417737484, "loss_total_v6": 0.8010995090007782, "step": 1204 }, { "epoch": 0.27790990990990994, "grad_norm": 22.65066146850586, "learning_rate": 9.595941100372301e-06, "loss": 12.379581451416016, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.2686426490545273, "loss_total_v6": 0.7737238332629204, "step": 1205 }, { "epoch": 0.27814054054054055, "grad_norm": 26.85739517211914, "learning_rate": 9.595147639581613e-06, "loss": 12.681892395019531, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.27305650152266026, "loss_total_v6": 0.7926182821393013, "step": 1206 }, { "epoch": 0.27837117117117116, "grad_norm": 25.735721588134766, "learning_rate": 9.594353433355541e-06, "loss": 13.200056076049805, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.3115757629275322, "loss_total_v6": 0.8250034973025322, "step": 1207 }, { "epoch": 0.27860180180180183, "grad_norm": 22.83829116821289, "learning_rate": 9.593558481822923e-06, "loss": 12.457939147949219, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.26919124834239483, "loss_total_v6": 0.7786211967468262, "step": 1208 }, { "epoch": 0.27883243243243244, "grad_norm": 22.91425132751465, "learning_rate": 9.592762785112716e-06, "loss": 12.126248359680176, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2454850859940052, "loss_total_v6": 0.7578905001282692, "step": 1209 }, { "epoch": 0.27906306306306305, "grad_norm": 23.372377395629883, "learning_rate": 9.591966343354005e-06, "loss": 12.92099380493164, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.30069566890597343, "loss_total_v6": 0.8075621202588081, "step": 1210 }, { "epoch": 0.27929369369369367, "grad_norm": 21.720401763916016, "learning_rate": 9.591169156675985e-06, "loss": 12.897138595581055, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.28841670602560043, "loss_total_v6": 0.8060711324214935, "step": 1211 }, { "epoch": 0.27952432432432434, "grad_norm": 23.180110931396484, "learning_rate": 9.590371225207981e-06, "loss": 12.081019401550293, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.24973841197788715, "loss_total_v6": 0.7550637200474739, "step": 1212 }, { "epoch": 0.27975495495495495, "grad_norm": 21.39875602722168, "learning_rate": 9.589572549079434e-06, "loss": 12.789484024047852, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2950855493545532, "loss_total_v6": 0.7993427589535713, "step": 1213 }, { "epoch": 0.27998558558558556, "grad_norm": 25.666126251220703, "learning_rate": 9.588773128419907e-06, "loss": 13.418710708618164, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.31445374339818954, "loss_total_v6": 0.8386694118380547, "step": 1214 }, { "epoch": 0.28021621621621623, "grad_norm": 24.000186920166016, "learning_rate": 9.587972963359084e-06, "loss": 13.105175018310547, "loss_alignment": 1.02587890625, "loss_alignment_w": 0.512939453125, "loss_sft": 0.3062102869153023, "loss_total_v6": 0.8190734311938286, "step": 1215 }, { "epoch": 0.28044684684684684, "grad_norm": 21.377666473388672, "learning_rate": 9.587172054026769e-06, "loss": 12.596085548400879, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.27941233664751053, "loss_total_v6": 0.7872553244233131, "step": 1216 }, { "epoch": 0.28067747747747746, "grad_norm": 22.774688720703125, "learning_rate": 9.586370400552888e-06, "loss": 12.941471099853516, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.29843546636402607, "loss_total_v6": 0.8088419660925865, "step": 1217 }, { "epoch": 0.2809081081081081, "grad_norm": 24.883363723754883, "learning_rate": 9.585568003067487e-06, "loss": 13.198793411254883, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.3208199590444565, "loss_total_v6": 0.8249245434999466, "step": 1218 }, { "epoch": 0.28113873873873874, "grad_norm": 23.681964874267578, "learning_rate": 9.584764861700732e-06, "loss": 12.798980712890625, "loss_alignment": 1.04248046875, "loss_alignment_w": 0.521240234375, "loss_sft": 0.27878765389323235, "loss_total_v6": 0.7999363243579865, "step": 1219 }, { "epoch": 0.28136936936936935, "grad_norm": 23.692188262939453, "learning_rate": 9.583960976582914e-06, "loss": 12.245012283325195, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2548151556402445, "loss_total_v6": 0.7653132155537605, "step": 1220 }, { "epoch": 0.2816, "grad_norm": 21.290225982666016, "learning_rate": 9.583156347844435e-06, "loss": 12.68795394897461, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.28144621662795544, "loss_total_v6": 0.7929971218109131, "step": 1221 }, { "epoch": 0.28183063063063063, "grad_norm": 21.900318145751953, "learning_rate": 9.582350975615828e-06, "loss": 12.3561429977417, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.25596254877746105, "loss_total_v6": 0.7722589522600174, "step": 1222 }, { "epoch": 0.28206126126126124, "grad_norm": 23.6851806640625, "learning_rate": 9.581544860027742e-06, "loss": 12.453256607055664, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.27153838984668255, "loss_total_v6": 0.7783285602927208, "step": 1223 }, { "epoch": 0.2822918918918919, "grad_norm": 24.19277000427246, "learning_rate": 9.580738001210946e-06, "loss": 13.070271492004395, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.30515792965888977, "loss_total_v6": 0.8168919533491135, "step": 1224 }, { "epoch": 0.2825225225225225, "grad_norm": 23.16391944885254, "learning_rate": 9.57993039929633e-06, "loss": 12.5645751953125, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2695389110594988, "loss_total_v6": 0.7852860018610954, "step": 1225 }, { "epoch": 0.28275315315315314, "grad_norm": 24.615375518798828, "learning_rate": 9.579122054414907e-06, "loss": 12.520418167114258, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2759495619684458, "loss_total_v6": 0.7825260981917381, "step": 1226 }, { "epoch": 0.2829837837837838, "grad_norm": 23.068464279174805, "learning_rate": 9.578312966697807e-06, "loss": 12.468883514404297, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.28550028428435326, "loss_total_v6": 0.7793051972985268, "step": 1227 }, { "epoch": 0.2832144144144144, "grad_norm": 25.732940673828125, "learning_rate": 9.57750313627628e-06, "loss": 12.658269882202148, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2794536091387272, "loss_total_v6": 0.7911418452858925, "step": 1228 }, { "epoch": 0.28344504504504503, "grad_norm": 23.908016204833984, "learning_rate": 9.576692563281705e-06, "loss": 12.805652618408203, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.27790757082402706, "loss_total_v6": 0.8003532513976097, "step": 1229 }, { "epoch": 0.2836756756756757, "grad_norm": 22.930728912353516, "learning_rate": 9.575881247845568e-06, "loss": 12.957451820373535, "loss_alignment": 1.01904296875, "loss_alignment_w": 0.509521484375, "loss_sft": 0.2994953542947769, "loss_total_v6": 0.8098407685756683, "step": 1230 }, { "epoch": 0.2839063063063063, "grad_norm": 21.29362678527832, "learning_rate": 9.575069190099487e-06, "loss": 12.403898239135742, "loss_alignment": 1.04248046875, "loss_alignment_w": 0.521240234375, "loss_sft": 0.2532404977828264, "loss_total_v6": 0.7752436473965645, "step": 1231 }, { "epoch": 0.2841369369369369, "grad_norm": 22.47616958618164, "learning_rate": 9.574256390175192e-06, "loss": 12.817676544189453, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.289248738437891, "loss_total_v6": 0.8011048212647438, "step": 1232 }, { "epoch": 0.2843675675675676, "grad_norm": 22.824993133544922, "learning_rate": 9.573442848204542e-06, "loss": 13.239316940307617, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.3195074796676636, "loss_total_v6": 0.8274573013186455, "step": 1233 }, { "epoch": 0.2845981981981982, "grad_norm": 25.90969467163086, "learning_rate": 9.572628564319508e-06, "loss": 12.65672779083252, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2802422810345888, "loss_total_v6": 0.7910455018281937, "step": 1234 }, { "epoch": 0.2848288288288288, "grad_norm": 23.83655548095703, "learning_rate": 9.571813538652184e-06, "loss": 12.560647964477539, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.27019371651113033, "loss_total_v6": 0.785040520131588, "step": 1235 }, { "epoch": 0.2850594594594595, "grad_norm": 23.56353759765625, "learning_rate": 9.57099777133479e-06, "loss": 13.172380447387695, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.30932726711034775, "loss_total_v6": 0.8232738226652145, "step": 1236 }, { "epoch": 0.2852900900900901, "grad_norm": 26.163253784179688, "learning_rate": 9.570181262499659e-06, "loss": 12.34520149230957, "loss_alignment": 1.03076171875, "loss_alignment_w": 0.515380859375, "loss_sft": 0.25645360350608826, "loss_total_v6": 0.771575078368187, "step": 1237 }, { "epoch": 0.2855207207207207, "grad_norm": 24.891620635986328, "learning_rate": 9.569364012279247e-06, "loss": 12.456958770751953, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.26803140714764595, "loss_total_v6": 0.7785599678754807, "step": 1238 }, { "epoch": 0.2857513513513513, "grad_norm": 24.10719108581543, "learning_rate": 9.56854602080613e-06, "loss": 12.339469909667969, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.26493024080991745, "loss_total_v6": 0.771216869354248, "step": 1239 }, { "epoch": 0.285981981981982, "grad_norm": 25.10639190673828, "learning_rate": 9.567727288213005e-06, "loss": 12.352027893066406, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.2748398520052433, "loss_total_v6": 0.7720017209649086, "step": 1240 }, { "epoch": 0.2862126126126126, "grad_norm": 25.191686630249023, "learning_rate": 9.56690781463269e-06, "loss": 12.430065155029297, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2690818440169096, "loss_total_v6": 0.7768790945410728, "step": 1241 }, { "epoch": 0.2864432432432432, "grad_norm": 23.556259155273438, "learning_rate": 9.566087600198118e-06, "loss": 12.440703392028809, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.27751346677541733, "loss_total_v6": 0.7775439843535423, "step": 1242 }, { "epoch": 0.2866738738738739, "grad_norm": 25.1390380859375, "learning_rate": 9.565266645042349e-06, "loss": 12.359094619750977, "loss_alignment": 1.01806640625, "loss_alignment_w": 0.509033203125, "loss_sft": 0.26362387277185917, "loss_total_v6": 0.7724434211850166, "step": 1243 }, { "epoch": 0.2869045045045045, "grad_norm": 21.964941024780273, "learning_rate": 9.564444949298559e-06, "loss": 12.358564376831055, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2665203344076872, "loss_total_v6": 0.7724102512001991, "step": 1244 }, { "epoch": 0.2871351351351351, "grad_norm": 21.0086669921875, "learning_rate": 9.563622513100046e-06, "loss": 11.7991943359375, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.23892975971102715, "loss_total_v6": 0.7374496534466743, "step": 1245 }, { "epoch": 0.2873657657657658, "grad_norm": 22.555437088012695, "learning_rate": 9.562799336580227e-06, "loss": 12.385522842407227, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.265672305598855, "loss_total_v6": 0.7740951627492905, "step": 1246 }, { "epoch": 0.2875963963963964, "grad_norm": 25.09088706970215, "learning_rate": 9.56197541987264e-06, "loss": 12.688247680664062, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.28613369539380074, "loss_total_v6": 0.7930154427886009, "step": 1247 }, { "epoch": 0.287827027027027, "grad_norm": 29.549400329589844, "learning_rate": 9.561150763110945e-06, "loss": 12.052334785461426, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.24750311113893986, "loss_total_v6": 0.7532709315419197, "step": 1248 }, { "epoch": 0.2880576576576577, "grad_norm": 22.241615295410156, "learning_rate": 9.560325366428915e-06, "loss": 11.886222839355469, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.24186656437814236, "loss_total_v6": 0.7428888976573944, "step": 1249 }, { "epoch": 0.2882882882882883, "grad_norm": 23.435531616210938, "learning_rate": 9.55949922996045e-06, "loss": 12.904090881347656, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2922997511923313, "loss_total_v6": 0.8065056875348091, "step": 1250 }, { "epoch": 0.2885189189189189, "grad_norm": 21.38473129272461, "learning_rate": 9.55867235383957e-06, "loss": 12.732013702392578, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2900440376251936, "loss_total_v6": 0.7957508414983749, "step": 1251 }, { "epoch": 0.28874954954954957, "grad_norm": 22.8197078704834, "learning_rate": 9.557844738200408e-06, "loss": 12.247411727905273, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.25577392242848873, "loss_total_v6": 0.7654632553458214, "step": 1252 }, { "epoch": 0.2889801801801802, "grad_norm": 23.331317901611328, "learning_rate": 9.557016383177226e-06, "loss": 12.733067512512207, "loss_alignment": 1.01318359375, "loss_alignment_w": 0.506591796875, "loss_sft": 0.2891028542071581, "loss_total_v6": 0.7958167269825935, "step": 1253 }, { "epoch": 0.2892108108108108, "grad_norm": 24.45028305053711, "learning_rate": 9.556187288904401e-06, "loss": 12.33597183227539, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.26944185234606266, "loss_total_v6": 0.7709982246160507, "step": 1254 }, { "epoch": 0.28944144144144146, "grad_norm": 21.651247024536133, "learning_rate": 9.555357455516427e-06, "loss": 12.550134658813477, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2809197101742029, "loss_total_v6": 0.7843834385275841, "step": 1255 }, { "epoch": 0.2896720720720721, "grad_norm": 24.06441307067871, "learning_rate": 9.554526883147926e-06, "loss": 12.291681289672852, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2633777745068073, "loss_total_v6": 0.7682300582528114, "step": 1256 }, { "epoch": 0.2899027027027027, "grad_norm": 27.168472290039062, "learning_rate": 9.553695571933632e-06, "loss": 13.124191284179688, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.3142499513924122, "loss_total_v6": 0.8202619254589081, "step": 1257 }, { "epoch": 0.29013333333333335, "grad_norm": 23.752696990966797, "learning_rate": 9.552863522008404e-06, "loss": 12.396341323852539, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.2732302416115999, "loss_total_v6": 0.7747713550925255, "step": 1258 }, { "epoch": 0.29036396396396397, "grad_norm": 22.2841796875, "learning_rate": 9.55203073350722e-06, "loss": 11.794143676757812, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.23707295209169388, "loss_total_v6": 0.7371339797973633, "step": 1259 }, { "epoch": 0.2905945945945946, "grad_norm": 21.95760726928711, "learning_rate": 9.551197206565174e-06, "loss": 13.031936645507812, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.30549337714910507, "loss_total_v6": 0.8144960328936577, "step": 1260 }, { "epoch": 0.29082522522522525, "grad_norm": 23.37995719909668, "learning_rate": 9.550362941317484e-06, "loss": 12.878968238830566, "loss_alignment": 1.02783203125, "loss_alignment_w": 0.513916015625, "loss_sft": 0.2907752972096205, "loss_total_v6": 0.8049355149269104, "step": 1261 }, { "epoch": 0.29105585585585586, "grad_norm": 22.380756378173828, "learning_rate": 9.549527937899487e-06, "loss": 12.488292694091797, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.28138801269233227, "loss_total_v6": 0.7805182933807373, "step": 1262 }, { "epoch": 0.2912864864864865, "grad_norm": 22.824155807495117, "learning_rate": 9.548692196446638e-06, "loss": 12.58956527709961, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.28355191089212894, "loss_total_v6": 0.7868478074669838, "step": 1263 }, { "epoch": 0.29151711711711714, "grad_norm": 21.855884552001953, "learning_rate": 9.547855717094516e-06, "loss": 12.406124114990234, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2672650534659624, "loss_total_v6": 0.775382749736309, "step": 1264 }, { "epoch": 0.29174774774774775, "grad_norm": 23.647384643554688, "learning_rate": 9.547018499978811e-06, "loss": 12.060453414916992, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.2575472854077816, "loss_total_v6": 0.7537783533334732, "step": 1265 }, { "epoch": 0.29197837837837837, "grad_norm": 20.993080139160156, "learning_rate": 9.546180545235344e-06, "loss": 12.237836837768555, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2537869047373533, "loss_total_v6": 0.7648647800087929, "step": 1266 }, { "epoch": 0.29220900900900904, "grad_norm": 22.428829193115234, "learning_rate": 9.545341853000046e-06, "loss": 12.309120178222656, "loss_alignment": 1.01220703125, "loss_alignment_w": 0.506103515625, "loss_sft": 0.26187373138964176, "loss_total_v6": 0.7693200036883354, "step": 1267 }, { "epoch": 0.29243963963963965, "grad_norm": 23.183822631835938, "learning_rate": 9.544502423408974e-06, "loss": 12.600227355957031, "loss_alignment": 1.02099609375, "loss_alignment_w": 0.510498046875, "loss_sft": 0.27791641652584076, "loss_total_v6": 0.7875141948461533, "step": 1268 }, { "epoch": 0.29267027027027026, "grad_norm": 25.326011657714844, "learning_rate": 9.5436622565983e-06, "loss": 12.251045227050781, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.26895572803914547, "loss_total_v6": 0.7656903564929962, "step": 1269 }, { "epoch": 0.2929009009009009, "grad_norm": 21.888212203979492, "learning_rate": 9.54282135270432e-06, "loss": 12.318008422851562, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.25861457735300064, "loss_total_v6": 0.769875556230545, "step": 1270 }, { "epoch": 0.29313153153153154, "grad_norm": 24.334068298339844, "learning_rate": 9.541979711863445e-06, "loss": 12.856040954589844, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.29686499759554863, "loss_total_v6": 0.8035025596618652, "step": 1271 }, { "epoch": 0.29336216216216215, "grad_norm": 25.425024032592773, "learning_rate": 9.541137334212212e-06, "loss": 12.740301132202148, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.286289531737566, "loss_total_v6": 0.7962687760591507, "step": 1272 }, { "epoch": 0.29359279279279277, "grad_norm": 22.042648315429688, "learning_rate": 9.540294219887267e-06, "loss": 12.037381172180176, "loss_alignment": 0.9775390625, "loss_alignment_w": 0.48876953125, "loss_sft": 0.2630785070359707, "loss_total_v6": 0.752336323261261, "step": 1273 }, { "epoch": 0.29382342342342344, "grad_norm": 22.808549880981445, "learning_rate": 9.539450369025388e-06, "loss": 12.91565990447998, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2986685521900654, "loss_total_v6": 0.8072287365794182, "step": 1274 }, { "epoch": 0.29405405405405405, "grad_norm": 23.17459487915039, "learning_rate": 9.538605781763464e-06, "loss": 13.047906875610352, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.2986179757863283, "loss_total_v6": 0.8154942244291306, "step": 1275 }, { "epoch": 0.29428468468468466, "grad_norm": 23.14498519897461, "learning_rate": 9.537760458238505e-06, "loss": 12.186273574829102, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.25161710754036903, "loss_total_v6": 0.7616421356797218, "step": 1276 }, { "epoch": 0.29451531531531533, "grad_norm": 23.766216278076172, "learning_rate": 9.536914398587642e-06, "loss": 12.647626876831055, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.28394592367112637, "loss_total_v6": 0.7904767021536827, "step": 1277 }, { "epoch": 0.29474594594594594, "grad_norm": 22.31979751586914, "learning_rate": 9.536067602948126e-06, "loss": 12.027328491210938, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.24748139083385468, "loss_total_v6": 0.7517080754041672, "step": 1278 }, { "epoch": 0.29497657657657655, "grad_norm": 22.789575576782227, "learning_rate": 9.535220071457325e-06, "loss": 12.23672103881836, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.250451797619462, "loss_total_v6": 0.7647950425744057, "step": 1279 }, { "epoch": 0.2952072072072072, "grad_norm": 22.346113204956055, "learning_rate": 9.534371804252727e-06, "loss": 12.339591026306152, "loss_alignment": 1.02392578125, "loss_alignment_w": 0.511962890625, "loss_sft": 0.25900221802294254, "loss_total_v6": 0.7712244614958763, "step": 1280 }, { "epoch": 0.29543783783783784, "grad_norm": 21.676042556762695, "learning_rate": 9.53352280147194e-06, "loss": 12.76557731628418, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.28379523381590843, "loss_total_v6": 0.7978485673666, "step": 1281 }, { "epoch": 0.29566846846846845, "grad_norm": 23.358062744140625, "learning_rate": 9.532673063252696e-06, "loss": 12.531007766723633, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.28631599247455597, "loss_total_v6": 0.7831879556179047, "step": 1282 }, { "epoch": 0.2958990990990991, "grad_norm": 22.980972290039062, "learning_rate": 9.531822589732831e-06, "loss": 12.848560333251953, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.2865097336471081, "loss_total_v6": 0.8030349910259247, "step": 1283 }, { "epoch": 0.29612972972972973, "grad_norm": 23.69475746154785, "learning_rate": 9.53097138105032e-06, "loss": 12.436197280883789, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.264673812314868, "loss_total_v6": 0.7772623300552368, "step": 1284 }, { "epoch": 0.29636036036036034, "grad_norm": 24.027074813842773, "learning_rate": 9.530119437343245e-06, "loss": 13.046754837036133, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.2944566272199154, "loss_total_v6": 0.8154221922159195, "step": 1285 }, { "epoch": 0.296590990990991, "grad_norm": 24.827322006225586, "learning_rate": 9.52926675874981e-06, "loss": 12.237836837768555, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.25714388489723206, "loss_total_v6": 0.7648648172616959, "step": 1286 }, { "epoch": 0.2968216216216216, "grad_norm": 24.546472549438477, "learning_rate": 9.528413345408339e-06, "loss": 12.094697952270508, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.24613774381577969, "loss_total_v6": 0.7559186071157455, "step": 1287 }, { "epoch": 0.29705225225225224, "grad_norm": 23.911006927490234, "learning_rate": 9.527559197457273e-06, "loss": 11.988893508911133, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.2527848891913891, "loss_total_v6": 0.7493058741092682, "step": 1288 }, { "epoch": 0.2972828828828829, "grad_norm": 24.341981887817383, "learning_rate": 9.526704315035176e-06, "loss": 12.18065071105957, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.25939849577844143, "loss_total_v6": 0.761290617287159, "step": 1289 }, { "epoch": 0.2975135135135135, "grad_norm": 24.758493423461914, "learning_rate": 9.525848698280727e-06, "loss": 12.521520614624023, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.28024519234895706, "loss_total_v6": 0.7825950607657433, "step": 1290 }, { "epoch": 0.29774414414414413, "grad_norm": 23.64375114440918, "learning_rate": 9.524992347332729e-06, "loss": 12.332993507385254, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2635946925729513, "loss_total_v6": 0.7708121091127396, "step": 1291 }, { "epoch": 0.2979747747747748, "grad_norm": 24.37857437133789, "learning_rate": 9.524135262330098e-06, "loss": 12.358838081359863, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.26871948689222336, "loss_total_v6": 0.7724273800849915, "step": 1292 }, { "epoch": 0.2982054054054054, "grad_norm": 23.060598373413086, "learning_rate": 9.523277443411876e-06, "loss": 12.788309097290039, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.29763653315603733, "loss_total_v6": 0.7992692515254021, "step": 1293 }, { "epoch": 0.298436036036036, "grad_norm": 21.4982967376709, "learning_rate": 9.522418890717216e-06, "loss": 12.38910961151123, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.27550950460135937, "loss_total_v6": 0.7743193507194519, "step": 1294 }, { "epoch": 0.2986666666666667, "grad_norm": 25.19458770751953, "learning_rate": 9.521559604385396e-06, "loss": 12.818292617797852, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.29607734084129333, "loss_total_v6": 0.8011432737112045, "step": 1295 }, { "epoch": 0.2988972972972973, "grad_norm": 22.182680130004883, "learning_rate": 9.520699584555813e-06, "loss": 12.794957160949707, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2962058447301388, "loss_total_v6": 0.7996848374605179, "step": 1296 }, { "epoch": 0.2991279279279279, "grad_norm": 23.257099151611328, "learning_rate": 9.51983883136798e-06, "loss": 12.61149787902832, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.2843123562633991, "loss_total_v6": 0.7882186323404312, "step": 1297 }, { "epoch": 0.2993585585585586, "grad_norm": 23.104206085205078, "learning_rate": 9.51897734496153e-06, "loss": 13.156233787536621, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.31745809502899647, "loss_total_v6": 0.8222646266222, "step": 1298 }, { "epoch": 0.2995891891891892, "grad_norm": 23.11491584777832, "learning_rate": 9.518115125476219e-06, "loss": 12.438472747802734, "loss_alignment": 1.02978515625, "loss_alignment_w": 0.514892578125, "loss_sft": 0.2630460988730192, "loss_total_v6": 0.7774045616388321, "step": 1299 }, { "epoch": 0.2998198198198198, "grad_norm": 21.1754207611084, "learning_rate": 9.517252173051912e-06, "loss": 13.031646728515625, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2974643390625715, "loss_total_v6": 0.814477875828743, "step": 1300 }, { "epoch": 0.3000504504504504, "grad_norm": 19.87427520751953, "learning_rate": 9.516388487828604e-06, "loss": 12.042282104492188, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2503537479788065, "loss_total_v6": 0.7526425868272781, "step": 1301 }, { "epoch": 0.3002810810810811, "grad_norm": 24.165401458740234, "learning_rate": 9.5155240699464e-06, "loss": 12.629262924194336, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2854684330523014, "loss_total_v6": 0.7893289029598236, "step": 1302 }, { "epoch": 0.3005117117117117, "grad_norm": 21.021339416503906, "learning_rate": 9.514658919545533e-06, "loss": 12.561858177185059, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.28287308663129807, "loss_total_v6": 0.7851161435246468, "step": 1303 }, { "epoch": 0.3007423423423423, "grad_norm": 24.82170867919922, "learning_rate": 9.513793036766345e-06, "loss": 12.197486877441406, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.2661118805408478, "loss_total_v6": 0.7623429298400879, "step": 1304 }, { "epoch": 0.300972972972973, "grad_norm": 24.750001907348633, "learning_rate": 9.512926421749305e-06, "loss": 13.141704559326172, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.30965305492281914, "loss_total_v6": 0.8213565498590469, "step": 1305 }, { "epoch": 0.3012036036036036, "grad_norm": 22.916852951049805, "learning_rate": 9.512059074634995e-06, "loss": 12.853147506713867, "loss_alignment": 1.01904296875, "loss_alignment_w": 0.509521484375, "loss_sft": 0.29480726830661297, "loss_total_v6": 0.8033216670155525, "step": 1306 }, { "epoch": 0.3014342342342342, "grad_norm": 24.71062469482422, "learning_rate": 9.511190995564117e-06, "loss": 12.464384078979492, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.27288999035954475, "loss_total_v6": 0.7790239974856377, "step": 1307 }, { "epoch": 0.3016648648648649, "grad_norm": 21.96790885925293, "learning_rate": 9.510322184677494e-06, "loss": 12.423765182495117, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.253063078969717, "loss_total_v6": 0.7764853239059448, "step": 1308 }, { "epoch": 0.3018954954954955, "grad_norm": 25.57649803161621, "learning_rate": 9.509452642116068e-06, "loss": 13.107824325561523, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.3022713251411915, "loss_total_v6": 0.8192390576004982, "step": 1309 }, { "epoch": 0.3021261261261261, "grad_norm": 24.909597396850586, "learning_rate": 9.508582368020897e-06, "loss": 12.493523597717285, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.27841902896761894, "loss_total_v6": 0.7808451876044273, "step": 1310 }, { "epoch": 0.3023567567567568, "grad_norm": 24.084474563598633, "learning_rate": 9.507711362533161e-06, "loss": 12.565312385559082, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.2802356258034706, "loss_total_v6": 0.7853320613503456, "step": 1311 }, { "epoch": 0.3025873873873874, "grad_norm": 22.944950103759766, "learning_rate": 9.506839625794152e-06, "loss": 13.19389820098877, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.30966501124203205, "loss_total_v6": 0.8246186301112175, "step": 1312 }, { "epoch": 0.302818018018018, "grad_norm": 23.732885360717773, "learning_rate": 9.505967157945288e-06, "loss": 12.504181861877441, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.27820022962987423, "loss_total_v6": 0.7815113812685013, "step": 1313 }, { "epoch": 0.30304864864864867, "grad_norm": 24.25226402282715, "learning_rate": 9.505093959128102e-06, "loss": 12.727118492126465, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.28218501806259155, "loss_total_v6": 0.7954449132084846, "step": 1314 }, { "epoch": 0.3032792792792793, "grad_norm": 22.166501998901367, "learning_rate": 9.504220029484247e-06, "loss": 12.058843612670898, "loss_alignment": 1.01513671875, "loss_alignment_w": 0.507568359375, "loss_sft": 0.24537691660225391, "loss_total_v6": 0.7536776810884476, "step": 1315 }, { "epoch": 0.3035099099099099, "grad_norm": 21.736106872558594, "learning_rate": 9.503345369155494e-06, "loss": 12.11670970916748, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.26580872759222984, "loss_total_v6": 0.7572943419218063, "step": 1316 }, { "epoch": 0.30374054054054056, "grad_norm": 25.27269744873047, "learning_rate": 9.502469978283733e-06, "loss": 12.759441375732422, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.28788258880376816, "loss_total_v6": 0.7974651232361794, "step": 1317 }, { "epoch": 0.3039711711711712, "grad_norm": 24.069721221923828, "learning_rate": 9.501593857010968e-06, "loss": 13.078481674194336, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.3029550313949585, "loss_total_v6": 0.817405104637146, "step": 1318 }, { "epoch": 0.3042018018018018, "grad_norm": 21.31113624572754, "learning_rate": 9.50071700547933e-06, "loss": 12.161627769470215, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.2581486199051142, "loss_total_v6": 0.7601017355918884, "step": 1319 }, { "epoch": 0.30443243243243245, "grad_norm": 23.569847106933594, "learning_rate": 9.499839423831062e-06, "loss": 12.419017791748047, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.27296900376677513, "loss_total_v6": 0.7761886417865753, "step": 1320 }, { "epoch": 0.30466306306306307, "grad_norm": 25.12035369873047, "learning_rate": 9.498961112208526e-06, "loss": 12.244148254394531, "loss_alignment": 1.02392578125, "loss_alignment_w": 0.511962890625, "loss_sft": 0.2530675046145916, "loss_total_v6": 0.7652592658996582, "step": 1321 }, { "epoch": 0.3048936936936937, "grad_norm": 21.170757293701172, "learning_rate": 9.498082070754206e-06, "loss": 12.606186866760254, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2785025108605623, "loss_total_v6": 0.7878866568207741, "step": 1322 }, { "epoch": 0.30512432432432435, "grad_norm": 25.025306701660156, "learning_rate": 9.497202299610702e-06, "loss": 12.72846794128418, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.2780731562525034, "loss_total_v6": 0.7955292165279388, "step": 1323 }, { "epoch": 0.30535495495495496, "grad_norm": 21.334945678710938, "learning_rate": 9.49632179892073e-06, "loss": 12.954418182373047, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.3070724345743656, "loss_total_v6": 0.8096511736512184, "step": 1324 }, { "epoch": 0.3055855855855856, "grad_norm": 21.002174377441406, "learning_rate": 9.49544056882713e-06, "loss": 12.75657844543457, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2844382431358099, "loss_total_v6": 0.7972861379384995, "step": 1325 }, { "epoch": 0.30581621621621624, "grad_norm": 26.45697784423828, "learning_rate": 9.494558609472856e-06, "loss": 13.28560733795166, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.31187207251787186, "loss_total_v6": 0.8303504660725594, "step": 1326 }, { "epoch": 0.30604684684684685, "grad_norm": 22.74205207824707, "learning_rate": 9.49367592100098e-06, "loss": 12.4957857131958, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.27064115554094315, "loss_total_v6": 0.7809866070747375, "step": 1327 }, { "epoch": 0.30627747747747747, "grad_norm": 24.03850746154785, "learning_rate": 9.492792503554695e-06, "loss": 13.026318550109863, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.29801633208990097, "loss_total_v6": 0.8141448870301247, "step": 1328 }, { "epoch": 0.30650810810810813, "grad_norm": 22.763153076171875, "learning_rate": 9.491908357277313e-06, "loss": 12.611056327819824, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.27888317964971066, "loss_total_v6": 0.788191020488739, "step": 1329 }, { "epoch": 0.30673873873873875, "grad_norm": 21.212200164794922, "learning_rate": 9.491023482312258e-06, "loss": 12.430997848510742, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2711237296462059, "loss_total_v6": 0.7769373282790184, "step": 1330 }, { "epoch": 0.30696936936936936, "grad_norm": 34.006534576416016, "learning_rate": 9.490137878803078e-06, "loss": 12.746002197265625, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2813053224235773, "loss_total_v6": 0.7966251596808434, "step": 1331 }, { "epoch": 0.3072, "grad_norm": 21.45107650756836, "learning_rate": 9.489251546893441e-06, "loss": 13.119535446166992, "loss_alignment": 1.03076171875, "loss_alignment_w": 0.515380859375, "loss_sft": 0.3057498037815094, "loss_total_v6": 0.8199709877371788, "step": 1332 }, { "epoch": 0.30743063063063064, "grad_norm": 23.979650497436523, "learning_rate": 9.488364486727126e-06, "loss": 12.606359481811523, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.28661568462848663, "loss_total_v6": 0.7878974601626396, "step": 1333 }, { "epoch": 0.30766126126126125, "grad_norm": 23.538671493530273, "learning_rate": 9.487476698448033e-06, "loss": 12.485328674316406, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2672256901860237, "loss_total_v6": 0.7803330048918724, "step": 1334 }, { "epoch": 0.30789189189189187, "grad_norm": 21.906213760375977, "learning_rate": 9.486588182200184e-06, "loss": 12.541177749633789, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2696177177131176, "loss_total_v6": 0.783823624253273, "step": 1335 }, { "epoch": 0.30812252252252254, "grad_norm": 24.4071044921875, "learning_rate": 9.485698938127716e-06, "loss": 12.735528945922852, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2825123220682144, "loss_total_v6": 0.7959705889225006, "step": 1336 }, { "epoch": 0.30835315315315315, "grad_norm": 21.21377944946289, "learning_rate": 9.484808966374883e-06, "loss": 13.010151863098145, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2972500938922167, "loss_total_v6": 0.8131344988942146, "step": 1337 }, { "epoch": 0.30858378378378376, "grad_norm": 21.584224700927734, "learning_rate": 9.483918267086058e-06, "loss": 12.523326873779297, "loss_alignment": 1.02294921875, "loss_alignment_w": 0.511474609375, "loss_sft": 0.27184363827109337, "loss_total_v6": 0.7827078998088837, "step": 1338 }, { "epoch": 0.30881441441441443, "grad_norm": 22.962291717529297, "learning_rate": 9.483026840405734e-06, "loss": 12.38062572479248, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.26106321811676025, "loss_total_v6": 0.7737890630960464, "step": 1339 }, { "epoch": 0.30904504504504504, "grad_norm": 25.97719383239746, "learning_rate": 9.48213468647852e-06, "loss": 12.121333122253418, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.2623898535966873, "loss_total_v6": 0.7575833275914192, "step": 1340 }, { "epoch": 0.30927567567567565, "grad_norm": 22.356679916381836, "learning_rate": 9.48124180544914e-06, "loss": 12.18577766418457, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.2572318483144045, "loss_total_v6": 0.7616111487150192, "step": 1341 }, { "epoch": 0.3095063063063063, "grad_norm": 24.5649471282959, "learning_rate": 9.480348197462443e-06, "loss": 12.912745475769043, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2979675941169262, "loss_total_v6": 0.8070465847849846, "step": 1342 }, { "epoch": 0.30973693693693694, "grad_norm": 23.9169979095459, "learning_rate": 9.479453862663391e-06, "loss": 12.480485916137695, "loss_alignment": 1.02783203125, "loss_alignment_w": 0.513916015625, "loss_sft": 0.26684671826660633, "loss_total_v6": 0.7800303623080254, "step": 1343 }, { "epoch": 0.30996756756756755, "grad_norm": 26.803333282470703, "learning_rate": 9.478558801197065e-06, "loss": 12.623282432556152, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.28106631338596344, "loss_total_v6": 0.7889551445841789, "step": 1344 }, { "epoch": 0.3101981981981982, "grad_norm": 23.970169067382812, "learning_rate": 9.477663013208663e-06, "loss": 13.192224502563477, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.32330864295363426, "loss_total_v6": 0.8245140910148621, "step": 1345 }, { "epoch": 0.31042882882882883, "grad_norm": 24.571739196777344, "learning_rate": 9.476766498843504e-06, "loss": 12.70279312133789, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.28251093439757824, "loss_total_v6": 0.7939245253801346, "step": 1346 }, { "epoch": 0.31065945945945944, "grad_norm": 23.395999908447266, "learning_rate": 9.475869258247019e-06, "loss": 12.488624572753906, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.2798828613013029, "loss_total_v6": 0.7805389910936356, "step": 1347 }, { "epoch": 0.3108900900900901, "grad_norm": 23.813201904296875, "learning_rate": 9.474971291564764e-06, "loss": 12.188822746276855, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2529513519257307, "loss_total_v6": 0.7618014216423035, "step": 1348 }, { "epoch": 0.3111207207207207, "grad_norm": 22.607585906982422, "learning_rate": 9.474072598942408e-06, "loss": 12.657955169677734, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2831571325659752, "loss_total_v6": 0.7911222204566002, "step": 1349 }, { "epoch": 0.31135135135135134, "grad_norm": 22.394813537597656, "learning_rate": 9.473173180525737e-06, "loss": 12.449746131896973, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.26248412765562534, "loss_total_v6": 0.7781091034412384, "step": 1350 }, { "epoch": 0.311581981981982, "grad_norm": 26.14142608642578, "learning_rate": 9.472273036460659e-06, "loss": 12.423853874206543, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.26175087317824364, "loss_total_v6": 0.776490829885006, "step": 1351 }, { "epoch": 0.3118126126126126, "grad_norm": 23.555686950683594, "learning_rate": 9.4713721668932e-06, "loss": 12.023199081420898, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.23449745401740074, "loss_total_v6": 0.7514499574899673, "step": 1352 }, { "epoch": 0.31204324324324323, "grad_norm": 22.163230895996094, "learning_rate": 9.470470571969494e-06, "loss": 12.950448989868164, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2958532590419054, "loss_total_v6": 0.8094030469655991, "step": 1353 }, { "epoch": 0.3122738738738739, "grad_norm": 24.56324005126953, "learning_rate": 9.469568251835806e-06, "loss": 13.093035697937012, "loss_alignment": 1.03662109375, "loss_alignment_w": 0.518310546875, "loss_sft": 0.30004996806383133, "loss_total_v6": 0.8183147311210632, "step": 1354 }, { "epoch": 0.3125045045045045, "grad_norm": 20.744796752929688, "learning_rate": 9.46866520663851e-06, "loss": 12.407402992248535, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2596087958663702, "loss_total_v6": 0.7754626721143723, "step": 1355 }, { "epoch": 0.3127351351351351, "grad_norm": 28.080209732055664, "learning_rate": 9.4677614365241e-06, "loss": 12.493034362792969, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.27259010821580887, "loss_total_v6": 0.7808146104216576, "step": 1356 }, { "epoch": 0.3129657657657658, "grad_norm": 22.11823844909668, "learning_rate": 9.46685694163919e-06, "loss": 12.529166221618652, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.26261080615222454, "loss_total_v6": 0.7830728590488434, "step": 1357 }, { "epoch": 0.3131963963963964, "grad_norm": 23.64776039123535, "learning_rate": 9.465951722130504e-06, "loss": 12.704385757446289, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.27252452075481415, "loss_total_v6": 0.7940241619944572, "step": 1358 }, { "epoch": 0.313427027027027, "grad_norm": 22.474288940429688, "learning_rate": 9.465045778144894e-06, "loss": 12.93166446685791, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2919783778488636, "loss_total_v6": 0.8082290217280388, "step": 1359 }, { "epoch": 0.3136576576576577, "grad_norm": 25.50969886779785, "learning_rate": 9.46413910982932e-06, "loss": 12.398242950439453, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2597992289811373, "loss_total_v6": 0.7748901918530464, "step": 1360 }, { "epoch": 0.3138882882882883, "grad_norm": 25.387113571166992, "learning_rate": 9.463231717330869e-06, "loss": 12.93612289428711, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.295064652338624, "loss_total_v6": 0.8085077032446861, "step": 1361 }, { "epoch": 0.3141189189189189, "grad_norm": 24.578903198242188, "learning_rate": 9.462323600796736e-06, "loss": 12.857279777526855, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.29669826105237007, "loss_total_v6": 0.8035799711942673, "step": 1362 }, { "epoch": 0.3143495495495495, "grad_norm": 25.173707962036133, "learning_rate": 9.461414760374242e-06, "loss": 12.70583724975586, "loss_alignment": 1.02294921875, "loss_alignment_w": 0.511474609375, "loss_sft": 0.2816789299249649, "loss_total_v6": 0.7941148281097412, "step": 1363 }, { "epoch": 0.3145801801801802, "grad_norm": 24.95764923095703, "learning_rate": 9.460505196210814e-06, "loss": 12.891286849975586, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.295451533049345, "loss_total_v6": 0.8057054579257965, "step": 1364 }, { "epoch": 0.3148108108108108, "grad_norm": 41.209388732910156, "learning_rate": 9.45959490845401e-06, "loss": 13.133033752441406, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.3174729272723198, "loss_total_v6": 0.8208146169781685, "step": 1365 }, { "epoch": 0.3150414414414414, "grad_norm": 23.278669357299805, "learning_rate": 9.458683897251498e-06, "loss": 12.32625675201416, "loss_alignment": 1.00732421875, "loss_alignment_w": 0.503662109375, "loss_sft": 0.2651725374162197, "loss_total_v6": 0.7703910619020462, "step": 1366 }, { "epoch": 0.3152720720720721, "grad_norm": 22.496009826660156, "learning_rate": 9.457772162751061e-06, "loss": 12.323099136352539, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.2665926069021225, "loss_total_v6": 0.7701936736702919, "step": 1367 }, { "epoch": 0.3155027027027027, "grad_norm": 20.983301162719727, "learning_rate": 9.456859705100605e-06, "loss": 12.601716995239258, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2754766307771206, "loss_total_v6": 0.7876073569059372, "step": 1368 }, { "epoch": 0.3157333333333333, "grad_norm": 22.520788192749023, "learning_rate": 9.455946524448153e-06, "loss": 12.576665878295898, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.279800858348608, "loss_total_v6": 0.7860416769981384, "step": 1369 }, { "epoch": 0.315963963963964, "grad_norm": 23.713977813720703, "learning_rate": 9.45503262094184e-06, "loss": 12.197576522827148, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.25581776537001133, "loss_total_v6": 0.7623485028743744, "step": 1370 }, { "epoch": 0.3161945945945946, "grad_norm": 24.723007202148438, "learning_rate": 9.454117994729923e-06, "loss": 12.597864151000977, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.2621742710471153, "loss_total_v6": 0.7873665243387222, "step": 1371 }, { "epoch": 0.3164252252252252, "grad_norm": 22.7773494720459, "learning_rate": 9.453202645960775e-06, "loss": 13.086578369140625, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.30512423627078533, "loss_total_v6": 0.8179111331701279, "step": 1372 }, { "epoch": 0.3166558558558559, "grad_norm": 24.6280460357666, "learning_rate": 9.452286574782883e-06, "loss": 13.20254135131836, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.31340949796140194, "loss_total_v6": 0.8251587897539139, "step": 1373 }, { "epoch": 0.3168864864864865, "grad_norm": 22.055068969726562, "learning_rate": 9.451369781344859e-06, "loss": 12.460641860961914, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2668272126466036, "loss_total_v6": 0.7787900939583778, "step": 1374 }, { "epoch": 0.3171171171171171, "grad_norm": 23.25065803527832, "learning_rate": 9.450452265795423e-06, "loss": 13.003168106079102, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.30253565683960915, "loss_total_v6": 0.8126979842782021, "step": 1375 }, { "epoch": 0.31734774774774777, "grad_norm": 22.91633415222168, "learning_rate": 9.44953402828342e-06, "loss": 12.38469123840332, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.27454677410423756, "loss_total_v6": 0.7740432247519493, "step": 1376 }, { "epoch": 0.3175783783783784, "grad_norm": 24.13189125061035, "learning_rate": 9.448615068957807e-06, "loss": 12.780112266540527, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.28587860614061356, "loss_total_v6": 0.798757016658783, "step": 1377 }, { "epoch": 0.317809009009009, "grad_norm": 25.32760238647461, "learning_rate": 9.447695387967657e-06, "loss": 12.984723091125488, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2951572369784117, "loss_total_v6": 0.8115451782941818, "step": 1378 }, { "epoch": 0.31803963963963966, "grad_norm": 21.984127044677734, "learning_rate": 9.446774985462168e-06, "loss": 12.245537757873535, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.26090578734874725, "loss_total_v6": 0.7653460949659348, "step": 1379 }, { "epoch": 0.3182702702702703, "grad_norm": 23.70550537109375, "learning_rate": 9.445853861590647e-06, "loss": 12.37656021118164, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.27165813744068146, "loss_total_v6": 0.7735349908471107, "step": 1380 }, { "epoch": 0.3185009009009009, "grad_norm": 21.225473403930664, "learning_rate": 9.44493201650252e-06, "loss": 12.582761764526367, "loss_alignment": 1.02197265625, "loss_alignment_w": 0.510986328125, "loss_sft": 0.2753600403666496, "loss_total_v6": 0.7864226773381233, "step": 1381 }, { "epoch": 0.31873153153153155, "grad_norm": 23.852073669433594, "learning_rate": 9.444009450347334e-06, "loss": 12.076119422912598, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.24139071628451347, "loss_total_v6": 0.7547574490308762, "step": 1382 }, { "epoch": 0.31896216216216217, "grad_norm": 24.274070739746094, "learning_rate": 9.443086163274745e-06, "loss": 12.735475540161133, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2860032021999359, "loss_total_v6": 0.7959671914577484, "step": 1383 }, { "epoch": 0.3191927927927928, "grad_norm": 23.0888614654541, "learning_rate": 9.442162155434534e-06, "loss": 12.86513900756836, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.2992341108620167, "loss_total_v6": 0.8040711581707001, "step": 1384 }, { "epoch": 0.31942342342342345, "grad_norm": 24.655784606933594, "learning_rate": 9.441237426976596e-06, "loss": 12.40896987915039, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.26816016621887684, "loss_total_v6": 0.7755606472492218, "step": 1385 }, { "epoch": 0.31965405405405406, "grad_norm": 25.116352081298828, "learning_rate": 9.440311978050942e-06, "loss": 12.786697387695312, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2870073430240154, "loss_total_v6": 0.79916862398386, "step": 1386 }, { "epoch": 0.3198846846846847, "grad_norm": 24.728939056396484, "learning_rate": 9.439385808807698e-06, "loss": 12.192331314086914, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.2641874998807907, "loss_total_v6": 0.7620207518339157, "step": 1387 }, { "epoch": 0.32011531531531534, "grad_norm": 23.405057907104492, "learning_rate": 9.438458919397113e-06, "loss": 12.507621765136719, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2703433558344841, "loss_total_v6": 0.7817264050245285, "step": 1388 }, { "epoch": 0.32034594594594595, "grad_norm": 25.269319534301758, "learning_rate": 9.437531309969548e-06, "loss": 12.709295272827148, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.29526172019541264, "loss_total_v6": 0.794330932199955, "step": 1389 }, { "epoch": 0.32057657657657657, "grad_norm": 23.121082305908203, "learning_rate": 9.43660298067548e-06, "loss": 12.594202995300293, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2756783589720726, "loss_total_v6": 0.7871377095580101, "step": 1390 }, { "epoch": 0.32080720720720723, "grad_norm": 20.854509353637695, "learning_rate": 9.435673931665508e-06, "loss": 12.570901870727539, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.283514641225338, "loss_total_v6": 0.785681389272213, "step": 1391 }, { "epoch": 0.32103783783783785, "grad_norm": 24.461692810058594, "learning_rate": 9.434744163090341e-06, "loss": 12.535074234008789, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2727152146399021, "loss_total_v6": 0.7834421619772911, "step": 1392 }, { "epoch": 0.32126846846846846, "grad_norm": 24.16805648803711, "learning_rate": 9.433813675100809e-06, "loss": 12.40739631652832, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.26378925889730453, "loss_total_v6": 0.7754622548818588, "step": 1393 }, { "epoch": 0.3214990990990991, "grad_norm": 25.64105224609375, "learning_rate": 9.43288246784786e-06, "loss": 12.222415924072266, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.2685243524610996, "loss_total_v6": 0.7639009803533554, "step": 1394 }, { "epoch": 0.32172972972972974, "grad_norm": 24.781234741210938, "learning_rate": 9.431950541482554e-06, "loss": 12.96605110168457, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.29986485093832016, "loss_total_v6": 0.8103781789541245, "step": 1395 }, { "epoch": 0.32196036036036035, "grad_norm": 23.026031494140625, "learning_rate": 9.431017896156074e-06, "loss": 13.13486099243164, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.30811144411563873, "loss_total_v6": 0.8209288492798805, "step": 1396 }, { "epoch": 0.32219099099099097, "grad_norm": 22.34543228149414, "learning_rate": 9.430084532019711e-06, "loss": 12.41524600982666, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2676826175302267, "loss_total_v6": 0.7759528905153275, "step": 1397 }, { "epoch": 0.32242162162162163, "grad_norm": 23.275819778442383, "learning_rate": 9.429150449224882e-06, "loss": 12.358604431152344, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2600684370845556, "loss_total_v6": 0.7724127843976021, "step": 1398 }, { "epoch": 0.32265225225225225, "grad_norm": 22.551361083984375, "learning_rate": 9.428215647923113e-06, "loss": 12.411615371704102, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.2733608391135931, "loss_total_v6": 0.7757259458303452, "step": 1399 }, { "epoch": 0.32288288288288286, "grad_norm": 23.74951934814453, "learning_rate": 9.427280128266049e-06, "loss": 12.30112361907959, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.26442573964595795, "loss_total_v6": 0.7688202485442162, "step": 1400 }, { "epoch": 0.32311351351351353, "grad_norm": 23.118440628051758, "learning_rate": 9.426343890405456e-06, "loss": 12.476594924926758, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2657337784767151, "loss_total_v6": 0.7797871381044388, "step": 1401 }, { "epoch": 0.32334414414414414, "grad_norm": 23.136198043823242, "learning_rate": 9.425406934493208e-06, "loss": 12.719054222106934, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.28387825563549995, "loss_total_v6": 0.7949408888816833, "step": 1402 }, { "epoch": 0.32357477477477475, "grad_norm": 23.462533950805664, "learning_rate": 9.424469260681304e-06, "loss": 13.187524795532227, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.3086257614195347, "loss_total_v6": 0.8242202624678612, "step": 1403 }, { "epoch": 0.3238054054054054, "grad_norm": 23.53495216369629, "learning_rate": 9.423530869121855e-06, "loss": 13.052330017089844, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.3065696209669113, "loss_total_v6": 0.815770648419857, "step": 1404 }, { "epoch": 0.32403603603603603, "grad_norm": 21.904598236083984, "learning_rate": 9.422591759967085e-06, "loss": 12.92060375213623, "loss_alignment": 1.02978515625, "loss_alignment_w": 0.514892578125, "loss_sft": 0.2925231270492077, "loss_total_v6": 0.8075377643108368, "step": 1405 }, { "epoch": 0.32426666666666665, "grad_norm": 22.663951873779297, "learning_rate": 9.421651933369344e-06, "loss": 12.432181358337402, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.26355312019586563, "loss_total_v6": 0.7770113572478294, "step": 1406 }, { "epoch": 0.3244972972972973, "grad_norm": 24.971229553222656, "learning_rate": 9.42071138948109e-06, "loss": 12.947380065917969, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.29851483926177025, "loss_total_v6": 0.8092112392187119, "step": 1407 }, { "epoch": 0.32472792792792793, "grad_norm": 24.53562355041504, "learning_rate": 9.4197701284549e-06, "loss": 12.386129379272461, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2634977176785469, "loss_total_v6": 0.774133063852787, "step": 1408 }, { "epoch": 0.32495855855855854, "grad_norm": 21.352333068847656, "learning_rate": 9.418828150443469e-06, "loss": 12.038860321044922, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2420680895447731, "loss_total_v6": 0.7524288147687912, "step": 1409 }, { "epoch": 0.3251891891891892, "grad_norm": 23.78633689880371, "learning_rate": 9.417885455599604e-06, "loss": 13.092920303344727, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.3071838468313217, "loss_total_v6": 0.818307526409626, "step": 1410 }, { "epoch": 0.3254198198198198, "grad_norm": 23.252805709838867, "learning_rate": 9.416942044076233e-06, "loss": 12.360943794250488, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.26634868793189526, "loss_total_v6": 0.7725589945912361, "step": 1411 }, { "epoch": 0.32565045045045043, "grad_norm": 20.147092819213867, "learning_rate": 9.415997916026401e-06, "loss": 12.54074478149414, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2763350382447243, "loss_total_v6": 0.7837965935468674, "step": 1412 }, { "epoch": 0.3258810810810811, "grad_norm": 23.871257781982422, "learning_rate": 9.415053071603262e-06, "loss": 12.42405891418457, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.2708274442702532, "loss_total_v6": 0.776503711938858, "step": 1413 }, { "epoch": 0.3261117117117117, "grad_norm": 20.953765869140625, "learning_rate": 9.414107510960092e-06, "loss": 12.214107513427734, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.25230382941663265, "loss_total_v6": 0.7633817046880722, "step": 1414 }, { "epoch": 0.32634234234234233, "grad_norm": 22.64473533630371, "learning_rate": 9.413161234250284e-06, "loss": 12.972774505615234, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.3084180448204279, "loss_total_v6": 0.8107984289526939, "step": 1415 }, { "epoch": 0.326572972972973, "grad_norm": 22.738037109375, "learning_rate": 9.412214241627344e-06, "loss": 12.834949493408203, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.300063356757164, "loss_total_v6": 0.8021843433380127, "step": 1416 }, { "epoch": 0.3268036036036036, "grad_norm": 22.5750789642334, "learning_rate": 9.411266533244894e-06, "loss": 12.448006629943848, "loss_alignment": 0.9775390625, "loss_alignment_w": 0.48876953125, "loss_sft": 0.28892573714256287, "loss_total_v6": 0.7780004069209099, "step": 1417 }, { "epoch": 0.3270342342342342, "grad_norm": 23.8868350982666, "learning_rate": 9.410318109256677e-06, "loss": 12.706619262695312, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.2723283413797617, "loss_total_v6": 0.7941636741161346, "step": 1418 }, { "epoch": 0.3272648648648649, "grad_norm": 20.70743179321289, "learning_rate": 9.409368969816545e-06, "loss": 12.133139610290527, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.2542471420019865, "loss_total_v6": 0.7583212405443192, "step": 1419 }, { "epoch": 0.3274954954954955, "grad_norm": 25.721172332763672, "learning_rate": 9.40841911507847e-06, "loss": 12.432092666625977, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2693611029535532, "loss_total_v6": 0.7770057544112206, "step": 1420 }, { "epoch": 0.3277261261261261, "grad_norm": 22.669797897338867, "learning_rate": 9.407468545196541e-06, "loss": 12.286026000976562, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2656489070504904, "loss_total_v6": 0.7678766697645187, "step": 1421 }, { "epoch": 0.3279567567567568, "grad_norm": 22.76808738708496, "learning_rate": 9.406517260324962e-06, "loss": 11.9552583694458, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2427023109048605, "loss_total_v6": 0.7472036555409431, "step": 1422 }, { "epoch": 0.3281873873873874, "grad_norm": 23.095613479614258, "learning_rate": 9.405565260618051e-06, "loss": 12.619659423828125, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.28515810891985893, "loss_total_v6": 0.7887286692857742, "step": 1423 }, { "epoch": 0.328418018018018, "grad_norm": 24.181718826293945, "learning_rate": 9.404612546230244e-06, "loss": 12.068660736083984, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2438238300383091, "loss_total_v6": 0.754291333258152, "step": 1424 }, { "epoch": 0.3286486486486486, "grad_norm": 23.290163040161133, "learning_rate": 9.403659117316093e-06, "loss": 12.721266746520996, "loss_alignment": 1.02490234375, "loss_alignment_w": 0.512451171875, "loss_sft": 0.28230755031108856, "loss_total_v6": 0.7950791567564011, "step": 1425 }, { "epoch": 0.3288792792792793, "grad_norm": 22.03244972229004, "learning_rate": 9.402704974030266e-06, "loss": 12.454469680786133, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.28098308481276035, "loss_total_v6": 0.7784043252468109, "step": 1426 }, { "epoch": 0.3291099099099099, "grad_norm": 25.394012451171875, "learning_rate": 9.401750116527544e-06, "loss": 12.40041446685791, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.25877529941499233, "loss_total_v6": 0.7750258892774582, "step": 1427 }, { "epoch": 0.3293405405405405, "grad_norm": 28.385700225830078, "learning_rate": 9.400794544962828e-06, "loss": 12.487667083740234, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.27677135542035103, "loss_total_v6": 0.7804792374372482, "step": 1428 }, { "epoch": 0.3295711711711712, "grad_norm": 23.536190032958984, "learning_rate": 9.399838259491134e-06, "loss": 12.116485595703125, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.25175672583281994, "loss_total_v6": 0.7572803944349289, "step": 1429 }, { "epoch": 0.3298018018018018, "grad_norm": 23.852384567260742, "learning_rate": 9.398881260267589e-06, "loss": 12.184304237365723, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.25572067871689796, "loss_total_v6": 0.7615189924836159, "step": 1430 }, { "epoch": 0.3300324324324324, "grad_norm": 23.390613555908203, "learning_rate": 9.397923547447445e-06, "loss": 12.746317863464355, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2855517156422138, "loss_total_v6": 0.796644888818264, "step": 1431 }, { "epoch": 0.3302630630630631, "grad_norm": 24.642004013061523, "learning_rate": 9.396965121186058e-06, "loss": 12.943826675415039, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.29821645095944405, "loss_total_v6": 0.8089891523122787, "step": 1432 }, { "epoch": 0.3304936936936937, "grad_norm": 23.836172103881836, "learning_rate": 9.396005981638913e-06, "loss": 12.348370552062988, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2573688682168722, "loss_total_v6": 0.7717731520533562, "step": 1433 }, { "epoch": 0.3307243243243243, "grad_norm": 27.317787170410156, "learning_rate": 9.3950461289616e-06, "loss": 12.196496963500977, "loss_alignment": 1.02294921875, "loss_alignment_w": 0.511474609375, "loss_sft": 0.25272906199097633, "loss_total_v6": 0.762281060218811, "step": 1434 }, { "epoch": 0.33095495495495497, "grad_norm": 23.093381881713867, "learning_rate": 9.394085563309827e-06, "loss": 13.03462028503418, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.2981079965829849, "loss_total_v6": 0.8146637678146362, "step": 1435 }, { "epoch": 0.3311855855855856, "grad_norm": 23.08232307434082, "learning_rate": 9.393124284839424e-06, "loss": 12.80598258972168, "loss_alignment": 1.01513671875, "loss_alignment_w": 0.507568359375, "loss_sft": 0.2942704036831856, "loss_total_v6": 0.8003739193081856, "step": 1436 }, { "epoch": 0.3314162162162162, "grad_norm": 23.725793838500977, "learning_rate": 9.392162293706325e-06, "loss": 12.673249244689941, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2825260851532221, "loss_total_v6": 0.7920780926942825, "step": 1437 }, { "epoch": 0.33164684684684687, "grad_norm": 23.401079177856445, "learning_rate": 9.391199590066593e-06, "loss": 12.42044734954834, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2695488501340151, "loss_total_v6": 0.7762779742479324, "step": 1438 }, { "epoch": 0.3318774774774775, "grad_norm": 23.31825828552246, "learning_rate": 9.390236174076396e-06, "loss": 12.407204627990723, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.27020126581192017, "loss_total_v6": 0.7754502967000008, "step": 1439 }, { "epoch": 0.3321081081081081, "grad_norm": 24.090322494506836, "learning_rate": 9.389272045892023e-06, "loss": 12.721723556518555, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2887142840772867, "loss_total_v6": 0.7951076999306679, "step": 1440 }, { "epoch": 0.33233873873873876, "grad_norm": 21.394264221191406, "learning_rate": 9.388307205669878e-06, "loss": 12.160957336425781, "loss_alignment": 1.02294921875, "loss_alignment_w": 0.511474609375, "loss_sft": 0.24812739714980125, "loss_total_v6": 0.7600597813725471, "step": 1441 }, { "epoch": 0.33256936936936937, "grad_norm": 22.658105850219727, "learning_rate": 9.387341653566478e-06, "loss": 12.557671546936035, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.269443117082119, "loss_total_v6": 0.7848544716835022, "step": 1442 }, { "epoch": 0.3328, "grad_norm": 26.092853546142578, "learning_rate": 9.386375389738455e-06, "loss": 12.723905563354492, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2918261718004942, "loss_total_v6": 0.7952441200613976, "step": 1443 }, { "epoch": 0.33303063063063065, "grad_norm": 24.24036407470703, "learning_rate": 9.385408414342565e-06, "loss": 12.662906646728516, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.2893107049167156, "loss_total_v6": 0.7914316579699516, "step": 1444 }, { "epoch": 0.33326126126126127, "grad_norm": 23.22225570678711, "learning_rate": 9.384440727535666e-06, "loss": 12.33456039428711, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.25934384390711784, "loss_total_v6": 0.770909994840622, "step": 1445 }, { "epoch": 0.3334918918918919, "grad_norm": 19.671052932739258, "learning_rate": 9.383472329474743e-06, "loss": 12.859095573425293, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2968727760016918, "loss_total_v6": 0.803693450987339, "step": 1446 }, { "epoch": 0.33372252252252255, "grad_norm": 20.213430404663086, "learning_rate": 9.38250322031689e-06, "loss": 11.7916259765625, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.22858430072665215, "loss_total_v6": 0.7369766309857368, "step": 1447 }, { "epoch": 0.33395315315315316, "grad_norm": 21.575634002685547, "learning_rate": 9.381533400219319e-06, "loss": 12.542451858520508, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.284040542319417, "loss_total_v6": 0.7839032039046288, "step": 1448 }, { "epoch": 0.33418378378378377, "grad_norm": 22.417648315429688, "learning_rate": 9.380562869339354e-06, "loss": 12.658477783203125, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.29100227542221546, "loss_total_v6": 0.7911548912525177, "step": 1449 }, { "epoch": 0.33441441441441444, "grad_norm": 24.639345169067383, "learning_rate": 9.37959162783444e-06, "loss": 12.870719909667969, "loss_alignment": 1.02685546875, "loss_alignment_w": 0.513427734375, "loss_sft": 0.29228928685188293, "loss_total_v6": 0.8044200092554092, "step": 1450 }, { "epoch": 0.33464504504504505, "grad_norm": 23.52741050720215, "learning_rate": 9.378619675862131e-06, "loss": 12.867963790893555, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.2978695798665285, "loss_total_v6": 0.8042477667331696, "step": 1451 }, { "epoch": 0.33487567567567567, "grad_norm": 22.801475524902344, "learning_rate": 9.377647013580102e-06, "loss": 12.603170394897461, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.281136903911829, "loss_total_v6": 0.7876981571316719, "step": 1452 }, { "epoch": 0.3351063063063063, "grad_norm": 22.469738006591797, "learning_rate": 9.376673641146139e-06, "loss": 12.612902641296387, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.270560409873724, "loss_total_v6": 0.7883064076304436, "step": 1453 }, { "epoch": 0.33533693693693695, "grad_norm": 26.742725372314453, "learning_rate": 9.375699558718146e-06, "loss": 12.204048156738281, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.26786473765969276, "loss_total_v6": 0.762753039598465, "step": 1454 }, { "epoch": 0.33556756756756756, "grad_norm": 24.328611373901367, "learning_rate": 9.374724766454138e-06, "loss": 12.185745239257812, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.2593354545533657, "loss_total_v6": 0.7616090327501297, "step": 1455 }, { "epoch": 0.3357981981981982, "grad_norm": 23.94683074951172, "learning_rate": 9.37374926451225e-06, "loss": 12.369409561157227, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2565781082957983, "loss_total_v6": 0.7730881199240685, "step": 1456 }, { "epoch": 0.33602882882882884, "grad_norm": 23.554180145263672, "learning_rate": 9.372773053050733e-06, "loss": 12.978937149047852, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.3039661981165409, "loss_total_v6": 0.8111835867166519, "step": 1457 }, { "epoch": 0.33625945945945945, "grad_norm": 24.982460021972656, "learning_rate": 9.371796132227943e-06, "loss": 12.43900203704834, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.2679161299020052, "loss_total_v6": 0.7774376273155212, "step": 1458 }, { "epoch": 0.33649009009009007, "grad_norm": 22.19384002685547, "learning_rate": 9.370818502202366e-06, "loss": 12.055330276489258, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.23375904001295567, "loss_total_v6": 0.7534581422805786, "step": 1459 }, { "epoch": 0.33672072072072073, "grad_norm": 22.85521125793457, "learning_rate": 9.36984016313259e-06, "loss": 13.00547981262207, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.30292427353560925, "loss_total_v6": 0.8128425031900406, "step": 1460 }, { "epoch": 0.33695135135135135, "grad_norm": 22.592222213745117, "learning_rate": 9.368861115177327e-06, "loss": 12.34321117401123, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2552000880241394, "loss_total_v6": 0.7714507058262825, "step": 1461 }, { "epoch": 0.33718198198198196, "grad_norm": 22.40033531188965, "learning_rate": 9.367881358495399e-06, "loss": 12.58797836303711, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.27846313267946243, "loss_total_v6": 0.7867486625909805, "step": 1462 }, { "epoch": 0.33741261261261263, "grad_norm": 29.39097785949707, "learning_rate": 9.366900893245744e-06, "loss": 13.207544326782227, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.31904757767915726, "loss_total_v6": 0.8254714980721474, "step": 1463 }, { "epoch": 0.33764324324324324, "grad_norm": 23.057497024536133, "learning_rate": 9.365919719587413e-06, "loss": 12.25728988647461, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.26141138933598995, "loss_total_v6": 0.7660805881023407, "step": 1464 }, { "epoch": 0.33787387387387385, "grad_norm": 22.044511795043945, "learning_rate": 9.36493783767958e-06, "loss": 12.02469539642334, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.24594350159168243, "loss_total_v6": 0.7515434622764587, "step": 1465 }, { "epoch": 0.3381045045045045, "grad_norm": 22.42091941833496, "learning_rate": 9.363955247681521e-06, "loss": 12.272847175598145, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2602780647575855, "loss_total_v6": 0.7670529633760452, "step": 1466 }, { "epoch": 0.33833513513513513, "grad_norm": 22.118087768554688, "learning_rate": 9.362971949752642e-06, "loss": 12.619563102722168, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.27415051124989986, "loss_total_v6": 0.7887226715683937, "step": 1467 }, { "epoch": 0.33856576576576575, "grad_norm": 23.608747482299805, "learning_rate": 9.36198794405245e-06, "loss": 12.283526420593262, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.26830026134848595, "loss_total_v6": 0.76772041618824, "step": 1468 }, { "epoch": 0.3387963963963964, "grad_norm": 24.116424560546875, "learning_rate": 9.361003230740574e-06, "loss": 12.853072166442871, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2903318274766207, "loss_total_v6": 0.8033170402050018, "step": 1469 }, { "epoch": 0.33902702702702703, "grad_norm": 23.3371524810791, "learning_rate": 9.360017809976758e-06, "loss": 12.339923858642578, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2577259782701731, "loss_total_v6": 0.7712452784180641, "step": 1470 }, { "epoch": 0.33925765765765764, "grad_norm": 22.079059600830078, "learning_rate": 9.359031681920858e-06, "loss": 12.235276222229004, "loss_alignment": 0.9697265625, "loss_alignment_w": 0.48486328125, "loss_sft": 0.27842242270708084, "loss_total_v6": 0.7647047638893127, "step": 1471 }, { "epoch": 0.3394882882882883, "grad_norm": 21.28357696533203, "learning_rate": 9.358044846732848e-06, "loss": 12.913154602050781, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.29248473793268204, "loss_total_v6": 0.8070721253752708, "step": 1472 }, { "epoch": 0.3397189189189189, "grad_norm": 23.962377548217773, "learning_rate": 9.35705730457281e-06, "loss": 12.310611724853516, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.26056315191090107, "loss_total_v6": 0.7694132402539253, "step": 1473 }, { "epoch": 0.33994954954954953, "grad_norm": 23.124069213867188, "learning_rate": 9.356069055600949e-06, "loss": 12.380931854248047, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.2690932732075453, "loss_total_v6": 0.7738082557916641, "step": 1474 }, { "epoch": 0.3401801801801802, "grad_norm": 24.97834014892578, "learning_rate": 9.355080099977579e-06, "loss": 12.69873046875, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.27389529161155224, "loss_total_v6": 0.7936706990003586, "step": 1475 }, { "epoch": 0.3404108108108108, "grad_norm": 23.630298614501953, "learning_rate": 9.354090437863133e-06, "loss": 12.534412384033203, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2763512432575226, "loss_total_v6": 0.7834007889032364, "step": 1476 }, { "epoch": 0.34064144144144143, "grad_norm": 26.92420196533203, "learning_rate": 9.353100069418153e-06, "loss": 12.286953926086426, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.26837708055973053, "loss_total_v6": 0.7679345980286598, "step": 1477 }, { "epoch": 0.3408720720720721, "grad_norm": 22.51068115234375, "learning_rate": 9.352108994803303e-06, "loss": 12.348560333251953, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.26526949740946293, "loss_total_v6": 0.7717849761247635, "step": 1478 }, { "epoch": 0.3411027027027027, "grad_norm": 21.774322509765625, "learning_rate": 9.351117214179354e-06, "loss": 12.352120399475098, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.27216015197336674, "loss_total_v6": 0.7720075696706772, "step": 1479 }, { "epoch": 0.3413333333333333, "grad_norm": 24.213266372680664, "learning_rate": 9.350124727707197e-06, "loss": 12.806673049926758, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.29593096673488617, "loss_total_v6": 0.8004170432686806, "step": 1480 }, { "epoch": 0.341563963963964, "grad_norm": 24.416393280029297, "learning_rate": 9.349131535547834e-06, "loss": 12.663095474243164, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.28344782441854477, "loss_total_v6": 0.7914434149861336, "step": 1481 }, { "epoch": 0.3417945945945946, "grad_norm": 22.530197143554688, "learning_rate": 9.348137637862385e-06, "loss": 12.62969970703125, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.28456501476466656, "loss_total_v6": 0.7893562465906143, "step": 1482 }, { "epoch": 0.3420252252252252, "grad_norm": 26.275548934936523, "learning_rate": 9.347143034812078e-06, "loss": 13.072321891784668, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.30218861252069473, "loss_total_v6": 0.8170201405882835, "step": 1483 }, { "epoch": 0.34225585585585583, "grad_norm": 24.045560836791992, "learning_rate": 9.346147726558266e-06, "loss": 13.04916763305664, "loss_alignment": 1.02294921875, "loss_alignment_w": 0.511474609375, "loss_sft": 0.30348799005150795, "loss_total_v6": 0.815572939813137, "step": 1484 }, { "epoch": 0.3424864864864865, "grad_norm": 24.6905517578125, "learning_rate": 9.345151713262404e-06, "loss": 12.232609748840332, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.25689348205924034, "loss_total_v6": 0.7645381391048431, "step": 1485 }, { "epoch": 0.3427171171171171, "grad_norm": 21.446378707885742, "learning_rate": 9.344154995086072e-06, "loss": 12.59913444519043, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.28309708274900913, "loss_total_v6": 0.7874458506703377, "step": 1486 }, { "epoch": 0.3429477477477477, "grad_norm": 25.074216842651367, "learning_rate": 9.343157572190957e-06, "loss": 12.689033508300781, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.28564875572919846, "loss_total_v6": 0.7930645272135735, "step": 1487 }, { "epoch": 0.3431783783783784, "grad_norm": 25.0394344329834, "learning_rate": 9.342159444738865e-06, "loss": 13.197395324707031, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.3094716425985098, "loss_total_v6": 0.8248372301459312, "step": 1488 }, { "epoch": 0.343409009009009, "grad_norm": 21.258859634399414, "learning_rate": 9.341160612891716e-06, "loss": 12.909570693969727, "loss_alignment": 1.02099609375, "loss_alignment_w": 0.510498046875, "loss_sft": 0.2972808964550495, "loss_total_v6": 0.8068481385707855, "step": 1489 }, { "epoch": 0.3436396396396396, "grad_norm": 23.260316848754883, "learning_rate": 9.340161076811539e-06, "loss": 13.1654052734375, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.30924223735928535, "loss_total_v6": 0.8228378221392632, "step": 1490 }, { "epoch": 0.3438702702702703, "grad_norm": 24.544902801513672, "learning_rate": 9.33916083666048e-06, "loss": 13.044541358947754, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.3078070431947708, "loss_total_v6": 0.8152838572859764, "step": 1491 }, { "epoch": 0.3441009009009009, "grad_norm": 23.191776275634766, "learning_rate": 9.338159892600809e-06, "loss": 11.895515441894531, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.2455144114792347, "loss_total_v6": 0.7434697300195694, "step": 1492 }, { "epoch": 0.3443315315315315, "grad_norm": 23.66101837158203, "learning_rate": 9.337158244794891e-06, "loss": 12.437799453735352, "loss_alignment": 1.03173828125, "loss_alignment_w": 0.515869140625, "loss_sft": 0.26253094524145126, "loss_total_v6": 0.7773624658584595, "step": 1493 }, { "epoch": 0.3445621621621622, "grad_norm": 22.335901260375977, "learning_rate": 9.336155893405224e-06, "loss": 12.339143753051758, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2585164476186037, "loss_total_v6": 0.7711964994668961, "step": 1494 }, { "epoch": 0.3447927927927928, "grad_norm": 19.284696578979492, "learning_rate": 9.335152838594404e-06, "loss": 12.333890914916992, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.2648714389652014, "loss_total_v6": 0.7708681374788284, "step": 1495 }, { "epoch": 0.3450234234234234, "grad_norm": 25.542661666870117, "learning_rate": 9.334149080525154e-06, "loss": 12.625762939453125, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.2838764265179634, "loss_total_v6": 0.7891102135181427, "step": 1496 }, { "epoch": 0.34525405405405407, "grad_norm": 23.165220260620117, "learning_rate": 9.333144619360306e-06, "loss": 13.121969223022461, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.3147062472999096, "loss_total_v6": 0.8201230689883232, "step": 1497 }, { "epoch": 0.3454846846846847, "grad_norm": 21.649852752685547, "learning_rate": 9.332139455262803e-06, "loss": 12.559470176696777, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.27217998541891575, "loss_total_v6": 0.7849668487906456, "step": 1498 }, { "epoch": 0.3457153153153153, "grad_norm": 23.94564437866211, "learning_rate": 9.331133588395709e-06, "loss": 12.38235092163086, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.27772691659629345, "loss_total_v6": 0.7738969475030899, "step": 1499 }, { "epoch": 0.34594594594594597, "grad_norm": 22.347461700439453, "learning_rate": 9.330127018922195e-06, "loss": 12.689083099365234, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.27692391350865364, "loss_total_v6": 0.7930677384138107, "step": 1500 }, { "epoch": 0.34594594594594597, "eval_loss": 0.9673298597335815, "eval_loss_alignment": 1.0303224885844748, "eval_loss_alignment_w": 0.5151612442922374, "eval_loss_contrast": 0.0, "eval_loss_highlevel": 0.0, "eval_loss_sft": 0.45216861985555795, "eval_loss_total_v6": 0.9673298699398564, "eval_loss_wm_recon": 0.0, "eval_loss_wm_recon_w": 0.0, "eval_runtime": 302.1057, "eval_samples_per_second": 5.776, "eval_steps_per_second": 0.725, "step": 1500 }, { "epoch": 0.3461765765765766, "grad_norm": 25.50492286682129, "learning_rate": 9.329119747005548e-06, "loss": 12.929819107055664, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.29740202985703945, "loss_total_v6": 0.808113694190979, "step": 1501 }, { "epoch": 0.3464072072072072, "grad_norm": 23.573575973510742, "learning_rate": 9.328111772809173e-06, "loss": 12.287212371826172, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2611758355051279, "loss_total_v6": 0.7679507583379745, "step": 1502 }, { "epoch": 0.34663783783783786, "grad_norm": 22.6658935546875, "learning_rate": 9.327103096496584e-06, "loss": 12.38606071472168, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.270085196942091, "loss_total_v6": 0.7741287872195244, "step": 1503 }, { "epoch": 0.34686846846846847, "grad_norm": 24.034177780151367, "learning_rate": 9.326093718231413e-06, "loss": 12.375940322875977, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.2719246093183756, "loss_total_v6": 0.7734962403774261, "step": 1504 }, { "epoch": 0.3470990990990991, "grad_norm": 24.482913970947266, "learning_rate": 9.325083638177401e-06, "loss": 12.03286361694336, "loss_alignment": 1.01513671875, "loss_alignment_w": 0.507568359375, "loss_sft": 0.24598096683621407, "loss_total_v6": 0.7520539686083794, "step": 1505 }, { "epoch": 0.34732972972972975, "grad_norm": 23.172016143798828, "learning_rate": 9.324072856498404e-06, "loss": 12.472521781921387, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.2769233826547861, "loss_total_v6": 0.7795326113700867, "step": 1506 }, { "epoch": 0.34756036036036037, "grad_norm": 20.78577995300293, "learning_rate": 9.3230613733584e-06, "loss": 12.192174911499023, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2467369083315134, "loss_total_v6": 0.7620109543204308, "step": 1507 }, { "epoch": 0.347790990990991, "grad_norm": 23.836393356323242, "learning_rate": 9.322049188921467e-06, "loss": 12.609098434448242, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.283140005543828, "loss_total_v6": 0.7880686074495316, "step": 1508 }, { "epoch": 0.34802162162162165, "grad_norm": 23.171527862548828, "learning_rate": 9.321036303351809e-06, "loss": 12.842708587646484, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2979085221886635, "loss_total_v6": 0.8026692643761635, "step": 1509 }, { "epoch": 0.34825225225225226, "grad_norm": 23.609081268310547, "learning_rate": 9.320022716813735e-06, "loss": 12.444223403930664, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.27489532344043255, "loss_total_v6": 0.7777639701962471, "step": 1510 }, { "epoch": 0.34848288288288287, "grad_norm": 23.507570266723633, "learning_rate": 9.319008429471675e-06, "loss": 12.331424713134766, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.2711413577198982, "loss_total_v6": 0.7707140892744064, "step": 1511 }, { "epoch": 0.34871351351351354, "grad_norm": 24.315959930419922, "learning_rate": 9.317993441490163e-06, "loss": 12.384108543395996, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.25795452669262886, "loss_total_v6": 0.774006798863411, "step": 1512 }, { "epoch": 0.34894414414414415, "grad_norm": 23.534950256347656, "learning_rate": 9.316977753033858e-06, "loss": 12.523092269897461, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.28196084685623646, "loss_total_v6": 0.7826933041214943, "step": 1513 }, { "epoch": 0.34917477477477477, "grad_norm": 21.984724044799805, "learning_rate": 9.315961364267527e-06, "loss": 12.155298233032227, "loss_alignment": 1.01904296875, "loss_alignment_w": 0.509521484375, "loss_sft": 0.2488265447318554, "loss_total_v6": 0.75970608741045, "step": 1514 }, { "epoch": 0.3494054054054054, "grad_norm": 23.815475463867188, "learning_rate": 9.314944275356049e-06, "loss": 12.757732391357422, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2821605149656534, "loss_total_v6": 0.7973582819104195, "step": 1515 }, { "epoch": 0.34963603603603605, "grad_norm": 24.594823837280273, "learning_rate": 9.31392648646442e-06, "loss": 12.434423446655273, "loss_alignment": 1.02587890625, "loss_alignment_w": 0.512939453125, "loss_sft": 0.2657988518476486, "loss_total_v6": 0.7771514132618904, "step": 1516 }, { "epoch": 0.34986666666666666, "grad_norm": 26.615745544433594, "learning_rate": 9.312907997757747e-06, "loss": 13.323431015014648, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.3193324878811836, "loss_total_v6": 0.8327144309878349, "step": 1517 }, { "epoch": 0.35009729729729727, "grad_norm": 23.699800491333008, "learning_rate": 9.31188880940125e-06, "loss": 13.28272819519043, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.3161324169486761, "loss_total_v6": 0.8301705121994019, "step": 1518 }, { "epoch": 0.35032792792792794, "grad_norm": 21.963970184326172, "learning_rate": 9.310868921560268e-06, "loss": 12.563197135925293, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.27566303312778473, "loss_total_v6": 0.7851998060941696, "step": 1519 }, { "epoch": 0.35055855855855855, "grad_norm": 24.4898738861084, "learning_rate": 9.309848334400247e-06, "loss": 13.10023307800293, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.3137138858437538, "loss_total_v6": 0.8187645599246025, "step": 1520 }, { "epoch": 0.35078918918918917, "grad_norm": 23.56010627746582, "learning_rate": 9.30882704808675e-06, "loss": 12.616582870483398, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.28078502230346203, "loss_total_v6": 0.788536474108696, "step": 1521 }, { "epoch": 0.35101981981981983, "grad_norm": 23.482885360717773, "learning_rate": 9.307805062785451e-06, "loss": 12.389114379882812, "loss_alignment": 1.01318359375, "loss_alignment_w": 0.506591796875, "loss_sft": 0.26829239167273045, "loss_total_v6": 0.7743196338415146, "step": 1522 }, { "epoch": 0.35125045045045045, "grad_norm": 25.093761444091797, "learning_rate": 9.306782378662138e-06, "loss": 11.713926315307617, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.23184574767947197, "loss_total_v6": 0.7321204170584679, "step": 1523 }, { "epoch": 0.35148108108108106, "grad_norm": 26.334054946899414, "learning_rate": 9.305758995882716e-06, "loss": 12.910543441772461, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.28994111716747284, "loss_total_v6": 0.806908905506134, "step": 1524 }, { "epoch": 0.3517117117117117, "grad_norm": 24.869670867919922, "learning_rate": 9.3047349146132e-06, "loss": 12.46364688873291, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.26291041262447834, "loss_total_v6": 0.7789779230952263, "step": 1525 }, { "epoch": 0.35194234234234234, "grad_norm": 23.08758544921875, "learning_rate": 9.30371013501972e-06, "loss": 12.65969467163086, "loss_alignment": 1.01220703125, "loss_alignment_w": 0.506103515625, "loss_sft": 0.28405926562845707, "loss_total_v6": 0.7912308797240257, "step": 1526 }, { "epoch": 0.35217297297297295, "grad_norm": 23.11258316040039, "learning_rate": 9.302684657268513e-06, "loss": 12.371950149536133, "loss_alignment": 1.01806640625, "loss_alignment_w": 0.509033203125, "loss_sft": 0.2633439786732197, "loss_total_v6": 0.7732469215989113, "step": 1527 }, { "epoch": 0.3524036036036036, "grad_norm": 26.60556983947754, "learning_rate": 9.30165848152594e-06, "loss": 12.571109771728516, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.27951456792652607, "loss_total_v6": 0.785694383084774, "step": 1528 }, { "epoch": 0.35263423423423423, "grad_norm": 26.617210388183594, "learning_rate": 9.300631607958464e-06, "loss": 12.580137252807617, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.27869022637605667, "loss_total_v6": 0.7862586081027985, "step": 1529 }, { "epoch": 0.35286486486486485, "grad_norm": 25.389421463012695, "learning_rate": 9.299604036732672e-06, "loss": 13.134223937988281, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.31032987870275974, "loss_total_v6": 0.8208889663219452, "step": 1530 }, { "epoch": 0.3530954954954955, "grad_norm": 24.46512222290039, "learning_rate": 9.298575768015257e-06, "loss": 11.783782958984375, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.23023037053644657, "loss_total_v6": 0.7364864647388458, "step": 1531 }, { "epoch": 0.3533261261261261, "grad_norm": 22.872583389282227, "learning_rate": 9.297546801973027e-06, "loss": 13.039924621582031, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.3058704752475023, "loss_total_v6": 0.8149952441453934, "step": 1532 }, { "epoch": 0.35355675675675674, "grad_norm": 21.915958404541016, "learning_rate": 9.296517138772902e-06, "loss": 12.642171859741211, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.2884115446358919, "loss_total_v6": 0.7901357635855675, "step": 1533 }, { "epoch": 0.3537873873873874, "grad_norm": 25.294639587402344, "learning_rate": 9.295486778581919e-06, "loss": 12.82603645324707, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2944861464202404, "loss_total_v6": 0.8016272783279419, "step": 1534 }, { "epoch": 0.354018018018018, "grad_norm": 23.71316146850586, "learning_rate": 9.294455721567224e-06, "loss": 12.726563453674316, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2910461761057377, "loss_total_v6": 0.7954102084040642, "step": 1535 }, { "epoch": 0.35424864864864863, "grad_norm": 24.126623153686523, "learning_rate": 9.293423967896076e-06, "loss": 12.134466171264648, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.2603725269436836, "loss_total_v6": 0.7584041431546211, "step": 1536 }, { "epoch": 0.3544792792792793, "grad_norm": 20.621583938598633, "learning_rate": 9.292391517735851e-06, "loss": 12.973549842834473, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.30372102558612823, "loss_total_v6": 0.8108468800783157, "step": 1537 }, { "epoch": 0.3547099099099099, "grad_norm": 23.569622039794922, "learning_rate": 9.291358371254033e-06, "loss": 12.887359619140625, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.30460546165704727, "loss_total_v6": 0.8054599314928055, "step": 1538 }, { "epoch": 0.35494054054054053, "grad_norm": 23.985544204711914, "learning_rate": 9.290324528618225e-06, "loss": 13.068013191223145, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.30822116136550903, "loss_total_v6": 0.8167508393526077, "step": 1539 }, { "epoch": 0.3551711711711712, "grad_norm": 25.59369468688965, "learning_rate": 9.289289989996133e-06, "loss": 12.677505493164062, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.29106234200298786, "loss_total_v6": 0.7923440858721733, "step": 1540 }, { "epoch": 0.3554018018018018, "grad_norm": 21.512420654296875, "learning_rate": 9.288254755555591e-06, "loss": 12.011453628540039, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.24502437002956867, "loss_total_v6": 0.7507158666849136, "step": 1541 }, { "epoch": 0.3556324324324324, "grad_norm": 22.810819625854492, "learning_rate": 9.287218825464528e-06, "loss": 12.227612495422363, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.25085914321243763, "loss_total_v6": 0.7642258182168007, "step": 1542 }, { "epoch": 0.3558630630630631, "grad_norm": 22.777984619140625, "learning_rate": 9.286182199891e-06, "loss": 12.499872207641602, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.28041806630790234, "loss_total_v6": 0.7812420427799225, "step": 1543 }, { "epoch": 0.3560936936936937, "grad_norm": 22.26558494567871, "learning_rate": 9.285144879003173e-06, "loss": 12.47338581085205, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.257049385458231, "loss_total_v6": 0.7795866057276726, "step": 1544 }, { "epoch": 0.3563243243243243, "grad_norm": 24.840293884277344, "learning_rate": 9.284106862969318e-06, "loss": 12.588464736938477, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.27649462036788464, "loss_total_v6": 0.7867790535092354, "step": 1545 }, { "epoch": 0.35655495495495493, "grad_norm": 24.323013305664062, "learning_rate": 9.283068151957828e-06, "loss": 12.442293167114258, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.25512135215103626, "loss_total_v6": 0.7776433378458023, "step": 1546 }, { "epoch": 0.3567855855855856, "grad_norm": 23.8992977142334, "learning_rate": 9.282028746137203e-06, "loss": 12.978315353393555, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2961605302989483, "loss_total_v6": 0.8111447021365166, "step": 1547 }, { "epoch": 0.3570162162162162, "grad_norm": 23.739578247070312, "learning_rate": 9.28098864567606e-06, "loss": 13.313399314880371, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.3193005993962288, "loss_total_v6": 0.8320874646306038, "step": 1548 }, { "epoch": 0.3572468468468468, "grad_norm": 23.772022247314453, "learning_rate": 9.279947850743125e-06, "loss": 12.55802059173584, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.280100267380476, "loss_total_v6": 0.7848762944340706, "step": 1549 }, { "epoch": 0.3574774774774775, "grad_norm": 28.186458587646484, "learning_rate": 9.278906361507238e-06, "loss": 12.164072036743164, "loss_alignment": 0.9833984375, "loss_alignment_w": 0.49169921875, "loss_sft": 0.26957762986421585, "loss_total_v6": 0.7602545097470284, "step": 1550 }, { "epoch": 0.3577081081081081, "grad_norm": 24.245872497558594, "learning_rate": 9.277864178137352e-06, "loss": 12.761041641235352, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.2979313600808382, "loss_total_v6": 0.7975651174783707, "step": 1551 }, { "epoch": 0.3579387387387387, "grad_norm": 22.226442337036133, "learning_rate": 9.276821300802535e-06, "loss": 12.38490104675293, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.26375662349164486, "loss_total_v6": 0.7740562930703163, "step": 1552 }, { "epoch": 0.3581693693693694, "grad_norm": 24.24321174621582, "learning_rate": 9.275777729671959e-06, "loss": 12.834657669067383, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2875939477235079, "loss_total_v6": 0.8021660819649696, "step": 1553 }, { "epoch": 0.3584, "grad_norm": 24.881759643554688, "learning_rate": 9.27473346491492e-06, "loss": 12.781190872192383, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2927361987531185, "loss_total_v6": 0.7988244667649269, "step": 1554 }, { "epoch": 0.3586306306306306, "grad_norm": 27.203777313232422, "learning_rate": 9.27368850670082e-06, "loss": 12.64666748046875, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.28142932057380676, "loss_total_v6": 0.7904167324304581, "step": 1555 }, { "epoch": 0.3588612612612613, "grad_norm": 21.90923309326172, "learning_rate": 9.272642855199171e-06, "loss": 12.307714462280273, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.2694610580801964, "loss_total_v6": 0.7692321762442589, "step": 1556 }, { "epoch": 0.3590918918918919, "grad_norm": 21.201032638549805, "learning_rate": 9.271596510579606e-06, "loss": 12.770037651062012, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.27298084273934364, "loss_total_v6": 0.7981273233890533, "step": 1557 }, { "epoch": 0.3593225225225225, "grad_norm": 21.560070037841797, "learning_rate": 9.27054947301186e-06, "loss": 12.984729766845703, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.29775163903832436, "loss_total_v6": 0.8115455880761147, "step": 1558 }, { "epoch": 0.35955315315315317, "grad_norm": 24.378765106201172, "learning_rate": 9.269501742665792e-06, "loss": 12.523994445800781, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.27875182405114174, "loss_total_v6": 0.7827496379613876, "step": 1559 }, { "epoch": 0.3597837837837838, "grad_norm": 23.1345272064209, "learning_rate": 9.268453319711362e-06, "loss": 13.117932319641113, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.3048561327159405, "loss_total_v6": 0.8198707774281502, "step": 1560 }, { "epoch": 0.3600144144144144, "grad_norm": 20.752138137817383, "learning_rate": 9.267404204318651e-06, "loss": 12.743885040283203, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.29748459346592426, "loss_total_v6": 0.7964927926659584, "step": 1561 }, { "epoch": 0.36024504504504506, "grad_norm": 24.655393600463867, "learning_rate": 9.266354396657846e-06, "loss": 12.936405181884766, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.29642509296536446, "loss_total_v6": 0.8085253238677979, "step": 1562 }, { "epoch": 0.3604756756756757, "grad_norm": 24.449642181396484, "learning_rate": 9.265303896899251e-06, "loss": 12.57819938659668, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2717178761959076, "loss_total_v6": 0.7861374393105507, "step": 1563 }, { "epoch": 0.3607063063063063, "grad_norm": 23.256948471069336, "learning_rate": 9.264252705213281e-06, "loss": 12.580102920532227, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.2882400546222925, "loss_total_v6": 0.7862564250826836, "step": 1564 }, { "epoch": 0.36093693693693696, "grad_norm": 23.3167667388916, "learning_rate": 9.263200821770462e-06, "loss": 12.617804527282715, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2816089987754822, "loss_total_v6": 0.7886127904057503, "step": 1565 }, { "epoch": 0.36116756756756757, "grad_norm": 25.26683235168457, "learning_rate": 9.262148246741433e-06, "loss": 12.675712585449219, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.28205442428588867, "loss_total_v6": 0.7922320291399956, "step": 1566 }, { "epoch": 0.3613981981981982, "grad_norm": 21.07280158996582, "learning_rate": 9.261094980296944e-06, "loss": 12.190969467163086, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.24975916929543018, "loss_total_v6": 0.7619356587529182, "step": 1567 }, { "epoch": 0.36162882882882885, "grad_norm": 23.270816802978516, "learning_rate": 9.26004102260786e-06, "loss": 12.3001708984375, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.27527618780732155, "loss_total_v6": 0.7687606960535049, "step": 1568 }, { "epoch": 0.36185945945945946, "grad_norm": 24.035869598388672, "learning_rate": 9.258986373845156e-06, "loss": 12.869388580322266, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2958833947777748, "loss_total_v6": 0.804336778819561, "step": 1569 }, { "epoch": 0.3620900900900901, "grad_norm": 24.6597957611084, "learning_rate": 9.257931034179919e-06, "loss": 12.093964576721191, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2494183573871851, "loss_total_v6": 0.7558728232979774, "step": 1570 }, { "epoch": 0.36232072072072075, "grad_norm": 21.875120162963867, "learning_rate": 9.256875003783348e-06, "loss": 12.644472122192383, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.28332153521478176, "loss_total_v6": 0.7902795299887657, "step": 1571 }, { "epoch": 0.36255135135135136, "grad_norm": 21.858015060424805, "learning_rate": 9.255818282826755e-06, "loss": 12.810663223266602, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2943493667989969, "loss_total_v6": 0.8006664663553238, "step": 1572 }, { "epoch": 0.36278198198198197, "grad_norm": 23.769575119018555, "learning_rate": 9.254760871481566e-06, "loss": 12.129981994628906, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.23876048438251019, "loss_total_v6": 0.7581238746643066, "step": 1573 }, { "epoch": 0.36301261261261264, "grad_norm": 27.56369400024414, "learning_rate": 9.253702769919315e-06, "loss": 12.879627227783203, "loss_alignment": 1.02587890625, "loss_alignment_w": 0.512939453125, "loss_sft": 0.29139637015759945, "loss_total_v6": 0.8049767091870308, "step": 1574 }, { "epoch": 0.36324324324324325, "grad_norm": 21.83915138244629, "learning_rate": 9.252643978311649e-06, "loss": 12.233739852905273, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.2689421996474266, "loss_total_v6": 0.7646087184548378, "step": 1575 }, { "epoch": 0.36347387387387387, "grad_norm": 25.338972091674805, "learning_rate": 9.251584496830329e-06, "loss": 12.442626953125, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2723998725414276, "loss_total_v6": 0.7776641920208931, "step": 1576 }, { "epoch": 0.3637045045045045, "grad_norm": 23.801025390625, "learning_rate": 9.250524325647224e-06, "loss": 12.680679321289062, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.27676492743194103, "loss_total_v6": 0.7925424724817276, "step": 1577 }, { "epoch": 0.36393513513513515, "grad_norm": 21.448331832885742, "learning_rate": 9.24946346493432e-06, "loss": 12.071671485900879, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.24245553277432919, "loss_total_v6": 0.7544794529676437, "step": 1578 }, { "epoch": 0.36416576576576576, "grad_norm": 22.94542121887207, "learning_rate": 9.248401914863712e-06, "loss": 12.567928314208984, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.2809636779129505, "loss_total_v6": 0.7854955047369003, "step": 1579 }, { "epoch": 0.36439639639639637, "grad_norm": 25.537012100219727, "learning_rate": 9.247339675607606e-06, "loss": 12.490533828735352, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.2793461047112942, "loss_total_v6": 0.7806583419442177, "step": 1580 }, { "epoch": 0.36462702702702704, "grad_norm": 23.064472198486328, "learning_rate": 9.24627674733832e-06, "loss": 12.503190994262695, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.25909529626369476, "loss_total_v6": 0.7814494147896767, "step": 1581 }, { "epoch": 0.36485765765765765, "grad_norm": 26.101594924926758, "learning_rate": 9.245213130228288e-06, "loss": 12.763372421264648, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2853663694113493, "loss_total_v6": 0.7977107465267181, "step": 1582 }, { "epoch": 0.36508828828828827, "grad_norm": 25.696666717529297, "learning_rate": 9.24414882445005e-06, "loss": 12.299161911010742, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.2666833959519863, "loss_total_v6": 0.7686975747346878, "step": 1583 }, { "epoch": 0.36531891891891893, "grad_norm": 24.485721588134766, "learning_rate": 9.243083830176261e-06, "loss": 13.045099258422852, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.3048512041568756, "loss_total_v6": 0.8153187334537506, "step": 1584 }, { "epoch": 0.36554954954954955, "grad_norm": 23.14337921142578, "learning_rate": 9.242018147579685e-06, "loss": 12.752642631530762, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.28613013587892056, "loss_total_v6": 0.7970401793718338, "step": 1585 }, { "epoch": 0.36578018018018016, "grad_norm": 21.37570571899414, "learning_rate": 9.240951776833202e-06, "loss": 12.770792007446289, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.29721315018832684, "loss_total_v6": 0.7981744855642319, "step": 1586 }, { "epoch": 0.3660108108108108, "grad_norm": 21.753299713134766, "learning_rate": 9.2398847181098e-06, "loss": 12.319097518920898, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.2697909642010927, "loss_total_v6": 0.769943580031395, "step": 1587 }, { "epoch": 0.36624144144144144, "grad_norm": 23.116846084594727, "learning_rate": 9.238816971582578e-06, "loss": 12.834842681884766, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2848894353955984, "loss_total_v6": 0.8021776303648949, "step": 1588 }, { "epoch": 0.36647207207207205, "grad_norm": 20.78203582763672, "learning_rate": 9.23774853742475e-06, "loss": 12.455411911010742, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.272451251745224, "loss_total_v6": 0.7784631997346878, "step": 1589 }, { "epoch": 0.3667027027027027, "grad_norm": 20.748403549194336, "learning_rate": 9.236679415809641e-06, "loss": 12.807110786437988, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.29070933163166046, "loss_total_v6": 0.8004444241523743, "step": 1590 }, { "epoch": 0.36693333333333333, "grad_norm": 23.10670280456543, "learning_rate": 9.235609606910687e-06, "loss": 12.195555686950684, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.25518794171512127, "loss_total_v6": 0.7622222304344177, "step": 1591 }, { "epoch": 0.36716396396396395, "grad_norm": 22.249515533447266, "learning_rate": 9.23453911090143e-06, "loss": 12.499618530273438, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.27208616957068443, "loss_total_v6": 0.781226210296154, "step": 1592 }, { "epoch": 0.3673945945945946, "grad_norm": 22.824207305908203, "learning_rate": 9.233467927955533e-06, "loss": 11.766761779785156, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.2390084806829691, "loss_total_v6": 0.735422670841217, "step": 1593 }, { "epoch": 0.3676252252252252, "grad_norm": 22.16664695739746, "learning_rate": 9.232396058246763e-06, "loss": 12.598772048950195, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.27900038845837116, "loss_total_v6": 0.7874232456088066, "step": 1594 }, { "epoch": 0.36785585585585584, "grad_norm": 27.246540069580078, "learning_rate": 9.231323501949003e-06, "loss": 12.888713836669922, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.3026454281061888, "loss_total_v6": 0.8055446073412895, "step": 1595 }, { "epoch": 0.3680864864864865, "grad_norm": 26.579484939575195, "learning_rate": 9.230250259236245e-06, "loss": 12.889663696289062, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2958078235387802, "loss_total_v6": 0.8056039586663246, "step": 1596 }, { "epoch": 0.3683171171171171, "grad_norm": 20.852039337158203, "learning_rate": 9.22917633028259e-06, "loss": 12.196146011352539, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2519594393670559, "loss_total_v6": 0.7622591108083725, "step": 1597 }, { "epoch": 0.36854774774774773, "grad_norm": 22.91717529296875, "learning_rate": 9.22810171526226e-06, "loss": 12.761463165283203, "loss_alignment": 1.02587890625, "loss_alignment_w": 0.512939453125, "loss_sft": 0.2847130000591278, "loss_total_v6": 0.7975913882255554, "step": 1598 }, { "epoch": 0.3687783783783784, "grad_norm": 20.042158126831055, "learning_rate": 9.227026414349574e-06, "loss": 12.385414123535156, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.26775604486465454, "loss_total_v6": 0.7740884348750114, "step": 1599 }, { "epoch": 0.369009009009009, "grad_norm": 25.10894775390625, "learning_rate": 9.225950427718974e-06, "loss": 12.551267623901367, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.27333058044314384, "loss_total_v6": 0.7844542264938354, "step": 1600 }, { "epoch": 0.3692396396396396, "grad_norm": 21.81214714050293, "learning_rate": 9.22487375554501e-06, "loss": 12.156479835510254, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.259612075984478, "loss_total_v6": 0.7597799524664879, "step": 1601 }, { "epoch": 0.3694702702702703, "grad_norm": 23.213703155517578, "learning_rate": 9.223796398002337e-06, "loss": 12.911624908447266, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.3031618669629097, "loss_total_v6": 0.8069765791296959, "step": 1602 }, { "epoch": 0.3697009009009009, "grad_norm": 22.49921989440918, "learning_rate": 9.222718355265733e-06, "loss": 12.429530143737793, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.2753350269049406, "loss_total_v6": 0.7768456488847733, "step": 1603 }, { "epoch": 0.3699315315315315, "grad_norm": 20.756336212158203, "learning_rate": 9.221639627510076e-06, "loss": 12.820242881774902, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.29714533872902393, "loss_total_v6": 0.8012652024626732, "step": 1604 }, { "epoch": 0.3701621621621622, "grad_norm": 22.639352798461914, "learning_rate": 9.220560214910362e-06, "loss": 12.129618644714355, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.2566058300435543, "loss_total_v6": 0.7581011652946472, "step": 1605 }, { "epoch": 0.3703927927927928, "grad_norm": 24.694074630737305, "learning_rate": 9.219480117641693e-06, "loss": 12.355340957641602, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.26404535584151745, "loss_total_v6": 0.7722088322043419, "step": 1606 }, { "epoch": 0.3706234234234234, "grad_norm": 25.374744415283203, "learning_rate": 9.21839933587929e-06, "loss": 12.935322761535645, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.3031171150505543, "loss_total_v6": 0.8084576725959778, "step": 1607 }, { "epoch": 0.370854054054054, "grad_norm": 25.17452049255371, "learning_rate": 9.217317869798473e-06, "loss": 12.300629615783691, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.2669887989759445, "loss_total_v6": 0.7687893435359001, "step": 1608 }, { "epoch": 0.3710846846846847, "grad_norm": 24.856252670288086, "learning_rate": 9.216235719574684e-06, "loss": 13.064538955688477, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.3123374842107296, "loss_total_v6": 0.8165336698293686, "step": 1609 }, { "epoch": 0.3713153153153153, "grad_norm": 22.372100830078125, "learning_rate": 9.215152885383473e-06, "loss": 13.291037559509277, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.31691112741827965, "loss_total_v6": 0.830689825117588, "step": 1610 }, { "epoch": 0.3715459459459459, "grad_norm": 22.31008529663086, "learning_rate": 9.214069367400497e-06, "loss": 12.903888702392578, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.30013011768460274, "loss_total_v6": 0.8064930438995361, "step": 1611 }, { "epoch": 0.3717765765765766, "grad_norm": 24.370424270629883, "learning_rate": 9.21298516580153e-06, "loss": 12.739432334899902, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.29996813274919987, "loss_total_v6": 0.7962144762277603, "step": 1612 }, { "epoch": 0.3720072072072072, "grad_norm": 24.98027229309082, "learning_rate": 9.21190028076245e-06, "loss": 12.512519836425781, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.26512574777007103, "loss_total_v6": 0.7820325195789337, "step": 1613 }, { "epoch": 0.3722378378378378, "grad_norm": 21.762969970703125, "learning_rate": 9.210814712459251e-06, "loss": 12.41698169708252, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.26362541504204273, "loss_total_v6": 0.7760613560676575, "step": 1614 }, { "epoch": 0.3724684684684685, "grad_norm": 23.20863151550293, "learning_rate": 9.209728461068037e-06, "loss": 12.809423446655273, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.28973996080458164, "loss_total_v6": 0.800588957965374, "step": 1615 }, { "epoch": 0.3726990990990991, "grad_norm": 22.599994659423828, "learning_rate": 9.208641526765024e-06, "loss": 12.639516830444336, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.29109890572726727, "loss_total_v6": 0.7899697571992874, "step": 1616 }, { "epoch": 0.3729297297297297, "grad_norm": 22.092525482177734, "learning_rate": 9.207553909726532e-06, "loss": 12.757146835327148, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.28519091196358204, "loss_total_v6": 0.7973216623067856, "step": 1617 }, { "epoch": 0.3731603603603604, "grad_norm": 23.67367935180664, "learning_rate": 9.206465610129e-06, "loss": 12.63461685180664, "loss_alignment": 1.01220703125, "loss_alignment_w": 0.506103515625, "loss_sft": 0.2823546454310417, "loss_total_v6": 0.7896635979413986, "step": 1618 }, { "epoch": 0.373390990990991, "grad_norm": 25.228160858154297, "learning_rate": 9.205376628148974e-06, "loss": 12.487384796142578, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.27008561603724957, "loss_total_v6": 0.7804615944623947, "step": 1619 }, { "epoch": 0.3736216216216216, "grad_norm": 24.412153244018555, "learning_rate": 9.204286963963112e-06, "loss": 12.71646785736084, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.29285662807524204, "loss_total_v6": 0.7947792410850525, "step": 1620 }, { "epoch": 0.37385225225225227, "grad_norm": 22.065265655517578, "learning_rate": 9.20319661774818e-06, "loss": 12.604496955871582, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.2608950585126877, "loss_total_v6": 0.7877810522913933, "step": 1621 }, { "epoch": 0.3740828828828829, "grad_norm": 23.426509857177734, "learning_rate": 9.202105589681058e-06, "loss": 12.573999404907227, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.26970063894987106, "loss_total_v6": 0.7858749628067017, "step": 1622 }, { "epoch": 0.3743135135135135, "grad_norm": 23.267427444458008, "learning_rate": 9.201013879938733e-06, "loss": 12.156536102294922, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.25270343013107777, "loss_total_v6": 0.7597835212945938, "step": 1623 }, { "epoch": 0.37454414414414416, "grad_norm": 22.49944305419922, "learning_rate": 9.199921488698307e-06, "loss": 13.093894958496094, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.30680228024721146, "loss_total_v6": 0.8183683902025223, "step": 1624 }, { "epoch": 0.3747747747747748, "grad_norm": 23.447410583496094, "learning_rate": 9.198828416136991e-06, "loss": 13.2103271484375, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.31317903473973274, "loss_total_v6": 0.8256454691290855, "step": 1625 }, { "epoch": 0.3750054054054054, "grad_norm": 22.572677612304688, "learning_rate": 9.197734662432101e-06, "loss": 12.73777961730957, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.29019091837108135, "loss_total_v6": 0.7961112856864929, "step": 1626 }, { "epoch": 0.37523603603603606, "grad_norm": 22.515512466430664, "learning_rate": 9.196640227761073e-06, "loss": 12.566340446472168, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.28162734024226665, "loss_total_v6": 0.7853962630033493, "step": 1627 }, { "epoch": 0.37546666666666667, "grad_norm": 21.058515548706055, "learning_rate": 9.195545112301448e-06, "loss": 12.890401840209961, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.2840283680707216, "loss_total_v6": 0.8056500852108002, "step": 1628 }, { "epoch": 0.3756972972972973, "grad_norm": 21.61559295654297, "learning_rate": 9.194449316230876e-06, "loss": 12.754310607910156, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.2917275261133909, "loss_total_v6": 0.7971444204449654, "step": 1629 }, { "epoch": 0.37592792792792795, "grad_norm": 21.373159408569336, "learning_rate": 9.193352839727122e-06, "loss": 12.67984390258789, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.28069520369172096, "loss_total_v6": 0.7924902588129044, "step": 1630 }, { "epoch": 0.37615855855855856, "grad_norm": 21.83163833618164, "learning_rate": 9.192255682968056e-06, "loss": 12.83754825592041, "loss_alignment": 1.01806640625, "loss_alignment_w": 0.509033203125, "loss_sft": 0.29306939989328384, "loss_total_v6": 0.8023467659950256, "step": 1631 }, { "epoch": 0.3763891891891892, "grad_norm": 22.655187606811523, "learning_rate": 9.191157846131662e-06, "loss": 12.425149917602539, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2629763428121805, "loss_total_v6": 0.7765719071030617, "step": 1632 }, { "epoch": 0.37661981981981985, "grad_norm": 22.563501358032227, "learning_rate": 9.190059329396038e-06, "loss": 12.212455749511719, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2547945659607649, "loss_total_v6": 0.7632784396409988, "step": 1633 }, { "epoch": 0.37685045045045046, "grad_norm": 22.398330688476562, "learning_rate": 9.18896013293938e-06, "loss": 12.26906967163086, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.26042340137064457, "loss_total_v6": 0.7668168544769287, "step": 1634 }, { "epoch": 0.37708108108108107, "grad_norm": 23.7537841796875, "learning_rate": 9.18786025694001e-06, "loss": 13.1105318069458, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.3075674343854189, "loss_total_v6": 0.8194082602858543, "step": 1635 }, { "epoch": 0.37731171171171174, "grad_norm": 24.192276000976562, "learning_rate": 9.18675970157635e-06, "loss": 12.948362350463867, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2959212176501751, "loss_total_v6": 0.8092726841568947, "step": 1636 }, { "epoch": 0.37754234234234235, "grad_norm": 24.757253646850586, "learning_rate": 9.18565846702693e-06, "loss": 12.123188018798828, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.2537166401743889, "loss_total_v6": 0.7576991990208626, "step": 1637 }, { "epoch": 0.37777297297297296, "grad_norm": 23.29766273498535, "learning_rate": 9.1845565534704e-06, "loss": 12.177507400512695, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2543650474399328, "loss_total_v6": 0.7610941827297211, "step": 1638 }, { "epoch": 0.3780036036036036, "grad_norm": 23.310749053955078, "learning_rate": 9.183453961085513e-06, "loss": 13.235260009765625, "loss_alignment": 1.01904296875, "loss_alignment_w": 0.509521484375, "loss_sft": 0.3173770532011986, "loss_total_v6": 0.8272037357091904, "step": 1639 }, { "epoch": 0.37823423423423425, "grad_norm": 24.444257736206055, "learning_rate": 9.182350690051134e-06, "loss": 11.829066276550293, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.23376240767538548, "loss_total_v6": 0.7393166050314903, "step": 1640 }, { "epoch": 0.37846486486486486, "grad_norm": 25.699569702148438, "learning_rate": 9.18124674054624e-06, "loss": 12.31667709350586, "loss_alignment": 1.01318359375, "loss_alignment_w": 0.506591796875, "loss_sft": 0.26288014464080334, "loss_total_v6": 0.7697923555970192, "step": 1641 }, { "epoch": 0.37869549549549547, "grad_norm": 25.699411392211914, "learning_rate": 9.180142112749915e-06, "loss": 12.81692886352539, "loss_alignment": 1.02099609375, "loss_alignment_w": 0.510498046875, "loss_sft": 0.2900869734585285, "loss_total_v6": 0.8010580614209175, "step": 1642 }, { "epoch": 0.37892612612612614, "grad_norm": 25.307695388793945, "learning_rate": 9.179036806841352e-06, "loss": 12.39195442199707, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.26897348277270794, "loss_total_v6": 0.7744971215724945, "step": 1643 }, { "epoch": 0.37915675675675675, "grad_norm": 24.66751480102539, "learning_rate": 9.17793082299986e-06, "loss": 12.330707550048828, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.260216960683465, "loss_total_v6": 0.770669236779213, "step": 1644 }, { "epoch": 0.37938738738738736, "grad_norm": 23.150352478027344, "learning_rate": 9.176824161404851e-06, "loss": 12.08030891418457, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.24668797478079796, "loss_total_v6": 0.7550192624330521, "step": 1645 }, { "epoch": 0.37961801801801803, "grad_norm": 23.44923210144043, "learning_rate": 9.175716822235854e-06, "loss": 12.34385871887207, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2639837861061096, "loss_total_v6": 0.7714911103248596, "step": 1646 }, { "epoch": 0.37984864864864865, "grad_norm": 22.843486785888672, "learning_rate": 9.1746088056725e-06, "loss": 12.394269943237305, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.26570018008351326, "loss_total_v6": 0.7746418416500092, "step": 1647 }, { "epoch": 0.38007927927927926, "grad_norm": 26.025354385375977, "learning_rate": 9.173500111894536e-06, "loss": 12.991240501403809, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.29536619782447815, "loss_total_v6": 0.8119525164365768, "step": 1648 }, { "epoch": 0.3803099099099099, "grad_norm": 25.454177856445312, "learning_rate": 9.172390741081817e-06, "loss": 12.886984825134277, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.29339735209941864, "loss_total_v6": 0.8054365590214729, "step": 1649 }, { "epoch": 0.38054054054054054, "grad_norm": 23.96070671081543, "learning_rate": 9.171280693414307e-06, "loss": 12.636234283447266, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.2868655286729336, "loss_total_v6": 0.7897646799683571, "step": 1650 }, { "epoch": 0.38077117117117115, "grad_norm": 22.011568069458008, "learning_rate": 9.170169969072081e-06, "loss": 12.413589477539062, "loss_alignment": 1.02099609375, "loss_alignment_w": 0.510498046875, "loss_sft": 0.2653971128165722, "loss_total_v6": 0.7758493646979332, "step": 1651 }, { "epoch": 0.3810018018018018, "grad_norm": 25.045358657836914, "learning_rate": 9.169058568235324e-06, "loss": 13.12961483001709, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.3102554380893707, "loss_total_v6": 0.8206009268760681, "step": 1652 }, { "epoch": 0.38123243243243243, "grad_norm": 26.21046257019043, "learning_rate": 9.167946491084327e-06, "loss": 12.627809524536133, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.28469096310436726, "loss_total_v6": 0.7892380952835083, "step": 1653 }, { "epoch": 0.38146306306306305, "grad_norm": 22.119943618774414, "learning_rate": 9.166833737799497e-06, "loss": 12.084729194641113, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.24298170022666454, "loss_total_v6": 0.755295567214489, "step": 1654 }, { "epoch": 0.3816936936936937, "grad_norm": 24.394235610961914, "learning_rate": 9.165720308561347e-06, "loss": 12.801956176757812, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.28060625679790974, "loss_total_v6": 0.8001222610473633, "step": 1655 }, { "epoch": 0.3819243243243243, "grad_norm": 24.758380889892578, "learning_rate": 9.164606203550498e-06, "loss": 12.600544929504395, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.27462511137127876, "loss_total_v6": 0.7875340580940247, "step": 1656 }, { "epoch": 0.38215495495495494, "grad_norm": 24.137666702270508, "learning_rate": 9.163491422947684e-06, "loss": 12.285832405090332, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.25989940389990807, "loss_total_v6": 0.7678645104169846, "step": 1657 }, { "epoch": 0.3823855855855856, "grad_norm": 24.03942108154297, "learning_rate": 9.162375966933748e-06, "loss": 12.866518020629883, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2937204074114561, "loss_total_v6": 0.8041574135422707, "step": 1658 }, { "epoch": 0.3826162162162162, "grad_norm": 23.594131469726562, "learning_rate": 9.16125983568964e-06, "loss": 12.257024765014648, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2581752836704254, "loss_total_v6": 0.7660640850663185, "step": 1659 }, { "epoch": 0.38284684684684683, "grad_norm": 23.3885555267334, "learning_rate": 9.160143029396422e-06, "loss": 12.635950088500977, "loss_alignment": 1.03076171875, "loss_alignment_w": 0.515380859375, "loss_sft": 0.27523573115468025, "loss_total_v6": 0.7897468283772469, "step": 1660 }, { "epoch": 0.3830774774774775, "grad_norm": 24.010353088378906, "learning_rate": 9.159025548235265e-06, "loss": 12.90562629699707, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.3068305142223835, "loss_total_v6": 0.8066016286611557, "step": 1661 }, { "epoch": 0.3833081081081081, "grad_norm": 25.775901794433594, "learning_rate": 9.15790739238745e-06, "loss": 12.826900482177734, "loss_alignment": 1.00732421875, "loss_alignment_w": 0.503662109375, "loss_sft": 0.29908727295696735, "loss_total_v6": 0.8016812726855278, "step": 1662 }, { "epoch": 0.3835387387387387, "grad_norm": 23.903398513793945, "learning_rate": 9.15678856203437e-06, "loss": 12.743916511535645, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.2988904342055321, "loss_total_v6": 0.796494796872139, "step": 1663 }, { "epoch": 0.3837693693693694, "grad_norm": 22.930164337158203, "learning_rate": 9.155669057357515e-06, "loss": 13.043128967285156, "loss_alignment": 1.01806640625, "loss_alignment_w": 0.509033203125, "loss_sft": 0.30504848435521126, "loss_total_v6": 0.8151955679059029, "step": 1664 }, { "epoch": 0.384, "grad_norm": 22.909345626831055, "learning_rate": 9.1545488785385e-06, "loss": 12.43644905090332, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.28319851867854595, "loss_total_v6": 0.7772780954837799, "step": 1665 }, { "epoch": 0.3842306306306306, "grad_norm": 20.555850982666016, "learning_rate": 9.153428025759046e-06, "loss": 12.888044357299805, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.3020085133612156, "loss_total_v6": 0.8055027574300766, "step": 1666 }, { "epoch": 0.3844612612612613, "grad_norm": 21.2141170501709, "learning_rate": 9.152306499200973e-06, "loss": 12.318525314331055, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.2629345506429672, "loss_total_v6": 0.7699078172445297, "step": 1667 }, { "epoch": 0.3846918918918919, "grad_norm": 22.497947692871094, "learning_rate": 9.151184299046221e-06, "loss": 12.552934646606445, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2687198091298342, "loss_total_v6": 0.7845584079623222, "step": 1668 }, { "epoch": 0.3849225225225225, "grad_norm": 25.69698143005371, "learning_rate": 9.150061425476839e-06, "loss": 12.936355590820312, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.29897022992372513, "loss_total_v6": 0.8085222318768501, "step": 1669 }, { "epoch": 0.3851531531531531, "grad_norm": 21.898622512817383, "learning_rate": 9.148937878674975e-06, "loss": 12.556886672973633, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.28546156734228134, "loss_total_v6": 0.7848054245114326, "step": 1670 }, { "epoch": 0.3853837837837838, "grad_norm": 21.700176239013672, "learning_rate": 9.147813658822902e-06, "loss": 12.197429656982422, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.2672679666429758, "loss_total_v6": 0.7623393535614014, "step": 1671 }, { "epoch": 0.3856144144144144, "grad_norm": 22.671173095703125, "learning_rate": 9.146688766102985e-06, "loss": 12.645284652709961, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.2883924338966608, "loss_total_v6": 0.7903302982449532, "step": 1672 }, { "epoch": 0.385845045045045, "grad_norm": 22.77350616455078, "learning_rate": 9.145563200697713e-06, "loss": 12.659219741821289, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.27646126598119736, "loss_total_v6": 0.7912012189626694, "step": 1673 }, { "epoch": 0.3860756756756757, "grad_norm": 25.053747177124023, "learning_rate": 9.144436962789675e-06, "loss": 13.103861808776855, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.304937981069088, "loss_total_v6": 0.8189913257956505, "step": 1674 }, { "epoch": 0.3863063063063063, "grad_norm": 22.59798812866211, "learning_rate": 9.14331005256157e-06, "loss": 12.102888107299805, "loss_alignment": 0.98291015625, "loss_alignment_w": 0.491455078125, "loss_sft": 0.26552470587193966, "loss_total_v6": 0.756430484354496, "step": 1675 }, { "epoch": 0.3865369369369369, "grad_norm": 20.943801879882812, "learning_rate": 9.142182470196213e-06, "loss": 12.275181770324707, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.25581582076847553, "loss_total_v6": 0.7671988531947136, "step": 1676 }, { "epoch": 0.3867675675675676, "grad_norm": 23.384761810302734, "learning_rate": 9.14105421587652e-06, "loss": 13.238441467285156, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.31539391726255417, "loss_total_v6": 0.8274026066064835, "step": 1677 }, { "epoch": 0.3869981981981982, "grad_norm": 19.819787979125977, "learning_rate": 9.139925289785518e-06, "loss": 12.37125301361084, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2679085433483124, "loss_total_v6": 0.7732033431529999, "step": 1678 }, { "epoch": 0.3872288288288288, "grad_norm": 22.0500545501709, "learning_rate": 9.138795692106346e-06, "loss": 13.487703323364258, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.3318578228354454, "loss_total_v6": 0.8429814800620079, "step": 1679 }, { "epoch": 0.3874594594594595, "grad_norm": 23.774429321289062, "learning_rate": 9.13766542302225e-06, "loss": 12.691425323486328, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.27778744511306286, "loss_total_v6": 0.7932141050696373, "step": 1680 }, { "epoch": 0.3876900900900901, "grad_norm": 20.776290893554688, "learning_rate": 9.136534482716583e-06, "loss": 12.636067390441895, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2856191173195839, "loss_total_v6": 0.7897542342543602, "step": 1681 }, { "epoch": 0.3879207207207207, "grad_norm": 22.616682052612305, "learning_rate": 9.13540287137281e-06, "loss": 12.541438102722168, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.27772113122045994, "loss_total_v6": 0.7838398814201355, "step": 1682 }, { "epoch": 0.38815135135135137, "grad_norm": 21.586801528930664, "learning_rate": 9.134270589174503e-06, "loss": 12.477804183959961, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.26435986906290054, "loss_total_v6": 0.77986279129982, "step": 1683 }, { "epoch": 0.388381981981982, "grad_norm": 21.525882720947266, "learning_rate": 9.133137636305345e-06, "loss": 12.677309036254883, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2809487897902727, "loss_total_v6": 0.792331837117672, "step": 1684 }, { "epoch": 0.3886126126126126, "grad_norm": 26.660200119018555, "learning_rate": 9.132004012949124e-06, "loss": 12.434808731079102, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.26657070964574814, "loss_total_v6": 0.7771755456924438, "step": 1685 }, { "epoch": 0.38884324324324326, "grad_norm": 21.76096534729004, "learning_rate": 9.13086971928974e-06, "loss": 12.272745132446289, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.26375071704387665, "loss_total_v6": 0.7670465931296349, "step": 1686 }, { "epoch": 0.3890738738738739, "grad_norm": 24.086946487426758, "learning_rate": 9.1297347555112e-06, "loss": 13.109183311462402, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.30331752076745033, "loss_total_v6": 0.8193239644169807, "step": 1687 }, { "epoch": 0.3893045045045045, "grad_norm": 23.014116287231445, "learning_rate": 9.12859912179762e-06, "loss": 12.51852798461914, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.27563304640352726, "loss_total_v6": 0.7824079841375351, "step": 1688 }, { "epoch": 0.38953513513513516, "grad_norm": 23.592449188232422, "learning_rate": 9.127462818333231e-06, "loss": 12.647554397583008, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.2888089753687382, "loss_total_v6": 0.790472187101841, "step": 1689 }, { "epoch": 0.38976576576576577, "grad_norm": 25.190526962280273, "learning_rate": 9.126325845302357e-06, "loss": 12.693563461303711, "loss_alignment": 1.03173828125, "loss_alignment_w": 0.515869140625, "loss_sft": 0.2770360875874758, "loss_total_v6": 0.7933477535843849, "step": 1690 }, { "epoch": 0.3899963963963964, "grad_norm": 22.8938045501709, "learning_rate": 9.125188202889449e-06, "loss": 12.24587631225586, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.25216840393841267, "loss_total_v6": 0.7653673067688942, "step": 1691 }, { "epoch": 0.39022702702702705, "grad_norm": 24.374998092651367, "learning_rate": 9.124049891279052e-06, "loss": 12.961628913879395, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.29472095146775246, "loss_total_v6": 0.8101018071174622, "step": 1692 }, { "epoch": 0.39045765765765766, "grad_norm": 22.728166580200195, "learning_rate": 9.122910910655827e-06, "loss": 12.553644180297852, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.27343334443867207, "loss_total_v6": 0.7846027761697769, "step": 1693 }, { "epoch": 0.3906882882882883, "grad_norm": 24.885879516601562, "learning_rate": 9.121771261204544e-06, "loss": 13.409255981445312, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.3259020522236824, "loss_total_v6": 0.838078536093235, "step": 1694 }, { "epoch": 0.39091891891891895, "grad_norm": 23.449071884155273, "learning_rate": 9.120630943110078e-06, "loss": 12.538407325744629, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2812548838555813, "loss_total_v6": 0.7836504653096199, "step": 1695 }, { "epoch": 0.39114954954954956, "grad_norm": 26.43948745727539, "learning_rate": 9.119489956557416e-06, "loss": 12.730676651000977, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2919593648985028, "loss_total_v6": 0.7956672459840775, "step": 1696 }, { "epoch": 0.39138018018018017, "grad_norm": 22.677125930786133, "learning_rate": 9.118348301731649e-06, "loss": 12.753583908081055, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.2769116014242172, "loss_total_v6": 0.7970990017056465, "step": 1697 }, { "epoch": 0.3916108108108108, "grad_norm": 22.40236473083496, "learning_rate": 9.117205978817979e-06, "loss": 12.403639793395996, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.27055829763412476, "loss_total_v6": 0.7752274945378304, "step": 1698 }, { "epoch": 0.39184144144144145, "grad_norm": 22.221389770507812, "learning_rate": 9.116062988001717e-06, "loss": 12.572463035583496, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2743043154478073, "loss_total_v6": 0.7857789471745491, "step": 1699 }, { "epoch": 0.39207207207207206, "grad_norm": 25.51910972595215, "learning_rate": 9.114919329468283e-06, "loss": 12.70818042755127, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.28196271881461143, "loss_total_v6": 0.7942612618207932, "step": 1700 }, { "epoch": 0.3923027027027027, "grad_norm": 23.52404022216797, "learning_rate": 9.113775003403201e-06, "loss": 12.886831283569336, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.29434906877577305, "loss_total_v6": 0.8054269403219223, "step": 1701 }, { "epoch": 0.39253333333333335, "grad_norm": 22.317184448242188, "learning_rate": 9.112630009992109e-06, "loss": 12.843585014343262, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.29379767179489136, "loss_total_v6": 0.8027240559458733, "step": 1702 }, { "epoch": 0.39276396396396396, "grad_norm": 22.637802124023438, "learning_rate": 9.111484349420747e-06, "loss": 13.08907699584961, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.3046700619161129, "loss_total_v6": 0.8180672824382782, "step": 1703 }, { "epoch": 0.39299459459459457, "grad_norm": 25.741302490234375, "learning_rate": 9.11033802187497e-06, "loss": 13.087324142456055, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.303660299628973, "loss_total_v6": 0.8179577812552452, "step": 1704 }, { "epoch": 0.39322522522522524, "grad_norm": 25.945964813232422, "learning_rate": 9.109191027540739e-06, "loss": 12.787294387817383, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.28678525425493717, "loss_total_v6": 0.7992058917880058, "step": 1705 }, { "epoch": 0.39345585585585585, "grad_norm": 22.608564376831055, "learning_rate": 9.108043366604116e-06, "loss": 12.782340049743652, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.28977150097489357, "loss_total_v6": 0.7988962829113007, "step": 1706 }, { "epoch": 0.39368648648648646, "grad_norm": 26.881155014038086, "learning_rate": 9.106895039251282e-06, "loss": 12.583602905273438, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2711249142885208, "loss_total_v6": 0.786475233733654, "step": 1707 }, { "epoch": 0.39391711711711713, "grad_norm": 22.498756408691406, "learning_rate": 9.10574604566852e-06, "loss": 13.15035629272461, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.3190285563468933, "loss_total_v6": 0.8218972235918045, "step": 1708 }, { "epoch": 0.39414774774774775, "grad_norm": 22.73251724243164, "learning_rate": 9.104596386042223e-06, "loss": 12.54858684539795, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2767335530370474, "loss_total_v6": 0.78428665548563, "step": 1709 }, { "epoch": 0.39437837837837836, "grad_norm": 21.86678123474121, "learning_rate": 9.103446060558892e-06, "loss": 12.744611740112305, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2918232288211584, "loss_total_v6": 0.7965382039546967, "step": 1710 }, { "epoch": 0.394609009009009, "grad_norm": 22.916912078857422, "learning_rate": 9.102295069405133e-06, "loss": 13.13858413696289, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.31391362473368645, "loss_total_v6": 0.8211615085601807, "step": 1711 }, { "epoch": 0.39483963963963964, "grad_norm": 23.544462203979492, "learning_rate": 9.101143412767665e-06, "loss": 12.301322937011719, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2622713800519705, "loss_total_v6": 0.7688326686620712, "step": 1712 }, { "epoch": 0.39507027027027025, "grad_norm": 22.227073669433594, "learning_rate": 9.09999109083331e-06, "loss": 12.459577560424805, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.27941026724874973, "loss_total_v6": 0.7787236198782921, "step": 1713 }, { "epoch": 0.3953009009009009, "grad_norm": 23.701923370361328, "learning_rate": 9.098838103789004e-06, "loss": 12.333763122558594, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2639327719807625, "loss_total_v6": 0.7708602398633957, "step": 1714 }, { "epoch": 0.39553153153153153, "grad_norm": 22.543977737426758, "learning_rate": 9.097684451821783e-06, "loss": 12.400814056396484, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.26327106542885303, "loss_total_v6": 0.7750508412718773, "step": 1715 }, { "epoch": 0.39576216216216215, "grad_norm": 22.980783462524414, "learning_rate": 9.096530135118798e-06, "loss": 12.900981903076172, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2984988298267126, "loss_total_v6": 0.8063113540410995, "step": 1716 }, { "epoch": 0.3959927927927928, "grad_norm": 22.07779312133789, "learning_rate": 9.095375153867302e-06, "loss": 12.348518371582031, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2645649667829275, "loss_total_v6": 0.7717823833227158, "step": 1717 }, { "epoch": 0.3962234234234234, "grad_norm": 24.013357162475586, "learning_rate": 9.09421950825466e-06, "loss": 12.6539888381958, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.27912501618266106, "loss_total_v6": 0.7908743023872375, "step": 1718 }, { "epoch": 0.39645405405405404, "grad_norm": 21.63901710510254, "learning_rate": 9.093063198468346e-06, "loss": 12.70136547088623, "loss_alignment": 1.02197265625, "loss_alignment_w": 0.510986328125, "loss_sft": 0.2836424596607685, "loss_total_v6": 0.7938353419303894, "step": 1719 }, { "epoch": 0.3966846846846847, "grad_norm": 25.887353897094727, "learning_rate": 9.091906224695935e-06, "loss": 13.200230598449707, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.3163626454770565, "loss_total_v6": 0.8250143900513649, "step": 1720 }, { "epoch": 0.3969153153153153, "grad_norm": 22.632165908813477, "learning_rate": 9.090748587125118e-06, "loss": 12.27983283996582, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.26556691341102123, "loss_total_v6": 0.7674895152449608, "step": 1721 }, { "epoch": 0.39714594594594593, "grad_norm": 20.79855728149414, "learning_rate": 9.089590285943687e-06, "loss": 12.907968521118164, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.2871710453182459, "loss_total_v6": 0.8067480772733688, "step": 1722 }, { "epoch": 0.3973765765765766, "grad_norm": 20.485515594482422, "learning_rate": 9.088431321339546e-06, "loss": 12.419544219970703, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.26695949770510197, "loss_total_v6": 0.7762215435504913, "step": 1723 }, { "epoch": 0.3976072072072072, "grad_norm": 22.145729064941406, "learning_rate": 9.087271693500701e-06, "loss": 12.276470184326172, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.25668979436159134, "loss_total_v6": 0.7672794014215469, "step": 1724 }, { "epoch": 0.3978378378378378, "grad_norm": 22.67620086669922, "learning_rate": 9.086111402615274e-06, "loss": 12.962903022766113, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2948768436908722, "loss_total_v6": 0.8101814165711403, "step": 1725 }, { "epoch": 0.3980684684684685, "grad_norm": 22.021337509155273, "learning_rate": 9.084950448871489e-06, "loss": 12.439896583557129, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.26925379037857056, "loss_total_v6": 0.7774935588240623, "step": 1726 }, { "epoch": 0.3982990990990991, "grad_norm": 23.001766204833984, "learning_rate": 9.083788832457677e-06, "loss": 12.608463287353516, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.2863351944833994, "loss_total_v6": 0.7880289405584335, "step": 1727 }, { "epoch": 0.3985297297297297, "grad_norm": 23.65790367126465, "learning_rate": 9.08262655356228e-06, "loss": 12.440006256103516, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.26915382593870163, "loss_total_v6": 0.7775003537535667, "step": 1728 }, { "epoch": 0.39876036036036033, "grad_norm": 23.756969451904297, "learning_rate": 9.081463612373845e-06, "loss": 12.545701026916504, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2748137544840574, "loss_total_v6": 0.7841063290834427, "step": 1729 }, { "epoch": 0.398990990990991, "grad_norm": 23.45423698425293, "learning_rate": 9.080300009081025e-06, "loss": 12.6107177734375, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.27584074437618256, "loss_total_v6": 0.7881698533892632, "step": 1730 }, { "epoch": 0.3992216216216216, "grad_norm": 22.665178298950195, "learning_rate": 9.079135743872584e-06, "loss": 12.62660026550293, "loss_alignment": 1.01708984375, "loss_alignment_w": 0.508544921875, "loss_sft": 0.2805413752794266, "loss_total_v6": 0.7891625463962555, "step": 1731 }, { "epoch": 0.3994522522522522, "grad_norm": 21.607929229736328, "learning_rate": 9.077970816937394e-06, "loss": 12.302375793457031, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.26017044112086296, "loss_total_v6": 0.768898494541645, "step": 1732 }, { "epoch": 0.3996828828828829, "grad_norm": 22.352060317993164, "learning_rate": 9.076805228464426e-06, "loss": 12.628145217895508, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.27908141911029816, "loss_total_v6": 0.7892590314149857, "step": 1733 }, { "epoch": 0.3999135135135135, "grad_norm": 23.094074249267578, "learning_rate": 9.07563897864277e-06, "loss": 12.51006031036377, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2733185701072216, "loss_total_v6": 0.7818787768483162, "step": 1734 }, { "epoch": 0.4001441441441441, "grad_norm": 22.46261978149414, "learning_rate": 9.07447206766162e-06, "loss": 12.336530685424805, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2654331810772419, "loss_total_v6": 0.7710331603884697, "step": 1735 }, { "epoch": 0.4003747747747748, "grad_norm": 22.76926612854004, "learning_rate": 9.073304495710267e-06, "loss": 13.19624137878418, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.30874336138367653, "loss_total_v6": 0.82476507127285, "step": 1736 }, { "epoch": 0.4006054054054054, "grad_norm": 21.367097854614258, "learning_rate": 9.07213626297812e-06, "loss": 12.320816993713379, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.2668467089533806, "loss_total_v6": 0.7700510546565056, "step": 1737 }, { "epoch": 0.400836036036036, "grad_norm": 21.59682273864746, "learning_rate": 9.070967369654697e-06, "loss": 12.443267822265625, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.2753085754811764, "loss_total_v6": 0.7777042239904404, "step": 1738 }, { "epoch": 0.4010666666666667, "grad_norm": 26.66511344909668, "learning_rate": 9.069797815929614e-06, "loss": 13.21410083770752, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.32063228264451027, "loss_total_v6": 0.8258813098073006, "step": 1739 }, { "epoch": 0.4012972972972973, "grad_norm": 23.512569427490234, "learning_rate": 9.0686276019926e-06, "loss": 12.410722732543945, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.2737323436886072, "loss_total_v6": 0.7756701856851578, "step": 1740 }, { "epoch": 0.4015279279279279, "grad_norm": 24.17647933959961, "learning_rate": 9.067456728033487e-06, "loss": 12.976120948791504, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.30203537084162235, "loss_total_v6": 0.8110075816512108, "step": 1741 }, { "epoch": 0.4017585585585586, "grad_norm": 23.0234375, "learning_rate": 9.066285194242221e-06, "loss": 12.745118141174316, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2887573689222336, "loss_total_v6": 0.796569861471653, "step": 1742 }, { "epoch": 0.4019891891891892, "grad_norm": 22.8078556060791, "learning_rate": 9.06511300080885e-06, "loss": 12.49972915649414, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.27805921621620655, "loss_total_v6": 0.781233049929142, "step": 1743 }, { "epoch": 0.4022198198198198, "grad_norm": 22.963985443115234, "learning_rate": 9.063940147923529e-06, "loss": 12.520734786987305, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.28024186566472054, "loss_total_v6": 0.7825459390878677, "step": 1744 }, { "epoch": 0.40245045045045047, "grad_norm": 21.929580688476562, "learning_rate": 9.062766635776523e-06, "loss": 12.492507934570312, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.28268907591700554, "loss_total_v6": 0.7807817310094833, "step": 1745 }, { "epoch": 0.4026810810810811, "grad_norm": 22.24260139465332, "learning_rate": 9.061592464558198e-06, "loss": 12.518211364746094, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.27150869742035866, "loss_total_v6": 0.782388225197792, "step": 1746 }, { "epoch": 0.4029117117117117, "grad_norm": 20.134445190429688, "learning_rate": 9.060417634459032e-06, "loss": 12.145566940307617, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.24615848250687122, "loss_total_v6": 0.7590979263186455, "step": 1747 }, { "epoch": 0.40314234234234236, "grad_norm": 26.757810592651367, "learning_rate": 9.05924214566961e-06, "loss": 12.81777286529541, "loss_alignment": 1.01318359375, "loss_alignment_w": 0.506591796875, "loss_sft": 0.294061241671443, "loss_total_v6": 0.8011107817292213, "step": 1748 }, { "epoch": 0.403372972972973, "grad_norm": 20.996417999267578, "learning_rate": 9.058065998380623e-06, "loss": 12.08828353881836, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2520996890962124, "loss_total_v6": 0.7555176839232445, "step": 1749 }, { "epoch": 0.4036036036036036, "grad_norm": 23.319604873657227, "learning_rate": 9.056889192782865e-06, "loss": 12.735618591308594, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.29043720848858356, "loss_total_v6": 0.7959761470556259, "step": 1750 }, { "epoch": 0.40383423423423426, "grad_norm": 21.92844009399414, "learning_rate": 9.055711729067245e-06, "loss": 12.796844482421875, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.28944206051528454, "loss_total_v6": 0.7998027950525284, "step": 1751 }, { "epoch": 0.40406486486486487, "grad_norm": 26.46289825439453, "learning_rate": 9.05453360742477e-06, "loss": 12.570732116699219, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.27949094772338867, "loss_total_v6": 0.7856707572937012, "step": 1752 }, { "epoch": 0.4042954954954955, "grad_norm": 24.79344940185547, "learning_rate": 9.053354828046558e-06, "loss": 12.489261627197266, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.270645322278142, "loss_total_v6": 0.7805788218975067, "step": 1753 }, { "epoch": 0.40452612612612615, "grad_norm": 22.032360076904297, "learning_rate": 9.052175391123834e-06, "loss": 12.544734954833984, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.27527223713696003, "loss_total_v6": 0.7840459868311882, "step": 1754 }, { "epoch": 0.40475675675675676, "grad_norm": 21.679622650146484, "learning_rate": 9.05099529684793e-06, "loss": 11.951227188110352, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.24069556407630444, "loss_total_v6": 0.7469516545534134, "step": 1755 }, { "epoch": 0.4049873873873874, "grad_norm": 26.54281234741211, "learning_rate": 9.049814545410282e-06, "loss": 12.73360824584961, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.28582545183598995, "loss_total_v6": 0.7958504781126976, "step": 1756 }, { "epoch": 0.40521801801801804, "grad_norm": 22.92696762084961, "learning_rate": 9.048633137002435e-06, "loss": 12.868800163269043, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.29737250693142414, "loss_total_v6": 0.8043000102043152, "step": 1757 }, { "epoch": 0.40544864864864866, "grad_norm": 22.514480590820312, "learning_rate": 9.047451071816038e-06, "loss": 12.845123291015625, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.28910258412361145, "loss_total_v6": 0.8028202205896378, "step": 1758 }, { "epoch": 0.40567927927927927, "grad_norm": 24.849454879760742, "learning_rate": 9.046268350042851e-06, "loss": 12.392300605773926, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.2719858605414629, "loss_total_v6": 0.7745187878608704, "step": 1759 }, { "epoch": 0.4059099099099099, "grad_norm": 23.05341339111328, "learning_rate": 9.045084971874738e-06, "loss": 12.667067527770996, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.27910325676202774, "loss_total_v6": 0.791691742837429, "step": 1760 }, { "epoch": 0.40614054054054055, "grad_norm": 24.84390640258789, "learning_rate": 9.043900937503667e-06, "loss": 12.292610168457031, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.26378677040338516, "loss_total_v6": 0.7682881131768227, "step": 1761 }, { "epoch": 0.40637117117117116, "grad_norm": 26.103620529174805, "learning_rate": 9.042716247121716e-06, "loss": 12.77280044555664, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2902738843113184, "loss_total_v6": 0.7982999980449677, "step": 1762 }, { "epoch": 0.4066018018018018, "grad_norm": 24.502296447753906, "learning_rate": 9.041530900921067e-06, "loss": 12.609888076782227, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.28521886467933655, "loss_total_v6": 0.7881180495023727, "step": 1763 }, { "epoch": 0.40683243243243244, "grad_norm": 23.405221939086914, "learning_rate": 9.04034489909401e-06, "loss": 12.61361312866211, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.27211546152830124, "loss_total_v6": 0.7883508279919624, "step": 1764 }, { "epoch": 0.40706306306306306, "grad_norm": 23.13076400756836, "learning_rate": 9.039158241832944e-06, "loss": 13.004558563232422, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.3031718749552965, "loss_total_v6": 0.8127849102020264, "step": 1765 }, { "epoch": 0.40729369369369367, "grad_norm": 26.346485137939453, "learning_rate": 9.037970929330369e-06, "loss": 12.652957916259766, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2851793635636568, "loss_total_v6": 0.7908098474144936, "step": 1766 }, { "epoch": 0.40752432432432434, "grad_norm": 23.090576171875, "learning_rate": 9.036782961778892e-06, "loss": 12.801172256469727, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.29126888513565063, "loss_total_v6": 0.8000732138752937, "step": 1767 }, { "epoch": 0.40775495495495495, "grad_norm": 22.942218780517578, "learning_rate": 9.035594339371229e-06, "loss": 13.123645782470703, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.31182025000452995, "loss_total_v6": 0.8202278390526772, "step": 1768 }, { "epoch": 0.40798558558558556, "grad_norm": 25.379558563232422, "learning_rate": 9.034405062300201e-06, "loss": 12.089408874511719, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.2568697761744261, "loss_total_v6": 0.755588062107563, "step": 1769 }, { "epoch": 0.40821621621621623, "grad_norm": 21.66939926147461, "learning_rate": 9.033215130758735e-06, "loss": 12.447644233703613, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2690056152641773, "loss_total_v6": 0.7779777646064758, "step": 1770 }, { "epoch": 0.40844684684684684, "grad_norm": 24.587919235229492, "learning_rate": 9.032024544939866e-06, "loss": 13.3156099319458, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.31374718993902206, "loss_total_v6": 0.8322256058454514, "step": 1771 }, { "epoch": 0.40867747747747746, "grad_norm": 22.676362991333008, "learning_rate": 9.030833305036732e-06, "loss": 12.55698299407959, "loss_alignment": 0.9873046875, "loss_alignment_w": 0.49365234375, "loss_sft": 0.29117434844374657, "loss_total_v6": 0.7848114296793938, "step": 1772 }, { "epoch": 0.4089081081081081, "grad_norm": 24.679288864135742, "learning_rate": 9.02964141124258e-06, "loss": 13.006948471069336, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.3014444727450609, "loss_total_v6": 0.8129343315958977, "step": 1773 }, { "epoch": 0.40913873873873874, "grad_norm": 22.546199798583984, "learning_rate": 9.02844886375076e-06, "loss": 12.384374618530273, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2603973113000393, "loss_total_v6": 0.7740234211087227, "step": 1774 }, { "epoch": 0.40936936936936935, "grad_norm": 22.336288452148438, "learning_rate": 9.02725566275473e-06, "loss": 13.07868766784668, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.31355753540992737, "loss_total_v6": 0.8174180090427399, "step": 1775 }, { "epoch": 0.4096, "grad_norm": 22.497650146484375, "learning_rate": 9.026061808448055e-06, "loss": 12.38679313659668, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.2711380198597908, "loss_total_v6": 0.7741745337843895, "step": 1776 }, { "epoch": 0.40983063063063063, "grad_norm": 22.19707679748535, "learning_rate": 9.024867301024405e-06, "loss": 12.313468933105469, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2613368071615696, "loss_total_v6": 0.7695917934179306, "step": 1777 }, { "epoch": 0.41006126126126125, "grad_norm": 27.549259185791016, "learning_rate": 9.023672140677556e-06, "loss": 12.6539306640625, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2811814174056053, "loss_total_v6": 0.7908707186579704, "step": 1778 }, { "epoch": 0.4102918918918919, "grad_norm": 25.164457321166992, "learning_rate": 9.022476327601385e-06, "loss": 13.302183151245117, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.319835539907217, "loss_total_v6": 0.831386461853981, "step": 1779 }, { "epoch": 0.4105225225225225, "grad_norm": 22.48068618774414, "learning_rate": 9.021279861989884e-06, "loss": 12.695520401000977, "loss_alignment": 1.01708984375, "loss_alignment_w": 0.508544921875, "loss_sft": 0.2844367790967226, "loss_total_v6": 0.7934699803590775, "step": 1780 }, { "epoch": 0.41075315315315314, "grad_norm": 23.45261001586914, "learning_rate": 9.020082744037147e-06, "loss": 12.957040786743164, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.28695744276046753, "loss_total_v6": 0.8098150938749313, "step": 1781 }, { "epoch": 0.4109837837837838, "grad_norm": 22.643007278442383, "learning_rate": 9.018884973937372e-06, "loss": 13.150918960571289, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.30267584323883057, "loss_total_v6": 0.821932427585125, "step": 1782 }, { "epoch": 0.4112144144144144, "grad_norm": 23.570484161376953, "learning_rate": 9.017686551884864e-06, "loss": 12.678403854370117, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2851370647549629, "loss_total_v6": 0.7924002408981323, "step": 1783 }, { "epoch": 0.41144504504504503, "grad_norm": 26.480195999145508, "learning_rate": 9.016487478074032e-06, "loss": 12.488265991210938, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.2821950353682041, "loss_total_v6": 0.7805165722966194, "step": 1784 }, { "epoch": 0.4116756756756757, "grad_norm": 22.342044830322266, "learning_rate": 9.015287752699395e-06, "loss": 12.122709274291992, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.25745575316250324, "loss_total_v6": 0.7576693668961525, "step": 1785 }, { "epoch": 0.4119063063063063, "grad_norm": 21.047691345214844, "learning_rate": 9.014087375955574e-06, "loss": 12.00732421875, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.2508849762380123, "loss_total_v6": 0.7504577189683914, "step": 1786 }, { "epoch": 0.4121369369369369, "grad_norm": 23.732402801513672, "learning_rate": 9.012886348037296e-06, "loss": 13.199161529541016, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.30688123777508736, "loss_total_v6": 0.8249476179480553, "step": 1787 }, { "epoch": 0.4123675675675676, "grad_norm": 23.155858993530273, "learning_rate": 9.011684669139398e-06, "loss": 12.53018856048584, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.28296894021332264, "loss_total_v6": 0.7831367775797844, "step": 1788 }, { "epoch": 0.4125981981981982, "grad_norm": 23.016204833984375, "learning_rate": 9.010482339456814e-06, "loss": 12.379642486572266, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.2762148082256317, "loss_total_v6": 0.7737276554107666, "step": 1789 }, { "epoch": 0.4128288288288288, "grad_norm": 26.730161666870117, "learning_rate": 9.009279359184594e-06, "loss": 12.78456974029541, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2958312872797251, "loss_total_v6": 0.7990356236696243, "step": 1790 }, { "epoch": 0.41305945945945943, "grad_norm": 23.06850242614746, "learning_rate": 9.008075728517884e-06, "loss": 12.522865295410156, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.27350859716534615, "loss_total_v6": 0.782679095864296, "step": 1791 }, { "epoch": 0.4132900900900901, "grad_norm": 23.19362449645996, "learning_rate": 9.006871447651941e-06, "loss": 12.683945655822754, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.28710082545876503, "loss_total_v6": 0.7927465811371803, "step": 1792 }, { "epoch": 0.4135207207207207, "grad_norm": 25.08515739440918, "learning_rate": 9.005666516782127e-06, "loss": 12.403290748596191, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.2769299354404211, "loss_total_v6": 0.7752057015895844, "step": 1793 }, { "epoch": 0.4137513513513513, "grad_norm": 24.340707778930664, "learning_rate": 9.004460936103909e-06, "loss": 12.270809173583984, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.26440783590078354, "loss_total_v6": 0.766925536096096, "step": 1794 }, { "epoch": 0.413981981981982, "grad_norm": 23.099140167236328, "learning_rate": 9.003254705812858e-06, "loss": 12.112875938415527, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.25281281396746635, "loss_total_v6": 0.7570547685027122, "step": 1795 }, { "epoch": 0.4142126126126126, "grad_norm": 22.591238021850586, "learning_rate": 9.002047826104652e-06, "loss": 12.729901313781738, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.28464774787425995, "loss_total_v6": 0.7956188321113586, "step": 1796 }, { "epoch": 0.4144432432432432, "grad_norm": 23.21276092529297, "learning_rate": 9.000840297175072e-06, "loss": 12.311031341552734, "loss_alignment": 0.9873046875, "loss_alignment_w": 0.49365234375, "loss_sft": 0.27452065609395504, "loss_total_v6": 0.7694394737482071, "step": 1797 }, { "epoch": 0.4146738738738739, "grad_norm": 22.538331985473633, "learning_rate": 8.999632119220008e-06, "loss": 13.027873039245605, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.3045374918729067, "loss_total_v6": 0.8142420575022697, "step": 1798 }, { "epoch": 0.4149045045045045, "grad_norm": 20.924102783203125, "learning_rate": 8.998423292435455e-06, "loss": 12.183174133300781, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2504010070115328, "loss_total_v6": 0.7614484056830406, "step": 1799 }, { "epoch": 0.4151351351351351, "grad_norm": 22.175384521484375, "learning_rate": 8.997213817017508e-06, "loss": 12.27243423461914, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.26101524010300636, "loss_total_v6": 0.7670271918177605, "step": 1800 }, { "epoch": 0.4153657657657658, "grad_norm": 22.927021026611328, "learning_rate": 8.996003693162372e-06, "loss": 12.447494506835938, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.2737722434103489, "loss_total_v6": 0.7779684066772461, "step": 1801 }, { "epoch": 0.4155963963963964, "grad_norm": 25.051753997802734, "learning_rate": 8.994792921066356e-06, "loss": 12.504793167114258, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.2842046171426773, "loss_total_v6": 0.7815495729446411, "step": 1802 }, { "epoch": 0.415827027027027, "grad_norm": 24.822587966918945, "learning_rate": 8.993581500925877e-06, "loss": 12.532621383666992, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.2807710785418749, "loss_total_v6": 0.7832888066768646, "step": 1803 }, { "epoch": 0.4160576576576577, "grad_norm": 23.8215389251709, "learning_rate": 8.992369432937451e-06, "loss": 11.995323181152344, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.24709846638143063, "loss_total_v6": 0.7497077211737633, "step": 1804 }, { "epoch": 0.4162882882882883, "grad_norm": 22.813758850097656, "learning_rate": 8.991156717297702e-06, "loss": 12.58023452758789, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.27453064545989037, "loss_total_v6": 0.7862646356225014, "step": 1805 }, { "epoch": 0.4165189189189189, "grad_norm": 22.232093811035156, "learning_rate": 8.989943354203364e-06, "loss": 13.126819610595703, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.3134682346135378, "loss_total_v6": 0.820426233112812, "step": 1806 }, { "epoch": 0.41674954954954957, "grad_norm": 22.946413040161133, "learning_rate": 8.988729343851268e-06, "loss": 12.33935546875, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2606506012380123, "loss_total_v6": 0.7712097242474556, "step": 1807 }, { "epoch": 0.4169801801801802, "grad_norm": 21.683422088623047, "learning_rate": 8.987514686438353e-06, "loss": 12.576132774353027, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.27680722437798977, "loss_total_v6": 0.786008283495903, "step": 1808 }, { "epoch": 0.4172108108108108, "grad_norm": 24.046615600585938, "learning_rate": 8.986299382161665e-06, "loss": 12.793352127075195, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2959987074136734, "loss_total_v6": 0.7995845228433609, "step": 1809 }, { "epoch": 0.41744144144144146, "grad_norm": 23.930156707763672, "learning_rate": 8.985083431218354e-06, "loss": 12.005533218383789, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.246927872300148, "loss_total_v6": 0.7503458559513092, "step": 1810 }, { "epoch": 0.4176720720720721, "grad_norm": 21.828659057617188, "learning_rate": 8.983866833805673e-06, "loss": 12.328733444213867, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.26660904474556446, "loss_total_v6": 0.7705457881093025, "step": 1811 }, { "epoch": 0.4179027027027027, "grad_norm": 23.35115623474121, "learning_rate": 8.982649590120982e-06, "loss": 12.606023788452148, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2794536966830492, "loss_total_v6": 0.7878765240311623, "step": 1812 }, { "epoch": 0.41813333333333336, "grad_norm": 23.63296127319336, "learning_rate": 8.981431700361745e-06, "loss": 12.444812774658203, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.2746880482882261, "loss_total_v6": 0.7778008058667183, "step": 1813 }, { "epoch": 0.41836396396396397, "grad_norm": 21.15829849243164, "learning_rate": 8.980213164725531e-06, "loss": 12.261556625366211, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2607625834643841, "loss_total_v6": 0.7663473039865494, "step": 1814 }, { "epoch": 0.4185945945945946, "grad_norm": 23.717945098876953, "learning_rate": 8.978993983410011e-06, "loss": 12.46988296508789, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.282404150813818, "loss_total_v6": 0.779367670416832, "step": 1815 }, { "epoch": 0.41882522522522525, "grad_norm": 24.632265090942383, "learning_rate": 8.977774156612968e-06, "loss": 12.522331237792969, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2768931817263365, "loss_total_v6": 0.7826457098126411, "step": 1816 }, { "epoch": 0.41905585585585586, "grad_norm": 21.102556228637695, "learning_rate": 8.976553684532281e-06, "loss": 11.851313591003418, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.23989843390882015, "loss_total_v6": 0.7407071143388748, "step": 1817 }, { "epoch": 0.4192864864864865, "grad_norm": 22.35422134399414, "learning_rate": 8.975332567365942e-06, "loss": 12.386520385742188, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.270373335108161, "loss_total_v6": 0.7741575017571449, "step": 1818 }, { "epoch": 0.41951711711711714, "grad_norm": 20.54512596130371, "learning_rate": 8.97411080531204e-06, "loss": 11.88996696472168, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.24435892887413502, "loss_total_v6": 0.7431229427456856, "step": 1819 }, { "epoch": 0.41974774774774776, "grad_norm": 24.65509796142578, "learning_rate": 8.972888398568772e-06, "loss": 12.593294143676758, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.2843800950795412, "loss_total_v6": 0.7870808765292168, "step": 1820 }, { "epoch": 0.41997837837837837, "grad_norm": 20.19536590576172, "learning_rate": 8.971665347334442e-06, "loss": 12.473845481872559, "loss_alignment": 1.01806640625, "loss_alignment_w": 0.509033203125, "loss_sft": 0.2696970868855715, "loss_total_v6": 0.7796153277158737, "step": 1821 }, { "epoch": 0.420209009009009, "grad_norm": 26.319225311279297, "learning_rate": 8.970441651807454e-06, "loss": 12.805963516235352, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.2979160454124212, "loss_total_v6": 0.8003726974129677, "step": 1822 }, { "epoch": 0.42043963963963965, "grad_norm": 24.294485092163086, "learning_rate": 8.969217312186319e-06, "loss": 12.618356704711914, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2709318343549967, "loss_total_v6": 0.7886472567915916, "step": 1823 }, { "epoch": 0.42067027027027026, "grad_norm": 26.497114181518555, "learning_rate": 8.967992328669655e-06, "loss": 12.228687286376953, "loss_alignment": 1.02392578125, "loss_alignment_w": 0.511962890625, "loss_sft": 0.25440522097051144, "loss_total_v6": 0.7642929330468178, "step": 1824 }, { "epoch": 0.4209009009009009, "grad_norm": 20.974374771118164, "learning_rate": 8.966766701456177e-06, "loss": 12.831356048583984, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.3023564387112856, "loss_total_v6": 0.8019597381353378, "step": 1825 }, { "epoch": 0.42113153153153154, "grad_norm": 21.85319709777832, "learning_rate": 8.965540430744713e-06, "loss": 12.857744216918945, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2924700267612934, "loss_total_v6": 0.8036089465022087, "step": 1826 }, { "epoch": 0.42136216216216216, "grad_norm": 22.758289337158203, "learning_rate": 8.964313516734188e-06, "loss": 12.859210968017578, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2996876835823059, "loss_total_v6": 0.8037007302045822, "step": 1827 }, { "epoch": 0.42159279279279277, "grad_norm": 24.314237594604492, "learning_rate": 8.963085959623638e-06, "loss": 12.959735870361328, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2915203236043453, "loss_total_v6": 0.8099835216999054, "step": 1828 }, { "epoch": 0.42182342342342344, "grad_norm": 23.875478744506836, "learning_rate": 8.961857759612198e-06, "loss": 12.396150588989258, "loss_alignment": 1.01904296875, "loss_alignment_w": 0.509521484375, "loss_sft": 0.2654668390750885, "loss_total_v6": 0.7747594490647316, "step": 1829 }, { "epoch": 0.42205405405405405, "grad_norm": 25.342941284179688, "learning_rate": 8.96062891689911e-06, "loss": 13.294865608215332, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.32136183977127075, "loss_total_v6": 0.8309290856122971, "step": 1830 }, { "epoch": 0.42228468468468466, "grad_norm": 22.746553421020508, "learning_rate": 8.959399431683721e-06, "loss": 12.750799179077148, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.28915825486183167, "loss_total_v6": 0.7969249784946442, "step": 1831 }, { "epoch": 0.42251531531531533, "grad_norm": 24.674867630004883, "learning_rate": 8.95816930416548e-06, "loss": 11.954060554504395, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.2418187726289034, "loss_total_v6": 0.7471287772059441, "step": 1832 }, { "epoch": 0.42274594594594594, "grad_norm": 22.6810245513916, "learning_rate": 8.956938534543939e-06, "loss": 12.883951187133789, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2987619675695896, "loss_total_v6": 0.8052469789981842, "step": 1833 }, { "epoch": 0.42297657657657656, "grad_norm": 22.798830032348633, "learning_rate": 8.95570712301876e-06, "loss": 12.506763458251953, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2745773922652006, "loss_total_v6": 0.7816727235913277, "step": 1834 }, { "epoch": 0.4232072072072072, "grad_norm": 23.11501121520996, "learning_rate": 8.954475069789703e-06, "loss": 12.84516429901123, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.29481192864477634, "loss_total_v6": 0.8028227835893631, "step": 1835 }, { "epoch": 0.42343783783783784, "grad_norm": 25.482269287109375, "learning_rate": 8.953242375056634e-06, "loss": 12.619385719299316, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2828216888010502, "loss_total_v6": 0.7887116074562073, "step": 1836 }, { "epoch": 0.42366846846846845, "grad_norm": 23.535045623779297, "learning_rate": 8.952009039019526e-06, "loss": 12.344209671020508, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.266737075522542, "loss_total_v6": 0.7715130597352982, "step": 1837 }, { "epoch": 0.4238990990990991, "grad_norm": 23.88230323791504, "learning_rate": 8.950775061878453e-06, "loss": 12.625659942626953, "loss_alignment": 1.02197265625, "loss_alignment_w": 0.510986328125, "loss_sft": 0.2783005051314831, "loss_total_v6": 0.789103738963604, "step": 1838 }, { "epoch": 0.42412972972972973, "grad_norm": 24.769466400146484, "learning_rate": 8.949540443833591e-06, "loss": 12.473177909851074, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.265932260081172, "loss_total_v6": 0.7795736417174339, "step": 1839 }, { "epoch": 0.42436036036036034, "grad_norm": 23.061296463012695, "learning_rate": 8.948305185085226e-06, "loss": 12.1583890914917, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2573815807700157, "loss_total_v6": 0.75989930331707, "step": 1840 }, { "epoch": 0.424590990990991, "grad_norm": 22.968177795410156, "learning_rate": 8.947069285833741e-06, "loss": 12.589313507080078, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.2811710834503174, "loss_total_v6": 0.7868321165442467, "step": 1841 }, { "epoch": 0.4248216216216216, "grad_norm": 22.986408233642578, "learning_rate": 8.945832746279631e-06, "loss": 12.708065032958984, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2802007105201483, "loss_total_v6": 0.7942540422081947, "step": 1842 }, { "epoch": 0.42505225225225224, "grad_norm": 23.829639434814453, "learning_rate": 8.944595566623483e-06, "loss": 12.792143821716309, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2858676705509424, "loss_total_v6": 0.7995090112090111, "step": 1843 }, { "epoch": 0.4252828828828829, "grad_norm": 23.91489028930664, "learning_rate": 8.943357747066004e-06, "loss": 12.73620319366455, "loss_alignment": 1.02783203125, "loss_alignment_w": 0.513916015625, "loss_sft": 0.2823713440448046, "loss_total_v6": 0.7960126847028732, "step": 1844 }, { "epoch": 0.4255135135135135, "grad_norm": 26.37476921081543, "learning_rate": 8.942119287807991e-06, "loss": 12.343242645263672, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.27128479443490505, "loss_total_v6": 0.7714526578783989, "step": 1845 }, { "epoch": 0.42574414414414413, "grad_norm": 22.28016471862793, "learning_rate": 8.940880189050349e-06, "loss": 12.227259635925293, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.26421898417174816, "loss_total_v6": 0.7642037346959114, "step": 1846 }, { "epoch": 0.4259747747747748, "grad_norm": 24.911340713500977, "learning_rate": 8.93964045099409e-06, "loss": 12.307778358459473, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.27196747437119484, "loss_total_v6": 0.7692361325025558, "step": 1847 }, { "epoch": 0.4262054054054054, "grad_norm": 22.290424346923828, "learning_rate": 8.938400073840325e-06, "loss": 12.642956733703613, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2764366362243891, "loss_total_v6": 0.7901847884058952, "step": 1848 }, { "epoch": 0.426436036036036, "grad_norm": 23.167449951171875, "learning_rate": 8.937159057790272e-06, "loss": 12.24465560913086, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.25337386690080166, "loss_total_v6": 0.7652909830212593, "step": 1849 }, { "epoch": 0.4266666666666667, "grad_norm": 23.029462814331055, "learning_rate": 8.935917403045251e-06, "loss": 12.809772491455078, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2955448590219021, "loss_total_v6": 0.8006107732653618, "step": 1850 }, { "epoch": 0.4268972972972973, "grad_norm": 22.743227005004883, "learning_rate": 8.934675109806688e-06, "loss": 12.666984558105469, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2832942306995392, "loss_total_v6": 0.7916865274310112, "step": 1851 }, { "epoch": 0.4271279279279279, "grad_norm": 23.414024353027344, "learning_rate": 8.933432178276108e-06, "loss": 12.461091995239258, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2681982219219208, "loss_total_v6": 0.7788183018565178, "step": 1852 }, { "epoch": 0.42735855855855853, "grad_norm": 25.043779373168945, "learning_rate": 8.932188608655141e-06, "loss": 12.605133056640625, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.2889041770249605, "loss_total_v6": 0.7878208011388779, "step": 1853 }, { "epoch": 0.4275891891891892, "grad_norm": 22.13701057434082, "learning_rate": 8.930944401145527e-06, "loss": 12.836726188659668, "loss_alignment": 1.02490234375, "loss_alignment_w": 0.512451171875, "loss_sft": 0.29063769057393074, "loss_total_v6": 0.8022954240441322, "step": 1854 }, { "epoch": 0.4278198198198198, "grad_norm": 25.060312271118164, "learning_rate": 8.9296995559491e-06, "loss": 12.286346435546875, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.26547049172222614, "loss_total_v6": 0.7678966224193573, "step": 1855 }, { "epoch": 0.4280504504504504, "grad_norm": 22.41282844543457, "learning_rate": 8.928454073267801e-06, "loss": 12.459386825561523, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2641853205859661, "loss_total_v6": 0.7787116914987564, "step": 1856 }, { "epoch": 0.4282810810810811, "grad_norm": 22.823396682739258, "learning_rate": 8.927207953303678e-06, "loss": 12.267356872558594, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.26809830963611603, "loss_total_v6": 0.7667097747325897, "step": 1857 }, { "epoch": 0.4285117117117117, "grad_norm": 23.750757217407227, "learning_rate": 8.92596119625888e-06, "loss": 12.357460975646973, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.2738061994314194, "loss_total_v6": 0.7723413407802582, "step": 1858 }, { "epoch": 0.4287423423423423, "grad_norm": 22.354385375976562, "learning_rate": 8.924713802335656e-06, "loss": 12.242417335510254, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.2686148080974817, "loss_total_v6": 0.7651510536670685, "step": 1859 }, { "epoch": 0.428972972972973, "grad_norm": 23.44908905029297, "learning_rate": 8.923465771736361e-06, "loss": 12.898494720458984, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.28775373101234436, "loss_total_v6": 0.8061558902263641, "step": 1860 }, { "epoch": 0.4292036036036036, "grad_norm": 23.687074661254883, "learning_rate": 8.922217104663454e-06, "loss": 12.448948860168457, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2709792200475931, "loss_total_v6": 0.7780593037605286, "step": 1861 }, { "epoch": 0.4294342342342342, "grad_norm": 22.348854064941406, "learning_rate": 8.920967801319498e-06, "loss": 12.587493896484375, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2753048110753298, "loss_total_v6": 0.7867183610796928, "step": 1862 }, { "epoch": 0.4296648648648649, "grad_norm": 24.54994773864746, "learning_rate": 8.919717861907158e-06, "loss": 12.456732749938965, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2714352197945118, "loss_total_v6": 0.7785458192229271, "step": 1863 }, { "epoch": 0.4298954954954955, "grad_norm": 21.278976440429688, "learning_rate": 8.9184672866292e-06, "loss": 12.282013893127441, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2597675882279873, "loss_total_v6": 0.7676258683204651, "step": 1864 }, { "epoch": 0.4301261261261261, "grad_norm": 22.303316116333008, "learning_rate": 8.917216075688496e-06, "loss": 12.47248649597168, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.27594463527202606, "loss_total_v6": 0.779530443251133, "step": 1865 }, { "epoch": 0.4303567567567568, "grad_norm": 22.19400405883789, "learning_rate": 8.915964229288021e-06, "loss": 13.287012100219727, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.33106386847794056, "loss_total_v6": 0.8304382488131523, "step": 1866 }, { "epoch": 0.4305873873873874, "grad_norm": 20.61075782775879, "learning_rate": 8.91471174763085e-06, "loss": 12.377792358398438, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2643804010003805, "loss_total_v6": 0.7736119851469994, "step": 1867 }, { "epoch": 0.430818018018018, "grad_norm": 22.26385498046875, "learning_rate": 8.913458630920169e-06, "loss": 12.466634750366211, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.27266442589461803, "loss_total_v6": 0.7791646867990494, "step": 1868 }, { "epoch": 0.43104864864864867, "grad_norm": 25.581592559814453, "learning_rate": 8.912204879359255e-06, "loss": 12.148237228393555, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2502468265593052, "loss_total_v6": 0.7592648044228554, "step": 1869 }, { "epoch": 0.4312792792792793, "grad_norm": 21.826509475708008, "learning_rate": 8.910950493151498e-06, "loss": 12.02151107788086, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.24775861948728561, "loss_total_v6": 0.7513444274663925, "step": 1870 }, { "epoch": 0.4315099099099099, "grad_norm": 24.06782341003418, "learning_rate": 8.909695472500387e-06, "loss": 12.859332084655762, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.29740636236965656, "loss_total_v6": 0.8037082850933075, "step": 1871 }, { "epoch": 0.43174054054054056, "grad_norm": 23.45041275024414, "learning_rate": 8.908439817609514e-06, "loss": 12.347726821899414, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.26668222434818745, "loss_total_v6": 0.7717328816652298, "step": 1872 }, { "epoch": 0.4319711711711712, "grad_norm": 20.460803985595703, "learning_rate": 8.907183528682575e-06, "loss": 12.429498672485352, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.27160996943712234, "loss_total_v6": 0.7768436968326569, "step": 1873 }, { "epoch": 0.4322018018018018, "grad_norm": 22.525423049926758, "learning_rate": 8.905926605923365e-06, "loss": 12.731148719787598, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.279934398829937, "loss_total_v6": 0.7956967502832413, "step": 1874 }, { "epoch": 0.43243243243243246, "grad_norm": 24.16112518310547, "learning_rate": 8.90466904953579e-06, "loss": 11.789052963256836, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.23905878514051437, "loss_total_v6": 0.7368157580494881, "step": 1875 }, { "epoch": 0.43266306306306307, "grad_norm": 22.748044967651367, "learning_rate": 8.903410859723848e-06, "loss": 12.285871505737305, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.262144960463047, "loss_total_v6": 0.7678669914603233, "step": 1876 }, { "epoch": 0.4328936936936937, "grad_norm": 21.72636604309082, "learning_rate": 8.902152036691649e-06, "loss": 12.534662246704102, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.27755701169371605, "loss_total_v6": 0.783416360616684, "step": 1877 }, { "epoch": 0.43312432432432435, "grad_norm": 26.276153564453125, "learning_rate": 8.900892580643404e-06, "loss": 12.712081909179688, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2888593915849924, "loss_total_v6": 0.7945051267743111, "step": 1878 }, { "epoch": 0.43335495495495496, "grad_norm": 23.0001277923584, "learning_rate": 8.899632491783419e-06, "loss": 12.213123321533203, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.26307608745992184, "loss_total_v6": 0.7633202224969864, "step": 1879 }, { "epoch": 0.4335855855855856, "grad_norm": 20.836181640625, "learning_rate": 8.898371770316113e-06, "loss": 12.535242080688477, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.272710457444191, "loss_total_v6": 0.7834526672959328, "step": 1880 }, { "epoch": 0.43381621621621624, "grad_norm": 22.432767868041992, "learning_rate": 8.897110416446002e-06, "loss": 12.274002075195312, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2603807207196951, "loss_total_v6": 0.767125129699707, "step": 1881 }, { "epoch": 0.43404684684684686, "grad_norm": 23.15557098388672, "learning_rate": 8.895848430377705e-06, "loss": 12.773189544677734, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.28906229324638844, "loss_total_v6": 0.7983243837952614, "step": 1882 }, { "epoch": 0.43427747747747747, "grad_norm": 20.990032196044922, "learning_rate": 8.894585812315945e-06, "loss": 12.749088287353516, "loss_alignment": 1.01220703125, "loss_alignment_w": 0.506103515625, "loss_sft": 0.2907754685729742, "loss_total_v6": 0.7968179881572723, "step": 1883 }, { "epoch": 0.4345081081081081, "grad_norm": 27.969139099121094, "learning_rate": 8.893322562465547e-06, "loss": 12.811145782470703, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2908089179545641, "loss_total_v6": 0.8006966188549995, "step": 1884 }, { "epoch": 0.43473873873873875, "grad_norm": 23.108598709106445, "learning_rate": 8.89205868103144e-06, "loss": 13.108881950378418, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.3049770873039961, "loss_total_v6": 0.8193050920963287, "step": 1885 }, { "epoch": 0.43496936936936936, "grad_norm": 26.777956008911133, "learning_rate": 8.89079416821865e-06, "loss": 12.951258659362793, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.30185480788350105, "loss_total_v6": 0.8094536662101746, "step": 1886 }, { "epoch": 0.4352, "grad_norm": 26.950519561767578, "learning_rate": 8.889529024232311e-06, "loss": 12.694511413574219, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.27994872257113457, "loss_total_v6": 0.7934069633483887, "step": 1887 }, { "epoch": 0.43543063063063064, "grad_norm": 25.190746307373047, "learning_rate": 8.888263249277656e-06, "loss": 12.441072463989258, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.27695664018392563, "loss_total_v6": 0.777567021548748, "step": 1888 }, { "epoch": 0.43566126126126126, "grad_norm": 23.516029357910156, "learning_rate": 8.886996843560025e-06, "loss": 12.73311710357666, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2847572546452284, "loss_total_v6": 0.7958198338747025, "step": 1889 }, { "epoch": 0.43589189189189187, "grad_norm": 24.074569702148438, "learning_rate": 8.885729807284855e-06, "loss": 12.344583511352539, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.26613484509289265, "loss_total_v6": 0.7715364620089531, "step": 1890 }, { "epoch": 0.43612252252252254, "grad_norm": 23.373811721801758, "learning_rate": 8.884462140657689e-06, "loss": 13.050976753234863, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.30674435570836067, "loss_total_v6": 0.8156860396265984, "step": 1891 }, { "epoch": 0.43635315315315315, "grad_norm": 22.452695846557617, "learning_rate": 8.883193843884169e-06, "loss": 13.099332809448242, "loss_alignment": 1.01904296875, "loss_alignment_w": 0.509521484375, "loss_sft": 0.31062115728855133, "loss_total_v6": 0.8187083303928375, "step": 1892 }, { "epoch": 0.43658378378378376, "grad_norm": 23.785324096679688, "learning_rate": 8.881924917170041e-06, "loss": 12.287921905517578, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.26160168275237083, "loss_total_v6": 0.7679951190948486, "step": 1893 }, { "epoch": 0.43681441441441443, "grad_norm": 22.611610412597656, "learning_rate": 8.880655360721157e-06, "loss": 12.738523483276367, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.28280629590153694, "loss_total_v6": 0.7961577326059341, "step": 1894 }, { "epoch": 0.43704504504504504, "grad_norm": 26.40348243713379, "learning_rate": 8.879385174743462e-06, "loss": 12.745294570922852, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.285335224121809, "loss_total_v6": 0.7965809479355812, "step": 1895 }, { "epoch": 0.43727567567567566, "grad_norm": 20.93982696533203, "learning_rate": 8.878114359443012e-06, "loss": 12.336580276489258, "loss_alignment": 0.98583984375, "loss_alignment_w": 0.492919921875, "loss_sft": 0.2771855816245079, "loss_total_v6": 0.7710362896323204, "step": 1896 }, { "epoch": 0.4375063063063063, "grad_norm": 21.291133880615234, "learning_rate": 8.87684291502596e-06, "loss": 12.657445907592773, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2803023811429739, "loss_total_v6": 0.7910903543233871, "step": 1897 }, { "epoch": 0.43773693693693694, "grad_norm": 25.131103515625, "learning_rate": 8.875570841698564e-06, "loss": 12.552753448486328, "loss_alignment": 1.02197265625, "loss_alignment_w": 0.510986328125, "loss_sft": 0.27439997904002666, "loss_total_v6": 0.7845470607280731, "step": 1898 }, { "epoch": 0.43796756756756755, "grad_norm": 24.03199577331543, "learning_rate": 8.874298139667182e-06, "loss": 11.909698486328125, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.24421889148652554, "loss_total_v6": 0.7443562000989914, "step": 1899 }, { "epoch": 0.4381981981981982, "grad_norm": 22.19931983947754, "learning_rate": 8.873024809138272e-06, "loss": 12.54770278930664, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2781584206968546, "loss_total_v6": 0.7842314019799232, "step": 1900 }, { "epoch": 0.43842882882882883, "grad_norm": 23.87339973449707, "learning_rate": 8.871750850318402e-06, "loss": 12.690164566040039, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.28477352298796177, "loss_total_v6": 0.793135330080986, "step": 1901 }, { "epoch": 0.43865945945945944, "grad_norm": 22.655406951904297, "learning_rate": 8.870476263414235e-06, "loss": 12.827466011047363, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2960708886384964, "loss_total_v6": 0.8017166331410408, "step": 1902 }, { "epoch": 0.4388900900900901, "grad_norm": 23.519094467163086, "learning_rate": 8.869201048632531e-06, "loss": 13.256819725036621, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.3195180222392082, "loss_total_v6": 0.8285512179136276, "step": 1903 }, { "epoch": 0.4391207207207207, "grad_norm": 25.061424255371094, "learning_rate": 8.867925206180166e-06, "loss": 11.955739974975586, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.24470083601772785, "loss_total_v6": 0.7472337707877159, "step": 1904 }, { "epoch": 0.43935135135135134, "grad_norm": 21.256776809692383, "learning_rate": 8.866648736264108e-06, "loss": 12.53321361541748, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2764746733009815, "loss_total_v6": 0.7833258286118507, "step": 1905 }, { "epoch": 0.439581981981982, "grad_norm": 24.516366958618164, "learning_rate": 8.865371639091426e-06, "loss": 12.545594215393066, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.28277214244008064, "loss_total_v6": 0.7840996608138084, "step": 1906 }, { "epoch": 0.4398126126126126, "grad_norm": 23.85825538635254, "learning_rate": 8.864093914869298e-06, "loss": 12.536907196044922, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.27643083967268467, "loss_total_v6": 0.7835567072033882, "step": 1907 }, { "epoch": 0.44004324324324323, "grad_norm": 22.824424743652344, "learning_rate": 8.862815563804996e-06, "loss": 12.527135848999023, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2739280480891466, "loss_total_v6": 0.7829460129141808, "step": 1908 }, { "epoch": 0.4402738738738739, "grad_norm": 22.786306381225586, "learning_rate": 8.8615365861059e-06, "loss": 12.192039489746094, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2582793664187193, "loss_total_v6": 0.7620025053620338, "step": 1909 }, { "epoch": 0.4405045045045045, "grad_norm": 23.47221565246582, "learning_rate": 8.860256981979485e-06, "loss": 12.753381729125977, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2916085161268711, "loss_total_v6": 0.7970863953232765, "step": 1910 }, { "epoch": 0.4407351351351351, "grad_norm": 23.4421329498291, "learning_rate": 8.858976751633334e-06, "loss": 12.31658935546875, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.26482778042554855, "loss_total_v6": 0.7697868943214417, "step": 1911 }, { "epoch": 0.44096576576576574, "grad_norm": 19.630443572998047, "learning_rate": 8.857695895275127e-06, "loss": 12.226484298706055, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.25297058559954166, "loss_total_v6": 0.764155276119709, "step": 1912 }, { "epoch": 0.4411963963963964, "grad_norm": 23.457584381103516, "learning_rate": 8.856414413112651e-06, "loss": 12.881772994995117, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2941550016403198, "loss_total_v6": 0.8051108345389366, "step": 1913 }, { "epoch": 0.441427027027027, "grad_norm": 23.416255950927734, "learning_rate": 8.855132305353787e-06, "loss": 12.923208236694336, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.2921518385410309, "loss_total_v6": 0.8077005296945572, "step": 1914 }, { "epoch": 0.44165765765765763, "grad_norm": 23.13288688659668, "learning_rate": 8.853849572206523e-06, "loss": 12.541738510131836, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.27407773211598396, "loss_total_v6": 0.7838586419820786, "step": 1915 }, { "epoch": 0.4418882882882883, "grad_norm": 24.768888473510742, "learning_rate": 8.852566213878947e-06, "loss": 13.00753402709961, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.30566196888685226, "loss_total_v6": 0.8129709139466286, "step": 1916 }, { "epoch": 0.4421189189189189, "grad_norm": 20.893566131591797, "learning_rate": 8.851282230579248e-06, "loss": 12.8996000289917, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2975275069475174, "loss_total_v6": 0.8062250092625618, "step": 1917 }, { "epoch": 0.4423495495495495, "grad_norm": 21.634048461914062, "learning_rate": 8.849997622515718e-06, "loss": 12.60159683227539, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.2693960312753916, "loss_total_v6": 0.7875997796654701, "step": 1918 }, { "epoch": 0.4425801801801802, "grad_norm": 25.18819236755371, "learning_rate": 8.848712389896748e-06, "loss": 12.841885566711426, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.29749087803065777, "loss_total_v6": 0.8026178330183029, "step": 1919 }, { "epoch": 0.4428108108108108, "grad_norm": 21.581775665283203, "learning_rate": 8.84742653293083e-06, "loss": 12.884787559509277, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2927870210260153, "loss_total_v6": 0.8052992299199104, "step": 1920 }, { "epoch": 0.4430414414414414, "grad_norm": 22.799110412597656, "learning_rate": 8.846140051826562e-06, "loss": 12.351509094238281, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.26925326883792877, "loss_total_v6": 0.7719693332910538, "step": 1921 }, { "epoch": 0.4432720720720721, "grad_norm": 21.758834838867188, "learning_rate": 8.844852946792636e-06, "loss": 12.32321548461914, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.2707807905972004, "loss_total_v6": 0.7702009677886963, "step": 1922 }, { "epoch": 0.4435027027027027, "grad_norm": 23.588754653930664, "learning_rate": 8.843565218037853e-06, "loss": 12.29555606842041, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2565703894942999, "loss_total_v6": 0.768472246825695, "step": 1923 }, { "epoch": 0.4437333333333333, "grad_norm": 23.77959632873535, "learning_rate": 8.84227686577111e-06, "loss": 13.235116004943848, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.31372126936912537, "loss_total_v6": 0.8271947652101517, "step": 1924 }, { "epoch": 0.443963963963964, "grad_norm": 24.38857078552246, "learning_rate": 8.840987890201404e-06, "loss": 12.48593807220459, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2661499585956335, "loss_total_v6": 0.7803711369633675, "step": 1925 }, { "epoch": 0.4441945945945946, "grad_norm": 22.19658851623535, "learning_rate": 8.839698291537836e-06, "loss": 12.112166404724121, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.25668999552726746, "loss_total_v6": 0.7570104226469994, "step": 1926 }, { "epoch": 0.4444252252252252, "grad_norm": 22.400362014770508, "learning_rate": 8.838408069989613e-06, "loss": 12.83736801147461, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.29751376062631607, "loss_total_v6": 0.8023355081677437, "step": 1927 }, { "epoch": 0.4446558558558559, "grad_norm": 25.378568649291992, "learning_rate": 8.837117225766033e-06, "loss": 12.31814193725586, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.26266642287373543, "loss_total_v6": 0.769883818924427, "step": 1928 }, { "epoch": 0.4448864864864865, "grad_norm": 19.658000946044922, "learning_rate": 8.8358257590765e-06, "loss": 11.875106811523438, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.24291129782795906, "loss_total_v6": 0.7421941384673119, "step": 1929 }, { "epoch": 0.4451171171171171, "grad_norm": 22.04034996032715, "learning_rate": 8.834533670130518e-06, "loss": 12.816252708435059, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.29411880299448967, "loss_total_v6": 0.80101577937603, "step": 1930 }, { "epoch": 0.44534774774774777, "grad_norm": 23.13429832458496, "learning_rate": 8.833240959137695e-06, "loss": 13.033044815063477, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.3132530450820923, "loss_total_v6": 0.8145653083920479, "step": 1931 }, { "epoch": 0.4455783783783784, "grad_norm": 22.10660743713379, "learning_rate": 8.831947626307735e-06, "loss": 13.052579879760742, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.30556287430226803, "loss_total_v6": 0.8157862424850464, "step": 1932 }, { "epoch": 0.445809009009009, "grad_norm": 24.421022415161133, "learning_rate": 8.830653671850448e-06, "loss": 12.78912353515625, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.29353709146380424, "loss_total_v6": 0.799320213496685, "step": 1933 }, { "epoch": 0.44603963963963966, "grad_norm": 26.717540740966797, "learning_rate": 8.829359095975741e-06, "loss": 12.63490104675293, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.27262198366224766, "loss_total_v6": 0.7896813005208969, "step": 1934 }, { "epoch": 0.4462702702702703, "grad_norm": 22.97661018371582, "learning_rate": 8.828063898893623e-06, "loss": 12.94894790649414, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.30424331314861774, "loss_total_v6": 0.8093092367053032, "step": 1935 }, { "epoch": 0.4465009009009009, "grad_norm": 23.243459701538086, "learning_rate": 8.826768080814205e-06, "loss": 12.22325611114502, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.255088122561574, "loss_total_v6": 0.7639535069465637, "step": 1936 }, { "epoch": 0.44673153153153156, "grad_norm": 22.621231079101562, "learning_rate": 8.825471641947696e-06, "loss": 13.004643440246582, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.3017885908484459, "loss_total_v6": 0.8127901926636696, "step": 1937 }, { "epoch": 0.44696216216216217, "grad_norm": 24.51506233215332, "learning_rate": 8.824174582504409e-06, "loss": 12.626357078552246, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.2856988273561001, "loss_total_v6": 0.7891473323106766, "step": 1938 }, { "epoch": 0.4471927927927928, "grad_norm": 25.06289291381836, "learning_rate": 8.822876902694753e-06, "loss": 12.632513046264648, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2772792112082243, "loss_total_v6": 0.7895320355892181, "step": 1939 }, { "epoch": 0.44742342342342345, "grad_norm": 22.25835609436035, "learning_rate": 8.821578602729242e-06, "loss": 12.609630584716797, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2821815051138401, "loss_total_v6": 0.7881019189953804, "step": 1940 }, { "epoch": 0.44765405405405406, "grad_norm": 21.842113494873047, "learning_rate": 8.82027968281849e-06, "loss": 12.40561294555664, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2666228227317333, "loss_total_v6": 0.7753508687019348, "step": 1941 }, { "epoch": 0.4478846846846847, "grad_norm": 20.816116333007812, "learning_rate": 8.818980143173212e-06, "loss": 12.73666763305664, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2927458677440882, "loss_total_v6": 0.7960417419672012, "step": 1942 }, { "epoch": 0.4481153153153153, "grad_norm": 21.57670021057129, "learning_rate": 8.817679984004219e-06, "loss": 12.796102523803711, "loss_alignment": 1.01904296875, "loss_alignment_w": 0.509521484375, "loss_sft": 0.2890447787940502, "loss_total_v6": 0.7997564449906349, "step": 1943 }, { "epoch": 0.44834594594594596, "grad_norm": 22.386924743652344, "learning_rate": 8.816379205522428e-06, "loss": 12.114633560180664, "loss_alignment": 1.01220703125, "loss_alignment_w": 0.506103515625, "loss_sft": 0.24991670064628124, "loss_total_v6": 0.7571646198630333, "step": 1944 }, { "epoch": 0.44857657657657657, "grad_norm": 23.623929977416992, "learning_rate": 8.815077807938855e-06, "loss": 12.525483131408691, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.28084378875792027, "loss_total_v6": 0.7828426733613014, "step": 1945 }, { "epoch": 0.4488072072072072, "grad_norm": 23.09137725830078, "learning_rate": 8.813775791464611e-06, "loss": 12.439217567443848, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.2750401720404625, "loss_total_v6": 0.7774510979652405, "step": 1946 }, { "epoch": 0.44903783783783785, "grad_norm": 22.71744728088379, "learning_rate": 8.812473156310918e-06, "loss": 12.901535034179688, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.30166153237223625, "loss_total_v6": 0.806345984339714, "step": 1947 }, { "epoch": 0.44926846846846846, "grad_norm": 21.53713035583496, "learning_rate": 8.81116990268909e-06, "loss": 12.851731300354004, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.28544147312641144, "loss_total_v6": 0.8032332211732864, "step": 1948 }, { "epoch": 0.4494990990990991, "grad_norm": 23.689006805419922, "learning_rate": 8.809866030810544e-06, "loss": 12.07048225402832, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.260737556964159, "loss_total_v6": 0.75440514087677, "step": 1949 }, { "epoch": 0.44972972972972974, "grad_norm": 23.596664428710938, "learning_rate": 8.808561540886796e-06, "loss": 13.175973892211914, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.3107115179300308, "loss_total_v6": 0.8234983906149864, "step": 1950 }, { "epoch": 0.44996036036036036, "grad_norm": 22.497848510742188, "learning_rate": 8.807256433129465e-06, "loss": 12.496488571166992, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.26070586405694485, "loss_total_v6": 0.7810305655002594, "step": 1951 }, { "epoch": 0.45019099099099097, "grad_norm": 21.90045738220215, "learning_rate": 8.805950707750268e-06, "loss": 12.304192543029785, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.26376304775476456, "loss_total_v6": 0.769012063741684, "step": 1952 }, { "epoch": 0.45042162162162164, "grad_norm": 23.710371017456055, "learning_rate": 8.80464436496102e-06, "loss": 12.546377182006836, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.28341617062687874, "loss_total_v6": 0.7841485813260078, "step": 1953 }, { "epoch": 0.45065225225225225, "grad_norm": 22.59602165222168, "learning_rate": 8.803337404973644e-06, "loss": 12.580960273742676, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.28177819587290287, "loss_total_v6": 0.7863100469112396, "step": 1954 }, { "epoch": 0.45088288288288286, "grad_norm": 23.287168502807617, "learning_rate": 8.802029828000157e-06, "loss": 12.291131973266602, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.26152768172323704, "loss_total_v6": 0.768195778131485, "step": 1955 }, { "epoch": 0.45111351351351353, "grad_norm": 22.73733139038086, "learning_rate": 8.800721634252671e-06, "loss": 12.517520904541016, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.2901880834251642, "loss_total_v6": 0.7823450565338135, "step": 1956 }, { "epoch": 0.45134414414414414, "grad_norm": 23.580249786376953, "learning_rate": 8.799412823943412e-06, "loss": 12.673572540283203, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.28439260832965374, "loss_total_v6": 0.7920982763171196, "step": 1957 }, { "epoch": 0.45157477477477476, "grad_norm": 24.26970863342285, "learning_rate": 8.798103397284695e-06, "loss": 12.70918083190918, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2848023120313883, "loss_total_v6": 0.7943238168954849, "step": 1958 }, { "epoch": 0.4518054054054054, "grad_norm": 24.675554275512695, "learning_rate": 8.796793354488938e-06, "loss": 12.514877319335938, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.27291770093142986, "loss_total_v6": 0.7821797952055931, "step": 1959 }, { "epoch": 0.45203603603603604, "grad_norm": 23.512935638427734, "learning_rate": 8.795482695768658e-06, "loss": 11.992156982421875, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.24499328806996346, "loss_total_v6": 0.749509871006012, "step": 1960 }, { "epoch": 0.45226666666666665, "grad_norm": 22.327428817749023, "learning_rate": 8.794171421336475e-06, "loss": 12.9783296585083, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.30322629399597645, "loss_total_v6": 0.8111456111073494, "step": 1961 }, { "epoch": 0.4524972972972973, "grad_norm": 23.960187911987305, "learning_rate": 8.792859531405106e-06, "loss": 12.746854782104492, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2881029509007931, "loss_total_v6": 0.7966784164309502, "step": 1962 }, { "epoch": 0.45272792792792793, "grad_norm": 24.52110481262207, "learning_rate": 8.791547026187373e-06, "loss": 12.910451889038086, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.299746872857213, "loss_total_v6": 0.806903213262558, "step": 1963 }, { "epoch": 0.45295855855855854, "grad_norm": 23.525123596191406, "learning_rate": 8.790233905896187e-06, "loss": 12.79456901550293, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.295769527554512, "loss_total_v6": 0.7996604964137077, "step": 1964 }, { "epoch": 0.4531891891891892, "grad_norm": 23.855173110961914, "learning_rate": 8.788920170744567e-06, "loss": 12.358738899230957, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2610228806734085, "loss_total_v6": 0.7724211886525154, "step": 1965 }, { "epoch": 0.4534198198198198, "grad_norm": 22.141897201538086, "learning_rate": 8.787605820945633e-06, "loss": 12.205740928649902, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.2593493275344372, "loss_total_v6": 0.7628588229417801, "step": 1966 }, { "epoch": 0.45365045045045044, "grad_norm": 22.567270278930664, "learning_rate": 8.786290856712598e-06, "loss": 12.586440086364746, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.28166285529732704, "loss_total_v6": 0.7866524904966354, "step": 1967 }, { "epoch": 0.4538810810810811, "grad_norm": 21.762779235839844, "learning_rate": 8.784975278258783e-06, "loss": 11.726265907287598, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.23641635850071907, "loss_total_v6": 0.7328915894031525, "step": 1968 }, { "epoch": 0.4541117117117117, "grad_norm": 25.644695281982422, "learning_rate": 8.783659085797602e-06, "loss": 12.358213424682617, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.25558846443891525, "loss_total_v6": 0.772388368844986, "step": 1969 }, { "epoch": 0.45434234234234233, "grad_norm": 21.78114128112793, "learning_rate": 8.782342279542569e-06, "loss": 12.718961715698242, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.28320114873349667, "loss_total_v6": 0.7949351444840431, "step": 1970 }, { "epoch": 0.454572972972973, "grad_norm": 23.673168182373047, "learning_rate": 8.781024859707301e-06, "loss": 12.492827415466309, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.26609223149716854, "loss_total_v6": 0.7808016985654831, "step": 1971 }, { "epoch": 0.4548036036036036, "grad_norm": 22.736732482910156, "learning_rate": 8.779706826505513e-06, "loss": 12.998003005981445, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.29774194583296776, "loss_total_v6": 0.8123751431703568, "step": 1972 }, { "epoch": 0.4550342342342342, "grad_norm": 22.608186721801758, "learning_rate": 8.778388180151019e-06, "loss": 12.411865234375, "loss_alignment": 1.01318359375, "loss_alignment_w": 0.506591796875, "loss_sft": 0.2703857645392418, "loss_total_v6": 0.7757415845990181, "step": 1973 }, { "epoch": 0.45526486486486484, "grad_norm": 23.45413589477539, "learning_rate": 8.777068920857734e-06, "loss": 12.611713409423828, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2816859856247902, "loss_total_v6": 0.7882320284843445, "step": 1974 }, { "epoch": 0.4554954954954955, "grad_norm": 25.27228355407715, "learning_rate": 8.77574904883967e-06, "loss": 13.136082649230957, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.31380296871066093, "loss_total_v6": 0.8210051357746124, "step": 1975 }, { "epoch": 0.4557261261261261, "grad_norm": 20.2070255279541, "learning_rate": 8.774428564310939e-06, "loss": 12.71365737915039, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.29077363573014736, "loss_total_v6": 0.7946035638451576, "step": 1976 }, { "epoch": 0.45595675675675673, "grad_norm": 22.524921417236328, "learning_rate": 8.773107467485752e-06, "loss": 12.60330581665039, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2808096185326576, "loss_total_v6": 0.7877065911889076, "step": 1977 }, { "epoch": 0.4561873873873874, "grad_norm": 19.818456649780273, "learning_rate": 8.771785758578427e-06, "loss": 12.264952659606934, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2576484102755785, "loss_total_v6": 0.7665595337748528, "step": 1978 }, { "epoch": 0.456418018018018, "grad_norm": 24.483745574951172, "learning_rate": 8.770463437803367e-06, "loss": 13.015756607055664, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2971273083239794, "loss_total_v6": 0.8134847581386566, "step": 1979 }, { "epoch": 0.4566486486486486, "grad_norm": 21.072494506835938, "learning_rate": 8.769140505375084e-06, "loss": 12.122088432312012, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2462016772478819, "loss_total_v6": 0.7576305195689201, "step": 1980 }, { "epoch": 0.4568792792792793, "grad_norm": 22.95305633544922, "learning_rate": 8.767816961508191e-06, "loss": 12.390987396240234, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.26042918115854263, "loss_total_v6": 0.7744367197155952, "step": 1981 }, { "epoch": 0.4571099099099099, "grad_norm": 22.89564323425293, "learning_rate": 8.76649280641739e-06, "loss": 12.167295455932617, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.24959173426032066, "loss_total_v6": 0.7604560106992722, "step": 1982 }, { "epoch": 0.4573405405405405, "grad_norm": 23.695192337036133, "learning_rate": 8.765168040317494e-06, "loss": 12.461939811706543, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.28112957812845707, "loss_total_v6": 0.7788712382316589, "step": 1983 }, { "epoch": 0.4575711711711712, "grad_norm": 19.371706008911133, "learning_rate": 8.763842663423408e-06, "loss": 12.037372589111328, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.24963499419391155, "loss_total_v6": 0.7523357793688774, "step": 1984 }, { "epoch": 0.4578018018018018, "grad_norm": 22.227182388305664, "learning_rate": 8.762516675950134e-06, "loss": 12.83283805847168, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2908676937222481, "loss_total_v6": 0.80205237865448, "step": 1985 }, { "epoch": 0.4580324324324324, "grad_norm": 22.792301177978516, "learning_rate": 8.761190078112783e-06, "loss": 12.025015830993652, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.25514928810298443, "loss_total_v6": 0.7515635043382645, "step": 1986 }, { "epoch": 0.4582630630630631, "grad_norm": 26.403491973876953, "learning_rate": 8.759862870126552e-06, "loss": 12.941539764404297, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.3031089920550585, "loss_total_v6": 0.8088462725281715, "step": 1987 }, { "epoch": 0.4584936936936937, "grad_norm": 24.762540817260742, "learning_rate": 8.758535052206749e-06, "loss": 12.937758445739746, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.29528896883130074, "loss_total_v6": 0.8086098954081535, "step": 1988 }, { "epoch": 0.4587243243243243, "grad_norm": 22.977746963500977, "learning_rate": 8.757206624568774e-06, "loss": 12.734243392944336, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.29579864628612995, "loss_total_v6": 0.7958901822566986, "step": 1989 }, { "epoch": 0.458954954954955, "grad_norm": 21.453842163085938, "learning_rate": 8.755877587428127e-06, "loss": 12.291793823242188, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.26550582982599735, "loss_total_v6": 0.7682371512055397, "step": 1990 }, { "epoch": 0.4591855855855856, "grad_norm": 23.691076278686523, "learning_rate": 8.754547941000412e-06, "loss": 12.162788391113281, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.25762609392404556, "loss_total_v6": 0.7601742744445801, "step": 1991 }, { "epoch": 0.4594162162162162, "grad_norm": 23.966941833496094, "learning_rate": 8.753217685501317e-06, "loss": 12.67313003540039, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2768575865775347, "loss_total_v6": 0.7920705825090408, "step": 1992 }, { "epoch": 0.45964684684684687, "grad_norm": 25.489850997924805, "learning_rate": 8.75188682114665e-06, "loss": 12.485309600830078, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.27433513291180134, "loss_total_v6": 0.7803318053483963, "step": 1993 }, { "epoch": 0.4598774774774775, "grad_norm": 26.199556350708008, "learning_rate": 8.750555348152299e-06, "loss": 13.384809494018555, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.3273800816386938, "loss_total_v6": 0.8365506082773209, "step": 1994 }, { "epoch": 0.4601081081081081, "grad_norm": 21.697099685668945, "learning_rate": 8.749223266734262e-06, "loss": 12.628202438354492, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.28343375585973263, "loss_total_v6": 0.789262630045414, "step": 1995 }, { "epoch": 0.46033873873873876, "grad_norm": 22.625354766845703, "learning_rate": 8.747890577108633e-06, "loss": 12.795158386230469, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2954708244651556, "loss_total_v6": 0.7996974512934685, "step": 1996 }, { "epoch": 0.4605693693693694, "grad_norm": 22.43921661376953, "learning_rate": 8.746557279491603e-06, "loss": 12.445270538330078, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.275372764095664, "loss_total_v6": 0.7778294011950493, "step": 1997 }, { "epoch": 0.4608, "grad_norm": 21.890079498291016, "learning_rate": 8.745223374099464e-06, "loss": 12.506208419799805, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2747105583548546, "loss_total_v6": 0.7816380560398102, "step": 1998 }, { "epoch": 0.46103063063063066, "grad_norm": 23.224512100219727, "learning_rate": 8.743888861148602e-06, "loss": 12.741064071655273, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.29154047742486, "loss_total_v6": 0.7963164672255516, "step": 1999 }, { "epoch": 0.46126126126126127, "grad_norm": 21.11588478088379, "learning_rate": 8.742553740855507e-06, "loss": 12.549397468566895, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.27254231460392475, "loss_total_v6": 0.7843373492360115, "step": 2000 }, { "epoch": 0.46126126126126127, "eval_loss": 0.9497801065444946, "eval_loss_alignment": 1.0139015232591324, "eval_loss_alignment_w": 0.5069507616295662, "eval_loss_contrast": 0.0, "eval_loss_highlevel": 0.0, "eval_loss_sft": 0.4428293346065909, "eval_loss_total_v6": 0.9497801000464997, "eval_loss_wm_recon": 0.0, "eval_loss_wm_recon_w": 0.0, "eval_runtime": 305.1122, "eval_samples_per_second": 5.719, "eval_steps_per_second": 0.718, "step": 2000 }, { "epoch": 0.4614918918918919, "grad_norm": 25.204906463623047, "learning_rate": 8.741218013436764e-06, "loss": 12.93905258178711, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.3058984540402889, "loss_total_v6": 0.8086907938122749, "step": 2001 }, { "epoch": 0.46172252252252255, "grad_norm": 21.045839309692383, "learning_rate": 8.739881679109059e-06, "loss": 12.264548301696777, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2564329281449318, "loss_total_v6": 0.766534261405468, "step": 2002 }, { "epoch": 0.46195315315315316, "grad_norm": 23.60470962524414, "learning_rate": 8.738544738089177e-06, "loss": 12.756422996520996, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.28381818532943726, "loss_total_v6": 0.7972764372825623, "step": 2003 }, { "epoch": 0.4621837837837838, "grad_norm": 22.96561622619629, "learning_rate": 8.737207190593995e-06, "loss": 12.411970138549805, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.27289473824203014, "loss_total_v6": 0.775748111307621, "step": 2004 }, { "epoch": 0.4624144144144144, "grad_norm": 26.390941619873047, "learning_rate": 8.735869036840496e-06, "loss": 12.987318992614746, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.3011788874864578, "loss_total_v6": 0.8117074370384216, "step": 2005 }, { "epoch": 0.46264504504504506, "grad_norm": 26.127079010009766, "learning_rate": 8.734530277045759e-06, "loss": 12.7620267868042, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2909432705491781, "loss_total_v6": 0.7976266592741013, "step": 2006 }, { "epoch": 0.46287567567567567, "grad_norm": 23.973169326782227, "learning_rate": 8.733190911426957e-06, "loss": 12.009757995605469, "loss_alignment": 1.01513671875, "loss_alignment_w": 0.507568359375, "loss_sft": 0.24421645142138004, "loss_total_v6": 0.750609889626503, "step": 2007 }, { "epoch": 0.4631063063063063, "grad_norm": 21.82257652282715, "learning_rate": 8.731850940201369e-06, "loss": 12.259847640991211, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2533620595932007, "loss_total_v6": 0.7662404626607895, "step": 2008 }, { "epoch": 0.46333693693693695, "grad_norm": 24.0062198638916, "learning_rate": 8.730510363586368e-06, "loss": 12.36807632446289, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.26993771456182003, "loss_total_v6": 0.7730047628283501, "step": 2009 }, { "epoch": 0.46356756756756756, "grad_norm": 22.674692153930664, "learning_rate": 8.729169181799423e-06, "loss": 11.971197128295898, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.242935536429286, "loss_total_v6": 0.7481998354196548, "step": 2010 }, { "epoch": 0.4637981981981982, "grad_norm": 23.143049240112305, "learning_rate": 8.727827395058103e-06, "loss": 12.059529304504395, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.2546819541603327, "loss_total_v6": 0.7537206187844276, "step": 2011 }, { "epoch": 0.46402882882882884, "grad_norm": 22.471755981445312, "learning_rate": 8.72648500358008e-06, "loss": 12.369183540344238, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.2757137008011341, "loss_total_v6": 0.7730739638209343, "step": 2012 }, { "epoch": 0.46425945945945946, "grad_norm": 22.197853088378906, "learning_rate": 8.725142007583115e-06, "loss": 12.13662338256836, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.24839185923337936, "loss_total_v6": 0.7585389763116837, "step": 2013 }, { "epoch": 0.46449009009009007, "grad_norm": 21.67507553100586, "learning_rate": 8.723798407285078e-06, "loss": 12.452587127685547, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.2578552197664976, "loss_total_v6": 0.7782867252826691, "step": 2014 }, { "epoch": 0.46472072072072074, "grad_norm": 22.752321243286133, "learning_rate": 8.722454202903923e-06, "loss": 12.706378936767578, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2910054214298725, "loss_total_v6": 0.7941486984491348, "step": 2015 }, { "epoch": 0.46495135135135135, "grad_norm": 22.057294845581055, "learning_rate": 8.721109394657717e-06, "loss": 12.341760635375977, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2640968207269907, "loss_total_v6": 0.7713600248098373, "step": 2016 }, { "epoch": 0.46518198198198196, "grad_norm": 24.588607788085938, "learning_rate": 8.719763982764613e-06, "loss": 12.331923484802246, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.2696313224732876, "loss_total_v6": 0.7707452103495598, "step": 2017 }, { "epoch": 0.46541261261261263, "grad_norm": 24.93829345703125, "learning_rate": 8.718417967442868e-06, "loss": 12.493952751159668, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.2841679137200117, "loss_total_v6": 0.780872032046318, "step": 2018 }, { "epoch": 0.46564324324324324, "grad_norm": 21.825820922851562, "learning_rate": 8.717071348910837e-06, "loss": 12.800439834594727, "loss_alignment": 1.01318359375, "loss_alignment_w": 0.506591796875, "loss_sft": 0.29210815764963627, "loss_total_v6": 0.8000274896621704, "step": 2019 }, { "epoch": 0.46587387387387386, "grad_norm": 21.12969970703125, "learning_rate": 8.715724127386971e-06, "loss": 12.406516075134277, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.26292554289102554, "loss_total_v6": 0.7754072472453117, "step": 2020 }, { "epoch": 0.4661045045045045, "grad_norm": 25.134090423583984, "learning_rate": 8.71437630308982e-06, "loss": 12.082773208618164, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.2552496399730444, "loss_total_v6": 0.7551733329892159, "step": 2021 }, { "epoch": 0.46633513513513514, "grad_norm": 23.91106414794922, "learning_rate": 8.71302787623803e-06, "loss": 13.179788589477539, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.3139101341366768, "loss_total_v6": 0.8237367868423462, "step": 2022 }, { "epoch": 0.46656576576576575, "grad_norm": 24.318096160888672, "learning_rate": 8.711678847050347e-06, "loss": 12.841575622558594, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2922072671353817, "loss_total_v6": 0.8025984913110733, "step": 2023 }, { "epoch": 0.4667963963963964, "grad_norm": 23.975563049316406, "learning_rate": 8.710329215745612e-06, "loss": 12.65364933013916, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.28796917386353016, "loss_total_v6": 0.7908530831336975, "step": 2024 }, { "epoch": 0.46702702702702703, "grad_norm": 20.777711868286133, "learning_rate": 8.708978982542766e-06, "loss": 12.261600494384766, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.25812555104494095, "loss_total_v6": 0.7663500234484673, "step": 2025 }, { "epoch": 0.46725765765765764, "grad_norm": 22.571016311645508, "learning_rate": 8.707628147660847e-06, "loss": 12.714797973632812, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2908144500106573, "loss_total_v6": 0.794674888253212, "step": 2026 }, { "epoch": 0.4674882882882883, "grad_norm": 23.904563903808594, "learning_rate": 8.70627671131899e-06, "loss": 12.43043041229248, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2668921146541834, "loss_total_v6": 0.7769019082188606, "step": 2027 }, { "epoch": 0.4677189189189189, "grad_norm": 23.00613021850586, "learning_rate": 8.70492467373643e-06, "loss": 12.52089786529541, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.282083123922348, "loss_total_v6": 0.7825561538338661, "step": 2028 }, { "epoch": 0.46794954954954954, "grad_norm": 25.467815399169922, "learning_rate": 8.703572035132497e-06, "loss": 12.647480010986328, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2756206542253494, "loss_total_v6": 0.7904674634337425, "step": 2029 }, { "epoch": 0.4681801801801802, "grad_norm": 22.181779861450195, "learning_rate": 8.702218795726619e-06, "loss": 12.22885513305664, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.2609465476125479, "loss_total_v6": 0.7643034756183624, "step": 2030 }, { "epoch": 0.4684108108108108, "grad_norm": 23.88681983947754, "learning_rate": 8.700864955738318e-06, "loss": 12.228538513183594, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.25729513727128506, "loss_total_v6": 0.7642836719751358, "step": 2031 }, { "epoch": 0.46864144144144143, "grad_norm": 22.792314529418945, "learning_rate": 8.699510515387222e-06, "loss": 12.483451843261719, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2721285577863455, "loss_total_v6": 0.7802156880497932, "step": 2032 }, { "epoch": 0.4688720720720721, "grad_norm": 23.12485694885254, "learning_rate": 8.69815547489305e-06, "loss": 12.666200637817383, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2850762605667114, "loss_total_v6": 0.7916375249624252, "step": 2033 }, { "epoch": 0.4691027027027027, "grad_norm": 26.043094635009766, "learning_rate": 8.696799834475618e-06, "loss": 12.95036506652832, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.3070022314786911, "loss_total_v6": 0.8093978241086006, "step": 2034 }, { "epoch": 0.4693333333333333, "grad_norm": 22.30657958984375, "learning_rate": 8.695443594354843e-06, "loss": 12.534523010253906, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.27643437683582306, "loss_total_v6": 0.7834076657891273, "step": 2035 }, { "epoch": 0.46956396396396394, "grad_norm": 24.102235794067383, "learning_rate": 8.694086754750738e-06, "loss": 12.455194473266602, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2719494625926018, "loss_total_v6": 0.7784497067332268, "step": 2036 }, { "epoch": 0.4697945945945946, "grad_norm": 22.180652618408203, "learning_rate": 8.692729315883409e-06, "loss": 12.572214126586914, "loss_alignment": 1.02294921875, "loss_alignment_w": 0.511474609375, "loss_sft": 0.27348005771636963, "loss_total_v6": 0.7857633680105209, "step": 2037 }, { "epoch": 0.4700252252252252, "grad_norm": 23.81884765625, "learning_rate": 8.691371277973066e-06, "loss": 12.283748626708984, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.26582692191004753, "loss_total_v6": 0.7677342593669891, "step": 2038 }, { "epoch": 0.47025585585585583, "grad_norm": 24.445491790771484, "learning_rate": 8.690012641240014e-06, "loss": 12.568509101867676, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.26954064331948757, "loss_total_v6": 0.7855318412184715, "step": 2039 }, { "epoch": 0.4704864864864865, "grad_norm": 28.840293884277344, "learning_rate": 8.688653405904653e-06, "loss": 12.327472686767578, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.2759297117590904, "loss_total_v6": 0.7704670503735542, "step": 2040 }, { "epoch": 0.4707171171171171, "grad_norm": 21.83655548095703, "learning_rate": 8.687293572187478e-06, "loss": 12.408485412597656, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.278109023347497, "loss_total_v6": 0.7755302935838699, "step": 2041 }, { "epoch": 0.4709477477477477, "grad_norm": 22.799278259277344, "learning_rate": 8.685933140309088e-06, "loss": 12.950361251831055, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.3036298118531704, "loss_total_v6": 0.8093976303935051, "step": 2042 }, { "epoch": 0.4711783783783784, "grad_norm": 22.79295539855957, "learning_rate": 8.684572110490174e-06, "loss": 12.447556495666504, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2646360583603382, "loss_total_v6": 0.7779722884297371, "step": 2043 }, { "epoch": 0.471409009009009, "grad_norm": 24.239877700805664, "learning_rate": 8.683210482951528e-06, "loss": 13.419806480407715, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.3340076394379139, "loss_total_v6": 0.8387379199266434, "step": 2044 }, { "epoch": 0.4716396396396396, "grad_norm": 20.51837730407715, "learning_rate": 8.681848257914031e-06, "loss": 12.063830375671387, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2488319668918848, "loss_total_v6": 0.7539894059300423, "step": 2045 }, { "epoch": 0.4718702702702703, "grad_norm": 20.699464797973633, "learning_rate": 8.680485435598674e-06, "loss": 12.60870361328125, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2834053486585617, "loss_total_v6": 0.7880439981818199, "step": 2046 }, { "epoch": 0.4721009009009009, "grad_norm": 23.755138397216797, "learning_rate": 8.679122016226529e-06, "loss": 12.256492614746094, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.25736377015709877, "loss_total_v6": 0.7660307660698891, "step": 2047 }, { "epoch": 0.4723315315315315, "grad_norm": 21.559579849243164, "learning_rate": 8.677758000018777e-06, "loss": 12.27071762084961, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.26408172212541103, "loss_total_v6": 0.7669198513031006, "step": 2048 }, { "epoch": 0.4725621621621622, "grad_norm": 23.334884643554688, "learning_rate": 8.676393387196692e-06, "loss": 12.478238105773926, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.27918798476457596, "loss_total_v6": 0.779889889061451, "step": 2049 }, { "epoch": 0.4727927927927928, "grad_norm": 23.940231323242188, "learning_rate": 8.675028177981643e-06, "loss": 12.81313705444336, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.29549573734402657, "loss_total_v6": 0.8008210510015488, "step": 2050 }, { "epoch": 0.4730234234234234, "grad_norm": 23.846811294555664, "learning_rate": 8.6736623725951e-06, "loss": 13.039567947387695, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.30343736335635185, "loss_total_v6": 0.8149729892611504, "step": 2051 }, { "epoch": 0.4732540540540541, "grad_norm": 22.80781364440918, "learning_rate": 8.672295971258624e-06, "loss": 12.79777717590332, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.28531941026449203, "loss_total_v6": 0.7998610436916351, "step": 2052 }, { "epoch": 0.4734846846846847, "grad_norm": 22.84781837463379, "learning_rate": 8.670928974193879e-06, "loss": 12.24422836303711, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.2672020774334669, "loss_total_v6": 0.7652642205357552, "step": 2053 }, { "epoch": 0.4737153153153153, "grad_norm": 21.93843650817871, "learning_rate": 8.66956138162262e-06, "loss": 13.080978393554688, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.3110913895070553, "loss_total_v6": 0.8175611570477486, "step": 2054 }, { "epoch": 0.47394594594594597, "grad_norm": 21.761579513549805, "learning_rate": 8.668193193766703e-06, "loss": 12.464006423950195, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.27652846090495586, "loss_total_v6": 0.7790003940463066, "step": 2055 }, { "epoch": 0.4741765765765766, "grad_norm": 21.22752571105957, "learning_rate": 8.666824410848075e-06, "loss": 12.41943359375, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.27592465467751026, "loss_total_v6": 0.7762145847082138, "step": 2056 }, { "epoch": 0.4744072072072072, "grad_norm": 20.86997413635254, "learning_rate": 8.665455033088787e-06, "loss": 13.231377601623535, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.32844121009111404, "loss_total_v6": 0.8269610926508904, "step": 2057 }, { "epoch": 0.47463783783783786, "grad_norm": 25.507593154907227, "learning_rate": 8.66408506071098e-06, "loss": 12.257490158081055, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.2652539275586605, "loss_total_v6": 0.7660931423306465, "step": 2058 }, { "epoch": 0.4748684684684685, "grad_norm": 21.990921020507812, "learning_rate": 8.662714493936895e-06, "loss": 11.972053527832031, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.24358413368463516, "loss_total_v6": 0.748253345489502, "step": 2059 }, { "epoch": 0.4750990990990991, "grad_norm": 25.847408294677734, "learning_rate": 8.661343332988869e-06, "loss": 12.919251441955566, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2995797246694565, "loss_total_v6": 0.8074532449245453, "step": 2060 }, { "epoch": 0.47532972972972976, "grad_norm": 23.162260055541992, "learning_rate": 8.659971578089333e-06, "loss": 12.616525650024414, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2749983072280884, "loss_total_v6": 0.7885328456759453, "step": 2061 }, { "epoch": 0.47556036036036037, "grad_norm": 22.386646270751953, "learning_rate": 8.658599229460818e-06, "loss": 12.386724472045898, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.26554902270436287, "loss_total_v6": 0.7741702571511269, "step": 2062 }, { "epoch": 0.475790990990991, "grad_norm": 22.260448455810547, "learning_rate": 8.657226287325951e-06, "loss": 12.664263725280762, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.2931643985211849, "loss_total_v6": 0.7915164530277252, "step": 2063 }, { "epoch": 0.47602162162162165, "grad_norm": 24.587488174438477, "learning_rate": 8.65585275190745e-06, "loss": 13.16889476776123, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.31020802818238735, "loss_total_v6": 0.8230559453368187, "step": 2064 }, { "epoch": 0.47625225225225226, "grad_norm": 23.933256149291992, "learning_rate": 8.654478623428136e-06, "loss": 12.58542251586914, "loss_alignment": 1.01220703125, "loss_alignment_w": 0.506103515625, "loss_sft": 0.2795241344720125, "loss_total_v6": 0.7865889519453049, "step": 2065 }, { "epoch": 0.4764828828828829, "grad_norm": 23.46670150756836, "learning_rate": 8.653103902110923e-06, "loss": 12.33151626586914, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2587720789015293, "loss_total_v6": 0.7707197368144989, "step": 2066 }, { "epoch": 0.4767135135135135, "grad_norm": 24.67242431640625, "learning_rate": 8.65172858817882e-06, "loss": 12.385683059692383, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.2729302942752838, "loss_total_v6": 0.7741051986813545, "step": 2067 }, { "epoch": 0.47694414414414416, "grad_norm": 22.889829635620117, "learning_rate": 8.650352681854934e-06, "loss": 12.325090408325195, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2639247067272663, "loss_total_v6": 0.7703181505203247, "step": 2068 }, { "epoch": 0.47717477477477477, "grad_norm": 24.593475341796875, "learning_rate": 8.648976183362468e-06, "loss": 12.708454132080078, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.2882511969655752, "loss_total_v6": 0.7942784056067467, "step": 2069 }, { "epoch": 0.4774054054054054, "grad_norm": 22.475500106811523, "learning_rate": 8.64759909292472e-06, "loss": 12.631178855895996, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2788132783025503, "loss_total_v6": 0.7894486635923386, "step": 2070 }, { "epoch": 0.47763603603603605, "grad_norm": 25.80147933959961, "learning_rate": 8.646221410765088e-06, "loss": 12.405990600585938, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.26284691132605076, "loss_total_v6": 0.7753743827342987, "step": 2071 }, { "epoch": 0.47786666666666666, "grad_norm": 23.671483993530273, "learning_rate": 8.644843137107058e-06, "loss": 12.211523056030273, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.26138913072645664, "loss_total_v6": 0.7632201835513115, "step": 2072 }, { "epoch": 0.4780972972972973, "grad_norm": 20.860218048095703, "learning_rate": 8.64346427217422e-06, "loss": 12.969592094421387, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2992774583399296, "loss_total_v6": 0.8105995059013367, "step": 2073 }, { "epoch": 0.47832792792792794, "grad_norm": 23.7706356048584, "learning_rate": 8.642084816190254e-06, "loss": 12.09827995300293, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.25284662656486034, "loss_total_v6": 0.7561425119638443, "step": 2074 }, { "epoch": 0.47855855855855856, "grad_norm": 20.36954116821289, "learning_rate": 8.640704769378943e-06, "loss": 12.23765754699707, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.25847541354596615, "loss_total_v6": 0.7648535892367363, "step": 2075 }, { "epoch": 0.47878918918918917, "grad_norm": 20.70817756652832, "learning_rate": 8.639324131964156e-06, "loss": 12.56396484375, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.28727724961936474, "loss_total_v6": 0.785247839987278, "step": 2076 }, { "epoch": 0.47901981981981984, "grad_norm": 22.00005340576172, "learning_rate": 8.637942904169867e-06, "loss": 12.227540969848633, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.26496896892786026, "loss_total_v6": 0.764221303164959, "step": 2077 }, { "epoch": 0.47925045045045045, "grad_norm": 21.388877868652344, "learning_rate": 8.63656108622014e-06, "loss": 12.209273338317871, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.26123325899243355, "loss_total_v6": 0.7630795761942863, "step": 2078 }, { "epoch": 0.47948108108108106, "grad_norm": 22.33111000061035, "learning_rate": 8.635178678339137e-06, "loss": 12.404806137084961, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.27070751786231995, "loss_total_v6": 0.775300420820713, "step": 2079 }, { "epoch": 0.47971171171171173, "grad_norm": 22.35472297668457, "learning_rate": 8.633795680751116e-06, "loss": 12.771749496459961, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.29422126710414886, "loss_total_v6": 0.7982343286275864, "step": 2080 }, { "epoch": 0.47994234234234234, "grad_norm": 22.843814849853516, "learning_rate": 8.63241209368043e-06, "loss": 12.757019996643066, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.28681566566228867, "loss_total_v6": 0.7973137125372887, "step": 2081 }, { "epoch": 0.48017297297297296, "grad_norm": 22.772836685180664, "learning_rate": 8.631027917351528e-06, "loss": 12.725055694580078, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.2927829772233963, "loss_total_v6": 0.7953159660100937, "step": 2082 }, { "epoch": 0.4804036036036036, "grad_norm": 22.450868606567383, "learning_rate": 8.629643151988951e-06, "loss": 12.491512298583984, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.27684386260807514, "loss_total_v6": 0.7807195633649826, "step": 2083 }, { "epoch": 0.48063423423423424, "grad_norm": 22.701133728027344, "learning_rate": 8.628257797817344e-06, "loss": 13.203584671020508, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.31858644261956215, "loss_total_v6": 0.8252240121364594, "step": 2084 }, { "epoch": 0.48086486486486485, "grad_norm": 26.719314575195312, "learning_rate": 8.626871855061438e-06, "loss": 12.77958869934082, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.29718307219445705, "loss_total_v6": 0.7987242341041565, "step": 2085 }, { "epoch": 0.4810954954954955, "grad_norm": 20.999605178833008, "learning_rate": 8.625485323946068e-06, "loss": 12.822446823120117, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2945823073387146, "loss_total_v6": 0.8014029562473297, "step": 2086 }, { "epoch": 0.48132612612612613, "grad_norm": 22.73838996887207, "learning_rate": 8.624098204696156e-06, "loss": 12.121186256408691, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.2581387311220169, "loss_total_v6": 0.7575741112232208, "step": 2087 }, { "epoch": 0.48155675675675674, "grad_norm": 21.625736236572266, "learning_rate": 8.622710497536727e-06, "loss": 12.627674102783203, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.28185964189469814, "loss_total_v6": 0.7892296463251114, "step": 2088 }, { "epoch": 0.4817873873873874, "grad_norm": 22.554887771606445, "learning_rate": 8.621322202692894e-06, "loss": 12.282670974731445, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.26681239530444145, "loss_total_v6": 0.7676669061183929, "step": 2089 }, { "epoch": 0.482018018018018, "grad_norm": 22.71713638305664, "learning_rate": 8.619933320389872e-06, "loss": 12.240682601928711, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.2626928240060806, "loss_total_v6": 0.7650426551699638, "step": 2090 }, { "epoch": 0.48224864864864864, "grad_norm": 23.346538543701172, "learning_rate": 8.618543850852969e-06, "loss": 12.230119705200195, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.2620936334133148, "loss_total_v6": 0.7643824592232704, "step": 2091 }, { "epoch": 0.4824792792792793, "grad_norm": 24.132740020751953, "learning_rate": 8.617153794307588e-06, "loss": 12.934101104736328, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.303178034722805, "loss_total_v6": 0.8083812892436981, "step": 2092 }, { "epoch": 0.4827099099099099, "grad_norm": 23.544313430786133, "learning_rate": 8.615763150979227e-06, "loss": 12.294837951660156, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.26462790556252, "loss_total_v6": 0.7684273421764374, "step": 2093 }, { "epoch": 0.48294054054054053, "grad_norm": 23.728271484375, "learning_rate": 8.614371921093479e-06, "loss": 13.077133178710938, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.3100424036383629, "loss_total_v6": 0.8173208311200142, "step": 2094 }, { "epoch": 0.4831711711711712, "grad_norm": 22.42043113708496, "learning_rate": 8.612980104876032e-06, "loss": 12.68347454071045, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.2932054400444031, "loss_total_v6": 0.7927171587944031, "step": 2095 }, { "epoch": 0.4834018018018018, "grad_norm": 23.678974151611328, "learning_rate": 8.61158770255267e-06, "loss": 12.760015487670898, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2931064162403345, "loss_total_v6": 0.7975009456276894, "step": 2096 }, { "epoch": 0.4836324324324324, "grad_norm": 22.194337844848633, "learning_rate": 8.610194714349271e-06, "loss": 12.588001251220703, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2832710910588503, "loss_total_v6": 0.7867500707507133, "step": 2097 }, { "epoch": 0.48386306306306304, "grad_norm": 23.71234893798828, "learning_rate": 8.608801140491811e-06, "loss": 12.334815979003906, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2599396575242281, "loss_total_v6": 0.7709259763360023, "step": 2098 }, { "epoch": 0.4840936936936937, "grad_norm": 21.805124282836914, "learning_rate": 8.607406981206357e-06, "loss": 12.136091232299805, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.2566441800445318, "loss_total_v6": 0.7585057392716408, "step": 2099 }, { "epoch": 0.4843243243243243, "grad_norm": 22.729997634887695, "learning_rate": 8.606012236719073e-06, "loss": 12.479461669921875, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.27479361556470394, "loss_total_v6": 0.7799663469195366, "step": 2100 }, { "epoch": 0.48455495495495493, "grad_norm": 22.560232162475586, "learning_rate": 8.604616907256218e-06, "loss": 12.466320037841797, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.26760938204824924, "loss_total_v6": 0.7791450247168541, "step": 2101 }, { "epoch": 0.4847855855855856, "grad_norm": 23.11582374572754, "learning_rate": 8.603220993044145e-06, "loss": 12.193343162536621, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.26127522997558117, "loss_total_v6": 0.7620839402079582, "step": 2102 }, { "epoch": 0.4850162162162162, "grad_norm": 24.899845123291016, "learning_rate": 8.6018244943093e-06, "loss": 12.340113639831543, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.267564507201314, "loss_total_v6": 0.7712571173906326, "step": 2103 }, { "epoch": 0.4852468468468468, "grad_norm": 23.09221076965332, "learning_rate": 8.600427411278234e-06, "loss": 12.358643531799316, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2659149784594774, "loss_total_v6": 0.7724152430891991, "step": 2104 }, { "epoch": 0.4854774774774775, "grad_norm": 23.99712562561035, "learning_rate": 8.599029744177577e-06, "loss": 12.94035530090332, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.30140223167836666, "loss_total_v6": 0.8087722137570381, "step": 2105 }, { "epoch": 0.4857081081081081, "grad_norm": 19.307960510253906, "learning_rate": 8.597631493234065e-06, "loss": 11.975042343139648, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.2511104438453913, "loss_total_v6": 0.7484401389956474, "step": 2106 }, { "epoch": 0.4859387387387387, "grad_norm": 22.128498077392578, "learning_rate": 8.596232658674524e-06, "loss": 12.31629467010498, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2624747157096863, "loss_total_v6": 0.7697684243321419, "step": 2107 }, { "epoch": 0.4861693693693694, "grad_norm": 22.133752822875977, "learning_rate": 8.594833240725877e-06, "loss": 12.306139945983887, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2629997003823519, "loss_total_v6": 0.7691337540745735, "step": 2108 }, { "epoch": 0.4864, "grad_norm": 29.722131729125977, "learning_rate": 8.593433239615141e-06, "loss": 12.36157512664795, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.27200336940586567, "loss_total_v6": 0.7725984677672386, "step": 2109 }, { "epoch": 0.4866306306306306, "grad_norm": 21.13983917236328, "learning_rate": 8.59203265556943e-06, "loss": 12.345442771911621, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2665089797228575, "loss_total_v6": 0.7715901508927345, "step": 2110 }, { "epoch": 0.4868612612612613, "grad_norm": 20.902006149291992, "learning_rate": 8.590631488815945e-06, "loss": 12.332096099853516, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2646219711750746, "loss_total_v6": 0.7707559987902641, "step": 2111 }, { "epoch": 0.4870918918918919, "grad_norm": 22.529598236083984, "learning_rate": 8.58922973958199e-06, "loss": 12.342828750610352, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.27145733684301376, "loss_total_v6": 0.7714268416166306, "step": 2112 }, { "epoch": 0.4873225225225225, "grad_norm": 22.738224029541016, "learning_rate": 8.587827408094958e-06, "loss": 12.278412818908691, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.2693233620375395, "loss_total_v6": 0.767400786280632, "step": 2113 }, { "epoch": 0.4875531531531532, "grad_norm": 24.72222137451172, "learning_rate": 8.58642449458234e-06, "loss": 12.48136043548584, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.28264848701655865, "loss_total_v6": 0.7800850421190262, "step": 2114 }, { "epoch": 0.4877837837837838, "grad_norm": 26.347396850585938, "learning_rate": 8.585020999271718e-06, "loss": 12.885991096496582, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.3029330335557461, "loss_total_v6": 0.805374450981617, "step": 2115 }, { "epoch": 0.4880144144144144, "grad_norm": 26.49907112121582, "learning_rate": 8.583616922390772e-06, "loss": 11.992783546447754, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.24922848865389824, "loss_total_v6": 0.7495489493012428, "step": 2116 }, { "epoch": 0.48824504504504507, "grad_norm": 26.433399200439453, "learning_rate": 8.582212264167274e-06, "loss": 12.14052963256836, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2521303091198206, "loss_total_v6": 0.7587831020355225, "step": 2117 }, { "epoch": 0.4884756756756757, "grad_norm": 22.223487854003906, "learning_rate": 8.58080702482909e-06, "loss": 12.956286430358887, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.3081199564039707, "loss_total_v6": 0.8097679167985916, "step": 2118 }, { "epoch": 0.4887063063063063, "grad_norm": 23.20102882385254, "learning_rate": 8.579401204604184e-06, "loss": 12.441469192504883, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.2751962002366781, "loss_total_v6": 0.7775918543338776, "step": 2119 }, { "epoch": 0.48893693693693696, "grad_norm": 23.610254287719727, "learning_rate": 8.577994803720605e-06, "loss": 12.2238130569458, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.2635000329464674, "loss_total_v6": 0.7639883011579514, "step": 2120 }, { "epoch": 0.4891675675675676, "grad_norm": 23.574840545654297, "learning_rate": 8.576587822406511e-06, "loss": 12.916976928710938, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.30032248981297016, "loss_total_v6": 0.8073110207915306, "step": 2121 }, { "epoch": 0.4893981981981982, "grad_norm": 22.03323745727539, "learning_rate": 8.57518026089014e-06, "loss": 12.537035942077637, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.2775527834892273, "loss_total_v6": 0.7835647240281105, "step": 2122 }, { "epoch": 0.48962882882882885, "grad_norm": 22.178707122802734, "learning_rate": 8.573772119399832e-06, "loss": 12.612586975097656, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2829308584332466, "loss_total_v6": 0.7882867008447647, "step": 2123 }, { "epoch": 0.48985945945945947, "grad_norm": 26.4283390045166, "learning_rate": 8.572363398164017e-06, "loss": 12.931051254272461, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.296258307993412, "loss_total_v6": 0.8081906586885452, "step": 2124 }, { "epoch": 0.4900900900900901, "grad_norm": 21.878395080566406, "learning_rate": 8.570954097411224e-06, "loss": 12.20682430267334, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.261446425691247, "loss_total_v6": 0.7629265263676643, "step": 2125 }, { "epoch": 0.4903207207207207, "grad_norm": 23.131301879882812, "learning_rate": 8.569544217370071e-06, "loss": 13.373491287231445, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.32243073359131813, "loss_total_v6": 0.8358432129025459, "step": 2126 }, { "epoch": 0.49055135135135136, "grad_norm": 22.474159240722656, "learning_rate": 8.568133758269272e-06, "loss": 11.989222526550293, "loss_alignment": 0.9873046875, "loss_alignment_w": 0.49365234375, "loss_sft": 0.25681847520172596, "loss_total_v6": 0.7493264302611351, "step": 2127 }, { "epoch": 0.490781981981982, "grad_norm": 25.209213256835938, "learning_rate": 8.566722720337634e-06, "loss": 12.344093322753906, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.26358651369810104, "loss_total_v6": 0.7715058103203773, "step": 2128 }, { "epoch": 0.4910126126126126, "grad_norm": 22.94028091430664, "learning_rate": 8.565311103804063e-06, "loss": 13.264078140258789, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.3167520649731159, "loss_total_v6": 0.8290048837661743, "step": 2129 }, { "epoch": 0.49124324324324325, "grad_norm": 21.514467239379883, "learning_rate": 8.56389890889755e-06, "loss": 11.92258358001709, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.242308109998703, "loss_total_v6": 0.7451614812016487, "step": 2130 }, { "epoch": 0.49147387387387387, "grad_norm": 23.017759323120117, "learning_rate": 8.562486135847186e-06, "loss": 12.154902458190918, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2500378768891096, "loss_total_v6": 0.75968137383461, "step": 2131 }, { "epoch": 0.4917045045045045, "grad_norm": 23.06482696533203, "learning_rate": 8.561072784882156e-06, "loss": 12.485231399536133, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2659378834068775, "loss_total_v6": 0.7803269550204277, "step": 2132 }, { "epoch": 0.49193513513513515, "grad_norm": 22.072059631347656, "learning_rate": 8.559658856231734e-06, "loss": 12.825382232666016, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.30501962825655937, "loss_total_v6": 0.8015863671898842, "step": 2133 }, { "epoch": 0.49216576576576576, "grad_norm": 23.527210235595703, "learning_rate": 8.558244350125293e-06, "loss": 13.061860084533691, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.3143063597381115, "loss_total_v6": 0.8163662850856781, "step": 2134 }, { "epoch": 0.4923963963963964, "grad_norm": 25.44088363647461, "learning_rate": 8.556829266792297e-06, "loss": 12.530118942260742, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.28102669678628445, "loss_total_v6": 0.7831323891878128, "step": 2135 }, { "epoch": 0.49262702702702704, "grad_norm": 24.402067184448242, "learning_rate": 8.555413606462302e-06, "loss": 12.859590530395508, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2909070495516062, "loss_total_v6": 0.8037244230508804, "step": 2136 }, { "epoch": 0.49285765765765766, "grad_norm": 22.928123474121094, "learning_rate": 8.553997369364964e-06, "loss": 12.750936508178711, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.29282892867922783, "loss_total_v6": 0.7969335466623306, "step": 2137 }, { "epoch": 0.49308828828828827, "grad_norm": 21.158212661743164, "learning_rate": 8.552580555730023e-06, "loss": 11.994363784790039, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.24682492204010487, "loss_total_v6": 0.7496477589011192, "step": 2138 }, { "epoch": 0.49331891891891894, "grad_norm": 21.326906204223633, "learning_rate": 8.551163165787322e-06, "loss": 12.5783109664917, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2725793980062008, "loss_total_v6": 0.7861444428563118, "step": 2139 }, { "epoch": 0.49354954954954955, "grad_norm": 22.27639389038086, "learning_rate": 8.549745199766792e-06, "loss": 12.103311538696289, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.25465638376772404, "loss_total_v6": 0.7564569488167763, "step": 2140 }, { "epoch": 0.49378018018018016, "grad_norm": 22.53899574279785, "learning_rate": 8.548326657898458e-06, "loss": 12.64673137664795, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.295532388612628, "loss_total_v6": 0.7904207035899162, "step": 2141 }, { "epoch": 0.49401081081081083, "grad_norm": 22.98369598388672, "learning_rate": 8.54690754041244e-06, "loss": 12.475086212158203, "loss_alignment": 0.98291015625, "loss_alignment_w": 0.491455078125, "loss_sft": 0.28663562797009945, "loss_total_v6": 0.7796928584575653, "step": 2142 }, { "epoch": 0.49424144144144144, "grad_norm": 25.869037628173828, "learning_rate": 8.545487847538949e-06, "loss": 12.541444778442383, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.2858086787164211, "loss_total_v6": 0.7838403135538101, "step": 2143 }, { "epoch": 0.49447207207207206, "grad_norm": 22.275299072265625, "learning_rate": 8.544067579508292e-06, "loss": 12.204967498779297, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.2659689914435148, "loss_total_v6": 0.7628104388713837, "step": 2144 }, { "epoch": 0.4947027027027027, "grad_norm": 21.70368194580078, "learning_rate": 8.542646736550868e-06, "loss": 13.06224536895752, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.31297232769429684, "loss_total_v6": 0.8163903281092644, "step": 2145 }, { "epoch": 0.49493333333333334, "grad_norm": 25.161842346191406, "learning_rate": 8.541225318897168e-06, "loss": 12.391883850097656, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.26320127584040165, "loss_total_v6": 0.7744927629828453, "step": 2146 }, { "epoch": 0.49516396396396395, "grad_norm": 20.95164680480957, "learning_rate": 8.539803326777782e-06, "loss": 12.279518127441406, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.2685075104236603, "loss_total_v6": 0.7674699053168297, "step": 2147 }, { "epoch": 0.4953945945945946, "grad_norm": 24.550825119018555, "learning_rate": 8.538380760423384e-06, "loss": 12.401712417602539, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2646700404584408, "loss_total_v6": 0.7751070335507393, "step": 2148 }, { "epoch": 0.49562522522522523, "grad_norm": 25.662311553955078, "learning_rate": 8.536957620064747e-06, "loss": 12.90262508392334, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.29867781326174736, "loss_total_v6": 0.8064140677452087, "step": 2149 }, { "epoch": 0.49585585585585584, "grad_norm": 22.318307876586914, "learning_rate": 8.535533905932739e-06, "loss": 12.617998123168945, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.27953062392771244, "loss_total_v6": 0.7886249050498009, "step": 2150 }, { "epoch": 0.4960864864864865, "grad_norm": 23.430089950561523, "learning_rate": 8.534109618258315e-06, "loss": 12.784173011779785, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.3031764645129442, "loss_total_v6": 0.7990108132362366, "step": 2151 }, { "epoch": 0.4963171171171171, "grad_norm": 24.65864372253418, "learning_rate": 8.532684757272527e-06, "loss": 12.945605278015137, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.30113522708415985, "loss_total_v6": 0.8091003075242043, "step": 2152 }, { "epoch": 0.49654774774774774, "grad_norm": 22.89068031311035, "learning_rate": 8.53125932320652e-06, "loss": 12.954975128173828, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.31296661868691444, "loss_total_v6": 0.8096859976649284, "step": 2153 }, { "epoch": 0.4967783783783784, "grad_norm": 21.202640533447266, "learning_rate": 8.52983331629153e-06, "loss": 12.578636169433594, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.27806235663592815, "loss_total_v6": 0.7861647754907608, "step": 2154 }, { "epoch": 0.497009009009009, "grad_norm": 22.21455192565918, "learning_rate": 8.528406736758888e-06, "loss": 12.752753257751465, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2880749609321356, "loss_total_v6": 0.7970471009612083, "step": 2155 }, { "epoch": 0.49723963963963963, "grad_norm": 23.767728805541992, "learning_rate": 8.526979584840016e-06, "loss": 12.588237762451172, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.27794531360268593, "loss_total_v6": 0.78676488250494, "step": 2156 }, { "epoch": 0.49747027027027024, "grad_norm": 20.099040985107422, "learning_rate": 8.52555186076643e-06, "loss": 12.920570373535156, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2961220517754555, "loss_total_v6": 0.8075356557965279, "step": 2157 }, { "epoch": 0.4977009009009009, "grad_norm": 25.00852394104004, "learning_rate": 8.524123564769741e-06, "loss": 12.887682914733887, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2974998354911804, "loss_total_v6": 0.8054801970720291, "step": 2158 }, { "epoch": 0.4979315315315315, "grad_norm": 21.819068908691406, "learning_rate": 8.522694697081648e-06, "loss": 12.593474388122559, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.27514454536139965, "loss_total_v6": 0.7870921418070793, "step": 2159 }, { "epoch": 0.49816216216216214, "grad_norm": 22.65491485595703, "learning_rate": 8.521265257933948e-06, "loss": 12.877056121826172, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.2996585760265589, "loss_total_v6": 0.8048160523176193, "step": 2160 }, { "epoch": 0.4983927927927928, "grad_norm": 21.21263885498047, "learning_rate": 8.519835247558523e-06, "loss": 12.850586891174316, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.294387873262167, "loss_total_v6": 0.803161695599556, "step": 2161 }, { "epoch": 0.4986234234234234, "grad_norm": 22.218130111694336, "learning_rate": 8.518404666187356e-06, "loss": 12.012290000915527, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.25586448796093464, "loss_total_v6": 0.7507680952548981, "step": 2162 }, { "epoch": 0.49885405405405403, "grad_norm": 22.608604431152344, "learning_rate": 8.51697351405252e-06, "loss": 12.754817962646484, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.2922474816441536, "loss_total_v6": 0.7971761152148247, "step": 2163 }, { "epoch": 0.4990846846846847, "grad_norm": 24.753589630126953, "learning_rate": 8.515541791386176e-06, "loss": 12.574075698852539, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.28690207935869694, "loss_total_v6": 0.7858797386288643, "step": 2164 }, { "epoch": 0.4993153153153153, "grad_norm": 23.542261123657227, "learning_rate": 8.514109498420586e-06, "loss": 12.921009063720703, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.31046218425035477, "loss_total_v6": 0.8075630590319633, "step": 2165 }, { "epoch": 0.4995459459459459, "grad_norm": 19.77745819091797, "learning_rate": 8.512676635388098e-06, "loss": 12.268505096435547, "loss_alignment": 0.9892578125, "loss_alignment_w": 0.49462890625, "loss_sft": 0.2721984088420868, "loss_total_v6": 0.7667815387248993, "step": 2166 }, { "epoch": 0.4997765765765766, "grad_norm": 23.875152587890625, "learning_rate": 8.511243202521151e-06, "loss": 12.796802520751953, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.29943398013710976, "loss_total_v6": 0.7998001649975777, "step": 2167 }, { "epoch": 0.5000072072072073, "grad_norm": 22.525590896606445, "learning_rate": 8.509809200052286e-06, "loss": 12.548961639404297, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.27477335557341576, "loss_total_v6": 0.784310095012188, "step": 2168 }, { "epoch": 0.5002378378378378, "grad_norm": 24.73114776611328, "learning_rate": 8.508374628214124e-06, "loss": 12.64391803741455, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.28285965509712696, "loss_total_v6": 0.79024488478899, "step": 2169 }, { "epoch": 0.5004684684684685, "grad_norm": 22.827896118164062, "learning_rate": 8.506939487239389e-06, "loss": 12.13836669921875, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.25240704976022243, "loss_total_v6": 0.7586478963494301, "step": 2170 }, { "epoch": 0.5006990990990992, "grad_norm": 23.591102600097656, "learning_rate": 8.505503777360889e-06, "loss": 12.873644828796387, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2920143473893404, "loss_total_v6": 0.8046028316020966, "step": 2171 }, { "epoch": 0.5009297297297297, "grad_norm": 23.04712677001953, "learning_rate": 8.504067498811533e-06, "loss": 12.559310913085938, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.2819356508553028, "loss_total_v6": 0.7849569097161293, "step": 2172 }, { "epoch": 0.5011603603603604, "grad_norm": 24.995525360107422, "learning_rate": 8.502630651824312e-06, "loss": 12.67477798461914, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.2919600363820791, "loss_total_v6": 0.7921736314892769, "step": 2173 }, { "epoch": 0.5013909909909909, "grad_norm": 23.162416458129883, "learning_rate": 8.50119323663232e-06, "loss": 12.437393188476562, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.2719506900757551, "loss_total_v6": 0.7773370370268822, "step": 2174 }, { "epoch": 0.5016216216216216, "grad_norm": 20.751197814941406, "learning_rate": 8.499755253468732e-06, "loss": 12.200277328491211, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2544606998562813, "loss_total_v6": 0.7625173181295395, "step": 2175 }, { "epoch": 0.5018522522522523, "grad_norm": 22.62650489807129, "learning_rate": 8.498316702566828e-06, "loss": 12.654550552368164, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2883154135197401, "loss_total_v6": 0.7909094095230103, "step": 2176 }, { "epoch": 0.5020828828828828, "grad_norm": 24.652029037475586, "learning_rate": 8.496877584159967e-06, "loss": 12.302618026733398, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2581714168190956, "loss_total_v6": 0.7689136043190956, "step": 2177 }, { "epoch": 0.5023135135135135, "grad_norm": 26.233057022094727, "learning_rate": 8.49543789848161e-06, "loss": 12.743546485900879, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.27588758058845997, "loss_total_v6": 0.7964716777205467, "step": 2178 }, { "epoch": 0.5025441441441442, "grad_norm": 21.836681365966797, "learning_rate": 8.493997645765302e-06, "loss": 12.310651779174805, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2614812105894089, "loss_total_v6": 0.7694157510995865, "step": 2179 }, { "epoch": 0.5027747747747747, "grad_norm": 22.64543342590332, "learning_rate": 8.492556826244687e-06, "loss": 12.944299697875977, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.30207591876387596, "loss_total_v6": 0.8090186938643456, "step": 2180 }, { "epoch": 0.5030054054054054, "grad_norm": 23.82621192932129, "learning_rate": 8.491115440153495e-06, "loss": 12.408586502075195, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.27367505617439747, "loss_total_v6": 0.7755366712808609, "step": 2181 }, { "epoch": 0.5032360360360361, "grad_norm": 23.79024314880371, "learning_rate": 8.489673487725556e-06, "loss": 13.221346855163574, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.31333368085324764, "loss_total_v6": 0.8263341709971428, "step": 2182 }, { "epoch": 0.5034666666666666, "grad_norm": 22.955507278442383, "learning_rate": 8.488230969194785e-06, "loss": 12.377596855163574, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.26584835536777973, "loss_total_v6": 0.7735998257994652, "step": 2183 }, { "epoch": 0.5036972972972973, "grad_norm": 21.16216468811035, "learning_rate": 8.48678788479519e-06, "loss": 12.692344665527344, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2833380475640297, "loss_total_v6": 0.7932715266942978, "step": 2184 }, { "epoch": 0.503927927927928, "grad_norm": 23.51577377319336, "learning_rate": 8.485344234760869e-06, "loss": 12.757786750793457, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2916701827198267, "loss_total_v6": 0.7973616868257523, "step": 2185 }, { "epoch": 0.5041585585585585, "grad_norm": 21.14394760131836, "learning_rate": 8.483900019326017e-06, "loss": 12.50757884979248, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.27885500714182854, "loss_total_v6": 0.7817236632108688, "step": 2186 }, { "epoch": 0.5043891891891892, "grad_norm": 22.788469314575195, "learning_rate": 8.482455238724919e-06, "loss": 12.735410690307617, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.28956975042819977, "loss_total_v6": 0.7959631904959679, "step": 2187 }, { "epoch": 0.5046198198198198, "grad_norm": 22.515932083129883, "learning_rate": 8.481009893191947e-06, "loss": 12.6466703414917, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2873040810227394, "loss_total_v6": 0.79041688144207, "step": 2188 }, { "epoch": 0.5048504504504504, "grad_norm": 24.091930389404297, "learning_rate": 8.479563982961572e-06, "loss": 12.534102439880371, "loss_alignment": 1.00732421875, "loss_alignment_w": 0.503662109375, "loss_sft": 0.27813238464295864, "loss_total_v6": 0.7833813950419426, "step": 2189 }, { "epoch": 0.5050810810810811, "grad_norm": 22.285249710083008, "learning_rate": 8.478117508268352e-06, "loss": 12.822802543640137, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2873871196061373, "loss_total_v6": 0.8014251589775085, "step": 2190 }, { "epoch": 0.5053117117117117, "grad_norm": 23.917430877685547, "learning_rate": 8.476670469346935e-06, "loss": 12.428621292114258, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.280069375410676, "loss_total_v6": 0.7767887935042381, "step": 2191 }, { "epoch": 0.5055423423423423, "grad_norm": 20.42778205871582, "learning_rate": 8.475222866432065e-06, "loss": 12.290559768676758, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2574788089841604, "loss_total_v6": 0.7681599631905556, "step": 2192 }, { "epoch": 0.505772972972973, "grad_norm": 21.722087860107422, "learning_rate": 8.473774699758574e-06, "loss": 13.097314834594727, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.3076874390244484, "loss_total_v6": 0.8185821771621704, "step": 2193 }, { "epoch": 0.5060036036036036, "grad_norm": 24.419876098632812, "learning_rate": 8.47232596956139e-06, "loss": 13.181695938110352, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.30809367075562477, "loss_total_v6": 0.8238560035824776, "step": 2194 }, { "epoch": 0.5062342342342342, "grad_norm": 23.487545013427734, "learning_rate": 8.470876676075528e-06, "loss": 12.702779769897461, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2921384982764721, "loss_total_v6": 0.7939237728714943, "step": 2195 }, { "epoch": 0.5064648648648649, "grad_norm": 25.68149185180664, "learning_rate": 8.469426819536093e-06, "loss": 12.67840576171875, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.2938194200396538, "loss_total_v6": 0.7924003824591637, "step": 2196 }, { "epoch": 0.5066954954954955, "grad_norm": 24.87109375, "learning_rate": 8.467976400178289e-06, "loss": 12.599859237670898, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.2868809010833502, "loss_total_v6": 0.7874912098050117, "step": 2197 }, { "epoch": 0.5069261261261261, "grad_norm": 21.348739624023438, "learning_rate": 8.4665254182374e-06, "loss": 12.208715438842773, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.26305997371673584, "loss_total_v6": 0.763044685125351, "step": 2198 }, { "epoch": 0.5071567567567568, "grad_norm": 23.76201057434082, "learning_rate": 8.465073873948812e-06, "loss": 12.459908485412598, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.26737647131085396, "loss_total_v6": 0.7787442728877068, "step": 2199 }, { "epoch": 0.5073873873873874, "grad_norm": 24.8475284576416, "learning_rate": 8.463621767547998e-06, "loss": 12.573107719421387, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2802803497761488, "loss_total_v6": 0.7858192548155785, "step": 2200 }, { "epoch": 0.507618018018018, "grad_norm": 21.33452606201172, "learning_rate": 8.46216909927052e-06, "loss": 12.22050666809082, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.26391902565956116, "loss_total_v6": 0.7637816891074181, "step": 2201 }, { "epoch": 0.5078486486486486, "grad_norm": 22.59601593017578, "learning_rate": 8.460715869352035e-06, "loss": 12.291990280151367, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.2706450019031763, "loss_total_v6": 0.7682493925094604, "step": 2202 }, { "epoch": 0.5080792792792793, "grad_norm": 22.90372657775879, "learning_rate": 8.459262078028287e-06, "loss": 12.96358871459961, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.29116606898605824, "loss_total_v6": 0.810224287211895, "step": 2203 }, { "epoch": 0.5083099099099099, "grad_norm": 23.81117057800293, "learning_rate": 8.457807725535117e-06, "loss": 12.131745338439941, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.25747113674879074, "loss_total_v6": 0.7582340911030769, "step": 2204 }, { "epoch": 0.5085405405405405, "grad_norm": 19.301055908203125, "learning_rate": 8.45635281210845e-06, "loss": 11.868560791015625, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.24375342950224876, "loss_total_v6": 0.7417850717902184, "step": 2205 }, { "epoch": 0.5087711711711712, "grad_norm": 21.182865142822266, "learning_rate": 8.454897337984307e-06, "loss": 12.194299697875977, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.25614703446626663, "loss_total_v6": 0.7621437534689903, "step": 2206 }, { "epoch": 0.5090018018018018, "grad_norm": 23.513504028320312, "learning_rate": 8.453441303398797e-06, "loss": 12.078105926513672, "loss_alignment": 0.9697265625, "loss_alignment_w": 0.48486328125, "loss_sft": 0.2701862119138241, "loss_total_v6": 0.7548816204071045, "step": 2207 }, { "epoch": 0.5092324324324324, "grad_norm": 24.12320327758789, "learning_rate": 8.451984708588122e-06, "loss": 12.68400764465332, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.281397869810462, "loss_total_v6": 0.7927504405379295, "step": 2208 }, { "epoch": 0.5094630630630631, "grad_norm": 21.57454490661621, "learning_rate": 8.450527553788576e-06, "loss": 12.614660263061523, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.28003923036158085, "loss_total_v6": 0.7884162664413452, "step": 2209 }, { "epoch": 0.5096936936936937, "grad_norm": 24.496360778808594, "learning_rate": 8.44906983923654e-06, "loss": 12.471296310424805, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.27274214290082455, "loss_total_v6": 0.7794560045003891, "step": 2210 }, { "epoch": 0.5099243243243243, "grad_norm": 23.011266708374023, "learning_rate": 8.447611565168487e-06, "loss": 12.18274211883545, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.26313032396137714, "loss_total_v6": 0.761421374976635, "step": 2211 }, { "epoch": 0.510154954954955, "grad_norm": 22.5140438079834, "learning_rate": 8.446152731820984e-06, "loss": 11.960099220275879, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2407618034631014, "loss_total_v6": 0.7475061938166618, "step": 2212 }, { "epoch": 0.5103855855855856, "grad_norm": 23.657316207885742, "learning_rate": 8.444693339430683e-06, "loss": 12.618178367614746, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.28317355178296566, "loss_total_v6": 0.7886361703276634, "step": 2213 }, { "epoch": 0.5106162162162162, "grad_norm": 23.450801849365234, "learning_rate": 8.443233388234334e-06, "loss": 12.799016952514648, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.2989314515143633, "loss_total_v6": 0.7999385371804237, "step": 2214 }, { "epoch": 0.5108468468468469, "grad_norm": 25.441869735717773, "learning_rate": 8.44177287846877e-06, "loss": 12.404790878295898, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.27488747984170914, "loss_total_v6": 0.7752994447946548, "step": 2215 }, { "epoch": 0.5110774774774774, "grad_norm": 22.105981826782227, "learning_rate": 8.440311810370921e-06, "loss": 12.274222373962402, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.26719995215535164, "loss_total_v6": 0.7671389207243919, "step": 2216 }, { "epoch": 0.5113081081081081, "grad_norm": 23.416919708251953, "learning_rate": 8.438850184177805e-06, "loss": 13.038019180297852, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.3009906895458698, "loss_total_v6": 0.8148761838674545, "step": 2217 }, { "epoch": 0.5115387387387388, "grad_norm": 23.264192581176758, "learning_rate": 8.437388000126527e-06, "loss": 12.81623649597168, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.29521646350622177, "loss_total_v6": 0.8010147958993912, "step": 2218 }, { "epoch": 0.5117693693693693, "grad_norm": 23.161436080932617, "learning_rate": 8.435925258454287e-06, "loss": 12.272897720336914, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.26575908064842224, "loss_total_v6": 0.7670560702681541, "step": 2219 }, { "epoch": 0.512, "grad_norm": 21.324899673461914, "learning_rate": 8.434461959398377e-06, "loss": 12.002923965454102, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.24810750409960747, "loss_total_v6": 0.7501827105879784, "step": 2220 }, { "epoch": 0.5122306306306307, "grad_norm": 25.509376525878906, "learning_rate": 8.432998103196175e-06, "loss": 12.723125457763672, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.29040408693253994, "loss_total_v6": 0.7951953783631325, "step": 2221 }, { "epoch": 0.5124612612612612, "grad_norm": 22.372716903686523, "learning_rate": 8.43153369008515e-06, "loss": 12.526079177856445, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2765780072659254, "loss_total_v6": 0.7828799188137054, "step": 2222 }, { "epoch": 0.5126918918918919, "grad_norm": 23.271461486816406, "learning_rate": 8.430068720302863e-06, "loss": 12.990121841430664, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.3018270693719387, "loss_total_v6": 0.8118825927376747, "step": 2223 }, { "epoch": 0.5129225225225226, "grad_norm": 22.285377502441406, "learning_rate": 8.428603194086965e-06, "loss": 12.440113067626953, "loss_alignment": 0.9853515625, "loss_alignment_w": 0.49267578125, "loss_sft": 0.28451085835695267, "loss_total_v6": 0.7775070816278458, "step": 2224 }, { "epoch": 0.5131531531531531, "grad_norm": 24.29505157470703, "learning_rate": 8.4271371116752e-06, "loss": 12.841028213500977, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.301374115049839, "loss_total_v6": 0.8025643080472946, "step": 2225 }, { "epoch": 0.5133837837837838, "grad_norm": 21.508344650268555, "learning_rate": 8.425670473305395e-06, "loss": 12.081936836242676, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.24738486111164093, "loss_total_v6": 0.7551210597157478, "step": 2226 }, { "epoch": 0.5136144144144145, "grad_norm": 21.843055725097656, "learning_rate": 8.424203279215473e-06, "loss": 12.960784912109375, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.30734821036458015, "loss_total_v6": 0.810049019753933, "step": 2227 }, { "epoch": 0.513845045045045, "grad_norm": 24.629053115844727, "learning_rate": 8.422735529643445e-06, "loss": 12.69436264038086, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.293794397264719, "loss_total_v6": 0.7933976799249649, "step": 2228 }, { "epoch": 0.5140756756756757, "grad_norm": 22.991579055786133, "learning_rate": 8.421267224827411e-06, "loss": 12.850509643554688, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.3002881985157728, "loss_total_v6": 0.803156815469265, "step": 2229 }, { "epoch": 0.5143063063063064, "grad_norm": 22.17257308959961, "learning_rate": 8.419798365005569e-06, "loss": 12.626619338989258, "loss_alignment": 0.990234375, "loss_alignment_w": 0.4951171875, "loss_sft": 0.2926579900085926, "loss_total_v6": 0.7891636937856674, "step": 2230 }, { "epoch": 0.5145369369369369, "grad_norm": 22.597257614135742, "learning_rate": 8.418328950416195e-06, "loss": 12.33367919921875, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.27611921913921833, "loss_total_v6": 0.7708549499511719, "step": 2231 }, { "epoch": 0.5147675675675676, "grad_norm": 21.388906478881836, "learning_rate": 8.416858981297663e-06, "loss": 12.80797004699707, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.30069646053016186, "loss_total_v6": 0.8004981130361557, "step": 2232 }, { "epoch": 0.5149981981981981, "grad_norm": 22.96133041381836, "learning_rate": 8.415388457888433e-06, "loss": 12.473337173461914, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.27160321176052094, "loss_total_v6": 0.7795835509896278, "step": 2233 }, { "epoch": 0.5152288288288288, "grad_norm": 23.596689224243164, "learning_rate": 8.41391738042706e-06, "loss": 12.819324493408203, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.29133532755076885, "loss_total_v6": 0.8012077733874321, "step": 2234 }, { "epoch": 0.5154594594594595, "grad_norm": 23.760515213012695, "learning_rate": 8.412445749152182e-06, "loss": 12.368366241455078, "loss_alignment": 0.98388671875, "loss_alignment_w": 0.491943359375, "loss_sft": 0.2807132415473461, "loss_total_v6": 0.7730228453874588, "step": 2235 }, { "epoch": 0.51569009009009, "grad_norm": 21.84771728515625, "learning_rate": 8.410973564302534e-06, "loss": 13.202796936035156, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.3178506065160036, "loss_total_v6": 0.8251748234033585, "step": 2236 }, { "epoch": 0.5159207207207207, "grad_norm": 21.863689422607422, "learning_rate": 8.409500826116934e-06, "loss": 12.539372444152832, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.283054631203413, "loss_total_v6": 0.7837107628583908, "step": 2237 }, { "epoch": 0.5161513513513514, "grad_norm": 23.77992820739746, "learning_rate": 8.408027534834293e-06, "loss": 13.175049781799316, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.31376656889915466, "loss_total_v6": 0.8234406188130379, "step": 2238 }, { "epoch": 0.5163819819819819, "grad_norm": 23.32583999633789, "learning_rate": 8.406553690693615e-06, "loss": 12.740764617919922, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.2950007636100054, "loss_total_v6": 0.7962977662682533, "step": 2239 }, { "epoch": 0.5166126126126126, "grad_norm": 23.791738510131836, "learning_rate": 8.405079293933986e-06, "loss": 12.737004280090332, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.29903823137283325, "loss_total_v6": 0.7960627898573875, "step": 2240 }, { "epoch": 0.5168432432432433, "grad_norm": 21.03898048400879, "learning_rate": 8.40360434479459e-06, "loss": 12.637361526489258, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.29575544595718384, "loss_total_v6": 0.7898350432515144, "step": 2241 }, { "epoch": 0.5170738738738738, "grad_norm": 25.080177307128906, "learning_rate": 8.402128843514695e-06, "loss": 12.862619400024414, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.3057142421603203, "loss_total_v6": 0.8039137050509453, "step": 2242 }, { "epoch": 0.5173045045045045, "grad_norm": 20.689682006835938, "learning_rate": 8.400652790333659e-06, "loss": 12.253890037536621, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.25402732752263546, "loss_total_v6": 0.7658681273460388, "step": 2243 }, { "epoch": 0.5175351351351352, "grad_norm": 22.229198455810547, "learning_rate": 8.399176185490932e-06, "loss": 12.792381286621094, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.2816100250929594, "loss_total_v6": 0.7995238527655602, "step": 2244 }, { "epoch": 0.5177657657657657, "grad_norm": 21.927953720092773, "learning_rate": 8.397699029226052e-06, "loss": 12.565793991088867, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.2901228219270706, "loss_total_v6": 0.7853621020913124, "step": 2245 }, { "epoch": 0.5179963963963964, "grad_norm": 25.481840133666992, "learning_rate": 8.396221321778645e-06, "loss": 12.841760635375977, "loss_alignment": 0.982421875, "loss_alignment_w": 0.4912109375, "loss_sft": 0.3107429314404726, "loss_total_v6": 0.8026100099086761, "step": 2246 }, { "epoch": 0.518227027027027, "grad_norm": 27.261821746826172, "learning_rate": 8.394743063388433e-06, "loss": 12.784713745117188, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.28889752365648746, "loss_total_v6": 0.7990446016192436, "step": 2247 }, { "epoch": 0.5184576576576576, "grad_norm": 22.83901596069336, "learning_rate": 8.393264254295218e-06, "loss": 12.300055503845215, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.25488322600722313, "loss_total_v6": 0.7687534838914871, "step": 2248 }, { "epoch": 0.5186882882882883, "grad_norm": 26.863975524902344, "learning_rate": 8.391784894738898e-06, "loss": 12.536819458007812, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.275540255010128, "loss_total_v6": 0.7835511714220047, "step": 2249 }, { "epoch": 0.518918918918919, "grad_norm": 23.480010986328125, "learning_rate": 8.390304984959455e-06, "loss": 12.213871002197266, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.26066617108881474, "loss_total_v6": 0.7633669674396515, "step": 2250 }, { "epoch": 0.5191495495495495, "grad_norm": 24.30265235900879, "learning_rate": 8.388824525196967e-06, "loss": 12.655628204345703, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.28552933037281036, "loss_total_v6": 0.7909767404198647, "step": 2251 }, { "epoch": 0.5193801801801802, "grad_norm": 19.899921417236328, "learning_rate": 8.387343515691594e-06, "loss": 12.618074417114258, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2836400046944618, "loss_total_v6": 0.7886296287178993, "step": 2252 }, { "epoch": 0.5196108108108108, "grad_norm": 21.81645965576172, "learning_rate": 8.385861956683594e-06, "loss": 12.832252502441406, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.29450841434299946, "loss_total_v6": 0.8020157366991043, "step": 2253 }, { "epoch": 0.5198414414414414, "grad_norm": 22.41352081298828, "learning_rate": 8.384379848413304e-06, "loss": 12.47627067565918, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.2792022991925478, "loss_total_v6": 0.7797668874263763, "step": 2254 }, { "epoch": 0.5200720720720721, "grad_norm": 21.5006103515625, "learning_rate": 8.382897191121157e-06, "loss": 12.301188468933105, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.26896160282194614, "loss_total_v6": 0.7688242793083191, "step": 2255 }, { "epoch": 0.5203027027027027, "grad_norm": 22.8327579498291, "learning_rate": 8.381413985047672e-06, "loss": 12.951128005981445, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.30154140293598175, "loss_total_v6": 0.8094454854726791, "step": 2256 }, { "epoch": 0.5205333333333333, "grad_norm": 20.484912872314453, "learning_rate": 8.379930230433461e-06, "loss": 12.853185653686523, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.29714434780180454, "loss_total_v6": 0.8033241555094719, "step": 2257 }, { "epoch": 0.520763963963964, "grad_norm": 22.15376853942871, "learning_rate": 8.378445927519218e-06, "loss": 12.470733642578125, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.28518869169056416, "loss_total_v6": 0.7794208526611328, "step": 2258 }, { "epoch": 0.5209945945945946, "grad_norm": 19.741825103759766, "learning_rate": 8.376961076545733e-06, "loss": 12.26800537109375, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.26552962325513363, "loss_total_v6": 0.7667503207921982, "step": 2259 }, { "epoch": 0.5212252252252252, "grad_norm": 23.748085021972656, "learning_rate": 8.375475677753882e-06, "loss": 12.242179870605469, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2511591110378504, "loss_total_v6": 0.7651362046599388, "step": 2260 }, { "epoch": 0.5214558558558559, "grad_norm": 24.42142677307129, "learning_rate": 8.373989731384627e-06, "loss": 12.268244743347168, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.26612443663179874, "loss_total_v6": 0.7667652890086174, "step": 2261 }, { "epoch": 0.5216864864864865, "grad_norm": 25.219058990478516, "learning_rate": 8.372503237679024e-06, "loss": 13.183732986450195, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.3226863518357277, "loss_total_v6": 0.823983334004879, "step": 2262 }, { "epoch": 0.5219171171171171, "grad_norm": 25.157615661621094, "learning_rate": 8.371016196878215e-06, "loss": 12.34949779510498, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2658468782901764, "loss_total_v6": 0.7718435823917389, "step": 2263 }, { "epoch": 0.5221477477477477, "grad_norm": 21.201885223388672, "learning_rate": 8.36952860922343e-06, "loss": 12.689057350158691, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.29001436568796635, "loss_total_v6": 0.7930660992860794, "step": 2264 }, { "epoch": 0.5223783783783784, "grad_norm": 21.540273666381836, "learning_rate": 8.36804047495599e-06, "loss": 12.256501197814941, "loss_alignment": 0.98583984375, "loss_alignment_w": 0.492919921875, "loss_sft": 0.27335553616285324, "loss_total_v6": 0.7660313099622726, "step": 2265 }, { "epoch": 0.522609009009009, "grad_norm": 22.420377731323242, "learning_rate": 8.366551794317304e-06, "loss": 13.009237289428711, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.3071111459285021, "loss_total_v6": 0.8130773529410362, "step": 2266 }, { "epoch": 0.5228396396396396, "grad_norm": 21.97037696838379, "learning_rate": 8.365062567548868e-06, "loss": 12.761981010437012, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.29335132241249084, "loss_total_v6": 0.7976238057017326, "step": 2267 }, { "epoch": 0.5230702702702703, "grad_norm": 24.957195281982422, "learning_rate": 8.363572794892268e-06, "loss": 13.232135772705078, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.32394151762127876, "loss_total_v6": 0.8270085230469704, "step": 2268 }, { "epoch": 0.5233009009009009, "grad_norm": 25.921541213989258, "learning_rate": 8.36208247658918e-06, "loss": 12.415891647338867, "loss_alignment": 0.98193359375, "loss_alignment_w": 0.490966796875, "loss_sft": 0.284110926091671, "loss_total_v6": 0.7759932279586792, "step": 2269 }, { "epoch": 0.5235315315315315, "grad_norm": 21.991113662719727, "learning_rate": 8.360591612881363e-06, "loss": 12.621600151062012, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.28698841854929924, "loss_total_v6": 0.7888500019907951, "step": 2270 }, { "epoch": 0.5237621621621622, "grad_norm": 24.992267608642578, "learning_rate": 8.359100204010673e-06, "loss": 12.792243957519531, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.3027501180768013, "loss_total_v6": 0.7995152547955513, "step": 2271 }, { "epoch": 0.5239927927927928, "grad_norm": 21.147397994995117, "learning_rate": 8.357608250219046e-06, "loss": 12.59079647064209, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.28838961385190487, "loss_total_v6": 0.78692477196455, "step": 2272 }, { "epoch": 0.5242234234234234, "grad_norm": 21.654272079467773, "learning_rate": 8.356115751748514e-06, "loss": 12.520309448242188, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.27540872246026993, "loss_total_v6": 0.7825193032622337, "step": 2273 }, { "epoch": 0.5244540540540541, "grad_norm": 20.615768432617188, "learning_rate": 8.354622708841192e-06, "loss": 11.860567092895508, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.24638189561665058, "loss_total_v6": 0.7412854507565498, "step": 2274 }, { "epoch": 0.5246846846846847, "grad_norm": 22.574848175048828, "learning_rate": 8.35312912173928e-06, "loss": 11.915977478027344, "loss_alignment": 0.99267578125, "loss_alignment_w": 0.496337890625, "loss_sft": 0.24793768115341663, "loss_total_v6": 0.7447485774755478, "step": 2275 }, { "epoch": 0.5249153153153153, "grad_norm": 22.02884292602539, "learning_rate": 8.35163499068508e-06, "loss": 12.578025817871094, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.27887867763638496, "loss_total_v6": 0.7861265763640404, "step": 2276 }, { "epoch": 0.525145945945946, "grad_norm": 22.248390197753906, "learning_rate": 8.350140315920964e-06, "loss": 12.299564361572266, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.26297020353376865, "loss_total_v6": 0.7687227874994278, "step": 2277 }, { "epoch": 0.5253765765765765, "grad_norm": 24.252796173095703, "learning_rate": 8.348645097689409e-06, "loss": 11.717913627624512, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.22602191753685474, "loss_total_v6": 0.7323696240782738, "step": 2278 }, { "epoch": 0.5256072072072072, "grad_norm": 23.726478576660156, "learning_rate": 8.34714933623297e-06, "loss": 12.349481582641602, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.26946230977773666, "loss_total_v6": 0.7718426510691643, "step": 2279 }, { "epoch": 0.5258378378378379, "grad_norm": 28.391128540039062, "learning_rate": 8.345653031794292e-06, "loss": 12.57742977142334, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.28363265469670296, "loss_total_v6": 0.7860893309116364, "step": 2280 }, { "epoch": 0.5260684684684684, "grad_norm": 23.876548767089844, "learning_rate": 8.34415618461611e-06, "loss": 12.847203254699707, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2961447611451149, "loss_total_v6": 0.8029502108693123, "step": 2281 }, { "epoch": 0.5262990990990991, "grad_norm": 22.06159210205078, "learning_rate": 8.342658794941245e-06, "loss": 12.659167289733887, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2859947420656681, "loss_total_v6": 0.7911979779601097, "step": 2282 }, { "epoch": 0.5265297297297298, "grad_norm": 23.61046600341797, "learning_rate": 8.34116086301261e-06, "loss": 12.259024620056152, "loss_alignment": 1.02392578125, "loss_alignment_w": 0.511962890625, "loss_sft": 0.2550348322838545, "loss_total_v6": 0.7661890238523483, "step": 2283 }, { "epoch": 0.5267603603603603, "grad_norm": 23.140552520751953, "learning_rate": 8.339662389073197e-06, "loss": 12.522727966308594, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.27675003558397293, "loss_total_v6": 0.7826704829931259, "step": 2284 }, { "epoch": 0.526990990990991, "grad_norm": 20.830772399902344, "learning_rate": 8.338163373366099e-06, "loss": 12.760275840759277, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2903303727507591, "loss_total_v6": 0.797517217695713, "step": 2285 }, { "epoch": 0.5272216216216217, "grad_norm": 26.76498794555664, "learning_rate": 8.336663816134483e-06, "loss": 12.10403823852539, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.242861058562994, "loss_total_v6": 0.7565023899078369, "step": 2286 }, { "epoch": 0.5274522522522522, "grad_norm": 23.505111694335938, "learning_rate": 8.335163717621615e-06, "loss": 12.434619903564453, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.2787201087921858, "loss_total_v6": 0.7771636992692947, "step": 2287 }, { "epoch": 0.5276828828828829, "grad_norm": 22.71406364440918, "learning_rate": 8.333663078070845e-06, "loss": 12.113579750061035, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2418551966547966, "loss_total_v6": 0.7570987343788147, "step": 2288 }, { "epoch": 0.5279135135135136, "grad_norm": 22.025814056396484, "learning_rate": 8.332161897725607e-06, "loss": 13.026826858520508, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.30933962017297745, "loss_total_v6": 0.8141766488552094, "step": 2289 }, { "epoch": 0.5281441441441441, "grad_norm": 23.35102653503418, "learning_rate": 8.330660176829427e-06, "loss": 12.309393882751465, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.26424070447683334, "loss_total_v6": 0.7693371251225471, "step": 2290 }, { "epoch": 0.5283747747747748, "grad_norm": 23.551105499267578, "learning_rate": 8.32915791562592e-06, "loss": 12.9470796585083, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.3028142061084509, "loss_total_v6": 0.8091924339532852, "step": 2291 }, { "epoch": 0.5286054054054055, "grad_norm": 21.181699752807617, "learning_rate": 8.327655114358782e-06, "loss": 12.558716773986816, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.27527626417577267, "loss_total_v6": 0.7849198058247566, "step": 2292 }, { "epoch": 0.528836036036036, "grad_norm": 23.175901412963867, "learning_rate": 8.326151773271805e-06, "loss": 12.44729232788086, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.2770708054304123, "loss_total_v6": 0.7779557853937149, "step": 2293 }, { "epoch": 0.5290666666666667, "grad_norm": 21.864355087280273, "learning_rate": 8.324647892608861e-06, "loss": 11.97077751159668, "loss_alignment": 0.96142578125, "loss_alignment_w": 0.480712890625, "loss_sft": 0.26758271269500256, "loss_total_v6": 0.7481735572218895, "step": 2294 }, { "epoch": 0.5292972972972972, "grad_norm": 24.44074249267578, "learning_rate": 8.323143472613916e-06, "loss": 12.39666748046875, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.2692222725600004, "loss_total_v6": 0.7747917398810387, "step": 2295 }, { "epoch": 0.5295279279279279, "grad_norm": 30.111928939819336, "learning_rate": 8.321638513531018e-06, "loss": 12.52267074584961, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.28719882667064667, "loss_total_v6": 0.7826669663190842, "step": 2296 }, { "epoch": 0.5297585585585586, "grad_norm": 22.437774658203125, "learning_rate": 8.320133015604306e-06, "loss": 12.68959903717041, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.28244931623339653, "loss_total_v6": 0.7930999398231506, "step": 2297 }, { "epoch": 0.5299891891891891, "grad_norm": 23.186784744262695, "learning_rate": 8.318626979078006e-06, "loss": 12.557143211364746, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.28367700800299644, "loss_total_v6": 0.7848214358091354, "step": 2298 }, { "epoch": 0.5302198198198198, "grad_norm": 22.542139053344727, "learning_rate": 8.31712040419643e-06, "loss": 13.125090599060059, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.31984514370560646, "loss_total_v6": 0.8203181400895119, "step": 2299 }, { "epoch": 0.5304504504504505, "grad_norm": 27.48114585876465, "learning_rate": 8.315613291203977e-06, "loss": 12.145424842834473, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2549538966268301, "loss_total_v6": 0.7590890228748322, "step": 2300 }, { "epoch": 0.530681081081081, "grad_norm": 22.58332061767578, "learning_rate": 8.314105640345135e-06, "loss": 12.829562187194824, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.2974988892674446, "loss_total_v6": 0.8018476292490959, "step": 2301 }, { "epoch": 0.5309117117117117, "grad_norm": 22.335708618164062, "learning_rate": 8.31259745186448e-06, "loss": 12.570562362670898, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.28840675204992294, "loss_total_v6": 0.7856601625680923, "step": 2302 }, { "epoch": 0.5311423423423424, "grad_norm": 21.42266273498535, "learning_rate": 8.311088726006674e-06, "loss": 12.525968551635742, "loss_alignment": 1.01708984375, "loss_alignment_w": 0.508544921875, "loss_sft": 0.27446542493999004, "loss_total_v6": 0.7828729823231697, "step": 2303 }, { "epoch": 0.5313729729729729, "grad_norm": 24.119680404663086, "learning_rate": 8.309579463016461e-06, "loss": 12.798614501953125, "loss_alignment": 0.98388671875, "loss_alignment_w": 0.491943359375, "loss_sft": 0.30578797683119774, "loss_total_v6": 0.7999133691191673, "step": 2304 }, { "epoch": 0.5316036036036036, "grad_norm": 21.093870162963867, "learning_rate": 8.308069663138682e-06, "loss": 12.878349304199219, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.3080096859484911, "loss_total_v6": 0.8048968762159348, "step": 2305 }, { "epoch": 0.5318342342342343, "grad_norm": 21.58739471435547, "learning_rate": 8.30655932661826e-06, "loss": 12.652421951293945, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.2888995911926031, "loss_total_v6": 0.790776401758194, "step": 2306 }, { "epoch": 0.5320648648648648, "grad_norm": 22.927576065063477, "learning_rate": 8.305048453700204e-06, "loss": 12.358187675476074, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.27334803342819214, "loss_total_v6": 0.7723867371678352, "step": 2307 }, { "epoch": 0.5322954954954955, "grad_norm": 20.67194175720215, "learning_rate": 8.303537044629611e-06, "loss": 12.446928024291992, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.26983057521283627, "loss_total_v6": 0.7779329866170883, "step": 2308 }, { "epoch": 0.5325261261261262, "grad_norm": 22.288732528686523, "learning_rate": 8.302025099651666e-06, "loss": 12.730866432189941, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.29586223885416985, "loss_total_v6": 0.7956791520118713, "step": 2309 }, { "epoch": 0.5327567567567567, "grad_norm": 22.88485336303711, "learning_rate": 8.30051261901164e-06, "loss": 11.97734260559082, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.2490263506770134, "loss_total_v6": 0.7485838606953621, "step": 2310 }, { "epoch": 0.5329873873873874, "grad_norm": 24.833925247192383, "learning_rate": 8.29899960295489e-06, "loss": 12.54367446899414, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.27813554741442204, "loss_total_v6": 0.7839796617627144, "step": 2311 }, { "epoch": 0.533218018018018, "grad_norm": 22.999832153320312, "learning_rate": 8.297486051726864e-06, "loss": 12.413429260253906, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.2724213656038046, "loss_total_v6": 0.7758393213152885, "step": 2312 }, { "epoch": 0.5334486486486486, "grad_norm": 27.798673629760742, "learning_rate": 8.295971965573088e-06, "loss": 12.776376724243164, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.28579769283533096, "loss_total_v6": 0.7985235303640366, "step": 2313 }, { "epoch": 0.5336792792792793, "grad_norm": 24.118457794189453, "learning_rate": 8.294457344739188e-06, "loss": 12.854813575744629, "loss_alignment": 1.02197265625, "loss_alignment_w": 0.510986328125, "loss_sft": 0.2916155569255352, "loss_total_v6": 0.8034258559346199, "step": 2314 }, { "epoch": 0.5339099099099099, "grad_norm": 24.54739761352539, "learning_rate": 8.292942189470863e-06, "loss": 12.467079162597656, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2763695679605007, "loss_total_v6": 0.7791924253106117, "step": 2315 }, { "epoch": 0.5341405405405405, "grad_norm": 22.492115020751953, "learning_rate": 8.291426500013909e-06, "loss": 12.264835357666016, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.26668955385684967, "loss_total_v6": 0.7665522545576096, "step": 2316 }, { "epoch": 0.5343711711711712, "grad_norm": 21.839902877807617, "learning_rate": 8.289910276614202e-06, "loss": 12.224921226501465, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.2622112352401018, "loss_total_v6": 0.7640575394034386, "step": 2317 }, { "epoch": 0.5346018018018018, "grad_norm": 25.861120223999023, "learning_rate": 8.288393519517708e-06, "loss": 11.972707748413086, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.24402177520096302, "loss_total_v6": 0.7482942268252373, "step": 2318 }, { "epoch": 0.5348324324324324, "grad_norm": 20.831483840942383, "learning_rate": 8.28687622897048e-06, "loss": 12.544747352600098, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.28033881075680256, "loss_total_v6": 0.7840467095375061, "step": 2319 }, { "epoch": 0.5350630630630631, "grad_norm": 22.632747650146484, "learning_rate": 8.285358405218655e-06, "loss": 12.663070678710938, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.28034874238073826, "loss_total_v6": 0.7914418876171112, "step": 2320 }, { "epoch": 0.5352936936936937, "grad_norm": 21.634187698364258, "learning_rate": 8.283840048508458e-06, "loss": 12.700540542602539, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.28906884603202343, "loss_total_v6": 0.7937838137149811, "step": 2321 }, { "epoch": 0.5355243243243243, "grad_norm": 22.04832649230957, "learning_rate": 8.2823211590862e-06, "loss": 12.076509475708008, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.25261509977281094, "loss_total_v6": 0.7547818347811699, "step": 2322 }, { "epoch": 0.535754954954955, "grad_norm": 23.682697296142578, "learning_rate": 8.280801737198281e-06, "loss": 13.097740173339844, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.3085227459669113, "loss_total_v6": 0.8186088129878044, "step": 2323 }, { "epoch": 0.5359855855855856, "grad_norm": 21.28583526611328, "learning_rate": 8.279281783091182e-06, "loss": 12.673710823059082, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.2867053207010031, "loss_total_v6": 0.7921069338917732, "step": 2324 }, { "epoch": 0.5362162162162162, "grad_norm": 22.771257400512695, "learning_rate": 8.277761297011475e-06, "loss": 12.704741477966309, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.28995697759091854, "loss_total_v6": 0.7940463572740555, "step": 2325 }, { "epoch": 0.5364468468468468, "grad_norm": 25.006372451782227, "learning_rate": 8.276240279205817e-06, "loss": 12.23686408996582, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.268191447481513, "loss_total_v6": 0.7648040130734444, "step": 2326 }, { "epoch": 0.5366774774774775, "grad_norm": 23.970308303833008, "learning_rate": 8.27471872992095e-06, "loss": 13.097625732421875, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.3208141718059778, "loss_total_v6": 0.8186016231775284, "step": 2327 }, { "epoch": 0.5369081081081081, "grad_norm": 21.01490592956543, "learning_rate": 8.273196649403703e-06, "loss": 12.323570251464844, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2600760441273451, "loss_total_v6": 0.7702231034636497, "step": 2328 }, { "epoch": 0.5371387387387387, "grad_norm": 21.600208282470703, "learning_rate": 8.271674037900994e-06, "loss": 12.296916961669922, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.27342485450208187, "loss_total_v6": 0.7685572803020477, "step": 2329 }, { "epoch": 0.5373693693693694, "grad_norm": 22.395336151123047, "learning_rate": 8.270150895659824e-06, "loss": 12.779031753540039, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.29127378575503826, "loss_total_v6": 0.7986895367503166, "step": 2330 }, { "epoch": 0.5376, "grad_norm": 20.96824073791504, "learning_rate": 8.268627222927276e-06, "loss": 12.65867805480957, "loss_alignment": 0.9892578125, "loss_alignment_w": 0.49462890625, "loss_sft": 0.29682840034365654, "loss_total_v6": 0.7911673858761787, "step": 2331 }, { "epoch": 0.5378306306306306, "grad_norm": 22.46871566772461, "learning_rate": 8.267103019950529e-06, "loss": 12.69025707244873, "loss_alignment": 0.97802734375, "loss_alignment_w": 0.489013671875, "loss_sft": 0.30428003706038, "loss_total_v6": 0.7931410819292068, "step": 2332 }, { "epoch": 0.5380612612612613, "grad_norm": 22.125022888183594, "learning_rate": 8.265578286976839e-06, "loss": 12.252896308898926, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.25915320217609406, "loss_total_v6": 0.765806034207344, "step": 2333 }, { "epoch": 0.5382918918918919, "grad_norm": 22.81772804260254, "learning_rate": 8.264053024253554e-06, "loss": 12.041996002197266, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.24974078871309757, "loss_total_v6": 0.7526246905326843, "step": 2334 }, { "epoch": 0.5385225225225225, "grad_norm": 22.026771545410156, "learning_rate": 8.262527232028106e-06, "loss": 12.607115745544434, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.28536603040993214, "loss_total_v6": 0.7879447191953659, "step": 2335 }, { "epoch": 0.5387531531531532, "grad_norm": 22.666170120239258, "learning_rate": 8.26100091054801e-06, "loss": 12.092988967895508, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.25759713537991047, "loss_total_v6": 0.7558118477463722, "step": 2336 }, { "epoch": 0.5389837837837838, "grad_norm": 20.676727294921875, "learning_rate": 8.259474060060872e-06, "loss": 11.794118881225586, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.23614059761166573, "loss_total_v6": 0.7371324226260185, "step": 2337 }, { "epoch": 0.5392144144144144, "grad_norm": 22.643596649169922, "learning_rate": 8.25794668081438e-06, "loss": 12.443292617797852, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.27515753731131554, "loss_total_v6": 0.7777057588100433, "step": 2338 }, { "epoch": 0.5394450450450451, "grad_norm": 22.696794509887695, "learning_rate": 8.256418773056309e-06, "loss": 12.768829345703125, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.28959846124053, "loss_total_v6": 0.7980518639087677, "step": 2339 }, { "epoch": 0.5396756756756756, "grad_norm": 22.8935546875, "learning_rate": 8.25489033703452e-06, "loss": 12.375299453735352, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2625003941357136, "loss_total_v6": 0.7734562009572983, "step": 2340 }, { "epoch": 0.5399063063063063, "grad_norm": 22.537744522094727, "learning_rate": 8.253361372996958e-06, "loss": 12.79095458984375, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2931022848933935, "loss_total_v6": 0.7994346916675568, "step": 2341 }, { "epoch": 0.540136936936937, "grad_norm": 24.491147994995117, "learning_rate": 8.251831881191656e-06, "loss": 12.296370506286621, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2620076406747103, "loss_total_v6": 0.7685231566429138, "step": 2342 }, { "epoch": 0.5403675675675675, "grad_norm": 23.710681915283203, "learning_rate": 8.250301861866734e-06, "loss": 12.346854209899902, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.2679553125053644, "loss_total_v6": 0.7716784179210663, "step": 2343 }, { "epoch": 0.5405981981981982, "grad_norm": 21.300064086914062, "learning_rate": 8.248771315270391e-06, "loss": 12.242539405822754, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.261328749358654, "loss_total_v6": 0.7651587277650833, "step": 2344 }, { "epoch": 0.5408288288288289, "grad_norm": 23.7994327545166, "learning_rate": 8.247240241650918e-06, "loss": 13.057859420776367, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.31210314854979515, "loss_total_v6": 0.8161162212491035, "step": 2345 }, { "epoch": 0.5410594594594594, "grad_norm": 22.92929458618164, "learning_rate": 8.24570864125669e-06, "loss": 12.655373573303223, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.28084427677094936, "loss_total_v6": 0.790960855782032, "step": 2346 }, { "epoch": 0.5412900900900901, "grad_norm": 21.67525863647461, "learning_rate": 8.244176514336167e-06, "loss": 12.28157901763916, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.2639671266078949, "loss_total_v6": 0.7675986886024475, "step": 2347 }, { "epoch": 0.5415207207207208, "grad_norm": 21.763378143310547, "learning_rate": 8.242643861137891e-06, "loss": 12.483631134033203, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2770836241543293, "loss_total_v6": 0.780226930975914, "step": 2348 }, { "epoch": 0.5417513513513513, "grad_norm": 21.377201080322266, "learning_rate": 8.241110681910495e-06, "loss": 12.597015380859375, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2770595531910658, "loss_total_v6": 0.7873134762048721, "step": 2349 }, { "epoch": 0.541981981981982, "grad_norm": 25.028663635253906, "learning_rate": 8.239576976902694e-06, "loss": 12.55327033996582, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.28410638496279716, "loss_total_v6": 0.784579411149025, "step": 2350 }, { "epoch": 0.5422126126126127, "grad_norm": 20.661666870117188, "learning_rate": 8.238042746363291e-06, "loss": 11.984390258789062, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.24377540685236454, "loss_total_v6": 0.7490244135260582, "step": 2351 }, { "epoch": 0.5424432432432432, "grad_norm": 23.41115379333496, "learning_rate": 8.23650799054117e-06, "loss": 12.04619026184082, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.25754083693027496, "loss_total_v6": 0.7528868392109871, "step": 2352 }, { "epoch": 0.5426738738738739, "grad_norm": 24.03182029724121, "learning_rate": 8.234972709685303e-06, "loss": 12.33110237121582, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.2771330773830414, "loss_total_v6": 0.770693875849247, "step": 2353 }, { "epoch": 0.5429045045045046, "grad_norm": 20.7380313873291, "learning_rate": 8.233436904044748e-06, "loss": 12.067582130432129, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.253583000972867, "loss_total_v6": 0.7542238608002663, "step": 2354 }, { "epoch": 0.5431351351351351, "grad_norm": 21.930707931518555, "learning_rate": 8.231900573868645e-06, "loss": 12.451064109802246, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.27460571378469467, "loss_total_v6": 0.7781915292143822, "step": 2355 }, { "epoch": 0.5433657657657658, "grad_norm": 22.760099411010742, "learning_rate": 8.230363719406223e-06, "loss": 12.4859037399292, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2693826351314783, "loss_total_v6": 0.7803689762949944, "step": 2356 }, { "epoch": 0.5435963963963963, "grad_norm": 25.789196014404297, "learning_rate": 8.228826340906795e-06, "loss": 12.357879638671875, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2555217668414116, "loss_total_v6": 0.7723674923181534, "step": 2357 }, { "epoch": 0.543827027027027, "grad_norm": 23.65781021118164, "learning_rate": 8.227288438619754e-06, "loss": 13.12053108215332, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.29900659807026386, "loss_total_v6": 0.8200331851840019, "step": 2358 }, { "epoch": 0.5440576576576577, "grad_norm": 20.87714195251465, "learning_rate": 8.225750012794585e-06, "loss": 12.413087844848633, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2724915239959955, "loss_total_v6": 0.775817945599556, "step": 2359 }, { "epoch": 0.5442882882882882, "grad_norm": 24.789649963378906, "learning_rate": 8.224211063680854e-06, "loss": 12.850210189819336, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2915261723101139, "loss_total_v6": 0.8031381368637085, "step": 2360 }, { "epoch": 0.5445189189189189, "grad_norm": 22.4388427734375, "learning_rate": 8.222671591528212e-06, "loss": 13.005266189575195, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.3136683702468872, "loss_total_v6": 0.8128291219472885, "step": 2361 }, { "epoch": 0.5447495495495496, "grad_norm": 23.115285873413086, "learning_rate": 8.2211315965864e-06, "loss": 12.920175552368164, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.3095251675695181, "loss_total_v6": 0.8075110018253326, "step": 2362 }, { "epoch": 0.5449801801801801, "grad_norm": 23.80994415283203, "learning_rate": 8.219591079105234e-06, "loss": 12.707834243774414, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.29660474322736263, "loss_total_v6": 0.7942396476864815, "step": 2363 }, { "epoch": 0.5452108108108108, "grad_norm": 22.215145111083984, "learning_rate": 8.218050039334624e-06, "loss": 12.803194046020508, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.3024274446070194, "loss_total_v6": 0.8001996576786041, "step": 2364 }, { "epoch": 0.5454414414414415, "grad_norm": 23.523773193359375, "learning_rate": 8.216508477524558e-06, "loss": 12.25666618347168, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2565353587269783, "loss_total_v6": 0.7660415843129158, "step": 2365 }, { "epoch": 0.545672072072072, "grad_norm": 24.192014694213867, "learning_rate": 8.214966393925115e-06, "loss": 12.714349746704102, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.29200710728764534, "loss_total_v6": 0.7946468964219093, "step": 2366 }, { "epoch": 0.5459027027027027, "grad_norm": 23.106307983398438, "learning_rate": 8.213423788786453e-06, "loss": 12.444969177246094, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.26835015416145325, "loss_total_v6": 0.7778106182813644, "step": 2367 }, { "epoch": 0.5461333333333334, "grad_norm": 22.311946868896484, "learning_rate": 8.211880662358818e-06, "loss": 12.585837364196777, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.27146284841001034, "loss_total_v6": 0.7866148129105568, "step": 2368 }, { "epoch": 0.5463639639639639, "grad_norm": 21.984277725219727, "learning_rate": 8.21033701489254e-06, "loss": 12.584640502929688, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.28805062361061573, "loss_total_v6": 0.7865400239825249, "step": 2369 }, { "epoch": 0.5465945945945946, "grad_norm": 24.906238555908203, "learning_rate": 8.20879284663803e-06, "loss": 12.305213928222656, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.26373531483113766, "loss_total_v6": 0.7690758556127548, "step": 2370 }, { "epoch": 0.5468252252252253, "grad_norm": 24.251625061035156, "learning_rate": 8.20724815784579e-06, "loss": 12.31238079071045, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2630235552787781, "loss_total_v6": 0.7695237994194031, "step": 2371 }, { "epoch": 0.5470558558558558, "grad_norm": 25.590688705444336, "learning_rate": 8.2057029487664e-06, "loss": 12.788681030273438, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2940893638879061, "loss_total_v6": 0.7992926016449928, "step": 2372 }, { "epoch": 0.5472864864864865, "grad_norm": 22.47159767150879, "learning_rate": 8.204157219650533e-06, "loss": 12.719438552856445, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.3041965253651142, "loss_total_v6": 0.7949649542570114, "step": 2373 }, { "epoch": 0.5475171171171171, "grad_norm": 22.29364585876465, "learning_rate": 8.202610970748934e-06, "loss": 12.370902061462402, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.26892418414354324, "loss_total_v6": 0.7731813937425613, "step": 2374 }, { "epoch": 0.5477477477477477, "grad_norm": 22.626482009887695, "learning_rate": 8.20106420231244e-06, "loss": 12.802237510681152, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.29185434617102146, "loss_total_v6": 0.800139844417572, "step": 2375 }, { "epoch": 0.5479783783783784, "grad_norm": 24.365835189819336, "learning_rate": 8.199516914591975e-06, "loss": 12.972566604614258, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.30842024087905884, "loss_total_v6": 0.8107853680849075, "step": 2376 }, { "epoch": 0.548209009009009, "grad_norm": 22.271892547607422, "learning_rate": 8.197969107838541e-06, "loss": 12.236303329467773, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2561020143330097, "loss_total_v6": 0.7647689953446388, "step": 2377 }, { "epoch": 0.5484396396396396, "grad_norm": 22.12565040588379, "learning_rate": 8.19642078230323e-06, "loss": 12.476070404052734, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2752988487482071, "loss_total_v6": 0.7797544077038765, "step": 2378 }, { "epoch": 0.5486702702702703, "grad_norm": 26.36653709411621, "learning_rate": 8.19487193823721e-06, "loss": 13.188362121582031, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.32666830718517303, "loss_total_v6": 0.8242726102471352, "step": 2379 }, { "epoch": 0.5489009009009009, "grad_norm": 26.07146644592285, "learning_rate": 8.19332257589174e-06, "loss": 12.59310245513916, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.28285749070346355, "loss_total_v6": 0.7870688959956169, "step": 2380 }, { "epoch": 0.5491315315315315, "grad_norm": 22.987804412841797, "learning_rate": 8.191772695518164e-06, "loss": 12.936991691589355, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.30518978647887707, "loss_total_v6": 0.8085619807243347, "step": 2381 }, { "epoch": 0.5493621621621622, "grad_norm": 21.058109283447266, "learning_rate": 8.190222297367901e-06, "loss": 12.536187171936035, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2794833779335022, "loss_total_v6": 0.7835116982460022, "step": 2382 }, { "epoch": 0.5495927927927928, "grad_norm": 24.134946823120117, "learning_rate": 8.188671381692466e-06, "loss": 12.535063743591309, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.2849826067686081, "loss_total_v6": 0.7834414690732956, "step": 2383 }, { "epoch": 0.5498234234234234, "grad_norm": 23.60434341430664, "learning_rate": 8.18711994874345e-06, "loss": 12.604887008666992, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.287561297416687, "loss_total_v6": 0.787805400788784, "step": 2384 }, { "epoch": 0.550054054054054, "grad_norm": 22.24051856994629, "learning_rate": 8.185567998772528e-06, "loss": 12.638601303100586, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2838242668658495, "loss_total_v6": 0.789912536740303, "step": 2385 }, { "epoch": 0.5502846846846847, "grad_norm": 25.18804931640625, "learning_rate": 8.184015532031465e-06, "loss": 13.453683853149414, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.3409772738814354, "loss_total_v6": 0.8408552333712578, "step": 2386 }, { "epoch": 0.5505153153153153, "grad_norm": 23.790979385375977, "learning_rate": 8.1824625487721e-06, "loss": 12.398504257202148, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.26953539811074734, "loss_total_v6": 0.7749064862728119, "step": 2387 }, { "epoch": 0.550745945945946, "grad_norm": 23.768346786499023, "learning_rate": 8.18090904924637e-06, "loss": 12.17335033416748, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.26309272833168507, "loss_total_v6": 0.7608343958854675, "step": 2388 }, { "epoch": 0.5509765765765766, "grad_norm": 21.21285057067871, "learning_rate": 8.179355033706279e-06, "loss": 12.930755615234375, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.30753135308623314, "loss_total_v6": 0.8081722110509872, "step": 2389 }, { "epoch": 0.5512072072072072, "grad_norm": 20.975278854370117, "learning_rate": 8.177800502403928e-06, "loss": 12.291706085205078, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.27967570535838604, "loss_total_v6": 0.7682316228747368, "step": 2390 }, { "epoch": 0.5514378378378378, "grad_norm": 20.890399932861328, "learning_rate": 8.176245455591493e-06, "loss": 13.295485496520996, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.31924910470843315, "loss_total_v6": 0.8309678360819817, "step": 2391 }, { "epoch": 0.5516684684684685, "grad_norm": 23.136669158935547, "learning_rate": 8.174689893521239e-06, "loss": 13.281366348266602, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.3243634030222893, "loss_total_v6": 0.8300854191184044, "step": 2392 }, { "epoch": 0.5518990990990991, "grad_norm": 21.2564754486084, "learning_rate": 8.173133816445513e-06, "loss": 12.255191802978516, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.26936741918325424, "loss_total_v6": 0.765949435532093, "step": 2393 }, { "epoch": 0.5521297297297297, "grad_norm": 24.22365951538086, "learning_rate": 8.171577224616746e-06, "loss": 12.093426704406738, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.2624462489038706, "loss_total_v6": 0.7558391764760017, "step": 2394 }, { "epoch": 0.5523603603603604, "grad_norm": 21.912158966064453, "learning_rate": 8.17002011828745e-06, "loss": 12.21310806274414, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.2655622996389866, "loss_total_v6": 0.7633192762732506, "step": 2395 }, { "epoch": 0.552590990990991, "grad_norm": 22.312541961669922, "learning_rate": 8.168462497710227e-06, "loss": 13.077564239501953, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.3099777828902006, "loss_total_v6": 0.817347802221775, "step": 2396 }, { "epoch": 0.5528216216216216, "grad_norm": 23.727006912231445, "learning_rate": 8.166904363137752e-06, "loss": 12.697425842285156, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.2944893781095743, "loss_total_v6": 0.7935891225934029, "step": 2397 }, { "epoch": 0.5530522522522523, "grad_norm": 21.702869415283203, "learning_rate": 8.165345714822791e-06, "loss": 13.033918380737305, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2968891877681017, "loss_total_v6": 0.8146199136972427, "step": 2398 }, { "epoch": 0.5532828828828829, "grad_norm": 22.632984161376953, "learning_rate": 8.163786553018192e-06, "loss": 12.887847900390625, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.30225566774606705, "loss_total_v6": 0.8054905235767365, "step": 2399 }, { "epoch": 0.5535135135135135, "grad_norm": 22.301681518554688, "learning_rate": 8.162226877976886e-06, "loss": 13.327072143554688, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.33455945551395416, "loss_total_v6": 0.8329420164227486, "step": 2400 }, { "epoch": 0.5537441441441442, "grad_norm": 22.661577224731445, "learning_rate": 8.160666689951888e-06, "loss": 12.73021411895752, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.288558317348361, "loss_total_v6": 0.7956383898854256, "step": 2401 }, { "epoch": 0.5539747747747747, "grad_norm": 21.439937591552734, "learning_rate": 8.159105989196294e-06, "loss": 12.818164825439453, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.2814514581114054, "loss_total_v6": 0.801135316491127, "step": 2402 }, { "epoch": 0.5542054054054054, "grad_norm": 25.586763381958008, "learning_rate": 8.157544775963285e-06, "loss": 12.500625610351562, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.2818688992410898, "loss_total_v6": 0.7812890559434891, "step": 2403 }, { "epoch": 0.5544360360360361, "grad_norm": 23.027873992919922, "learning_rate": 8.155983050506123e-06, "loss": 12.646200180053711, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2859319429844618, "loss_total_v6": 0.7903875336050987, "step": 2404 }, { "epoch": 0.5546666666666666, "grad_norm": 22.06444549560547, "learning_rate": 8.154420813078155e-06, "loss": 12.320138931274414, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.26834545470774174, "loss_total_v6": 0.7700086534023285, "step": 2405 }, { "epoch": 0.5548972972972973, "grad_norm": 21.75238037109375, "learning_rate": 8.152858063932811e-06, "loss": 12.239340782165527, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.2658285479992628, "loss_total_v6": 0.7649587988853455, "step": 2406 }, { "epoch": 0.555127927927928, "grad_norm": 22.96917724609375, "learning_rate": 8.151294803323607e-06, "loss": 13.259016990661621, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.33039747923612595, "loss_total_v6": 0.8286885321140289, "step": 2407 }, { "epoch": 0.5553585585585585, "grad_norm": 23.109046936035156, "learning_rate": 8.149731031504136e-06, "loss": 12.401338577270508, "loss_alignment": 0.98291015625, "loss_alignment_w": 0.491455078125, "loss_sft": 0.2835980039089918, "loss_total_v6": 0.7750835940241814, "step": 2408 }, { "epoch": 0.5555891891891892, "grad_norm": 21.429662704467773, "learning_rate": 8.148166748728075e-06, "loss": 12.176673889160156, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.2623543068766594, "loss_total_v6": 0.7610420659184456, "step": 2409 }, { "epoch": 0.5558198198198199, "grad_norm": 22.061185836791992, "learning_rate": 8.146601955249187e-06, "loss": 12.53976821899414, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.2877943702042103, "loss_total_v6": 0.7837354987859726, "step": 2410 }, { "epoch": 0.5560504504504504, "grad_norm": 21.81653594970703, "learning_rate": 8.145036651321319e-06, "loss": 12.87278938293457, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.29701149091124535, "loss_total_v6": 0.8045493215322495, "step": 2411 }, { "epoch": 0.5562810810810811, "grad_norm": 22.352523803710938, "learning_rate": 8.143470837198394e-06, "loss": 12.26861572265625, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.26367565989494324, "loss_total_v6": 0.7667884454131126, "step": 2412 }, { "epoch": 0.5565117117117118, "grad_norm": 23.85882568359375, "learning_rate": 8.141904513134425e-06, "loss": 12.417394638061523, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.27614820562303066, "loss_total_v6": 0.7760871648788452, "step": 2413 }, { "epoch": 0.5567423423423423, "grad_norm": 24.37571907043457, "learning_rate": 8.140337679383502e-06, "loss": 13.107892990112305, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.3153675477951765, "loss_total_v6": 0.8192432671785355, "step": 2414 }, { "epoch": 0.556972972972973, "grad_norm": 21.783382415771484, "learning_rate": 8.138770336199803e-06, "loss": 12.326091766357422, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.2720744702965021, "loss_total_v6": 0.7703807502985001, "step": 2415 }, { "epoch": 0.5572036036036037, "grad_norm": 21.138835906982422, "learning_rate": 8.137202483837584e-06, "loss": 12.162951469421387, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2531958743929863, "loss_total_v6": 0.7601844295859337, "step": 2416 }, { "epoch": 0.5574342342342342, "grad_norm": 23.159765243530273, "learning_rate": 8.135634122551188e-06, "loss": 12.49818229675293, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.27616207860410213, "loss_total_v6": 0.7811364307999611, "step": 2417 }, { "epoch": 0.5576648648648649, "grad_norm": 20.364551544189453, "learning_rate": 8.134065252595035e-06, "loss": 12.197490692138672, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.25868105702102184, "loss_total_v6": 0.7623431980609894, "step": 2418 }, { "epoch": 0.5578954954954954, "grad_norm": 21.495779037475586, "learning_rate": 8.132495874223633e-06, "loss": 12.632109642028809, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.29074285738170147, "loss_total_v6": 0.7895068675279617, "step": 2419 }, { "epoch": 0.5581261261261261, "grad_norm": 21.945968627929688, "learning_rate": 8.13092598769157e-06, "loss": 12.44985580444336, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.26969317346811295, "loss_total_v6": 0.7781160101294518, "step": 2420 }, { "epoch": 0.5583567567567568, "grad_norm": 22.7623291015625, "learning_rate": 8.129355593253513e-06, "loss": 12.817422866821289, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.3021418731659651, "loss_total_v6": 0.8010889738798141, "step": 2421 }, { "epoch": 0.5585873873873873, "grad_norm": 21.71157455444336, "learning_rate": 8.127784691164223e-06, "loss": 12.255204200744629, "loss_alignment": 0.97705078125, "loss_alignment_w": 0.488525390625, "loss_sft": 0.2764483243227005, "loss_total_v6": 0.7659502476453781, "step": 2422 }, { "epoch": 0.558818018018018, "grad_norm": 22.9461669921875, "learning_rate": 8.126213281678527e-06, "loss": 12.589740753173828, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.28598901629447937, "loss_total_v6": 0.7868587523698807, "step": 2423 }, { "epoch": 0.5590486486486487, "grad_norm": 21.799121856689453, "learning_rate": 8.124641365051346e-06, "loss": 12.684070587158203, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.2913048155605793, "loss_total_v6": 0.7927543967962265, "step": 2424 }, { "epoch": 0.5592792792792792, "grad_norm": 25.79604148864746, "learning_rate": 8.123068941537681e-06, "loss": 12.746292114257812, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2925996296107769, "loss_total_v6": 0.7966432496905327, "step": 2425 }, { "epoch": 0.5595099099099099, "grad_norm": 21.067493438720703, "learning_rate": 8.121496011392613e-06, "loss": 12.624654769897461, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.29480878077447414, "loss_total_v6": 0.7890409231185913, "step": 2426 }, { "epoch": 0.5597405405405406, "grad_norm": 21.151620864868164, "learning_rate": 8.119922574871305e-06, "loss": 12.55868148803711, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2773034404963255, "loss_total_v6": 0.7849176004528999, "step": 2427 }, { "epoch": 0.5599711711711711, "grad_norm": 22.342695236206055, "learning_rate": 8.118348632229006e-06, "loss": 12.169853210449219, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.2625231761485338, "loss_total_v6": 0.760615848004818, "step": 2428 }, { "epoch": 0.5602018018018018, "grad_norm": 28.919151306152344, "learning_rate": 8.116774183721045e-06, "loss": 12.816060066223145, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.30172091349959373, "loss_total_v6": 0.8010037764906883, "step": 2429 }, { "epoch": 0.5604324324324325, "grad_norm": 22.56797981262207, "learning_rate": 8.115199229602831e-06, "loss": 12.574749946594238, "loss_alignment": 0.98486328125, "loss_alignment_w": 0.492431640625, "loss_sft": 0.292727317661047, "loss_total_v6": 0.7859218791127205, "step": 2430 }, { "epoch": 0.560663063063063, "grad_norm": 24.20073127746582, "learning_rate": 8.113623770129853e-06, "loss": 12.50795841217041, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.2801909949630499, "loss_total_v6": 0.7817474007606506, "step": 2431 }, { "epoch": 0.5608936936936937, "grad_norm": 23.804101943969727, "learning_rate": 8.112047805557693e-06, "loss": 13.6392240524292, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.33684176206588745, "loss_total_v6": 0.8524514734745026, "step": 2432 }, { "epoch": 0.5611243243243244, "grad_norm": 24.131946563720703, "learning_rate": 8.110471336142003e-06, "loss": 12.135319709777832, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.263309795409441, "loss_total_v6": 0.7584574893116951, "step": 2433 }, { "epoch": 0.5613549549549549, "grad_norm": 24.156902313232422, "learning_rate": 8.108894362138521e-06, "loss": 12.324771881103516, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.27032882906496525, "loss_total_v6": 0.7702983096241951, "step": 2434 }, { "epoch": 0.5615855855855856, "grad_norm": 23.370962142944336, "learning_rate": 8.107316883803071e-06, "loss": 12.585678100585938, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.283369954675436, "loss_total_v6": 0.7866048291325569, "step": 2435 }, { "epoch": 0.5618162162162162, "grad_norm": 22.689008712768555, "learning_rate": 8.105738901391553e-06, "loss": 13.143838882446289, "loss_alignment": 1.01318359375, "loss_alignment_w": 0.506591796875, "loss_sft": 0.3139368202537298, "loss_total_v6": 0.8214899078011513, "step": 2436 }, { "epoch": 0.5620468468468468, "grad_norm": 24.97484016418457, "learning_rate": 8.10416041515995e-06, "loss": 12.328079223632812, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.26971153542399406, "loss_total_v6": 0.7705049887299538, "step": 2437 }, { "epoch": 0.5622774774774775, "grad_norm": 21.603431701660156, "learning_rate": 8.10258142536433e-06, "loss": 13.01129150390625, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.3064766302704811, "loss_total_v6": 0.8132057413458824, "step": 2438 }, { "epoch": 0.5625081081081081, "grad_norm": 22.600624084472656, "learning_rate": 8.101001932260837e-06, "loss": 12.478919982910156, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.27657556906342506, "loss_total_v6": 0.779932513833046, "step": 2439 }, { "epoch": 0.5627387387387387, "grad_norm": 23.122331619262695, "learning_rate": 8.099421936105702e-06, "loss": 12.247819900512695, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.26536662317812443, "loss_total_v6": 0.7654886990785599, "step": 2440 }, { "epoch": 0.5629693693693694, "grad_norm": 22.042293548583984, "learning_rate": 8.097841437155237e-06, "loss": 12.034181594848633, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.24859633669257164, "loss_total_v6": 0.7521363645792007, "step": 2441 }, { "epoch": 0.5632, "grad_norm": 24.941835403442383, "learning_rate": 8.096260435665831e-06, "loss": 12.681259155273438, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.28487306647002697, "loss_total_v6": 0.7925787270069122, "step": 2442 }, { "epoch": 0.5634306306306306, "grad_norm": 22.25433921813965, "learning_rate": 8.09467893189396e-06, "loss": 12.356393814086914, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.26388221606612206, "loss_total_v6": 0.772274561226368, "step": 2443 }, { "epoch": 0.5636612612612613, "grad_norm": 22.98162078857422, "learning_rate": 8.093096926096178e-06, "loss": 13.194425582885742, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.3143519163131714, "loss_total_v6": 0.8246515840291977, "step": 2444 }, { "epoch": 0.5638918918918919, "grad_norm": 23.17856788635254, "learning_rate": 8.091514418529124e-06, "loss": 12.316478729248047, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.27624959498643875, "loss_total_v6": 0.7697798758745193, "step": 2445 }, { "epoch": 0.5641225225225225, "grad_norm": 24.47858238220215, "learning_rate": 8.089931409449511e-06, "loss": 12.30507755279541, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.2734161261469126, "loss_total_v6": 0.7690673470497131, "step": 2446 }, { "epoch": 0.5643531531531532, "grad_norm": 21.98017120361328, "learning_rate": 8.088347899114146e-06, "loss": 12.145917892456055, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.25788390822708607, "loss_total_v6": 0.759119875729084, "step": 2447 }, { "epoch": 0.5645837837837838, "grad_norm": 21.637943267822266, "learning_rate": 8.0867638877799e-06, "loss": 12.129329681396484, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.2593953460454941, "loss_total_v6": 0.7580831050872803, "step": 2448 }, { "epoch": 0.5648144144144144, "grad_norm": 23.295080184936523, "learning_rate": 8.085179375703745e-06, "loss": 12.349300384521484, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.27312824688851833, "loss_total_v6": 0.7718312367796898, "step": 2449 }, { "epoch": 0.565045045045045, "grad_norm": 22.20286750793457, "learning_rate": 8.083594363142717e-06, "loss": 12.55473518371582, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2723570875823498, "loss_total_v6": 0.7846709340810776, "step": 2450 }, { "epoch": 0.5652756756756757, "grad_norm": 23.889148712158203, "learning_rate": 8.082008850353942e-06, "loss": 12.912382125854492, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.30415524542331696, "loss_total_v6": 0.8070238828659058, "step": 2451 }, { "epoch": 0.5655063063063063, "grad_norm": 26.346445083618164, "learning_rate": 8.080422837594627e-06, "loss": 12.499276161193848, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.2854314520955086, "loss_total_v6": 0.7812047749757767, "step": 2452 }, { "epoch": 0.5657369369369369, "grad_norm": 21.637434005737305, "learning_rate": 8.078836325122057e-06, "loss": 12.855342864990234, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.29982731118798256, "loss_total_v6": 0.8034588992595673, "step": 2453 }, { "epoch": 0.5659675675675676, "grad_norm": 21.956867218017578, "learning_rate": 8.077249313193602e-06, "loss": 12.271623611450195, "loss_alignment": 0.98291015625, "loss_alignment_w": 0.491455078125, "loss_sft": 0.27600966952741146, "loss_total_v6": 0.7669764682650566, "step": 2454 }, { "epoch": 0.5661981981981982, "grad_norm": 23.30752182006836, "learning_rate": 8.075661802066707e-06, "loss": 12.554397583007812, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2794618308544159, "loss_total_v6": 0.7846498191356659, "step": 2455 }, { "epoch": 0.5664288288288288, "grad_norm": 20.609180450439453, "learning_rate": 8.074073791998907e-06, "loss": 12.675973892211914, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.28654151409864426, "loss_total_v6": 0.792248323559761, "step": 2456 }, { "epoch": 0.5666594594594595, "grad_norm": 24.15949821472168, "learning_rate": 8.072485283247807e-06, "loss": 12.129434585571289, "loss_alignment": 0.98583984375, "loss_alignment_w": 0.492919921875, "loss_sft": 0.2646661791950464, "loss_total_v6": 0.7580896392464638, "step": 2457 }, { "epoch": 0.5668900900900901, "grad_norm": 23.290674209594727, "learning_rate": 8.070896276071102e-06, "loss": 12.860843658447266, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.30526756681501865, "loss_total_v6": 0.8038027286529541, "step": 2458 }, { "epoch": 0.5671207207207207, "grad_norm": 23.0158634185791, "learning_rate": 8.069306770726563e-06, "loss": 12.676396369934082, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.29930904507637024, "loss_total_v6": 0.7922747507691383, "step": 2459 }, { "epoch": 0.5673513513513514, "grad_norm": 23.261505126953125, "learning_rate": 8.067716767472045e-06, "loss": 13.017279624938965, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.3153652809560299, "loss_total_v6": 0.8135799840092659, "step": 2460 }, { "epoch": 0.567581981981982, "grad_norm": 21.562036514282227, "learning_rate": 8.06612626656548e-06, "loss": 12.026570320129395, "loss_alignment": 0.9912109375, "loss_alignment_w": 0.49560546875, "loss_sft": 0.25613149255514145, "loss_total_v6": 0.7516606524586678, "step": 2461 }, { "epoch": 0.5678126126126126, "grad_norm": 22.080875396728516, "learning_rate": 8.064535268264883e-06, "loss": 12.542745590209961, "loss_alignment": 0.9853515625, "loss_alignment_w": 0.49267578125, "loss_sft": 0.29080329835414886, "loss_total_v6": 0.783921554684639, "step": 2462 }, { "epoch": 0.5680432432432433, "grad_norm": 20.322582244873047, "learning_rate": 8.06294377282835e-06, "loss": 12.123495101928711, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.2630742508918047, "loss_total_v6": 0.7577183917164803, "step": 2463 }, { "epoch": 0.5682738738738738, "grad_norm": 23.090179443359375, "learning_rate": 8.061351780514057e-06, "loss": 12.512286186218262, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.2876941990107298, "loss_total_v6": 0.7820179089903831, "step": 2464 }, { "epoch": 0.5685045045045045, "grad_norm": 23.081998825073242, "learning_rate": 8.05975929158026e-06, "loss": 12.722479820251465, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.2979015540331602, "loss_total_v6": 0.7951549738645554, "step": 2465 }, { "epoch": 0.5687351351351352, "grad_norm": 23.83824920654297, "learning_rate": 8.058166306285297e-06, "loss": 12.690756797790527, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.2884725611656904, "loss_total_v6": 0.7931722924113274, "step": 2466 }, { "epoch": 0.5689657657657657, "grad_norm": 21.6550350189209, "learning_rate": 8.056572824887584e-06, "loss": 12.65962028503418, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.2920502759516239, "loss_total_v6": 0.7912262976169586, "step": 2467 }, { "epoch": 0.5691963963963964, "grad_norm": 20.142560958862305, "learning_rate": 8.054978847645623e-06, "loss": 12.248136520385742, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.26376903615891933, "loss_total_v6": 0.7655085474252701, "step": 2468 }, { "epoch": 0.5694270270270271, "grad_norm": 23.126245498657227, "learning_rate": 8.053384374817989e-06, "loss": 12.455514907836914, "loss_alignment": 0.9892578125, "loss_alignment_w": 0.49462890625, "loss_sft": 0.28344400972127914, "loss_total_v6": 0.7784696742892265, "step": 2469 }, { "epoch": 0.5696576576576576, "grad_norm": 21.224166870117188, "learning_rate": 8.051789406663342e-06, "loss": 12.896923065185547, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2994964215904474, "loss_total_v6": 0.8060576841235161, "step": 2470 }, { "epoch": 0.5698882882882883, "grad_norm": 22.276920318603516, "learning_rate": 8.05019394344042e-06, "loss": 12.687042236328125, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.29202463664114475, "loss_total_v6": 0.7929401472210884, "step": 2471 }, { "epoch": 0.570118918918919, "grad_norm": 20.23021125793457, "learning_rate": 8.048597985408047e-06, "loss": 12.600006103515625, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2808323297649622, "loss_total_v6": 0.7875004038214684, "step": 2472 }, { "epoch": 0.5703495495495495, "grad_norm": 19.923532485961914, "learning_rate": 8.047001532825119e-06, "loss": 12.006244659423828, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.26308559253811836, "loss_total_v6": 0.7503902837634087, "step": 2473 }, { "epoch": 0.5705801801801802, "grad_norm": 21.190448760986328, "learning_rate": 8.045404585950618e-06, "loss": 12.55309772491455, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.28202040679752827, "loss_total_v6": 0.78456861525774, "step": 2474 }, { "epoch": 0.5708108108108109, "grad_norm": 24.043132781982422, "learning_rate": 8.043807145043604e-06, "loss": 12.230493545532227, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.25898895412683487, "loss_total_v6": 0.7644058465957642, "step": 2475 }, { "epoch": 0.5710414414414414, "grad_norm": 22.61359214782715, "learning_rate": 8.042209210363216e-06, "loss": 12.653609275817871, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.29167453572154045, "loss_total_v6": 0.7908505648374557, "step": 2476 }, { "epoch": 0.5712720720720721, "grad_norm": 23.030353546142578, "learning_rate": 8.040610782168678e-06, "loss": 13.054242134094238, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.3099239803850651, "loss_total_v6": 0.8158901631832123, "step": 2477 }, { "epoch": 0.5715027027027026, "grad_norm": 23.80879020690918, "learning_rate": 8.039011860719288e-06, "loss": 12.272974014282227, "loss_alignment": 0.9853515625, "loss_alignment_w": 0.49267578125, "loss_sft": 0.27374424785375595, "loss_total_v6": 0.767060898244381, "step": 2478 }, { "epoch": 0.5717333333333333, "grad_norm": 20.69883155822754, "learning_rate": 8.037412446274428e-06, "loss": 12.43227481842041, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2744232006371021, "loss_total_v6": 0.7770171836018562, "step": 2479 }, { "epoch": 0.571963963963964, "grad_norm": 23.200603485107422, "learning_rate": 8.035812539093557e-06, "loss": 12.646797180175781, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.27547120302915573, "loss_total_v6": 0.7904248312115669, "step": 2480 }, { "epoch": 0.5721945945945945, "grad_norm": 24.754241943359375, "learning_rate": 8.034212139436217e-06, "loss": 12.28675651550293, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.26493155770003796, "loss_total_v6": 0.7679222822189331, "step": 2481 }, { "epoch": 0.5724252252252252, "grad_norm": 20.31218147277832, "learning_rate": 8.032611247562028e-06, "loss": 12.80435848236084, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.3030799776315689, "loss_total_v6": 0.8002723976969719, "step": 2482 }, { "epoch": 0.5726558558558559, "grad_norm": 22.426279067993164, "learning_rate": 8.031009863730692e-06, "loss": 13.079923629760742, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2988947778940201, "loss_total_v6": 0.8174952417612076, "step": 2483 }, { "epoch": 0.5728864864864864, "grad_norm": 22.847185134887695, "learning_rate": 8.029407988201986e-06, "loss": 12.945621490478516, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.30206699669361115, "loss_total_v6": 0.8091013059020042, "step": 2484 }, { "epoch": 0.5731171171171171, "grad_norm": 28.078571319580078, "learning_rate": 8.027805621235771e-06, "loss": 11.885051727294922, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.2415491994470358, "loss_total_v6": 0.7428156733512878, "step": 2485 }, { "epoch": 0.5733477477477478, "grad_norm": 23.031274795532227, "learning_rate": 8.02620276309199e-06, "loss": 12.157554626464844, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.26001504622399807, "loss_total_v6": 0.7598472163081169, "step": 2486 }, { "epoch": 0.5735783783783783, "grad_norm": 21.840179443359375, "learning_rate": 8.024599414030657e-06, "loss": 12.749320983886719, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.3009066730737686, "loss_total_v6": 0.7968325614929199, "step": 2487 }, { "epoch": 0.573809009009009, "grad_norm": 25.934326171875, "learning_rate": 8.022995574311876e-06, "loss": 12.979269027709961, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.3002790417522192, "loss_total_v6": 0.8112043216824532, "step": 2488 }, { "epoch": 0.5740396396396397, "grad_norm": 24.243083953857422, "learning_rate": 8.021391244195822e-06, "loss": 12.204313278198242, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.26402074471116066, "loss_total_v6": 0.7627695500850677, "step": 2489 }, { "epoch": 0.5742702702702702, "grad_norm": 24.281925201416016, "learning_rate": 8.019786423942752e-06, "loss": 12.428230285644531, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.27047779969871044, "loss_total_v6": 0.7767644077539444, "step": 2490 }, { "epoch": 0.5745009009009009, "grad_norm": 24.093584060668945, "learning_rate": 8.018181113813009e-06, "loss": 12.687957763671875, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.29118165001273155, "loss_total_v6": 0.7929974049329758, "step": 2491 }, { "epoch": 0.5747315315315316, "grad_norm": 22.980239868164062, "learning_rate": 8.016575314067005e-06, "loss": 12.267168045043945, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2630816772580147, "loss_total_v6": 0.7666980102658272, "step": 2492 }, { "epoch": 0.5749621621621621, "grad_norm": 25.370067596435547, "learning_rate": 8.01496902496524e-06, "loss": 12.702911376953125, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2870807610452175, "loss_total_v6": 0.7939319610595703, "step": 2493 }, { "epoch": 0.5751927927927928, "grad_norm": 24.00690460205078, "learning_rate": 8.013362246768288e-06, "loss": 12.789470672607422, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.29815178737044334, "loss_total_v6": 0.7993419542908669, "step": 2494 }, { "epoch": 0.5754234234234235, "grad_norm": 21.208606719970703, "learning_rate": 8.011754979736804e-06, "loss": 12.50526237487793, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.281533095985651, "loss_total_v6": 0.781578853726387, "step": 2495 }, { "epoch": 0.575654054054054, "grad_norm": 24.755388259887695, "learning_rate": 8.010147224131523e-06, "loss": 12.601511001586914, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.2909208592027426, "loss_total_v6": 0.7875944375991821, "step": 2496 }, { "epoch": 0.5758846846846847, "grad_norm": 22.031890869140625, "learning_rate": 8.00853898021326e-06, "loss": 12.72746467590332, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.2952986788004637, "loss_total_v6": 0.7954665273427963, "step": 2497 }, { "epoch": 0.5761153153153153, "grad_norm": 23.093730926513672, "learning_rate": 8.006930248242905e-06, "loss": 12.853116989135742, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.29887955635786057, "loss_total_v6": 0.8033198490738869, "step": 2498 }, { "epoch": 0.5763459459459459, "grad_norm": 23.57525062561035, "learning_rate": 8.005321028481434e-06, "loss": 12.675237655639648, "loss_alignment": 0.9853515625, "loss_alignment_w": 0.49267578125, "loss_sft": 0.2987789511680603, "loss_total_v6": 0.7922023981809616, "step": 2499 }, { "epoch": 0.5765765765765766, "grad_norm": 21.13600730895996, "learning_rate": 8.003711321189895e-06, "loss": 12.648093223571777, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.2908720541745424, "loss_total_v6": 0.7905058115720749, "step": 2500 }, { "epoch": 0.5765765765765766, "eval_loss": 0.9429246187210083, "eval_loss_alignment": 1.0095850278253424, "eval_loss_alignment_w": 0.5047925139126712, "eval_loss_contrast": 0.0, "eval_loss_highlevel": 0.0, "eval_loss_sft": 0.43813209751935583, "eval_loss_total_v6": 0.9429246073580224, "eval_loss_wm_recon": 0.0, "eval_loss_wm_recon_w": 0.0, "eval_runtime": 360.7558, "eval_samples_per_second": 4.837, "eval_steps_per_second": 0.607, "step": 2500 }, { "epoch": 0.5768072072072072, "grad_norm": 21.993303298950195, "learning_rate": 8.002101126629422e-06, "loss": 12.471492767333984, "loss_alignment": 1.00732421875, "loss_alignment_w": 0.503662109375, "loss_sft": 0.2764165513217449, "loss_total_v6": 0.7794682756066322, "step": 2501 }, { "epoch": 0.5770378378378378, "grad_norm": 20.564376831054688, "learning_rate": 8.00049044506122e-06, "loss": 12.5150146484375, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2775497827678919, "loss_total_v6": 0.7821884304285049, "step": 2502 }, { "epoch": 0.5772684684684685, "grad_norm": 26.071537017822266, "learning_rate": 7.99887927674658e-06, "loss": 13.195486068725586, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.32120826840400696, "loss_total_v6": 0.8247178494930267, "step": 2503 }, { "epoch": 0.5774990990990991, "grad_norm": 25.444177627563477, "learning_rate": 7.997267621946871e-06, "loss": 12.145389556884766, "loss_alignment": 0.98388671875, "loss_alignment_w": 0.491943359375, "loss_sft": 0.2671586852520704, "loss_total_v6": 0.7590868100523949, "step": 2504 }, { "epoch": 0.5777297297297297, "grad_norm": 21.387306213378906, "learning_rate": 7.995655480923536e-06, "loss": 12.297107696533203, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.27458119578659534, "loss_total_v6": 0.7685692310333252, "step": 2505 }, { "epoch": 0.5779603603603604, "grad_norm": 24.17843246459961, "learning_rate": 7.994042853938102e-06, "loss": 11.851012229919434, "loss_alignment": 0.97802734375, "loss_alignment_w": 0.489013671875, "loss_sft": 0.2526054345071316, "loss_total_v6": 0.740688294172287, "step": 2506 }, { "epoch": 0.578190990990991, "grad_norm": 22.64731788635254, "learning_rate": 7.992429741252174e-06, "loss": 12.59103012084961, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.28632904775440693, "loss_total_v6": 0.7869394049048424, "step": 2507 }, { "epoch": 0.5784216216216216, "grad_norm": 22.66516876220703, "learning_rate": 7.990816143127431e-06, "loss": 12.33708381652832, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.2752181179821491, "loss_total_v6": 0.7710677236318588, "step": 2508 }, { "epoch": 0.5786522522522523, "grad_norm": 22.208925247192383, "learning_rate": 7.989202059825639e-06, "loss": 12.421218872070312, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2646837569773197, "loss_total_v6": 0.7763262018561363, "step": 2509 }, { "epoch": 0.5788828828828829, "grad_norm": 25.850393295288086, "learning_rate": 7.987587491608636e-06, "loss": 12.781909942626953, "loss_alignment": 0.9892578125, "loss_alignment_w": 0.49462890625, "loss_sft": 0.30329446122050285, "loss_total_v6": 0.7988693937659264, "step": 2510 }, { "epoch": 0.5791135135135135, "grad_norm": 22.542736053466797, "learning_rate": 7.985972438738343e-06, "loss": 12.128480911254883, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.2591286674141884, "loss_total_v6": 0.758030042052269, "step": 2511 }, { "epoch": 0.5793441441441441, "grad_norm": 23.652326583862305, "learning_rate": 7.984356901476755e-06, "loss": 12.117433547973633, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.24284370988607407, "loss_total_v6": 0.7573395743966103, "step": 2512 }, { "epoch": 0.5795747747747748, "grad_norm": 20.02838706970215, "learning_rate": 7.982740880085948e-06, "loss": 12.807560920715332, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.29780227318406105, "loss_total_v6": 0.8004725500941277, "step": 2513 }, { "epoch": 0.5798054054054054, "grad_norm": 23.12791633605957, "learning_rate": 7.981124374828079e-06, "loss": 12.869610786437988, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.29554638639092445, "loss_total_v6": 0.8043506741523743, "step": 2514 }, { "epoch": 0.580036036036036, "grad_norm": 21.761659622192383, "learning_rate": 7.97950738596538e-06, "loss": 12.729784965515137, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.28522027283906937, "loss_total_v6": 0.7956115379929543, "step": 2515 }, { "epoch": 0.5802666666666667, "grad_norm": 23.84274673461914, "learning_rate": 7.977889913760163e-06, "loss": 12.521381378173828, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.27340060845017433, "loss_total_v6": 0.7825863659381866, "step": 2516 }, { "epoch": 0.5804972972972973, "grad_norm": 23.144973754882812, "learning_rate": 7.976271958474816e-06, "loss": 11.982925415039062, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.2564554139971733, "loss_total_v6": 0.7489328384399414, "step": 2517 }, { "epoch": 0.5807279279279279, "grad_norm": 24.25390625, "learning_rate": 7.974653520371813e-06, "loss": 13.089787483215332, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.31187089532613754, "loss_total_v6": 0.8181117326021194, "step": 2518 }, { "epoch": 0.5809585585585586, "grad_norm": 20.816617965698242, "learning_rate": 7.973034599713693e-06, "loss": 13.043862342834473, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.3164773229509592, "loss_total_v6": 0.8152413889765739, "step": 2519 }, { "epoch": 0.5811891891891892, "grad_norm": 22.36139678955078, "learning_rate": 7.971415196763088e-06, "loss": 12.167840957641602, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.2568126618862152, "loss_total_v6": 0.7604900225996971, "step": 2520 }, { "epoch": 0.5814198198198198, "grad_norm": 21.72694969177246, "learning_rate": 7.969795311782698e-06, "loss": 12.051353454589844, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.25755836069583893, "loss_total_v6": 0.7532095909118652, "step": 2521 }, { "epoch": 0.5816504504504505, "grad_norm": 22.652467727661133, "learning_rate": 7.968174945035301e-06, "loss": 12.623390197753906, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.27837225794792175, "loss_total_v6": 0.788961872458458, "step": 2522 }, { "epoch": 0.581881081081081, "grad_norm": 22.57976722717285, "learning_rate": 7.966554096783763e-06, "loss": 12.876266479492188, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.303820613771677, "loss_total_v6": 0.8047666549682617, "step": 2523 }, { "epoch": 0.5821117117117117, "grad_norm": 22.714590072631836, "learning_rate": 7.964932767291019e-06, "loss": 12.905977249145508, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.3052808418869972, "loss_total_v6": 0.8066235929727554, "step": 2524 }, { "epoch": 0.5823423423423424, "grad_norm": 23.695953369140625, "learning_rate": 7.963310956820085e-06, "loss": 12.616715431213379, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.2866526562720537, "loss_total_v6": 0.7885447293519974, "step": 2525 }, { "epoch": 0.582572972972973, "grad_norm": 22.802797317504883, "learning_rate": 7.961688665634054e-06, "loss": 12.034114837646484, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.24891254492104053, "loss_total_v6": 0.7521321773529053, "step": 2526 }, { "epoch": 0.5828036036036036, "grad_norm": 22.325376510620117, "learning_rate": 7.960065893996099e-06, "loss": 12.82235336303711, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.30322808399796486, "loss_total_v6": 0.801397055387497, "step": 2527 }, { "epoch": 0.5830342342342343, "grad_norm": 20.761621475219727, "learning_rate": 7.958442642169469e-06, "loss": 12.455484390258789, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.2828775756061077, "loss_total_v6": 0.7784678041934967, "step": 2528 }, { "epoch": 0.5832648648648648, "grad_norm": 22.87957763671875, "learning_rate": 7.956818910417489e-06, "loss": 12.095376014709473, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.24828580394387245, "loss_total_v6": 0.755961000919342, "step": 2529 }, { "epoch": 0.5834954954954955, "grad_norm": 23.440160751342773, "learning_rate": 7.95519469900357e-06, "loss": 12.15926742553711, "loss_alignment": 0.970703125, "loss_alignment_w": 0.4853515625, "loss_sft": 0.27515193074941635, "loss_total_v6": 0.7599542066454887, "step": 2530 }, { "epoch": 0.5837261261261262, "grad_norm": 24.661108016967773, "learning_rate": 7.953570008191192e-06, "loss": 12.059829711914062, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.2535409778356552, "loss_total_v6": 0.7537393420934677, "step": 2531 }, { "epoch": 0.5839567567567567, "grad_norm": 23.743389129638672, "learning_rate": 7.951944838243916e-06, "loss": 12.453401565551758, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.2807789705693722, "loss_total_v6": 0.7783375829458237, "step": 2532 }, { "epoch": 0.5841873873873874, "grad_norm": 19.568517684936523, "learning_rate": 7.950319189425384e-06, "loss": 12.16193675994873, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.2641341518610716, "loss_total_v6": 0.760121077299118, "step": 2533 }, { "epoch": 0.5844180180180181, "grad_norm": 24.094820022583008, "learning_rate": 7.948693061999307e-06, "loss": 12.055191040039062, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.2531899958848953, "loss_total_v6": 0.7534494176506996, "step": 2534 }, { "epoch": 0.5846486486486486, "grad_norm": 23.616992950439453, "learning_rate": 7.947066456229485e-06, "loss": 12.12067699432373, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2511183097958565, "loss_total_v6": 0.7575422823429108, "step": 2535 }, { "epoch": 0.5848792792792793, "grad_norm": 20.626169204711914, "learning_rate": 7.945439372379782e-06, "loss": 12.431005477905273, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.2767547331750393, "loss_total_v6": 0.7769378274679184, "step": 2536 }, { "epoch": 0.58510990990991, "grad_norm": 23.658523559570312, "learning_rate": 7.943811810714155e-06, "loss": 12.099624633789062, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.25932410545647144, "loss_total_v6": 0.7562265396118164, "step": 2537 }, { "epoch": 0.5853405405405405, "grad_norm": 23.18024444580078, "learning_rate": 7.942183771496628e-06, "loss": 12.496428489685059, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.2799129206687212, "loss_total_v6": 0.7810267806053162, "step": 2538 }, { "epoch": 0.5855711711711712, "grad_norm": 22.637922286987305, "learning_rate": 7.940555254991305e-06, "loss": 12.347549438476562, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.27509403601288795, "loss_total_v6": 0.7717218548059464, "step": 2539 }, { "epoch": 0.5858018018018017, "grad_norm": 22.398183822631836, "learning_rate": 7.938926261462366e-06, "loss": 12.106419563293457, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.2599623668938875, "loss_total_v6": 0.7566512078046799, "step": 2540 }, { "epoch": 0.5860324324324324, "grad_norm": 25.052610397338867, "learning_rate": 7.937296791174072e-06, "loss": 12.713298797607422, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.2909953501075506, "loss_total_v6": 0.7945811748504639, "step": 2541 }, { "epoch": 0.5862630630630631, "grad_norm": 21.90016746520996, "learning_rate": 7.935666844390761e-06, "loss": 12.160083770751953, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2558853458613157, "loss_total_v6": 0.7600052133202553, "step": 2542 }, { "epoch": 0.5864936936936936, "grad_norm": 21.185731887817383, "learning_rate": 7.934036421376845e-06, "loss": 12.546131134033203, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.28874140605330467, "loss_total_v6": 0.7841332405805588, "step": 2543 }, { "epoch": 0.5867243243243243, "grad_norm": 23.668596267700195, "learning_rate": 7.932405522396814e-06, "loss": 12.345458984375, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.269699078053236, "loss_total_v6": 0.7715911418199539, "step": 2544 }, { "epoch": 0.586954954954955, "grad_norm": 23.34402084350586, "learning_rate": 7.930774147715235e-06, "loss": 12.502180099487305, "loss_alignment": 0.98486328125, "loss_alignment_w": 0.492431640625, "loss_sft": 0.2875202652066946, "loss_total_v6": 0.781386248767376, "step": 2545 }, { "epoch": 0.5871855855855855, "grad_norm": 22.878250122070312, "learning_rate": 7.929142297596757e-06, "loss": 12.546289443969727, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.27558291517198086, "loss_total_v6": 0.7841430753469467, "step": 2546 }, { "epoch": 0.5874162162162162, "grad_norm": 22.598834991455078, "learning_rate": 7.927509972306103e-06, "loss": 12.661238670349121, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.284552501514554, "loss_total_v6": 0.7913274243474007, "step": 2547 }, { "epoch": 0.5876468468468469, "grad_norm": 23.067214965820312, "learning_rate": 7.925877172108067e-06, "loss": 12.7821044921875, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.3016433250159025, "loss_total_v6": 0.7988815009593964, "step": 2548 }, { "epoch": 0.5878774774774774, "grad_norm": 23.147266387939453, "learning_rate": 7.92424389726753e-06, "loss": 12.556344985961914, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.28202501498162746, "loss_total_v6": 0.7847715839743614, "step": 2549 }, { "epoch": 0.5881081081081081, "grad_norm": 20.569292068481445, "learning_rate": 7.922610148049445e-06, "loss": 12.797642707824707, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.30073766596615314, "loss_total_v6": 0.7998526617884636, "step": 2550 }, { "epoch": 0.5883387387387388, "grad_norm": 22.982147216796875, "learning_rate": 7.92097592471884e-06, "loss": 12.75132942199707, "loss_alignment": 0.96484375, "loss_alignment_w": 0.482421875, "loss_sft": 0.3138954360038042, "loss_total_v6": 0.7969581410288811, "step": 2551 }, { "epoch": 0.5885693693693693, "grad_norm": 21.20234489440918, "learning_rate": 7.919341227540828e-06, "loss": 12.379101753234863, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.2740295082330704, "loss_total_v6": 0.7736938297748566, "step": 2552 }, { "epoch": 0.5888, "grad_norm": 21.52509307861328, "learning_rate": 7.917706056780588e-06, "loss": 12.896925926208496, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.30570695735514164, "loss_total_v6": 0.8060579001903534, "step": 2553 }, { "epoch": 0.5890306306306307, "grad_norm": 22.82732391357422, "learning_rate": 7.916070412703383e-06, "loss": 12.549456596374512, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.2860042341053486, "loss_total_v6": 0.7843410149216652, "step": 2554 }, { "epoch": 0.5892612612612612, "grad_norm": 22.13241195678711, "learning_rate": 7.914434295574552e-06, "loss": 13.05551815032959, "loss_alignment": 0.98583984375, "loss_alignment_w": 0.492919921875, "loss_sft": 0.32410280406475067, "loss_total_v6": 0.81596989184618, "step": 2555 }, { "epoch": 0.5894918918918919, "grad_norm": 29.25642967224121, "learning_rate": 7.912797705659508e-06, "loss": 12.538190841674805, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2815007269382477, "loss_total_v6": 0.7836369499564171, "step": 2556 }, { "epoch": 0.5897225225225226, "grad_norm": 21.98497200012207, "learning_rate": 7.911160643223742e-06, "loss": 12.797611236572266, "loss_alignment": 0.9755859375, "loss_alignment_w": 0.48779296875, "loss_sft": 0.3128206767141819, "loss_total_v6": 0.7998507022857666, "step": 2557 }, { "epoch": 0.5899531531531531, "grad_norm": 22.264036178588867, "learning_rate": 7.909523108532824e-06, "loss": 12.287041664123535, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.2632098738104105, "loss_total_v6": 0.7679400816559792, "step": 2558 }, { "epoch": 0.5901837837837838, "grad_norm": 20.940950393676758, "learning_rate": 7.907885101852397e-06, "loss": 12.341444969177246, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2629632242023945, "loss_total_v6": 0.7713402956724167, "step": 2559 }, { "epoch": 0.5904144144144144, "grad_norm": 22.26945686340332, "learning_rate": 7.906246623448184e-06, "loss": 12.801270484924316, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.3059540055692196, "loss_total_v6": 0.8000794053077698, "step": 2560 }, { "epoch": 0.590645045045045, "grad_norm": 23.690509796142578, "learning_rate": 7.90460767358598e-06, "loss": 12.771391868591309, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.29999724216759205, "loss_total_v6": 0.7982119843363762, "step": 2561 }, { "epoch": 0.5908756756756757, "grad_norm": 22.194067001342773, "learning_rate": 7.902968252531661e-06, "loss": 12.695646286010742, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.2982386350631714, "loss_total_v6": 0.793477900326252, "step": 2562 }, { "epoch": 0.5911063063063063, "grad_norm": 25.045461654663086, "learning_rate": 7.90132836055118e-06, "loss": 12.860158920288086, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.28500681184232235, "loss_total_v6": 0.803759902715683, "step": 2563 }, { "epoch": 0.5913369369369369, "grad_norm": 28.74834442138672, "learning_rate": 7.899687997910558e-06, "loss": 12.593666076660156, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.27833032235503197, "loss_total_v6": 0.7871041297912598, "step": 2564 }, { "epoch": 0.5915675675675676, "grad_norm": 21.785974502563477, "learning_rate": 7.898047164875903e-06, "loss": 12.623306274414062, "loss_alignment": 0.982421875, "loss_alignment_w": 0.4912109375, "loss_sft": 0.2970132678747177, "loss_total_v6": 0.7889566347002983, "step": 2565 }, { "epoch": 0.5917981981981982, "grad_norm": 21.877023696899414, "learning_rate": 7.896405861713393e-06, "loss": 12.526236534118652, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.28456827625632286, "loss_total_v6": 0.7828897833824158, "step": 2566 }, { "epoch": 0.5920288288288288, "grad_norm": 22.671085357666016, "learning_rate": 7.894764088689287e-06, "loss": 12.315177917480469, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.27485605515539646, "loss_total_v6": 0.7696985825896263, "step": 2567 }, { "epoch": 0.5922594594594595, "grad_norm": 21.030773162841797, "learning_rate": 7.893121846069913e-06, "loss": 12.780839920043945, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.29800906777381897, "loss_total_v6": 0.798802524805069, "step": 2568 }, { "epoch": 0.5924900900900901, "grad_norm": 22.271406173706055, "learning_rate": 7.891479134121682e-06, "loss": 12.93750286102295, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.30607620999217033, "loss_total_v6": 0.8085939213633537, "step": 2569 }, { "epoch": 0.5927207207207207, "grad_norm": 20.696147918701172, "learning_rate": 7.889835953111075e-06, "loss": 12.528834342956543, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.28697370178997517, "loss_total_v6": 0.7830521836876869, "step": 2570 }, { "epoch": 0.5929513513513514, "grad_norm": 22.968704223632812, "learning_rate": 7.888192303304658e-06, "loss": 12.32805061340332, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.27125086821615696, "loss_total_v6": 0.7705031856894493, "step": 2571 }, { "epoch": 0.593181981981982, "grad_norm": 21.145828247070312, "learning_rate": 7.886548184969063e-06, "loss": 12.891830444335938, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.3056478314101696, "loss_total_v6": 0.8057393878698349, "step": 2572 }, { "epoch": 0.5934126126126126, "grad_norm": 21.616004943847656, "learning_rate": 7.884903598371004e-06, "loss": 12.5582275390625, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.2804489377886057, "loss_total_v6": 0.7848892509937286, "step": 2573 }, { "epoch": 0.5936432432432432, "grad_norm": 24.48699188232422, "learning_rate": 7.883258543777271e-06, "loss": 12.546195983886719, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.2866092137992382, "loss_total_v6": 0.7841372489929199, "step": 2574 }, { "epoch": 0.5938738738738739, "grad_norm": 25.470102310180664, "learning_rate": 7.881613021454728e-06, "loss": 12.742389678955078, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.289471872150898, "loss_total_v6": 0.7963993549346924, "step": 2575 }, { "epoch": 0.5941045045045045, "grad_norm": 24.88486099243164, "learning_rate": 7.879967031670313e-06, "loss": 12.925702095031738, "loss_alignment": 0.9912109375, "loss_alignment_w": 0.49560546875, "loss_sft": 0.31112175062298775, "loss_total_v6": 0.8078563809394836, "step": 2576 }, { "epoch": 0.5943351351351351, "grad_norm": 23.394363403320312, "learning_rate": 7.878320574691047e-06, "loss": 12.865251541137695, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.3055430091917515, "loss_total_v6": 0.8040782138705254, "step": 2577 }, { "epoch": 0.5945657657657658, "grad_norm": 23.790481567382812, "learning_rate": 7.876673650784017e-06, "loss": 12.580406188964844, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.29042576998472214, "loss_total_v6": 0.7862753868103027, "step": 2578 }, { "epoch": 0.5947963963963964, "grad_norm": 21.498306274414062, "learning_rate": 7.875026260216395e-06, "loss": 12.457499504089355, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.27252072654664516, "loss_total_v6": 0.7785937264561653, "step": 2579 }, { "epoch": 0.595027027027027, "grad_norm": 23.690410614013672, "learning_rate": 7.87337840325542e-06, "loss": 13.201826095581055, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.3255261480808258, "loss_total_v6": 0.8251141682267189, "step": 2580 }, { "epoch": 0.5952576576576577, "grad_norm": 22.39194107055664, "learning_rate": 7.871730080168414e-06, "loss": 12.7611722946167, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.28211613558232784, "loss_total_v6": 0.7975732833147049, "step": 2581 }, { "epoch": 0.5954882882882883, "grad_norm": 23.832273483276367, "learning_rate": 7.870081291222772e-06, "loss": 12.867307662963867, "loss_alignment": 0.984375, "loss_alignment_w": 0.4921875, "loss_sft": 0.3118361011147499, "loss_total_v6": 0.8042066991329193, "step": 2582 }, { "epoch": 0.5957189189189189, "grad_norm": 22.230703353881836, "learning_rate": 7.868432036685964e-06, "loss": 11.931344985961914, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.24871505983173847, "loss_total_v6": 0.745709054172039, "step": 2583 }, { "epoch": 0.5959495495495496, "grad_norm": 21.44774627685547, "learning_rate": 7.866782316825535e-06, "loss": 12.150899887084961, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.2584241423755884, "loss_total_v6": 0.7594312280416489, "step": 2584 }, { "epoch": 0.5961801801801802, "grad_norm": 29.335073471069336, "learning_rate": 7.865132131909106e-06, "loss": 12.642780303955078, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.29356124624609947, "loss_total_v6": 0.7901737913489342, "step": 2585 }, { "epoch": 0.5964108108108108, "grad_norm": 21.37993812561035, "learning_rate": 7.863481482204378e-06, "loss": 12.44161605834961, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2717569116503, "loss_total_v6": 0.7776010408997536, "step": 2586 }, { "epoch": 0.5966414414414415, "grad_norm": 22.038249969482422, "learning_rate": 7.861830367979117e-06, "loss": 12.201736450195312, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.2732897009700537, "loss_total_v6": 0.76260856539011, "step": 2587 }, { "epoch": 0.596872072072072, "grad_norm": 24.476472854614258, "learning_rate": 7.860178789501173e-06, "loss": 12.961448669433594, "loss_alignment": 0.9912109375, "loss_alignment_w": 0.49560546875, "loss_sft": 0.3158888500183821, "loss_total_v6": 0.8100905418395996, "step": 2588 }, { "epoch": 0.5971027027027027, "grad_norm": 22.183380126953125, "learning_rate": 7.85852674703847e-06, "loss": 12.794279098510742, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.29475958459079266, "loss_total_v6": 0.7996423989534378, "step": 2589 }, { "epoch": 0.5973333333333334, "grad_norm": 22.78269386291504, "learning_rate": 7.856874240859006e-06, "loss": 12.616621017456055, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.293284272775054, "loss_total_v6": 0.7885387912392616, "step": 2590 }, { "epoch": 0.5975639639639639, "grad_norm": 23.363950729370117, "learning_rate": 7.855221271230852e-06, "loss": 12.866806983947754, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.3008337430655956, "loss_total_v6": 0.804175429046154, "step": 2591 }, { "epoch": 0.5977945945945946, "grad_norm": 23.80318832397461, "learning_rate": 7.85356783842216e-06, "loss": 12.778411865234375, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2965755332261324, "loss_total_v6": 0.7986507192254066, "step": 2592 }, { "epoch": 0.5980252252252253, "grad_norm": 24.637653350830078, "learning_rate": 7.85191394270115e-06, "loss": 12.499002456665039, "loss_alignment": 0.9912109375, "loss_alignment_w": 0.49560546875, "loss_sft": 0.2841478921473026, "loss_total_v6": 0.7811876758933067, "step": 2593 }, { "epoch": 0.5982558558558558, "grad_norm": 20.98017120361328, "learning_rate": 7.850259584336124e-06, "loss": 12.096527099609375, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.2542628925293684, "loss_total_v6": 0.7560329213738441, "step": 2594 }, { "epoch": 0.5984864864864865, "grad_norm": 21.814483642578125, "learning_rate": 7.848604763595452e-06, "loss": 12.706568717956543, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.28921674005687237, "loss_total_v6": 0.7941605746746063, "step": 2595 }, { "epoch": 0.5987171171171172, "grad_norm": 21.04720687866211, "learning_rate": 7.846949480747588e-06, "loss": 12.431013107299805, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2727726586163044, "loss_total_v6": 0.7769383043050766, "step": 2596 }, { "epoch": 0.5989477477477477, "grad_norm": 21.165985107421875, "learning_rate": 7.845293736061052e-06, "loss": 13.206530570983887, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.3236381411552429, "loss_total_v6": 0.8254081532359123, "step": 2597 }, { "epoch": 0.5991783783783784, "grad_norm": 22.976158142089844, "learning_rate": 7.843637529804445e-06, "loss": 12.17835807800293, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.257607314735651, "loss_total_v6": 0.7611473426222801, "step": 2598 }, { "epoch": 0.5994090090090091, "grad_norm": 22.750545501708984, "learning_rate": 7.841980862246437e-06, "loss": 12.241493225097656, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.26080557703971863, "loss_total_v6": 0.7650932893157005, "step": 2599 }, { "epoch": 0.5996396396396396, "grad_norm": 21.057703018188477, "learning_rate": 7.84032373365578e-06, "loss": 12.587358474731445, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.28616056218743324, "loss_total_v6": 0.7867098897695541, "step": 2600 }, { "epoch": 0.5998702702702703, "grad_norm": 23.79821014404297, "learning_rate": 7.838666144301294e-06, "loss": 12.143655776977539, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.2652040682733059, "loss_total_v6": 0.7589784786105156, "step": 2601 }, { "epoch": 0.6001009009009008, "grad_norm": 23.84341812133789, "learning_rate": 7.837008094451882e-06, "loss": 12.22171401977539, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.26384190283715725, "loss_total_v6": 0.7638571634888649, "step": 2602 }, { "epoch": 0.6003315315315315, "grad_norm": 26.099227905273438, "learning_rate": 7.835349584376511e-06, "loss": 12.096435546875, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.2590789720416069, "loss_total_v6": 0.7560272067785263, "step": 2603 }, { "epoch": 0.6005621621621622, "grad_norm": 24.171363830566406, "learning_rate": 7.833690614344232e-06, "loss": 12.441733360290527, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.28467316180467606, "loss_total_v6": 0.7776083126664162, "step": 2604 }, { "epoch": 0.6007927927927927, "grad_norm": 24.80536460876465, "learning_rate": 7.832031184624165e-06, "loss": 13.145811080932617, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.3177374601364136, "loss_total_v6": 0.821613185107708, "step": 2605 }, { "epoch": 0.6010234234234234, "grad_norm": 22.0263729095459, "learning_rate": 7.830371295485507e-06, "loss": 12.161062240600586, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.260691961273551, "loss_total_v6": 0.7600663751363754, "step": 2606 }, { "epoch": 0.6012540540540541, "grad_norm": 21.313289642333984, "learning_rate": 7.82871094719753e-06, "loss": 12.933496475219727, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.30558163672685623, "loss_total_v6": 0.8083434924483299, "step": 2607 }, { "epoch": 0.6014846846846846, "grad_norm": 24.30315399169922, "learning_rate": 7.827050140029578e-06, "loss": 12.204105377197266, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.26625082083046436, "loss_total_v6": 0.7627565786242485, "step": 2608 }, { "epoch": 0.6017153153153153, "grad_norm": 19.685426712036133, "learning_rate": 7.825388874251072e-06, "loss": 12.801238059997559, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.297285046428442, "loss_total_v6": 0.8000773787498474, "step": 2609 }, { "epoch": 0.601945945945946, "grad_norm": 23.769472122192383, "learning_rate": 7.823727150131506e-06, "loss": 12.493666648864746, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.28315825015306473, "loss_total_v6": 0.7808541357517242, "step": 2610 }, { "epoch": 0.6021765765765765, "grad_norm": 21.511581420898438, "learning_rate": 7.822064967940449e-06, "loss": 12.515385627746582, "loss_alignment": 0.96484375, "loss_alignment_w": 0.482421875, "loss_sft": 0.2998965587466955, "loss_total_v6": 0.7822116166353226, "step": 2611 }, { "epoch": 0.6024072072072072, "grad_norm": 23.09720230102539, "learning_rate": 7.820402327947543e-06, "loss": 12.016504287719727, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.2543121464550495, "loss_total_v6": 0.7510315179824829, "step": 2612 }, { "epoch": 0.6026378378378379, "grad_norm": 21.60006332397461, "learning_rate": 7.818739230422505e-06, "loss": 12.167281150817871, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.26350683346390724, "loss_total_v6": 0.7604550719261169, "step": 2613 }, { "epoch": 0.6028684684684684, "grad_norm": 22.200946807861328, "learning_rate": 7.817075675635129e-06, "loss": 12.509271621704102, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.28378263488411903, "loss_total_v6": 0.7818294763565063, "step": 2614 }, { "epoch": 0.6030990990990991, "grad_norm": 24.254587173461914, "learning_rate": 7.815411663855279e-06, "loss": 12.889463424682617, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.31009285151958466, "loss_total_v6": 0.8055914863944054, "step": 2615 }, { "epoch": 0.6033297297297298, "grad_norm": 22.025814056396484, "learning_rate": 7.813747195352896e-06, "loss": 12.397433280944824, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.27509896270930767, "loss_total_v6": 0.7748395651578903, "step": 2616 }, { "epoch": 0.6035603603603603, "grad_norm": 21.922380447387695, "learning_rate": 7.812082270397989e-06, "loss": 12.704578399658203, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.3001702129840851, "loss_total_v6": 0.7940361872315407, "step": 2617 }, { "epoch": 0.603790990990991, "grad_norm": 22.9821720123291, "learning_rate": 7.810416889260653e-06, "loss": 12.576070785522461, "loss_alignment": 0.984375, "loss_alignment_w": 0.4921875, "loss_sft": 0.2954190857708454, "loss_total_v6": 0.7860044166445732, "step": 2618 }, { "epoch": 0.6040216216216217, "grad_norm": 19.682483673095703, "learning_rate": 7.808751052211048e-06, "loss": 12.762649536132812, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2907075975090265, "loss_total_v6": 0.7976655811071396, "step": 2619 }, { "epoch": 0.6042522522522522, "grad_norm": 23.464345932006836, "learning_rate": 7.807084759519405e-06, "loss": 12.171571731567383, "loss_alignment": 0.99267578125, "loss_alignment_w": 0.496337890625, "loss_sft": 0.26569752395153046, "loss_total_v6": 0.7607231959700584, "step": 2620 }, { "epoch": 0.6044828828828829, "grad_norm": 20.393356323242188, "learning_rate": 7.80541801145604e-06, "loss": 12.51667308807373, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.28046101331710815, "loss_total_v6": 0.7822920754551888, "step": 2621 }, { "epoch": 0.6047135135135135, "grad_norm": 23.629064559936523, "learning_rate": 7.803750808291332e-06, "loss": 12.368572235107422, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.2786815259605646, "loss_total_v6": 0.7730357572436333, "step": 2622 }, { "epoch": 0.6049441441441441, "grad_norm": 20.82347297668457, "learning_rate": 7.802083150295744e-06, "loss": 12.208602905273438, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.2636327687650919, "loss_total_v6": 0.7630377113819122, "step": 2623 }, { "epoch": 0.6051747747747748, "grad_norm": 23.022924423217773, "learning_rate": 7.800415037739802e-06, "loss": 12.71904182434082, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.29469596967101097, "loss_total_v6": 0.7949400916695595, "step": 2624 }, { "epoch": 0.6054054054054054, "grad_norm": 23.0616455078125, "learning_rate": 7.798746470894113e-06, "loss": 13.274675369262695, "loss_alignment": 1.01416015625, "loss_alignment_w": 0.507080078125, "loss_sft": 0.3230906669050455, "loss_total_v6": 0.8296672105789185, "step": 2625 }, { "epoch": 0.605636036036036, "grad_norm": 25.06573486328125, "learning_rate": 7.797077450029353e-06, "loss": 12.070901870727539, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.2534547857940197, "loss_total_v6": 0.7544313371181488, "step": 2626 }, { "epoch": 0.6058666666666667, "grad_norm": 21.735069274902344, "learning_rate": 7.79540797541628e-06, "loss": 12.190173149108887, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2572471909224987, "loss_total_v6": 0.761885866522789, "step": 2627 }, { "epoch": 0.6060972972972973, "grad_norm": 23.880645751953125, "learning_rate": 7.793738047325718e-06, "loss": 12.21253776550293, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.27005854435265064, "loss_total_v6": 0.7632836252450943, "step": 2628 }, { "epoch": 0.6063279279279279, "grad_norm": 21.768247604370117, "learning_rate": 7.792067666028565e-06, "loss": 12.421770095825195, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.27475846372544765, "loss_total_v6": 0.7763606533408165, "step": 2629 }, { "epoch": 0.6065585585585586, "grad_norm": 23.5372257232666, "learning_rate": 7.790396831795792e-06, "loss": 13.148030281066895, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.3237813077867031, "loss_total_v6": 0.8217518925666809, "step": 2630 }, { "epoch": 0.6067891891891892, "grad_norm": 21.5302734375, "learning_rate": 7.788725544898452e-06, "loss": 11.971246719360352, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.2580143120139837, "loss_total_v6": 0.7482029274106026, "step": 2631 }, { "epoch": 0.6070198198198198, "grad_norm": 21.78987693786621, "learning_rate": 7.78705380560766e-06, "loss": 12.721219062805176, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2909868471324444, "loss_total_v6": 0.7950762212276459, "step": 2632 }, { "epoch": 0.6072504504504505, "grad_norm": 20.88488006591797, "learning_rate": 7.78538161419461e-06, "loss": 12.335918426513672, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.2706896960735321, "loss_total_v6": 0.7709948867559433, "step": 2633 }, { "epoch": 0.6074810810810811, "grad_norm": 21.632354736328125, "learning_rate": 7.78370897093057e-06, "loss": 12.521490097045898, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.28613313660025597, "loss_total_v6": 0.7825931012630463, "step": 2634 }, { "epoch": 0.6077117117117117, "grad_norm": 23.03496742248535, "learning_rate": 7.78203587608688e-06, "loss": 13.155637741088867, "loss_alignment": 1.01708984375, "loss_alignment_w": 0.508544921875, "loss_sft": 0.3135451152920723, "loss_total_v6": 0.8222273662686348, "step": 2635 }, { "epoch": 0.6079423423423423, "grad_norm": 23.400310516357422, "learning_rate": 7.780362329934953e-06, "loss": 12.521126747131348, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.2896810211241245, "loss_total_v6": 0.782570406794548, "step": 2636 }, { "epoch": 0.608172972972973, "grad_norm": 20.171886444091797, "learning_rate": 7.778688332746274e-06, "loss": 12.61784553527832, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.29246059991419315, "loss_total_v6": 0.7886153683066368, "step": 2637 }, { "epoch": 0.6084036036036036, "grad_norm": 19.72056007385254, "learning_rate": 7.777013884792403e-06, "loss": 12.153650283813477, "loss_alignment": 0.9521484375, "loss_alignment_w": 0.47607421875, "loss_sft": 0.28378838300704956, "loss_total_v6": 0.7596031874418259, "step": 2638 }, { "epoch": 0.6086342342342342, "grad_norm": 22.30448341369629, "learning_rate": 7.775338986344975e-06, "loss": 13.164129257202148, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.3243907243013382, "loss_total_v6": 0.8227580487728119, "step": 2639 }, { "epoch": 0.6088648648648649, "grad_norm": 21.57400894165039, "learning_rate": 7.773663637675695e-06, "loss": 13.05159854888916, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.31056744046509266, "loss_total_v6": 0.8157248944044113, "step": 2640 }, { "epoch": 0.6090954954954955, "grad_norm": 23.27686309814453, "learning_rate": 7.77198783905634e-06, "loss": 12.267061233520508, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.26777468249201775, "loss_total_v6": 0.7666913345456123, "step": 2641 }, { "epoch": 0.6093261261261261, "grad_norm": 21.43515968322754, "learning_rate": 7.770311590758766e-06, "loss": 12.436142921447754, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.27995975874364376, "loss_total_v6": 0.7772589325904846, "step": 2642 }, { "epoch": 0.6095567567567568, "grad_norm": 23.810495376586914, "learning_rate": 7.768634893054894e-06, "loss": 12.509347915649414, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.28165112622082233, "loss_total_v6": 0.7818342223763466, "step": 2643 }, { "epoch": 0.6097873873873874, "grad_norm": 19.899614334106445, "learning_rate": 7.76695774621672e-06, "loss": 12.917905807495117, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.3091086093336344, "loss_total_v6": 0.807369090616703, "step": 2644 }, { "epoch": 0.610018018018018, "grad_norm": 24.690080642700195, "learning_rate": 7.765280150516322e-06, "loss": 12.477624893188477, "loss_alignment": 0.97900390625, "loss_alignment_w": 0.489501953125, "loss_sft": 0.29163133539259434, "loss_total_v6": 0.7798515409231186, "step": 2645 }, { "epoch": 0.6102486486486487, "grad_norm": 21.28955078125, "learning_rate": 7.763602106225836e-06, "loss": 12.45765495300293, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.2818535827100277, "loss_total_v6": 0.7786034345626831, "step": 2646 }, { "epoch": 0.6104792792792793, "grad_norm": 21.983631134033203, "learning_rate": 7.761923613617481e-06, "loss": 11.8883056640625, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.24933629669249058, "loss_total_v6": 0.7430191487073898, "step": 2647 }, { "epoch": 0.6107099099099099, "grad_norm": 25.966617584228516, "learning_rate": 7.760244672963548e-06, "loss": 12.12187385559082, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.258181681856513, "loss_total_v6": 0.7576171010732651, "step": 2648 }, { "epoch": 0.6109405405405406, "grad_norm": 23.097225189208984, "learning_rate": 7.758565284536395e-06, "loss": 12.331287384033203, "loss_alignment": 0.982421875, "loss_alignment_w": 0.4912109375, "loss_sft": 0.2796166092157364, "loss_total_v6": 0.7707054391503334, "step": 2649 }, { "epoch": 0.6111711711711711, "grad_norm": 22.637115478515625, "learning_rate": 7.75688544860846e-06, "loss": 12.89026165008545, "loss_alignment": 0.98486328125, "loss_alignment_w": 0.492431640625, "loss_sft": 0.31327079981565475, "loss_total_v6": 0.8056413605809212, "step": 2650 }, { "epoch": 0.6114018018018018, "grad_norm": 21.95665168762207, "learning_rate": 7.755205165452245e-06, "loss": 11.888124465942383, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.2497673910111189, "loss_total_v6": 0.7430077418684959, "step": 2651 }, { "epoch": 0.6116324324324325, "grad_norm": 21.80414581298828, "learning_rate": 7.753524435340334e-06, "loss": 12.113081932067871, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.2568235211074352, "loss_total_v6": 0.7570676505565643, "step": 2652 }, { "epoch": 0.611863063063063, "grad_norm": 22.48381233215332, "learning_rate": 7.751843258545377e-06, "loss": 12.105694770812988, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.26747019216418266, "loss_total_v6": 0.7566059455275536, "step": 2653 }, { "epoch": 0.6120936936936937, "grad_norm": 20.618438720703125, "learning_rate": 7.750161635340097e-06, "loss": 12.454693794250488, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.2784183267503977, "loss_total_v6": 0.7784183323383331, "step": 2654 }, { "epoch": 0.6123243243243244, "grad_norm": 22.91752052307129, "learning_rate": 7.748479565997293e-06, "loss": 12.531888961791992, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.28632537461817265, "loss_total_v6": 0.7832430973649025, "step": 2655 }, { "epoch": 0.6125549549549549, "grad_norm": 25.749828338623047, "learning_rate": 7.746797050789834e-06, "loss": 12.801815032958984, "loss_alignment": 0.96435546875, "loss_alignment_w": 0.482177734375, "loss_sft": 0.3185917753726244, "loss_total_v6": 0.8001133948564529, "step": 2656 }, { "epoch": 0.6127855855855856, "grad_norm": 23.449359893798828, "learning_rate": 7.74511408999066e-06, "loss": 12.724104881286621, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.28388874791562557, "loss_total_v6": 0.7952565401792526, "step": 2657 }, { "epoch": 0.6130162162162163, "grad_norm": 21.73177146911621, "learning_rate": 7.743430683872785e-06, "loss": 12.299139022827148, "loss_alignment": 0.96044921875, "loss_alignment_w": 0.480224609375, "loss_sft": 0.2878612242639065, "loss_total_v6": 0.7686961963772774, "step": 2658 }, { "epoch": 0.6132468468468468, "grad_norm": 20.999755859375, "learning_rate": 7.741746832709296e-06, "loss": 12.398693084716797, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.28207469917833805, "loss_total_v6": 0.7749183475971222, "step": 2659 }, { "epoch": 0.6134774774774775, "grad_norm": 20.931978225708008, "learning_rate": 7.740062536773352e-06, "loss": 12.804510116577148, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.29393421672284603, "loss_total_v6": 0.80028185993433, "step": 2660 }, { "epoch": 0.6137081081081082, "grad_norm": 22.18168830871582, "learning_rate": 7.738377796338179e-06, "loss": 12.080449104309082, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.2620471268892288, "loss_total_v6": 0.7550280764698982, "step": 2661 }, { "epoch": 0.6139387387387387, "grad_norm": 23.023530960083008, "learning_rate": 7.736692611677084e-06, "loss": 12.432504653930664, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.2823415379971266, "loss_total_v6": 0.7770315110683441, "step": 2662 }, { "epoch": 0.6141693693693694, "grad_norm": 19.048791885375977, "learning_rate": 7.735006983063443e-06, "loss": 12.487220764160156, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2734627826139331, "loss_total_v6": 0.7804512903094292, "step": 2663 }, { "epoch": 0.6144, "grad_norm": 20.15619468688965, "learning_rate": 7.733320910770695e-06, "loss": 12.565573692321777, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.28087748773396015, "loss_total_v6": 0.7853483259677887, "step": 2664 }, { "epoch": 0.6146306306306306, "grad_norm": 22.316282272338867, "learning_rate": 7.731634395072366e-06, "loss": 11.972978591918945, "loss_alignment": 0.9873046875, "loss_alignment_w": 0.49365234375, "loss_sft": 0.25356012396514416, "loss_total_v6": 0.7483111321926117, "step": 2665 }, { "epoch": 0.6148612612612613, "grad_norm": 21.12688636779785, "learning_rate": 7.729947436242041e-06, "loss": 12.418354034423828, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2710659969598055, "loss_total_v6": 0.7761471718549728, "step": 2666 }, { "epoch": 0.6150918918918918, "grad_norm": 23.313499450683594, "learning_rate": 7.728260034553387e-06, "loss": 13.060646057128906, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.3116516601294279, "loss_total_v6": 0.8162903487682343, "step": 2667 }, { "epoch": 0.6153225225225225, "grad_norm": 21.51626205444336, "learning_rate": 7.726572190280134e-06, "loss": 12.395503044128418, "loss_alignment": 0.99267578125, "loss_alignment_w": 0.496337890625, "loss_sft": 0.27824369817972183, "loss_total_v6": 0.7747189030051231, "step": 2668 }, { "epoch": 0.6155531531531532, "grad_norm": 21.797924041748047, "learning_rate": 7.724883903696092e-06, "loss": 12.36909008026123, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.26333304680883884, "loss_total_v6": 0.7730681374669075, "step": 2669 }, { "epoch": 0.6157837837837837, "grad_norm": 21.936016082763672, "learning_rate": 7.723195175075136e-06, "loss": 12.458980560302734, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.27986119501292706, "loss_total_v6": 0.7786863073706627, "step": 2670 }, { "epoch": 0.6160144144144144, "grad_norm": 21.025545120239258, "learning_rate": 7.721506004691216e-06, "loss": 12.355768203735352, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.27098423615098, "loss_total_v6": 0.7722354978322983, "step": 2671 }, { "epoch": 0.6162450450450451, "grad_norm": 23.510841369628906, "learning_rate": 7.719816392818354e-06, "loss": 12.256525993347168, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.27090043388307095, "loss_total_v6": 0.766032874584198, "step": 2672 }, { "epoch": 0.6164756756756756, "grad_norm": 21.061017990112305, "learning_rate": 7.718126339730642e-06, "loss": 12.772985458374023, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2916434910148382, "loss_total_v6": 0.7983115538954735, "step": 2673 }, { "epoch": 0.6167063063063063, "grad_norm": 19.26534652709961, "learning_rate": 7.716435845702242e-06, "loss": 12.368575096130371, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.27755249105393887, "loss_total_v6": 0.7730359137058258, "step": 2674 }, { "epoch": 0.616936936936937, "grad_norm": 23.979246139526367, "learning_rate": 7.714744911007395e-06, "loss": 12.341949462890625, "loss_alignment": 0.9833984375, "loss_alignment_w": 0.49169921875, "loss_sft": 0.282281881198287, "loss_total_v6": 0.7713718488812447, "step": 2675 }, { "epoch": 0.6171675675675675, "grad_norm": 22.21875762939453, "learning_rate": 7.713053535920403e-06, "loss": 12.186853408813477, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.2648521903902292, "loss_total_v6": 0.7616783604025841, "step": 2676 }, { "epoch": 0.6173981981981982, "grad_norm": 24.33641242980957, "learning_rate": 7.71136172071565e-06, "loss": 12.69084358215332, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2892103809863329, "loss_total_v6": 0.7931776866316795, "step": 2677 }, { "epoch": 0.6176288288288289, "grad_norm": 22.20269203186035, "learning_rate": 7.709669465667578e-06, "loss": 12.583654403686523, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.2895454354584217, "loss_total_v6": 0.7864784225821495, "step": 2678 }, { "epoch": 0.6178594594594594, "grad_norm": 22.332448959350586, "learning_rate": 7.707976771050718e-06, "loss": 12.387272834777832, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.2741587981581688, "loss_total_v6": 0.7742045745253563, "step": 2679 }, { "epoch": 0.6180900900900901, "grad_norm": 22.829893112182617, "learning_rate": 7.706283637139658e-06, "loss": 12.248886108398438, "loss_alignment": 0.98193359375, "loss_alignment_w": 0.490966796875, "loss_sft": 0.27516839653253555, "loss_total_v6": 0.7655553743243217, "step": 2680 }, { "epoch": 0.6183207207207208, "grad_norm": 21.30900001525879, "learning_rate": 7.704590064209062e-06, "loss": 12.57675552368164, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.2814391050487757, "loss_total_v6": 0.7860472723841667, "step": 2681 }, { "epoch": 0.6185513513513513, "grad_norm": 22.930166244506836, "learning_rate": 7.702896052533666e-06, "loss": 12.562314987182617, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.29556641541421413, "loss_total_v6": 0.7851446568965912, "step": 2682 }, { "epoch": 0.618781981981982, "grad_norm": 21.131977081298828, "learning_rate": 7.701201602388276e-06, "loss": 13.094791412353516, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.3100778590887785, "loss_total_v6": 0.8184244483709335, "step": 2683 }, { "epoch": 0.6190126126126126, "grad_norm": 25.035062789916992, "learning_rate": 7.69950671404777e-06, "loss": 12.552385330200195, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.2896815165877342, "loss_total_v6": 0.7845240607857704, "step": 2684 }, { "epoch": 0.6192432432432432, "grad_norm": 24.198759078979492, "learning_rate": 7.697811387787097e-06, "loss": 12.411401748657227, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.27189792692661285, "loss_total_v6": 0.7757126241922379, "step": 2685 }, { "epoch": 0.6194738738738739, "grad_norm": 23.19799041748047, "learning_rate": 7.696115623881275e-06, "loss": 12.490694046020508, "loss_alignment": 0.984375, "loss_alignment_w": 0.4921875, "loss_sft": 0.28861824609339237, "loss_total_v6": 0.7806684076786041, "step": 2686 }, { "epoch": 0.6197045045045045, "grad_norm": 22.940956115722656, "learning_rate": 7.694419422605399e-06, "loss": 12.456659317016602, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.28838310204446316, "loss_total_v6": 0.7785411775112152, "step": 2687 }, { "epoch": 0.6199351351351351, "grad_norm": 22.955734252929688, "learning_rate": 7.692722784234624e-06, "loss": 12.305150985717773, "loss_alignment": 0.9873046875, "loss_alignment_w": 0.49365234375, "loss_sft": 0.27548058331012726, "loss_total_v6": 0.7690718993544579, "step": 2688 }, { "epoch": 0.6201657657657658, "grad_norm": 25.353837966918945, "learning_rate": 7.691025709044189e-06, "loss": 12.326828956604004, "loss_alignment": 0.974609375, "loss_alignment_w": 0.4873046875, "loss_sft": 0.28371721506118774, "loss_total_v6": 0.7704268097877502, "step": 2689 }, { "epoch": 0.6203963963963964, "grad_norm": 22.524335861206055, "learning_rate": 7.689328197309394e-06, "loss": 12.764799118041992, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.3041323758661747, "loss_total_v6": 0.7977999523282051, "step": 2690 }, { "epoch": 0.620627027027027, "grad_norm": 22.666276931762695, "learning_rate": 7.687630249305613e-06, "loss": 11.994307518005371, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.25457279570400715, "loss_total_v6": 0.7496442124247551, "step": 2691 }, { "epoch": 0.6208576576576577, "grad_norm": 22.575082778930664, "learning_rate": 7.685931865308293e-06, "loss": 12.777843475341797, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.2918708510696888, "loss_total_v6": 0.7986152172088623, "step": 2692 }, { "epoch": 0.6210882882882883, "grad_norm": 22.18335723876953, "learning_rate": 7.684233045592946e-06, "loss": 11.964640617370605, "loss_alignment": 0.9716796875, "loss_alignment_w": 0.48583984375, "loss_sft": 0.26169075816869736, "loss_total_v6": 0.7477900460362434, "step": 2693 }, { "epoch": 0.6213189189189189, "grad_norm": 23.323482513427734, "learning_rate": 7.682533790435165e-06, "loss": 12.433516502380371, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.27024361677467823, "loss_total_v6": 0.7770948112010956, "step": 2694 }, { "epoch": 0.6215495495495496, "grad_norm": 23.189205169677734, "learning_rate": 7.680834100110601e-06, "loss": 12.861039161682129, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.30199920013546944, "loss_total_v6": 0.8038149550557137, "step": 2695 }, { "epoch": 0.6217801801801802, "grad_norm": 23.75860595703125, "learning_rate": 7.679133974894984e-06, "loss": 12.352355003356934, "loss_alignment": 0.97900390625, "loss_alignment_w": 0.489501953125, "loss_sft": 0.2829475421458483, "loss_total_v6": 0.7720222324132919, "step": 2696 }, { "epoch": 0.6220108108108108, "grad_norm": 23.32773208618164, "learning_rate": 7.677433415064112e-06, "loss": 12.495950698852539, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.27498489804565907, "loss_total_v6": 0.7809968888759613, "step": 2697 }, { "epoch": 0.6222414414414414, "grad_norm": 21.941232681274414, "learning_rate": 7.675732420893854e-06, "loss": 12.903888702392578, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.3067066762596369, "loss_total_v6": 0.8064929991960526, "step": 2698 }, { "epoch": 0.6224720720720721, "grad_norm": 23.03485870361328, "learning_rate": 7.674030992660149e-06, "loss": 12.444833755493164, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.28035033494234085, "loss_total_v6": 0.777802087366581, "step": 2699 }, { "epoch": 0.6227027027027027, "grad_norm": 21.396753311157227, "learning_rate": 7.672329130639007e-06, "loss": 12.658321380615234, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.28208135813474655, "loss_total_v6": 0.7911450639367104, "step": 2700 }, { "epoch": 0.6229333333333333, "grad_norm": 22.60293960571289, "learning_rate": 7.670626835106507e-06, "loss": 12.353549003601074, "loss_alignment": 0.9912109375, "loss_alignment_w": 0.49560546875, "loss_sft": 0.275896267965436, "loss_total_v6": 0.7720968052744865, "step": 2701 }, { "epoch": 0.623163963963964, "grad_norm": 22.1002197265625, "learning_rate": 7.6689241063388e-06, "loss": 12.665128707885742, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.2876337505877018, "loss_total_v6": 0.7915705144405365, "step": 2702 }, { "epoch": 0.6233945945945946, "grad_norm": 22.194719314575195, "learning_rate": 7.667220944612103e-06, "loss": 12.836091995239258, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.30069939233362675, "loss_total_v6": 0.8022557571530342, "step": 2703 }, { "epoch": 0.6236252252252252, "grad_norm": 22.128231048583984, "learning_rate": 7.665517350202715e-06, "loss": 12.292428970336914, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.2777678072452545, "loss_total_v6": 0.7682768255472183, "step": 2704 }, { "epoch": 0.6238558558558559, "grad_norm": 21.494691848754883, "learning_rate": 7.663813323386988e-06, "loss": 12.004510879516602, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.250907551497221, "loss_total_v6": 0.7502819150686264, "step": 2705 }, { "epoch": 0.6240864864864865, "grad_norm": 21.199371337890625, "learning_rate": 7.66210886444136e-06, "loss": 12.530781745910645, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.2766583636403084, "loss_total_v6": 0.7831738814711571, "step": 2706 }, { "epoch": 0.6243171171171171, "grad_norm": 20.238866806030273, "learning_rate": 7.66040397364233e-06, "loss": 12.33665657043457, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.27641215175390244, "loss_total_v6": 0.7710410431027412, "step": 2707 }, { "epoch": 0.6245477477477478, "grad_norm": 22.251089096069336, "learning_rate": 7.658698651266467e-06, "loss": 12.738779067993164, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.2959142383188009, "loss_total_v6": 0.796173669397831, "step": 2708 }, { "epoch": 0.6247783783783784, "grad_norm": 23.362842559814453, "learning_rate": 7.656992897590416e-06, "loss": 12.350704193115234, "loss_alignment": 0.97802734375, "loss_alignment_w": 0.489013671875, "loss_sft": 0.282813822850585, "loss_total_v6": 0.7719190344214439, "step": 2709 }, { "epoch": 0.625009009009009, "grad_norm": 21.065061569213867, "learning_rate": 7.655286712890883e-06, "loss": 12.300978660583496, "loss_alignment": 0.98583984375, "loss_alignment_w": 0.492919921875, "loss_sft": 0.275875985622406, "loss_total_v6": 0.7688111588358879, "step": 2710 }, { "epoch": 0.6252396396396397, "grad_norm": 19.56519317626953, "learning_rate": 7.653580097444655e-06, "loss": 11.756284713745117, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.24145111627876759, "loss_total_v6": 0.7347677946090698, "step": 2711 }, { "epoch": 0.6254702702702702, "grad_norm": 23.61752700805664, "learning_rate": 7.651873051528582e-06, "loss": 12.54002571105957, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.28359899297356606, "loss_total_v6": 0.7837515994906425, "step": 2712 }, { "epoch": 0.6257009009009009, "grad_norm": 20.20368194580078, "learning_rate": 7.650165575419582e-06, "loss": 12.502191543579102, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.28405725210905075, "loss_total_v6": 0.7813869565725327, "step": 2713 }, { "epoch": 0.6259315315315316, "grad_norm": 22.25172996520996, "learning_rate": 7.648457669394645e-06, "loss": 12.078475952148438, "loss_alignment": 0.978515625, "loss_alignment_w": 0.4892578125, "loss_sft": 0.26665402203798294, "loss_total_v6": 0.7549047395586967, "step": 2714 }, { "epoch": 0.6261621621621621, "grad_norm": 22.025243759155273, "learning_rate": 7.646749333730837e-06, "loss": 12.229944229125977, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.2715431787073612, "loss_total_v6": 0.7643715366721153, "step": 2715 }, { "epoch": 0.6263927927927928, "grad_norm": 25.318552017211914, "learning_rate": 7.645040568705281e-06, "loss": 12.48817253112793, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.28246390633285046, "loss_total_v6": 0.7805107906460762, "step": 2716 }, { "epoch": 0.6266234234234235, "grad_norm": 27.515535354614258, "learning_rate": 7.643331374595184e-06, "loss": 12.047871589660645, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.25439577922225, "loss_total_v6": 0.7529919818043709, "step": 2717 }, { "epoch": 0.626854054054054, "grad_norm": 23.18694496154785, "learning_rate": 7.64162175167781e-06, "loss": 11.941282272338867, "loss_alignment": 0.95361328125, "loss_alignment_w": 0.476806640625, "loss_sft": 0.2682875916361809, "loss_total_v6": 0.7463301569223404, "step": 2718 }, { "epoch": 0.6270846846846847, "grad_norm": 22.009998321533203, "learning_rate": 7.639911700230502e-06, "loss": 12.565315246582031, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.28926895558834076, "loss_total_v6": 0.7853321731090546, "step": 2719 }, { "epoch": 0.6273153153153154, "grad_norm": 24.123754501342773, "learning_rate": 7.638201220530664e-06, "loss": 12.910701751708984, "loss_alignment": 0.9814453125, "loss_alignment_w": 0.49072265625, "loss_sft": 0.3148381616920233, "loss_total_v6": 0.8069188296794891, "step": 2720 }, { "epoch": 0.6275459459459459, "grad_norm": 22.46843910217285, "learning_rate": 7.636490312855778e-06, "loss": 13.294878005981445, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.3293124809861183, "loss_total_v6": 0.8309299275279045, "step": 2721 }, { "epoch": 0.6277765765765766, "grad_norm": 20.559904098510742, "learning_rate": 7.634778977483389e-06, "loss": 12.066758155822754, "loss_alignment": 0.98583984375, "loss_alignment_w": 0.492919921875, "loss_sft": 0.26103879138827324, "loss_total_v6": 0.7541723474860191, "step": 2722 }, { "epoch": 0.6280072072072072, "grad_norm": 19.16739273071289, "learning_rate": 7.633067214691115e-06, "loss": 11.74066162109375, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.24357227422297, "loss_total_v6": 0.7337913811206818, "step": 2723 }, { "epoch": 0.6282378378378378, "grad_norm": 22.182741165161133, "learning_rate": 7.63135502475664e-06, "loss": 11.851899147033691, "loss_alignment": 0.97607421875, "loss_alignment_w": 0.488037109375, "loss_sft": 0.25173002295196056, "loss_total_v6": 0.7407436966896057, "step": 2724 }, { "epoch": 0.6284684684684685, "grad_norm": 21.831989288330078, "learning_rate": 7.62964240795772e-06, "loss": 12.579418182373047, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.29181368462741375, "loss_total_v6": 0.7862136960029602, "step": 2725 }, { "epoch": 0.628699099099099, "grad_norm": 24.299169540405273, "learning_rate": 7.627929364572184e-06, "loss": 12.159261703491211, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.2661032360047102, "loss_total_v6": 0.7599539011716843, "step": 2726 }, { "epoch": 0.6289297297297297, "grad_norm": 22.441097259521484, "learning_rate": 7.6262158948779185e-06, "loss": 12.466477394104004, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.2827101908624172, "loss_total_v6": 0.7791548818349838, "step": 2727 }, { "epoch": 0.6291603603603604, "grad_norm": 23.70136260986328, "learning_rate": 7.6245019991528934e-06, "loss": 12.857290267944336, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.30647980980575085, "loss_total_v6": 0.8035806342959404, "step": 2728 }, { "epoch": 0.6293909909909909, "grad_norm": 20.667203903198242, "learning_rate": 7.622787677675136e-06, "loss": 12.259414672851562, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.2708978820592165, "loss_total_v6": 0.7662134543061256, "step": 2729 }, { "epoch": 0.6296216216216216, "grad_norm": 21.125080108642578, "learning_rate": 7.621072930722749e-06, "loss": 12.298713684082031, "loss_alignment": 0.9755859375, "loss_alignment_w": 0.48779296875, "loss_sft": 0.27916764467954636, "loss_total_v6": 0.7686695903539658, "step": 2730 }, { "epoch": 0.6298522522522523, "grad_norm": 21.39300537109375, "learning_rate": 7.619357758573904e-06, "loss": 12.62549877166748, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.2860571909695864, "loss_total_v6": 0.7890936881303787, "step": 2731 }, { "epoch": 0.6300828828828828, "grad_norm": 22.598512649536133, "learning_rate": 7.617642161506837e-06, "loss": 12.171720504760742, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.25681097619235516, "loss_total_v6": 0.7607324942946434, "step": 2732 }, { "epoch": 0.6303135135135135, "grad_norm": 22.691286087036133, "learning_rate": 7.615926139799859e-06, "loss": 13.142982482910156, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.3219246566295624, "loss_total_v6": 0.8214363530278206, "step": 2733 }, { "epoch": 0.6305441441441442, "grad_norm": 21.123756408691406, "learning_rate": 7.614209693731347e-06, "loss": 12.608814239501953, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.2920791916549206, "loss_total_v6": 0.7880508750677109, "step": 2734 }, { "epoch": 0.6307747747747747, "grad_norm": 25.72956085205078, "learning_rate": 7.612492823579744e-06, "loss": 12.964807510375977, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.31075819209218025, "loss_total_v6": 0.8103004693984985, "step": 2735 }, { "epoch": 0.6310054054054054, "grad_norm": 21.068262100219727, "learning_rate": 7.610775529623568e-06, "loss": 12.1485595703125, "loss_alignment": 0.97314453125, "loss_alignment_w": 0.486572265625, "loss_sft": 0.2733077257871628, "loss_total_v6": 0.7592849284410477, "step": 2736 }, { "epoch": 0.6312360360360361, "grad_norm": 23.9156551361084, "learning_rate": 7.609057812141399e-06, "loss": 12.717696189880371, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.29662601090967655, "loss_total_v6": 0.7948560044169426, "step": 2737 }, { "epoch": 0.6314666666666666, "grad_norm": 24.049333572387695, "learning_rate": 7.607339671411893e-06, "loss": 12.21164321899414, "loss_alignment": 0.98291015625, "loss_alignment_w": 0.491455078125, "loss_sft": 0.271208044141531, "loss_total_v6": 0.7632276937365532, "step": 2738 }, { "epoch": 0.6316972972972973, "grad_norm": 22.853853225708008, "learning_rate": 7.605621107713768e-06, "loss": 12.581380844116211, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.28506983630359173, "loss_total_v6": 0.7863363027572632, "step": 2739 }, { "epoch": 0.631927927927928, "grad_norm": 22.379358291625977, "learning_rate": 7.603902121325813e-06, "loss": 12.19502067565918, "loss_alignment": 0.982421875, "loss_alignment_w": 0.4912109375, "loss_sft": 0.27227485179901123, "loss_total_v6": 0.7621887922286987, "step": 2740 }, { "epoch": 0.6321585585585585, "grad_norm": 22.73175811767578, "learning_rate": 7.6021827125268865e-06, "loss": 12.243335723876953, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.26847382821142673, "loss_total_v6": 0.765208438038826, "step": 2741 }, { "epoch": 0.6323891891891892, "grad_norm": 21.89320182800293, "learning_rate": 7.6004628815959155e-06, "loss": 12.358848571777344, "loss_alignment": 0.98095703125, "loss_alignment_w": 0.490478515625, "loss_sft": 0.28281930834054947, "loss_total_v6": 0.7724280878901482, "step": 2742 }, { "epoch": 0.6326198198198199, "grad_norm": 21.669052124023438, "learning_rate": 7.598742628811894e-06, "loss": 12.27273178100586, "loss_alignment": 0.9755859375, "loss_alignment_w": 0.48779296875, "loss_sft": 0.2780930809676647, "loss_total_v6": 0.7670457065105438, "step": 2743 }, { "epoch": 0.6328504504504504, "grad_norm": 23.184795379638672, "learning_rate": 7.597021954453887e-06, "loss": 12.328573226928711, "loss_alignment": 0.98095703125, "loss_alignment_w": 0.490478515625, "loss_sft": 0.28095761872828007, "loss_total_v6": 0.770535871386528, "step": 2744 }, { "epoch": 0.6330810810810811, "grad_norm": 24.493886947631836, "learning_rate": 7.595300858801025e-06, "loss": 12.507104873657227, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.2907273266464472, "loss_total_v6": 0.781694084405899, "step": 2745 }, { "epoch": 0.6333117117117117, "grad_norm": 20.828580856323242, "learning_rate": 7.5935793421325076e-06, "loss": 12.236410140991211, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.2665914185345173, "loss_total_v6": 0.7647756487131119, "step": 2746 }, { "epoch": 0.6335423423423423, "grad_norm": 23.159488677978516, "learning_rate": 7.591857404727605e-06, "loss": 12.25894546508789, "loss_alignment": 0.97802734375, "loss_alignment_w": 0.489013671875, "loss_sft": 0.27825378999114037, "loss_total_v6": 0.7661840468645096, "step": 2747 }, { "epoch": 0.633772972972973, "grad_norm": 22.15401840209961, "learning_rate": 7.590135046865652e-06, "loss": 12.406994819641113, "loss_alignment": 0.98046875, "loss_alignment_w": 0.490234375, "loss_sft": 0.2844398468732834, "loss_total_v6": 0.775437168776989, "step": 2748 }, { "epoch": 0.6340036036036036, "grad_norm": 21.871814727783203, "learning_rate": 7.588412268826053e-06, "loss": 12.482223510742188, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2760343551635742, "loss_total_v6": 0.7801389470696449, "step": 2749 }, { "epoch": 0.6342342342342342, "grad_norm": 23.122760772705078, "learning_rate": 7.586689070888284e-06, "loss": 12.353435516357422, "loss_alignment": 0.98046875, "loss_alignment_w": 0.490234375, "loss_sft": 0.2839152179658413, "loss_total_v6": 0.7720896750688553, "step": 2750 }, { "epoch": 0.6344648648648649, "grad_norm": 20.462871551513672, "learning_rate": 7.5849654533318855e-06, "loss": 12.4991455078125, "loss_alignment": 0.98486328125, "loss_alignment_w": 0.492431640625, "loss_sft": 0.28905483707785606, "loss_total_v6": 0.7811965644359589, "step": 2751 }, { "epoch": 0.6346954954954955, "grad_norm": 20.51488494873047, "learning_rate": 7.583241416436462e-06, "loss": 12.281654357910156, "loss_alignment": 0.98486328125, "loss_alignment_w": 0.492431640625, "loss_sft": 0.274424085393548, "loss_total_v6": 0.7676034048199654, "step": 2752 }, { "epoch": 0.6349261261261261, "grad_norm": 21.91701889038086, "learning_rate": 7.5815169604816965e-06, "loss": 12.750134468078613, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.2982872258871794, "loss_total_v6": 0.7968834191560745, "step": 2753 }, { "epoch": 0.6351567567567568, "grad_norm": 23.458227157592773, "learning_rate": 7.57979208574733e-06, "loss": 12.607378005981445, "loss_alignment": 0.9892578125, "loss_alignment_w": 0.49462890625, "loss_sft": 0.29392733611166477, "loss_total_v6": 0.7879611700773239, "step": 2754 }, { "epoch": 0.6353873873873874, "grad_norm": 20.637475967407227, "learning_rate": 7.578066792513179e-06, "loss": 12.636548042297363, "loss_alignment": 0.98388671875, "loss_alignment_w": 0.491943359375, "loss_sft": 0.2974594607949257, "loss_total_v6": 0.7897842526435852, "step": 2755 }, { "epoch": 0.635618018018018, "grad_norm": 20.01071548461914, "learning_rate": 7.576341081059125e-06, "loss": 12.484930038452148, "loss_alignment": 0.97265625, "loss_alignment_w": 0.486328125, "loss_sft": 0.29439200833439827, "loss_total_v6": 0.7803081125020981, "step": 2756 }, { "epoch": 0.6358486486486487, "grad_norm": 21.88738441467285, "learning_rate": 7.5746149516651114e-06, "loss": 12.575037002563477, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.29013608396053314, "loss_total_v6": 0.7859398499131203, "step": 2757 }, { "epoch": 0.6360792792792793, "grad_norm": 21.243513107299805, "learning_rate": 7.57288840461116e-06, "loss": 12.418270111083984, "loss_alignment": 0.96875, "loss_alignment_w": 0.484375, "loss_sft": 0.2907903268933296, "loss_total_v6": 0.776141881942749, "step": 2758 }, { "epoch": 0.6363099099099099, "grad_norm": 21.84957504272461, "learning_rate": 7.571161440177353e-06, "loss": 12.752664566040039, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.2952562179416418, "loss_total_v6": 0.7970414981245995, "step": 2759 }, { "epoch": 0.6365405405405405, "grad_norm": 24.214061737060547, "learning_rate": 7.5694340586438446e-06, "loss": 12.885010719299316, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.3004761040210724, "loss_total_v6": 0.8053131327033043, "step": 2760 }, { "epoch": 0.6367711711711712, "grad_norm": 22.787269592285156, "learning_rate": 7.5677062602908515e-06, "loss": 13.018786430358887, "loss_alignment": 0.9912109375, "loss_alignment_w": 0.49560546875, "loss_sft": 0.31747356429696083, "loss_total_v6": 0.813674122095108, "step": 2761 }, { "epoch": 0.6370018018018018, "grad_norm": 21.835372924804688, "learning_rate": 7.565978045398663e-06, "loss": 12.291513442993164, "loss_alignment": 0.978515625, "loss_alignment_w": 0.4892578125, "loss_sft": 0.2789312396198511, "loss_total_v6": 0.7682195752859116, "step": 2762 }, { "epoch": 0.6372324324324324, "grad_norm": 21.980796813964844, "learning_rate": 7.5642494142476346e-06, "loss": 12.618306159973145, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.29244363866746426, "loss_total_v6": 0.7886441722512245, "step": 2763 }, { "epoch": 0.6374630630630631, "grad_norm": 23.14885139465332, "learning_rate": 7.562520367118187e-06, "loss": 12.72951889038086, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.29333662055432796, "loss_total_v6": 0.7955949008464813, "step": 2764 }, { "epoch": 0.6376936936936937, "grad_norm": 22.72296905517578, "learning_rate": 7.5607909042908115e-06, "loss": 12.661996841430664, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.289497934281826, "loss_total_v6": 0.7913747727870941, "step": 2765 }, { "epoch": 0.6379243243243243, "grad_norm": 24.785308837890625, "learning_rate": 7.559061026046066e-06, "loss": 12.43899917602539, "loss_alignment": 0.97900390625, "loss_alignment_w": 0.489501953125, "loss_sft": 0.28897315450012684, "loss_total_v6": 0.7774375006556511, "step": 2766 }, { "epoch": 0.638154954954955, "grad_norm": 20.84404945373535, "learning_rate": 7.557330732664574e-06, "loss": 12.403424263000488, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.2788303606212139, "loss_total_v6": 0.7752140015363693, "step": 2767 }, { "epoch": 0.6383855855855856, "grad_norm": 19.399112701416016, "learning_rate": 7.555600024427028e-06, "loss": 12.514391899108887, "loss_alignment": 0.97607421875, "loss_alignment_w": 0.488037109375, "loss_sft": 0.2944023236632347, "loss_total_v6": 0.7821494936943054, "step": 2768 }, { "epoch": 0.6386162162162162, "grad_norm": 22.60846519470215, "learning_rate": 7.553868901614187e-06, "loss": 12.861692428588867, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.3020247295498848, "loss_total_v6": 0.8038557842373848, "step": 2769 }, { "epoch": 0.6388468468468469, "grad_norm": 21.91567039489746, "learning_rate": 7.5521373645068794e-06, "loss": 11.979373931884766, "loss_alignment": 0.97216796875, "loss_alignment_w": 0.486083984375, "loss_sft": 0.26274897158145905, "loss_total_v6": 0.7487108632922173, "step": 2770 }, { "epoch": 0.6390774774774775, "grad_norm": 21.962188720703125, "learning_rate": 7.550405413385999e-06, "loss": 12.430480003356934, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.27936168015003204, "loss_total_v6": 0.7769050151109695, "step": 2771 }, { "epoch": 0.6393081081081081, "grad_norm": 20.944765090942383, "learning_rate": 7.548673048532504e-06, "loss": 12.467984199523926, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.2800424247980118, "loss_total_v6": 0.7792489975690842, "step": 2772 }, { "epoch": 0.6395387387387388, "grad_norm": 22.05233383178711, "learning_rate": 7.546940270227425e-06, "loss": 12.636573791503906, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.290533559396863, "loss_total_v6": 0.7897858768701553, "step": 2773 }, { "epoch": 0.6397693693693693, "grad_norm": 21.71693992614746, "learning_rate": 7.545207078751858e-06, "loss": 12.62995433807373, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.29243917763233185, "loss_total_v6": 0.7893721461296082, "step": 2774 }, { "epoch": 0.64, "grad_norm": 19.181833267211914, "learning_rate": 7.543473474386962e-06, "loss": 12.365474700927734, "loss_alignment": 0.9853515625, "loss_alignment_w": 0.49267578125, "loss_sft": 0.2787016090005636, "loss_total_v6": 0.7728421911597252, "step": 2775 }, { "epoch": 0.6402306306306307, "grad_norm": 24.629348754882812, "learning_rate": 7.54173945741397e-06, "loss": 12.790310859680176, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.29673943668603897, "loss_total_v6": 0.7993944436311722, "step": 2776 }, { "epoch": 0.6404612612612612, "grad_norm": 22.555721282958984, "learning_rate": 7.5400050281141765e-06, "loss": 12.612573623657227, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.2934891451150179, "loss_total_v6": 0.7882858961820602, "step": 2777 }, { "epoch": 0.6406918918918919, "grad_norm": 22.792795181274414, "learning_rate": 7.538270186768945e-06, "loss": 12.384956359863281, "loss_alignment": 0.97705078125, "loss_alignment_w": 0.488525390625, "loss_sft": 0.2855496369302273, "loss_total_v6": 0.7740597873926163, "step": 2778 }, { "epoch": 0.6409225225225226, "grad_norm": 19.35626792907715, "learning_rate": 7.536534933659705e-06, "loss": 12.481054306030273, "loss_alignment": 0.97900390625, "loss_alignment_w": 0.489501953125, "loss_sft": 0.2904113568365574, "loss_total_v6": 0.7800658643245697, "step": 2779 }, { "epoch": 0.6411531531531531, "grad_norm": 22.2405948638916, "learning_rate": 7.534799269067952e-06, "loss": 12.216329574584961, "loss_alignment": 0.97412109375, "loss_alignment_w": 0.487060546875, "loss_sft": 0.27682623267173767, "loss_total_v6": 0.7635205835103989, "step": 2780 }, { "epoch": 0.6413837837837838, "grad_norm": 20.508304595947266, "learning_rate": 7.5330631932752515e-06, "loss": 12.153409957885742, "loss_alignment": 0.97705078125, "loss_alignment_w": 0.488525390625, "loss_sft": 0.27107796631753445, "loss_total_v6": 0.7595881000161171, "step": 2781 }, { "epoch": 0.6416144144144145, "grad_norm": 22.310012817382812, "learning_rate": 7.531326706563236e-06, "loss": 12.263919830322266, "loss_alignment": 0.98046875, "loss_alignment_w": 0.490234375, "loss_sft": 0.2764285020530224, "loss_total_v6": 0.766495019197464, "step": 2782 }, { "epoch": 0.641845045045045, "grad_norm": 23.358722686767578, "learning_rate": 7.529589809213595e-06, "loss": 12.255760192871094, "loss_alignment": 0.9638671875, "loss_alignment_w": 0.48193359375, "loss_sft": 0.2830901965498924, "loss_total_v6": 0.7659850418567657, "step": 2783 }, { "epoch": 0.6420756756756757, "grad_norm": 23.102041244506836, "learning_rate": 7.527852501508099e-06, "loss": 12.038698196411133, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.2555314563214779, "loss_total_v6": 0.752418652176857, "step": 2784 }, { "epoch": 0.6423063063063063, "grad_norm": 19.489023208618164, "learning_rate": 7.526114783728575e-06, "loss": 12.339981079101562, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.2805261183530092, "loss_total_v6": 0.7712487876415253, "step": 2785 }, { "epoch": 0.6425369369369369, "grad_norm": 20.837005615234375, "learning_rate": 7.52437665615692e-06, "loss": 12.330999374389648, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.2665218226611614, "loss_total_v6": 0.7706874683499336, "step": 2786 }, { "epoch": 0.6427675675675676, "grad_norm": 23.168399810791016, "learning_rate": 7.522638119075097e-06, "loss": 12.4967622756958, "loss_alignment": 0.9765625, "loss_alignment_w": 0.48828125, "loss_sft": 0.29259856045246124, "loss_total_v6": 0.7810476645827293, "step": 2787 }, { "epoch": 0.6429981981981981, "grad_norm": 21.256103515625, "learning_rate": 7.520899172765136e-06, "loss": 12.691174507141113, "loss_alignment": 0.982421875, "loss_alignment_w": 0.4912109375, "loss_sft": 0.3016975298523903, "loss_total_v6": 0.7931984066963196, "step": 2788 }, { "epoch": 0.6432288288288288, "grad_norm": 23.821826934814453, "learning_rate": 7.519159817509132e-06, "loss": 12.556541442871094, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.2878355644643307, "loss_total_v6": 0.784783810377121, "step": 2789 }, { "epoch": 0.6434594594594595, "grad_norm": 20.677181243896484, "learning_rate": 7.517420053589246e-06, "loss": 12.16220760345459, "loss_alignment": 0.9833984375, "loss_alignment_w": 0.49169921875, "loss_sft": 0.26833190210163593, "loss_total_v6": 0.7601379454135895, "step": 2790 }, { "epoch": 0.64369009009009, "grad_norm": 24.03567123413086, "learning_rate": 7.51567988128771e-06, "loss": 12.200699806213379, "loss_alignment": 0.97265625, "loss_alignment_w": 0.486328125, "loss_sft": 0.27647505700588226, "loss_total_v6": 0.7625437378883362, "step": 2791 }, { "epoch": 0.6439207207207207, "grad_norm": 25.5277042388916, "learning_rate": 7.513939300886816e-06, "loss": 12.28904914855957, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.26943882368505, "loss_total_v6": 0.7680655419826508, "step": 2792 }, { "epoch": 0.6441513513513514, "grad_norm": 22.01666831970215, "learning_rate": 7.512198312668926e-06, "loss": 12.864189147949219, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.30181459337472916, "loss_total_v6": 0.8040118366479874, "step": 2793 }, { "epoch": 0.6443819819819819, "grad_norm": 25.27899932861328, "learning_rate": 7.510456916916466e-06, "loss": 12.372489929199219, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.2835040111094713, "loss_total_v6": 0.7732806354761124, "step": 2794 }, { "epoch": 0.6446126126126126, "grad_norm": 22.860713958740234, "learning_rate": 7.508715113911933e-06, "loss": 12.473443984985352, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.2807651683688164, "loss_total_v6": 0.7795902490615845, "step": 2795 }, { "epoch": 0.6448432432432433, "grad_norm": 22.191757202148438, "learning_rate": 7.506972903937879e-06, "loss": 12.102977752685547, "loss_alignment": 0.98193359375, "loss_alignment_w": 0.490966796875, "loss_sft": 0.2655150853097439, "loss_total_v6": 0.7564361318945885, "step": 2796 }, { "epoch": 0.6450738738738738, "grad_norm": 22.78098487854004, "learning_rate": 7.505230287276937e-06, "loss": 12.4906005859375, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.28296663984656334, "loss_total_v6": 0.7806625738739967, "step": 2797 }, { "epoch": 0.6453045045045045, "grad_norm": 21.457489013671875, "learning_rate": 7.503487264211793e-06, "loss": 12.066862106323242, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.2551554664969444, "loss_total_v6": 0.754178911447525, "step": 2798 }, { "epoch": 0.6455351351351352, "grad_norm": 22.187191009521484, "learning_rate": 7.5017438350252056e-06, "loss": 12.874282836914062, "loss_alignment": 0.9765625, "loss_alignment_w": 0.48828125, "loss_sft": 0.3157510682940483, "loss_total_v6": 0.8046426773071289, "step": 2799 }, { "epoch": 0.6457657657657657, "grad_norm": 28.2624454498291, "learning_rate": 7.500000000000001e-06, "loss": 13.405617713928223, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.3447328694164753, "loss_total_v6": 0.8378511220216751, "step": 2800 }, { "epoch": 0.6459963963963964, "grad_norm": 23.40953254699707, "learning_rate": 7.4982557594190645e-06, "loss": 12.61440372467041, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.2914367504417896, "loss_total_v6": 0.788400262594223, "step": 2801 }, { "epoch": 0.6462270270270271, "grad_norm": 24.48843002319336, "learning_rate": 7.49651111356535e-06, "loss": 12.290648460388184, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.2734297998249531, "loss_total_v6": 0.7681655138731003, "step": 2802 }, { "epoch": 0.6464576576576576, "grad_norm": 21.74254035949707, "learning_rate": 7.494766062721881e-06, "loss": 12.367480278015137, "loss_alignment": 0.9775390625, "loss_alignment_w": 0.48876953125, "loss_sft": 0.28403013199567795, "loss_total_v6": 0.7729675024747849, "step": 2803 }, { "epoch": 0.6466882882882883, "grad_norm": 23.208852767944336, "learning_rate": 7.4930206071717435e-06, "loss": 12.236202239990234, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.26824167743325233, "loss_total_v6": 0.7647626623511314, "step": 2804 }, { "epoch": 0.646918918918919, "grad_norm": 22.837848663330078, "learning_rate": 7.491274747198086e-06, "loss": 12.686413764953613, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.29853137396275997, "loss_total_v6": 0.7929008826613426, "step": 2805 }, { "epoch": 0.6471495495495495, "grad_norm": 22.607465744018555, "learning_rate": 7.48952848308413e-06, "loss": 12.224575996398926, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.2775857839733362, "loss_total_v6": 0.7640359774231911, "step": 2806 }, { "epoch": 0.6473801801801802, "grad_norm": 20.176645278930664, "learning_rate": 7.4877818151131555e-06, "loss": 12.535266876220703, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.2820503655821085, "loss_total_v6": 0.7834541723132133, "step": 2807 }, { "epoch": 0.6476108108108108, "grad_norm": 23.835529327392578, "learning_rate": 7.486034743568512e-06, "loss": 12.693845748901367, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.2953032124787569, "loss_total_v6": 0.7933653444051743, "step": 2808 }, { "epoch": 0.6478414414414414, "grad_norm": 19.98453140258789, "learning_rate": 7.484287268733614e-06, "loss": 12.372580528259277, "loss_alignment": 0.9638671875, "loss_alignment_w": 0.48193359375, "loss_sft": 0.29121536388993263, "loss_total_v6": 0.7732862755656242, "step": 2809 }, { "epoch": 0.6480720720720721, "grad_norm": 22.489377975463867, "learning_rate": 7.482539390891941e-06, "loss": 12.02233600616455, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.2574080005288124, "loss_total_v6": 0.7513960227370262, "step": 2810 }, { "epoch": 0.6483027027027027, "grad_norm": 23.399126052856445, "learning_rate": 7.480791110327035e-06, "loss": 13.034408569335938, "loss_alignment": 0.9775390625, "loss_alignment_w": 0.48876953125, "loss_sft": 0.32653718441724777, "loss_total_v6": 0.8146505728363991, "step": 2811 }, { "epoch": 0.6485333333333333, "grad_norm": 21.48811912536621, "learning_rate": 7.479042427322509e-06, "loss": 12.624292373657227, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2852645944803953, "loss_total_v6": 0.7890182733535767, "step": 2812 }, { "epoch": 0.648763963963964, "grad_norm": 22.288198471069336, "learning_rate": 7.477293342162038e-06, "loss": 12.75782585144043, "loss_alignment": 0.9853515625, "loss_alignment_w": 0.49267578125, "loss_sft": 0.30487142503261566, "loss_total_v6": 0.7973640710115433, "step": 2813 }, { "epoch": 0.6489945945945946, "grad_norm": 33.076141357421875, "learning_rate": 7.475543855129363e-06, "loss": 12.325019836425781, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.28255130536854267, "loss_total_v6": 0.7703137472271919, "step": 2814 }, { "epoch": 0.6492252252252252, "grad_norm": 23.23664093017578, "learning_rate": 7.473793966508288e-06, "loss": 12.660863876342773, "loss_alignment": 0.98388671875, "loss_alignment_w": 0.491943359375, "loss_sft": 0.29739222675561905, "loss_total_v6": 0.791303962469101, "step": 2815 }, { "epoch": 0.6494558558558559, "grad_norm": 20.190231323242188, "learning_rate": 7.472043676582685e-06, "loss": 12.22052001953125, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.25790792517364025, "loss_total_v6": 0.7637825533747673, "step": 2816 }, { "epoch": 0.6496864864864865, "grad_norm": 22.90365219116211, "learning_rate": 7.47029298563649e-06, "loss": 12.386835098266602, "loss_alignment": 0.98193359375, "loss_alignment_w": 0.490966796875, "loss_sft": 0.2834545858204365, "loss_total_v6": 0.7741772383451462, "step": 2817 }, { "epoch": 0.6499171171171171, "grad_norm": 21.799495697021484, "learning_rate": 7.468541893953703e-06, "loss": 12.045135498046875, "loss_alignment": 0.98046875, "loss_alignment_w": 0.490234375, "loss_sft": 0.2631816677749157, "loss_total_v6": 0.7528209686279297, "step": 2818 }, { "epoch": 0.6501477477477478, "grad_norm": 21.486209869384766, "learning_rate": 7.4667904018183935e-06, "loss": 12.631988525390625, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.2963199634104967, "loss_total_v6": 0.7894992604851723, "step": 2819 }, { "epoch": 0.6503783783783784, "grad_norm": 23.46438980102539, "learning_rate": 7.465038509514688e-06, "loss": 12.586566925048828, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.29854704067111015, "loss_total_v6": 0.7866604402661324, "step": 2820 }, { "epoch": 0.650609009009009, "grad_norm": 23.62499237060547, "learning_rate": 7.463286217326786e-06, "loss": 12.496438980102539, "loss_alignment": 0.966796875, "loss_alignment_w": 0.4833984375, "loss_sft": 0.2960573323071003, "loss_total_v6": 0.7810274288058281, "step": 2821 }, { "epoch": 0.6508396396396396, "grad_norm": 24.730125427246094, "learning_rate": 7.461533525538947e-06, "loss": 12.973597526550293, "loss_alignment": 0.982421875, "loss_alignment_w": 0.4912109375, "loss_sft": 0.31945581920444965, "loss_total_v6": 0.8108498826622963, "step": 2822 }, { "epoch": 0.6510702702702703, "grad_norm": 23.509151458740234, "learning_rate": 7.459780434435496e-06, "loss": 12.356071472167969, "loss_alignment": 0.9833984375, "loss_alignment_w": 0.49169921875, "loss_sft": 0.2805246990174055, "loss_total_v6": 0.7722544595599174, "step": 2823 }, { "epoch": 0.6513009009009009, "grad_norm": 21.9525089263916, "learning_rate": 7.4580269443008245e-06, "loss": 12.408222198486328, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.27330140583217144, "loss_total_v6": 0.7755139172077179, "step": 2824 }, { "epoch": 0.6515315315315315, "grad_norm": 22.88713264465332, "learning_rate": 7.4562730554193875e-06, "loss": 12.80545425415039, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.299379613250494, "loss_total_v6": 0.8003409132361412, "step": 2825 }, { "epoch": 0.6517621621621622, "grad_norm": 23.216415405273438, "learning_rate": 7.454518768075705e-06, "loss": 12.9760160446167, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.31080265156924725, "loss_total_v6": 0.8110010027885437, "step": 2826 }, { "epoch": 0.6519927927927928, "grad_norm": 21.97073745727539, "learning_rate": 7.452764082554361e-06, "loss": 11.678609848022461, "loss_alignment": 0.97802734375, "loss_alignment_w": 0.489013671875, "loss_sft": 0.2407468818128109, "loss_total_v6": 0.7299131229519844, "step": 2827 }, { "epoch": 0.6522234234234234, "grad_norm": 22.02021026611328, "learning_rate": 7.451008999140005e-06, "loss": 12.278671264648438, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.2741919159889221, "loss_total_v6": 0.7674169912934303, "step": 2828 }, { "epoch": 0.6524540540540541, "grad_norm": 21.230571746826172, "learning_rate": 7.449253518117351e-06, "loss": 12.361169815063477, "loss_alignment": 0.98291015625, "loss_alignment_w": 0.491455078125, "loss_sft": 0.28210986219346523, "loss_total_v6": 0.7725730836391449, "step": 2829 }, { "epoch": 0.6526846846846847, "grad_norm": 22.238393783569336, "learning_rate": 7.447497639771175e-06, "loss": 12.243391036987305, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.2757099438458681, "loss_total_v6": 0.7652119174599648, "step": 2830 }, { "epoch": 0.6529153153153153, "grad_norm": 22.39800262451172, "learning_rate": 7.445741364386323e-06, "loss": 12.325666427612305, "loss_alignment": 0.98291015625, "loss_alignment_w": 0.491455078125, "loss_sft": 0.28033341094851494, "loss_total_v6": 0.770354151725769, "step": 2831 }, { "epoch": 0.653145945945946, "grad_norm": 25.528423309326172, "learning_rate": 7.443984692247701e-06, "loss": 12.210750579833984, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.2682530414313078, "loss_total_v6": 0.7631718590855598, "step": 2832 }, { "epoch": 0.6533765765765766, "grad_norm": 22.80916404724121, "learning_rate": 7.442227623640277e-06, "loss": 12.061539649963379, "loss_alignment": 0.97607421875, "loss_alignment_w": 0.488037109375, "loss_sft": 0.26564124785363674, "loss_total_v6": 0.7538461834192276, "step": 2833 }, { "epoch": 0.6536072072072072, "grad_norm": 21.857004165649414, "learning_rate": 7.440470158849091e-06, "loss": 12.213561058044434, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.2748678866773844, "loss_total_v6": 0.7633475363254547, "step": 2834 }, { "epoch": 0.6538378378378379, "grad_norm": 20.454694747924805, "learning_rate": 7.438712298159242e-06, "loss": 12.038688659667969, "loss_alignment": 0.97412109375, "loss_alignment_w": 0.487060546875, "loss_sft": 0.2652354445308447, "loss_total_v6": 0.752418041229248, "step": 2835 }, { "epoch": 0.6540684684684684, "grad_norm": 23.425039291381836, "learning_rate": 7.436954041855893e-06, "loss": 12.6378173828125, "loss_alignment": 0.966796875, "loss_alignment_w": 0.4833984375, "loss_sft": 0.30652615427970886, "loss_total_v6": 0.7898635938763618, "step": 2836 }, { "epoch": 0.6542990990990991, "grad_norm": 21.2974910736084, "learning_rate": 7.435195390224272e-06, "loss": 12.24258804321289, "loss_alignment": 0.97119140625, "loss_alignment_w": 0.485595703125, "loss_sft": 0.27874207124114037, "loss_total_v6": 0.7651617228984833, "step": 2837 }, { "epoch": 0.6545297297297298, "grad_norm": 21.76021385192871, "learning_rate": 7.433436343549674e-06, "loss": 12.582836151123047, "loss_alignment": 0.98486328125, "loss_alignment_w": 0.492431640625, "loss_sft": 0.29456018656492233, "loss_total_v6": 0.7864272221922874, "step": 2838 }, { "epoch": 0.6547603603603603, "grad_norm": 21.67563819885254, "learning_rate": 7.431676902117453e-06, "loss": 12.295650482177734, "loss_alignment": 0.98046875, "loss_alignment_w": 0.490234375, "loss_sft": 0.27880834601819515, "loss_total_v6": 0.7684781551361084, "step": 2839 }, { "epoch": 0.654990990990991, "grad_norm": 21.665334701538086, "learning_rate": 7.42991706621303e-06, "loss": 11.986518859863281, "loss_alignment": 0.97265625, "loss_alignment_w": 0.486328125, "loss_sft": 0.2623715829104185, "loss_total_v6": 0.7491574734449387, "step": 2840 }, { "epoch": 0.6552216216216217, "grad_norm": 22.237276077270508, "learning_rate": 7.428156836121892e-06, "loss": 12.134357452392578, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.27016187831759453, "loss_total_v6": 0.7583973333239555, "step": 2841 }, { "epoch": 0.6554522522522522, "grad_norm": 21.51613998413086, "learning_rate": 7.426396212129585e-06, "loss": 12.469955444335938, "loss_alignment": 0.97216796875, "loss_alignment_w": 0.486083984375, "loss_sft": 0.2919607385993004, "loss_total_v6": 0.7793722450733185, "step": 2842 }, { "epoch": 0.6556828828828829, "grad_norm": 22.72909927368164, "learning_rate": 7.424635194521722e-06, "loss": 11.878555297851562, "loss_alignment": 0.9736328125, "loss_alignment_w": 0.48681640625, "loss_sft": 0.25650881230831146, "loss_total_v6": 0.7424096986651421, "step": 2843 }, { "epoch": 0.6559135135135136, "grad_norm": 20.087722778320312, "learning_rate": 7.422873783583982e-06, "loss": 12.141009330749512, "loss_alignment": 0.984375, "loss_alignment_w": 0.4921875, "loss_sft": 0.26564902625977993, "loss_total_v6": 0.7588130906224251, "step": 2844 }, { "epoch": 0.6561441441441441, "grad_norm": 21.36869239807129, "learning_rate": 7.421111979602099e-06, "loss": 12.196990966796875, "loss_alignment": 0.9697265625, "loss_alignment_w": 0.48486328125, "loss_sft": 0.2766552399843931, "loss_total_v6": 0.7623119875788689, "step": 2845 }, { "epoch": 0.6563747747747748, "grad_norm": 22.87839126586914, "learning_rate": 7.419349782861882e-06, "loss": 12.4749174118042, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.28491610288619995, "loss_total_v6": 0.7796823233366013, "step": 2846 }, { "epoch": 0.6566054054054054, "grad_norm": 21.57476806640625, "learning_rate": 7.417587193649197e-06, "loss": 12.055716514587402, "loss_alignment": 0.966796875, "loss_alignment_w": 0.4833984375, "loss_sft": 0.27069422602653503, "loss_total_v6": 0.7534823045134544, "step": 2847 }, { "epoch": 0.656836036036036, "grad_norm": 23.36111068725586, "learning_rate": 7.415824212249977e-06, "loss": 12.465167045593262, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.2791644874960184, "loss_total_v6": 0.7790729478001595, "step": 2848 }, { "epoch": 0.6570666666666667, "grad_norm": 25.114627838134766, "learning_rate": 7.414060838950214e-06, "loss": 12.869901657104492, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.31035030633211136, "loss_total_v6": 0.8043688982725143, "step": 2849 }, { "epoch": 0.6572972972972972, "grad_norm": 26.446643829345703, "learning_rate": 7.412297074035968e-06, "loss": 12.706697463989258, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.29474847577512264, "loss_total_v6": 0.7941686138510704, "step": 2850 }, { "epoch": 0.6575279279279279, "grad_norm": 23.74821662902832, "learning_rate": 7.41053291779336e-06, "loss": 11.860311508178711, "loss_alignment": 0.96728515625, "loss_alignment_w": 0.483642578125, "loss_sft": 0.25715390406548977, "loss_total_v6": 0.7412694841623306, "step": 2851 }, { "epoch": 0.6577585585585586, "grad_norm": 21.157928466796875, "learning_rate": 7.408768370508577e-06, "loss": 12.363643646240234, "loss_alignment": 0.982421875, "loss_alignment_w": 0.4912109375, "loss_sft": 0.2821118850260973, "loss_total_v6": 0.7727277278900146, "step": 2852 }, { "epoch": 0.6579891891891891, "grad_norm": 24.340856552124023, "learning_rate": 7.407003432467867e-06, "loss": 11.968379974365234, "loss_alignment": 0.9609375, "loss_alignment_w": 0.48046875, "loss_sft": 0.2655561435967684, "loss_total_v6": 0.7480237782001495, "step": 2853 }, { "epoch": 0.6582198198198198, "grad_norm": 21.647706985473633, "learning_rate": 7.405238103957544e-06, "loss": 11.88914966583252, "loss_alignment": 0.970703125, "loss_alignment_w": 0.4853515625, "loss_sft": 0.25718627125024796, "loss_total_v6": 0.7430718764662743, "step": 2854 }, { "epoch": 0.6584504504504505, "grad_norm": 24.94748306274414, "learning_rate": 7.403472385263979e-06, "loss": 12.60710620880127, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.2921402845531702, "loss_total_v6": 0.7879441380500793, "step": 2855 }, { "epoch": 0.658681081081081, "grad_norm": 30.004663467407227, "learning_rate": 7.401706276673616e-06, "loss": 12.190069198608398, "loss_alignment": 0.9619140625, "loss_alignment_w": 0.48095703125, "loss_sft": 0.28038820438086987, "loss_total_v6": 0.7618792876601219, "step": 2856 }, { "epoch": 0.6589117117117117, "grad_norm": 22.583223342895508, "learning_rate": 7.399939778472955e-06, "loss": 13.031706809997559, "loss_alignment": 0.97900390625, "loss_alignment_w": 0.489501953125, "loss_sft": 0.3237590491771698, "loss_total_v6": 0.8144817054271698, "step": 2857 }, { "epoch": 0.6591423423423424, "grad_norm": 23.303707122802734, "learning_rate": 7.39817289094856e-06, "loss": 12.707548141479492, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.2938555181026459, "loss_total_v6": 0.7942217215895653, "step": 2858 }, { "epoch": 0.6593729729729729, "grad_norm": 19.81562042236328, "learning_rate": 7.396405614387065e-06, "loss": 12.12971019744873, "loss_alignment": 0.9677734375, "loss_alignment_w": 0.48388671875, "loss_sft": 0.27385398000478745, "loss_total_v6": 0.7581069096922874, "step": 2859 }, { "epoch": 0.6596036036036036, "grad_norm": 21.62485694885254, "learning_rate": 7.3946379490751545e-06, "loss": 12.543657302856445, "loss_alignment": 0.978515625, "loss_alignment_w": 0.4892578125, "loss_sft": 0.2954684402793646, "loss_total_v6": 0.7839785888791084, "step": 2860 }, { "epoch": 0.6598342342342343, "grad_norm": 21.386356353759766, "learning_rate": 7.392869895299587e-06, "loss": 12.820950508117676, "loss_alignment": 0.982421875, "loss_alignment_w": 0.4912109375, "loss_sft": 0.30980854108929634, "loss_total_v6": 0.8013093918561935, "step": 2861 }, { "epoch": 0.6600648648648648, "grad_norm": 24.129304885864258, "learning_rate": 7.3911014533471805e-06, "loss": 11.973916053771973, "loss_alignment": 0.95556640625, "loss_alignment_w": 0.477783203125, "loss_sft": 0.27069341763854027, "loss_total_v6": 0.7483697682619095, "step": 2862 }, { "epoch": 0.6602954954954955, "grad_norm": 21.32354164123535, "learning_rate": 7.389332623504815e-06, "loss": 12.755244255065918, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.3022839520126581, "loss_total_v6": 0.7972027882933617, "step": 2863 }, { "epoch": 0.6605261261261262, "grad_norm": 22.42266082763672, "learning_rate": 7.387563406059433e-06, "loss": 11.950725555419922, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.2525508087128401, "loss_total_v6": 0.7469203397631645, "step": 2864 }, { "epoch": 0.6607567567567567, "grad_norm": 22.438169479370117, "learning_rate": 7.3857938012980425e-06, "loss": 13.102495193481445, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.31449616700410843, "loss_total_v6": 0.8189059868454933, "step": 2865 }, { "epoch": 0.6609873873873874, "grad_norm": 23.064632415771484, "learning_rate": 7.3840238095077135e-06, "loss": 12.546810150146484, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.27875872515141964, "loss_total_v6": 0.7841756045818329, "step": 2866 }, { "epoch": 0.661218018018018, "grad_norm": 27.0670166015625, "learning_rate": 7.3822534309755756e-06, "loss": 12.805130004882812, "loss_alignment": 0.97412109375, "loss_alignment_w": 0.487060546875, "loss_sft": 0.3132142722606659, "loss_total_v6": 0.8003205955028534, "step": 2867 }, { "epoch": 0.6614486486486486, "grad_norm": 21.181381225585938, "learning_rate": 7.380482665988826e-06, "loss": 12.576894760131836, "loss_alignment": 0.978515625, "loss_alignment_w": 0.4892578125, "loss_sft": 0.29717952385544777, "loss_total_v6": 0.7860558778047562, "step": 2868 }, { "epoch": 0.6616792792792793, "grad_norm": 20.468313217163086, "learning_rate": 7.378711514834722e-06, "loss": 12.186963081359863, "loss_alignment": 0.98583984375, "loss_alignment_w": 0.492919921875, "loss_sft": 0.26839908584952354, "loss_total_v6": 0.7616851851344109, "step": 2869 }, { "epoch": 0.6619099099099099, "grad_norm": 24.859663009643555, "learning_rate": 7.376939977800581e-06, "loss": 12.480341911315918, "loss_alignment": 0.978515625, "loss_alignment_w": 0.4892578125, "loss_sft": 0.2905956692993641, "loss_total_v6": 0.7800213620066643, "step": 2870 }, { "epoch": 0.6621405405405405, "grad_norm": 24.39521598815918, "learning_rate": 7.375168055173789e-06, "loss": 12.125673294067383, "loss_alignment": 0.9755859375, "loss_alignment_w": 0.48779296875, "loss_sft": 0.27035157196223736, "loss_total_v6": 0.7578546106815338, "step": 2871 }, { "epoch": 0.6623711711711712, "grad_norm": 24.931411743164062, "learning_rate": 7.373395747241792e-06, "loss": 12.094829559326172, "loss_alignment": 0.95947265625, "loss_alignment_w": 0.479736328125, "loss_sft": 0.2750765774399042, "loss_total_v6": 0.7559268027544022, "step": 2872 }, { "epoch": 0.6626018018018018, "grad_norm": 24.106225967407227, "learning_rate": 7.371623054292095e-06, "loss": 12.030743598937988, "loss_alignment": 0.990234375, "loss_alignment_w": 0.4951171875, "loss_sft": 0.2575977109372616, "loss_total_v6": 0.7519214376807213, "step": 2873 }, { "epoch": 0.6628324324324324, "grad_norm": 21.133329391479492, "learning_rate": 7.369849976612268e-06, "loss": 12.54345703125, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.2854462582617998, "loss_total_v6": 0.7839661315083504, "step": 2874 }, { "epoch": 0.6630630630630631, "grad_norm": 23.848379135131836, "learning_rate": 7.368076514489947e-06, "loss": 12.783682823181152, "loss_alignment": 0.966796875, "loss_alignment_w": 0.4833984375, "loss_sft": 0.31539864651858807, "loss_total_v6": 0.7989801913499832, "step": 2875 }, { "epoch": 0.6632936936936937, "grad_norm": 22.341293334960938, "learning_rate": 7.3663026682128255e-06, "loss": 12.352383613586426, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.28151488676667213, "loss_total_v6": 0.7720239385962486, "step": 2876 }, { "epoch": 0.6635243243243243, "grad_norm": 23.98525619506836, "learning_rate": 7.364528438068659e-06, "loss": 11.736189842224121, "loss_alignment": 0.990234375, "loss_alignment_w": 0.4951171875, "loss_sft": 0.23882192373275757, "loss_total_v6": 0.733511857688427, "step": 2877 }, { "epoch": 0.663754954954955, "grad_norm": 22.243127822875977, "learning_rate": 7.362753824345271e-06, "loss": 12.260704040527344, "loss_alignment": 0.95458984375, "loss_alignment_w": 0.477294921875, "loss_sft": 0.2893194518983364, "loss_total_v6": 0.7662939578294754, "step": 2878 }, { "epoch": 0.6639855855855856, "grad_norm": 23.388957977294922, "learning_rate": 7.36097882733054e-06, "loss": 12.715058326721191, "loss_alignment": 0.984375, "loss_alignment_w": 0.4921875, "loss_sft": 0.3039837460964918, "loss_total_v6": 0.7946911528706551, "step": 2879 }, { "epoch": 0.6642162162162162, "grad_norm": 22.47502899169922, "learning_rate": 7.35920344731241e-06, "loss": 12.476933479309082, "loss_alignment": 0.978515625, "loss_alignment_w": 0.4892578125, "loss_sft": 0.2910082880407572, "loss_total_v6": 0.7798083275556564, "step": 2880 }, { "epoch": 0.6644468468468469, "grad_norm": 23.391460418701172, "learning_rate": 7.357427684578893e-06, "loss": 12.116033554077148, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.2577098719775677, "loss_total_v6": 0.7572521045804024, "step": 2881 }, { "epoch": 0.6646774774774775, "grad_norm": 21.434267044067383, "learning_rate": 7.355651539418049e-06, "loss": 12.647032737731934, "loss_alignment": 0.98388671875, "loss_alignment_w": 0.491943359375, "loss_sft": 0.298023147508502, "loss_total_v6": 0.7904395386576653, "step": 2882 }, { "epoch": 0.6649081081081081, "grad_norm": 23.438674926757812, "learning_rate": 7.353875012118013e-06, "loss": 12.554887771606445, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.29635338112711906, "loss_total_v6": 0.7846804410219193, "step": 2883 }, { "epoch": 0.6651387387387387, "grad_norm": 24.345399856567383, "learning_rate": 7.352098102966978e-06, "loss": 12.546371459960938, "loss_alignment": 0.9853515625, "loss_alignment_w": 0.49267578125, "loss_sft": 0.2924184575676918, "loss_total_v6": 0.7841481864452362, "step": 2884 }, { "epoch": 0.6653693693693694, "grad_norm": 26.331130981445312, "learning_rate": 7.350320812253196e-06, "loss": 12.391056060791016, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.275326082482934, "loss_total_v6": 0.7744410336017609, "step": 2885 }, { "epoch": 0.6656, "grad_norm": 21.22938346862793, "learning_rate": 7.348543140264984e-06, "loss": 12.046506881713867, "loss_alignment": 0.970703125, "loss_alignment_w": 0.4853515625, "loss_sft": 0.26615127362310886, "loss_total_v6": 0.7529066503047943, "step": 2886 }, { "epoch": 0.6658306306306306, "grad_norm": 21.960891723632812, "learning_rate": 7.346765087290719e-06, "loss": 11.975977897644043, "loss_alignment": 0.97119140625, "loss_alignment_w": 0.485595703125, "loss_sft": 0.26336068101227283, "loss_total_v6": 0.7484986335039139, "step": 2887 }, { "epoch": 0.6660612612612613, "grad_norm": 24.05299949645996, "learning_rate": 7.344986653618844e-06, "loss": 12.342382431030273, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.27814328484237194, "loss_total_v6": 0.7713988795876503, "step": 2888 }, { "epoch": 0.6662918918918919, "grad_norm": 22.901830673217773, "learning_rate": 7.343207839537858e-06, "loss": 12.710206985473633, "loss_alignment": 0.98486328125, "loss_alignment_w": 0.492431640625, "loss_sft": 0.3024751450866461, "loss_total_v6": 0.7943879812955856, "step": 2889 }, { "epoch": 0.6665225225225225, "grad_norm": 22.685007095336914, "learning_rate": 7.341428645336324e-06, "loss": 12.054071426391602, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.26303828321397305, "loss_total_v6": 0.7533794939517975, "step": 2890 }, { "epoch": 0.6667531531531532, "grad_norm": 24.759658813476562, "learning_rate": 7.3396490713028674e-06, "loss": 12.329870223999023, "loss_alignment": 0.98583984375, "loss_alignment_w": 0.492919921875, "loss_sft": 0.27560653537511826, "loss_total_v6": 0.7706169337034225, "step": 2891 }, { "epoch": 0.6669837837837838, "grad_norm": 21.302154541015625, "learning_rate": 7.337869117726176e-06, "loss": 12.101152420043945, "loss_alignment": 0.97314453125, "loss_alignment_w": 0.486572265625, "loss_sft": 0.27025330252945423, "loss_total_v6": 0.756322018802166, "step": 2892 }, { "epoch": 0.6672144144144144, "grad_norm": 21.477771759033203, "learning_rate": 7.336088784894997e-06, "loss": 12.361943244934082, "loss_alignment": 0.9765625, "loss_alignment_w": 0.48828125, "loss_sft": 0.28414182364940643, "loss_total_v6": 0.7726214602589607, "step": 2893 }, { "epoch": 0.6674450450450451, "grad_norm": 21.91358184814453, "learning_rate": 7.33430807309814e-06, "loss": 11.575233459472656, "loss_alignment": 0.96533203125, "loss_alignment_w": 0.482666015625, "loss_sft": 0.24043505638837814, "loss_total_v6": 0.7234520465135574, "step": 2894 }, { "epoch": 0.6676756756756757, "grad_norm": 23.059627532958984, "learning_rate": 7.332526982624476e-06, "loss": 12.283774375915527, "loss_alignment": 0.98388671875, "loss_alignment_w": 0.491943359375, "loss_sft": 0.2750448063015938, "loss_total_v6": 0.7677358835935593, "step": 2895 }, { "epoch": 0.6679063063063063, "grad_norm": 21.829484939575195, "learning_rate": 7.330745513762937e-06, "loss": 12.424570083618164, "loss_alignment": 0.96728515625, "loss_alignment_w": 0.483642578125, "loss_sft": 0.29304564744234085, "loss_total_v6": 0.7765356004238129, "step": 2896 }, { "epoch": 0.668136936936937, "grad_norm": 23.29634666442871, "learning_rate": 7.32896366680252e-06, "loss": 12.68918228149414, "loss_alignment": 0.97265625, "loss_alignment_w": 0.486328125, "loss_sft": 0.3074476011097431, "loss_total_v6": 0.793073870241642, "step": 2897 }, { "epoch": 0.6683675675675675, "grad_norm": 22.288663864135742, "learning_rate": 7.327181442032277e-06, "loss": 12.318470001220703, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.27564174868166447, "loss_total_v6": 0.7699044197797775, "step": 2898 }, { "epoch": 0.6685981981981982, "grad_norm": 25.76631736755371, "learning_rate": 7.325398839741325e-06, "loss": 12.691831588745117, "loss_alignment": 0.9814453125, "loss_alignment_w": 0.49072265625, "loss_sft": 0.3019369952380657, "loss_total_v6": 0.7932394966483116, "step": 2899 }, { "epoch": 0.6688288288288289, "grad_norm": 22.749069213867188, "learning_rate": 7.323615860218844e-06, "loss": 12.127145767211914, "loss_alignment": 0.9716796875, "loss_alignment_w": 0.48583984375, "loss_sft": 0.27167948335409164, "loss_total_v6": 0.7579466104507446, "step": 2900 }, { "epoch": 0.6690594594594594, "grad_norm": 24.347431182861328, "learning_rate": 7.321832503754071e-06, "loss": 12.495441436767578, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.28290300257503986, "loss_total_v6": 0.7809651121497154, "step": 2901 }, { "epoch": 0.6692900900900901, "grad_norm": 22.51470375061035, "learning_rate": 7.3200487706363054e-06, "loss": 12.233091354370117, "loss_alignment": 0.9755859375, "loss_alignment_w": 0.48779296875, "loss_sft": 0.27795018441975117, "loss_total_v6": 0.7645682170987129, "step": 2902 }, { "epoch": 0.6695207207207208, "grad_norm": 21.232955932617188, "learning_rate": 7.3182646611549125e-06, "loss": 12.090433120727539, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2543856408447027, "loss_total_v6": 0.7556521147489548, "step": 2903 }, { "epoch": 0.6697513513513513, "grad_norm": 21.798805236816406, "learning_rate": 7.31648017559931e-06, "loss": 12.129863739013672, "loss_alignment": 0.97607421875, "loss_alignment_w": 0.488037109375, "loss_sft": 0.2709186151623726, "loss_total_v6": 0.7581164762377739, "step": 2904 }, { "epoch": 0.669981981981982, "grad_norm": 22.995914459228516, "learning_rate": 7.314695314258983e-06, "loss": 11.973889350891113, "loss_alignment": 0.9462890625, "loss_alignment_w": 0.47314453125, "loss_sft": 0.27452164702117443, "loss_total_v6": 0.748368076980114, "step": 2905 }, { "epoch": 0.6702126126126126, "grad_norm": 23.89446258544922, "learning_rate": 7.312910077423477e-06, "loss": 12.3638916015625, "loss_alignment": 0.9873046875, "loss_alignment_w": 0.49365234375, "loss_sft": 0.27875521779060364, "loss_total_v6": 0.7727432325482368, "step": 2906 }, { "epoch": 0.6704432432432432, "grad_norm": 24.1834716796875, "learning_rate": 7.311124465382395e-06, "loss": 12.794812202453613, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.30373458564281464, "loss_total_v6": 0.7996757328510284, "step": 2907 }, { "epoch": 0.6706738738738739, "grad_norm": 24.662817001342773, "learning_rate": 7.309338478425405e-06, "loss": 12.82931900024414, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.30369411036372185, "loss_total_v6": 0.8018324971199036, "step": 2908 }, { "epoch": 0.6709045045045045, "grad_norm": 22.928251266479492, "learning_rate": 7.307552116842233e-06, "loss": 12.562702178955078, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.2917301580309868, "loss_total_v6": 0.7851689010858536, "step": 2909 }, { "epoch": 0.6711351351351351, "grad_norm": 22.378198623657227, "learning_rate": 7.305765380922668e-06, "loss": 12.459534645080566, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.2794686034321785, "loss_total_v6": 0.7787209153175354, "step": 2910 }, { "epoch": 0.6713657657657658, "grad_norm": 22.805335998535156, "learning_rate": 7.303978270956557e-06, "loss": 11.926656723022461, "loss_alignment": 0.9814453125, "loss_alignment_w": 0.49072265625, "loss_sft": 0.2539914771914482, "loss_total_v6": 0.745416022837162, "step": 2911 }, { "epoch": 0.6715963963963963, "grad_norm": 21.519515991210938, "learning_rate": 7.302190787233808e-06, "loss": 12.12685775756836, "loss_alignment": 0.98095703125, "loss_alignment_w": 0.490478515625, "loss_sft": 0.26841140910983086, "loss_total_v6": 0.7579286247491837, "step": 2912 }, { "epoch": 0.671827027027027, "grad_norm": 22.217798233032227, "learning_rate": 7.300402930044394e-06, "loss": 11.947132110595703, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.2565987389534712, "loss_total_v6": 0.7466957867145538, "step": 2913 }, { "epoch": 0.6720576576576577, "grad_norm": 21.508092880249023, "learning_rate": 7.298614699678341e-06, "loss": 12.29173755645752, "loss_alignment": 0.98291015625, "loss_alignment_w": 0.491455078125, "loss_sft": 0.27734311670064926, "loss_total_v6": 0.7682336047291756, "step": 2914 }, { "epoch": 0.6722882882882882, "grad_norm": 22.556474685668945, "learning_rate": 7.296826096425743e-06, "loss": 12.497453689575195, "loss_alignment": 0.9638671875, "loss_alignment_w": 0.48193359375, "loss_sft": 0.29889787547290325, "loss_total_v6": 0.7810908779501915, "step": 2915 }, { "epoch": 0.6725189189189189, "grad_norm": 21.61472511291504, "learning_rate": 7.29503712057675e-06, "loss": 12.721355438232422, "loss_alignment": 0.9775390625, "loss_alignment_w": 0.48876953125, "loss_sft": 0.30607109144330025, "loss_total_v6": 0.7950847372412682, "step": 2916 }, { "epoch": 0.6727495495495496, "grad_norm": 21.321176528930664, "learning_rate": 7.293247772421577e-06, "loss": 12.934386253356934, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.31277843192219734, "loss_total_v6": 0.8083991557359695, "step": 2917 }, { "epoch": 0.6729801801801801, "grad_norm": 24.181921005249023, "learning_rate": 7.291458052250488e-06, "loss": 11.959417343139648, "loss_alignment": 0.97607421875, "loss_alignment_w": 0.488037109375, "loss_sft": 0.2579464390873909, "loss_total_v6": 0.747463621199131, "step": 2918 }, { "epoch": 0.6732108108108108, "grad_norm": 20.7012882232666, "learning_rate": 7.289667960353826e-06, "loss": 11.463644981384277, "loss_alignment": 0.974609375, "loss_alignment_w": 0.4873046875, "loss_sft": 0.2291120756417513, "loss_total_v6": 0.7164777964353561, "step": 2919 }, { "epoch": 0.6734414414414415, "grad_norm": 20.40899658203125, "learning_rate": 7.287877497021978e-06, "loss": 12.690656661987305, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.2982320114970207, "loss_total_v6": 0.7931660711765289, "step": 2920 }, { "epoch": 0.673672072072072, "grad_norm": 22.00999641418457, "learning_rate": 7.286086662545395e-06, "loss": 12.28514575958252, "loss_alignment": 0.9814453125, "loss_alignment_w": 0.49072265625, "loss_sft": 0.2756646368652582, "loss_total_v6": 0.7678216174244881, "step": 2921 }, { "epoch": 0.6739027027027027, "grad_norm": 21.94966697692871, "learning_rate": 7.284295457214596e-06, "loss": 12.145933151245117, "loss_alignment": 0.970703125, "loss_alignment_w": 0.4853515625, "loss_sft": 0.2715872675180435, "loss_total_v6": 0.7591208443045616, "step": 2922 }, { "epoch": 0.6741333333333334, "grad_norm": 22.536746978759766, "learning_rate": 7.282503881320149e-06, "loss": 12.117833137512207, "loss_alignment": 0.978515625, "loss_alignment_w": 0.4892578125, "loss_sft": 0.2686255779117346, "loss_total_v6": 0.7573645636439323, "step": 2923 }, { "epoch": 0.6743639639639639, "grad_norm": 22.256683349609375, "learning_rate": 7.2807119351526915e-06, "loss": 12.195476531982422, "loss_alignment": 0.96923828125, "loss_alignment_w": 0.484619140625, "loss_sft": 0.27726248651742935, "loss_total_v6": 0.7622173130512238, "step": 2924 }, { "epoch": 0.6745945945945946, "grad_norm": 23.18370246887207, "learning_rate": 7.2789196190029155e-06, "loss": 12.311399459838867, "loss_alignment": 0.97412109375, "loss_alignment_w": 0.487060546875, "loss_sft": 0.28331749327480793, "loss_total_v6": 0.7694624811410904, "step": 2925 }, { "epoch": 0.6748252252252253, "grad_norm": 23.412837982177734, "learning_rate": 7.2771269331615744e-06, "loss": 12.3712158203125, "loss_alignment": 0.974609375, "loss_alignment_w": 0.4873046875, "loss_sft": 0.2873611357063055, "loss_total_v6": 0.7732009589672089, "step": 2926 }, { "epoch": 0.6750558558558558, "grad_norm": 22.95995330810547, "learning_rate": 7.275333877919483e-06, "loss": 12.273845672607422, "loss_alignment": 0.94970703125, "loss_alignment_w": 0.474853515625, "loss_sft": 0.29065962694585323, "loss_total_v6": 0.7671153321862221, "step": 2927 }, { "epoch": 0.6752864864864865, "grad_norm": 22.837234497070312, "learning_rate": 7.273540453567512e-06, "loss": 12.142107009887695, "loss_alignment": 0.97705078125, "loss_alignment_w": 0.488525390625, "loss_sft": 0.27044777572155, "loss_total_v6": 0.7588816657662392, "step": 2928 }, { "epoch": 0.6755171171171171, "grad_norm": 23.118772506713867, "learning_rate": 7.271746660396599e-06, "loss": 11.744569778442383, "loss_alignment": 0.9765625, "loss_alignment_w": 0.48828125, "loss_sft": 0.24502190947532654, "loss_total_v6": 0.7340355738997459, "step": 2929 }, { "epoch": 0.6757477477477477, "grad_norm": 22.508363723754883, "learning_rate": 7.269952498697734e-06, "loss": 13.1004056930542, "loss_alignment": 0.9697265625, "loss_alignment_w": 0.48486328125, "loss_sft": 0.33325594291090965, "loss_total_v6": 0.818775363266468, "step": 2930 }, { "epoch": 0.6759783783783784, "grad_norm": 22.466154098510742, "learning_rate": 7.2681579687619715e-06, "loss": 12.058773040771484, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.25608418695628643, "loss_total_v6": 0.7536732852458954, "step": 2931 }, { "epoch": 0.676209009009009, "grad_norm": 20.959402084350586, "learning_rate": 7.266363070880424e-06, "loss": 12.534990310668945, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.29393493570387363, "loss_total_v6": 0.7834369093179703, "step": 2932 }, { "epoch": 0.6764396396396396, "grad_norm": 24.527183532714844, "learning_rate": 7.2645678053442645e-06, "loss": 12.581037521362305, "loss_alignment": 0.96630859375, "loss_alignment_w": 0.483154296875, "loss_sft": 0.30264173820614815, "loss_total_v6": 0.7863148227334023, "step": 2933 }, { "epoch": 0.6766702702702703, "grad_norm": 24.09444236755371, "learning_rate": 7.262772172444723e-06, "loss": 12.54616641998291, "loss_alignment": 0.96875, "loss_alignment_w": 0.484375, "loss_sft": 0.300386019051075, "loss_total_v6": 0.7841354236006737, "step": 2934 }, { "epoch": 0.6769009009009009, "grad_norm": 23.469125747680664, "learning_rate": 7.260976172473094e-06, "loss": 12.190604209899902, "loss_alignment": 0.9638671875, "loss_alignment_w": 0.48193359375, "loss_sft": 0.2796892747282982, "loss_total_v6": 0.7619127631187439, "step": 2935 }, { "epoch": 0.6771315315315315, "grad_norm": 22.24032974243164, "learning_rate": 7.259179805720726e-06, "loss": 12.110214233398438, "loss_alignment": 0.9638671875, "loss_alignment_w": 0.48193359375, "loss_sft": 0.27521413937211037, "loss_total_v6": 0.7568883299827576, "step": 2936 }, { "epoch": 0.6773621621621622, "grad_norm": 22.19389533996582, "learning_rate": 7.257383072479032e-06, "loss": 12.363336563110352, "loss_alignment": 0.98583984375, "loss_alignment_w": 0.492919921875, "loss_sft": 0.2801395133137703, "loss_total_v6": 0.772708497941494, "step": 2937 }, { "epoch": 0.6775927927927928, "grad_norm": 23.93843650817871, "learning_rate": 7.25558597303948e-06, "loss": 12.627889633178711, "loss_alignment": 0.99267578125, "loss_alignment_w": 0.496337890625, "loss_sft": 0.2912725545465946, "loss_total_v6": 0.7892431169748306, "step": 2938 }, { "epoch": 0.6778234234234234, "grad_norm": 22.764293670654297, "learning_rate": 7.253788507693602e-06, "loss": 12.47040843963623, "loss_alignment": 0.9619140625, "loss_alignment_w": 0.48095703125, "loss_sft": 0.29923696257174015, "loss_total_v6": 0.7794005498290062, "step": 2939 }, { "epoch": 0.6780540540540541, "grad_norm": 22.841737747192383, "learning_rate": 7.251990676732985e-06, "loss": 12.276464462280273, "loss_alignment": 0.9599609375, "loss_alignment_w": 0.47998046875, "loss_sft": 0.2877716328948736, "loss_total_v6": 0.7672790437936783, "step": 2940 }, { "epoch": 0.6782846846846847, "grad_norm": 22.86682891845703, "learning_rate": 7.2501924804492784e-06, "loss": 12.27908706665039, "loss_alignment": 0.984375, "loss_alignment_w": 0.4921875, "loss_sft": 0.2742178700864315, "loss_total_v6": 0.7674429640173912, "step": 2941 }, { "epoch": 0.6785153153153153, "grad_norm": 21.58182144165039, "learning_rate": 7.248393919134189e-06, "loss": 12.19426155090332, "loss_alignment": 0.9716796875, "loss_alignment_w": 0.48583984375, "loss_sft": 0.276240449398756, "loss_total_v6": 0.7621413320302963, "step": 2942 }, { "epoch": 0.678745945945946, "grad_norm": 22.323307037353516, "learning_rate": 7.246594993079483e-06, "loss": 12.202613830566406, "loss_alignment": 0.97265625, "loss_alignment_w": 0.486328125, "loss_sft": 0.2771592363715172, "loss_total_v6": 0.7626633867621422, "step": 2943 }, { "epoch": 0.6789765765765766, "grad_norm": 25.347904205322266, "learning_rate": 7.24479570257699e-06, "loss": 11.739957809448242, "loss_alignment": 0.95068359375, "loss_alignment_w": 0.475341796875, "loss_sft": 0.25936685502529144, "loss_total_v6": 0.7337473630905151, "step": 2944 }, { "epoch": 0.6792072072072072, "grad_norm": 22.49755859375, "learning_rate": 7.242996047918589e-06, "loss": 12.081315994262695, "loss_alignment": 0.9677734375, "loss_alignment_w": 0.48388671875, "loss_sft": 0.27012742310762405, "loss_total_v6": 0.7550822347402573, "step": 2945 }, { "epoch": 0.6794378378378378, "grad_norm": 20.61044692993164, "learning_rate": 7.2411960293962294e-06, "loss": 12.194814682006836, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.26481563597917557, "loss_total_v6": 0.7621758878231049, "step": 2946 }, { "epoch": 0.6796684684684685, "grad_norm": 23.42919921875, "learning_rate": 7.239395647301912e-06, "loss": 12.263540267944336, "loss_alignment": 0.96728515625, "loss_alignment_w": 0.483642578125, "loss_sft": 0.2836069092154503, "loss_total_v6": 0.766471303999424, "step": 2947 }, { "epoch": 0.6798990990990991, "grad_norm": 24.219959259033203, "learning_rate": 7.237594901927699e-06, "loss": 12.455941200256348, "loss_alignment": 0.96337890625, "loss_alignment_w": 0.481689453125, "loss_sft": 0.2966695688664913, "loss_total_v6": 0.7784963473677635, "step": 2948 }, { "epoch": 0.6801297297297297, "grad_norm": 24.667436599731445, "learning_rate": 7.2357937935657105e-06, "loss": 11.83835220336914, "loss_alignment": 0.95556640625, "loss_alignment_w": 0.477783203125, "loss_sft": 0.2615187019109726, "loss_total_v6": 0.7398969978094101, "step": 2949 }, { "epoch": 0.6803603603603604, "grad_norm": 22.971479415893555, "learning_rate": 7.2339923225081296e-06, "loss": 12.25065803527832, "loss_alignment": 0.96630859375, "loss_alignment_w": 0.483154296875, "loss_sft": 0.2821608018130064, "loss_total_v6": 0.7656660676002502, "step": 2950 }, { "epoch": 0.680590990990991, "grad_norm": 23.594724655151367, "learning_rate": 7.232190489047193e-06, "loss": 12.649406433105469, "loss_alignment": 0.97119140625, "loss_alignment_w": 0.485595703125, "loss_sft": 0.30351204983890057, "loss_total_v6": 0.7905878648161888, "step": 2951 }, { "epoch": 0.6808216216216216, "grad_norm": 23.50196647644043, "learning_rate": 7.2303882934751965e-06, "loss": 12.093050003051758, "loss_alignment": 0.970703125, "loss_alignment_w": 0.4853515625, "loss_sft": 0.270021541044116, "loss_total_v6": 0.7558156624436378, "step": 2952 }, { "epoch": 0.6810522522522523, "grad_norm": 23.602109909057617, "learning_rate": 7.228585736084499e-06, "loss": 11.891511917114258, "loss_alignment": 0.96533203125, "loss_alignment_w": 0.482666015625, "loss_sft": 0.26059927605092525, "loss_total_v6": 0.7432195097208023, "step": 2953 }, { "epoch": 0.6812828828828829, "grad_norm": 22.693323135375977, "learning_rate": 7.226782817167515e-06, "loss": 12.406402587890625, "loss_alignment": 0.978515625, "loss_alignment_w": 0.4892578125, "loss_sft": 0.28641698509454727, "loss_total_v6": 0.7754001468420029, "step": 2954 }, { "epoch": 0.6815135135135135, "grad_norm": 22.103900909423828, "learning_rate": 7.224979537016716e-06, "loss": 12.274612426757812, "loss_alignment": 0.9658203125, "loss_alignment_w": 0.48291015625, "loss_sft": 0.2826204337179661, "loss_total_v6": 0.7671632841229439, "step": 2955 }, { "epoch": 0.6817441441441442, "grad_norm": 22.990476608276367, "learning_rate": 7.223175895924638e-06, "loss": 11.635080337524414, "loss_alignment": 0.96533203125, "loss_alignment_w": 0.482666015625, "loss_sft": 0.24451125599443913, "loss_total_v6": 0.7271925434470177, "step": 2956 }, { "epoch": 0.6819747747747748, "grad_norm": 23.152273178100586, "learning_rate": 7.2213718941838704e-06, "loss": 12.146513938903809, "loss_alignment": 0.9892578125, "loss_alignment_w": 0.49462890625, "loss_sft": 0.2651081010699272, "loss_total_v6": 0.7591571435332298, "step": 2957 }, { "epoch": 0.6822054054054054, "grad_norm": 23.207075119018555, "learning_rate": 7.21956753208706e-06, "loss": 12.393768310546875, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.2725811079144478, "loss_total_v6": 0.7746105194091797, "step": 2958 }, { "epoch": 0.6824360360360361, "grad_norm": 25.42705726623535, "learning_rate": 7.217762809926917e-06, "loss": 12.789701461791992, "loss_alignment": 0.98046875, "loss_alignment_w": 0.490234375, "loss_sft": 0.3100527133792639, "loss_total_v6": 0.7993563115596771, "step": 2959 }, { "epoch": 0.6826666666666666, "grad_norm": 22.54009437561035, "learning_rate": 7.215957727996208e-06, "loss": 12.174041748046875, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.27224540896713734, "loss_total_v6": 0.7608776390552521, "step": 2960 }, { "epoch": 0.6828972972972973, "grad_norm": 23.407651901245117, "learning_rate": 7.214152286587755e-06, "loss": 12.022041320800781, "loss_alignment": 0.951171875, "loss_alignment_w": 0.4755859375, "loss_sft": 0.2755627389997244, "loss_total_v6": 0.7513775527477264, "step": 2961 }, { "epoch": 0.683127927927928, "grad_norm": 27.670080184936523, "learning_rate": 7.212346485994443e-06, "loss": 12.147321701049805, "loss_alignment": 0.9560546875, "loss_alignment_w": 0.47802734375, "loss_sft": 0.2802189327776432, "loss_total_v6": 0.7592076063156128, "step": 2962 }, { "epoch": 0.6833585585585585, "grad_norm": 23.341163635253906, "learning_rate": 7.210540326509213e-06, "loss": 12.17013931274414, "loss_alignment": 0.98388671875, "loss_alignment_w": 0.491943359375, "loss_sft": 0.2683851830661297, "loss_total_v6": 0.7606337368488312, "step": 2963 }, { "epoch": 0.6835891891891892, "grad_norm": 22.920915603637695, "learning_rate": 7.208733808425064e-06, "loss": 12.368231773376465, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.27614254131913185, "loss_total_v6": 0.773014485836029, "step": 2964 }, { "epoch": 0.6838198198198199, "grad_norm": 24.622034072875977, "learning_rate": 7.206926932035052e-06, "loss": 12.569385528564453, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.29677125811576843, "loss_total_v6": 0.7855865880846977, "step": 2965 }, { "epoch": 0.6840504504504504, "grad_norm": 24.384469985961914, "learning_rate": 7.205119697632297e-06, "loss": 12.351297378540039, "loss_alignment": 0.9609375, "loss_alignment_w": 0.48046875, "loss_sft": 0.29107535257935524, "loss_total_v6": 0.771956093609333, "step": 2966 }, { "epoch": 0.6842810810810811, "grad_norm": 23.44341278076172, "learning_rate": 7.2033121055099675e-06, "loss": 12.12735366821289, "loss_alignment": 0.962890625, "loss_alignment_w": 0.4814453125, "loss_sft": 0.27710932679474354, "loss_total_v6": 0.7579595670104027, "step": 2967 }, { "epoch": 0.6845117117117117, "grad_norm": 23.406160354614258, "learning_rate": 7.201504155961296e-06, "loss": 12.193084716796875, "loss_alignment": 0.9697265625, "loss_alignment_w": 0.48486328125, "loss_sft": 0.2780742160975933, "loss_total_v6": 0.7620677724480629, "step": 2968 }, { "epoch": 0.6847423423423423, "grad_norm": 23.594703674316406, "learning_rate": 7.199695849279576e-06, "loss": 12.326294898986816, "loss_alignment": 0.958984375, "loss_alignment_w": 0.4794921875, "loss_sft": 0.2927933558821678, "loss_total_v6": 0.7703934386372566, "step": 2969 }, { "epoch": 0.684972972972973, "grad_norm": 24.90220069885254, "learning_rate": 7.197887185758153e-06, "loss": 12.553789138793945, "loss_alignment": 0.96630859375, "loss_alignment_w": 0.483154296875, "loss_sft": 0.3025866951793432, "loss_total_v6": 0.7846118435263634, "step": 2970 }, { "epoch": 0.6852036036036036, "grad_norm": 21.028003692626953, "learning_rate": 7.196078165690431e-06, "loss": 12.365862846374512, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.27404133789241314, "loss_total_v6": 0.7728664129972458, "step": 2971 }, { "epoch": 0.6854342342342342, "grad_norm": 23.262495040893555, "learning_rate": 7.194268789369875e-06, "loss": 12.673558235168457, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.2992995399981737, "loss_total_v6": 0.7920973598957062, "step": 2972 }, { "epoch": 0.6856648648648649, "grad_norm": 23.803125381469727, "learning_rate": 7.192459057090007e-06, "loss": 12.319358825683594, "loss_alignment": 0.9609375, "loss_alignment_w": 0.48046875, "loss_sft": 0.2895827069878578, "loss_total_v6": 0.769959919154644, "step": 2973 }, { "epoch": 0.6858954954954954, "grad_norm": 22.31180191040039, "learning_rate": 7.190648969144403e-06, "loss": 12.33380126953125, "loss_alignment": 0.96875, "loss_alignment_w": 0.484375, "loss_sft": 0.2869758903980255, "loss_total_v6": 0.7708625867962837, "step": 2974 }, { "epoch": 0.6861261261261261, "grad_norm": 24.246238708496094, "learning_rate": 7.188838525826702e-06, "loss": 12.556872367858887, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.28495713137090206, "loss_total_v6": 0.7848045155405998, "step": 2975 }, { "epoch": 0.6863567567567568, "grad_norm": 24.695781707763672, "learning_rate": 7.1870277274306e-06, "loss": 12.285839080810547, "loss_alignment": 0.95556640625, "loss_alignment_w": 0.477783203125, "loss_sft": 0.28947137482464314, "loss_total_v6": 0.7678649425506592, "step": 2976 }, { "epoch": 0.6865873873873873, "grad_norm": 22.785621643066406, "learning_rate": 7.185216574249844e-06, "loss": 12.878998756408691, "loss_alignment": 0.984375, "loss_alignment_w": 0.4921875, "loss_sft": 0.313543364405632, "loss_total_v6": 0.8049374222755432, "step": 2977 }, { "epoch": 0.686818018018018, "grad_norm": 22.517765045166016, "learning_rate": 7.1834050665782474e-06, "loss": 13.288081169128418, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.33010828867554665, "loss_total_v6": 0.8305050656199455, "step": 2978 }, { "epoch": 0.6870486486486487, "grad_norm": 24.429536819458008, "learning_rate": 7.181593204709676e-06, "loss": 12.181395530700684, "loss_alignment": 0.9853515625, "loss_alignment_w": 0.49267578125, "loss_sft": 0.2681578751653433, "loss_total_v6": 0.7613372132182121, "step": 2979 }, { "epoch": 0.6872792792792792, "grad_norm": 23.601669311523438, "learning_rate": 7.179780988938051e-06, "loss": 12.30968189239502, "loss_alignment": 0.96484375, "loss_alignment_w": 0.482421875, "loss_sft": 0.2880471646785736, "loss_total_v6": 0.7693551257252693, "step": 2980 }, { "epoch": 0.6875099099099099, "grad_norm": 21.922130584716797, "learning_rate": 7.177968419557359e-06, "loss": 12.349893569946289, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.27870425768196583, "loss_total_v6": 0.7718683406710625, "step": 2981 }, { "epoch": 0.6877405405405406, "grad_norm": 21.856447219848633, "learning_rate": 7.176155496861639e-06, "loss": 12.54150104522705, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.27906783670186996, "loss_total_v6": 0.7838438153266907, "step": 2982 }, { "epoch": 0.6879711711711711, "grad_norm": 23.19557762145996, "learning_rate": 7.174342221144983e-06, "loss": 12.175047874450684, "loss_alignment": 0.9755859375, "loss_alignment_w": 0.48779296875, "loss_sft": 0.2723387908190489, "loss_total_v6": 0.7609404847025871, "step": 2983 }, { "epoch": 0.6882018018018018, "grad_norm": 23.719032287597656, "learning_rate": 7.17252859270155e-06, "loss": 12.391876220703125, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.28511236421763897, "loss_total_v6": 0.7744922339916229, "step": 2984 }, { "epoch": 0.6884324324324325, "grad_norm": 22.795246124267578, "learning_rate": 7.170714611825547e-06, "loss": 11.959615707397461, "loss_alignment": 0.95703125, "loss_alignment_w": 0.478515625, "loss_sft": 0.26967751421034336, "loss_total_v6": 0.7474759891629219, "step": 2985 }, { "epoch": 0.688663063063063, "grad_norm": 22.736644744873047, "learning_rate": 7.168900278811243e-06, "loss": 12.206962585449219, "loss_alignment": 0.9775390625, "loss_alignment_w": 0.48876953125, "loss_sft": 0.2741808835417032, "loss_total_v6": 0.7629351541399956, "step": 2986 }, { "epoch": 0.6888936936936937, "grad_norm": 22.785526275634766, "learning_rate": 7.167085593952964e-06, "loss": 12.715234756469727, "loss_alignment": 0.95947265625, "loss_alignment_w": 0.479736328125, "loss_sft": 0.3162170648574829, "loss_total_v6": 0.7947021424770355, "step": 2987 }, { "epoch": 0.6891243243243244, "grad_norm": 22.548810958862305, "learning_rate": 7.165270557545094e-06, "loss": 12.568541526794434, "loss_alignment": 0.9736328125, "loss_alignment_w": 0.48681640625, "loss_sft": 0.29865637235343456, "loss_total_v6": 0.7855338081717491, "step": 2988 }, { "epoch": 0.6893549549549549, "grad_norm": 24.150146484375, "learning_rate": 7.163455169882068e-06, "loss": 12.272151947021484, "loss_alignment": 0.97412109375, "loss_alignment_w": 0.487060546875, "loss_sft": 0.28033043816685677, "loss_total_v6": 0.767009548842907, "step": 2989 }, { "epoch": 0.6895855855855856, "grad_norm": 20.483747482299805, "learning_rate": 7.161639431258387e-06, "loss": 12.430815696716309, "loss_alignment": 0.9833984375, "loss_alignment_w": 0.49169921875, "loss_sft": 0.285898145288229, "loss_total_v6": 0.7769259810447693, "step": 2990 }, { "epoch": 0.6898162162162162, "grad_norm": 21.669307708740234, "learning_rate": 7.159823341968601e-06, "loss": 12.39935302734375, "loss_alignment": 0.97900390625, "loss_alignment_w": 0.489501953125, "loss_sft": 0.284282686188817, "loss_total_v6": 0.774959571659565, "step": 2991 }, { "epoch": 0.6900468468468468, "grad_norm": 21.69596290588379, "learning_rate": 7.158006902307322e-06, "loss": 12.390398025512695, "loss_alignment": 0.978515625, "loss_alignment_w": 0.4892578125, "loss_sft": 0.28543200716376305, "loss_total_v6": 0.774399884045124, "step": 2992 }, { "epoch": 0.6902774774774775, "grad_norm": 22.629257202148438, "learning_rate": 7.156190112569215e-06, "loss": 12.437631607055664, "loss_alignment": 0.9765625, "loss_alignment_w": 0.48828125, "loss_sft": 0.28797210939228535, "loss_total_v6": 0.7773519828915596, "step": 2993 }, { "epoch": 0.6905081081081081, "grad_norm": 21.55914306640625, "learning_rate": 7.154372973049007e-06, "loss": 12.16578483581543, "loss_alignment": 0.97705078125, "loss_alignment_w": 0.488525390625, "loss_sft": 0.27327049151062965, "loss_total_v6": 0.7603615745902061, "step": 2994 }, { "epoch": 0.6907387387387387, "grad_norm": 20.779294967651367, "learning_rate": 7.1525554840414765e-06, "loss": 12.470213890075684, "loss_alignment": 0.9599609375, "loss_alignment_w": 0.47998046875, "loss_sft": 0.298995915800333, "loss_total_v6": 0.7793883681297302, "step": 2995 }, { "epoch": 0.6909693693693694, "grad_norm": 21.933046340942383, "learning_rate": 7.15073764584146e-06, "loss": 12.13336181640625, "loss_alignment": 0.96533203125, "loss_alignment_w": 0.482666015625, "loss_sft": 0.27525710314512253, "loss_total_v6": 0.7583351209759712, "step": 2996 }, { "epoch": 0.6912, "grad_norm": 23.80824851989746, "learning_rate": 7.148919458743853e-06, "loss": 12.60224437713623, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.2981230765581131, "loss_total_v6": 0.7876402735710144, "step": 2997 }, { "epoch": 0.6914306306306306, "grad_norm": 23.099763870239258, "learning_rate": 7.147100923043606e-06, "loss": 12.641767501831055, "loss_alignment": 0.9716796875, "loss_alignment_w": 0.48583984375, "loss_sft": 0.30362975038588047, "loss_total_v6": 0.7901104390621185, "step": 2998 }, { "epoch": 0.6916612612612613, "grad_norm": 21.684263229370117, "learning_rate": 7.1452820390357225e-06, "loss": 11.95109748840332, "loss_alignment": 0.98388671875, "loss_alignment_w": 0.491943359375, "loss_sft": 0.25414567440748215, "loss_total_v6": 0.7469435855746269, "step": 2999 }, { "epoch": 0.6918918918918919, "grad_norm": 22.934865951538086, "learning_rate": 7.143462807015271e-06, "loss": 12.357133865356445, "loss_alignment": 0.97119140625, "loss_alignment_w": 0.485595703125, "loss_sft": 0.2868625372648239, "loss_total_v6": 0.772320918738842, "step": 3000 }, { "epoch": 0.6918918918918919, "eval_loss": 0.9422372579574585, "eval_loss_alignment": 0.9797842644121004, "eval_loss_alignment_w": 0.4898921322060502, "eval_loss_contrast": 0.0, "eval_loss_highlevel": 0.0, "eval_loss_sft": 0.4523450953707303, "eval_loss_total_v6": 0.9422372244468563, "eval_loss_wm_recon": 0.0, "eval_loss_wm_recon_w": 0.0, "eval_runtime": 348.6373, "eval_samples_per_second": 5.005, "eval_steps_per_second": 0.628, "step": 3000 }, { "epoch": 0.6921225225225225, "grad_norm": 22.06238555908203, "learning_rate": 7.141643227277371e-06, "loss": 12.557337760925293, "loss_alignment": 0.947265625, "loss_alignment_w": 0.4736328125, "loss_sft": 0.3104988969862461, "loss_total_v6": 0.784833624958992, "step": 3001 }, { "epoch": 0.6923531531531532, "grad_norm": 23.368316650390625, "learning_rate": 7.139823300117194e-06, "loss": 11.599892616271973, "loss_alignment": 0.9619140625, "loss_alignment_w": 0.48095703125, "loss_sft": 0.24244937859475613, "loss_total_v6": 0.7249933108687401, "step": 3002 }, { "epoch": 0.6925837837837838, "grad_norm": 21.530254364013672, "learning_rate": 7.1380030258299775e-06, "loss": 11.832895278930664, "loss_alignment": 0.9619140625, "loss_alignment_w": 0.48095703125, "loss_sft": 0.2582327499985695, "loss_total_v6": 0.7395559847354889, "step": 3003 }, { "epoch": 0.6928144144144144, "grad_norm": 23.32497787475586, "learning_rate": 7.1361824047110084e-06, "loss": 13.051804542541504, "loss_alignment": 0.970703125, "loss_alignment_w": 0.4853515625, "loss_sft": 0.330233633518219, "loss_total_v6": 0.8157377764582634, "step": 3004 }, { "epoch": 0.693045045045045, "grad_norm": 21.705955505371094, "learning_rate": 7.134361437055633e-06, "loss": 12.268756866455078, "loss_alignment": 0.97705078125, "loss_alignment_w": 0.488525390625, "loss_sft": 0.2779056951403618, "loss_total_v6": 0.766797311604023, "step": 3005 }, { "epoch": 0.6932756756756757, "grad_norm": 26.59788703918457, "learning_rate": 7.132540123159253e-06, "loss": 12.13215446472168, "loss_alignment": 0.96435546875, "loss_alignment_w": 0.482177734375, "loss_sft": 0.27412876300513744, "loss_total_v6": 0.758259616792202, "step": 3006 }, { "epoch": 0.6935063063063063, "grad_norm": 23.653018951416016, "learning_rate": 7.1307184633173256e-06, "loss": 12.42025375366211, "loss_alignment": 0.962890625, "loss_alignment_w": 0.4814453125, "loss_sft": 0.2954614423215389, "loss_total_v6": 0.776265874505043, "step": 3007 }, { "epoch": 0.6937369369369369, "grad_norm": 23.675052642822266, "learning_rate": 7.128896457825364e-06, "loss": 12.008968353271484, "loss_alignment": 0.9736328125, "loss_alignment_w": 0.48681640625, "loss_sft": 0.264049232006073, "loss_total_v6": 0.7505604848265648, "step": 3008 }, { "epoch": 0.6939675675675676, "grad_norm": 25.30384635925293, "learning_rate": 7.127074106978939e-06, "loss": 12.446039199829102, "loss_alignment": 0.9501953125, "loss_alignment_w": 0.47509765625, "loss_sft": 0.3028713706880808, "loss_total_v6": 0.7778774574398994, "step": 3009 }, { "epoch": 0.6941981981981982, "grad_norm": 24.192827224731445, "learning_rate": 7.125251411073677e-06, "loss": 12.133371353149414, "loss_alignment": 0.9560546875, "loss_alignment_w": 0.47802734375, "loss_sft": 0.2795912176370621, "loss_total_v6": 0.7583357393741608, "step": 3010 }, { "epoch": 0.6944288288288288, "grad_norm": 21.785404205322266, "learning_rate": 7.123428370405258e-06, "loss": 12.314878463745117, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.27500519528985023, "loss_total_v6": 0.7696798741817474, "step": 3011 }, { "epoch": 0.6946594594594595, "grad_norm": 22.454936981201172, "learning_rate": 7.121604985269423e-06, "loss": 12.214970588684082, "loss_alignment": 0.9658203125, "loss_alignment_w": 0.48291015625, "loss_sft": 0.28131894767284393, "loss_total_v6": 0.7634356543421745, "step": 3012 }, { "epoch": 0.6948900900900901, "grad_norm": 21.886606216430664, "learning_rate": 7.119781255961963e-06, "loss": 11.780338287353516, "loss_alignment": 0.9501953125, "loss_alignment_w": 0.47509765625, "loss_sft": 0.2629435248672962, "loss_total_v6": 0.7362711802124977, "step": 3013 }, { "epoch": 0.6951207207207207, "grad_norm": 25.26970672607422, "learning_rate": 7.1179571827787275e-06, "loss": 12.155762672424316, "loss_alignment": 0.978515625, "loss_alignment_w": 0.4892578125, "loss_sft": 0.2699585501104593, "loss_total_v6": 0.7597351521253586, "step": 3014 }, { "epoch": 0.6953513513513514, "grad_norm": 23.92538833618164, "learning_rate": 7.116132766015622e-06, "loss": 12.096494674682617, "loss_alignment": 0.96240234375, "loss_alignment_w": 0.481201171875, "loss_sft": 0.2750891223549843, "loss_total_v6": 0.7560309022665024, "step": 3015 }, { "epoch": 0.695581981981982, "grad_norm": 23.032785415649414, "learning_rate": 7.11430800596861e-06, "loss": 12.92476749420166, "loss_alignment": 0.96728515625, "loss_alignment_w": 0.483642578125, "loss_sft": 0.32337718829512596, "loss_total_v6": 0.8077979683876038, "step": 3016 }, { "epoch": 0.6958126126126126, "grad_norm": 23.48740577697754, "learning_rate": 7.1124829029337066e-06, "loss": 11.893014907836914, "loss_alignment": 0.9453125, "loss_alignment_w": 0.47265625, "loss_sft": 0.27114550210535526, "loss_total_v6": 0.7433134615421295, "step": 3017 }, { "epoch": 0.6960432432432433, "grad_norm": 24.433996200561523, "learning_rate": 7.110657457206982e-06, "loss": 12.97325325012207, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.3212806098163128, "loss_total_v6": 0.8108283430337906, "step": 3018 }, { "epoch": 0.6962738738738738, "grad_norm": 22.461835861206055, "learning_rate": 7.108831669084569e-06, "loss": 12.158482551574707, "loss_alignment": 0.94873046875, "loss_alignment_w": 0.474365234375, "loss_sft": 0.2854789234697819, "loss_total_v6": 0.7599051892757416, "step": 3019 }, { "epoch": 0.6965045045045045, "grad_norm": 24.848417282104492, "learning_rate": 7.107005538862647e-06, "loss": 12.512659072875977, "loss_alignment": 0.939453125, "loss_alignment_w": 0.4697265625, "loss_sft": 0.31225357577204704, "loss_total_v6": 0.7820411697030067, "step": 3020 }, { "epoch": 0.6967351351351352, "grad_norm": 21.91623306274414, "learning_rate": 7.105179066837456e-06, "loss": 11.985949516296387, "loss_alignment": 0.96337890625, "loss_alignment_w": 0.481689453125, "loss_sft": 0.2665931712836027, "loss_total_v6": 0.749121867120266, "step": 3021 }, { "epoch": 0.6969657657657657, "grad_norm": 22.787729263305664, "learning_rate": 7.103352253305291e-06, "loss": 11.966768264770508, "loss_alignment": 0.9736328125, "loss_alignment_w": 0.48681640625, "loss_sft": 0.2614269945770502, "loss_total_v6": 0.7479229494929314, "step": 3022 }, { "epoch": 0.6971963963963964, "grad_norm": 21.445858001708984, "learning_rate": 7.101525098562503e-06, "loss": 12.159881591796875, "loss_alignment": 0.97900390625, "loss_alignment_w": 0.489501953125, "loss_sft": 0.2709178775548935, "loss_total_v6": 0.7599925845861435, "step": 3023 }, { "epoch": 0.6974270270270271, "grad_norm": 22.63262176513672, "learning_rate": 7.099697602905494e-06, "loss": 12.229948043823242, "loss_alignment": 0.9326171875, "loss_alignment_w": 0.46630859375, "loss_sft": 0.29841412231326103, "loss_total_v6": 0.7643717601895332, "step": 3024 }, { "epoch": 0.6976576576576576, "grad_norm": 23.071393966674805, "learning_rate": 7.09786976663073e-06, "loss": 11.841169357299805, "loss_alignment": 0.98486328125, "loss_alignment_w": 0.492431640625, "loss_sft": 0.24848063476383686, "loss_total_v6": 0.7400730475783348, "step": 3025 }, { "epoch": 0.6978882882882883, "grad_norm": 28.034629821777344, "learning_rate": 7.0960415900347205e-06, "loss": 12.49581241607666, "loss_alignment": 0.966796875, "loss_alignment_w": 0.4833984375, "loss_sft": 0.296582767739892, "loss_total_v6": 0.7809882834553719, "step": 3026 }, { "epoch": 0.698118918918919, "grad_norm": 22.205434799194336, "learning_rate": 7.094213073414039e-06, "loss": 12.015334129333496, "loss_alignment": 0.9677734375, "loss_alignment_w": 0.48388671875, "loss_sft": 0.266873262822628, "loss_total_v6": 0.7509583458304405, "step": 3027 }, { "epoch": 0.6983495495495495, "grad_norm": 24.383485794067383, "learning_rate": 7.092384217065315e-06, "loss": 12.041488647460938, "loss_alignment": 0.93896484375, "loss_alignment_w": 0.469482421875, "loss_sft": 0.2819509208202362, "loss_total_v6": 0.7525930181145668, "step": 3028 }, { "epoch": 0.6985801801801802, "grad_norm": 22.683481216430664, "learning_rate": 7.090555021285224e-06, "loss": 12.688960075378418, "loss_alignment": 0.97265625, "loss_alignment_w": 0.486328125, "loss_sft": 0.30659456364810467, "loss_total_v6": 0.7930600196123123, "step": 3029 }, { "epoch": 0.6988108108108108, "grad_norm": 24.208818435668945, "learning_rate": 7.088725486370506e-06, "loss": 11.892184257507324, "loss_alignment": 0.9423828125, "loss_alignment_w": 0.47119140625, "loss_sft": 0.2721768952906132, "loss_total_v6": 0.7432615011930466, "step": 3030 }, { "epoch": 0.6990414414414414, "grad_norm": 24.24894142150879, "learning_rate": 7.086895612617951e-06, "loss": 12.174095153808594, "loss_alignment": 0.9755859375, "loss_alignment_w": 0.48779296875, "loss_sft": 0.2714705765247345, "loss_total_v6": 0.7608809918165207, "step": 3031 }, { "epoch": 0.6992720720720721, "grad_norm": 23.56214141845703, "learning_rate": 7.085065400324407e-06, "loss": 12.011087417602539, "loss_alignment": 0.9501953125, "loss_alignment_w": 0.47509765625, "loss_sft": 0.2777773439884186, "loss_total_v6": 0.7506929934024811, "step": 3032 }, { "epoch": 0.6995027027027027, "grad_norm": 25.609710693359375, "learning_rate": 7.083234849786773e-06, "loss": 12.019513130187988, "loss_alignment": 0.95263671875, "loss_alignment_w": 0.476318359375, "loss_sft": 0.273909367620945, "loss_total_v6": 0.7512195631861687, "step": 3033 }, { "epoch": 0.6997333333333333, "grad_norm": 21.431480407714844, "learning_rate": 7.081403961302007e-06, "loss": 12.408376693725586, "loss_alignment": 0.9638671875, "loss_alignment_w": 0.48193359375, "loss_sft": 0.29462752863764763, "loss_total_v6": 0.7755235135555267, "step": 3034 }, { "epoch": 0.699963963963964, "grad_norm": 24.240076065063477, "learning_rate": 7.079572735167119e-06, "loss": 13.287708282470703, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.33353349566459656, "loss_total_v6": 0.8304817527532578, "step": 3035 }, { "epoch": 0.7001945945945945, "grad_norm": 23.05207633972168, "learning_rate": 7.0777411716791734e-06, "loss": 12.233288764953613, "loss_alignment": 0.9716796875, "loss_alignment_w": 0.48583984375, "loss_sft": 0.2780540566891432, "loss_total_v6": 0.7645805403590202, "step": 3036 }, { "epoch": 0.7004252252252252, "grad_norm": 22.457624435424805, "learning_rate": 7.075909271135293e-06, "loss": 12.015735626220703, "loss_alignment": 0.9833984375, "loss_alignment_w": 0.49169921875, "loss_sft": 0.2584755811840296, "loss_total_v6": 0.7509834915399551, "step": 3037 }, { "epoch": 0.7006558558558559, "grad_norm": 25.130842208862305, "learning_rate": 7.074077033832653e-06, "loss": 12.011987686157227, "loss_alignment": 0.95703125, "loss_alignment_w": 0.478515625, "loss_sft": 0.27191311679780483, "loss_total_v6": 0.7507492080330849, "step": 3038 }, { "epoch": 0.7008864864864864, "grad_norm": 22.00851058959961, "learning_rate": 7.072244460068481e-06, "loss": 12.224206924438477, "loss_alignment": 0.9599609375, "loss_alignment_w": 0.47998046875, "loss_sft": 0.2841087058186531, "loss_total_v6": 0.764012910425663, "step": 3039 }, { "epoch": 0.7011171171171171, "grad_norm": 24.9760684967041, "learning_rate": 7.07041155014006e-06, "loss": 12.813912391662598, "loss_alignment": 0.99267578125, "loss_alignment_w": 0.496337890625, "loss_sft": 0.3030667956918478, "loss_total_v6": 0.800869531929493, "step": 3040 }, { "epoch": 0.7013477477477478, "grad_norm": 21.3630313873291, "learning_rate": 7.068578304344734e-06, "loss": 12.151166915893555, "loss_alignment": 0.98486328125, "loss_alignment_w": 0.492431640625, "loss_sft": 0.26741297356784344, "loss_total_v6": 0.7594478875398636, "step": 3041 }, { "epoch": 0.7015783783783783, "grad_norm": 20.779373168945312, "learning_rate": 7.066744722979893e-06, "loss": 12.237957954406738, "loss_alignment": 0.958984375, "loss_alignment_w": 0.4794921875, "loss_sft": 0.284907104447484, "loss_total_v6": 0.7648723423480988, "step": 3042 }, { "epoch": 0.701809009009009, "grad_norm": 24.562166213989258, "learning_rate": 7.0649108063429835e-06, "loss": 12.645246505737305, "loss_alignment": 0.9833984375, "loss_alignment_w": 0.49169921875, "loss_sft": 0.2980488818138838, "loss_total_v6": 0.7903279215097427, "step": 3043 }, { "epoch": 0.7020396396396397, "grad_norm": 22.586273193359375, "learning_rate": 7.0630765547315125e-06, "loss": 11.759679794311523, "loss_alignment": 0.966796875, "loss_alignment_w": 0.4833984375, "loss_sft": 0.25155098363757133, "loss_total_v6": 0.7349799498915672, "step": 3044 }, { "epoch": 0.7022702702702702, "grad_norm": 23.89826202392578, "learning_rate": 7.061241968443033e-06, "loss": 12.724616050720215, "loss_alignment": 0.958984375, "loss_alignment_w": 0.4794921875, "loss_sft": 0.3153996206820011, "loss_total_v6": 0.795288510620594, "step": 3045 }, { "epoch": 0.7025009009009009, "grad_norm": 21.711809158325195, "learning_rate": 7.059407047775154e-06, "loss": 12.322427749633789, "loss_alignment": 0.9873046875, "loss_alignment_w": 0.49365234375, "loss_sft": 0.2761026341468096, "loss_total_v6": 0.7701517194509506, "step": 3046 }, { "epoch": 0.7027315315315316, "grad_norm": 22.130271911621094, "learning_rate": 7.057571793025545e-06, "loss": 11.975666046142578, "loss_alignment": 0.9541015625, "loss_alignment_w": 0.47705078125, "loss_sft": 0.27155046351253986, "loss_total_v6": 0.7484791800379753, "step": 3047 }, { "epoch": 0.7029621621621621, "grad_norm": 24.32940673828125, "learning_rate": 7.055736204491922e-06, "loss": 12.715757369995117, "loss_alignment": 0.97021484375, "loss_alignment_w": 0.485107421875, "loss_sft": 0.3093984965234995, "loss_total_v6": 0.7947348058223724, "step": 3048 }, { "epoch": 0.7031927927927928, "grad_norm": 22.783924102783203, "learning_rate": 7.053900282472059e-06, "loss": 11.883438110351562, "loss_alignment": 0.9638671875, "loss_alignment_w": 0.48193359375, "loss_sft": 0.26075075939297676, "loss_total_v6": 0.7427148818969727, "step": 3049 }, { "epoch": 0.7034234234234235, "grad_norm": 23.26387596130371, "learning_rate": 7.052064027263785e-06, "loss": 12.012956619262695, "loss_alignment": 0.94091796875, "loss_alignment_w": 0.470458984375, "loss_sft": 0.2804271373897791, "loss_total_v6": 0.7508098408579826, "step": 3050 }, { "epoch": 0.703654054054054, "grad_norm": 21.558834075927734, "learning_rate": 7.050227439164979e-06, "loss": 12.63692855834961, "loss_alignment": 0.96728515625, "loss_alignment_w": 0.483642578125, "loss_sft": 0.3056619241833687, "loss_total_v6": 0.7898080796003342, "step": 3051 }, { "epoch": 0.7038846846846847, "grad_norm": 22.46664810180664, "learning_rate": 7.048390518473579e-06, "loss": 12.486461639404297, "loss_alignment": 0.9560546875, "loss_alignment_w": 0.47802734375, "loss_sft": 0.3019187208265066, "loss_total_v6": 0.7804038152098656, "step": 3052 }, { "epoch": 0.7041153153153153, "grad_norm": 22.11797523498535, "learning_rate": 7.046553265487573e-06, "loss": 12.154051780700684, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.26286308094859123, "loss_total_v6": 0.7596282139420509, "step": 3053 }, { "epoch": 0.7043459459459459, "grad_norm": 23.004310607910156, "learning_rate": 7.044715680505005e-06, "loss": 12.125738143920898, "loss_alignment": 0.95703125, "loss_alignment_w": 0.478515625, "loss_sft": 0.27873264253139496, "loss_total_v6": 0.7578586265444756, "step": 3054 }, { "epoch": 0.7045765765765766, "grad_norm": 23.39501953125, "learning_rate": 7.042877763823973e-06, "loss": 13.029764175415039, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.32635362818837166, "loss_total_v6": 0.8143602460622787, "step": 3055 }, { "epoch": 0.7048072072072072, "grad_norm": 25.20073699951172, "learning_rate": 7.041039515742626e-06, "loss": 12.244766235351562, "loss_alignment": 0.96826171875, "loss_alignment_w": 0.484130859375, "loss_sft": 0.2821436356753111, "loss_total_v6": 0.765297919511795, "step": 3056 }, { "epoch": 0.7050378378378378, "grad_norm": 21.990680694580078, "learning_rate": 7.039200936559171e-06, "loss": 12.681232452392578, "loss_alignment": 0.96630859375, "loss_alignment_w": 0.483154296875, "loss_sft": 0.30956011079251766, "loss_total_v6": 0.7925770580768585, "step": 3057 }, { "epoch": 0.7052684684684685, "grad_norm": 23.838422775268555, "learning_rate": 7.037362026571863e-06, "loss": 12.054688453674316, "loss_alignment": 0.98095703125, "loss_alignment_w": 0.490478515625, "loss_sft": 0.2637482024729252, "loss_total_v6": 0.7534180134534836, "step": 3058 }, { "epoch": 0.7054990990990991, "grad_norm": 24.201061248779297, "learning_rate": 7.035522786079019e-06, "loss": 12.784841537475586, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.3043626546859741, "loss_total_v6": 0.7990525737404823, "step": 3059 }, { "epoch": 0.7057297297297297, "grad_norm": 22.549922943115234, "learning_rate": 7.033683215379002e-06, "loss": 12.218453407287598, "loss_alignment": 0.9609375, "loss_alignment_w": 0.48046875, "loss_sft": 0.2822232600301504, "loss_total_v6": 0.7636533156037331, "step": 3060 }, { "epoch": 0.7059603603603604, "grad_norm": 24.417217254638672, "learning_rate": 7.031843314770233e-06, "loss": 12.009868621826172, "loss_alignment": 0.962890625, "loss_alignment_w": 0.4814453125, "loss_sft": 0.2698581013828516, "loss_total_v6": 0.7506168112158775, "step": 3061 }, { "epoch": 0.706190990990991, "grad_norm": 22.040210723876953, "learning_rate": 7.0300030845511815e-06, "loss": 11.890851974487305, "loss_alignment": 0.96142578125, "loss_alignment_w": 0.480712890625, "loss_sft": 0.2618244048207998, "loss_total_v6": 0.743178203701973, "step": 3062 }, { "epoch": 0.7064216216216216, "grad_norm": 23.825576782226562, "learning_rate": 7.0281625250203764e-06, "loss": 12.10482120513916, "loss_alignment": 0.9638671875, "loss_alignment_w": 0.48193359375, "loss_sft": 0.2749533951282501, "loss_total_v6": 0.7565513253211975, "step": 3063 }, { "epoch": 0.7066522522522523, "grad_norm": 23.291542053222656, "learning_rate": 7.026321636476398e-06, "loss": 12.099081039428711, "loss_alignment": 0.9501953125, "loss_alignment_w": 0.47509765625, "loss_sft": 0.28114067390561104, "loss_total_v6": 0.7561925649642944, "step": 3064 }, { "epoch": 0.7068828828828829, "grad_norm": 22.159454345703125, "learning_rate": 7.024480419217878e-06, "loss": 11.98023796081543, "loss_alignment": 0.95263671875, "loss_alignment_w": 0.476318359375, "loss_sft": 0.2722786758095026, "loss_total_v6": 0.7487648651003838, "step": 3065 }, { "epoch": 0.7071135135135135, "grad_norm": 23.138412475585938, "learning_rate": 7.022638873543504e-06, "loss": 12.21507453918457, "loss_alignment": 0.970703125, "loss_alignment_w": 0.4853515625, "loss_sft": 0.27799904718995094, "loss_total_v6": 0.7634421810507774, "step": 3066 }, { "epoch": 0.7073441441441441, "grad_norm": 22.035526275634766, "learning_rate": 7.020796999752016e-06, "loss": 12.480464935302734, "loss_alignment": 0.96240234375, "loss_alignment_w": 0.481201171875, "loss_sft": 0.29843120090663433, "loss_total_v6": 0.7800290957093239, "step": 3067 }, { "epoch": 0.7075747747747748, "grad_norm": 27.709177017211914, "learning_rate": 7.018954798142205e-06, "loss": 11.954586029052734, "loss_alignment": 0.95751953125, "loss_alignment_w": 0.478759765625, "loss_sft": 0.267120148986578, "loss_total_v6": 0.7471616491675377, "step": 3068 }, { "epoch": 0.7078054054054054, "grad_norm": 23.276187896728516, "learning_rate": 7.0171122690129186e-06, "loss": 12.512642860412598, "loss_alignment": 0.98046875, "loss_alignment_w": 0.490234375, "loss_sft": 0.29148536548018456, "loss_total_v6": 0.782040186226368, "step": 3069 }, { "epoch": 0.708036036036036, "grad_norm": 27.65260124206543, "learning_rate": 7.015269412663057e-06, "loss": 12.361743927001953, "loss_alignment": 0.96533203125, "loss_alignment_w": 0.482666015625, "loss_sft": 0.2898972313851118, "loss_total_v6": 0.7726090028882027, "step": 3070 }, { "epoch": 0.7082666666666667, "grad_norm": 21.72434425354004, "learning_rate": 7.0134262293915704e-06, "loss": 12.374566078186035, "loss_alignment": 0.9765625, "loss_alignment_w": 0.48828125, "loss_sft": 0.285800576210022, "loss_total_v6": 0.7734103798866272, "step": 3071 }, { "epoch": 0.7084972972972973, "grad_norm": 21.725194931030273, "learning_rate": 7.011582719497466e-06, "loss": 12.031438827514648, "loss_alignment": 0.9541015625, "loss_alignment_w": 0.47705078125, "loss_sft": 0.274776853621006, "loss_total_v6": 0.7519649341702461, "step": 3072 }, { "epoch": 0.7087279279279279, "grad_norm": 22.93776512145996, "learning_rate": 7.009738883279802e-06, "loss": 12.567106246948242, "loss_alignment": 0.966796875, "loss_alignment_w": 0.4833984375, "loss_sft": 0.30192364379763603, "loss_total_v6": 0.7854441478848457, "step": 3073 }, { "epoch": 0.7089585585585586, "grad_norm": 20.254858016967773, "learning_rate": 7.007894721037688e-06, "loss": 12.249241828918457, "loss_alignment": 0.97216796875, "loss_alignment_w": 0.486083984375, "loss_sft": 0.2789748031646013, "loss_total_v6": 0.7655775919556618, "step": 3074 }, { "epoch": 0.7091891891891892, "grad_norm": 23.098142623901367, "learning_rate": 7.006050233070289e-06, "loss": 12.156160354614258, "loss_alignment": 0.98583984375, "loss_alignment_w": 0.492919921875, "loss_sft": 0.26661125011742115, "loss_total_v6": 0.7597600519657135, "step": 3075 }, { "epoch": 0.7094198198198198, "grad_norm": 23.119035720825195, "learning_rate": 7.004205419676826e-06, "loss": 12.551406860351562, "loss_alignment": 0.943359375, "loss_alignment_w": 0.4716796875, "loss_sft": 0.3119745757430792, "loss_total_v6": 0.7844629809260368, "step": 3076 }, { "epoch": 0.7096504504504505, "grad_norm": 24.550594329833984, "learning_rate": 7.0023602811565625e-06, "loss": 12.3627347946167, "loss_alignment": 0.96337890625, "loss_alignment_w": 0.481689453125, "loss_sft": 0.29052373208105564, "loss_total_v6": 0.7726709321141243, "step": 3077 }, { "epoch": 0.7098810810810811, "grad_norm": 23.061187744140625, "learning_rate": 7.000514817808824e-06, "loss": 11.907501220703125, "loss_alignment": 0.96826171875, "loss_alignment_w": 0.484130859375, "loss_sft": 0.2598743084818125, "loss_total_v6": 0.7442187815904617, "step": 3078 }, { "epoch": 0.7101117117117117, "grad_norm": 22.064716339111328, "learning_rate": 6.998669029932989e-06, "loss": 11.733073234558105, "loss_alignment": 0.95947265625, "loss_alignment_w": 0.479736328125, "loss_sft": 0.2551218830049038, "loss_total_v6": 0.7333170622587204, "step": 3079 }, { "epoch": 0.7103423423423424, "grad_norm": 24.88426399230957, "learning_rate": 6.9968229178284775e-06, "loss": 11.75634479522705, "loss_alignment": 0.9521484375, "loss_alignment_w": 0.47607421875, "loss_sft": 0.25778177566826344, "loss_total_v6": 0.7347715198993683, "step": 3080 }, { "epoch": 0.710572972972973, "grad_norm": 23.643218994140625, "learning_rate": 6.994976481794778e-06, "loss": 11.883451461791992, "loss_alignment": 0.93505859375, "loss_alignment_w": 0.467529296875, "loss_sft": 0.27575102634727955, "loss_total_v6": 0.7427157461643219, "step": 3081 }, { "epoch": 0.7108036036036036, "grad_norm": 24.191781997680664, "learning_rate": 6.993129722131421e-06, "loss": 11.96574592590332, "loss_alignment": 0.95068359375, "loss_alignment_w": 0.475341796875, "loss_sft": 0.2740126810967922, "loss_total_v6": 0.7478591427206993, "step": 3082 }, { "epoch": 0.7110342342342343, "grad_norm": 23.641155242919922, "learning_rate": 6.991282639137988e-06, "loss": 12.293335914611816, "loss_alignment": 0.96337890625, "loss_alignment_w": 0.481689453125, "loss_sft": 0.2869950272142887, "loss_total_v6": 0.7683335021138191, "step": 3083 }, { "epoch": 0.7112648648648648, "grad_norm": 23.66542625427246, "learning_rate": 6.989435233114124e-06, "loss": 11.954275131225586, "loss_alignment": 0.96240234375, "loss_alignment_w": 0.481201171875, "loss_sft": 0.26601734571158886, "loss_total_v6": 0.7471422329545021, "step": 3084 }, { "epoch": 0.7114954954954955, "grad_norm": 24.332265853881836, "learning_rate": 6.987587504359514e-06, "loss": 12.39927864074707, "loss_alignment": 0.966796875, "loss_alignment_w": 0.4833984375, "loss_sft": 0.29041207768023014, "loss_total_v6": 0.7749549150466919, "step": 3085 }, { "epoch": 0.7117261261261262, "grad_norm": 22.668821334838867, "learning_rate": 6.985739453173903e-06, "loss": 12.300846099853516, "loss_alignment": 0.9521484375, "loss_alignment_w": 0.47607421875, "loss_sft": 0.292103061452508, "loss_total_v6": 0.7688028812408447, "step": 3086 }, { "epoch": 0.7119567567567567, "grad_norm": 23.13733673095703, "learning_rate": 6.983891079857088e-06, "loss": 12.265680313110352, "loss_alignment": 0.96630859375, "loss_alignment_w": 0.483154296875, "loss_sft": 0.2832218613475561, "loss_total_v6": 0.7666050419211388, "step": 3087 }, { "epoch": 0.7121873873873874, "grad_norm": 24.96211814880371, "learning_rate": 6.982042384708911e-06, "loss": 12.402462005615234, "loss_alignment": 0.97802734375, "loss_alignment_w": 0.489013671875, "loss_sft": 0.28760503232479095, "loss_total_v6": 0.7751538902521133, "step": 3088 }, { "epoch": 0.7124180180180181, "grad_norm": 23.103384017944336, "learning_rate": 6.980193368029276e-06, "loss": 11.924989700317383, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.2576409373432398, "loss_total_v6": 0.7453118413686752, "step": 3089 }, { "epoch": 0.7126486486486486, "grad_norm": 24.668806076049805, "learning_rate": 6.978344030118134e-06, "loss": 12.536080360412598, "loss_alignment": 0.95068359375, "loss_alignment_w": 0.475341796875, "loss_sft": 0.3084531631320715, "loss_total_v6": 0.7835050523281097, "step": 3090 }, { "epoch": 0.7128792792792793, "grad_norm": 22.85938262939453, "learning_rate": 6.976494371275488e-06, "loss": 12.791722297668457, "loss_alignment": 0.982421875, "loss_alignment_w": 0.4912109375, "loss_sft": 0.30775286816060543, "loss_total_v6": 0.7994826212525368, "step": 3091 }, { "epoch": 0.7131099099099099, "grad_norm": 21.393234252929688, "learning_rate": 6.974644391801395e-06, "loss": 12.268392562866211, "loss_alignment": 0.9658203125, "loss_alignment_w": 0.48291015625, "loss_sft": 0.28445952013134956, "loss_total_v6": 0.7667745649814606, "step": 3092 }, { "epoch": 0.7133405405405405, "grad_norm": 23.658174514770508, "learning_rate": 6.972794091995961e-06, "loss": 11.837614059448242, "loss_alignment": 0.974609375, "loss_alignment_w": 0.4873046875, "loss_sft": 0.25227151811122894, "loss_total_v6": 0.7398508787155151, "step": 3093 }, { "epoch": 0.7135711711711712, "grad_norm": 22.42273712158203, "learning_rate": 6.970943472159349e-06, "loss": 11.926834106445312, "loss_alignment": 0.95947265625, "loss_alignment_w": 0.479736328125, "loss_sft": 0.26569079607725143, "loss_total_v6": 0.7454271167516708, "step": 3094 }, { "epoch": 0.7138018018018017, "grad_norm": 23.129236221313477, "learning_rate": 6.969092532591767e-06, "loss": 12.341115951538086, "loss_alignment": 0.96875, "loss_alignment_w": 0.484375, "loss_sft": 0.28645641542971134, "loss_total_v6": 0.771319717168808, "step": 3095 }, { "epoch": 0.7140324324324324, "grad_norm": 21.38312339782715, "learning_rate": 6.9672412735934805e-06, "loss": 11.734310150146484, "loss_alignment": 0.9404296875, "loss_alignment_w": 0.47021484375, "loss_sft": 0.2616231646388769, "loss_total_v6": 0.7333943918347359, "step": 3096 }, { "epoch": 0.7142630630630631, "grad_norm": 23.412906646728516, "learning_rate": 6.965389695464805e-06, "loss": 12.360172271728516, "loss_alignment": 0.98388671875, "loss_alignment_w": 0.491943359375, "loss_sft": 0.2808115780353546, "loss_total_v6": 0.7725107818841934, "step": 3097 }, { "epoch": 0.7144936936936936, "grad_norm": 22.045183181762695, "learning_rate": 6.963537798506108e-06, "loss": 12.094761848449707, "loss_alignment": 0.95458984375, "loss_alignment_w": 0.477294921875, "loss_sft": 0.2777273990213871, "loss_total_v6": 0.7559226080775261, "step": 3098 }, { "epoch": 0.7147243243243243, "grad_norm": 24.028034210205078, "learning_rate": 6.961685583017808e-06, "loss": 12.265405654907227, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.27229465171694756, "loss_total_v6": 0.7665878906846046, "step": 3099 }, { "epoch": 0.714954954954955, "grad_norm": 24.217391967773438, "learning_rate": 6.959833049300376e-06, "loss": 12.203730583190918, "loss_alignment": 0.97021484375, "loss_alignment_w": 0.485107421875, "loss_sft": 0.2790906075388193, "loss_total_v6": 0.7627331838011742, "step": 3100 }, { "epoch": 0.7151855855855855, "grad_norm": 23.854198455810547, "learning_rate": 6.957980197654335e-06, "loss": 12.438060760498047, "loss_alignment": 0.9853515625, "loss_alignment_w": 0.49267578125, "loss_sft": 0.2848861366510391, "loss_total_v6": 0.7773788124322891, "step": 3101 }, { "epoch": 0.7154162162162162, "grad_norm": 21.7246150970459, "learning_rate": 6.9561270283802555e-06, "loss": 12.289554595947266, "loss_alignment": 0.95703125, "loss_alignment_w": 0.478515625, "loss_sft": 0.28950523398816586, "loss_total_v6": 0.7680971398949623, "step": 3102 }, { "epoch": 0.7156468468468469, "grad_norm": 23.56899642944336, "learning_rate": 6.954273541778767e-06, "loss": 11.725963592529297, "loss_alignment": 0.9443359375, "loss_alignment_w": 0.47216796875, "loss_sft": 0.25840066373348236, "loss_total_v6": 0.7328727096319199, "step": 3103 }, { "epoch": 0.7158774774774774, "grad_norm": 25.22998046875, "learning_rate": 6.952419738150546e-06, "loss": 12.20018196105957, "loss_alignment": 0.94921875, "loss_alignment_w": 0.474609375, "loss_sft": 0.28753582015633583, "loss_total_v6": 0.7625113874673843, "step": 3104 }, { "epoch": 0.7161081081081081, "grad_norm": 27.64411735534668, "learning_rate": 6.950565617796317e-06, "loss": 12.03005313873291, "loss_alignment": 0.9306640625, "loss_alignment_w": 0.46533203125, "loss_sft": 0.2858443930745125, "loss_total_v6": 0.7518783435225487, "step": 3105 }, { "epoch": 0.7163387387387388, "grad_norm": 21.420202255249023, "learning_rate": 6.948711181016863e-06, "loss": 11.947040557861328, "loss_alignment": 0.96142578125, "loss_alignment_w": 0.480712890625, "loss_sft": 0.2664807066321373, "loss_total_v6": 0.7466900423169136, "step": 3106 }, { "epoch": 0.7165693693693693, "grad_norm": 22.628385543823242, "learning_rate": 6.946856428113014e-06, "loss": 12.378562927246094, "loss_alignment": 0.9560546875, "loss_alignment_w": 0.47802734375, "loss_sft": 0.2960600517690182, "loss_total_v6": 0.7736601606011391, "step": 3107 }, { "epoch": 0.7168, "grad_norm": 22.835859298706055, "learning_rate": 6.9450013593856515e-06, "loss": 11.475872039794922, "loss_alignment": 0.9609375, "loss_alignment_w": 0.48046875, "loss_sft": 0.23634606413543224, "loss_total_v6": 0.7172420471906662, "step": 3108 }, { "epoch": 0.7170306306306307, "grad_norm": 25.49479103088379, "learning_rate": 6.943145975135711e-06, "loss": 12.430508613586426, "loss_alignment": 0.96044921875, "loss_alignment_w": 0.480224609375, "loss_sft": 0.29637700878083706, "loss_total_v6": 0.776906780898571, "step": 3109 }, { "epoch": 0.7172612612612612, "grad_norm": 22.734336853027344, "learning_rate": 6.941290275664175e-06, "loss": 11.796455383300781, "loss_alignment": 0.9736328125, "loss_alignment_w": 0.48681640625, "loss_sft": 0.24994327500462532, "loss_total_v6": 0.73727847635746, "step": 3110 }, { "epoch": 0.7174918918918919, "grad_norm": 23.505701065063477, "learning_rate": 6.939434261272081e-06, "loss": 12.66150951385498, "loss_alignment": 0.953125, "loss_alignment_w": 0.4765625, "loss_sft": 0.3150870092213154, "loss_total_v6": 0.7913443148136139, "step": 3111 }, { "epoch": 0.7177225225225226, "grad_norm": 24.12978744506836, "learning_rate": 6.9375779322605154e-06, "loss": 12.505438804626465, "loss_alignment": 0.97021484375, "loss_alignment_w": 0.485107421875, "loss_sft": 0.2979778479784727, "loss_total_v6": 0.7815899178385735, "step": 3112 }, { "epoch": 0.7179531531531531, "grad_norm": 22.032596588134766, "learning_rate": 6.935721288930617e-06, "loss": 12.28945541381836, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.2788026798516512, "loss_total_v6": 0.768091008067131, "step": 3113 }, { "epoch": 0.7181837837837838, "grad_norm": 21.31465721130371, "learning_rate": 6.933864331583572e-06, "loss": 12.441232681274414, "loss_alignment": 0.95556640625, "loss_alignment_w": 0.477783203125, "loss_sft": 0.2987867332994938, "loss_total_v6": 0.7775770202279091, "step": 3114 }, { "epoch": 0.7184144144144144, "grad_norm": 23.698219299316406, "learning_rate": 6.932007060520622e-06, "loss": 11.421932220458984, "loss_alignment": 0.95068359375, "loss_alignment_w": 0.475341796875, "loss_sft": 0.23859001509845257, "loss_total_v6": 0.7138708010315895, "step": 3115 }, { "epoch": 0.718645045045045, "grad_norm": 24.760379791259766, "learning_rate": 6.930149476043059e-06, "loss": 11.785842895507812, "loss_alignment": 0.94677734375, "loss_alignment_w": 0.473388671875, "loss_sft": 0.26347067952156067, "loss_total_v6": 0.7366151735186577, "step": 3116 }, { "epoch": 0.7188756756756757, "grad_norm": 28.98386001586914, "learning_rate": 6.928291578452222e-06, "loss": 12.6649169921875, "loss_alignment": 0.96533203125, "loss_alignment_w": 0.482666015625, "loss_sft": 0.30918124690651894, "loss_total_v6": 0.7915573492646217, "step": 3117 }, { "epoch": 0.7191063063063063, "grad_norm": 25.137046813964844, "learning_rate": 6.926433368049506e-06, "loss": 11.878473281860352, "loss_alignment": 0.96875, "loss_alignment_w": 0.484375, "loss_sft": 0.2582279350608587, "loss_total_v6": 0.7424045652151108, "step": 3118 }, { "epoch": 0.7193369369369369, "grad_norm": 23.725881576538086, "learning_rate": 6.924574845136352e-06, "loss": 12.389781951904297, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.2875602785497904, "loss_total_v6": 0.7743613943457603, "step": 3119 }, { "epoch": 0.7195675675675676, "grad_norm": 21.632633209228516, "learning_rate": 6.922716010014256e-06, "loss": 12.014942169189453, "loss_alignment": 0.962890625, "loss_alignment_w": 0.4814453125, "loss_sft": 0.2685120142996311, "loss_total_v6": 0.7509338855743408, "step": 3120 }, { "epoch": 0.7197981981981982, "grad_norm": 22.089439392089844, "learning_rate": 6.920856862984759e-06, "loss": 12.666004180908203, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.2921897638589144, "loss_total_v6": 0.7916252166032791, "step": 3121 }, { "epoch": 0.7200288288288288, "grad_norm": 21.6865234375, "learning_rate": 6.91899740434946e-06, "loss": 12.060157775878906, "loss_alignment": 0.93798828125, "loss_alignment_w": 0.468994140625, "loss_sft": 0.28417063504457474, "loss_total_v6": 0.7537598833441734, "step": 3122 }, { "epoch": 0.7202594594594595, "grad_norm": 21.92119789123535, "learning_rate": 6.917137634410004e-06, "loss": 11.714753150939941, "loss_alignment": 0.9453125, "loss_alignment_w": 0.47265625, "loss_sft": 0.2592259254306555, "loss_total_v6": 0.7321720719337463, "step": 3123 }, { "epoch": 0.7204900900900901, "grad_norm": 23.616479873657227, "learning_rate": 6.915277553468083e-06, "loss": 12.079832077026367, "loss_alignment": 0.9443359375, "loss_alignment_w": 0.47216796875, "loss_sft": 0.28189077228307724, "loss_total_v6": 0.7549895346164703, "step": 3124 }, { "epoch": 0.7207207207207207, "grad_norm": 22.355915069580078, "learning_rate": 6.913417161825449e-06, "loss": 12.011798858642578, "loss_alignment": 0.9736328125, "loss_alignment_w": 0.48681640625, "loss_sft": 0.26450083404779434, "loss_total_v6": 0.7507374063134193, "step": 3125 }, { "epoch": 0.7209513513513514, "grad_norm": 23.305875778198242, "learning_rate": 6.911556459783898e-06, "loss": 11.89249324798584, "loss_alignment": 0.939453125, "loss_alignment_w": 0.4697265625, "loss_sft": 0.27381360717117786, "loss_total_v6": 0.7432807981967926, "step": 3126 }, { "epoch": 0.721181981981982, "grad_norm": 23.279991149902344, "learning_rate": 6.9096954476452724e-06, "loss": 12.126965522766113, "loss_alignment": 0.9775390625, "loss_alignment_w": 0.48876953125, "loss_sft": 0.27011181600391865, "loss_total_v6": 0.7579353377223015, "step": 3127 }, { "epoch": 0.7214126126126126, "grad_norm": 22.769027709960938, "learning_rate": 6.9078341257114765e-06, "loss": 12.257964134216309, "loss_alignment": 0.962890625, "loss_alignment_w": 0.4814453125, "loss_sft": 0.2848300226032734, "loss_total_v6": 0.7661227732896805, "step": 3128 }, { "epoch": 0.7216432432432432, "grad_norm": 22.29218101501465, "learning_rate": 6.905972494284456e-06, "loss": 12.39011287689209, "loss_alignment": 0.94677734375, "loss_alignment_w": 0.473388671875, "loss_sft": 0.3002456985414028, "loss_total_v6": 0.774382047355175, "step": 3129 }, { "epoch": 0.7218738738738739, "grad_norm": 24.946617126464844, "learning_rate": 6.904110553666206e-06, "loss": 12.237804412841797, "loss_alignment": 0.95703125, "loss_alignment_w": 0.478515625, "loss_sft": 0.2857673205435276, "loss_total_v6": 0.7648627832531929, "step": 3130 }, { "epoch": 0.7221045045045045, "grad_norm": 23.559968948364258, "learning_rate": 6.9022483041587785e-06, "loss": 12.849653244018555, "loss_alignment": 0.982421875, "loss_alignment_w": 0.4912109375, "loss_sft": 0.31206026300787926, "loss_total_v6": 0.8031033352017403, "step": 3131 }, { "epoch": 0.7223351351351351, "grad_norm": 24.436376571655273, "learning_rate": 6.900385746064268e-06, "loss": 11.891372680664062, "loss_alignment": 0.97119140625, "loss_alignment_w": 0.485595703125, "loss_sft": 0.25744721479713917, "loss_total_v6": 0.7432107850909233, "step": 3132 }, { "epoch": 0.7225657657657658, "grad_norm": 23.164501190185547, "learning_rate": 6.8985228796848284e-06, "loss": 12.483771324157715, "loss_alignment": 0.970703125, "loss_alignment_w": 0.4853515625, "loss_sft": 0.29596756398677826, "loss_total_v6": 0.7802357375621796, "step": 3133 }, { "epoch": 0.7227963963963964, "grad_norm": 26.616992950439453, "learning_rate": 6.896659705322654e-06, "loss": 12.778837203979492, "loss_alignment": 0.95556640625, "loss_alignment_w": 0.477783203125, "loss_sft": 0.3197955060750246, "loss_total_v6": 0.7986773252487183, "step": 3134 }, { "epoch": 0.723027027027027, "grad_norm": 23.088226318359375, "learning_rate": 6.894796223279994e-06, "loss": 12.401032447814941, "loss_alignment": 0.9677734375, "loss_alignment_w": 0.48388671875, "loss_sft": 0.2924595531076193, "loss_total_v6": 0.775064542889595, "step": 3135 }, { "epoch": 0.7232576576576577, "grad_norm": 25.916950225830078, "learning_rate": 6.892932433859147e-06, "loss": 11.871187210083008, "loss_alignment": 0.9541015625, "loss_alignment_w": 0.47705078125, "loss_sft": 0.26437963359057903, "loss_total_v6": 0.7419492155313492, "step": 3136 }, { "epoch": 0.7234882882882883, "grad_norm": 24.412639617919922, "learning_rate": 6.891068337362462e-06, "loss": 12.169952392578125, "loss_alignment": 0.96240234375, "loss_alignment_w": 0.481201171875, "loss_sft": 0.27758977748453617, "loss_total_v6": 0.7606219723820686, "step": 3137 }, { "epoch": 0.7237189189189189, "grad_norm": 25.8095760345459, "learning_rate": 6.889203934092337e-06, "loss": 12.133016586303711, "loss_alignment": 0.96484375, "loss_alignment_w": 0.482421875, "loss_sft": 0.2756933346390724, "loss_total_v6": 0.7583135887980461, "step": 3138 }, { "epoch": 0.7239495495495496, "grad_norm": 25.116641998291016, "learning_rate": 6.887339224351217e-06, "loss": 12.11992359161377, "loss_alignment": 0.97216796875, "loss_alignment_w": 0.486083984375, "loss_sft": 0.2718079462647438, "loss_total_v6": 0.7574952244758606, "step": 3139 }, { "epoch": 0.7241801801801802, "grad_norm": 22.08954429626465, "learning_rate": 6.885474208441602e-06, "loss": 11.756927490234375, "loss_alignment": 0.9521484375, "loss_alignment_w": 0.47607421875, "loss_sft": 0.2588558550924063, "loss_total_v6": 0.7348080202937126, "step": 3140 }, { "epoch": 0.7244108108108108, "grad_norm": 24.617507934570312, "learning_rate": 6.88360888666604e-06, "loss": 11.629745483398438, "loss_alignment": 0.9521484375, "loss_alignment_w": 0.47607421875, "loss_sft": 0.2495336439460516, "loss_total_v6": 0.7268590927124023, "step": 3141 }, { "epoch": 0.7246414414414415, "grad_norm": 22.813945770263672, "learning_rate": 6.8817432593271275e-06, "loss": 12.282308578491211, "loss_alignment": 0.94384765625, "loss_alignment_w": 0.471923828125, "loss_sft": 0.29529321007430553, "loss_total_v6": 0.7676442861557007, "step": 3142 }, { "epoch": 0.724872072072072, "grad_norm": 24.225448608398438, "learning_rate": 6.8798773267275065e-06, "loss": 12.04464054107666, "loss_alignment": 0.95751953125, "loss_alignment_w": 0.478759765625, "loss_sft": 0.2742133680731058, "loss_total_v6": 0.7527900189161301, "step": 3143 }, { "epoch": 0.7251027027027027, "grad_norm": 26.51407241821289, "learning_rate": 6.878011089169878e-06, "loss": 12.053150177001953, "loss_alignment": 0.94921875, "loss_alignment_w": 0.474609375, "loss_sft": 0.2780105900019407, "loss_total_v6": 0.7533218637108803, "step": 3144 }, { "epoch": 0.7253333333333334, "grad_norm": 26.143564224243164, "learning_rate": 6.876144546956987e-06, "loss": 12.296855926513672, "loss_alignment": 0.9521484375, "loss_alignment_w": 0.47607421875, "loss_sft": 0.2920520231127739, "loss_total_v6": 0.7685534879565239, "step": 3145 }, { "epoch": 0.7255639639639639, "grad_norm": 22.766883850097656, "learning_rate": 6.874277700391623e-06, "loss": 12.164803504943848, "loss_alignment": 0.95703125, "loss_alignment_w": 0.478515625, "loss_sft": 0.282074511051178, "loss_total_v6": 0.7603002190589905, "step": 3146 }, { "epoch": 0.7257945945945946, "grad_norm": 24.195388793945312, "learning_rate": 6.872410549776636e-06, "loss": 11.624333381652832, "loss_alignment": 0.9248046875, "loss_alignment_w": 0.46240234375, "loss_sft": 0.2641946878284216, "loss_total_v6": 0.7265207916498184, "step": 3147 }, { "epoch": 0.7260252252252253, "grad_norm": 24.534191131591797, "learning_rate": 6.870543095414919e-06, "loss": 12.515368461608887, "loss_alignment": 0.951171875, "loss_alignment_w": 0.4755859375, "loss_sft": 0.3053886145353317, "loss_total_v6": 0.7822105213999748, "step": 3148 }, { "epoch": 0.7262558558558558, "grad_norm": 27.2171573638916, "learning_rate": 6.8686753376094095e-06, "loss": 11.816505432128906, "loss_alignment": 0.9423828125, "loss_alignment_w": 0.47119140625, "loss_sft": 0.2674927730113268, "loss_total_v6": 0.7385315671563148, "step": 3149 }, { "epoch": 0.7264864864864865, "grad_norm": 22.876943588256836, "learning_rate": 6.8668072766631054e-06, "loss": 11.488252639770508, "loss_alignment": 0.94921875, "loss_alignment_w": 0.474609375, "loss_sft": 0.24336067773401737, "loss_total_v6": 0.7180158421397209, "step": 3150 }, { "epoch": 0.7267171171171171, "grad_norm": 22.64739227294922, "learning_rate": 6.8649389128790455e-06, "loss": 12.43968677520752, "loss_alignment": 0.9609375, "loss_alignment_w": 0.48046875, "loss_sft": 0.29641660302877426, "loss_total_v6": 0.7774804383516312, "step": 3151 }, { "epoch": 0.7269477477477477, "grad_norm": 22.85974884033203, "learning_rate": 6.863070246560319e-06, "loss": 12.031152725219727, "loss_alignment": 0.982421875, "loss_alignment_w": 0.4912109375, "loss_sft": 0.26094966381788254, "loss_total_v6": 0.7519469931721687, "step": 3152 }, { "epoch": 0.7271783783783784, "grad_norm": 23.10259246826172, "learning_rate": 6.861201278010068e-06, "loss": 11.981782913208008, "loss_alignment": 0.95703125, "loss_alignment_w": 0.478515625, "loss_sft": 0.27136817015707493, "loss_total_v6": 0.7488614246249199, "step": 3153 }, { "epoch": 0.727409009009009, "grad_norm": 24.706653594970703, "learning_rate": 6.859332007531477e-06, "loss": 12.410709381103516, "loss_alignment": 0.9658203125, "loss_alignment_w": 0.48291015625, "loss_sft": 0.2921793982386589, "loss_total_v6": 0.7756693810224533, "step": 3154 }, { "epoch": 0.7276396396396396, "grad_norm": 24.15196990966797, "learning_rate": 6.8574624354277866e-06, "loss": 12.210051536560059, "loss_alignment": 0.95849609375, "loss_alignment_w": 0.479248046875, "loss_sft": 0.28523820266127586, "loss_total_v6": 0.7631282061338425, "step": 3155 }, { "epoch": 0.7278702702702703, "grad_norm": 25.405017852783203, "learning_rate": 6.855592562002281e-06, "loss": 12.070079803466797, "loss_alignment": 0.9609375, "loss_alignment_w": 0.48046875, "loss_sft": 0.27337718196213245, "loss_total_v6": 0.754380002617836, "step": 3156 }, { "epoch": 0.7281009009009008, "grad_norm": 24.361881256103516, "learning_rate": 6.853722387558298e-06, "loss": 12.516979217529297, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.2873161230236292, "loss_total_v6": 0.7823112234473228, "step": 3157 }, { "epoch": 0.7283315315315315, "grad_norm": 22.784669876098633, "learning_rate": 6.851851912399219e-06, "loss": 12.175875663757324, "loss_alignment": 0.9443359375, "loss_alignment_w": 0.47216796875, "loss_sft": 0.28882422111928463, "loss_total_v6": 0.7609921991825104, "step": 3158 }, { "epoch": 0.7285621621621622, "grad_norm": 26.062519073486328, "learning_rate": 6.849981136828478e-06, "loss": 12.024723052978516, "loss_alignment": 0.95068359375, "loss_alignment_w": 0.475341796875, "loss_sft": 0.27653905004262924, "loss_total_v6": 0.7515451535582542, "step": 3159 }, { "epoch": 0.7287927927927927, "grad_norm": 23.638431549072266, "learning_rate": 6.848110061149555e-06, "loss": 11.965889930725098, "loss_alignment": 0.9609375, "loss_alignment_w": 0.48046875, "loss_sft": 0.26788765378296375, "loss_total_v6": 0.7478680983185768, "step": 3160 }, { "epoch": 0.7290234234234234, "grad_norm": 21.1898250579834, "learning_rate": 6.8462386856659824e-06, "loss": 11.782470703125, "loss_alignment": 0.94091796875, "loss_alignment_w": 0.470458984375, "loss_sft": 0.2648773957043886, "loss_total_v6": 0.7364044785499573, "step": 3161 }, { "epoch": 0.7292540540540541, "grad_norm": 23.04637336730957, "learning_rate": 6.844367010681338e-06, "loss": 12.135538101196289, "loss_alignment": 0.9677734375, "loss_alignment_w": 0.48388671875, "loss_sft": 0.27492013014853, "loss_total_v6": 0.7584711834788322, "step": 3162 }, { "epoch": 0.7294846846846846, "grad_norm": 22.64874267578125, "learning_rate": 6.842495036499248e-06, "loss": 11.452049255371094, "loss_alignment": 0.94580078125, "loss_alignment_w": 0.472900390625, "loss_sft": 0.24320362508296967, "loss_total_v6": 0.7157530561089516, "step": 3163 }, { "epoch": 0.7297153153153153, "grad_norm": 23.601821899414062, "learning_rate": 6.840622763423391e-06, "loss": 12.518135070800781, "loss_alignment": 0.97119140625, "loss_alignment_w": 0.485595703125, "loss_sft": 0.2976116891950369, "loss_total_v6": 0.7823834344744682, "step": 3164 }, { "epoch": 0.729945945945946, "grad_norm": 22.594512939453125, "learning_rate": 6.838750191757488e-06, "loss": 12.170327186584473, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.27048734948039055, "loss_total_v6": 0.7606454491615295, "step": 3165 }, { "epoch": 0.7301765765765765, "grad_norm": 24.265316009521484, "learning_rate": 6.836877321805316e-06, "loss": 12.783834457397461, "loss_alignment": 0.9560546875, "loss_alignment_w": 0.47802734375, "loss_sft": 0.3212827779352665, "loss_total_v6": 0.7989896833896637, "step": 3166 }, { "epoch": 0.7304072072072072, "grad_norm": 27.75029945373535, "learning_rate": 6.835004153870694e-06, "loss": 12.061568260192871, "loss_alignment": 0.982421875, "loss_alignment_w": 0.4912109375, "loss_sft": 0.2634762916713953, "loss_total_v6": 0.7538479939103127, "step": 3167 }, { "epoch": 0.7306378378378379, "grad_norm": 24.274639129638672, "learning_rate": 6.833130688257489e-06, "loss": 12.14105224609375, "loss_alignment": 0.9443359375, "loss_alignment_w": 0.47216796875, "loss_sft": 0.2877311874181032, "loss_total_v6": 0.7588157951831818, "step": 3168 }, { "epoch": 0.7308684684684684, "grad_norm": 22.781307220458984, "learning_rate": 6.831256925269624e-06, "loss": 12.095617294311523, "loss_alignment": 0.95068359375, "loss_alignment_w": 0.475341796875, "loss_sft": 0.2820991761982441, "loss_total_v6": 0.7559761330485344, "step": 3169 }, { "epoch": 0.7310990990990991, "grad_norm": 23.809816360473633, "learning_rate": 6.829382865211063e-06, "loss": 12.291383743286133, "loss_alignment": 0.94140625, "loss_alignment_w": 0.470703125, "loss_sft": 0.29651655070483685, "loss_total_v6": 0.7682114914059639, "step": 3170 }, { "epoch": 0.7313297297297298, "grad_norm": 24.508779525756836, "learning_rate": 6.827508508385817e-06, "loss": 12.167329788208008, "loss_alignment": 0.9501953125, "loss_alignment_w": 0.47509765625, "loss_sft": 0.2851621713489294, "loss_total_v6": 0.7604581564664841, "step": 3171 }, { "epoch": 0.7315603603603603, "grad_norm": 22.26992416381836, "learning_rate": 6.825633855097954e-06, "loss": 12.261587142944336, "loss_alignment": 0.9638671875, "loss_alignment_w": 0.48193359375, "loss_sft": 0.2847665883600712, "loss_total_v6": 0.766349196434021, "step": 3172 }, { "epoch": 0.731790990990991, "grad_norm": 22.434364318847656, "learning_rate": 6.823758905651582e-06, "loss": 12.163410186767578, "loss_alignment": 0.9453125, "loss_alignment_w": 0.47265625, "loss_sft": 0.2872059550136328, "loss_total_v6": 0.760213166475296, "step": 3173 }, { "epoch": 0.7320216216216217, "grad_norm": 23.592714309692383, "learning_rate": 6.821883660350858e-06, "loss": 12.422792434692383, "loss_alignment": 0.95703125, "loss_alignment_w": 0.478515625, "loss_sft": 0.2968407869338989, "loss_total_v6": 0.7764245197176933, "step": 3174 }, { "epoch": 0.7322522522522522, "grad_norm": 22.167285919189453, "learning_rate": 6.820008119499992e-06, "loss": 11.744691848754883, "loss_alignment": 0.93896484375, "loss_alignment_w": 0.469482421875, "loss_sft": 0.2634622696787119, "loss_total_v6": 0.734043300151825, "step": 3175 }, { "epoch": 0.7324828828828829, "grad_norm": 24.894437789916992, "learning_rate": 6.818132283403236e-06, "loss": 12.35335922241211, "loss_alignment": 0.9677734375, "loss_alignment_w": 0.48388671875, "loss_sft": 0.28842712193727493, "loss_total_v6": 0.7720849439501762, "step": 3176 }, { "epoch": 0.7327135135135135, "grad_norm": 23.611173629760742, "learning_rate": 6.816256152364893e-06, "loss": 12.973939895629883, "loss_alignment": 0.97216796875, "loss_alignment_w": 0.486083984375, "loss_sft": 0.325092438608408, "loss_total_v6": 0.8108712285757065, "step": 3177 }, { "epoch": 0.7329441441441441, "grad_norm": 31.124544143676758, "learning_rate": 6.814379726689313e-06, "loss": 12.31900405883789, "loss_alignment": 0.943359375, "loss_alignment_w": 0.4716796875, "loss_sft": 0.2984564397484064, "loss_total_v6": 0.7699377462267876, "step": 3178 }, { "epoch": 0.7331747747747748, "grad_norm": 22.88240623474121, "learning_rate": 6.812503006680895e-06, "loss": 11.506268501281738, "loss_alignment": 0.9609375, "loss_alignment_w": 0.48046875, "loss_sft": 0.2377117220312357, "loss_total_v6": 0.719141811132431, "step": 3179 }, { "epoch": 0.7334054054054054, "grad_norm": 25.012487411499023, "learning_rate": 6.810625992644085e-06, "loss": 12.686370849609375, "loss_alignment": 0.96728515625, "loss_alignment_w": 0.483642578125, "loss_sft": 0.3095608204603195, "loss_total_v6": 0.7928981930017471, "step": 3180 }, { "epoch": 0.733636036036036, "grad_norm": 23.720518112182617, "learning_rate": 6.808748684883374e-06, "loss": 12.453910827636719, "loss_alignment": 0.9658203125, "loss_alignment_w": 0.48291015625, "loss_sft": 0.2953066173940897, "loss_total_v6": 0.7783693745732307, "step": 3181 }, { "epoch": 0.7338666666666667, "grad_norm": 23.511695861816406, "learning_rate": 6.806871083703307e-06, "loss": 12.156609535217285, "loss_alignment": 0.94482421875, "loss_alignment_w": 0.472412109375, "loss_sft": 0.2880168743431568, "loss_total_v6": 0.7597881108522415, "step": 3182 }, { "epoch": 0.7340972972972973, "grad_norm": 23.356422424316406, "learning_rate": 6.80499318940847e-06, "loss": 12.491668701171875, "loss_alignment": 0.95068359375, "loss_alignment_w": 0.475341796875, "loss_sft": 0.30538754165172577, "loss_total_v6": 0.7807293236255646, "step": 3183 }, { "epoch": 0.7343279279279279, "grad_norm": 22.284223556518555, "learning_rate": 6.8031150023035e-06, "loss": 11.866703987121582, "loss_alignment": 0.9296875, "loss_alignment_w": 0.46484375, "loss_sft": 0.2785800099372864, "loss_total_v6": 0.7416689619421959, "step": 3184 }, { "epoch": 0.7345585585585586, "grad_norm": 23.219881057739258, "learning_rate": 6.8012365226930825e-06, "loss": 12.731391906738281, "loss_alignment": 0.9677734375, "loss_alignment_w": 0.48388671875, "loss_sft": 0.3124356288462877, "loss_total_v6": 0.7957120016217232, "step": 3185 }, { "epoch": 0.7347891891891892, "grad_norm": 24.02411460876465, "learning_rate": 6.799357750881944e-06, "loss": 11.624250411987305, "loss_alignment": 0.93505859375, "loss_alignment_w": 0.467529296875, "loss_sft": 0.2596119213849306, "loss_total_v6": 0.7265156134963036, "step": 3186 }, { "epoch": 0.7350198198198198, "grad_norm": 25.14107894897461, "learning_rate": 6.7974786871748674e-06, "loss": 12.536731719970703, "loss_alignment": 0.9677734375, "loss_alignment_w": 0.48388671875, "loss_sft": 0.2992470283061266, "loss_total_v6": 0.7835457623004913, "step": 3187 }, { "epoch": 0.7352504504504505, "grad_norm": 22.982152938842773, "learning_rate": 6.795599331876679e-06, "loss": 11.32621955871582, "loss_alignment": 0.9560546875, "loss_alignment_w": 0.47802734375, "loss_sft": 0.23056330811232328, "loss_total_v6": 0.7078887224197388, "step": 3188 }, { "epoch": 0.7354810810810811, "grad_norm": 22.51236343383789, "learning_rate": 6.79371968529225e-06, "loss": 11.67171573638916, "loss_alignment": 0.9453125, "loss_alignment_w": 0.47265625, "loss_sft": 0.2570853941142559, "loss_total_v6": 0.7294822409749031, "step": 3189 }, { "epoch": 0.7357117117117117, "grad_norm": 26.529191970825195, "learning_rate": 6.7918397477265e-06, "loss": 12.4530029296875, "loss_alignment": 0.95556640625, "loss_alignment_w": 0.477783203125, "loss_sft": 0.3003769312053919, "loss_total_v6": 0.7783127129077911, "step": 3190 }, { "epoch": 0.7359423423423423, "grad_norm": 23.980741500854492, "learning_rate": 6.7899595194844025e-06, "loss": 11.895832061767578, "loss_alignment": 0.955078125, "loss_alignment_w": 0.4775390625, "loss_sft": 0.26637760549783707, "loss_total_v6": 0.7434894815087318, "step": 3191 }, { "epoch": 0.736172972972973, "grad_norm": 24.695539474487305, "learning_rate": 6.788079000870966e-06, "loss": 12.474390029907227, "loss_alignment": 0.95654296875, "loss_alignment_w": 0.478271484375, "loss_sft": 0.30200352892279625, "loss_total_v6": 0.7796493917703629, "step": 3192 }, { "epoch": 0.7364036036036036, "grad_norm": 25.654159545898438, "learning_rate": 6.7861981921912555e-06, "loss": 12.482175827026367, "loss_alignment": 0.9658203125, "loss_alignment_w": 0.48291015625, "loss_sft": 0.29730214178562164, "loss_total_v6": 0.7801360040903091, "step": 3193 }, { "epoch": 0.7366342342342342, "grad_norm": 24.618886947631836, "learning_rate": 6.784317093750381e-06, "loss": 12.212296485900879, "loss_alignment": 0.962890625, "loss_alignment_w": 0.4814453125, "loss_sft": 0.28124336898326874, "loss_total_v6": 0.7632685154676437, "step": 3194 }, { "epoch": 0.7368648648648649, "grad_norm": 27.24838638305664, "learning_rate": 6.782435705853495e-06, "loss": 12.79422378540039, "loss_alignment": 0.95947265625, "loss_alignment_w": 0.479736328125, "loss_sft": 0.3193076252937317, "loss_total_v6": 0.7996390387415886, "step": 3195 }, { "epoch": 0.7370954954954955, "grad_norm": 23.634756088256836, "learning_rate": 6.780554028805803e-06, "loss": 12.268709182739258, "loss_alignment": 0.96240234375, "loss_alignment_w": 0.481201171875, "loss_sft": 0.2854710351675749, "loss_total_v6": 0.7667943015694618, "step": 3196 }, { "epoch": 0.7373261261261261, "grad_norm": 27.26429557800293, "learning_rate": 6.7786720629125565e-06, "loss": 12.530064582824707, "loss_alignment": 0.96044921875, "loss_alignment_w": 0.480224609375, "loss_sft": 0.30127172358334064, "loss_total_v6": 0.783129058778286, "step": 3197 }, { "epoch": 0.7375567567567568, "grad_norm": 22.606510162353516, "learning_rate": 6.776789808479049e-06, "loss": 12.208456039428711, "loss_alignment": 0.9521484375, "loss_alignment_w": 0.47607421875, "loss_sft": 0.28600829280912876, "loss_total_v6": 0.7630285397171974, "step": 3198 }, { "epoch": 0.7377873873873874, "grad_norm": 26.478912353515625, "learning_rate": 6.774907265810626e-06, "loss": 12.177063941955566, "loss_alignment": 0.958984375, "loss_alignment_w": 0.4794921875, "loss_sft": 0.2827949933707714, "loss_total_v6": 0.7610664740204811, "step": 3199 }, { "epoch": 0.738018018018018, "grad_norm": 22.743967056274414, "learning_rate": 6.773024435212678e-06, "loss": 11.886887550354004, "loss_alignment": 0.919921875, "loss_alignment_w": 0.4599609375, "loss_sft": 0.283412029966712, "loss_total_v6": 0.7429304718971252, "step": 3200 }, { "epoch": 0.7382486486486487, "grad_norm": 23.87235450744629, "learning_rate": 6.771141316990643e-06, "loss": 11.89337158203125, "loss_alignment": 0.95166015625, "loss_alignment_w": 0.475830078125, "loss_sft": 0.26752088218927383, "loss_total_v6": 0.7433357015252113, "step": 3201 }, { "epoch": 0.7384792792792793, "grad_norm": 21.750877380371094, "learning_rate": 6.7692579114500015e-06, "loss": 11.884733200073242, "loss_alignment": 0.951171875, "loss_alignment_w": 0.4755859375, "loss_sft": 0.26672161370515823, "loss_total_v6": 0.742795817553997, "step": 3202 }, { "epoch": 0.7387099099099099, "grad_norm": 25.777875900268555, "learning_rate": 6.767374218896286e-06, "loss": 11.595129013061523, "loss_alignment": 0.9248046875, "loss_alignment_w": 0.46240234375, "loss_sft": 0.26264411956071854, "loss_total_v6": 0.7246955186128616, "step": 3203 }, { "epoch": 0.7389405405405406, "grad_norm": 23.931352615356445, "learning_rate": 6.765490239635075e-06, "loss": 12.087322235107422, "loss_alignment": 0.9609375, "loss_alignment_w": 0.48046875, "loss_sft": 0.2759959641844034, "loss_total_v6": 0.7554576322436333, "step": 3204 }, { "epoch": 0.7391711711711711, "grad_norm": 25.23208236694336, "learning_rate": 6.7636059739719876e-06, "loss": 11.937332153320312, "loss_alignment": 0.91845703125, "loss_alignment_w": 0.459228515625, "loss_sft": 0.28560350462794304, "loss_total_v6": 0.7460832297801971, "step": 3205 }, { "epoch": 0.7394018018018018, "grad_norm": 23.92264747619629, "learning_rate": 6.761721422212697e-06, "loss": 12.150484085083008, "loss_alignment": 0.953125, "loss_alignment_w": 0.4765625, "loss_sft": 0.2811185158789158, "loss_total_v6": 0.7594052478671074, "step": 3206 }, { "epoch": 0.7396324324324325, "grad_norm": 24.910120010375977, "learning_rate": 6.75983658466292e-06, "loss": 12.43228530883789, "loss_alignment": 0.951171875, "loss_alignment_w": 0.4755859375, "loss_sft": 0.3019964434206486, "loss_total_v6": 0.7770178094506264, "step": 3207 }, { "epoch": 0.739863063063063, "grad_norm": 22.73103141784668, "learning_rate": 6.757951461628416e-06, "loss": 12.013851165771484, "loss_alignment": 0.96728515625, "loss_alignment_w": 0.483642578125, "loss_sft": 0.26835223846137524, "loss_total_v6": 0.7508656904101372, "step": 3208 }, { "epoch": 0.7400936936936937, "grad_norm": 26.125537872314453, "learning_rate": 6.756066053414998e-06, "loss": 11.749353408813477, "loss_alignment": 0.962890625, "loss_alignment_w": 0.4814453125, "loss_sft": 0.2526909075677395, "loss_total_v6": 0.7343345731496811, "step": 3209 }, { "epoch": 0.7403243243243244, "grad_norm": 24.723237991333008, "learning_rate": 6.75418036032852e-06, "loss": 12.833696365356445, "loss_alignment": 0.953125, "loss_alignment_w": 0.4765625, "loss_sft": 0.3258792143315077, "loss_total_v6": 0.8021060079336166, "step": 3210 }, { "epoch": 0.7405549549549549, "grad_norm": 24.86876678466797, "learning_rate": 6.75229438267488e-06, "loss": 12.332137107849121, "loss_alignment": 0.9658203125, "loss_alignment_w": 0.48291015625, "loss_sft": 0.28626150637865067, "loss_total_v6": 0.7707585766911507, "step": 3211 }, { "epoch": 0.7407855855855856, "grad_norm": 25.307941436767578, "learning_rate": 6.7504081207600295e-06, "loss": 12.113409042358398, "loss_alignment": 0.9619140625, "loss_alignment_w": 0.48095703125, "loss_sft": 0.27495609782636166, "loss_total_v6": 0.7570880874991417, "step": 3212 }, { "epoch": 0.7410162162162162, "grad_norm": 25.41567039489746, "learning_rate": 6.748521574889961e-06, "loss": 12.189936637878418, "loss_alignment": 0.947265625, "loss_alignment_w": 0.4736328125, "loss_sft": 0.2875515818595886, "loss_total_v6": 0.7618710324168205, "step": 3213 }, { "epoch": 0.7412468468468468, "grad_norm": 23.530672073364258, "learning_rate": 6.746634745370715e-06, "loss": 11.723989486694336, "loss_alignment": 0.93310546875, "loss_alignment_w": 0.466552734375, "loss_sft": 0.2665628660470247, "loss_total_v6": 0.732749380171299, "step": 3214 }, { "epoch": 0.7414774774774775, "grad_norm": 23.314241409301758, "learning_rate": 6.7447476325083764e-06, "loss": 12.068260192871094, "loss_alignment": 0.96826171875, "loss_alignment_w": 0.484130859375, "loss_sft": 0.27201225608587265, "loss_total_v6": 0.7542662620544434, "step": 3215 }, { "epoch": 0.741708108108108, "grad_norm": 25.753456115722656, "learning_rate": 6.7428602366090764e-06, "loss": 12.14443588256836, "loss_alignment": 0.97021484375, "loss_alignment_w": 0.485107421875, "loss_sft": 0.27457597106695175, "loss_total_v6": 0.7590272724628448, "step": 3216 }, { "epoch": 0.7419387387387387, "grad_norm": 24.32484245300293, "learning_rate": 6.740972557978995e-06, "loss": 12.116437911987305, "loss_alignment": 0.94580078125, "loss_alignment_w": 0.472900390625, "loss_sft": 0.28495682775974274, "loss_total_v6": 0.7572773694992065, "step": 3217 }, { "epoch": 0.7421693693693694, "grad_norm": 22.243972778320312, "learning_rate": 6.739084596924353e-06, "loss": 11.286636352539062, "loss_alignment": 0.93896484375, "loss_alignment_w": 0.469482421875, "loss_sft": 0.2354746311903, "loss_total_v6": 0.7054148390889168, "step": 3218 }, { "epoch": 0.7424, "grad_norm": 21.706449508666992, "learning_rate": 6.737196353751423e-06, "loss": 12.333333969116211, "loss_alignment": 0.9375, "loss_alignment_w": 0.46875, "loss_sft": 0.3022512588649988, "loss_total_v6": 0.7708333879709244, "step": 3219 }, { "epoch": 0.7426306306306306, "grad_norm": 21.620731353759766, "learning_rate": 6.735307828766515e-06, "loss": 11.726123809814453, "loss_alignment": 0.9443359375, "loss_alignment_w": 0.47216796875, "loss_sft": 0.26062317937612534, "loss_total_v6": 0.7328827008605003, "step": 3220 }, { "epoch": 0.7428612612612613, "grad_norm": 23.535728454589844, "learning_rate": 6.733419022275995e-06, "loss": 12.229698181152344, "loss_alignment": 0.93310546875, "loss_alignment_w": 0.466552734375, "loss_sft": 0.29723875783383846, "loss_total_v6": 0.7643560990691185, "step": 3221 }, { "epoch": 0.7430918918918918, "grad_norm": 27.37643814086914, "learning_rate": 6.731529934586267e-06, "loss": 12.295443534851074, "loss_alignment": 0.96337890625, "loss_alignment_w": 0.481689453125, "loss_sft": 0.28659265860915184, "loss_total_v6": 0.7684652134776115, "step": 3222 }, { "epoch": 0.7433225225225225, "grad_norm": 23.84072494506836, "learning_rate": 6.729640566003785e-06, "loss": 11.810469627380371, "loss_alignment": 0.94970703125, "loss_alignment_w": 0.474853515625, "loss_sft": 0.26406374759972095, "loss_total_v6": 0.7381543219089508, "step": 3223 }, { "epoch": 0.7435531531531532, "grad_norm": 22.2059383392334, "learning_rate": 6.7277509168350445e-06, "loss": 12.044204711914062, "loss_alignment": 0.9521484375, "loss_alignment_w": 0.47607421875, "loss_sft": 0.2756509967148304, "loss_total_v6": 0.7527627721428871, "step": 3224 }, { "epoch": 0.7437837837837837, "grad_norm": 22.05384635925293, "learning_rate": 6.7258609873865895e-06, "loss": 12.05897331237793, "loss_alignment": 0.931640625, "loss_alignment_w": 0.4658203125, "loss_sft": 0.28736204095184803, "loss_total_v6": 0.7536858916282654, "step": 3225 }, { "epoch": 0.7440144144144144, "grad_norm": 23.750761032104492, "learning_rate": 6.72397077796501e-06, "loss": 11.922179222106934, "loss_alignment": 0.90380859375, "loss_alignment_w": 0.451904296875, "loss_sft": 0.2931556235998869, "loss_total_v6": 0.7451361790299416, "step": 3226 }, { "epoch": 0.7442450450450451, "grad_norm": 24.51534652709961, "learning_rate": 6.722080288876936e-06, "loss": 12.662532806396484, "loss_alignment": 0.96435546875, "loss_alignment_w": 0.482177734375, "loss_sft": 0.30862025544047356, "loss_total_v6": 0.7914083302021027, "step": 3227 }, { "epoch": 0.7444756756756756, "grad_norm": 22.690519332885742, "learning_rate": 6.720189520429052e-06, "loss": 11.990272521972656, "loss_alignment": 0.95849609375, "loss_alignment_w": 0.479248046875, "loss_sft": 0.2702355496585369, "loss_total_v6": 0.7493920102715492, "step": 3228 }, { "epoch": 0.7447063063063063, "grad_norm": 24.116924285888672, "learning_rate": 6.718298472928082e-06, "loss": 11.643857955932617, "loss_alignment": 0.93212890625, "loss_alignment_w": 0.466064453125, "loss_sft": 0.26086792163550854, "loss_total_v6": 0.7277410998940468, "step": 3229 }, { "epoch": 0.744936936936937, "grad_norm": 25.524093627929688, "learning_rate": 6.716407146680793e-06, "loss": 12.465141296386719, "loss_alignment": 0.95849609375, "loss_alignment_w": 0.479248046875, "loss_sft": 0.3013033978641033, "loss_total_v6": 0.7790713459253311, "step": 3230 }, { "epoch": 0.7451675675675675, "grad_norm": 23.378082275390625, "learning_rate": 6.714515541994002e-06, "loss": 12.442360877990723, "loss_alignment": 0.9541015625, "loss_alignment_w": 0.47705078125, "loss_sft": 0.3004594389349222, "loss_total_v6": 0.7776475325226784, "step": 3231 }, { "epoch": 0.7453981981981982, "grad_norm": 23.655166625976562, "learning_rate": 6.712623659174569e-06, "loss": 12.674474716186523, "loss_alignment": 0.95458984375, "loss_alignment_w": 0.477294921875, "loss_sft": 0.3145545683801174, "loss_total_v6": 0.7921546772122383, "step": 3232 }, { "epoch": 0.7456288288288289, "grad_norm": 23.820556640625, "learning_rate": 6.7107314985294e-06, "loss": 12.181327819824219, "loss_alignment": 0.92919921875, "loss_alignment_w": 0.464599609375, "loss_sft": 0.2969164941459894, "loss_total_v6": 0.7613329887390137, "step": 3233 }, { "epoch": 0.7458594594594594, "grad_norm": 25.981271743774414, "learning_rate": 6.708839060365449e-06, "loss": 12.287789344787598, "loss_alignment": 0.95458984375, "loss_alignment_w": 0.477294921875, "loss_sft": 0.29072244837880135, "loss_total_v6": 0.7679868340492249, "step": 3234 }, { "epoch": 0.7460900900900901, "grad_norm": 23.514860153198242, "learning_rate": 6.706946344989704e-06, "loss": 12.236804962158203, "loss_alignment": 0.9697265625, "loss_alignment_w": 0.48486328125, "loss_sft": 0.28031850419938564, "loss_total_v6": 0.7648003324866295, "step": 3235 }, { "epoch": 0.7463207207207208, "grad_norm": 26.875215530395508, "learning_rate": 6.705053352709212e-06, "loss": 12.43221664428711, "loss_alignment": 0.95361328125, "loss_alignment_w": 0.476806640625, "loss_sft": 0.3001764062792063, "loss_total_v6": 0.7770135775208473, "step": 3236 }, { "epoch": 0.7465513513513513, "grad_norm": 23.48318099975586, "learning_rate": 6.703160083831056e-06, "loss": 11.89145278930664, "loss_alignment": 0.943359375, "loss_alignment_w": 0.4716796875, "loss_sft": 0.2712614480406046, "loss_total_v6": 0.743215762078762, "step": 3237 }, { "epoch": 0.746781981981982, "grad_norm": 23.87840461730957, "learning_rate": 6.701266538662368e-06, "loss": 11.331622123718262, "loss_alignment": 0.919921875, "loss_alignment_w": 0.4599609375, "loss_sft": 0.24765514582395554, "loss_total_v6": 0.7082263827323914, "step": 3238 }, { "epoch": 0.7470126126126126, "grad_norm": 23.562034606933594, "learning_rate": 6.699372717510321e-06, "loss": 12.353282928466797, "loss_alignment": 0.94873046875, "loss_alignment_w": 0.474365234375, "loss_sft": 0.2980048190802336, "loss_total_v6": 0.7720801383256912, "step": 3239 }, { "epoch": 0.7472432432432432, "grad_norm": 58.76903533935547, "learning_rate": 6.697478620682137e-06, "loss": 12.562740325927734, "loss_alignment": 0.9599609375, "loss_alignment_w": 0.47998046875, "loss_sft": 0.3037565052509308, "loss_total_v6": 0.7851712852716446, "step": 3240 }, { "epoch": 0.7474738738738739, "grad_norm": 21.8107967376709, "learning_rate": 6.69558424848508e-06, "loss": 11.355453491210938, "loss_alignment": 0.9296875, "loss_alignment_w": 0.46484375, "loss_sft": 0.24497895501554012, "loss_total_v6": 0.709715873003006, "step": 3241 }, { "epoch": 0.7477045045045045, "grad_norm": 24.737712860107422, "learning_rate": 6.693689601226458e-06, "loss": 12.343807220458984, "loss_alignment": 0.95458984375, "loss_alignment_w": 0.477294921875, "loss_sft": 0.2936284504830837, "loss_total_v6": 0.7714879438281059, "step": 3242 }, { "epoch": 0.7479351351351351, "grad_norm": 23.957231521606445, "learning_rate": 6.691794679213627e-06, "loss": 12.331716537475586, "loss_alignment": 0.95849609375, "loss_alignment_w": 0.479248046875, "loss_sft": 0.29181995801627636, "loss_total_v6": 0.7707323133945465, "step": 3243 }, { "epoch": 0.7481657657657658, "grad_norm": 22.15886116027832, "learning_rate": 6.689899482753984e-06, "loss": 11.866817474365234, "loss_alignment": 0.97314453125, "loss_alignment_w": 0.486572265625, "loss_sft": 0.2542035598307848, "loss_total_v6": 0.7416761070489883, "step": 3244 }, { "epoch": 0.7483963963963964, "grad_norm": 25.07353401184082, "learning_rate": 6.688004012154975e-06, "loss": 12.377840042114258, "loss_alignment": 0.96044921875, "loss_alignment_w": 0.480224609375, "loss_sft": 0.29252067022025585, "loss_total_v6": 0.7736150324344635, "step": 3245 }, { "epoch": 0.748627027027027, "grad_norm": 23.56062889099121, "learning_rate": 6.686108267724085e-06, "loss": 11.871347427368164, "loss_alignment": 0.92822265625, "loss_alignment_w": 0.464111328125, "loss_sft": 0.2782599199563265, "loss_total_v6": 0.741959273815155, "step": 3246 }, { "epoch": 0.7488576576576577, "grad_norm": 22.64150047302246, "learning_rate": 6.684212249768845e-06, "loss": 12.109363555908203, "loss_alignment": 0.95458984375, "loss_alignment_w": 0.477294921875, "loss_sft": 0.28007432259619236, "loss_total_v6": 0.7568351924419403, "step": 3247 }, { "epoch": 0.7490882882882883, "grad_norm": 24.77558135986328, "learning_rate": 6.6823159585968355e-06, "loss": 12.561208724975586, "loss_alignment": 0.947265625, "loss_alignment_w": 0.4736328125, "loss_sft": 0.31194629333913326, "loss_total_v6": 0.7850755751132965, "step": 3248 }, { "epoch": 0.7493189189189189, "grad_norm": 25.245471954345703, "learning_rate": 6.680419394515672e-06, "loss": 11.76544189453125, "loss_alignment": 0.9384765625, "loss_alignment_w": 0.46923828125, "loss_sft": 0.2667884938418865, "loss_total_v6": 0.7353401556611061, "step": 3249 }, { "epoch": 0.7495495495495496, "grad_norm": 23.731224060058594, "learning_rate": 6.678522557833025e-06, "loss": 11.911531448364258, "loss_alignment": 0.97119140625, "loss_alignment_w": 0.485595703125, "loss_sft": 0.2589360885322094, "loss_total_v6": 0.7444707602262497, "step": 3250 }, { "epoch": 0.7497801801801802, "grad_norm": 23.421894073486328, "learning_rate": 6.6766254488566e-06, "loss": 11.966959953308105, "loss_alignment": 0.9580078125, "loss_alignment_w": 0.47900390625, "loss_sft": 0.26941939257085323, "loss_total_v6": 0.7479349970817566, "step": 3251 }, { "epoch": 0.7500108108108108, "grad_norm": 24.801599502563477, "learning_rate": 6.674728067894149e-06, "loss": 11.763202667236328, "loss_alignment": 0.93408203125, "loss_alignment_w": 0.467041015625, "loss_sft": 0.2670909631997347, "loss_total_v6": 0.7352001294493675, "step": 3252 }, { "epoch": 0.7502414414414414, "grad_norm": 23.06279945373535, "learning_rate": 6.672830415253475e-06, "loss": 11.756026268005371, "loss_alignment": 0.9658203125, "loss_alignment_w": 0.48291015625, "loss_sft": 0.25055975280702114, "loss_total_v6": 0.7347516342997551, "step": 3253 }, { "epoch": 0.7504720720720721, "grad_norm": 23.857154846191406, "learning_rate": 6.670932491242414e-06, "loss": 11.724747657775879, "loss_alignment": 0.92431640625, "loss_alignment_w": 0.462158203125, "loss_sft": 0.2722559757530689, "loss_total_v6": 0.7327967509627342, "step": 3254 }, { "epoch": 0.7507027027027027, "grad_norm": 23.477405548095703, "learning_rate": 6.669034296168855e-06, "loss": 12.383260726928711, "loss_alignment": 0.9580078125, "loss_alignment_w": 0.47900390625, "loss_sft": 0.295087231323123, "loss_total_v6": 0.7739538177847862, "step": 3255 }, { "epoch": 0.7509333333333333, "grad_norm": 23.371234893798828, "learning_rate": 6.667135830340727e-06, "loss": 11.641739845275879, "loss_alignment": 0.93701171875, "loss_alignment_w": 0.468505859375, "loss_sft": 0.25859936233609915, "loss_total_v6": 0.7276087403297424, "step": 3256 }, { "epoch": 0.751163963963964, "grad_norm": 25.518686294555664, "learning_rate": 6.665237094066002e-06, "loss": 12.126752853393555, "loss_alignment": 0.94091796875, "loss_alignment_w": 0.470458984375, "loss_sft": 0.28747834637761116, "loss_total_v6": 0.7579220607876778, "step": 3257 }, { "epoch": 0.7513945945945946, "grad_norm": 24.87741470336914, "learning_rate": 6.663338087652697e-06, "loss": 12.293787956237793, "loss_alignment": 0.95703125, "loss_alignment_w": 0.478515625, "loss_sft": 0.29065476544201374, "loss_total_v6": 0.7683617100119591, "step": 3258 }, { "epoch": 0.7516252252252252, "grad_norm": 25.419851303100586, "learning_rate": 6.661438811408876e-06, "loss": 11.876131057739258, "loss_alignment": 0.9658203125, "loss_alignment_w": 0.48291015625, "loss_sft": 0.25972952134907246, "loss_total_v6": 0.7422582060098648, "step": 3259 }, { "epoch": 0.7518558558558559, "grad_norm": 24.36643409729004, "learning_rate": 6.659539265642643e-06, "loss": 12.970306396484375, "loss_alignment": 0.9638671875, "loss_alignment_w": 0.48193359375, "loss_sft": 0.327566159889102, "loss_total_v6": 0.8106441646814346, "step": 3260 }, { "epoch": 0.7520864864864865, "grad_norm": 24.044038772583008, "learning_rate": 6.657639450662145e-06, "loss": 12.21931266784668, "loss_alignment": 0.91943359375, "loss_alignment_w": 0.459716796875, "loss_sft": 0.3045700341463089, "loss_total_v6": 0.7637070268392563, "step": 3261 }, { "epoch": 0.7523171171171171, "grad_norm": 26.534229278564453, "learning_rate": 6.655739366775575e-06, "loss": 12.156232833862305, "loss_alignment": 0.92822265625, "loss_alignment_w": 0.464111328125, "loss_sft": 0.29514969885349274, "loss_total_v6": 0.7597645744681358, "step": 3262 }, { "epoch": 0.7525477477477478, "grad_norm": 23.82567596435547, "learning_rate": 6.653839014291171e-06, "loss": 12.073081970214844, "loss_alignment": 0.9208984375, "loss_alignment_w": 0.46044921875, "loss_sft": 0.29396579414606094, "loss_total_v6": 0.7545676082372665, "step": 3263 }, { "epoch": 0.7527783783783784, "grad_norm": 27.211828231811523, "learning_rate": 6.65193839351721e-06, "loss": 12.028026580810547, "loss_alignment": 0.9296875, "loss_alignment_w": 0.46484375, "loss_sft": 0.2866180371493101, "loss_total_v6": 0.7517516613006592, "step": 3264 }, { "epoch": 0.753009009009009, "grad_norm": 24.67951202392578, "learning_rate": 6.650037504762015e-06, "loss": 12.547645568847656, "loss_alignment": 0.94091796875, "loss_alignment_w": 0.470458984375, "loss_sft": 0.31279231980443, "loss_total_v6": 0.7842278629541397, "step": 3265 }, { "epoch": 0.7532396396396397, "grad_norm": 22.518110275268555, "learning_rate": 6.648136348333955e-06, "loss": 12.105241775512695, "loss_alignment": 0.93505859375, "loss_alignment_w": 0.467529296875, "loss_sft": 0.2890330869704485, "loss_total_v6": 0.7565776333212852, "step": 3266 }, { "epoch": 0.7534702702702702, "grad_norm": 23.533294677734375, "learning_rate": 6.646234924541438e-06, "loss": 12.03079605102539, "loss_alignment": 0.9345703125, "loss_alignment_w": 0.46728515625, "loss_sft": 0.28390714153647423, "loss_total_v6": 0.7519247159361839, "step": 3267 }, { "epoch": 0.7537009009009009, "grad_norm": 22.53984832763672, "learning_rate": 6.644333233692917e-06, "loss": 11.140458106994629, "loss_alignment": 0.93310546875, "loss_alignment_w": 0.466552734375, "loss_sft": 0.22969534248113632, "loss_total_v6": 0.6962786018848419, "step": 3268 }, { "epoch": 0.7539315315315316, "grad_norm": 23.90743637084961, "learning_rate": 6.64243127609689e-06, "loss": 11.917976379394531, "loss_alignment": 0.9580078125, "loss_alignment_w": 0.47900390625, "loss_sft": 0.2668613772839308, "loss_total_v6": 0.744873471558094, "step": 3269 }, { "epoch": 0.7541621621621621, "grad_norm": 24.112178802490234, "learning_rate": 6.640529052061897e-06, "loss": 11.842851638793945, "loss_alignment": 0.923828125, "loss_alignment_w": 0.4619140625, "loss_sft": 0.27907284907996655, "loss_total_v6": 0.7401781901717186, "step": 3270 }, { "epoch": 0.7543927927927928, "grad_norm": 23.794269561767578, "learning_rate": 6.638626561896519e-06, "loss": 11.87605094909668, "loss_alignment": 0.9453125, "loss_alignment_w": 0.47265625, "loss_sft": 0.26965801790356636, "loss_total_v6": 0.7422531992197037, "step": 3271 }, { "epoch": 0.7546234234234235, "grad_norm": 29.34269905090332, "learning_rate": 6.636723805909384e-06, "loss": 12.383909225463867, "loss_alignment": 0.95263671875, "loss_alignment_w": 0.476318359375, "loss_sft": 0.2984694242477417, "loss_total_v6": 0.7739943340420723, "step": 3272 }, { "epoch": 0.754854054054054, "grad_norm": 23.191232681274414, "learning_rate": 6.634820784409163e-06, "loss": 12.180948257446289, "loss_alignment": 0.955078125, "loss_alignment_w": 0.4775390625, "loss_sft": 0.2830225098878145, "loss_total_v6": 0.7613092586398125, "step": 3273 }, { "epoch": 0.7550846846846847, "grad_norm": 24.350801467895508, "learning_rate": 6.6329174977045644e-06, "loss": 12.330839157104492, "loss_alignment": 0.95947265625, "loss_alignment_w": 0.479736328125, "loss_sft": 0.2907274980098009, "loss_total_v6": 0.7706774696707726, "step": 3274 }, { "epoch": 0.7553153153153153, "grad_norm": 23.027196884155273, "learning_rate": 6.631013946104348e-06, "loss": 12.12576961517334, "loss_alignment": 0.95556640625, "loss_alignment_w": 0.477783203125, "loss_sft": 0.2817863468080759, "loss_total_v6": 0.757860578596592, "step": 3275 }, { "epoch": 0.7555459459459459, "grad_norm": 23.32649803161621, "learning_rate": 6.629110129917309e-06, "loss": 12.163192749023438, "loss_alignment": 0.95654296875, "loss_alignment_w": 0.478271484375, "loss_sft": 0.2815618757158518, "loss_total_v6": 0.7601995542645454, "step": 3276 }, { "epoch": 0.7557765765765766, "grad_norm": 25.563215255737305, "learning_rate": 6.62720604945229e-06, "loss": 11.706274032592773, "loss_alignment": 0.9111328125, "loss_alignment_w": 0.45556640625, "loss_sft": 0.277357405051589, "loss_total_v6": 0.7316420674324036, "step": 3277 }, { "epoch": 0.7560072072072072, "grad_norm": 21.349096298217773, "learning_rate": 6.625301705018178e-06, "loss": 11.77874755859375, "loss_alignment": 0.9521484375, "loss_alignment_w": 0.47607421875, "loss_sft": 0.2613182161003351, "loss_total_v6": 0.7361717596650124, "step": 3278 }, { "epoch": 0.7562378378378378, "grad_norm": 25.677345275878906, "learning_rate": 6.623397096923895e-06, "loss": 11.682960510253906, "loss_alignment": 0.9365234375, "loss_alignment_w": 0.46826171875, "loss_sft": 0.26209117472171783, "loss_total_v6": 0.730185016989708, "step": 3279 }, { "epoch": 0.7564684684684685, "grad_norm": 25.370859146118164, "learning_rate": 6.6214922254784145e-06, "loss": 12.02648639678955, "loss_alignment": 0.94970703125, "loss_alignment_w": 0.474853515625, "loss_sft": 0.2767560575157404, "loss_total_v6": 0.7516553848981857, "step": 3280 }, { "epoch": 0.756699099099099, "grad_norm": 21.616012573242188, "learning_rate": 6.619587090990748e-06, "loss": 12.3186674118042, "loss_alignment": 0.96923828125, "loss_alignment_w": 0.484619140625, "loss_sft": 0.28660980239510536, "loss_total_v6": 0.7699167132377625, "step": 3281 }, { "epoch": 0.7569297297297297, "grad_norm": 35.340877532958984, "learning_rate": 6.617681693769951e-06, "loss": 12.338080406188965, "loss_alignment": 0.94091796875, "loss_alignment_w": 0.470458984375, "loss_sft": 0.3011898510158062, "loss_total_v6": 0.7711300477385521, "step": 3282 }, { "epoch": 0.7571603603603604, "grad_norm": 24.722843170166016, "learning_rate": 6.615776034125121e-06, "loss": 12.25283432006836, "loss_alignment": 0.9482421875, "loss_alignment_w": 0.47412109375, "loss_sft": 0.2927033733576536, "loss_total_v6": 0.7658021450042725, "step": 3283 }, { "epoch": 0.7573909909909909, "grad_norm": 23.51349639892578, "learning_rate": 6.613870112365399e-06, "loss": 12.375709533691406, "loss_alignment": 0.97119140625, "loss_alignment_w": 0.485595703125, "loss_sft": 0.28765728883445263, "loss_total_v6": 0.7734818682074547, "step": 3284 }, { "epoch": 0.7576216216216216, "grad_norm": 22.925504684448242, "learning_rate": 6.611963928799968e-06, "loss": 12.121725082397461, "loss_alignment": 0.94580078125, "loss_alignment_w": 0.472900390625, "loss_sft": 0.2843412756919861, "loss_total_v6": 0.7576078623533249, "step": 3285 }, { "epoch": 0.7578522522522523, "grad_norm": 23.42659568786621, "learning_rate": 6.6100574837380525e-06, "loss": 11.84648609161377, "loss_alignment": 0.92919921875, "loss_alignment_w": 0.464599609375, "loss_sft": 0.27574474550783634, "loss_total_v6": 0.7404054030776024, "step": 3286 }, { "epoch": 0.7580828828828828, "grad_norm": 24.230867385864258, "learning_rate": 6.608150777488922e-06, "loss": 12.109002113342285, "loss_alignment": 0.92919921875, "loss_alignment_w": 0.464599609375, "loss_sft": 0.2921519801020622, "loss_total_v6": 0.756812609732151, "step": 3287 }, { "epoch": 0.7583135135135135, "grad_norm": 23.900236129760742, "learning_rate": 6.606243810361885e-06, "loss": 12.132963180541992, "loss_alignment": 0.94189453125, "loss_alignment_w": 0.470947265625, "loss_sft": 0.28733244724571705, "loss_total_v6": 0.7583102434873581, "step": 3288 }, { "epoch": 0.7585441441441442, "grad_norm": 24.828489303588867, "learning_rate": 6.604336582666298e-06, "loss": 12.072421073913574, "loss_alignment": 0.9482421875, "loss_alignment_w": 0.47412109375, "loss_sft": 0.2806493956595659, "loss_total_v6": 0.7545263394713402, "step": 3289 }, { "epoch": 0.7587747747747747, "grad_norm": 26.96017837524414, "learning_rate": 6.602429094711549e-06, "loss": 12.73432731628418, "loss_alignment": 0.92822265625, "loss_alignment_w": 0.464111328125, "loss_sft": 0.33247081376612186, "loss_total_v6": 0.795895479619503, "step": 3290 }, { "epoch": 0.7590054054054054, "grad_norm": 26.3837833404541, "learning_rate": 6.600521346807081e-06, "loss": 12.213420867919922, "loss_alignment": 0.9423828125, "loss_alignment_w": 0.47119140625, "loss_sft": 0.2906672991812229, "loss_total_v6": 0.7633388563990593, "step": 3291 }, { "epoch": 0.7592360360360361, "grad_norm": 22.882173538208008, "learning_rate": 6.598613339262369e-06, "loss": 12.101689338684082, "loss_alignment": 0.9326171875, "loss_alignment_w": 0.46630859375, "loss_sft": 0.28972654789686203, "loss_total_v6": 0.7563555911183357, "step": 3292 }, { "epoch": 0.7594666666666666, "grad_norm": 26.200992584228516, "learning_rate": 6.596705072386938e-06, "loss": 11.969268798828125, "loss_alignment": 0.94189453125, "loss_alignment_w": 0.470947265625, "loss_sft": 0.2776813171803951, "loss_total_v6": 0.7480792999267578, "step": 3293 }, { "epoch": 0.7596972972972973, "grad_norm": 23.574892044067383, "learning_rate": 6.594796546490351e-06, "loss": 11.797431945800781, "loss_alignment": 0.92822265625, "loss_alignment_w": 0.464111328125, "loss_sft": 0.273518119007349, "loss_total_v6": 0.73733951151371, "step": 3294 }, { "epoch": 0.759927927927928, "grad_norm": 23.91178321838379, "learning_rate": 6.592887761882212e-06, "loss": 11.737396240234375, "loss_alignment": 0.9697265625, "loss_alignment_w": 0.48486328125, "loss_sft": 0.24919695314019918, "loss_total_v6": 0.7335872501134872, "step": 3295 }, { "epoch": 0.7601585585585585, "grad_norm": 23.689617156982422, "learning_rate": 6.590978718872166e-06, "loss": 12.226634979248047, "loss_alignment": 0.9150390625, "loss_alignment_w": 0.45751953125, "loss_sft": 0.3063094783574343, "loss_total_v6": 0.7641647011041641, "step": 3296 }, { "epoch": 0.7603891891891892, "grad_norm": 35.27545928955078, "learning_rate": 6.5890694177699085e-06, "loss": 11.67244815826416, "loss_alignment": 0.9326171875, "loss_alignment_w": 0.46630859375, "loss_sft": 0.2637534588575363, "loss_total_v6": 0.7295280173420906, "step": 3297 }, { "epoch": 0.7606198198198199, "grad_norm": 26.609088897705078, "learning_rate": 6.587159858885166e-06, "loss": 11.791723251342773, "loss_alignment": 0.89990234375, "loss_alignment_w": 0.449951171875, "loss_sft": 0.2870773430913687, "loss_total_v6": 0.7369827106595039, "step": 3298 }, { "epoch": 0.7608504504504504, "grad_norm": 24.773681640625, "learning_rate": 6.585250042527712e-06, "loss": 11.893524169921875, "loss_alignment": 0.9697265625, "loss_alignment_w": 0.48486328125, "loss_sft": 0.2586955353617668, "loss_total_v6": 0.743345208466053, "step": 3299 }, { "epoch": 0.7610810810810811, "grad_norm": 24.038759231567383, "learning_rate": 6.583339969007364e-06, "loss": 12.232856750488281, "loss_alignment": 0.94970703125, "loss_alignment_w": 0.474853515625, "loss_sft": 0.28904391825199127, "loss_total_v6": 0.7645535543560982, "step": 3300 }, { "epoch": 0.7613117117117117, "grad_norm": 24.120872497558594, "learning_rate": 6.581429638633975e-06, "loss": 12.156084060668945, "loss_alignment": 0.9423828125, "loss_alignment_w": 0.47119140625, "loss_sft": 0.2894335947930813, "loss_total_v6": 0.7597552537918091, "step": 3301 }, { "epoch": 0.7615423423423423, "grad_norm": 21.62392234802246, "learning_rate": 6.5795190517174446e-06, "loss": 11.68124771118164, "loss_alignment": 0.93505859375, "loss_alignment_w": 0.467529296875, "loss_sft": 0.2619230952113867, "loss_total_v6": 0.7300779819488525, "step": 3302 }, { "epoch": 0.761772972972973, "grad_norm": 22.222307205200195, "learning_rate": 6.577608208567714e-06, "loss": 12.091753005981445, "loss_alignment": 0.93798828125, "loss_alignment_w": 0.468994140625, "loss_sft": 0.2865725476294756, "loss_total_v6": 0.7557345330715179, "step": 3303 }, { "epoch": 0.7620036036036036, "grad_norm": 24.573593139648438, "learning_rate": 6.575697109494763e-06, "loss": 11.862342834472656, "loss_alignment": 0.9306640625, "loss_alignment_w": 0.46533203125, "loss_sft": 0.2758507840335369, "loss_total_v6": 0.7413964197039604, "step": 3304 }, { "epoch": 0.7622342342342342, "grad_norm": 23.523845672607422, "learning_rate": 6.573785754808615e-06, "loss": 11.965429306030273, "loss_alignment": 0.91796875, "loss_alignment_w": 0.458984375, "loss_sft": 0.2883361466228962, "loss_total_v6": 0.7478393167257309, "step": 3305 }, { "epoch": 0.7624648648648649, "grad_norm": 22.381656646728516, "learning_rate": 6.571874144819333e-06, "loss": 11.950428009033203, "loss_alignment": 0.95068359375, "loss_alignment_w": 0.475341796875, "loss_sft": 0.2710106149315834, "loss_total_v6": 0.7469017282128334, "step": 3306 }, { "epoch": 0.7626954954954955, "grad_norm": 24.503767013549805, "learning_rate": 6.569962279837025e-06, "loss": 12.475357055664062, "loss_alignment": 0.93701171875, "loss_alignment_w": 0.468505859375, "loss_sft": 0.3106088750064373, "loss_total_v6": 0.7797098159790039, "step": 3307 }, { "epoch": 0.7629261261261261, "grad_norm": 25.092348098754883, "learning_rate": 6.568050160171837e-06, "loss": 12.005678176879883, "loss_alignment": 0.9453125, "loss_alignment_w": 0.47265625, "loss_sft": 0.27757659554481506, "loss_total_v6": 0.7503549084067345, "step": 3308 }, { "epoch": 0.7631567567567568, "grad_norm": 22.508716583251953, "learning_rate": 6.566137786133957e-06, "loss": 12.092827796936035, "loss_alignment": 0.95947265625, "loss_alignment_w": 0.479736328125, "loss_sft": 0.2751193791627884, "loss_total_v6": 0.7558017522096634, "step": 3309 }, { "epoch": 0.7633873873873874, "grad_norm": 24.1284122467041, "learning_rate": 6.564225158033616e-06, "loss": 12.21178150177002, "loss_alignment": 0.93359375, "loss_alignment_w": 0.466796875, "loss_sft": 0.29538662545382977, "loss_total_v6": 0.7632363438606262, "step": 3310 }, { "epoch": 0.763618018018018, "grad_norm": 24.096473693847656, "learning_rate": 6.562312276181083e-06, "loss": 12.445234298706055, "loss_alignment": 0.94189453125, "loss_alignment_w": 0.470947265625, "loss_sft": 0.3067120090126991, "loss_total_v6": 0.7778271436691284, "step": 3311 }, { "epoch": 0.7638486486486487, "grad_norm": 22.9943790435791, "learning_rate": 6.560399140886673e-06, "loss": 12.18989086151123, "loss_alignment": 0.9453125, "loss_alignment_w": 0.47265625, "loss_sft": 0.28878474421799183, "loss_total_v6": 0.7618682160973549, "step": 3312 }, { "epoch": 0.7640792792792793, "grad_norm": 25.476144790649414, "learning_rate": 6.558485752460735e-06, "loss": 12.195615768432617, "loss_alignment": 0.93212890625, "loss_alignment_w": 0.466064453125, "loss_sft": 0.2962073013186455, "loss_total_v6": 0.7622260004281998, "step": 3313 }, { "epoch": 0.7643099099099099, "grad_norm": 24.240407943725586, "learning_rate": 6.556572111213667e-06, "loss": 11.905435562133789, "loss_alignment": 0.92236328125, "loss_alignment_w": 0.461181640625, "loss_sft": 0.28309115022420883, "loss_total_v6": 0.7440896779298782, "step": 3314 }, { "epoch": 0.7645405405405405, "grad_norm": 27.08007049560547, "learning_rate": 6.554658217455902e-06, "loss": 12.123445510864258, "loss_alignment": 0.96875, "loss_alignment_w": 0.484375, "loss_sft": 0.27392017655074596, "loss_total_v6": 0.7577153369784355, "step": 3315 }, { "epoch": 0.7647711711711712, "grad_norm": 23.430221557617188, "learning_rate": 6.552744071497918e-06, "loss": 11.499467849731445, "loss_alignment": 0.9306640625, "loss_alignment_w": 0.46533203125, "loss_sft": 0.2528964262455702, "loss_total_v6": 0.7187167257070541, "step": 3316 }, { "epoch": 0.7650018018018018, "grad_norm": 25.4312686920166, "learning_rate": 6.550829673650231e-06, "loss": 12.210732460021973, "loss_alignment": 0.91064453125, "loss_alignment_w": 0.455322265625, "loss_sft": 0.30710080079734325, "loss_total_v6": 0.7631707787513733, "step": 3317 }, { "epoch": 0.7652324324324324, "grad_norm": 25.09666633605957, "learning_rate": 6.548915024223399e-06, "loss": 12.737603187561035, "loss_alignment": 0.93896484375, "loss_alignment_w": 0.469482421875, "loss_sft": 0.32648043520748615, "loss_total_v6": 0.7961001843214035, "step": 3318 }, { "epoch": 0.7654630630630631, "grad_norm": 24.345998764038086, "learning_rate": 6.547000123528022e-06, "loss": 12.002331733703613, "loss_alignment": 0.9404296875, "loss_alignment_w": 0.47021484375, "loss_sft": 0.28038865700364113, "loss_total_v6": 0.7501457408070564, "step": 3319 }, { "epoch": 0.7656936936936937, "grad_norm": 24.16305160522461, "learning_rate": 6.545084971874738e-06, "loss": 11.371203422546387, "loss_alignment": 0.9287109375, "loss_alignment_w": 0.46435546875, "loss_sft": 0.24591747857630253, "loss_total_v6": 0.7107001915574074, "step": 3320 }, { "epoch": 0.7659243243243243, "grad_norm": 27.59426498413086, "learning_rate": 6.543169569574228e-06, "loss": 12.802237510681152, "loss_alignment": 0.9501953125, "loss_alignment_w": 0.47509765625, "loss_sft": 0.3258509673178196, "loss_total_v6": 0.8001398667693138, "step": 3321 }, { "epoch": 0.766154954954955, "grad_norm": 26.462526321411133, "learning_rate": 6.541253916937214e-06, "loss": 11.956613540649414, "loss_alignment": 0.9521484375, "loss_alignment_w": 0.47607421875, "loss_sft": 0.2712141592055559, "loss_total_v6": 0.7472883760929108, "step": 3322 }, { "epoch": 0.7663855855855856, "grad_norm": 23.472999572753906, "learning_rate": 6.539338014274454e-06, "loss": 11.557766914367676, "loss_alignment": 0.92919921875, "loss_alignment_w": 0.464599609375, "loss_sft": 0.25791345350444317, "loss_total_v6": 0.7223604470491409, "step": 3323 }, { "epoch": 0.7666162162162162, "grad_norm": 23.8176212310791, "learning_rate": 6.537421861896754e-06, "loss": 11.773395538330078, "loss_alignment": 0.9453125, "loss_alignment_w": 0.47265625, "loss_sft": 0.2627232186496258, "loss_total_v6": 0.7358372583985329, "step": 3324 }, { "epoch": 0.7668468468468469, "grad_norm": 23.5146484375, "learning_rate": 6.5355054601149545e-06, "loss": 11.986739158630371, "loss_alignment": 0.9521484375, "loss_alignment_w": 0.47607421875, "loss_sft": 0.2737073563039303, "loss_total_v6": 0.7491712272167206, "step": 3325 }, { "epoch": 0.7670774774774775, "grad_norm": 26.80533790588379, "learning_rate": 6.5335888092399415e-06, "loss": 12.047881126403809, "loss_alignment": 0.958984375, "loss_alignment_w": 0.4794921875, "loss_sft": 0.2725085262209177, "loss_total_v6": 0.7529925629496574, "step": 3326 }, { "epoch": 0.7673081081081081, "grad_norm": 25.056621551513672, "learning_rate": 6.531671909582634e-06, "loss": 12.890304565429688, "loss_alignment": 0.9580078125, "loss_alignment_w": 0.47900390625, "loss_sft": 0.32737254351377487, "loss_total_v6": 0.8056440204381943, "step": 3327 }, { "epoch": 0.7675387387387388, "grad_norm": 25.265073776245117, "learning_rate": 6.529754761453999e-06, "loss": 11.756213188171387, "loss_alignment": 0.9482421875, "loss_alignment_w": 0.47412109375, "loss_sft": 0.2603523004800081, "loss_total_v6": 0.7347633466124535, "step": 3328 }, { "epoch": 0.7677693693693693, "grad_norm": 25.801862716674805, "learning_rate": 6.52783736516504e-06, "loss": 11.942366600036621, "loss_alignment": 0.9365234375, "loss_alignment_w": 0.46826171875, "loss_sft": 0.27822774462401867, "loss_total_v6": 0.7463979125022888, "step": 3329 }, { "epoch": 0.768, "grad_norm": 25.335046768188477, "learning_rate": 6.5259197210268e-06, "loss": 12.501441955566406, "loss_alignment": 0.97119140625, "loss_alignment_w": 0.485595703125, "loss_sft": 0.295576561242342, "loss_total_v6": 0.7813401073217392, "step": 3330 }, { "epoch": 0.7682306306306307, "grad_norm": 31.395212173461914, "learning_rate": 6.524001829350366e-06, "loss": 12.258512496948242, "loss_alignment": 0.9091796875, "loss_alignment_w": 0.45458984375, "loss_sft": 0.310804245993495, "loss_total_v6": 0.7661570459604263, "step": 3331 }, { "epoch": 0.7684612612612612, "grad_norm": 24.117799758911133, "learning_rate": 6.522083690446863e-06, "loss": 12.13093376159668, "loss_alignment": 0.9140625, "loss_alignment_w": 0.45703125, "loss_sft": 0.30046542175114155, "loss_total_v6": 0.7581833228468895, "step": 3332 }, { "epoch": 0.7686918918918919, "grad_norm": 23.76357078552246, "learning_rate": 6.520165304627452e-06, "loss": 12.506250381469727, "loss_alignment": 0.96728515625, "loss_alignment_w": 0.483642578125, "loss_sft": 0.2967468425631523, "loss_total_v6": 0.7816406115889549, "step": 3333 }, { "epoch": 0.7689225225225226, "grad_norm": 22.4146785736084, "learning_rate": 6.5182466722033426e-06, "loss": 12.381143569946289, "loss_alignment": 0.96142578125, "loss_alignment_w": 0.480712890625, "loss_sft": 0.2922846581786871, "loss_total_v6": 0.773821510374546, "step": 3334 }, { "epoch": 0.7691531531531531, "grad_norm": 21.802234649658203, "learning_rate": 6.516327793485776e-06, "loss": 11.229228019714355, "loss_alignment": 0.923828125, "loss_alignment_w": 0.4619140625, "loss_sft": 0.24050773680210114, "loss_total_v6": 0.7018267288804054, "step": 3335 }, { "epoch": 0.7693837837837838, "grad_norm": 24.175437927246094, "learning_rate": 6.5144086687860385e-06, "loss": 12.260683059692383, "loss_alignment": 0.9130859375, "loss_alignment_w": 0.45654296875, "loss_sft": 0.30991761572659016, "loss_total_v6": 0.7662927061319351, "step": 3336 }, { "epoch": 0.7696144144144144, "grad_norm": 23.95783805847168, "learning_rate": 6.512489298415457e-06, "loss": 11.990274429321289, "loss_alignment": 0.92041015625, "loss_alignment_w": 0.460205078125, "loss_sft": 0.2887598145753145, "loss_total_v6": 0.7493921518325806, "step": 3337 }, { "epoch": 0.769845045045045, "grad_norm": 25.314760208129883, "learning_rate": 6.5105696826853935e-06, "loss": 12.206236839294434, "loss_alignment": 0.93505859375, "loss_alignment_w": 0.467529296875, "loss_sft": 0.296001348644495, "loss_total_v6": 0.7628897875547409, "step": 3338 }, { "epoch": 0.7700756756756757, "grad_norm": 22.979894638061523, "learning_rate": 6.508649821907254e-06, "loss": 11.991012573242188, "loss_alignment": 0.9267578125, "loss_alignment_w": 0.46337890625, "loss_sft": 0.2865934446454048, "loss_total_v6": 0.7494382709264755, "step": 3339 }, { "epoch": 0.7703063063063063, "grad_norm": 28.112754821777344, "learning_rate": 6.50672971639248e-06, "loss": 12.076518058776855, "loss_alignment": 0.93505859375, "loss_alignment_w": 0.467529296875, "loss_sft": 0.28668851777911186, "loss_total_v6": 0.7547823786735535, "step": 3340 }, { "epoch": 0.7705369369369369, "grad_norm": 27.85814666748047, "learning_rate": 6.50480936645256e-06, "loss": 12.358945846557617, "loss_alignment": 0.96435546875, "loss_alignment_w": 0.482177734375, "loss_sft": 0.2904852665960789, "loss_total_v6": 0.7724341154098511, "step": 3341 }, { "epoch": 0.7707675675675676, "grad_norm": 24.721410751342773, "learning_rate": 6.502888772399013e-06, "loss": 11.58290958404541, "loss_alignment": 0.92578125, "loss_alignment_w": 0.462890625, "loss_sft": 0.25989681482315063, "loss_total_v6": 0.7239318415522575, "step": 3342 }, { "epoch": 0.7709981981981981, "grad_norm": 22.89617919921875, "learning_rate": 6.5009679345434056e-06, "loss": 11.607609748840332, "loss_alignment": 0.9150390625, "loss_alignment_w": 0.45751953125, "loss_sft": 0.2669185344129801, "loss_total_v6": 0.7254756391048431, "step": 3343 }, { "epoch": 0.7712288288288288, "grad_norm": 24.652694702148438, "learning_rate": 6.499046853197338e-06, "loss": 12.099960327148438, "loss_alignment": 0.923828125, "loss_alignment_w": 0.4619140625, "loss_sft": 0.2951879482716322, "loss_total_v6": 0.756247490644455, "step": 3344 }, { "epoch": 0.7714594594594595, "grad_norm": 25.20789909362793, "learning_rate": 6.497125528672453e-06, "loss": 11.660985946655273, "loss_alignment": 0.91259765625, "loss_alignment_w": 0.456298828125, "loss_sft": 0.2709869220852852, "loss_total_v6": 0.7288116365671158, "step": 3345 }, { "epoch": 0.77169009009009, "grad_norm": 23.305997848510742, "learning_rate": 6.495203961280434e-06, "loss": 12.225688934326172, "loss_alignment": 0.978515625, "loss_alignment_w": 0.4892578125, "loss_sft": 0.2746798936277628, "loss_total_v6": 0.764105536043644, "step": 3346 }, { "epoch": 0.7719207207207207, "grad_norm": 24.184524536132812, "learning_rate": 6.493282151333e-06, "loss": 12.399608612060547, "loss_alignment": 0.93896484375, "loss_alignment_w": 0.469482421875, "loss_sft": 0.3053252808749676, "loss_total_v6": 0.7749755457043648, "step": 3347 }, { "epoch": 0.7721513513513514, "grad_norm": 23.740537643432617, "learning_rate": 6.491360099141913e-06, "loss": 12.203022003173828, "loss_alignment": 0.943359375, "loss_alignment_w": 0.4716796875, "loss_sft": 0.2909634094685316, "loss_total_v6": 0.7626888751983643, "step": 3348 }, { "epoch": 0.7723819819819819, "grad_norm": 23.628318786621094, "learning_rate": 6.489437805018972e-06, "loss": 11.925220489501953, "loss_alignment": 0.943359375, "loss_alignment_w": 0.4716796875, "loss_sft": 0.2751266937702894, "loss_total_v6": 0.7453262880444527, "step": 3349 }, { "epoch": 0.7726126126126126, "grad_norm": 22.0504150390625, "learning_rate": 6.487515269276015e-06, "loss": 11.511236190795898, "loss_alignment": 0.912109375, "loss_alignment_w": 0.4560546875, "loss_sft": 0.2626499105244875, "loss_total_v6": 0.719452254474163, "step": 3350 }, { "epoch": 0.7728432432432433, "grad_norm": 26.009231567382812, "learning_rate": 6.485592492224923e-06, "loss": 12.270362854003906, "loss_alignment": 0.9521484375, "loss_alignment_w": 0.47607421875, "loss_sft": 0.2916474938392639, "loss_total_v6": 0.7668977230787277, "step": 3351 }, { "epoch": 0.7730738738738738, "grad_norm": 24.842660903930664, "learning_rate": 6.483669474177609e-06, "loss": 12.107733726501465, "loss_alignment": 0.94384765625, "loss_alignment_w": 0.471923828125, "loss_sft": 0.2836193200200796, "loss_total_v6": 0.756733350455761, "step": 3352 }, { "epoch": 0.7733045045045045, "grad_norm": 24.500818252563477, "learning_rate": 6.481746215446033e-06, "loss": 11.580708503723145, "loss_alignment": 0.939453125, "loss_alignment_w": 0.4697265625, "loss_sft": 0.255105322226882, "loss_total_v6": 0.7237942665815353, "step": 3353 }, { "epoch": 0.7735351351351352, "grad_norm": 26.254880905151367, "learning_rate": 6.47982271634219e-06, "loss": 11.969499588012695, "loss_alignment": 0.9248046875, "loss_alignment_w": 0.46240234375, "loss_sft": 0.28584392555058, "loss_total_v6": 0.7480937242507935, "step": 3354 }, { "epoch": 0.7737657657657657, "grad_norm": 23.076942443847656, "learning_rate": 6.477898977178112e-06, "loss": 12.403119087219238, "loss_alignment": 0.951171875, "loss_alignment_w": 0.4755859375, "loss_sft": 0.3008754923939705, "loss_total_v6": 0.7751949355006218, "step": 3355 }, { "epoch": 0.7739963963963964, "grad_norm": 23.566064834594727, "learning_rate": 6.475974998265875e-06, "loss": 12.069372177124023, "loss_alignment": 0.93212890625, "loss_alignment_w": 0.466064453125, "loss_sft": 0.2875998690724373, "loss_total_v6": 0.7543357089161873, "step": 3356 }, { "epoch": 0.7742270270270271, "grad_norm": 26.728790283203125, "learning_rate": 6.47405077991759e-06, "loss": 12.022132873535156, "loss_alignment": 0.9375, "loss_alignment_w": 0.46875, "loss_sft": 0.28403712064027786, "loss_total_v6": 0.7513833269476891, "step": 3357 }, { "epoch": 0.7744576576576576, "grad_norm": 23.454551696777344, "learning_rate": 6.472126322445409e-06, "loss": 11.972832679748535, "loss_alignment": 0.9404296875, "loss_alignment_w": 0.47021484375, "loss_sft": 0.27894171699881554, "loss_total_v6": 0.7483020573854446, "step": 3358 }, { "epoch": 0.7746882882882883, "grad_norm": 22.797948837280273, "learning_rate": 6.47020162616152e-06, "loss": 12.219001770019531, "loss_alignment": 0.92333984375, "loss_alignment_w": 0.461669921875, "loss_sft": 0.3018345460295677, "loss_total_v6": 0.7636875882744789, "step": 3359 }, { "epoch": 0.774918918918919, "grad_norm": 26.073732376098633, "learning_rate": 6.468276691378155e-06, "loss": 12.329014778137207, "loss_alignment": 0.93017578125, "loss_alignment_w": 0.465087890625, "loss_sft": 0.3051856458187103, "loss_total_v6": 0.7705634534358978, "step": 3360 }, { "epoch": 0.7751495495495495, "grad_norm": 27.19049835205078, "learning_rate": 6.466351518407578e-06, "loss": 12.020292282104492, "loss_alignment": 0.93115234375, "loss_alignment_w": 0.465576171875, "loss_sft": 0.2854174915701151, "loss_total_v6": 0.7512682974338531, "step": 3361 }, { "epoch": 0.7753801801801802, "grad_norm": 24.232637405395508, "learning_rate": 6.464426107562095e-06, "loss": 11.922264099121094, "loss_alignment": 0.92138671875, "loss_alignment_w": 0.460693359375, "loss_sft": 0.28545526787638664, "loss_total_v6": 0.7451415359973907, "step": 3362 }, { "epoch": 0.7756108108108108, "grad_norm": 28.161054611206055, "learning_rate": 6.462500459154054e-06, "loss": 12.084891319274902, "loss_alignment": 0.90478515625, "loss_alignment_w": 0.452392578125, "loss_sft": 0.30356927402317524, "loss_total_v6": 0.7553057298064232, "step": 3363 }, { "epoch": 0.7758414414414414, "grad_norm": 26.58737564086914, "learning_rate": 6.460574573495835e-06, "loss": 12.198152542114258, "loss_alignment": 0.9111328125, "loss_alignment_w": 0.45556640625, "loss_sft": 0.3079930543899536, "loss_total_v6": 0.7623845487833023, "step": 3364 }, { "epoch": 0.7760720720720721, "grad_norm": 24.540584564208984, "learning_rate": 6.45864845089986e-06, "loss": 12.068424224853516, "loss_alignment": 0.95361328125, "loss_alignment_w": 0.476806640625, "loss_sft": 0.27840061858296394, "loss_total_v6": 0.7542764917016029, "step": 3365 }, { "epoch": 0.7763027027027027, "grad_norm": 26.41848373413086, "learning_rate": 6.456722091678591e-06, "loss": 12.501559257507324, "loss_alignment": 0.94091796875, "loss_alignment_w": 0.470458984375, "loss_sft": 0.31046124547719955, "loss_total_v6": 0.781347468495369, "step": 3366 }, { "epoch": 0.7765333333333333, "grad_norm": 28.37410545349121, "learning_rate": 6.4547954961445235e-06, "loss": 11.840932846069336, "loss_alignment": 0.92626953125, "loss_alignment_w": 0.463134765625, "loss_sft": 0.27806790731847286, "loss_total_v6": 0.7400582805275917, "step": 3367 }, { "epoch": 0.776763963963964, "grad_norm": 22.45695686340332, "learning_rate": 6.452868664610197e-06, "loss": 11.765730857849121, "loss_alignment": 0.94873046875, "loss_alignment_w": 0.474365234375, "loss_sft": 0.2602147478610277, "loss_total_v6": 0.7353581786155701, "step": 3368 }, { "epoch": 0.7769945945945946, "grad_norm": 25.54767608642578, "learning_rate": 6.450941597388184e-06, "loss": 11.504382133483887, "loss_alignment": 0.9111328125, "loss_alignment_w": 0.45556640625, "loss_sft": 0.2639915067702532, "loss_total_v6": 0.7190238460898399, "step": 3369 }, { "epoch": 0.7772252252252252, "grad_norm": 25.618160247802734, "learning_rate": 6.449014294791101e-06, "loss": 12.29353141784668, "loss_alignment": 0.93212890625, "loss_alignment_w": 0.466064453125, "loss_sft": 0.30189976654946804, "loss_total_v6": 0.7683456838130951, "step": 3370 }, { "epoch": 0.7774558558558559, "grad_norm": 23.998926162719727, "learning_rate": 6.447086757131596e-06, "loss": 12.150742530822754, "loss_alignment": 0.9541015625, "loss_alignment_w": 0.47705078125, "loss_sft": 0.28188238106667995, "loss_total_v6": 0.7594214156270027, "step": 3371 }, { "epoch": 0.7776864864864865, "grad_norm": 28.01923179626465, "learning_rate": 6.445158984722358e-06, "loss": 12.852972030639648, "loss_alignment": 0.943359375, "loss_alignment_w": 0.4716796875, "loss_sft": 0.33175307139754295, "loss_total_v6": 0.8033106923103333, "step": 3372 }, { "epoch": 0.7779171171171171, "grad_norm": 24.20229721069336, "learning_rate": 6.4432309778761205e-06, "loss": 12.134614944458008, "loss_alignment": 0.9462890625, "loss_alignment_w": 0.47314453125, "loss_sft": 0.2846127636730671, "loss_total_v6": 0.7584133967757225, "step": 3373 }, { "epoch": 0.7781477477477478, "grad_norm": 23.323627471923828, "learning_rate": 6.441302736905641e-06, "loss": 11.776622772216797, "loss_alignment": 0.923828125, "loss_alignment_w": 0.4619140625, "loss_sft": 0.27421638555824757, "loss_total_v6": 0.7360389158129692, "step": 3374 }, { "epoch": 0.7783783783783784, "grad_norm": 22.485965728759766, "learning_rate": 6.439374262123731e-06, "loss": 11.937152862548828, "loss_alignment": 0.94287109375, "loss_alignment_w": 0.471435546875, "loss_sft": 0.27416348084807396, "loss_total_v6": 0.7460720464587212, "step": 3375 }, { "epoch": 0.778609009009009, "grad_norm": 24.628480911254883, "learning_rate": 6.437445553843229e-06, "loss": 12.07953929901123, "loss_alignment": 0.9384765625, "loss_alignment_w": 0.46923828125, "loss_sft": 0.2852903641760349, "loss_total_v6": 0.7549711838364601, "step": 3376 }, { "epoch": 0.7788396396396396, "grad_norm": 23.26119041442871, "learning_rate": 6.435516612377011e-06, "loss": 11.944055557250977, "loss_alignment": 0.943359375, "loss_alignment_w": 0.4716796875, "loss_sft": 0.27441173419356346, "loss_total_v6": 0.7465034127235413, "step": 3377 }, { "epoch": 0.7790702702702703, "grad_norm": 25.265993118286133, "learning_rate": 6.4335874380380005e-06, "loss": 12.024860382080078, "loss_alignment": 0.93017578125, "loss_alignment_w": 0.465087890625, "loss_sft": 0.2860081326216459, "loss_total_v6": 0.7515538111329079, "step": 3378 }, { "epoch": 0.7793009009009009, "grad_norm": 24.91043472290039, "learning_rate": 6.431658031139148e-06, "loss": 11.970369338989258, "loss_alignment": 0.9306640625, "loss_alignment_w": 0.46533203125, "loss_sft": 0.2827092073857784, "loss_total_v6": 0.7481480464339256, "step": 3379 }, { "epoch": 0.7795315315315315, "grad_norm": 24.51613426208496, "learning_rate": 6.429728391993446e-06, "loss": 12.22940444946289, "loss_alignment": 0.95361328125, "loss_alignment_w": 0.476806640625, "loss_sft": 0.2874396424740553, "loss_total_v6": 0.764337807893753, "step": 3380 }, { "epoch": 0.7797621621621622, "grad_norm": 24.41210174560547, "learning_rate": 6.427798520913929e-06, "loss": 11.310052871704102, "loss_alignment": 0.92333984375, "loss_alignment_w": 0.461669921875, "loss_sft": 0.2443386521190405, "loss_total_v6": 0.7068783342838287, "step": 3381 }, { "epoch": 0.7799927927927928, "grad_norm": 23.63738250732422, "learning_rate": 6.425868418213663e-06, "loss": 11.86609172821045, "loss_alignment": 0.92431640625, "loss_alignment_w": 0.462158203125, "loss_sft": 0.2779466584324837, "loss_total_v6": 0.7416307404637337, "step": 3382 }, { "epoch": 0.7802234234234234, "grad_norm": 25.810501098632812, "learning_rate": 6.423938084205753e-06, "loss": 12.036242485046387, "loss_alignment": 0.95068359375, "loss_alignment_w": 0.475341796875, "loss_sft": 0.27732008323073387, "loss_total_v6": 0.7522651553153992, "step": 3383 }, { "epoch": 0.7804540540540541, "grad_norm": 27.250364303588867, "learning_rate": 6.4220075192033425e-06, "loss": 12.031444549560547, "loss_alignment": 0.919921875, "loss_alignment_w": 0.4599609375, "loss_sft": 0.2911956403404474, "loss_total_v6": 0.7519652470946312, "step": 3384 }, { "epoch": 0.7806846846846847, "grad_norm": 25.265220642089844, "learning_rate": 6.420076723519615e-06, "loss": 11.976219177246094, "loss_alignment": 0.95068359375, "loss_alignment_w": 0.475341796875, "loss_sft": 0.27262259274721146, "loss_total_v6": 0.7485136985778809, "step": 3385 }, { "epoch": 0.7809153153153153, "grad_norm": 23.709253311157227, "learning_rate": 6.418145697467784e-06, "loss": 11.63353443145752, "loss_alignment": 0.92626953125, "loss_alignment_w": 0.463134765625, "loss_sft": 0.2637933362275362, "loss_total_v6": 0.7270959317684174, "step": 3386 }, { "epoch": 0.781145945945946, "grad_norm": 22.547496795654297, "learning_rate": 6.416214441361106e-06, "loss": 11.145055770874023, "loss_alignment": 0.93505859375, "loss_alignment_w": 0.467529296875, "loss_sft": 0.22946393489837646, "loss_total_v6": 0.6965660005807877, "step": 3387 }, { "epoch": 0.7813765765765766, "grad_norm": 24.849029541015625, "learning_rate": 6.414282955512876e-06, "loss": 12.20143985748291, "loss_alignment": 0.94140625, "loss_alignment_w": 0.470703125, "loss_sft": 0.2922378461807966, "loss_total_v6": 0.7625899910926819, "step": 3388 }, { "epoch": 0.7816072072072072, "grad_norm": 31.315128326416016, "learning_rate": 6.412351240236423e-06, "loss": 11.72204303741455, "loss_alignment": 0.94091796875, "loss_alignment_w": 0.470458984375, "loss_sft": 0.26238235272467136, "loss_total_v6": 0.7326277196407318, "step": 3389 }, { "epoch": 0.7818378378378379, "grad_norm": 25.38868522644043, "learning_rate": 6.410419295845114e-06, "loss": 12.31071662902832, "loss_alignment": 0.96142578125, "loss_alignment_w": 0.480712890625, "loss_sft": 0.2888442184776068, "loss_total_v6": 0.7694198116660118, "step": 3390 }, { "epoch": 0.7820684684684684, "grad_norm": 25.225208282470703, "learning_rate": 6.408487122652353e-06, "loss": 12.360934257507324, "loss_alignment": 0.93017578125, "loss_alignment_w": 0.465087890625, "loss_sft": 0.30756204575300217, "loss_total_v6": 0.7725583910942078, "step": 3391 }, { "epoch": 0.7822990990990991, "grad_norm": 24.69028663635254, "learning_rate": 6.406554720971583e-06, "loss": 12.263188362121582, "loss_alignment": 0.93115234375, "loss_alignment_w": 0.465576171875, "loss_sft": 0.30180390179157257, "loss_total_v6": 0.7664492651820183, "step": 3392 }, { "epoch": 0.7825297297297298, "grad_norm": 22.432044982910156, "learning_rate": 6.404622091116282e-06, "loss": 11.399042129516602, "loss_alignment": 0.93115234375, "loss_alignment_w": 0.465576171875, "loss_sft": 0.24721498601138592, "loss_total_v6": 0.7124401703476906, "step": 3393 }, { "epoch": 0.7827603603603603, "grad_norm": 24.774681091308594, "learning_rate": 6.402689233399965e-06, "loss": 11.59225845336914, "loss_alignment": 0.91943359375, "loss_alignment_w": 0.459716796875, "loss_sft": 0.2639753632247448, "loss_total_v6": 0.7245161384344101, "step": 3394 }, { "epoch": 0.782990990990991, "grad_norm": 25.67101287841797, "learning_rate": 6.400756148136185e-06, "loss": 11.51749038696289, "loss_alignment": 0.94189453125, "loss_alignment_w": 0.470947265625, "loss_sft": 0.24968935549259186, "loss_total_v6": 0.719843178987503, "step": 3395 }, { "epoch": 0.7832216216216216, "grad_norm": 24.950624465942383, "learning_rate": 6.398822835638531e-06, "loss": 11.972784996032715, "loss_alignment": 0.92041015625, "loss_alignment_w": 0.460205078125, "loss_sft": 0.28736154921352863, "loss_total_v6": 0.7482990622520447, "step": 3396 }, { "epoch": 0.7834522522522522, "grad_norm": 24.962501525878906, "learning_rate": 6.396889296220631e-06, "loss": 11.913040161132812, "loss_alignment": 0.90966796875, "loss_alignment_w": 0.454833984375, "loss_sft": 0.2903260998427868, "loss_total_v6": 0.744564987719059, "step": 3397 }, { "epoch": 0.7836828828828829, "grad_norm": 24.356510162353516, "learning_rate": 6.3949555301961474e-06, "loss": 12.46340560913086, "loss_alignment": 0.94677734375, "loss_alignment_w": 0.473388671875, "loss_sft": 0.30452135019004345, "loss_total_v6": 0.7789628803730011, "step": 3398 }, { "epoch": 0.7839135135135135, "grad_norm": 23.245986938476562, "learning_rate": 6.393021537878778e-06, "loss": 12.358570098876953, "loss_alignment": 0.9228515625, "loss_alignment_w": 0.46142578125, "loss_sft": 0.3111526481807232, "loss_total_v6": 0.7724105939269066, "step": 3399 }, { "epoch": 0.7841441441441441, "grad_norm": 23.880537033081055, "learning_rate": 6.391087319582264e-06, "loss": 12.282825469970703, "loss_alignment": 0.94384765625, "loss_alignment_w": 0.471923828125, "loss_sft": 0.2960732374340296, "loss_total_v6": 0.7676766365766525, "step": 3400 }, { "epoch": 0.7843747747747748, "grad_norm": 22.786731719970703, "learning_rate": 6.3891528756203746e-06, "loss": 11.807833671569824, "loss_alignment": 0.94921875, "loss_alignment_w": 0.474609375, "loss_sft": 0.2645857036113739, "loss_total_v6": 0.7379896193742752, "step": 3401 }, { "epoch": 0.7846054054054054, "grad_norm": 25.136215209960938, "learning_rate": 6.387218206306922e-06, "loss": 11.946357727050781, "loss_alignment": 0.90869140625, "loss_alignment_w": 0.454345703125, "loss_sft": 0.2930646017193794, "loss_total_v6": 0.7466473802924156, "step": 3402 }, { "epoch": 0.784836036036036, "grad_norm": 27.041793823242188, "learning_rate": 6.385283311955752e-06, "loss": 11.861871719360352, "loss_alignment": 0.92919921875, "loss_alignment_w": 0.464599609375, "loss_sft": 0.27759138867259026, "loss_total_v6": 0.7413670122623444, "step": 3403 }, { "epoch": 0.7850666666666667, "grad_norm": 22.09276008605957, "learning_rate": 6.383348192880747e-06, "loss": 11.715744018554688, "loss_alignment": 0.91796875, "loss_alignment_w": 0.458984375, "loss_sft": 0.27331069484353065, "loss_total_v6": 0.732234038412571, "step": 3404 }, { "epoch": 0.7852972972972972, "grad_norm": 26.911396026611328, "learning_rate": 6.381412849395827e-06, "loss": 12.455714225769043, "loss_alignment": 0.95654296875, "loss_alignment_w": 0.478271484375, "loss_sft": 0.3014618866145611, "loss_total_v6": 0.7784821465611458, "step": 3405 }, { "epoch": 0.7855279279279279, "grad_norm": 25.728792190551758, "learning_rate": 6.379477281814948e-06, "loss": 11.854145050048828, "loss_alignment": 0.921875, "loss_alignment_w": 0.4609375, "loss_sft": 0.28048059344291687, "loss_total_v6": 0.7408840283751488, "step": 3406 }, { "epoch": 0.7857585585585586, "grad_norm": 24.80284881591797, "learning_rate": 6.377541490452103e-06, "loss": 11.429733276367188, "loss_alignment": 0.90283203125, "loss_alignment_w": 0.451416015625, "loss_sft": 0.2628507725894451, "loss_total_v6": 0.7143583372235298, "step": 3407 }, { "epoch": 0.7859891891891891, "grad_norm": 25.273780822753906, "learning_rate": 6.375605475621318e-06, "loss": 12.019843101501465, "loss_alignment": 0.908203125, "loss_alignment_w": 0.4541015625, "loss_sft": 0.29733701050281525, "loss_total_v6": 0.7512402087450027, "step": 3408 }, { "epoch": 0.7862198198198198, "grad_norm": 26.567842483520508, "learning_rate": 6.373669237636659e-06, "loss": 11.814226150512695, "loss_alignment": 0.9326171875, "loss_alignment_w": 0.46630859375, "loss_sft": 0.2718516420572996, "loss_total_v6": 0.7383891120553017, "step": 3409 }, { "epoch": 0.7864504504504505, "grad_norm": 25.31184959411621, "learning_rate": 6.37173277681223e-06, "loss": 12.328485488891602, "loss_alignment": 0.9482421875, "loss_alignment_w": 0.47412109375, "loss_sft": 0.2950053960084915, "loss_total_v6": 0.7705302909016609, "step": 3410 }, { "epoch": 0.786681081081081, "grad_norm": 23.48567771911621, "learning_rate": 6.369796093462164e-06, "loss": 11.486274719238281, "loss_alignment": 0.93505859375, "loss_alignment_w": 0.467529296875, "loss_sft": 0.2508969586342573, "loss_total_v6": 0.7178921774029732, "step": 3411 }, { "epoch": 0.7869117117117117, "grad_norm": 23.832120895385742, "learning_rate": 6.367859187900635e-06, "loss": 11.661491394042969, "loss_alignment": 0.921875, "loss_alignment_w": 0.4609375, "loss_sft": 0.2675395254045725, "loss_total_v6": 0.7288432642817497, "step": 3412 }, { "epoch": 0.7871423423423424, "grad_norm": 23.57682228088379, "learning_rate": 6.3659220604418525e-06, "loss": 12.78803539276123, "loss_alignment": 0.94921875, "loss_alignment_w": 0.474609375, "loss_sft": 0.3243834339082241, "loss_total_v6": 0.7992521971464157, "step": 3413 }, { "epoch": 0.7873729729729729, "grad_norm": 23.338098526000977, "learning_rate": 6.363984711400064e-06, "loss": 11.95170783996582, "loss_alignment": 0.9248046875, "loss_alignment_w": 0.46240234375, "loss_sft": 0.28517451137304306, "loss_total_v6": 0.7469817101955414, "step": 3414 }, { "epoch": 0.7876036036036036, "grad_norm": 27.393720626831055, "learning_rate": 6.362047141089547e-06, "loss": 11.807937622070312, "loss_alignment": 0.92626953125, "loss_alignment_w": 0.463134765625, "loss_sft": 0.2754105981439352, "loss_total_v6": 0.7379960715770721, "step": 3415 }, { "epoch": 0.7878342342342343, "grad_norm": 24.04515838623047, "learning_rate": 6.3601093498246215e-06, "loss": 11.711065292358398, "loss_alignment": 0.9140625, "loss_alignment_w": 0.45703125, "loss_sft": 0.2754596397280693, "loss_total_v6": 0.7319415658712387, "step": 3416 }, { "epoch": 0.7880648648648648, "grad_norm": 25.002304077148438, "learning_rate": 6.35817133791964e-06, "loss": 12.667459487915039, "loss_alignment": 0.95703125, "loss_alignment_w": 0.478515625, "loss_sft": 0.31266656145453453, "loss_total_v6": 0.7917162552475929, "step": 3417 }, { "epoch": 0.7882954954954955, "grad_norm": 23.800430297851562, "learning_rate": 6.356233105688988e-06, "loss": 12.282272338867188, "loss_alignment": 0.93310546875, "loss_alignment_w": 0.466552734375, "loss_sft": 0.3007078282535076, "loss_total_v6": 0.7676420360803604, "step": 3418 }, { "epoch": 0.7885261261261262, "grad_norm": 25.25019073486328, "learning_rate": 6.354294653447095e-06, "loss": 11.793724060058594, "loss_alignment": 0.9296875, "loss_alignment_w": 0.46484375, "loss_sft": 0.27215717919170856, "loss_total_v6": 0.7371077463030815, "step": 3419 }, { "epoch": 0.7887567567567567, "grad_norm": 23.2078914642334, "learning_rate": 6.35235598150842e-06, "loss": 11.792807579040527, "loss_alignment": 0.93798828125, "loss_alignment_w": 0.468994140625, "loss_sft": 0.26758328825235367, "loss_total_v6": 0.7370504513382912, "step": 3420 }, { "epoch": 0.7889873873873874, "grad_norm": 22.924232482910156, "learning_rate": 6.350417090187454e-06, "loss": 11.501819610595703, "loss_alignment": 0.9140625, "loss_alignment_w": 0.45703125, "loss_sft": 0.26287004351615906, "loss_total_v6": 0.7188636958599091, "step": 3421 }, { "epoch": 0.789218018018018, "grad_norm": 24.492963790893555, "learning_rate": 6.348477979798735e-06, "loss": 12.158674240112305, "loss_alignment": 0.95654296875, "loss_alignment_w": 0.478271484375, "loss_sft": 0.2817982845008373, "loss_total_v6": 0.759917177259922, "step": 3422 }, { "epoch": 0.7894486486486486, "grad_norm": 25.502092361450195, "learning_rate": 6.346538650656825e-06, "loss": 11.745891571044922, "loss_alignment": 0.9326171875, "loss_alignment_w": 0.46630859375, "loss_sft": 0.2677027750760317, "loss_total_v6": 0.7341181859374046, "step": 3423 }, { "epoch": 0.7896792792792793, "grad_norm": 27.050508499145508, "learning_rate": 6.344599103076329e-06, "loss": 12.026646614074707, "loss_alignment": 0.92333984375, "loss_alignment_w": 0.461669921875, "loss_sft": 0.29057533480226994, "loss_total_v6": 0.7516654208302498, "step": 3424 }, { "epoch": 0.78990990990991, "grad_norm": 25.842226028442383, "learning_rate": 6.342659337371884e-06, "loss": 12.12474250793457, "loss_alignment": 0.94287109375, "loss_alignment_w": 0.471435546875, "loss_sft": 0.2876731473952532, "loss_total_v6": 0.7577964216470718, "step": 3425 }, { "epoch": 0.7901405405405405, "grad_norm": 26.446672439575195, "learning_rate": 6.340719353858165e-06, "loss": 12.099893569946289, "loss_alignment": 0.916015625, "loss_alignment_w": 0.4580078125, "loss_sft": 0.2967554423958063, "loss_total_v6": 0.7562433406710625, "step": 3426 }, { "epoch": 0.7903711711711712, "grad_norm": 23.537858963012695, "learning_rate": 6.338779152849876e-06, "loss": 12.23118782043457, "loss_alignment": 0.94775390625, "loss_alignment_w": 0.473876953125, "loss_sft": 0.2911673691123724, "loss_total_v6": 0.7644492238759995, "step": 3427 }, { "epoch": 0.7906018018018018, "grad_norm": 24.79803466796875, "learning_rate": 6.336838734661766e-06, "loss": 12.068727493286133, "loss_alignment": 0.912109375, "loss_alignment_w": 0.4560546875, "loss_sft": 0.29764570854604244, "loss_total_v6": 0.7542954832315445, "step": 3428 }, { "epoch": 0.7908324324324324, "grad_norm": 23.97466278076172, "learning_rate": 6.334898099608612e-06, "loss": 11.830564498901367, "loss_alignment": 0.94287109375, "loss_alignment_w": 0.471435546875, "loss_sft": 0.2677458841353655, "loss_total_v6": 0.7394103333353996, "step": 3429 }, { "epoch": 0.7910630630630631, "grad_norm": 26.58675193786621, "learning_rate": 6.332957248005227e-06, "loss": 11.694701194763184, "loss_alignment": 0.916015625, "loss_alignment_w": 0.4580078125, "loss_sft": 0.2727279495447874, "loss_total_v6": 0.7309188395738602, "step": 3430 }, { "epoch": 0.7912936936936937, "grad_norm": 26.426942825317383, "learning_rate": 6.331016180166463e-06, "loss": 11.937154769897461, "loss_alignment": 0.91748046875, "loss_alignment_w": 0.458740234375, "loss_sft": 0.2871946133673191, "loss_total_v6": 0.7460722029209137, "step": 3431 }, { "epoch": 0.7915243243243243, "grad_norm": 30.444202423095703, "learning_rate": 6.329074896407202e-06, "loss": 11.737508773803711, "loss_alignment": 0.90234375, "loss_alignment_w": 0.451171875, "loss_sft": 0.2824377119541168, "loss_total_v6": 0.7335943281650543, "step": 3432 }, { "epoch": 0.791754954954955, "grad_norm": 25.818819046020508, "learning_rate": 6.327133397042364e-06, "loss": 11.822332382202148, "loss_alignment": 0.90185546875, "loss_alignment_w": 0.450927734375, "loss_sft": 0.28735766373574734, "loss_total_v6": 0.7388957589864731, "step": 3433 }, { "epoch": 0.7919855855855856, "grad_norm": 24.217914581298828, "learning_rate": 6.325191682386906e-06, "loss": 12.108567237854004, "loss_alignment": 0.93701171875, "loss_alignment_w": 0.468505859375, "loss_sft": 0.28690632060170174, "loss_total_v6": 0.7567854598164558, "step": 3434 }, { "epoch": 0.7922162162162162, "grad_norm": 24.65545082092285, "learning_rate": 6.323249752755813e-06, "loss": 12.485269546508789, "loss_alignment": 0.91943359375, "loss_alignment_w": 0.459716796875, "loss_sft": 0.3206735886633396, "loss_total_v6": 0.7803293392062187, "step": 3435 }, { "epoch": 0.7924468468468469, "grad_norm": 25.7728271484375, "learning_rate": 6.3213076084641135e-06, "loss": 12.276050567626953, "loss_alignment": 0.9443359375, "loss_alignment_w": 0.47216796875, "loss_sft": 0.29433755576610565, "loss_total_v6": 0.7672531977295876, "step": 3436 }, { "epoch": 0.7926774774774775, "grad_norm": 24.128416061401367, "learning_rate": 6.3193652498268656e-06, "loss": 12.206451416015625, "loss_alignment": 0.95751953125, "loss_alignment_w": 0.478759765625, "loss_sft": 0.28507422655820847, "loss_total_v6": 0.762903168797493, "step": 3437 }, { "epoch": 0.7929081081081081, "grad_norm": 24.579622268676758, "learning_rate": 6.317422677159161e-06, "loss": 11.720571517944336, "loss_alignment": 0.935546875, "loss_alignment_w": 0.4677734375, "loss_sft": 0.2656320836395025, "loss_total_v6": 0.7325357720255852, "step": 3438 }, { "epoch": 0.7931387387387387, "grad_norm": 24.488304138183594, "learning_rate": 6.315479890776129e-06, "loss": 12.292688369750977, "loss_alignment": 0.9365234375, "loss_alignment_w": 0.46826171875, "loss_sft": 0.3013741336762905, "loss_total_v6": 0.768293060362339, "step": 3439 }, { "epoch": 0.7933693693693694, "grad_norm": 22.71251678466797, "learning_rate": 6.313536890992935e-06, "loss": 11.702474594116211, "loss_alignment": 0.93603515625, "loss_alignment_w": 0.468017578125, "loss_sft": 0.264806205406785, "loss_total_v6": 0.7314046993851662, "step": 3440 }, { "epoch": 0.7936, "grad_norm": 24.59747886657715, "learning_rate": 6.3115936781247756e-06, "loss": 11.845711708068848, "loss_alignment": 0.9052734375, "loss_alignment_w": 0.45263671875, "loss_sft": 0.2869267724454403, "loss_total_v6": 0.7403569892048836, "step": 3441 }, { "epoch": 0.7938306306306306, "grad_norm": 24.256494522094727, "learning_rate": 6.309650252486882e-06, "loss": 11.792031288146973, "loss_alignment": 0.92919921875, "loss_alignment_w": 0.464599609375, "loss_sft": 0.2724786438047886, "loss_total_v6": 0.7370019480586052, "step": 3442 }, { "epoch": 0.7940612612612613, "grad_norm": 27.334260940551758, "learning_rate": 6.307706614394522e-06, "loss": 11.792364120483398, "loss_alignment": 0.900390625, "loss_alignment_w": 0.4501953125, "loss_sft": 0.28568298928439617, "loss_total_v6": 0.7370227351784706, "step": 3443 }, { "epoch": 0.7942918918918919, "grad_norm": 30.695676803588867, "learning_rate": 6.3057627641629995e-06, "loss": 12.722513198852539, "loss_alignment": 0.9501953125, "loss_alignment_w": 0.47509765625, "loss_sft": 0.3194796033203602, "loss_total_v6": 0.7951570972800255, "step": 3444 }, { "epoch": 0.7945225225225225, "grad_norm": 23.700651168823242, "learning_rate": 6.303818702107646e-06, "loss": 11.405534744262695, "loss_alignment": 0.93017578125, "loss_alignment_w": 0.465087890625, "loss_sft": 0.2481852676719427, "loss_total_v6": 0.7128459513187408, "step": 3445 }, { "epoch": 0.7947531531531532, "grad_norm": 24.97010612487793, "learning_rate": 6.301874428543834e-06, "loss": 12.087389945983887, "loss_alignment": 0.94091796875, "loss_alignment_w": 0.470458984375, "loss_sft": 0.2839347831904888, "loss_total_v6": 0.7554618865251541, "step": 3446 }, { "epoch": 0.7949837837837838, "grad_norm": 28.601566314697266, "learning_rate": 6.2999299437869685e-06, "loss": 11.629548072814941, "loss_alignment": 0.9287109375, "loss_alignment_w": 0.46435546875, "loss_sft": 0.2632847595959902, "loss_total_v6": 0.7268467620015144, "step": 3447 }, { "epoch": 0.7952144144144144, "grad_norm": 25.390304565429688, "learning_rate": 6.2979852481524875e-06, "loss": 12.049286842346191, "loss_alignment": 0.93212890625, "loss_alignment_w": 0.466064453125, "loss_sft": 0.2876721192151308, "loss_total_v6": 0.7530804350972176, "step": 3448 }, { "epoch": 0.7954450450450451, "grad_norm": 24.817564010620117, "learning_rate": 6.296040341955863e-06, "loss": 11.631061553955078, "loss_alignment": 0.9150390625, "loss_alignment_w": 0.45751953125, "loss_sft": 0.2696812227368355, "loss_total_v6": 0.7269413247704506, "step": 3449 }, { "epoch": 0.7956756756756757, "grad_norm": 24.831666946411133, "learning_rate": 6.294095225512604e-06, "loss": 11.844839096069336, "loss_alignment": 0.93798828125, "loss_alignment_w": 0.468994140625, "loss_sft": 0.2700570859014988, "loss_total_v6": 0.7403024658560753, "step": 3450 }, { "epoch": 0.7959063063063063, "grad_norm": 29.299625396728516, "learning_rate": 6.292149899138252e-06, "loss": 11.897632598876953, "loss_alignment": 0.912109375, "loss_alignment_w": 0.4560546875, "loss_sft": 0.2884629350155592, "loss_total_v6": 0.7436020895838737, "step": 3451 }, { "epoch": 0.796136936936937, "grad_norm": 22.83928871154785, "learning_rate": 6.29020436314838e-06, "loss": 11.612468719482422, "loss_alignment": 0.91650390625, "loss_alignment_w": 0.458251953125, "loss_sft": 0.26787828654050827, "loss_total_v6": 0.7257792949676514, "step": 3452 }, { "epoch": 0.7963675675675675, "grad_norm": 24.432558059692383, "learning_rate": 6.288258617858599e-06, "loss": 12.205134391784668, "loss_alignment": 0.93359375, "loss_alignment_w": 0.466796875, "loss_sft": 0.2961461413651705, "loss_total_v6": 0.7628209143877029, "step": 3453 }, { "epoch": 0.7965981981981982, "grad_norm": 26.048675537109375, "learning_rate": 6.286312663584553e-06, "loss": 11.735578536987305, "loss_alignment": 0.9033203125, "loss_alignment_w": 0.45166015625, "loss_sft": 0.28326305374503136, "loss_total_v6": 0.7334736436605453, "step": 3454 }, { "epoch": 0.7968288288288289, "grad_norm": 26.672945022583008, "learning_rate": 6.284366500641914e-06, "loss": 12.101316452026367, "loss_alignment": 0.93896484375, "loss_alignment_w": 0.469482421875, "loss_sft": 0.28680405765771866, "loss_total_v6": 0.7563322857022285, "step": 3455 }, { "epoch": 0.7970594594594594, "grad_norm": 22.101472854614258, "learning_rate": 6.282420129346402e-06, "loss": 12.268325805664062, "loss_alignment": 0.9384765625, "loss_alignment_w": 0.46923828125, "loss_sft": 0.29808146320283413, "loss_total_v6": 0.7667704001069069, "step": 3456 }, { "epoch": 0.7972900900900901, "grad_norm": 25.751941680908203, "learning_rate": 6.2804735500137535e-06, "loss": 12.094117164611816, "loss_alignment": 0.94482421875, "loss_alignment_w": 0.472412109375, "loss_sft": 0.2824631407856941, "loss_total_v6": 0.7558823376893997, "step": 3457 }, { "epoch": 0.7975207207207207, "grad_norm": 24.299362182617188, "learning_rate": 6.2785267629597514e-06, "loss": 12.303262710571289, "loss_alignment": 0.94970703125, "loss_alignment_w": 0.474853515625, "loss_sft": 0.2936731446534395, "loss_total_v6": 0.7689539194107056, "step": 3458 }, { "epoch": 0.7977513513513513, "grad_norm": 22.65322494506836, "learning_rate": 6.276579768500208e-06, "loss": 12.009197235107422, "loss_alignment": 0.9189453125, "loss_alignment_w": 0.45947265625, "loss_sft": 0.29215497337281704, "loss_total_v6": 0.7505747899413109, "step": 3459 }, { "epoch": 0.797981981981982, "grad_norm": 25.862207412719727, "learning_rate": 6.274632566950967e-06, "loss": 11.541219711303711, "loss_alignment": 0.921875, "loss_alignment_w": 0.4609375, "loss_sft": 0.2604650519788265, "loss_total_v6": 0.7213262394070625, "step": 3460 }, { "epoch": 0.7982126126126126, "grad_norm": 23.54939842224121, "learning_rate": 6.272685158627909e-06, "loss": 12.13058853149414, "loss_alignment": 0.8828125, "loss_alignment_w": 0.44140625, "loss_sft": 0.3167402744293213, "loss_total_v6": 0.7581617683172226, "step": 3461 }, { "epoch": 0.7984432432432432, "grad_norm": 27.34246826171875, "learning_rate": 6.270737543846948e-06, "loss": 12.20014762878418, "loss_alignment": 0.9423828125, "loss_alignment_w": 0.47119140625, "loss_sft": 0.29082953929901123, "loss_total_v6": 0.7625092417001724, "step": 3462 }, { "epoch": 0.7986738738738739, "grad_norm": 24.320749282836914, "learning_rate": 6.268789722924029e-06, "loss": 11.915709495544434, "loss_alignment": 0.92138671875, "loss_alignment_w": 0.460693359375, "loss_sft": 0.2845115214586258, "loss_total_v6": 0.7447318509221077, "step": 3463 }, { "epoch": 0.7989045045045045, "grad_norm": 24.937623977661133, "learning_rate": 6.266841696175132e-06, "loss": 11.53298568725586, "loss_alignment": 0.919921875, "loss_alignment_w": 0.4599609375, "loss_sft": 0.2610643208026886, "loss_total_v6": 0.7208116352558136, "step": 3464 }, { "epoch": 0.7991351351351351, "grad_norm": 22.031112670898438, "learning_rate": 6.264893463916272e-06, "loss": 11.108292579650879, "loss_alignment": 0.9150390625, "loss_alignment_w": 0.45751953125, "loss_sft": 0.2375879678875208, "loss_total_v6": 0.6942682713270187, "step": 3465 }, { "epoch": 0.7993657657657658, "grad_norm": 23.64391326904297, "learning_rate": 6.262945026463495e-06, "loss": 11.858818054199219, "loss_alignment": 0.92041015625, "loss_alignment_w": 0.460205078125, "loss_sft": 0.280574319884181, "loss_total_v6": 0.7411761060357094, "step": 3466 }, { "epoch": 0.7995963963963963, "grad_norm": 25.415239334106445, "learning_rate": 6.26099638413288e-06, "loss": 11.911890029907227, "loss_alignment": 0.9384765625, "loss_alignment_w": 0.46923828125, "loss_sft": 0.275209030136466, "loss_total_v6": 0.7444931045174599, "step": 3467 }, { "epoch": 0.799827027027027, "grad_norm": 22.576173782348633, "learning_rate": 6.25904753724054e-06, "loss": 11.628862380981445, "loss_alignment": 0.91259765625, "loss_alignment_w": 0.456298828125, "loss_sft": 0.27039823308587074, "loss_total_v6": 0.7268038913607597, "step": 3468 }, { "epoch": 0.8000576576576577, "grad_norm": 24.028331756591797, "learning_rate": 6.257098486102624e-06, "loss": 12.140169143676758, "loss_alignment": 0.91650390625, "loss_alignment_w": 0.458251953125, "loss_sft": 0.30107317864894867, "loss_total_v6": 0.7587605714797974, "step": 3469 }, { "epoch": 0.8002882882882882, "grad_norm": 25.19887924194336, "learning_rate": 6.2551492310353094e-06, "loss": 12.39750862121582, "loss_alignment": 0.95947265625, "loss_alignment_w": 0.479736328125, "loss_sft": 0.29519950971007347, "loss_total_v6": 0.7748443186283112, "step": 3470 }, { "epoch": 0.8005189189189189, "grad_norm": 24.48992919921875, "learning_rate": 6.253199772354809e-06, "loss": 11.691598892211914, "loss_alignment": 0.92919921875, "loss_alignment_w": 0.464599609375, "loss_sft": 0.2665067594498396, "loss_total_v6": 0.7307249158620834, "step": 3471 }, { "epoch": 0.8007495495495496, "grad_norm": 27.88046646118164, "learning_rate": 6.251250110377368e-06, "loss": 11.919652938842773, "loss_alignment": 0.94482421875, "loss_alignment_w": 0.472412109375, "loss_sft": 0.27316127344965935, "loss_total_v6": 0.7449782937765121, "step": 3472 }, { "epoch": 0.8009801801801801, "grad_norm": 25.001293182373047, "learning_rate": 6.249300245419267e-06, "loss": 12.348577499389648, "loss_alignment": 0.9423828125, "loss_alignment_w": 0.47119140625, "loss_sft": 0.2993891853839159, "loss_total_v6": 0.7717860490083694, "step": 3473 }, { "epoch": 0.8012108108108108, "grad_norm": 26.35373306274414, "learning_rate": 6.247350177796816e-06, "loss": 11.880793571472168, "loss_alignment": 0.9072265625, "loss_alignment_w": 0.45361328125, "loss_sft": 0.2873799167573452, "loss_total_v6": 0.7425496056675911, "step": 3474 }, { "epoch": 0.8014414414414415, "grad_norm": 24.30645751953125, "learning_rate": 6.2453999078263596e-06, "loss": 11.949501037597656, "loss_alignment": 0.91796875, "loss_alignment_w": 0.458984375, "loss_sft": 0.2886833418160677, "loss_total_v6": 0.7468437626957893, "step": 3475 }, { "epoch": 0.801672072072072, "grad_norm": 23.208005905151367, "learning_rate": 6.243449435824276e-06, "loss": 11.907883644104004, "loss_alignment": 0.93017578125, "loss_alignment_w": 0.465087890625, "loss_sft": 0.2781477142125368, "loss_total_v6": 0.7442427054047585, "step": 3476 }, { "epoch": 0.8019027027027027, "grad_norm": 27.97944450378418, "learning_rate": 6.2414987621069725e-06, "loss": 12.392878532409668, "loss_alignment": 0.93115234375, "loss_alignment_w": 0.465576171875, "loss_sft": 0.30946701020002365, "loss_total_v6": 0.7745548784732819, "step": 3477 }, { "epoch": 0.8021333333333334, "grad_norm": 30.49297332763672, "learning_rate": 6.239547886990895e-06, "loss": 11.60879135131836, "loss_alignment": 0.9189453125, "loss_alignment_w": 0.45947265625, "loss_sft": 0.26613778807222843, "loss_total_v6": 0.7255494222044945, "step": 3478 }, { "epoch": 0.8023639639639639, "grad_norm": 24.329397201538086, "learning_rate": 6.237596810792518e-06, "loss": 11.988090515136719, "loss_alignment": 0.9521484375, "loss_alignment_w": 0.47607421875, "loss_sft": 0.27293733786791563, "loss_total_v6": 0.7492556795477867, "step": 3479 }, { "epoch": 0.8025945945945946, "grad_norm": 24.89215087890625, "learning_rate": 6.235645533828348e-06, "loss": 11.572929382324219, "loss_alignment": 0.92138671875, "loss_alignment_w": 0.460693359375, "loss_sft": 0.2624468430876732, "loss_total_v6": 0.7233080714941025, "step": 3480 }, { "epoch": 0.8028252252252253, "grad_norm": 23.639257431030273, "learning_rate": 6.23369405641493e-06, "loss": 12.52269172668457, "loss_alignment": 0.951171875, "loss_alignment_w": 0.4755859375, "loss_sft": 0.30793676897883415, "loss_total_v6": 0.7826682329177856, "step": 3481 }, { "epoch": 0.8030558558558558, "grad_norm": 23.97050666809082, "learning_rate": 6.2317423788688315e-06, "loss": 11.627225875854492, "loss_alignment": 0.91845703125, "loss_alignment_w": 0.459228515625, "loss_sft": 0.2676104549318552, "loss_total_v6": 0.7267016097903252, "step": 3482 }, { "epoch": 0.8032864864864865, "grad_norm": 24.001602172851562, "learning_rate": 6.229790501506661e-06, "loss": 12.187944412231445, "loss_alignment": 0.880859375, "loss_alignment_w": 0.4404296875, "loss_sft": 0.32108787819743156, "loss_total_v6": 0.7617465034127235, "step": 3483 }, { "epoch": 0.8035171171171172, "grad_norm": 24.658437728881836, "learning_rate": 6.227838424645057e-06, "loss": 11.312162399291992, "loss_alignment": 0.90380859375, "loss_alignment_w": 0.451904296875, "loss_sft": 0.25521269626915455, "loss_total_v6": 0.7070101648569107, "step": 3484 }, { "epoch": 0.8037477477477477, "grad_norm": 24.684925079345703, "learning_rate": 6.225886148600688e-06, "loss": 11.605170249938965, "loss_alignment": 0.90625, "loss_alignment_w": 0.453125, "loss_sft": 0.2716335617005825, "loss_total_v6": 0.7253231406211853, "step": 3485 }, { "epoch": 0.8039783783783784, "grad_norm": 29.21168327331543, "learning_rate": 6.223933673690259e-06, "loss": 12.228693008422852, "loss_alignment": 0.93701171875, "loss_alignment_w": 0.468505859375, "loss_sft": 0.2956806421279907, "loss_total_v6": 0.7642933055758476, "step": 3486 }, { "epoch": 0.804209009009009, "grad_norm": 28.82124900817871, "learning_rate": 6.221981000230503e-06, "loss": 12.095294952392578, "loss_alignment": 0.95751953125, "loss_alignment_w": 0.478759765625, "loss_sft": 0.2778522726148367, "loss_total_v6": 0.7559559270739555, "step": 3487 }, { "epoch": 0.8044396396396396, "grad_norm": 25.074888229370117, "learning_rate": 6.220028128538188e-06, "loss": 11.709274291992188, "loss_alignment": 0.935546875, "loss_alignment_w": 0.4677734375, "loss_sft": 0.26469709537923336, "loss_total_v6": 0.7318296581506729, "step": 3488 }, { "epoch": 0.8046702702702703, "grad_norm": 25.924524307250977, "learning_rate": 6.218075058930113e-06, "loss": 12.215553283691406, "loss_alignment": 0.91748046875, "loss_alignment_w": 0.458740234375, "loss_sft": 0.3041978031396866, "loss_total_v6": 0.7634721100330353, "step": 3489 }, { "epoch": 0.8049009009009009, "grad_norm": 22.587495803833008, "learning_rate": 6.21612179172311e-06, "loss": 11.792238235473633, "loss_alignment": 0.92138671875, "loss_alignment_w": 0.460693359375, "loss_sft": 0.2764741163700819, "loss_total_v6": 0.7370148748159409, "step": 3490 }, { "epoch": 0.8051315315315315, "grad_norm": 25.313156127929688, "learning_rate": 6.2141683272340425e-06, "loss": 11.306066513061523, "loss_alignment": 0.900390625, "loss_alignment_w": 0.4501953125, "loss_sft": 0.25550306402146816, "loss_total_v6": 0.7066291719675064, "step": 3491 }, { "epoch": 0.8053621621621622, "grad_norm": 23.54275131225586, "learning_rate": 6.212214665779805e-06, "loss": 12.002374649047852, "loss_alignment": 0.94140625, "loss_alignment_w": 0.470703125, "loss_sft": 0.2809711694717407, "loss_total_v6": 0.7501484081149101, "step": 3492 }, { "epoch": 0.8055927927927928, "grad_norm": 24.2623348236084, "learning_rate": 6.210260807677326e-06, "loss": 12.440207481384277, "loss_alignment": 0.923828125, "loss_alignment_w": 0.4619140625, "loss_sft": 0.31605667993426323, "loss_total_v6": 0.7775129675865173, "step": 3493 }, { "epoch": 0.8058234234234234, "grad_norm": 24.542743682861328, "learning_rate": 6.208306753243565e-06, "loss": 11.877864837646484, "loss_alignment": 0.923828125, "loss_alignment_w": 0.4619140625, "loss_sft": 0.2800404652953148, "loss_total_v6": 0.7423665225505829, "step": 3494 }, { "epoch": 0.8060540540540541, "grad_norm": 38.880252838134766, "learning_rate": 6.206352502795512e-06, "loss": 12.132904052734375, "loss_alignment": 0.9287109375, "loss_alignment_w": 0.46435546875, "loss_sft": 0.29315755143761635, "loss_total_v6": 0.758306473493576, "step": 3495 }, { "epoch": 0.8062846846846847, "grad_norm": 25.929914474487305, "learning_rate": 6.20439805665019e-06, "loss": 11.384803771972656, "loss_alignment": 0.9482421875, "loss_alignment_w": 0.47412109375, "loss_sft": 0.23744438216090202, "loss_total_v6": 0.711550235748291, "step": 3496 }, { "epoch": 0.8065153153153153, "grad_norm": 23.694242477416992, "learning_rate": 6.202443415124655e-06, "loss": 12.005977630615234, "loss_alignment": 0.9423828125, "loss_alignment_w": 0.47119140625, "loss_sft": 0.27877022325992584, "loss_total_v6": 0.7503735944628716, "step": 3497 }, { "epoch": 0.806745945945946, "grad_norm": 23.762737274169922, "learning_rate": 6.200488578535992e-06, "loss": 11.752455711364746, "loss_alignment": 0.93212890625, "loss_alignment_w": 0.466064453125, "loss_sft": 0.2662057727575302, "loss_total_v6": 0.734528511762619, "step": 3498 }, { "epoch": 0.8069765765765766, "grad_norm": 23.159629821777344, "learning_rate": 6.19853354720132e-06, "loss": 11.958091735839844, "loss_alignment": 0.9404296875, "loss_alignment_w": 0.47021484375, "loss_sft": 0.2781729828566313, "loss_total_v6": 0.7473807260394096, "step": 3499 }, { "epoch": 0.8072072072072072, "grad_norm": 28.162633895874023, "learning_rate": 6.1965783214377895e-06, "loss": 11.611865997314453, "loss_alignment": 0.91064453125, "loss_alignment_w": 0.455322265625, "loss_sft": 0.2694732993841171, "loss_total_v6": 0.7257416024804115, "step": 3500 }, { "epoch": 0.8072072072072072, "eval_loss": 0.9080885648727417, "eval_loss_alignment": 0.9329917594178082, "eval_loss_alignment_w": 0.4664958797089041, "eval_loss_contrast": 0.0, "eval_loss_highlevel": 0.0, "eval_loss_sft": 0.4415926767946923, "eval_loss_total_v6": 0.9080885553128644, "eval_loss_wm_recon": 0.0, "eval_loss_wm_recon_w": 0.0, "eval_runtime": 309.3134, "eval_samples_per_second": 5.642, "eval_steps_per_second": 0.708, "step": 3500 }, { "epoch": 0.8074378378378378, "grad_norm": 23.727766036987305, "learning_rate": 6.194622901562581e-06, "loss": 12.321141242980957, "loss_alignment": 0.9453125, "loss_alignment_w": 0.47265625, "loss_sft": 0.29900201596319675, "loss_total_v6": 0.7700713276863098, "step": 3501 }, { "epoch": 0.8076684684684685, "grad_norm": 23.105615615844727, "learning_rate": 6.192667287892905e-06, "loss": 12.117755889892578, "loss_alignment": 0.9453125, "loss_alignment_w": 0.47265625, "loss_sft": 0.2838947791606188, "loss_total_v6": 0.7573597356677055, "step": 3502 }, { "epoch": 0.8078990990990991, "grad_norm": 26.268735885620117, "learning_rate": 6.190711480746009e-06, "loss": 11.550561904907227, "loss_alignment": 0.91845703125, "loss_alignment_w": 0.459228515625, "loss_sft": 0.2642075438052416, "loss_total_v6": 0.7219101414084435, "step": 3503 }, { "epoch": 0.8081297297297297, "grad_norm": 24.71919822692871, "learning_rate": 6.188755480439166e-06, "loss": 11.726117134094238, "loss_alignment": 0.9072265625, "loss_alignment_w": 0.45361328125, "loss_sft": 0.2785213701426983, "loss_total_v6": 0.7328823357820511, "step": 3504 }, { "epoch": 0.8083603603603604, "grad_norm": 28.221757888793945, "learning_rate": 6.186799287289684e-06, "loss": 11.869956970214844, "loss_alignment": 0.9453125, "loss_alignment_w": 0.47265625, "loss_sft": 0.2693838570266962, "loss_total_v6": 0.7418722882866859, "step": 3505 }, { "epoch": 0.808590990990991, "grad_norm": 30.61217498779297, "learning_rate": 6.184842901614901e-06, "loss": 11.947281837463379, "loss_alignment": 0.90234375, "loss_alignment_w": 0.451171875, "loss_sft": 0.2943888586014509, "loss_total_v6": 0.746705137193203, "step": 3506 }, { "epoch": 0.8088216216216216, "grad_norm": 25.48411750793457, "learning_rate": 6.182886323732188e-06, "loss": 12.1251220703125, "loss_alignment": 0.92138671875, "loss_alignment_w": 0.460693359375, "loss_sft": 0.29790502041578293, "loss_total_v6": 0.7578201815485954, "step": 3507 }, { "epoch": 0.8090522522522523, "grad_norm": 27.4683780670166, "learning_rate": 6.180929553958943e-06, "loss": 11.659626007080078, "loss_alignment": 0.9267578125, "loss_alignment_w": 0.46337890625, "loss_sft": 0.26508832909166813, "loss_total_v6": 0.7287266254425049, "step": 3508 }, { "epoch": 0.8092828828828829, "grad_norm": 25.396038055419922, "learning_rate": 6.178972592612597e-06, "loss": 12.365650177001953, "loss_alignment": 0.93115234375, "loss_alignment_w": 0.465576171875, "loss_sft": 0.308329775929451, "loss_total_v6": 0.7728531062602997, "step": 3509 }, { "epoch": 0.8095135135135135, "grad_norm": 25.810123443603516, "learning_rate": 6.177015440010617e-06, "loss": 11.934259414672852, "loss_alignment": 0.91064453125, "loss_alignment_w": 0.455322265625, "loss_sft": 0.2905536759644747, "loss_total_v6": 0.7458912059664726, "step": 3510 }, { "epoch": 0.8097441441441442, "grad_norm": 24.00266456604004, "learning_rate": 6.175058096470493e-06, "loss": 11.86484432220459, "loss_alignment": 0.91748046875, "loss_alignment_w": 0.458740234375, "loss_sft": 0.28275151550769806, "loss_total_v6": 0.7415527775883675, "step": 3511 }, { "epoch": 0.8099747747747748, "grad_norm": 22.801464080810547, "learning_rate": 6.173100562309751e-06, "loss": 12.175308227539062, "loss_alignment": 0.9169921875, "loss_alignment_w": 0.45849609375, "loss_sft": 0.3035593740642071, "loss_total_v6": 0.760956808924675, "step": 3512 }, { "epoch": 0.8102054054054054, "grad_norm": 24.689804077148438, "learning_rate": 6.171142837845947e-06, "loss": 12.137937545776367, "loss_alignment": 0.919921875, "loss_alignment_w": 0.4599609375, "loss_sft": 0.29913320392370224, "loss_total_v6": 0.7586211189627647, "step": 3513 }, { "epoch": 0.8104360360360361, "grad_norm": 25.353498458862305, "learning_rate": 6.169184923396667e-06, "loss": 11.9878511428833, "loss_alignment": 0.91259765625, "loss_alignment_w": 0.456298828125, "loss_sft": 0.2938421405851841, "loss_total_v6": 0.7492407038807869, "step": 3514 }, { "epoch": 0.8106666666666666, "grad_norm": 30.76123809814453, "learning_rate": 6.1672268192795285e-06, "loss": 12.140508651733398, "loss_alignment": 0.91748046875, "loss_alignment_w": 0.458740234375, "loss_sft": 0.30052979849278927, "loss_total_v6": 0.7587817460298538, "step": 3515 }, { "epoch": 0.8108972972972973, "grad_norm": 24.73855209350586, "learning_rate": 6.165268525812178e-06, "loss": 11.97705078125, "loss_alignment": 0.9140625, "loss_alignment_w": 0.45703125, "loss_sft": 0.2917022928595543, "loss_total_v6": 0.7485656887292862, "step": 3516 }, { "epoch": 0.811127927927928, "grad_norm": 29.101911544799805, "learning_rate": 6.1633100433122985e-06, "loss": 11.459989547729492, "loss_alignment": 0.916015625, "loss_alignment_w": 0.4580078125, "loss_sft": 0.2580279726535082, "loss_total_v6": 0.7162493765354156, "step": 3517 }, { "epoch": 0.8113585585585585, "grad_norm": 26.019437789916992, "learning_rate": 6.161351372097596e-06, "loss": 11.809660911560059, "loss_alignment": 0.91943359375, "loss_alignment_w": 0.459716796875, "loss_sft": 0.2799128796905279, "loss_total_v6": 0.7381037920713425, "step": 3518 }, { "epoch": 0.8115891891891892, "grad_norm": 25.07984733581543, "learning_rate": 6.159392512485811e-06, "loss": 11.610187530517578, "loss_alignment": 0.89990234375, "loss_alignment_w": 0.449951171875, "loss_sft": 0.2755177263170481, "loss_total_v6": 0.7256367355585098, "step": 3519 }, { "epoch": 0.8118198198198198, "grad_norm": 23.6868953704834, "learning_rate": 6.157433464794717e-06, "loss": 11.699914932250977, "loss_alignment": 0.923828125, "loss_alignment_w": 0.4619140625, "loss_sft": 0.26942218840122223, "loss_total_v6": 0.7312447056174278, "step": 3520 }, { "epoch": 0.8120504504504504, "grad_norm": 26.790481567382812, "learning_rate": 6.15547422934211e-06, "loss": 11.457854270935059, "loss_alignment": 0.87646484375, "loss_alignment_w": 0.438232421875, "loss_sft": 0.27611348032951355, "loss_total_v6": 0.7161159068346024, "step": 3521 }, { "epoch": 0.8122810810810811, "grad_norm": 25.00580596923828, "learning_rate": 6.1535148064458275e-06, "loss": 12.123635292053223, "loss_alignment": 0.9228515625, "loss_alignment_w": 0.46142578125, "loss_sft": 0.2948365658521652, "loss_total_v6": 0.7577271834015846, "step": 3522 }, { "epoch": 0.8125117117117117, "grad_norm": 23.88765525817871, "learning_rate": 6.151555196423728e-06, "loss": 12.566140174865723, "loss_alignment": 0.93603515625, "loss_alignment_w": 0.468017578125, "loss_sft": 0.3185715787112713, "loss_total_v6": 0.7853837609291077, "step": 3523 }, { "epoch": 0.8127423423423423, "grad_norm": 24.970659255981445, "learning_rate": 6.149595399593703e-06, "loss": 11.76854419708252, "loss_alignment": 0.9345703125, "loss_alignment_w": 0.46728515625, "loss_sft": 0.27018666826188564, "loss_total_v6": 0.7355339899659157, "step": 3524 }, { "epoch": 0.812972972972973, "grad_norm": 25.797380447387695, "learning_rate": 6.147635416273679e-06, "loss": 12.316180229187012, "loss_alignment": 0.92822265625, "loss_alignment_w": 0.464111328125, "loss_sft": 0.30606188625097275, "loss_total_v6": 0.7697612568736076, "step": 3525 }, { "epoch": 0.8132036036036036, "grad_norm": 25.484783172607422, "learning_rate": 6.145675246781606e-06, "loss": 11.587773323059082, "loss_alignment": 0.9140625, "loss_alignment_w": 0.45703125, "loss_sft": 0.26673151925206184, "loss_total_v6": 0.7242358028888702, "step": 3526 }, { "epoch": 0.8134342342342342, "grad_norm": 25.84285545349121, "learning_rate": 6.143714891435468e-06, "loss": 11.667388916015625, "loss_alignment": 0.9140625, "loss_alignment_w": 0.45703125, "loss_sft": 0.2726841550320387, "loss_total_v6": 0.7292118594050407, "step": 3527 }, { "epoch": 0.8136648648648649, "grad_norm": 24.712039947509766, "learning_rate": 6.141754350553279e-06, "loss": 11.719144821166992, "loss_alignment": 0.92333984375, "loss_alignment_w": 0.461669921875, "loss_sft": 0.269952654838562, "loss_total_v6": 0.7324465587735176, "step": 3528 }, { "epoch": 0.8138954954954954, "grad_norm": 27.736963272094727, "learning_rate": 6.139793624453083e-06, "loss": 12.328880310058594, "loss_alignment": 0.908203125, "loss_alignment_w": 0.4541015625, "loss_sft": 0.316346675157547, "loss_total_v6": 0.7705550417304039, "step": 3529 }, { "epoch": 0.8141261261261261, "grad_norm": 21.963970184326172, "learning_rate": 6.13783271345295e-06, "loss": 11.892863273620605, "loss_alignment": 0.92822265625, "loss_alignment_w": 0.464111328125, "loss_sft": 0.280123358592391, "loss_total_v6": 0.743303932249546, "step": 3530 }, { "epoch": 0.8143567567567568, "grad_norm": 25.639984130859375, "learning_rate": 6.135871617870988e-06, "loss": 12.430511474609375, "loss_alignment": 0.93798828125, "loss_alignment_w": 0.468994140625, "loss_sft": 0.30710411444306374, "loss_total_v6": 0.7769069597125053, "step": 3531 }, { "epoch": 0.8145873873873873, "grad_norm": 26.415552139282227, "learning_rate": 6.133910338025329e-06, "loss": 11.774072647094727, "loss_alignment": 0.9150390625, "loss_alignment_w": 0.45751953125, "loss_sft": 0.28012996912002563, "loss_total_v6": 0.7358794957399368, "step": 3532 }, { "epoch": 0.814818018018018, "grad_norm": 21.9886531829834, "learning_rate": 6.131948874234132e-06, "loss": 11.156782150268555, "loss_alignment": 0.919921875, "loss_alignment_w": 0.4599609375, "loss_sft": 0.23785674013197422, "loss_total_v6": 0.6972988545894623, "step": 3533 }, { "epoch": 0.8150486486486487, "grad_norm": 28.091371536254883, "learning_rate": 6.129987226815597e-06, "loss": 11.700818061828613, "loss_alignment": 0.90966796875, "loss_alignment_w": 0.454833984375, "loss_sft": 0.27697067707777023, "loss_total_v6": 0.7313011288642883, "step": 3534 }, { "epoch": 0.8152792792792792, "grad_norm": 24.795730590820312, "learning_rate": 6.128025396087943e-06, "loss": 11.491874694824219, "loss_alignment": 0.9140625, "loss_alignment_w": 0.45703125, "loss_sft": 0.2611803822219372, "loss_total_v6": 0.7182421386241913, "step": 3535 }, { "epoch": 0.8155099099099099, "grad_norm": 24.77429962158203, "learning_rate": 6.1260633823694224e-06, "loss": 11.76297378540039, "loss_alignment": 0.91748046875, "loss_alignment_w": 0.458740234375, "loss_sft": 0.2753928005695343, "loss_total_v6": 0.7351858988404274, "step": 3536 }, { "epoch": 0.8157405405405406, "grad_norm": 23.249595642089844, "learning_rate": 6.1241011859783196e-06, "loss": 11.885824203491211, "loss_alignment": 0.9267578125, "loss_alignment_w": 0.46337890625, "loss_sft": 0.27904257737100124, "loss_total_v6": 0.7428639680147171, "step": 3537 }, { "epoch": 0.8159711711711711, "grad_norm": 24.0679874420166, "learning_rate": 6.122138807232944e-06, "loss": 12.271662712097168, "loss_alignment": 0.9111328125, "loss_alignment_w": 0.45556640625, "loss_sft": 0.31147358752787113, "loss_total_v6": 0.7669789418578148, "step": 3538 }, { "epoch": 0.8162018018018018, "grad_norm": 22.34117889404297, "learning_rate": 6.120176246451641e-06, "loss": 11.697579383850098, "loss_alignment": 0.90771484375, "loss_alignment_w": 0.453857421875, "loss_sft": 0.2767225205898285, "loss_total_v6": 0.7310987263917923, "step": 3539 }, { "epoch": 0.8164324324324325, "grad_norm": 25.02363395690918, "learning_rate": 6.118213503952779e-06, "loss": 11.835442543029785, "loss_alignment": 0.93212890625, "loss_alignment_w": 0.466064453125, "loss_sft": 0.27290299721062183, "loss_total_v6": 0.7397151440382004, "step": 3540 }, { "epoch": 0.816663063063063, "grad_norm": 25.340017318725586, "learning_rate": 6.116250580054758e-06, "loss": 11.458955764770508, "loss_alignment": 0.8955078125, "loss_alignment_w": 0.44775390625, "loss_sft": 0.26762212812900543, "loss_total_v6": 0.7161847427487373, "step": 3541 }, { "epoch": 0.8168936936936937, "grad_norm": 23.33395004272461, "learning_rate": 6.11428747507601e-06, "loss": 11.655966758728027, "loss_alignment": 0.90087890625, "loss_alignment_w": 0.450439453125, "loss_sft": 0.27795170433819294, "loss_total_v6": 0.7284979447722435, "step": 3542 }, { "epoch": 0.8171243243243244, "grad_norm": 26.877700805664062, "learning_rate": 6.112324189334992e-06, "loss": 11.462373733520508, "loss_alignment": 0.9169921875, "loss_alignment_w": 0.45849609375, "loss_sft": 0.2576428968459368, "loss_total_v6": 0.7163983583450317, "step": 3543 }, { "epoch": 0.8173549549549549, "grad_norm": 24.05406379699707, "learning_rate": 6.110360723150195e-06, "loss": 12.025094985961914, "loss_alignment": 0.9140625, "loss_alignment_w": 0.45703125, "loss_sft": 0.2943846005946398, "loss_total_v6": 0.7515684217214584, "step": 3544 }, { "epoch": 0.8175855855855856, "grad_norm": 23.83767318725586, "learning_rate": 6.108397076840137e-06, "loss": 12.086613655090332, "loss_alignment": 0.931640625, "loss_alignment_w": 0.4658203125, "loss_sft": 0.2914241198450327, "loss_total_v6": 0.7554133459925652, "step": 3545 }, { "epoch": 0.8178162162162163, "grad_norm": 23.375553131103516, "learning_rate": 6.1064332507233625e-06, "loss": 11.690106391906738, "loss_alignment": 0.9052734375, "loss_alignment_w": 0.45263671875, "loss_sft": 0.27858998253941536, "loss_total_v6": 0.73063163459301, "step": 3546 }, { "epoch": 0.8180468468468468, "grad_norm": 27.43617057800293, "learning_rate": 6.1044692451184515e-06, "loss": 12.214481353759766, "loss_alignment": 0.93896484375, "loss_alignment_w": 0.469482421875, "loss_sft": 0.29485344141721725, "loss_total_v6": 0.7634051069617271, "step": 3547 }, { "epoch": 0.8182774774774775, "grad_norm": 24.06975555419922, "learning_rate": 6.102505060344006e-06, "loss": 12.352415084838867, "loss_alignment": 0.93994140625, "loss_alignment_w": 0.469970703125, "loss_sft": 0.30216208659112453, "loss_total_v6": 0.772025965154171, "step": 3548 }, { "epoch": 0.8185081081081081, "grad_norm": 23.06983184814453, "learning_rate": 6.100540696718663e-06, "loss": 11.427212715148926, "loss_alignment": 0.927734375, "loss_alignment_w": 0.4638671875, "loss_sft": 0.24964695237576962, "loss_total_v6": 0.7142008021473885, "step": 3549 }, { "epoch": 0.8187387387387387, "grad_norm": 22.92333221435547, "learning_rate": 6.0985761545610865e-06, "loss": 11.984207153320312, "loss_alignment": 0.9296875, "loss_alignment_w": 0.46484375, "loss_sft": 0.2832994796335697, "loss_total_v6": 0.7490129768848419, "step": 3550 }, { "epoch": 0.8189693693693694, "grad_norm": 25.083417892456055, "learning_rate": 6.096611434189967e-06, "loss": 11.980751037597656, "loss_alignment": 0.9404296875, "loss_alignment_w": 0.47021484375, "loss_sft": 0.2780479732900858, "loss_total_v6": 0.7487969025969505, "step": 3551 }, { "epoch": 0.8192, "grad_norm": 26.1021785736084, "learning_rate": 6.094646535924026e-06, "loss": 12.32429313659668, "loss_alignment": 0.95654296875, "loss_alignment_w": 0.478271484375, "loss_sft": 0.2918900717049837, "loss_total_v6": 0.7702683433890343, "step": 3552 }, { "epoch": 0.8194306306306306, "grad_norm": 26.632726669311523, "learning_rate": 6.092681460082015e-06, "loss": 12.060311317443848, "loss_alignment": 0.92724609375, "loss_alignment_w": 0.463623046875, "loss_sft": 0.2894597891718149, "loss_total_v6": 0.7537694498896599, "step": 3553 }, { "epoch": 0.8196612612612613, "grad_norm": 26.380678176879883, "learning_rate": 6.090716206982714e-06, "loss": 11.677791595458984, "loss_alignment": 0.89306640625, "loss_alignment_w": 0.446533203125, "loss_sft": 0.2831914834678173, "loss_total_v6": 0.7298619896173477, "step": 3554 }, { "epoch": 0.8198918918918919, "grad_norm": 29.807323455810547, "learning_rate": 6.088750776944928e-06, "loss": 12.010480880737305, "loss_alignment": 0.90087890625, "loss_alignment_w": 0.450439453125, "loss_sft": 0.30058184638619423, "loss_total_v6": 0.7506550624966621, "step": 3555 }, { "epoch": 0.8201225225225225, "grad_norm": 27.68819808959961, "learning_rate": 6.086785170287496e-06, "loss": 12.045478820800781, "loss_alignment": 0.9091796875, "loss_alignment_w": 0.45458984375, "loss_sft": 0.29777958430349827, "loss_total_v6": 0.7528424188494682, "step": 3556 }, { "epoch": 0.8203531531531532, "grad_norm": 22.615421295166016, "learning_rate": 6.084819387329283e-06, "loss": 12.010038375854492, "loss_alignment": 0.91552734375, "loss_alignment_w": 0.457763671875, "loss_sft": 0.291993984952569, "loss_total_v6": 0.7506273910403252, "step": 3557 }, { "epoch": 0.8205837837837838, "grad_norm": 24.539682388305664, "learning_rate": 6.082853428389179e-06, "loss": 11.712053298950195, "loss_alignment": 0.90478515625, "loss_alignment_w": 0.452392578125, "loss_sft": 0.2803889587521553, "loss_total_v6": 0.7320033460855484, "step": 3558 }, { "epoch": 0.8208144144144144, "grad_norm": 24.79148292541504, "learning_rate": 6.080887293786113e-06, "loss": 12.273429870605469, "loss_alignment": 0.91064453125, "loss_alignment_w": 0.455322265625, "loss_sft": 0.3128352388739586, "loss_total_v6": 0.7670893892645836, "step": 3559 }, { "epoch": 0.821045045045045, "grad_norm": 24.90812873840332, "learning_rate": 6.078920983839032e-06, "loss": 12.01374626159668, "loss_alignment": 0.923828125, "loss_alignment_w": 0.4619140625, "loss_sft": 0.28911297395825386, "loss_total_v6": 0.7508591488003731, "step": 3560 }, { "epoch": 0.8212756756756757, "grad_norm": 26.066837310791016, "learning_rate": 6.076954498866914e-06, "loss": 11.934818267822266, "loss_alignment": 0.93310546875, "loss_alignment_w": 0.466552734375, "loss_sft": 0.28024319373071194, "loss_total_v6": 0.745926171541214, "step": 3561 }, { "epoch": 0.8215063063063063, "grad_norm": 26.802610397338867, "learning_rate": 6.0749878391887705e-06, "loss": 11.56682014465332, "loss_alignment": 0.90625, "loss_alignment_w": 0.453125, "loss_sft": 0.27013700269162655, "loss_total_v6": 0.7229262813925743, "step": 3562 }, { "epoch": 0.821736936936937, "grad_norm": 36.42044448852539, "learning_rate": 6.073021005123634e-06, "loss": 11.642508506774902, "loss_alignment": 0.908203125, "loss_alignment_w": 0.4541015625, "loss_sft": 0.27328058890998363, "loss_total_v6": 0.7276568114757538, "step": 3563 }, { "epoch": 0.8219675675675676, "grad_norm": 26.065324783325195, "learning_rate": 6.07105399699057e-06, "loss": 11.877378463745117, "loss_alignment": 0.9287109375, "loss_alignment_w": 0.46435546875, "loss_sft": 0.2783011104911566, "loss_total_v6": 0.742336131632328, "step": 3564 }, { "epoch": 0.8221981981981982, "grad_norm": 26.502256393432617, "learning_rate": 6.069086815108671e-06, "loss": 11.649801254272461, "loss_alignment": 0.91357421875, "loss_alignment_w": 0.456787109375, "loss_sft": 0.27214942313730717, "loss_total_v6": 0.7281125336885452, "step": 3565 }, { "epoch": 0.8224288288288288, "grad_norm": 24.163002014160156, "learning_rate": 6.067119459797061e-06, "loss": 11.152902603149414, "loss_alignment": 0.92431640625, "loss_alignment_w": 0.462158203125, "loss_sft": 0.23485244065523148, "loss_total_v6": 0.6970564126968384, "step": 3566 }, { "epoch": 0.8226594594594595, "grad_norm": 24.801240921020508, "learning_rate": 6.065151931374884e-06, "loss": 11.945167541503906, "loss_alignment": 0.95458984375, "loss_alignment_w": 0.477294921875, "loss_sft": 0.2696137186139822, "loss_total_v6": 0.7465729489922523, "step": 3567 }, { "epoch": 0.8228900900900901, "grad_norm": 24.793149948120117, "learning_rate": 6.063184230161319e-06, "loss": 11.509737014770508, "loss_alignment": 0.921875, "loss_alignment_w": 0.4609375, "loss_sft": 0.25842106714844704, "loss_total_v6": 0.7193585634231567, "step": 3568 }, { "epoch": 0.8231207207207207, "grad_norm": 24.915918350219727, "learning_rate": 6.061216356475571e-06, "loss": 11.974797248840332, "loss_alignment": 0.92724609375, "loss_alignment_w": 0.463623046875, "loss_sft": 0.2849085722118616, "loss_total_v6": 0.7484248280525208, "step": 3569 }, { "epoch": 0.8233513513513514, "grad_norm": 23.664854049682617, "learning_rate": 6.059248310636873e-06, "loss": 12.378076553344727, "loss_alignment": 0.939453125, "loss_alignment_w": 0.4697265625, "loss_sft": 0.3033233378082514, "loss_total_v6": 0.7736297473311424, "step": 3570 }, { "epoch": 0.823581981981982, "grad_norm": 28.71236228942871, "learning_rate": 6.057280092964486e-06, "loss": 11.010150909423828, "loss_alignment": 0.8876953125, "loss_alignment_w": 0.44384765625, "loss_sft": 0.24395104497671127, "loss_total_v6": 0.6881344169378281, "step": 3571 }, { "epoch": 0.8238126126126126, "grad_norm": 28.5324764251709, "learning_rate": 6.055311703777699e-06, "loss": 11.754077911376953, "loss_alignment": 0.89892578125, "loss_alignment_w": 0.449462890625, "loss_sft": 0.28533482924103737, "loss_total_v6": 0.734629899263382, "step": 3572 }, { "epoch": 0.8240432432432433, "grad_norm": 26.455745697021484, "learning_rate": 6.053343143395827e-06, "loss": 12.342567443847656, "loss_alignment": 0.9248046875, "loss_alignment_w": 0.46240234375, "loss_sft": 0.3081231154501438, "loss_total_v6": 0.7714104726910591, "step": 3573 }, { "epoch": 0.8242738738738739, "grad_norm": 23.82367706298828, "learning_rate": 6.0513744121382164e-06, "loss": 12.380722045898438, "loss_alignment": 0.935546875, "loss_alignment_w": 0.4677734375, "loss_sft": 0.30672355368733406, "loss_total_v6": 0.7737950757145882, "step": 3574 }, { "epoch": 0.8245045045045045, "grad_norm": 24.120824813842773, "learning_rate": 6.049405510324237e-06, "loss": 11.759785652160645, "loss_alignment": 0.90234375, "loss_alignment_w": 0.451171875, "loss_sft": 0.2845013365149498, "loss_total_v6": 0.7349865883588791, "step": 3575 }, { "epoch": 0.8247351351351352, "grad_norm": 25.518964767456055, "learning_rate": 6.047436438273293e-06, "loss": 12.086371421813965, "loss_alignment": 0.9326171875, "loss_alignment_w": 0.46630859375, "loss_sft": 0.28860138915479183, "loss_total_v6": 0.7553982436656952, "step": 3576 }, { "epoch": 0.8249657657657657, "grad_norm": 22.803112030029297, "learning_rate": 6.045467196304807e-06, "loss": 11.999839782714844, "loss_alignment": 0.9140625, "loss_alignment_w": 0.45703125, "loss_sft": 0.29346222057938576, "loss_total_v6": 0.7499899417161942, "step": 3577 }, { "epoch": 0.8251963963963964, "grad_norm": 24.019847869873047, "learning_rate": 6.043497784738237e-06, "loss": 12.10580825805664, "loss_alignment": 0.912109375, "loss_alignment_w": 0.4560546875, "loss_sft": 0.30032943561673164, "loss_total_v6": 0.7566129863262177, "step": 3578 }, { "epoch": 0.825427027027027, "grad_norm": 22.272626876831055, "learning_rate": 6.0415282038930645e-06, "loss": 11.870716094970703, "loss_alignment": 0.91357421875, "loss_alignment_w": 0.456787109375, "loss_sft": 0.2835610006004572, "loss_total_v6": 0.7419197857379913, "step": 3579 }, { "epoch": 0.8256576576576576, "grad_norm": 27.821805953979492, "learning_rate": 6.039558454088796e-06, "loss": 11.876840591430664, "loss_alignment": 0.8896484375, "loss_alignment_w": 0.44482421875, "loss_sft": 0.2973105050623417, "loss_total_v6": 0.7423025518655777, "step": 3580 }, { "epoch": 0.8258882882882883, "grad_norm": 24.568782806396484, "learning_rate": 6.037588535644976e-06, "loss": 12.046217918395996, "loss_alignment": 0.90673828125, "loss_alignment_w": 0.453369140625, "loss_sft": 0.2998551856726408, "loss_total_v6": 0.752888634800911, "step": 3581 }, { "epoch": 0.8261189189189189, "grad_norm": 26.337146759033203, "learning_rate": 6.035618448881163e-06, "loss": 12.216253280639648, "loss_alignment": 0.935546875, "loss_alignment_w": 0.4677734375, "loss_sft": 0.2955745458602905, "loss_total_v6": 0.763515830039978, "step": 3582 }, { "epoch": 0.8263495495495495, "grad_norm": 26.701709747314453, "learning_rate": 6.033648194116951e-06, "loss": 12.426462173461914, "loss_alignment": 0.9296875, "loss_alignment_w": 0.46484375, "loss_sft": 0.3114439342170954, "loss_total_v6": 0.7766539007425308, "step": 3583 }, { "epoch": 0.8265801801801802, "grad_norm": 23.84823989868164, "learning_rate": 6.031677771671962e-06, "loss": 11.399749755859375, "loss_alignment": 0.8818359375, "loss_alignment_w": 0.44091796875, "loss_sft": 0.2722530849277973, "loss_total_v6": 0.7124843895435333, "step": 3584 }, { "epoch": 0.8268108108108108, "grad_norm": 24.85213851928711, "learning_rate": 6.029707181865837e-06, "loss": 11.294474601745605, "loss_alignment": 0.92724609375, "loss_alignment_w": 0.463623046875, "loss_sft": 0.24296825006604195, "loss_total_v6": 0.7059046477079391, "step": 3585 }, { "epoch": 0.8270414414414414, "grad_norm": 23.202232360839844, "learning_rate": 6.027736425018254e-06, "loss": 12.03731918334961, "loss_alignment": 0.900390625, "loss_alignment_w": 0.4501953125, "loss_sft": 0.30219811759889126, "loss_total_v6": 0.7523323968052864, "step": 3586 }, { "epoch": 0.8272720720720721, "grad_norm": 24.594663619995117, "learning_rate": 6.0257655014489115e-06, "loss": 11.757779121398926, "loss_alignment": 0.94970703125, "loss_alignment_w": 0.474853515625, "loss_sft": 0.25962619949132204, "loss_total_v6": 0.7348611727356911, "step": 3587 }, { "epoch": 0.8275027027027027, "grad_norm": 28.45660400390625, "learning_rate": 6.0237944114775385e-06, "loss": 12.217432975769043, "loss_alignment": 0.9140625, "loss_alignment_w": 0.45703125, "loss_sft": 0.30672615952789783, "loss_total_v6": 0.7635895684361458, "step": 3588 }, { "epoch": 0.8277333333333333, "grad_norm": 27.00086212158203, "learning_rate": 6.021823155423888e-06, "loss": 11.552401542663574, "loss_alignment": 0.90673828125, "loss_alignment_w": 0.453369140625, "loss_sft": 0.268823828548193, "loss_total_v6": 0.7220251113176346, "step": 3589 }, { "epoch": 0.827963963963964, "grad_norm": 24.23578453063965, "learning_rate": 6.019851733607744e-06, "loss": 11.50327205657959, "loss_alignment": 0.892578125, "loss_alignment_w": 0.4462890625, "loss_sft": 0.27309267595410347, "loss_total_v6": 0.7189544886350632, "step": 3590 }, { "epoch": 0.8281945945945945, "grad_norm": 23.317346572875977, "learning_rate": 6.017880146348914e-06, "loss": 11.785801887512207, "loss_alignment": 0.916015625, "loss_alignment_w": 0.4580078125, "loss_sft": 0.2782691475003958, "loss_total_v6": 0.7366126328706741, "step": 3591 }, { "epoch": 0.8284252252252252, "grad_norm": 24.007062911987305, "learning_rate": 6.0159083939672326e-06, "loss": 11.591413497924805, "loss_alignment": 0.880859375, "loss_alignment_w": 0.4404296875, "loss_sft": 0.2841709367930889, "loss_total_v6": 0.7244633138179779, "step": 3592 }, { "epoch": 0.8286558558558559, "grad_norm": 26.32392692565918, "learning_rate": 6.013936476782563e-06, "loss": 12.279058456420898, "loss_alignment": 0.927734375, "loss_alignment_w": 0.4638671875, "loss_sft": 0.3032229896634817, "loss_total_v6": 0.767441138625145, "step": 3593 }, { "epoch": 0.8288864864864864, "grad_norm": 27.580646514892578, "learning_rate": 6.011964395114794e-06, "loss": 11.433645248413086, "loss_alignment": 0.9072265625, "loss_alignment_w": 0.45361328125, "loss_sft": 0.26118785236030817, "loss_total_v6": 0.7146027684211731, "step": 3594 }, { "epoch": 0.8291171171171171, "grad_norm": 24.633136749267578, "learning_rate": 6.009992149283841e-06, "loss": 12.098825454711914, "loss_alignment": 0.931640625, "loss_alignment_w": 0.4658203125, "loss_sft": 0.2896848861128092, "loss_total_v6": 0.7561765685677528, "step": 3595 }, { "epoch": 0.8293477477477478, "grad_norm": 25.2099609375, "learning_rate": 6.0080197396096465e-06, "loss": 11.621377944946289, "loss_alignment": 0.9140625, "loss_alignment_w": 0.45703125, "loss_sft": 0.26930486410856247, "loss_total_v6": 0.7263360992074013, "step": 3596 }, { "epoch": 0.8295783783783783, "grad_norm": 24.53080177307129, "learning_rate": 6.006047166412178e-06, "loss": 11.803430557250977, "loss_alignment": 0.9208984375, "loss_alignment_w": 0.46044921875, "loss_sft": 0.27750933915376663, "loss_total_v6": 0.7377144396305084, "step": 3597 }, { "epoch": 0.829809009009009, "grad_norm": 23.824262619018555, "learning_rate": 6.004074430011433e-06, "loss": 11.992870330810547, "loss_alignment": 0.95166015625, "loss_alignment_w": 0.475830078125, "loss_sft": 0.27396841533482075, "loss_total_v6": 0.7495543509721756, "step": 3598 }, { "epoch": 0.8300396396396397, "grad_norm": 25.269054412841797, "learning_rate": 6.002101530727432e-06, "loss": 11.732982635498047, "loss_alignment": 0.90966796875, "loss_alignment_w": 0.454833984375, "loss_sft": 0.27814168483018875, "loss_total_v6": 0.7333113625645638, "step": 3599 }, { "epoch": 0.8302702702702702, "grad_norm": 22.747941970825195, "learning_rate": 6.000128468880223e-06, "loss": 12.027026176452637, "loss_alignment": 0.90087890625, "loss_alignment_w": 0.450439453125, "loss_sft": 0.30086819268763065, "loss_total_v6": 0.7516891062259674, "step": 3600 }, { "epoch": 0.8305009009009009, "grad_norm": 22.117687225341797, "learning_rate": 5.998155244789881e-06, "loss": 11.996366500854492, "loss_alignment": 0.91552734375, "loss_alignment_w": 0.457763671875, "loss_sft": 0.29271121323108673, "loss_total_v6": 0.7497729659080505, "step": 3601 }, { "epoch": 0.8307315315315316, "grad_norm": 23.911603927612305, "learning_rate": 5.996181858776508e-06, "loss": 11.623231887817383, "loss_alignment": 0.91845703125, "loss_alignment_w": 0.459228515625, "loss_sft": 0.2680779807269573, "loss_total_v6": 0.7264519855380058, "step": 3602 }, { "epoch": 0.8309621621621621, "grad_norm": 26.57411003112793, "learning_rate": 5.99420831116023e-06, "loss": 11.430070877075195, "loss_alignment": 0.888671875, "loss_alignment_w": 0.4443359375, "loss_sft": 0.270241878926754, "loss_total_v6": 0.7143794298171997, "step": 3603 }, { "epoch": 0.8311927927927928, "grad_norm": 23.737478256225586, "learning_rate": 5.9922346022611995e-06, "loss": 11.75700569152832, "loss_alignment": 0.91796875, "loss_alignment_w": 0.458984375, "loss_sft": 0.27668297104537487, "loss_total_v6": 0.73481285572052, "step": 3604 }, { "epoch": 0.8314234234234235, "grad_norm": 25.078907012939453, "learning_rate": 5.990260732399598e-06, "loss": 11.982131958007812, "loss_alignment": 0.92138671875, "loss_alignment_w": 0.460693359375, "loss_sft": 0.28831195645034313, "loss_total_v6": 0.7488832622766495, "step": 3605 }, { "epoch": 0.831654054054054, "grad_norm": 31.075363159179688, "learning_rate": 5.988286701895631e-06, "loss": 12.21557903289795, "loss_alignment": 0.939453125, "loss_alignment_w": 0.4697265625, "loss_sft": 0.293182535097003, "loss_total_v6": 0.76347366720438, "step": 3606 }, { "epoch": 0.8318846846846847, "grad_norm": 28.12286949157715, "learning_rate": 5.9863125110695295e-06, "loss": 11.94339656829834, "loss_alignment": 0.91064453125, "loss_alignment_w": 0.455322265625, "loss_sft": 0.29030080512166023, "loss_total_v6": 0.7464622929692268, "step": 3607 }, { "epoch": 0.8321153153153154, "grad_norm": 26.16986846923828, "learning_rate": 5.984338160241552e-06, "loss": 12.038475036621094, "loss_alignment": 0.91650390625, "loss_alignment_w": 0.458251953125, "loss_sft": 0.2947935685515404, "loss_total_v6": 0.752404659986496, "step": 3608 }, { "epoch": 0.8323459459459459, "grad_norm": 23.275232315063477, "learning_rate": 5.982363649731981e-06, "loss": 11.540412902832031, "loss_alignment": 0.916015625, "loss_alignment_w": 0.4580078125, "loss_sft": 0.26413774117827415, "loss_total_v6": 0.721275806427002, "step": 3609 }, { "epoch": 0.8325765765765766, "grad_norm": 25.084379196166992, "learning_rate": 5.980388979861127e-06, "loss": 12.313039779663086, "loss_alignment": 0.92236328125, "loss_alignment_w": 0.461181640625, "loss_sft": 0.3081391956657171, "loss_total_v6": 0.7695650085806847, "step": 3610 }, { "epoch": 0.8328072072072072, "grad_norm": 27.221088409423828, "learning_rate": 5.978414150949324e-06, "loss": 11.723752975463867, "loss_alignment": 0.904296875, "loss_alignment_w": 0.4521484375, "loss_sft": 0.2801130339503288, "loss_total_v6": 0.7327345460653305, "step": 3611 }, { "epoch": 0.8330378378378378, "grad_norm": 25.739913940429688, "learning_rate": 5.976439163316936e-06, "loss": 12.122623443603516, "loss_alignment": 0.91650390625, "loss_alignment_w": 0.458251953125, "loss_sft": 0.29916784539818764, "loss_total_v6": 0.7576639279723167, "step": 3612 }, { "epoch": 0.8332684684684685, "grad_norm": 24.245210647583008, "learning_rate": 5.974464017284349e-06, "loss": 10.927654266357422, "loss_alignment": 0.892578125, "loss_alignment_w": 0.4462890625, "loss_sft": 0.23543811589479446, "loss_total_v6": 0.6829784065485001, "step": 3613 }, { "epoch": 0.8334990990990991, "grad_norm": 26.954490661621094, "learning_rate": 5.972488713171975e-06, "loss": 12.007488250732422, "loss_alignment": 0.92333984375, "loss_alignment_w": 0.461669921875, "loss_sft": 0.2873637229204178, "loss_total_v6": 0.7504679709672928, "step": 3614 }, { "epoch": 0.8337297297297297, "grad_norm": 33.413917541503906, "learning_rate": 5.970513251300251e-06, "loss": 11.943102836608887, "loss_alignment": 0.92138671875, "loss_alignment_w": 0.460693359375, "loss_sft": 0.28675767965614796, "loss_total_v6": 0.746443934738636, "step": 3615 }, { "epoch": 0.8339603603603604, "grad_norm": 25.527528762817383, "learning_rate": 5.968537631989645e-06, "loss": 12.089239120483398, "loss_alignment": 0.93212890625, "loss_alignment_w": 0.466064453125, "loss_sft": 0.2898487262427807, "loss_total_v6": 0.7555774450302124, "step": 3616 }, { "epoch": 0.834190990990991, "grad_norm": 24.306421279907227, "learning_rate": 5.966561855560642e-06, "loss": 11.173639297485352, "loss_alignment": 0.896484375, "loss_alignment_w": 0.4482421875, "loss_sft": 0.2490573963150382, "loss_total_v6": 0.6983524262905121, "step": 3617 }, { "epoch": 0.8344216216216216, "grad_norm": 23.11979103088379, "learning_rate": 5.96458592233376e-06, "loss": 12.354549407958984, "loss_alignment": 0.908203125, "loss_alignment_w": 0.4541015625, "loss_sft": 0.317828930914402, "loss_total_v6": 0.7721593603491783, "step": 3618 }, { "epoch": 0.8346522522522523, "grad_norm": 26.376222610473633, "learning_rate": 5.962609832629538e-06, "loss": 11.701908111572266, "loss_alignment": 0.9248046875, "loss_alignment_w": 0.46240234375, "loss_sft": 0.2692873626947403, "loss_total_v6": 0.7313692346215248, "step": 3619 }, { "epoch": 0.8348828828828829, "grad_norm": 23.252948760986328, "learning_rate": 5.9606335867685424e-06, "loss": 12.172804832458496, "loss_alignment": 0.90576171875, "loss_alignment_w": 0.452880859375, "loss_sft": 0.30724806152284145, "loss_total_v6": 0.7608002871274948, "step": 3620 }, { "epoch": 0.8351135135135135, "grad_norm": 25.074920654296875, "learning_rate": 5.9586571850713645e-06, "loss": 11.621222496032715, "loss_alignment": 0.8955078125, "loss_alignment_w": 0.44775390625, "loss_sft": 0.2798694968223572, "loss_total_v6": 0.7263264134526253, "step": 3621 }, { "epoch": 0.8353441441441442, "grad_norm": 24.009138107299805, "learning_rate": 5.9566806278586206e-06, "loss": 11.983830451965332, "loss_alignment": 0.9072265625, "loss_alignment_w": 0.45361328125, "loss_sft": 0.29398756474256516, "loss_total_v6": 0.7489893659949303, "step": 3622 }, { "epoch": 0.8355747747747748, "grad_norm": 26.356231689453125, "learning_rate": 5.9547039154509526e-06, "loss": 11.780254364013672, "loss_alignment": 0.92333984375, "loss_alignment_w": 0.461669921875, "loss_sft": 0.27476388588547707, "loss_total_v6": 0.7362659499049187, "step": 3623 }, { "epoch": 0.8358054054054054, "grad_norm": 29.652965545654297, "learning_rate": 5.952727048169025e-06, "loss": 12.395179748535156, "loss_alignment": 0.93603515625, "loss_alignment_w": 0.468017578125, "loss_sft": 0.3068336993455887, "loss_total_v6": 0.7746987268328667, "step": 3624 }, { "epoch": 0.836036036036036, "grad_norm": 24.39432716369629, "learning_rate": 5.950750026333534e-06, "loss": 11.160662651062012, "loss_alignment": 0.89013671875, "loss_alignment_w": 0.445068359375, "loss_sft": 0.2518779318779707, "loss_total_v6": 0.6975413784384727, "step": 3625 }, { "epoch": 0.8362666666666667, "grad_norm": 25.74016761779785, "learning_rate": 5.948772850265193e-06, "loss": 11.588201522827148, "loss_alignment": 0.89697265625, "loss_alignment_w": 0.448486328125, "loss_sft": 0.27617296390235424, "loss_total_v6": 0.7242625504732132, "step": 3626 }, { "epoch": 0.8364972972972973, "grad_norm": 24.85071563720703, "learning_rate": 5.946795520284745e-06, "loss": 12.389235496520996, "loss_alignment": 0.93505859375, "loss_alignment_w": 0.467529296875, "loss_sft": 0.306538512930274, "loss_total_v6": 0.7743272185325623, "step": 3627 }, { "epoch": 0.8367279279279279, "grad_norm": 24.430007934570312, "learning_rate": 5.9448180367129595e-06, "loss": 12.071632385253906, "loss_alignment": 0.92431640625, "loss_alignment_w": 0.462158203125, "loss_sft": 0.29346317797899246, "loss_total_v6": 0.7544769868254662, "step": 3628 }, { "epoch": 0.8369585585585586, "grad_norm": 25.53422737121582, "learning_rate": 5.942840399870624e-06, "loss": 11.858059883117676, "loss_alignment": 0.90625, "loss_alignment_w": 0.453125, "loss_sft": 0.28779010847210884, "loss_total_v6": 0.7411287426948547, "step": 3629 }, { "epoch": 0.8371891891891892, "grad_norm": 24.10359764099121, "learning_rate": 5.940862610078557e-06, "loss": 11.450240135192871, "loss_alignment": 0.92333984375, "loss_alignment_w": 0.461669921875, "loss_sft": 0.25230685994029045, "loss_total_v6": 0.7156399860978127, "step": 3630 }, { "epoch": 0.8374198198198198, "grad_norm": 25.111846923828125, "learning_rate": 5.938884667657602e-06, "loss": 11.508548736572266, "loss_alignment": 0.912109375, "loss_alignment_w": 0.4560546875, "loss_sft": 0.26336690224707127, "loss_total_v6": 0.7192842736840248, "step": 3631 }, { "epoch": 0.8376504504504505, "grad_norm": 22.57262420654297, "learning_rate": 5.936906572928625e-06, "loss": 11.612911224365234, "loss_alignment": 0.92919921875, "loss_alignment_w": 0.464599609375, "loss_sft": 0.2606580164283514, "loss_total_v6": 0.7258069738745689, "step": 3632 }, { "epoch": 0.8378810810810811, "grad_norm": 24.170679092407227, "learning_rate": 5.934928326212513e-06, "loss": 11.540204048156738, "loss_alignment": 0.91796875, "loss_alignment_w": 0.458984375, "loss_sft": 0.2613781616091728, "loss_total_v6": 0.7212627828121185, "step": 3633 }, { "epoch": 0.8381117117117117, "grad_norm": 24.610458374023438, "learning_rate": 5.932949927830186e-06, "loss": 11.946493148803711, "loss_alignment": 0.90966796875, "loss_alignment_w": 0.454833984375, "loss_sft": 0.2916692662984133, "loss_total_v6": 0.7466558516025543, "step": 3634 }, { "epoch": 0.8383423423423424, "grad_norm": 24.753963470458984, "learning_rate": 5.930971378102585e-06, "loss": 11.691279411315918, "loss_alignment": 0.91796875, "loss_alignment_w": 0.458984375, "loss_sft": 0.2715985160320997, "loss_total_v6": 0.7307049706578255, "step": 3635 }, { "epoch": 0.838572972972973, "grad_norm": 25.410749435424805, "learning_rate": 5.92899267735067e-06, "loss": 11.929426193237305, "loss_alignment": 0.9248046875, "loss_alignment_w": 0.46240234375, "loss_sft": 0.2819813769310713, "loss_total_v6": 0.7455891519784927, "step": 3636 }, { "epoch": 0.8388036036036036, "grad_norm": 26.338226318359375, "learning_rate": 5.927013825895435e-06, "loss": 11.113534927368164, "loss_alignment": 0.88818359375, "loss_alignment_w": 0.444091796875, "loss_sft": 0.25048889964818954, "loss_total_v6": 0.6945959478616714, "step": 3637 }, { "epoch": 0.8390342342342343, "grad_norm": 25.065866470336914, "learning_rate": 5.925034824057892e-06, "loss": 11.962703704833984, "loss_alignment": 0.90478515625, "loss_alignment_w": 0.452392578125, "loss_sft": 0.29384211637079716, "loss_total_v6": 0.7476690113544464, "step": 3638 }, { "epoch": 0.8392648648648648, "grad_norm": 27.637434005737305, "learning_rate": 5.923055672159079e-06, "loss": 11.847734451293945, "loss_alignment": 0.9130859375, "loss_alignment_w": 0.45654296875, "loss_sft": 0.28380309604108334, "loss_total_v6": 0.7404834106564522, "step": 3639 }, { "epoch": 0.8394954954954955, "grad_norm": 24.4916934967041, "learning_rate": 5.921076370520058e-06, "loss": 11.645495414733887, "loss_alignment": 0.8818359375, "loss_alignment_w": 0.44091796875, "loss_sft": 0.28607103414833546, "loss_total_v6": 0.7278434783220291, "step": 3640 }, { "epoch": 0.8397261261261261, "grad_norm": 25.44816780090332, "learning_rate": 5.919096919461918e-06, "loss": 11.645662307739258, "loss_alignment": 0.90087890625, "loss_alignment_w": 0.450439453125, "loss_sft": 0.27776538766920567, "loss_total_v6": 0.7278539016842842, "step": 3641 }, { "epoch": 0.8399567567567567, "grad_norm": 23.093318939208984, "learning_rate": 5.91711731930577e-06, "loss": 10.883807182312012, "loss_alignment": 0.89208984375, "loss_alignment_w": 0.446044921875, "loss_sft": 0.23385732248425484, "loss_total_v6": 0.6802379488945007, "step": 3642 }, { "epoch": 0.8401873873873874, "grad_norm": 23.833568572998047, "learning_rate": 5.915137570372745e-06, "loss": 11.83159065246582, "loss_alignment": 0.89404296875, "loss_alignment_w": 0.447021484375, "loss_sft": 0.2920714486390352, "loss_total_v6": 0.739474393427372, "step": 3643 }, { "epoch": 0.840418018018018, "grad_norm": 24.95167350769043, "learning_rate": 5.913157672984007e-06, "loss": 12.075494766235352, "loss_alignment": 0.91455078125, "loss_alignment_w": 0.457275390625, "loss_sft": 0.2980381101369858, "loss_total_v6": 0.7547184228897095, "step": 3644 }, { "epoch": 0.8406486486486486, "grad_norm": 24.1223201751709, "learning_rate": 5.911177627460739e-06, "loss": 11.832054138183594, "loss_alignment": 0.884765625, "loss_alignment_w": 0.4423828125, "loss_sft": 0.29705951921641827, "loss_total_v6": 0.739503376185894, "step": 3645 }, { "epoch": 0.8408792792792793, "grad_norm": 26.086997985839844, "learning_rate": 5.909197434124145e-06, "loss": 11.847196578979492, "loss_alignment": 0.90625, "loss_alignment_w": 0.453125, "loss_sft": 0.28795041888952255, "loss_total_v6": 0.7404497787356377, "step": 3646 }, { "epoch": 0.8411099099099099, "grad_norm": 26.060161590576172, "learning_rate": 5.907217093295461e-06, "loss": 12.52676010131836, "loss_alignment": 0.91259765625, "loss_alignment_w": 0.456298828125, "loss_sft": 0.32633379101753235, "loss_total_v6": 0.7829225435853004, "step": 3647 }, { "epoch": 0.8413405405405405, "grad_norm": 22.72506332397461, "learning_rate": 5.905236605295941e-06, "loss": 11.650091171264648, "loss_alignment": 0.90087890625, "loss_alignment_w": 0.450439453125, "loss_sft": 0.2771114259958267, "loss_total_v6": 0.7281307280063629, "step": 3648 }, { "epoch": 0.8415711711711712, "grad_norm": 25.45795249938965, "learning_rate": 5.903255970446862e-06, "loss": 11.992227554321289, "loss_alignment": 0.92041015625, "loss_alignment_w": 0.460205078125, "loss_sft": 0.28895818442106247, "loss_total_v6": 0.7495142370462418, "step": 3649 }, { "epoch": 0.8418018018018018, "grad_norm": 25.077499389648438, "learning_rate": 5.90127518906953e-06, "loss": 12.277568817138672, "loss_alignment": 0.908203125, "loss_alignment_w": 0.4541015625, "loss_sft": 0.312285203486681, "loss_total_v6": 0.7673480808734894, "step": 3650 }, { "epoch": 0.8420324324324324, "grad_norm": 23.467592239379883, "learning_rate": 5.89929426148527e-06, "loss": 12.380571365356445, "loss_alignment": 0.93115234375, "loss_alignment_w": 0.465576171875, "loss_sft": 0.3075991403311491, "loss_total_v6": 0.7737856805324554, "step": 3651 }, { "epoch": 0.8422630630630631, "grad_norm": 24.108909606933594, "learning_rate": 5.897313188015433e-06, "loss": 12.108573913574219, "loss_alignment": 0.93359375, "loss_alignment_w": 0.466796875, "loss_sft": 0.29066042229533195, "loss_total_v6": 0.7567859143018723, "step": 3652 }, { "epoch": 0.8424936936936936, "grad_norm": 27.389909744262695, "learning_rate": 5.8953319689813944e-06, "loss": 11.968140602111816, "loss_alignment": 0.92724609375, "loss_alignment_w": 0.463623046875, "loss_sft": 0.28469093330204487, "loss_total_v6": 0.7480087950825691, "step": 3653 }, { "epoch": 0.8427243243243243, "grad_norm": 25.876750946044922, "learning_rate": 5.893350604704552e-06, "loss": 11.492260932922363, "loss_alignment": 0.89453125, "loss_alignment_w": 0.447265625, "loss_sft": 0.271443173289299, "loss_total_v6": 0.7182662934064865, "step": 3654 }, { "epoch": 0.842954954954955, "grad_norm": 25.10433006286621, "learning_rate": 5.891369095506326e-06, "loss": 11.756113052368164, "loss_alignment": 0.89501953125, "loss_alignment_w": 0.447509765625, "loss_sft": 0.28854427486658096, "loss_total_v6": 0.7347570955753326, "step": 3655 }, { "epoch": 0.8431855855855855, "grad_norm": 25.546663284301758, "learning_rate": 5.889387441708162e-06, "loss": 11.469583511352539, "loss_alignment": 0.89990234375, "loss_alignment_w": 0.449951171875, "loss_sft": 0.267294529825449, "loss_total_v6": 0.7168489769101143, "step": 3656 }, { "epoch": 0.8434162162162162, "grad_norm": 23.678544998168945, "learning_rate": 5.887405643631528e-06, "loss": 11.776421546936035, "loss_alignment": 0.884765625, "loss_alignment_w": 0.4423828125, "loss_sft": 0.2939639277756214, "loss_total_v6": 0.7360263094305992, "step": 3657 }, { "epoch": 0.8436468468468469, "grad_norm": 26.288843154907227, "learning_rate": 5.885423701597918e-06, "loss": 12.126026153564453, "loss_alignment": 0.90966796875, "loss_alignment_w": 0.454833984375, "loss_sft": 0.30290530249476433, "loss_total_v6": 0.7578766047954559, "step": 3658 }, { "epoch": 0.8438774774774774, "grad_norm": 25.862051010131836, "learning_rate": 5.883441615928843e-06, "loss": 11.590829849243164, "loss_alignment": 0.9150390625, "loss_alignment_w": 0.45751953125, "loss_sft": 0.267822852358222, "loss_total_v6": 0.7244268730282784, "step": 3659 }, { "epoch": 0.8441081081081081, "grad_norm": 26.207401275634766, "learning_rate": 5.8814593869458455e-06, "loss": 11.898946762084961, "loss_alignment": 0.8955078125, "loss_alignment_w": 0.44775390625, "loss_sft": 0.2961133811622858, "loss_total_v6": 0.743684209883213, "step": 3660 }, { "epoch": 0.8443387387387388, "grad_norm": 26.397369384765625, "learning_rate": 5.879477014970485e-06, "loss": 11.437568664550781, "loss_alignment": 0.9169921875, "loss_alignment_w": 0.45849609375, "loss_sft": 0.25552794337272644, "loss_total_v6": 0.7148480415344238, "step": 3661 }, { "epoch": 0.8445693693693693, "grad_norm": 26.94974136352539, "learning_rate": 5.877494500324347e-06, "loss": 11.736364364624023, "loss_alignment": 0.892578125, "loss_alignment_w": 0.4462890625, "loss_sft": 0.2872794587165117, "loss_total_v6": 0.7335227504372597, "step": 3662 }, { "epoch": 0.8448, "grad_norm": 26.562950134277344, "learning_rate": 5.875511843329039e-06, "loss": 12.348352432250977, "loss_alignment": 0.90966796875, "loss_alignment_w": 0.454833984375, "loss_sft": 0.3173958398401737, "loss_total_v6": 0.7717720568180084, "step": 3663 }, { "epoch": 0.8450306306306307, "grad_norm": 24.940570831298828, "learning_rate": 5.8735290443061944e-06, "loss": 12.03346061706543, "loss_alignment": 0.9228515625, "loss_alignment_w": 0.46142578125, "loss_sft": 0.291199604049325, "loss_total_v6": 0.7520913034677505, "step": 3664 }, { "epoch": 0.8452612612612612, "grad_norm": 24.000083923339844, "learning_rate": 5.871546103577464e-06, "loss": 11.627610206604004, "loss_alignment": 0.90380859375, "loss_alignment_w": 0.451904296875, "loss_sft": 0.2731733750551939, "loss_total_v6": 0.726725623011589, "step": 3665 }, { "epoch": 0.8454918918918919, "grad_norm": 23.31669807434082, "learning_rate": 5.869563021464528e-06, "loss": 11.91729736328125, "loss_alignment": 0.94873046875, "loss_alignment_w": 0.474365234375, "loss_sft": 0.27121351659297943, "loss_total_v6": 0.7448311001062393, "step": 3666 }, { "epoch": 0.8457225225225226, "grad_norm": 24.4226131439209, "learning_rate": 5.867579798289086e-06, "loss": 11.984333992004395, "loss_alignment": 0.90576171875, "loss_alignment_w": 0.452880859375, "loss_sft": 0.29614003747701645, "loss_total_v6": 0.7490208968520164, "step": 3667 }, { "epoch": 0.8459531531531531, "grad_norm": 23.449201583862305, "learning_rate": 5.865596434372858e-06, "loss": 11.837244033813477, "loss_alignment": 0.87890625, "loss_alignment_w": 0.439453125, "loss_sft": 0.3000237010419369, "loss_total_v6": 0.7398277819156647, "step": 3668 }, { "epoch": 0.8461837837837838, "grad_norm": 23.66835594177246, "learning_rate": 5.863612930037594e-06, "loss": 11.608710289001465, "loss_alignment": 0.931640625, "loss_alignment_w": 0.4658203125, "loss_sft": 0.25929684564471245, "loss_total_v6": 0.7255443781614304, "step": 3669 }, { "epoch": 0.8464144144144145, "grad_norm": 26.131454467773438, "learning_rate": 5.86162928560506e-06, "loss": 12.116649627685547, "loss_alignment": 0.92041015625, "loss_alignment_w": 0.460205078125, "loss_sft": 0.2966887652873993, "loss_total_v6": 0.7572905793786049, "step": 3670 }, { "epoch": 0.846645045045045, "grad_norm": 26.939579010009766, "learning_rate": 5.859645501397048e-06, "loss": 12.250089645385742, "loss_alignment": 0.92041015625, "loss_alignment_w": 0.460205078125, "loss_sft": 0.3061883933842182, "loss_total_v6": 0.7656305730342865, "step": 3671 }, { "epoch": 0.8468756756756757, "grad_norm": 24.96284294128418, "learning_rate": 5.8576615777353725e-06, "loss": 11.809518814086914, "loss_alignment": 0.8984375, "loss_alignment_w": 0.44921875, "loss_sft": 0.28945598751306534, "loss_total_v6": 0.7380949109792709, "step": 3672 }, { "epoch": 0.8471063063063063, "grad_norm": 25.38872528076172, "learning_rate": 5.85567751494187e-06, "loss": 12.139826774597168, "loss_alignment": 0.94580078125, "loss_alignment_w": 0.472900390625, "loss_sft": 0.2851673997938633, "loss_total_v6": 0.7587391659617424, "step": 3673 }, { "epoch": 0.8473369369369369, "grad_norm": 24.500890731811523, "learning_rate": 5.853693313338399e-06, "loss": 11.332748413085938, "loss_alignment": 0.87255859375, "loss_alignment_w": 0.436279296875, "loss_sft": 0.27212423644959927, "loss_total_v6": 0.7082967460155487, "step": 3674 }, { "epoch": 0.8475675675675676, "grad_norm": 23.986257553100586, "learning_rate": 5.851708973246841e-06, "loss": 11.320435523986816, "loss_alignment": 0.88916015625, "loss_alignment_w": 0.444580078125, "loss_sft": 0.2635574769228697, "loss_total_v6": 0.7075272127985954, "step": 3675 }, { "epoch": 0.8477981981981982, "grad_norm": 24.670473098754883, "learning_rate": 5.849724494989104e-06, "loss": 11.604116439819336, "loss_alignment": 0.9208984375, "loss_alignment_w": 0.46044921875, "loss_sft": 0.26517421938478947, "loss_total_v6": 0.7252572700381279, "step": 3676 }, { "epoch": 0.8480288288288288, "grad_norm": 24.618385314941406, "learning_rate": 5.8477398788871086e-06, "loss": 11.297922134399414, "loss_alignment": 0.89306640625, "loss_alignment_w": 0.446533203125, "loss_sft": 0.26054823212325573, "loss_total_v6": 0.7061201333999634, "step": 3677 }, { "epoch": 0.8482594594594595, "grad_norm": 28.109882354736328, "learning_rate": 5.845755125262808e-06, "loss": 11.859537124633789, "loss_alignment": 0.93310546875, "loss_alignment_w": 0.466552734375, "loss_sft": 0.27532447315752506, "loss_total_v6": 0.7412210702896118, "step": 3678 }, { "epoch": 0.8484900900900901, "grad_norm": 24.16754913330078, "learning_rate": 5.843770234438172e-06, "loss": 11.950016975402832, "loss_alignment": 0.90673828125, "loss_alignment_w": 0.453369140625, "loss_sft": 0.29352214746177197, "loss_total_v6": 0.7468760311603546, "step": 3679 }, { "epoch": 0.8487207207207207, "grad_norm": 25.970144271850586, "learning_rate": 5.841785206735192e-06, "loss": 11.959638595581055, "loss_alignment": 0.87109375, "loss_alignment_w": 0.435546875, "loss_sft": 0.3125103488564491, "loss_total_v6": 0.7474774047732353, "step": 3680 }, { "epoch": 0.8489513513513514, "grad_norm": 26.04503059387207, "learning_rate": 5.83980004247589e-06, "loss": 12.138813018798828, "loss_alignment": 0.8896484375, "loss_alignment_w": 0.44482421875, "loss_sft": 0.31443145498633385, "loss_total_v6": 0.7586758211255074, "step": 3681 }, { "epoch": 0.849181981981982, "grad_norm": 22.20587730407715, "learning_rate": 5.8378147419822984e-06, "loss": 11.619854927062988, "loss_alignment": 0.89501953125, "loss_alignment_w": 0.447509765625, "loss_sft": 0.27894476242363453, "loss_total_v6": 0.7262409031391144, "step": 3682 }, { "epoch": 0.8494126126126126, "grad_norm": 25.813655853271484, "learning_rate": 5.835829305576477e-06, "loss": 12.324289321899414, "loss_alignment": 0.92626953125, "loss_alignment_w": 0.463134765625, "loss_sft": 0.30753004737198353, "loss_total_v6": 0.7702680677175522, "step": 3683 }, { "epoch": 0.8496432432432433, "grad_norm": 26.94569206237793, "learning_rate": 5.8338437335805124e-06, "loss": 11.656293869018555, "loss_alignment": 0.92578125, "loss_alignment_w": 0.462890625, "loss_sft": 0.26642123237252235, "loss_total_v6": 0.7285184189677238, "step": 3684 }, { "epoch": 0.8498738738738739, "grad_norm": 24.774959564208984, "learning_rate": 5.831858026316504e-06, "loss": 11.957897186279297, "loss_alignment": 0.90625, "loss_alignment_w": 0.453125, "loss_sft": 0.2944266330450773, "loss_total_v6": 0.7473685592412949, "step": 3685 }, { "epoch": 0.8501045045045045, "grad_norm": 24.208091735839844, "learning_rate": 5.829872184106579e-06, "loss": 11.812505722045898, "loss_alignment": 0.88818359375, "loss_alignment_w": 0.444091796875, "loss_sft": 0.29466280341148376, "loss_total_v6": 0.7382815927267075, "step": 3686 }, { "epoch": 0.8503351351351351, "grad_norm": 25.066024780273438, "learning_rate": 5.82788620727289e-06, "loss": 11.062037467956543, "loss_alignment": 0.90185546875, "loss_alignment_w": 0.450927734375, "loss_sft": 0.2408920992165804, "loss_total_v6": 0.6913773566484451, "step": 3687 }, { "epoch": 0.8505657657657658, "grad_norm": 23.07847785949707, "learning_rate": 5.825900096137599e-06, "loss": 11.543191909790039, "loss_alignment": 0.89013671875, "loss_alignment_w": 0.445068359375, "loss_sft": 0.2771288361400366, "loss_total_v6": 0.7214495316147804, "step": 3688 }, { "epoch": 0.8507963963963964, "grad_norm": 26.82064437866211, "learning_rate": 5.823913851022901e-06, "loss": 11.935404777526855, "loss_alignment": 0.89208984375, "loss_alignment_w": 0.446044921875, "loss_sft": 0.3007723428308964, "loss_total_v6": 0.7459627911448479, "step": 3689 }, { "epoch": 0.851027027027027, "grad_norm": 23.970258712768555, "learning_rate": 5.8219274722510096e-06, "loss": 11.676036834716797, "loss_alignment": 0.884765625, "loss_alignment_w": 0.4423828125, "loss_sft": 0.2868659496307373, "loss_total_v6": 0.7297522649168968, "step": 3690 }, { "epoch": 0.8512576576576577, "grad_norm": 24.684221267700195, "learning_rate": 5.819940960144161e-06, "loss": 11.558277130126953, "loss_alignment": 0.87939453125, "loss_alignment_w": 0.439697265625, "loss_sft": 0.2817795444279909, "loss_total_v6": 0.7223923206329346, "step": 3691 }, { "epoch": 0.8514882882882883, "grad_norm": 28.503429412841797, "learning_rate": 5.81795431502461e-06, "loss": 11.961502075195312, "loss_alignment": 0.91162109375, "loss_alignment_w": 0.455810546875, "loss_sft": 0.2933092229068279, "loss_total_v6": 0.747593879699707, "step": 3692 }, { "epoch": 0.8517189189189189, "grad_norm": 24.721935272216797, "learning_rate": 5.815967537214634e-06, "loss": 11.712308883666992, "loss_alignment": 0.935546875, "loss_alignment_w": 0.4677734375, "loss_sft": 0.2637880854308605, "loss_total_v6": 0.7320192754268646, "step": 3693 }, { "epoch": 0.8519495495495496, "grad_norm": 26.282405853271484, "learning_rate": 5.813980627036535e-06, "loss": 11.712507247924805, "loss_alignment": 0.91455078125, "loss_alignment_w": 0.457275390625, "loss_sft": 0.27454270981252193, "loss_total_v6": 0.7320317178964615, "step": 3694 }, { "epoch": 0.8521801801801802, "grad_norm": 25.131488800048828, "learning_rate": 5.811993584812631e-06, "loss": 11.713224411010742, "loss_alignment": 0.9052734375, "loss_alignment_w": 0.45263671875, "loss_sft": 0.2798975817859173, "loss_total_v6": 0.732076533138752, "step": 3695 }, { "epoch": 0.8524108108108108, "grad_norm": 26.363323211669922, "learning_rate": 5.810006410865267e-06, "loss": 11.184934616088867, "loss_alignment": 0.8779296875, "loss_alignment_w": 0.43896484375, "loss_sft": 0.2604597229510546, "loss_total_v6": 0.6990583688020706, "step": 3696 }, { "epoch": 0.8526414414414415, "grad_norm": 26.849266052246094, "learning_rate": 5.8080191055168064e-06, "loss": 11.825709342956543, "loss_alignment": 0.8974609375, "loss_alignment_w": 0.44873046875, "loss_sft": 0.29107824712991714, "loss_total_v6": 0.7391068190336227, "step": 3697 }, { "epoch": 0.852872072072072, "grad_norm": 26.433931350708008, "learning_rate": 5.806031669089634e-06, "loss": 11.984532356262207, "loss_alignment": 0.873046875, "loss_alignment_w": 0.4365234375, "loss_sft": 0.31186898052692413, "loss_total_v6": 0.7490332499146461, "step": 3698 }, { "epoch": 0.8531027027027027, "grad_norm": 24.286582946777344, "learning_rate": 5.804044101906156e-06, "loss": 12.208568572998047, "loss_alignment": 0.93017578125, "loss_alignment_w": 0.465087890625, "loss_sft": 0.2978408560156822, "loss_total_v6": 0.7630355283617973, "step": 3699 }, { "epoch": 0.8533333333333334, "grad_norm": 26.6956729888916, "learning_rate": 5.8020564042888015e-06, "loss": 12.076447486877441, "loss_alignment": 0.9267578125, "loss_alignment_w": 0.46337890625, "loss_sft": 0.2920094281435013, "loss_total_v6": 0.7547779977321625, "step": 3700 }, { "epoch": 0.853563963963964, "grad_norm": 26.0217342376709, "learning_rate": 5.800068576560019e-06, "loss": 11.599867820739746, "loss_alignment": 0.900390625, "loss_alignment_w": 0.4501953125, "loss_sft": 0.27555936574935913, "loss_total_v6": 0.7249917313456535, "step": 3701 }, { "epoch": 0.8537945945945946, "grad_norm": 25.15800666809082, "learning_rate": 5.798080619042276e-06, "loss": 11.661083221435547, "loss_alignment": 0.916015625, "loss_alignment_w": 0.4580078125, "loss_sft": 0.2733122855424881, "loss_total_v6": 0.7288176789879799, "step": 3702 }, { "epoch": 0.8540252252252252, "grad_norm": 25.9604549407959, "learning_rate": 5.796092532058067e-06, "loss": 12.514986991882324, "loss_alignment": 0.8984375, "loss_alignment_w": 0.44921875, "loss_sft": 0.33400552719831467, "loss_total_v6": 0.7821866646409035, "step": 3703 }, { "epoch": 0.8542558558558558, "grad_norm": 25.532167434692383, "learning_rate": 5.794104315929904e-06, "loss": 12.060096740722656, "loss_alignment": 0.91796875, "loss_alignment_w": 0.458984375, "loss_sft": 0.2940240055322647, "loss_total_v6": 0.7537560760974884, "step": 3704 }, { "epoch": 0.8544864864864865, "grad_norm": 29.590879440307617, "learning_rate": 5.7921159709803145e-06, "loss": 12.400102615356445, "loss_alignment": 0.9130859375, "loss_alignment_w": 0.45654296875, "loss_sft": 0.31846342608332634, "loss_total_v6": 0.7750064134597778, "step": 3705 }, { "epoch": 0.8547171171171171, "grad_norm": 24.95232582092285, "learning_rate": 5.790127497531859e-06, "loss": 11.524337768554688, "loss_alignment": 0.8935546875, "loss_alignment_w": 0.44677734375, "loss_sft": 0.2741956654936075, "loss_total_v6": 0.7202711179852486, "step": 3706 }, { "epoch": 0.8549477477477477, "grad_norm": 25.714033126831055, "learning_rate": 5.7881388959071095e-06, "loss": 11.333558082580566, "loss_alignment": 0.90185546875, "loss_alignment_w": 0.450927734375, "loss_sft": 0.25587852485477924, "loss_total_v6": 0.7083474099636078, "step": 3707 }, { "epoch": 0.8551783783783784, "grad_norm": 26.216554641723633, "learning_rate": 5.786150166428661e-06, "loss": 12.620622634887695, "loss_alignment": 0.896484375, "loss_alignment_w": 0.4482421875, "loss_sft": 0.34130968153476715, "loss_total_v6": 0.7887889295816422, "step": 3708 }, { "epoch": 0.855409009009009, "grad_norm": 25.48872184753418, "learning_rate": 5.78416130941913e-06, "loss": 11.512453079223633, "loss_alignment": 0.8857421875, "loss_alignment_w": 0.44287109375, "loss_sft": 0.2772370632737875, "loss_total_v6": 0.719528317451477, "step": 3709 }, { "epoch": 0.8556396396396396, "grad_norm": 26.620702743530273, "learning_rate": 5.782172325201155e-06, "loss": 11.445247650146484, "loss_alignment": 0.9150390625, "loss_alignment_w": 0.45751953125, "loss_sft": 0.2588308062404394, "loss_total_v6": 0.7153280079364777, "step": 3710 }, { "epoch": 0.8558702702702703, "grad_norm": 27.53969383239746, "learning_rate": 5.7801832140973915e-06, "loss": 12.003931045532227, "loss_alignment": 0.91748046875, "loss_alignment_w": 0.458740234375, "loss_sft": 0.29278727248311043, "loss_total_v6": 0.7502457424998283, "step": 3711 }, { "epoch": 0.8561009009009009, "grad_norm": 31.466089248657227, "learning_rate": 5.778193976430518e-06, "loss": 11.708906173706055, "loss_alignment": 0.90966796875, "loss_alignment_w": 0.454833984375, "loss_sft": 0.27721684612333775, "loss_total_v6": 0.7318066880106926, "step": 3712 }, { "epoch": 0.8563315315315315, "grad_norm": 24.880691528320312, "learning_rate": 5.776204612523236e-06, "loss": 11.635024070739746, "loss_alignment": 0.900390625, "loss_alignment_w": 0.4501953125, "loss_sft": 0.27652065455913544, "loss_total_v6": 0.7271889895200729, "step": 3713 }, { "epoch": 0.8565621621621622, "grad_norm": 24.309011459350586, "learning_rate": 5.77421512269826e-06, "loss": 11.887317657470703, "loss_alignment": 0.8935546875, "loss_alignment_w": 0.44677734375, "loss_sft": 0.29643943533301353, "loss_total_v6": 0.7429573684930801, "step": 3714 }, { "epoch": 0.8567927927927927, "grad_norm": 32.75691604614258, "learning_rate": 5.772225507278333e-06, "loss": 11.453773498535156, "loss_alignment": 0.9150390625, "loss_alignment_w": 0.45751953125, "loss_sft": 0.2570443358272314, "loss_total_v6": 0.7158608511090279, "step": 3715 }, { "epoch": 0.8570234234234234, "grad_norm": 26.64501190185547, "learning_rate": 5.770235766586215e-06, "loss": 11.94553279876709, "loss_alignment": 0.8837890625, "loss_alignment_w": 0.44189453125, "loss_sft": 0.304029842838645, "loss_total_v6": 0.7465957552194595, "step": 3716 }, { "epoch": 0.8572540540540541, "grad_norm": 25.01849365234375, "learning_rate": 5.768245900944683e-06, "loss": 11.567667007446289, "loss_alignment": 0.9091796875, "loss_alignment_w": 0.45458984375, "loss_sft": 0.2672143969684839, "loss_total_v6": 0.7229791730642319, "step": 3717 }, { "epoch": 0.8574846846846846, "grad_norm": 24.929439544677734, "learning_rate": 5.766255910676542e-06, "loss": 11.954583168029785, "loss_alignment": 0.89501953125, "loss_alignment_w": 0.447509765625, "loss_sft": 0.2987361680716276, "loss_total_v6": 0.7471614480018616, "step": 3718 }, { "epoch": 0.8577153153153153, "grad_norm": 25.865144729614258, "learning_rate": 5.76426579610461e-06, "loss": 12.199084281921387, "loss_alignment": 0.9033203125, "loss_alignment_w": 0.45166015625, "loss_sft": 0.3101112190634012, "loss_total_v6": 0.7624427825212479, "step": 3719 }, { "epoch": 0.857945945945946, "grad_norm": 25.610157012939453, "learning_rate": 5.762275557551728e-06, "loss": 11.875762939453125, "loss_alignment": 0.91259765625, "loss_alignment_w": 0.456298828125, "loss_sft": 0.28637884743511677, "loss_total_v6": 0.7422351613640785, "step": 3720 }, { "epoch": 0.8581765765765765, "grad_norm": 25.55715560913086, "learning_rate": 5.760285195340758e-06, "loss": 11.36027717590332, "loss_alignment": 0.90185546875, "loss_alignment_w": 0.450927734375, "loss_sft": 0.25860122963786125, "loss_total_v6": 0.7100172787904739, "step": 3721 }, { "epoch": 0.8584072072072072, "grad_norm": 28.889984130859375, "learning_rate": 5.758294709794579e-06, "loss": 11.564516067504883, "loss_alignment": 0.90380859375, "loss_alignment_w": 0.451904296875, "loss_sft": 0.27161039412021637, "loss_total_v6": 0.7227822914719582, "step": 3722 }, { "epoch": 0.8586378378378379, "grad_norm": 25.837539672851562, "learning_rate": 5.7563041012360975e-06, "loss": 11.552495956420898, "loss_alignment": 0.8876953125, "loss_alignment_w": 0.44384765625, "loss_sft": 0.2778477407991886, "loss_total_v6": 0.7220310494303703, "step": 3723 }, { "epoch": 0.8588684684684684, "grad_norm": 25.188631057739258, "learning_rate": 5.754313369988227e-06, "loss": 12.001474380493164, "loss_alignment": 0.88427734375, "loss_alignment_w": 0.442138671875, "loss_sft": 0.30758727341890335, "loss_total_v6": 0.7500921711325645, "step": 3724 }, { "epoch": 0.8590990990990991, "grad_norm": 23.604679107666016, "learning_rate": 5.752322516373916e-06, "loss": 11.325080871582031, "loss_alignment": 0.88818359375, "loss_alignment_w": 0.444091796875, "loss_sft": 0.2622609194368124, "loss_total_v6": 0.7078175917267799, "step": 3725 }, { "epoch": 0.8593297297297298, "grad_norm": 25.7387638092041, "learning_rate": 5.750331540716121e-06, "loss": 11.665350914001465, "loss_alignment": 0.89599609375, "loss_alignment_w": 0.447998046875, "loss_sft": 0.28038444742560387, "loss_total_v6": 0.729084424674511, "step": 3726 }, { "epoch": 0.8595603603603603, "grad_norm": 23.21384620666504, "learning_rate": 5.748340443337821e-06, "loss": 11.206809997558594, "loss_alignment": 0.8798828125, "loss_alignment_w": 0.43994140625, "loss_sft": 0.2611098550260067, "loss_total_v6": 0.7004256397485733, "step": 3727 }, { "epoch": 0.859790990990991, "grad_norm": 25.067049026489258, "learning_rate": 5.746349224562021e-06, "loss": 11.286911964416504, "loss_alignment": 0.888671875, "loss_alignment_w": 0.4443359375, "loss_sft": 0.25937181152403355, "loss_total_v6": 0.7054320126771927, "step": 3728 }, { "epoch": 0.8600216216216217, "grad_norm": 24.824060440063477, "learning_rate": 5.744357884711739e-06, "loss": 11.039260864257812, "loss_alignment": 0.89453125, "loss_alignment_w": 0.447265625, "loss_sft": 0.24282551743090153, "loss_total_v6": 0.6899538412690163, "step": 3729 }, { "epoch": 0.8602522522522522, "grad_norm": 26.92272186279297, "learning_rate": 5.742366424110013e-06, "loss": 11.709648132324219, "loss_alignment": 0.90234375, "loss_alignment_w": 0.451171875, "loss_sft": 0.2789416406303644, "loss_total_v6": 0.7318530380725861, "step": 3730 }, { "epoch": 0.8604828828828829, "grad_norm": 34.40030288696289, "learning_rate": 5.740374843079905e-06, "loss": 11.477949142456055, "loss_alignment": 0.90625, "loss_alignment_w": 0.453125, "loss_sft": 0.26301081851124763, "loss_total_v6": 0.7173717767000198, "step": 3731 }, { "epoch": 0.8607135135135136, "grad_norm": 26.190418243408203, "learning_rate": 5.738383141944493e-06, "loss": 11.207574844360352, "loss_alignment": 0.88720703125, "loss_alignment_w": 0.443603515625, "loss_sft": 0.2562290336936712, "loss_total_v6": 0.7004734203219414, "step": 3732 }, { "epoch": 0.8609441441441441, "grad_norm": 27.72197151184082, "learning_rate": 5.7363913210268755e-06, "loss": 11.954647064208984, "loss_alignment": 0.91943359375, "loss_alignment_w": 0.459716796875, "loss_sft": 0.2870367057621479, "loss_total_v6": 0.7471654787659645, "step": 3733 }, { "epoch": 0.8611747747747748, "grad_norm": 25.796405792236328, "learning_rate": 5.734399380650168e-06, "loss": 11.619665145874023, "loss_alignment": 0.9052734375, "loss_alignment_w": 0.45263671875, "loss_sft": 0.27319567278027534, "loss_total_v6": 0.7262290716171265, "step": 3734 }, { "epoch": 0.8614054054054054, "grad_norm": 26.949413299560547, "learning_rate": 5.732407321137511e-06, "loss": 11.713024139404297, "loss_alignment": 0.91748046875, "loss_alignment_w": 0.458740234375, "loss_sft": 0.2740561906248331, "loss_total_v6": 0.7320640087127686, "step": 3735 }, { "epoch": 0.861636036036036, "grad_norm": 24.932897567749023, "learning_rate": 5.730415142812059e-06, "loss": 11.676790237426758, "loss_alignment": 0.8798828125, "loss_alignment_w": 0.43994140625, "loss_sft": 0.28886615112423897, "loss_total_v6": 0.7297993674874306, "step": 3736 }, { "epoch": 0.8618666666666667, "grad_norm": 25.467872619628906, "learning_rate": 5.728422845996987e-06, "loss": 11.51048469543457, "loss_alignment": 0.89208984375, "loss_alignment_w": 0.446044921875, "loss_sft": 0.27357398346066475, "loss_total_v6": 0.7194052785634995, "step": 3737 }, { "epoch": 0.8620972972972973, "grad_norm": 27.0095157623291, "learning_rate": 5.726430431015494e-06, "loss": 11.200918197631836, "loss_alignment": 0.880859375, "loss_alignment_w": 0.4404296875, "loss_sft": 0.2606805022805929, "loss_total_v6": 0.7000573724508286, "step": 3738 }, { "epoch": 0.8623279279279279, "grad_norm": 24.04359245300293, "learning_rate": 5.724437898190789e-06, "loss": 12.324975967407227, "loss_alignment": 0.927734375, "loss_alignment_w": 0.4638671875, "loss_sft": 0.3048873897641897, "loss_total_v6": 0.7703109681606293, "step": 3739 }, { "epoch": 0.8625585585585586, "grad_norm": 27.756607055664062, "learning_rate": 5.722445247846107e-06, "loss": 12.353151321411133, "loss_alignment": 0.9091796875, "loss_alignment_w": 0.45458984375, "loss_sft": 0.3175126202404499, "loss_total_v6": 0.7720719575881958, "step": 3740 }, { "epoch": 0.8627891891891892, "grad_norm": 24.60100746154785, "learning_rate": 5.7204524803047024e-06, "loss": 11.956803321838379, "loss_alignment": 0.89404296875, "loss_alignment_w": 0.447021484375, "loss_sft": 0.29975991509854794, "loss_total_v6": 0.7473002001643181, "step": 3741 }, { "epoch": 0.8630198198198198, "grad_norm": 23.17465591430664, "learning_rate": 5.718459595889843e-06, "loss": 11.661052703857422, "loss_alignment": 0.86962890625, "loss_alignment_w": 0.434814453125, "loss_sft": 0.2953135557472706, "loss_total_v6": 0.7288157790899277, "step": 3742 }, { "epoch": 0.8632504504504505, "grad_norm": 24.793642044067383, "learning_rate": 5.716466594924824e-06, "loss": 11.75483512878418, "loss_alignment": 0.89892578125, "loss_alignment_w": 0.449462890625, "loss_sft": 0.2850311566144228, "loss_total_v6": 0.73467718064785, "step": 3743 }, { "epoch": 0.8634810810810811, "grad_norm": 25.522441864013672, "learning_rate": 5.714473477732948e-06, "loss": 11.373394966125488, "loss_alignment": 0.87353515625, "loss_alignment_w": 0.436767578125, "loss_sft": 0.27326090447604656, "loss_total_v6": 0.7108371928334236, "step": 3744 }, { "epoch": 0.8637117117117117, "grad_norm": 26.100845336914062, "learning_rate": 5.71248024463755e-06, "loss": 11.322616577148438, "loss_alignment": 0.892578125, "loss_alignment_w": 0.4462890625, "loss_sft": 0.2602911461144686, "loss_total_v6": 0.7076635509729385, "step": 3745 }, { "epoch": 0.8639423423423424, "grad_norm": 25.520158767700195, "learning_rate": 5.7104868959619705e-06, "loss": 11.503864288330078, "loss_alignment": 0.9189453125, "loss_alignment_w": 0.45947265625, "loss_sft": 0.25964089669287205, "loss_total_v6": 0.7189914956688881, "step": 3746 }, { "epoch": 0.864172972972973, "grad_norm": 26.16757583618164, "learning_rate": 5.70849343202958e-06, "loss": 11.711868286132812, "loss_alignment": 0.9423828125, "loss_alignment_w": 0.47119140625, "loss_sft": 0.2602814957499504, "loss_total_v6": 0.7319917231798172, "step": 3747 }, { "epoch": 0.8644036036036036, "grad_norm": 27.67926597595215, "learning_rate": 5.706499853163761e-06, "loss": 11.34929084777832, "loss_alignment": 0.9140625, "loss_alignment_w": 0.45703125, "loss_sft": 0.25269618444144726, "loss_total_v6": 0.7093307077884674, "step": 3748 }, { "epoch": 0.8646342342342342, "grad_norm": 27.252361297607422, "learning_rate": 5.704506159687914e-06, "loss": 12.038347244262695, "loss_alignment": 0.90234375, "loss_alignment_w": 0.451171875, "loss_sft": 0.30101120471954346, "loss_total_v6": 0.7523966878652573, "step": 3749 }, { "epoch": 0.8648648648648649, "grad_norm": 25.19200897216797, "learning_rate": 5.7025123519254644e-06, "loss": 11.910381317138672, "loss_alignment": 0.90380859375, "loss_alignment_w": 0.451904296875, "loss_sft": 0.29205205477774143, "loss_total_v6": 0.7443988472223282, "step": 3750 }, { "epoch": 0.8650954954954955, "grad_norm": 25.04826545715332, "learning_rate": 5.700518430199851e-06, "loss": 12.041400909423828, "loss_alignment": 0.9013671875, "loss_alignment_w": 0.45068359375, "loss_sft": 0.30310947448015213, "loss_total_v6": 0.7525875866413116, "step": 3751 }, { "epoch": 0.8653261261261261, "grad_norm": 27.1201229095459, "learning_rate": 5.698524394834531e-06, "loss": 12.453658103942871, "loss_alignment": 0.9189453125, "loss_alignment_w": 0.45947265625, "loss_sft": 0.31878944113850594, "loss_total_v6": 0.7783536538481712, "step": 3752 }, { "epoch": 0.8655567567567568, "grad_norm": 24.475454330444336, "learning_rate": 5.696530246152983e-06, "loss": 11.415778160095215, "loss_alignment": 0.88525390625, "loss_alignment_w": 0.442626953125, "loss_sft": 0.27114905416965485, "loss_total_v6": 0.7134861126542091, "step": 3753 }, { "epoch": 0.8657873873873874, "grad_norm": 25.137056350708008, "learning_rate": 5.694535984478702e-06, "loss": 12.28413200378418, "loss_alignment": 0.91943359375, "loss_alignment_w": 0.459716796875, "loss_sft": 0.3078888673335314, "loss_total_v6": 0.7677582427859306, "step": 3754 }, { "epoch": 0.866018018018018, "grad_norm": 26.29396629333496, "learning_rate": 5.6925416101352e-06, "loss": 11.515016555786133, "loss_alignment": 0.8837890625, "loss_alignment_w": 0.44189453125, "loss_sft": 0.2774888128042221, "loss_total_v6": 0.7196885198354721, "step": 3755 }, { "epoch": 0.8662486486486487, "grad_norm": 25.24698829650879, "learning_rate": 5.690547123446011e-06, "loss": 11.990961074829102, "loss_alignment": 0.89599609375, "loss_alignment_w": 0.447998046875, "loss_sft": 0.30131492018699646, "loss_total_v6": 0.7494350448250771, "step": 3756 }, { "epoch": 0.8664792792792793, "grad_norm": 24.33738136291504, "learning_rate": 5.688552524734686e-06, "loss": 11.924473762512207, "loss_alignment": 0.87060546875, "loss_alignment_w": 0.435302734375, "loss_sft": 0.30953436344861984, "loss_total_v6": 0.7452796176075935, "step": 3757 }, { "epoch": 0.8667099099099099, "grad_norm": 24.16144371032715, "learning_rate": 5.686557814324791e-06, "loss": 11.564085006713867, "loss_alignment": 0.86865234375, "loss_alignment_w": 0.434326171875, "loss_sft": 0.28856646083295345, "loss_total_v6": 0.7227553054690361, "step": 3758 }, { "epoch": 0.8669405405405406, "grad_norm": 33.72822952270508, "learning_rate": 5.684562992539913e-06, "loss": 12.143898010253906, "loss_alignment": 0.93212890625, "loss_alignment_w": 0.466064453125, "loss_sft": 0.29285289719700813, "loss_total_v6": 0.7589936479926109, "step": 3759 }, { "epoch": 0.8671711711711712, "grad_norm": 27.58354949951172, "learning_rate": 5.682568059703659e-06, "loss": 11.630134582519531, "loss_alignment": 0.90625, "loss_alignment_w": 0.453125, "loss_sft": 0.2740941345691681, "loss_total_v6": 0.7268834188580513, "step": 3760 }, { "epoch": 0.8674018018018018, "grad_norm": 26.00593376159668, "learning_rate": 5.680573016139648e-06, "loss": 11.363363265991211, "loss_alignment": 0.8740234375, "loss_alignment_w": 0.43701171875, "loss_sft": 0.27341216057538986, "loss_total_v6": 0.7102102562785149, "step": 3761 }, { "epoch": 0.8676324324324325, "grad_norm": 25.63845443725586, "learning_rate": 5.678577862171523e-06, "loss": 11.86703109741211, "loss_alignment": 0.90087890625, "loss_alignment_w": 0.450439453125, "loss_sft": 0.2924707029014826, "loss_total_v6": 0.7416894733905792, "step": 3762 }, { "epoch": 0.867863063063063, "grad_norm": 26.166255950927734, "learning_rate": 5.6765825981229415e-06, "loss": 11.848932266235352, "loss_alignment": 0.9013671875, "loss_alignment_w": 0.45068359375, "loss_sft": 0.2905308548361063, "loss_total_v6": 0.7405583038926125, "step": 3763 }, { "epoch": 0.8680936936936937, "grad_norm": 31.2166805267334, "learning_rate": 5.674587224317579e-06, "loss": 11.884599685668945, "loss_alignment": 0.9033203125, "loss_alignment_w": 0.45166015625, "loss_sft": 0.2906543090939522, "loss_total_v6": 0.7427874654531479, "step": 3764 }, { "epoch": 0.8683243243243243, "grad_norm": 24.824230194091797, "learning_rate": 5.672591741079132e-06, "loss": 11.844532012939453, "loss_alignment": 0.87939453125, "loss_alignment_w": 0.439697265625, "loss_sft": 0.30063177831470966, "loss_total_v6": 0.7402832955121994, "step": 3765 }, { "epoch": 0.8685549549549549, "grad_norm": 25.79355239868164, "learning_rate": 5.67059614873131e-06, "loss": 11.69416618347168, "loss_alignment": 0.88671875, "loss_alignment_w": 0.443359375, "loss_sft": 0.2866562157869339, "loss_total_v6": 0.730885349214077, "step": 3766 }, { "epoch": 0.8687855855855856, "grad_norm": 24.21979522705078, "learning_rate": 5.668600447597844e-06, "loss": 10.927270889282227, "loss_alignment": 0.89013671875, "loss_alignment_w": 0.445068359375, "loss_sft": 0.23655858635902405, "loss_total_v6": 0.682954452931881, "step": 3767 }, { "epoch": 0.8690162162162162, "grad_norm": 23.189790725708008, "learning_rate": 5.6666046380024795e-06, "loss": 11.521833419799805, "loss_alignment": 0.876953125, "loss_alignment_w": 0.4384765625, "loss_sft": 0.28325545974075794, "loss_total_v6": 0.7201145887374878, "step": 3768 }, { "epoch": 0.8692468468468468, "grad_norm": 23.09012794494629, "learning_rate": 5.664608720268984e-06, "loss": 11.2642240524292, "loss_alignment": 0.88525390625, "loss_alignment_w": 0.442626953125, "loss_sft": 0.26150910183787346, "loss_total_v6": 0.7040139883756638, "step": 3769 }, { "epoch": 0.8694774774774775, "grad_norm": 27.4941463470459, "learning_rate": 5.662612694721139e-06, "loss": 11.597447395324707, "loss_alignment": 0.87451171875, "loss_alignment_w": 0.437255859375, "loss_sft": 0.28759985603392124, "loss_total_v6": 0.7248404398560524, "step": 3770 }, { "epoch": 0.8697081081081081, "grad_norm": 27.14716339111328, "learning_rate": 5.6606165616827415e-06, "loss": 11.582319259643555, "loss_alignment": 0.8994140625, "loss_alignment_w": 0.44970703125, "loss_sft": 0.27502717077732086, "loss_total_v6": 0.7238949909806252, "step": 3771 }, { "epoch": 0.8699387387387387, "grad_norm": 24.246633529663086, "learning_rate": 5.658620321477613e-06, "loss": 11.62808609008789, "loss_alignment": 0.89697265625, "loss_alignment_w": 0.448486328125, "loss_sft": 0.2801001351326704, "loss_total_v6": 0.726755402982235, "step": 3772 }, { "epoch": 0.8701693693693694, "grad_norm": 25.385913848876953, "learning_rate": 5.656623974429584e-06, "loss": 11.644170761108398, "loss_alignment": 0.90576171875, "loss_alignment_w": 0.452880859375, "loss_sft": 0.2743915505707264, "loss_total_v6": 0.7277606949210167, "step": 3773 }, { "epoch": 0.8704, "grad_norm": 25.30933952331543, "learning_rate": 5.65462752086251e-06, "loss": 11.651124954223633, "loss_alignment": 0.88720703125, "loss_alignment_w": 0.443603515625, "loss_sft": 0.28430188819766045, "loss_total_v6": 0.7281953319907188, "step": 3774 }, { "epoch": 0.8706306306306306, "grad_norm": 24.8275203704834, "learning_rate": 5.65263096110026e-06, "loss": 11.444896697998047, "loss_alignment": 0.8857421875, "loss_alignment_w": 0.44287109375, "loss_sft": 0.2718245666474104, "loss_total_v6": 0.715306006371975, "step": 3775 }, { "epoch": 0.8708612612612613, "grad_norm": 25.029006958007812, "learning_rate": 5.650634295466717e-06, "loss": 11.739051818847656, "loss_alignment": 0.88330078125, "loss_alignment_w": 0.441650390625, "loss_sft": 0.2919335290789604, "loss_total_v6": 0.7336907163262367, "step": 3776 }, { "epoch": 0.8710918918918918, "grad_norm": 24.212688446044922, "learning_rate": 5.648637524285789e-06, "loss": 11.593734741210938, "loss_alignment": 0.87841796875, "loss_alignment_w": 0.439208984375, "loss_sft": 0.28619289584457874, "loss_total_v6": 0.724608413875103, "step": 3777 }, { "epoch": 0.8713225225225225, "grad_norm": 25.996179580688477, "learning_rate": 5.646640647881393e-06, "loss": 11.9587984085083, "loss_alignment": 0.89990234375, "loss_alignment_w": 0.449951171875, "loss_sft": 0.2987097054719925, "loss_total_v6": 0.7474249079823494, "step": 3778 }, { "epoch": 0.8715531531531532, "grad_norm": 25.077762603759766, "learning_rate": 5.6446436665774706e-06, "loss": 11.906208038330078, "loss_alignment": 0.9189453125, "loss_alignment_w": 0.45947265625, "loss_sft": 0.28394812531769276, "loss_total_v6": 0.7441379502415657, "step": 3779 }, { "epoch": 0.8717837837837837, "grad_norm": 25.712215423583984, "learning_rate": 5.642646580697974e-06, "loss": 11.366340637207031, "loss_alignment": 0.875, "loss_alignment_w": 0.4375, "loss_sft": 0.27213332429528236, "loss_total_v6": 0.7103962600231171, "step": 3780 }, { "epoch": 0.8720144144144144, "grad_norm": 24.882354736328125, "learning_rate": 5.640649390566876e-06, "loss": 11.39299488067627, "loss_alignment": 0.8662109375, "loss_alignment_w": 0.43310546875, "loss_sft": 0.27875835448503494, "loss_total_v6": 0.7120621800422668, "step": 3781 }, { "epoch": 0.8722450450450451, "grad_norm": 26.427194595336914, "learning_rate": 5.638652096508166e-06, "loss": 12.272161483764648, "loss_alignment": 0.9130859375, "loss_alignment_w": 0.45654296875, "loss_sft": 0.30999408662319183, "loss_total_v6": 0.7670101001858711, "step": 3782 }, { "epoch": 0.8724756756756756, "grad_norm": 25.461692810058594, "learning_rate": 5.636654698845847e-06, "loss": 11.682648658752441, "loss_alignment": 0.87890625, "loss_alignment_w": 0.439453125, "loss_sft": 0.29045298881828785, "loss_total_v6": 0.7301655262708664, "step": 3783 }, { "epoch": 0.8727063063063063, "grad_norm": 22.577268600463867, "learning_rate": 5.634657197903944e-06, "loss": 11.281776428222656, "loss_alignment": 0.869140625, "loss_alignment_w": 0.4345703125, "loss_sft": 0.2699303422123194, "loss_total_v6": 0.7051110118627548, "step": 3784 }, { "epoch": 0.872936936936937, "grad_norm": 26.286733627319336, "learning_rate": 5.632659594006498e-06, "loss": 11.711969375610352, "loss_alignment": 0.90283203125, "loss_alignment_w": 0.451416015625, "loss_sft": 0.2805515509098768, "loss_total_v6": 0.7319981157779694, "step": 3785 }, { "epoch": 0.8731675675675675, "grad_norm": 23.627361297607422, "learning_rate": 5.63066188747756e-06, "loss": 11.397981643676758, "loss_alignment": 0.89697265625, "loss_alignment_w": 0.448486328125, "loss_sft": 0.2627888713032007, "loss_total_v6": 0.7123738378286362, "step": 3786 }, { "epoch": 0.8733981981981982, "grad_norm": 56.863739013671875, "learning_rate": 5.628664078641205e-06, "loss": 11.901244163513184, "loss_alignment": 0.912109375, "loss_alignment_w": 0.4560546875, "loss_sft": 0.2867049928754568, "loss_total_v6": 0.7438277453184128, "step": 3787 }, { "epoch": 0.8736288288288289, "grad_norm": 26.40709686279297, "learning_rate": 5.626666167821522e-06, "loss": 11.346617698669434, "loss_alignment": 0.86962890625, "loss_alignment_w": 0.434814453125, "loss_sft": 0.2737998627126217, "loss_total_v6": 0.7091636285185814, "step": 3788 }, { "epoch": 0.8738594594594594, "grad_norm": 27.10664939880371, "learning_rate": 5.6246681553426175e-06, "loss": 11.157354354858398, "loss_alignment": 0.8798828125, "loss_alignment_w": 0.43994140625, "loss_sft": 0.2571643888950348, "loss_total_v6": 0.6973346844315529, "step": 3789 }, { "epoch": 0.8740900900900901, "grad_norm": 26.2903995513916, "learning_rate": 5.622670041528612e-06, "loss": 11.736379623413086, "loss_alignment": 0.9130859375, "loss_alignment_w": 0.45654296875, "loss_sft": 0.2760957442224026, "loss_total_v6": 0.7335237264633179, "step": 3790 }, { "epoch": 0.8743207207207208, "grad_norm": 24.081735610961914, "learning_rate": 5.620671826703645e-06, "loss": 11.460418701171875, "loss_alignment": 0.88916015625, "loss_alignment_w": 0.444580078125, "loss_sft": 0.2714824192225933, "loss_total_v6": 0.7162761315703392, "step": 3791 }, { "epoch": 0.8745513513513513, "grad_norm": 26.521881103515625, "learning_rate": 5.6186735111918735e-06, "loss": 11.363618850708008, "loss_alignment": 0.88671875, "loss_alignment_w": 0.443359375, "loss_sft": 0.26682098954916, "loss_total_v6": 0.7102261334657669, "step": 3792 }, { "epoch": 0.874781981981982, "grad_norm": 23.97198486328125, "learning_rate": 5.616675095317463e-06, "loss": 11.998188018798828, "loss_alignment": 0.90185546875, "loss_alignment_w": 0.450927734375, "loss_sft": 0.2989590559154749, "loss_total_v6": 0.749886766076088, "step": 3793 }, { "epoch": 0.8750126126126127, "grad_norm": 28.459156036376953, "learning_rate": 5.614676579404608e-06, "loss": 11.9420166015625, "loss_alignment": 0.90185546875, "loss_alignment_w": 0.450927734375, "loss_sft": 0.29709623754024506, "loss_total_v6": 0.7463760152459145, "step": 3794 }, { "epoch": 0.8752432432432432, "grad_norm": 25.420602798461914, "learning_rate": 5.612677963777508e-06, "loss": 12.101511001586914, "loss_alignment": 0.89208984375, "loss_alignment_w": 0.446044921875, "loss_sft": 0.31040636636316776, "loss_total_v6": 0.7563444748520851, "step": 3795 }, { "epoch": 0.8754738738738739, "grad_norm": 24.97661018371582, "learning_rate": 5.6106792487603845e-06, "loss": 11.592117309570312, "loss_alignment": 0.87109375, "loss_alignment_w": 0.435546875, "loss_sft": 0.2882890831679106, "loss_total_v6": 0.7245073616504669, "step": 3796 }, { "epoch": 0.8757045045045045, "grad_norm": 23.194011688232422, "learning_rate": 5.608680434677474e-06, "loss": 11.34298038482666, "loss_alignment": 0.8720703125, "loss_alignment_w": 0.43603515625, "loss_sft": 0.272046634927392, "loss_total_v6": 0.7089362740516663, "step": 3797 }, { "epoch": 0.8759351351351351, "grad_norm": 26.132034301757812, "learning_rate": 5.606681521853027e-06, "loss": 11.315723419189453, "loss_alignment": 0.9013671875, "loss_alignment_w": 0.45068359375, "loss_sft": 0.2569764256477356, "loss_total_v6": 0.7072327435016632, "step": 3798 }, { "epoch": 0.8761657657657658, "grad_norm": 28.93242835998535, "learning_rate": 5.604682510611312e-06, "loss": 11.60683822631836, "loss_alignment": 0.90966796875, "loss_alignment_w": 0.454833984375, "loss_sft": 0.27161572501063347, "loss_total_v6": 0.7254273816943169, "step": 3799 }, { "epoch": 0.8763963963963964, "grad_norm": 24.021554946899414, "learning_rate": 5.6026834012766155e-06, "loss": 11.582855224609375, "loss_alignment": 0.8876953125, "loss_alignment_w": 0.44384765625, "loss_sft": 0.27966883033514023, "loss_total_v6": 0.7239284738898277, "step": 3800 }, { "epoch": 0.876627027027027, "grad_norm": 26.5701847076416, "learning_rate": 5.600684194173236e-06, "loss": 11.926645278930664, "loss_alignment": 0.89208984375, "loss_alignment_w": 0.446044921875, "loss_sft": 0.2996144723147154, "loss_total_v6": 0.7454153001308441, "step": 3801 }, { "epoch": 0.8768576576576577, "grad_norm": 24.156850814819336, "learning_rate": 5.59868488962549e-06, "loss": 11.7186279296875, "loss_alignment": 0.888671875, "loss_alignment_w": 0.4443359375, "loss_sft": 0.28804774209856987, "loss_total_v6": 0.732414223253727, "step": 3802 }, { "epoch": 0.8770882882882883, "grad_norm": 25.31525993347168, "learning_rate": 5.596685487957707e-06, "loss": 11.248089790344238, "loss_alignment": 0.8681640625, "loss_alignment_w": 0.43408203125, "loss_sft": 0.26915245689451694, "loss_total_v6": 0.7030056044459343, "step": 3803 }, { "epoch": 0.8773189189189189, "grad_norm": 27.481678009033203, "learning_rate": 5.5946859894942385e-06, "loss": 11.90727424621582, "loss_alignment": 0.88525390625, "loss_alignment_w": 0.442626953125, "loss_sft": 0.30260002613067627, "loss_total_v6": 0.74420465528965, "step": 3804 }, { "epoch": 0.8775495495495496, "grad_norm": 25.510616302490234, "learning_rate": 5.5926863945594435e-06, "loss": 11.935547828674316, "loss_alignment": 0.91552734375, "loss_alignment_w": 0.457763671875, "loss_sft": 0.28787238523364067, "loss_total_v6": 0.7459717467427254, "step": 3805 }, { "epoch": 0.8777801801801802, "grad_norm": 23.52370834350586, "learning_rate": 5.590686703477705e-06, "loss": 11.186285018920898, "loss_alignment": 0.880859375, "loss_alignment_w": 0.4404296875, "loss_sft": 0.25880471989512444, "loss_total_v6": 0.6991428583860397, "step": 3806 }, { "epoch": 0.8780108108108108, "grad_norm": 26.992706298828125, "learning_rate": 5.588686916573413e-06, "loss": 12.070819854736328, "loss_alignment": 0.89306640625, "loss_alignment_w": 0.446533203125, "loss_sft": 0.3082440812140703, "loss_total_v6": 0.7544263005256653, "step": 3807 }, { "epoch": 0.8782414414414415, "grad_norm": 28.41925621032715, "learning_rate": 5.586687034170981e-06, "loss": 11.46146011352539, "loss_alignment": 0.8798828125, "loss_alignment_w": 0.43994140625, "loss_sft": 0.2763083204627037, "loss_total_v6": 0.7163412794470787, "step": 3808 }, { "epoch": 0.8784720720720721, "grad_norm": 26.83738136291504, "learning_rate": 5.5846870565948345e-06, "loss": 11.825767517089844, "loss_alignment": 0.89013671875, "loss_alignment_w": 0.445068359375, "loss_sft": 0.29355382919311523, "loss_total_v6": 0.7391104623675346, "step": 3809 }, { "epoch": 0.8787027027027027, "grad_norm": 23.936582565307617, "learning_rate": 5.582686984169412e-06, "loss": 11.769460678100586, "loss_alignment": 0.888671875, "loss_alignment_w": 0.4443359375, "loss_sft": 0.2929184976965189, "loss_total_v6": 0.7355912253260612, "step": 3810 }, { "epoch": 0.8789333333333333, "grad_norm": 25.248836517333984, "learning_rate": 5.580686817219171e-06, "loss": 10.841802597045898, "loss_alignment": 0.89697265625, "loss_alignment_w": 0.448486328125, "loss_sft": 0.22981300577521324, "loss_total_v6": 0.6776126995682716, "step": 3811 }, { "epoch": 0.879163963963964, "grad_norm": 25.409650802612305, "learning_rate": 5.5786865560685855e-06, "loss": 11.283350944519043, "loss_alignment": 0.89306640625, "loss_alignment_w": 0.446533203125, "loss_sft": 0.25783700309693813, "loss_total_v6": 0.7052094638347626, "step": 3812 }, { "epoch": 0.8793945945945946, "grad_norm": 29.21004867553711, "learning_rate": 5.576686201042139e-06, "loss": 12.3325777053833, "loss_alignment": 0.90771484375, "loss_alignment_w": 0.453857421875, "loss_sft": 0.3156927227973938, "loss_total_v6": 0.7707861214876175, "step": 3813 }, { "epoch": 0.8796252252252252, "grad_norm": 24.837404251098633, "learning_rate": 5.5746857524643335e-06, "loss": 11.945077896118164, "loss_alignment": 0.9091796875, "loss_alignment_w": 0.45458984375, "loss_sft": 0.2922217193990946, "loss_total_v6": 0.7465673983097076, "step": 3814 }, { "epoch": 0.8798558558558559, "grad_norm": 24.94133758544922, "learning_rate": 5.572685210659688e-06, "loss": 11.79883861541748, "loss_alignment": 0.88623046875, "loss_alignment_w": 0.443115234375, "loss_sft": 0.2934119217097759, "loss_total_v6": 0.7374274134635925, "step": 3815 }, { "epoch": 0.8800864864864865, "grad_norm": 27.636472702026367, "learning_rate": 5.570684575952737e-06, "loss": 11.795392990112305, "loss_alignment": 0.89453125, "loss_alignment_w": 0.447265625, "loss_sft": 0.2899921350181103, "loss_total_v6": 0.7372120320796967, "step": 3816 }, { "epoch": 0.8803171171171171, "grad_norm": 22.248451232910156, "learning_rate": 5.568683848668025e-06, "loss": 11.651739120483398, "loss_alignment": 0.88720703125, "loss_alignment_w": 0.443603515625, "loss_sft": 0.28490484692156315, "loss_total_v6": 0.7282337099313736, "step": 3817 }, { "epoch": 0.8805477477477478, "grad_norm": 26.616357803344727, "learning_rate": 5.566683029130116e-06, "loss": 11.638479232788086, "loss_alignment": 0.93115234375, "loss_alignment_w": 0.465576171875, "loss_sft": 0.2626680415123701, "loss_total_v6": 0.7274049893021584, "step": 3818 }, { "epoch": 0.8807783783783784, "grad_norm": 25.28487777709961, "learning_rate": 5.564682117663586e-06, "loss": 12.151938438415527, "loss_alignment": 0.896484375, "loss_alignment_w": 0.4482421875, "loss_sft": 0.31245942786335945, "loss_total_v6": 0.7594961673021317, "step": 3819 }, { "epoch": 0.881009009009009, "grad_norm": 24.733795166015625, "learning_rate": 5.562681114593028e-06, "loss": 11.555211067199707, "loss_alignment": 0.89208984375, "loss_alignment_w": 0.446044921875, "loss_sft": 0.27569798938930035, "loss_total_v6": 0.7222006916999817, "step": 3820 }, { "epoch": 0.8812396396396397, "grad_norm": 24.28887367248535, "learning_rate": 5.56068002024305e-06, "loss": 11.81309700012207, "loss_alignment": 0.87255859375, "loss_alignment_w": 0.436279296875, "loss_sft": 0.302634347230196, "loss_total_v6": 0.7383185997605324, "step": 3821 }, { "epoch": 0.8814702702702703, "grad_norm": 25.616985321044922, "learning_rate": 5.558678834938274e-06, "loss": 11.880668640136719, "loss_alignment": 0.908203125, "loss_alignment_w": 0.4541015625, "loss_sft": 0.2889742814004421, "loss_total_v6": 0.7425417825579643, "step": 3822 }, { "epoch": 0.8817009009009009, "grad_norm": 25.262500762939453, "learning_rate": 5.556677559003338e-06, "loss": 11.568258285522461, "loss_alignment": 0.89990234375, "loss_alignment_w": 0.449951171875, "loss_sft": 0.2715238332748413, "loss_total_v6": 0.7230161428451538, "step": 3823 }, { "epoch": 0.8819315315315315, "grad_norm": 23.45688819885254, "learning_rate": 5.554676192762891e-06, "loss": 11.2659330368042, "loss_alignment": 0.90771484375, "loss_alignment_w": 0.453857421875, "loss_sft": 0.2498971614986658, "loss_total_v6": 0.7041208148002625, "step": 3824 }, { "epoch": 0.8821621621621621, "grad_norm": 27.530824661254883, "learning_rate": 5.5526747365416e-06, "loss": 11.317002296447754, "loss_alignment": 0.8916015625, "loss_alignment_w": 0.44580078125, "loss_sft": 0.2602911405265331, "loss_total_v6": 0.7073126658797264, "step": 3825 }, { "epoch": 0.8823927927927928, "grad_norm": 25.445283889770508, "learning_rate": 5.550673190664148e-06, "loss": 11.434298515319824, "loss_alignment": 0.8974609375, "loss_alignment_w": 0.44873046875, "loss_sft": 0.2651808112859726, "loss_total_v6": 0.7146436795592308, "step": 3826 }, { "epoch": 0.8826234234234234, "grad_norm": 24.2083683013916, "learning_rate": 5.548671555455226e-06, "loss": 11.814724922180176, "loss_alignment": 0.86474609375, "loss_alignment_w": 0.432373046875, "loss_sft": 0.3059557154774666, "loss_total_v6": 0.7384203225374222, "step": 3827 }, { "epoch": 0.882854054054054, "grad_norm": 26.197513580322266, "learning_rate": 5.5466698312395495e-06, "loss": 11.616228103637695, "loss_alignment": 0.92724609375, "loss_alignment_w": 0.463623046875, "loss_sft": 0.2624217700213194, "loss_total_v6": 0.7260142862796783, "step": 3828 }, { "epoch": 0.8830846846846847, "grad_norm": 25.070566177368164, "learning_rate": 5.544668018341839e-06, "loss": 11.437792778015137, "loss_alignment": 0.86083984375, "loss_alignment_w": 0.430419921875, "loss_sft": 0.2842589970678091, "loss_total_v6": 0.7148620262742043, "step": 3829 }, { "epoch": 0.8833153153153153, "grad_norm": 25.303232192993164, "learning_rate": 5.542666117086832e-06, "loss": 12.067667007446289, "loss_alignment": 0.888671875, "loss_alignment_w": 0.4443359375, "loss_sft": 0.30969490855932236, "loss_total_v6": 0.7542291954159737, "step": 3830 }, { "epoch": 0.8835459459459459, "grad_norm": 25.801616668701172, "learning_rate": 5.540664127799286e-06, "loss": 11.601314544677734, "loss_alignment": 0.9013671875, "loss_alignment_w": 0.45068359375, "loss_sft": 0.2740018181502819, "loss_total_v6": 0.7250821217894554, "step": 3831 }, { "epoch": 0.8837765765765766, "grad_norm": 28.491914749145508, "learning_rate": 5.538662050803965e-06, "loss": 12.200470924377441, "loss_alignment": 0.91455078125, "loss_alignment_w": 0.457275390625, "loss_sft": 0.3055744804441929, "loss_total_v6": 0.7625294476747513, "step": 3832 }, { "epoch": 0.8840072072072072, "grad_norm": 23.753862380981445, "learning_rate": 5.5366598864256496e-06, "loss": 11.054893493652344, "loss_alignment": 0.88037109375, "loss_alignment_w": 0.440185546875, "loss_sft": 0.2507453225553036, "loss_total_v6": 0.6909308731555939, "step": 3833 }, { "epoch": 0.8842378378378378, "grad_norm": 27.158607482910156, "learning_rate": 5.5346576349891405e-06, "loss": 11.763113021850586, "loss_alignment": 0.88720703125, "loss_alignment_w": 0.443603515625, "loss_sft": 0.2916215732693672, "loss_total_v6": 0.735194593667984, "step": 3834 }, { "epoch": 0.8844684684684685, "grad_norm": 27.178863525390625, "learning_rate": 5.532655296819242e-06, "loss": 11.332683563232422, "loss_alignment": 0.9033203125, "loss_alignment_w": 0.45166015625, "loss_sft": 0.2557933088392019, "loss_total_v6": 0.7082927078008652, "step": 3835 }, { "epoch": 0.884699099099099, "grad_norm": 25.824033737182617, "learning_rate": 5.53065287224078e-06, "loss": 11.980018615722656, "loss_alignment": 0.86865234375, "loss_alignment_w": 0.434326171875, "loss_sft": 0.3138146325945854, "loss_total_v6": 0.7487511485815048, "step": 3836 }, { "epoch": 0.8849297297297297, "grad_norm": 26.783300399780273, "learning_rate": 5.528650361578593e-06, "loss": 11.350286483764648, "loss_alignment": 0.884765625, "loss_alignment_w": 0.4423828125, "loss_sft": 0.266094533726573, "loss_total_v6": 0.7093929052352905, "step": 3837 }, { "epoch": 0.8851603603603604, "grad_norm": 26.41507339477539, "learning_rate": 5.526647765157532e-06, "loss": 11.351632118225098, "loss_alignment": 0.90087890625, "loss_alignment_w": 0.450439453125, "loss_sft": 0.2602582275867462, "loss_total_v6": 0.7094769701361656, "step": 3838 }, { "epoch": 0.885390990990991, "grad_norm": 26.09130096435547, "learning_rate": 5.524645083302464e-06, "loss": 11.576202392578125, "loss_alignment": 0.896484375, "loss_alignment_w": 0.4482421875, "loss_sft": 0.2757740244269371, "loss_total_v6": 0.7235126793384552, "step": 3839 }, { "epoch": 0.8856216216216216, "grad_norm": 25.382366180419922, "learning_rate": 5.522642316338268e-06, "loss": 11.853619575500488, "loss_alignment": 0.896484375, "loss_alignment_w": 0.4482421875, "loss_sft": 0.292029183357954, "loss_total_v6": 0.7408512085676193, "step": 3840 }, { "epoch": 0.8858522522522523, "grad_norm": 30.92147445678711, "learning_rate": 5.5206394645898375e-06, "loss": 11.295707702636719, "loss_alignment": 0.89453125, "loss_alignment_w": 0.447265625, "loss_sft": 0.2596468813717365, "loss_total_v6": 0.7059817165136337, "step": 3841 }, { "epoch": 0.8860828828828828, "grad_norm": 23.654306411743164, "learning_rate": 5.518636528382078e-06, "loss": 11.569793701171875, "loss_alignment": 0.87841796875, "loss_alignment_w": 0.439208984375, "loss_sft": 0.2849254682660103, "loss_total_v6": 0.7231121212244034, "step": 3842 }, { "epoch": 0.8863135135135135, "grad_norm": 25.936140060424805, "learning_rate": 5.516633508039914e-06, "loss": 11.328155517578125, "loss_alignment": 0.87109375, "loss_alignment_w": 0.435546875, "loss_sft": 0.27348520793020725, "loss_total_v6": 0.7080097049474716, "step": 3843 }, { "epoch": 0.8865441441441442, "grad_norm": 25.217586517333984, "learning_rate": 5.514630403888278e-06, "loss": 12.325398445129395, "loss_alignment": 0.92431640625, "loss_alignment_w": 0.462158203125, "loss_sft": 0.3071873430162668, "loss_total_v6": 0.7703373730182648, "step": 3844 }, { "epoch": 0.8867747747747747, "grad_norm": 25.77528190612793, "learning_rate": 5.512627216252117e-06, "loss": 11.552943229675293, "loss_alignment": 0.90087890625, "loss_alignment_w": 0.450439453125, "loss_sft": 0.27155850268900394, "loss_total_v6": 0.7220589816570282, "step": 3845 }, { "epoch": 0.8870054054054054, "grad_norm": 28.41666603088379, "learning_rate": 5.510623945456393e-06, "loss": 11.003339767456055, "loss_alignment": 0.888671875, "loss_alignment_w": 0.4443359375, "loss_sft": 0.24274719133973122, "loss_total_v6": 0.6877087280154228, "step": 3846 }, { "epoch": 0.8872360360360361, "grad_norm": 25.605327606201172, "learning_rate": 5.508620591826084e-06, "loss": 12.162948608398438, "loss_alignment": 0.9150390625, "loss_alignment_w": 0.45751953125, "loss_sft": 0.30374809354543686, "loss_total_v6": 0.7601842656731606, "step": 3847 }, { "epoch": 0.8874666666666666, "grad_norm": 24.524141311645508, "learning_rate": 5.506617155686177e-06, "loss": 11.38586139678955, "loss_alignment": 0.88916015625, "loss_alignment_w": 0.444580078125, "loss_sft": 0.26508314348757267, "loss_total_v6": 0.7116163372993469, "step": 3848 }, { "epoch": 0.8876972972972973, "grad_norm": 26.120615005493164, "learning_rate": 5.504613637361671e-06, "loss": 11.701305389404297, "loss_alignment": 0.8955078125, "loss_alignment_w": 0.44775390625, "loss_sft": 0.2838218454271555, "loss_total_v6": 0.7313315868377686, "step": 3849 }, { "epoch": 0.887927927927928, "grad_norm": 27.302000045776367, "learning_rate": 5.502610037177586e-06, "loss": 11.415594100952148, "loss_alignment": 0.88037109375, "loss_alignment_w": 0.440185546875, "loss_sft": 0.2727549895644188, "loss_total_v6": 0.7134745866060257, "step": 3850 }, { "epoch": 0.8881585585585585, "grad_norm": 26.620534896850586, "learning_rate": 5.5006063554589485e-06, "loss": 11.511807441711426, "loss_alignment": 0.9033203125, "loss_alignment_w": 0.45166015625, "loss_sft": 0.26753788255155087, "loss_total_v6": 0.7194879427552223, "step": 3851 }, { "epoch": 0.8883891891891892, "grad_norm": 25.273162841796875, "learning_rate": 5.498602592530799e-06, "loss": 11.383808135986328, "loss_alignment": 0.90185546875, "loss_alignment_w": 0.450927734375, "loss_sft": 0.2598889209330082, "loss_total_v6": 0.7114880308508873, "step": 3852 }, { "epoch": 0.8886198198198199, "grad_norm": 26.94394302368164, "learning_rate": 5.496598748718196e-06, "loss": 11.813894271850586, "loss_alignment": 0.89208984375, "loss_alignment_w": 0.446044921875, "loss_sft": 0.29154529981315136, "loss_total_v6": 0.7383684143424034, "step": 3853 }, { "epoch": 0.8888504504504504, "grad_norm": 23.63068199157715, "learning_rate": 5.494594824346204e-06, "loss": 11.407264709472656, "loss_alignment": 0.8896484375, "loss_alignment_w": 0.44482421875, "loss_sft": 0.2683434318751097, "loss_total_v6": 0.7129540517926216, "step": 3854 }, { "epoch": 0.8890810810810811, "grad_norm": 24.567907333374023, "learning_rate": 5.4925908197399066e-06, "loss": 11.944025039672852, "loss_alignment": 0.89990234375, "loss_alignment_w": 0.449951171875, "loss_sft": 0.2984119579195976, "loss_total_v6": 0.746501550078392, "step": 3855 }, { "epoch": 0.8893117117117117, "grad_norm": 22.33935546875, "learning_rate": 5.490586735224398e-06, "loss": 11.538675308227539, "loss_alignment": 0.86865234375, "loss_alignment_w": 0.434326171875, "loss_sft": 0.28607807122170925, "loss_total_v6": 0.72116719186306, "step": 3856 }, { "epoch": 0.8895423423423423, "grad_norm": 25.54315948486328, "learning_rate": 5.488582571124783e-06, "loss": 11.521924018859863, "loss_alignment": 0.88916015625, "loss_alignment_w": 0.444580078125, "loss_sft": 0.27518922835588455, "loss_total_v6": 0.7201202362775803, "step": 3857 }, { "epoch": 0.889772972972973, "grad_norm": 28.24310302734375, "learning_rate": 5.486578327766184e-06, "loss": 11.936223983764648, "loss_alignment": 0.91748046875, "loss_alignment_w": 0.458740234375, "loss_sft": 0.28774676844477654, "loss_total_v6": 0.7460139691829681, "step": 3858 }, { "epoch": 0.8900036036036036, "grad_norm": 25.749631881713867, "learning_rate": 5.484574005473733e-06, "loss": 11.668437957763672, "loss_alignment": 0.8779296875, "loss_alignment_w": 0.43896484375, "loss_sft": 0.2913500405848026, "loss_total_v6": 0.7292773276567459, "step": 3859 }, { "epoch": 0.8902342342342342, "grad_norm": 25.88726043701172, "learning_rate": 5.482569604572577e-06, "loss": 11.89284896850586, "loss_alignment": 0.89208984375, "loss_alignment_w": 0.446044921875, "loss_sft": 0.2964494153857231, "loss_total_v6": 0.7433030754327774, "step": 3860 }, { "epoch": 0.8904648648648649, "grad_norm": 32.299903869628906, "learning_rate": 5.480565125387873e-06, "loss": 11.219121932983398, "loss_alignment": 0.88525390625, "loss_alignment_w": 0.442626953125, "loss_sft": 0.2570575699210167, "loss_total_v6": 0.7011951506137848, "step": 3861 }, { "epoch": 0.8906954954954955, "grad_norm": 25.933441162109375, "learning_rate": 5.4785605682447905e-06, "loss": 11.23941421508789, "loss_alignment": 0.89306640625, "loss_alignment_w": 0.446533203125, "loss_sft": 0.25531974993646145, "loss_total_v6": 0.7024633511900902, "step": 3862 }, { "epoch": 0.8909261261261261, "grad_norm": 24.11219596862793, "learning_rate": 5.476555933468519e-06, "loss": 11.096748352050781, "loss_alignment": 0.91650390625, "loss_alignment_w": 0.458251953125, "loss_sft": 0.23579833935946226, "loss_total_v6": 0.6935467645525932, "step": 3863 }, { "epoch": 0.8911567567567568, "grad_norm": 27.671894073486328, "learning_rate": 5.47455122138425e-06, "loss": 12.475077629089355, "loss_alignment": 0.9306640625, "loss_alignment_w": 0.46533203125, "loss_sft": 0.3142230175435543, "loss_total_v6": 0.7796923667192459, "step": 3864 }, { "epoch": 0.8913873873873874, "grad_norm": 26.109434127807617, "learning_rate": 5.472546432317194e-06, "loss": 11.24728775024414, "loss_alignment": 0.8642578125, "loss_alignment_w": 0.43212890625, "loss_sft": 0.2704450860619545, "loss_total_v6": 0.702955462038517, "step": 3865 }, { "epoch": 0.891618018018018, "grad_norm": 29.7059326171875, "learning_rate": 5.470541566592573e-06, "loss": 12.010589599609375, "loss_alignment": 0.87451171875, "loss_alignment_w": 0.437255859375, "loss_sft": 0.31268880516290665, "loss_total_v6": 0.7506618276238441, "step": 3866 }, { "epoch": 0.8918486486486487, "grad_norm": 39.83811950683594, "learning_rate": 5.46853662453562e-06, "loss": 11.887442588806152, "loss_alignment": 0.8994140625, "loss_alignment_w": 0.44970703125, "loss_sft": 0.2932886704802513, "loss_total_v6": 0.7429651767015457, "step": 3867 }, { "epoch": 0.8920792792792793, "grad_norm": 27.59174346923828, "learning_rate": 5.466531606471581e-06, "loss": 11.648612976074219, "loss_alignment": 0.9169921875, "loss_alignment_w": 0.45849609375, "loss_sft": 0.2688402831554413, "loss_total_v6": 0.7280383035540581, "step": 3868 }, { "epoch": 0.8923099099099099, "grad_norm": 26.02617645263672, "learning_rate": 5.464526512725716e-06, "loss": 11.384687423706055, "loss_alignment": 0.8701171875, "loss_alignment_w": 0.43505859375, "loss_sft": 0.2755688428878784, "loss_total_v6": 0.7115429639816284, "step": 3869 }, { "epoch": 0.8925405405405406, "grad_norm": 24.867969512939453, "learning_rate": 5.462521343623296e-06, "loss": 11.462589263916016, "loss_alignment": 0.8896484375, "loss_alignment_w": 0.44482421875, "loss_sft": 0.27235052920877934, "loss_total_v6": 0.7164118066430092, "step": 3870 }, { "epoch": 0.8927711711711712, "grad_norm": 26.279781341552734, "learning_rate": 5.460516099489601e-06, "loss": 11.167829513549805, "loss_alignment": 0.87939453125, "loss_alignment_w": 0.439697265625, "loss_sft": 0.25752912648022175, "loss_total_v6": 0.6979893445968628, "step": 3871 }, { "epoch": 0.8930018018018018, "grad_norm": 26.838871002197266, "learning_rate": 5.458510780649932e-06, "loss": 11.556690216064453, "loss_alignment": 0.8681640625, "loss_alignment_w": 0.43408203125, "loss_sft": 0.28831792064011097, "loss_total_v6": 0.7222931385040283, "step": 3872 }, { "epoch": 0.8932324324324324, "grad_norm": 26.2478084564209, "learning_rate": 5.456505387429592e-06, "loss": 11.804515838623047, "loss_alignment": 0.90673828125, "loss_alignment_w": 0.453369140625, "loss_sft": 0.28419950418174267, "loss_total_v6": 0.7377822622656822, "step": 3873 }, { "epoch": 0.8934630630630631, "grad_norm": 27.284910202026367, "learning_rate": 5.4544999201539e-06, "loss": 11.556587219238281, "loss_alignment": 0.865234375, "loss_alignment_w": 0.4326171875, "loss_sft": 0.2895321883261204, "loss_total_v6": 0.7222867161035538, "step": 3874 }, { "epoch": 0.8936936936936937, "grad_norm": 24.244537353515625, "learning_rate": 5.45249437914819e-06, "loss": 11.808601379394531, "loss_alignment": 0.93505859375, "loss_alignment_w": 0.467529296875, "loss_sft": 0.2700047381222248, "loss_total_v6": 0.7380375638604164, "step": 3875 }, { "epoch": 0.8939243243243243, "grad_norm": 32.8610725402832, "learning_rate": 5.450488764737805e-06, "loss": 11.307073593139648, "loss_alignment": 0.8818359375, "loss_alignment_w": 0.44091796875, "loss_sft": 0.266231931746006, "loss_total_v6": 0.7066921144723892, "step": 3876 }, { "epoch": 0.894154954954955, "grad_norm": 25.66650390625, "learning_rate": 5.448483077248099e-06, "loss": 11.716134071350098, "loss_alignment": 0.89208984375, "loss_alignment_w": 0.446044921875, "loss_sft": 0.2855268120765686, "loss_total_v6": 0.7322583943605423, "step": 3877 }, { "epoch": 0.8943855855855856, "grad_norm": 27.173961639404297, "learning_rate": 5.44647731700444e-06, "loss": 11.175567626953125, "loss_alignment": 0.8720703125, "loss_alignment_w": 0.43603515625, "loss_sft": 0.2626056484878063, "loss_total_v6": 0.6984729617834091, "step": 3878 }, { "epoch": 0.8946162162162162, "grad_norm": 24.63669204711914, "learning_rate": 5.4444714843322085e-06, "loss": 11.045469284057617, "loss_alignment": 0.8916015625, "loss_alignment_w": 0.44580078125, "loss_sft": 0.24422063492238522, "loss_total_v6": 0.6903418526053429, "step": 3879 }, { "epoch": 0.8948468468468469, "grad_norm": 27.240428924560547, "learning_rate": 5.442465579556793e-06, "loss": 11.962203025817871, "loss_alignment": 0.89208984375, "loss_alignment_w": 0.446044921875, "loss_sft": 0.30069249123334885, "loss_total_v6": 0.7476376816630363, "step": 3880 }, { "epoch": 0.8950774774774775, "grad_norm": 28.0047664642334, "learning_rate": 5.4404596030035964e-06, "loss": 11.4013090133667, "loss_alignment": 0.8974609375, "loss_alignment_w": 0.44873046875, "loss_sft": 0.2632562890648842, "loss_total_v6": 0.7125818431377411, "step": 3881 }, { "epoch": 0.8953081081081081, "grad_norm": 25.521347045898438, "learning_rate": 5.438453554998036e-06, "loss": 11.72401237487793, "loss_alignment": 0.87353515625, "loss_alignment_w": 0.436767578125, "loss_sft": 0.29523555003106594, "loss_total_v6": 0.7327507883310318, "step": 3882 }, { "epoch": 0.8955387387387388, "grad_norm": 26.804590225219727, "learning_rate": 5.436447435865532e-06, "loss": 12.021331787109375, "loss_alignment": 0.888671875, "loss_alignment_w": 0.4443359375, "loss_sft": 0.3078517448157072, "loss_total_v6": 0.751333199441433, "step": 3883 }, { "epoch": 0.8957693693693694, "grad_norm": 77.94468688964844, "learning_rate": 5.434441245931524e-06, "loss": 11.858772277832031, "loss_alignment": 0.896484375, "loss_alignment_w": 0.4482421875, "loss_sft": 0.29248851351439953, "loss_total_v6": 0.7411732524633408, "step": 3884 }, { "epoch": 0.896, "grad_norm": 26.67194175720215, "learning_rate": 5.432434985521463e-06, "loss": 11.53532600402832, "loss_alignment": 0.91162109375, "loss_alignment_w": 0.455810546875, "loss_sft": 0.264063922688365, "loss_total_v6": 0.720957837998867, "step": 3885 }, { "epoch": 0.8962306306306306, "grad_norm": 25.199459075927734, "learning_rate": 5.430428654960806e-06, "loss": 11.65234375, "loss_alignment": 0.88330078125, "loss_alignment_w": 0.441650390625, "loss_sft": 0.2871093489229679, "loss_total_v6": 0.7282714545726776, "step": 3886 }, { "epoch": 0.8964612612612612, "grad_norm": 26.37202262878418, "learning_rate": 5.428422254575028e-06, "loss": 12.50619125366211, "loss_alignment": 0.91259765625, "loss_alignment_w": 0.456298828125, "loss_sft": 0.32585686817765236, "loss_total_v6": 0.7816369161009789, "step": 3887 }, { "epoch": 0.8966918918918919, "grad_norm": 29.301061630249023, "learning_rate": 5.426415784689608e-06, "loss": 11.896814346313477, "loss_alignment": 0.8857421875, "loss_alignment_w": 0.44287109375, "loss_sft": 0.3019309900701046, "loss_total_v6": 0.7435508742928505, "step": 3888 }, { "epoch": 0.8969225225225225, "grad_norm": 25.209453582763672, "learning_rate": 5.424409245630042e-06, "loss": 11.468618392944336, "loss_alignment": 0.90185546875, "loss_alignment_w": 0.450927734375, "loss_sft": 0.2653420902788639, "loss_total_v6": 0.7167886272072792, "step": 3889 }, { "epoch": 0.8971531531531531, "grad_norm": 24.134593963623047, "learning_rate": 5.4224026377218365e-06, "loss": 11.6650390625, "loss_alignment": 0.87646484375, "loss_alignment_w": 0.438232421875, "loss_sft": 0.2909851260483265, "loss_total_v6": 0.7290649488568306, "step": 3890 }, { "epoch": 0.8973837837837838, "grad_norm": 25.965147018432617, "learning_rate": 5.420395961290506e-06, "loss": 11.492471694946289, "loss_alignment": 0.91943359375, "loss_alignment_w": 0.459716796875, "loss_sft": 0.25972237810492516, "loss_total_v6": 0.718279518187046, "step": 3891 }, { "epoch": 0.8976144144144144, "grad_norm": 24.582611083984375, "learning_rate": 5.41838921666158e-06, "loss": 11.657630920410156, "loss_alignment": 0.87353515625, "loss_alignment_w": 0.436767578125, "loss_sft": 0.29233790189027786, "loss_total_v6": 0.7286019250750542, "step": 3892 }, { "epoch": 0.897845045045045, "grad_norm": 25.79593849182129, "learning_rate": 5.416382404160593e-06, "loss": 11.302179336547852, "loss_alignment": 0.88720703125, "loss_alignment_w": 0.443603515625, "loss_sft": 0.26165348663926125, "loss_total_v6": 0.7063861712813377, "step": 3893 }, { "epoch": 0.8980756756756757, "grad_norm": 28.80843162536621, "learning_rate": 5.4143755241131e-06, "loss": 11.842936515808105, "loss_alignment": 0.8701171875, "loss_alignment_w": 0.43505859375, "loss_sft": 0.3039500489830971, "loss_total_v6": 0.740183562040329, "step": 3894 }, { "epoch": 0.8983063063063063, "grad_norm": 26.9193115234375, "learning_rate": 5.41236857684466e-06, "loss": 11.945902824401855, "loss_alignment": 0.888671875, "loss_alignment_w": 0.4443359375, "loss_sft": 0.3034884352236986, "loss_total_v6": 0.7466189116239548, "step": 3895 }, { "epoch": 0.8985369369369369, "grad_norm": 23.94438362121582, "learning_rate": 5.4103615626808426e-06, "loss": 11.798027992248535, "loss_alignment": 0.90966796875, "loss_alignment_w": 0.454833984375, "loss_sft": 0.2825275082141161, "loss_total_v6": 0.7373767420649529, "step": 3896 }, { "epoch": 0.8987675675675676, "grad_norm": 24.676225662231445, "learning_rate": 5.408354481947234e-06, "loss": 11.286382675170898, "loss_alignment": 0.89453125, "loss_alignment_w": 0.447265625, "loss_sft": 0.2581486012786627, "loss_total_v6": 0.7053989470005035, "step": 3897 }, { "epoch": 0.8989981981981982, "grad_norm": 26.185710906982422, "learning_rate": 5.406347334969424e-06, "loss": 12.139495849609375, "loss_alignment": 0.89013671875, "loss_alignment_w": 0.445068359375, "loss_sft": 0.31366534531116486, "loss_total_v6": 0.7587184458971024, "step": 3898 }, { "epoch": 0.8992288288288288, "grad_norm": 27.687578201293945, "learning_rate": 5.404340122073017e-06, "loss": 12.081268310546875, "loss_alignment": 0.9404296875, "loss_alignment_w": 0.47021484375, "loss_sft": 0.2848034147173166, "loss_total_v6": 0.7550792843103409, "step": 3899 }, { "epoch": 0.8994594594594595, "grad_norm": 28.034440994262695, "learning_rate": 5.402332843583631e-06, "loss": 11.682601928710938, "loss_alignment": 0.88720703125, "loss_alignment_w": 0.443603515625, "loss_sft": 0.28645235672593117, "loss_total_v6": 0.7301626652479172, "step": 3900 }, { "epoch": 0.89969009009009, "grad_norm": 27.097179412841797, "learning_rate": 5.400325499826886e-06, "loss": 12.27033805847168, "loss_alignment": 0.9130859375, "loss_alignment_w": 0.45654296875, "loss_sft": 0.3098801113665104, "loss_total_v6": 0.7668961063027382, "step": 3901 }, { "epoch": 0.8999207207207207, "grad_norm": 26.848615646362305, "learning_rate": 5.398318091128423e-06, "loss": 11.531627655029297, "loss_alignment": 0.88916015625, "loss_alignment_w": 0.444580078125, "loss_sft": 0.27590250596404076, "loss_total_v6": 0.7207267358899117, "step": 3902 }, { "epoch": 0.9001513513513514, "grad_norm": 27.837875366210938, "learning_rate": 5.396310617813884e-06, "loss": 11.336381912231445, "loss_alignment": 0.85888671875, "loss_alignment_w": 0.429443359375, "loss_sft": 0.279172046110034, "loss_total_v6": 0.7085238471627235, "step": 3903 }, { "epoch": 0.9003819819819819, "grad_norm": 26.085485458374023, "learning_rate": 5.39430308020893e-06, "loss": 11.522871971130371, "loss_alignment": 0.87841796875, "loss_alignment_w": 0.439208984375, "loss_sft": 0.2809551991522312, "loss_total_v6": 0.7201794907450676, "step": 3904 }, { "epoch": 0.9006126126126126, "grad_norm": 23.329570770263672, "learning_rate": 5.392295478639226e-06, "loss": 11.633979797363281, "loss_alignment": 0.87939453125, "loss_alignment_w": 0.439697265625, "loss_sft": 0.28596158139407635, "loss_total_v6": 0.7271237149834633, "step": 3905 }, { "epoch": 0.9008432432432433, "grad_norm": 26.667024612426758, "learning_rate": 5.390287813430449e-06, "loss": 11.90313720703125, "loss_alignment": 0.87939453125, "loss_alignment_w": 0.439697265625, "loss_sft": 0.30440134182572365, "loss_total_v6": 0.7439460381865501, "step": 3906 }, { "epoch": 0.9010738738738738, "grad_norm": 26.772886276245117, "learning_rate": 5.388280084908289e-06, "loss": 12.02960205078125, "loss_alignment": 0.8720703125, "loss_alignment_w": 0.43603515625, "loss_sft": 0.3152198698371649, "loss_total_v6": 0.7518501281738281, "step": 3907 }, { "epoch": 0.9013045045045045, "grad_norm": 25.603315353393555, "learning_rate": 5.386272293398445e-06, "loss": 11.22503662109375, "loss_alignment": 0.89794921875, "loss_alignment_w": 0.448974609375, "loss_sft": 0.2513847481459379, "loss_total_v6": 0.7015648260712624, "step": 3908 }, { "epoch": 0.9015351351351352, "grad_norm": 28.820194244384766, "learning_rate": 5.384264439226622e-06, "loss": 11.219197273254395, "loss_alignment": 0.87939453125, "loss_alignment_w": 0.439697265625, "loss_sft": 0.2620366159826517, "loss_total_v6": 0.7011998519301414, "step": 3909 }, { "epoch": 0.9017657657657657, "grad_norm": 27.24315071105957, "learning_rate": 5.382256522718542e-06, "loss": 11.674129486083984, "loss_alignment": 0.86962890625, "loss_alignment_w": 0.434814453125, "loss_sft": 0.29439138248562813, "loss_total_v6": 0.7296330779790878, "step": 3910 }, { "epoch": 0.9019963963963964, "grad_norm": 27.290271759033203, "learning_rate": 5.380248544199932e-06, "loss": 11.666377067565918, "loss_alignment": 0.8837890625, "loss_alignment_w": 0.44189453125, "loss_sft": 0.2881542854011059, "loss_total_v6": 0.7291485667228699, "step": 3911 }, { "epoch": 0.9022270270270271, "grad_norm": 25.992605209350586, "learning_rate": 5.378240503996531e-06, "loss": 11.191116333007812, "loss_alignment": 0.8447265625, "loss_alignment_w": 0.42236328125, "loss_sft": 0.2767304889857769, "loss_total_v6": 0.6994447261095047, "step": 3912 }, { "epoch": 0.9024576576576576, "grad_norm": 28.642610549926758, "learning_rate": 5.376232402434088e-06, "loss": 12.021248817443848, "loss_alignment": 0.90185546875, "loss_alignment_w": 0.450927734375, "loss_sft": 0.30014093592762947, "loss_total_v6": 0.7513280659914017, "step": 3913 }, { "epoch": 0.9026882882882883, "grad_norm": 25.90138053894043, "learning_rate": 5.374224239838365e-06, "loss": 12.295845031738281, "loss_alignment": 0.912109375, "loss_alignment_w": 0.4560546875, "loss_sft": 0.3128628507256508, "loss_total_v6": 0.7684902995824814, "step": 3914 }, { "epoch": 0.902918918918919, "grad_norm": 24.0093994140625, "learning_rate": 5.372216016535125e-06, "loss": 11.485240936279297, "loss_alignment": 0.89404296875, "loss_alignment_w": 0.447021484375, "loss_sft": 0.2710959855467081, "loss_total_v6": 0.7178275734186172, "step": 3915 }, { "epoch": 0.9031495495495495, "grad_norm": 26.37392807006836, "learning_rate": 5.370207732850151e-06, "loss": 11.782951354980469, "loss_alignment": 0.87646484375, "loss_alignment_w": 0.438232421875, "loss_sft": 0.2979273423552513, "loss_total_v6": 0.7364344298839569, "step": 3916 }, { "epoch": 0.9033801801801802, "grad_norm": 25.65019989013672, "learning_rate": 5.368199389109229e-06, "loss": 11.557486534118652, "loss_alignment": 0.87646484375, "loss_alignment_w": 0.438232421875, "loss_sft": 0.2833628263324499, "loss_total_v6": 0.7223429083824158, "step": 3917 }, { "epoch": 0.9036108108108108, "grad_norm": 25.97389793395996, "learning_rate": 5.366190985638159e-06, "loss": 11.665276527404785, "loss_alignment": 0.86328125, "loss_alignment_w": 0.431640625, "loss_sft": 0.296477809548378, "loss_total_v6": 0.7290797457098961, "step": 3918 }, { "epoch": 0.9038414414414414, "grad_norm": 32.293888092041016, "learning_rate": 5.364182522762747e-06, "loss": 11.477945327758789, "loss_alignment": 0.875, "loss_alignment_w": 0.4375, "loss_sft": 0.27950537018477917, "loss_total_v6": 0.7173715978860855, "step": 3919 }, { "epoch": 0.9040720720720721, "grad_norm": 28.5564022064209, "learning_rate": 5.362174000808813e-06, "loss": 11.766831398010254, "loss_alignment": 0.88427734375, "loss_alignment_w": 0.442138671875, "loss_sft": 0.29414280131459236, "loss_total_v6": 0.7354269623756409, "step": 3920 }, { "epoch": 0.9043027027027027, "grad_norm": 26.56691551208496, "learning_rate": 5.36016542010218e-06, "loss": 11.231042861938477, "loss_alignment": 0.8779296875, "loss_alignment_w": 0.43896484375, "loss_sft": 0.2623802162706852, "loss_total_v6": 0.701940156519413, "step": 3921 }, { "epoch": 0.9045333333333333, "grad_norm": 25.594192504882812, "learning_rate": 5.35815678096869e-06, "loss": 11.986050605773926, "loss_alignment": 0.8759765625, "loss_alignment_w": 0.43798828125, "loss_sft": 0.31080421060323715, "loss_total_v6": 0.7491281926631927, "step": 3922 }, { "epoch": 0.904763963963964, "grad_norm": 25.261537551879883, "learning_rate": 5.356148083734183e-06, "loss": 11.969664573669434, "loss_alignment": 0.88427734375, "loss_alignment_w": 0.442138671875, "loss_sft": 0.30617899633944035, "loss_total_v6": 0.7481040358543396, "step": 3923 }, { "epoch": 0.9049945945945946, "grad_norm": 26.939720153808594, "learning_rate": 5.354139328724518e-06, "loss": 11.363829612731934, "loss_alignment": 0.884765625, "loss_alignment_w": 0.4423828125, "loss_sft": 0.2678412515670061, "loss_total_v6": 0.7102393135428429, "step": 3924 }, { "epoch": 0.9052252252252252, "grad_norm": 25.415342330932617, "learning_rate": 5.35213051626556e-06, "loss": 10.887104034423828, "loss_alignment": 0.876953125, "loss_alignment_w": 0.4384765625, "loss_sft": 0.241769066080451, "loss_total_v6": 0.6804439872503281, "step": 3925 }, { "epoch": 0.9054558558558559, "grad_norm": 27.025997161865234, "learning_rate": 5.350121646683183e-06, "loss": 11.275894165039062, "loss_alignment": 0.8974609375, "loss_alignment_w": 0.44873046875, "loss_sft": 0.2573404386639595, "loss_total_v6": 0.7047433853149414, "step": 3926 }, { "epoch": 0.9056864864864865, "grad_norm": 26.349681854248047, "learning_rate": 5.348112720303269e-06, "loss": 11.749963760375977, "loss_alignment": 0.89306640625, "loss_alignment_w": 0.446533203125, "loss_sft": 0.2885872144252062, "loss_total_v6": 0.7343727499246597, "step": 3927 }, { "epoch": 0.9059171171171171, "grad_norm": 27.248254776000977, "learning_rate": 5.346103737451711e-06, "loss": 11.87492561340332, "loss_alignment": 0.873046875, "loss_alignment_w": 0.4365234375, "loss_sft": 0.30608665384352207, "loss_total_v6": 0.7421828731894493, "step": 3928 }, { "epoch": 0.9061477477477478, "grad_norm": 26.014177322387695, "learning_rate": 5.344094698454413e-06, "loss": 11.530869483947754, "loss_alignment": 0.853515625, "loss_alignment_w": 0.4267578125, "loss_sft": 0.29386053420603275, "loss_total_v6": 0.7206793874502182, "step": 3929 }, { "epoch": 0.9063783783783784, "grad_norm": 25.075870513916016, "learning_rate": 5.342085603637282e-06, "loss": 11.512092590332031, "loss_alignment": 0.880859375, "loss_alignment_w": 0.4404296875, "loss_sft": 0.2796559054404497, "loss_total_v6": 0.7195057794451714, "step": 3930 }, { "epoch": 0.906609009009009, "grad_norm": 24.000877380371094, "learning_rate": 5.3400764533262415e-06, "loss": 11.383222579956055, "loss_alignment": 0.8544921875, "loss_alignment_w": 0.42724609375, "loss_sft": 0.2843426764011383, "loss_total_v6": 0.7114514335989952, "step": 3931 }, { "epoch": 0.9068396396396396, "grad_norm": 24.666656494140625, "learning_rate": 5.338067247847219e-06, "loss": 11.834705352783203, "loss_alignment": 0.8828125, "loss_alignment_w": 0.44140625, "loss_sft": 0.2977440394461155, "loss_total_v6": 0.7396690621972084, "step": 3932 }, { "epoch": 0.9070702702702703, "grad_norm": 24.729305267333984, "learning_rate": 5.336057987526152e-06, "loss": 11.767152786254883, "loss_alignment": 0.890625, "loss_alignment_w": 0.4453125, "loss_sft": 0.2906075417995453, "loss_total_v6": 0.7354470416903496, "step": 3933 }, { "epoch": 0.9073009009009009, "grad_norm": 30.018388748168945, "learning_rate": 5.33404867268899e-06, "loss": 12.206204414367676, "loss_alignment": 0.89794921875, "loss_alignment_w": 0.448974609375, "loss_sft": 0.31432516127824783, "loss_total_v6": 0.7628877609968185, "step": 3934 }, { "epoch": 0.9075315315315315, "grad_norm": 25.203702926635742, "learning_rate": 5.332039303661683e-06, "loss": 11.571184158325195, "loss_alignment": 0.86962890625, "loss_alignment_w": 0.434814453125, "loss_sft": 0.28907118551433086, "loss_total_v6": 0.7231990173459053, "step": 3935 }, { "epoch": 0.9077621621621622, "grad_norm": 27.31273078918457, "learning_rate": 5.330029880770203e-06, "loss": 11.285224914550781, "loss_alignment": 0.88720703125, "loss_alignment_w": 0.443603515625, "loss_sft": 0.2620587833225727, "loss_total_v6": 0.7053266018629074, "step": 3936 }, { "epoch": 0.9079927927927928, "grad_norm": 28.323253631591797, "learning_rate": 5.328020404340517e-06, "loss": 11.347408294677734, "loss_alignment": 0.87744140625, "loss_alignment_w": 0.438720703125, "loss_sft": 0.27104165963828564, "loss_total_v6": 0.7092130407691002, "step": 3937 }, { "epoch": 0.9082234234234234, "grad_norm": 27.4581241607666, "learning_rate": 5.326010874698608e-06, "loss": 11.164654731750488, "loss_alignment": 0.8505859375, "loss_alignment_w": 0.42529296875, "loss_sft": 0.2715519480407238, "loss_total_v6": 0.6977909430861473, "step": 3938 }, { "epoch": 0.9084540540540541, "grad_norm": 24.890214920043945, "learning_rate": 5.324001292170469e-06, "loss": 11.270599365234375, "loss_alignment": 0.9033203125, "loss_alignment_w": 0.45166015625, "loss_sft": 0.2537899259477854, "loss_total_v6": 0.7044124901294708, "step": 3939 }, { "epoch": 0.9086846846846847, "grad_norm": 31.886539459228516, "learning_rate": 5.3219916570820976e-06, "loss": 11.889045715332031, "loss_alignment": 0.900390625, "loss_alignment_w": 0.4501953125, "loss_sft": 0.29319049045443535, "loss_total_v6": 0.7430653497576714, "step": 3940 }, { "epoch": 0.9089153153153153, "grad_norm": 26.231943130493164, "learning_rate": 5.319981969759502e-06, "loss": 11.937915802001953, "loss_alignment": 0.88671875, "loss_alignment_w": 0.443359375, "loss_sft": 0.30210423097014427, "loss_total_v6": 0.7461197525262833, "step": 3941 }, { "epoch": 0.909145945945946, "grad_norm": 24.59029769897461, "learning_rate": 5.317972230528698e-06, "loss": 11.7101469039917, "loss_alignment": 0.90625, "loss_alignment_w": 0.453125, "loss_sft": 0.2789575234055519, "loss_total_v6": 0.7318841814994812, "step": 3942 }, { "epoch": 0.9093765765765766, "grad_norm": 25.479206085205078, "learning_rate": 5.3159624397157075e-06, "loss": 11.435722351074219, "loss_alignment": 0.89111328125, "loss_alignment_w": 0.445556640625, "loss_sft": 0.2684283684939146, "loss_total_v6": 0.7147326841950417, "step": 3943 }, { "epoch": 0.9096072072072072, "grad_norm": 25.61235237121582, "learning_rate": 5.3139525976465675e-06, "loss": 11.661087036132812, "loss_alignment": 0.90625, "loss_alignment_w": 0.453125, "loss_sft": 0.2772035412490368, "loss_total_v6": 0.7288179397583008, "step": 3944 }, { "epoch": 0.9098378378378379, "grad_norm": 26.99464225769043, "learning_rate": 5.311942704647317e-06, "loss": 12.267422676086426, "loss_alignment": 0.91455078125, "loss_alignment_w": 0.457275390625, "loss_sft": 0.3098810501396656, "loss_total_v6": 0.7667139321565628, "step": 3945 }, { "epoch": 0.9100684684684685, "grad_norm": 25.6917667388916, "learning_rate": 5.309932761044006e-06, "loss": 11.750534057617188, "loss_alignment": 0.86181640625, "loss_alignment_w": 0.430908203125, "loss_sft": 0.3037595637142658, "loss_total_v6": 0.7344083562493324, "step": 3946 }, { "epoch": 0.9102990990990991, "grad_norm": 24.4786376953125, "learning_rate": 5.307922767162692e-06, "loss": 11.4395170211792, "loss_alignment": 0.86865234375, "loss_alignment_w": 0.434326171875, "loss_sft": 0.28175752982497215, "loss_total_v6": 0.7149698138237, "step": 3947 }, { "epoch": 0.9105297297297297, "grad_norm": 29.53768539428711, "learning_rate": 5.305912723329442e-06, "loss": 11.797597885131836, "loss_alignment": 0.8916015625, "loss_alignment_w": 0.44580078125, "loss_sft": 0.29177800565958023, "loss_total_v6": 0.7373499274253845, "step": 3948 }, { "epoch": 0.9107603603603603, "grad_norm": 25.32670021057129, "learning_rate": 5.303902629870327e-06, "loss": 11.677563667297363, "loss_alignment": 0.87744140625, "loss_alignment_w": 0.438720703125, "loss_sft": 0.29074557311832905, "loss_total_v6": 0.7298477292060852, "step": 3949 }, { "epoch": 0.910990990990991, "grad_norm": 26.38591194152832, "learning_rate": 5.301892487111431e-06, "loss": 11.944917678833008, "loss_alignment": 0.8955078125, "loss_alignment_w": 0.44775390625, "loss_sft": 0.2988034524023533, "loss_total_v6": 0.746557354927063, "step": 3950 }, { "epoch": 0.9112216216216216, "grad_norm": 29.486841201782227, "learning_rate": 5.2998822953788445e-06, "loss": 11.506675720214844, "loss_alignment": 0.89990234375, "loss_alignment_w": 0.449951171875, "loss_sft": 0.26808691769838333, "loss_total_v6": 0.7191672325134277, "step": 3951 }, { "epoch": 0.9114522522522522, "grad_norm": 28.84946632385254, "learning_rate": 5.297872054998663e-06, "loss": 11.947944641113281, "loss_alignment": 0.90771484375, "loss_alignment_w": 0.453857421875, "loss_sft": 0.2916531562805176, "loss_total_v6": 0.7467465251684189, "step": 3952 }, { "epoch": 0.9116828828828829, "grad_norm": 24.80313491821289, "learning_rate": 5.295861766296996e-06, "loss": 11.588417053222656, "loss_alignment": 0.8740234375, "loss_alignment_w": 0.43701171875, "loss_sft": 0.2872795853763819, "loss_total_v6": 0.7242760434746742, "step": 3953 }, { "epoch": 0.9119135135135135, "grad_norm": 23.980152130126953, "learning_rate": 5.2938514295999544e-06, "loss": 11.14290714263916, "loss_alignment": 0.87060546875, "loss_alignment_w": 0.435302734375, "loss_sft": 0.26318892650306225, "loss_total_v6": 0.6964316964149475, "step": 3954 }, { "epoch": 0.9121441441441441, "grad_norm": 29.152372360229492, "learning_rate": 5.291841045233659e-06, "loss": 11.773117065429688, "loss_alignment": 0.8837890625, "loss_alignment_w": 0.44189453125, "loss_sft": 0.29444412887096405, "loss_total_v6": 0.7358198314905167, "step": 3955 }, { "epoch": 0.9123747747747748, "grad_norm": 22.393569946289062, "learning_rate": 5.289830613524242e-06, "loss": 11.515294075012207, "loss_alignment": 0.86376953125, "loss_alignment_w": 0.431884765625, "loss_sft": 0.2868598233908415, "loss_total_v6": 0.7197058573365211, "step": 3956 }, { "epoch": 0.9126054054054054, "grad_norm": 25.504016876220703, "learning_rate": 5.287820134797837e-06, "loss": 11.914390563964844, "loss_alignment": 0.931640625, "loss_alignment_w": 0.4658203125, "loss_sft": 0.280004046857357, "loss_total_v6": 0.7446494251489639, "step": 3957 }, { "epoch": 0.912836036036036, "grad_norm": 38.24471664428711, "learning_rate": 5.2858096093805896e-06, "loss": 11.378559112548828, "loss_alignment": 0.86279296875, "loss_alignment_w": 0.431396484375, "loss_sft": 0.28069427236914635, "loss_total_v6": 0.7111599668860435, "step": 3958 }, { "epoch": 0.9130666666666667, "grad_norm": 28.816692352294922, "learning_rate": 5.283799037598655e-06, "loss": 11.816398620605469, "loss_alignment": 0.89892578125, "loss_alignment_w": 0.449462890625, "loss_sft": 0.2902064397931099, "loss_total_v6": 0.7385249435901642, "step": 3959 }, { "epoch": 0.9132972972972973, "grad_norm": 25.54609489440918, "learning_rate": 5.281788419778187e-06, "loss": 11.01524543762207, "loss_alignment": 0.88232421875, "loss_alignment_w": 0.441162109375, "loss_sft": 0.24748907797038555, "loss_total_v6": 0.6884528398513794, "step": 3960 }, { "epoch": 0.9135279279279279, "grad_norm": 24.065567016601562, "learning_rate": 5.279777756245357e-06, "loss": 11.880987167358398, "loss_alignment": 0.9033203125, "loss_alignment_w": 0.45166015625, "loss_sft": 0.29048955254256725, "loss_total_v6": 0.7425617054104805, "step": 3961 }, { "epoch": 0.9137585585585586, "grad_norm": 23.682430267333984, "learning_rate": 5.277767047326337e-06, "loss": 11.75655746459961, "loss_alignment": 0.88134765625, "loss_alignment_w": 0.440673828125, "loss_sft": 0.29365322180092335, "loss_total_v6": 0.7347848117351532, "step": 3962 }, { "epoch": 0.9139891891891891, "grad_norm": 26.63176918029785, "learning_rate": 5.275756293347312e-06, "loss": 11.606721878051758, "loss_alignment": 0.849609375, "loss_alignment_w": 0.4248046875, "loss_sft": 0.2999287135899067, "loss_total_v6": 0.7254200801253319, "step": 3963 }, { "epoch": 0.9142198198198198, "grad_norm": 27.484142303466797, "learning_rate": 5.273745494634467e-06, "loss": 11.489236831665039, "loss_alignment": 0.87744140625, "loss_alignment_w": 0.438720703125, "loss_sft": 0.28030259162187576, "loss_total_v6": 0.718077264726162, "step": 3964 }, { "epoch": 0.9144504504504505, "grad_norm": 25.682445526123047, "learning_rate": 5.271734651514001e-06, "loss": 10.71541976928711, "loss_alignment": 0.86669921875, "loss_alignment_w": 0.433349609375, "loss_sft": 0.23479255847632885, "loss_total_v6": 0.6697137877345085, "step": 3965 }, { "epoch": 0.914681081081081, "grad_norm": 33.113460540771484, "learning_rate": 5.269723764312117e-06, "loss": 11.507131576538086, "loss_alignment": 0.875, "loss_alignment_w": 0.4375, "loss_sft": 0.28122268058359623, "loss_total_v6": 0.7191957086324692, "step": 3966 }, { "epoch": 0.9149117117117117, "grad_norm": 29.257354736328125, "learning_rate": 5.267712833355025e-06, "loss": 10.834268569946289, "loss_alignment": 0.88330078125, "loss_alignment_w": 0.441650390625, "loss_sft": 0.2357355197891593, "loss_total_v6": 0.6771417781710625, "step": 3967 }, { "epoch": 0.9151423423423424, "grad_norm": 25.869266510009766, "learning_rate": 5.265701858968944e-06, "loss": 11.522197723388672, "loss_alignment": 0.873046875, "loss_alignment_w": 0.4365234375, "loss_sft": 0.28323241509497166, "loss_total_v6": 0.7201373279094696, "step": 3968 }, { "epoch": 0.9153729729729729, "grad_norm": 25.767772674560547, "learning_rate": 5.263690841480098e-06, "loss": 11.095391273498535, "loss_alignment": 0.8583984375, "loss_alignment_w": 0.42919921875, "loss_sft": 0.26423225551843643, "loss_total_v6": 0.6934619694948196, "step": 3969 }, { "epoch": 0.9156036036036036, "grad_norm": 27.262969970703125, "learning_rate": 5.2616797812147205e-06, "loss": 11.550752639770508, "loss_alignment": 0.89013671875, "loss_alignment_w": 0.445068359375, "loss_sft": 0.2775250729173422, "loss_total_v6": 0.7219220697879791, "step": 3970 }, { "epoch": 0.9158342342342343, "grad_norm": 26.2028865814209, "learning_rate": 5.259668678499047e-06, "loss": 11.421186447143555, "loss_alignment": 0.88671875, "loss_alignment_w": 0.443359375, "loss_sft": 0.2700528409332037, "loss_total_v6": 0.7138241901993752, "step": 3971 }, { "epoch": 0.9160648648648648, "grad_norm": 26.653409957885742, "learning_rate": 5.257657533659326e-06, "loss": 11.902217864990234, "loss_alignment": 0.86474609375, "loss_alignment_w": 0.432373046875, "loss_sft": 0.31192754209041595, "loss_total_v6": 0.7438885942101479, "step": 3972 }, { "epoch": 0.9162954954954955, "grad_norm": 25.17473793029785, "learning_rate": 5.255646347021811e-06, "loss": 11.3479642868042, "loss_alignment": 0.8876953125, "loss_alignment_w": 0.44384765625, "loss_sft": 0.2657053116708994, "loss_total_v6": 0.709247775375843, "step": 3973 }, { "epoch": 0.9165261261261262, "grad_norm": 27.541494369506836, "learning_rate": 5.2536351189127565e-06, "loss": 11.469654083251953, "loss_alignment": 0.8720703125, "loss_alignment_w": 0.43603515625, "loss_sft": 0.28080296143889427, "loss_total_v6": 0.7168533504009247, "step": 3974 }, { "epoch": 0.9167567567567567, "grad_norm": 35.631980895996094, "learning_rate": 5.251623849658434e-06, "loss": 10.999469757080078, "loss_alignment": 0.89013671875, "loss_alignment_w": 0.445068359375, "loss_sft": 0.24203229509294033, "loss_total_v6": 0.6874668523669243, "step": 3975 }, { "epoch": 0.9169873873873874, "grad_norm": 28.08428955078125, "learning_rate": 5.249612539585114e-06, "loss": 11.325713157653809, "loss_alignment": 0.88037109375, "loss_alignment_w": 0.440185546875, "loss_sft": 0.2684344854205847, "loss_total_v6": 0.7078571021556854, "step": 3976 }, { "epoch": 0.917218018018018, "grad_norm": 26.803586959838867, "learning_rate": 5.247601189019072e-06, "loss": 11.866189956665039, "loss_alignment": 0.87646484375, "loss_alignment_w": 0.438232421875, "loss_sft": 0.3038317207247019, "loss_total_v6": 0.7416368946433067, "step": 3977 }, { "epoch": 0.9174486486486486, "grad_norm": 24.212045669555664, "learning_rate": 5.245589798286601e-06, "loss": 12.237066268920898, "loss_alignment": 0.916015625, "loss_alignment_w": 0.4580078125, "loss_sft": 0.3056492079049349, "loss_total_v6": 0.7648166567087173, "step": 3978 }, { "epoch": 0.9176792792792793, "grad_norm": 25.553178787231445, "learning_rate": 5.2435783677139895e-06, "loss": 11.603899955749512, "loss_alignment": 0.892578125, "loss_alignment_w": 0.4462890625, "loss_sft": 0.27813071571290493, "loss_total_v6": 0.7252437546849251, "step": 3979 }, { "epoch": 0.91790990990991, "grad_norm": 26.817764282226562, "learning_rate": 5.241566897627536e-06, "loss": 10.898592948913574, "loss_alignment": 0.87890625, "loss_alignment_w": 0.439453125, "loss_sft": 0.24270080402493477, "loss_total_v6": 0.6811620742082596, "step": 3980 }, { "epoch": 0.9181405405405405, "grad_norm": 26.171749114990234, "learning_rate": 5.2395553883535475e-06, "loss": 11.379332542419434, "loss_alignment": 0.86572265625, "loss_alignment_w": 0.432861328125, "loss_sft": 0.2776450663805008, "loss_total_v6": 0.7112082913517952, "step": 3981 }, { "epoch": 0.9183711711711712, "grad_norm": 25.483272552490234, "learning_rate": 5.237543840218333e-06, "loss": 11.131025314331055, "loss_alignment": 0.86083984375, "loss_alignment_w": 0.430419921875, "loss_sft": 0.2652691565454006, "loss_total_v6": 0.6956891119480133, "step": 3982 }, { "epoch": 0.9186018018018018, "grad_norm": 27.614511489868164, "learning_rate": 5.235532253548213e-06, "loss": 11.548747062683105, "loss_alignment": 0.88818359375, "loss_alignment_w": 0.444091796875, "loss_sft": 0.27778122015297413, "loss_total_v6": 0.7217966988682747, "step": 3983 }, { "epoch": 0.9188324324324324, "grad_norm": 24.655017852783203, "learning_rate": 5.233520628669512e-06, "loss": 12.045631408691406, "loss_alignment": 0.8798828125, "loss_alignment_w": 0.43994140625, "loss_sft": 0.3112473040819168, "loss_total_v6": 0.7528519332408905, "step": 3984 }, { "epoch": 0.9190630630630631, "grad_norm": 25.51428985595703, "learning_rate": 5.23150896590856e-06, "loss": 11.34241008758545, "loss_alignment": 0.89013671875, "loss_alignment_w": 0.445068359375, "loss_sft": 0.26412221416831017, "loss_total_v6": 0.7089006453752518, "step": 3985 }, { "epoch": 0.9192936936936937, "grad_norm": 27.5350399017334, "learning_rate": 5.22949726559169e-06, "loss": 11.563175201416016, "loss_alignment": 0.86376953125, "loss_alignment_w": 0.431884765625, "loss_sft": 0.29018804244697094, "loss_total_v6": 0.7226984351873398, "step": 3986 }, { "epoch": 0.9195243243243243, "grad_norm": 24.147497177124023, "learning_rate": 5.227485528045249e-06, "loss": 10.6306734085083, "loss_alignment": 0.84716796875, "loss_alignment_w": 0.423583984375, "loss_sft": 0.23990230448544025, "loss_total_v6": 0.6644170731306076, "step": 3987 }, { "epoch": 0.919754954954955, "grad_norm": 24.53484344482422, "learning_rate": 5.225473753595585e-06, "loss": 11.22220516204834, "loss_alignment": 0.8642578125, "loss_alignment_w": 0.43212890625, "loss_sft": 0.26907581835985184, "loss_total_v6": 0.7013878375291824, "step": 3988 }, { "epoch": 0.9199855855855856, "grad_norm": 29.246749877929688, "learning_rate": 5.2234619425690506e-06, "loss": 11.530561447143555, "loss_alignment": 0.88818359375, "loss_alignment_w": 0.444091796875, "loss_sft": 0.27608006075024605, "loss_total_v6": 0.720660112798214, "step": 3989 }, { "epoch": 0.9202162162162162, "grad_norm": 29.32939338684082, "learning_rate": 5.221450095292008e-06, "loss": 12.024544715881348, "loss_alignment": 0.888671875, "loss_alignment_w": 0.4443359375, "loss_sft": 0.30658775568008423, "loss_total_v6": 0.751534029841423, "step": 3990 }, { "epoch": 0.9204468468468469, "grad_norm": 26.18902015686035, "learning_rate": 5.219438212090823e-06, "loss": 12.352447509765625, "loss_alignment": 0.90087890625, "loss_alignment_w": 0.450439453125, "loss_sft": 0.32201576977968216, "loss_total_v6": 0.772027961909771, "step": 3991 }, { "epoch": 0.9206774774774775, "grad_norm": 25.402828216552734, "learning_rate": 5.217426293291869e-06, "loss": 11.422019958496094, "loss_alignment": 0.8955078125, "loss_alignment_w": 0.44775390625, "loss_sft": 0.26644282788038254, "loss_total_v6": 0.7138762846589088, "step": 3992 }, { "epoch": 0.9209081081081081, "grad_norm": 25.35210609436035, "learning_rate": 5.215414339221522e-06, "loss": 11.225061416625977, "loss_alignment": 0.88671875, "loss_alignment_w": 0.443359375, "loss_sft": 0.25823751278221607, "loss_total_v6": 0.7015663981437683, "step": 3993 }, { "epoch": 0.9211387387387387, "grad_norm": 27.584461212158203, "learning_rate": 5.213402350206167e-06, "loss": 11.541118621826172, "loss_alignment": 0.884765625, "loss_alignment_w": 0.4423828125, "loss_sft": 0.2777011785656214, "loss_total_v6": 0.7213199585676193, "step": 3994 }, { "epoch": 0.9213693693693694, "grad_norm": 25.26320457458496, "learning_rate": 5.211390326572196e-06, "loss": 11.737466812133789, "loss_alignment": 0.869140625, "loss_alignment_w": 0.4345703125, "loss_sft": 0.29871612414717674, "loss_total_v6": 0.7335916310548782, "step": 3995 }, { "epoch": 0.9216, "grad_norm": 28.410926818847656, "learning_rate": 5.209378268645998e-06, "loss": 11.093849182128906, "loss_alignment": 0.85986328125, "loss_alignment_w": 0.429931640625, "loss_sft": 0.2628387752920389, "loss_total_v6": 0.6933655217289925, "step": 3996 }, { "epoch": 0.9218306306306306, "grad_norm": 27.069164276123047, "learning_rate": 5.207366176753979e-06, "loss": 11.81894302368164, "loss_alignment": 0.87353515625, "loss_alignment_w": 0.436767578125, "loss_sft": 0.30074146017432213, "loss_total_v6": 0.7386839687824249, "step": 3997 }, { "epoch": 0.9220612612612613, "grad_norm": 27.568363189697266, "learning_rate": 5.205354051222546e-06, "loss": 11.768484115600586, "loss_alignment": 0.888671875, "loss_alignment_w": 0.4443359375, "loss_sft": 0.2913316283375025, "loss_total_v6": 0.7355302274227142, "step": 3998 }, { "epoch": 0.9222918918918919, "grad_norm": 28.267610549926758, "learning_rate": 5.203341892378104e-06, "loss": 11.945947647094727, "loss_alignment": 0.890625, "loss_alignment_w": 0.4453125, "loss_sft": 0.300790436565876, "loss_total_v6": 0.7466217055916786, "step": 3999 }, { "epoch": 0.9225225225225225, "grad_norm": 27.42057228088379, "learning_rate": 5.201329700547077e-06, "loss": 11.414398193359375, "loss_alignment": 0.86572265625, "loss_alignment_w": 0.432861328125, "loss_sft": 0.2795315384864807, "loss_total_v6": 0.7133999243378639, "step": 4000 }, { "epoch": 0.9225225225225225, "eval_loss": 0.8610600233078003, "eval_loss_alignment": 0.8869216431221462, "eval_loss_alignment_w": 0.4434608215610731, "eval_loss_contrast": 0.0, "eval_loss_highlevel": 0.0, "eval_loss_sft": 0.4175991363415163, "eval_loss_total_v6": 0.8610599621041725, "eval_loss_wm_recon": 0.0, "eval_loss_wm_recon_w": 0.0, "eval_runtime": 335.5153, "eval_samples_per_second": 5.201, "eval_steps_per_second": 0.653, "step": 4000 }, { "epoch": 0.9227531531531532, "grad_norm": 25.065174102783203, "learning_rate": 5.199317476055883e-06, "loss": 11.539581298828125, "loss_alignment": 0.87890625, "loss_alignment_w": 0.439453125, "loss_sft": 0.28059571608901024, "loss_total_v6": 0.7212238013744354, "step": 4001 }, { "epoch": 0.9229837837837838, "grad_norm": 25.917522430419922, "learning_rate": 5.19730521923095e-06, "loss": 11.248576164245605, "loss_alignment": 0.8876953125, "loss_alignment_w": 0.44384765625, "loss_sft": 0.2588984575122595, "loss_total_v6": 0.7030360549688339, "step": 4002 }, { "epoch": 0.9232144144144144, "grad_norm": 28.165395736694336, "learning_rate": 5.195292930398716e-06, "loss": 11.462852478027344, "loss_alignment": 0.87109375, "loss_alignment_w": 0.435546875, "loss_sft": 0.280423641204834, "loss_total_v6": 0.7164282649755478, "step": 4003 }, { "epoch": 0.9234450450450451, "grad_norm": 30.290128707885742, "learning_rate": 5.193280609885611e-06, "loss": 11.381486892700195, "loss_alignment": 0.9033203125, "loss_alignment_w": 0.45166015625, "loss_sft": 0.260674636811018, "loss_total_v6": 0.7113429680466652, "step": 4004 }, { "epoch": 0.9236756756756757, "grad_norm": 25.899560928344727, "learning_rate": 5.191268258018084e-06, "loss": 11.419790267944336, "loss_alignment": 0.884765625, "loss_alignment_w": 0.4423828125, "loss_sft": 0.27158296667039394, "loss_total_v6": 0.7137369140982628, "step": 4005 }, { "epoch": 0.9239063063063063, "grad_norm": 25.30133819580078, "learning_rate": 5.1892558751225815e-06, "loss": 11.366809844970703, "loss_alignment": 0.8486328125, "loss_alignment_w": 0.42431640625, "loss_sft": 0.28540730476379395, "loss_total_v6": 0.710425615310669, "step": 4006 }, { "epoch": 0.924136936936937, "grad_norm": 27.351720809936523, "learning_rate": 5.187243461525559e-06, "loss": 11.483972549438477, "loss_alignment": 0.875, "loss_alignment_w": 0.4375, "loss_sft": 0.2803551200777292, "loss_total_v6": 0.7177483066916466, "step": 4007 }, { "epoch": 0.9243675675675675, "grad_norm": 24.299654006958008, "learning_rate": 5.18523101755347e-06, "loss": 11.297268867492676, "loss_alignment": 0.86962890625, "loss_alignment_w": 0.434814453125, "loss_sft": 0.2717226482927799, "loss_total_v6": 0.7060793414711952, "step": 4008 }, { "epoch": 0.9245981981981982, "grad_norm": 29.392208099365234, "learning_rate": 5.183218543532782e-06, "loss": 11.61266803741455, "loss_alignment": 0.87744140625, "loss_alignment_w": 0.438720703125, "loss_sft": 0.2867048252373934, "loss_total_v6": 0.7257917374372482, "step": 4009 }, { "epoch": 0.9248288288288288, "grad_norm": 29.10157585144043, "learning_rate": 5.1812060397899624e-06, "loss": 11.432504653930664, "loss_alignment": 0.87939453125, "loss_alignment_w": 0.439697265625, "loss_sft": 0.27459016628563404, "loss_total_v6": 0.7145315483212471, "step": 4010 }, { "epoch": 0.9250594594594594, "grad_norm": 27.274797439575195, "learning_rate": 5.179193506651483e-06, "loss": 11.670364379882812, "loss_alignment": 0.876953125, "loss_alignment_w": 0.4384765625, "loss_sft": 0.29047873243689537, "loss_total_v6": 0.7293978035449982, "step": 4011 }, { "epoch": 0.9252900900900901, "grad_norm": 27.03717613220215, "learning_rate": 5.177180944443821e-06, "loss": 11.410306930541992, "loss_alignment": 0.86474609375, "loss_alignment_w": 0.432373046875, "loss_sft": 0.28101533837616444, "loss_total_v6": 0.7131442278623581, "step": 4012 }, { "epoch": 0.9255207207207207, "grad_norm": 27.051624298095703, "learning_rate": 5.175168353493463e-06, "loss": 11.434144973754883, "loss_alignment": 0.86767578125, "loss_alignment_w": 0.433837890625, "loss_sft": 0.27930073998868465, "loss_total_v6": 0.7146340236067772, "step": 4013 }, { "epoch": 0.9257513513513513, "grad_norm": 26.080448150634766, "learning_rate": 5.173155734126891e-06, "loss": 11.908778190612793, "loss_alignment": 0.8818359375, "loss_alignment_w": 0.44091796875, "loss_sft": 0.30431145057082176, "loss_total_v6": 0.7442986443638802, "step": 4014 }, { "epoch": 0.925981981981982, "grad_norm": 26.100170135498047, "learning_rate": 5.1711430866705995e-06, "loss": 11.828346252441406, "loss_alignment": 0.90283203125, "loss_alignment_w": 0.451416015625, "loss_sft": 0.2887100949883461, "loss_total_v6": 0.7392716258764267, "step": 4015 }, { "epoch": 0.9262126126126126, "grad_norm": 25.177261352539062, "learning_rate": 5.169130411451083e-06, "loss": 11.83354377746582, "loss_alignment": 0.90087890625, "loss_alignment_w": 0.450439453125, "loss_sft": 0.2892333399504423, "loss_total_v6": 0.7395965084433556, "step": 4016 }, { "epoch": 0.9264432432432432, "grad_norm": 25.31311798095703, "learning_rate": 5.167117708794845e-06, "loss": 11.096741676330566, "loss_alignment": 0.85498046875, "loss_alignment_w": 0.427490234375, "loss_sft": 0.265369426459074, "loss_total_v6": 0.693546324968338, "step": 4017 }, { "epoch": 0.9266738738738739, "grad_norm": 25.958192825317383, "learning_rate": 5.165104979028388e-06, "loss": 11.689275741577148, "loss_alignment": 0.88427734375, "loss_alignment_w": 0.442138671875, "loss_sft": 0.2882122080773115, "loss_total_v6": 0.7305797711014748, "step": 4018 }, { "epoch": 0.9269045045045045, "grad_norm": 28.7810115814209, "learning_rate": 5.163092222478224e-06, "loss": 11.201258659362793, "loss_alignment": 0.86572265625, "loss_alignment_w": 0.432861328125, "loss_sft": 0.2663628626614809, "loss_total_v6": 0.7000786811113358, "step": 4019 }, { "epoch": 0.9271351351351351, "grad_norm": 25.620784759521484, "learning_rate": 5.1610794394708665e-06, "loss": 11.895459175109863, "loss_alignment": 0.86669921875, "loss_alignment_w": 0.433349609375, "loss_sft": 0.30901793763041496, "loss_total_v6": 0.7434661686420441, "step": 4020 }, { "epoch": 0.9273657657657658, "grad_norm": 26.8702449798584, "learning_rate": 5.1590666303328295e-06, "loss": 12.314512252807617, "loss_alignment": 0.88671875, "loss_alignment_w": 0.443359375, "loss_sft": 0.32580942660570145, "loss_total_v6": 0.7696570530533791, "step": 4021 }, { "epoch": 0.9275963963963963, "grad_norm": 27.103261947631836, "learning_rate": 5.157053795390642e-06, "loss": 11.182151794433594, "loss_alignment": 0.8623046875, "loss_alignment_w": 0.43115234375, "loss_sft": 0.26909015513956547, "loss_total_v6": 0.6988844498991966, "step": 4022 }, { "epoch": 0.927827027027027, "grad_norm": 22.204940795898438, "learning_rate": 5.155040934970827e-06, "loss": 11.084396362304688, "loss_alignment": 0.86572265625, "loss_alignment_w": 0.432861328125, "loss_sft": 0.2602796442806721, "loss_total_v6": 0.6927747279405594, "step": 4023 }, { "epoch": 0.9280576576576577, "grad_norm": 27.324893951416016, "learning_rate": 5.1530280493999165e-06, "loss": 11.383222579956055, "loss_alignment": 0.8720703125, "loss_alignment_w": 0.43603515625, "loss_sft": 0.2765453923493624, "loss_total_v6": 0.7114514037966728, "step": 4024 }, { "epoch": 0.9282882882882882, "grad_norm": 41.330039978027344, "learning_rate": 5.151015139004445e-06, "loss": 11.639801979064941, "loss_alignment": 0.89453125, "loss_alignment_w": 0.447265625, "loss_sft": 0.28141219541430473, "loss_total_v6": 0.7274876311421394, "step": 4025 }, { "epoch": 0.9285189189189189, "grad_norm": 26.38829803466797, "learning_rate": 5.149002204110951e-06, "loss": 11.732625961303711, "loss_alignment": 0.92578125, "loss_alignment_w": 0.462890625, "loss_sft": 0.2710393890738487, "loss_total_v6": 0.7332891374826431, "step": 4026 }, { "epoch": 0.9287495495495496, "grad_norm": 25.770132064819336, "learning_rate": 5.146989245045978e-06, "loss": 10.970584869384766, "loss_alignment": 0.88916015625, "loss_alignment_w": 0.444580078125, "loss_sft": 0.24033381417393684, "loss_total_v6": 0.6856615915894508, "step": 4027 }, { "epoch": 0.9289801801801801, "grad_norm": 24.306304931640625, "learning_rate": 5.144976262136074e-06, "loss": 11.651638984680176, "loss_alignment": 0.87451171875, "loss_alignment_w": 0.437255859375, "loss_sft": 0.29190235771238804, "loss_total_v6": 0.7282274514436722, "step": 4028 }, { "epoch": 0.9292108108108108, "grad_norm": 35.40489196777344, "learning_rate": 5.142963255707787e-06, "loss": 11.247228622436523, "loss_alignment": 0.8896484375, "loss_alignment_w": 0.44482421875, "loss_sft": 0.2580360621213913, "loss_total_v6": 0.7029517889022827, "step": 4029 }, { "epoch": 0.9294414414414415, "grad_norm": 27.299190521240234, "learning_rate": 5.140950226087673e-06, "loss": 11.837141036987305, "loss_alignment": 0.896484375, "loss_alignment_w": 0.4482421875, "loss_sft": 0.2911671865731478, "loss_total_v6": 0.7398213446140289, "step": 4030 }, { "epoch": 0.929672072072072, "grad_norm": 26.467721939086914, "learning_rate": 5.13893717360229e-06, "loss": 11.182523727416992, "loss_alignment": 0.86328125, "loss_alignment_w": 0.431640625, "loss_sft": 0.267740149050951, "loss_total_v6": 0.698907732963562, "step": 4031 }, { "epoch": 0.9299027027027027, "grad_norm": 26.686279296875, "learning_rate": 5.136924098578201e-06, "loss": 12.004051208496094, "loss_alignment": 0.87939453125, "loss_alignment_w": 0.439697265625, "loss_sft": 0.30912161991000175, "loss_total_v6": 0.750253215432167, "step": 4032 }, { "epoch": 0.9301333333333334, "grad_norm": 26.40962791442871, "learning_rate": 5.134911001341971e-06, "loss": 11.073822021484375, "loss_alignment": 0.859375, "loss_alignment_w": 0.4296875, "loss_sft": 0.26346402801573277, "loss_total_v6": 0.6921139061450958, "step": 4033 }, { "epoch": 0.9303639639639639, "grad_norm": 29.3872127532959, "learning_rate": 5.132897882220168e-06, "loss": 11.261823654174805, "loss_alignment": 0.87646484375, "loss_alignment_w": 0.438232421875, "loss_sft": 0.26595200784504414, "loss_total_v6": 0.7038639858365059, "step": 4034 }, { "epoch": 0.9305945945945946, "grad_norm": 24.488088607788086, "learning_rate": 5.130884741539367e-06, "loss": 11.779159545898438, "loss_alignment": 0.89501953125, "loss_alignment_w": 0.447509765625, "loss_sft": 0.289694769307971, "loss_total_v6": 0.7361974269151688, "step": 4035 }, { "epoch": 0.9308252252252253, "grad_norm": 24.641199111938477, "learning_rate": 5.1288715796261425e-06, "loss": 11.369794845581055, "loss_alignment": 0.87255859375, "loss_alignment_w": 0.436279296875, "loss_sft": 0.27512634359300137, "loss_total_v6": 0.7106122002005577, "step": 4036 }, { "epoch": 0.9310558558558558, "grad_norm": 31.76946258544922, "learning_rate": 5.126858396807075e-06, "loss": 11.308688163757324, "loss_alignment": 0.8525390625, "loss_alignment_w": 0.42626953125, "loss_sft": 0.28030986711382866, "loss_total_v6": 0.7067930176854134, "step": 4037 }, { "epoch": 0.9312864864864865, "grad_norm": 31.773439407348633, "learning_rate": 5.124845193408747e-06, "loss": 11.411430358886719, "loss_alignment": 0.87353515625, "loss_alignment_w": 0.436767578125, "loss_sft": 0.27649264223873615, "loss_total_v6": 0.7132144197821617, "step": 4038 }, { "epoch": 0.9315171171171172, "grad_norm": 25.449447631835938, "learning_rate": 5.122831969757747e-06, "loss": 11.480219841003418, "loss_alignment": 0.857421875, "loss_alignment_w": 0.4287109375, "loss_sft": 0.2888333220034838, "loss_total_v6": 0.7175137475132942, "step": 4039 }, { "epoch": 0.9317477477477477, "grad_norm": 24.48404884338379, "learning_rate": 5.120818726180662e-06, "loss": 11.293170928955078, "loss_alignment": 0.8740234375, "loss_alignment_w": 0.43701171875, "loss_sft": 0.26884198747575283, "loss_total_v6": 0.7058231979608536, "step": 4040 }, { "epoch": 0.9319783783783784, "grad_norm": 28.70362663269043, "learning_rate": 5.118805463004087e-06, "loss": 11.476387023925781, "loss_alignment": 0.833984375, "loss_alignment_w": 0.4169921875, "loss_sft": 0.30125858820974827, "loss_total_v6": 0.7172742411494255, "step": 4041 }, { "epoch": 0.932209009009009, "grad_norm": 24.397296905517578, "learning_rate": 5.1167921805546176e-06, "loss": 11.87887191772461, "loss_alignment": 0.8857421875, "loss_alignment_w": 0.44287109375, "loss_sft": 0.299832995980978, "loss_total_v6": 0.7424294352531433, "step": 4042 }, { "epoch": 0.9324396396396396, "grad_norm": 26.761127471923828, "learning_rate": 5.1147788791588516e-06, "loss": 12.213197708129883, "loss_alignment": 0.87548828125, "loss_alignment_w": 0.437744140625, "loss_sft": 0.32602324336767197, "loss_total_v6": 0.7633248493075371, "step": 4043 }, { "epoch": 0.9326702702702703, "grad_norm": 24.78718376159668, "learning_rate": 5.112765559143395e-06, "loss": 11.657837867736816, "loss_alignment": 0.8720703125, "loss_alignment_w": 0.43603515625, "loss_sft": 0.29433445259928703, "loss_total_v6": 0.728614829480648, "step": 4044 }, { "epoch": 0.9329009009009009, "grad_norm": 24.88588523864746, "learning_rate": 5.110752220834849e-06, "loss": 11.171366691589355, "loss_alignment": 0.859375, "loss_alignment_w": 0.4296875, "loss_sft": 0.268995963037014, "loss_total_v6": 0.6982104256749153, "step": 4045 }, { "epoch": 0.9331315315315315, "grad_norm": 29.01752281188965, "learning_rate": 5.108738864559824e-06, "loss": 11.454959869384766, "loss_alignment": 0.8955078125, "loss_alignment_w": 0.44775390625, "loss_sft": 0.26737236231565475, "loss_total_v6": 0.7159349769353867, "step": 4046 }, { "epoch": 0.9333621621621622, "grad_norm": 27.69818115234375, "learning_rate": 5.106725490644932e-06, "loss": 11.757862091064453, "loss_alignment": 0.880859375, "loss_alignment_w": 0.4404296875, "loss_sft": 0.29562691040337086, "loss_total_v6": 0.7348663732409477, "step": 4047 }, { "epoch": 0.9335927927927928, "grad_norm": 25.555801391601562, "learning_rate": 5.1047120994167855e-06, "loss": 11.234903335571289, "loss_alignment": 0.8623046875, "loss_alignment_w": 0.43115234375, "loss_sft": 0.2711817231029272, "loss_total_v6": 0.7021815031766891, "step": 4048 }, { "epoch": 0.9338234234234234, "grad_norm": 25.382911682128906, "learning_rate": 5.102698691202003e-06, "loss": 11.521649360656738, "loss_alignment": 0.92041015625, "loss_alignment_w": 0.460205078125, "loss_sft": 0.26017266511917114, "loss_total_v6": 0.720103070139885, "step": 4049 }, { "epoch": 0.9340540540540541, "grad_norm": 30.998027801513672, "learning_rate": 5.100685266327202e-06, "loss": 11.289176940917969, "loss_alignment": 0.88427734375, "loss_alignment_w": 0.442138671875, "loss_sft": 0.2633738275617361, "loss_total_v6": 0.7055735215544701, "step": 4050 }, { "epoch": 0.9342846846846847, "grad_norm": 27.684871673583984, "learning_rate": 5.098671825119008e-06, "loss": 11.717278480529785, "loss_alignment": 0.90625, "loss_alignment_w": 0.453125, "loss_sft": 0.2790523134171963, "loss_total_v6": 0.7323299050331116, "step": 4051 }, { "epoch": 0.9345153153153153, "grad_norm": 26.99294090270996, "learning_rate": 5.096658367904043e-06, "loss": 11.87299919128418, "loss_alignment": 0.89794921875, "loss_alignment_w": 0.448974609375, "loss_sft": 0.29244697466492653, "loss_total_v6": 0.7420624494552612, "step": 4052 }, { "epoch": 0.934745945945946, "grad_norm": 26.7664794921875, "learning_rate": 5.094644895008936e-06, "loss": 11.164552688598633, "loss_alignment": 0.84912109375, "loss_alignment_w": 0.424560546875, "loss_sft": 0.2731325179338455, "loss_total_v6": 0.6977846026420593, "step": 4053 }, { "epoch": 0.9349765765765766, "grad_norm": 30.008983612060547, "learning_rate": 5.092631406760319e-06, "loss": 11.803560256958008, "loss_alignment": 0.86328125, "loss_alignment_w": 0.431640625, "loss_sft": 0.3058224655687809, "loss_total_v6": 0.7377225011587143, "step": 4054 }, { "epoch": 0.9352072072072072, "grad_norm": 28.12686538696289, "learning_rate": 5.09061790348482e-06, "loss": 11.21406078338623, "loss_alignment": 0.88623046875, "loss_alignment_w": 0.443115234375, "loss_sft": 0.257733054459095, "loss_total_v6": 0.7008787766098976, "step": 4055 }, { "epoch": 0.9354378378378378, "grad_norm": 26.896366119384766, "learning_rate": 5.0886043855090795e-06, "loss": 11.073347091674805, "loss_alignment": 0.888671875, "loss_alignment_w": 0.4443359375, "loss_sft": 0.2476414069533348, "loss_total_v6": 0.6920841559767723, "step": 4056 }, { "epoch": 0.9356684684684685, "grad_norm": 24.05824851989746, "learning_rate": 5.086590853159732e-06, "loss": 11.503429412841797, "loss_alignment": 0.8720703125, "loss_alignment_w": 0.43603515625, "loss_sft": 0.2836310938000679, "loss_total_v6": 0.7189643457531929, "step": 4057 }, { "epoch": 0.9358990990990991, "grad_norm": 23.984006881713867, "learning_rate": 5.084577306763416e-06, "loss": 11.939470291137695, "loss_alignment": 0.9091796875, "loss_alignment_w": 0.45458984375, "loss_sft": 0.2915965598076582, "loss_total_v6": 0.746216893196106, "step": 4058 }, { "epoch": 0.9361297297297297, "grad_norm": 26.163894653320312, "learning_rate": 5.082563746646777e-06, "loss": 11.409956932067871, "loss_alignment": 0.86669921875, "loss_alignment_w": 0.433349609375, "loss_sft": 0.28024571388959885, "loss_total_v6": 0.7131223008036613, "step": 4059 }, { "epoch": 0.9363603603603604, "grad_norm": 34.464385986328125, "learning_rate": 5.080550173136457e-06, "loss": 11.663958549499512, "loss_alignment": 0.87646484375, "loss_alignment_w": 0.438232421875, "loss_sft": 0.2903835028409958, "loss_total_v6": 0.7289974167943001, "step": 4060 }, { "epoch": 0.936590990990991, "grad_norm": 27.071775436401367, "learning_rate": 5.078536586559104e-06, "loss": 12.113557815551758, "loss_alignment": 0.89208984375, "loss_alignment_w": 0.446044921875, "loss_sft": 0.3100758548825979, "loss_total_v6": 0.7570973485708237, "step": 4061 }, { "epoch": 0.9368216216216216, "grad_norm": 25.323745727539062, "learning_rate": 5.0765229872413655e-06, "loss": 10.974964141845703, "loss_alignment": 0.86865234375, "loss_alignment_w": 0.434326171875, "loss_sft": 0.25115131959319115, "loss_total_v6": 0.6859352588653564, "step": 4062 }, { "epoch": 0.9370522522522523, "grad_norm": 28.889484405517578, "learning_rate": 5.074509375509893e-06, "loss": 11.634384155273438, "loss_alignment": 0.87060546875, "loss_alignment_w": 0.435302734375, "loss_sft": 0.2929143775254488, "loss_total_v6": 0.727148987352848, "step": 4063 }, { "epoch": 0.9372828828828829, "grad_norm": 33.35722351074219, "learning_rate": 5.072495751691338e-06, "loss": 11.355856895446777, "loss_alignment": 0.89013671875, "loss_alignment_w": 0.445068359375, "loss_sft": 0.2650541812181473, "loss_total_v6": 0.7097410708665848, "step": 4064 }, { "epoch": 0.9375135135135135, "grad_norm": 28.617671966552734, "learning_rate": 5.070482116112356e-06, "loss": 11.496625900268555, "loss_alignment": 0.89306640625, "loss_alignment_w": 0.446533203125, "loss_sft": 0.2728603966534138, "loss_total_v6": 0.7185391038656235, "step": 4065 }, { "epoch": 0.9377441441441442, "grad_norm": 23.967897415161133, "learning_rate": 5.068468469099605e-06, "loss": 11.521501541137695, "loss_alignment": 0.87646484375, "loss_alignment_w": 0.438232421875, "loss_sft": 0.2811747957020998, "loss_total_v6": 0.720093846321106, "step": 4066 }, { "epoch": 0.9379747747747748, "grad_norm": 26.29917335510254, "learning_rate": 5.066454810979741e-06, "loss": 11.174161911010742, "loss_alignment": 0.87646484375, "loss_alignment_w": 0.438232421875, "loss_sft": 0.26000015065073967, "loss_total_v6": 0.6983851417899132, "step": 4067 }, { "epoch": 0.9382054054054054, "grad_norm": 31.618404388427734, "learning_rate": 5.064441142079425e-06, "loss": 11.57099437713623, "loss_alignment": 0.83935546875, "loss_alignment_w": 0.419677734375, "loss_sft": 0.3033415749669075, "loss_total_v6": 0.7231871709227562, "step": 4068 }, { "epoch": 0.938436036036036, "grad_norm": 26.16530990600586, "learning_rate": 5.06242746272532e-06, "loss": 11.667076110839844, "loss_alignment": 0.884765625, "loss_alignment_w": 0.4423828125, "loss_sft": 0.2868552450090647, "loss_total_v6": 0.7291922643780708, "step": 4069 }, { "epoch": 0.9386666666666666, "grad_norm": 27.362897872924805, "learning_rate": 5.0604137732440875e-06, "loss": 11.01079273223877, "loss_alignment": 0.8447265625, "loss_alignment_w": 0.42236328125, "loss_sft": 0.2661316879093647, "loss_total_v6": 0.6881745308637619, "step": 4070 }, { "epoch": 0.9388972972972973, "grad_norm": 26.89129638671875, "learning_rate": 5.058400073962395e-06, "loss": 11.270975112915039, "loss_alignment": 0.857421875, "loss_alignment_w": 0.4287109375, "loss_sft": 0.2759843226522207, "loss_total_v6": 0.7044358924031258, "step": 4071 }, { "epoch": 0.9391279279279279, "grad_norm": 26.025375366210938, "learning_rate": 5.056386365206908e-06, "loss": 11.723329544067383, "loss_alignment": 0.8642578125, "loss_alignment_w": 0.43212890625, "loss_sft": 0.30044189281761646, "loss_total_v6": 0.732708141207695, "step": 4072 }, { "epoch": 0.9393585585585585, "grad_norm": 32.325660705566406, "learning_rate": 5.054372647304297e-06, "loss": 11.142455101013184, "loss_alignment": 0.84814453125, "loss_alignment_w": 0.424072265625, "loss_sft": 0.27295677922666073, "loss_total_v6": 0.696403443813324, "step": 4073 }, { "epoch": 0.9395891891891892, "grad_norm": 26.262311935424805, "learning_rate": 5.05235892058123e-06, "loss": 11.143463134765625, "loss_alignment": 0.84912109375, "loss_alignment_w": 0.424560546875, "loss_sft": 0.2710971534252167, "loss_total_v6": 0.696466401219368, "step": 4074 }, { "epoch": 0.9398198198198198, "grad_norm": 23.69660758972168, "learning_rate": 5.050345185364378e-06, "loss": 10.750125885009766, "loss_alignment": 0.8681640625, "loss_alignment_w": 0.43408203125, "loss_sft": 0.23726682178676128, "loss_total_v6": 0.6718828901648521, "step": 4075 }, { "epoch": 0.9400504504504504, "grad_norm": 26.288944244384766, "learning_rate": 5.0483314419804165e-06, "loss": 11.626986503601074, "loss_alignment": 0.87353515625, "loss_alignment_w": 0.436767578125, "loss_sft": 0.29030052572488785, "loss_total_v6": 0.7266866341233253, "step": 4076 }, { "epoch": 0.9402810810810811, "grad_norm": 27.86788558959961, "learning_rate": 5.046317690756014e-06, "loss": 11.601951599121094, "loss_alignment": 0.85107421875, "loss_alignment_w": 0.425537109375, "loss_sft": 0.2995848748832941, "loss_total_v6": 0.7251219898462296, "step": 4077 }, { "epoch": 0.9405117117117117, "grad_norm": 26.46165657043457, "learning_rate": 5.044303932017853e-06, "loss": 10.989819526672363, "loss_alignment": 0.875, "loss_alignment_w": 0.4375, "loss_sft": 0.2477920688688755, "loss_total_v6": 0.6868637353181839, "step": 4078 }, { "epoch": 0.9407423423423423, "grad_norm": 26.78510093688965, "learning_rate": 5.042290166092607e-06, "loss": 11.10142707824707, "loss_alignment": 0.89306640625, "loss_alignment_w": 0.446533203125, "loss_sft": 0.24697029404342175, "loss_total_v6": 0.6938392147421837, "step": 4079 }, { "epoch": 0.940972972972973, "grad_norm": 27.39710235595703, "learning_rate": 5.0402763933069496e-06, "loss": 11.723441123962402, "loss_alignment": 0.87353515625, "loss_alignment_w": 0.436767578125, "loss_sft": 0.29603904299438, "loss_total_v6": 0.7327150702476501, "step": 4080 }, { "epoch": 0.9412036036036036, "grad_norm": 25.490018844604492, "learning_rate": 5.038262613987567e-06, "loss": 11.137256622314453, "loss_alignment": 0.87939453125, "loss_alignment_w": 0.439697265625, "loss_sft": 0.2566711772233248, "loss_total_v6": 0.6960785388946533, "step": 4081 }, { "epoch": 0.9414342342342342, "grad_norm": 36.687679290771484, "learning_rate": 5.036248828461133e-06, "loss": 11.440347671508789, "loss_alignment": 0.880859375, "loss_alignment_w": 0.4404296875, "loss_sft": 0.27431738935410976, "loss_total_v6": 0.7150217592716217, "step": 4082 }, { "epoch": 0.9416648648648649, "grad_norm": 25.86151885986328, "learning_rate": 5.034235037054333e-06, "loss": 11.492715835571289, "loss_alignment": 0.88037109375, "loss_alignment_w": 0.440185546875, "loss_sft": 0.277712507173419, "loss_total_v6": 0.7182947546243668, "step": 4083 }, { "epoch": 0.9418954954954954, "grad_norm": 24.735349655151367, "learning_rate": 5.032221240093846e-06, "loss": 11.228805541992188, "loss_alignment": 0.8642578125, "loss_alignment_w": 0.43212890625, "loss_sft": 0.2700986657291651, "loss_total_v6": 0.7018003389239311, "step": 4084 }, { "epoch": 0.9421261261261261, "grad_norm": 30.98955535888672, "learning_rate": 5.030207437906354e-06, "loss": 11.652868270874023, "loss_alignment": 0.896484375, "loss_alignment_w": 0.4482421875, "loss_sft": 0.278231056407094, "loss_total_v6": 0.7283042892813683, "step": 4085 }, { "epoch": 0.9423567567567568, "grad_norm": 28.572105407714844, "learning_rate": 5.028193630818544e-06, "loss": 11.465812683105469, "loss_alignment": 0.87890625, "loss_alignment_w": 0.439453125, "loss_sft": 0.27795356698334217, "loss_total_v6": 0.7166132628917694, "step": 4086 }, { "epoch": 0.9425873873873873, "grad_norm": 25.908267974853516, "learning_rate": 5.0261798191570975e-06, "loss": 11.041463851928711, "loss_alignment": 0.880859375, "loss_alignment_w": 0.4404296875, "loss_sft": 0.25085199251770973, "loss_total_v6": 0.690091498196125, "step": 4087 }, { "epoch": 0.942818018018018, "grad_norm": 25.23584747314453, "learning_rate": 5.024166003248703e-06, "loss": 11.146730422973633, "loss_alignment": 0.85595703125, "loss_alignment_w": 0.427978515625, "loss_sft": 0.26797495409846306, "loss_total_v6": 0.6966706439852715, "step": 4088 }, { "epoch": 0.9430486486486487, "grad_norm": 26.10209083557129, "learning_rate": 5.022152183420042e-06, "loss": 10.878459930419922, "loss_alignment": 0.85546875, "loss_alignment_w": 0.427734375, "loss_sft": 0.25290183536708355, "loss_total_v6": 0.6799037456512451, "step": 4089 }, { "epoch": 0.9432792792792792, "grad_norm": 27.21964454650879, "learning_rate": 5.020138359997803e-06, "loss": 12.151606559753418, "loss_alignment": 0.9150390625, "loss_alignment_w": 0.45751953125, "loss_sft": 0.3012386690825224, "loss_total_v6": 0.759475402534008, "step": 4090 }, { "epoch": 0.9435099099099099, "grad_norm": 26.89110565185547, "learning_rate": 5.018124533308675e-06, "loss": 11.626238822937012, "loss_alignment": 0.86572265625, "loss_alignment_w": 0.432861328125, "loss_sft": 0.2937328089028597, "loss_total_v6": 0.726639948785305, "step": 4091 }, { "epoch": 0.9437405405405406, "grad_norm": 25.287500381469727, "learning_rate": 5.016110703679341e-06, "loss": 11.224525451660156, "loss_alignment": 0.86181640625, "loss_alignment_w": 0.430908203125, "loss_sft": 0.27079255133867264, "loss_total_v6": 0.7015329077839851, "step": 4092 }, { "epoch": 0.9439711711711711, "grad_norm": 25.388195037841797, "learning_rate": 5.014096871436492e-06, "loss": 10.960771560668945, "loss_alignment": 0.85986328125, "loss_alignment_w": 0.429931640625, "loss_sft": 0.25478092208504677, "loss_total_v6": 0.6850482746958733, "step": 4093 }, { "epoch": 0.9442018018018018, "grad_norm": 25.515527725219727, "learning_rate": 5.012083036906817e-06, "loss": 11.371506690979004, "loss_alignment": 0.8681640625, "loss_alignment_w": 0.43408203125, "loss_sft": 0.2763320077210665, "loss_total_v6": 0.7107191905379295, "step": 4094 }, { "epoch": 0.9444324324324325, "grad_norm": 28.17436408996582, "learning_rate": 5.010069200417006e-06, "loss": 11.07203483581543, "loss_alignment": 0.8515625, "loss_alignment_w": 0.42578125, "loss_sft": 0.26545803248882294, "loss_total_v6": 0.6920021995902061, "step": 4095 }, { "epoch": 0.944663063063063, "grad_norm": 25.665924072265625, "learning_rate": 5.0080553622937434e-06, "loss": 11.688965797424316, "loss_alignment": 0.8603515625, "loss_alignment_w": 0.43017578125, "loss_sft": 0.2996674496680498, "loss_total_v6": 0.730560377240181, "step": 4096 }, { "epoch": 0.9448936936936937, "grad_norm": 28.764278411865234, "learning_rate": 5.0060415228637225e-06, "loss": 11.331807136535645, "loss_alignment": 0.84619140625, "loss_alignment_w": 0.423095703125, "loss_sft": 0.28581363148987293, "loss_total_v6": 0.708237960934639, "step": 4097 }, { "epoch": 0.9451243243243244, "grad_norm": 27.619041442871094, "learning_rate": 5.004027682453633e-06, "loss": 10.856180191040039, "loss_alignment": 0.845703125, "loss_alignment_w": 0.4228515625, "loss_sft": 0.25408802181482315, "loss_total_v6": 0.6785112544894218, "step": 4098 }, { "epoch": 0.9453549549549549, "grad_norm": 27.80766487121582, "learning_rate": 5.002013841390161e-06, "loss": 11.071233749389648, "loss_alignment": 0.85595703125, "loss_alignment_w": 0.427978515625, "loss_sft": 0.26284446381032467, "loss_total_v6": 0.6919521316885948, "step": 4099 }, { "epoch": 0.9455855855855856, "grad_norm": 26.55846405029297, "learning_rate": 5e-06, "loss": 11.576108932495117, "loss_alignment": 0.86669921875, "loss_alignment_w": 0.433349609375, "loss_sft": 0.2902487702667713, "loss_total_v6": 0.7235068306326866, "step": 4100 }, { "epoch": 0.9458162162162163, "grad_norm": 26.757537841796875, "learning_rate": 4.997986158609841e-06, "loss": 11.431994438171387, "loss_alignment": 0.8447265625, "loss_alignment_w": 0.42236328125, "loss_sft": 0.2909309361129999, "loss_total_v6": 0.7144996598362923, "step": 4101 }, { "epoch": 0.9460468468468468, "grad_norm": 27.639389038085938, "learning_rate": 4.9959723175463705e-06, "loss": 11.25886344909668, "loss_alignment": 0.8623046875, "loss_alignment_w": 0.43115234375, "loss_sft": 0.27321328967809677, "loss_total_v6": 0.7036789730191231, "step": 4102 }, { "epoch": 0.9462774774774775, "grad_norm": 26.856304168701172, "learning_rate": 4.993958477136278e-06, "loss": 11.867387771606445, "loss_alignment": 0.88916015625, "loss_alignment_w": 0.444580078125, "loss_sft": 0.2968875356018543, "loss_total_v6": 0.7417117580771446, "step": 4103 }, { "epoch": 0.9465081081081081, "grad_norm": 26.362537384033203, "learning_rate": 4.991944637706258e-06, "loss": 10.901179313659668, "loss_alignment": 0.84228515625, "loss_alignment_w": 0.421142578125, "loss_sft": 0.26001328602433205, "loss_total_v6": 0.6813237071037292, "step": 4104 }, { "epoch": 0.9467387387387387, "grad_norm": 29.52037239074707, "learning_rate": 4.989930799582996e-06, "loss": 11.343764305114746, "loss_alignment": 0.91357421875, "loss_alignment_w": 0.456787109375, "loss_sft": 0.25190827064216137, "loss_total_v6": 0.7089852839708328, "step": 4105 }, { "epoch": 0.9469693693693694, "grad_norm": 35.24380874633789, "learning_rate": 4.987916963093184e-06, "loss": 11.419516563415527, "loss_alignment": 0.87255859375, "loss_alignment_w": 0.436279296875, "loss_sft": 0.27860015258193016, "loss_total_v6": 0.7137197926640511, "step": 4106 }, { "epoch": 0.9472, "grad_norm": 29.195642471313477, "learning_rate": 4.985903128563509e-06, "loss": 12.190298080444336, "loss_alignment": 0.88623046875, "loss_alignment_w": 0.443115234375, "loss_sft": 0.319449782371521, "loss_total_v6": 0.7618936449289322, "step": 4107 }, { "epoch": 0.9474306306306306, "grad_norm": 26.756864547729492, "learning_rate": 4.98388929632066e-06, "loss": 11.196035385131836, "loss_alignment": 0.85302734375, "loss_alignment_w": 0.426513671875, "loss_sft": 0.2742608916014433, "loss_total_v6": 0.6997522041201591, "step": 4108 }, { "epoch": 0.9476612612612613, "grad_norm": 25.17734718322754, "learning_rate": 4.981875466691327e-06, "loss": 11.2492094039917, "loss_alignment": 0.86181640625, "loss_alignment_w": 0.430908203125, "loss_sft": 0.27065681107342243, "loss_total_v6": 0.7030755952000618, "step": 4109 }, { "epoch": 0.9478918918918919, "grad_norm": 27.517032623291016, "learning_rate": 4.979861640002198e-06, "loss": 11.82027816772461, "loss_alignment": 0.8984375, "loss_alignment_w": 0.44921875, "loss_sft": 0.2906167097389698, "loss_total_v6": 0.7387673407793045, "step": 4110 }, { "epoch": 0.9481225225225225, "grad_norm": 29.710567474365234, "learning_rate": 4.977847816579959e-06, "loss": 11.45566463470459, "loss_alignment": 0.8896484375, "loss_alignment_w": 0.44482421875, "loss_sft": 0.27097171545028687, "loss_total_v6": 0.7159790471196175, "step": 4111 }, { "epoch": 0.9483531531531532, "grad_norm": 27.754653930664062, "learning_rate": 4.9758339967512995e-06, "loss": 11.604342460632324, "loss_alignment": 0.8798828125, "loss_alignment_w": 0.43994140625, "loss_sft": 0.28574196994304657, "loss_total_v6": 0.7252714037895203, "step": 4112 }, { "epoch": 0.9485837837837838, "grad_norm": 27.166215896606445, "learning_rate": 4.9738201808429025e-06, "loss": 11.305252075195312, "loss_alignment": 0.88232421875, "loss_alignment_w": 0.441162109375, "loss_sft": 0.2657976485788822, "loss_total_v6": 0.7065782845020294, "step": 4113 }, { "epoch": 0.9488144144144144, "grad_norm": 25.69838523864746, "learning_rate": 4.9718063691814574e-06, "loss": 11.798083305358887, "loss_alignment": 0.8701171875, "loss_alignment_w": 0.43505859375, "loss_sft": 0.3023674339056015, "loss_total_v6": 0.7373802363872528, "step": 4114 }, { "epoch": 0.949045045045045, "grad_norm": 28.122892379760742, "learning_rate": 4.9697925620936464e-06, "loss": 11.03524398803711, "loss_alignment": 0.861328125, "loss_alignment_w": 0.4306640625, "loss_sft": 0.2597253564745188, "loss_total_v6": 0.6897027716040611, "step": 4115 }, { "epoch": 0.9492756756756757, "grad_norm": 28.636322021484375, "learning_rate": 4.9677787599061565e-06, "loss": 11.514252662658691, "loss_alignment": 0.8935546875, "loss_alignment_w": 0.44677734375, "loss_sft": 0.2738247737288475, "loss_total_v6": 0.7196408063173294, "step": 4116 }, { "epoch": 0.9495063063063063, "grad_norm": 25.640459060668945, "learning_rate": 4.965764962945669e-06, "loss": 11.693609237670898, "loss_alignment": 0.8857421875, "loss_alignment_w": 0.44287109375, "loss_sft": 0.28755221888422966, "loss_total_v6": 0.7308505550026894, "step": 4117 }, { "epoch": 0.949736936936937, "grad_norm": 27.289886474609375, "learning_rate": 4.963751171538866e-06, "loss": 11.930980682373047, "loss_alignment": 0.86572265625, "loss_alignment_w": 0.432861328125, "loss_sft": 0.3124739732593298, "loss_total_v6": 0.7456862479448318, "step": 4118 }, { "epoch": 0.9499675675675676, "grad_norm": 26.21134376525879, "learning_rate": 4.961737386012435e-06, "loss": 11.598427772521973, "loss_alignment": 0.8525390625, "loss_alignment_w": 0.42626953125, "loss_sft": 0.2989679090678692, "loss_total_v6": 0.7249017432332039, "step": 4119 }, { "epoch": 0.9501981981981982, "grad_norm": 26.559656143188477, "learning_rate": 4.959723606693051e-06, "loss": 11.35623550415039, "loss_alignment": 0.8515625, "loss_alignment_w": 0.42578125, "loss_sft": 0.28541779331862926, "loss_total_v6": 0.70976472645998, "step": 4120 }, { "epoch": 0.9504288288288288, "grad_norm": 28.785194396972656, "learning_rate": 4.957709833907396e-06, "loss": 11.192641258239746, "loss_alignment": 0.87158203125, "loss_alignment_w": 0.435791015625, "loss_sft": 0.26306239888072014, "loss_total_v6": 0.6995400786399841, "step": 4121 }, { "epoch": 0.9506594594594595, "grad_norm": 26.713417053222656, "learning_rate": 4.955696067982149e-06, "loss": 11.624658584594727, "loss_alignment": 0.8623046875, "loss_alignment_w": 0.43115234375, "loss_sft": 0.29592292942106724, "loss_total_v6": 0.7265411913394928, "step": 4122 }, { "epoch": 0.9508900900900901, "grad_norm": 27.449203491210938, "learning_rate": 4.953682309243986e-06, "loss": 11.158307075500488, "loss_alignment": 0.85693359375, "loss_alignment_w": 0.428466796875, "loss_sft": 0.2687290143221617, "loss_total_v6": 0.6973941922187805, "step": 4123 }, { "epoch": 0.9511207207207207, "grad_norm": 29.45429801940918, "learning_rate": 4.951668558019586e-06, "loss": 11.592583656311035, "loss_alignment": 0.8564453125, "loss_alignment_w": 0.42822265625, "loss_sft": 0.29719884134829044, "loss_total_v6": 0.7245364859700203, "step": 4124 }, { "epoch": 0.9513513513513514, "grad_norm": 24.96506118774414, "learning_rate": 4.949654814635623e-06, "loss": 11.537551879882812, "loss_alignment": 0.88134765625, "loss_alignment_w": 0.440673828125, "loss_sft": 0.2804231494665146, "loss_total_v6": 0.721097007393837, "step": 4125 }, { "epoch": 0.951581981981982, "grad_norm": 26.85976219177246, "learning_rate": 4.9476410794187726e-06, "loss": 11.52007007598877, "loss_alignment": 0.85986328125, "loss_alignment_w": 0.429931640625, "loss_sft": 0.2886383831501007, "loss_total_v6": 0.7200043648481369, "step": 4126 }, { "epoch": 0.9518126126126126, "grad_norm": 28.02090835571289, "learning_rate": 4.945627352695704e-06, "loss": 11.729191780090332, "loss_alignment": 0.84423828125, "loss_alignment_w": 0.422119140625, "loss_sft": 0.3109248522669077, "loss_total_v6": 0.7330745086073875, "step": 4127 }, { "epoch": 0.9520432432432433, "grad_norm": 25.248117446899414, "learning_rate": 4.943613634793092e-06, "loss": 11.377914428710938, "loss_alignment": 0.86865234375, "loss_alignment_w": 0.434326171875, "loss_sft": 0.27571011520922184, "loss_total_v6": 0.7111196666955948, "step": 4128 }, { "epoch": 0.9522738738738739, "grad_norm": 30.152311325073242, "learning_rate": 4.941599926037606e-06, "loss": 11.855354309082031, "loss_alignment": 0.85546875, "loss_alignment_w": 0.427734375, "loss_sft": 0.31159253790974617, "loss_total_v6": 0.7409595996141434, "step": 4129 }, { "epoch": 0.9525045045045045, "grad_norm": 30.96675682067871, "learning_rate": 4.939586226755913e-06, "loss": 11.132011413574219, "loss_alignment": 0.845703125, "loss_alignment_w": 0.4228515625, "loss_sft": 0.2729297149926424, "loss_total_v6": 0.6957507580518723, "step": 4130 }, { "epoch": 0.9527351351351351, "grad_norm": 26.85824203491211, "learning_rate": 4.937572537274683e-06, "loss": 11.628812789916992, "loss_alignment": 0.87451171875, "loss_alignment_w": 0.437255859375, "loss_sft": 0.28849209658801556, "loss_total_v6": 0.7268008142709732, "step": 4131 }, { "epoch": 0.9529657657657657, "grad_norm": 39.55439376831055, "learning_rate": 4.935558857920576e-06, "loss": 11.521389961242676, "loss_alignment": 0.88671875, "loss_alignment_w": 0.443359375, "loss_sft": 0.27642231062054634, "loss_total_v6": 0.7200868651270866, "step": 4132 }, { "epoch": 0.9531963963963964, "grad_norm": 27.26169776916504, "learning_rate": 4.933545189020261e-06, "loss": 11.87484073638916, "loss_alignment": 0.853515625, "loss_alignment_w": 0.4267578125, "loss_sft": 0.315969068557024, "loss_total_v6": 0.7421775385737419, "step": 4133 }, { "epoch": 0.953427027027027, "grad_norm": 24.35125732421875, "learning_rate": 4.931531530900396e-06, "loss": 11.409996032714844, "loss_alignment": 0.8564453125, "loss_alignment_w": 0.42822265625, "loss_sft": 0.28516145795583725, "loss_total_v6": 0.7131247445940971, "step": 4134 }, { "epoch": 0.9536576576576576, "grad_norm": 24.122819900512695, "learning_rate": 4.929517883887645e-06, "loss": 11.087942123413086, "loss_alignment": 0.85888671875, "loss_alignment_w": 0.429443359375, "loss_sft": 0.263278316706419, "loss_total_v6": 0.6929963380098343, "step": 4135 }, { "epoch": 0.9538882882882883, "grad_norm": 27.346933364868164, "learning_rate": 4.927504248308663e-06, "loss": 11.416399955749512, "loss_alignment": 0.88916015625, "loss_alignment_w": 0.444580078125, "loss_sft": 0.2691432423889637, "loss_total_v6": 0.7135249599814415, "step": 4136 }, { "epoch": 0.9541189189189189, "grad_norm": 26.88717269897461, "learning_rate": 4.925490624490107e-06, "loss": 11.30385971069336, "loss_alignment": 0.861328125, "loss_alignment_w": 0.4306640625, "loss_sft": 0.2770631667226553, "loss_total_v6": 0.7064912542700768, "step": 4137 }, { "epoch": 0.9543495495495495, "grad_norm": 26.103044509887695, "learning_rate": 4.923477012758635e-06, "loss": 11.501193046569824, "loss_alignment": 0.8828125, "loss_alignment_w": 0.44140625, "loss_sft": 0.27747935988008976, "loss_total_v6": 0.7188245728611946, "step": 4138 }, { "epoch": 0.9545801801801802, "grad_norm": 26.71021270751953, "learning_rate": 4.921463413440898e-06, "loss": 11.739724159240723, "loss_alignment": 0.88037109375, "loss_alignment_w": 0.440185546875, "loss_sft": 0.29357774555683136, "loss_total_v6": 0.7337327525019646, "step": 4139 }, { "epoch": 0.9548108108108108, "grad_norm": 27.935319900512695, "learning_rate": 4.919449826863544e-06, "loss": 11.026872634887695, "loss_alignment": 0.85498046875, "loss_alignment_w": 0.427490234375, "loss_sft": 0.2626353446394205, "loss_total_v6": 0.6891795322299004, "step": 4140 }, { "epoch": 0.9550414414414414, "grad_norm": 27.617259979248047, "learning_rate": 4.917436253353225e-06, "loss": 11.22392463684082, "loss_alignment": 0.8515625, "loss_alignment_w": 0.42578125, "loss_sft": 0.2769652493298054, "loss_total_v6": 0.7014952674508095, "step": 4141 }, { "epoch": 0.9552720720720721, "grad_norm": 28.010515213012695, "learning_rate": 4.915422693236584e-06, "loss": 11.615198135375977, "loss_alignment": 0.88427734375, "loss_alignment_w": 0.442138671875, "loss_sft": 0.2847877237945795, "loss_total_v6": 0.7259498313069344, "step": 4142 }, { "epoch": 0.9555027027027027, "grad_norm": 28.475433349609375, "learning_rate": 4.9134091468402696e-06, "loss": 10.944931030273438, "loss_alignment": 0.869140625, "loss_alignment_w": 0.4345703125, "loss_sft": 0.25014403834939003, "loss_total_v6": 0.6840582266449928, "step": 4143 }, { "epoch": 0.9557333333333333, "grad_norm": 24.188770294189453, "learning_rate": 4.911395614490922e-06, "loss": 11.407150268554688, "loss_alignment": 0.85986328125, "loss_alignment_w": 0.429931640625, "loss_sft": 0.282938938587904, "loss_total_v6": 0.7129468619823456, "step": 4144 }, { "epoch": 0.955963963963964, "grad_norm": 25.975622177124023, "learning_rate": 4.909382096515182e-06, "loss": 11.509714126586914, "loss_alignment": 0.88232421875, "loss_alignment_w": 0.441162109375, "loss_sft": 0.27819507010281086, "loss_total_v6": 0.7193571701645851, "step": 4145 }, { "epoch": 0.9561945945945945, "grad_norm": 28.013412475585938, "learning_rate": 4.907368593239684e-06, "loss": 11.582015037536621, "loss_alignment": 0.87451171875, "loss_alignment_w": 0.437255859375, "loss_sft": 0.2875813916325569, "loss_total_v6": 0.7238759547472, "step": 4146 }, { "epoch": 0.9564252252252252, "grad_norm": 26.366931915283203, "learning_rate": 4.905355104991064e-06, "loss": 10.912906646728516, "loss_alignment": 0.8525390625, "loss_alignment_w": 0.42626953125, "loss_sft": 0.2553903963416815, "loss_total_v6": 0.6820566579699516, "step": 4147 }, { "epoch": 0.9566558558558559, "grad_norm": 27.24469566345215, "learning_rate": 4.903341632095958e-06, "loss": 11.551689147949219, "loss_alignment": 0.873046875, "loss_alignment_w": 0.4365234375, "loss_sft": 0.28565542213618755, "loss_total_v6": 0.721980519592762, "step": 4148 }, { "epoch": 0.9568864864864864, "grad_norm": 27.594911575317383, "learning_rate": 4.901328174880994e-06, "loss": 11.45477294921875, "loss_alignment": 0.87255859375, "loss_alignment_w": 0.436279296875, "loss_sft": 0.279811792075634, "loss_total_v6": 0.7159232646226883, "step": 4149 }, { "epoch": 0.9571171171171171, "grad_norm": 30.806114196777344, "learning_rate": 4.8993147336728e-06, "loss": 11.415594100952148, "loss_alignment": 0.8779296875, "loss_alignment_w": 0.43896484375, "loss_sft": 0.2759440392255783, "loss_total_v6": 0.7134745940566063, "step": 4150 }, { "epoch": 0.9573477477477478, "grad_norm": 27.681116104125977, "learning_rate": 4.8973013087980005e-06, "loss": 11.530458450317383, "loss_alignment": 0.8896484375, "loss_alignment_w": 0.44482421875, "loss_sft": 0.27572261542081833, "loss_total_v6": 0.720653660595417, "step": 4151 }, { "epoch": 0.9575783783783783, "grad_norm": 25.94159698486328, "learning_rate": 4.895287900583216e-06, "loss": 10.311368942260742, "loss_alignment": 0.869140625, "loss_alignment_w": 0.4345703125, "loss_sft": 0.2106074020266533, "loss_total_v6": 0.6444605737924576, "step": 4152 }, { "epoch": 0.957809009009009, "grad_norm": 25.403522491455078, "learning_rate": 4.893274509355069e-06, "loss": 11.683329582214355, "loss_alignment": 0.88427734375, "loss_alignment_w": 0.442138671875, "loss_sft": 0.2878557834774256, "loss_total_v6": 0.7302080914378166, "step": 4153 }, { "epoch": 0.9580396396396397, "grad_norm": 27.256967544555664, "learning_rate": 4.891261135440177e-06, "loss": 11.608306884765625, "loss_alignment": 0.8466796875, "loss_alignment_w": 0.42333984375, "loss_sft": 0.30109598487615585, "loss_total_v6": 0.7255192026495934, "step": 4154 }, { "epoch": 0.9582702702702702, "grad_norm": 27.599346160888672, "learning_rate": 4.889247779165152e-06, "loss": 11.422069549560547, "loss_alignment": 0.8828125, "loss_alignment_w": 0.44140625, "loss_sft": 0.2733276039361954, "loss_total_v6": 0.713879369199276, "step": 4155 }, { "epoch": 0.9585009009009009, "grad_norm": 27.906436920166016, "learning_rate": 4.887234440856608e-06, "loss": 10.958189964294434, "loss_alignment": 0.85595703125, "loss_alignment_w": 0.427978515625, "loss_sft": 0.25709144957363605, "loss_total_v6": 0.6848868802189827, "step": 4156 }, { "epoch": 0.9587315315315316, "grad_norm": 28.2574462890625, "learning_rate": 4.885221120841148e-06, "loss": 11.85693073272705, "loss_alignment": 0.88134765625, "loss_alignment_w": 0.440673828125, "loss_sft": 0.3014982305467129, "loss_total_v6": 0.7410581484436989, "step": 4157 }, { "epoch": 0.9589621621621621, "grad_norm": 27.12353515625, "learning_rate": 4.883207819445385e-06, "loss": 11.069660186767578, "loss_alignment": 0.84765625, "loss_alignment_w": 0.423828125, "loss_sft": 0.2672169376164675, "loss_total_v6": 0.691853754222393, "step": 4158 }, { "epoch": 0.9591927927927928, "grad_norm": 24.416378021240234, "learning_rate": 4.881194536995915e-06, "loss": 11.25706672668457, "loss_alignment": 0.83251953125, "loss_alignment_w": 0.416259765625, "loss_sft": 0.2884512897580862, "loss_total_v6": 0.7035666406154633, "step": 4159 }, { "epoch": 0.9594234234234235, "grad_norm": 26.54612159729004, "learning_rate": 4.87918127381934e-06, "loss": 11.3138427734375, "loss_alignment": 0.85498046875, "loss_alignment_w": 0.427490234375, "loss_sft": 0.2810135055333376, "loss_total_v6": 0.7071151956915855, "step": 4160 }, { "epoch": 0.959654054054054, "grad_norm": 25.959692001342773, "learning_rate": 4.877168030242256e-06, "loss": 11.27692985534668, "loss_alignment": 0.876953125, "loss_alignment_w": 0.4384765625, "loss_sft": 0.26672833040356636, "loss_total_v6": 0.7048081606626511, "step": 4161 }, { "epoch": 0.9598846846846847, "grad_norm": 28.059326171875, "learning_rate": 4.875154806591253e-06, "loss": 11.612285614013672, "loss_alignment": 0.84619140625, "loss_alignment_w": 0.423095703125, "loss_sft": 0.30210759304463863, "loss_total_v6": 0.7257678732275963, "step": 4162 }, { "epoch": 0.9601153153153154, "grad_norm": 28.166088104248047, "learning_rate": 4.873141603192926e-06, "loss": 10.900317192077637, "loss_alignment": 0.8603515625, "loss_alignment_w": 0.43017578125, "loss_sft": 0.25175014697015285, "loss_total_v6": 0.6812698021531105, "step": 4163 }, { "epoch": 0.9603459459459459, "grad_norm": 33.52153778076172, "learning_rate": 4.871128420373859e-06, "loss": 11.493584632873535, "loss_alignment": 0.84375, "loss_alignment_w": 0.421875, "loss_sft": 0.2963061761111021, "loss_total_v6": 0.7183490470051765, "step": 4164 }, { "epoch": 0.9605765765765766, "grad_norm": 26.63762664794922, "learning_rate": 4.869115258460636e-06, "loss": 11.299182891845703, "loss_alignment": 0.85888671875, "loss_alignment_w": 0.429443359375, "loss_sft": 0.2766182217746973, "loss_total_v6": 0.7061989009380341, "step": 4165 }, { "epoch": 0.9608072072072072, "grad_norm": 25.706584930419922, "learning_rate": 4.867102117779835e-06, "loss": 11.558683395385742, "loss_alignment": 0.8701171875, "loss_alignment_w": 0.43505859375, "loss_sft": 0.2880915515124798, "loss_total_v6": 0.7224177420139313, "step": 4166 }, { "epoch": 0.9610378378378378, "grad_norm": 25.65275001525879, "learning_rate": 4.865088998658031e-06, "loss": 10.97189712524414, "loss_alignment": 0.86181640625, "loss_alignment_w": 0.430908203125, "loss_sft": 0.2546675596386194, "loss_total_v6": 0.6857435926795006, "step": 4167 }, { "epoch": 0.9612684684684685, "grad_norm": 26.52950668334961, "learning_rate": 4.8630759014218e-06, "loss": 11.227472305297852, "loss_alignment": 0.9052734375, "loss_alignment_w": 0.45263671875, "loss_sft": 0.2486530840396881, "loss_total_v6": 0.7017170563340187, "step": 4168 }, { "epoch": 0.9614990990990991, "grad_norm": 25.756132125854492, "learning_rate": 4.861062826397712e-06, "loss": 11.172384262084961, "loss_alignment": 0.8408203125, "loss_alignment_w": 0.42041015625, "loss_sft": 0.27818430960178375, "loss_total_v6": 0.6982740238308907, "step": 4169 }, { "epoch": 0.9617297297297297, "grad_norm": 25.930362701416016, "learning_rate": 4.859049773912329e-06, "loss": 11.48751449584961, "loss_alignment": 0.86376953125, "loss_alignment_w": 0.431884765625, "loss_sft": 0.2868325933814049, "loss_total_v6": 0.7179696932435036, "step": 4170 }, { "epoch": 0.9619603603603604, "grad_norm": 30.77796745300293, "learning_rate": 4.857036744292216e-06, "loss": 11.441829681396484, "loss_alignment": 0.857421875, "loss_alignment_w": 0.4287109375, "loss_sft": 0.286113478243351, "loss_total_v6": 0.7151143103837967, "step": 4171 }, { "epoch": 0.962190990990991, "grad_norm": 29.62744140625, "learning_rate": 4.855023737863927e-06, "loss": 12.314961433410645, "loss_alignment": 0.85400390625, "loss_alignment_w": 0.427001953125, "loss_sft": 0.3428356945514679, "loss_total_v6": 0.7696851044893265, "step": 4172 }, { "epoch": 0.9624216216216216, "grad_norm": 28.39826202392578, "learning_rate": 4.853010754954023e-06, "loss": 11.082498550415039, "loss_alignment": 0.85498046875, "loss_alignment_w": 0.427490234375, "loss_sft": 0.26502860337495804, "loss_total_v6": 0.6926561444997787, "step": 4173 }, { "epoch": 0.9626522522522523, "grad_norm": 25.0953426361084, "learning_rate": 4.85099779588905e-06, "loss": 11.509700775146484, "loss_alignment": 0.876953125, "loss_alignment_w": 0.4384765625, "loss_sft": 0.2793234083801508, "loss_total_v6": 0.7193563431501389, "step": 4174 }, { "epoch": 0.9628828828828829, "grad_norm": 29.815898895263672, "learning_rate": 4.848984860995557e-06, "loss": 11.594011306762695, "loss_alignment": 0.84814453125, "loss_alignment_w": 0.424072265625, "loss_sft": 0.30107229948043823, "loss_total_v6": 0.7246257439255714, "step": 4175 }, { "epoch": 0.9631135135135135, "grad_norm": 24.842288970947266, "learning_rate": 4.846971950600084e-06, "loss": 11.4331693649292, "loss_alignment": 0.89697265625, "loss_alignment_w": 0.448486328125, "loss_sft": 0.2665139827877283, "loss_total_v6": 0.7145730704069138, "step": 4176 }, { "epoch": 0.9633441441441442, "grad_norm": 27.223838806152344, "learning_rate": 4.844959065029173e-06, "loss": 11.31222915649414, "loss_alignment": 0.8486328125, "loss_alignment_w": 0.42431640625, "loss_sft": 0.28207228519022465, "loss_total_v6": 0.7070143073797226, "step": 4177 }, { "epoch": 0.9635747747747748, "grad_norm": 27.294166564941406, "learning_rate": 4.842946204609359e-06, "loss": 11.370407104492188, "loss_alignment": 0.865234375, "loss_alignment_w": 0.4326171875, "loss_sft": 0.2787505052983761, "loss_total_v6": 0.7106504887342453, "step": 4178 }, { "epoch": 0.9638054054054054, "grad_norm": 26.6804141998291, "learning_rate": 4.840933369667171e-06, "loss": 11.537635803222656, "loss_alignment": 0.8779296875, "loss_alignment_w": 0.43896484375, "loss_sft": 0.28338854759931564, "loss_total_v6": 0.7211021780967712, "step": 4179 }, { "epoch": 0.964036036036036, "grad_norm": 35.89686584472656, "learning_rate": 4.838920560529137e-06, "loss": 11.688064575195312, "loss_alignment": 0.86474609375, "loss_alignment_w": 0.432373046875, "loss_sft": 0.2988939341157675, "loss_total_v6": 0.7305040284991264, "step": 4180 }, { "epoch": 0.9642666666666667, "grad_norm": 29.494722366333008, "learning_rate": 4.8369077775217765e-06, "loss": 11.832000732421875, "loss_alignment": 0.86865234375, "loss_alignment_w": 0.434326171875, "loss_sft": 0.30503652803599834, "loss_total_v6": 0.7395000457763672, "step": 4181 }, { "epoch": 0.9644972972972973, "grad_norm": 26.782930374145508, "learning_rate": 4.834895020971613e-06, "loss": 11.592181205749512, "loss_alignment": 0.890625, "loss_alignment_w": 0.4453125, "loss_sft": 0.27898515202105045, "loss_total_v6": 0.7245112881064415, "step": 4182 }, { "epoch": 0.9647279279279279, "grad_norm": 24.937395095825195, "learning_rate": 4.832882291205156e-06, "loss": 11.310590744018555, "loss_alignment": 0.8798828125, "loss_alignment_w": 0.43994140625, "loss_sft": 0.26764193549752235, "loss_total_v6": 0.7069119736552238, "step": 4183 }, { "epoch": 0.9649585585585586, "grad_norm": 28.59354019165039, "learning_rate": 4.830869588548918e-06, "loss": 11.467479705810547, "loss_alignment": 0.890625, "loss_alignment_w": 0.4453125, "loss_sft": 0.2715117484331131, "loss_total_v6": 0.7167174816131592, "step": 4184 }, { "epoch": 0.9651891891891892, "grad_norm": 23.983997344970703, "learning_rate": 4.828856913329403e-06, "loss": 11.269543647766113, "loss_alignment": 0.8515625, "loss_alignment_w": 0.42578125, "loss_sft": 0.27769545279443264, "loss_total_v6": 0.7043464779853821, "step": 4185 }, { "epoch": 0.9654198198198198, "grad_norm": 26.68711280822754, "learning_rate": 4.82684426587311e-06, "loss": 11.621379852294922, "loss_alignment": 0.87646484375, "loss_alignment_w": 0.438232421875, "loss_sft": 0.2882717065513134, "loss_total_v6": 0.7263362631201744, "step": 4186 }, { "epoch": 0.9656504504504505, "grad_norm": 23.413347244262695, "learning_rate": 4.824831646506539e-06, "loss": 11.439567565917969, "loss_alignment": 0.8623046875, "loss_alignment_w": 0.43115234375, "loss_sft": 0.2842020858079195, "loss_total_v6": 0.714972972869873, "step": 4187 }, { "epoch": 0.9658810810810811, "grad_norm": 26.4743709564209, "learning_rate": 4.82281905555618e-06, "loss": 11.573404312133789, "loss_alignment": 0.8837890625, "loss_alignment_w": 0.44189453125, "loss_sft": 0.28129066340625286, "loss_total_v6": 0.7233377546072006, "step": 4188 }, { "epoch": 0.9661117117117117, "grad_norm": 26.010421752929688, "learning_rate": 4.820806493348519e-06, "loss": 11.482789993286133, "loss_alignment": 0.87646484375, "loss_alignment_w": 0.438232421875, "loss_sft": 0.2802811935544014, "loss_total_v6": 0.7176744043827057, "step": 4189 }, { "epoch": 0.9663423423423424, "grad_norm": 41.12852478027344, "learning_rate": 4.81879396021004e-06, "loss": 11.494855880737305, "loss_alignment": 0.85595703125, "loss_alignment_w": 0.427978515625, "loss_sft": 0.290556775406003, "loss_total_v6": 0.7184284776449203, "step": 4190 }, { "epoch": 0.966572972972973, "grad_norm": 29.565595626831055, "learning_rate": 4.816781456467218e-06, "loss": 10.892557144165039, "loss_alignment": 0.86474609375, "loss_alignment_w": 0.432373046875, "loss_sft": 0.2484270315617323, "loss_total_v6": 0.6807848215103149, "step": 4191 }, { "epoch": 0.9668036036036036, "grad_norm": 26.884782791137695, "learning_rate": 4.814768982446532e-06, "loss": 11.061847686767578, "loss_alignment": 0.84033203125, "loss_alignment_w": 0.420166015625, "loss_sft": 0.27086370810866356, "loss_total_v6": 0.6913654282689095, "step": 4192 }, { "epoch": 0.9670342342342342, "grad_norm": 27.65949821472168, "learning_rate": 4.812756538474443e-06, "loss": 11.622798919677734, "loss_alignment": 0.8564453125, "loss_alignment_w": 0.42822265625, "loss_sft": 0.2963864952325821, "loss_total_v6": 0.7264249175786972, "step": 4193 }, { "epoch": 0.9672648648648648, "grad_norm": 26.860944747924805, "learning_rate": 4.81074412487742e-06, "loss": 11.601043701171875, "loss_alignment": 0.84619140625, "loss_alignment_w": 0.423095703125, "loss_sft": 0.30221363343298435, "loss_total_v6": 0.7250652238726616, "step": 4194 }, { "epoch": 0.9674954954954955, "grad_norm": 26.467798233032227, "learning_rate": 4.808731741981917e-06, "loss": 11.23880672454834, "loss_alignment": 0.86328125, "loss_alignment_w": 0.431640625, "loss_sft": 0.2703575864434242, "loss_total_v6": 0.7024254351854324, "step": 4195 }, { "epoch": 0.9677261261261261, "grad_norm": 31.387500762939453, "learning_rate": 4.806719390114389e-06, "loss": 11.765792846679688, "loss_alignment": 0.8525390625, "loss_alignment_w": 0.42626953125, "loss_sft": 0.31020640581846237, "loss_total_v6": 0.7353620603680611, "step": 4196 }, { "epoch": 0.9679567567567567, "grad_norm": 31.00080680847168, "learning_rate": 4.804707069601287e-06, "loss": 11.485761642456055, "loss_alignment": 0.84912109375, "loss_alignment_w": 0.424560546875, "loss_sft": 0.2932537626475096, "loss_total_v6": 0.7178600654006004, "step": 4197 }, { "epoch": 0.9681873873873874, "grad_norm": 28.77654266357422, "learning_rate": 4.802694780769051e-06, "loss": 11.688263893127441, "loss_alignment": 0.86865234375, "loss_alignment_w": 0.434326171875, "loss_sft": 0.2964496724307537, "loss_total_v6": 0.7305164784193039, "step": 4198 }, { "epoch": 0.968418018018018, "grad_norm": 28.77507209777832, "learning_rate": 4.800682523944119e-06, "loss": 11.217336654663086, "loss_alignment": 0.83447265625, "loss_alignment_w": 0.417236328125, "loss_sft": 0.28470165096223354, "loss_total_v6": 0.7010835036635399, "step": 4199 }, { "epoch": 0.9686486486486486, "grad_norm": 27.206069946289062, "learning_rate": 4.798670299452926e-06, "loss": 11.234574317932129, "loss_alignment": 0.85546875, "loss_alignment_w": 0.427734375, "loss_sft": 0.274716405197978, "loss_total_v6": 0.7021608650684357, "step": 4200 }, { "epoch": 0.9688792792792793, "grad_norm": 24.68424415588379, "learning_rate": 4.796658107621897e-06, "loss": 11.274742126464844, "loss_alignment": 0.87841796875, "loss_alignment_w": 0.439208984375, "loss_sft": 0.2658133953809738, "loss_total_v6": 0.7046714052557945, "step": 4201 }, { "epoch": 0.9691099099099099, "grad_norm": 26.66562271118164, "learning_rate": 4.794645948777456e-06, "loss": 11.036623001098633, "loss_alignment": 0.86181640625, "loss_alignment_w": 0.430908203125, "loss_sft": 0.25816356018185616, "loss_total_v6": 0.6897889152169228, "step": 4202 }, { "epoch": 0.9693405405405405, "grad_norm": 25.608415603637695, "learning_rate": 4.792633823246022e-06, "loss": 10.776193618774414, "loss_alignment": 0.8359375, "loss_alignment_w": 0.41796875, "loss_sft": 0.2556043919175863, "loss_total_v6": 0.6735121086239815, "step": 4203 }, { "epoch": 0.9695711711711712, "grad_norm": 24.798593521118164, "learning_rate": 4.7906217313540035e-06, "loss": 11.240253448486328, "loss_alignment": 0.875, "loss_alignment_w": 0.4375, "loss_sft": 0.2653362974524498, "loss_total_v6": 0.7025158479809761, "step": 4204 }, { "epoch": 0.9698018018018018, "grad_norm": 25.26759910583496, "learning_rate": 4.788609673427807e-06, "loss": 10.852258682250977, "loss_alignment": 0.84619140625, "loss_alignment_w": 0.423095703125, "loss_sft": 0.2552772704511881, "loss_total_v6": 0.6782661899924278, "step": 4205 }, { "epoch": 0.9700324324324324, "grad_norm": 24.33165168762207, "learning_rate": 4.786597649793833e-06, "loss": 11.084714889526367, "loss_alignment": 0.8642578125, "loss_alignment_w": 0.43212890625, "loss_sft": 0.26075728982686996, "loss_total_v6": 0.6927946582436562, "step": 4206 }, { "epoch": 0.9702630630630631, "grad_norm": 31.67454719543457, "learning_rate": 4.78458566077848e-06, "loss": 11.017436981201172, "loss_alignment": 0.85693359375, "loss_alignment_w": 0.428466796875, "loss_sft": 0.2602909132838249, "loss_total_v6": 0.6885898485779762, "step": 4207 }, { "epoch": 0.9704936936936936, "grad_norm": 26.472293853759766, "learning_rate": 4.782573706708133e-06, "loss": 11.080211639404297, "loss_alignment": 0.8603515625, "loss_alignment_w": 0.43017578125, "loss_sft": 0.2630851473659277, "loss_total_v6": 0.6925132498145103, "step": 4208 }, { "epoch": 0.9707243243243243, "grad_norm": 28.99308967590332, "learning_rate": 4.780561787909179e-06, "loss": 11.024395942687988, "loss_alignment": 0.86328125, "loss_alignment_w": 0.431640625, "loss_sft": 0.25794872641563416, "loss_total_v6": 0.6890247911214828, "step": 4209 }, { "epoch": 0.970954954954955, "grad_norm": 26.941696166992188, "learning_rate": 4.778549904707996e-06, "loss": 11.185623168945312, "loss_alignment": 0.86376953125, "loss_alignment_w": 0.431884765625, "loss_sft": 0.26575188525021076, "loss_total_v6": 0.6991015002131462, "step": 4210 }, { "epoch": 0.9711855855855855, "grad_norm": 26.169647216796875, "learning_rate": 4.776538057430951e-06, "loss": 11.568216323852539, "loss_alignment": 0.892578125, "loss_alignment_w": 0.4462890625, "loss_sft": 0.2761293426156044, "loss_total_v6": 0.7230135053396225, "step": 4211 }, { "epoch": 0.9714162162162162, "grad_norm": 26.656673431396484, "learning_rate": 4.774526246404417e-06, "loss": 10.977533340454102, "loss_alignment": 0.8779296875, "loss_alignment_w": 0.43896484375, "loss_sft": 0.2476040031760931, "loss_total_v6": 0.6860958561301231, "step": 4212 }, { "epoch": 0.9716468468468469, "grad_norm": 24.654571533203125, "learning_rate": 4.772514471954752e-06, "loss": 10.875484466552734, "loss_alignment": 0.86083984375, "loss_alignment_w": 0.430419921875, "loss_sft": 0.250625379383564, "loss_total_v6": 0.6797178015112877, "step": 4213 }, { "epoch": 0.9718774774774774, "grad_norm": 25.938762664794922, "learning_rate": 4.770502734408311e-06, "loss": 10.934228897094727, "loss_alignment": 0.8583984375, "loss_alignment_w": 0.42919921875, "loss_sft": 0.25434268079698086, "loss_total_v6": 0.683389313519001, "step": 4214 }, { "epoch": 0.9721081081081081, "grad_norm": 26.49209213256836, "learning_rate": 4.768491034091444e-06, "loss": 11.167282104492188, "loss_alignment": 0.86083984375, "loss_alignment_w": 0.430419921875, "loss_sft": 0.2677640300244093, "loss_total_v6": 0.6979550942778587, "step": 4215 }, { "epoch": 0.9723387387387388, "grad_norm": 33.478416442871094, "learning_rate": 4.766479371330488e-06, "loss": 11.626381874084473, "loss_alignment": 0.83544921875, "loss_alignment_w": 0.417724609375, "loss_sft": 0.30915315076708794, "loss_total_v6": 0.7266488745808601, "step": 4216 }, { "epoch": 0.9725693693693693, "grad_norm": 27.09726333618164, "learning_rate": 4.7644677464517874e-06, "loss": 11.376054763793945, "loss_alignment": 0.86767578125, "loss_alignment_w": 0.433837890625, "loss_sft": 0.2751972209662199, "loss_total_v6": 0.7110034748911858, "step": 4217 }, { "epoch": 0.9728, "grad_norm": 28.88778305053711, "learning_rate": 4.762456159781668e-06, "loss": 11.508441925048828, "loss_alignment": 0.83544921875, "loss_alignment_w": 0.417724609375, "loss_sft": 0.30211757868528366, "loss_total_v6": 0.7192776277661324, "step": 4218 }, { "epoch": 0.9730306306306307, "grad_norm": 27.7652645111084, "learning_rate": 4.760444611646455e-06, "loss": 11.252087593078613, "loss_alignment": 0.86669921875, "loss_alignment_w": 0.433349609375, "loss_sft": 0.2708061821758747, "loss_total_v6": 0.7032554969191551, "step": 4219 }, { "epoch": 0.9732612612612612, "grad_norm": 27.232524871826172, "learning_rate": 4.758433102372466e-06, "loss": 11.76495361328125, "loss_alignment": 0.86279296875, "loss_alignment_w": 0.431396484375, "loss_sft": 0.30383685417473316, "loss_total_v6": 0.7353096306324005, "step": 4220 }, { "epoch": 0.9734918918918919, "grad_norm": 29.446544647216797, "learning_rate": 4.756421632286011e-06, "loss": 12.582574844360352, "loss_alignment": 0.8818359375, "loss_alignment_w": 0.44091796875, "loss_sft": 0.3450657371431589, "loss_total_v6": 0.7864109501242638, "step": 4221 }, { "epoch": 0.9737225225225226, "grad_norm": 25.98943328857422, "learning_rate": 4.7544102017134e-06, "loss": 11.82988452911377, "loss_alignment": 0.8701171875, "loss_alignment_w": 0.43505859375, "loss_sft": 0.3044922798871994, "loss_total_v6": 0.73936777561903, "step": 4222 }, { "epoch": 0.9739531531531531, "grad_norm": 27.56830406188965, "learning_rate": 4.752398810980929e-06, "loss": 11.213627815246582, "loss_alignment": 0.8623046875, "loss_alignment_w": 0.43115234375, "loss_sft": 0.27024869062006474, "loss_total_v6": 0.7008517310023308, "step": 4223 }, { "epoch": 0.9741837837837838, "grad_norm": 29.167556762695312, "learning_rate": 4.7503874604148896e-06, "loss": 11.795339584350586, "loss_alignment": 0.88232421875, "loss_alignment_w": 0.441162109375, "loss_sft": 0.2959703505039215, "loss_total_v6": 0.7372087612748146, "step": 4224 }, { "epoch": 0.9744144144144145, "grad_norm": 29.26735496520996, "learning_rate": 4.748376150341566e-06, "loss": 11.027517318725586, "loss_alignment": 0.81298828125, "loss_alignment_w": 0.406494140625, "loss_sft": 0.2822832316160202, "loss_total_v6": 0.6892198547720909, "step": 4225 }, { "epoch": 0.974645045045045, "grad_norm": 26.36391258239746, "learning_rate": 4.746364881087244e-06, "loss": 10.968017578125, "loss_alignment": 0.841796875, "loss_alignment_w": 0.4208984375, "loss_sft": 0.2640991024672985, "loss_total_v6": 0.6855010688304901, "step": 4226 }, { "epoch": 0.9748756756756757, "grad_norm": 28.140159606933594, "learning_rate": 4.744353652978191e-06, "loss": 11.673526763916016, "loss_alignment": 0.84912109375, "loss_alignment_w": 0.424560546875, "loss_sft": 0.30610300600528717, "loss_total_v6": 0.7295954301953316, "step": 4227 }, { "epoch": 0.9751063063063063, "grad_norm": 28.741174697875977, "learning_rate": 4.742342466340675e-06, "loss": 10.69550895690918, "loss_alignment": 0.8310546875, "loss_alignment_w": 0.41552734375, "loss_sft": 0.25266727060079575, "loss_total_v6": 0.6684692725539207, "step": 4228 }, { "epoch": 0.9753369369369369, "grad_norm": 24.360658645629883, "learning_rate": 4.740331321500955e-06, "loss": 11.452095031738281, "loss_alignment": 0.89208984375, "loss_alignment_w": 0.446044921875, "loss_sft": 0.2694669011980295, "loss_total_v6": 0.7157559469342232, "step": 4229 }, { "epoch": 0.9755675675675676, "grad_norm": 31.99755096435547, "learning_rate": 4.738320218785281e-06, "loss": 11.219158172607422, "loss_alignment": 0.841796875, "loss_alignment_w": 0.4208984375, "loss_sft": 0.28026837669312954, "loss_total_v6": 0.7011973410844803, "step": 4230 }, { "epoch": 0.9757981981981982, "grad_norm": 27.631019592285156, "learning_rate": 4.736309158519902e-06, "loss": 11.366559982299805, "loss_alignment": 0.849609375, "loss_alignment_w": 0.4248046875, "loss_sft": 0.2846744582056999, "loss_total_v6": 0.710409939289093, "step": 4231 }, { "epoch": 0.9760288288288288, "grad_norm": 28.394302368164062, "learning_rate": 4.734298141031057e-06, "loss": 11.650199890136719, "loss_alignment": 0.86767578125, "loss_alignment_w": 0.433837890625, "loss_sft": 0.29478786140680313, "loss_total_v6": 0.7281375080347061, "step": 4232 }, { "epoch": 0.9762594594594595, "grad_norm": 26.01399803161621, "learning_rate": 4.732287166644976e-06, "loss": 11.2141752243042, "loss_alignment": 0.84130859375, "loss_alignment_w": 0.420654296875, "loss_sft": 0.28105558827519417, "loss_total_v6": 0.7008859217166901, "step": 4233 }, { "epoch": 0.9764900900900901, "grad_norm": 27.19676399230957, "learning_rate": 4.730276235687885e-06, "loss": 11.926054000854492, "loss_alignment": 0.85693359375, "loss_alignment_w": 0.428466796875, "loss_sft": 0.31784238293766975, "loss_total_v6": 0.7453784048557281, "step": 4234 }, { "epoch": 0.9767207207207207, "grad_norm": 29.041608810424805, "learning_rate": 4.728265348486e-06, "loss": 10.827142715454102, "loss_alignment": 0.83447265625, "loss_alignment_w": 0.417236328125, "loss_sft": 0.2596279475837946, "loss_total_v6": 0.676696427166462, "step": 4235 }, { "epoch": 0.9769513513513514, "grad_norm": 29.434776306152344, "learning_rate": 4.726254505365535e-06, "loss": 11.416583061218262, "loss_alignment": 0.8740234375, "loss_alignment_w": 0.43701171875, "loss_sft": 0.27716557309031487, "loss_total_v6": 0.7135364413261414, "step": 4236 }, { "epoch": 0.977181981981982, "grad_norm": 28.0848445892334, "learning_rate": 4.72424370665269e-06, "loss": 11.118074417114258, "loss_alignment": 0.8330078125, "loss_alignment_w": 0.41650390625, "loss_sft": 0.2787572368979454, "loss_total_v6": 0.6948796883225441, "step": 4237 }, { "epoch": 0.9774126126126126, "grad_norm": 30.019779205322266, "learning_rate": 4.7222329526736635e-06, "loss": 11.054696083068848, "loss_alignment": 0.84375, "loss_alignment_w": 0.421875, "loss_sft": 0.26782282441854477, "loss_total_v6": 0.6909185200929642, "step": 4238 }, { "epoch": 0.9776432432432433, "grad_norm": 27.876134872436523, "learning_rate": 4.720222243754645e-06, "loss": 11.208837509155273, "loss_alignment": 0.857421875, "loss_alignment_w": 0.4287109375, "loss_sft": 0.2709716521203518, "loss_total_v6": 0.7005523145198822, "step": 4239 }, { "epoch": 0.9778738738738739, "grad_norm": 31.584354400634766, "learning_rate": 4.718211580221813e-06, "loss": 11.61099624633789, "loss_alignment": 0.8759765625, "loss_alignment_w": 0.43798828125, "loss_sft": 0.287531154230237, "loss_total_v6": 0.7256873026490211, "step": 4240 }, { "epoch": 0.9781045045045045, "grad_norm": 27.691356658935547, "learning_rate": 4.716200962401347e-06, "loss": 12.003595352172852, "loss_alignment": 0.88330078125, "loss_alignment_w": 0.441650390625, "loss_sft": 0.30776562355458736, "loss_total_v6": 0.7502247393131256, "step": 4241 }, { "epoch": 0.9783351351351351, "grad_norm": 27.656410217285156, "learning_rate": 4.714190390619411e-06, "loss": 11.592533111572266, "loss_alignment": 0.869140625, "loss_alignment_w": 0.4345703125, "loss_sft": 0.2911683917045593, "loss_total_v6": 0.7245332673192024, "step": 4242 }, { "epoch": 0.9785657657657658, "grad_norm": 27.3007869720459, "learning_rate": 4.712179865202164e-06, "loss": 11.473541259765625, "loss_alignment": 0.86083984375, "loss_alignment_w": 0.430419921875, "loss_sft": 0.28504372388124466, "loss_total_v6": 0.7170963436365128, "step": 4243 }, { "epoch": 0.9787963963963964, "grad_norm": 30.1173152923584, "learning_rate": 4.7101693864757606e-06, "loss": 11.413816452026367, "loss_alignment": 0.85498046875, "loss_alignment_w": 0.427490234375, "loss_sft": 0.28463732451200485, "loss_total_v6": 0.7133635208010674, "step": 4244 }, { "epoch": 0.979027027027027, "grad_norm": 25.767017364501953, "learning_rate": 4.708158954766341e-06, "loss": 11.25350570678711, "loss_alignment": 0.875, "loss_alignment_w": 0.4375, "loss_sft": 0.26527953892946243, "loss_total_v6": 0.703344076871872, "step": 4245 }, { "epoch": 0.9792576576576577, "grad_norm": 28.245380401611328, "learning_rate": 4.706148570400047e-06, "loss": 11.295730590820312, "loss_alignment": 0.8603515625, "loss_alignment_w": 0.43017578125, "loss_sft": 0.27470871433615685, "loss_total_v6": 0.7059831470251083, "step": 4246 }, { "epoch": 0.9794882882882883, "grad_norm": 29.91977310180664, "learning_rate": 4.704138233703006e-06, "loss": 11.825931549072266, "loss_alignment": 0.8828125, "loss_alignment_w": 0.44140625, "loss_sft": 0.2985079810023308, "loss_total_v6": 0.7391207441687584, "step": 4247 }, { "epoch": 0.9797189189189189, "grad_norm": 26.705087661743164, "learning_rate": 4.702127945001339e-06, "loss": 10.546627044677734, "loss_alignment": 0.83984375, "loss_alignment_w": 0.419921875, "loss_sft": 0.24092078022658825, "loss_total_v6": 0.659164197742939, "step": 4248 }, { "epoch": 0.9799495495495496, "grad_norm": 24.855789184570312, "learning_rate": 4.700117704621158e-06, "loss": 10.905824661254883, "loss_alignment": 0.8427734375, "loss_alignment_w": 0.42138671875, "loss_sft": 0.2591439578682184, "loss_total_v6": 0.681614026427269, "step": 4249 }, { "epoch": 0.9801801801801802, "grad_norm": 26.185611724853516, "learning_rate": 4.69810751288857e-06, "loss": 11.16441822052002, "loss_alignment": 0.85400390625, "loss_alignment_w": 0.427001953125, "loss_sft": 0.27057583071291447, "loss_total_v6": 0.6977761760354042, "step": 4250 }, { "epoch": 0.9804108108108108, "grad_norm": 27.585216522216797, "learning_rate": 4.696097370129675e-06, "loss": 10.716495513916016, "loss_alignment": 0.8271484375, "loss_alignment_w": 0.41357421875, "loss_sft": 0.2579615246504545, "loss_total_v6": 0.6697809621691704, "step": 4251 }, { "epoch": 0.9806414414414414, "grad_norm": 25.237646102905273, "learning_rate": 4.69408727667056e-06, "loss": 11.51345443725586, "loss_alignment": 0.8837890625, "loss_alignment_w": 0.44189453125, "loss_sft": 0.27881027199327946, "loss_total_v6": 0.7195908948779106, "step": 4252 }, { "epoch": 0.980872072072072, "grad_norm": 27.6003360748291, "learning_rate": 4.69207723283731e-06, "loss": 11.509753227233887, "loss_alignment": 0.884765625, "loss_alignment_w": 0.4423828125, "loss_sft": 0.2774497717618942, "loss_total_v6": 0.7193595692515373, "step": 4253 }, { "epoch": 0.9811027027027027, "grad_norm": 26.05771827697754, "learning_rate": 4.690067238955996e-06, "loss": 11.214834213256836, "loss_alignment": 0.84228515625, "loss_alignment_w": 0.421142578125, "loss_sft": 0.2791284304112196, "loss_total_v6": 0.7009271457791328, "step": 4254 }, { "epoch": 0.9813333333333333, "grad_norm": 26.330080032348633, "learning_rate": 4.688057295352684e-06, "loss": 11.448413848876953, "loss_alignment": 0.85205078125, "loss_alignment_w": 0.426025390625, "loss_sft": 0.2902633734047413, "loss_total_v6": 0.715525858104229, "step": 4255 }, { "epoch": 0.981563963963964, "grad_norm": 31.85994529724121, "learning_rate": 4.686047402353433e-06, "loss": 11.801677703857422, "loss_alignment": 0.85595703125, "loss_alignment_w": 0.427978515625, "loss_sft": 0.30982462875545025, "loss_total_v6": 0.7376048117876053, "step": 4256 }, { "epoch": 0.9817945945945946, "grad_norm": 28.633235931396484, "learning_rate": 4.684037560284294e-06, "loss": 11.140594482421875, "loss_alignment": 0.8623046875, "loss_alignment_w": 0.43115234375, "loss_sft": 0.2650432363152504, "loss_total_v6": 0.6962871253490448, "step": 4257 }, { "epoch": 0.9820252252252252, "grad_norm": 25.2777156829834, "learning_rate": 4.682027769471305e-06, "loss": 11.173913955688477, "loss_alignment": 0.853515625, "loss_alignment_w": 0.4267578125, "loss_sft": 0.2711540646851063, "loss_total_v6": 0.698369599878788, "step": 4258 }, { "epoch": 0.9822558558558558, "grad_norm": 29.77121353149414, "learning_rate": 4.680018030240501e-06, "loss": 11.55281925201416, "loss_alignment": 0.91259765625, "loss_alignment_w": 0.456298828125, "loss_sft": 0.26561506278812885, "loss_total_v6": 0.722051240503788, "step": 4259 }, { "epoch": 0.9824864864864865, "grad_norm": 25.191347122192383, "learning_rate": 4.678008342917903e-06, "loss": 11.159520149230957, "loss_alignment": 0.8740234375, "loss_alignment_w": 0.43701171875, "loss_sft": 0.2610686384141445, "loss_total_v6": 0.6974700093269348, "step": 4260 }, { "epoch": 0.9827171171171171, "grad_norm": 27.442794799804688, "learning_rate": 4.675998707829532e-06, "loss": 10.834138870239258, "loss_alignment": 0.82421875, "loss_alignment_w": 0.412109375, "loss_sft": 0.26513112522661686, "loss_total_v6": 0.6771337166428566, "step": 4261 }, { "epoch": 0.9829477477477477, "grad_norm": 25.55242156982422, "learning_rate": 4.673989125301393e-06, "loss": 11.300204277038574, "loss_alignment": 0.86669921875, "loss_alignment_w": 0.433349609375, "loss_sft": 0.2737828828394413, "loss_total_v6": 0.7062627375125885, "step": 4262 }, { "epoch": 0.9831783783783784, "grad_norm": 24.24431800842285, "learning_rate": 4.671979595659486e-06, "loss": 11.302289962768555, "loss_alignment": 0.8525390625, "loss_alignment_w": 0.42626953125, "loss_sft": 0.27963535487651825, "loss_total_v6": 0.7063931450247765, "step": 4263 }, { "epoch": 0.983409009009009, "grad_norm": 27.298219680786133, "learning_rate": 4.6699701192298e-06, "loss": 10.809865951538086, "loss_alignment": 0.84814453125, "loss_alignment_w": 0.424072265625, "loss_sft": 0.2514528427273035, "loss_total_v6": 0.6756166517734528, "step": 4264 }, { "epoch": 0.9836396396396396, "grad_norm": 27.91352081298828, "learning_rate": 4.6679606963383166e-06, "loss": 10.976418495178223, "loss_alignment": 0.8515625, "loss_alignment_w": 0.42578125, "loss_sft": 0.2613740377128124, "loss_total_v6": 0.6860261410474777, "step": 4265 }, { "epoch": 0.9838702702702703, "grad_norm": 23.877349853515625, "learning_rate": 4.665951327311013e-06, "loss": 11.273223876953125, "loss_alignment": 0.85546875, "loss_alignment_w": 0.427734375, "loss_sft": 0.2771930228918791, "loss_total_v6": 0.7045764774084091, "step": 4266 }, { "epoch": 0.9841009009009009, "grad_norm": 25.67732810974121, "learning_rate": 4.66394201247385e-06, "loss": 11.76345443725586, "loss_alignment": 0.87841796875, "loss_alignment_w": 0.439208984375, "loss_sft": 0.29596108570694923, "loss_total_v6": 0.7352158427238464, "step": 4267 }, { "epoch": 0.9843315315315315, "grad_norm": 28.28881072998047, "learning_rate": 4.661932752152783e-06, "loss": 11.538423538208008, "loss_alignment": 0.85693359375, "loss_alignment_w": 0.428466796875, "loss_sft": 0.2940274141728878, "loss_total_v6": 0.7211514636874199, "step": 4268 }, { "epoch": 0.9845621621621622, "grad_norm": 25.336870193481445, "learning_rate": 4.659923546673761e-06, "loss": 10.927845001220703, "loss_alignment": 0.85986328125, "loss_alignment_w": 0.429931640625, "loss_sft": 0.2539131660014391, "loss_total_v6": 0.6829903200268745, "step": 4269 }, { "epoch": 0.9847927927927927, "grad_norm": 30.485841751098633, "learning_rate": 4.657914396362719e-06, "loss": 11.185274124145508, "loss_alignment": 0.86083984375, "loss_alignment_w": 0.430419921875, "loss_sft": 0.26925478503108025, "loss_total_v6": 0.6990796327590942, "step": 4270 }, { "epoch": 0.9850234234234234, "grad_norm": 25.251371383666992, "learning_rate": 4.655905301545589e-06, "loss": 11.384162902832031, "loss_alignment": 0.8720703125, "loss_alignment_w": 0.43603515625, "loss_sft": 0.2748951856046915, "loss_total_v6": 0.7115101739764214, "step": 4271 }, { "epoch": 0.9852540540540541, "grad_norm": 27.586584091186523, "learning_rate": 4.653896262548291e-06, "loss": 11.742016792297363, "loss_alignment": 0.88916015625, "loss_alignment_w": 0.444580078125, "loss_sft": 0.2892959900200367, "loss_total_v6": 0.7338760569691658, "step": 4272 }, { "epoch": 0.9854846846846846, "grad_norm": 28.821287155151367, "learning_rate": 4.651887279696734e-06, "loss": 11.76991081237793, "loss_alignment": 0.88232421875, "loss_alignment_w": 0.441162109375, "loss_sft": 0.29398429952561855, "loss_total_v6": 0.7356194481253624, "step": 4273 }, { "epoch": 0.9857153153153153, "grad_norm": 31.339189529418945, "learning_rate": 4.649878353316818e-06, "loss": 11.152812957763672, "loss_alignment": 0.859375, "loss_alignment_w": 0.4296875, "loss_sft": 0.26751588843762875, "loss_total_v6": 0.6970508322119713, "step": 4274 }, { "epoch": 0.985945945945946, "grad_norm": 27.25634765625, "learning_rate": 4.647869483734441e-06, "loss": 11.347265243530273, "loss_alignment": 0.86181640625, "loss_alignment_w": 0.430908203125, "loss_sft": 0.2774413377046585, "loss_total_v6": 0.7092040330171585, "step": 4275 }, { "epoch": 0.9861765765765765, "grad_norm": 28.952037811279297, "learning_rate": 4.645860671275483e-06, "loss": 12.468498229980469, "loss_alignment": 0.9169921875, "loss_alignment_w": 0.45849609375, "loss_sft": 0.3201594240963459, "loss_total_v6": 0.7792811393737793, "step": 4276 }, { "epoch": 0.9864072072072072, "grad_norm": 37.10963439941406, "learning_rate": 4.643851916265819e-06, "loss": 10.915922164916992, "loss_alignment": 0.82861328125, "loss_alignment_w": 0.414306640625, "loss_sft": 0.2691896799951792, "loss_total_v6": 0.6822451129555702, "step": 4277 }, { "epoch": 0.9866378378378379, "grad_norm": 26.297449111938477, "learning_rate": 4.6418432190313135e-06, "loss": 11.586687088012695, "loss_alignment": 0.83984375, "loss_alignment_w": 0.419921875, "loss_sft": 0.30427658930420876, "loss_total_v6": 0.7241679579019547, "step": 4278 }, { "epoch": 0.9868684684684684, "grad_norm": 26.43995475769043, "learning_rate": 4.63983457989782e-06, "loss": 11.251798629760742, "loss_alignment": 0.876953125, "loss_alignment_w": 0.4384765625, "loss_sft": 0.2644098959863186, "loss_total_v6": 0.7032374143600464, "step": 4279 }, { "epoch": 0.9870990990990991, "grad_norm": 26.685516357421875, "learning_rate": 4.637825999191189e-06, "loss": 11.169246673583984, "loss_alignment": 0.81005859375, "loss_alignment_w": 0.405029296875, "loss_sft": 0.2927739769220352, "loss_total_v6": 0.6980779469013214, "step": 4280 }, { "epoch": 0.9873297297297298, "grad_norm": 25.892980575561523, "learning_rate": 4.635817477237253e-06, "loss": 11.354830741882324, "loss_alignment": 0.84912109375, "loss_alignment_w": 0.424560546875, "loss_sft": 0.2847044039517641, "loss_total_v6": 0.7096769288182259, "step": 4281 }, { "epoch": 0.9875603603603603, "grad_norm": 26.891010284423828, "learning_rate": 4.6338090143618435e-06, "loss": 11.61322021484375, "loss_alignment": 0.8564453125, "loss_alignment_w": 0.42822265625, "loss_sft": 0.297512024641037, "loss_total_v6": 0.7258262261748314, "step": 4282 }, { "epoch": 0.987790990990991, "grad_norm": 25.529747009277344, "learning_rate": 4.631800610890772e-06, "loss": 11.59600830078125, "loss_alignment": 0.85595703125, "loss_alignment_w": 0.427978515625, "loss_sft": 0.2971382364630699, "loss_total_v6": 0.7247505113482475, "step": 4283 }, { "epoch": 0.9880216216216217, "grad_norm": 26.637832641601562, "learning_rate": 4.62979226714985e-06, "loss": 11.207460403442383, "loss_alignment": 0.8525390625, "loss_alignment_w": 0.42626953125, "loss_sft": 0.27474608831107616, "loss_total_v6": 0.7004663050174713, "step": 4284 }, { "epoch": 0.9882522522522522, "grad_norm": 33.281253814697266, "learning_rate": 4.627783983464875e-06, "loss": 10.871997833251953, "loss_alignment": 0.85302734375, "loss_alignment_w": 0.426513671875, "loss_sft": 0.2537491209805012, "loss_total_v6": 0.6794998496770859, "step": 4285 }, { "epoch": 0.9884828828828829, "grad_norm": 29.212156295776367, "learning_rate": 4.625775760161637e-06, "loss": 11.279109001159668, "loss_alignment": 0.84716796875, "loss_alignment_w": 0.423583984375, "loss_sft": 0.28139087557792664, "loss_total_v6": 0.7049443200230598, "step": 4286 }, { "epoch": 0.9887135135135136, "grad_norm": 26.000837326049805, "learning_rate": 4.623767597565913e-06, "loss": 11.517744064331055, "loss_alignment": 0.8505859375, "loss_alignment_w": 0.42529296875, "loss_sft": 0.2950390838086605, "loss_total_v6": 0.7198590189218521, "step": 4287 }, { "epoch": 0.9889441441441441, "grad_norm": 26.623779296875, "learning_rate": 4.621759496003472e-06, "loss": 11.033796310424805, "loss_alignment": 0.8564453125, "loss_alignment_w": 0.42822265625, "loss_sft": 0.26239665783941746, "loss_total_v6": 0.6896122321486473, "step": 4288 }, { "epoch": 0.9891747747747748, "grad_norm": 24.837509155273438, "learning_rate": 4.61975145580007e-06, "loss": 10.897727966308594, "loss_alignment": 0.8583984375, "loss_alignment_w": 0.42919921875, "loss_sft": 0.2516188304871321, "loss_total_v6": 0.6811079531908035, "step": 4289 }, { "epoch": 0.9894054054054054, "grad_norm": 25.416614532470703, "learning_rate": 4.61774347728146e-06, "loss": 11.817026138305664, "loss_alignment": 0.86328125, "loss_alignment_w": 0.431640625, "loss_sft": 0.3067251928150654, "loss_total_v6": 0.7385641559958458, "step": 4290 }, { "epoch": 0.989636036036036, "grad_norm": 28.785037994384766, "learning_rate": 4.61573556077338e-06, "loss": 11.584066390991211, "loss_alignment": 0.86474609375, "loss_alignment_w": 0.432373046875, "loss_sft": 0.2910665012896061, "loss_total_v6": 0.7240041196346283, "step": 4291 }, { "epoch": 0.9898666666666667, "grad_norm": 29.106176376342773, "learning_rate": 4.613727706601558e-06, "loss": 11.896515846252441, "loss_alignment": 0.85595703125, "loss_alignment_w": 0.427978515625, "loss_sft": 0.3152028303593397, "loss_total_v6": 0.7435322627425194, "step": 4292 }, { "epoch": 0.9900972972972973, "grad_norm": 26.895851135253906, "learning_rate": 4.611719915091712e-06, "loss": 11.626893997192383, "loss_alignment": 0.86767578125, "loss_alignment_w": 0.433837890625, "loss_sft": 0.2913781013339758, "loss_total_v6": 0.7266808450222015, "step": 4293 }, { "epoch": 0.9903279279279279, "grad_norm": 30.453004837036133, "learning_rate": 4.609712186569552e-06, "loss": 11.188953399658203, "loss_alignment": 0.85009765625, "loss_alignment_w": 0.425048828125, "loss_sft": 0.27354361675679684, "loss_total_v6": 0.6993096023797989, "step": 4294 }, { "epoch": 0.9905585585585586, "grad_norm": 32.369224548339844, "learning_rate": 4.6077045213607765e-06, "loss": 11.295845031738281, "loss_alignment": 0.8681640625, "loss_alignment_w": 0.43408203125, "loss_sft": 0.2719541061669588, "loss_total_v6": 0.7059903591871262, "step": 4295 }, { "epoch": 0.9907891891891892, "grad_norm": 28.729965209960938, "learning_rate": 4.605696919791071e-06, "loss": 11.65298843383789, "loss_alignment": 0.8427734375, "loss_alignment_w": 0.42138671875, "loss_sft": 0.3071997333317995, "loss_total_v6": 0.7283117324113846, "step": 4296 }, { "epoch": 0.9910198198198198, "grad_norm": 29.66866111755371, "learning_rate": 4.603689382186117e-06, "loss": 11.290641784667969, "loss_alignment": 0.8720703125, "loss_alignment_w": 0.43603515625, "loss_sft": 0.26985883712768555, "loss_total_v6": 0.705665111541748, "step": 4297 }, { "epoch": 0.9912504504504505, "grad_norm": 24.057220458984375, "learning_rate": 4.6016819088715795e-06, "loss": 11.665823936462402, "loss_alignment": 0.86669921875, "loss_alignment_w": 0.433349609375, "loss_sft": 0.29501670598983765, "loss_total_v6": 0.7291140258312225, "step": 4298 }, { "epoch": 0.9914810810810811, "grad_norm": 28.7033634185791, "learning_rate": 4.599674500173114e-06, "loss": 11.102092742919922, "loss_alignment": 0.84423828125, "loss_alignment_w": 0.422119140625, "loss_sft": 0.2719142362475395, "loss_total_v6": 0.6938807889819145, "step": 4299 }, { "epoch": 0.9917117117117117, "grad_norm": 26.844301223754883, "learning_rate": 4.597667156416371e-06, "loss": 11.735169410705566, "loss_alignment": 0.8857421875, "loss_alignment_w": 0.44287109375, "loss_sft": 0.2902871035039425, "loss_total_v6": 0.7334481030702591, "step": 4300 }, { "epoch": 0.9919423423423424, "grad_norm": 27.890792846679688, "learning_rate": 4.595659877926984e-06, "loss": 11.106051445007324, "loss_alignment": 0.8203125, "loss_alignment_w": 0.41015625, "loss_sft": 0.28404830023646355, "loss_total_v6": 0.6941282227635384, "step": 4301 }, { "epoch": 0.992172972972973, "grad_norm": 28.16690444946289, "learning_rate": 4.593652665030578e-06, "loss": 11.254935264587402, "loss_alignment": 0.8564453125, "loss_alignment_w": 0.42822265625, "loss_sft": 0.2737154122442007, "loss_total_v6": 0.703433446586132, "step": 4302 }, { "epoch": 0.9924036036036036, "grad_norm": 28.993377685546875, "learning_rate": 4.591645518052769e-06, "loss": 10.962934494018555, "loss_alignment": 0.837890625, "loss_alignment_w": 0.4189453125, "loss_sft": 0.26527679339051247, "loss_total_v6": 0.6851834133267403, "step": 4303 }, { "epoch": 0.9926342342342342, "grad_norm": 29.139976501464844, "learning_rate": 4.589638437319157e-06, "loss": 10.995542526245117, "loss_alignment": 0.83544921875, "loss_alignment_w": 0.417724609375, "loss_sft": 0.27027502842247486, "loss_total_v6": 0.6872214451432228, "step": 4304 }, { "epoch": 0.9928648648648649, "grad_norm": 27.791866302490234, "learning_rate": 4.587631423155341e-06, "loss": 11.199681282043457, "loss_alignment": 0.84375, "loss_alignment_w": 0.421875, "loss_sft": 0.2785628531128168, "loss_total_v6": 0.6999800801277161, "step": 4305 }, { "epoch": 0.9930954954954955, "grad_norm": 29.648818969726562, "learning_rate": 4.585624475886901e-06, "loss": 11.001482009887695, "loss_alignment": 0.84033203125, "loss_alignment_w": 0.420166015625, "loss_sft": 0.266999339684844, "loss_total_v6": 0.687592588365078, "step": 4306 }, { "epoch": 0.9933261261261261, "grad_norm": 27.50870132446289, "learning_rate": 4.583617595839409e-06, "loss": 11.136527061462402, "loss_alignment": 0.8408203125, "loss_alignment_w": 0.42041015625, "loss_sft": 0.27443262562155724, "loss_total_v6": 0.6960329562425613, "step": 4307 }, { "epoch": 0.9935567567567568, "grad_norm": 26.547182083129883, "learning_rate": 4.581610783338424e-06, "loss": 11.01667308807373, "loss_alignment": 0.83544921875, "loss_alignment_w": 0.417724609375, "loss_sft": 0.2720839325338602, "loss_total_v6": 0.6885420605540276, "step": 4308 }, { "epoch": 0.9937873873873874, "grad_norm": 25.424623489379883, "learning_rate": 4.579604038709495e-06, "loss": 11.87877082824707, "loss_alignment": 0.87646484375, "loss_alignment_w": 0.438232421875, "loss_sft": 0.3055030591785908, "loss_total_v6": 0.7424231916666031, "step": 4309 }, { "epoch": 0.994018018018018, "grad_norm": 27.75236701965332, "learning_rate": 4.577597362278165e-06, "loss": 11.205204963684082, "loss_alignment": 0.82666015625, "loss_alignment_w": 0.413330078125, "loss_sft": 0.28684261813759804, "loss_total_v6": 0.7003252953290939, "step": 4310 }, { "epoch": 0.9942486486486487, "grad_norm": 28.37531852722168, "learning_rate": 4.5755907543699586e-06, "loss": 11.362641334533691, "loss_alignment": 0.86279296875, "loss_alignment_w": 0.431396484375, "loss_sft": 0.2776394374668598, "loss_total_v6": 0.7101650685071945, "step": 4311 }, { "epoch": 0.9944792792792793, "grad_norm": 28.87403106689453, "learning_rate": 4.573584215310394e-06, "loss": 11.15054702758789, "loss_alignment": 0.82568359375, "loss_alignment_w": 0.412841796875, "loss_sft": 0.2835333254188299, "loss_total_v6": 0.696909174323082, "step": 4312 }, { "epoch": 0.9947099099099099, "grad_norm": 24.89498519897461, "learning_rate": 4.571577745424975e-06, "loss": 10.609821319580078, "loss_alignment": 0.83544921875, "loss_alignment_w": 0.417724609375, "loss_sft": 0.24697609804570675, "loss_total_v6": 0.6631137877702713, "step": 4313 }, { "epoch": 0.9949405405405405, "grad_norm": 27.539304733276367, "learning_rate": 4.569571345039195e-06, "loss": 11.050500869750977, "loss_alignment": 0.8212890625, "loss_alignment_w": 0.41064453125, "loss_sft": 0.2791725266724825, "loss_total_v6": 0.690656304359436, "step": 4314 }, { "epoch": 0.9951711711711712, "grad_norm": 26.56369400024414, "learning_rate": 4.567565014478538e-06, "loss": 11.49972915649414, "loss_alignment": 0.857421875, "loss_alignment_w": 0.4287109375, "loss_sft": 0.2899916097521782, "loss_total_v6": 0.7187330648303032, "step": 4315 }, { "epoch": 0.9954018018018018, "grad_norm": 27.173585891723633, "learning_rate": 4.565558754068477e-06, "loss": 11.512531280517578, "loss_alignment": 0.85205078125, "loss_alignment_w": 0.426025390625, "loss_sft": 0.2930348105728626, "loss_total_v6": 0.7195331826806068, "step": 4316 }, { "epoch": 0.9956324324324324, "grad_norm": 28.442684173583984, "learning_rate": 4.56355256413447e-06, "loss": 11.793935775756836, "loss_alignment": 0.85205078125, "loss_alignment_w": 0.426025390625, "loss_sft": 0.3120722100138664, "loss_total_v6": 0.7371210306882858, "step": 4317 }, { "epoch": 0.995863063063063, "grad_norm": 29.609161376953125, "learning_rate": 4.561546445001968e-06, "loss": 11.209729194641113, "loss_alignment": 0.83642578125, "loss_alignment_w": 0.418212890625, "loss_sft": 0.2821662910282612, "loss_total_v6": 0.7006080523133278, "step": 4318 }, { "epoch": 0.9960936936936937, "grad_norm": 27.269807815551758, "learning_rate": 4.5595403969964035e-06, "loss": 11.40085220336914, "loss_alignment": 0.86181640625, "loss_alignment_w": 0.430908203125, "loss_sft": 0.2803938388824463, "loss_total_v6": 0.7125532701611519, "step": 4319 }, { "epoch": 0.9963243243243243, "grad_norm": 29.981245040893555, "learning_rate": 4.557534420443209e-06, "loss": 11.650022506713867, "loss_alignment": 0.86474609375, "loss_alignment_w": 0.432373046875, "loss_sft": 0.2956313081085682, "loss_total_v6": 0.7281263917684555, "step": 4320 }, { "epoch": 0.9965549549549549, "grad_norm": 26.591461181640625, "learning_rate": 4.555528515667793e-06, "loss": 10.960517883300781, "loss_alignment": 0.82763671875, "loss_alignment_w": 0.413818359375, "loss_sft": 0.27171754464507103, "loss_total_v6": 0.685032345354557, "step": 4321 }, { "epoch": 0.9967855855855856, "grad_norm": 26.363916397094727, "learning_rate": 4.5535226829955615e-06, "loss": 11.101114273071289, "loss_alignment": 0.80859375, "loss_alignment_w": 0.404296875, "loss_sft": 0.28908029943704605, "loss_total_v6": 0.6938196793198586, "step": 4322 }, { "epoch": 0.9970162162162162, "grad_norm": 27.906845092773438, "learning_rate": 4.5515169227519016e-06, "loss": 11.503454208374023, "loss_alignment": 0.8447265625, "loss_alignment_w": 0.42236328125, "loss_sft": 0.29570238664746284, "loss_total_v6": 0.71896593272686, "step": 4323 }, { "epoch": 0.9972468468468468, "grad_norm": 24.173677444458008, "learning_rate": 4.549511235262196e-06, "loss": 10.907445907592773, "loss_alignment": 0.84814453125, "loss_alignment_w": 0.424072265625, "loss_sft": 0.25781095400452614, "loss_total_v6": 0.6817153766751289, "step": 4324 }, { "epoch": 0.9974774774774775, "grad_norm": 26.276992797851562, "learning_rate": 4.547505620851812e-06, "loss": 11.489849090576172, "loss_alignment": 0.87890625, "loss_alignment_w": 0.439453125, "loss_sft": 0.27695342153310776, "loss_total_v6": 0.7181155309081078, "step": 4325 }, { "epoch": 0.9977081081081081, "grad_norm": 27.77890968322754, "learning_rate": 4.545500079846103e-06, "loss": 11.376564025878906, "loss_alignment": 0.86767578125, "loss_alignment_w": 0.433837890625, "loss_sft": 0.27612931840121746, "loss_total_v6": 0.711035281419754, "step": 4326 }, { "epoch": 0.9979387387387387, "grad_norm": 38.41559982299805, "learning_rate": 4.543494612570411e-06, "loss": 11.354930877685547, "loss_alignment": 0.87109375, "loss_alignment_w": 0.435546875, "loss_sft": 0.2725189011543989, "loss_total_v6": 0.7096832245588303, "step": 4327 }, { "epoch": 0.9981693693693694, "grad_norm": 28.00627326965332, "learning_rate": 4.541489219350069e-06, "loss": 11.005616188049316, "loss_alignment": 0.85986328125, "loss_alignment_w": 0.429931640625, "loss_sft": 0.2588502001017332, "loss_total_v6": 0.6878510415554047, "step": 4328 }, { "epoch": 0.9984, "grad_norm": 25.527244567871094, "learning_rate": 4.539483900510399e-06, "loss": 11.109559059143066, "loss_alignment": 0.853515625, "loss_alignment_w": 0.4267578125, "loss_sft": 0.2676811795681715, "loss_total_v6": 0.6943474411964417, "step": 4329 }, { "epoch": 0.9986306306306306, "grad_norm": 26.619022369384766, "learning_rate": 4.537478656376705e-06, "loss": 11.33087158203125, "loss_alignment": 0.86669921875, "loss_alignment_w": 0.433349609375, "loss_sft": 0.27547072619199753, "loss_total_v6": 0.7081794440746307, "step": 4330 }, { "epoch": 0.9988612612612613, "grad_norm": 26.26848602294922, "learning_rate": 4.5354734872742855e-06, "loss": 11.602178573608398, "loss_alignment": 0.89599609375, "loss_alignment_w": 0.447998046875, "loss_sft": 0.2774585746228695, "loss_total_v6": 0.7251361683011055, "step": 4331 }, { "epoch": 0.9990918918918918, "grad_norm": 22.329360961914062, "learning_rate": 4.533468393528421e-06, "loss": 11.373550415039062, "loss_alignment": 0.857421875, "loss_alignment_w": 0.4287109375, "loss_sft": 0.2827616110444069, "loss_total_v6": 0.710846945643425, "step": 4332 }, { "epoch": 0.9993225225225225, "grad_norm": 28.09711265563965, "learning_rate": 4.531463375464382e-06, "loss": 11.309536933898926, "loss_alignment": 0.84765625, "loss_alignment_w": 0.423828125, "loss_sft": 0.2824228275567293, "loss_total_v6": 0.7068460509181023, "step": 4333 }, { "epoch": 0.9995531531531532, "grad_norm": 26.718334197998047, "learning_rate": 4.529458433407429e-06, "loss": 10.459373474121094, "loss_alignment": 0.8017578125, "loss_alignment_w": 0.40087890625, "loss_sft": 0.251458577811718, "loss_total_v6": 0.6537107899785042, "step": 4334 }, { "epoch": 0.9997837837837837, "grad_norm": 30.66808319091797, "learning_rate": 4.527453567682808e-06, "loss": 10.817717552185059, "loss_alignment": 0.859375, "loss_alignment_w": 0.4296875, "loss_sft": 0.2463282998651266, "loss_total_v6": 0.6761073619127274, "step": 4335 }, { "epoch": 1.0, "grad_norm": 25.383861541748047, "learning_rate": 4.525448778615752e-06, "loss": 10.005655288696289, "loss_alignment": 0.8307291666666666, "loss_alignment_w": 0.4153645833333333, "loss_sft": 0.25150002439816793, "loss_total_v6": 0.6670436859130859, "step": 4336 }, { "epoch": 1.0002306306306306, "grad_norm": 24.461278915405273, "learning_rate": 4.523444066531484e-06, "loss": 10.226125717163086, "loss_alignment": 0.857421875, "loss_alignment_w": 0.4287109375, "loss_sft": 0.21109331212937832, "loss_total_v6": 0.6391328647732735, "step": 4337 }, { "epoch": 1.0004612612612613, "grad_norm": 24.54036521911621, "learning_rate": 4.521439431755209e-06, "loss": 10.178466796875, "loss_alignment": 0.83935546875, "loss_alignment_w": 0.419677734375, "loss_sft": 0.21726993471384048, "loss_total_v6": 0.6361541971564293, "step": 4338 }, { "epoch": 1.000691891891892, "grad_norm": 24.184324264526367, "learning_rate": 4.51943487461213e-06, "loss": 9.838737487792969, "loss_alignment": 0.8515625, "loss_alignment_w": 0.42578125, "loss_sft": 0.18883465230464935, "loss_total_v6": 0.61492108553648, "step": 4339 }, { "epoch": 1.0009225225225225, "grad_norm": 28.21491241455078, "learning_rate": 4.517430395427424e-06, "loss": 10.535022735595703, "loss_alignment": 0.859375, "loss_alignment_w": 0.4296875, "loss_sft": 0.2291176225990057, "loss_total_v6": 0.6584389060735703, "step": 4340 }, { "epoch": 1.0011531531531532, "grad_norm": 26.574993133544922, "learning_rate": 4.515425994526269e-06, "loss": 10.084651947021484, "loss_alignment": 0.8427734375, "loss_alignment_w": 0.42138671875, "loss_sft": 0.20878194086253643, "loss_total_v6": 0.6302907466888428, "step": 4341 }, { "epoch": 1.0013837837837838, "grad_norm": 27.290260314941406, "learning_rate": 4.513421672233817e-06, "loss": 10.457733154296875, "loss_alignment": 0.87939453125, "loss_alignment_w": 0.439697265625, "loss_sft": 0.2136211283504963, "loss_total_v6": 0.6536083146929741, "step": 4342 }, { "epoch": 1.0016144144144143, "grad_norm": 26.203529357910156, "learning_rate": 4.511417428875217e-06, "loss": 10.070375442504883, "loss_alignment": 0.84619140625, "loss_alignment_w": 0.423095703125, "loss_sft": 0.20633327215909958, "loss_total_v6": 0.6293984577059746, "step": 4343 }, { "epoch": 1.0018450450450451, "grad_norm": 24.588552474975586, "learning_rate": 4.509413264775604e-06, "loss": 10.189889907836914, "loss_alignment": 0.85107421875, "loss_alignment_w": 0.425537109375, "loss_sft": 0.21136153303086758, "loss_total_v6": 0.6368680968880653, "step": 4344 }, { "epoch": 1.0020756756756757, "grad_norm": 28.943410873413086, "learning_rate": 4.507409180260095e-06, "loss": 9.920629501342773, "loss_alignment": 0.8271484375, "loss_alignment_w": 0.41357421875, "loss_sft": 0.2071822863072157, "loss_total_v6": 0.6200393661856651, "step": 4345 }, { "epoch": 1.0023063063063062, "grad_norm": 26.04715919494629, "learning_rate": 4.505405175653797e-06, "loss": 10.134405136108398, "loss_alignment": 0.845703125, "loss_alignment_w": 0.4228515625, "loss_sft": 0.21068607456982136, "loss_total_v6": 0.6334002986550331, "step": 4346 }, { "epoch": 1.002536936936937, "grad_norm": 25.64447021484375, "learning_rate": 4.5034012512818065e-06, "loss": 9.8034086227417, "loss_alignment": 0.8486328125, "loss_alignment_w": 0.42431640625, "loss_sft": 0.18844240345060825, "loss_total_v6": 0.6127130389213562, "step": 4347 }, { "epoch": 1.0027675675675676, "grad_norm": 23.588481903076172, "learning_rate": 4.5013974074692014e-06, "loss": 9.760469436645508, "loss_alignment": 0.86328125, "loss_alignment_w": 0.431640625, "loss_sft": 0.1783277103677392, "loss_total_v6": 0.6100293770432472, "step": 4348 }, { "epoch": 1.0029981981981981, "grad_norm": 24.15664291381836, "learning_rate": 4.499393644541052e-06, "loss": 9.844615936279297, "loss_alignment": 0.8603515625, "loss_alignment_w": 0.43017578125, "loss_sft": 0.18572313990443945, "loss_total_v6": 0.6152885407209396, "step": 4349 }, { "epoch": 1.003228828828829, "grad_norm": 26.15625762939453, "learning_rate": 4.497389962822416e-06, "loss": 10.269499778747559, "loss_alignment": 0.869140625, "loss_alignment_w": 0.4345703125, "loss_sft": 0.20847888477146626, "loss_total_v6": 0.6418437361717224, "step": 4350 }, { "epoch": 1.0034594594594595, "grad_norm": 27.644145965576172, "learning_rate": 4.495386362638331e-06, "loss": 9.886016845703125, "loss_alignment": 0.83203125, "loss_alignment_w": 0.416015625, "loss_sft": 0.20211977884173393, "loss_total_v6": 0.6178760007023811, "step": 4351 }, { "epoch": 1.00369009009009, "grad_norm": 24.485130310058594, "learning_rate": 4.493382844313826e-06, "loss": 9.684837341308594, "loss_alignment": 0.8330078125, "loss_alignment_w": 0.41650390625, "loss_sft": 0.18872218765318394, "loss_total_v6": 0.6053023859858513, "step": 4352 }, { "epoch": 1.0039207207207208, "grad_norm": 27.965707778930664, "learning_rate": 4.491379408173917e-06, "loss": 10.234840393066406, "loss_alignment": 0.84716796875, "loss_alignment_w": 0.423583984375, "loss_sft": 0.21581890620291233, "loss_total_v6": 0.6396775618195534, "step": 4353 }, { "epoch": 1.0041513513513514, "grad_norm": 29.32892608642578, "learning_rate": 4.4893760545436074e-06, "loss": 10.451337814331055, "loss_alignment": 0.84716796875, "loss_alignment_w": 0.423583984375, "loss_sft": 0.2298077493906021, "loss_total_v6": 0.6532086357474327, "step": 4354 }, { "epoch": 1.004381981981982, "grad_norm": 25.65121841430664, "learning_rate": 4.487372783747884e-06, "loss": 9.914344787597656, "loss_alignment": 0.82861328125, "loss_alignment_w": 0.414306640625, "loss_sft": 0.20579766016453505, "loss_total_v6": 0.6196465268731117, "step": 4355 }, { "epoch": 1.0046126126126127, "grad_norm": 28.56806755065918, "learning_rate": 4.485369596111724e-06, "loss": 10.457588195800781, "loss_alignment": 0.84619140625, "loss_alignment_w": 0.423095703125, "loss_sft": 0.23169378004968166, "loss_total_v6": 0.6535992920398712, "step": 4356 }, { "epoch": 1.0048432432432433, "grad_norm": 27.41791343688965, "learning_rate": 4.483366491960088e-06, "loss": 10.017480850219727, "loss_alignment": 0.8564453125, "loss_alignment_w": 0.42822265625, "loss_sft": 0.19828188605606556, "loss_total_v6": 0.6260925456881523, "step": 4357 }, { "epoch": 1.0050738738738738, "grad_norm": 29.185619354248047, "learning_rate": 4.4813634716179225e-06, "loss": 10.135011672973633, "loss_alignment": 0.87060546875, "loss_alignment_w": 0.435302734375, "loss_sft": 0.1986238043755293, "loss_total_v6": 0.6334382593631744, "step": 4358 }, { "epoch": 1.0053045045045046, "grad_norm": 28.24945068359375, "learning_rate": 4.479360535410164e-06, "loss": 9.745054244995117, "loss_alignment": 0.8447265625, "loss_alignment_w": 0.42236328125, "loss_sft": 0.1870993161574006, "loss_total_v6": 0.609065867960453, "step": 4359 }, { "epoch": 1.0055351351351351, "grad_norm": 31.122753143310547, "learning_rate": 4.477357683661734e-06, "loss": 10.32841682434082, "loss_alignment": 0.86962890625, "loss_alignment_w": 0.434814453125, "loss_sft": 0.2117033787071705, "loss_total_v6": 0.6455260068178177, "step": 4360 }, { "epoch": 1.0057657657657657, "grad_norm": 25.764745712280273, "learning_rate": 4.4753549166975375e-06, "loss": 10.317808151245117, "loss_alignment": 0.84912109375, "loss_alignment_w": 0.424560546875, "loss_sft": 0.22076028026640415, "loss_total_v6": 0.6448630541563034, "step": 4361 }, { "epoch": 1.0059963963963965, "grad_norm": 26.106689453125, "learning_rate": 4.4733522348424695e-06, "loss": 9.864042282104492, "loss_alignment": 0.82470703125, "loss_alignment_w": 0.412353515625, "loss_sft": 0.20451539754867554, "loss_total_v6": 0.6165026873350143, "step": 4362 }, { "epoch": 1.006227027027027, "grad_norm": 26.671672821044922, "learning_rate": 4.471349638421408e-06, "loss": 10.331052780151367, "loss_alignment": 0.8447265625, "loss_alignment_w": 0.42236328125, "loss_sft": 0.22273245081305504, "loss_total_v6": 0.6456908136606216, "step": 4363 }, { "epoch": 1.0064576576576576, "grad_norm": 29.323942184448242, "learning_rate": 4.469347127759222e-06, "loss": 9.489168167114258, "loss_alignment": 0.8232421875, "loss_alignment_w": 0.41162109375, "loss_sft": 0.18129934230819345, "loss_total_v6": 0.5930730476975441, "step": 4364 }, { "epoch": 1.0066882882882884, "grad_norm": 35.93355941772461, "learning_rate": 4.467344703180761e-06, "loss": 10.188526153564453, "loss_alignment": 0.86328125, "loss_alignment_w": 0.431640625, "loss_sft": 0.20442508719861507, "loss_total_v6": 0.6367828920483589, "step": 4365 }, { "epoch": 1.006918918918919, "grad_norm": 29.99700927734375, "learning_rate": 4.465342365010863e-06, "loss": 9.948915481567383, "loss_alignment": 0.83447265625, "loss_alignment_w": 0.417236328125, "loss_sft": 0.20388421975076199, "loss_total_v6": 0.6218072101473808, "step": 4366 }, { "epoch": 1.0071495495495495, "grad_norm": 28.382848739624023, "learning_rate": 4.463340113574351e-06, "loss": 9.812334060668945, "loss_alignment": 0.845703125, "loss_alignment_w": 0.4228515625, "loss_sft": 0.19176210835576057, "loss_total_v6": 0.6132708862423897, "step": 4367 }, { "epoch": 1.0073801801801803, "grad_norm": 26.61302375793457, "learning_rate": 4.461337949196036e-06, "loss": 10.256671905517578, "loss_alignment": 0.859375, "loss_alignment_w": 0.4296875, "loss_sft": 0.2105457279831171, "loss_total_v6": 0.6410419642925262, "step": 4368 }, { "epoch": 1.0076108108108108, "grad_norm": 25.962081909179688, "learning_rate": 4.459335872200716e-06, "loss": 10.645026206970215, "loss_alignment": 0.8798828125, "loss_alignment_w": 0.43994140625, "loss_sft": 0.22364847175776958, "loss_total_v6": 0.6653141379356384, "step": 4369 }, { "epoch": 1.0078414414414414, "grad_norm": 47.5501594543457, "learning_rate": 4.45733388291317e-06, "loss": 9.888208389282227, "loss_alignment": 0.873046875, "loss_alignment_w": 0.4365234375, "loss_sft": 0.18156589940190315, "loss_total_v6": 0.6180130317807198, "step": 4370 }, { "epoch": 1.0080720720720722, "grad_norm": 28.755756378173828, "learning_rate": 4.4553319816581635e-06, "loss": 9.794998168945312, "loss_alignment": 0.837890625, "loss_alignment_w": 0.4189453125, "loss_sft": 0.19281483441591263, "loss_total_v6": 0.612187385559082, "step": 4371 }, { "epoch": 1.0083027027027027, "grad_norm": 24.708473205566406, "learning_rate": 4.453330168760451e-06, "loss": 10.080850601196289, "loss_alignment": 0.8798828125, "loss_alignment_w": 0.43994140625, "loss_sft": 0.19082890823483467, "loss_total_v6": 0.6300531402230263, "step": 4372 }, { "epoch": 1.0085333333333333, "grad_norm": 25.995325088500977, "learning_rate": 4.451328444544774e-06, "loss": 10.191169738769531, "loss_alignment": 0.84033203125, "loss_alignment_w": 0.420166015625, "loss_sft": 0.21620226837694645, "loss_total_v6": 0.6369480937719345, "step": 4373 }, { "epoch": 1.008763963963964, "grad_norm": 29.1247501373291, "learning_rate": 4.449326809335853e-06, "loss": 9.750102043151855, "loss_alignment": 0.8193359375, "loss_alignment_w": 0.40966796875, "loss_sft": 0.19994233176112175, "loss_total_v6": 0.6093814224004745, "step": 4374 }, { "epoch": 1.0089945945945946, "grad_norm": 29.50464630126953, "learning_rate": 4.447325263458401e-06, "loss": 10.174408912658691, "loss_alignment": 0.85693359375, "loss_alignment_w": 0.428466796875, "loss_sft": 0.20801358949393034, "loss_total_v6": 0.6359005495905876, "step": 4375 }, { "epoch": 1.0092252252252252, "grad_norm": 28.130512237548828, "learning_rate": 4.445323807237112e-06, "loss": 10.092361450195312, "loss_alignment": 0.8359375, "loss_alignment_w": 0.41796875, "loss_sft": 0.21297170594334602, "loss_total_v6": 0.6307726204395294, "step": 4376 }, { "epoch": 1.009455855855856, "grad_norm": 28.661211013793945, "learning_rate": 4.443322440996663e-06, "loss": 9.670989990234375, "loss_alignment": 0.83984375, "loss_alignment_w": 0.419921875, "loss_sft": 0.18518641777336597, "loss_total_v6": 0.6044368892908096, "step": 4377 }, { "epoch": 1.0096864864864865, "grad_norm": 28.108407974243164, "learning_rate": 4.441321165061726e-06, "loss": 9.80009651184082, "loss_alignment": 0.85107421875, "loss_alignment_w": 0.425537109375, "loss_sft": 0.18654161598533392, "loss_total_v6": 0.6125060021877289, "step": 4378 }, { "epoch": 1.009917117117117, "grad_norm": 36.722476959228516, "learning_rate": 4.439319979756951e-06, "loss": 9.726188659667969, "loss_alignment": 0.85595703125, "loss_alignment_w": 0.427978515625, "loss_sft": 0.1794963264837861, "loss_total_v6": 0.6078868210315704, "step": 4379 }, { "epoch": 1.0101477477477478, "grad_norm": 29.706451416015625, "learning_rate": 4.4373188854069736e-06, "loss": 10.364200592041016, "loss_alignment": 0.8583984375, "loss_alignment_w": 0.42919921875, "loss_sft": 0.21825812757015228, "loss_total_v6": 0.6477625146508217, "step": 4380 }, { "epoch": 1.0103783783783784, "grad_norm": 24.227203369140625, "learning_rate": 4.435317882336416e-06, "loss": 10.343664169311523, "loss_alignment": 0.8671875, "loss_alignment_w": 0.43359375, "loss_sft": 0.21381600387394428, "loss_total_v6": 0.646478958427906, "step": 4381 }, { "epoch": 1.010609009009009, "grad_norm": 38.07234573364258, "learning_rate": 4.433316970869885e-06, "loss": 9.79226303100586, "loss_alignment": 0.87060546875, "loss_alignment_w": 0.435302734375, "loss_sft": 0.1778581477701664, "loss_total_v6": 0.6120164543390274, "step": 4382 }, { "epoch": 1.0108396396396397, "grad_norm": 25.583877563476562, "learning_rate": 4.431316151331976e-06, "loss": 10.248800277709961, "loss_alignment": 0.8720703125, "loss_alignment_w": 0.43603515625, "loss_sft": 0.2042859774082899, "loss_total_v6": 0.6405500397086143, "step": 4383 }, { "epoch": 1.0110702702702703, "grad_norm": 25.1676025390625, "learning_rate": 4.429315424047264e-06, "loss": 9.862373352050781, "loss_alignment": 0.857421875, "loss_alignment_w": 0.4287109375, "loss_sft": 0.1864209845662117, "loss_total_v6": 0.6163983792066574, "step": 4384 }, { "epoch": 1.0113009009009009, "grad_norm": 28.010740280151367, "learning_rate": 4.4273147893403126e-06, "loss": 9.929437637329102, "loss_alignment": 0.84326171875, "loss_alignment_w": 0.421630859375, "loss_sft": 0.19900478795170784, "loss_total_v6": 0.6205898746848106, "step": 4385 }, { "epoch": 1.0115315315315316, "grad_norm": 26.464881896972656, "learning_rate": 4.425314247535668e-06, "loss": 9.759136199951172, "loss_alignment": 0.8115234375, "loss_alignment_w": 0.40576171875, "loss_sft": 0.20531335473060608, "loss_total_v6": 0.6099459603428841, "step": 4386 }, { "epoch": 1.0117621621621622, "grad_norm": 29.598039627075195, "learning_rate": 4.423313798957862e-06, "loss": 10.389589309692383, "loss_alignment": 0.849609375, "loss_alignment_w": 0.4248046875, "loss_sft": 0.22423946112394333, "loss_total_v6": 0.6493493095040321, "step": 4387 }, { "epoch": 1.0119927927927928, "grad_norm": 24.933273315429688, "learning_rate": 4.421313443931416e-06, "loss": 10.052627563476562, "loss_alignment": 0.84521484375, "loss_alignment_w": 0.422607421875, "loss_sft": 0.20507140643894672, "loss_total_v6": 0.6282891854643822, "step": 4388 }, { "epoch": 1.0122234234234235, "grad_norm": 28.00162124633789, "learning_rate": 4.41931318278083e-06, "loss": 9.926828384399414, "loss_alignment": 0.84130859375, "loss_alignment_w": 0.420654296875, "loss_sft": 0.1978803426027298, "loss_total_v6": 0.6204267367720604, "step": 4389 }, { "epoch": 1.012454054054054, "grad_norm": 28.175077438354492, "learning_rate": 4.417313015830589e-06, "loss": 9.832383155822754, "loss_alignment": 0.83349609375, "loss_alignment_w": 0.416748046875, "loss_sft": 0.1977148875594139, "loss_total_v6": 0.6145239621400833, "step": 4390 }, { "epoch": 1.0126846846846846, "grad_norm": 25.82244110107422, "learning_rate": 4.415312943405168e-06, "loss": 10.017605781555176, "loss_alignment": 0.85009765625, "loss_alignment_w": 0.425048828125, "loss_sft": 0.20027335174381733, "loss_total_v6": 0.6261003389954567, "step": 4391 }, { "epoch": 1.0129153153153154, "grad_norm": 27.17519187927246, "learning_rate": 4.41331296582902e-06, "loss": 9.73260498046875, "loss_alignment": 0.8349609375, "loss_alignment_w": 0.41748046875, "loss_sft": 0.19028854742646217, "loss_total_v6": 0.6082878261804581, "step": 4392 }, { "epoch": 1.013145945945946, "grad_norm": 28.96082305908203, "learning_rate": 4.4113130834265875e-06, "loss": 9.995560646057129, "loss_alignment": 0.84326171875, "loss_alignment_w": 0.421630859375, "loss_sft": 0.20397666655480862, "loss_total_v6": 0.62472253292799, "step": 4393 }, { "epoch": 1.0133765765765765, "grad_norm": 39.16897964477539, "learning_rate": 4.409313296522298e-06, "loss": 10.122393608093262, "loss_alignment": 0.84423828125, "loss_alignment_w": 0.422119140625, "loss_sft": 0.21004221681505442, "loss_total_v6": 0.63264961540699, "step": 4394 }, { "epoch": 1.0136072072072073, "grad_norm": 27.04820442199707, "learning_rate": 4.407313605440558e-06, "loss": 9.95826530456543, "loss_alignment": 0.82763671875, "loss_alignment_w": 0.413818359375, "loss_sft": 0.20803917944431305, "loss_total_v6": 0.6223916038870811, "step": 4395 }, { "epoch": 1.0138378378378379, "grad_norm": 26.527740478515625, "learning_rate": 4.405314010505764e-06, "loss": 10.405729293823242, "loss_alignment": 0.8408203125, "loss_alignment_w": 0.42041015625, "loss_sft": 0.23148907162249088, "loss_total_v6": 0.6503580883145332, "step": 4396 }, { "epoch": 1.0140684684684684, "grad_norm": 28.284318923950195, "learning_rate": 4.403314512042293e-06, "loss": 9.886220932006836, "loss_alignment": 0.830078125, "loss_alignment_w": 0.4150390625, "loss_sft": 0.20141539722681046, "loss_total_v6": 0.6178887560963631, "step": 4397 }, { "epoch": 1.014299099099099, "grad_norm": 28.376150131225586, "learning_rate": 4.401315110374512e-06, "loss": 10.34051513671875, "loss_alignment": 0.8427734375, "loss_alignment_w": 0.42138671875, "loss_sft": 0.2260094452649355, "loss_total_v6": 0.6462822332978249, "step": 4398 }, { "epoch": 1.0145297297297298, "grad_norm": 25.55301856994629, "learning_rate": 4.399315805826765e-06, "loss": 9.904358863830566, "loss_alignment": 0.8203125, "loss_alignment_w": 0.41015625, "loss_sft": 0.2092934064567089, "loss_total_v6": 0.6190224438905716, "step": 4399 }, { "epoch": 1.0147603603603603, "grad_norm": 28.40309715270996, "learning_rate": 4.397316598723385e-06, "loss": 10.55699348449707, "loss_alignment": 0.859375, "loss_alignment_w": 0.4296875, "loss_sft": 0.2284155786037445, "loss_total_v6": 0.6598120629787445, "step": 4400 }, { "epoch": 1.0149909909909909, "grad_norm": 29.916234970092773, "learning_rate": 4.3953174893886885e-06, "loss": 10.092058181762695, "loss_alignment": 0.8525390625, "loss_alignment_w": 0.42626953125, "loss_sft": 0.20375169441103935, "loss_total_v6": 0.6307536512613297, "step": 4401 }, { "epoch": 1.0152216216216217, "grad_norm": 55.66022872924805, "learning_rate": 4.393318478146975e-06, "loss": 9.674736022949219, "loss_alignment": 0.8232421875, "loss_alignment_w": 0.41162109375, "loss_sft": 0.1923785451799631, "loss_total_v6": 0.6046710014343262, "step": 4402 }, { "epoch": 1.0154522522522522, "grad_norm": 28.855989456176758, "learning_rate": 4.391319565322528e-06, "loss": 9.893288612365723, "loss_alignment": 0.83642578125, "loss_alignment_w": 0.418212890625, "loss_sft": 0.20023969002068043, "loss_total_v6": 0.6183305233716965, "step": 4403 }, { "epoch": 1.0156828828828828, "grad_norm": 27.928104400634766, "learning_rate": 4.389320751239617e-06, "loss": 9.99981689453125, "loss_alignment": 0.81884765625, "loss_alignment_w": 0.409423828125, "loss_sft": 0.21518328413367271, "loss_total_v6": 0.6249885633587837, "step": 4404 }, { "epoch": 1.0159135135135136, "grad_norm": 27.495697021484375, "learning_rate": 4.387322036222493e-06, "loss": 10.080047607421875, "loss_alignment": 0.86474609375, "loss_alignment_w": 0.432373046875, "loss_sft": 0.19779779016971588, "loss_total_v6": 0.6300030052661896, "step": 4405 }, { "epoch": 1.0161441441441441, "grad_norm": 23.0249080657959, "learning_rate": 4.385323420595395e-06, "loss": 10.231101989746094, "loss_alignment": 0.869140625, "loss_alignment_w": 0.4345703125, "loss_sft": 0.2047972772270441, "loss_total_v6": 0.6394438743591309, "step": 4406 }, { "epoch": 1.0163747747747747, "grad_norm": 27.367868423461914, "learning_rate": 4.383324904682537e-06, "loss": 9.96391487121582, "loss_alignment": 0.88525390625, "loss_alignment_w": 0.442626953125, "loss_sft": 0.18001089245080948, "loss_total_v6": 0.6227446496486664, "step": 4407 }, { "epoch": 1.0166054054054054, "grad_norm": 29.443588256835938, "learning_rate": 4.381326488808129e-06, "loss": 10.427215576171875, "loss_alignment": 0.88525390625, "loss_alignment_w": 0.442626953125, "loss_sft": 0.20960805378854275, "loss_total_v6": 0.651700958609581, "step": 4408 }, { "epoch": 1.016836036036036, "grad_norm": 24.878982543945312, "learning_rate": 4.379328173296356e-06, "loss": 9.812591552734375, "loss_alignment": 0.87109375, "loss_alignment_w": 0.435546875, "loss_sft": 0.1779995560646057, "loss_total_v6": 0.613287016749382, "step": 4409 }, { "epoch": 1.0170666666666666, "grad_norm": 29.646774291992188, "learning_rate": 4.37732995847139e-06, "loss": 9.756523132324219, "loss_alignment": 0.859375, "loss_alignment_w": 0.4296875, "loss_sft": 0.18117858842015266, "loss_total_v6": 0.6097827106714249, "step": 4410 }, { "epoch": 1.0172972972972973, "grad_norm": 34.88642120361328, "learning_rate": 4.375331844657385e-06, "loss": 10.270608901977539, "loss_alignment": 0.87158203125, "loss_alignment_w": 0.435791015625, "loss_sft": 0.2051454372704029, "loss_total_v6": 0.641913004219532, "step": 4411 }, { "epoch": 1.017527927927928, "grad_norm": 28.525766372680664, "learning_rate": 4.373333832178478e-06, "loss": 9.747289657592773, "loss_alignment": 0.82763671875, "loss_alignment_w": 0.413818359375, "loss_sft": 0.19627231173217297, "loss_total_v6": 0.6092056557536125, "step": 4412 }, { "epoch": 1.0177585585585585, "grad_norm": 27.599077224731445, "learning_rate": 4.3713359213587965e-06, "loss": 9.967554092407227, "loss_alignment": 0.87744140625, "loss_alignment_w": 0.438720703125, "loss_sft": 0.18509064055979252, "loss_total_v6": 0.6229721307754517, "step": 4413 }, { "epoch": 1.0179891891891892, "grad_norm": 24.744338989257812, "learning_rate": 4.3693381125224415e-06, "loss": 9.537611961364746, "loss_alignment": 0.83447265625, "loss_alignment_w": 0.417236328125, "loss_sft": 0.17909333668649197, "loss_total_v6": 0.5961007624864578, "step": 4414 }, { "epoch": 1.0182198198198198, "grad_norm": 28.2424373626709, "learning_rate": 4.367340405993505e-06, "loss": 10.13135814666748, "loss_alignment": 0.84228515625, "loss_alignment_w": 0.421142578125, "loss_sft": 0.21299806702882051, "loss_total_v6": 0.6332098692655563, "step": 4415 }, { "epoch": 1.0184504504504504, "grad_norm": 27.08316993713379, "learning_rate": 4.365342802096057e-06, "loss": 9.781064987182617, "loss_alignment": 0.84716796875, "loss_alignment_w": 0.423583984375, "loss_sft": 0.1879767468199134, "loss_total_v6": 0.6113165840506554, "step": 4416 }, { "epoch": 1.0186810810810811, "grad_norm": 28.64457130432129, "learning_rate": 4.363345301154154e-06, "loss": 9.843071937561035, "loss_alignment": 0.83251953125, "loss_alignment_w": 0.416259765625, "loss_sft": 0.1993442326784134, "loss_total_v6": 0.6151919737458229, "step": 4417 }, { "epoch": 1.0189117117117117, "grad_norm": 25.225332260131836, "learning_rate": 4.361347903491836e-06, "loss": 9.915721893310547, "loss_alignment": 0.853515625, "loss_alignment_w": 0.4267578125, "loss_sft": 0.19280692376196384, "loss_total_v6": 0.6197325959801674, "step": 4418 }, { "epoch": 1.0191423423423422, "grad_norm": 30.49974822998047, "learning_rate": 4.359350609433126e-06, "loss": 9.713412284851074, "loss_alignment": 0.830078125, "loss_alignment_w": 0.4150390625, "loss_sft": 0.1918661082163453, "loss_total_v6": 0.6070882827043533, "step": 4419 }, { "epoch": 1.019372972972973, "grad_norm": 28.12120819091797, "learning_rate": 4.357353419302028e-06, "loss": 9.829415321350098, "loss_alignment": 0.85302734375, "loss_alignment_w": 0.426513671875, "loss_sft": 0.18823674693703651, "loss_total_v6": 0.6143384575843811, "step": 4420 }, { "epoch": 1.0196036036036036, "grad_norm": 29.03469467163086, "learning_rate": 4.355356333422529e-06, "loss": 9.842811584472656, "loss_alignment": 0.86083984375, "loss_alignment_w": 0.430419921875, "loss_sft": 0.1854119896888733, "loss_total_v6": 0.6151757687330246, "step": 4421 }, { "epoch": 1.0198342342342341, "grad_norm": 24.520721435546875, "learning_rate": 4.353359352118607e-06, "loss": 9.680195808410645, "loss_alignment": 0.830078125, "loss_alignment_w": 0.4150390625, "loss_sft": 0.18867618963122368, "loss_total_v6": 0.6050122380256653, "step": 4422 }, { "epoch": 1.020064864864865, "grad_norm": 30.12879180908203, "learning_rate": 4.351362475714214e-06, "loss": 10.11980152130127, "loss_alignment": 0.84765625, "loss_alignment_w": 0.423828125, "loss_sft": 0.2088425438851118, "loss_total_v6": 0.6324875801801682, "step": 4423 }, { "epoch": 1.0202954954954955, "grad_norm": 27.656591415405273, "learning_rate": 4.349365704533285e-06, "loss": 10.280292510986328, "loss_alignment": 0.8583984375, "loss_alignment_w": 0.42919921875, "loss_sft": 0.21418887376785278, "loss_total_v6": 0.6425183266401291, "step": 4424 }, { "epoch": 1.020526126126126, "grad_norm": 28.57402229309082, "learning_rate": 4.347369038899744e-06, "loss": 9.945096015930176, "loss_alignment": 0.83544921875, "loss_alignment_w": 0.417724609375, "loss_sft": 0.20448476448655128, "loss_total_v6": 0.6215684935450554, "step": 4425 }, { "epoch": 1.0207567567567568, "grad_norm": 29.89179229736328, "learning_rate": 4.345372479137491e-06, "loss": 10.143695831298828, "loss_alignment": 0.837890625, "loss_alignment_w": 0.4189453125, "loss_sft": 0.21373865753412247, "loss_total_v6": 0.6339809820055962, "step": 4426 }, { "epoch": 1.0209873873873874, "grad_norm": 31.52591896057129, "learning_rate": 4.343376025570416e-06, "loss": 9.942024230957031, "loss_alignment": 0.87548828125, "loss_alignment_w": 0.437744140625, "loss_sft": 0.18405963946133852, "loss_total_v6": 0.6213765442371368, "step": 4427 }, { "epoch": 1.021218018018018, "grad_norm": 23.170488357543945, "learning_rate": 4.34137967852239e-06, "loss": 9.781634330749512, "loss_alignment": 0.8369140625, "loss_alignment_w": 0.41845703125, "loss_sft": 0.19284932874143124, "loss_total_v6": 0.6113521233201027, "step": 4428 }, { "epoch": 1.0214486486486487, "grad_norm": 25.440580368041992, "learning_rate": 4.339383438317261e-06, "loss": 9.990673065185547, "loss_alignment": 0.83642578125, "loss_alignment_w": 0.418212890625, "loss_sft": 0.20544122159481049, "loss_total_v6": 0.6244170516729355, "step": 4429 }, { "epoch": 1.0216792792792793, "grad_norm": 32.5225944519043, "learning_rate": 4.337387305278864e-06, "loss": 10.070684432983398, "loss_alignment": 0.8349609375, "loss_alignment_w": 0.41748046875, "loss_sft": 0.21202885173261166, "loss_total_v6": 0.629417784512043, "step": 4430 }, { "epoch": 1.0219099099099098, "grad_norm": 28.692405700683594, "learning_rate": 4.335391279731017e-06, "loss": 9.651230812072754, "loss_alignment": 0.828125, "loss_alignment_w": 0.4140625, "loss_sft": 0.1906195655465126, "loss_total_v6": 0.6032019406557083, "step": 4431 }, { "epoch": 1.0221405405405406, "grad_norm": 30.735538482666016, "learning_rate": 4.333395361997521e-06, "loss": 9.686299324035645, "loss_alignment": 0.83544921875, "loss_alignment_w": 0.417724609375, "loss_sft": 0.18835571967065334, "loss_total_v6": 0.6053936779499054, "step": 4432 }, { "epoch": 1.0223711711711712, "grad_norm": 28.748992919921875, "learning_rate": 4.3313995524021575e-06, "loss": 10.094060897827148, "loss_alignment": 0.85107421875, "loss_alignment_w": 0.425537109375, "loss_sft": 0.20616563223302364, "loss_total_v6": 0.6308787688612938, "step": 4433 }, { "epoch": 1.0226018018018017, "grad_norm": 28.19923973083496, "learning_rate": 4.329403851268692e-06, "loss": 9.873555183410645, "loss_alignment": 0.86865234375, "loss_alignment_w": 0.434326171875, "loss_sft": 0.18231325410306454, "loss_total_v6": 0.6170971691608429, "step": 4434 }, { "epoch": 1.0228324324324325, "grad_norm": 26.540203094482422, "learning_rate": 4.327408258920871e-06, "loss": 9.860650062561035, "loss_alignment": 0.8583984375, "loss_alignment_w": 0.42919921875, "loss_sft": 0.18783905915915966, "loss_total_v6": 0.6162905991077423, "step": 4435 }, { "epoch": 1.023063063063063, "grad_norm": 23.462902069091797, "learning_rate": 4.325412775682422e-06, "loss": 9.466194152832031, "loss_alignment": 0.82861328125, "loss_alignment_w": 0.414306640625, "loss_sft": 0.17919211275875568, "loss_total_v6": 0.5916371643543243, "step": 4436 }, { "epoch": 1.0232936936936936, "grad_norm": 24.349266052246094, "learning_rate": 4.323417401877059e-06, "loss": 10.084047317504883, "loss_alignment": 0.8603515625, "loss_alignment_w": 0.43017578125, "loss_sft": 0.2002907693386078, "loss_total_v6": 0.630252942442894, "step": 4437 }, { "epoch": 1.0235243243243244, "grad_norm": 25.43955421447754, "learning_rate": 4.321422137828479e-06, "loss": 9.369820594787598, "loss_alignment": 0.8154296875, "loss_alignment_w": 0.40771484375, "loss_sft": 0.17783790081739426, "loss_total_v6": 0.5856137871742249, "step": 4438 }, { "epoch": 1.023754954954955, "grad_norm": 29.332117080688477, "learning_rate": 4.319426983860353e-06, "loss": 10.22396183013916, "loss_alignment": 0.859375, "loss_alignment_w": 0.4296875, "loss_sft": 0.21001203171908855, "loss_total_v6": 0.6389976292848587, "step": 4439 }, { "epoch": 1.0239855855855855, "grad_norm": 26.6148681640625, "learning_rate": 4.3174319402963436e-06, "loss": 9.8233642578125, "loss_alignment": 0.82666015625, "loss_alignment_w": 0.413330078125, "loss_sft": 0.2005539145320654, "loss_total_v6": 0.6139602959156036, "step": 4440 }, { "epoch": 1.0242162162162163, "grad_norm": 26.47930335998535, "learning_rate": 4.3154370074600874e-06, "loss": 10.00228214263916, "loss_alignment": 0.845703125, "loss_alignment_w": 0.4228515625, "loss_sft": 0.20253522880375385, "loss_total_v6": 0.6251426637172699, "step": 4441 }, { "epoch": 1.0244468468468468, "grad_norm": 45.466121673583984, "learning_rate": 4.313442185675211e-06, "loss": 10.129373550415039, "loss_alignment": 0.82666015625, "loss_alignment_w": 0.413330078125, "loss_sft": 0.21995407156646252, "loss_total_v6": 0.6330857947468758, "step": 4442 }, { "epoch": 1.0246774774774774, "grad_norm": 29.148395538330078, "learning_rate": 4.311447475265315e-06, "loss": 10.179306030273438, "loss_alignment": 0.87841796875, "loss_alignment_w": 0.439208984375, "loss_sft": 0.19699759501963854, "loss_total_v6": 0.6362065821886063, "step": 4443 }, { "epoch": 1.0249081081081082, "grad_norm": 28.247713088989258, "learning_rate": 4.30945287655399e-06, "loss": 9.856674194335938, "loss_alignment": 0.8564453125, "loss_alignment_w": 0.42822265625, "loss_sft": 0.18783476389944553, "loss_total_v6": 0.6160421594977379, "step": 4444 }, { "epoch": 1.0251387387387387, "grad_norm": 29.840282440185547, "learning_rate": 4.3074583898648016e-06, "loss": 10.145047187805176, "loss_alignment": 0.84375, "loss_alignment_w": 0.421875, "loss_sft": 0.21261773444712162, "loss_total_v6": 0.6340654790401459, "step": 4445 }, { "epoch": 1.0253693693693693, "grad_norm": 25.049880981445312, "learning_rate": 4.305464015521299e-06, "loss": 9.971309661865234, "loss_alignment": 0.830078125, "loss_alignment_w": 0.4150390625, "loss_sft": 0.20847292058169842, "loss_total_v6": 0.6232068166136742, "step": 4446 }, { "epoch": 1.0256, "grad_norm": 27.88433265686035, "learning_rate": 4.303469753847018e-06, "loss": 9.937204360961914, "loss_alignment": 0.8388671875, "loss_alignment_w": 0.41943359375, "loss_sft": 0.20055831782519817, "loss_total_v6": 0.6210752949118614, "step": 4447 }, { "epoch": 1.0258306306306306, "grad_norm": 27.35247802734375, "learning_rate": 4.301475605165471e-06, "loss": 10.313421249389648, "loss_alignment": 0.85986328125, "loss_alignment_w": 0.429931640625, "loss_sft": 0.21554219350218773, "loss_total_v6": 0.6445888206362724, "step": 4448 }, { "epoch": 1.0260612612612612, "grad_norm": 26.65496253967285, "learning_rate": 4.299481569800151e-06, "loss": 10.082040786743164, "loss_alignment": 0.84619140625, "loss_alignment_w": 0.423095703125, "loss_sft": 0.2065740842372179, "loss_total_v6": 0.6301275491714478, "step": 4449 }, { "epoch": 1.026291891891892, "grad_norm": 24.782468795776367, "learning_rate": 4.297487648074538e-06, "loss": 9.748132705688477, "loss_alignment": 0.837890625, "loss_alignment_w": 0.4189453125, "loss_sft": 0.18945852480828762, "loss_total_v6": 0.6092583239078522, "step": 4450 }, { "epoch": 1.0265225225225225, "grad_norm": 27.285985946655273, "learning_rate": 4.295493840312087e-06, "loss": 9.97477912902832, "loss_alignment": 0.8623046875, "loss_alignment_w": 0.43115234375, "loss_sft": 0.19332417473196983, "loss_total_v6": 0.6234236732125282, "step": 4451 }, { "epoch": 1.026753153153153, "grad_norm": 32.379154205322266, "learning_rate": 4.293500146836241e-06, "loss": 9.982122421264648, "loss_alignment": 0.806640625, "loss_alignment_w": 0.4033203125, "loss_sft": 0.22068443521857262, "loss_total_v6": 0.6238826885819435, "step": 4452 }, { "epoch": 1.0269837837837839, "grad_norm": 29.032508850097656, "learning_rate": 4.291506567970422e-06, "loss": 9.721622467041016, "loss_alignment": 0.8583984375, "loss_alignment_w": 0.42919921875, "loss_sft": 0.1792413927614689, "loss_total_v6": 0.6076014190912247, "step": 4453 }, { "epoch": 1.0272144144144144, "grad_norm": 29.842069625854492, "learning_rate": 4.289513104038032e-06, "loss": 10.114648818969727, "loss_alignment": 0.849609375, "loss_alignment_w": 0.4248046875, "loss_sft": 0.2064758539199829, "loss_total_v6": 0.6321655288338661, "step": 4454 }, { "epoch": 1.027445045045045, "grad_norm": 25.768878936767578, "learning_rate": 4.287519755362453e-06, "loss": 9.902788162231445, "loss_alignment": 0.85595703125, "loss_alignment_w": 0.427978515625, "loss_sft": 0.18986237049102783, "loss_total_v6": 0.6189242452383041, "step": 4455 }, { "epoch": 1.0276756756756757, "grad_norm": 24.46603012084961, "learning_rate": 4.285526522267052e-06, "loss": 9.553237915039062, "loss_alignment": 0.8212890625, "loss_alignment_w": 0.41064453125, "loss_sft": 0.18580719083547592, "loss_total_v6": 0.5970773473381996, "step": 4456 }, { "epoch": 1.0279063063063063, "grad_norm": 26.85508155822754, "learning_rate": 4.283533405075179e-06, "loss": 9.892683029174805, "loss_alignment": 0.83349609375, "loss_alignment_w": 0.416748046875, "loss_sft": 0.2006443478167057, "loss_total_v6": 0.6182926595211029, "step": 4457 }, { "epoch": 1.0281369369369369, "grad_norm": 27.578720092773438, "learning_rate": 4.2815404041101575e-06, "loss": 10.251710891723633, "loss_alignment": 0.87353515625, "loss_alignment_w": 0.436767578125, "loss_sft": 0.20437633246183395, "loss_total_v6": 0.6407319381833076, "step": 4458 }, { "epoch": 1.0283675675675676, "grad_norm": 28.251985549926758, "learning_rate": 4.279547519695299e-06, "loss": 9.659151077270508, "loss_alignment": 0.83056640625, "loss_alignment_w": 0.415283203125, "loss_sft": 0.1880170190706849, "loss_total_v6": 0.6036969535052776, "step": 4459 }, { "epoch": 1.0285981981981982, "grad_norm": 31.1573486328125, "learning_rate": 4.277554752153895e-06, "loss": 9.737686157226562, "loss_alignment": 0.8359375, "loss_alignment_w": 0.41796875, "loss_sft": 0.1910028476268053, "loss_total_v6": 0.6086053848266602, "step": 4460 }, { "epoch": 1.0288288288288288, "grad_norm": 24.94127082824707, "learning_rate": 4.275562101809213e-06, "loss": 9.955645561218262, "loss_alignment": 0.83642578125, "loss_alignment_w": 0.418212890625, "loss_sft": 0.20338936150074005, "loss_total_v6": 0.6222278401255608, "step": 4461 }, { "epoch": 1.0290594594594595, "grad_norm": 33.23536682128906, "learning_rate": 4.273569568984508e-06, "loss": 10.668020248413086, "loss_alignment": 0.888671875, "loss_alignment_w": 0.4443359375, "loss_sft": 0.22326983511447906, "loss_total_v6": 0.6667512729763985, "step": 4462 }, { "epoch": 1.02929009009009, "grad_norm": 25.251628875732422, "learning_rate": 4.2715771540030136e-06, "loss": 10.101430892944336, "loss_alignment": 0.861328125, "loss_alignment_w": 0.4306640625, "loss_sft": 0.20044650696218014, "loss_total_v6": 0.6313394382596016, "step": 4463 }, { "epoch": 1.0295207207207207, "grad_norm": 27.869089126586914, "learning_rate": 4.269584857187942e-06, "loss": 9.600509643554688, "loss_alignment": 0.8203125, "loss_alignment_w": 0.41015625, "loss_sft": 0.1889448594301939, "loss_total_v6": 0.6000318825244904, "step": 4464 }, { "epoch": 1.0297513513513514, "grad_norm": 33.47806930541992, "learning_rate": 4.267592678862491e-06, "loss": 10.077657699584961, "loss_alignment": 0.833984375, "loss_alignment_w": 0.4169921875, "loss_sft": 0.2133497279137373, "loss_total_v6": 0.6298536285758018, "step": 4465 }, { "epoch": 1.029981981981982, "grad_norm": 28.611867904663086, "learning_rate": 4.265600619349832e-06, "loss": 9.963767051696777, "loss_alignment": 0.8447265625, "loss_alignment_w": 0.42236328125, "loss_sft": 0.20055521931499243, "loss_total_v6": 0.6227354034781456, "step": 4466 }, { "epoch": 1.0302126126126125, "grad_norm": 27.9197940826416, "learning_rate": 4.263608678973127e-06, "loss": 10.477542877197266, "loss_alignment": 0.87158203125, "loss_alignment_w": 0.435791015625, "loss_sft": 0.2187654823064804, "loss_total_v6": 0.6548464298248291, "step": 4467 }, { "epoch": 1.0304432432432433, "grad_norm": 28.543996810913086, "learning_rate": 4.2616168580555086e-06, "loss": 10.314717292785645, "loss_alignment": 0.85498046875, "loss_alignment_w": 0.427490234375, "loss_sft": 0.21887332946062088, "loss_total_v6": 0.6446698606014252, "step": 4468 }, { "epoch": 1.0306738738738739, "grad_norm": 36.99138641357422, "learning_rate": 4.259625156920096e-06, "loss": 9.650604248046875, "loss_alignment": 0.83935546875, "loss_alignment_w": 0.419677734375, "loss_sft": 0.18450741469860077, "loss_total_v6": 0.6031628176569939, "step": 4469 }, { "epoch": 1.0309045045045044, "grad_norm": 29.50480842590332, "learning_rate": 4.257633575889987e-06, "loss": 10.35603141784668, "loss_alignment": 0.89599609375, "loss_alignment_w": 0.447998046875, "loss_sft": 0.19928445294499397, "loss_total_v6": 0.6472519859671593, "step": 4470 }, { "epoch": 1.0311351351351352, "grad_norm": 31.297653198242188, "learning_rate": 4.255642115288262e-06, "loss": 9.163983345031738, "loss_alignment": 0.8154296875, "loss_alignment_w": 0.40771484375, "loss_sft": 0.16298941895365715, "loss_total_v6": 0.572748951613903, "step": 4471 }, { "epoch": 1.0313657657657658, "grad_norm": 26.209545135498047, "learning_rate": 4.25365077543798e-06, "loss": 10.43901252746582, "loss_alignment": 0.892578125, "loss_alignment_w": 0.4462890625, "loss_sft": 0.20679006353020668, "loss_total_v6": 0.6524382531642914, "step": 4472 }, { "epoch": 1.0315963963963963, "grad_norm": 28.40774917602539, "learning_rate": 4.25165955666218e-06, "loss": 9.601531028747559, "loss_alignment": 0.8056640625, "loss_alignment_w": 0.40283203125, "loss_sft": 0.1965922750532627, "loss_total_v6": 0.6000956892967224, "step": 4473 }, { "epoch": 1.031827027027027, "grad_norm": 29.428958892822266, "learning_rate": 4.249668459283882e-06, "loss": 9.977336883544922, "loss_alignment": 0.85205078125, "loss_alignment_w": 0.426025390625, "loss_sft": 0.19764968566596508, "loss_total_v6": 0.6235835403203964, "step": 4474 }, { "epoch": 1.0320576576576577, "grad_norm": 25.012407302856445, "learning_rate": 4.247677483626085e-06, "loss": 10.254656791687012, "loss_alignment": 0.861328125, "loss_alignment_w": 0.4306640625, "loss_sft": 0.21074026450514793, "loss_total_v6": 0.6409160494804382, "step": 4475 }, { "epoch": 1.0322882882882882, "grad_norm": 27.19542694091797, "learning_rate": 4.245686630011773e-06, "loss": 9.902698516845703, "loss_alignment": 0.84033203125, "loss_alignment_w": 0.420166015625, "loss_sft": 0.19864584505558014, "loss_total_v6": 0.6189186722040176, "step": 4476 }, { "epoch": 1.032518918918919, "grad_norm": 28.65757179260254, "learning_rate": 4.243695898763904e-06, "loss": 9.539745330810547, "loss_alignment": 0.85791015625, "loss_alignment_w": 0.428955078125, "loss_sft": 0.16833183076232672, "loss_total_v6": 0.5962340459227562, "step": 4477 }, { "epoch": 1.0327495495495496, "grad_norm": 28.12579345703125, "learning_rate": 4.2417052902054215e-06, "loss": 9.875656127929688, "loss_alignment": 0.8056640625, "loss_alignment_w": 0.40283203125, "loss_sft": 0.215006859973073, "loss_total_v6": 0.6172285377979279, "step": 4478 }, { "epoch": 1.0329801801801801, "grad_norm": 32.01594543457031, "learning_rate": 4.239714804659245e-06, "loss": 9.830856323242188, "loss_alignment": 0.84814453125, "loss_alignment_w": 0.424072265625, "loss_sft": 0.19110392779111862, "loss_total_v6": 0.6144285276532173, "step": 4479 }, { "epoch": 1.033210810810811, "grad_norm": 35.59156799316406, "learning_rate": 4.237724442448273e-06, "loss": 9.78817367553711, "loss_alignment": 0.8408203125, "loss_alignment_w": 0.42041015625, "loss_sft": 0.19072508439421654, "loss_total_v6": 0.6117608770728111, "step": 4480 }, { "epoch": 1.0334414414414415, "grad_norm": 28.06830596923828, "learning_rate": 4.235734203895391e-06, "loss": 9.563276290893555, "loss_alignment": 0.8427734375, "loss_alignment_w": 0.42138671875, "loss_sft": 0.17703524231910706, "loss_total_v6": 0.5977047979831696, "step": 4481 }, { "epoch": 1.033672072072072, "grad_norm": 25.573101043701172, "learning_rate": 4.23374408932346e-06, "loss": 9.664950370788574, "loss_alignment": 0.85107421875, "loss_alignment_w": 0.425537109375, "loss_sft": 0.17852225434035063, "loss_total_v6": 0.6040593683719635, "step": 4482 }, { "epoch": 1.0339027027027028, "grad_norm": 27.097131729125977, "learning_rate": 4.2317540990553175e-06, "loss": 9.90266227722168, "loss_alignment": 0.80908203125, "loss_alignment_w": 0.404541015625, "loss_sft": 0.2141617275774479, "loss_total_v6": 0.618916392326355, "step": 4483 }, { "epoch": 1.0341333333333333, "grad_norm": 23.607561111450195, "learning_rate": 4.229764233413788e-06, "loss": 10.283443450927734, "loss_alignment": 0.8876953125, "loss_alignment_w": 0.44384765625, "loss_sft": 0.1994321458041668, "loss_total_v6": 0.6427152156829834, "step": 4484 }, { "epoch": 1.034363963963964, "grad_norm": 26.467334747314453, "learning_rate": 4.227774492721667e-06, "loss": 10.531500816345215, "loss_alignment": 0.86376953125, "loss_alignment_w": 0.431884765625, "loss_sft": 0.22563211619853973, "loss_total_v6": 0.6582188010215759, "step": 4485 }, { "epoch": 1.0345945945945947, "grad_norm": 27.34777069091797, "learning_rate": 4.225784877301741e-06, "loss": 9.66050910949707, "loss_alignment": 0.837890625, "loss_alignment_w": 0.4189453125, "loss_sft": 0.18601139355450869, "loss_total_v6": 0.6037817895412445, "step": 4486 }, { "epoch": 1.0348252252252252, "grad_norm": 31.402273178100586, "learning_rate": 4.223795387476766e-06, "loss": 10.589521408081055, "loss_alignment": 0.875, "loss_alignment_w": 0.4375, "loss_sft": 0.2245892435312271, "loss_total_v6": 0.6618451252579689, "step": 4487 }, { "epoch": 1.0350558558558558, "grad_norm": 27.71827507019043, "learning_rate": 4.221806023569483e-06, "loss": 9.90948486328125, "loss_alignment": 0.83740234375, "loss_alignment_w": 0.418701171875, "loss_sft": 0.20047378912568092, "loss_total_v6": 0.6193428114056587, "step": 4488 }, { "epoch": 1.0352864864864866, "grad_norm": 27.14574432373047, "learning_rate": 4.219816785902609e-06, "loss": 10.299579620361328, "loss_alignment": 0.85986328125, "loss_alignment_w": 0.429931640625, "loss_sft": 0.2139294482767582, "loss_total_v6": 0.6437237709760666, "step": 4489 }, { "epoch": 1.0355171171171171, "grad_norm": 23.22188949584961, "learning_rate": 4.217827674798845e-06, "loss": 9.503559112548828, "loss_alignment": 0.83349609375, "loss_alignment_w": 0.416748046875, "loss_sft": 0.1782314833253622, "loss_total_v6": 0.5939724445343018, "step": 4490 }, { "epoch": 1.0357477477477477, "grad_norm": 27.28573989868164, "learning_rate": 4.2158386905808705e-06, "loss": 9.690247535705566, "loss_alignment": 0.85205078125, "loss_alignment_w": 0.426025390625, "loss_sft": 0.180271215736866, "loss_total_v6": 0.6056404784321785, "step": 4491 }, { "epoch": 1.0359783783783785, "grad_norm": 28.82509994506836, "learning_rate": 4.213849833571341e-06, "loss": 9.902308464050293, "loss_alignment": 0.8173828125, "loss_alignment_w": 0.40869140625, "loss_sft": 0.21061484515666962, "loss_total_v6": 0.6188942715525627, "step": 4492 }, { "epoch": 1.036209009009009, "grad_norm": 24.747432708740234, "learning_rate": 4.211861104092892e-06, "loss": 9.784727096557617, "loss_alignment": 0.8505859375, "loss_alignment_w": 0.42529296875, "loss_sft": 0.18553536012768745, "loss_total_v6": 0.6115454807877541, "step": 4493 }, { "epoch": 1.0364396396396396, "grad_norm": 26.914901733398438, "learning_rate": 4.209872502468143e-06, "loss": 9.44534683227539, "loss_alignment": 0.80224609375, "loss_alignment_w": 0.401123046875, "loss_sft": 0.18742581643164158, "loss_total_v6": 0.5903341397643089, "step": 4494 }, { "epoch": 1.0366702702702704, "grad_norm": 24.95761489868164, "learning_rate": 4.2078840290196854e-06, "loss": 9.602749824523926, "loss_alignment": 0.8095703125, "loss_alignment_w": 0.40478515625, "loss_sft": 0.19523412734270096, "loss_total_v6": 0.6001718491315842, "step": 4495 }, { "epoch": 1.036900900900901, "grad_norm": 24.903478622436523, "learning_rate": 4.205895684070099e-06, "loss": 10.327106475830078, "loss_alignment": 0.86328125, "loss_alignment_w": 0.431640625, "loss_sft": 0.2140781618654728, "loss_total_v6": 0.6454441472887993, "step": 4496 }, { "epoch": 1.0371315315315315, "grad_norm": 27.56637191772461, "learning_rate": 4.203907467941934e-06, "loss": 9.710956573486328, "loss_alignment": 0.841796875, "loss_alignment_w": 0.4208984375, "loss_sft": 0.18618897534906864, "loss_total_v6": 0.6069348081946373, "step": 4497 }, { "epoch": 1.0373621621621623, "grad_norm": 27.2434139251709, "learning_rate": 4.201919380957725e-06, "loss": 10.062187194824219, "loss_alignment": 0.82958984375, "loss_alignment_w": 0.414794921875, "loss_sft": 0.21387810818850994, "loss_total_v6": 0.6288866773247719, "step": 4498 }, { "epoch": 1.0375927927927928, "grad_norm": 26.722553253173828, "learning_rate": 4.199931423439983e-06, "loss": 10.358291625976562, "loss_alignment": 0.87060546875, "loss_alignment_w": 0.435302734375, "loss_sft": 0.212700879201293, "loss_total_v6": 0.6473932638764381, "step": 4499 }, { "epoch": 1.0378234234234234, "grad_norm": 25.208993911743164, "learning_rate": 4.1979435957111984e-06, "loss": 10.622387886047363, "loss_alignment": 0.84130859375, "loss_alignment_w": 0.420654296875, "loss_sft": 0.24390110187232494, "loss_total_v6": 0.6638992726802826, "step": 4500 }, { "epoch": 1.0378234234234234, "eval_loss": 0.8589687347412109, "eval_loss_alignment": 0.8464567280251142, "eval_loss_alignment_w": 0.4232283640125571, "eval_loss_contrast": 0.0, "eval_loss_highlevel": 0.0, "eval_loss_sft": 0.4357403880623105, "eval_loss_total_v6": 0.8589687495062884, "eval_loss_wm_recon": 0.0, "eval_loss_wm_recon_w": 0.0, "eval_runtime": 331.0171, "eval_samples_per_second": 5.272, "eval_steps_per_second": 0.662, "step": 4500 }, { "epoch": 1.0380540540540542, "grad_norm": 27.158937454223633, "learning_rate": 4.1959558980938445e-06, "loss": 9.613994598388672, "loss_alignment": 0.80029296875, "loss_alignment_w": 0.400146484375, "loss_sft": 0.20039251632988453, "loss_total_v6": 0.600874699652195, "step": 4501 }, { "epoch": 1.0382846846846847, "grad_norm": 28.245803833007812, "learning_rate": 4.1939683309103665e-06, "loss": 10.338142395019531, "loss_alignment": 0.85400390625, "loss_alignment_w": 0.427001953125, "loss_sft": 0.2183232121169567, "loss_total_v6": 0.6461338996887207, "step": 4502 }, { "epoch": 1.0385153153153153, "grad_norm": 26.707807540893555, "learning_rate": 4.191980894483195e-06, "loss": 9.91874885559082, "loss_alignment": 0.83642578125, "loss_alignment_w": 0.418212890625, "loss_sft": 0.20123586989939213, "loss_total_v6": 0.6199218034744263, "step": 4503 }, { "epoch": 1.038745945945946, "grad_norm": 28.72686767578125, "learning_rate": 4.189993589134735e-06, "loss": 9.92210578918457, "loss_alignment": 0.83251953125, "loss_alignment_w": 0.416259765625, "loss_sft": 0.20283424481749535, "loss_total_v6": 0.6201316118240356, "step": 4504 }, { "epoch": 1.0389765765765766, "grad_norm": 28.22842788696289, "learning_rate": 4.18800641518737e-06, "loss": 9.790353775024414, "loss_alignment": 0.82470703125, "loss_alignment_w": 0.412353515625, "loss_sft": 0.1991315856575966, "loss_total_v6": 0.6118970736861229, "step": 4505 }, { "epoch": 1.0392072072072072, "grad_norm": 28.010223388671875, "learning_rate": 4.186019372963467e-06, "loss": 10.005346298217773, "loss_alignment": 0.82861328125, "loss_alignment_w": 0.414306640625, "loss_sft": 0.21087490394711494, "loss_total_v6": 0.6253341361880302, "step": 4506 }, { "epoch": 1.039437837837838, "grad_norm": 28.756980895996094, "learning_rate": 4.1840324627853676e-06, "loss": 10.213949203491211, "loss_alignment": 0.86669921875, "loss_alignment_w": 0.433349609375, "loss_sft": 0.20488490350544453, "loss_total_v6": 0.6383718550205231, "step": 4507 }, { "epoch": 1.0396684684684685, "grad_norm": 30.33485984802246, "learning_rate": 4.1820456849753915e-06, "loss": 9.96295166015625, "loss_alignment": 0.8564453125, "loss_alignment_w": 0.42822265625, "loss_sft": 0.19447707757353783, "loss_total_v6": 0.6226844787597656, "step": 4508 }, { "epoch": 1.039899099099099, "grad_norm": 25.84773826599121, "learning_rate": 4.1800590398558414e-06, "loss": 9.820289611816406, "loss_alignment": 0.8251953125, "loss_alignment_w": 0.41259765625, "loss_sft": 0.20130771026015282, "loss_total_v6": 0.6137680560350418, "step": 4509 }, { "epoch": 1.0401297297297298, "grad_norm": 23.761247634887695, "learning_rate": 4.17807252774899e-06, "loss": 10.122678756713867, "loss_alignment": 0.845703125, "loss_alignment_w": 0.4228515625, "loss_sft": 0.2098310999572277, "loss_total_v6": 0.6326673924922943, "step": 4510 }, { "epoch": 1.0403603603603604, "grad_norm": 38.24666976928711, "learning_rate": 4.176086148977101e-06, "loss": 10.24120044708252, "loss_alignment": 0.84521484375, "loss_alignment_w": 0.422607421875, "loss_sft": 0.21644527278840542, "loss_total_v6": 0.6400750279426575, "step": 4511 }, { "epoch": 1.040590990990991, "grad_norm": 30.078697204589844, "learning_rate": 4.174099903862403e-06, "loss": 9.597865104675293, "loss_alignment": 0.8447265625, "loss_alignment_w": 0.42236328125, "loss_sft": 0.17761007603257895, "loss_total_v6": 0.5998665615916252, "step": 4512 }, { "epoch": 1.0408216216216217, "grad_norm": 25.718101501464844, "learning_rate": 4.172113792727114e-06, "loss": 10.152121543884277, "loss_alignment": 0.8544921875, "loss_alignment_w": 0.42724609375, "loss_sft": 0.2072004759684205, "loss_total_v6": 0.6345076113939285, "step": 4513 }, { "epoch": 1.0410522522522523, "grad_norm": 29.042478561401367, "learning_rate": 4.1701278158934214e-06, "loss": 9.90748405456543, "loss_alignment": 0.845703125, "loss_alignment_w": 0.4228515625, "loss_sft": 0.19595413468778133, "loss_total_v6": 0.6192176938056946, "step": 4514 }, { "epoch": 1.0412828828828828, "grad_norm": 27.19244956970215, "learning_rate": 4.168141973683496e-06, "loss": 9.594586372375488, "loss_alignment": 0.8232421875, "loss_alignment_w": 0.41162109375, "loss_sft": 0.18840674683451653, "loss_total_v6": 0.5996616557240486, "step": 4515 }, { "epoch": 1.0415135135135136, "grad_norm": 26.950969696044922, "learning_rate": 4.166156266419489e-06, "loss": 10.211544036865234, "loss_alignment": 0.845703125, "loss_alignment_w": 0.4228515625, "loss_sft": 0.21594981849193573, "loss_total_v6": 0.6382215097546577, "step": 4516 }, { "epoch": 1.0417441441441442, "grad_norm": 29.293315887451172, "learning_rate": 4.164170694423525e-06, "loss": 10.308723449707031, "loss_alignment": 0.85791015625, "loss_alignment_w": 0.428955078125, "loss_sft": 0.21544703096151352, "loss_total_v6": 0.644295260310173, "step": 4517 }, { "epoch": 1.0419747747747747, "grad_norm": 28.94830322265625, "learning_rate": 4.162185258017704e-06, "loss": 9.972456932067871, "loss_alignment": 0.837890625, "loss_alignment_w": 0.4189453125, "loss_sft": 0.20389072597026825, "loss_total_v6": 0.6232785657048225, "step": 4518 }, { "epoch": 1.0422054054054053, "grad_norm": 28.439802169799805, "learning_rate": 4.160199957524111e-06, "loss": 10.383401870727539, "loss_alignment": 0.86767578125, "loss_alignment_w": 0.433837890625, "loss_sft": 0.21517046820372343, "loss_total_v6": 0.6489625945687294, "step": 4519 }, { "epoch": 1.042436036036036, "grad_norm": 31.302574157714844, "learning_rate": 4.158214793264808e-06, "loss": 10.286945343017578, "loss_alignment": 0.86279296875, "loss_alignment_w": 0.431396484375, "loss_sft": 0.2115528341382742, "loss_total_v6": 0.642934076488018, "step": 4520 }, { "epoch": 1.0426666666666666, "grad_norm": 25.658262252807617, "learning_rate": 4.15622976556183e-06, "loss": 9.999967575073242, "loss_alignment": 0.84130859375, "loss_alignment_w": 0.420654296875, "loss_sft": 0.2045114804059267, "loss_total_v6": 0.6249979585409164, "step": 4521 }, { "epoch": 1.0428972972972974, "grad_norm": 23.914260864257812, "learning_rate": 4.154244874737195e-06, "loss": 10.280359268188477, "loss_alignment": 0.876953125, "loss_alignment_w": 0.4384765625, "loss_sft": 0.2031913734972477, "loss_total_v6": 0.6425224468111992, "step": 4522 }, { "epoch": 1.043127927927928, "grad_norm": 27.9320125579834, "learning_rate": 4.152260121112893e-06, "loss": 10.004903793334961, "loss_alignment": 0.8125, "loss_alignment_w": 0.40625, "loss_sft": 0.21971258893609047, "loss_total_v6": 0.6253064572811127, "step": 4523 }, { "epoch": 1.0433585585585585, "grad_norm": 29.838045120239258, "learning_rate": 4.150275505010898e-06, "loss": 10.890070915222168, "loss_alignment": 0.8720703125, "loss_alignment_w": 0.43603515625, "loss_sft": 0.2461506798863411, "loss_total_v6": 0.6806294322013855, "step": 4524 }, { "epoch": 1.043589189189189, "grad_norm": 26.57573890686035, "learning_rate": 4.148291026753159e-06, "loss": 10.103057861328125, "loss_alignment": 0.8564453125, "loss_alignment_w": 0.42822265625, "loss_sft": 0.20437814109027386, "loss_total_v6": 0.6314411088824272, "step": 4525 }, { "epoch": 1.0438198198198199, "grad_norm": 26.872591018676758, "learning_rate": 4.146306686661602e-06, "loss": 9.666576385498047, "loss_alignment": 0.802734375, "loss_alignment_w": 0.4013671875, "loss_sft": 0.2027175072580576, "loss_total_v6": 0.6041609942913055, "step": 4526 }, { "epoch": 1.0440504504504504, "grad_norm": 27.938222885131836, "learning_rate": 4.144322485058131e-06, "loss": 10.001312255859375, "loss_alignment": 0.8701171875, "loss_alignment_w": 0.43505859375, "loss_sft": 0.1898098113015294, "loss_total_v6": 0.6250820159912109, "step": 4527 }, { "epoch": 1.044281081081081, "grad_norm": 29.09166717529297, "learning_rate": 4.142338422264629e-06, "loss": 9.915560722351074, "loss_alignment": 0.85595703125, "loss_alignment_w": 0.427978515625, "loss_sft": 0.19172878563404083, "loss_total_v6": 0.6197225600481033, "step": 4528 }, { "epoch": 1.0445117117117118, "grad_norm": 26.890161514282227, "learning_rate": 4.140354498602952e-06, "loss": 10.267156600952148, "loss_alignment": 0.86279296875, "loss_alignment_w": 0.431396484375, "loss_sft": 0.21020921505987644, "loss_total_v6": 0.6416972503066063, "step": 4529 }, { "epoch": 1.0447423423423423, "grad_norm": 28.07464599609375, "learning_rate": 4.138370714394941e-06, "loss": 9.78533935546875, "loss_alignment": 0.810546875, "loss_alignment_w": 0.4052734375, "loss_sft": 0.20611194148659706, "loss_total_v6": 0.6115837544202805, "step": 4530 }, { "epoch": 1.0449729729729729, "grad_norm": 26.114469528198242, "learning_rate": 4.136387069962408e-06, "loss": 9.04435920715332, "loss_alignment": 0.82763671875, "loss_alignment_w": 0.413818359375, "loss_sft": 0.1502334512770176, "loss_total_v6": 0.5652725026011467, "step": 4531 }, { "epoch": 1.0452036036036036, "grad_norm": 27.70185089111328, "learning_rate": 4.134403565627144e-06, "loss": 10.060304641723633, "loss_alignment": 0.814453125, "loss_alignment_w": 0.4072265625, "loss_sft": 0.22160353139042854, "loss_total_v6": 0.6287690475583076, "step": 4532 }, { "epoch": 1.0454342342342342, "grad_norm": 24.258989334106445, "learning_rate": 4.132420201710917e-06, "loss": 10.052841186523438, "loss_alignment": 0.82373046875, "loss_alignment_w": 0.411865234375, "loss_sft": 0.21549130976200104, "loss_total_v6": 0.6283025816082954, "step": 4533 }, { "epoch": 1.0456648648648648, "grad_norm": 27.819313049316406, "learning_rate": 4.130436978535472e-06, "loss": 10.177545547485352, "loss_alignment": 0.84619140625, "loss_alignment_w": 0.423095703125, "loss_sft": 0.2131686992943287, "loss_total_v6": 0.6360965594649315, "step": 4534 }, { "epoch": 1.0458954954954955, "grad_norm": 29.693805694580078, "learning_rate": 4.1284538964225364e-06, "loss": 9.783153533935547, "loss_alignment": 0.826171875, "loss_alignment_w": 0.4130859375, "loss_sft": 0.19891045801341534, "loss_total_v6": 0.6114470958709717, "step": 4535 }, { "epoch": 1.046126126126126, "grad_norm": 27.701095581054688, "learning_rate": 4.126470955693806e-06, "loss": 10.397331237792969, "loss_alignment": 0.8623046875, "loss_alignment_w": 0.43115234375, "loss_sft": 0.21799424849450588, "loss_total_v6": 0.6498332321643829, "step": 4536 }, { "epoch": 1.0463567567567567, "grad_norm": 24.2929744720459, "learning_rate": 4.124488156670962e-06, "loss": 9.973977088928223, "loss_alignment": 0.830078125, "loss_alignment_w": 0.4150390625, "loss_sft": 0.20821242965757847, "loss_total_v6": 0.6233735755085945, "step": 4537 }, { "epoch": 1.0465873873873874, "grad_norm": 30.38290023803711, "learning_rate": 4.122505499675655e-06, "loss": 10.023380279541016, "loss_alignment": 0.84130859375, "loss_alignment_w": 0.420654296875, "loss_sft": 0.20566958747804165, "loss_total_v6": 0.6264612227678299, "step": 4538 }, { "epoch": 1.046818018018018, "grad_norm": 26.11377716064453, "learning_rate": 4.120522985029516e-06, "loss": 10.229127883911133, "loss_alignment": 0.86572265625, "loss_alignment_w": 0.432861328125, "loss_sft": 0.20630657114088535, "loss_total_v6": 0.6393204927444458, "step": 4539 }, { "epoch": 1.0470486486486486, "grad_norm": 29.039926528930664, "learning_rate": 4.118540613054155e-06, "loss": 9.582066535949707, "loss_alignment": 0.83154296875, "loss_alignment_w": 0.415771484375, "loss_sft": 0.18242104351520538, "loss_total_v6": 0.5988791808485985, "step": 4540 }, { "epoch": 1.0472792792792793, "grad_norm": 25.61064338684082, "learning_rate": 4.116558384071158e-06, "loss": 9.76241683959961, "loss_alignment": 0.83544921875, "loss_alignment_w": 0.417724609375, "loss_sft": 0.1927316291257739, "loss_total_v6": 0.6101510599255562, "step": 4541 }, { "epoch": 1.0475099099099099, "grad_norm": 24.555789947509766, "learning_rate": 4.114576298402085e-06, "loss": 10.353719711303711, "loss_alignment": 0.849609375, "loss_alignment_w": 0.4248046875, "loss_sft": 0.22201289609074593, "loss_total_v6": 0.6471075192093849, "step": 4542 }, { "epoch": 1.0477405405405404, "grad_norm": 27.10494041442871, "learning_rate": 4.112594356368473e-06, "loss": 10.123658180236816, "loss_alignment": 0.84716796875, "loss_alignment_w": 0.423583984375, "loss_sft": 0.2092514494433999, "loss_total_v6": 0.6327286288142204, "step": 4543 }, { "epoch": 1.0479711711711712, "grad_norm": 27.506519317626953, "learning_rate": 4.1106125582918385e-06, "loss": 10.338613510131836, "loss_alignment": 0.87646484375, "loss_alignment_w": 0.438232421875, "loss_sft": 0.20733587630093098, "loss_total_v6": 0.6461633816361427, "step": 4544 }, { "epoch": 1.0482018018018018, "grad_norm": 27.299259185791016, "learning_rate": 4.108630904493676e-06, "loss": 10.03294849395752, "loss_alignment": 0.88525390625, "loss_alignment_w": 0.442626953125, "loss_sft": 0.18450860772281885, "loss_total_v6": 0.6270592883229256, "step": 4545 }, { "epoch": 1.0484324324324323, "grad_norm": 28.063369750976562, "learning_rate": 4.106649395295449e-06, "loss": 10.087156295776367, "loss_alignment": 0.84228515625, "loss_alignment_w": 0.421142578125, "loss_sft": 0.20867904275655746, "loss_total_v6": 0.6304472461342812, "step": 4546 }, { "epoch": 1.0486630630630631, "grad_norm": 24.45861053466797, "learning_rate": 4.104668031018607e-06, "loss": 9.989776611328125, "loss_alignment": 0.86376953125, "loss_alignment_w": 0.431884765625, "loss_sft": 0.1923237070441246, "loss_total_v6": 0.6243610680103302, "step": 4547 }, { "epoch": 1.0488936936936937, "grad_norm": 31.01615333557129, "learning_rate": 4.1026868119845685e-06, "loss": 10.226213455200195, "loss_alignment": 0.85009765625, "loss_alignment_w": 0.425048828125, "loss_sft": 0.2120448648929596, "loss_total_v6": 0.6391383782029152, "step": 4548 }, { "epoch": 1.0491243243243242, "grad_norm": 27.639305114746094, "learning_rate": 4.10070573851473e-06, "loss": 9.669088363647461, "loss_alignment": 0.82177734375, "loss_alignment_w": 0.410888671875, "loss_sft": 0.19487890973687172, "loss_total_v6": 0.6043180152773857, "step": 4549 }, { "epoch": 1.049354954954955, "grad_norm": 27.070432662963867, "learning_rate": 4.098724810930472e-06, "loss": 10.023448944091797, "loss_alignment": 0.83154296875, "loss_alignment_w": 0.415771484375, "loss_sft": 0.20990066044032574, "loss_total_v6": 0.6264655962586403, "step": 4550 }, { "epoch": 1.0495855855855856, "grad_norm": 27.384620666503906, "learning_rate": 4.09674402955314e-06, "loss": 9.990236282348633, "loss_alignment": 0.84765625, "loss_alignment_w": 0.423828125, "loss_sft": 0.1998444553464651, "loss_total_v6": 0.6243897303938866, "step": 4551 }, { "epoch": 1.0498162162162161, "grad_norm": 27.2574405670166, "learning_rate": 4.0947633947040616e-06, "loss": 10.205045700073242, "loss_alignment": 0.849609375, "loss_alignment_w": 0.4248046875, "loss_sft": 0.21365154907107353, "loss_total_v6": 0.6378153711557388, "step": 4552 }, { "epoch": 1.050046846846847, "grad_norm": 26.245059967041016, "learning_rate": 4.092782906704541e-06, "loss": 10.210485458374023, "loss_alignment": 0.85693359375, "loss_alignment_w": 0.428466796875, "loss_sft": 0.20974958315491676, "loss_total_v6": 0.6381553336977959, "step": 4553 }, { "epoch": 1.0502774774774775, "grad_norm": 30.621814727783203, "learning_rate": 4.0908025658758555e-06, "loss": 9.83772087097168, "loss_alignment": 0.818359375, "loss_alignment_w": 0.4091796875, "loss_sft": 0.20464027114212513, "loss_total_v6": 0.6148575469851494, "step": 4554 }, { "epoch": 1.050508108108108, "grad_norm": 25.491079330444336, "learning_rate": 4.088822372539263e-06, "loss": 10.194881439208984, "loss_alignment": 0.84619140625, "loss_alignment_w": 0.423095703125, "loss_sft": 0.2138402396813035, "loss_total_v6": 0.6371800750494003, "step": 4555 }, { "epoch": 1.0507387387387388, "grad_norm": 27.744958877563477, "learning_rate": 4.086842327015995e-06, "loss": 9.902488708496094, "loss_alignment": 0.859375, "loss_alignment_w": 0.4296875, "loss_sft": 0.19121695309877396, "loss_total_v6": 0.6189055219292641, "step": 4556 }, { "epoch": 1.0509693693693694, "grad_norm": 24.34824562072754, "learning_rate": 4.084862429627257e-06, "loss": 10.07331657409668, "loss_alignment": 0.84521484375, "loss_alignment_w": 0.422607421875, "loss_sft": 0.20805826224386692, "loss_total_v6": 0.6295822933316231, "step": 4557 }, { "epoch": 1.0512, "grad_norm": 29.006519317626953, "learning_rate": 4.0828826806942335e-06, "loss": 9.88912582397461, "loss_alignment": 0.806640625, "loss_alignment_w": 0.4033203125, "loss_sft": 0.21429230459034443, "loss_total_v6": 0.6180703565478325, "step": 4558 }, { "epoch": 1.0514306306306307, "grad_norm": 28.57862663269043, "learning_rate": 4.080903080538082e-06, "loss": 10.278000831604004, "loss_alignment": 0.86572265625, "loss_alignment_w": 0.432861328125, "loss_sft": 0.20975788310170174, "loss_total_v6": 0.642375074326992, "step": 4559 }, { "epoch": 1.0516612612612612, "grad_norm": 26.843563079833984, "learning_rate": 4.0789236294799425e-06, "loss": 10.125999450683594, "loss_alignment": 0.822265625, "loss_alignment_w": 0.4111328125, "loss_sft": 0.2215132713317871, "loss_total_v6": 0.6328749433159828, "step": 4560 }, { "epoch": 1.0518918918918918, "grad_norm": 30.006166458129883, "learning_rate": 4.076944327840923e-06, "loss": 10.239120483398438, "loss_alignment": 0.8515625, "loss_alignment_w": 0.42578125, "loss_sft": 0.21243949979543686, "loss_total_v6": 0.6399450078606606, "step": 4561 }, { "epoch": 1.0521225225225226, "grad_norm": 28.310861587524414, "learning_rate": 4.07496517594211e-06, "loss": 10.316068649291992, "loss_alignment": 0.857421875, "loss_alignment_w": 0.4287109375, "loss_sft": 0.21541774459183216, "loss_total_v6": 0.6447542905807495, "step": 4562 }, { "epoch": 1.0523531531531531, "grad_norm": 26.20424461364746, "learning_rate": 4.072986174104568e-06, "loss": 9.911723136901855, "loss_alignment": 0.84423828125, "loss_alignment_w": 0.422119140625, "loss_sft": 0.19760766066610813, "loss_total_v6": 0.619482696056366, "step": 4563 }, { "epoch": 1.0525837837837837, "grad_norm": 33.69473648071289, "learning_rate": 4.071007322649331e-06, "loss": 9.500625610351562, "loss_alignment": 0.794921875, "loss_alignment_w": 0.3974609375, "loss_sft": 0.19684696570038795, "loss_total_v6": 0.5937890857458115, "step": 4564 }, { "epoch": 1.0528144144144145, "grad_norm": 28.568511962890625, "learning_rate": 4.069028621897417e-06, "loss": 10.25026798248291, "loss_alignment": 0.86279296875, "loss_alignment_w": 0.431396484375, "loss_sft": 0.2088790237903595, "loss_total_v6": 0.6406417563557625, "step": 4565 }, { "epoch": 1.053045045045045, "grad_norm": 26.70285415649414, "learning_rate": 4.067050072169815e-06, "loss": 9.508371353149414, "loss_alignment": 0.83984375, "loss_alignment_w": 0.419921875, "loss_sft": 0.17487013526260853, "loss_total_v6": 0.5942732244729996, "step": 4566 }, { "epoch": 1.0532756756756756, "grad_norm": 27.139713287353516, "learning_rate": 4.065071673787488e-06, "loss": 9.367086410522461, "loss_alignment": 0.79443359375, "loss_alignment_w": 0.397216796875, "loss_sft": 0.18892797268927097, "loss_total_v6": 0.5854428708553314, "step": 4567 }, { "epoch": 1.0535063063063064, "grad_norm": 26.907283782958984, "learning_rate": 4.063093427071376e-06, "loss": 9.689428329467773, "loss_alignment": 0.82373046875, "loss_alignment_w": 0.411865234375, "loss_sft": 0.1941665429621935, "loss_total_v6": 0.6055892556905746, "step": 4568 }, { "epoch": 1.053736936936937, "grad_norm": 26.906692504882812, "learning_rate": 4.061115332342399e-06, "loss": 10.2919921875, "loss_alignment": 0.86474609375, "loss_alignment_w": 0.432373046875, "loss_sft": 0.21095275320112705, "loss_total_v6": 0.6432494968175888, "step": 4569 }, { "epoch": 1.0539675675675675, "grad_norm": 29.374004364013672, "learning_rate": 4.059137389921444e-06, "loss": 9.50063705444336, "loss_alignment": 0.818359375, "loss_alignment_w": 0.4091796875, "loss_sft": 0.18561720848083496, "loss_total_v6": 0.5937898308038712, "step": 4570 }, { "epoch": 1.0541981981981983, "grad_norm": 28.168426513671875, "learning_rate": 4.057159600129378e-06, "loss": 9.770931243896484, "loss_alignment": 0.814453125, "loss_alignment_w": 0.4072265625, "loss_sft": 0.20385340228676796, "loss_total_v6": 0.6106832325458527, "step": 4571 }, { "epoch": 1.0544288288288288, "grad_norm": 32.85749435424805, "learning_rate": 4.055181963287044e-06, "loss": 9.751596450805664, "loss_alignment": 0.830078125, "loss_alignment_w": 0.4150390625, "loss_sft": 0.19445096142590046, "loss_total_v6": 0.6094747632741928, "step": 4572 }, { "epoch": 1.0546594594594594, "grad_norm": 29.182235717773438, "learning_rate": 4.053204479715254e-06, "loss": 9.920801162719727, "loss_alignment": 0.86181640625, "loss_alignment_w": 0.430908203125, "loss_sft": 0.18964539468288422, "loss_total_v6": 0.6200500577688217, "step": 4573 }, { "epoch": 1.0548900900900902, "grad_norm": 26.024051666259766, "learning_rate": 4.051227149734807e-06, "loss": 9.497650146484375, "loss_alignment": 0.869140625, "loss_alignment_w": 0.4345703125, "loss_sft": 0.1592922117561102, "loss_total_v6": 0.5936030969023705, "step": 4574 }, { "epoch": 1.0551207207207207, "grad_norm": 24.8500919342041, "learning_rate": 4.049249973666468e-06, "loss": 9.731393814086914, "loss_alignment": 0.8193359375, "loss_alignment_w": 0.40966796875, "loss_sft": 0.1980711631476879, "loss_total_v6": 0.6082121133804321, "step": 4575 }, { "epoch": 1.0553513513513513, "grad_norm": 24.09532928466797, "learning_rate": 4.047272951830976e-06, "loss": 9.989959716796875, "loss_alignment": 0.84912109375, "loss_alignment_w": 0.424560546875, "loss_sft": 0.19967463240027428, "loss_total_v6": 0.6243725270032883, "step": 4576 }, { "epoch": 1.055581981981982, "grad_norm": 26.06340217590332, "learning_rate": 4.04529608454905e-06, "loss": 10.131002426147461, "loss_alignment": 0.8447265625, "loss_alignment_w": 0.42236328125, "loss_sft": 0.21067177690565586, "loss_total_v6": 0.6331876516342163, "step": 4577 }, { "epoch": 1.0558126126126126, "grad_norm": 32.45259094238281, "learning_rate": 4.04331937214138e-06, "loss": 9.98845100402832, "loss_alignment": 0.82958984375, "loss_alignment_w": 0.414794921875, "loss_sft": 0.20992577448487282, "loss_total_v6": 0.6242781654000282, "step": 4578 }, { "epoch": 1.0560432432432432, "grad_norm": 25.225038528442383, "learning_rate": 4.041342814928636e-06, "loss": 9.592884063720703, "loss_alignment": 0.83203125, "loss_alignment_w": 0.416015625, "loss_sft": 0.183493809774518, "loss_total_v6": 0.5995552316308022, "step": 4579 }, { "epoch": 1.056273873873874, "grad_norm": 29.236351013183594, "learning_rate": 4.039366413231458e-06, "loss": 9.482969284057617, "loss_alignment": 0.8037109375, "loss_alignment_w": 0.40185546875, "loss_sft": 0.19014347903430462, "loss_total_v6": 0.5926855951547623, "step": 4580 }, { "epoch": 1.0565045045045045, "grad_norm": 31.170879364013672, "learning_rate": 4.037390167370464e-06, "loss": 9.773024559020996, "loss_alignment": 0.8466796875, "loss_alignment_w": 0.42333984375, "loss_sft": 0.18738265708088875, "loss_total_v6": 0.6108140647411346, "step": 4581 }, { "epoch": 1.056735135135135, "grad_norm": 27.644174575805664, "learning_rate": 4.035414077666244e-06, "loss": 9.435487747192383, "loss_alignment": 0.8271484375, "loss_alignment_w": 0.41357421875, "loss_sft": 0.17673886194825172, "loss_total_v6": 0.5897179767489433, "step": 4582 }, { "epoch": 1.0569657657657658, "grad_norm": 27.133480072021484, "learning_rate": 4.033438144439359e-06, "loss": 10.456039428710938, "loss_alignment": 0.88818359375, "loss_alignment_w": 0.444091796875, "loss_sft": 0.20902921818196774, "loss_total_v6": 0.6535024866461754, "step": 4583 }, { "epoch": 1.0571963963963964, "grad_norm": 27.744556427001953, "learning_rate": 4.031462368010357e-06, "loss": 9.794109344482422, "loss_alignment": 0.84375, "loss_alignment_w": 0.421875, "loss_sft": 0.18957016244530678, "loss_total_v6": 0.6121318116784096, "step": 4584 }, { "epoch": 1.057427027027027, "grad_norm": 25.842370986938477, "learning_rate": 4.02948674869975e-06, "loss": 9.741387367248535, "loss_alignment": 0.81689453125, "loss_alignment_w": 0.408447265625, "loss_sft": 0.19921452179551125, "loss_total_v6": 0.608836717903614, "step": 4585 }, { "epoch": 1.0576576576576577, "grad_norm": 34.47038269042969, "learning_rate": 4.027511286828028e-06, "loss": 10.375509262084961, "loss_alignment": 0.857421875, "loss_alignment_w": 0.4287109375, "loss_sft": 0.21969735994935036, "loss_total_v6": 0.6484693214297295, "step": 4586 }, { "epoch": 1.0578882882882883, "grad_norm": 26.376461029052734, "learning_rate": 4.025535982715654e-06, "loss": 9.918537139892578, "loss_alignment": 0.84765625, "loss_alignment_w": 0.423828125, "loss_sft": 0.19644663482904434, "loss_total_v6": 0.6199085563421249, "step": 4587 }, { "epoch": 1.0581189189189188, "grad_norm": 36.219444274902344, "learning_rate": 4.023560836683065e-06, "loss": 9.525054931640625, "loss_alignment": 0.83740234375, "loss_alignment_w": 0.418701171875, "loss_sft": 0.1774387350305915, "loss_total_v6": 0.5953159406781197, "step": 4588 }, { "epoch": 1.0583495495495496, "grad_norm": 28.89129066467285, "learning_rate": 4.021585849050677e-06, "loss": 10.069660186767578, "loss_alignment": 0.85888671875, "loss_alignment_w": 0.429443359375, "loss_sft": 0.19989516958594322, "loss_total_v6": 0.6293537840247154, "step": 4589 }, { "epoch": 1.0585801801801802, "grad_norm": 25.62660789489746, "learning_rate": 4.019611020138875e-06, "loss": 10.08237075805664, "loss_alignment": 0.84765625, "loss_alignment_w": 0.423828125, "loss_sft": 0.20555715262889862, "loss_total_v6": 0.6301482170820236, "step": 4590 }, { "epoch": 1.0588108108108107, "grad_norm": 32.65183639526367, "learning_rate": 4.017636350268022e-06, "loss": 10.283117294311523, "loss_alignment": 0.83056640625, "loss_alignment_w": 0.415283203125, "loss_sft": 0.22695383429527283, "loss_total_v6": 0.642694815993309, "step": 4591 }, { "epoch": 1.0590414414414415, "grad_norm": 27.500823974609375, "learning_rate": 4.01566183975845e-06, "loss": 10.357009887695312, "loss_alignment": 0.83837890625, "loss_alignment_w": 0.419189453125, "loss_sft": 0.22819996811449528, "loss_total_v6": 0.6473131477832794, "step": 4592 }, { "epoch": 1.059272072072072, "grad_norm": 27.312841415405273, "learning_rate": 4.0136874889304705e-06, "loss": 9.56668472290039, "loss_alignment": 0.81591796875, "loss_alignment_w": 0.407958984375, "loss_sft": 0.1888296715915203, "loss_total_v6": 0.5979178175330162, "step": 4593 }, { "epoch": 1.0595027027027026, "grad_norm": 30.38519859313965, "learning_rate": 4.0117132981043695e-06, "loss": 9.739995002746582, "loss_alignment": 0.8388671875, "loss_alignment_w": 0.41943359375, "loss_sft": 0.18894988670945168, "loss_total_v6": 0.6087497100234032, "step": 4594 }, { "epoch": 1.0597333333333334, "grad_norm": 26.066307067871094, "learning_rate": 4.009739267600403e-06, "loss": 9.824285507202148, "loss_alignment": 0.8203125, "loss_alignment_w": 0.41015625, "loss_sft": 0.20207635685801506, "loss_total_v6": 0.6140178665518761, "step": 4595 }, { "epoch": 1.059963963963964, "grad_norm": 28.246633529663086, "learning_rate": 4.007765397738801e-06, "loss": 9.843035697937012, "loss_alignment": 0.80615234375, "loss_alignment_w": 0.403076171875, "loss_sft": 0.21254080720245838, "loss_total_v6": 0.6151897385716438, "step": 4596 }, { "epoch": 1.0601945945945945, "grad_norm": 27.78767967224121, "learning_rate": 4.005791688839773e-06, "loss": 10.081506729125977, "loss_alignment": 0.86669921875, "loss_alignment_w": 0.433349609375, "loss_sft": 0.1977363433688879, "loss_total_v6": 0.6300941556692123, "step": 4597 }, { "epoch": 1.0604252252252253, "grad_norm": 28.203001022338867, "learning_rate": 4.003818141223493e-06, "loss": 9.789311408996582, "loss_alignment": 0.84619140625, "loss_alignment_w": 0.423095703125, "loss_sft": 0.18824797868728638, "loss_total_v6": 0.611831970512867, "step": 4598 }, { "epoch": 1.0606558558558559, "grad_norm": 27.78836441040039, "learning_rate": 4.0018447552101194e-06, "loss": 10.109315872192383, "loss_alignment": 0.82470703125, "loss_alignment_w": 0.412353515625, "loss_sft": 0.2187615316361189, "loss_total_v6": 0.6318322345614433, "step": 4599 }, { "epoch": 1.0608864864864864, "grad_norm": 28.932058334350586, "learning_rate": 3.999871531119779e-06, "loss": 9.538738250732422, "loss_alignment": 0.81201171875, "loss_alignment_w": 0.406005859375, "loss_sft": 0.19021105580031872, "loss_total_v6": 0.5961711332201958, "step": 4600 }, { "epoch": 1.0611171171171172, "grad_norm": 30.459226608276367, "learning_rate": 3.997898469272571e-06, "loss": 10.388660430908203, "loss_alignment": 0.86328125, "loss_alignment_w": 0.431640625, "loss_sft": 0.21766588278114796, "loss_total_v6": 0.6492912545800209, "step": 4601 }, { "epoch": 1.0613477477477478, "grad_norm": 25.431739807128906, "learning_rate": 3.995925569988569e-06, "loss": 9.451484680175781, "loss_alignment": 0.8134765625, "loss_alignment_w": 0.40673828125, "loss_sft": 0.1839032331481576, "loss_total_v6": 0.5907178148627281, "step": 4602 }, { "epoch": 1.0615783783783783, "grad_norm": 28.995302200317383, "learning_rate": 3.993952833587823e-06, "loss": 9.883848190307617, "loss_alignment": 0.8505859375, "loss_alignment_w": 0.42529296875, "loss_sft": 0.19142519123852253, "loss_total_v6": 0.6177405044436455, "step": 4603 }, { "epoch": 1.061809009009009, "grad_norm": 29.35468101501465, "learning_rate": 3.991980260390356e-06, "loss": 10.023783683776855, "loss_alignment": 0.8359375, "loss_alignment_w": 0.41796875, "loss_sft": 0.20979944616556168, "loss_total_v6": 0.6264864653348923, "step": 4604 }, { "epoch": 1.0620396396396397, "grad_norm": 29.89459800720215, "learning_rate": 3.990007850716161e-06, "loss": 9.334915161132812, "loss_alignment": 0.8134765625, "loss_alignment_w": 0.40673828125, "loss_sft": 0.177792533300817, "loss_total_v6": 0.5834321975708008, "step": 4605 }, { "epoch": 1.0622702702702702, "grad_norm": 25.813282012939453, "learning_rate": 3.9880356048852085e-06, "loss": 9.781087875366211, "loss_alignment": 0.83154296875, "loss_alignment_w": 0.415771484375, "loss_sft": 0.19550067838281393, "loss_total_v6": 0.6113179475069046, "step": 4606 }, { "epoch": 1.062500900900901, "grad_norm": 27.298749923706055, "learning_rate": 3.986063523217439e-06, "loss": 10.06047534942627, "loss_alignment": 0.84619140625, "loss_alignment_w": 0.423095703125, "loss_sft": 0.20508892461657524, "loss_total_v6": 0.6287797093391418, "step": 4607 }, { "epoch": 1.0627315315315315, "grad_norm": 28.52643394470215, "learning_rate": 3.984091606032768e-06, "loss": 10.406587600708008, "loss_alignment": 0.85302734375, "loss_alignment_w": 0.426513671875, "loss_sft": 0.2247219942510128, "loss_total_v6": 0.6504116803407669, "step": 4608 }, { "epoch": 1.062962162162162, "grad_norm": 26.159799575805664, "learning_rate": 3.982119853651087e-06, "loss": 9.557685852050781, "loss_alignment": 0.826171875, "loss_alignment_w": 0.4130859375, "loss_sft": 0.18585632368922234, "loss_total_v6": 0.5973553508520126, "step": 4609 }, { "epoch": 1.0631927927927929, "grad_norm": 29.39994239807129, "learning_rate": 3.980148266392257e-06, "loss": 9.846742630004883, "loss_alignment": 0.876953125, "loss_alignment_w": 0.4384765625, "loss_sft": 0.1772804856300354, "loss_total_v6": 0.6154213771224022, "step": 4610 }, { "epoch": 1.0634234234234234, "grad_norm": 27.24390411376953, "learning_rate": 3.978176844576113e-06, "loss": 10.108137130737305, "loss_alignment": 0.8095703125, "loss_alignment_w": 0.40478515625, "loss_sft": 0.22817888297140598, "loss_total_v6": 0.6317585781216621, "step": 4611 }, { "epoch": 1.063654054054054, "grad_norm": 25.57326889038086, "learning_rate": 3.9762055885224614e-06, "loss": 10.281060218811035, "loss_alignment": 0.84375, "loss_alignment_w": 0.421875, "loss_sft": 0.21995881758630276, "loss_total_v6": 0.6425662487745285, "step": 4612 }, { "epoch": 1.0638846846846848, "grad_norm": 28.83965492248535, "learning_rate": 3.974234498551089e-06, "loss": 10.089521408081055, "loss_alignment": 0.83837890625, "loss_alignment_w": 0.419189453125, "loss_sft": 0.21168028563261032, "loss_total_v6": 0.6305950880050659, "step": 4613 }, { "epoch": 1.0641153153153153, "grad_norm": 26.620826721191406, "learning_rate": 3.972263574981748e-06, "loss": 9.940670013427734, "loss_alignment": 0.8525390625, "loss_alignment_w": 0.42626953125, "loss_sft": 0.19634987320750952, "loss_total_v6": 0.6212918907403946, "step": 4614 }, { "epoch": 1.064345945945946, "grad_norm": 27.53184700012207, "learning_rate": 3.970292818134164e-06, "loss": 10.089468955993652, "loss_alignment": 0.83447265625, "loss_alignment_w": 0.417236328125, "loss_sft": 0.2131113037467003, "loss_total_v6": 0.6305917873978615, "step": 4615 }, { "epoch": 1.0645765765765767, "grad_norm": 27.061830520629883, "learning_rate": 3.968322228328041e-06, "loss": 10.058749198913574, "loss_alignment": 0.86083984375, "loss_alignment_w": 0.430419921875, "loss_sft": 0.19840451888740063, "loss_total_v6": 0.6286718547344208, "step": 4616 }, { "epoch": 1.0648072072072072, "grad_norm": 27.350374221801758, "learning_rate": 3.966351805883048e-06, "loss": 10.398737907409668, "loss_alignment": 0.87841796875, "loss_alignment_w": 0.439208984375, "loss_sft": 0.2110172864049673, "loss_total_v6": 0.649921104311943, "step": 4617 }, { "epoch": 1.0650378378378378, "grad_norm": 27.708099365234375, "learning_rate": 3.964381551118837e-06, "loss": 10.146738052368164, "loss_alignment": 0.8603515625, "loss_alignment_w": 0.43017578125, "loss_sft": 0.20362908765673637, "loss_total_v6": 0.6341711133718491, "step": 4618 }, { "epoch": 1.0652684684684686, "grad_norm": 25.52079963684082, "learning_rate": 3.962411464355026e-06, "loss": 9.966068267822266, "loss_alignment": 0.83203125, "loss_alignment_w": 0.416015625, "loss_sft": 0.20584130473434925, "loss_total_v6": 0.6228792741894722, "step": 4619 }, { "epoch": 1.0654990990990991, "grad_norm": 27.634185791015625, "learning_rate": 3.960441545911205e-06, "loss": 9.652205467224121, "loss_alignment": 0.8095703125, "loss_alignment_w": 0.40478515625, "loss_sft": 0.19823355413973331, "loss_total_v6": 0.6032628417015076, "step": 4620 }, { "epoch": 1.0657297297297297, "grad_norm": 24.517183303833008, "learning_rate": 3.958471796106939e-06, "loss": 9.845942497253418, "loss_alignment": 0.87939453125, "loss_alignment_w": 0.439697265625, "loss_sft": 0.17585723102092743, "loss_total_v6": 0.6153713837265968, "step": 4621 }, { "epoch": 1.0659603603603605, "grad_norm": 29.608741760253906, "learning_rate": 3.956502215261763e-06, "loss": 9.324793815612793, "loss_alignment": 0.7939453125, "loss_alignment_w": 0.39697265625, "loss_sft": 0.1862847562879324, "loss_total_v6": 0.582799643278122, "step": 4622 }, { "epoch": 1.066190990990991, "grad_norm": 28.29273223876953, "learning_rate": 3.9545328036951935e-06, "loss": 9.980775833129883, "loss_alignment": 0.8369140625, "loss_alignment_w": 0.41845703125, "loss_sft": 0.206272192299366, "loss_total_v6": 0.6237984523177147, "step": 4623 }, { "epoch": 1.0664216216216216, "grad_norm": 27.634401321411133, "learning_rate": 3.952563561726708e-06, "loss": 9.19759750366211, "loss_alignment": 0.81494140625, "loss_alignment_w": 0.407470703125, "loss_sft": 0.16715025529265404, "loss_total_v6": 0.57484982162714, "step": 4624 }, { "epoch": 1.0666522522522524, "grad_norm": 30.988037109375, "learning_rate": 3.9505944896757635e-06, "loss": 10.255891799926758, "loss_alignment": 0.849609375, "loss_alignment_w": 0.4248046875, "loss_sft": 0.21632586419582367, "loss_total_v6": 0.6409932225942612, "step": 4625 }, { "epoch": 1.066882882882883, "grad_norm": 28.43682861328125, "learning_rate": 3.948625587861786e-06, "loss": 9.720105171203613, "loss_alignment": 0.7939453125, "loss_alignment_w": 0.39697265625, "loss_sft": 0.21047282963991165, "loss_total_v6": 0.6075065433979034, "step": 4626 }, { "epoch": 1.0671135135135135, "grad_norm": 26.34648895263672, "learning_rate": 3.946656856604174e-06, "loss": 10.010239601135254, "loss_alignment": 0.84912109375, "loss_alignment_w": 0.424560546875, "loss_sft": 0.20069794915616512, "loss_total_v6": 0.6256399676203728, "step": 4627 }, { "epoch": 1.0673441441441442, "grad_norm": 28.425537109375, "learning_rate": 3.9446882962223025e-06, "loss": 9.496097564697266, "loss_alignment": 0.84521484375, "loss_alignment_w": 0.422607421875, "loss_sft": 0.1717379279434681, "loss_total_v6": 0.5935061126947403, "step": 4628 }, { "epoch": 1.0675747747747748, "grad_norm": 24.992691040039062, "learning_rate": 3.942719907035516e-06, "loss": 9.343595504760742, "loss_alignment": 0.79150390625, "loss_alignment_w": 0.395751953125, "loss_sft": 0.18840592354536057, "loss_total_v6": 0.5839747563004494, "step": 4629 }, { "epoch": 1.0678054054054054, "grad_norm": 28.165082931518555, "learning_rate": 3.940751689363129e-06, "loss": 9.414345741271973, "loss_alignment": 0.810546875, "loss_alignment_w": 0.4052734375, "loss_sft": 0.18358092568814754, "loss_total_v6": 0.5883965939283371, "step": 4630 }, { "epoch": 1.0680360360360361, "grad_norm": 29.150840759277344, "learning_rate": 3.9387836435244315e-06, "loss": 9.803574562072754, "loss_alignment": 0.80126953125, "loss_alignment_w": 0.400634765625, "loss_sft": 0.21161559037864208, "loss_total_v6": 0.6127233877778053, "step": 4631 }, { "epoch": 1.0682666666666667, "grad_norm": 25.71839141845703, "learning_rate": 3.936815769838682e-06, "loss": 9.73788070678711, "loss_alignment": 0.83154296875, "loss_alignment_w": 0.415771484375, "loss_sft": 0.19286133907735348, "loss_total_v6": 0.6086175814270973, "step": 4632 }, { "epoch": 1.0684972972972973, "grad_norm": 27.71229362487793, "learning_rate": 3.934848068625117e-06, "loss": 9.769615173339844, "loss_alignment": 0.82080078125, "loss_alignment_w": 0.410400390625, "loss_sft": 0.20093296840786934, "loss_total_v6": 0.6106009408831596, "step": 4633 }, { "epoch": 1.0687279279279278, "grad_norm": 32.31028366088867, "learning_rate": 3.932880540202941e-06, "loss": 10.09538745880127, "loss_alignment": 0.84130859375, "loss_alignment_w": 0.420654296875, "loss_sft": 0.21035321988165379, "loss_total_v6": 0.6309617087244987, "step": 4634 }, { "epoch": 1.0689585585585586, "grad_norm": 24.756410598754883, "learning_rate": 3.93091318489133e-06, "loss": 9.543338775634766, "loss_alignment": 0.826171875, "loss_alignment_w": 0.4130859375, "loss_sft": 0.18381522223353386, "loss_total_v6": 0.5964586734771729, "step": 4635 }, { "epoch": 1.0691891891891891, "grad_norm": 35.907108306884766, "learning_rate": 3.928946003009432e-06, "loss": 9.590744018554688, "loss_alignment": 0.84423828125, "loss_alignment_w": 0.422119140625, "loss_sft": 0.17675301991403103, "loss_total_v6": 0.5994214788079262, "step": 4636 }, { "epoch": 1.06941981981982, "grad_norm": 27.47796058654785, "learning_rate": 3.926978994876366e-06, "loss": 9.471649169921875, "loss_alignment": 0.84814453125, "loss_alignment_w": 0.424072265625, "loss_sft": 0.1690043956041336, "loss_total_v6": 0.5919780284166336, "step": 4637 }, { "epoch": 1.0696504504504505, "grad_norm": 26.80740737915039, "learning_rate": 3.925012160811231e-06, "loss": 9.590873718261719, "loss_alignment": 0.822265625, "loss_alignment_w": 0.4111328125, "loss_sft": 0.18925806786864996, "loss_total_v6": 0.5994295962154865, "step": 4638 }, { "epoch": 1.069881081081081, "grad_norm": 26.345821380615234, "learning_rate": 3.923045501133088e-06, "loss": 10.012378692626953, "loss_alignment": 0.8603515625, "loss_alignment_w": 0.43017578125, "loss_sft": 0.19536901824176311, "loss_total_v6": 0.6257736757397652, "step": 4639 }, { "epoch": 1.0701117117117116, "grad_norm": 29.619050979614258, "learning_rate": 3.92107901616097e-06, "loss": 9.64331340789795, "loss_alignment": 0.8447265625, "loss_alignment_w": 0.42236328125, "loss_sft": 0.1803590776398778, "loss_total_v6": 0.6027070805430412, "step": 4640 }, { "epoch": 1.0703423423423424, "grad_norm": 35.57518005371094, "learning_rate": 3.919112706213889e-06, "loss": 9.83188247680664, "loss_alignment": 0.833984375, "loss_alignment_w": 0.4169921875, "loss_sft": 0.1967069897800684, "loss_total_v6": 0.6144926473498344, "step": 4641 }, { "epoch": 1.070572972972973, "grad_norm": 30.952564239501953, "learning_rate": 3.917146571610821e-06, "loss": 10.125555038452148, "loss_alignment": 0.83447265625, "loss_alignment_w": 0.417236328125, "loss_sft": 0.21585498750209808, "loss_total_v6": 0.6328471750020981, "step": 4642 }, { "epoch": 1.0708036036036037, "grad_norm": 32.312076568603516, "learning_rate": 3.915180612670719e-06, "loss": 10.410222053527832, "loss_alignment": 0.8515625, "loss_alignment_w": 0.42578125, "loss_sft": 0.22530009783804417, "loss_total_v6": 0.6506388410925865, "step": 4643 }, { "epoch": 1.0710342342342343, "grad_norm": 29.38865089416504, "learning_rate": 3.913214829712506e-06, "loss": 9.599666595458984, "loss_alignment": 0.8046875, "loss_alignment_w": 0.40234375, "loss_sft": 0.1975744366645813, "loss_total_v6": 0.5999792143702507, "step": 4644 }, { "epoch": 1.0712648648648648, "grad_norm": 26.857419967651367, "learning_rate": 3.911249223055073e-06, "loss": 9.257661819458008, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.1916257357224822, "loss_total_v6": 0.5786038935184479, "step": 4645 }, { "epoch": 1.0714954954954954, "grad_norm": 29.292110443115234, "learning_rate": 3.909283793017289e-06, "loss": 10.055176734924316, "loss_alignment": 0.8251953125, "loss_alignment_w": 0.41259765625, "loss_sft": 0.21679691597819328, "loss_total_v6": 0.6284485384821892, "step": 4646 }, { "epoch": 1.0717261261261262, "grad_norm": 29.256330490112305, "learning_rate": 3.907318539917985e-06, "loss": 9.721658706665039, "loss_alignment": 0.8173828125, "loss_alignment_w": 0.40869140625, "loss_sft": 0.1998125147074461, "loss_total_v6": 0.6076036393642426, "step": 4647 }, { "epoch": 1.0719567567567567, "grad_norm": 24.85477066040039, "learning_rate": 3.905353464075975e-06, "loss": 9.8324556350708, "loss_alignment": 0.828125, "loss_alignment_w": 0.4140625, "loss_sft": 0.20107636228203773, "loss_total_v6": 0.6145284995436668, "step": 4648 }, { "epoch": 1.0721873873873875, "grad_norm": 30.89145851135254, "learning_rate": 3.903388565810034e-06, "loss": 9.628486633300781, "loss_alignment": 0.82421875, "loss_alignment_w": 0.412109375, "loss_sft": 0.1895641889423132, "loss_total_v6": 0.6017803847789764, "step": 4649 }, { "epoch": 1.072418018018018, "grad_norm": 26.72823143005371, "learning_rate": 3.901423845438916e-06, "loss": 10.518898963928223, "loss_alignment": 0.84326171875, "loss_alignment_w": 0.421630859375, "loss_sft": 0.23714307136833668, "loss_total_v6": 0.6574311852455139, "step": 4650 }, { "epoch": 1.0726486486486486, "grad_norm": 30.22020721435547, "learning_rate": 3.8994593032813386e-06, "loss": 9.778552055358887, "loss_alignment": 0.84033203125, "loss_alignment_w": 0.420166015625, "loss_sft": 0.19061201065778732, "loss_total_v6": 0.611159510910511, "step": 4651 }, { "epoch": 1.0728792792792792, "grad_norm": 29.077106475830078, "learning_rate": 3.897494939655996e-06, "loss": 10.022420883178711, "loss_alignment": 0.8359375, "loss_alignment_w": 0.41796875, "loss_sft": 0.20948539394885302, "loss_total_v6": 0.6264012902975082, "step": 4652 }, { "epoch": 1.07310990990991, "grad_norm": 37.44532775878906, "learning_rate": 3.89553075488155e-06, "loss": 9.721792221069336, "loss_alignment": 0.83447265625, "loss_alignment_w": 0.417236328125, "loss_sft": 0.19081825017929077, "loss_total_v6": 0.6076120734214783, "step": 4653 }, { "epoch": 1.0733405405405405, "grad_norm": 31.044099807739258, "learning_rate": 3.893566749276639e-06, "loss": 9.821720123291016, "loss_alignment": 0.8544921875, "loss_alignment_w": 0.42724609375, "loss_sft": 0.1862605269998312, "loss_total_v6": 0.6138575673103333, "step": 4654 }, { "epoch": 1.073571171171171, "grad_norm": 31.335309982299805, "learning_rate": 3.8916029231598655e-06, "loss": 9.35862922668457, "loss_alignment": 0.81884765625, "loss_alignment_w": 0.409423828125, "loss_sft": 0.1750479405745864, "loss_total_v6": 0.5849142745137215, "step": 4655 }, { "epoch": 1.0738018018018018, "grad_norm": 26.485328674316406, "learning_rate": 3.8896392768498076e-06, "loss": 9.99552059173584, "loss_alignment": 0.822265625, "loss_alignment_w": 0.4111328125, "loss_sft": 0.21341937221586704, "loss_total_v6": 0.62472003698349, "step": 4656 }, { "epoch": 1.0740324324324324, "grad_norm": 28.833585739135742, "learning_rate": 3.887675810665009e-06, "loss": 9.793876647949219, "loss_alignment": 0.81005859375, "loss_alignment_w": 0.405029296875, "loss_sft": 0.20711859315633774, "loss_total_v6": 0.6121173426508904, "step": 4657 }, { "epoch": 1.074263063063063, "grad_norm": 30.45791244506836, "learning_rate": 3.885712524923992e-06, "loss": 9.262259483337402, "loss_alignment": 0.783203125, "loss_alignment_w": 0.3916015625, "loss_sft": 0.1871065590530634, "loss_total_v6": 0.578891210258007, "step": 4658 }, { "epoch": 1.0744936936936937, "grad_norm": 28.059062957763672, "learning_rate": 3.883749419945244e-06, "loss": 9.778608322143555, "loss_alignment": 0.841796875, "loss_alignment_w": 0.4208984375, "loss_sft": 0.19050878565758467, "loss_total_v6": 0.6111630648374557, "step": 4659 }, { "epoch": 1.0747243243243243, "grad_norm": 28.519311904907227, "learning_rate": 3.881786496047224e-06, "loss": 9.667789459228516, "loss_alignment": 0.802734375, "loss_alignment_w": 0.4013671875, "loss_sft": 0.20286970771849155, "loss_total_v6": 0.6042369082570076, "step": 4660 }, { "epoch": 1.0749549549549549, "grad_norm": 29.415607452392578, "learning_rate": 3.8798237535483604e-06, "loss": 9.549007415771484, "loss_alignment": 0.826171875, "loss_alignment_w": 0.4130859375, "loss_sft": 0.18468831665813923, "loss_total_v6": 0.5968129336833954, "step": 4661 }, { "epoch": 1.0751855855855856, "grad_norm": 26.933237075805664, "learning_rate": 3.877861192767056e-06, "loss": 9.673059463500977, "loss_alignment": 0.8173828125, "loss_alignment_w": 0.40869140625, "loss_sft": 0.19613422639667988, "loss_total_v6": 0.604566253721714, "step": 4662 }, { "epoch": 1.0754162162162162, "grad_norm": 29.173025131225586, "learning_rate": 3.875898814021682e-06, "loss": 9.785503387451172, "loss_alignment": 0.8193359375, "loss_alignment_w": 0.40966796875, "loss_sft": 0.2012240905314684, "loss_total_v6": 0.6115939617156982, "step": 4663 }, { "epoch": 1.0756468468468467, "grad_norm": 27.79627799987793, "learning_rate": 3.873936617630578e-06, "loss": 9.465524673461914, "loss_alignment": 0.8046875, "loss_alignment_w": 0.40234375, "loss_sft": 0.18970933370292187, "loss_total_v6": 0.591595321893692, "step": 4664 }, { "epoch": 1.0758774774774775, "grad_norm": 26.837596893310547, "learning_rate": 3.8719746039120595e-06, "loss": 9.68238639831543, "loss_alignment": 0.8349609375, "loss_alignment_w": 0.41748046875, "loss_sft": 0.1871346142143011, "loss_total_v6": 0.6051491647958755, "step": 4665 }, { "epoch": 1.076108108108108, "grad_norm": 25.220890045166016, "learning_rate": 3.870012773184404e-06, "loss": 9.771625518798828, "loss_alignment": 0.83349609375, "loss_alignment_w": 0.416748046875, "loss_sft": 0.19318506866693497, "loss_total_v6": 0.6107265576720238, "step": 4666 }, { "epoch": 1.0763387387387386, "grad_norm": 27.315397262573242, "learning_rate": 3.8680511257658685e-06, "loss": 9.294662475585938, "loss_alignment": 0.81689453125, "loss_alignment_w": 0.408447265625, "loss_sft": 0.17251490242779255, "loss_total_v6": 0.5809163898229599, "step": 4667 }, { "epoch": 1.0765693693693694, "grad_norm": 25.061105728149414, "learning_rate": 3.866089661974674e-06, "loss": 10.280658721923828, "loss_alignment": 0.8427734375, "loss_alignment_w": 0.42138671875, "loss_sft": 0.2216122131794691, "loss_total_v6": 0.6425411626696587, "step": 4668 }, { "epoch": 1.0768, "grad_norm": 28.00990104675293, "learning_rate": 3.864128382129014e-06, "loss": 10.113160133361816, "loss_alignment": 0.84375, "loss_alignment_w": 0.421875, "loss_sft": 0.21094517596065998, "loss_total_v6": 0.6320724785327911, "step": 4669 }, { "epoch": 1.0770306306306305, "grad_norm": 29.066869735717773, "learning_rate": 3.8621672865470505e-06, "loss": 9.959146499633789, "loss_alignment": 0.87841796875, "loss_alignment_w": 0.439208984375, "loss_sft": 0.18334440514445305, "loss_total_v6": 0.622446596622467, "step": 4670 }, { "epoch": 1.0772612612612613, "grad_norm": 28.98397445678711, "learning_rate": 3.860206375546918e-06, "loss": 9.708710670471191, "loss_alignment": 0.8134765625, "loss_alignment_w": 0.40673828125, "loss_sft": 0.2001782264560461, "loss_total_v6": 0.6067944020032883, "step": 4671 }, { "epoch": 1.0774918918918919, "grad_norm": 30.852514266967773, "learning_rate": 3.8582456494467214e-06, "loss": 9.793027877807617, "loss_alignment": 0.83935546875, "loss_alignment_w": 0.419677734375, "loss_sft": 0.19215760007500648, "loss_total_v6": 0.6120642125606537, "step": 4672 }, { "epoch": 1.0777225225225224, "grad_norm": 39.5285758972168, "learning_rate": 3.8562851085645335e-06, "loss": 9.550307273864746, "loss_alignment": 0.80712890625, "loss_alignment_w": 0.403564453125, "loss_sft": 0.19305510260164738, "loss_total_v6": 0.5968942046165466, "step": 4673 }, { "epoch": 1.0779531531531532, "grad_norm": 24.09596061706543, "learning_rate": 3.854324753218396e-06, "loss": 10.222095489501953, "loss_alignment": 0.86669921875, "loss_alignment_w": 0.433349609375, "loss_sft": 0.20542455092072487, "loss_total_v6": 0.6388809978961945, "step": 4674 }, { "epoch": 1.0781837837837838, "grad_norm": 26.89997673034668, "learning_rate": 3.852364583726324e-06, "loss": 9.553842544555664, "loss_alignment": 0.8076171875, "loss_alignment_w": 0.40380859375, "loss_sft": 0.19391690287739038, "loss_total_v6": 0.5971151292324066, "step": 4675 }, { "epoch": 1.0784144144144143, "grad_norm": 26.29710578918457, "learning_rate": 3.850404600406298e-06, "loss": 9.909914016723633, "loss_alignment": 0.8154296875, "loss_alignment_w": 0.40771484375, "loss_sft": 0.21138009149581194, "loss_total_v6": 0.6193696111440659, "step": 4676 }, { "epoch": 1.078645045045045, "grad_norm": 27.997041702270508, "learning_rate": 3.848444803576273e-06, "loss": 10.055837631225586, "loss_alignment": 0.81982421875, "loss_alignment_w": 0.409912109375, "loss_sft": 0.21895918156951666, "loss_total_v6": 0.6284898519515991, "step": 4677 }, { "epoch": 1.0788756756756757, "grad_norm": 34.06844711303711, "learning_rate": 3.846485193554175e-06, "loss": 9.65190601348877, "loss_alignment": 0.83935546875, "loss_alignment_w": 0.419677734375, "loss_sft": 0.1837647706270218, "loss_total_v6": 0.6032441258430481, "step": 4678 }, { "epoch": 1.0791063063063062, "grad_norm": 25.315853118896484, "learning_rate": 3.844525770657892e-06, "loss": 9.065308570861816, "loss_alignment": 0.81689453125, "loss_alignment_w": 0.408447265625, "loss_sft": 0.15888219326734543, "loss_total_v6": 0.5665817856788635, "step": 4679 }, { "epoch": 1.079336936936937, "grad_norm": 28.475845336914062, "learning_rate": 3.842566535205286e-06, "loss": 9.428972244262695, "loss_alignment": 0.78515625, "loss_alignment_w": 0.392578125, "loss_sft": 0.19738877564668655, "loss_total_v6": 0.589310772716999, "step": 4680 }, { "epoch": 1.0795675675675676, "grad_norm": 29.262426376342773, "learning_rate": 3.8406074875141895e-06, "loss": 9.891745567321777, "loss_alignment": 0.84130859375, "loss_alignment_w": 0.420654296875, "loss_sft": 0.19835797976702452, "loss_total_v6": 0.6182340681552887, "step": 4681 }, { "epoch": 1.0797981981981981, "grad_norm": 43.440433502197266, "learning_rate": 3.8386486279024055e-06, "loss": 10.260015487670898, "loss_alignment": 0.85302734375, "loss_alignment_w": 0.426513671875, "loss_sft": 0.21438636630773544, "loss_total_v6": 0.6412510201334953, "step": 4682 }, { "epoch": 1.080028828828829, "grad_norm": 25.970529556274414, "learning_rate": 3.836689956687702e-06, "loss": 9.454611778259277, "loss_alignment": 0.78857421875, "loss_alignment_w": 0.394287109375, "loss_sft": 0.19813676737248898, "loss_total_v6": 0.590913251042366, "step": 4683 }, { "epoch": 1.0802594594594594, "grad_norm": 27.545923233032227, "learning_rate": 3.834731474187823e-06, "loss": 10.063163757324219, "loss_alignment": 0.85546875, "loss_alignment_w": 0.427734375, "loss_sft": 0.20145747996866703, "loss_total_v6": 0.6289477348327637, "step": 4684 }, { "epoch": 1.08049009009009, "grad_norm": 31.36359405517578, "learning_rate": 3.832773180720475e-06, "loss": 9.469324111938477, "loss_alignment": 0.81396484375, "loss_alignment_w": 0.406982421875, "loss_sft": 0.18347701989114285, "loss_total_v6": 0.591832734644413, "step": 4685 }, { "epoch": 1.0807207207207208, "grad_norm": 27.688289642333984, "learning_rate": 3.830815076603335e-06, "loss": 9.820576667785645, "loss_alignment": 0.828125, "loss_alignment_w": 0.4140625, "loss_sft": 0.19879275187849998, "loss_total_v6": 0.6137860342860222, "step": 4686 }, { "epoch": 1.0809513513513513, "grad_norm": 29.52079963684082, "learning_rate": 3.828857162154053e-06, "loss": 9.94534683227539, "loss_alignment": 0.830078125, "loss_alignment_w": 0.4150390625, "loss_sft": 0.20654512103646994, "loss_total_v6": 0.6215841919183731, "step": 4687 }, { "epoch": 1.081181981981982, "grad_norm": 26.83411979675293, "learning_rate": 3.82689943769025e-06, "loss": 9.632879257202148, "loss_alignment": 0.82177734375, "loss_alignment_w": 0.410888671875, "loss_sft": 0.19208182394504547, "loss_total_v6": 0.6020549535751343, "step": 4688 }, { "epoch": 1.0814126126126127, "grad_norm": 31.172088623046875, "learning_rate": 3.824941903529508e-06, "loss": 9.587677001953125, "loss_alignment": 0.80224609375, "loss_alignment_w": 0.401123046875, "loss_sft": 0.19841188937425613, "loss_total_v6": 0.5992297902703285, "step": 4689 }, { "epoch": 1.0816432432432432, "grad_norm": 29.314414978027344, "learning_rate": 3.822984559989385e-06, "loss": 9.869675636291504, "loss_alignment": 0.80859375, "loss_alignment_w": 0.404296875, "loss_sft": 0.21300032921135426, "loss_total_v6": 0.6168547049164772, "step": 4690 }, { "epoch": 1.0818738738738738, "grad_norm": 29.554237365722656, "learning_rate": 3.821027407387402e-06, "loss": 9.880731582641602, "loss_alignment": 0.79638671875, "loss_alignment_w": 0.398193359375, "loss_sft": 0.21872674487531185, "loss_total_v6": 0.6175456866621971, "step": 4691 }, { "epoch": 1.0821045045045046, "grad_norm": 28.187654495239258, "learning_rate": 3.819070446041059e-06, "loss": 9.709386825561523, "loss_alignment": 0.82080078125, "loss_alignment_w": 0.410400390625, "loss_sft": 0.1973823681473732, "loss_total_v6": 0.6068367138504982, "step": 4692 }, { "epoch": 1.0823351351351351, "grad_norm": 30.5041561126709, "learning_rate": 3.817113676267813e-06, "loss": 9.766196250915527, "loss_alignment": 0.80810546875, "loss_alignment_w": 0.404052734375, "loss_sft": 0.2063345331698656, "loss_total_v6": 0.6103872656822205, "step": 4693 }, { "epoch": 1.0825657657657657, "grad_norm": 26.726987838745117, "learning_rate": 3.8151570983851e-06, "loss": 9.945323944091797, "loss_alignment": 0.8310546875, "loss_alignment_w": 0.41552734375, "loss_sft": 0.20541449636220932, "loss_total_v6": 0.6215827167034149, "step": 4694 }, { "epoch": 1.0827963963963965, "grad_norm": 26.785980224609375, "learning_rate": 3.813200712710317e-06, "loss": 10.020453453063965, "loss_alignment": 0.826171875, "loss_alignment_w": 0.4130859375, "loss_sft": 0.21279567666351795, "loss_total_v6": 0.6262783333659172, "step": 4695 }, { "epoch": 1.083027027027027, "grad_norm": 27.60404396057129, "learning_rate": 3.811244519560834e-06, "loss": 9.54301929473877, "loss_alignment": 0.814453125, "loss_alignment_w": 0.4072265625, "loss_sft": 0.1899293102324009, "loss_total_v6": 0.5964387282729149, "step": 4696 }, { "epoch": 1.0832576576576576, "grad_norm": 28.434846878051758, "learning_rate": 3.8092885192539926e-06, "loss": 9.540154457092285, "loss_alignment": 0.81201171875, "loss_alignment_w": 0.406005859375, "loss_sft": 0.1903300816193223, "loss_total_v6": 0.5962596535682678, "step": 4697 }, { "epoch": 1.0834882882882884, "grad_norm": 28.69058609008789, "learning_rate": 3.8073327121070968e-06, "loss": 9.647201538085938, "loss_alignment": 0.83447265625, "loss_alignment_w": 0.417236328125, "loss_sft": 0.18517968989908695, "loss_total_v6": 0.6029500663280487, "step": 4698 }, { "epoch": 1.083718918918919, "grad_norm": 32.66124725341797, "learning_rate": 3.8053770984374215e-06, "loss": 9.286515235900879, "loss_alignment": 0.8046875, "loss_alignment_w": 0.40234375, "loss_sft": 0.17864329926669598, "loss_total_v6": 0.5804072171449661, "step": 4699 }, { "epoch": 1.0839495495495495, "grad_norm": 30.011293411254883, "learning_rate": 3.803421678562213e-06, "loss": 10.031734466552734, "loss_alignment": 0.82177734375, "loss_alignment_w": 0.410888671875, "loss_sft": 0.2148587740957737, "loss_total_v6": 0.6269834265112877, "step": 4700 }, { "epoch": 1.0841801801801803, "grad_norm": 28.594524383544922, "learning_rate": 3.8014664527986807e-06, "loss": 9.668631553649902, "loss_alignment": 0.80029296875, "loss_alignment_w": 0.400146484375, "loss_sft": 0.20334956981241703, "loss_total_v6": 0.6042894944548607, "step": 4701 }, { "epoch": 1.0844108108108108, "grad_norm": 28.503192901611328, "learning_rate": 3.7995114214640085e-06, "loss": 9.86817741394043, "loss_alignment": 0.8056640625, "loss_alignment_w": 0.40283203125, "loss_sft": 0.2144173476845026, "loss_total_v6": 0.6167610809206963, "step": 4702 }, { "epoch": 1.0846414414414414, "grad_norm": 28.98988914489746, "learning_rate": 3.797556584875347e-06, "loss": 9.770508766174316, "loss_alignment": 0.806640625, "loss_alignment_w": 0.4033203125, "loss_sft": 0.20722969621419907, "loss_total_v6": 0.6106568053364754, "step": 4703 }, { "epoch": 1.0848720720720721, "grad_norm": 24.41766929626465, "learning_rate": 3.7956019433498124e-06, "loss": 10.158016204833984, "loss_alignment": 0.8583984375, "loss_alignment_w": 0.42919921875, "loss_sft": 0.20543264411389828, "loss_total_v6": 0.6348759979009628, "step": 4704 }, { "epoch": 1.0851027027027027, "grad_norm": 26.81346321105957, "learning_rate": 3.7936474972044905e-06, "loss": 9.55074691772461, "loss_alignment": 0.8203125, "loss_alignment_w": 0.41015625, "loss_sft": 0.18752840906381607, "loss_total_v6": 0.5969216972589493, "step": 4705 }, { "epoch": 1.0853333333333333, "grad_norm": 26.226945877075195, "learning_rate": 3.791693246756436e-06, "loss": 9.99193000793457, "loss_alignment": 0.82275390625, "loss_alignment_w": 0.411376953125, "loss_sft": 0.21400367841124535, "loss_total_v6": 0.6244956031441689, "step": 4706 }, { "epoch": 1.085563963963964, "grad_norm": 28.230684280395508, "learning_rate": 3.7897391923226753e-06, "loss": 9.627337455749512, "loss_alignment": 0.78125, "loss_alignment_w": 0.390625, "loss_sft": 0.21216695569455624, "loss_total_v6": 0.6017085760831833, "step": 4707 }, { "epoch": 1.0857945945945946, "grad_norm": 29.123287200927734, "learning_rate": 3.787785334220196e-06, "loss": 9.486701965332031, "loss_alignment": 0.78271484375, "loss_alignment_w": 0.391357421875, "loss_sft": 0.2007069457322359, "loss_total_v6": 0.5929188579320908, "step": 4708 }, { "epoch": 1.0860252252252252, "grad_norm": 32.366825103759766, "learning_rate": 3.7858316727659595e-06, "loss": 9.46120834350586, "loss_alignment": 0.83203125, "loss_alignment_w": 0.416015625, "loss_sft": 0.17494363337755203, "loss_total_v6": 0.591325506567955, "step": 4709 }, { "epoch": 1.086255855855856, "grad_norm": 25.73163604736328, "learning_rate": 3.78387820827689e-06, "loss": 10.042957305908203, "loss_alignment": 0.82275390625, "loss_alignment_w": 0.411376953125, "loss_sft": 0.21504143811762333, "loss_total_v6": 0.6276848912239075, "step": 4710 }, { "epoch": 1.0864864864864865, "grad_norm": 26.027124404907227, "learning_rate": 3.7819249410698877e-06, "loss": 10.246301651000977, "loss_alignment": 0.8740234375, "loss_alignment_w": 0.43701171875, "loss_sft": 0.20387045480310917, "loss_total_v6": 0.6403939127922058, "step": 4711 }, { "epoch": 1.086717117117117, "grad_norm": 28.47724723815918, "learning_rate": 3.779971871461813e-06, "loss": 10.289819717407227, "loss_alignment": 0.8388671875, "loss_alignment_w": 0.41943359375, "loss_sft": 0.22410734184086323, "loss_total_v6": 0.6431136950850487, "step": 4712 }, { "epoch": 1.0869477477477478, "grad_norm": 27.795516967773438, "learning_rate": 3.778018999769499e-06, "loss": 9.314177513122559, "loss_alignment": 0.80615234375, "loss_alignment_w": 0.403076171875, "loss_sft": 0.1791514977812767, "loss_total_v6": 0.5821361094713211, "step": 4713 }, { "epoch": 1.0871783783783784, "grad_norm": 23.943267822265625, "learning_rate": 3.776066326309743e-06, "loss": 9.935159683227539, "loss_alignment": 0.818359375, "loss_alignment_w": 0.4091796875, "loss_sft": 0.21054707653820515, "loss_total_v6": 0.6209474802017212, "step": 4714 }, { "epoch": 1.087409009009009, "grad_norm": 27.92660140991211, "learning_rate": 3.774113851399312e-06, "loss": 9.43320083618164, "loss_alignment": 0.830078125, "loss_alignment_w": 0.4150390625, "loss_sft": 0.17410872504115105, "loss_total_v6": 0.5895750373601913, "step": 4715 }, { "epoch": 1.0876396396396397, "grad_norm": 28.669706344604492, "learning_rate": 3.7721615753549446e-06, "loss": 9.72812271118164, "loss_alignment": 0.8056640625, "loss_alignment_w": 0.40283203125, "loss_sft": 0.20502310618758202, "loss_total_v6": 0.6080077067017555, "step": 4716 }, { "epoch": 1.0878702702702703, "grad_norm": 31.393945693969727, "learning_rate": 3.7702094984933406e-06, "loss": 9.638788223266602, "loss_alignment": 0.8232421875, "loss_alignment_w": 0.41162109375, "loss_sft": 0.19055905379354954, "loss_total_v6": 0.602424293756485, "step": 4717 }, { "epoch": 1.0881009009009008, "grad_norm": 28.087501525878906, "learning_rate": 3.7682576211311706e-06, "loss": 9.880993843078613, "loss_alignment": 0.81640625, "loss_alignment_w": 0.408203125, "loss_sft": 0.20993882417678833, "loss_total_v6": 0.6175621002912521, "step": 4718 }, { "epoch": 1.0883315315315316, "grad_norm": 29.189952850341797, "learning_rate": 3.7663059435850735e-06, "loss": 9.374166488647461, "loss_alignment": 0.8056640625, "loss_alignment_w": 0.40283203125, "loss_sft": 0.1822752058506012, "loss_total_v6": 0.5858854278922081, "step": 4719 }, { "epoch": 1.0885621621621622, "grad_norm": 28.535446166992188, "learning_rate": 3.7643544661716518e-06, "loss": 10.212974548339844, "loss_alignment": 0.8349609375, "loss_alignment_w": 0.41748046875, "loss_sft": 0.22173075005412102, "loss_total_v6": 0.6383109390735626, "step": 4720 }, { "epoch": 1.0887927927927927, "grad_norm": 28.329219818115234, "learning_rate": 3.762403189207483e-06, "loss": 9.82348918914795, "loss_alignment": 0.8056640625, "loss_alignment_w": 0.40283203125, "loss_sft": 0.21128867380321026, "loss_total_v6": 0.6139681041240692, "step": 4721 }, { "epoch": 1.0890234234234235, "grad_norm": 29.42761993408203, "learning_rate": 3.7604521130091065e-06, "loss": 9.947542190551758, "loss_alignment": 0.8544921875, "loss_alignment_w": 0.42724609375, "loss_sft": 0.1953755235299468, "loss_total_v6": 0.6217213422060013, "step": 4722 }, { "epoch": 1.089254054054054, "grad_norm": 30.470666885375977, "learning_rate": 3.7585012378930296e-06, "loss": 9.476150512695312, "loss_alignment": 0.802734375, "loss_alignment_w": 0.4013671875, "loss_sft": 0.19086174294352531, "loss_total_v6": 0.5922594591975212, "step": 4723 }, { "epoch": 1.0894846846846846, "grad_norm": 29.83234214782715, "learning_rate": 3.756550564175727e-06, "loss": 8.816438674926758, "loss_alignment": 0.7880859375, "loss_alignment_w": 0.39404296875, "loss_sft": 0.15669455379247665, "loss_total_v6": 0.5510274097323418, "step": 4724 }, { "epoch": 1.0897153153153154, "grad_norm": 31.254152297973633, "learning_rate": 3.7546000921736413e-06, "loss": 9.951452255249023, "loss_alignment": 0.80712890625, "loss_alignment_w": 0.403564453125, "loss_sft": 0.21776046231389046, "loss_total_v6": 0.621965765953064, "step": 4725 }, { "epoch": 1.089945945945946, "grad_norm": 29.32950782775879, "learning_rate": 3.752649822203186e-06, "loss": 10.358501434326172, "loss_alignment": 0.837890625, "loss_alignment_w": 0.4189453125, "loss_sft": 0.22830847091972828, "loss_total_v6": 0.6474063843488693, "step": 4726 }, { "epoch": 1.0901765765765765, "grad_norm": 35.46923065185547, "learning_rate": 3.7506997545807344e-06, "loss": 10.501991271972656, "loss_alignment": 0.830078125, "loss_alignment_w": 0.4150390625, "loss_sft": 0.24183888360857964, "loss_total_v6": 0.656374417245388, "step": 4727 }, { "epoch": 1.0904072072072073, "grad_norm": 27.539337158203125, "learning_rate": 3.748749889622634e-06, "loss": 10.202449798583984, "loss_alignment": 0.873046875, "loss_alignment_w": 0.4365234375, "loss_sft": 0.2018926339223981, "loss_total_v6": 0.6376530975103378, "step": 4728 }, { "epoch": 1.0906378378378379, "grad_norm": 30.507951736450195, "learning_rate": 3.7468002276451943e-06, "loss": 9.547492027282715, "loss_alignment": 0.80615234375, "loss_alignment_w": 0.403076171875, "loss_sft": 0.1942219529300928, "loss_total_v6": 0.5967182815074921, "step": 4729 }, { "epoch": 1.0908684684684684, "grad_norm": 27.520511627197266, "learning_rate": 3.744850768964692e-06, "loss": 10.004011154174805, "loss_alignment": 0.806640625, "loss_alignment_w": 0.4033203125, "loss_sft": 0.221365787088871, "loss_total_v6": 0.6252506822347641, "step": 4730 }, { "epoch": 1.0910990990990992, "grad_norm": 27.316272735595703, "learning_rate": 3.742901513897377e-06, "loss": 10.164484024047852, "loss_alignment": 0.81591796875, "loss_alignment_w": 0.407958984375, "loss_sft": 0.22690930031239986, "loss_total_v6": 0.6352802589535713, "step": 4731 }, { "epoch": 1.0913297297297297, "grad_norm": 28.125045776367188, "learning_rate": 3.7409524627594607e-06, "loss": 10.072586059570312, "loss_alignment": 0.8154296875, "loss_alignment_w": 0.40771484375, "loss_sft": 0.2226151954382658, "loss_total_v6": 0.6295365989208221, "step": 4732 }, { "epoch": 1.0915603603603603, "grad_norm": 25.9344425201416, "learning_rate": 3.739003615867122e-06, "loss": 9.687944412231445, "loss_alignment": 0.8251953125, "loss_alignment_w": 0.41259765625, "loss_sft": 0.19306676276028156, "loss_total_v6": 0.6054965928196907, "step": 4733 }, { "epoch": 1.091790990990991, "grad_norm": 29.86565399169922, "learning_rate": 3.7370549735365075e-06, "loss": 9.251744270324707, "loss_alignment": 0.7978515625, "loss_alignment_w": 0.39892578125, "loss_sft": 0.17923194076865911, "loss_total_v6": 0.5782340243458748, "step": 4734 }, { "epoch": 1.0920216216216216, "grad_norm": 26.05658531188965, "learning_rate": 3.7351065360837284e-06, "loss": 9.561321258544922, "loss_alignment": 0.81494140625, "loss_alignment_w": 0.407470703125, "loss_sft": 0.19133258797228336, "loss_total_v6": 0.597582571208477, "step": 4735 }, { "epoch": 1.0922522522522522, "grad_norm": 33.71010208129883, "learning_rate": 3.7331583038248688e-06, "loss": 10.120258331298828, "loss_alignment": 0.8525390625, "loss_alignment_w": 0.42626953125, "loss_sft": 0.20606346800923347, "loss_total_v6": 0.6325161159038544, "step": 4736 }, { "epoch": 1.092482882882883, "grad_norm": 34.44001770019531, "learning_rate": 3.7312102770759724e-06, "loss": 9.78143310546875, "loss_alignment": 0.82763671875, "loss_alignment_w": 0.413818359375, "loss_sft": 0.19602582789957523, "loss_total_v6": 0.6113395392894745, "step": 4737 }, { "epoch": 1.0927135135135135, "grad_norm": 35.265499114990234, "learning_rate": 3.7292624561530544e-06, "loss": 9.462379455566406, "loss_alignment": 0.82666015625, "loss_alignment_w": 0.413330078125, "loss_sft": 0.17725992668420076, "loss_total_v6": 0.5913987085223198, "step": 4738 }, { "epoch": 1.092944144144144, "grad_norm": 27.539268493652344, "learning_rate": 3.727314841372093e-06, "loss": 9.545839309692383, "loss_alignment": 0.82373046875, "loss_alignment_w": 0.411865234375, "loss_sft": 0.18508541956543922, "loss_total_v6": 0.5966149643063545, "step": 4739 }, { "epoch": 1.0931747747747749, "grad_norm": 31.600154876708984, "learning_rate": 3.725367433049033e-06, "loss": 9.312334060668945, "loss_alignment": 0.78955078125, "loss_alignment_w": 0.394775390625, "loss_sft": 0.18719972856342793, "loss_total_v6": 0.5820209085941315, "step": 4740 }, { "epoch": 1.0934054054054054, "grad_norm": 31.209163665771484, "learning_rate": 3.7234202314997935e-06, "loss": 10.080829620361328, "loss_alignment": 0.8115234375, "loss_alignment_w": 0.40576171875, "loss_sft": 0.2241833321750164, "loss_total_v6": 0.6300518661737442, "step": 4741 }, { "epoch": 1.093636036036036, "grad_norm": 26.867876052856445, "learning_rate": 3.7214732370402494e-06, "loss": 10.09062671661377, "loss_alignment": 0.85888671875, "loss_alignment_w": 0.429443359375, "loss_sft": 0.20099193416535854, "loss_total_v6": 0.6306641772389412, "step": 4742 }, { "epoch": 1.0938666666666668, "grad_norm": 27.096181869506836, "learning_rate": 3.7195264499862477e-06, "loss": 9.842615127563477, "loss_alignment": 0.8212890625, "loss_alignment_w": 0.41064453125, "loss_sft": 0.20432052668184042, "loss_total_v6": 0.6151634156703949, "step": 4743 }, { "epoch": 1.0940972972972973, "grad_norm": 31.267742156982422, "learning_rate": 3.7175798706536014e-06, "loss": 9.41340446472168, "loss_alignment": 0.8125, "loss_alignment_w": 0.40625, "loss_sft": 0.18227085284888744, "loss_total_v6": 0.5883377641439438, "step": 4744 }, { "epoch": 1.0943279279279279, "grad_norm": 27.83660316467285, "learning_rate": 3.7156334993580854e-06, "loss": 9.939244270324707, "loss_alignment": 0.81591796875, "loss_alignment_w": 0.407958984375, "loss_sft": 0.21303017251193523, "loss_total_v6": 0.6212027594447136, "step": 4745 }, { "epoch": 1.0945585585585587, "grad_norm": 26.159908294677734, "learning_rate": 3.7136873364154493e-06, "loss": 9.608865737915039, "loss_alignment": 0.80810546875, "loss_alignment_w": 0.404052734375, "loss_sft": 0.19526544585824013, "loss_total_v6": 0.6005541384220123, "step": 4746 }, { "epoch": 1.0947891891891892, "grad_norm": 29.938560485839844, "learning_rate": 3.7117413821414028e-06, "loss": 9.850542068481445, "loss_alignment": 0.82763671875, "loss_alignment_w": 0.413818359375, "loss_sft": 0.1997042652219534, "loss_total_v6": 0.6156588718295097, "step": 4747 }, { "epoch": 1.0950198198198198, "grad_norm": 28.42997932434082, "learning_rate": 3.7097956368516225e-06, "loss": 9.419516563415527, "loss_alignment": 0.82763671875, "loss_alignment_w": 0.413818359375, "loss_sft": 0.17529813013970852, "loss_total_v6": 0.5887197628617287, "step": 4748 }, { "epoch": 1.0952504504504506, "grad_norm": 31.0455379486084, "learning_rate": 3.7078501008617505e-06, "loss": 9.94354248046875, "loss_alignment": 0.8466796875, "loss_alignment_w": 0.42333984375, "loss_sft": 0.19912339001893997, "loss_total_v6": 0.6214713752269745, "step": 4749 }, { "epoch": 1.095481081081081, "grad_norm": 30.861881256103516, "learning_rate": 3.705904774487396e-06, "loss": 9.653179168701172, "loss_alignment": 0.8271484375, "loss_alignment_w": 0.41357421875, "loss_sft": 0.1900546494871378, "loss_total_v6": 0.6033236682415009, "step": 4750 }, { "epoch": 1.0957117117117117, "grad_norm": 32.18990707397461, "learning_rate": 3.7039596580441385e-06, "loss": 9.929205894470215, "loss_alignment": 0.83056640625, "loss_alignment_w": 0.415283203125, "loss_sft": 0.20478864759206772, "loss_total_v6": 0.6205753833055496, "step": 4751 }, { "epoch": 1.0959423423423424, "grad_norm": 28.181415557861328, "learning_rate": 3.702014751847514e-06, "loss": 9.619260787963867, "loss_alignment": 0.814453125, "loss_alignment_w": 0.4072265625, "loss_sft": 0.19377887435257435, "loss_total_v6": 0.6012037917971611, "step": 4752 }, { "epoch": 1.096172972972973, "grad_norm": 25.908559799194336, "learning_rate": 3.7000700562130336e-06, "loss": 9.484014511108398, "loss_alignment": 0.8173828125, "loss_alignment_w": 0.40869140625, "loss_sft": 0.1835254654288292, "loss_total_v6": 0.5927509218454361, "step": 4753 }, { "epoch": 1.0964036036036036, "grad_norm": 26.856473922729492, "learning_rate": 3.6981255714561687e-06, "loss": 8.942831039428711, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.17237605899572372, "loss_total_v6": 0.558926947414875, "step": 4754 }, { "epoch": 1.0966342342342341, "grad_norm": 27.797332763671875, "learning_rate": 3.6961812978923557e-06, "loss": 9.995007514953613, "loss_alignment": 0.8642578125, "loss_alignment_w": 0.43212890625, "loss_sft": 0.1924522491171956, "loss_total_v6": 0.624687984585762, "step": 4755 }, { "epoch": 1.096864864864865, "grad_norm": 24.88245391845703, "learning_rate": 3.6942372358370026e-06, "loss": 9.916351318359375, "loss_alignment": 0.8212890625, "loss_alignment_w": 0.41064453125, "loss_sft": 0.20906637236475945, "loss_total_v6": 0.6197719499468803, "step": 4756 }, { "epoch": 1.0970954954954955, "grad_norm": 29.574182510375977, "learning_rate": 3.692293385605479e-06, "loss": 9.347921371459961, "loss_alignment": 0.80517578125, "loss_alignment_w": 0.402587890625, "loss_sft": 0.1808790024369955, "loss_total_v6": 0.5842451006174088, "step": 4757 }, { "epoch": 1.0973261261261262, "grad_norm": 28.945035934448242, "learning_rate": 3.6903497475131196e-06, "loss": 9.459390640258789, "loss_alignment": 0.79248046875, "loss_alignment_w": 0.396240234375, "loss_sft": 0.19526159390807152, "loss_total_v6": 0.5912119150161743, "step": 4758 }, { "epoch": 1.0975567567567568, "grad_norm": 29.95607566833496, "learning_rate": 3.6884063218752252e-06, "loss": 9.40420150756836, "loss_alignment": 0.78759765625, "loss_alignment_w": 0.393798828125, "loss_sft": 0.19431471452116966, "loss_total_v6": 0.5877625942230225, "step": 4759 }, { "epoch": 1.0977873873873873, "grad_norm": 26.664297103881836, "learning_rate": 3.6864631090070656e-06, "loss": 9.294692039489746, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.1963814962655306, "loss_total_v6": 0.5809182599186897, "step": 4760 }, { "epoch": 1.098018018018018, "grad_norm": 27.159469604492188, "learning_rate": 3.684520109223872e-06, "loss": 9.24847412109375, "loss_alignment": 0.80322265625, "loss_alignment_w": 0.401611328125, "loss_sft": 0.1766929319128394, "loss_total_v6": 0.5780296139419079, "step": 4761 }, { "epoch": 1.0982486486486487, "grad_norm": 28.601354598999023, "learning_rate": 3.6825773228408405e-06, "loss": 9.579533576965332, "loss_alignment": 0.8046875, "loss_alignment_w": 0.40234375, "loss_sft": 0.1960566360503435, "loss_total_v6": 0.5987208187580109, "step": 4762 }, { "epoch": 1.0984792792792792, "grad_norm": 27.248262405395508, "learning_rate": 3.680634750173137e-06, "loss": 9.684608459472656, "loss_alignment": 0.818359375, "loss_alignment_w": 0.4091796875, "loss_sft": 0.1958795227110386, "loss_total_v6": 0.6052880957722664, "step": 4763 }, { "epoch": 1.09870990990991, "grad_norm": 27.718841552734375, "learning_rate": 3.6786923915358865e-06, "loss": 9.912429809570312, "loss_alignment": 0.83984375, "loss_alignment_w": 0.419921875, "loss_sft": 0.19994070753455162, "loss_total_v6": 0.6195269003510475, "step": 4764 }, { "epoch": 1.0989405405405406, "grad_norm": 28.507902145385742, "learning_rate": 3.676750247244186e-06, "loss": 9.497940063476562, "loss_alignment": 0.8310546875, "loss_alignment_w": 0.41552734375, "loss_sft": 0.17867378797382116, "loss_total_v6": 0.5936212837696075, "step": 4765 }, { "epoch": 1.0991711711711711, "grad_norm": 33.98233413696289, "learning_rate": 3.6748083176130955e-06, "loss": 9.806076049804688, "loss_alignment": 0.82275390625, "loss_alignment_w": 0.411376953125, "loss_sft": 0.2008619848638773, "loss_total_v6": 0.6128797978162766, "step": 4766 }, { "epoch": 1.0994018018018017, "grad_norm": 27.20705223083496, "learning_rate": 3.6728666029576374e-06, "loss": 9.531588554382324, "loss_alignment": 0.82275390625, "loss_alignment_w": 0.411376953125, "loss_sft": 0.18401164934039116, "loss_total_v6": 0.5957242697477341, "step": 4767 }, { "epoch": 1.0996324324324325, "grad_norm": 31.769573211669922, "learning_rate": 3.6709251035928002e-06, "loss": 9.385916709899902, "loss_alignment": 0.8076171875, "loss_alignment_w": 0.40380859375, "loss_sft": 0.1824297159910202, "loss_total_v6": 0.5866197794675827, "step": 4768 }, { "epoch": 1.099863063063063, "grad_norm": 26.89481544494629, "learning_rate": 3.6689838198335382e-06, "loss": 9.349781036376953, "loss_alignment": 0.78466796875, "loss_alignment_w": 0.392333984375, "loss_sft": 0.19216467253863811, "loss_total_v6": 0.5843613371253014, "step": 4769 }, { "epoch": 1.1000936936936938, "grad_norm": 29.842021942138672, "learning_rate": 3.6670427519947737e-06, "loss": 9.777840614318848, "loss_alignment": 0.84228515625, "loss_alignment_w": 0.421142578125, "loss_sft": 0.18971304967999458, "loss_total_v6": 0.6111150309443474, "step": 4770 }, { "epoch": 1.1003243243243244, "grad_norm": 30.650739669799805, "learning_rate": 3.665101900391389e-06, "loss": 9.787622451782227, "loss_alignment": 0.81201171875, "loss_alignment_w": 0.406005859375, "loss_sft": 0.2044235747307539, "loss_total_v6": 0.6117264479398727, "step": 4771 }, { "epoch": 1.100554954954955, "grad_norm": 28.227190017700195, "learning_rate": 3.6631612653382354e-06, "loss": 9.787135124206543, "loss_alignment": 0.80517578125, "loss_alignment_w": 0.402587890625, "loss_sft": 0.20894018560647964, "loss_total_v6": 0.6116959452629089, "step": 4772 }, { "epoch": 1.1007855855855855, "grad_norm": 30.949289321899414, "learning_rate": 3.6612208471501244e-06, "loss": 10.022650718688965, "loss_alignment": 0.86376953125, "loss_alignment_w": 0.431884765625, "loss_sft": 0.19286770559847355, "loss_total_v6": 0.6264156550168991, "step": 4773 }, { "epoch": 1.1010162162162163, "grad_norm": 27.42755889892578, "learning_rate": 3.6592806461418375e-06, "loss": 9.785236358642578, "loss_alignment": 0.83984375, "loss_alignment_w": 0.419921875, "loss_sft": 0.19257097877562046, "loss_total_v6": 0.6115773245692253, "step": 4774 }, { "epoch": 1.1012468468468468, "grad_norm": 30.522010803222656, "learning_rate": 3.657340662628116e-06, "loss": 9.510021209716797, "loss_alignment": 0.81689453125, "loss_alignment_w": 0.408447265625, "loss_sft": 0.1861579231917858, "loss_total_v6": 0.5943762883543968, "step": 4775 }, { "epoch": 1.1014774774774774, "grad_norm": 29.67198944091797, "learning_rate": 3.655400896923672e-06, "loss": 9.169219970703125, "loss_alignment": 0.765625, "loss_alignment_w": 0.3828125, "loss_sft": 0.1896381676197052, "loss_total_v6": 0.5730762779712677, "step": 4776 }, { "epoch": 1.1017081081081082, "grad_norm": 29.081138610839844, "learning_rate": 3.653461349343176e-06, "loss": 9.927459716796875, "loss_alignment": 0.818359375, "loss_alignment_w": 0.4091796875, "loss_sft": 0.21119504049420357, "loss_total_v6": 0.6204662695527077, "step": 4777 }, { "epoch": 1.1019387387387387, "grad_norm": 28.97766876220703, "learning_rate": 3.651522020201268e-06, "loss": 9.706195831298828, "loss_alignment": 0.82470703125, "loss_alignment_w": 0.412353515625, "loss_sft": 0.19417686574161053, "loss_total_v6": 0.606637179851532, "step": 4778 }, { "epoch": 1.1021693693693693, "grad_norm": 29.877830505371094, "learning_rate": 3.649582909812546e-06, "loss": 9.446231842041016, "loss_alignment": 0.8017578125, "loss_alignment_w": 0.40087890625, "loss_sft": 0.18973944522440434, "loss_total_v6": 0.5903894901275635, "step": 4779 }, { "epoch": 1.1024, "grad_norm": 29.560394287109375, "learning_rate": 3.6476440184915817e-06, "loss": 9.396771430969238, "loss_alignment": 0.81787109375, "loss_alignment_w": 0.408935546875, "loss_sft": 0.1792934061959386, "loss_total_v6": 0.587298184633255, "step": 4780 }, { "epoch": 1.1026306306306306, "grad_norm": 27.65850830078125, "learning_rate": 3.6457053465529056e-06, "loss": 9.865657806396484, "loss_alignment": 0.802734375, "loss_alignment_w": 0.4013671875, "loss_sft": 0.21421410888433456, "loss_total_v6": 0.6166036054491997, "step": 4781 }, { "epoch": 1.1028612612612612, "grad_norm": 29.045146942138672, "learning_rate": 3.643766894311013e-06, "loss": 9.666783332824707, "loss_alignment": 0.7919921875, "loss_alignment_w": 0.39599609375, "loss_sft": 0.20765904523432255, "loss_total_v6": 0.6041739583015442, "step": 4782 }, { "epoch": 1.103091891891892, "grad_norm": 27.562110900878906, "learning_rate": 3.6418286620803623e-06, "loss": 9.406045913696289, "loss_alignment": 0.837890625, "loss_alignment_w": 0.4189453125, "loss_sft": 0.16977181006222963, "loss_total_v6": 0.5878778770565987, "step": 4783 }, { "epoch": 1.1033225225225225, "grad_norm": 29.803882598876953, "learning_rate": 3.639890650175379e-06, "loss": 9.805994033813477, "loss_alignment": 0.79052734375, "loss_alignment_w": 0.395263671875, "loss_sft": 0.21716843731701374, "loss_total_v6": 0.6128746047616005, "step": 4784 }, { "epoch": 1.103553153153153, "grad_norm": 30.044769287109375, "learning_rate": 3.637952858910454e-06, "loss": 9.837968826293945, "loss_alignment": 0.82421875, "loss_alignment_w": 0.412109375, "loss_sft": 0.20383178442716599, "loss_total_v6": 0.614873044192791, "step": 4785 }, { "epoch": 1.1037837837837838, "grad_norm": 32.92253494262695, "learning_rate": 3.6360152885999377e-06, "loss": 9.982693672180176, "loss_alignment": 0.8291015625, "loss_alignment_w": 0.41455078125, "loss_sft": 0.2094743587076664, "loss_total_v6": 0.6239183396100998, "step": 4786 }, { "epoch": 1.1040144144144144, "grad_norm": 30.89983367919922, "learning_rate": 3.6340779395581487e-06, "loss": 9.452975273132324, "loss_alignment": 0.81689453125, "loss_alignment_w": 0.408447265625, "loss_sft": 0.18227213248610497, "loss_total_v6": 0.5908109545707703, "step": 4787 }, { "epoch": 1.104245045045045, "grad_norm": 29.15049934387207, "learning_rate": 3.6321408120993677e-06, "loss": 9.566475868225098, "loss_alignment": 0.80419921875, "loss_alignment_w": 0.402099609375, "loss_sft": 0.19507273193448782, "loss_total_v6": 0.5979047454893589, "step": 4788 }, { "epoch": 1.1044756756756757, "grad_norm": 29.800939559936523, "learning_rate": 3.630203906537838e-06, "loss": 9.583978652954102, "loss_alignment": 0.81982421875, "loss_alignment_w": 0.409912109375, "loss_sft": 0.18867459706962109, "loss_total_v6": 0.598998673260212, "step": 4789 }, { "epoch": 1.1047063063063063, "grad_norm": 28.687862396240234, "learning_rate": 3.6282672231877714e-06, "loss": 9.625872611999512, "loss_alignment": 0.80859375, "loss_alignment_w": 0.404296875, "loss_sft": 0.197991531342268, "loss_total_v6": 0.6016170457005501, "step": 4790 }, { "epoch": 1.1049369369369368, "grad_norm": 26.352127075195312, "learning_rate": 3.6263307623633415e-06, "loss": 10.42211627960205, "loss_alignment": 0.859375, "loss_alignment_w": 0.4296875, "loss_sft": 0.2216184437274933, "loss_total_v6": 0.6513822451233864, "step": 4791 }, { "epoch": 1.1051675675675676, "grad_norm": 30.849950790405273, "learning_rate": 3.624394524378684e-06, "loss": 9.929481506347656, "loss_alignment": 0.8310546875, "loss_alignment_w": 0.41552734375, "loss_sft": 0.20489734038710594, "loss_total_v6": 0.6205925568938255, "step": 4792 }, { "epoch": 1.1053981981981982, "grad_norm": 26.349565505981445, "learning_rate": 3.6224585095479005e-06, "loss": 9.203624725341797, "loss_alignment": 0.8330078125, "loss_alignment_w": 0.41650390625, "loss_sft": 0.159027811139822, "loss_total_v6": 0.5752265378832817, "step": 4793 }, { "epoch": 1.1056288288288287, "grad_norm": 33.0752067565918, "learning_rate": 3.620522718185053e-06, "loss": 9.908124923706055, "loss_alignment": 0.80908203125, "loss_alignment_w": 0.404541015625, "loss_sft": 0.2154492549598217, "loss_total_v6": 0.6192578226327896, "step": 4794 }, { "epoch": 1.1058594594594595, "grad_norm": 32.9907341003418, "learning_rate": 3.6185871506041748e-06, "loss": 9.517197608947754, "loss_alignment": 0.78515625, "loss_alignment_w": 0.392578125, "loss_sft": 0.20198732241988182, "loss_total_v6": 0.5948248654603958, "step": 4795 }, { "epoch": 1.10609009009009, "grad_norm": 29.298446655273438, "learning_rate": 3.616651807119255e-06, "loss": 9.729594230651855, "loss_alignment": 0.8173828125, "loss_alignment_w": 0.40869140625, "loss_sft": 0.2000033389776945, "loss_total_v6": 0.6080996543169022, "step": 4796 }, { "epoch": 1.1063207207207206, "grad_norm": 28.98293685913086, "learning_rate": 3.614716688044251e-06, "loss": 9.373834609985352, "loss_alignment": 0.7958984375, "loss_alignment_w": 0.39794921875, "loss_sft": 0.18756447359919548, "loss_total_v6": 0.5858646556735039, "step": 4797 }, { "epoch": 1.1065513513513514, "grad_norm": 26.553627014160156, "learning_rate": 3.61278179369308e-06, "loss": 9.439668655395508, "loss_alignment": 0.8046875, "loss_alignment_w": 0.40234375, "loss_sft": 0.18772711418569088, "loss_total_v6": 0.5899793058633804, "step": 4798 }, { "epoch": 1.106781981981982, "grad_norm": 30.699066162109375, "learning_rate": 3.610847124379625e-06, "loss": 9.407014846801758, "loss_alignment": 0.79833984375, "loss_alignment_w": 0.399169921875, "loss_sft": 0.1884786132723093, "loss_total_v6": 0.5879384353756905, "step": 4799 }, { "epoch": 1.1070126126126125, "grad_norm": 24.64507293701172, "learning_rate": 3.6089126804177373e-06, "loss": 10.120637893676758, "loss_alignment": 0.8291015625, "loss_alignment_w": 0.41455078125, "loss_sft": 0.21872151270508766, "loss_total_v6": 0.6325398907065392, "step": 4800 }, { "epoch": 1.1072432432432433, "grad_norm": 27.480730056762695, "learning_rate": 3.6069784621212226e-06, "loss": 9.991741180419922, "loss_alignment": 0.814453125, "loss_alignment_w": 0.4072265625, "loss_sft": 0.21800498478114605, "loss_total_v6": 0.6244838684797287, "step": 4801 }, { "epoch": 1.1074738738738739, "grad_norm": 29.881362915039062, "learning_rate": 3.6050444698038547e-06, "loss": 9.75684928894043, "loss_alignment": 0.82861328125, "loss_alignment_w": 0.414306640625, "loss_sft": 0.19548122212290764, "loss_total_v6": 0.6098031252622604, "step": 4802 }, { "epoch": 1.1077045045045044, "grad_norm": 31.275253295898438, "learning_rate": 3.603110703779371e-06, "loss": 9.775323867797852, "loss_alignment": 0.81396484375, "loss_alignment_w": 0.406982421875, "loss_sft": 0.20315133593976498, "loss_total_v6": 0.6109577342867851, "step": 4803 }, { "epoch": 1.1079351351351352, "grad_norm": 28.845949172973633, "learning_rate": 3.6011771643614697e-06, "loss": 10.197636604309082, "loss_alignment": 0.8056640625, "loss_alignment_w": 0.40283203125, "loss_sft": 0.23397094011306763, "loss_total_v6": 0.637352280318737, "step": 4804 }, { "epoch": 1.1081657657657658, "grad_norm": 25.277666091918945, "learning_rate": 3.599243851863816e-06, "loss": 9.73692512512207, "loss_alignment": 0.81591796875, "loss_alignment_w": 0.407958984375, "loss_sft": 0.2021399475634098, "loss_total_v6": 0.6085578054189682, "step": 4805 }, { "epoch": 1.1083963963963963, "grad_norm": 28.79935073852539, "learning_rate": 3.5973107666000374e-06, "loss": 9.88433837890625, "loss_alignment": 0.82177734375, "loss_alignment_w": 0.410888671875, "loss_sft": 0.20679090730845928, "loss_total_v6": 0.6177711337804794, "step": 4806 }, { "epoch": 1.108627027027027, "grad_norm": 30.300939559936523, "learning_rate": 3.595377908883721e-06, "loss": 9.444788932800293, "loss_alignment": 0.79296875, "loss_alignment_w": 0.396484375, "loss_sft": 0.19328085891902447, "loss_total_v6": 0.5902992859482765, "step": 4807 }, { "epoch": 1.1088576576576576, "grad_norm": 28.51726722717285, "learning_rate": 3.593445279028418e-06, "loss": 9.910872459411621, "loss_alignment": 0.8232421875, "loss_alignment_w": 0.41162109375, "loss_sft": 0.20747270248830318, "loss_total_v6": 0.6194294989109039, "step": 4808 }, { "epoch": 1.1090882882882882, "grad_norm": 27.420780181884766, "learning_rate": 3.5915128773476476e-06, "loss": 9.48134708404541, "loss_alignment": 0.81494140625, "loss_alignment_w": 0.407470703125, "loss_sft": 0.18508296366780996, "loss_total_v6": 0.5925841927528381, "step": 4809 }, { "epoch": 1.109318918918919, "grad_norm": 26.596277236938477, "learning_rate": 3.589580704154888e-06, "loss": 9.003650665283203, "loss_alignment": 0.7861328125, "loss_alignment_w": 0.39306640625, "loss_sft": 0.16949394904077053, "loss_total_v6": 0.562728188931942, "step": 4810 }, { "epoch": 1.1095495495495495, "grad_norm": 28.2959041595459, "learning_rate": 3.587648759763579e-06, "loss": 10.13827896118164, "loss_alignment": 0.8369140625, "loss_alignment_w": 0.41845703125, "loss_sft": 0.21437669172883034, "loss_total_v6": 0.6336424052715302, "step": 4811 }, { "epoch": 1.10978018018018, "grad_norm": 30.669757843017578, "learning_rate": 3.585717044487126e-06, "loss": 9.705730438232422, "loss_alignment": 0.83837890625, "loss_alignment_w": 0.419189453125, "loss_sft": 0.1872356366366148, "loss_total_v6": 0.6066081747412682, "step": 4812 }, { "epoch": 1.1100108108108109, "grad_norm": 24.0860595703125, "learning_rate": 3.5837855586388947e-06, "loss": 9.999103546142578, "loss_alignment": 0.81396484375, "loss_alignment_w": 0.406982421875, "loss_sft": 0.21732068434357643, "loss_total_v6": 0.6249439716339111, "step": 4813 }, { "epoch": 1.1102414414414414, "grad_norm": 29.397869110107422, "learning_rate": 3.5818543025322183e-06, "loss": 9.749078750610352, "loss_alignment": 0.798828125, "loss_alignment_w": 0.3994140625, "loss_sft": 0.2114902725443244, "loss_total_v6": 0.6093174070119858, "step": 4814 }, { "epoch": 1.110472072072072, "grad_norm": 29.08426856994629, "learning_rate": 3.579923276480387e-06, "loss": 9.22201919555664, "loss_alignment": 0.806640625, "loss_alignment_w": 0.4033203125, "loss_sft": 0.17276601027697325, "loss_total_v6": 0.5763762295246124, "step": 4815 }, { "epoch": 1.1107027027027028, "grad_norm": 29.423917770385742, "learning_rate": 3.5779924807966583e-06, "loss": 9.409139633178711, "loss_alignment": 0.796875, "loss_alignment_w": 0.3984375, "loss_sft": 0.18934377282857895, "loss_total_v6": 0.5880712196230888, "step": 4816 }, { "epoch": 1.1109333333333333, "grad_norm": 24.50779151916504, "learning_rate": 3.576061915794248e-06, "loss": 10.002848625183105, "loss_alignment": 0.85791015625, "loss_alignment_w": 0.428955078125, "loss_sft": 0.19643662124872208, "loss_total_v6": 0.6251780614256859, "step": 4817 }, { "epoch": 1.1111639639639639, "grad_norm": 26.33169937133789, "learning_rate": 3.5741315817863366e-06, "loss": 9.921783447265625, "loss_alignment": 0.8017578125, "loss_alignment_w": 0.40087890625, "loss_sft": 0.2199955079704523, "loss_total_v6": 0.6201115027070045, "step": 4818 }, { "epoch": 1.1113945945945947, "grad_norm": 30.227764129638672, "learning_rate": 3.5722014790860715e-06, "loss": 9.45389175415039, "loss_alignment": 0.8466796875, "loss_alignment_w": 0.42333984375, "loss_sft": 0.1674215253442526, "loss_total_v6": 0.5908681899309158, "step": 4819 }, { "epoch": 1.1116252252252252, "grad_norm": 32.65107727050781, "learning_rate": 3.5702716080065546e-06, "loss": 9.61886215209961, "loss_alignment": 0.798828125, "loss_alignment_w": 0.3994140625, "loss_sft": 0.20138336718082428, "loss_total_v6": 0.6011788919568062, "step": 4820 }, { "epoch": 1.1118558558558558, "grad_norm": 26.522493362426758, "learning_rate": 3.5683419688608544e-06, "loss": 9.123315811157227, "loss_alignment": 0.79052734375, "loss_alignment_w": 0.395263671875, "loss_sft": 0.17381436005234718, "loss_total_v6": 0.5702072009444237, "step": 4821 }, { "epoch": 1.1120864864864866, "grad_norm": 30.173093795776367, "learning_rate": 3.566412561962003e-06, "loss": 9.593417167663574, "loss_alignment": 0.82568359375, "loss_alignment_w": 0.412841796875, "loss_sft": 0.18750969879329205, "loss_total_v6": 0.5995885580778122, "step": 4822 }, { "epoch": 1.1123171171171171, "grad_norm": 29.582643508911133, "learning_rate": 3.564483387622989e-06, "loss": 9.75387191772461, "loss_alignment": 0.8134765625, "loss_alignment_w": 0.40673828125, "loss_sft": 0.2030770555138588, "loss_total_v6": 0.6096169948577881, "step": 4823 }, { "epoch": 1.1125477477477477, "grad_norm": 36.667503356933594, "learning_rate": 3.562554446156773e-06, "loss": 9.553339004516602, "loss_alignment": 0.8203125, "loss_alignment_w": 0.41015625, "loss_sft": 0.18759882263839245, "loss_total_v6": 0.5970836728811264, "step": 4824 }, { "epoch": 1.1127783783783785, "grad_norm": 28.453128814697266, "learning_rate": 3.56062573787627e-06, "loss": 9.459518432617188, "loss_alignment": 0.81689453125, "loss_alignment_w": 0.408447265625, "loss_sft": 0.1829405091702938, "loss_total_v6": 0.5912199020385742, "step": 4825 }, { "epoch": 1.113009009009009, "grad_norm": 29.930944442749023, "learning_rate": 3.5586972630943593e-06, "loss": 10.15488052368164, "loss_alignment": 0.81689453125, "loss_alignment_w": 0.408447265625, "loss_sft": 0.22537825629115105, "loss_total_v6": 0.6346800401806831, "step": 4826 }, { "epoch": 1.1132396396396396, "grad_norm": 28.337697982788086, "learning_rate": 3.5567690221238824e-06, "loss": 9.572118759155273, "loss_alignment": 0.8154296875, "loss_alignment_w": 0.40771484375, "loss_sft": 0.1913360394537449, "loss_total_v6": 0.5982574224472046, "step": 4827 }, { "epoch": 1.1134702702702703, "grad_norm": 42.61563491821289, "learning_rate": 3.5548410152776414e-06, "loss": 9.164928436279297, "loss_alignment": 0.796875, "loss_alignment_w": 0.3984375, "loss_sft": 0.1736838761717081, "loss_total_v6": 0.572808027267456, "step": 4828 }, { "epoch": 1.113700900900901, "grad_norm": 29.88977813720703, "learning_rate": 3.5529132428684064e-06, "loss": 9.454988479614258, "loss_alignment": 0.822265625, "loss_alignment_w": 0.4111328125, "loss_sft": 0.17926985304802656, "loss_total_v6": 0.5909367203712463, "step": 4829 }, { "epoch": 1.1139315315315315, "grad_norm": 29.5238037109375, "learning_rate": 3.5509857052089013e-06, "loss": 9.116079330444336, "loss_alignment": 0.79345703125, "loss_alignment_w": 0.396728515625, "loss_sft": 0.17244661692529917, "loss_total_v6": 0.5697549730539322, "step": 4830 }, { "epoch": 1.1141621621621622, "grad_norm": 28.097347259521484, "learning_rate": 3.5490584026118175e-06, "loss": 10.040546417236328, "loss_alignment": 0.83935546875, "loss_alignment_w": 0.419677734375, "loss_sft": 0.20761223323643208, "loss_total_v6": 0.6275341287255287, "step": 4831 }, { "epoch": 1.1143927927927928, "grad_norm": 27.253263473510742, "learning_rate": 3.5471313353898056e-06, "loss": 9.62009334564209, "loss_alignment": 0.78466796875, "loss_alignment_w": 0.392333984375, "loss_sft": 0.20884550921618938, "loss_total_v6": 0.6012557968497276, "step": 4832 }, { "epoch": 1.1146234234234234, "grad_norm": 28.280696868896484, "learning_rate": 3.5452045038554773e-06, "loss": 9.84508228302002, "loss_alignment": 0.81103515625, "loss_alignment_w": 0.405517578125, "loss_sft": 0.2102273479104042, "loss_total_v6": 0.6153176575899124, "step": 4833 }, { "epoch": 1.1148540540540541, "grad_norm": 27.40287208557129, "learning_rate": 3.5432779083214095e-06, "loss": 9.887577056884766, "loss_alignment": 0.82080078125, "loss_alignment_w": 0.410400390625, "loss_sft": 0.20745106972754002, "loss_total_v6": 0.6179735288023949, "step": 4834 }, { "epoch": 1.1150846846846847, "grad_norm": 26.168333053588867, "learning_rate": 3.541351549100141e-06, "loss": 9.5499267578125, "loss_alignment": 0.8125, "loss_alignment_w": 0.40625, "loss_sft": 0.1905594188719988, "loss_total_v6": 0.5968704372644424, "step": 4835 }, { "epoch": 1.1153153153153152, "grad_norm": 34.532203674316406, "learning_rate": 3.539425426504166e-06, "loss": 9.991464614868164, "loss_alignment": 0.81982421875, "loss_alignment_w": 0.409912109375, "loss_sft": 0.21455439180135727, "loss_total_v6": 0.6244665086269379, "step": 4836 }, { "epoch": 1.115545945945946, "grad_norm": 27.941078186035156, "learning_rate": 3.537499540845948e-06, "loss": 9.459503173828125, "loss_alignment": 0.8349609375, "loss_alignment_w": 0.41748046875, "loss_sft": 0.17413518391549587, "loss_total_v6": 0.5912189334630966, "step": 4837 }, { "epoch": 1.1157765765765766, "grad_norm": 32.846797943115234, "learning_rate": 3.5355738924379044e-06, "loss": 9.412636756896973, "loss_alignment": 0.7958984375, "loss_alignment_w": 0.39794921875, "loss_sft": 0.18977599777281284, "loss_total_v6": 0.5882898047566414, "step": 4838 }, { "epoch": 1.1160072072072071, "grad_norm": 26.804153442382812, "learning_rate": 3.5336484815924245e-06, "loss": 9.723450660705566, "loss_alignment": 0.8017578125, "loss_alignment_w": 0.40087890625, "loss_sft": 0.2077065259218216, "loss_total_v6": 0.6077156737446785, "step": 4839 }, { "epoch": 1.116237837837838, "grad_norm": 32.37357711791992, "learning_rate": 3.5317233086218474e-06, "loss": 9.816795349121094, "loss_alignment": 0.83447265625, "loss_alignment_w": 0.417236328125, "loss_sft": 0.19680169969797134, "loss_total_v6": 0.6135497316718102, "step": 4840 }, { "epoch": 1.1164684684684685, "grad_norm": 26.822189331054688, "learning_rate": 3.5297983738384813e-06, "loss": 9.69300651550293, "loss_alignment": 0.8125, "loss_alignment_w": 0.40625, "loss_sft": 0.1999138668179512, "loss_total_v6": 0.6058129221200943, "step": 4841 }, { "epoch": 1.116699099099099, "grad_norm": 25.766658782958984, "learning_rate": 3.527873677554593e-06, "loss": 9.413736343383789, "loss_alignment": 0.8056640625, "loss_alignment_w": 0.40283203125, "loss_sft": 0.18635044619441032, "loss_total_v6": 0.5883585065603256, "step": 4842 }, { "epoch": 1.1169297297297298, "grad_norm": 36.369285583496094, "learning_rate": 3.52594922008241e-06, "loss": 9.219720840454102, "loss_alignment": 0.771484375, "loss_alignment_w": 0.3857421875, "loss_sft": 0.19103965628892183, "loss_total_v6": 0.5762325264513493, "step": 4843 }, { "epoch": 1.1171603603603604, "grad_norm": 25.327848434448242, "learning_rate": 3.524025001734126e-06, "loss": 9.629084587097168, "loss_alignment": 0.82275390625, "loss_alignment_w": 0.411376953125, "loss_sft": 0.1898304633796215, "loss_total_v6": 0.6018177941441536, "step": 4844 }, { "epoch": 1.117390990990991, "grad_norm": 27.488771438598633, "learning_rate": 3.52210102282189e-06, "loss": 9.21278190612793, "loss_alignment": 0.7822265625, "loss_alignment_w": 0.39111328125, "loss_sft": 0.1857995092868805, "loss_total_v6": 0.5757988914847374, "step": 4845 }, { "epoch": 1.1176216216216217, "grad_norm": 29.158098220825195, "learning_rate": 3.5201772836578118e-06, "loss": 9.742864608764648, "loss_alignment": 0.81591796875, "loss_alignment_w": 0.407958984375, "loss_sft": 0.20060388185083866, "loss_total_v6": 0.6089290753006935, "step": 4846 }, { "epoch": 1.1178522522522523, "grad_norm": 33.907962799072266, "learning_rate": 3.5182537845539688e-06, "loss": 9.7863187789917, "loss_alignment": 0.818359375, "loss_alignment_w": 0.4091796875, "loss_sft": 0.2030298411846161, "loss_total_v6": 0.6116449385881424, "step": 4847 }, { "epoch": 1.1180828828828828, "grad_norm": 30.050872802734375, "learning_rate": 3.5163305258223913e-06, "loss": 9.38934326171875, "loss_alignment": 0.81982421875, "loss_alignment_w": 0.409912109375, "loss_sft": 0.17769997380673885, "loss_total_v6": 0.5868339091539383, "step": 4848 }, { "epoch": 1.1183135135135136, "grad_norm": 27.346628189086914, "learning_rate": 3.5144075077750783e-06, "loss": 10.114446640014648, "loss_alignment": 0.82421875, "loss_alignment_w": 0.412109375, "loss_sft": 0.21914325095713139, "loss_total_v6": 0.6321529150009155, "step": 4849 }, { "epoch": 1.1185441441441442, "grad_norm": 26.84129524230957, "learning_rate": 3.5124847307239863e-06, "loss": 9.704103469848633, "loss_alignment": 0.79345703125, "loss_alignment_w": 0.396728515625, "loss_sft": 0.20938122831285, "loss_total_v6": 0.6065064668655396, "step": 4850 }, { "epoch": 1.1187747747747747, "grad_norm": 32.7711067199707, "learning_rate": 3.5105621949810304e-06, "loss": 9.589899063110352, "loss_alignment": 0.81201171875, "loss_alignment_w": 0.406005859375, "loss_sft": 0.192462554667145, "loss_total_v6": 0.5993686728179455, "step": 4851 }, { "epoch": 1.1190054054054055, "grad_norm": 26.3369197845459, "learning_rate": 3.5086399008580885e-06, "loss": 9.80984115600586, "loss_alignment": 0.81689453125, "loss_alignment_w": 0.408447265625, "loss_sft": 0.20422527939081192, "loss_total_v6": 0.6131150722503662, "step": 4852 }, { "epoch": 1.119236036036036, "grad_norm": 28.397886276245117, "learning_rate": 3.506717848667e-06, "loss": 9.974048614501953, "loss_alignment": 0.82275390625, "loss_alignment_w": 0.411376953125, "loss_sft": 0.21255039237439632, "loss_total_v6": 0.6233780160546303, "step": 4853 }, { "epoch": 1.1194666666666666, "grad_norm": 34.00255584716797, "learning_rate": 3.5047960387195673e-06, "loss": 9.766654968261719, "loss_alignment": 0.82373046875, "loss_alignment_w": 0.411865234375, "loss_sft": 0.1978029739111662, "loss_total_v6": 0.6104158908128738, "step": 4854 }, { "epoch": 1.1196972972972974, "grad_norm": 27.353437423706055, "learning_rate": 3.5028744713275474e-06, "loss": 9.237709045410156, "loss_alignment": 0.82080078125, "loss_alignment_w": 0.410400390625, "loss_sft": 0.16585775837302208, "loss_total_v6": 0.5773568041622639, "step": 4855 }, { "epoch": 1.119927927927928, "grad_norm": 29.023096084594727, "learning_rate": 3.5009531468026646e-06, "loss": 9.425789833068848, "loss_alignment": 0.7763671875, "loss_alignment_w": 0.38818359375, "loss_sft": 0.20181330665946007, "loss_total_v6": 0.5891118720173836, "step": 4856 }, { "epoch": 1.1201585585585585, "grad_norm": 30.207542419433594, "learning_rate": 3.4990320654565957e-06, "loss": 9.997695922851562, "loss_alignment": 0.83740234375, "loss_alignment_w": 0.418701171875, "loss_sft": 0.20552915334701538, "loss_total_v6": 0.6248559281229973, "step": 4857 }, { "epoch": 1.1203891891891893, "grad_norm": 28.7703857421875, "learning_rate": 3.497111227600988e-06, "loss": 9.795120239257812, "loss_alignment": 0.81201171875, "loss_alignment_w": 0.406005859375, "loss_sft": 0.20649432577192783, "loss_total_v6": 0.6121950149536133, "step": 4858 }, { "epoch": 1.1206198198198198, "grad_norm": 24.097383499145508, "learning_rate": 3.495190633547442e-06, "loss": 9.706199645996094, "loss_alignment": 0.81884765625, "loss_alignment_w": 0.409423828125, "loss_sft": 0.1966033410280943, "loss_total_v6": 0.6066375374794006, "step": 4859 }, { "epoch": 1.1208504504504504, "grad_norm": 27.438655853271484, "learning_rate": 3.4932702836075216e-06, "loss": 9.487314224243164, "loss_alignment": 0.791015625, "loss_alignment_w": 0.3955078125, "loss_sft": 0.19726619496941566, "loss_total_v6": 0.592957116663456, "step": 4860 }, { "epoch": 1.1210810810810812, "grad_norm": 30.77764129638672, "learning_rate": 3.4913501780927485e-06, "loss": 9.570825576782227, "loss_alignment": 0.802734375, "loss_alignment_w": 0.4013671875, "loss_sft": 0.19572602771222591, "loss_total_v6": 0.598176620900631, "step": 4861 }, { "epoch": 1.1213117117117117, "grad_norm": 25.497873306274414, "learning_rate": 3.4894303173146064e-06, "loss": 9.803167343139648, "loss_alignment": 0.82421875, "loss_alignment_w": 0.412109375, "loss_sft": 0.20136685110628605, "loss_total_v6": 0.6126980111002922, "step": 4862 }, { "epoch": 1.1215423423423423, "grad_norm": 27.305797576904297, "learning_rate": 3.487510701584544e-06, "loss": 9.305803298950195, "loss_alignment": 0.7861328125, "loss_alignment_w": 0.39306640625, "loss_sft": 0.18940075859427452, "loss_total_v6": 0.5816127061843872, "step": 4863 }, { "epoch": 1.121772972972973, "grad_norm": 29.779298782348633, "learning_rate": 3.4855913312139624e-06, "loss": 9.137899398803711, "loss_alignment": 0.78076171875, "loss_alignment_w": 0.390380859375, "loss_sft": 0.18044794350862503, "loss_total_v6": 0.5711186975240707, "step": 4864 }, { "epoch": 1.1220036036036036, "grad_norm": 27.715774536132812, "learning_rate": 3.483672206514226e-06, "loss": 9.154396057128906, "loss_alignment": 0.79931640625, "loss_alignment_w": 0.399658203125, "loss_sft": 0.17201851960271597, "loss_total_v6": 0.5721497163176537, "step": 4865 }, { "epoch": 1.1222342342342342, "grad_norm": 30.62535285949707, "learning_rate": 3.4817533277966604e-06, "loss": 9.912768363952637, "loss_alignment": 0.8076171875, "loss_alignment_w": 0.40380859375, "loss_sft": 0.21557155158370733, "loss_total_v6": 0.6195480152964592, "step": 4866 }, { "epoch": 1.122464864864865, "grad_norm": 28.176271438598633, "learning_rate": 3.4798346953725487e-06, "loss": 9.359373092651367, "loss_alignment": 0.7841796875, "loss_alignment_w": 0.39208984375, "loss_sft": 0.1926726195961237, "loss_total_v6": 0.584960825741291, "step": 4867 }, { "epoch": 1.1226954954954955, "grad_norm": 27.465118408203125, "learning_rate": 3.4779163095531387e-06, "loss": 9.20920181274414, "loss_alignment": 0.77392578125, "loss_alignment_w": 0.386962890625, "loss_sft": 0.18906998075544834, "loss_total_v6": 0.5755750983953476, "step": 4868 }, { "epoch": 1.122926126126126, "grad_norm": 26.890949249267578, "learning_rate": 3.4759981706496346e-06, "loss": 9.090747833251953, "loss_alignment": 0.8017578125, "loss_alignment_w": 0.40087890625, "loss_sft": 0.16593477502465248, "loss_total_v6": 0.5681717023253441, "step": 4869 }, { "epoch": 1.1231567567567569, "grad_norm": 28.406221389770508, "learning_rate": 3.474080278973202e-06, "loss": 9.345900535583496, "loss_alignment": 0.80322265625, "loss_alignment_w": 0.401611328125, "loss_sft": 0.18356028944253922, "loss_total_v6": 0.5841187760233879, "step": 4870 }, { "epoch": 1.1233873873873874, "grad_norm": 30.121713638305664, "learning_rate": 3.4721626348349625e-06, "loss": 10.055463790893555, "loss_alignment": 0.82861328125, "loss_alignment_w": 0.414306640625, "loss_sft": 0.21292392630130053, "loss_total_v6": 0.6284665018320084, "step": 4871 }, { "epoch": 1.123618018018018, "grad_norm": 27.762939453125, "learning_rate": 3.470245238546002e-06, "loss": 9.333099365234375, "loss_alignment": 0.78955078125, "loss_alignment_w": 0.394775390625, "loss_sft": 0.18832971341907978, "loss_total_v6": 0.583318717777729, "step": 4872 }, { "epoch": 1.1238486486486488, "grad_norm": 29.109750747680664, "learning_rate": 3.468328090417368e-06, "loss": 9.505585670471191, "loss_alignment": 0.80712890625, "loss_alignment_w": 0.403564453125, "loss_sft": 0.18995479494333267, "loss_total_v6": 0.5940991044044495, "step": 4873 }, { "epoch": 1.1240792792792793, "grad_norm": 29.3408145904541, "learning_rate": 3.4664111907600606e-06, "loss": 10.137445449829102, "loss_alignment": 0.84130859375, "loss_alignment_w": 0.420654296875, "loss_sft": 0.21354633197188377, "loss_total_v6": 0.6335903033614159, "step": 4874 }, { "epoch": 1.1243099099099099, "grad_norm": 28.675453186035156, "learning_rate": 3.464494539885047e-06, "loss": 9.373857498168945, "loss_alignment": 0.83740234375, "loss_alignment_w": 0.418701171875, "loss_sft": 0.167546390555799, "loss_total_v6": 0.5858660787343979, "step": 4875 }, { "epoch": 1.1245405405405404, "grad_norm": 27.115022659301758, "learning_rate": 3.4625781381032484e-06, "loss": 10.092048645019531, "loss_alignment": 0.83935546875, "loss_alignment_w": 0.419677734375, "loss_sft": 0.21212817169725895, "loss_total_v6": 0.6307530403137207, "step": 4876 }, { "epoch": 1.1247711711711712, "grad_norm": 27.994619369506836, "learning_rate": 3.4606619857255475e-06, "loss": 9.97521686553955, "loss_alignment": 0.84130859375, "loss_alignment_w": 0.420654296875, "loss_sft": 0.20305616408586502, "loss_total_v6": 0.6234510689973831, "step": 4877 }, { "epoch": 1.1250018018018018, "grad_norm": 28.967313766479492, "learning_rate": 3.4587460830627884e-06, "loss": 9.66523265838623, "loss_alignment": 0.8369140625, "loss_alignment_w": 0.41845703125, "loss_sft": 0.1848265491425991, "loss_total_v6": 0.6040770336985588, "step": 4878 }, { "epoch": 1.1252324324324325, "grad_norm": 29.926443099975586, "learning_rate": 3.456830430425774e-06, "loss": 9.997735977172852, "loss_alignment": 0.8046875, "loss_alignment_w": 0.40234375, "loss_sft": 0.22251472063362598, "loss_total_v6": 0.6248584613204002, "step": 4879 }, { "epoch": 1.125463063063063, "grad_norm": 34.83522033691406, "learning_rate": 3.4549150281252635e-06, "loss": 9.384861946105957, "loss_alignment": 0.79296875, "loss_alignment_w": 0.396484375, "loss_sft": 0.1904509561136365, "loss_total_v6": 0.5865538641810417, "step": 4880 }, { "epoch": 1.1256936936936937, "grad_norm": 27.376998901367188, "learning_rate": 3.45299987647198e-06, "loss": 9.656916618347168, "loss_alignment": 0.81591796875, "loss_alignment_w": 0.407958984375, "loss_sft": 0.19463703129440546, "loss_total_v6": 0.6035573072731495, "step": 4881 }, { "epoch": 1.1259243243243242, "grad_norm": 28.39662742614746, "learning_rate": 3.4510849757766014e-06, "loss": 9.726849555969238, "loss_alignment": 0.82763671875, "loss_alignment_w": 0.413818359375, "loss_sft": 0.19415553100407124, "loss_total_v6": 0.6079281195998192, "step": 4882 }, { "epoch": 1.126154954954955, "grad_norm": 26.750715255737305, "learning_rate": 3.44917032634977e-06, "loss": 9.153735160827637, "loss_alignment": 0.79638671875, "loss_alignment_w": 0.398193359375, "loss_sft": 0.1734420908614993, "loss_total_v6": 0.5721084773540497, "step": 4883 }, { "epoch": 1.1263855855855855, "grad_norm": 29.343217849731445, "learning_rate": 3.447255928502083e-06, "loss": 9.447839736938477, "loss_alignment": 0.81689453125, "loss_alignment_w": 0.408447265625, "loss_sft": 0.18190539069473743, "loss_total_v6": 0.5904899686574936, "step": 4884 }, { "epoch": 1.1266162162162163, "grad_norm": 32.96486282348633, "learning_rate": 3.4453417825440993e-06, "loss": 9.245964050292969, "loss_alignment": 0.79150390625, "loss_alignment_w": 0.395751953125, "loss_sft": 0.18273112643510103, "loss_total_v6": 0.5778727307915688, "step": 4885 }, { "epoch": 1.1268468468468469, "grad_norm": 28.183801651000977, "learning_rate": 3.443427888786335e-06, "loss": 9.68472957611084, "loss_alignment": 0.8330078125, "loss_alignment_w": 0.41650390625, "loss_sft": 0.18847124464809895, "loss_total_v6": 0.6052955985069275, "step": 4886 }, { "epoch": 1.1270774774774774, "grad_norm": 29.116920471191406, "learning_rate": 3.441514247539265e-06, "loss": 9.794224739074707, "loss_alignment": 0.8115234375, "loss_alignment_w": 0.40576171875, "loss_sft": 0.2070487178862095, "loss_total_v6": 0.612139068543911, "step": 4887 }, { "epoch": 1.127308108108108, "grad_norm": 27.826839447021484, "learning_rate": 3.439600859113329e-06, "loss": 9.629936218261719, "loss_alignment": 0.796875, "loss_alignment_w": 0.3984375, "loss_sft": 0.20376918837428093, "loss_total_v6": 0.6018710061907768, "step": 4888 }, { "epoch": 1.1275387387387388, "grad_norm": 27.387710571289062, "learning_rate": 3.4376877238189184e-06, "loss": 9.430173873901367, "loss_alignment": 0.79296875, "loss_alignment_w": 0.396484375, "loss_sft": 0.19346603378653526, "loss_total_v6": 0.5893858298659325, "step": 4889 }, { "epoch": 1.1277693693693693, "grad_norm": 27.576845169067383, "learning_rate": 3.4357748419663857e-06, "loss": 9.683062553405762, "loss_alignment": 0.82666015625, "loss_alignment_w": 0.413330078125, "loss_sft": 0.1935703083872795, "loss_total_v6": 0.6051914021372795, "step": 4890 }, { "epoch": 1.1280000000000001, "grad_norm": 26.64919090270996, "learning_rate": 3.433862213866045e-06, "loss": 9.791566848754883, "loss_alignment": 0.810546875, "loss_alignment_w": 0.4052734375, "loss_sft": 0.20677581802010536, "loss_total_v6": 0.6119729429483414, "step": 4891 }, { "epoch": 1.1282306306306307, "grad_norm": 28.651533126831055, "learning_rate": 3.4319498398281638e-06, "loss": 9.396020889282227, "loss_alignment": 0.83203125, "loss_alignment_w": 0.416015625, "loss_sft": 0.17015234753489494, "loss_total_v6": 0.5872513651847839, "step": 4892 }, { "epoch": 1.1284612612612612, "grad_norm": 27.826461791992188, "learning_rate": 3.4300377201629753e-06, "loss": 9.293207168579102, "loss_alignment": 0.78564453125, "loss_alignment_w": 0.392822265625, "loss_sft": 0.18876607529819012, "loss_total_v6": 0.5808254182338715, "step": 4893 }, { "epoch": 1.1286918918918918, "grad_norm": 27.765771865844727, "learning_rate": 3.428125855180668e-06, "loss": 9.16567325592041, "loss_alignment": 0.79931640625, "loss_alignment_w": 0.399658203125, "loss_sft": 0.17260123509913683, "loss_total_v6": 0.5728545561432838, "step": 4894 }, { "epoch": 1.1289225225225226, "grad_norm": 28.43672752380371, "learning_rate": 3.4262142451913865e-06, "loss": 9.576030731201172, "loss_alignment": 0.796875, "loss_alignment_w": 0.3984375, "loss_sft": 0.20058324001729488, "loss_total_v6": 0.5985019579529762, "step": 4895 }, { "epoch": 1.1291531531531531, "grad_norm": 31.7490234375, "learning_rate": 3.4243028905052387e-06, "loss": 9.556640625, "loss_alignment": 0.822265625, "loss_alignment_w": 0.4111328125, "loss_sft": 0.18537897244095802, "loss_total_v6": 0.5972899943590164, "step": 4896 }, { "epoch": 1.129383783783784, "grad_norm": 32.03483200073242, "learning_rate": 3.4223917914322867e-06, "loss": 9.217803955078125, "loss_alignment": 0.80322265625, "loss_alignment_w": 0.401611328125, "loss_sft": 0.17346382606774569, "loss_total_v6": 0.5761127807199955, "step": 4897 }, { "epoch": 1.1296144144144145, "grad_norm": 26.77668571472168, "learning_rate": 3.4204809482825563e-06, "loss": 9.517146110534668, "loss_alignment": 0.7978515625, "loss_alignment_w": 0.39892578125, "loss_sft": 0.19684188440442085, "loss_total_v6": 0.5948216319084167, "step": 4898 }, { "epoch": 1.129845045045045, "grad_norm": 30.521921157836914, "learning_rate": 3.4185703613660264e-06, "loss": 9.272672653198242, "loss_alignment": 0.8115234375, "loss_alignment_w": 0.40576171875, "loss_sft": 0.17356670461595058, "loss_total_v6": 0.5795420408248901, "step": 4899 }, { "epoch": 1.1300756756756756, "grad_norm": 30.229862213134766, "learning_rate": 3.416660030992639e-06, "loss": 9.887114524841309, "loss_alignment": 0.85107421875, "loss_alignment_w": 0.425537109375, "loss_sft": 0.19306367449462414, "loss_total_v6": 0.6179446503520012, "step": 4900 }, { "epoch": 1.1303063063063064, "grad_norm": 27.962799072265625, "learning_rate": 3.41474995747229e-06, "loss": 9.767661094665527, "loss_alignment": 0.82763671875, "loss_alignment_w": 0.413818359375, "loss_sft": 0.19533294811844826, "loss_total_v6": 0.6104788258671761, "step": 4901 }, { "epoch": 1.130536936936937, "grad_norm": 32.445255279541016, "learning_rate": 3.4128401411148348e-06, "loss": 9.549875259399414, "loss_alignment": 0.81640625, "loss_alignment_w": 0.408203125, "loss_sft": 0.18881666474044323, "loss_total_v6": 0.5968671813607216, "step": 4902 }, { "epoch": 1.1307675675675677, "grad_norm": 31.931974411010742, "learning_rate": 3.410930582230093e-06, "loss": 9.631587982177734, "loss_alignment": 0.82275390625, "loss_alignment_w": 0.411376953125, "loss_sft": 0.191070344299078, "loss_total_v6": 0.6019742414355278, "step": 4903 }, { "epoch": 1.1309981981981982, "grad_norm": 31.105762481689453, "learning_rate": 3.409021281127835e-06, "loss": 9.740577697753906, "loss_alignment": 0.80615234375, "loss_alignment_w": 0.403076171875, "loss_sft": 0.20654914155602455, "loss_total_v6": 0.608786091208458, "step": 4904 }, { "epoch": 1.1312288288288288, "grad_norm": 28.93968963623047, "learning_rate": 3.4071122381177906e-06, "loss": 9.78325366973877, "loss_alignment": 0.8134765625, "loss_alignment_w": 0.40673828125, "loss_sft": 0.20433357171714306, "loss_total_v6": 0.6114533245563507, "step": 4905 }, { "epoch": 1.1314594594594594, "grad_norm": 26.00908660888672, "learning_rate": 3.40520345350965e-06, "loss": 9.614479064941406, "loss_alignment": 0.7900390625, "loss_alignment_w": 0.39501953125, "loss_sft": 0.2060990259051323, "loss_total_v6": 0.6009049415588379, "step": 4906 }, { "epoch": 1.1316900900900901, "grad_norm": 28.443256378173828, "learning_rate": 3.403294927613062e-06, "loss": 9.207509994506836, "loss_alignment": 0.796875, "loss_alignment_w": 0.3984375, "loss_sft": 0.1764825927093625, "loss_total_v6": 0.5754693858325481, "step": 4907 }, { "epoch": 1.1319207207207207, "grad_norm": 28.533878326416016, "learning_rate": 3.401386660737631e-06, "loss": 9.454294204711914, "loss_alignment": 0.7958984375, "loss_alignment_w": 0.39794921875, "loss_sft": 0.19390543084591627, "loss_total_v6": 0.5908933505415916, "step": 4908 }, { "epoch": 1.1321513513513515, "grad_norm": 27.260562896728516, "learning_rate": 3.3994786531929213e-06, "loss": 9.615652084350586, "loss_alignment": 0.80615234375, "loss_alignment_w": 0.403076171875, "loss_sft": 0.19781051576137543, "loss_total_v6": 0.6009782254695892, "step": 4909 }, { "epoch": 1.132381981981982, "grad_norm": 33.398704528808594, "learning_rate": 3.397570905288453e-06, "loss": 9.442667007446289, "loss_alignment": 0.79150390625, "loss_alignment_w": 0.395751953125, "loss_sft": 0.19442997686564922, "loss_total_v6": 0.5901666693389416, "step": 4910 }, { "epoch": 1.1326126126126126, "grad_norm": 28.68450355529785, "learning_rate": 3.3956634173337044e-06, "loss": 9.237335205078125, "loss_alignment": 0.8017578125, "loss_alignment_w": 0.40087890625, "loss_sft": 0.17680545058101416, "loss_total_v6": 0.5773333981633186, "step": 4911 }, { "epoch": 1.1328432432432431, "grad_norm": 29.714813232421875, "learning_rate": 3.393756189638115e-06, "loss": 9.446476936340332, "loss_alignment": 0.7861328125, "loss_alignment_w": 0.39306640625, "loss_sft": 0.19791820645332336, "loss_total_v6": 0.5904047787189484, "step": 4912 }, { "epoch": 1.133073873873874, "grad_norm": 27.075925827026367, "learning_rate": 3.391849222511079e-06, "loss": 9.477874755859375, "loss_alignment": 0.7998046875, "loss_alignment_w": 0.39990234375, "loss_sft": 0.1924953442066908, "loss_total_v6": 0.5923671498894691, "step": 4913 }, { "epoch": 1.1333045045045045, "grad_norm": 25.20987319946289, "learning_rate": 3.3899425162619483e-06, "loss": 9.346735000610352, "loss_alignment": 0.806640625, "loss_alignment_w": 0.4033203125, "loss_sft": 0.18004189431667328, "loss_total_v6": 0.5841709300875664, "step": 4914 }, { "epoch": 1.133535135135135, "grad_norm": 29.81510353088379, "learning_rate": 3.3880360712000337e-06, "loss": 9.433374404907227, "loss_alignment": 0.79052734375, "loss_alignment_w": 0.395263671875, "loss_sft": 0.19351350236684084, "loss_total_v6": 0.5895858630537987, "step": 4915 }, { "epoch": 1.1337657657657658, "grad_norm": 30.61638641357422, "learning_rate": 3.3861298876346015e-06, "loss": 9.33344841003418, "loss_alignment": 0.80615234375, "loss_alignment_w": 0.403076171875, "loss_sft": 0.17910461500287056, "loss_total_v6": 0.5833404809236526, "step": 4916 }, { "epoch": 1.1339963963963964, "grad_norm": 37.39975357055664, "learning_rate": 3.3842239658748806e-06, "loss": 9.526165962219238, "loss_alignment": 0.8193359375, "loss_alignment_w": 0.40966796875, "loss_sft": 0.18558004312217236, "loss_total_v6": 0.59538534283638, "step": 4917 }, { "epoch": 1.134227027027027, "grad_norm": 36.38225173950195, "learning_rate": 3.3823183062300502e-06, "loss": 9.567995071411133, "loss_alignment": 0.763671875, "loss_alignment_w": 0.3818359375, "loss_sft": 0.21570595353841782, "loss_total_v6": 0.597999669611454, "step": 4918 }, { "epoch": 1.1344576576576577, "grad_norm": 28.87718963623047, "learning_rate": 3.3804129090092542e-06, "loss": 9.51201343536377, "loss_alignment": 0.82470703125, "loss_alignment_w": 0.412353515625, "loss_sft": 0.18287975620478392, "loss_total_v6": 0.594500832259655, "step": 4919 }, { "epoch": 1.1346882882882883, "grad_norm": 30.401287078857422, "learning_rate": 3.378507774521587e-06, "loss": 9.577653884887695, "loss_alignment": 0.8369140625, "loss_alignment_w": 0.41845703125, "loss_sft": 0.1793681737035513, "loss_total_v6": 0.5986034125089645, "step": 4920 }, { "epoch": 1.1349189189189188, "grad_norm": 31.20085334777832, "learning_rate": 3.376602903076105e-06, "loss": 9.793190002441406, "loss_alignment": 0.85595703125, "loss_alignment_w": 0.427978515625, "loss_sft": 0.1843400038778782, "loss_total_v6": 0.6120743826031685, "step": 4921 }, { "epoch": 1.1351495495495496, "grad_norm": 28.167266845703125, "learning_rate": 3.3746982949818243e-06, "loss": 9.3889799118042, "loss_alignment": 0.80859375, "loss_alignment_w": 0.404296875, "loss_sft": 0.1819803575053811, "loss_total_v6": 0.586811289191246, "step": 4922 }, { "epoch": 1.1353801801801802, "grad_norm": 27.335203170776367, "learning_rate": 3.372793950547711e-06, "loss": 9.300104141235352, "loss_alignment": 0.8095703125, "loss_alignment_w": 0.40478515625, "loss_sft": 0.17547948472201824, "loss_total_v6": 0.5812564864754677, "step": 4923 }, { "epoch": 1.1356108108108107, "grad_norm": 31.166658401489258, "learning_rate": 3.3708898700826925e-06, "loss": 9.275211334228516, "loss_alignment": 0.78515625, "loss_alignment_w": 0.392578125, "loss_sft": 0.18718360736966133, "loss_total_v6": 0.5797006785869598, "step": 4924 }, { "epoch": 1.1358414414414415, "grad_norm": 41.04751205444336, "learning_rate": 3.3689860538956547e-06, "loss": 9.771232604980469, "loss_alignment": 0.8408203125, "loss_alignment_w": 0.42041015625, "loss_sft": 0.1895594261586666, "loss_total_v6": 0.6107020005583763, "step": 4925 }, { "epoch": 1.136072072072072, "grad_norm": 30.410110473632812, "learning_rate": 3.3670825022954364e-06, "loss": 9.571256637573242, "loss_alignment": 0.81103515625, "loss_alignment_w": 0.405517578125, "loss_sft": 0.19213665835559368, "loss_total_v6": 0.5982035398483276, "step": 4926 }, { "epoch": 1.1363027027027026, "grad_norm": 26.05481719970703, "learning_rate": 3.365179215590838e-06, "loss": 9.516974449157715, "loss_alignment": 0.81640625, "loss_alignment_w": 0.408203125, "loss_sft": 0.18689768388867378, "loss_total_v6": 0.5948108956217766, "step": 4927 }, { "epoch": 1.1365333333333334, "grad_norm": 30.279661178588867, "learning_rate": 3.3632761940906167e-06, "loss": 10.213542938232422, "loss_alignment": 0.84228515625, "loss_alignment_w": 0.421142578125, "loss_sft": 0.2167766373604536, "loss_total_v6": 0.6383464634418488, "step": 4928 }, { "epoch": 1.136763963963964, "grad_norm": 32.10470962524414, "learning_rate": 3.3613734381034834e-06, "loss": 10.33787727355957, "loss_alignment": 0.84521484375, "loss_alignment_w": 0.422607421875, "loss_sft": 0.2229605708271265, "loss_total_v6": 0.6461173295974731, "step": 4929 }, { "epoch": 1.1369945945945945, "grad_norm": 28.732637405395508, "learning_rate": 3.359470947938106e-06, "loss": 9.369222640991211, "loss_alignment": 0.81201171875, "loss_alignment_w": 0.406005859375, "loss_sft": 0.1803487315773964, "loss_total_v6": 0.5855763852596283, "step": 4930 }, { "epoch": 1.1372252252252253, "grad_norm": 30.739028930664062, "learning_rate": 3.3575687239031108e-06, "loss": 9.495489120483398, "loss_alignment": 0.794921875, "loss_alignment_w": 0.3974609375, "loss_sft": 0.19664799980819225, "loss_total_v6": 0.593468077480793, "step": 4931 }, { "epoch": 1.1374558558558558, "grad_norm": 29.18057632446289, "learning_rate": 3.355666766307084e-06, "loss": 9.73359489440918, "loss_alignment": 0.82080078125, "loss_alignment_w": 0.410400390625, "loss_sft": 0.19782731775194407, "loss_total_v6": 0.6083497479557991, "step": 4932 }, { "epoch": 1.1376864864864864, "grad_norm": 27.12116050720215, "learning_rate": 3.3537650754585638e-06, "loss": 9.131092071533203, "loss_alignment": 0.77783203125, "loss_alignment_w": 0.388916015625, "loss_sft": 0.18153304792940617, "loss_total_v6": 0.5706932172179222, "step": 4933 }, { "epoch": 1.1379171171171172, "grad_norm": 32.34579849243164, "learning_rate": 3.351863651666047e-06, "loss": 9.424850463867188, "loss_alignment": 0.837890625, "loss_alignment_w": 0.4189453125, "loss_sft": 0.17067237850278616, "loss_total_v6": 0.5890531167387962, "step": 4934 }, { "epoch": 1.1381477477477477, "grad_norm": 30.81993293762207, "learning_rate": 3.349962495237987e-06, "loss": 9.33837890625, "loss_alignment": 0.81689453125, "loss_alignment_w": 0.408447265625, "loss_sft": 0.1748199136927724, "loss_total_v6": 0.5836486592888832, "step": 4935 }, { "epoch": 1.1383783783783783, "grad_norm": 37.365699768066406, "learning_rate": 3.3480616064827914e-06, "loss": 9.49451732635498, "loss_alignment": 0.791015625, "loss_alignment_w": 0.3955078125, "loss_sft": 0.19791480340063572, "loss_total_v6": 0.5934073552489281, "step": 4936 }, { "epoch": 1.138609009009009, "grad_norm": 30.1567440032959, "learning_rate": 3.3461609857088303e-06, "loss": 9.476242065429688, "loss_alignment": 0.80078125, "loss_alignment_w": 0.400390625, "loss_sft": 0.1908979369327426, "loss_total_v6": 0.5922651141881943, "step": 4937 }, { "epoch": 1.1388396396396396, "grad_norm": 32.780086517333984, "learning_rate": 3.3442606332244258e-06, "loss": 9.745235443115234, "loss_alignment": 0.8251953125, "loss_alignment_w": 0.41259765625, "loss_sft": 0.19716615788638592, "loss_total_v6": 0.609077200293541, "step": 4938 }, { "epoch": 1.1390702702702702, "grad_norm": 26.26007652282715, "learning_rate": 3.3423605493378564e-06, "loss": 9.454023361206055, "loss_alignment": 0.8134765625, "loss_alignment_w": 0.40673828125, "loss_sft": 0.1847790740430355, "loss_total_v6": 0.5908764749765396, "step": 4939 }, { "epoch": 1.139300900900901, "grad_norm": 27.604206085205078, "learning_rate": 3.340460734357359e-06, "loss": 9.44205379486084, "loss_alignment": 0.8251953125, "loss_alignment_w": 0.41259765625, "loss_sft": 0.1775459460914135, "loss_total_v6": 0.5901283621788025, "step": 4940 }, { "epoch": 1.1395315315315315, "grad_norm": 29.052724838256836, "learning_rate": 3.338561188591124e-06, "loss": 9.730962753295898, "loss_alignment": 0.814453125, "loss_alignment_w": 0.4072265625, "loss_sft": 0.20089759305119514, "loss_total_v6": 0.6081851795315742, "step": 4941 }, { "epoch": 1.139762162162162, "grad_norm": 29.709850311279297, "learning_rate": 3.336661912347303e-06, "loss": 9.07283878326416, "loss_alignment": 0.78076171875, "loss_alignment_w": 0.390380859375, "loss_sft": 0.17609170265495777, "loss_total_v6": 0.5670524090528488, "step": 4942 }, { "epoch": 1.1399927927927929, "grad_norm": 27.799331665039062, "learning_rate": 3.3347629059339993e-06, "loss": 9.402946472167969, "loss_alignment": 0.80126953125, "loss_alignment_w": 0.400634765625, "loss_sft": 0.18691209889948368, "loss_total_v6": 0.5876841843128204, "step": 4943 }, { "epoch": 1.1402234234234234, "grad_norm": 26.027414321899414, "learning_rate": 3.3328641696592756e-06, "loss": 9.438716888427734, "loss_alignment": 0.81005859375, "loss_alignment_w": 0.405029296875, "loss_sft": 0.18545505590736866, "loss_total_v6": 0.5899197608232498, "step": 4944 }, { "epoch": 1.140454054054054, "grad_norm": 27.376501083374023, "learning_rate": 3.330965703831146e-06, "loss": 9.68248176574707, "loss_alignment": 0.81103515625, "loss_alignment_w": 0.405517578125, "loss_sft": 0.1984473317861557, "loss_total_v6": 0.6051550880074501, "step": 4945 }, { "epoch": 1.1406846846846848, "grad_norm": 28.91275405883789, "learning_rate": 3.3290675087575857e-06, "loss": 9.141587257385254, "loss_alignment": 0.79638671875, "loss_alignment_w": 0.398193359375, "loss_sft": 0.17166046053171158, "loss_total_v6": 0.5713491812348366, "step": 4946 }, { "epoch": 1.1409153153153153, "grad_norm": 30.18077278137207, "learning_rate": 3.3271695847465264e-06, "loss": 9.558839797973633, "loss_alignment": 0.8173828125, "loss_alignment_w": 0.40869140625, "loss_sft": 0.1887513753026724, "loss_total_v6": 0.5974275097250938, "step": 4947 }, { "epoch": 1.1411459459459459, "grad_norm": 30.570209503173828, "learning_rate": 3.325271932105851e-06, "loss": 9.554713249206543, "loss_alignment": 0.77783203125, "loss_alignment_w": 0.388916015625, "loss_sft": 0.20784159377217293, "loss_total_v6": 0.5971695855259895, "step": 4948 }, { "epoch": 1.1413765765765767, "grad_norm": 29.877206802368164, "learning_rate": 3.323374551143402e-06, "loss": 9.799365043640137, "loss_alignment": 0.7978515625, "loss_alignment_w": 0.39892578125, "loss_sft": 0.2136413287371397, "loss_total_v6": 0.6124603003263474, "step": 4949 }, { "epoch": 1.1416072072072072, "grad_norm": 29.78211784362793, "learning_rate": 3.3214774421669777e-06, "loss": 9.074298858642578, "loss_alignment": 0.7861328125, "loss_alignment_w": 0.39306640625, "loss_sft": 0.1739399628713727, "loss_total_v6": 0.5671437010169029, "step": 4950 }, { "epoch": 1.1418378378378378, "grad_norm": 26.35651206970215, "learning_rate": 3.319580605484328e-06, "loss": 9.766773223876953, "loss_alignment": 0.78515625, "loss_alignment_w": 0.392578125, "loss_sft": 0.2162582352757454, "loss_total_v6": 0.6104232892394066, "step": 4951 }, { "epoch": 1.1420684684684685, "grad_norm": 45.03878402709961, "learning_rate": 3.3176840414031653e-06, "loss": 9.908224105834961, "loss_alignment": 0.82421875, "loss_alignment_w": 0.412109375, "loss_sft": 0.2067731749266386, "loss_total_v6": 0.6192640066146851, "step": 4952 }, { "epoch": 1.142299099099099, "grad_norm": 31.35543441772461, "learning_rate": 3.315787750231156e-06, "loss": 9.567042350769043, "loss_alignment": 0.78759765625, "loss_alignment_w": 0.393798828125, "loss_sft": 0.20383615419268608, "loss_total_v6": 0.5979401767253876, "step": 4953 }, { "epoch": 1.1425297297297297, "grad_norm": 29.836946487426758, "learning_rate": 3.313891732275918e-06, "loss": 9.348766326904297, "loss_alignment": 0.7998046875, "loss_alignment_w": 0.39990234375, "loss_sft": 0.18410558439791203, "loss_total_v6": 0.5842978432774544, "step": 4954 }, { "epoch": 1.1427603603603604, "grad_norm": 28.59748077392578, "learning_rate": 3.3119959878450257e-06, "loss": 9.47777271270752, "loss_alignment": 0.81787109375, "loss_alignment_w": 0.408935546875, "loss_sft": 0.18351676128804684, "loss_total_v6": 0.5923607647418976, "step": 4955 }, { "epoch": 1.142990990990991, "grad_norm": 28.01904296875, "learning_rate": 3.310100517246016e-06, "loss": 9.430196762084961, "loss_alignment": 0.78662109375, "loss_alignment_w": 0.393310546875, "loss_sft": 0.1968854833394289, "loss_total_v6": 0.5893873162567616, "step": 4956 }, { "epoch": 1.1432216216216216, "grad_norm": 30.60626220703125, "learning_rate": 3.3082053207863746e-06, "loss": 9.764793395996094, "loss_alignment": 0.83642578125, "loss_alignment_w": 0.418212890625, "loss_sft": 0.19188835751265287, "loss_total_v6": 0.61029963940382, "step": 4957 }, { "epoch": 1.1434522522522523, "grad_norm": 26.46676254272461, "learning_rate": 3.3063103987735433e-06, "loss": 9.707416534423828, "loss_alignment": 0.8271484375, "loss_alignment_w": 0.41357421875, "loss_sft": 0.19266626238822937, "loss_total_v6": 0.6067135185003281, "step": 4958 }, { "epoch": 1.143682882882883, "grad_norm": 31.289384841918945, "learning_rate": 3.3044157515149233e-06, "loss": 9.738983154296875, "loss_alignment": 0.81103515625, "loss_alignment_w": 0.405517578125, "loss_sft": 0.20271103084087372, "loss_total_v6": 0.6086863800883293, "step": 4959 }, { "epoch": 1.1439135135135134, "grad_norm": 31.444480895996094, "learning_rate": 3.3025213793178647e-06, "loss": 9.729907989501953, "loss_alignment": 0.80859375, "loss_alignment_w": 0.404296875, "loss_sft": 0.20403606817126274, "loss_total_v6": 0.6081192940473557, "step": 4960 }, { "epoch": 1.1441441441441442, "grad_norm": 30.33631706237793, "learning_rate": 3.30062728248968e-06, "loss": 9.334516525268555, "loss_alignment": 0.81298828125, "loss_alignment_w": 0.406494140625, "loss_sft": 0.17579927295446396, "loss_total_v6": 0.5834072902798653, "step": 4961 }, { "epoch": 1.1443747747747748, "grad_norm": 27.307416915893555, "learning_rate": 3.2987334613376333e-06, "loss": 9.311739921569824, "loss_alignment": 0.7861328125, "loss_alignment_w": 0.39306640625, "loss_sft": 0.1887952722609043, "loss_total_v6": 0.5819837376475334, "step": 4962 }, { "epoch": 1.1446054054054053, "grad_norm": 29.605684280395508, "learning_rate": 3.2968399161689457e-06, "loss": 9.924732208251953, "loss_alignment": 0.8310546875, "loss_alignment_w": 0.41552734375, "loss_sft": 0.20560763776302338, "loss_total_v6": 0.6202957481145859, "step": 4963 }, { "epoch": 1.1448360360360361, "grad_norm": 27.41769790649414, "learning_rate": 3.2949466472907897e-06, "loss": 9.731654167175293, "loss_alignment": 0.8154296875, "loss_alignment_w": 0.40771484375, "loss_sft": 0.2005898281931877, "loss_total_v6": 0.6082283779978752, "step": 4964 }, { "epoch": 1.1450666666666667, "grad_norm": 28.05479621887207, "learning_rate": 3.293053655010296e-06, "loss": 9.786982536315918, "loss_alignment": 0.82763671875, "loss_alignment_w": 0.413818359375, "loss_sft": 0.19852418266236782, "loss_total_v6": 0.6116864010691643, "step": 4965 }, { "epoch": 1.1452972972972972, "grad_norm": 27.88741683959961, "learning_rate": 3.2911609396345533e-06, "loss": 9.537056922912598, "loss_alignment": 0.8173828125, "loss_alignment_w": 0.40869140625, "loss_sft": 0.18706952966749668, "loss_total_v6": 0.5960660800337791, "step": 4966 }, { "epoch": 1.145527927927928, "grad_norm": 25.648550033569336, "learning_rate": 3.2892685014706005e-06, "loss": 9.799785614013672, "loss_alignment": 0.810546875, "loss_alignment_w": 0.4052734375, "loss_sft": 0.20594669040292501, "loss_total_v6": 0.6124865747988224, "step": 4967 }, { "epoch": 1.1457585585585586, "grad_norm": 31.025983810424805, "learning_rate": 3.287376340825432e-06, "loss": 10.140331268310547, "loss_alignment": 0.8349609375, "loss_alignment_w": 0.41748046875, "loss_sft": 0.21636656112968922, "loss_total_v6": 0.6337707191705704, "step": 4968 }, { "epoch": 1.1459891891891891, "grad_norm": 30.20549201965332, "learning_rate": 3.2854844580060007e-06, "loss": 9.079627990722656, "loss_alignment": 0.7998046875, "loss_alignment_w": 0.39990234375, "loss_sft": 0.16769647505134344, "loss_total_v6": 0.5674767382442951, "step": 4969 }, { "epoch": 1.14621981981982, "grad_norm": 29.831748962402344, "learning_rate": 3.2835928533192086e-06, "loss": 9.455673217773438, "loss_alignment": 0.7880859375, "loss_alignment_w": 0.39404296875, "loss_sft": 0.19733336754143238, "loss_total_v6": 0.5909796133637428, "step": 4970 }, { "epoch": 1.1464504504504505, "grad_norm": 29.01312255859375, "learning_rate": 3.28170152707192e-06, "loss": 8.753011703491211, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.1633199006319046, "loss_total_v6": 0.5470632053911686, "step": 4971 }, { "epoch": 1.146681081081081, "grad_norm": 34.599979400634766, "learning_rate": 3.2798104795709484e-06, "loss": 9.669175148010254, "loss_alignment": 0.79443359375, "loss_alignment_w": 0.397216796875, "loss_sft": 0.20701506175100803, "loss_total_v6": 0.6043234318494797, "step": 4972 }, { "epoch": 1.1469117117117118, "grad_norm": 28.41455078125, "learning_rate": 3.2779197111230654e-06, "loss": 9.538289070129395, "loss_alignment": 0.80810546875, "loss_alignment_w": 0.404052734375, "loss_sft": 0.19138842169195414, "loss_total_v6": 0.5961430668830872, "step": 4973 }, { "epoch": 1.1471423423423424, "grad_norm": 32.05223846435547, "learning_rate": 3.276029222034993e-06, "loss": 9.652727127075195, "loss_alignment": 0.80908203125, "loss_alignment_w": 0.404541015625, "loss_sft": 0.1981288269162178, "loss_total_v6": 0.6032954379916191, "step": 4974 }, { "epoch": 1.147372972972973, "grad_norm": 29.924741744995117, "learning_rate": 3.274139012613411e-06, "loss": 9.382057189941406, "loss_alignment": 0.78466796875, "loss_alignment_w": 0.392333984375, "loss_sft": 0.19352575950324535, "loss_total_v6": 0.5863785445690155, "step": 4975 }, { "epoch": 1.1476036036036037, "grad_norm": 30.931987762451172, "learning_rate": 3.2722490831649568e-06, "loss": 9.831356048583984, "loss_alignment": 0.85693359375, "loss_alignment_w": 0.428466796875, "loss_sft": 0.1848790915682912, "loss_total_v6": 0.6144597753882408, "step": 4976 }, { "epoch": 1.1478342342342343, "grad_norm": 29.007232666015625, "learning_rate": 3.2703594339962164e-06, "loss": 9.52086067199707, "loss_alignment": 0.79736328125, "loss_alignment_w": 0.398681640625, "loss_sft": 0.19699776358902454, "loss_total_v6": 0.5950537994503975, "step": 4977 }, { "epoch": 1.1480648648648648, "grad_norm": 31.9674129486084, "learning_rate": 3.268470065413734e-06, "loss": 9.616218566894531, "loss_alignment": 0.79150390625, "loss_alignment_w": 0.395751953125, "loss_sft": 0.20578050799667835, "loss_total_v6": 0.6010137051343918, "step": 4978 }, { "epoch": 1.1482954954954956, "grad_norm": 27.504411697387695, "learning_rate": 3.266580977724007e-06, "loss": 9.617827415466309, "loss_alignment": 0.8037109375, "loss_alignment_w": 0.40185546875, "loss_sft": 0.19910616055130959, "loss_total_v6": 0.6011142209172249, "step": 4979 }, { "epoch": 1.1485261261261261, "grad_norm": 30.25712013244629, "learning_rate": 3.2646921712334854e-06, "loss": 9.723771095275879, "loss_alignment": 0.791015625, "loss_alignment_w": 0.3955078125, "loss_sft": 0.21350961923599243, "loss_total_v6": 0.6077356934547424, "step": 4980 }, { "epoch": 1.1487567567567567, "grad_norm": 26.428550720214844, "learning_rate": 3.262803646248579e-06, "loss": 9.50796127319336, "loss_alignment": 0.81640625, "loss_alignment_w": 0.408203125, "loss_sft": 0.1859376672655344, "loss_total_v6": 0.5942476093769073, "step": 4981 }, { "epoch": 1.1489873873873875, "grad_norm": 32.95578384399414, "learning_rate": 3.2609154030756486e-06, "loss": 9.400617599487305, "loss_alignment": 0.80859375, "loss_alignment_w": 0.404296875, "loss_sft": 0.18414199352264404, "loss_total_v6": 0.5875385999679565, "step": 4982 }, { "epoch": 1.149218018018018, "grad_norm": 27.528371810913086, "learning_rate": 3.2590274420210067e-06, "loss": 9.774184226989746, "loss_alignment": 0.80126953125, "loss_alignment_w": 0.400634765625, "loss_sft": 0.20793243497610092, "loss_total_v6": 0.6108864992856979, "step": 4983 }, { "epoch": 1.1494486486486486, "grad_norm": 27.388097763061523, "learning_rate": 3.2571397633909252e-06, "loss": 10.2554931640625, "loss_alignment": 0.8056640625, "loss_alignment_w": 0.40283203125, "loss_sft": 0.23824309185147285, "loss_total_v6": 0.6409683153033257, "step": 4984 }, { "epoch": 1.1496792792792794, "grad_norm": 33.73674392700195, "learning_rate": 3.255252367491625e-06, "loss": 9.774227142333984, "loss_alignment": 0.806640625, "loss_alignment_w": 0.4033203125, "loss_sft": 0.2068364303559065, "loss_total_v6": 0.6108891740441322, "step": 4985 }, { "epoch": 1.14990990990991, "grad_norm": 31.124788284301758, "learning_rate": 3.2533652546292866e-06, "loss": 9.555137634277344, "loss_alignment": 0.80224609375, "loss_alignment_w": 0.401123046875, "loss_sft": 0.1961646294221282, "loss_total_v6": 0.5971961095929146, "step": 4986 }, { "epoch": 1.1501405405405405, "grad_norm": 25.380659103393555, "learning_rate": 3.2514784251100394e-06, "loss": 9.5333251953125, "loss_alignment": 0.7939453125, "loss_alignment_w": 0.39697265625, "loss_sft": 0.1977920699864626, "loss_total_v6": 0.5958328098058701, "step": 4987 }, { "epoch": 1.1503711711711713, "grad_norm": 26.8619327545166, "learning_rate": 3.249591879239972e-06, "loss": 9.510246276855469, "loss_alignment": 0.796875, "loss_alignment_w": 0.3984375, "loss_sft": 0.19659377075731754, "loss_total_v6": 0.5943904221057892, "step": 4988 }, { "epoch": 1.1506018018018018, "grad_norm": 35.409690856933594, "learning_rate": 3.247705617325122e-06, "loss": 9.557913780212402, "loss_alignment": 0.80078125, "loss_alignment_w": 0.400390625, "loss_sft": 0.1978639978915453, "loss_total_v6": 0.5973696112632751, "step": 4989 }, { "epoch": 1.1508324324324324, "grad_norm": 30.448091506958008, "learning_rate": 3.2458196396714812e-06, "loss": 9.850072860717773, "loss_alignment": 0.818359375, "loss_alignment_w": 0.4091796875, "loss_sft": 0.20500025898218155, "loss_total_v6": 0.615629531443119, "step": 4990 }, { "epoch": 1.151063063063063, "grad_norm": 25.713054656982422, "learning_rate": 3.243933946585003e-06, "loss": 9.296656608581543, "loss_alignment": 0.7998046875, "loss_alignment_w": 0.39990234375, "loss_sft": 0.18106239009648561, "loss_total_v6": 0.581041045486927, "step": 4991 }, { "epoch": 1.1512936936936937, "grad_norm": 27.298728942871094, "learning_rate": 3.242048538371585e-06, "loss": 9.12039566040039, "loss_alignment": 0.7978515625, "loss_alignment_w": 0.39892578125, "loss_sft": 0.17016814462840557, "loss_total_v6": 0.5700247287750244, "step": 4992 }, { "epoch": 1.1515243243243243, "grad_norm": 28.86494255065918, "learning_rate": 3.240163415337082e-06, "loss": 9.560131072998047, "loss_alignment": 0.7734375, "loss_alignment_w": 0.38671875, "loss_sft": 0.21033164486289024, "loss_total_v6": 0.5975081846117973, "step": 4993 }, { "epoch": 1.151754954954955, "grad_norm": 27.534866333007812, "learning_rate": 3.238278577787305e-06, "loss": 9.133905410766602, "loss_alignment": 0.7724609375, "loss_alignment_w": 0.38623046875, "loss_sft": 0.18421132303774357, "loss_total_v6": 0.5708690211176872, "step": 4994 }, { "epoch": 1.1519855855855856, "grad_norm": 31.694477081298828, "learning_rate": 3.2363940260280137e-06, "loss": 9.829216003417969, "loss_alignment": 0.84716796875, "loss_alignment_w": 0.423583984375, "loss_sft": 0.19126081839203835, "loss_total_v6": 0.6143259927630424, "step": 4995 }, { "epoch": 1.1522162162162162, "grad_norm": 27.551721572875977, "learning_rate": 3.2345097603649267e-06, "loss": 9.44390869140625, "loss_alignment": 0.79248046875, "loss_alignment_w": 0.396240234375, "loss_sft": 0.1933936644345522, "loss_total_v6": 0.5902442336082458, "step": 4996 }, { "epoch": 1.1524468468468467, "grad_norm": 29.02454948425293, "learning_rate": 3.2326257811037154e-06, "loss": 9.378899574279785, "loss_alignment": 0.8017578125, "loss_alignment_w": 0.40087890625, "loss_sft": 0.18620260432362556, "loss_total_v6": 0.5861812382936478, "step": 4997 }, { "epoch": 1.1526774774774775, "grad_norm": 24.795082092285156, "learning_rate": 3.2307420885500006e-06, "loss": 9.432886123657227, "loss_alignment": 0.810546875, "loss_alignment_w": 0.4052734375, "loss_sft": 0.18440406396985054, "loss_total_v6": 0.5895554199814796, "step": 4998 }, { "epoch": 1.152908108108108, "grad_norm": 29.221359252929688, "learning_rate": 3.228858683009359e-06, "loss": 9.404085159301758, "loss_alignment": 0.81982421875, "loss_alignment_w": 0.409912109375, "loss_sft": 0.1779347751289606, "loss_total_v6": 0.5877553299069405, "step": 4999 }, { "epoch": 1.1531387387387388, "grad_norm": 28.917409896850586, "learning_rate": 3.226975564787322e-06, "loss": 9.581413269042969, "loss_alignment": 0.8046875, "loss_alignment_w": 0.40234375, "loss_sft": 0.1954722236841917, "loss_total_v6": 0.5988383032381535, "step": 5000 }, { "epoch": 1.1531387387387388, "eval_loss": 0.8553000092506409, "eval_loss_alignment": 0.8308205800513698, "eval_loss_alignment_w": 0.4154102900256849, "eval_loss_contrast": 0.0, "eval_loss_highlevel": 0.0, "eval_loss_sft": 0.43988966367730536, "eval_loss_total_v6": 0.8552999567903884, "eval_loss_wm_recon": 0.0, "eval_loss_wm_recon_w": 0.0, "eval_runtime": 298.0475, "eval_samples_per_second": 5.855, "eval_steps_per_second": 0.735, "step": 5000 }, { "epoch": 1.1533693693693694, "grad_norm": 27.543149948120117, "learning_rate": 3.2250927341893745e-06, "loss": 9.436673164367676, "loss_alignment": 0.81005859375, "loss_alignment_w": 0.405029296875, "loss_sft": 0.18473223969340324, "loss_total_v6": 0.5897920578718185, "step": 5001 }, { "epoch": 1.1536, "grad_norm": 29.909217834472656, "learning_rate": 3.2232101915209512e-06, "loss": 9.841812133789062, "loss_alignment": 0.8369140625, "loss_alignment_w": 0.41845703125, "loss_sft": 0.1960154166445136, "loss_total_v6": 0.6151133179664612, "step": 5002 }, { "epoch": 1.1538306306306305, "grad_norm": 29.98517417907715, "learning_rate": 3.221327937087445e-06, "loss": 9.409037590026855, "loss_alignment": 0.80078125, "loss_alignment_w": 0.400390625, "loss_sft": 0.18802517838776112, "loss_total_v6": 0.5880648270249367, "step": 5003 }, { "epoch": 1.1540612612612613, "grad_norm": 29.641685485839844, "learning_rate": 3.2194459711941967e-06, "loss": 9.12525749206543, "loss_alignment": 0.80029296875, "loss_alignment_w": 0.400146484375, "loss_sft": 0.17039570584893227, "loss_total_v6": 0.5703285709023476, "step": 5004 }, { "epoch": 1.1542918918918919, "grad_norm": 27.116464614868164, "learning_rate": 3.2175642941465056e-06, "loss": 9.379368782043457, "loss_alignment": 0.78271484375, "loss_alignment_w": 0.391357421875, "loss_sft": 0.1949752103537321, "loss_total_v6": 0.5862105637788773, "step": 5005 }, { "epoch": 1.1545225225225226, "grad_norm": 30.467449188232422, "learning_rate": 3.215682906249621e-06, "loss": 9.515073776245117, "loss_alignment": 0.814453125, "loss_alignment_w": 0.4072265625, "loss_sft": 0.18749605678021908, "loss_total_v6": 0.5946920961141586, "step": 5006 }, { "epoch": 1.1547531531531532, "grad_norm": 24.54671859741211, "learning_rate": 3.2138018078087457e-06, "loss": 9.188640594482422, "loss_alignment": 0.81689453125, "loss_alignment_w": 0.408447265625, "loss_sft": 0.16662100329995155, "loss_total_v6": 0.5742900520563126, "step": 5007 }, { "epoch": 1.1549837837837837, "grad_norm": 27.88858413696289, "learning_rate": 3.2119209991290346e-06, "loss": 9.30660629272461, "loss_alignment": 0.80712890625, "loss_alignment_w": 0.403564453125, "loss_sft": 0.17831204924732447, "loss_total_v6": 0.5816628709435463, "step": 5008 }, { "epoch": 1.1552144144144143, "grad_norm": 27.611080169677734, "learning_rate": 3.2100404805155975e-06, "loss": 9.22713565826416, "loss_alignment": 0.806640625, "loss_alignment_w": 0.4033203125, "loss_sft": 0.17426070384681225, "loss_total_v6": 0.5766959935426712, "step": 5009 }, { "epoch": 1.155445045045045, "grad_norm": 27.24020767211914, "learning_rate": 3.2081602522734987e-06, "loss": 9.087013244628906, "loss_alignment": 0.78125, "loss_alignment_w": 0.390625, "loss_sft": 0.17600112967193127, "loss_total_v6": 0.5679383724927902, "step": 5010 }, { "epoch": 1.1556756756756756, "grad_norm": 28.521665573120117, "learning_rate": 3.2062803147077516e-06, "loss": 9.136664390563965, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.18687100522220135, "loss_total_v6": 0.5710415169596672, "step": 5011 }, { "epoch": 1.1559063063063064, "grad_norm": 26.272539138793945, "learning_rate": 3.2044006681233226e-06, "loss": 9.593300819396973, "loss_alignment": 0.81982421875, "loss_alignment_w": 0.409912109375, "loss_sft": 0.18934875912964344, "loss_total_v6": 0.5995813086628914, "step": 5012 }, { "epoch": 1.156136936936937, "grad_norm": 26.020694732666016, "learning_rate": 3.2025213128251342e-06, "loss": 9.739313125610352, "loss_alignment": 0.8310546875, "loss_alignment_w": 0.41552734375, "loss_sft": 0.1935002040117979, "loss_total_v6": 0.6087071150541306, "step": 5013 }, { "epoch": 1.1563675675675675, "grad_norm": 32.385894775390625, "learning_rate": 3.2006422491180567e-06, "loss": 9.188813209533691, "loss_alignment": 0.7978515625, "loss_alignment_w": 0.39892578125, "loss_sft": 0.17540554516017437, "loss_total_v6": 0.5743008106946945, "step": 5014 }, { "epoch": 1.156598198198198, "grad_norm": 28.670995712280273, "learning_rate": 3.19876347730692e-06, "loss": 9.62143611907959, "loss_alignment": 0.77880859375, "loss_alignment_w": 0.389404296875, "loss_sft": 0.21283571794629097, "loss_total_v6": 0.6013397350907326, "step": 5015 }, { "epoch": 1.1568288288288289, "grad_norm": 27.65319061279297, "learning_rate": 3.1968849976965016e-06, "loss": 9.5842866897583, "loss_alignment": 0.82177734375, "loss_alignment_w": 0.410888671875, "loss_sft": 0.18834285158663988, "loss_total_v6": 0.599017895758152, "step": 5016 }, { "epoch": 1.1570594594594594, "grad_norm": 29.047565460205078, "learning_rate": 3.1950068105915312e-06, "loss": 9.479496002197266, "loss_alignment": 0.84619140625, "loss_alignment_w": 0.423095703125, "loss_sft": 0.1696169190108776, "loss_total_v6": 0.5924684777855873, "step": 5017 }, { "epoch": 1.1572900900900902, "grad_norm": 32.57596969604492, "learning_rate": 3.193128916296695e-06, "loss": 9.984169006347656, "loss_alignment": 0.8125, "loss_alignment_w": 0.40625, "loss_sft": 0.21695180237293243, "loss_total_v6": 0.6240105256438255, "step": 5018 }, { "epoch": 1.1575207207207208, "grad_norm": 25.857738494873047, "learning_rate": 3.191251315116626e-06, "loss": 9.110652923583984, "loss_alignment": 0.7890625, "loss_alignment_w": 0.39453125, "loss_sft": 0.17384688928723335, "loss_total_v6": 0.569415770471096, "step": 5019 }, { "epoch": 1.1577513513513513, "grad_norm": 28.859895706176758, "learning_rate": 3.189374007355917e-06, "loss": 9.459970474243164, "loss_alignment": 0.82373046875, "loss_alignment_w": 0.411865234375, "loss_sft": 0.1804968323558569, "loss_total_v6": 0.5912481844425201, "step": 5020 }, { "epoch": 1.1579819819819819, "grad_norm": 28.050874710083008, "learning_rate": 3.1874969933191057e-06, "loss": 9.47380256652832, "loss_alignment": 0.806640625, "loss_alignment_w": 0.4033203125, "loss_sft": 0.18867023568600416, "loss_total_v6": 0.5921126529574394, "step": 5021 }, { "epoch": 1.1582126126126127, "grad_norm": 28.477981567382812, "learning_rate": 3.185620273310689e-06, "loss": 9.562639236450195, "loss_alignment": 0.80078125, "loss_alignment_w": 0.400390625, "loss_sft": 0.19703018106520176, "loss_total_v6": 0.5976649597287178, "step": 5022 }, { "epoch": 1.1584432432432432, "grad_norm": 33.217037200927734, "learning_rate": 3.183743847635109e-06, "loss": 9.362665176391602, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.19734921865165234, "loss_total_v6": 0.5851665884256363, "step": 5023 }, { "epoch": 1.158673873873874, "grad_norm": 28.826736450195312, "learning_rate": 3.181867716596765e-06, "loss": 9.36346435546875, "loss_alignment": 0.7646484375, "loss_alignment_w": 0.38232421875, "loss_sft": 0.20286181941628456, "loss_total_v6": 0.5852165594696999, "step": 5024 }, { "epoch": 1.1589045045045046, "grad_norm": 28.674610137939453, "learning_rate": 3.1799918805000097e-06, "loss": 9.603791236877441, "loss_alignment": 0.7861328125, "loss_alignment_w": 0.39306640625, "loss_sft": 0.2088032402098179, "loss_total_v6": 0.6002369448542595, "step": 5025 }, { "epoch": 1.159135135135135, "grad_norm": 28.20955467224121, "learning_rate": 3.1781163396491434e-06, "loss": 10.098575592041016, "loss_alignment": 0.869140625, "loss_alignment_w": 0.4345703125, "loss_sft": 0.1966821886599064, "loss_total_v6": 0.6311609447002411, "step": 5026 }, { "epoch": 1.1593657657657657, "grad_norm": 30.107473373413086, "learning_rate": 3.17624109434842e-06, "loss": 9.465324401855469, "loss_alignment": 0.77783203125, "loss_alignment_w": 0.388916015625, "loss_sft": 0.20136977173388004, "loss_total_v6": 0.5915827862918377, "step": 5027 }, { "epoch": 1.1595963963963964, "grad_norm": 28.00020408630371, "learning_rate": 3.174366144902048e-06, "loss": 10.14931869506836, "loss_alignment": 0.8369140625, "loss_alignment_w": 0.41845703125, "loss_sft": 0.21676036529242992, "loss_total_v6": 0.6343323811888695, "step": 5028 }, { "epoch": 1.159827027027027, "grad_norm": 28.974821090698242, "learning_rate": 3.172491491614183e-06, "loss": 9.417362213134766, "loss_alignment": 0.76904296875, "loss_alignment_w": 0.384521484375, "loss_sft": 0.20304132252931595, "loss_total_v6": 0.5885851159691811, "step": 5029 }, { "epoch": 1.1600576576576578, "grad_norm": 28.579130172729492, "learning_rate": 3.170617134788939e-06, "loss": 9.649229049682617, "loss_alignment": 0.837890625, "loss_alignment_w": 0.4189453125, "loss_sft": 0.1847418760880828, "loss_total_v6": 0.6030768230557442, "step": 5030 }, { "epoch": 1.1602882882882883, "grad_norm": 35.85271072387695, "learning_rate": 3.168743074730377e-06, "loss": 9.277982711791992, "loss_alignment": 0.7783203125, "loss_alignment_w": 0.38916015625, "loss_sft": 0.19092739187180996, "loss_total_v6": 0.5798739269375801, "step": 5031 }, { "epoch": 1.160518918918919, "grad_norm": 32.68257141113281, "learning_rate": 3.1668693117425128e-06, "loss": 9.613690376281738, "loss_alignment": 0.78076171875, "loss_alignment_w": 0.390380859375, "loss_sft": 0.2101696189492941, "loss_total_v6": 0.600855641067028, "step": 5032 }, { "epoch": 1.1607495495495495, "grad_norm": 32.0135612487793, "learning_rate": 3.164995846129309e-06, "loss": 9.72231674194336, "loss_alignment": 0.7861328125, "loss_alignment_w": 0.39306640625, "loss_sft": 0.2137849722057581, "loss_total_v6": 0.6076448485255241, "step": 5033 }, { "epoch": 1.1609801801801802, "grad_norm": 27.228288650512695, "learning_rate": 3.1631226781946843e-06, "loss": 9.548964500427246, "loss_alignment": 0.81494140625, "loss_alignment_w": 0.407470703125, "loss_sft": 0.18953795731067657, "loss_total_v6": 0.5968102961778641, "step": 5034 }, { "epoch": 1.1612108108108108, "grad_norm": 27.780590057373047, "learning_rate": 3.1612498082425125e-06, "loss": 9.744507789611816, "loss_alignment": 0.7958984375, "loss_alignment_w": 0.39794921875, "loss_sft": 0.21205906011164188, "loss_total_v6": 0.6090317443013191, "step": 5035 }, { "epoch": 1.1614414414414413, "grad_norm": 29.257339477539062, "learning_rate": 3.1593772365766107e-06, "loss": 9.445274353027344, "loss_alignment": 0.80810546875, "loss_alignment_w": 0.404052734375, "loss_sft": 0.18710092455148697, "loss_total_v6": 0.590329684317112, "step": 5036 }, { "epoch": 1.1616720720720721, "grad_norm": 26.68314552307129, "learning_rate": 3.1575049635007538e-06, "loss": 9.684982299804688, "loss_alignment": 0.80810546875, "loss_alignment_w": 0.404052734375, "loss_sft": 0.20234208554029465, "loss_total_v6": 0.6053114235401154, "step": 5037 }, { "epoch": 1.1619027027027027, "grad_norm": 28.938486099243164, "learning_rate": 3.155632989318665e-06, "loss": 10.108468055725098, "loss_alignment": 0.8486328125, "loss_alignment_w": 0.42431640625, "loss_sft": 0.20810369960963726, "loss_total_v6": 0.6317792385816574, "step": 5038 }, { "epoch": 1.1621333333333332, "grad_norm": 28.185062408447266, "learning_rate": 3.153761314334019e-06, "loss": 9.483304977416992, "loss_alignment": 0.78125, "loss_alignment_w": 0.390625, "loss_sft": 0.20163903757929802, "loss_total_v6": 0.5927065685391426, "step": 5039 }, { "epoch": 1.162363963963964, "grad_norm": 27.070125579833984, "learning_rate": 3.151889938850445e-06, "loss": 9.527849197387695, "loss_alignment": 0.8173828125, "loss_alignment_w": 0.40869140625, "loss_sft": 0.18589894846081734, "loss_total_v6": 0.5954906120896339, "step": 5040 }, { "epoch": 1.1625945945945946, "grad_norm": 27.679153442382812, "learning_rate": 3.1500188631715244e-06, "loss": 10.051029205322266, "loss_alignment": 0.8251953125, "loss_alignment_w": 0.41259765625, "loss_sft": 0.21531702391803265, "loss_total_v6": 0.6281893402338028, "step": 5041 }, { "epoch": 1.1628252252252251, "grad_norm": 25.97692108154297, "learning_rate": 3.1481480876007823e-06, "loss": 9.676567077636719, "loss_alignment": 0.80419921875, "loss_alignment_w": 0.402099609375, "loss_sft": 0.20325043611228466, "loss_total_v6": 0.6047854796051979, "step": 5042 }, { "epoch": 1.163055855855856, "grad_norm": 30.533113479614258, "learning_rate": 3.146277612441704e-06, "loss": 9.596287727355957, "loss_alignment": 0.8017578125, "loss_alignment_w": 0.40087890625, "loss_sft": 0.19856864772737026, "loss_total_v6": 0.5997679978609085, "step": 5043 }, { "epoch": 1.1632864864864865, "grad_norm": 31.909488677978516, "learning_rate": 3.1444074379977185e-06, "loss": 9.76050090789795, "loss_alignment": 0.81787109375, "loss_alignment_w": 0.408935546875, "loss_sft": 0.20126356557011604, "loss_total_v6": 0.61003128439188, "step": 5044 }, { "epoch": 1.163517117117117, "grad_norm": 26.0400447845459, "learning_rate": 3.1425375645722147e-06, "loss": 9.327592849731445, "loss_alignment": 0.78076171875, "loss_alignment_w": 0.390380859375, "loss_sft": 0.192212151363492, "loss_total_v6": 0.5829745009541512, "step": 5045 }, { "epoch": 1.1637477477477478, "grad_norm": 24.9808292388916, "learning_rate": 3.140667992468524e-06, "loss": 9.289512634277344, "loss_alignment": 0.82861328125, "loss_alignment_w": 0.414306640625, "loss_sft": 0.16705081891268492, "loss_total_v6": 0.580594539642334, "step": 5046 }, { "epoch": 1.1639783783783784, "grad_norm": 31.98677635192871, "learning_rate": 3.1387987219899344e-06, "loss": 9.35057544708252, "loss_alignment": 0.80517578125, "loss_alignment_w": 0.402587890625, "loss_sft": 0.18006834480911493, "loss_total_v6": 0.5844109877943993, "step": 5047 }, { "epoch": 1.164209009009009, "grad_norm": 28.29001808166504, "learning_rate": 3.1369297534396823e-06, "loss": 9.500551223754883, "loss_alignment": 0.80029296875, "loss_alignment_w": 0.400146484375, "loss_sft": 0.19311919622123241, "loss_total_v6": 0.5937844812870026, "step": 5048 }, { "epoch": 1.1644396396396397, "grad_norm": 25.717859268188477, "learning_rate": 3.1350610871209553e-06, "loss": 9.568519592285156, "loss_alignment": 0.80419921875, "loss_alignment_w": 0.402099609375, "loss_sft": 0.1962685389444232, "loss_total_v6": 0.5980324521660805, "step": 5049 }, { "epoch": 1.1646702702702703, "grad_norm": 31.145044326782227, "learning_rate": 3.1331927233368954e-06, "loss": 9.654173851013184, "loss_alignment": 0.81689453125, "loss_alignment_w": 0.408447265625, "loss_sft": 0.19527428969740868, "loss_total_v6": 0.603385865688324, "step": 5050 }, { "epoch": 1.1649009009009008, "grad_norm": 25.448850631713867, "learning_rate": 3.1313246623905918e-06, "loss": 9.012126922607422, "loss_alignment": 0.77197265625, "loss_alignment_w": 0.385986328125, "loss_sft": 0.17693592607975006, "loss_total_v6": 0.5632579475641251, "step": 5051 }, { "epoch": 1.1651315315315316, "grad_norm": 28.2034854888916, "learning_rate": 3.1294569045850844e-06, "loss": 9.449934005737305, "loss_alignment": 0.80322265625, "loss_alignment_w": 0.401611328125, "loss_sft": 0.19132881425321102, "loss_total_v6": 0.5906208381056786, "step": 5052 }, { "epoch": 1.1653621621621622, "grad_norm": 29.605180740356445, "learning_rate": 3.127589450223364e-06, "loss": 9.036596298217773, "loss_alignment": 0.7548828125, "loss_alignment_w": 0.37744140625, "loss_sft": 0.18774260487407446, "loss_total_v6": 0.564787283539772, "step": 5053 }, { "epoch": 1.1655927927927927, "grad_norm": 32.172245025634766, "learning_rate": 3.125722299608378e-06, "loss": 9.151790618896484, "loss_alignment": 0.76708984375, "loss_alignment_w": 0.383544921875, "loss_sft": 0.18825892824679613, "loss_total_v6": 0.5719869434833527, "step": 5054 }, { "epoch": 1.1658234234234235, "grad_norm": 30.106178283691406, "learning_rate": 3.1238554530430152e-06, "loss": 9.777214050292969, "loss_alignment": 0.802734375, "loss_alignment_w": 0.4013671875, "loss_sft": 0.2097391877323389, "loss_total_v6": 0.6110758557915688, "step": 5055 }, { "epoch": 1.166054054054054, "grad_norm": 23.336774826049805, "learning_rate": 3.1219889108301238e-06, "loss": 9.227341651916504, "loss_alignment": 0.81103515625, "loss_alignment_w": 0.405517578125, "loss_sft": 0.17122177965939045, "loss_total_v6": 0.5767088532447815, "step": 5056 }, { "epoch": 1.1662846846846846, "grad_norm": 26.73682403564453, "learning_rate": 3.1201226732724956e-06, "loss": 9.459964752197266, "loss_alignment": 0.79541015625, "loss_alignment_w": 0.397705078125, "loss_sft": 0.19352742657065392, "loss_total_v6": 0.5912477746605873, "step": 5057 }, { "epoch": 1.1665153153153154, "grad_norm": 27.243600845336914, "learning_rate": 3.1182567406728746e-06, "loss": 9.407843589782715, "loss_alignment": 0.7998046875, "loss_alignment_w": 0.39990234375, "loss_sft": 0.18781321682035923, "loss_total_v6": 0.5879902169108391, "step": 5058 }, { "epoch": 1.166745945945946, "grad_norm": 23.578187942504883, "learning_rate": 3.1163911133339603e-06, "loss": 9.414651870727539, "loss_alignment": 0.7783203125, "loss_alignment_w": 0.38916015625, "loss_sft": 0.19893511570990086, "loss_total_v6": 0.5884157195687294, "step": 5059 }, { "epoch": 1.1669765765765765, "grad_norm": 30.627418518066406, "learning_rate": 3.114525791558398e-06, "loss": 9.330122947692871, "loss_alignment": 0.80712890625, "loss_alignment_w": 0.403564453125, "loss_sft": 0.18046848475933075, "loss_total_v6": 0.5831326879560947, "step": 5060 }, { "epoch": 1.1672072072072073, "grad_norm": 29.4888973236084, "learning_rate": 3.1126607756487836e-06, "loss": 9.958600997924805, "loss_alignment": 0.82568359375, "loss_alignment_w": 0.412841796875, "loss_sft": 0.2093571051955223, "loss_total_v6": 0.6224125325679779, "step": 5061 }, { "epoch": 1.1674378378378378, "grad_norm": 26.56478500366211, "learning_rate": 3.110796065907665e-06, "loss": 9.770181655883789, "loss_alignment": 0.83203125, "loss_alignment_w": 0.416015625, "loss_sft": 0.1958720088005066, "loss_total_v6": 0.6106363981962204, "step": 5062 }, { "epoch": 1.1676684684684684, "grad_norm": 28.5311222076416, "learning_rate": 3.1089316626375383e-06, "loss": 9.607767105102539, "loss_alignment": 0.80615234375, "loss_alignment_w": 0.403076171875, "loss_sft": 0.196417436003685, "loss_total_v6": 0.6004854440689087, "step": 5063 }, { "epoch": 1.1678990990990992, "grad_norm": 27.541595458984375, "learning_rate": 3.1070675661408533e-06, "loss": 9.549646377563477, "loss_alignment": 0.81884765625, "loss_alignment_w": 0.409423828125, "loss_sft": 0.18741381168365479, "loss_total_v6": 0.5968528911471367, "step": 5064 }, { "epoch": 1.1681297297297297, "grad_norm": 29.061521530151367, "learning_rate": 3.1052037767200073e-06, "loss": 9.489198684692383, "loss_alignment": 0.8251953125, "loss_alignment_w": 0.41259765625, "loss_sft": 0.18070615455508232, "loss_total_v6": 0.5930749252438545, "step": 5065 }, { "epoch": 1.1683603603603603, "grad_norm": 27.25901222229004, "learning_rate": 3.103340294677348e-06, "loss": 9.564977645874023, "loss_alignment": 0.79833984375, "loss_alignment_w": 0.399169921875, "loss_sft": 0.19827493652701378, "loss_total_v6": 0.5978110805153847, "step": 5066 }, { "epoch": 1.168590990990991, "grad_norm": 28.06563377380371, "learning_rate": 3.101477120315173e-06, "loss": 8.96273136138916, "loss_alignment": 0.77294921875, "loss_alignment_w": 0.386474609375, "loss_sft": 0.17291790712624788, "loss_total_v6": 0.5601706989109516, "step": 5067 }, { "epoch": 1.1688216216216216, "grad_norm": 30.981992721557617, "learning_rate": 3.099614253935731e-06, "loss": 9.674858093261719, "loss_alignment": 0.81689453125, "loss_alignment_w": 0.408447265625, "loss_sft": 0.19659753423184156, "loss_total_v6": 0.6046785935759544, "step": 5068 }, { "epoch": 1.1690522522522522, "grad_norm": 29.986312866210938, "learning_rate": 3.0977516958412236e-06, "loss": 9.778749465942383, "loss_alignment": 0.83203125, "loss_alignment_w": 0.416015625, "loss_sft": 0.19552240148186684, "loss_total_v6": 0.6111718341708183, "step": 5069 }, { "epoch": 1.169282882882883, "grad_norm": 33.165985107421875, "learning_rate": 3.0958894463337957e-06, "loss": 10.2781982421875, "loss_alignment": 0.8505859375, "loss_alignment_w": 0.42529296875, "loss_sft": 0.21663663163781166, "loss_total_v6": 0.6423873752355576, "step": 5070 }, { "epoch": 1.1695135135135135, "grad_norm": 26.67398452758789, "learning_rate": 3.094027505715547e-06, "loss": 9.192832946777344, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.18748239241540432, "loss_total_v6": 0.574552096426487, "step": 5071 }, { "epoch": 1.169744144144144, "grad_norm": 27.55837631225586, "learning_rate": 3.092165874288525e-06, "loss": 9.429594039916992, "loss_alignment": 0.77880859375, "loss_alignment_w": 0.389404296875, "loss_sft": 0.19994529150426388, "loss_total_v6": 0.5893495976924896, "step": 5072 }, { "epoch": 1.1699747747747749, "grad_norm": 25.892974853515625, "learning_rate": 3.090304552354727e-06, "loss": 9.706992149353027, "loss_alignment": 0.80810546875, "loss_alignment_w": 0.404052734375, "loss_sft": 0.20232907123863697, "loss_total_v6": 0.6066869720816612, "step": 5073 }, { "epoch": 1.1702054054054054, "grad_norm": 28.324451446533203, "learning_rate": 3.0884435402161034e-06, "loss": 9.457063674926758, "loss_alignment": 0.80029296875, "loss_alignment_w": 0.400146484375, "loss_sft": 0.1904775220900774, "loss_total_v6": 0.5910664647817612, "step": 5074 }, { "epoch": 1.170436036036036, "grad_norm": 26.904706954956055, "learning_rate": 3.0865828381745515e-06, "loss": 9.738088607788086, "loss_alignment": 0.80859375, "loss_alignment_w": 0.404296875, "loss_sft": 0.20468463376164436, "loss_total_v6": 0.6086305528879166, "step": 5075 }, { "epoch": 1.1706666666666667, "grad_norm": 28.982728958129883, "learning_rate": 3.0847224465319183e-06, "loss": 9.329526901245117, "loss_alignment": 0.83544921875, "loss_alignment_w": 0.417724609375, "loss_sft": 0.16508087143301964, "loss_total_v6": 0.5830954164266586, "step": 5076 }, { "epoch": 1.1708972972972973, "grad_norm": 29.44859504699707, "learning_rate": 3.082862365589999e-06, "loss": 9.699530601501465, "loss_alignment": 0.81103515625, "loss_alignment_w": 0.405517578125, "loss_sft": 0.2007031086832285, "loss_total_v6": 0.6062206849455833, "step": 5077 }, { "epoch": 1.1711279279279279, "grad_norm": 31.21261215209961, "learning_rate": 3.081002595650541e-06, "loss": 9.632423400878906, "loss_alignment": 0.837890625, "loss_alignment_w": 0.4189453125, "loss_sft": 0.18422550056129694, "loss_total_v6": 0.6020264253020287, "step": 5078 }, { "epoch": 1.1713585585585586, "grad_norm": 30.22932243347168, "learning_rate": 3.079143137015242e-06, "loss": 9.290620803833008, "loss_alignment": 0.7705078125, "loss_alignment_w": 0.38525390625, "loss_sft": 0.19539465568959713, "loss_total_v6": 0.5806638151407242, "step": 5079 }, { "epoch": 1.1715891891891892, "grad_norm": 29.082870483398438, "learning_rate": 3.0772839899857465e-06, "loss": 9.841713905334473, "loss_alignment": 0.82958984375, "loss_alignment_w": 0.414794921875, "loss_sft": 0.20048005692660809, "loss_total_v6": 0.6151071190834045, "step": 5080 }, { "epoch": 1.1718198198198198, "grad_norm": 32.43397903442383, "learning_rate": 3.0754251548636503e-06, "loss": 9.456918716430664, "loss_alignment": 0.7734375, "loss_alignment_w": 0.38671875, "loss_sft": 0.2037283256649971, "loss_total_v6": 0.5910574197769165, "step": 5081 }, { "epoch": 1.1720504504504505, "grad_norm": 29.681392669677734, "learning_rate": 3.0735666319504975e-06, "loss": 9.47982120513916, "loss_alignment": 0.8076171875, "loss_alignment_w": 0.40380859375, "loss_sft": 0.18880234844982624, "loss_total_v6": 0.5924888625741005, "step": 5082 }, { "epoch": 1.172281081081081, "grad_norm": 29.299840927124023, "learning_rate": 3.0717084215477793e-06, "loss": 9.875850677490234, "loss_alignment": 0.81591796875, "loss_alignment_w": 0.407958984375, "loss_sft": 0.20941904187202454, "loss_total_v6": 0.6172407045960426, "step": 5083 }, { "epoch": 1.1725117117117116, "grad_norm": 31.555246353149414, "learning_rate": 3.0698505239569425e-06, "loss": 9.087591171264648, "loss_alignment": 0.79736328125, "loss_alignment_w": 0.398681640625, "loss_sft": 0.1697963047772646, "loss_total_v6": 0.5679744184017181, "step": 5084 }, { "epoch": 1.1727423423423424, "grad_norm": 26.023696899414062, "learning_rate": 3.0679929394793796e-06, "loss": 9.172933578491211, "loss_alignment": 0.78369140625, "loss_alignment_w": 0.391845703125, "loss_sft": 0.1811269372701645, "loss_total_v6": 0.5733083188533783, "step": 5085 }, { "epoch": 1.172972972972973, "grad_norm": 30.51306915283203, "learning_rate": 3.06613566841643e-06, "loss": 9.636244773864746, "loss_alignment": 0.84228515625, "loss_alignment_w": 0.421142578125, "loss_sft": 0.1815956849604845, "loss_total_v6": 0.6022652760148048, "step": 5086 }, { "epoch": 1.1732036036036035, "grad_norm": 29.15662956237793, "learning_rate": 3.064278711069386e-06, "loss": 9.290140151977539, "loss_alignment": 0.7890625, "loss_alignment_w": 0.39453125, "loss_sft": 0.18588889949023724, "loss_total_v6": 0.5806337669491768, "step": 5087 }, { "epoch": 1.1734342342342343, "grad_norm": 27.869155883789062, "learning_rate": 3.0624220677394854e-06, "loss": 9.417510986328125, "loss_alignment": 0.79296875, "loss_alignment_w": 0.396484375, "loss_sft": 0.1924610361456871, "loss_total_v6": 0.5885944738984108, "step": 5088 }, { "epoch": 1.1736648648648649, "grad_norm": 30.777385711669922, "learning_rate": 3.0605657387279198e-06, "loss": 9.431676864624023, "loss_alignment": 0.77587890625, "loss_alignment_w": 0.387939453125, "loss_sft": 0.20212018489837646, "loss_total_v6": 0.5894798040390015, "step": 5089 }, { "epoch": 1.1738954954954954, "grad_norm": 26.768718719482422, "learning_rate": 3.0587097243358254e-06, "loss": 9.192972183227539, "loss_alignment": 0.7626953125, "loss_alignment_w": 0.38134765625, "loss_sft": 0.19432702846825123, "loss_total_v6": 0.5745607912540436, "step": 5090 }, { "epoch": 1.1741261261261262, "grad_norm": 30.281782150268555, "learning_rate": 3.056854024864291e-06, "loss": 9.280186653137207, "loss_alignment": 0.81298828125, "loss_alignment_w": 0.406494140625, "loss_sft": 0.17258677165955305, "loss_total_v6": 0.5800116956233978, "step": 5091 }, { "epoch": 1.1743567567567568, "grad_norm": 27.972137451171875, "learning_rate": 3.05499864061435e-06, "loss": 9.734488487243652, "loss_alignment": 0.783203125, "loss_alignment_w": 0.3916015625, "loss_sft": 0.21775002591311932, "loss_total_v6": 0.6084055304527283, "step": 5092 }, { "epoch": 1.1745873873873873, "grad_norm": 32.50519943237305, "learning_rate": 3.0531435718869873e-06, "loss": 9.674509048461914, "loss_alignment": 0.80810546875, "loss_alignment_w": 0.404052734375, "loss_sft": 0.20112286694347858, "loss_total_v6": 0.604656808078289, "step": 5093 }, { "epoch": 1.174818018018018, "grad_norm": 29.125152587890625, "learning_rate": 3.0512888189831386e-06, "loss": 9.999001502990723, "loss_alignment": 0.83349609375, "loss_alignment_w": 0.416748046875, "loss_sft": 0.20796065777540207, "loss_total_v6": 0.6249375864863396, "step": 5094 }, { "epoch": 1.1750486486486487, "grad_norm": 27.389623641967773, "learning_rate": 3.0494343822036856e-06, "loss": 9.857965469360352, "loss_alignment": 0.837890625, "loss_alignment_w": 0.4189453125, "loss_sft": 0.19777266681194305, "loss_total_v6": 0.6161228567361832, "step": 5095 }, { "epoch": 1.1752792792792792, "grad_norm": 27.762306213378906, "learning_rate": 3.0475802618494564e-06, "loss": 9.509904861450195, "loss_alignment": 0.81005859375, "loss_alignment_w": 0.405029296875, "loss_sft": 0.19072829093784094, "loss_total_v6": 0.5943690463900566, "step": 5096 }, { "epoch": 1.17550990990991, "grad_norm": 26.459383010864258, "learning_rate": 3.045726458221235e-06, "loss": 9.175739288330078, "loss_alignment": 0.80126953125, "loss_alignment_w": 0.400634765625, "loss_sft": 0.17258956003934145, "loss_total_v6": 0.5734837427735329, "step": 5097 }, { "epoch": 1.1757405405405406, "grad_norm": 23.764619827270508, "learning_rate": 3.0438729716197444e-06, "loss": 9.654287338256836, "loss_alignment": 0.81640625, "loss_alignment_w": 0.408203125, "loss_sft": 0.1951440554112196, "loss_total_v6": 0.6033929660916328, "step": 5098 }, { "epoch": 1.1759711711711711, "grad_norm": 29.773653030395508, "learning_rate": 3.0420198023456664e-06, "loss": 9.472063064575195, "loss_alignment": 0.79248046875, "loss_alignment_w": 0.396240234375, "loss_sft": 0.1967250443994999, "loss_total_v6": 0.5920039936900139, "step": 5099 }, { "epoch": 1.176201801801802, "grad_norm": 28.543354034423828, "learning_rate": 3.040166950699626e-06, "loss": 9.162942886352539, "loss_alignment": 0.78759765625, "loss_alignment_w": 0.393798828125, "loss_sft": 0.17949545197188854, "loss_total_v6": 0.5726839080452919, "step": 5100 }, { "epoch": 1.1764324324324325, "grad_norm": 30.255916595458984, "learning_rate": 3.0383144169821944e-06, "loss": 9.664176940917969, "loss_alignment": 0.81103515625, "loss_alignment_w": 0.405517578125, "loss_sft": 0.19721174612641335, "loss_total_v6": 0.6040110737085342, "step": 5101 }, { "epoch": 1.176663063063063, "grad_norm": 31.928468704223633, "learning_rate": 3.0364622014938927e-06, "loss": 9.604427337646484, "loss_alignment": 0.79736328125, "loss_alignment_w": 0.398681640625, "loss_sft": 0.19968772120773792, "loss_total_v6": 0.6002767011523247, "step": 5102 }, { "epoch": 1.1768936936936938, "grad_norm": 34.479522705078125, "learning_rate": 3.034610304535196e-06, "loss": 9.58631420135498, "loss_alignment": 0.82421875, "loss_alignment_w": 0.412109375, "loss_sft": 0.1877371808513999, "loss_total_v6": 0.5991446375846863, "step": 5103 }, { "epoch": 1.1771243243243243, "grad_norm": 25.529272079467773, "learning_rate": 3.032758726406521e-06, "loss": 9.409818649291992, "loss_alignment": 0.79736328125, "loss_alignment_w": 0.398681640625, "loss_sft": 0.18915732018649578, "loss_total_v6": 0.5881136506795883, "step": 5104 }, { "epoch": 1.177354954954955, "grad_norm": 27.731369018554688, "learning_rate": 3.030907467408235e-06, "loss": 9.356990814208984, "loss_alignment": 0.81982421875, "loss_alignment_w": 0.409912109375, "loss_sft": 0.17500665597617626, "loss_total_v6": 0.5848119631409645, "step": 5105 }, { "epoch": 1.1775855855855857, "grad_norm": 26.279626846313477, "learning_rate": 3.029056527840654e-06, "loss": 9.609874725341797, "loss_alignment": 0.7861328125, "loss_alignment_w": 0.39306640625, "loss_sft": 0.2081763856112957, "loss_total_v6": 0.6006171852350235, "step": 5106 }, { "epoch": 1.1778162162162162, "grad_norm": 29.50487518310547, "learning_rate": 3.0272059080040393e-06, "loss": 9.648406982421875, "loss_alignment": 0.798828125, "loss_alignment_w": 0.3994140625, "loss_sft": 0.20214648731052876, "loss_total_v6": 0.6030253991484642, "step": 5107 }, { "epoch": 1.1780468468468468, "grad_norm": 29.920164108276367, "learning_rate": 3.025355608198606e-06, "loss": 9.286121368408203, "loss_alignment": 0.8056640625, "loss_alignment_w": 0.40283203125, "loss_sft": 0.17802361398935318, "loss_total_v6": 0.5803826078772545, "step": 5108 }, { "epoch": 1.1782774774774776, "grad_norm": 28.504302978515625, "learning_rate": 3.023505628724512e-06, "loss": 9.05465316772461, "loss_alignment": 0.7568359375, "loss_alignment_w": 0.37841796875, "loss_sft": 0.1883370652794838, "loss_total_v6": 0.5659158006310463, "step": 5109 }, { "epoch": 1.1785081081081081, "grad_norm": 30.13155174255371, "learning_rate": 3.0216559698818672e-06, "loss": 9.361038208007812, "loss_alignment": 0.794921875, "loss_alignment_w": 0.3974609375, "loss_sft": 0.1874513290822506, "loss_total_v6": 0.5850648805499077, "step": 5110 }, { "epoch": 1.1787387387387387, "grad_norm": 29.552846908569336, "learning_rate": 3.0198066319707246e-06, "loss": 10.189706802368164, "loss_alignment": 0.82421875, "loss_alignment_w": 0.412109375, "loss_sft": 0.22486941143870354, "loss_total_v6": 0.636856697499752, "step": 5111 }, { "epoch": 1.1789693693693692, "grad_norm": 33.90107345581055, "learning_rate": 3.017957615291088e-06, "loss": 9.042413711547852, "loss_alignment": 0.7822265625, "loss_alignment_w": 0.39111328125, "loss_sft": 0.1739154877141118, "loss_total_v6": 0.5651508308947086, "step": 5112 }, { "epoch": 1.1792, "grad_norm": 31.873323440551758, "learning_rate": 3.016108920142914e-06, "loss": 9.736566543579102, "loss_alignment": 0.8115234375, "loss_alignment_w": 0.40576171875, "loss_sft": 0.20313993096351624, "loss_total_v6": 0.6085354536771774, "step": 5113 }, { "epoch": 1.1794306306306306, "grad_norm": 29.45583152770996, "learning_rate": 3.0142605468260976e-06, "loss": 9.762002944946289, "loss_alignment": 0.8076171875, "loss_alignment_w": 0.40380859375, "loss_sft": 0.20711007155478, "loss_total_v6": 0.6101251915097237, "step": 5114 }, { "epoch": 1.1796612612612614, "grad_norm": 31.409698486328125, "learning_rate": 3.012412495640487e-06, "loss": 9.929937362670898, "loss_alignment": 0.8310546875, "loss_alignment_w": 0.41552734375, "loss_sft": 0.20532258786261082, "loss_total_v6": 0.6206210404634476, "step": 5115 }, { "epoch": 1.179891891891892, "grad_norm": 30.207843780517578, "learning_rate": 3.010564766885878e-06, "loss": 9.303433418273926, "loss_alignment": 0.80908203125, "loss_alignment_w": 0.404541015625, "loss_sft": 0.17614536732435226, "loss_total_v6": 0.5814645737409592, "step": 5116 }, { "epoch": 1.1801225225225225, "grad_norm": 26.305959701538086, "learning_rate": 3.008717360862012e-06, "loss": 9.119500160217285, "loss_alignment": 0.80078125, "loss_alignment_w": 0.400390625, "loss_sft": 0.16898304782807827, "loss_total_v6": 0.5699687600135803, "step": 5117 }, { "epoch": 1.180353153153153, "grad_norm": 29.14403533935547, "learning_rate": 3.006870277868581e-06, "loss": 9.373668670654297, "loss_alignment": 0.8046875, "loss_alignment_w": 0.40234375, "loss_sft": 0.18308331444859505, "loss_total_v6": 0.5858542993664742, "step": 5118 }, { "epoch": 1.1805837837837838, "grad_norm": 33.64267349243164, "learning_rate": 3.0050235182052234e-06, "loss": 9.497957229614258, "loss_alignment": 0.7705078125, "loss_alignment_w": 0.38525390625, "loss_sft": 0.20775808952748775, "loss_total_v6": 0.5936223566532135, "step": 5119 }, { "epoch": 1.1808144144144144, "grad_norm": 29.616147994995117, "learning_rate": 3.0031770821715233e-06, "loss": 9.825364112854004, "loss_alignment": 0.8349609375, "loss_alignment_w": 0.41748046875, "loss_sft": 0.19542985409498215, "loss_total_v6": 0.6140852496027946, "step": 5120 }, { "epoch": 1.1810450450450451, "grad_norm": 25.76966667175293, "learning_rate": 3.0013309700670146e-06, "loss": 10.028326988220215, "loss_alignment": 0.828125, "loss_alignment_w": 0.4140625, "loss_sft": 0.2130283461883664, "loss_total_v6": 0.6267704293131828, "step": 5121 }, { "epoch": 1.1812756756756757, "grad_norm": 32.21657943725586, "learning_rate": 2.9994851821911754e-06, "loss": 9.200005531311035, "loss_alignment": 0.79443359375, "loss_alignment_w": 0.397216796875, "loss_sft": 0.17818027082830667, "loss_total_v6": 0.5750003457069397, "step": 5122 }, { "epoch": 1.1815063063063063, "grad_norm": 30.193758010864258, "learning_rate": 2.9976397188434388e-06, "loss": 9.3065767288208, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.19425567612051964, "loss_total_v6": 0.5816610604524612, "step": 5123 }, { "epoch": 1.1817369369369368, "grad_norm": 32.46367263793945, "learning_rate": 2.9957945803231757e-06, "loss": 9.628340721130371, "loss_alignment": 0.8154296875, "loss_alignment_w": 0.40771484375, "loss_sft": 0.1942700818181038, "loss_total_v6": 0.60177131742239, "step": 5124 }, { "epoch": 1.1819675675675676, "grad_norm": 30.885875701904297, "learning_rate": 2.993949766929711e-06, "loss": 9.547003746032715, "loss_alignment": 0.78662109375, "loss_alignment_w": 0.393310546875, "loss_sft": 0.20218697749078274, "loss_total_v6": 0.5966877043247223, "step": 5125 }, { "epoch": 1.1821981981981982, "grad_norm": 29.41395378112793, "learning_rate": 2.992105278962314e-06, "loss": 9.470827102661133, "loss_alignment": 0.79296875, "loss_alignment_w": 0.396484375, "loss_sft": 0.1950914105400443, "loss_total_v6": 0.5919267162680626, "step": 5126 }, { "epoch": 1.182428828828829, "grad_norm": 28.39815902709961, "learning_rate": 2.9902611167202e-06, "loss": 9.56796646118164, "loss_alignment": 0.8046875, "loss_alignment_w": 0.40234375, "loss_sft": 0.19588305056095123, "loss_total_v6": 0.5979979038238525, "step": 5127 }, { "epoch": 1.1826594594594595, "grad_norm": 29.78813934326172, "learning_rate": 2.9884172805025345e-06, "loss": 9.565200805664062, "loss_alignment": 0.8095703125, "loss_alignment_w": 0.40478515625, "loss_sft": 0.19319251365959644, "loss_total_v6": 0.5978250652551651, "step": 5128 }, { "epoch": 1.18289009009009, "grad_norm": 29.429433822631836, "learning_rate": 2.986573770608431e-06, "loss": 9.581464767456055, "loss_alignment": 0.8271484375, "loss_alignment_w": 0.41357421875, "loss_sft": 0.18410767428576946, "loss_total_v6": 0.5988415479660034, "step": 5129 }, { "epoch": 1.1831207207207206, "grad_norm": 29.24380111694336, "learning_rate": 2.9847305873369447e-06, "loss": 9.842113494873047, "loss_alignment": 0.82421875, "loss_alignment_w": 0.412109375, "loss_sft": 0.20290067605674267, "loss_total_v6": 0.615132138133049, "step": 5130 }, { "epoch": 1.1833513513513514, "grad_norm": 30.561159133911133, "learning_rate": 2.982887730987083e-06, "loss": 9.427789688110352, "loss_alignment": 0.7666015625, "loss_alignment_w": 0.38330078125, "loss_sft": 0.20558508671820164, "loss_total_v6": 0.5892368033528328, "step": 5131 }, { "epoch": 1.183581981981982, "grad_norm": 30.68246841430664, "learning_rate": 2.981045201857796e-06, "loss": 9.373821258544922, "loss_alignment": 0.79248046875, "loss_alignment_w": 0.396240234375, "loss_sft": 0.1893489034846425, "loss_total_v6": 0.5858638063073158, "step": 5132 }, { "epoch": 1.1838126126126127, "grad_norm": 27.619686126708984, "learning_rate": 2.9792030002479868e-06, "loss": 9.493903160095215, "loss_alignment": 0.7802734375, "loss_alignment_w": 0.39013671875, "loss_sft": 0.20315594039857388, "loss_total_v6": 0.5933689251542091, "step": 5133 }, { "epoch": 1.1840432432432433, "grad_norm": 29.675630569458008, "learning_rate": 2.9773611264564973e-06, "loss": 10.185420989990234, "loss_alignment": 0.84912109375, "loss_alignment_w": 0.424560546875, "loss_sft": 0.21114331856369972, "loss_total_v6": 0.6365888342261314, "step": 5134 }, { "epoch": 1.1842738738738738, "grad_norm": 31.720542907714844, "learning_rate": 2.9755195807821236e-06, "loss": 9.258503913879395, "loss_alignment": 0.78955078125, "loss_alignment_w": 0.394775390625, "loss_sft": 0.1834538597613573, "loss_total_v6": 0.5786564946174622, "step": 5135 }, { "epoch": 1.1845045045045044, "grad_norm": 30.521728515625, "learning_rate": 2.9736783635236044e-06, "loss": 9.21810245513916, "loss_alignment": 0.76806640625, "loss_alignment_w": 0.384033203125, "loss_sft": 0.19232705049216747, "loss_total_v6": 0.5761313810944557, "step": 5136 }, { "epoch": 1.1847351351351352, "grad_norm": 26.117002487182617, "learning_rate": 2.9718374749796244e-06, "loss": 9.634034156799316, "loss_alignment": 0.80810546875, "loss_alignment_w": 0.404052734375, "loss_sft": 0.19834905862808228, "loss_total_v6": 0.6021271273493767, "step": 5137 }, { "epoch": 1.1849657657657657, "grad_norm": 28.32009506225586, "learning_rate": 2.9699969154488206e-06, "loss": 9.705477714538574, "loss_alignment": 0.8076171875, "loss_alignment_w": 0.40380859375, "loss_sft": 0.20278378389775753, "loss_total_v6": 0.6065923720598221, "step": 5138 }, { "epoch": 1.1851963963963965, "grad_norm": 28.389598846435547, "learning_rate": 2.9681566852297692e-06, "loss": 8.938505172729492, "loss_alignment": 0.7822265625, "loss_alignment_w": 0.39111328125, "loss_sft": 0.16728386282920837, "loss_total_v6": 0.5586565360426903, "step": 5139 }, { "epoch": 1.185427027027027, "grad_norm": 29.35165023803711, "learning_rate": 2.966316784621e-06, "loss": 9.650773048400879, "loss_alignment": 0.81396484375, "loss_alignment_w": 0.406982421875, "loss_sft": 0.19568735174834728, "loss_total_v6": 0.6031732857227325, "step": 5140 }, { "epoch": 1.1856576576576576, "grad_norm": 29.466144561767578, "learning_rate": 2.964477213920983e-06, "loss": 9.768688201904297, "loss_alignment": 0.7998046875, "loss_alignment_w": 0.39990234375, "loss_sft": 0.20977092906832695, "loss_total_v6": 0.6105430200695992, "step": 5141 }, { "epoch": 1.1858882882882882, "grad_norm": 31.229568481445312, "learning_rate": 2.9626379734281373e-06, "loss": 9.875086784362793, "loss_alignment": 0.810546875, "loss_alignment_w": 0.4052734375, "loss_sft": 0.21315544843673706, "loss_total_v6": 0.6171929389238358, "step": 5142 }, { "epoch": 1.186118918918919, "grad_norm": 36.68936538696289, "learning_rate": 2.960799063440831e-06, "loss": 9.528822898864746, "loss_alignment": 0.79248046875, "loss_alignment_w": 0.396240234375, "loss_sft": 0.19841093756258488, "loss_total_v6": 0.5955514311790466, "step": 5143 }, { "epoch": 1.1863495495495495, "grad_norm": 27.651813507080078, "learning_rate": 2.9589604842573762e-06, "loss": 9.287309646606445, "loss_alignment": 0.79638671875, "loss_alignment_w": 0.398193359375, "loss_sft": 0.18185156397521496, "loss_total_v6": 0.580456905066967, "step": 5144 }, { "epoch": 1.1865801801801803, "grad_norm": 30.074138641357422, "learning_rate": 2.957122236176029e-06, "loss": 9.45051097869873, "loss_alignment": 0.798828125, "loss_alignment_w": 0.3994140625, "loss_sft": 0.19098349288105965, "loss_total_v6": 0.5906569585204124, "step": 5145 }, { "epoch": 1.1868108108108109, "grad_norm": 32.59885025024414, "learning_rate": 2.9552843194949966e-06, "loss": 9.374244689941406, "loss_alignment": 0.78369140625, "loss_alignment_w": 0.391845703125, "loss_sft": 0.1944718286395073, "loss_total_v6": 0.5858902856707573, "step": 5146 }, { "epoch": 1.1870414414414414, "grad_norm": 26.97109603881836, "learning_rate": 2.9534467345124273e-06, "loss": 8.875580787658691, "loss_alignment": 0.7578125, "loss_alignment_w": 0.37890625, "loss_sft": 0.17494781874120235, "loss_total_v6": 0.55472382158041, "step": 5147 }, { "epoch": 1.187272072072072, "grad_norm": 28.18146324157715, "learning_rate": 2.951609481526422e-06, "loss": 8.974698066711426, "loss_alignment": 0.79150390625, "loss_alignment_w": 0.395751953125, "loss_sft": 0.16577706206589937, "loss_total_v6": 0.5609186664223671, "step": 5148 }, { "epoch": 1.1875027027027028, "grad_norm": 26.812313079833984, "learning_rate": 2.9497725608350213e-06, "loss": 9.623221397399902, "loss_alignment": 0.8037109375, "loss_alignment_w": 0.40185546875, "loss_sft": 0.19897024892270565, "loss_total_v6": 0.6014513447880745, "step": 5149 }, { "epoch": 1.1877333333333333, "grad_norm": 28.268417358398438, "learning_rate": 2.947935972736217e-06, "loss": 9.693181037902832, "loss_alignment": 0.814453125, "loss_alignment_w": 0.4072265625, "loss_sft": 0.19775797240436077, "loss_total_v6": 0.605823777616024, "step": 5150 }, { "epoch": 1.187963963963964, "grad_norm": 31.495527267456055, "learning_rate": 2.9460997175279414e-06, "loss": 9.5973482131958, "loss_alignment": 0.791015625, "loss_alignment_w": 0.3955078125, "loss_sft": 0.20400601625442505, "loss_total_v6": 0.5998342707753181, "step": 5151 }, { "epoch": 1.1881945945945946, "grad_norm": 27.865930557250977, "learning_rate": 2.9442637955080787e-06, "loss": 9.490447044372559, "loss_alignment": 0.81396484375, "loss_alignment_w": 0.406982421875, "loss_sft": 0.1872996874153614, "loss_total_v6": 0.5931529551744461, "step": 5152 }, { "epoch": 1.1884252252252252, "grad_norm": 28.278898239135742, "learning_rate": 2.9424282069744564e-06, "loss": 9.294975280761719, "loss_alignment": 0.7861328125, "loss_alignment_w": 0.39306640625, "loss_sft": 0.18716761842370033, "loss_total_v6": 0.5809359401464462, "step": 5153 }, { "epoch": 1.1886558558558558, "grad_norm": 29.576881408691406, "learning_rate": 2.9405929522248476e-06, "loss": 8.866069793701172, "loss_alignment": 0.771484375, "loss_alignment_w": 0.3857421875, "loss_sft": 0.16818878706544638, "loss_total_v6": 0.5541293397545815, "step": 5154 }, { "epoch": 1.1888864864864865, "grad_norm": 25.857070922851562, "learning_rate": 2.9387580315569695e-06, "loss": 9.38819694519043, "loss_alignment": 0.81689453125, "loss_alignment_w": 0.408447265625, "loss_sft": 0.17918476182967424, "loss_total_v6": 0.5867622792720795, "step": 5155 }, { "epoch": 1.189117117117117, "grad_norm": 30.796100616455078, "learning_rate": 2.936923445268488e-06, "loss": 9.274417877197266, "loss_alignment": 0.779296875, "loss_alignment_w": 0.3896484375, "loss_sft": 0.19099454581737518, "loss_total_v6": 0.5796511545777321, "step": 5156 }, { "epoch": 1.1893477477477477, "grad_norm": 29.159303665161133, "learning_rate": 2.9350891936570157e-06, "loss": 9.677274703979492, "loss_alignment": 0.7978515625, "loss_alignment_w": 0.39892578125, "loss_sft": 0.20665154233574867, "loss_total_v6": 0.6048296317458153, "step": 5157 }, { "epoch": 1.1895783783783784, "grad_norm": 32.592063903808594, "learning_rate": 2.933255277020108e-06, "loss": 9.79463005065918, "loss_alignment": 0.7939453125, "loss_alignment_w": 0.39697265625, "loss_sft": 0.21455085463821888, "loss_total_v6": 0.6121643930673599, "step": 5158 }, { "epoch": 1.189809009009009, "grad_norm": 35.232513427734375, "learning_rate": 2.9314216956552677e-06, "loss": 9.884413719177246, "loss_alignment": 0.85986328125, "loss_alignment_w": 0.429931640625, "loss_sft": 0.18776784650981426, "loss_total_v6": 0.6177758276462555, "step": 5159 }, { "epoch": 1.1900396396396395, "grad_norm": 26.418569564819336, "learning_rate": 2.9295884498599415e-06, "loss": 9.198261260986328, "loss_alignment": 0.80810546875, "loss_alignment_w": 0.404052734375, "loss_sft": 0.1705639660358429, "loss_total_v6": 0.5748913660645485, "step": 5160 }, { "epoch": 1.1902702702702703, "grad_norm": 28.039396286010742, "learning_rate": 2.9277555399315214e-06, "loss": 9.392343521118164, "loss_alignment": 0.80810546875, "loss_alignment_w": 0.404052734375, "loss_sft": 0.18301446922123432, "loss_total_v6": 0.5870214551687241, "step": 5161 }, { "epoch": 1.1905009009009009, "grad_norm": 37.18626022338867, "learning_rate": 2.925922966167349e-06, "loss": 10.12997817993164, "loss_alignment": 0.8017578125, "loss_alignment_w": 0.40087890625, "loss_sft": 0.23181750252842903, "loss_total_v6": 0.6331236511468887, "step": 5162 }, { "epoch": 1.1907315315315314, "grad_norm": 34.69364929199219, "learning_rate": 2.9240907288647087e-06, "loss": 9.421075820922852, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.2048756293952465, "loss_total_v6": 0.5888172686100006, "step": 5163 }, { "epoch": 1.1909621621621622, "grad_norm": 32.20720291137695, "learning_rate": 2.9222588283208274e-06, "loss": 9.730890274047852, "loss_alignment": 0.78515625, "loss_alignment_w": 0.392578125, "loss_sft": 0.21583144180476665, "loss_total_v6": 0.6081806719303131, "step": 5164 }, { "epoch": 1.1911927927927928, "grad_norm": 27.581253051757812, "learning_rate": 2.9204272648328835e-06, "loss": 9.73208236694336, "loss_alignment": 0.80517578125, "loss_alignment_w": 0.402587890625, "loss_sft": 0.20638443902134895, "loss_total_v6": 0.6082551404833794, "step": 5165 }, { "epoch": 1.1914234234234233, "grad_norm": 29.85249137878418, "learning_rate": 2.918596038697995e-06, "loss": 9.634832382202148, "loss_alignment": 0.82861328125, "loss_alignment_w": 0.414306640625, "loss_sft": 0.18681753426790237, "loss_total_v6": 0.6021770387887955, "step": 5166 }, { "epoch": 1.1916540540540541, "grad_norm": 34.836978912353516, "learning_rate": 2.916765150213227e-06, "loss": 9.492110252380371, "loss_alignment": 0.8056640625, "loss_alignment_w": 0.40283203125, "loss_sft": 0.1904248120263219, "loss_total_v6": 0.5932568609714508, "step": 5167 }, { "epoch": 1.1918846846846847, "grad_norm": 31.447629928588867, "learning_rate": 2.914934599675594e-06, "loss": 8.971261978149414, "loss_alignment": 0.76220703125, "loss_alignment_w": 0.381103515625, "loss_sft": 0.18042433355003595, "loss_total_v6": 0.5607038587331772, "step": 5168 }, { "epoch": 1.1921153153153152, "grad_norm": 29.90482521057129, "learning_rate": 2.9131043873820504e-06, "loss": 9.176718711853027, "loss_alignment": 0.79638671875, "loss_alignment_w": 0.398193359375, "loss_sft": 0.17513794545084238, "loss_total_v6": 0.5735449343919754, "step": 5169 }, { "epoch": 1.192345945945946, "grad_norm": 29.169780731201172, "learning_rate": 2.911274513629497e-06, "loss": 9.504847526550293, "loss_alignment": 0.7841796875, "loss_alignment_w": 0.39208984375, "loss_sft": 0.20242087915539742, "loss_total_v6": 0.5940529555082321, "step": 5170 }, { "epoch": 1.1925765765765766, "grad_norm": 30.139028549194336, "learning_rate": 2.909444978714777e-06, "loss": 9.283533096313477, "loss_alignment": 0.7578125, "loss_alignment_w": 0.37890625, "loss_sft": 0.20149763487279415, "loss_total_v6": 0.5802207961678505, "step": 5171 }, { "epoch": 1.1928072072072071, "grad_norm": 34.12568283081055, "learning_rate": 2.9076157829346883e-06, "loss": 9.76829719543457, "loss_alignment": 0.8134765625, "loss_alignment_w": 0.40673828125, "loss_sft": 0.2033683229237795, "loss_total_v6": 0.6105185970664024, "step": 5172 }, { "epoch": 1.193037837837838, "grad_norm": 30.995630264282227, "learning_rate": 2.9057869265859604e-06, "loss": 9.76204776763916, "loss_alignment": 0.818359375, "loss_alignment_w": 0.4091796875, "loss_sft": 0.20117721892893314, "loss_total_v6": 0.6101280078291893, "step": 5173 }, { "epoch": 1.1932684684684685, "grad_norm": 29.37936019897461, "learning_rate": 2.9039584099652816e-06, "loss": 9.715599060058594, "loss_alignment": 0.7998046875, "loss_alignment_w": 0.39990234375, "loss_sft": 0.20703275315463543, "loss_total_v6": 0.6072249934077263, "step": 5174 }, { "epoch": 1.193499099099099, "grad_norm": 31.313011169433594, "learning_rate": 2.9021302333692738e-06, "loss": 9.274873733520508, "loss_alignment": 0.7763671875, "loss_alignment_w": 0.38818359375, "loss_sft": 0.1916790921241045, "loss_total_v6": 0.5796796157956123, "step": 5175 }, { "epoch": 1.1937297297297298, "grad_norm": 28.43672752380371, "learning_rate": 2.900302397094506e-06, "loss": 9.144545555114746, "loss_alignment": 0.80224609375, "loss_alignment_w": 0.401123046875, "loss_sft": 0.1704872939735651, "loss_total_v6": 0.5715340748429298, "step": 5176 }, { "epoch": 1.1939603603603604, "grad_norm": 25.992048263549805, "learning_rate": 2.898474901437499e-06, "loss": 9.767401695251465, "loss_alignment": 0.7880859375, "loss_alignment_w": 0.39404296875, "loss_sft": 0.21660278737545013, "loss_total_v6": 0.6104626208543777, "step": 5177 }, { "epoch": 1.194190990990991, "grad_norm": 31.86124610900879, "learning_rate": 2.8966477466947107e-06, "loss": 9.715877532958984, "loss_alignment": 0.7900390625, "loss_alignment_w": 0.39501953125, "loss_sft": 0.2131231315433979, "loss_total_v6": 0.6072423830628395, "step": 5178 }, { "epoch": 1.1944216216216217, "grad_norm": 25.265504837036133, "learning_rate": 2.8948209331625454e-06, "loss": 9.26826000213623, "loss_alignment": 0.79638671875, "loss_alignment_w": 0.398193359375, "loss_sft": 0.18099659867584705, "loss_total_v6": 0.5792662501335144, "step": 5179 }, { "epoch": 1.1946522522522522, "grad_norm": 30.59939956665039, "learning_rate": 2.8929944611373555e-06, "loss": 9.588546752929688, "loss_alignment": 0.771484375, "loss_alignment_w": 0.3857421875, "loss_sft": 0.21218395605683327, "loss_total_v6": 0.5992841795086861, "step": 5180 }, { "epoch": 1.1948828828828828, "grad_norm": 30.212989807128906, "learning_rate": 2.8911683309154326e-06, "loss": 9.356033325195312, "loss_alignment": 0.76953125, "loss_alignment_w": 0.384765625, "loss_sft": 0.19991019368171692, "loss_total_v6": 0.5847520902752876, "step": 5181 }, { "epoch": 1.1951135135135136, "grad_norm": 29.806413650512695, "learning_rate": 2.889342542793019e-06, "loss": 9.379342079162598, "loss_alignment": 0.814453125, "loss_alignment_w": 0.4072265625, "loss_sft": 0.17933323606848717, "loss_total_v6": 0.5862088650465012, "step": 5182 }, { "epoch": 1.1953441441441441, "grad_norm": 23.69651985168457, "learning_rate": 2.8875170970662964e-06, "loss": 9.70376205444336, "loss_alignment": 0.8134765625, "loss_alignment_w": 0.40673828125, "loss_sft": 0.2001588884741068, "loss_total_v6": 0.6064851880073547, "step": 5183 }, { "epoch": 1.1955747747747747, "grad_norm": 29.0710391998291, "learning_rate": 2.8856919940313933e-06, "loss": 9.95199203491211, "loss_alignment": 0.8359375, "loss_alignment_w": 0.41796875, "loss_sft": 0.2027643583714962, "loss_total_v6": 0.621999554336071, "step": 5184 }, { "epoch": 1.1958054054054055, "grad_norm": 29.14024543762207, "learning_rate": 2.8838672339843786e-06, "loss": 9.909982681274414, "loss_alignment": 0.81103515625, "loss_alignment_w": 0.405517578125, "loss_sft": 0.21327651292085648, "loss_total_v6": 0.6193739101290703, "step": 5185 }, { "epoch": 1.196036036036036, "grad_norm": 26.885576248168945, "learning_rate": 2.8820428172212733e-06, "loss": 9.10055923461914, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.18090649787336588, "loss_total_v6": 0.5687849447131157, "step": 5186 }, { "epoch": 1.1962666666666666, "grad_norm": 29.396102905273438, "learning_rate": 2.880218744038039e-06, "loss": 9.631562232971191, "loss_alignment": 0.81298828125, "loss_alignment_w": 0.406494140625, "loss_sft": 0.19653135910630226, "loss_total_v6": 0.6019726395606995, "step": 5187 }, { "epoch": 1.1964972972972974, "grad_norm": 30.684261322021484, "learning_rate": 2.8783950147305796e-06, "loss": 9.872550964355469, "loss_alignment": 0.83642578125, "loss_alignment_w": 0.418212890625, "loss_sft": 0.19786028936505318, "loss_total_v6": 0.6170344725251198, "step": 5188 }, { "epoch": 1.196727927927928, "grad_norm": 29.82796859741211, "learning_rate": 2.8765716295947438e-06, "loss": 9.750971794128418, "loss_alignment": 0.81298828125, "loss_alignment_w": 0.406494140625, "loss_sft": 0.20466585457324982, "loss_total_v6": 0.6094357520341873, "step": 5189 }, { "epoch": 1.1969585585585585, "grad_norm": 35.359066009521484, "learning_rate": 2.8747485889263244e-06, "loss": 9.119792938232422, "loss_alignment": 0.837890625, "loss_alignment_w": 0.4189453125, "loss_sft": 0.15171314496546984, "loss_total_v6": 0.5699870586395264, "step": 5190 }, { "epoch": 1.1971891891891893, "grad_norm": 30.85201644897461, "learning_rate": 2.8729258930210626e-06, "loss": 9.935999870300293, "loss_alignment": 0.84375, "loss_alignment_w": 0.421875, "loss_sft": 0.20020836405456066, "loss_total_v6": 0.6209999769926071, "step": 5191 }, { "epoch": 1.1974198198198198, "grad_norm": 31.49979019165039, "learning_rate": 2.871103542174637e-06, "loss": 9.806055068969727, "loss_alignment": 0.81103515625, "loss_alignment_w": 0.405517578125, "loss_sft": 0.20658272225409746, "loss_total_v6": 0.6128784641623497, "step": 5192 }, { "epoch": 1.1976504504504504, "grad_norm": 30.578243255615234, "learning_rate": 2.869281536682676e-06, "loss": 9.553581237792969, "loss_alignment": 0.80029296875, "loss_alignment_w": 0.400146484375, "loss_sft": 0.19626569375395775, "loss_total_v6": 0.59709881991148, "step": 5193 }, { "epoch": 1.1978810810810812, "grad_norm": 29.222612380981445, "learning_rate": 2.867459876840749e-06, "loss": 9.389351844787598, "loss_alignment": 0.787109375, "loss_alignment_w": 0.3935546875, "loss_sft": 0.19292890653014183, "loss_total_v6": 0.5868345163762569, "step": 5194 }, { "epoch": 1.1981117117117117, "grad_norm": 29.4608154296875, "learning_rate": 2.8656385629443694e-06, "loss": 9.354708671569824, "loss_alignment": 0.775390625, "loss_alignment_w": 0.3876953125, "loss_sft": 0.19851514138281345, "loss_total_v6": 0.5846693068742752, "step": 5195 }, { "epoch": 1.1983423423423423, "grad_norm": 27.601171493530273, "learning_rate": 2.863817595288993e-06, "loss": 9.777196884155273, "loss_alignment": 0.79052734375, "loss_alignment_w": 0.395263671875, "loss_sft": 0.21631464175879955, "loss_total_v6": 0.6110747754573822, "step": 5196 }, { "epoch": 1.198572972972973, "grad_norm": 25.525066375732422, "learning_rate": 2.861996974170025e-06, "loss": 9.246415138244629, "loss_alignment": 0.763671875, "loss_alignment_w": 0.3818359375, "loss_sft": 0.19530207477509975, "loss_total_v6": 0.5779009610414505, "step": 5197 }, { "epoch": 1.1988036036036036, "grad_norm": 28.320308685302734, "learning_rate": 2.8601766998828074e-06, "loss": 9.507488250732422, "loss_alignment": 0.78857421875, "loss_alignment_w": 0.394287109375, "loss_sft": 0.2004649732261896, "loss_total_v6": 0.5942180380225182, "step": 5198 }, { "epoch": 1.1990342342342342, "grad_norm": 27.374210357666016, "learning_rate": 2.858356772722632e-06, "loss": 9.335247039794922, "loss_alignment": 0.81005859375, "loss_alignment_w": 0.405029296875, "loss_sft": 0.17859148513525724, "loss_total_v6": 0.5834529250860214, "step": 5199 }, { "epoch": 1.199264864864865, "grad_norm": 28.35289764404297, "learning_rate": 2.8565371929847286e-06, "loss": 9.655835151672363, "loss_alignment": 0.78564453125, "loss_alignment_w": 0.392822265625, "loss_sft": 0.21053009293973446, "loss_total_v6": 0.6034896895289421, "step": 5200 }, { "epoch": 1.1994954954954955, "grad_norm": 35.199745178222656, "learning_rate": 2.854717960964277e-06, "loss": 9.609784126281738, "loss_alignment": 0.822265625, "loss_alignment_w": 0.4111328125, "loss_sft": 0.19031797535717487, "loss_total_v6": 0.6006115227937698, "step": 5201 }, { "epoch": 1.199726126126126, "grad_norm": 30.93567657470703, "learning_rate": 2.852899076956397e-06, "loss": 9.311970710754395, "loss_alignment": 0.77978515625, "loss_alignment_w": 0.389892578125, "loss_sft": 0.19119007140398026, "loss_total_v6": 0.5819981843233109, "step": 5202 }, { "epoch": 1.1999567567567568, "grad_norm": 28.827280044555664, "learning_rate": 2.851080541256147e-06, "loss": 9.195733070373535, "loss_alignment": 0.79150390625, "loss_alignment_w": 0.395751953125, "loss_sft": 0.1795306857675314, "loss_total_v6": 0.5747333317995071, "step": 5203 }, { "epoch": 1.2001873873873874, "grad_norm": 29.15018081665039, "learning_rate": 2.849262354158541e-06, "loss": 9.316740989685059, "loss_alignment": 0.767578125, "loss_alignment_w": 0.3837890625, "loss_sft": 0.19985003769397736, "loss_total_v6": 0.582296296954155, "step": 5204 }, { "epoch": 1.200418018018018, "grad_norm": 25.351585388183594, "learning_rate": 2.8474445159585235e-06, "loss": 9.363476753234863, "loss_alignment": 0.802734375, "loss_alignment_w": 0.4013671875, "loss_sft": 0.1832397449761629, "loss_total_v6": 0.585217297077179, "step": 5205 }, { "epoch": 1.2006486486486487, "grad_norm": 31.63669204711914, "learning_rate": 2.8456270269509934e-06, "loss": 9.577527046203613, "loss_alignment": 0.7734375, "loss_alignment_w": 0.38671875, "loss_sft": 0.21172411553561687, "loss_total_v6": 0.598595455288887, "step": 5206 }, { "epoch": 1.2008792792792793, "grad_norm": 29.348405838012695, "learning_rate": 2.843809887430785e-06, "loss": 9.547479629516602, "loss_alignment": 0.8251953125, "loss_alignment_w": 0.41259765625, "loss_sft": 0.18378412164747715, "loss_total_v6": 0.5967174619436264, "step": 5207 }, { "epoch": 1.2011099099099098, "grad_norm": 28.050724029541016, "learning_rate": 2.84199309769268e-06, "loss": 9.487464904785156, "loss_alignment": 0.78857421875, "loss_alignment_w": 0.394287109375, "loss_sft": 0.19791650213301182, "loss_total_v6": 0.5929665639996529, "step": 5208 }, { "epoch": 1.2013405405405406, "grad_norm": 30.911903381347656, "learning_rate": 2.8401766580313993e-06, "loss": 9.50745964050293, "loss_alignment": 0.80615234375, "loss_alignment_w": 0.403076171875, "loss_sft": 0.19107901491224766, "loss_total_v6": 0.5942162126302719, "step": 5209 }, { "epoch": 1.2015711711711712, "grad_norm": 28.969680786132812, "learning_rate": 2.838360568741613e-06, "loss": 9.111438751220703, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.1821358222514391, "loss_total_v6": 0.5694649294018745, "step": 5210 }, { "epoch": 1.2018018018018017, "grad_norm": 29.32398223876953, "learning_rate": 2.836544830117932e-06, "loss": 9.200061798095703, "loss_alignment": 0.7841796875, "loss_alignment_w": 0.39208984375, "loss_sft": 0.18335654214024544, "loss_total_v6": 0.5750038847327232, "step": 5211 }, { "epoch": 1.2020324324324325, "grad_norm": 26.30609703063965, "learning_rate": 2.8347294424549075e-06, "loss": 8.529836654663086, "loss_alignment": 0.7431640625, "loss_alignment_w": 0.37158203125, "loss_sft": 0.16167009249329567, "loss_total_v6": 0.5331148132681847, "step": 5212 }, { "epoch": 1.202263063063063, "grad_norm": 33.47696304321289, "learning_rate": 2.832914406047037e-06, "loss": 9.672889709472656, "loss_alignment": 0.78466796875, "loss_alignment_w": 0.392333984375, "loss_sft": 0.2115655317902565, "loss_total_v6": 0.6045556366443634, "step": 5213 }, { "epoch": 1.2024936936936936, "grad_norm": 27.31724739074707, "learning_rate": 2.831099721188759e-06, "loss": 9.697120666503906, "loss_alignment": 0.77978515625, "loss_alignment_w": 0.389892578125, "loss_sft": 0.2147125918418169, "loss_total_v6": 0.6060700193047523, "step": 5214 }, { "epoch": 1.2027243243243244, "grad_norm": 28.029157638549805, "learning_rate": 2.8292853881744554e-06, "loss": 9.475621223449707, "loss_alignment": 0.77734375, "loss_alignment_w": 0.388671875, "loss_sft": 0.20457679592072964, "loss_total_v6": 0.5922263190150261, "step": 5215 }, { "epoch": 1.202954954954955, "grad_norm": 27.658279418945312, "learning_rate": 2.827471407298451e-06, "loss": 9.068777084350586, "loss_alignment": 0.787109375, "loss_alignment_w": 0.3935546875, "loss_sft": 0.1749986633658409, "loss_total_v6": 0.5667986050248146, "step": 5216 }, { "epoch": 1.2031855855855855, "grad_norm": 28.103361129760742, "learning_rate": 2.8256577788550178e-06, "loss": 9.460875511169434, "loss_alignment": 0.7705078125, "loss_alignment_w": 0.38525390625, "loss_sft": 0.2059440016746521, "loss_total_v6": 0.591304711997509, "step": 5217 }, { "epoch": 1.2034162162162163, "grad_norm": 26.641714096069336, "learning_rate": 2.8238445031383634e-06, "loss": 9.740217208862305, "loss_alignment": 0.78076171875, "loss_alignment_w": 0.390380859375, "loss_sft": 0.21813854482024908, "loss_total_v6": 0.6087635606527328, "step": 5218 }, { "epoch": 1.2036468468468469, "grad_norm": 28.63962173461914, "learning_rate": 2.8220315804426425e-06, "loss": 9.814037322998047, "loss_alignment": 0.82958984375, "loss_alignment_w": 0.414794921875, "loss_sft": 0.19766687415540218, "loss_total_v6": 0.6133773326873779, "step": 5219 }, { "epoch": 1.2038774774774774, "grad_norm": 27.80547332763672, "learning_rate": 2.820219011061949e-06, "loss": 9.618642807006836, "loss_alignment": 0.82958984375, "loss_alignment_w": 0.414794921875, "loss_sft": 0.18684331979602575, "loss_total_v6": 0.6011651903390884, "step": 5220 }, { "epoch": 1.2041081081081082, "grad_norm": 26.080324172973633, "learning_rate": 2.818406795290327e-06, "loss": 9.323423385620117, "loss_alignment": 0.79736328125, "loss_alignment_w": 0.398681640625, "loss_sft": 0.1848258040845394, "loss_total_v6": 0.5827139765024185, "step": 5221 }, { "epoch": 1.2043387387387388, "grad_norm": 27.107187271118164, "learning_rate": 2.8165949334217534e-06, "loss": 9.289514541625977, "loss_alignment": 0.78466796875, "loss_alignment_w": 0.392333984375, "loss_sft": 0.18783338740468025, "loss_total_v6": 0.5805946290493011, "step": 5222 }, { "epoch": 1.2045693693693693, "grad_norm": 29.911096572875977, "learning_rate": 2.814783425750157e-06, "loss": 9.003608703613281, "loss_alignment": 0.76123046875, "loss_alignment_w": 0.380615234375, "loss_sft": 0.18233916722238064, "loss_total_v6": 0.5627254992723465, "step": 5223 }, { "epoch": 1.2048, "grad_norm": 28.34607696533203, "learning_rate": 2.8129722725694024e-06, "loss": 9.58365249633789, "loss_alignment": 0.8203125, "loss_alignment_w": 0.41015625, "loss_sft": 0.18805908411741257, "loss_total_v6": 0.598978266119957, "step": 5224 }, { "epoch": 1.2050306306306307, "grad_norm": 31.73677635192871, "learning_rate": 2.8111614741732975e-06, "loss": 9.394550323486328, "loss_alignment": 0.81640625, "loss_alignment_w": 0.408203125, "loss_sft": 0.17837642692029476, "loss_total_v6": 0.5871593877673149, "step": 5225 }, { "epoch": 1.2052612612612612, "grad_norm": 31.545503616333008, "learning_rate": 2.8093510308555975e-06, "loss": 9.305418014526367, "loss_alignment": 0.7841796875, "loss_alignment_w": 0.39208984375, "loss_sft": 0.18908682838082314, "loss_total_v6": 0.5815886631608009, "step": 5226 }, { "epoch": 1.205491891891892, "grad_norm": 32.07501220703125, "learning_rate": 2.8075409429099953e-06, "loss": 9.232015609741211, "loss_alignment": 0.7919921875, "loss_alignment_w": 0.39599609375, "loss_sft": 0.18115746974945068, "loss_total_v6": 0.5770009756088257, "step": 5227 }, { "epoch": 1.2057225225225225, "grad_norm": 28.6357364654541, "learning_rate": 2.8057312106301256e-06, "loss": 9.184486389160156, "loss_alignment": 0.78173828125, "loss_alignment_w": 0.390869140625, "loss_sft": 0.1827950607985258, "loss_total_v6": 0.5740304067730904, "step": 5228 }, { "epoch": 1.205953153153153, "grad_norm": 33.38047409057617, "learning_rate": 2.8039218343095704e-06, "loss": 9.581357955932617, "loss_alignment": 0.81787109375, "loss_alignment_w": 0.408935546875, "loss_sft": 0.18968570791184902, "loss_total_v6": 0.5988348796963692, "step": 5229 }, { "epoch": 1.2061837837837839, "grad_norm": 28.7991943359375, "learning_rate": 2.8021128142418473e-06, "loss": 8.973464965820312, "loss_alignment": 0.77978515625, "loss_alignment_w": 0.389892578125, "loss_sft": 0.1710558058694005, "loss_total_v6": 0.5608415603637695, "step": 5230 }, { "epoch": 1.2064144144144144, "grad_norm": 33.143287658691406, "learning_rate": 2.800304150720424e-06, "loss": 9.311824798583984, "loss_alignment": 0.765625, "loss_alignment_w": 0.3828125, "loss_sft": 0.19871875829994678, "loss_total_v6": 0.5819890275597572, "step": 5231 }, { "epoch": 1.206645045045045, "grad_norm": 30.755165100097656, "learning_rate": 2.7984958440387045e-06, "loss": 9.439207077026367, "loss_alignment": 0.78564453125, "loss_alignment_w": 0.392822265625, "loss_sft": 0.1972503326833248, "loss_total_v6": 0.5899504944682121, "step": 5232 }, { "epoch": 1.2068756756756756, "grad_norm": 30.745323181152344, "learning_rate": 2.7966878944900354e-06, "loss": 10.051010131835938, "loss_alignment": 0.79443359375, "loss_alignment_w": 0.397216796875, "loss_sft": 0.2302846796810627, "loss_total_v6": 0.6281881257891655, "step": 5233 }, { "epoch": 1.2071063063063063, "grad_norm": 28.859342575073242, "learning_rate": 2.794880302367705e-06, "loss": 10.017276763916016, "loss_alignment": 0.814453125, "loss_alignment_w": 0.4072265625, "loss_sft": 0.21921944804489613, "loss_total_v6": 0.6260797902941704, "step": 5234 }, { "epoch": 1.2073369369369369, "grad_norm": 29.037246704101562, "learning_rate": 2.7930730679649465e-06, "loss": 9.204643249511719, "loss_alignment": 0.7939453125, "loss_alignment_w": 0.39697265625, "loss_sft": 0.1783022591844201, "loss_total_v6": 0.57529017329216, "step": 5235 }, { "epoch": 1.2075675675675677, "grad_norm": 32.2974853515625, "learning_rate": 2.791266191574936e-06, "loss": 9.18001937866211, "loss_alignment": 0.75439453125, "loss_alignment_w": 0.377197265625, "loss_sft": 0.19644707068800926, "loss_total_v6": 0.5737511739134789, "step": 5236 }, { "epoch": 1.2077981981981982, "grad_norm": 27.5631046295166, "learning_rate": 2.789459673490788e-06, "loss": 9.672492027282715, "loss_alignment": 0.81982421875, "loss_alignment_w": 0.409912109375, "loss_sft": 0.19561044499278069, "loss_total_v6": 0.6045307293534279, "step": 5237 }, { "epoch": 1.2080288288288288, "grad_norm": 29.47732925415039, "learning_rate": 2.7876535140055584e-06, "loss": 9.682601928710938, "loss_alignment": 0.7919921875, "loss_alignment_w": 0.39599609375, "loss_sft": 0.2095479741692543, "loss_total_v6": 0.6051625832915306, "step": 5238 }, { "epoch": 1.2082594594594593, "grad_norm": 31.254806518554688, "learning_rate": 2.7858477134122474e-06, "loss": 9.500326156616211, "loss_alignment": 0.8095703125, "loss_alignment_w": 0.40478515625, "loss_sft": 0.19040434807538986, "loss_total_v6": 0.5937704294919968, "step": 5239 }, { "epoch": 1.2084900900900901, "grad_norm": 28.07364273071289, "learning_rate": 2.7840422720037943e-06, "loss": 9.689220428466797, "loss_alignment": 0.78857421875, "loss_alignment_w": 0.394287109375, "loss_sft": 0.21193000860512257, "loss_total_v6": 0.605576254427433, "step": 5240 }, { "epoch": 1.2087207207207207, "grad_norm": 29.176843643188477, "learning_rate": 2.782237190073083e-06, "loss": 9.954368591308594, "loss_alignment": 0.78955078125, "loss_alignment_w": 0.394775390625, "loss_sft": 0.2267012819647789, "loss_total_v6": 0.6221480518579483, "step": 5241 }, { "epoch": 1.2089513513513515, "grad_norm": 30.903043746948242, "learning_rate": 2.7804324679129412e-06, "loss": 9.409053802490234, "loss_alignment": 0.78662109375, "loss_alignment_w": 0.393310546875, "loss_sft": 0.1950299944728613, "loss_total_v6": 0.5880658701062202, "step": 5242 }, { "epoch": 1.209181981981982, "grad_norm": 33.68586730957031, "learning_rate": 2.7786281058161325e-06, "loss": 9.231729507446289, "loss_alignment": 0.7724609375, "loss_alignment_w": 0.38623046875, "loss_sft": 0.19214115291833878, "loss_total_v6": 0.5769830867648125, "step": 5243 }, { "epoch": 1.2094126126126126, "grad_norm": 29.974992752075195, "learning_rate": 2.776824104075364e-06, "loss": 8.927370071411133, "loss_alignment": 0.79052734375, "loss_alignment_w": 0.395263671875, "loss_sft": 0.16329205967485905, "loss_total_v6": 0.557960644364357, "step": 5244 }, { "epoch": 1.2096432432432431, "grad_norm": 29.293272018432617, "learning_rate": 2.7750204629832846e-06, "loss": 9.433307647705078, "loss_alignment": 0.79443359375, "loss_alignment_w": 0.397216796875, "loss_sft": 0.1924869604408741, "loss_total_v6": 0.5895816832780838, "step": 5245 }, { "epoch": 1.209873873873874, "grad_norm": 33.42106628417969, "learning_rate": 2.773217182832487e-06, "loss": 9.316554069519043, "loss_alignment": 0.80615234375, "loss_alignment_w": 0.403076171875, "loss_sft": 0.1796662174165249, "loss_total_v6": 0.5822846367955208, "step": 5246 }, { "epoch": 1.2101045045045045, "grad_norm": 32.489158630371094, "learning_rate": 2.7714142639155016e-06, "loss": 9.164960861206055, "loss_alignment": 0.7607421875, "loss_alignment_w": 0.38037109375, "loss_sft": 0.1936291754245758, "loss_total_v6": 0.5728100910782814, "step": 5247 }, { "epoch": 1.2103351351351352, "grad_norm": 29.46417808532715, "learning_rate": 2.769611706524805e-06, "loss": 9.07991886138916, "loss_alignment": 0.77294921875, "loss_alignment_w": 0.386474609375, "loss_sft": 0.18068465404212475, "loss_total_v6": 0.5674949586391449, "step": 5248 }, { "epoch": 1.2105657657657658, "grad_norm": 29.257387161254883, "learning_rate": 2.767809510952808e-06, "loss": 9.794137954711914, "loss_alignment": 0.79443359375, "loss_alignment_w": 0.397216796875, "loss_sft": 0.21458114869892597, "loss_total_v6": 0.6121336072683334, "step": 5249 }, { "epoch": 1.2107963963963964, "grad_norm": 28.660383224487305, "learning_rate": 2.766007677491871e-06, "loss": 9.249972343444824, "loss_alignment": 0.7763671875, "loss_alignment_w": 0.38818359375, "loss_sft": 0.18969552777707577, "loss_total_v6": 0.5781232565641403, "step": 5250 }, { "epoch": 1.211027027027027, "grad_norm": 37.050514221191406, "learning_rate": 2.76420620643429e-06, "loss": 9.153602600097656, "loss_alignment": 0.79638671875, "loss_alignment_w": 0.398193359375, "loss_sft": 0.1748986067250371, "loss_total_v6": 0.5721001401543617, "step": 5251 }, { "epoch": 1.2112576576576577, "grad_norm": 25.689908981323242, "learning_rate": 2.7624050980723032e-06, "loss": 9.65587043762207, "loss_alignment": 0.80810546875, "loss_alignment_w": 0.404052734375, "loss_sft": 0.20004946365952492, "loss_total_v6": 0.6034918501973152, "step": 5252 }, { "epoch": 1.2114882882882883, "grad_norm": 30.868488311767578, "learning_rate": 2.7606043526980886e-06, "loss": 9.466581344604492, "loss_alignment": 0.7763671875, "loss_alignment_w": 0.38818359375, "loss_sft": 0.20419489592313766, "loss_total_v6": 0.5916613340377808, "step": 5253 }, { "epoch": 1.211718918918919, "grad_norm": 30.675403594970703, "learning_rate": 2.75880397060377e-06, "loss": 9.424505233764648, "loss_alignment": 0.79541015625, "loss_alignment_w": 0.397705078125, "loss_sft": 0.19202840141952038, "loss_total_v6": 0.5890315845608711, "step": 5254 }, { "epoch": 1.2119495495495496, "grad_norm": 27.047212600708008, "learning_rate": 2.757003952081411e-06, "loss": 9.689844131469727, "loss_alignment": 0.78515625, "loss_alignment_w": 0.392578125, "loss_sft": 0.21152651123702526, "loss_total_v6": 0.6056152731180191, "step": 5255 }, { "epoch": 1.2121801801801801, "grad_norm": 28.03197479248047, "learning_rate": 2.755204297423012e-06, "loss": 8.89541244506836, "loss_alignment": 0.79345703125, "loss_alignment_w": 0.396728515625, "loss_sft": 0.15923475287854671, "loss_total_v6": 0.5559632927179337, "step": 5256 }, { "epoch": 1.2124108108108107, "grad_norm": 31.938186645507812, "learning_rate": 2.753405006920518e-06, "loss": 9.438992500305176, "loss_alignment": 0.76953125, "loss_alignment_w": 0.384765625, "loss_sft": 0.20584281347692013, "loss_total_v6": 0.5899370387196541, "step": 5257 }, { "epoch": 1.2126414414414415, "grad_norm": 29.03769302368164, "learning_rate": 2.751606080865814e-06, "loss": 9.855031967163086, "loss_alignment": 0.7958984375, "loss_alignment_w": 0.39794921875, "loss_sft": 0.21855482272803783, "loss_total_v6": 0.6159394755959511, "step": 5258 }, { "epoch": 1.212872072072072, "grad_norm": 28.28386688232422, "learning_rate": 2.7498075195507236e-06, "loss": 8.914934158325195, "loss_alignment": 0.744140625, "loss_alignment_w": 0.3720703125, "loss_sft": 0.18608958553522825, "loss_total_v6": 0.5571833401918411, "step": 5259 }, { "epoch": 1.2131027027027028, "grad_norm": 29.77192497253418, "learning_rate": 2.748009323267016e-06, "loss": 8.944998741149902, "loss_alignment": 0.791015625, "loss_alignment_w": 0.3955078125, "loss_sft": 0.1630968665704131, "loss_total_v6": 0.5590624511241913, "step": 5260 }, { "epoch": 1.2133333333333334, "grad_norm": 28.040685653686523, "learning_rate": 2.7462114923064e-06, "loss": 9.617058753967285, "loss_alignment": 0.7724609375, "loss_alignment_w": 0.38623046875, "loss_sft": 0.21582753956317902, "loss_total_v6": 0.6010661870241165, "step": 5261 }, { "epoch": 1.213563963963964, "grad_norm": 29.732311248779297, "learning_rate": 2.744414026960522e-06, "loss": 9.347806930541992, "loss_alignment": 0.779296875, "loss_alignment_w": 0.3896484375, "loss_sft": 0.19471161626279354, "loss_total_v6": 0.5842379778623581, "step": 5262 }, { "epoch": 1.2137945945945945, "grad_norm": 29.882122039794922, "learning_rate": 2.7426169275209713e-06, "loss": 9.854280471801758, "loss_alignment": 0.80419921875, "loss_alignment_w": 0.402099609375, "loss_sft": 0.21368610300123692, "loss_total_v6": 0.6158925145864487, "step": 5263 }, { "epoch": 1.2140252252252253, "grad_norm": 27.736373901367188, "learning_rate": 2.7408201942792755e-06, "loss": 9.618709564208984, "loss_alignment": 0.78662109375, "loss_alignment_w": 0.393310546875, "loss_sft": 0.20903368294239044, "loss_total_v6": 0.6011692956089973, "step": 5264 }, { "epoch": 1.2142558558558558, "grad_norm": 28.25923728942871, "learning_rate": 2.739023827526909e-06, "loss": 9.403956413269043, "loss_alignment": 0.8046875, "loss_alignment_w": 0.40234375, "loss_sft": 0.18503731954842806, "loss_total_v6": 0.587747260928154, "step": 5265 }, { "epoch": 1.2144864864864866, "grad_norm": 28.302597045898438, "learning_rate": 2.7372278275552776e-06, "loss": 9.79090404510498, "loss_alignment": 0.77783203125, "loss_alignment_w": 0.388916015625, "loss_sft": 0.22234407626092434, "loss_total_v6": 0.6119314953684807, "step": 5266 }, { "epoch": 1.2147171171171172, "grad_norm": 27.793251037597656, "learning_rate": 2.735432194655737e-06, "loss": 9.861381530761719, "loss_alignment": 0.81005859375, "loss_alignment_w": 0.405029296875, "loss_sft": 0.21213102899491787, "loss_total_v6": 0.6163363456726074, "step": 5267 }, { "epoch": 1.2149477477477477, "grad_norm": 27.8704776763916, "learning_rate": 2.7336369291195773e-06, "loss": 9.303796768188477, "loss_alignment": 0.76171875, "loss_alignment_w": 0.380859375, "loss_sft": 0.20042956620454788, "loss_total_v6": 0.5814872905611992, "step": 5268 }, { "epoch": 1.2151783783783783, "grad_norm": 32.50087356567383, "learning_rate": 2.7318420312380285e-06, "loss": 9.940001487731934, "loss_alignment": 0.8359375, "loss_alignment_w": 0.41796875, "loss_sft": 0.20209116488695145, "loss_total_v6": 0.6212500929832458, "step": 5269 }, { "epoch": 1.215409009009009, "grad_norm": 28.84967803955078, "learning_rate": 2.7300475013022666e-06, "loss": 9.128622055053711, "loss_alignment": 0.79931640625, "loss_alignment_w": 0.399658203125, "loss_sft": 0.1718877125531435, "loss_total_v6": 0.570538841187954, "step": 5270 }, { "epoch": 1.2156396396396396, "grad_norm": 32.530120849609375, "learning_rate": 2.7282533396034026e-06, "loss": 9.475367546081543, "loss_alignment": 0.8349609375, "loss_alignment_w": 0.41748046875, "loss_sft": 0.17512675374746323, "loss_total_v6": 0.5922104939818382, "step": 5271 }, { "epoch": 1.2158702702702704, "grad_norm": 28.711444854736328, "learning_rate": 2.726459546432488e-06, "loss": 9.570684432983398, "loss_alignment": 0.81689453125, "loss_alignment_w": 0.408447265625, "loss_sft": 0.1902240263298154, "loss_total_v6": 0.5981677547097206, "step": 5272 }, { "epoch": 1.216100900900901, "grad_norm": 30.582263946533203, "learning_rate": 2.7246661220805194e-06, "loss": 9.348703384399414, "loss_alignment": 0.7880859375, "loss_alignment_w": 0.39404296875, "loss_sft": 0.19110550917685032, "loss_total_v6": 0.5842939615249634, "step": 5273 }, { "epoch": 1.2163315315315315, "grad_norm": 27.54128646850586, "learning_rate": 2.7228730668384255e-06, "loss": 9.635875701904297, "loss_alignment": 0.794921875, "loss_alignment_w": 0.3974609375, "loss_sft": 0.20566638559103012, "loss_total_v6": 0.6022422984242439, "step": 5274 }, { "epoch": 1.216562162162162, "grad_norm": 31.736478805541992, "learning_rate": 2.7210803809970853e-06, "loss": 9.242477416992188, "loss_alignment": 0.78955078125, "loss_alignment_w": 0.394775390625, "loss_sft": 0.18304729461669922, "loss_total_v6": 0.5776548162102699, "step": 5275 }, { "epoch": 1.2167927927927928, "grad_norm": 29.30035972595215, "learning_rate": 2.71928806484731e-06, "loss": 9.274898529052734, "loss_alignment": 0.7783203125, "loss_alignment_w": 0.38916015625, "loss_sft": 0.19041421823203564, "loss_total_v6": 0.5796811729669571, "step": 5276 }, { "epoch": 1.2170234234234234, "grad_norm": 34.017295837402344, "learning_rate": 2.717496118679853e-06, "loss": 9.321756362915039, "loss_alignment": 0.77587890625, "loss_alignment_w": 0.387939453125, "loss_sft": 0.1943651270121336, "loss_total_v6": 0.5826097726821899, "step": 5277 }, { "epoch": 1.217254054054054, "grad_norm": 33.195777893066406, "learning_rate": 2.7157045427854056e-06, "loss": 9.681900024414062, "loss_alignment": 0.82421875, "loss_alignment_w": 0.412109375, "loss_sft": 0.19322301726788282, "loss_total_v6": 0.6051187515258789, "step": 5278 }, { "epoch": 1.2174846846846847, "grad_norm": 30.643312454223633, "learning_rate": 2.7139133374546047e-06, "loss": 9.02769947052002, "loss_alignment": 0.75341796875, "loss_alignment_w": 0.376708984375, "loss_sft": 0.18834621459245682, "loss_total_v6": 0.5642312243580818, "step": 5279 }, { "epoch": 1.2177153153153153, "grad_norm": 30.951982498168945, "learning_rate": 2.712122502978024e-06, "loss": 9.561949729919434, "loss_alignment": 0.85693359375, "loss_alignment_w": 0.428466796875, "loss_sft": 0.17126077227294445, "loss_total_v6": 0.597621850669384, "step": 5280 }, { "epoch": 1.2179459459459459, "grad_norm": 29.646913528442383, "learning_rate": 2.710332039646175e-06, "loss": 9.85264778137207, "loss_alignment": 0.83056640625, "loss_alignment_w": 0.415283203125, "loss_sft": 0.20162120275199413, "loss_total_v6": 0.6157905086874962, "step": 5281 }, { "epoch": 1.2181765765765766, "grad_norm": 36.97547149658203, "learning_rate": 2.708541947749512e-06, "loss": 9.67940616607666, "loss_alignment": 0.7919921875, "loss_alignment_w": 0.39599609375, "loss_sft": 0.20930243842303753, "loss_total_v6": 0.6049628481268883, "step": 5282 }, { "epoch": 1.2184072072072072, "grad_norm": 28.2856388092041, "learning_rate": 2.7067522275784275e-06, "loss": 9.7652006149292, "loss_alignment": 0.82275390625, "loss_alignment_w": 0.411376953125, "loss_sft": 0.19974153768271208, "loss_total_v6": 0.610325038433075, "step": 5283 }, { "epoch": 1.2186378378378377, "grad_norm": 30.565174102783203, "learning_rate": 2.7049628794232506e-06, "loss": 9.200302124023438, "loss_alignment": 0.80224609375, "loss_alignment_w": 0.401123046875, "loss_sft": 0.17432310245931149, "loss_total_v6": 0.5750189051032066, "step": 5284 }, { "epoch": 1.2188684684684685, "grad_norm": 32.962242126464844, "learning_rate": 2.7031739035742575e-06, "loss": 9.366722106933594, "loss_alignment": 0.7763671875, "loss_alignment_w": 0.38818359375, "loss_sft": 0.19767898507416248, "loss_total_v6": 0.5854200944304466, "step": 5285 }, { "epoch": 1.219099099099099, "grad_norm": 27.739179611206055, "learning_rate": 2.7013853003216605e-06, "loss": 9.639031410217285, "loss_alignment": 0.8173828125, "loss_alignment_w": 0.40869140625, "loss_sft": 0.19390065409243107, "loss_total_v6": 0.6024394780397415, "step": 5286 }, { "epoch": 1.2193297297297296, "grad_norm": 28.750505447387695, "learning_rate": 2.699597069955609e-06, "loss": 9.500123977661133, "loss_alignment": 0.80859375, "loss_alignment_w": 0.404296875, "loss_sft": 0.1899796798825264, "loss_total_v6": 0.5937577672302723, "step": 5287 }, { "epoch": 1.2195603603603604, "grad_norm": 28.256582260131836, "learning_rate": 2.697809212766195e-06, "loss": 9.463974952697754, "loss_alignment": 0.80322265625, "loss_alignment_w": 0.401611328125, "loss_sft": 0.18901737220585346, "loss_total_v6": 0.5914984345436096, "step": 5288 }, { "epoch": 1.219790990990991, "grad_norm": 27.42481231689453, "learning_rate": 2.696021729043445e-06, "loss": 9.597976684570312, "loss_alignment": 0.80908203125, "loss_alignment_w": 0.404541015625, "loss_sft": 0.1943254554644227, "loss_total_v6": 0.5998735502362251, "step": 5289 }, { "epoch": 1.2200216216216215, "grad_norm": 24.346179962158203, "learning_rate": 2.694234619077334e-06, "loss": 9.222990036010742, "loss_alignment": 0.7861328125, "loss_alignment_w": 0.39306640625, "loss_sft": 0.18303471058607101, "loss_total_v6": 0.576436810195446, "step": 5290 }, { "epoch": 1.2202522522522523, "grad_norm": 31.183374404907227, "learning_rate": 2.692447883157767e-06, "loss": 9.896756172180176, "loss_alignment": 0.85107421875, "loss_alignment_w": 0.425537109375, "loss_sft": 0.19319328106939793, "loss_total_v6": 0.6185472831130028, "step": 5291 }, { "epoch": 1.2204828828828829, "grad_norm": 26.904502868652344, "learning_rate": 2.690661521574596e-06, "loss": 9.765058517456055, "loss_alignment": 0.81005859375, "loss_alignment_w": 0.405029296875, "loss_sft": 0.2059430330991745, "loss_total_v6": 0.610316202044487, "step": 5292 }, { "epoch": 1.2207135135135134, "grad_norm": 30.521808624267578, "learning_rate": 2.6888755346176066e-06, "loss": 9.521039962768555, "loss_alignment": 0.81103515625, "loss_alignment_w": 0.405517578125, "loss_sft": 0.1898068655282259, "loss_total_v6": 0.5950650349259377, "step": 5293 }, { "epoch": 1.2209441441441442, "grad_norm": 75.80821228027344, "learning_rate": 2.687089922576524e-06, "loss": 9.63699722290039, "loss_alignment": 0.80615234375, "loss_alignment_w": 0.403076171875, "loss_sft": 0.19932768121361732, "loss_total_v6": 0.6023123115301132, "step": 5294 }, { "epoch": 1.2211747747747748, "grad_norm": 30.385988235473633, "learning_rate": 2.685304685741018e-06, "loss": 9.619251251220703, "loss_alignment": 0.82568359375, "loss_alignment_w": 0.412841796875, "loss_sft": 0.18814778048545122, "loss_total_v6": 0.6012031957507133, "step": 5295 }, { "epoch": 1.2214054054054053, "grad_norm": 29.3596248626709, "learning_rate": 2.683519824400693e-06, "loss": 8.922008514404297, "loss_alignment": 0.75732421875, "loss_alignment_w": 0.378662109375, "loss_sft": 0.17942125163972378, "loss_total_v6": 0.5576255843043327, "step": 5296 }, { "epoch": 1.221636036036036, "grad_norm": 31.59866714477539, "learning_rate": 2.6817353388450896e-06, "loss": 9.260899543762207, "loss_alignment": 0.759765625, "loss_alignment_w": 0.3798828125, "loss_sft": 0.19855720922350883, "loss_total_v6": 0.5788062363862991, "step": 5297 }, { "epoch": 1.2218666666666667, "grad_norm": 26.658903121948242, "learning_rate": 2.679951229363694e-06, "loss": 9.472818374633789, "loss_alignment": 0.8056640625, "loss_alignment_w": 0.40283203125, "loss_sft": 0.1901651481166482, "loss_total_v6": 0.5920511335134506, "step": 5298 }, { "epoch": 1.2220972972972972, "grad_norm": 28.712081909179688, "learning_rate": 2.6781674962459304e-06, "loss": 10.145729064941406, "loss_alignment": 0.82373046875, "loss_alignment_w": 0.411865234375, "loss_sft": 0.22097637690603733, "loss_total_v6": 0.6341080889105797, "step": 5299 }, { "epoch": 1.222327927927928, "grad_norm": 30.869876861572266, "learning_rate": 2.6763841397811576e-06, "loss": 9.671689987182617, "loss_alignment": 0.81005859375, "loss_alignment_w": 0.405029296875, "loss_sft": 0.20036687795072794, "loss_total_v6": 0.6044806614518166, "step": 5300 }, { "epoch": 1.2225585585585586, "grad_norm": 25.879756927490234, "learning_rate": 2.6746011602586764e-06, "loss": 9.866495132446289, "loss_alignment": 0.794921875, "loss_alignment_w": 0.3974609375, "loss_sft": 0.21885927952826023, "loss_total_v6": 0.6166559234261513, "step": 5301 }, { "epoch": 1.222789189189189, "grad_norm": 27.879074096679688, "learning_rate": 2.6728185579677258e-06, "loss": 9.677583694458008, "loss_alignment": 0.79052734375, "loss_alignment_w": 0.395263671875, "loss_sft": 0.2098141759634018, "loss_total_v6": 0.6048489660024643, "step": 5302 }, { "epoch": 1.2230198198198199, "grad_norm": 29.152231216430664, "learning_rate": 2.671036333197482e-06, "loss": 9.655860900878906, "loss_alignment": 0.8154296875, "loss_alignment_w": 0.40771484375, "loss_sft": 0.1945557203143835, "loss_total_v6": 0.6034912616014481, "step": 5303 }, { "epoch": 1.2232504504504504, "grad_norm": 30.303592681884766, "learning_rate": 2.6692544862370624e-06, "loss": 9.671648025512695, "loss_alignment": 0.7978515625, "loss_alignment_w": 0.39892578125, "loss_sft": 0.20449937507510185, "loss_total_v6": 0.6044779866933823, "step": 5304 }, { "epoch": 1.223481081081081, "grad_norm": 44.09015655517578, "learning_rate": 2.6674730173755246e-06, "loss": 10.018577575683594, "loss_alignment": 0.7900390625, "loss_alignment_w": 0.39501953125, "loss_sft": 0.23083635978400707, "loss_total_v6": 0.6261610835790634, "step": 5305 }, { "epoch": 1.2237117117117118, "grad_norm": 28.16429901123047, "learning_rate": 2.665691926901862e-06, "loss": 9.506338119506836, "loss_alignment": 0.779296875, "loss_alignment_w": 0.3896484375, "loss_sft": 0.20452816598117352, "loss_total_v6": 0.5941461101174355, "step": 5306 }, { "epoch": 1.2239423423423423, "grad_norm": 31.677974700927734, "learning_rate": 2.6639112151050056e-06, "loss": 9.028024673461914, "loss_alignment": 0.8232421875, "loss_alignment_w": 0.41162109375, "loss_sft": 0.15333231538534164, "loss_total_v6": 0.5642515160143375, "step": 5307 }, { "epoch": 1.224172972972973, "grad_norm": 32.38252639770508, "learning_rate": 2.662130882273825e-06, "loss": 9.7604341506958, "loss_alignment": 0.8154296875, "loss_alignment_w": 0.40771484375, "loss_sft": 0.20170192886143923, "loss_total_v6": 0.6100271344184875, "step": 5308 }, { "epoch": 1.2244036036036037, "grad_norm": 25.913654327392578, "learning_rate": 2.6603509286971342e-06, "loss": 9.444820404052734, "loss_alignment": 0.7802734375, "loss_alignment_w": 0.39013671875, "loss_sft": 0.2008664384484291, "loss_total_v6": 0.5903012380003929, "step": 5309 }, { "epoch": 1.2246342342342342, "grad_norm": 31.439456939697266, "learning_rate": 2.6585713546636773e-06, "loss": 9.644241333007812, "loss_alignment": 0.78076171875, "loss_alignment_w": 0.390380859375, "loss_sft": 0.21279619447886944, "loss_total_v6": 0.6027650535106659, "step": 5310 }, { "epoch": 1.2248648648648648, "grad_norm": 29.384702682495117, "learning_rate": 2.6567921604621438e-06, "loss": 9.618667602539062, "loss_alignment": 0.798828125, "loss_alignment_w": 0.3994140625, "loss_sft": 0.2016153298318386, "loss_total_v6": 0.601166732609272, "step": 5311 }, { "epoch": 1.2250954954954956, "grad_norm": 30.43545913696289, "learning_rate": 2.655013346381158e-06, "loss": 9.25770092010498, "loss_alignment": 0.79150390625, "loss_alignment_w": 0.395751953125, "loss_sft": 0.18332737497985363, "loss_total_v6": 0.5786063000559807, "step": 5312 }, { "epoch": 1.2253261261261261, "grad_norm": 31.787307739257812, "learning_rate": 2.6532349127092803e-06, "loss": 9.591890335083008, "loss_alignment": 0.8203125, "loss_alignment_w": 0.41015625, "loss_sft": 0.18935216590762138, "loss_total_v6": 0.599493145942688, "step": 5313 }, { "epoch": 1.2255567567567567, "grad_norm": 41.10680389404297, "learning_rate": 2.6514568597350178e-06, "loss": 9.419269561767578, "loss_alignment": 0.826171875, "loss_alignment_w": 0.4130859375, "loss_sft": 0.1763660404831171, "loss_total_v6": 0.5887043178081512, "step": 5314 }, { "epoch": 1.2257873873873875, "grad_norm": 25.04434585571289, "learning_rate": 2.6496791877468063e-06, "loss": 9.130069732666016, "loss_alignment": 0.7783203125, "loss_alignment_w": 0.38916015625, "loss_sft": 0.1815760489553213, "loss_total_v6": 0.5706294029951096, "step": 5315 }, { "epoch": 1.226018018018018, "grad_norm": 29.774770736694336, "learning_rate": 2.647901897033023e-06, "loss": 9.33322525024414, "loss_alignment": 0.77880859375, "loss_alignment_w": 0.389404296875, "loss_sft": 0.19352561607956886, "loss_total_v6": 0.58332659304142, "step": 5316 }, { "epoch": 1.2262486486486486, "grad_norm": 29.79582405090332, "learning_rate": 2.646124987881987e-06, "loss": 9.058141708374023, "loss_alignment": 0.7705078125, "loss_alignment_w": 0.38525390625, "loss_sft": 0.1812309492379427, "loss_total_v6": 0.5661338791251183, "step": 5317 }, { "epoch": 1.2264792792792794, "grad_norm": 26.4598331451416, "learning_rate": 2.644348460581951e-06, "loss": 9.474983215332031, "loss_alignment": 0.81494140625, "loss_alignment_w": 0.407470703125, "loss_sft": 0.18595168739557266, "loss_total_v6": 0.5921864286065102, "step": 5318 }, { "epoch": 1.22670990990991, "grad_norm": 32.502166748046875, "learning_rate": 2.642572315421109e-06, "loss": 8.881162643432617, "loss_alignment": 0.77294921875, "loss_alignment_w": 0.386474609375, "loss_sft": 0.16954411007463932, "loss_total_v6": 0.5550726726651192, "step": 5319 }, { "epoch": 1.2269405405405405, "grad_norm": 28.347572326660156, "learning_rate": 2.64079655268759e-06, "loss": 9.575141906738281, "loss_alignment": 0.7783203125, "loss_alignment_w": 0.38916015625, "loss_sft": 0.20867585204541683, "loss_total_v6": 0.5984463468194008, "step": 5320 }, { "epoch": 1.2271711711711712, "grad_norm": 31.593914031982422, "learning_rate": 2.6390211726694624e-06, "loss": 9.537144660949707, "loss_alignment": 0.7705078125, "loss_alignment_w": 0.38525390625, "loss_sft": 0.2115958370268345, "loss_total_v6": 0.5960715115070343, "step": 5321 }, { "epoch": 1.2274018018018018, "grad_norm": 30.042316436767578, "learning_rate": 2.637246175654731e-06, "loss": 9.535572052001953, "loss_alignment": 0.78857421875, "loss_alignment_w": 0.394287109375, "loss_sft": 0.20098423585295677, "loss_total_v6": 0.5959732234477997, "step": 5322 }, { "epoch": 1.2276324324324324, "grad_norm": 30.304872512817383, "learning_rate": 2.6354715619313403e-06, "loss": 9.617300033569336, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.21449980325996876, "loss_total_v6": 0.6010812371969223, "step": 5323 }, { "epoch": 1.2278630630630631, "grad_norm": 30.790771484375, "learning_rate": 2.6336973317871757e-06, "loss": 9.4760160446167, "loss_alignment": 0.7763671875, "loss_alignment_w": 0.38818359375, "loss_sft": 0.20510504208505154, "loss_total_v6": 0.5922510400414467, "step": 5324 }, { "epoch": 1.2280936936936937, "grad_norm": 25.80756187438965, "learning_rate": 2.631923485510054e-06, "loss": 9.474006652832031, "loss_alignment": 0.83349609375, "loss_alignment_w": 0.416748046875, "loss_sft": 0.17696429789066315, "loss_total_v6": 0.592125415802002, "step": 5325 }, { "epoch": 1.2283243243243243, "grad_norm": 28.67352867126465, "learning_rate": 2.630150023387733e-06, "loss": 10.119976043701172, "loss_alignment": 0.82763671875, "loss_alignment_w": 0.413818359375, "loss_sft": 0.21874112635850906, "loss_total_v6": 0.6324984505772591, "step": 5326 }, { "epoch": 1.228554954954955, "grad_norm": 29.87921714782715, "learning_rate": 2.6283769457079084e-06, "loss": 9.72035026550293, "loss_alignment": 0.86181640625, "loss_alignment_w": 0.430908203125, "loss_sft": 0.17630848940461874, "loss_total_v6": 0.6075218841433525, "step": 5327 }, { "epoch": 1.2287855855855856, "grad_norm": 28.699033737182617, "learning_rate": 2.6266042527582103e-06, "loss": 9.274324417114258, "loss_alignment": 0.75244140625, "loss_alignment_w": 0.376220703125, "loss_sft": 0.20380608178675175, "loss_total_v6": 0.5796453058719635, "step": 5328 }, { "epoch": 1.2290162162162162, "grad_norm": 31.414295196533203, "learning_rate": 2.6248319448262104e-06, "loss": 9.622979164123535, "loss_alignment": 0.79248046875, "loss_alignment_w": 0.396240234375, "loss_sft": 0.2053333129733801, "loss_total_v6": 0.601436197757721, "step": 5329 }, { "epoch": 1.229246846846847, "grad_norm": 33.21345138549805, "learning_rate": 2.6230600221994195e-06, "loss": 9.614581108093262, "loss_alignment": 0.8232421875, "loss_alignment_w": 0.41162109375, "loss_sft": 0.18999215960502625, "loss_total_v6": 0.6009113490581512, "step": 5330 }, { "epoch": 1.2294774774774775, "grad_norm": 30.59483528137207, "learning_rate": 2.621288485165281e-06, "loss": 9.205134391784668, "loss_alignment": 0.78271484375, "loss_alignment_w": 0.391357421875, "loss_sft": 0.18428390938788652, "loss_total_v6": 0.5753209292888641, "step": 5331 }, { "epoch": 1.229708108108108, "grad_norm": 28.549814224243164, "learning_rate": 2.6195173340111767e-06, "loss": 9.487396240234375, "loss_alignment": 0.83154296875, "loss_alignment_w": 0.415771484375, "loss_sft": 0.1773128556087613, "loss_total_v6": 0.592962272465229, "step": 5332 }, { "epoch": 1.2299387387387388, "grad_norm": 29.76786231994629, "learning_rate": 2.6177465690244252e-06, "loss": 9.308295249938965, "loss_alignment": 0.78857421875, "loss_alignment_w": 0.394287109375, "loss_sft": 0.18745083920657635, "loss_total_v6": 0.5817684605717659, "step": 5333 }, { "epoch": 1.2301693693693694, "grad_norm": 28.7386531829834, "learning_rate": 2.6159761904922886e-06, "loss": 9.14803695678711, "loss_alignment": 0.7529296875, "loss_alignment_w": 0.37646484375, "loss_sft": 0.1949212085455656, "loss_total_v6": 0.571752279996872, "step": 5334 }, { "epoch": 1.2304, "grad_norm": 28.422029495239258, "learning_rate": 2.614206198701958e-06, "loss": 9.482184410095215, "loss_alignment": 0.81494140625, "loss_alignment_w": 0.407470703125, "loss_sft": 0.18531840108335018, "loss_total_v6": 0.5926365107297897, "step": 5335 }, { "epoch": 1.2306306306306307, "grad_norm": 32.26991271972656, "learning_rate": 2.612436593940568e-06, "loss": 9.723482131958008, "loss_alignment": 0.80810546875, "loss_alignment_w": 0.404052734375, "loss_sft": 0.20368016883730888, "loss_total_v6": 0.6077176257967949, "step": 5336 }, { "epoch": 1.2308612612612613, "grad_norm": 32.3997917175293, "learning_rate": 2.6106673764951874e-06, "loss": 9.333561897277832, "loss_alignment": 0.7705078125, "loss_alignment_w": 0.38525390625, "loss_sft": 0.19755966402590275, "loss_total_v6": 0.5833476260304451, "step": 5337 }, { "epoch": 1.2310918918918918, "grad_norm": 30.241901397705078, "learning_rate": 2.6088985466528203e-06, "loss": 9.412254333496094, "loss_alignment": 0.76611328125, "loss_alignment_w": 0.383056640625, "loss_sft": 0.20656727999448776, "loss_total_v6": 0.5882658883929253, "step": 5338 }, { "epoch": 1.2313225225225226, "grad_norm": 34.01069641113281, "learning_rate": 2.6071301047004144e-06, "loss": 9.152900695800781, "loss_alignment": 0.8037109375, "loss_alignment_w": 0.40185546875, "loss_sft": 0.17071964964270592, "loss_total_v6": 0.5720563158392906, "step": 5339 }, { "epoch": 1.2315531531531532, "grad_norm": 30.873750686645508, "learning_rate": 2.605362050924848e-06, "loss": 9.515790939331055, "loss_alignment": 0.84765625, "loss_alignment_w": 0.423828125, "loss_sft": 0.171244528144598, "loss_total_v6": 0.5947369411587715, "step": 5340 }, { "epoch": 1.2317837837837837, "grad_norm": 36.45307540893555, "learning_rate": 2.603594385612937e-06, "loss": 8.995984077453613, "loss_alignment": 0.75732421875, "loss_alignment_w": 0.378662109375, "loss_sft": 0.18367843329906464, "loss_total_v6": 0.5622490048408508, "step": 5341 }, { "epoch": 1.2320144144144145, "grad_norm": 36.447845458984375, "learning_rate": 2.60182710905144e-06, "loss": 9.679265022277832, "loss_alignment": 0.8232421875, "loss_alignment_w": 0.41162109375, "loss_sft": 0.1927531687542796, "loss_total_v6": 0.6049540862441063, "step": 5342 }, { "epoch": 1.232245045045045, "grad_norm": 28.27307891845703, "learning_rate": 2.6000602215270454e-06, "loss": 9.567276954650879, "loss_alignment": 0.81005859375, "loss_alignment_w": 0.405029296875, "loss_sft": 0.19326123222708702, "loss_total_v6": 0.5979548245668411, "step": 5343 }, { "epoch": 1.2324756756756756, "grad_norm": 31.659503936767578, "learning_rate": 2.598293723326385e-06, "loss": 9.39110279083252, "loss_alignment": 0.7666015625, "loss_alignment_w": 0.38330078125, "loss_sft": 0.20359736867249012, "loss_total_v6": 0.5869439616799355, "step": 5344 }, { "epoch": 1.2327063063063064, "grad_norm": 32.292572021484375, "learning_rate": 2.5965276147360226e-06, "loss": 8.75123405456543, "loss_alignment": 0.755859375, "loss_alignment_w": 0.3779296875, "loss_sft": 0.16784748993813992, "loss_total_v6": 0.5469521321356297, "step": 5345 }, { "epoch": 1.232936936936937, "grad_norm": 29.72218894958496, "learning_rate": 2.5947618960424596e-06, "loss": 9.712125778198242, "loss_alignment": 0.80712890625, "loss_alignment_w": 0.403564453125, "loss_sft": 0.20303144305944443, "loss_total_v6": 0.6070078685879707, "step": 5346 }, { "epoch": 1.2331675675675675, "grad_norm": 28.637300491333008, "learning_rate": 2.5929965675321344e-06, "loss": 9.220773696899414, "loss_alignment": 0.79541015625, "loss_alignment_w": 0.397705078125, "loss_sft": 0.17856278084218502, "loss_total_v6": 0.5762983709573746, "step": 5347 }, { "epoch": 1.2333981981981983, "grad_norm": 31.965900421142578, "learning_rate": 2.5912316294914232e-06, "loss": 9.583881378173828, "loss_alignment": 0.77880859375, "loss_alignment_w": 0.389404296875, "loss_sft": 0.20984766818583012, "loss_total_v6": 0.5989925637841225, "step": 5348 }, { "epoch": 1.2336288288288288, "grad_norm": 33.34019088745117, "learning_rate": 2.5894670822066402e-06, "loss": 9.641561508178711, "loss_alignment": 0.80029296875, "loss_alignment_w": 0.400146484375, "loss_sft": 0.2019322793930769, "loss_total_v6": 0.6025975793600082, "step": 5349 }, { "epoch": 1.2338594594594594, "grad_norm": 30.4906063079834, "learning_rate": 2.587702925964034e-06, "loss": 9.696842193603516, "loss_alignment": 0.84130859375, "loss_alignment_w": 0.420654296875, "loss_sft": 0.18579509016126394, "loss_total_v6": 0.6060526594519615, "step": 5350 }, { "epoch": 1.2340900900900902, "grad_norm": 28.357938766479492, "learning_rate": 2.5859391610497888e-06, "loss": 9.889636993408203, "loss_alignment": 0.80419921875, "loss_alignment_w": 0.402099609375, "loss_sft": 0.21508712321519852, "loss_total_v6": 0.6181022599339485, "step": 5351 }, { "epoch": 1.2343207207207207, "grad_norm": 28.63344383239746, "learning_rate": 2.5841757877500245e-06, "loss": 9.331194877624512, "loss_alignment": 0.7900390625, "loss_alignment_w": 0.39501953125, "loss_sft": 0.1876919139176607, "loss_total_v6": 0.5831996873021126, "step": 5352 }, { "epoch": 1.2345513513513513, "grad_norm": 30.482837677001953, "learning_rate": 2.5824128063508043e-06, "loss": 9.49044418334961, "loss_alignment": 0.7998046875, "loss_alignment_w": 0.39990234375, "loss_sft": 0.19321989454329014, "loss_total_v6": 0.5931527614593506, "step": 5353 }, { "epoch": 1.2347819819819819, "grad_norm": 31.300918579101562, "learning_rate": 2.5806502171381183e-06, "loss": 9.336761474609375, "loss_alignment": 0.79052734375, "loss_alignment_w": 0.395263671875, "loss_sft": 0.18765831366181374, "loss_total_v6": 0.5835475921630859, "step": 5354 }, { "epoch": 1.2350126126126126, "grad_norm": 29.474912643432617, "learning_rate": 2.5788880203979024e-06, "loss": 9.399048805236816, "loss_alignment": 0.775390625, "loss_alignment_w": 0.3876953125, "loss_sft": 0.20041662268340588, "loss_total_v6": 0.5874405279755592, "step": 5355 }, { "epoch": 1.2352432432432432, "grad_norm": 28.4366455078125, "learning_rate": 2.5771262164160216e-06, "loss": 9.001710891723633, "loss_alignment": 0.77197265625, "loss_alignment_w": 0.385986328125, "loss_sft": 0.1766816172748804, "loss_total_v6": 0.5626069083809853, "step": 5356 }, { "epoch": 1.235473873873874, "grad_norm": 33.433414459228516, "learning_rate": 2.575364805478278e-06, "loss": 9.962724685668945, "loss_alignment": 0.83642578125, "loss_alignment_w": 0.418212890625, "loss_sft": 0.20302310585975647, "loss_total_v6": 0.6226703152060509, "step": 5357 }, { "epoch": 1.2357045045045045, "grad_norm": 33.7427864074707, "learning_rate": 2.573603787870416e-06, "loss": 9.492316246032715, "loss_alignment": 0.80224609375, "loss_alignment_w": 0.401123046875, "loss_sft": 0.19274181500077248, "loss_total_v6": 0.5932697504758835, "step": 5358 }, { "epoch": 1.235935135135135, "grad_norm": 40.525997161865234, "learning_rate": 2.5718431638781094e-06, "loss": 9.312171936035156, "loss_alignment": 0.78662109375, "loss_alignment_w": 0.393310546875, "loss_sft": 0.1889748526737094, "loss_total_v6": 0.5820107609033585, "step": 5359 }, { "epoch": 1.2361657657657656, "grad_norm": 29.603118896484375, "learning_rate": 2.57008293378697e-06, "loss": 9.809330940246582, "loss_alignment": 0.79833984375, "loss_alignment_w": 0.399169921875, "loss_sft": 0.21376066841185093, "loss_total_v6": 0.6130831837654114, "step": 5360 }, { "epoch": 1.2363963963963964, "grad_norm": 32.02178192138672, "learning_rate": 2.568323097882548e-06, "loss": 9.50180721282959, "loss_alignment": 0.7841796875, "loss_alignment_w": 0.39208984375, "loss_sft": 0.20169681124389172, "loss_total_v6": 0.5938629508018494, "step": 5361 }, { "epoch": 1.236627027027027, "grad_norm": 30.410125732421875, "learning_rate": 2.566563656450326e-06, "loss": 9.558650970458984, "loss_alignment": 0.7783203125, "loss_alignment_w": 0.38916015625, "loss_sft": 0.20776725560426712, "loss_total_v6": 0.5974156931042671, "step": 5362 }, { "epoch": 1.2368576576576578, "grad_norm": 28.324026107788086, "learning_rate": 2.5648046097757285e-06, "loss": 9.690811157226562, "loss_alignment": 0.7998046875, "loss_alignment_w": 0.39990234375, "loss_sft": 0.20496456511318684, "loss_total_v6": 0.605675645172596, "step": 5363 }, { "epoch": 1.2370882882882883, "grad_norm": 33.82666778564453, "learning_rate": 2.563045958144109e-06, "loss": 8.916080474853516, "loss_alignment": 0.7783203125, "loss_alignment_w": 0.38916015625, "loss_sft": 0.16804908867925406, "loss_total_v6": 0.5572550259530544, "step": 5364 }, { "epoch": 1.2373189189189189, "grad_norm": 26.418676376342773, "learning_rate": 2.5612877018407608e-06, "loss": 9.119251251220703, "loss_alignment": 0.77294921875, "loss_alignment_w": 0.386474609375, "loss_sft": 0.18411941640079021, "loss_total_v6": 0.5699531584978104, "step": 5365 }, { "epoch": 1.2375495495495494, "grad_norm": 28.913007736206055, "learning_rate": 2.5595298411509095e-06, "loss": 9.409852981567383, "loss_alignment": 0.7666015625, "loss_alignment_w": 0.38330078125, "loss_sft": 0.20467765256762505, "loss_total_v6": 0.588115781545639, "step": 5366 }, { "epoch": 1.2377801801801802, "grad_norm": 31.17060661315918, "learning_rate": 2.5577723763597222e-06, "loss": 9.299652099609375, "loss_alignment": 0.78759765625, "loss_alignment_w": 0.393798828125, "loss_sft": 0.1864375490695238, "loss_total_v6": 0.5812282040715218, "step": 5367 }, { "epoch": 1.2380108108108108, "grad_norm": 27.5810604095459, "learning_rate": 2.5560153077523007e-06, "loss": 9.2344970703125, "loss_alignment": 0.7724609375, "loss_alignment_w": 0.38623046875, "loss_sft": 0.18979642167687416, "loss_total_v6": 0.5771560370922089, "step": 5368 }, { "epoch": 1.2382414414414415, "grad_norm": 29.405805587768555, "learning_rate": 2.5542586356136777e-06, "loss": 9.934039115905762, "loss_alignment": 0.83251953125, "loss_alignment_w": 0.416259765625, "loss_sft": 0.2050449661910534, "loss_total_v6": 0.6208774819970131, "step": 5369 }, { "epoch": 1.238472072072072, "grad_norm": 28.68061637878418, "learning_rate": 2.552502360228827e-06, "loss": 9.263418197631836, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.195464501157403, "loss_total_v6": 0.5789636522531509, "step": 5370 }, { "epoch": 1.2387027027027027, "grad_norm": 30.36309051513672, "learning_rate": 2.5507464818826523e-06, "loss": 8.987642288208008, "loss_alignment": 0.77197265625, "loss_alignment_w": 0.385986328125, "loss_sft": 0.17461222037672997, "loss_total_v6": 0.5617276802659035, "step": 5371 }, { "epoch": 1.2389333333333332, "grad_norm": 27.145814895629883, "learning_rate": 2.548991000859997e-06, "loss": 9.150543212890625, "loss_alignment": 0.798828125, "loss_alignment_w": 0.3994140625, "loss_sft": 0.1740055400878191, "loss_total_v6": 0.571908988058567, "step": 5372 }, { "epoch": 1.239163963963964, "grad_norm": 27.770315170288086, "learning_rate": 2.5472359174456394e-06, "loss": 8.945923805236816, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.17055517807602882, "loss_total_v6": 0.5591202303767204, "step": 5373 }, { "epoch": 1.2393945945945946, "grad_norm": 30.419296264648438, "learning_rate": 2.545481231924296e-06, "loss": 9.274749755859375, "loss_alignment": 0.77587890625, "loss_alignment_w": 0.387939453125, "loss_sft": 0.19147295877337456, "loss_total_v6": 0.5796718373894691, "step": 5374 }, { "epoch": 1.2396252252252253, "grad_norm": 26.71524429321289, "learning_rate": 2.5437269445806146e-06, "loss": 9.107120513916016, "loss_alignment": 0.77783203125, "loss_alignment_w": 0.388916015625, "loss_sft": 0.179699232801795, "loss_total_v6": 0.569195069372654, "step": 5375 }, { "epoch": 1.239855855855856, "grad_norm": 28.329448699951172, "learning_rate": 2.541973055699178e-06, "loss": 8.89614486694336, "loss_alignment": 0.7705078125, "loss_alignment_w": 0.38525390625, "loss_sft": 0.1713807713240385, "loss_total_v6": 0.55600905418396, "step": 5376 }, { "epoch": 1.2400864864864865, "grad_norm": 28.813756942749023, "learning_rate": 2.5402195655645055e-06, "loss": 8.987412452697754, "loss_alignment": 0.7880859375, "loss_alignment_w": 0.39404296875, "loss_sft": 0.1682349108159542, "loss_total_v6": 0.561713308095932, "step": 5377 }, { "epoch": 1.240317117117117, "grad_norm": 28.84654426574707, "learning_rate": 2.5384664744610553e-06, "loss": 9.835026741027832, "loss_alignment": 0.78662109375, "loss_alignment_w": 0.393310546875, "loss_sft": 0.22121078707277775, "loss_total_v6": 0.6146891862154007, "step": 5378 }, { "epoch": 1.2405477477477478, "grad_norm": 27.834941864013672, "learning_rate": 2.5367137826732146e-06, "loss": 9.642264366149902, "loss_alignment": 0.81298828125, "loss_alignment_w": 0.406494140625, "loss_sft": 0.19553700741380453, "loss_total_v6": 0.6026414930820465, "step": 5379 }, { "epoch": 1.2407783783783783, "grad_norm": 30.91106414794922, "learning_rate": 2.534961490485313e-06, "loss": 9.152191162109375, "loss_alignment": 0.78515625, "loss_alignment_w": 0.392578125, "loss_sft": 0.17970845848321915, "loss_total_v6": 0.5720119252800941, "step": 5380 }, { "epoch": 1.2410090090090091, "grad_norm": 29.87221908569336, "learning_rate": 2.5332095981816094e-06, "loss": 9.847404479980469, "loss_alignment": 0.76953125, "loss_alignment_w": 0.384765625, "loss_sft": 0.229781623929739, "loss_total_v6": 0.6154627576470375, "step": 5381 }, { "epoch": 1.2412396396396397, "grad_norm": 29.043184280395508, "learning_rate": 2.5314581060462972e-06, "loss": 8.918712615966797, "loss_alignment": 0.7958984375, "loss_alignment_w": 0.39794921875, "loss_sft": 0.15947036165744066, "loss_total_v6": 0.5574195794761181, "step": 5382 }, { "epoch": 1.2414702702702702, "grad_norm": 30.72022247314453, "learning_rate": 2.5297070143635117e-06, "loss": 9.522298812866211, "loss_alignment": 0.7705078125, "loss_alignment_w": 0.38525390625, "loss_sft": 0.20880642905831337, "loss_total_v6": 0.59514369815588, "step": 5383 }, { "epoch": 1.2417009009009008, "grad_norm": 29.260610580444336, "learning_rate": 2.5279563234173177e-06, "loss": 9.034150123596191, "loss_alignment": 0.779296875, "loss_alignment_w": 0.3896484375, "loss_sft": 0.17541321087628603, "loss_total_v6": 0.564634408801794, "step": 5384 }, { "epoch": 1.2419315315315316, "grad_norm": 26.61260223388672, "learning_rate": 2.526206033491714e-06, "loss": 9.244834899902344, "loss_alignment": 0.80517578125, "loss_alignment_w": 0.402587890625, "loss_sft": 0.17487864568829536, "loss_total_v6": 0.5778022184967995, "step": 5385 }, { "epoch": 1.2421621621621621, "grad_norm": 29.68421745300293, "learning_rate": 2.5244561448706396e-06, "loss": 9.499289512634277, "loss_alignment": 0.7841796875, "loss_alignment_w": 0.39208984375, "loss_sft": 0.20172256790101528, "loss_total_v6": 0.5937056094408035, "step": 5386 }, { "epoch": 1.242392792792793, "grad_norm": 33.836063385009766, "learning_rate": 2.5227066578379624e-06, "loss": 9.520323753356934, "loss_alignment": 0.8095703125, "loss_alignment_w": 0.40478515625, "loss_sft": 0.19121164456009865, "loss_total_v6": 0.595020242035389, "step": 5387 }, { "epoch": 1.2426234234234235, "grad_norm": 28.332273483276367, "learning_rate": 2.5209575726774916e-06, "loss": 9.570884704589844, "loss_alignment": 0.80712890625, "loss_alignment_w": 0.403564453125, "loss_sft": 0.19466161914169788, "loss_total_v6": 0.5981802865862846, "step": 5388 }, { "epoch": 1.242854054054054, "grad_norm": 30.5451602935791, "learning_rate": 2.519208889672967e-06, "loss": 9.510448455810547, "loss_alignment": 0.78125, "loss_alignment_w": 0.390625, "loss_sft": 0.20380849577486515, "loss_total_v6": 0.5944029912352562, "step": 5389 }, { "epoch": 1.2430846846846846, "grad_norm": 30.829574584960938, "learning_rate": 2.517460609108063e-06, "loss": 9.447465896606445, "loss_alignment": 0.77294921875, "loss_alignment_w": 0.386474609375, "loss_sft": 0.20506012719124556, "loss_total_v6": 0.5904666222631931, "step": 5390 }, { "epoch": 1.2433153153153154, "grad_norm": 33.42844009399414, "learning_rate": 2.5157127312663877e-06, "loss": 9.796051025390625, "loss_alignment": 0.76171875, "loss_alignment_w": 0.380859375, "loss_sft": 0.23119541443884373, "loss_total_v6": 0.6122531443834305, "step": 5391 }, { "epoch": 1.243545945945946, "grad_norm": 30.658729553222656, "learning_rate": 2.513965256431488e-06, "loss": 9.669839859008789, "loss_alignment": 0.79638671875, "loss_alignment_w": 0.398193359375, "loss_sft": 0.20492045022547245, "loss_total_v6": 0.6043650209903717, "step": 5392 }, { "epoch": 1.2437765765765767, "grad_norm": 26.77716827392578, "learning_rate": 2.5122181848868454e-06, "loss": 8.988197326660156, "loss_alignment": 0.76953125, "loss_alignment_w": 0.384765625, "loss_sft": 0.17688987776637077, "loss_total_v6": 0.5617623031139374, "step": 5393 }, { "epoch": 1.2440072072072073, "grad_norm": 31.383136749267578, "learning_rate": 2.510471516915872e-06, "loss": 9.296330451965332, "loss_alignment": 0.765625, "loss_alignment_w": 0.3828125, "loss_sft": 0.1971705723553896, "loss_total_v6": 0.5810206606984138, "step": 5394 }, { "epoch": 1.2442378378378378, "grad_norm": 30.8861141204834, "learning_rate": 2.5087252528019157e-06, "loss": 9.398372650146484, "loss_alignment": 0.8037109375, "loss_alignment_w": 0.40185546875, "loss_sft": 0.18525289930403233, "loss_total_v6": 0.5873982757329941, "step": 5395 }, { "epoch": 1.2444684684684684, "grad_norm": 28.721879959106445, "learning_rate": 2.506979392828258e-06, "loss": 9.218360900878906, "loss_alignment": 0.7568359375, "loss_alignment_w": 0.37841796875, "loss_sft": 0.1971344593912363, "loss_total_v6": 0.5761475265026093, "step": 5396 }, { "epoch": 1.2446990990990991, "grad_norm": 27.841110229492188, "learning_rate": 2.5052339372781202e-06, "loss": 9.431511878967285, "loss_alignment": 0.77099609375, "loss_alignment_w": 0.385498046875, "loss_sft": 0.20301012322306633, "loss_total_v6": 0.5894694626331329, "step": 5397 }, { "epoch": 1.2449297297297297, "grad_norm": 29.49098777770996, "learning_rate": 2.5034888864346503e-06, "loss": 9.864262580871582, "loss_alignment": 0.79541015625, "loss_alignment_w": 0.397705078125, "loss_sft": 0.21856720559298992, "loss_total_v6": 0.6165164113044739, "step": 5398 }, { "epoch": 1.2451603603603603, "grad_norm": 30.36842155456543, "learning_rate": 2.5017442405809376e-06, "loss": 9.305870056152344, "loss_alignment": 0.822265625, "loss_alignment_w": 0.4111328125, "loss_sft": 0.1720099076628685, "loss_total_v6": 0.5816168636083603, "step": 5399 }, { "epoch": 1.245390990990991, "grad_norm": 29.430906295776367, "learning_rate": 2.5000000000000015e-06, "loss": 8.99854564666748, "loss_alignment": 0.7978515625, "loss_alignment_w": 0.39892578125, "loss_sft": 0.16514652967453003, "loss_total_v6": 0.5624090805649757, "step": 5400 }, { "epoch": 1.2456216216216216, "grad_norm": 29.378585815429688, "learning_rate": 2.498256164974794e-06, "loss": 9.019937515258789, "loss_alignment": 0.74560546875, "loss_alignment_w": 0.372802734375, "loss_sft": 0.190302524715662, "loss_total_v6": 0.5637461133301258, "step": 5401 }, { "epoch": 1.2458522522522522, "grad_norm": 30.76294708251953, "learning_rate": 2.4965127357882085e-06, "loss": 9.955849647521973, "loss_alignment": 0.84765625, "loss_alignment_w": 0.423828125, "loss_sft": 0.19893127214163542, "loss_total_v6": 0.6222406029701233, "step": 5402 }, { "epoch": 1.246082882882883, "grad_norm": 29.8426570892334, "learning_rate": 2.494769712723066e-06, "loss": 9.496246337890625, "loss_alignment": 0.75244140625, "loss_alignment_w": 0.376220703125, "loss_sft": 0.2162417694926262, "loss_total_v6": 0.5935153365135193, "step": 5403 }, { "epoch": 1.2463135135135135, "grad_norm": 30.993989944458008, "learning_rate": 2.493027096062121e-06, "loss": 9.57780647277832, "loss_alignment": 0.8369140625, "loss_alignment_w": 0.41845703125, "loss_sft": 0.17922506667673588, "loss_total_v6": 0.5986128821969032, "step": 5404 }, { "epoch": 1.246544144144144, "grad_norm": 29.900287628173828, "learning_rate": 2.49128488608807e-06, "loss": 8.938332557678223, "loss_alignment": 0.779296875, "loss_alignment_w": 0.3896484375, "loss_sft": 0.16789870336651802, "loss_total_v6": 0.5586457699537277, "step": 5405 }, { "epoch": 1.2467747747747748, "grad_norm": 26.05140495300293, "learning_rate": 2.4895430830835337e-06, "loss": 9.04754638671875, "loss_alignment": 0.80712890625, "loss_alignment_w": 0.403564453125, "loss_sft": 0.16176989115774632, "loss_total_v6": 0.5654716715216637, "step": 5406 }, { "epoch": 1.2470054054054054, "grad_norm": 29.046960830688477, "learning_rate": 2.487801687331075e-06, "loss": 9.723302841186523, "loss_alignment": 0.80126953125, "loss_alignment_w": 0.400634765625, "loss_sft": 0.20714792609214783, "loss_total_v6": 0.6077063903212547, "step": 5407 }, { "epoch": 1.247236036036036, "grad_norm": 28.496103286743164, "learning_rate": 2.4860606991131857e-06, "loss": 8.944324493408203, "loss_alignment": 0.76220703125, "loss_alignment_w": 0.381103515625, "loss_sft": 0.17898489721119404, "loss_total_v6": 0.5590203143656254, "step": 5408 }, { "epoch": 1.2474666666666667, "grad_norm": 28.812904357910156, "learning_rate": 2.4843201187122925e-06, "loss": 9.34716796875, "loss_alignment": 0.78466796875, "loss_alignment_w": 0.392333984375, "loss_sft": 0.1916808933019638, "loss_total_v6": 0.584197998046875, "step": 5409 }, { "epoch": 1.2476972972972973, "grad_norm": 25.779897689819336, "learning_rate": 2.4825799464107547e-06, "loss": 9.268047332763672, "loss_alignment": 0.76171875, "loss_alignment_w": 0.380859375, "loss_sft": 0.19772221148014069, "loss_total_v6": 0.5792529657483101, "step": 5410 }, { "epoch": 1.2479279279279278, "grad_norm": 27.997615814208984, "learning_rate": 2.480840182490869e-06, "loss": 9.111961364746094, "loss_alignment": 0.75634765625, "loss_alignment_w": 0.378173828125, "loss_sft": 0.19126277416944504, "loss_total_v6": 0.569497637450695, "step": 5411 }, { "epoch": 1.2481585585585586, "grad_norm": 32.67865753173828, "learning_rate": 2.4791008272348656e-06, "loss": 9.708223342895508, "loss_alignment": 0.8095703125, "loss_alignment_w": 0.40478515625, "loss_sft": 0.20155152771621943, "loss_total_v6": 0.6067639514803886, "step": 5412 }, { "epoch": 1.2483891891891892, "grad_norm": 29.928354263305664, "learning_rate": 2.4773618809249045e-06, "loss": 8.961934089660645, "loss_alignment": 0.7734375, "loss_alignment_w": 0.38671875, "loss_sft": 0.17353950440883636, "loss_total_v6": 0.5601209104061127, "step": 5413 }, { "epoch": 1.2486198198198197, "grad_norm": 30.698089599609375, "learning_rate": 2.475623343843082e-06, "loss": 9.166855812072754, "loss_alignment": 0.755859375, "loss_alignment_w": 0.3779296875, "loss_sft": 0.19534973613917828, "loss_total_v6": 0.5729284733533859, "step": 5414 }, { "epoch": 1.2488504504504505, "grad_norm": 28.403528213500977, "learning_rate": 2.4738852162714278e-06, "loss": 9.56002140045166, "loss_alignment": 0.81005859375, "loss_alignment_w": 0.405029296875, "loss_sft": 0.1917549055069685, "loss_total_v6": 0.5975013598799706, "step": 5415 }, { "epoch": 1.249081081081081, "grad_norm": 30.354753494262695, "learning_rate": 2.4721474984919027e-06, "loss": 9.392499923706055, "loss_alignment": 0.78125, "loss_alignment_w": 0.390625, "loss_sft": 0.1956280618906021, "loss_total_v6": 0.587031252682209, "step": 5416 }, { "epoch": 1.2493117117117116, "grad_norm": 29.624372482299805, "learning_rate": 2.470410190786404e-06, "loss": 9.637046813964844, "loss_alignment": 0.78564453125, "loss_alignment_w": 0.392822265625, "loss_sft": 0.20898965746164322, "loss_total_v6": 0.6023154556751251, "step": 5417 }, { "epoch": 1.2495423423423424, "grad_norm": 30.881771087646484, "learning_rate": 2.4686732934367664e-06, "loss": 8.934626579284668, "loss_alignment": 0.759765625, "loss_alignment_w": 0.3798828125, "loss_sft": 0.177432743832469, "loss_total_v6": 0.5584141835570335, "step": 5418 }, { "epoch": 1.249772972972973, "grad_norm": 26.603918075561523, "learning_rate": 2.466936806724749e-06, "loss": 9.1891450881958, "loss_alignment": 0.79833984375, "loss_alignment_w": 0.399169921875, "loss_sft": 0.17507538758218288, "loss_total_v6": 0.5743215903639793, "step": 5419 }, { "epoch": 1.2500036036036035, "grad_norm": 32.010223388671875, "learning_rate": 2.4652007309320497e-06, "loss": 9.45901107788086, "loss_alignment": 0.79638671875, "loss_alignment_w": 0.398193359375, "loss_sft": 0.19319313764572144, "loss_total_v6": 0.5911881625652313, "step": 5420 }, { "epoch": 1.2502342342342343, "grad_norm": 28.809694290161133, "learning_rate": 2.4634650663402967e-06, "loss": 8.835962295532227, "loss_alignment": 0.755859375, "loss_alignment_w": 0.3779296875, "loss_sft": 0.1738601941615343, "loss_total_v6": 0.552247665822506, "step": 5421 }, { "epoch": 1.2504648648648649, "grad_norm": 27.012605667114258, "learning_rate": 2.4617298132310567e-06, "loss": 9.481435775756836, "loss_alignment": 0.76708984375, "loss_alignment_w": 0.383544921875, "loss_sft": 0.20841919258236885, "loss_total_v6": 0.592589721083641, "step": 5422 }, { "epoch": 1.2506954954954954, "grad_norm": 31.846731185913086, "learning_rate": 2.459994971885824e-06, "loss": 9.488313674926758, "loss_alignment": 0.7734375, "loss_alignment_w": 0.38671875, "loss_sft": 0.20552262477576733, "loss_total_v6": 0.5930195823311806, "step": 5423 }, { "epoch": 1.2509261261261262, "grad_norm": 30.6555233001709, "learning_rate": 2.4582605425860312e-06, "loss": 9.41379165649414, "loss_alignment": 0.81298828125, "loss_alignment_w": 0.406494140625, "loss_sft": 0.1808454692363739, "loss_total_v6": 0.5883619636297226, "step": 5424 }, { "epoch": 1.2511567567567567, "grad_norm": 27.49359893798828, "learning_rate": 2.4565265256130393e-06, "loss": 9.607942581176758, "loss_alignment": 0.78662109375, "loss_alignment_w": 0.393310546875, "loss_sft": 0.20717056281864643, "loss_total_v6": 0.6004963740706444, "step": 5425 }, { "epoch": 1.2513873873873873, "grad_norm": 32.302101135253906, "learning_rate": 2.4547929212481436e-06, "loss": 9.764843940734863, "loss_alignment": 0.81640625, "loss_alignment_w": 0.408203125, "loss_sft": 0.20240479614585638, "loss_total_v6": 0.6103027611970901, "step": 5426 }, { "epoch": 1.251618018018018, "grad_norm": 28.313594818115234, "learning_rate": 2.453059729772577e-06, "loss": 8.81157398223877, "loss_alignment": 0.77392578125, "loss_alignment_w": 0.386962890625, "loss_sft": 0.16411142144352198, "loss_total_v6": 0.5507233738899231, "step": 5427 }, { "epoch": 1.2518486486486486, "grad_norm": 27.04300308227539, "learning_rate": 2.451326951467499e-06, "loss": 9.449358940124512, "loss_alignment": 0.77978515625, "loss_alignment_w": 0.389892578125, "loss_sft": 0.19944113120436668, "loss_total_v6": 0.5905849263072014, "step": 5428 }, { "epoch": 1.2520792792792792, "grad_norm": 27.817956924438477, "learning_rate": 2.449594586614003e-06, "loss": 9.106169700622559, "loss_alignment": 0.75927734375, "loss_alignment_w": 0.379638671875, "loss_sft": 0.19054978899657726, "loss_total_v6": 0.5691355988383293, "step": 5429 }, { "epoch": 1.25230990990991, "grad_norm": 30.268980026245117, "learning_rate": 2.4478626354931222e-06, "loss": 9.492147445678711, "loss_alignment": 0.798828125, "loss_alignment_w": 0.3994140625, "loss_sft": 0.19410458765923977, "loss_total_v6": 0.593259260058403, "step": 5430 }, { "epoch": 1.2525405405405405, "grad_norm": 27.8886775970459, "learning_rate": 2.4461310983858135e-06, "loss": 9.612735748291016, "loss_alignment": 0.8115234375, "loss_alignment_w": 0.40576171875, "loss_sft": 0.1942560076713562, "loss_total_v6": 0.6007959395647049, "step": 5431 }, { "epoch": 1.252771171171171, "grad_norm": 28.757770538330078, "learning_rate": 2.444399975572974e-06, "loss": 9.856849670410156, "loss_alignment": 0.79443359375, "loss_alignment_w": 0.397216796875, "loss_sft": 0.21914147399365902, "loss_total_v6": 0.6160530969500542, "step": 5432 }, { "epoch": 1.2530018018018019, "grad_norm": 30.089569091796875, "learning_rate": 2.4426692673354284e-06, "loss": 9.603591918945312, "loss_alignment": 0.783203125, "loss_alignment_w": 0.3916015625, "loss_sft": 0.20813463628292084, "loss_total_v6": 0.6002244725823402, "step": 5433 }, { "epoch": 1.2532324324324324, "grad_norm": 31.20802879333496, "learning_rate": 2.4409389739539374e-06, "loss": 9.036678314208984, "loss_alignment": 0.7607421875, "loss_alignment_w": 0.38037109375, "loss_sft": 0.18321593105793, "loss_total_v6": 0.5647924542427063, "step": 5434 }, { "epoch": 1.253463063063063, "grad_norm": 32.118587493896484, "learning_rate": 2.43920909570919e-06, "loss": 9.430623054504395, "loss_alignment": 0.7861328125, "loss_alignment_w": 0.39306640625, "loss_sft": 0.19569139275699854, "loss_total_v6": 0.5894139334559441, "step": 5435 }, { "epoch": 1.2536936936936938, "grad_norm": 29.890140533447266, "learning_rate": 2.4374796328818136e-06, "loss": 9.68599796295166, "loss_alignment": 0.81640625, "loss_alignment_w": 0.408203125, "loss_sft": 0.19706495851278305, "loss_total_v6": 0.6053748801350594, "step": 5436 }, { "epoch": 1.2539243243243243, "grad_norm": 31.913637161254883, "learning_rate": 2.4357505857523667e-06, "loss": 9.686626434326172, "loss_alignment": 0.83203125, "loss_alignment_w": 0.416015625, "loss_sft": 0.19032932445406914, "loss_total_v6": 0.6054141595959663, "step": 5437 }, { "epoch": 1.2541549549549549, "grad_norm": 30.613996505737305, "learning_rate": 2.434021954601339e-06, "loss": 9.255730628967285, "loss_alignment": 0.7666015625, "loss_alignment_w": 0.38330078125, "loss_sft": 0.19519764743745327, "loss_total_v6": 0.5784831643104553, "step": 5438 }, { "epoch": 1.2543855855855857, "grad_norm": 29.949344635009766, "learning_rate": 2.432293739709151e-06, "loss": 10.025686264038086, "loss_alignment": 0.8310546875, "loss_alignment_w": 0.41552734375, "loss_sft": 0.21100173890590668, "loss_total_v6": 0.6266054064035416, "step": 5439 }, { "epoch": 1.2546162162162162, "grad_norm": 29.08414077758789, "learning_rate": 2.430565941356157e-06, "loss": 9.238143920898438, "loss_alignment": 0.74072265625, "loss_alignment_w": 0.370361328125, "loss_sft": 0.20731262117624283, "loss_total_v6": 0.5773840136826038, "step": 5440 }, { "epoch": 1.2548468468468468, "grad_norm": 30.903907775878906, "learning_rate": 2.4288385598226483e-06, "loss": 9.651100158691406, "loss_alignment": 0.8212890625, "loss_alignment_w": 0.41064453125, "loss_sft": 0.19079446513205767, "loss_total_v6": 0.6031937301158905, "step": 5441 }, { "epoch": 1.2550774774774776, "grad_norm": 26.64326286315918, "learning_rate": 2.427111595388841e-06, "loss": 9.310710906982422, "loss_alignment": 0.7568359375, "loss_alignment_w": 0.37841796875, "loss_sft": 0.20327261835336685, "loss_total_v6": 0.5819194540381432, "step": 5442 }, { "epoch": 1.2553081081081081, "grad_norm": 30.076520919799805, "learning_rate": 2.4253850483348894e-06, "loss": 9.447684288024902, "loss_alignment": 0.7724609375, "loss_alignment_w": 0.38623046875, "loss_sft": 0.2037157490849495, "loss_total_v6": 0.5904802605509758, "step": 5443 }, { "epoch": 1.2555387387387387, "grad_norm": 34.11768341064453, "learning_rate": 2.4236589189408783e-06, "loss": 9.198335647583008, "loss_alignment": 0.77978515625, "loss_alignment_w": 0.389892578125, "loss_sft": 0.18511018436402082, "loss_total_v6": 0.5748959518969059, "step": 5444 }, { "epoch": 1.2557693693693694, "grad_norm": 33.11974334716797, "learning_rate": 2.4219332074868207e-06, "loss": 8.938520431518555, "loss_alignment": 0.79736328125, "loss_alignment_w": 0.398681640625, "loss_sft": 0.1612270874902606, "loss_total_v6": 0.5586575195193291, "step": 5445 }, { "epoch": 1.256, "grad_norm": 30.051244735717773, "learning_rate": 2.42020791425267e-06, "loss": 9.88587760925293, "loss_alignment": 0.7939453125, "loss_alignment_w": 0.39697265625, "loss_sft": 0.22062001004815102, "loss_total_v6": 0.6178673133254051, "step": 5446 }, { "epoch": 1.2562306306306306, "grad_norm": 29.667932510375977, "learning_rate": 2.418483039518304e-06, "loss": 9.510796546936035, "loss_alignment": 0.78857421875, "loss_alignment_w": 0.394287109375, "loss_sft": 0.2015720084309578, "loss_total_v6": 0.594424769282341, "step": 5447 }, { "epoch": 1.2564612612612613, "grad_norm": 30.00787925720215, "learning_rate": 2.4167585835635383e-06, "loss": 9.32822036743164, "loss_alignment": 0.765625, "loss_alignment_w": 0.3828125, "loss_sft": 0.19978926703333855, "loss_total_v6": 0.5830137506127357, "step": 5448 }, { "epoch": 1.256691891891892, "grad_norm": 29.78329086303711, "learning_rate": 2.4150345466681175e-06, "loss": 9.291925430297852, "loss_alignment": 0.80810546875, "loss_alignment_w": 0.404052734375, "loss_sft": 0.17730294167995453, "loss_total_v6": 0.5807453244924545, "step": 5449 }, { "epoch": 1.2569225225225225, "grad_norm": 28.416704177856445, "learning_rate": 2.4133109291117156e-06, "loss": 8.922246932983398, "loss_alignment": 0.7578125, "loss_alignment_w": 0.37890625, "loss_sft": 0.17940552905201912, "loss_total_v6": 0.55764040350914, "step": 5450 }, { "epoch": 1.2571531531531532, "grad_norm": 28.694942474365234, "learning_rate": 2.411587731173947e-06, "loss": 9.335216522216797, "loss_alignment": 0.787109375, "loss_alignment_w": 0.3935546875, "loss_sft": 0.18963698484003544, "loss_total_v6": 0.5834510624408722, "step": 5451 }, { "epoch": 1.2573837837837838, "grad_norm": 30.678089141845703, "learning_rate": 2.40986495313435e-06, "loss": 9.895610809326172, "loss_alignment": 0.8291015625, "loss_alignment_w": 0.41455078125, "loss_sft": 0.20404700003564358, "loss_total_v6": 0.6184757202863693, "step": 5452 }, { "epoch": 1.2576144144144143, "grad_norm": 31.009483337402344, "learning_rate": 2.408142595272396e-06, "loss": 9.490398406982422, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.20444752275943756, "loss_total_v6": 0.593149907886982, "step": 5453 }, { "epoch": 1.2578450450450451, "grad_norm": 31.778871536254883, "learning_rate": 2.4064206578674937e-06, "loss": 9.186625480651855, "loss_alignment": 0.78466796875, "loss_alignment_w": 0.392333984375, "loss_sft": 0.18135707825422287, "loss_total_v6": 0.5741640999913216, "step": 5454 }, { "epoch": 1.2580756756756757, "grad_norm": 30.980199813842773, "learning_rate": 2.4046991411989754e-06, "loss": 9.997869491577148, "loss_alignment": 0.80419921875, "loss_alignment_w": 0.402099609375, "loss_sft": 0.22262990288436413, "loss_total_v6": 0.6248668655753136, "step": 5455 }, { "epoch": 1.2583063063063062, "grad_norm": 31.35055923461914, "learning_rate": 2.402978045546114e-06, "loss": 9.073116302490234, "loss_alignment": 0.78955078125, "loss_alignment_w": 0.394775390625, "loss_sft": 0.17122626677155495, "loss_total_v6": 0.5670697651803493, "step": 5456 }, { "epoch": 1.2585369369369368, "grad_norm": 29.60733985900879, "learning_rate": 2.4012573711881076e-06, "loss": 9.24716854095459, "loss_alignment": 0.783203125, "loss_alignment_w": 0.3916015625, "loss_sft": 0.1858429480344057, "loss_total_v6": 0.5779480561614037, "step": 5457 }, { "epoch": 1.2587675675675676, "grad_norm": 26.948415756225586, "learning_rate": 2.399537118404087e-06, "loss": 8.995401382446289, "loss_alignment": 0.75, "loss_alignment_w": 0.375, "loss_sft": 0.18792980257421732, "loss_total_v6": 0.562212597578764, "step": 5458 }, { "epoch": 1.2589981981981981, "grad_norm": 29.24308204650879, "learning_rate": 2.397817287473115e-06, "loss": 9.188318252563477, "loss_alignment": 0.79248046875, "loss_alignment_w": 0.396240234375, "loss_sft": 0.17748034186661243, "loss_total_v6": 0.5742698982357979, "step": 5459 }, { "epoch": 1.259228828828829, "grad_norm": 26.15113639831543, "learning_rate": 2.3960978786741878e-06, "loss": 9.205493927001953, "loss_alignment": 0.8125, "loss_alignment_w": 0.40625, "loss_sft": 0.16857454180717468, "loss_total_v6": 0.5753433406352997, "step": 5460 }, { "epoch": 1.2594594594594595, "grad_norm": 25.908180236816406, "learning_rate": 2.394378892286233e-06, "loss": 9.690746307373047, "loss_alignment": 0.80078125, "loss_alignment_w": 0.400390625, "loss_sft": 0.20349576510488987, "loss_total_v6": 0.6056716591119766, "step": 5461 }, { "epoch": 1.25969009009009, "grad_norm": 30.706045150756836, "learning_rate": 2.392660328588108e-06, "loss": 8.95638656616211, "loss_alignment": 0.7734375, "loss_alignment_w": 0.38671875, "loss_sft": 0.17253657430410385, "loss_total_v6": 0.5597741194069386, "step": 5462 }, { "epoch": 1.2599207207207206, "grad_norm": 26.641992568969727, "learning_rate": 2.3909421878586014e-06, "loss": 9.36295223236084, "loss_alignment": 0.78076171875, "loss_alignment_w": 0.390380859375, "loss_sft": 0.19431538321077824, "loss_total_v6": 0.5851845219731331, "step": 5463 }, { "epoch": 1.2601513513513514, "grad_norm": 28.068260192871094, "learning_rate": 2.3892244703764345e-06, "loss": 9.403722763061523, "loss_alignment": 0.802734375, "loss_alignment_w": 0.4013671875, "loss_sft": 0.18622812628746033, "loss_total_v6": 0.5877326726913452, "step": 5464 }, { "epoch": 1.260381981981982, "grad_norm": 28.700817108154297, "learning_rate": 2.387507176420256e-06, "loss": 9.062236785888672, "loss_alignment": 0.77490234375, "loss_alignment_w": 0.387451171875, "loss_sft": 0.17832828871905804, "loss_total_v6": 0.566389799118042, "step": 5465 }, { "epoch": 1.2606126126126127, "grad_norm": 30.531307220458984, "learning_rate": 2.385790306268653e-06, "loss": 9.310934066772461, "loss_alignment": 0.7841796875, "loss_alignment_w": 0.39208984375, "loss_sft": 0.18966044671833515, "loss_total_v6": 0.5819334164261818, "step": 5466 }, { "epoch": 1.2608432432432433, "grad_norm": 27.970903396606445, "learning_rate": 2.3840738602001406e-06, "loss": 9.483760833740234, "loss_alignment": 0.798828125, "loss_alignment_w": 0.3994140625, "loss_sft": 0.19316841661930084, "loss_total_v6": 0.5927350372076035, "step": 5467 }, { "epoch": 1.2610738738738738, "grad_norm": 29.975215911865234, "learning_rate": 2.3823578384931633e-06, "loss": 9.060142517089844, "loss_alignment": 0.74755859375, "loss_alignment_w": 0.373779296875, "loss_sft": 0.192555938847363, "loss_total_v6": 0.5662589333951473, "step": 5468 }, { "epoch": 1.2613045045045044, "grad_norm": 27.869834899902344, "learning_rate": 2.380642241426098e-06, "loss": 9.006381034851074, "loss_alignment": 0.7666015625, "loss_alignment_w": 0.38330078125, "loss_sft": 0.17996420338749886, "loss_total_v6": 0.5628987923264503, "step": 5469 }, { "epoch": 1.2615351351351352, "grad_norm": 28.156787872314453, "learning_rate": 2.378927069277251e-06, "loss": 9.548419952392578, "loss_alignment": 0.7890625, "loss_alignment_w": 0.39453125, "loss_sft": 0.20091742929071188, "loss_total_v6": 0.5967762023210526, "step": 5470 }, { "epoch": 1.2617657657657657, "grad_norm": 30.926111221313477, "learning_rate": 2.3772123223248654e-06, "loss": 9.082014083862305, "loss_alignment": 0.78076171875, "loss_alignment_w": 0.390380859375, "loss_sft": 0.17774854600429535, "loss_total_v6": 0.5676258578896523, "step": 5471 }, { "epoch": 1.2619963963963965, "grad_norm": 29.248659133911133, "learning_rate": 2.3754980008471074e-06, "loss": 8.877567291259766, "loss_alignment": 0.724609375, "loss_alignment_w": 0.3623046875, "loss_sft": 0.1926652928814292, "loss_total_v6": 0.5548479072749615, "step": 5472 }, { "epoch": 1.262227027027027, "grad_norm": 38.33469772338867, "learning_rate": 2.373784105122082e-06, "loss": 9.485977172851562, "loss_alignment": 0.80029296875, "loss_alignment_w": 0.400146484375, "loss_sft": 0.19266602583229542, "loss_total_v6": 0.5928735509514809, "step": 5473 }, { "epoch": 1.2624576576576576, "grad_norm": 31.034650802612305, "learning_rate": 2.372070635427819e-06, "loss": 8.964244842529297, "loss_alignment": 0.763671875, "loss_alignment_w": 0.3818359375, "loss_sft": 0.17815466225147247, "loss_total_v6": 0.5602652654051781, "step": 5474 }, { "epoch": 1.2626882882882882, "grad_norm": 26.964447021484375, "learning_rate": 2.3703575920422793e-06, "loss": 9.65116024017334, "loss_alignment": 0.80712890625, "loss_alignment_w": 0.403564453125, "loss_sft": 0.1994957197457552, "loss_total_v6": 0.6031975224614143, "step": 5475 }, { "epoch": 1.262918918918919, "grad_norm": 28.993684768676758, "learning_rate": 2.3686449752433617e-06, "loss": 9.293376922607422, "loss_alignment": 0.783203125, "loss_alignment_w": 0.3916015625, "loss_sft": 0.18845631275326014, "loss_total_v6": 0.5808360651135445, "step": 5476 }, { "epoch": 1.2631495495495495, "grad_norm": 31.49225616455078, "learning_rate": 2.366932785308888e-06, "loss": 9.216242790222168, "loss_alignment": 0.7939453125, "loss_alignment_w": 0.39697265625, "loss_sft": 0.1791035421192646, "loss_total_v6": 0.5760151743888855, "step": 5477 }, { "epoch": 1.2633801801801803, "grad_norm": 28.836261749267578, "learning_rate": 2.3652210225166122e-06, "loss": 9.40622615814209, "loss_alignment": 0.81396484375, "loss_alignment_w": 0.406982421875, "loss_sft": 0.180769395083189, "loss_total_v6": 0.5878891423344612, "step": 5478 }, { "epoch": 1.2636108108108108, "grad_norm": 27.48086166381836, "learning_rate": 2.363509687144224e-06, "loss": 9.370064735412598, "loss_alignment": 0.78271484375, "loss_alignment_w": 0.391357421875, "loss_sft": 0.19474459625780582, "loss_total_v6": 0.5856290236115456, "step": 5479 }, { "epoch": 1.2638414414414414, "grad_norm": 28.89608383178711, "learning_rate": 2.3617987794693358e-06, "loss": 9.890382766723633, "loss_alignment": 0.822265625, "loss_alignment_w": 0.4111328125, "loss_sft": 0.20747384801506996, "loss_total_v6": 0.6181489005684853, "step": 5480 }, { "epoch": 1.264072072072072, "grad_norm": 28.723247528076172, "learning_rate": 2.3600882997694994e-06, "loss": 9.860494613647461, "loss_alignment": 0.7978515625, "loss_alignment_w": 0.39892578125, "loss_sft": 0.21694313548505306, "loss_total_v6": 0.6162809133529663, "step": 5481 }, { "epoch": 1.2643027027027027, "grad_norm": 32.114219665527344, "learning_rate": 2.3583782483221906e-06, "loss": 8.969974517822266, "loss_alignment": 0.763671875, "loss_alignment_w": 0.3818359375, "loss_sft": 0.17861965391784906, "loss_total_v6": 0.5606234222650528, "step": 5482 }, { "epoch": 1.2645333333333333, "grad_norm": 30.031612396240234, "learning_rate": 2.356668625404818e-06, "loss": 9.59371280670166, "loss_alignment": 0.79248046875, "loss_alignment_w": 0.396240234375, "loss_sft": 0.2044044267386198, "loss_total_v6": 0.5996070727705956, "step": 5483 }, { "epoch": 1.264763963963964, "grad_norm": 27.885744094848633, "learning_rate": 2.354959431294719e-06, "loss": 9.072288513183594, "loss_alignment": 0.77783203125, "loss_alignment_w": 0.388916015625, "loss_sft": 0.17779689095914364, "loss_total_v6": 0.567018061876297, "step": 5484 }, { "epoch": 1.2649945945945946, "grad_norm": 33.203956604003906, "learning_rate": 2.353250666269164e-06, "loss": 9.334976196289062, "loss_alignment": 0.7666015625, "loss_alignment_w": 0.38330078125, "loss_sft": 0.200486171990633, "loss_total_v6": 0.5834359973669052, "step": 5485 }, { "epoch": 1.2652252252252252, "grad_norm": 35.00156784057617, "learning_rate": 2.3515423306053553e-06, "loss": 9.055896759033203, "loss_alignment": 0.7802734375, "loss_alignment_w": 0.39013671875, "loss_sft": 0.1762535385787487, "loss_total_v6": 0.5659935213625431, "step": 5486 }, { "epoch": 1.2654558558558557, "grad_norm": 29.492969512939453, "learning_rate": 2.3498344245804207e-06, "loss": 9.179082870483398, "loss_alignment": 0.81005859375, "loss_alignment_w": 0.405029296875, "loss_sft": 0.1694721356034279, "loss_total_v6": 0.5736926794052124, "step": 5487 }, { "epoch": 1.2656864864864865, "grad_norm": 29.2797908782959, "learning_rate": 2.3481269484714208e-06, "loss": 9.120132446289062, "loss_alignment": 0.77099609375, "loss_alignment_w": 0.385498046875, "loss_sft": 0.18473911751061678, "loss_total_v6": 0.5700083076953888, "step": 5488 }, { "epoch": 1.265917117117117, "grad_norm": 30.686323165893555, "learning_rate": 2.346419902555347e-06, "loss": 9.55398178100586, "loss_alignment": 0.78271484375, "loss_alignment_w": 0.391357421875, "loss_sft": 0.20454578287899494, "loss_total_v6": 0.5971238911151886, "step": 5489 }, { "epoch": 1.2661477477477479, "grad_norm": 27.235374450683594, "learning_rate": 2.344713287109117e-06, "loss": 9.189783096313477, "loss_alignment": 0.76611328125, "loss_alignment_w": 0.383056640625, "loss_sft": 0.18973314575850964, "loss_total_v6": 0.5743614360690117, "step": 5490 }, { "epoch": 1.2663783783783784, "grad_norm": 28.10345458984375, "learning_rate": 2.3430071024095853e-06, "loss": 9.365263938903809, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.2012957977131009, "loss_total_v6": 0.5853290036320686, "step": 5491 }, { "epoch": 1.266609009009009, "grad_norm": 29.4051570892334, "learning_rate": 2.3413013487335332e-06, "loss": 9.336423873901367, "loss_alignment": 0.7841796875, "loss_alignment_w": 0.39208984375, "loss_sft": 0.19200116768479347, "loss_total_v6": 0.5835264474153519, "step": 5492 }, { "epoch": 1.2668396396396395, "grad_norm": 32.37001419067383, "learning_rate": 2.3395960263576717e-06, "loss": 9.591669082641602, "loss_alignment": 0.78759765625, "loss_alignment_w": 0.393798828125, "loss_sft": 0.2065502293407917, "loss_total_v6": 0.5994793400168419, "step": 5493 }, { "epoch": 1.2670702702702703, "grad_norm": 29.676111221313477, "learning_rate": 2.3378911355586392e-06, "loss": 9.250712394714355, "loss_alignment": 0.79248046875, "loss_alignment_w": 0.396240234375, "loss_sft": 0.1814715638756752, "loss_total_v6": 0.5781695693731308, "step": 5494 }, { "epoch": 1.2673009009009009, "grad_norm": 30.523550033569336, "learning_rate": 2.3361866766130114e-06, "loss": 9.308809280395508, "loss_alignment": 0.79345703125, "loss_alignment_w": 0.396728515625, "loss_sft": 0.18372926022857428, "loss_total_v6": 0.5818005576729774, "step": 5495 }, { "epoch": 1.2675315315315316, "grad_norm": 31.547027587890625, "learning_rate": 2.334482649797287e-06, "loss": 9.401496887207031, "loss_alignment": 0.77734375, "loss_alignment_w": 0.388671875, "loss_sft": 0.19945567287504673, "loss_total_v6": 0.5875934958457947, "step": 5496 }, { "epoch": 1.2677621621621622, "grad_norm": 33.31351852416992, "learning_rate": 2.332779055387896e-06, "loss": 9.072844505310059, "loss_alignment": 0.78271484375, "loss_alignment_w": 0.391357421875, "loss_sft": 0.17612261418253183, "loss_total_v6": 0.5670528002083302, "step": 5497 }, { "epoch": 1.2679927927927928, "grad_norm": 30.380130767822266, "learning_rate": 2.3310758936612025e-06, "loss": 9.653772354125977, "loss_alignment": 0.81005859375, "loss_alignment_w": 0.405029296875, "loss_sft": 0.1973091308027506, "loss_total_v6": 0.6033607646822929, "step": 5498 }, { "epoch": 1.2682234234234233, "grad_norm": 32.07852554321289, "learning_rate": 2.329373164893494e-06, "loss": 8.9310302734375, "loss_alignment": 0.76904296875, "loss_alignment_w": 0.384521484375, "loss_sft": 0.17340847942978144, "loss_total_v6": 0.5581893771886826, "step": 5499 }, { "epoch": 1.268454054054054, "grad_norm": 31.586843490600586, "learning_rate": 2.3276708693609947e-06, "loss": 9.327108383178711, "loss_alignment": 0.7646484375, "loss_alignment_w": 0.38232421875, "loss_sft": 0.1999486219137907, "loss_total_v6": 0.5829442292451859, "step": 5500 }, { "epoch": 1.268454054054054, "eval_loss": 0.8691052198410034, "eval_loss_alignment": 0.8014924907248858, "eval_loss_alignment_w": 0.4007462453624429, "eval_loss_contrast": 0.0, "eval_loss_highlevel": 0.0, "eval_loss_sft": 0.468359030357915, "eval_loss_total_v6": 0.8691052761796403, "eval_loss_wm_recon": 0.0, "eval_loss_wm_recon_w": 0.0, "eval_runtime": 299.5048, "eval_samples_per_second": 5.826, "eval_steps_per_second": 0.731, "step": 5500 }, { "epoch": 1.2686846846846846, "grad_norm": 31.01856231689453, "learning_rate": 2.3259690073398527e-06, "loss": 9.562511444091797, "loss_alignment": 0.779296875, "loss_alignment_w": 0.3896484375, "loss_sft": 0.2071692869067192, "loss_total_v6": 0.5976569429039955, "step": 5501 }, { "epoch": 1.2689153153153154, "grad_norm": 30.061405181884766, "learning_rate": 2.324267579106148e-06, "loss": 9.332290649414062, "loss_alignment": 0.765625, "loss_alignment_w": 0.3828125, "loss_sft": 0.19992158375680447, "loss_total_v6": 0.5832681506872177, "step": 5502 }, { "epoch": 1.269145945945946, "grad_norm": 33.99355697631836, "learning_rate": 2.322566584935889e-06, "loss": 9.097448348999023, "loss_alignment": 0.7646484375, "loss_alignment_w": 0.38232421875, "loss_sft": 0.1861747559159994, "loss_total_v6": 0.5685905069112778, "step": 5503 }, { "epoch": 1.2693765765765765, "grad_norm": 26.765371322631836, "learning_rate": 2.320866025105016e-06, "loss": 9.433195114135742, "loss_alignment": 0.78369140625, "loss_alignment_w": 0.391845703125, "loss_sft": 0.19907178729772568, "loss_total_v6": 0.5895747318863869, "step": 5504 }, { "epoch": 1.269607207207207, "grad_norm": 33.62285232543945, "learning_rate": 2.3191658998893997e-06, "loss": 9.833111763000488, "loss_alignment": 0.80908203125, "loss_alignment_w": 0.404541015625, "loss_sft": 0.21010475419461727, "loss_total_v6": 0.6145694702863693, "step": 5505 }, { "epoch": 1.2698378378378379, "grad_norm": 30.571760177612305, "learning_rate": 2.317466209564837e-06, "loss": 9.404315948486328, "loss_alignment": 0.8037109375, "loss_alignment_w": 0.40185546875, "loss_sft": 0.18612783588469028, "loss_total_v6": 0.5877697020769119, "step": 5506 }, { "epoch": 1.2700684684684684, "grad_norm": 31.04212188720703, "learning_rate": 2.315766954407055e-06, "loss": 9.449821472167969, "loss_alignment": 0.77734375, "loss_alignment_w": 0.388671875, "loss_sft": 0.20185036212205887, "loss_total_v6": 0.5906138047575951, "step": 5507 }, { "epoch": 1.2702990990990992, "grad_norm": 38.39234924316406, "learning_rate": 2.3140681346917106e-06, "loss": 9.167346954345703, "loss_alignment": 0.75439453125, "loss_alignment_w": 0.377197265625, "loss_sft": 0.19548728689551353, "loss_total_v6": 0.5729592144489288, "step": 5508 }, { "epoch": 1.2705297297297298, "grad_norm": 27.675018310546875, "learning_rate": 2.3123697506943886e-06, "loss": 9.337858200073242, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.19985760562121868, "loss_total_v6": 0.5836161449551582, "step": 5509 }, { "epoch": 1.2707603603603603, "grad_norm": 36.880226135253906, "learning_rate": 2.3106718026906073e-06, "loss": 9.780511856079102, "loss_alignment": 0.8203125, "loss_alignment_w": 0.41015625, "loss_sft": 0.20121729280799627, "loss_total_v6": 0.6112819761037827, "step": 5510 }, { "epoch": 1.2709909909909909, "grad_norm": 32.36723327636719, "learning_rate": 2.3089742909558123e-06, "loss": 9.668062210083008, "loss_alignment": 0.80810546875, "loss_alignment_w": 0.404052734375, "loss_sft": 0.1999874897301197, "loss_total_v6": 0.6042538657784462, "step": 5511 }, { "epoch": 1.2712216216216217, "grad_norm": 30.98086929321289, "learning_rate": 2.307277215765377e-06, "loss": 10.030776977539062, "loss_alignment": 0.796875, "loss_alignment_w": 0.3984375, "loss_sft": 0.2290048524737358, "loss_total_v6": 0.6269235238432884, "step": 5512 }, { "epoch": 1.2714522522522522, "grad_norm": 28.592967987060547, "learning_rate": 2.3055805773946044e-06, "loss": 9.621109008789062, "loss_alignment": 0.78564453125, "loss_alignment_w": 0.392822265625, "loss_sft": 0.208481777459383, "loss_total_v6": 0.6013193055987358, "step": 5513 }, { "epoch": 1.271682882882883, "grad_norm": 30.21257972717285, "learning_rate": 2.3038843761187254e-06, "loss": 9.341012954711914, "loss_alignment": 0.7763671875, "loss_alignment_w": 0.38818359375, "loss_sft": 0.19692674092948437, "loss_total_v6": 0.5838133171200752, "step": 5514 }, { "epoch": 1.2719135135135136, "grad_norm": 25.47195816040039, "learning_rate": 2.302188612212905e-06, "loss": 9.498027801513672, "loss_alignment": 0.791015625, "loss_alignment_w": 0.3955078125, "loss_sft": 0.19764591939747334, "loss_total_v6": 0.5936267748475075, "step": 5515 }, { "epoch": 1.2721441441441441, "grad_norm": 29.975406646728516, "learning_rate": 2.3004932859522306e-06, "loss": 9.427779197692871, "loss_alignment": 0.78466796875, "loss_alignment_w": 0.392333984375, "loss_sft": 0.1971921045333147, "loss_total_v6": 0.5892361924052238, "step": 5516 }, { "epoch": 1.2723747747747747, "grad_norm": 29.78416633605957, "learning_rate": 2.298798397611725e-06, "loss": 8.808771133422852, "loss_alignment": 0.7646484375, "loss_alignment_w": 0.38232421875, "loss_sft": 0.16918528638780117, "loss_total_v6": 0.5505481883883476, "step": 5517 }, { "epoch": 1.2726054054054055, "grad_norm": 32.06715774536133, "learning_rate": 2.2971039474663365e-06, "loss": 9.270710945129395, "loss_alignment": 0.80029296875, "loss_alignment_w": 0.400146484375, "loss_sft": 0.17925765179097652, "loss_total_v6": 0.5794194117188454, "step": 5518 }, { "epoch": 1.272836036036036, "grad_norm": 29.24905014038086, "learning_rate": 2.295409935790939e-06, "loss": 9.585332870483398, "loss_alignment": 0.78759765625, "loss_alignment_w": 0.393798828125, "loss_sft": 0.20539131481200457, "loss_total_v6": 0.599083349108696, "step": 5519 }, { "epoch": 1.2730666666666668, "grad_norm": 28.444686889648438, "learning_rate": 2.2937163628603437e-06, "loss": 8.915510177612305, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.1690815556794405, "loss_total_v6": 0.5572193786501884, "step": 5520 }, { "epoch": 1.2732972972972973, "grad_norm": 32.792755126953125, "learning_rate": 2.292023228949284e-06, "loss": 9.136282920837402, "loss_alignment": 0.77001953125, "loss_alignment_w": 0.385009765625, "loss_sft": 0.1856874842196703, "loss_total_v6": 0.5710176825523376, "step": 5521 }, { "epoch": 1.273527927927928, "grad_norm": 32.82509231567383, "learning_rate": 2.290330534332422e-06, "loss": 10.167302131652832, "loss_alignment": 0.8115234375, "loss_alignment_w": 0.40576171875, "loss_sft": 0.22975569032132626, "loss_total_v6": 0.635456383228302, "step": 5522 }, { "epoch": 1.2737585585585585, "grad_norm": 27.546306610107422, "learning_rate": 2.2886382792843537e-06, "loss": 9.051595687866211, "loss_alignment": 0.7822265625, "loss_alignment_w": 0.39111328125, "loss_sft": 0.1750082029029727, "loss_total_v6": 0.56572475284338, "step": 5523 }, { "epoch": 1.2739891891891892, "grad_norm": 28.371496200561523, "learning_rate": 2.2869464640795974e-06, "loss": 8.99425220489502, "loss_alignment": 0.71826171875, "loss_alignment_w": 0.359130859375, "loss_sft": 0.20369657222181559, "loss_total_v6": 0.5621407777070999, "step": 5524 }, { "epoch": 1.2742198198198198, "grad_norm": 28.919904708862305, "learning_rate": 2.2852550889926067e-06, "loss": 8.963157653808594, "loss_alignment": 0.7568359375, "loss_alignment_w": 0.37841796875, "loss_sft": 0.18203881196677685, "loss_total_v6": 0.5601973757147789, "step": 5525 }, { "epoch": 1.2744504504504506, "grad_norm": 28.79674530029297, "learning_rate": 2.2835641542977594e-06, "loss": 9.590325355529785, "loss_alignment": 0.826171875, "loss_alignment_w": 0.4130859375, "loss_sft": 0.18528701178729534, "loss_total_v6": 0.5993953049182892, "step": 5526 }, { "epoch": 1.2746810810810811, "grad_norm": 33.11351776123047, "learning_rate": 2.2818736602693614e-06, "loss": 9.596710205078125, "loss_alignment": 0.78271484375, "loss_alignment_w": 0.391357421875, "loss_sft": 0.20819279365241528, "loss_total_v6": 0.5997943431138992, "step": 5527 }, { "epoch": 1.2749117117117117, "grad_norm": 30.99070930480957, "learning_rate": 2.2801836071816476e-06, "loss": 9.1823091506958, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.190807169303298, "loss_total_v6": 0.5738943293690681, "step": 5528 }, { "epoch": 1.2751423423423422, "grad_norm": 31.5889835357666, "learning_rate": 2.2784939953087835e-06, "loss": 9.214987754821777, "loss_alignment": 0.7890625, "loss_alignment_w": 0.39453125, "loss_sft": 0.18073407001793385, "loss_total_v6": 0.575936708599329, "step": 5529 }, { "epoch": 1.275372972972973, "grad_norm": 33.89588928222656, "learning_rate": 2.2768048249248648e-06, "loss": 9.718151092529297, "loss_alignment": 0.78466796875, "loss_alignment_w": 0.392333984375, "loss_sft": 0.21482155844569206, "loss_total_v6": 0.6073844134807587, "step": 5530 }, { "epoch": 1.2756036036036036, "grad_norm": 33.03546142578125, "learning_rate": 2.2751160963039095e-06, "loss": 9.180486679077148, "loss_alignment": 0.77294921875, "loss_alignment_w": 0.386474609375, "loss_sft": 0.18751949071884155, "loss_total_v6": 0.5737804472446442, "step": 5531 }, { "epoch": 1.2758342342342344, "grad_norm": 34.07847213745117, "learning_rate": 2.273427809719867e-06, "loss": 9.291763305664062, "loss_alignment": 0.78564453125, "loss_alignment_w": 0.392822265625, "loss_sft": 0.18834022618830204, "loss_total_v6": 0.5807352513074875, "step": 5532 }, { "epoch": 1.276064864864865, "grad_norm": 31.02229881286621, "learning_rate": 2.271739965446616e-06, "loss": 9.350567817687988, "loss_alignment": 0.77978515625, "loss_alignment_w": 0.389892578125, "loss_sft": 0.19470098614692688, "loss_total_v6": 0.5844104886054993, "step": 5533 }, { "epoch": 1.2762954954954955, "grad_norm": 28.426246643066406, "learning_rate": 2.2700525637579603e-06, "loss": 8.75069808959961, "loss_alignment": 0.7529296875, "loss_alignment_w": 0.37646484375, "loss_sft": 0.17039281129837036, "loss_total_v6": 0.5469186678528786, "step": 5534 }, { "epoch": 1.276526126126126, "grad_norm": 31.101160049438477, "learning_rate": 2.2683656049276352e-06, "loss": 9.105953216552734, "loss_alignment": 0.787109375, "loss_alignment_w": 0.3935546875, "loss_sft": 0.17486551590263844, "loss_total_v6": 0.5691220909357071, "step": 5535 }, { "epoch": 1.2767567567567568, "grad_norm": 26.977964401245117, "learning_rate": 2.2666790892293063e-06, "loss": 9.218173027038574, "loss_alignment": 0.7607421875, "loss_alignment_w": 0.38037109375, "loss_sft": 0.19518486596643925, "loss_total_v6": 0.5761357992887497, "step": 5536 }, { "epoch": 1.2769873873873874, "grad_norm": 33.958465576171875, "learning_rate": 2.264993016936561e-06, "loss": 9.619085311889648, "loss_alignment": 0.7900390625, "loss_alignment_w": 0.39501953125, "loss_sft": 0.2060511838644743, "loss_total_v6": 0.6011927872896194, "step": 5537 }, { "epoch": 1.2772180180180182, "grad_norm": 36.36820983886719, "learning_rate": 2.263307388322918e-06, "loss": 9.113327980041504, "loss_alignment": 0.76416015625, "loss_alignment_w": 0.382080078125, "loss_sft": 0.18710621260106564, "loss_total_v6": 0.5695830285549164, "step": 5538 }, { "epoch": 1.2774486486486487, "grad_norm": 29.67780876159668, "learning_rate": 2.261622203661822e-06, "loss": 9.353588104248047, "loss_alignment": 0.76611328125, "loss_alignment_w": 0.383056640625, "loss_sft": 0.20181730203330517, "loss_total_v6": 0.5845993012189865, "step": 5539 }, { "epoch": 1.2776792792792793, "grad_norm": 34.21244430541992, "learning_rate": 2.2599374632266514e-06, "loss": 9.143803596496582, "loss_alignment": 0.75146484375, "loss_alignment_w": 0.375732421875, "loss_sft": 0.19633514992892742, "loss_total_v6": 0.571487732231617, "step": 5540 }, { "epoch": 1.2779099099099098, "grad_norm": 28.166475296020508, "learning_rate": 2.2582531672907048e-06, "loss": 9.120452880859375, "loss_alignment": 0.7607421875, "loss_alignment_w": 0.38037109375, "loss_sft": 0.19008450396358967, "loss_total_v6": 0.5700283274054527, "step": 5541 }, { "epoch": 1.2781405405405406, "grad_norm": 32.1484489440918, "learning_rate": 2.256569316127216e-06, "loss": 9.655157089233398, "loss_alignment": 0.822265625, "loss_alignment_w": 0.4111328125, "loss_sft": 0.19283328671008348, "loss_total_v6": 0.6034473031759262, "step": 5542 }, { "epoch": 1.2783711711711712, "grad_norm": 29.11457633972168, "learning_rate": 2.254885910009341e-06, "loss": 9.085803031921387, "loss_alignment": 0.7607421875, "loss_alignment_w": 0.38037109375, "loss_sft": 0.18677445128560066, "loss_total_v6": 0.5678626969456673, "step": 5543 }, { "epoch": 1.278601801801802, "grad_norm": 33.61402893066406, "learning_rate": 2.2532029492101674e-06, "loss": 9.11470890045166, "loss_alignment": 0.783203125, "loss_alignment_w": 0.3916015625, "loss_sft": 0.17799146380275488, "loss_total_v6": 0.56966932117939, "step": 5544 }, { "epoch": 1.2788324324324325, "grad_norm": 31.25881004333496, "learning_rate": 2.251520434002708e-06, "loss": 8.925742149353027, "loss_alignment": 0.7578125, "loss_alignment_w": 0.37890625, "loss_sft": 0.17808285914361477, "loss_total_v6": 0.557858869433403, "step": 5545 }, { "epoch": 1.279063063063063, "grad_norm": 27.762784957885742, "learning_rate": 2.2498383646599046e-06, "loss": 9.688755989074707, "loss_alignment": 0.830078125, "loss_alignment_w": 0.4150390625, "loss_sft": 0.19021829590201378, "loss_total_v6": 0.6055472493171692, "step": 5546 }, { "epoch": 1.2792936936936936, "grad_norm": 26.730548858642578, "learning_rate": 2.2481567414546246e-06, "loss": 9.357372283935547, "loss_alignment": 0.759765625, "loss_alignment_w": 0.3798828125, "loss_sft": 0.2057311963289976, "loss_total_v6": 0.5848357900977135, "step": 5547 }, { "epoch": 1.2795243243243244, "grad_norm": 32.03850173950195, "learning_rate": 2.2464755646596658e-06, "loss": 9.916208267211914, "loss_alignment": 0.83642578125, "loss_alignment_w": 0.418212890625, "loss_sft": 0.2020078618079424, "loss_total_v6": 0.6197630017995834, "step": 5548 }, { "epoch": 1.279754954954955, "grad_norm": 34.421348571777344, "learning_rate": 2.244794834547755e-06, "loss": 9.120279312133789, "loss_alignment": 0.76953125, "loss_alignment_w": 0.384765625, "loss_sft": 0.18512975424528122, "loss_total_v6": 0.5700174421072006, "step": 5549 }, { "epoch": 1.2799855855855855, "grad_norm": 29.889812469482422, "learning_rate": 2.243114551391542e-06, "loss": 9.418647766113281, "loss_alignment": 0.79638671875, "loss_alignment_w": 0.398193359375, "loss_sft": 0.19135711062699556, "loss_total_v6": 0.5886654630303383, "step": 5550 }, { "epoch": 1.2802162162162163, "grad_norm": 30.30117416381836, "learning_rate": 2.241434715463606e-06, "loss": 9.009197235107422, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.17653919383883476, "loss_total_v6": 0.5630748197436333, "step": 5551 }, { "epoch": 1.2804468468468468, "grad_norm": 29.34848403930664, "learning_rate": 2.2397553270364546e-06, "loss": 9.307842254638672, "loss_alignment": 0.77490234375, "loss_alignment_w": 0.387451171875, "loss_sft": 0.1940905787050724, "loss_total_v6": 0.5817401111125946, "step": 5552 }, { "epoch": 1.2806774774774774, "grad_norm": 37.43506622314453, "learning_rate": 2.2380763863825196e-06, "loss": 9.204116821289062, "loss_alignment": 0.77880859375, "loss_alignment_w": 0.389404296875, "loss_sft": 0.1855020383372903, "loss_total_v6": 0.5752572789788246, "step": 5553 }, { "epoch": 1.2809081081081082, "grad_norm": 30.083669662475586, "learning_rate": 2.236397893774164e-06, "loss": 9.494627952575684, "loss_alignment": 0.77392578125, "loss_alignment_w": 0.386962890625, "loss_sft": 0.206878574565053, "loss_total_v6": 0.5934142023324966, "step": 5554 }, { "epoch": 1.2811387387387387, "grad_norm": 28.848493576049805, "learning_rate": 2.23471984948368e-06, "loss": 8.968420028686523, "loss_alignment": 0.7548828125, "loss_alignment_w": 0.37744140625, "loss_sft": 0.18239817023277283, "loss_total_v6": 0.5605262145400047, "step": 5555 }, { "epoch": 1.2813693693693693, "grad_norm": 30.078645706176758, "learning_rate": 2.23304225378328e-06, "loss": 9.020532608032227, "loss_alignment": 0.75830078125, "loss_alignment_w": 0.379150390625, "loss_sft": 0.1848922586068511, "loss_total_v6": 0.5637832470238209, "step": 5556 }, { "epoch": 1.2816, "grad_norm": 28.66029167175293, "learning_rate": 2.2313651069451094e-06, "loss": 9.646411895751953, "loss_alignment": 0.794921875, "loss_alignment_w": 0.3974609375, "loss_sft": 0.20518040098249912, "loss_total_v6": 0.6029007360339165, "step": 5557 }, { "epoch": 1.2818306306306306, "grad_norm": 26.788206100463867, "learning_rate": 2.2296884092412356e-06, "loss": 9.187944412231445, "loss_alignment": 0.77001953125, "loss_alignment_w": 0.385009765625, "loss_sft": 0.18934358470141888, "loss_total_v6": 0.5742465406656265, "step": 5558 }, { "epoch": 1.2820612612612612, "grad_norm": 28.479530334472656, "learning_rate": 2.2280121609436605e-06, "loss": 9.264598846435547, "loss_alignment": 0.7822265625, "loss_alignment_w": 0.39111328125, "loss_sft": 0.1883819354698062, "loss_total_v6": 0.5790374577045441, "step": 5559 }, { "epoch": 1.282291891891892, "grad_norm": 29.529510498046875, "learning_rate": 2.2263363623243058e-06, "loss": 9.643081665039062, "loss_alignment": 0.80322265625, "loss_alignment_w": 0.401611328125, "loss_sft": 0.2003946267068386, "loss_total_v6": 0.6026926040649414, "step": 5560 }, { "epoch": 1.2825225225225225, "grad_norm": 26.20588493347168, "learning_rate": 2.224661013655026e-06, "loss": 9.710081100463867, "loss_alignment": 0.7861328125, "loss_alignment_w": 0.39306640625, "loss_sft": 0.21442406252026558, "loss_total_v6": 0.6068801209330559, "step": 5561 }, { "epoch": 1.282753153153153, "grad_norm": 32.05452346801758, "learning_rate": 2.2229861152075983e-06, "loss": 9.51618766784668, "loss_alignment": 0.8076171875, "loss_alignment_w": 0.40380859375, "loss_sft": 0.19145673885941505, "loss_total_v6": 0.5947617515921593, "step": 5562 }, { "epoch": 1.2829837837837839, "grad_norm": 32.951072692871094, "learning_rate": 2.2213116672537278e-06, "loss": 8.973381042480469, "loss_alignment": 0.79052734375, "loss_alignment_w": 0.395263671875, "loss_sft": 0.1645045755431056, "loss_total_v6": 0.5608363524079323, "step": 5563 }, { "epoch": 1.2832144144144144, "grad_norm": 28.75135040283203, "learning_rate": 2.2196376700650495e-06, "loss": 9.526803016662598, "loss_alignment": 0.7958984375, "loss_alignment_w": 0.39794921875, "loss_sft": 0.1964078452438116, "loss_total_v6": 0.5954251810908318, "step": 5564 }, { "epoch": 1.283445045045045, "grad_norm": 31.47119140625, "learning_rate": 2.217964123913123e-06, "loss": 9.609570503234863, "loss_alignment": 0.78564453125, "loss_alignment_w": 0.392822265625, "loss_sft": 0.20779111050069332, "loss_total_v6": 0.6005981341004372, "step": 5565 }, { "epoch": 1.2836756756756758, "grad_norm": 31.56613540649414, "learning_rate": 2.2162910290694307e-06, "loss": 9.692927360534668, "loss_alignment": 0.8203125, "loss_alignment_w": 0.41015625, "loss_sft": 0.19449208490550518, "loss_total_v6": 0.6058079600334167, "step": 5566 }, { "epoch": 1.2839063063063063, "grad_norm": 31.00172233581543, "learning_rate": 2.2146183858053916e-06, "loss": 9.015422821044922, "loss_alignment": 0.76708984375, "loss_alignment_w": 0.383544921875, "loss_sft": 0.17984270863234997, "loss_total_v6": 0.5634639263153076, "step": 5567 }, { "epoch": 1.2841369369369369, "grad_norm": 33.562660217285156, "learning_rate": 2.2129461943923407e-06, "loss": 9.54629135131836, "loss_alignment": 0.77734375, "loss_alignment_w": 0.388671875, "loss_sft": 0.20725415833294392, "loss_total_v6": 0.5966431871056557, "step": 5568 }, { "epoch": 1.2843675675675676, "grad_norm": 29.002426147460938, "learning_rate": 2.2112744551015496e-06, "loss": 9.264019012451172, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.191305885091424, "loss_total_v6": 0.5790011882781982, "step": 5569 }, { "epoch": 1.2845981981981982, "grad_norm": 30.01369285583496, "learning_rate": 2.209603168204209e-06, "loss": 9.349920272827148, "loss_alignment": 0.75146484375, "loss_alignment_w": 0.375732421875, "loss_sft": 0.20813401602208614, "loss_total_v6": 0.5843699797987938, "step": 5570 }, { "epoch": 1.2848288288288288, "grad_norm": 27.72995948791504, "learning_rate": 2.2079323339714387e-06, "loss": 9.598335266113281, "loss_alignment": 0.7900390625, "loss_alignment_w": 0.39501953125, "loss_sft": 0.2064480446279049, "loss_total_v6": 0.5998959168791771, "step": 5571 }, { "epoch": 1.2850594594594595, "grad_norm": 28.193798065185547, "learning_rate": 2.206261952674284e-06, "loss": 9.031228065490723, "loss_alignment": 0.78271484375, "loss_alignment_w": 0.391357421875, "loss_sft": 0.17315534874796867, "loss_total_v6": 0.5644517317414284, "step": 5572 }, { "epoch": 1.28529009009009, "grad_norm": 30.955036163330078, "learning_rate": 2.20459202458372e-06, "loss": 8.830352783203125, "loss_alignment": 0.7548828125, "loss_alignment_w": 0.37744140625, "loss_sft": 0.1751728132367134, "loss_total_v6": 0.5518970638513565, "step": 5573 }, { "epoch": 1.2855207207207207, "grad_norm": 32.64838409423828, "learning_rate": 2.202922549970647e-06, "loss": 9.03934383392334, "loss_alignment": 0.7783203125, "loss_alignment_w": 0.38916015625, "loss_sft": 0.1773704970255494, "loss_total_v6": 0.5649589747190475, "step": 5574 }, { "epoch": 1.2857513513513514, "grad_norm": 30.229171752929688, "learning_rate": 2.20125352910589e-06, "loss": 9.150161743164062, "loss_alignment": 0.7548828125, "loss_alignment_w": 0.37744140625, "loss_sft": 0.19436737895011902, "loss_total_v6": 0.5718850716948509, "step": 5575 }, { "epoch": 1.285981981981982, "grad_norm": 33.81978988647461, "learning_rate": 2.1995849622602017e-06, "loss": 9.352412223815918, "loss_alignment": 0.7939453125, "loss_alignment_w": 0.39697265625, "loss_sft": 0.18625614047050476, "loss_total_v6": 0.5845257863402367, "step": 5576 }, { "epoch": 1.2862126126126125, "grad_norm": 32.191017150878906, "learning_rate": 2.1979168497042602e-06, "loss": 9.036575317382812, "loss_alignment": 0.76123046875, "loss_alignment_w": 0.380615234375, "loss_sft": 0.1843691449612379, "loss_total_v6": 0.5647859871387482, "step": 5577 }, { "epoch": 1.286443243243243, "grad_norm": 30.151254653930664, "learning_rate": 2.1962491917086685e-06, "loss": 9.193603515625, "loss_alignment": 0.8115234375, "loss_alignment_w": 0.40576171875, "loss_sft": 0.16918945591896772, "loss_total_v6": 0.5746002197265625, "step": 5578 }, { "epoch": 1.2866738738738739, "grad_norm": 30.938961029052734, "learning_rate": 2.1945819885439608e-06, "loss": 9.114757537841797, "loss_alignment": 0.798828125, "loss_alignment_w": 0.3994140625, "loss_sft": 0.17092968337237835, "loss_total_v6": 0.5696723386645317, "step": 5579 }, { "epoch": 1.2869045045045044, "grad_norm": 31.7398681640625, "learning_rate": 2.192915240480596e-06, "loss": 9.825902938842773, "loss_alignment": 0.787109375, "loss_alignment_w": 0.3935546875, "loss_sft": 0.22089995071291924, "loss_total_v6": 0.6141189411282539, "step": 5580 }, { "epoch": 1.2871351351351352, "grad_norm": 31.402549743652344, "learning_rate": 2.191248947788955e-06, "loss": 9.331075668334961, "loss_alignment": 0.78125, "loss_alignment_w": 0.390625, "loss_sft": 0.1936353612691164, "loss_total_v6": 0.5831922367215157, "step": 5581 }, { "epoch": 1.2873657657657658, "grad_norm": 32.58018112182617, "learning_rate": 2.1895831107393485e-06, "loss": 9.016358375549316, "loss_alignment": 0.75146484375, "loss_alignment_w": 0.375732421875, "loss_sft": 0.18899542279541492, "loss_total_v6": 0.5635223984718323, "step": 5582 }, { "epoch": 1.2875963963963963, "grad_norm": 32.97007369995117, "learning_rate": 2.187917729602011e-06, "loss": 9.442657470703125, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.2013416923582554, "loss_total_v6": 0.5901661440730095, "step": 5583 }, { "epoch": 1.287827027027027, "grad_norm": 27.36176109313965, "learning_rate": 2.186252804647107e-06, "loss": 9.167031288146973, "loss_alignment": 0.76513671875, "loss_alignment_w": 0.382568359375, "loss_sft": 0.1905084252357483, "loss_total_v6": 0.5729394480586052, "step": 5584 }, { "epoch": 1.2880576576576577, "grad_norm": 30.106765747070312, "learning_rate": 2.1845883361447218e-06, "loss": 8.918966293334961, "loss_alignment": 0.78515625, "loss_alignment_w": 0.392578125, "loss_sft": 0.16673410683870316, "loss_total_v6": 0.5574353970587254, "step": 5585 }, { "epoch": 1.2882882882882882, "grad_norm": 27.50729751586914, "learning_rate": 2.1829243243648718e-06, "loss": 9.207799911499023, "loss_alignment": 0.771484375, "loss_alignment_w": 0.3857421875, "loss_sft": 0.18948588520288467, "loss_total_v6": 0.5754874721169472, "step": 5586 }, { "epoch": 1.288518918918919, "grad_norm": 30.52655792236328, "learning_rate": 2.1812607695774944e-06, "loss": 9.336250305175781, "loss_alignment": 0.80615234375, "loss_alignment_w": 0.403076171875, "loss_sft": 0.17998171597719193, "loss_total_v6": 0.5835156217217445, "step": 5587 }, { "epoch": 1.2887495495495496, "grad_norm": 27.66676902770996, "learning_rate": 2.179597672052458e-06, "loss": 9.807281494140625, "loss_alignment": 0.8291015625, "loss_alignment_w": 0.41455078125, "loss_sft": 0.19814492017030716, "loss_total_v6": 0.6129550933837891, "step": 5588 }, { "epoch": 1.2889801801801801, "grad_norm": 29.72879981994629, "learning_rate": 2.177935032059553e-06, "loss": 9.34189224243164, "loss_alignment": 0.7734375, "loss_alignment_w": 0.38671875, "loss_sft": 0.19772933796048164, "loss_total_v6": 0.5838682278990746, "step": 5589 }, { "epoch": 1.2892108108108107, "grad_norm": 28.174318313598633, "learning_rate": 2.1762728498684967e-06, "loss": 9.043783187866211, "loss_alignment": 0.77001953125, "loss_alignment_w": 0.385009765625, "loss_sft": 0.18092861026525497, "loss_total_v6": 0.56523647159338, "step": 5590 }, { "epoch": 1.2894414414414415, "grad_norm": 28.998802185058594, "learning_rate": 2.1746111257489295e-06, "loss": 9.37936782836914, "loss_alignment": 0.77587890625, "loss_alignment_w": 0.387939453125, "loss_sft": 0.19737070053815842, "loss_total_v6": 0.5862104520201683, "step": 5591 }, { "epoch": 1.289672072072072, "grad_norm": 32.645713806152344, "learning_rate": 2.172949859970422e-06, "loss": 10.121808052062988, "loss_alignment": 0.8115234375, "loss_alignment_w": 0.40576171875, "loss_sft": 0.22666824236512184, "loss_total_v6": 0.632613018155098, "step": 5592 }, { "epoch": 1.2899027027027028, "grad_norm": 30.007232666015625, "learning_rate": 2.171289052802471e-06, "loss": 9.099748611450195, "loss_alignment": 0.75390625, "loss_alignment_w": 0.376953125, "loss_sft": 0.19143021292984486, "loss_total_v6": 0.5687342882156372, "step": 5593 }, { "epoch": 1.2901333333333334, "grad_norm": 32.13261795043945, "learning_rate": 2.1696287045144936e-06, "loss": 9.632888793945312, "loss_alignment": 0.80859375, "loss_alignment_w": 0.404296875, "loss_sft": 0.197941767051816, "loss_total_v6": 0.6020555421710014, "step": 5594 }, { "epoch": 1.290363963963964, "grad_norm": 38.004940032958984, "learning_rate": 2.1679688153758373e-06, "loss": 9.2568359375, "loss_alignment": 0.76904296875, "loss_alignment_w": 0.384521484375, "loss_sft": 0.1934814415872097, "loss_total_v6": 0.5785522386431694, "step": 5595 }, { "epoch": 1.2905945945945945, "grad_norm": 32.038516998291016, "learning_rate": 2.166309385655771e-06, "loss": 9.40980339050293, "loss_alignment": 0.806640625, "loss_alignment_w": 0.4033203125, "loss_sft": 0.1854179948568344, "loss_total_v6": 0.5881126895546913, "step": 5596 }, { "epoch": 1.2908252252252252, "grad_norm": 30.339332580566406, "learning_rate": 2.16465041562349e-06, "loss": 9.432073593139648, "loss_alignment": 0.77783203125, "loss_alignment_w": 0.388916015625, "loss_sft": 0.20017664693295956, "loss_total_v6": 0.5895046219229698, "step": 5597 }, { "epoch": 1.2910558558558558, "grad_norm": 32.38603973388672, "learning_rate": 2.1629919055481185e-06, "loss": 9.582447052001953, "loss_alignment": 0.7861328125, "loss_alignment_w": 0.39306640625, "loss_sft": 0.20530243404209614, "loss_total_v6": 0.5989029183983803, "step": 5598 }, { "epoch": 1.2912864864864866, "grad_norm": 31.377826690673828, "learning_rate": 2.1613338556987057e-06, "loss": 9.499481201171875, "loss_alignment": 0.79296875, "loss_alignment_w": 0.396484375, "loss_sft": 0.19767564721405506, "loss_total_v6": 0.593717522919178, "step": 5599 }, { "epoch": 1.2915171171171171, "grad_norm": 27.262760162353516, "learning_rate": 2.159676266344222e-06, "loss": 9.179695129394531, "loss_alignment": 0.74755859375, "loss_alignment_w": 0.373779296875, "loss_sft": 0.19949388317763805, "loss_total_v6": 0.5737309157848358, "step": 5600 }, { "epoch": 1.2917477477477477, "grad_norm": 32.548606872558594, "learning_rate": 2.158019137753565e-06, "loss": 9.182746887207031, "loss_alignment": 0.775390625, "loss_alignment_w": 0.3876953125, "loss_sft": 0.18529559299349785, "loss_total_v6": 0.5739216953516006, "step": 5601 }, { "epoch": 1.2919783783783783, "grad_norm": 31.28896713256836, "learning_rate": 2.1563624701955573e-06, "loss": 9.135116577148438, "loss_alignment": 0.7705078125, "loss_alignment_w": 0.38525390625, "loss_sft": 0.1863927487283945, "loss_total_v6": 0.5709447339177132, "step": 5602 }, { "epoch": 1.292209009009009, "grad_norm": 29.008028030395508, "learning_rate": 2.1547062639389494e-06, "loss": 9.132837295532227, "loss_alignment": 0.79443359375, "loss_alignment_w": 0.397216796875, "loss_sft": 0.17358552478253841, "loss_total_v6": 0.5708023235201836, "step": 5603 }, { "epoch": 1.2924396396396396, "grad_norm": 29.1551456451416, "learning_rate": 2.153050519252412e-06, "loss": 9.385205268859863, "loss_alignment": 0.79296875, "loss_alignment_w": 0.396484375, "loss_sft": 0.1893280390650034, "loss_total_v6": 0.5865753255784512, "step": 5604 }, { "epoch": 1.2926702702702704, "grad_norm": 39.871334075927734, "learning_rate": 2.1513952364045475e-06, "loss": 9.515284538269043, "loss_alignment": 0.77880859375, "loss_alignment_w": 0.389404296875, "loss_sft": 0.20537729375064373, "loss_total_v6": 0.5947052836418152, "step": 5605 }, { "epoch": 1.292900900900901, "grad_norm": 31.71645164489746, "learning_rate": 2.1497404156638783e-06, "loss": 9.961013793945312, "loss_alignment": 0.8564453125, "loss_alignment_w": 0.42822265625, "loss_sft": 0.19403556175529957, "loss_total_v6": 0.6225633844733238, "step": 5606 }, { "epoch": 1.2931315315315315, "grad_norm": 27.330583572387695, "learning_rate": 2.1480860572988505e-06, "loss": 9.363344192504883, "loss_alignment": 0.78759765625, "loss_alignment_w": 0.393798828125, "loss_sft": 0.19159330427646637, "loss_total_v6": 0.585209034383297, "step": 5607 }, { "epoch": 1.293362162162162, "grad_norm": 26.317792892456055, "learning_rate": 2.146432161577842e-06, "loss": 8.849533081054688, "loss_alignment": 0.759765625, "loss_alignment_w": 0.3798828125, "loss_sft": 0.1738690882921219, "loss_total_v6": 0.5530957579612732, "step": 5608 }, { "epoch": 1.2935927927927928, "grad_norm": 31.53095054626465, "learning_rate": 2.14477872876915e-06, "loss": 9.833518028259277, "loss_alignment": 0.8017578125, "loss_alignment_w": 0.40087890625, "loss_sft": 0.2134107854217291, "loss_total_v6": 0.614594891667366, "step": 5609 }, { "epoch": 1.2938234234234234, "grad_norm": 29.064685821533203, "learning_rate": 2.1431257591409952e-06, "loss": 9.071630477905273, "loss_alignment": 0.76806640625, "loss_alignment_w": 0.384033203125, "loss_sft": 0.18323364108800888, "loss_total_v6": 0.5669769048690796, "step": 5610 }, { "epoch": 1.2940540540540542, "grad_norm": 31.679243087768555, "learning_rate": 2.1414732529615316e-06, "loss": 9.475547790527344, "loss_alignment": 0.7626953125, "loss_alignment_w": 0.38134765625, "loss_sft": 0.21146917343139648, "loss_total_v6": 0.5922217518091202, "step": 5611 }, { "epoch": 1.2942846846846847, "grad_norm": 35.36518096923828, "learning_rate": 2.1398212104988273e-06, "loss": 9.17386531829834, "loss_alignment": 0.78564453125, "loss_alignment_w": 0.392822265625, "loss_sft": 0.18008660152554512, "loss_total_v6": 0.5733666270971298, "step": 5612 }, { "epoch": 1.2945153153153153, "grad_norm": 29.268951416015625, "learning_rate": 2.138169632020885e-06, "loss": 9.191656112670898, "loss_alignment": 0.78955078125, "loss_alignment_w": 0.394775390625, "loss_sft": 0.1788181271404028, "loss_total_v6": 0.5744785293936729, "step": 5613 }, { "epoch": 1.2947459459459458, "grad_norm": 30.846872329711914, "learning_rate": 2.1365185177956248e-06, "loss": 9.46829891204834, "loss_alignment": 0.75634765625, "loss_alignment_w": 0.378173828125, "loss_sft": 0.214174659922719, "loss_total_v6": 0.5917686596512794, "step": 5614 }, { "epoch": 1.2949765765765766, "grad_norm": 29.000822067260742, "learning_rate": 2.134867868090895e-06, "loss": 8.922344207763672, "loss_alignment": 0.7490234375, "loss_alignment_w": 0.37451171875, "loss_sft": 0.182875357568264, "loss_total_v6": 0.5576464831829071, "step": 5615 }, { "epoch": 1.2952072072072072, "grad_norm": 31.396501541137695, "learning_rate": 2.133217683174466e-06, "loss": 9.002506256103516, "loss_alignment": 0.7763671875, "loss_alignment_w": 0.38818359375, "loss_sft": 0.1746408585458994, "loss_total_v6": 0.5626566112041473, "step": 5616 }, { "epoch": 1.295437837837838, "grad_norm": 32.4105110168457, "learning_rate": 2.131567963314036e-06, "loss": 9.91934585571289, "loss_alignment": 0.79541015625, "loss_alignment_w": 0.397705078125, "loss_sft": 0.22217772156000137, "loss_total_v6": 0.6199590861797333, "step": 5617 }, { "epoch": 1.2956684684684685, "grad_norm": 28.18872833251953, "learning_rate": 2.1299187087772284e-06, "loss": 9.003536224365234, "loss_alignment": 0.77978515625, "loss_alignment_w": 0.389892578125, "loss_sft": 0.17288947012275457, "loss_total_v6": 0.5627210102975368, "step": 5618 }, { "epoch": 1.295899099099099, "grad_norm": 31.85196876525879, "learning_rate": 2.128269919831587e-06, "loss": 9.143021583557129, "loss_alignment": 0.783203125, "loss_alignment_w": 0.3916015625, "loss_sft": 0.18020351976156235, "loss_total_v6": 0.57143884152174, "step": 5619 }, { "epoch": 1.2961297297297296, "grad_norm": 27.8999080657959, "learning_rate": 2.1266215967445823e-06, "loss": 8.861786842346191, "loss_alignment": 0.7841796875, "loss_alignment_w": 0.39208984375, "loss_sft": 0.1625195387750864, "loss_total_v6": 0.55386171489954, "step": 5620 }, { "epoch": 1.2963603603603604, "grad_norm": 35.54248046875, "learning_rate": 2.124973739783609e-06, "loss": 9.588139533996582, "loss_alignment": 0.78173828125, "loss_alignment_w": 0.390869140625, "loss_sft": 0.20875579491257668, "loss_total_v6": 0.5992587506771088, "step": 5621 }, { "epoch": 1.296590990990991, "grad_norm": 29.32504653930664, "learning_rate": 2.1233263492159843e-06, "loss": 8.765290260314941, "loss_alignment": 0.7705078125, "loss_alignment_w": 0.38525390625, "loss_sft": 0.16356853488832712, "loss_total_v6": 0.5478306151926517, "step": 5622 }, { "epoch": 1.2968216216216217, "grad_norm": 30.96978759765625, "learning_rate": 2.121679425308954e-06, "loss": 8.654691696166992, "loss_alignment": 0.74853515625, "loss_alignment_w": 0.374267578125, "loss_sft": 0.1672151880338788, "loss_total_v6": 0.5409182123839855, "step": 5623 }, { "epoch": 1.2970522522522523, "grad_norm": 33.554141998291016, "learning_rate": 2.120032968329687e-06, "loss": 9.168041229248047, "loss_alignment": 0.8134765625, "loss_alignment_w": 0.40673828125, "loss_sft": 0.16771391965448856, "loss_total_v6": 0.5730025991797447, "step": 5624 }, { "epoch": 1.2972828828828828, "grad_norm": 26.61310386657715, "learning_rate": 2.1183869785452744e-06, "loss": 8.99693489074707, "loss_alignment": 0.75634765625, "loss_alignment_w": 0.378173828125, "loss_sft": 0.1829291693866253, "loss_total_v6": 0.5623084232211113, "step": 5625 }, { "epoch": 1.2975135135135134, "grad_norm": 29.30427360534668, "learning_rate": 2.116741456222731e-06, "loss": 9.168862342834473, "loss_alignment": 0.78857421875, "loss_alignment_w": 0.394287109375, "loss_sft": 0.1787210199981928, "loss_total_v6": 0.5730538889765739, "step": 5626 }, { "epoch": 1.2977441441441442, "grad_norm": 29.595834732055664, "learning_rate": 2.115096401628997e-06, "loss": 9.835224151611328, "loss_alignment": 0.79638671875, "loss_alignment_w": 0.398193359375, "loss_sft": 0.21458550542593002, "loss_total_v6": 0.6147014871239662, "step": 5627 }, { "epoch": 1.2979747747747747, "grad_norm": 30.875022888183594, "learning_rate": 2.113451815030939e-06, "loss": 9.178403854370117, "loss_alignment": 0.775390625, "loss_alignment_w": 0.3876953125, "loss_sft": 0.18506985530257225, "loss_total_v6": 0.5736501887440681, "step": 5628 }, { "epoch": 1.2982054054054055, "grad_norm": 29.412240982055664, "learning_rate": 2.111807696695343e-06, "loss": 8.978519439697266, "loss_alignment": 0.791015625, "loss_alignment_w": 0.3955078125, "loss_sft": 0.1655276156961918, "loss_total_v6": 0.5611574724316597, "step": 5629 }, { "epoch": 1.298436036036036, "grad_norm": 27.800281524658203, "learning_rate": 2.1101640468889255e-06, "loss": 8.860177040100098, "loss_alignment": 0.7490234375, "loss_alignment_w": 0.37451171875, "loss_sft": 0.1804853230714798, "loss_total_v6": 0.5537610799074173, "step": 5630 }, { "epoch": 1.2986666666666666, "grad_norm": 34.18750762939453, "learning_rate": 2.1085208658783206e-06, "loss": 9.392882347106934, "loss_alignment": 0.7578125, "loss_alignment_w": 0.37890625, "loss_sft": 0.208301511593163, "loss_total_v6": 0.5870551690459251, "step": 5631 }, { "epoch": 1.2988972972972972, "grad_norm": 41.996829986572266, "learning_rate": 2.1068781539300874e-06, "loss": 8.9736328125, "loss_alignment": 0.76513671875, "loss_alignment_w": 0.382568359375, "loss_sft": 0.1786956861615181, "loss_total_v6": 0.5608520582318306, "step": 5632 }, { "epoch": 1.299127927927928, "grad_norm": 33.3647346496582, "learning_rate": 2.1052359113107147e-06, "loss": 9.196131706237793, "loss_alignment": 0.75830078125, "loss_alignment_w": 0.379150390625, "loss_sft": 0.1962334495037794, "loss_total_v6": 0.5747582241892815, "step": 5633 }, { "epoch": 1.2993585585585585, "grad_norm": 31.76365852355957, "learning_rate": 2.103594138286607e-06, "loss": 9.298229217529297, "loss_alignment": 0.7626953125, "loss_alignment_w": 0.38134765625, "loss_sft": 0.19934913516044617, "loss_total_v6": 0.5811393111944199, "step": 5634 }, { "epoch": 1.2995891891891893, "grad_norm": 28.85186767578125, "learning_rate": 2.101952835124097e-06, "loss": 8.984920501708984, "loss_alignment": 0.75634765625, "loss_alignment_w": 0.378173828125, "loss_sft": 0.18397881090641022, "loss_total_v6": 0.5615575313568115, "step": 5635 }, { "epoch": 1.2998198198198199, "grad_norm": 30.913944244384766, "learning_rate": 2.1003120020894413e-06, "loss": 9.071998596191406, "loss_alignment": 0.74658203125, "loss_alignment_w": 0.373291015625, "loss_sft": 0.19456334691494703, "loss_total_v6": 0.5669998750090599, "step": 5636 }, { "epoch": 1.3000504504504504, "grad_norm": 31.773942947387695, "learning_rate": 2.0986716394488212e-06, "loss": 9.172496795654297, "loss_alignment": 0.79443359375, "loss_alignment_w": 0.397216796875, "loss_sft": 0.1756828222423792, "loss_total_v6": 0.5732810795307159, "step": 5637 }, { "epoch": 1.300281081081081, "grad_norm": 33.990577697753906, "learning_rate": 2.097031747468339e-06, "loss": 9.422780990600586, "loss_alignment": 0.76318359375, "loss_alignment_w": 0.381591796875, "loss_sft": 0.20733203180134296, "loss_total_v6": 0.588923841714859, "step": 5638 }, { "epoch": 1.3005117117117118, "grad_norm": 31.588680267333984, "learning_rate": 2.0953923264140214e-06, "loss": 9.369386672973633, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.19805922731757164, "loss_total_v6": 0.5855866819620132, "step": 5639 }, { "epoch": 1.3007423423423423, "grad_norm": 51.425636291503906, "learning_rate": 2.0937533765518187e-06, "loss": 9.401164054870605, "loss_alignment": 0.78515625, "loss_alignment_w": 0.392578125, "loss_sft": 0.1945368517190218, "loss_total_v6": 0.5875727459788322, "step": 5640 }, { "epoch": 1.300972972972973, "grad_norm": 28.808753967285156, "learning_rate": 2.0921148981476034e-06, "loss": 9.505195617675781, "loss_alignment": 0.78125, "loss_alignment_w": 0.390625, "loss_sft": 0.204075388610363, "loss_total_v6": 0.5940747857093811, "step": 5641 }, { "epoch": 1.3012036036036037, "grad_norm": 30.110090255737305, "learning_rate": 2.090476891467176e-06, "loss": 9.096272468566895, "loss_alignment": 0.77197265625, "loss_alignment_w": 0.385986328125, "loss_sft": 0.1831563487648964, "loss_total_v6": 0.5685170516371727, "step": 5642 }, { "epoch": 1.3014342342342342, "grad_norm": 30.2713680267334, "learning_rate": 2.088839356776258e-06, "loss": 9.679041862487793, "loss_alignment": 0.8076171875, "loss_alignment_w": 0.40380859375, "loss_sft": 0.20142148807644844, "loss_total_v6": 0.6049401462078094, "step": 5643 }, { "epoch": 1.3016648648648648, "grad_norm": 25.433242797851562, "learning_rate": 2.087202294340494e-06, "loss": 9.66423511505127, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.2169602196663618, "loss_total_v6": 0.6040146574378014, "step": 5644 }, { "epoch": 1.3018954954954955, "grad_norm": 29.65664291381836, "learning_rate": 2.0855657044254503e-06, "loss": 8.812034606933594, "loss_alignment": 0.7470703125, "loss_alignment_w": 0.37353515625, "loss_sft": 0.17836141027510166, "loss_total_v6": 0.5507521629333496, "step": 5645 }, { "epoch": 1.302126126126126, "grad_norm": 28.8093318939209, "learning_rate": 2.0839295872966177e-06, "loss": 9.152982711791992, "loss_alignment": 0.7734375, "loss_alignment_w": 0.38671875, "loss_sft": 0.1857851892709732, "loss_total_v6": 0.5720614269375801, "step": 5646 }, { "epoch": 1.3023567567567569, "grad_norm": 35.21427917480469, "learning_rate": 2.0822939432194134e-06, "loss": 9.560517311096191, "loss_alignment": 0.8046875, "loss_alignment_w": 0.40234375, "loss_sft": 0.19494441710412502, "loss_total_v6": 0.5975323170423508, "step": 5647 }, { "epoch": 1.3025873873873874, "grad_norm": 28.067951202392578, "learning_rate": 2.0806587724591725e-06, "loss": 9.006418228149414, "loss_alignment": 0.76318359375, "loss_alignment_w": 0.381591796875, "loss_sft": 0.18182813003659248, "loss_total_v6": 0.5629011616110802, "step": 5648 }, { "epoch": 1.302818018018018, "grad_norm": 27.500328063964844, "learning_rate": 2.0790240752811596e-06, "loss": 9.484217643737793, "loss_alignment": 0.78369140625, "loss_alignment_w": 0.391845703125, "loss_sft": 0.19971248134970665, "loss_total_v6": 0.5927636176347733, "step": 5649 }, { "epoch": 1.3030486486486486, "grad_norm": 33.07665252685547, "learning_rate": 2.077389851950557e-06, "loss": 9.087396621704102, "loss_alignment": 0.755859375, "loss_alignment_w": 0.3779296875, "loss_sft": 0.1900020968168974, "loss_total_v6": 0.567962296307087, "step": 5650 }, { "epoch": 1.3032792792792793, "grad_norm": 29.339126586914062, "learning_rate": 2.0757561027324708e-06, "loss": 9.716470718383789, "loss_alignment": 0.77490234375, "loss_alignment_w": 0.387451171875, "loss_sft": 0.21988931484520435, "loss_total_v6": 0.607279472053051, "step": 5651 }, { "epoch": 1.30350990990991, "grad_norm": 30.34438705444336, "learning_rate": 2.0741228278919347e-06, "loss": 9.533830642700195, "loss_alignment": 0.7900390625, "loss_alignment_w": 0.39501953125, "loss_sft": 0.2009822279214859, "loss_total_v6": 0.5958644226193428, "step": 5652 }, { "epoch": 1.3037405405405407, "grad_norm": 32.90998458862305, "learning_rate": 2.0724900276939004e-06, "loss": 9.477992057800293, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.20377891138195992, "loss_total_v6": 0.5923744887113571, "step": 5653 }, { "epoch": 1.3039711711711712, "grad_norm": 30.702421188354492, "learning_rate": 2.070857702403243e-06, "loss": 9.748270034790039, "loss_alignment": 0.77099609375, "loss_alignment_w": 0.385498046875, "loss_sft": 0.22378410957753658, "loss_total_v6": 0.6092668995261192, "step": 5654 }, { "epoch": 1.3042018018018018, "grad_norm": 30.473154067993164, "learning_rate": 2.069225852284766e-06, "loss": 9.046899795532227, "loss_alignment": 0.78857421875, "loss_alignment_w": 0.394287109375, "loss_sft": 0.17093049548566341, "loss_total_v6": 0.5654312074184418, "step": 5655 }, { "epoch": 1.3044324324324323, "grad_norm": 29.53094482421875, "learning_rate": 2.0675944776031875e-06, "loss": 8.929728507995605, "loss_alignment": 0.76513671875, "loss_alignment_w": 0.382568359375, "loss_sft": 0.17561596166342497, "loss_total_v6": 0.5581080168485641, "step": 5656 }, { "epoch": 1.3046630630630631, "grad_norm": 29.899293899536133, "learning_rate": 2.065963578623157e-06, "loss": 9.149736404418945, "loss_alignment": 0.77197265625, "loss_alignment_w": 0.385986328125, "loss_sft": 0.18629943393170834, "loss_total_v6": 0.5718585178256035, "step": 5657 }, { "epoch": 1.3048936936936937, "grad_norm": 28.900192260742188, "learning_rate": 2.0643331556092407e-06, "loss": 8.927343368530273, "loss_alignment": 0.75439453125, "loss_alignment_w": 0.377197265625, "loss_sft": 0.1803039237856865, "loss_total_v6": 0.5579589530825615, "step": 5658 }, { "epoch": 1.3051243243243245, "grad_norm": 31.69975471496582, "learning_rate": 2.062703208825929e-06, "loss": 8.982675552368164, "loss_alignment": 0.7587890625, "loss_alignment_w": 0.37939453125, "loss_sft": 0.18244991078972816, "loss_total_v6": 0.5614171996712685, "step": 5659 }, { "epoch": 1.305354954954955, "grad_norm": 28.829669952392578, "learning_rate": 2.061073738537635e-06, "loss": 9.660009384155273, "loss_alignment": 0.8056640625, "loss_alignment_w": 0.40283203125, "loss_sft": 0.20114741660654545, "loss_total_v6": 0.603750579059124, "step": 5660 }, { "epoch": 1.3055855855855856, "grad_norm": 34.31674575805664, "learning_rate": 2.0594447450086958e-06, "loss": 9.371379852294922, "loss_alignment": 0.80029296875, "loss_alignment_w": 0.400146484375, "loss_sft": 0.18612939678132534, "loss_total_v6": 0.5857112929224968, "step": 5661 }, { "epoch": 1.3058162162162161, "grad_norm": 35.06880569458008, "learning_rate": 2.057816228503373e-06, "loss": 9.637828826904297, "loss_alignment": 0.8203125, "loss_alignment_w": 0.41015625, "loss_sft": 0.19319984503090382, "loss_total_v6": 0.602364294230938, "step": 5662 }, { "epoch": 1.306046846846847, "grad_norm": 30.675153732299805, "learning_rate": 2.056188189285846e-06, "loss": 8.943099021911621, "loss_alignment": 0.7529296875, "loss_alignment_w": 0.37646484375, "loss_sft": 0.18214312847703695, "loss_total_v6": 0.5589436665177345, "step": 5663 }, { "epoch": 1.3062774774774775, "grad_norm": 31.50975227355957, "learning_rate": 2.054560627620219e-06, "loss": 9.098424911499023, "loss_alignment": 0.75830078125, "loss_alignment_w": 0.379150390625, "loss_sft": 0.18863142281770706, "loss_total_v6": 0.5686515718698502, "step": 5664 }, { "epoch": 1.3065081081081082, "grad_norm": 29.555431365966797, "learning_rate": 2.0529335437705193e-06, "loss": 9.10835075378418, "loss_alignment": 0.7578125, "loss_alignment_w": 0.37890625, "loss_sft": 0.18932809866964817, "loss_total_v6": 0.5692719519138336, "step": 5665 }, { "epoch": 1.3067387387387388, "grad_norm": 29.99942398071289, "learning_rate": 2.0513069380006943e-06, "loss": 9.60934829711914, "loss_alignment": 0.76806640625, "loss_alignment_w": 0.384033203125, "loss_sft": 0.21655108593404293, "loss_total_v6": 0.6005842834711075, "step": 5666 }, { "epoch": 1.3069693693693694, "grad_norm": 33.52456283569336, "learning_rate": 2.0496808105746167e-06, "loss": 9.21399974822998, "loss_alignment": 0.76806640625, "loss_alignment_w": 0.384033203125, "loss_sft": 0.19237583875656128, "loss_total_v6": 0.5758749954402447, "step": 5667 }, { "epoch": 1.3072, "grad_norm": 31.814197540283203, "learning_rate": 2.0480551617560834e-06, "loss": 9.53632640838623, "loss_alignment": 0.779296875, "loss_alignment_w": 0.3896484375, "loss_sft": 0.20667717792093754, "loss_total_v6": 0.5960204154253006, "step": 5668 }, { "epoch": 1.3074306306306307, "grad_norm": 28.540800094604492, "learning_rate": 2.046429991808809e-06, "loss": 9.206083297729492, "loss_alignment": 0.8154296875, "loss_alignment_w": 0.40771484375, "loss_sft": 0.16713132616132498, "loss_total_v6": 0.5753802433609962, "step": 5669 }, { "epoch": 1.3076612612612613, "grad_norm": 27.696002960205078, "learning_rate": 2.044805300996432e-06, "loss": 9.659235000610352, "loss_alignment": 0.78125, "loss_alignment_w": 0.390625, "loss_sft": 0.21361124515533447, "loss_total_v6": 0.6037021987140179, "step": 5670 }, { "epoch": 1.3078918918918918, "grad_norm": 36.40863037109375, "learning_rate": 2.043181089582511e-06, "loss": 9.320245742797852, "loss_alignment": 0.80322265625, "loss_alignment_w": 0.401611328125, "loss_sft": 0.18009530566632748, "loss_total_v6": 0.5825153365731239, "step": 5671 }, { "epoch": 1.3081225225225226, "grad_norm": 28.730792999267578, "learning_rate": 2.0415573578305343e-06, "loss": 9.088680267333984, "loss_alignment": 0.77490234375, "loss_alignment_w": 0.387451171875, "loss_sft": 0.18236134573817253, "loss_total_v6": 0.5680424943566322, "step": 5672 }, { "epoch": 1.3083531531531531, "grad_norm": 30.503225326538086, "learning_rate": 2.0399341060039023e-06, "loss": 9.011646270751953, "loss_alignment": 0.75341796875, "loss_alignment_w": 0.376708984375, "loss_sft": 0.18705292232334614, "loss_total_v6": 0.5632278621196747, "step": 5673 }, { "epoch": 1.3085837837837837, "grad_norm": 25.938953399658203, "learning_rate": 2.0383113343659477e-06, "loss": 9.77039909362793, "loss_alignment": 0.8203125, "loss_alignment_w": 0.41015625, "loss_sft": 0.2003411129117012, "loss_total_v6": 0.6106499582529068, "step": 5674 }, { "epoch": 1.3088144144144145, "grad_norm": 29.788402557373047, "learning_rate": 2.036689043179917e-06, "loss": 9.831440925598145, "loss_alignment": 0.76953125, "loss_alignment_w": 0.384765625, "loss_sft": 0.23053866811096668, "loss_total_v6": 0.6144650653004646, "step": 5675 }, { "epoch": 1.309045045045045, "grad_norm": 29.537946701049805, "learning_rate": 2.0350672327089813e-06, "loss": 9.780694007873535, "loss_alignment": 0.822265625, "loss_alignment_w": 0.4111328125, "loss_sft": 0.2004810068756342, "loss_total_v6": 0.6112933829426765, "step": 5676 }, { "epoch": 1.3092756756756756, "grad_norm": 27.809118270874023, "learning_rate": 2.033445903216238e-06, "loss": 9.034795761108398, "loss_alignment": 0.7666015625, "loss_alignment_w": 0.38330078125, "loss_sft": 0.18183168955147266, "loss_total_v6": 0.5646746978163719, "step": 5677 }, { "epoch": 1.3095063063063064, "grad_norm": 31.00922393798828, "learning_rate": 2.0318250549647e-06, "loss": 9.178499221801758, "loss_alignment": 0.7890625, "loss_alignment_w": 0.39453125, "loss_sft": 0.17837730888277292, "loss_total_v6": 0.5736562311649323, "step": 5678 }, { "epoch": 1.309736936936937, "grad_norm": 32.01834487915039, "learning_rate": 2.0302046882173043e-06, "loss": 9.198223114013672, "loss_alignment": 0.7802734375, "loss_alignment_w": 0.39013671875, "loss_sft": 0.1846758835017681, "loss_total_v6": 0.5748888999223709, "step": 5679 }, { "epoch": 1.3099675675675675, "grad_norm": 28.57377052307129, "learning_rate": 2.028584803236914e-06, "loss": 8.549860954284668, "loss_alignment": 0.7783203125, "loss_alignment_w": 0.38916015625, "loss_sft": 0.145480802282691, "loss_total_v6": 0.5343662835657597, "step": 5680 }, { "epoch": 1.3101981981981983, "grad_norm": 27.57318878173828, "learning_rate": 2.0269654002863065e-06, "loss": 9.819746017456055, "loss_alignment": 0.810546875, "loss_alignment_w": 0.4052734375, "loss_sft": 0.20958987437188625, "loss_total_v6": 0.6137341558933258, "step": 5681 }, { "epoch": 1.3104288288288288, "grad_norm": 27.2864990234375, "learning_rate": 2.0253464796281885e-06, "loss": 9.35581111907959, "loss_alignment": 0.77099609375, "loss_alignment_w": 0.385498046875, "loss_sft": 0.20015573129057884, "loss_total_v6": 0.5847382321953773, "step": 5682 }, { "epoch": 1.3106594594594594, "grad_norm": 28.64563751220703, "learning_rate": 2.0237280415251835e-06, "loss": 8.877853393554688, "loss_alignment": 0.76708984375, "loss_alignment_w": 0.383544921875, "loss_sft": 0.17080215271562338, "loss_total_v6": 0.554865874350071, "step": 5683 }, { "epoch": 1.3108900900900902, "grad_norm": 32.98578643798828, "learning_rate": 2.0221100862398374e-06, "loss": 9.59364128112793, "loss_alignment": 0.79931640625, "loss_alignment_w": 0.399658203125, "loss_sft": 0.19970020838081837, "loss_total_v6": 0.5996025651693344, "step": 5684 }, { "epoch": 1.3111207207207207, "grad_norm": 25.62026023864746, "learning_rate": 2.0204926140346204e-06, "loss": 9.224654197692871, "loss_alignment": 0.78759765625, "loss_alignment_w": 0.393798828125, "loss_sft": 0.1831387896090746, "loss_total_v6": 0.576540894806385, "step": 5685 }, { "epoch": 1.3113513513513513, "grad_norm": 34.764732360839844, "learning_rate": 2.0188756251719204e-06, "loss": 8.975027084350586, "loss_alignment": 0.76123046875, "loss_alignment_w": 0.380615234375, "loss_sft": 0.1803239556029439, "loss_total_v6": 0.5609391704201698, "step": 5686 }, { "epoch": 1.311581981981982, "grad_norm": 28.625951766967773, "learning_rate": 2.0172591199140525e-06, "loss": 9.150003433227539, "loss_alignment": 0.7431640625, "loss_alignment_w": 0.37158203125, "loss_sft": 0.19980494305491447, "loss_total_v6": 0.5718752592802048, "step": 5687 }, { "epoch": 1.3118126126126126, "grad_norm": 28.62319564819336, "learning_rate": 2.0156430985232466e-06, "loss": 9.276873588562012, "loss_alignment": 0.7236328125, "loss_alignment_w": 0.36181640625, "loss_sft": 0.21769827418029308, "loss_total_v6": 0.5798046067357063, "step": 5688 }, { "epoch": 1.3120432432432432, "grad_norm": 32.830474853515625, "learning_rate": 2.0140275612616595e-06, "loss": 9.14160442352295, "loss_alignment": 0.76904296875, "loss_alignment_w": 0.384521484375, "loss_sft": 0.18661513179540634, "loss_total_v6": 0.5713502503931522, "step": 5689 }, { "epoch": 1.312273873873874, "grad_norm": 26.677295684814453, "learning_rate": 2.0124125083913636e-06, "loss": 8.513221740722656, "loss_alignment": 0.7490234375, "loss_alignment_w": 0.37451171875, "loss_sft": 0.15747310686856508, "loss_total_v6": 0.5320763550698757, "step": 5690 }, { "epoch": 1.3125045045045045, "grad_norm": 32.148841857910156, "learning_rate": 2.01079794017436e-06, "loss": 8.898791313171387, "loss_alignment": 0.74560546875, "loss_alignment_w": 0.372802734375, "loss_sft": 0.18430253211408854, "loss_total_v6": 0.5561744719743729, "step": 5691 }, { "epoch": 1.312735135135135, "grad_norm": 28.612150192260742, "learning_rate": 2.0091838568725685e-06, "loss": 9.424600601196289, "loss_alignment": 0.794921875, "loss_alignment_w": 0.3974609375, "loss_sft": 0.1916071893647313, "loss_total_v6": 0.5890375971794128, "step": 5692 }, { "epoch": 1.3129657657657658, "grad_norm": 30.45772361755371, "learning_rate": 2.007570258747828e-06, "loss": 9.350414276123047, "loss_alignment": 0.7783203125, "loss_alignment_w": 0.38916015625, "loss_sft": 0.19513393938541412, "loss_total_v6": 0.5844009071588516, "step": 5693 }, { "epoch": 1.3131963963963964, "grad_norm": 33.031158447265625, "learning_rate": 2.005957146061899e-06, "loss": 10.336753845214844, "loss_alignment": 0.83203125, "loss_alignment_w": 0.416015625, "loss_sft": 0.23087070882320404, "loss_total_v6": 0.6460470929741859, "step": 5694 }, { "epoch": 1.313427027027027, "grad_norm": 32.81485366821289, "learning_rate": 2.004344519076465e-06, "loss": 9.025070190429688, "loss_alignment": 0.7880859375, "loss_alignment_w": 0.39404296875, "loss_sft": 0.17104634270071983, "loss_total_v6": 0.5640669278800488, "step": 5695 }, { "epoch": 1.3136576576576577, "grad_norm": 27.043865203857422, "learning_rate": 2.0027323780531312e-06, "loss": 8.815797805786133, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.16213243827223778, "loss_total_v6": 0.5509873926639557, "step": 5696 }, { "epoch": 1.3138882882882883, "grad_norm": 33.06968688964844, "learning_rate": 2.0011207232534196e-06, "loss": 9.39528751373291, "loss_alignment": 0.76806640625, "loss_alignment_w": 0.384033203125, "loss_sft": 0.20285181887447834, "loss_total_v6": 0.5872054845094681, "step": 5697 }, { "epoch": 1.3141189189189189, "grad_norm": 27.693004608154297, "learning_rate": 1.9995095549387812e-06, "loss": 8.879032135009766, "loss_alignment": 0.77978515625, "loss_alignment_w": 0.389892578125, "loss_sft": 0.16480278875678778, "loss_total_v6": 0.5549395233392715, "step": 5698 }, { "epoch": 1.3143495495495494, "grad_norm": 27.918285369873047, "learning_rate": 1.9978988733705807e-06, "loss": 8.769643783569336, "loss_alignment": 0.75244140625, "loss_alignment_w": 0.376220703125, "loss_sft": 0.17162260133773088, "loss_total_v6": 0.5481026917695999, "step": 5699 }, { "epoch": 1.3145801801801802, "grad_norm": 28.483049392700195, "learning_rate": 1.996288678810105e-06, "loss": 9.703060150146484, "loss_alignment": 0.78857421875, "loss_alignment_w": 0.394287109375, "loss_sft": 0.2115896064788103, "loss_total_v6": 0.6064412668347359, "step": 5700 }, { "epoch": 1.3148108108108107, "grad_norm": 29.31020736694336, "learning_rate": 1.9946789715185677e-06, "loss": 9.422112464904785, "loss_alignment": 0.75732421875, "loss_alignment_w": 0.378662109375, "loss_sft": 0.20924332737922668, "loss_total_v6": 0.5888819992542267, "step": 5701 }, { "epoch": 1.3150414414414415, "grad_norm": 31.16141128540039, "learning_rate": 1.9930697517570968e-06, "loss": 9.265932083129883, "loss_alignment": 0.751953125, "loss_alignment_w": 0.3759765625, "loss_sft": 0.2023507822304964, "loss_total_v6": 0.5791207812726498, "step": 5702 }, { "epoch": 1.315272072072072, "grad_norm": 35.57752227783203, "learning_rate": 1.9914610197867416e-06, "loss": 9.460354804992676, "loss_alignment": 0.76611328125, "loss_alignment_w": 0.383056640625, "loss_sft": 0.208886893466115, "loss_total_v6": 0.5912721678614616, "step": 5703 }, { "epoch": 1.3155027027027026, "grad_norm": 38.80131149291992, "learning_rate": 1.9898527758684787e-06, "loss": 9.609809875488281, "loss_alignment": 0.7998046875, "loss_alignment_w": 0.39990234375, "loss_sft": 0.20020727813243866, "loss_total_v6": 0.60061314702034, "step": 5704 }, { "epoch": 1.3157333333333332, "grad_norm": 35.64584732055664, "learning_rate": 1.988245020263197e-06, "loss": 8.770413398742676, "loss_alignment": 0.76611328125, "loss_alignment_w": 0.383056640625, "loss_sft": 0.1647889968007803, "loss_total_v6": 0.5481508113443851, "step": 5705 }, { "epoch": 1.315963963963964, "grad_norm": 26.98053550720215, "learning_rate": 1.9866377532317137e-06, "loss": 9.374754905700684, "loss_alignment": 0.796875, "loss_alignment_w": 0.3984375, "loss_sft": 0.18690486252307892, "loss_total_v6": 0.5859221592545509, "step": 5706 }, { "epoch": 1.3161945945945945, "grad_norm": 31.274808883666992, "learning_rate": 1.9850309750347617e-06, "loss": 9.314674377441406, "loss_alignment": 0.81689453125, "loss_alignment_w": 0.408447265625, "loss_sft": 0.1742539368569851, "loss_total_v6": 0.5821671560406685, "step": 5707 }, { "epoch": 1.3164252252252253, "grad_norm": 29.947357177734375, "learning_rate": 1.9834246859329962e-06, "loss": 9.301474571228027, "loss_alignment": 0.794921875, "loss_alignment_w": 0.3974609375, "loss_sft": 0.18453735206276178, "loss_total_v6": 0.581342164427042, "step": 5708 }, { "epoch": 1.3166558558558559, "grad_norm": 28.899715423583984, "learning_rate": 1.9818188861869926e-06, "loss": 9.42303466796875, "loss_alignment": 0.78515625, "loss_alignment_w": 0.392578125, "loss_sft": 0.19613265711814165, "loss_total_v6": 0.5889396965503693, "step": 5709 }, { "epoch": 1.3168864864864864, "grad_norm": 30.946651458740234, "learning_rate": 1.9802135760572474e-06, "loss": 9.50528621673584, "loss_alignment": 0.791015625, "loss_alignment_w": 0.3955078125, "loss_sft": 0.19913716055452824, "loss_total_v6": 0.59408038854599, "step": 5710 }, { "epoch": 1.317117117117117, "grad_norm": 28.360383987426758, "learning_rate": 1.9786087558041795e-06, "loss": 9.546722412109375, "loss_alignment": 0.80126953125, "loss_alignment_w": 0.400634765625, "loss_sft": 0.19577602483332157, "loss_total_v6": 0.5966701880097389, "step": 5711 }, { "epoch": 1.3173477477477478, "grad_norm": 29.769865036010742, "learning_rate": 1.977004425688126e-06, "loss": 9.475067138671875, "loss_alignment": 0.8154296875, "loss_alignment_w": 0.40771484375, "loss_sft": 0.18426320515573025, "loss_total_v6": 0.5921916738152504, "step": 5712 }, { "epoch": 1.3175783783783783, "grad_norm": 26.00351333618164, "learning_rate": 1.975400585969344e-06, "loss": 8.575296401977539, "loss_alignment": 0.76416015625, "loss_alignment_w": 0.382080078125, "loss_sft": 0.1534334234893322, "loss_total_v6": 0.5359560251235962, "step": 5713 }, { "epoch": 1.317809009009009, "grad_norm": 31.398540496826172, "learning_rate": 1.973797236908013e-06, "loss": 9.539883613586426, "loss_alignment": 0.8212890625, "loss_alignment_w": 0.41064453125, "loss_sft": 0.18683416210114956, "loss_total_v6": 0.596242718398571, "step": 5714 }, { "epoch": 1.3180396396396397, "grad_norm": 30.35353660583496, "learning_rate": 1.9721943787642296e-06, "loss": 9.044342041015625, "loss_alignment": 0.77294921875, "loss_alignment_w": 0.386474609375, "loss_sft": 0.17858313769102097, "loss_total_v6": 0.5652713775634766, "step": 5715 }, { "epoch": 1.3182702702702702, "grad_norm": 35.08470153808594, "learning_rate": 1.970592011798015e-06, "loss": 9.450838088989258, "loss_alignment": 0.77783203125, "loss_alignment_w": 0.388916015625, "loss_sft": 0.20089165307581425, "loss_total_v6": 0.590677410364151, "step": 5716 }, { "epoch": 1.3185009009009008, "grad_norm": 30.194671630859375, "learning_rate": 1.96899013626931e-06, "loss": 8.93731689453125, "loss_alignment": 0.7607421875, "loss_alignment_w": 0.38037109375, "loss_sft": 0.17889789305627346, "loss_total_v6": 0.558582354336977, "step": 5717 }, { "epoch": 1.3187315315315316, "grad_norm": 28.235044479370117, "learning_rate": 1.967388752437973e-06, "loss": 9.57863998413086, "loss_alignment": 0.759765625, "loss_alignment_w": 0.3798828125, "loss_sft": 0.21936203725636005, "loss_total_v6": 0.5986650139093399, "step": 5718 }, { "epoch": 1.318962162162162, "grad_norm": 24.12108039855957, "learning_rate": 1.9657878605637854e-06, "loss": 9.414422035217285, "loss_alignment": 0.830078125, "loss_alignment_w": 0.4150390625, "loss_sft": 0.17363696359097958, "loss_total_v6": 0.5884013548493385, "step": 5719 }, { "epoch": 1.319192792792793, "grad_norm": 25.761634826660156, "learning_rate": 1.9641874609064443e-06, "loss": 9.235893249511719, "loss_alignment": 0.7939453125, "loss_alignment_w": 0.39697265625, "loss_sft": 0.18068265728652477, "loss_total_v6": 0.5772433206439018, "step": 5720 }, { "epoch": 1.3194234234234234, "grad_norm": 27.476268768310547, "learning_rate": 1.9625875537255745e-06, "loss": 9.185745239257812, "loss_alignment": 0.79931640625, "loss_alignment_w": 0.399658203125, "loss_sft": 0.1737794540822506, "loss_total_v6": 0.5741090551018715, "step": 5721 }, { "epoch": 1.319654054054054, "grad_norm": 32.89216232299805, "learning_rate": 1.9609881392807124e-06, "loss": 9.377177238464355, "loss_alignment": 0.77197265625, "loss_alignment_w": 0.385986328125, "loss_sft": 0.2001635693013668, "loss_total_v6": 0.5860736072063446, "step": 5722 }, { "epoch": 1.3198846846846846, "grad_norm": 29.094682693481445, "learning_rate": 1.959389217831323e-06, "loss": 9.574377059936523, "loss_alignment": 0.822265625, "loss_alignment_w": 0.4111328125, "loss_sft": 0.18647221568971872, "loss_total_v6": 0.5983985215425491, "step": 5723 }, { "epoch": 1.3201153153153153, "grad_norm": 33.375186920166016, "learning_rate": 1.9577907896367846e-06, "loss": 8.820700645446777, "loss_alignment": 0.7509765625, "loss_alignment_w": 0.37548828125, "loss_sft": 0.17580550536513329, "loss_total_v6": 0.5512937903404236, "step": 5724 }, { "epoch": 1.320345945945946, "grad_norm": 32.067901611328125, "learning_rate": 1.956192854956397e-06, "loss": 9.64394474029541, "loss_alignment": 0.78515625, "loss_alignment_w": 0.392578125, "loss_sft": 0.21152645163238049, "loss_total_v6": 0.6027465462684631, "step": 5725 }, { "epoch": 1.3205765765765767, "grad_norm": 31.842077255249023, "learning_rate": 1.954595414049383e-06, "loss": 9.492222785949707, "loss_alignment": 0.81005859375, "loss_alignment_w": 0.405029296875, "loss_sft": 0.18780736811459064, "loss_total_v6": 0.5932639017701149, "step": 5726 }, { "epoch": 1.3208072072072072, "grad_norm": 38.851810455322266, "learning_rate": 1.9529984671748825e-06, "loss": 9.014384269714355, "loss_alignment": 0.77880859375, "loss_alignment_w": 0.389404296875, "loss_sft": 0.173598013818264, "loss_total_v6": 0.5633990429341793, "step": 5727 }, { "epoch": 1.3210378378378378, "grad_norm": 31.88385581970215, "learning_rate": 1.9514020145919536e-06, "loss": 9.055206298828125, "loss_alignment": 0.740234375, "loss_alignment_w": 0.3701171875, "loss_sft": 0.19584844633936882, "loss_total_v6": 0.565950371325016, "step": 5728 }, { "epoch": 1.3212684684684683, "grad_norm": 32.26390075683594, "learning_rate": 1.9498060565595804e-06, "loss": 8.999968528747559, "loss_alignment": 0.771484375, "loss_alignment_w": 0.3857421875, "loss_sft": 0.1768473731353879, "loss_total_v6": 0.5624980069696903, "step": 5729 }, { "epoch": 1.3214990990990991, "grad_norm": 29.88195037841797, "learning_rate": 1.948210593336659e-06, "loss": 9.174112319946289, "loss_alignment": 0.7763671875, "loss_alignment_w": 0.38818359375, "loss_sft": 0.18524422496557236, "loss_total_v6": 0.5733820348978043, "step": 5730 }, { "epoch": 1.3217297297297297, "grad_norm": 26.91161346435547, "learning_rate": 1.946615625182013e-06, "loss": 9.287995338439941, "loss_alignment": 0.78369140625, "loss_alignment_w": 0.391845703125, "loss_sft": 0.1877232175320387, "loss_total_v6": 0.5804997012019157, "step": 5731 }, { "epoch": 1.3219603603603605, "grad_norm": 29.739301681518555, "learning_rate": 1.945021152354379e-06, "loss": 9.069377899169922, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.17830159049481153, "loss_total_v6": 0.5668361485004425, "step": 5732 }, { "epoch": 1.322190990990991, "grad_norm": 30.709735870361328, "learning_rate": 1.9434271751124173e-06, "loss": 9.069015502929688, "loss_alignment": 0.775390625, "loss_alignment_w": 0.3876953125, "loss_sft": 0.17916391976177692, "loss_total_v6": 0.5668134763836861, "step": 5733 }, { "epoch": 1.3224216216216216, "grad_norm": 32.033241271972656, "learning_rate": 1.9418336937147047e-06, "loss": 8.796586036682129, "loss_alignment": 0.74365234375, "loss_alignment_w": 0.371826171875, "loss_sft": 0.17683129105716944, "loss_total_v6": 0.5497866310179234, "step": 5734 }, { "epoch": 1.3226522522522521, "grad_norm": 29.78329849243164, "learning_rate": 1.94024070841974e-06, "loss": 8.857170104980469, "loss_alignment": 0.7529296875, "loss_alignment_w": 0.37646484375, "loss_sft": 0.1768031157553196, "loss_total_v6": 0.5535731315612793, "step": 5735 }, { "epoch": 1.322882882882883, "grad_norm": 30.51393699645996, "learning_rate": 1.938648219485944e-06, "loss": 9.615039825439453, "loss_alignment": 0.8427734375, "loss_alignment_w": 0.42138671875, "loss_sft": 0.18020939454436302, "loss_total_v6": 0.6009399741888046, "step": 5736 }, { "epoch": 1.3231135135135135, "grad_norm": 28.94622802734375, "learning_rate": 1.9370562271716514e-06, "loss": 9.132902145385742, "loss_alignment": 0.783203125, "loss_alignment_w": 0.3916015625, "loss_sft": 0.1792201278731227, "loss_total_v6": 0.5708064138889313, "step": 5737 }, { "epoch": 1.3233441441441443, "grad_norm": 26.165319442749023, "learning_rate": 1.9354647317351187e-06, "loss": 9.752840042114258, "loss_alignment": 0.78173828125, "loss_alignment_w": 0.390869140625, "loss_sft": 0.21859178319573402, "loss_total_v6": 0.6095524579286575, "step": 5738 }, { "epoch": 1.3235747747747748, "grad_norm": 32.47040939331055, "learning_rate": 1.933873733434521e-06, "loss": 8.851667404174805, "loss_alignment": 0.748046875, "loss_alignment_w": 0.3740234375, "loss_sft": 0.17874805815517902, "loss_total_v6": 0.5532292537391186, "step": 5739 }, { "epoch": 1.3238054054054054, "grad_norm": 30.941158294677734, "learning_rate": 1.9322832325279563e-06, "loss": 8.828409194946289, "loss_alignment": 0.75244140625, "loss_alignment_w": 0.376220703125, "loss_sft": 0.17608896270394325, "loss_total_v6": 0.5517756044864655, "step": 5740 }, { "epoch": 1.324036036036036, "grad_norm": 31.06538200378418, "learning_rate": 1.9306932292734365e-06, "loss": 9.298280715942383, "loss_alignment": 0.76611328125, "loss_alignment_w": 0.383056640625, "loss_sft": 0.19820801354944706, "loss_total_v6": 0.5811425596475601, "step": 5741 }, { "epoch": 1.3242666666666667, "grad_norm": 37.92540740966797, "learning_rate": 1.9291037239288986e-06, "loss": 9.693838119506836, "loss_alignment": 0.81201171875, "loss_alignment_w": 0.406005859375, "loss_sft": 0.2007898036390543, "loss_total_v6": 0.6058648750185966, "step": 5742 }, { "epoch": 1.3244972972972973, "grad_norm": 28.34012222290039, "learning_rate": 1.927514716752194e-06, "loss": 8.96591567993164, "loss_alignment": 0.7783203125, "loss_alignment_w": 0.38916015625, "loss_sft": 0.1703550973907113, "loss_total_v6": 0.5603697597980499, "step": 5743 }, { "epoch": 1.324727927927928, "grad_norm": 27.734081268310547, "learning_rate": 1.9259262080010938e-06, "loss": 9.983367919921875, "loss_alignment": 0.79296875, "loss_alignment_w": 0.396484375, "loss_sft": 0.2279796600341797, "loss_total_v6": 0.6239605024456978, "step": 5744 }, { "epoch": 1.3249585585585586, "grad_norm": 33.821434020996094, "learning_rate": 1.924338197933293e-06, "loss": 9.124588012695312, "loss_alignment": 0.7841796875, "loss_alignment_w": 0.39208984375, "loss_sft": 0.1785935815423727, "loss_total_v6": 0.5702866911888123, "step": 5745 }, { "epoch": 1.3251891891891892, "grad_norm": 35.89584732055664, "learning_rate": 1.9227506868064e-06, "loss": 9.736519813537598, "loss_alignment": 0.8095703125, "loss_alignment_w": 0.40478515625, "loss_sft": 0.20153480395674706, "loss_total_v6": 0.6085324883460999, "step": 5746 }, { "epoch": 1.3254198198198197, "grad_norm": 31.604520797729492, "learning_rate": 1.9211636748779434e-06, "loss": 9.23946762084961, "loss_alignment": 0.77001953125, "loss_alignment_w": 0.385009765625, "loss_sft": 0.19247219059616327, "loss_total_v6": 0.5774666890501976, "step": 5747 }, { "epoch": 1.3256504504504505, "grad_norm": 28.908754348754883, "learning_rate": 1.9195771624053745e-06, "loss": 9.071931838989258, "loss_alignment": 0.7509765625, "loss_alignment_w": 0.37548828125, "loss_sft": 0.1935063824057579, "loss_total_v6": 0.5669957473874092, "step": 5748 }, { "epoch": 1.325881081081081, "grad_norm": 24.625478744506836, "learning_rate": 1.9179911496460584e-06, "loss": 9.180638313293457, "loss_alignment": 0.7607421875, "loss_alignment_w": 0.38037109375, "loss_sft": 0.19338829815387726, "loss_total_v6": 0.5737899020314217, "step": 5749 }, { "epoch": 1.3261117117117118, "grad_norm": 30.358928680419922, "learning_rate": 1.9164056368572847e-06, "loss": 9.559816360473633, "loss_alignment": 0.7734375, "loss_alignment_w": 0.38671875, "loss_sft": 0.21038836799561977, "loss_total_v6": 0.5974885672330856, "step": 5750 }, { "epoch": 1.3263423423423424, "grad_norm": 34.95475769042969, "learning_rate": 1.9148206242962575e-06, "loss": 9.869302749633789, "loss_alignment": 0.791015625, "loss_alignment_w": 0.3955078125, "loss_sft": 0.22130834124982357, "loss_total_v6": 0.6168314069509506, "step": 5751 }, { "epoch": 1.326572972972973, "grad_norm": 30.2581844329834, "learning_rate": 1.913236112220101e-06, "loss": 8.977434158325195, "loss_alignment": 0.779296875, "loss_alignment_w": 0.3896484375, "loss_sft": 0.17168531566858292, "loss_total_v6": 0.5610896125435829, "step": 5752 }, { "epoch": 1.3268036036036035, "grad_norm": 31.114973068237305, "learning_rate": 1.9116521008858564e-06, "loss": 9.28112506866455, "loss_alignment": 0.77783203125, "loss_alignment_w": 0.388916015625, "loss_sft": 0.1900404067710042, "loss_total_v6": 0.5800703167915344, "step": 5753 }, { "epoch": 1.3270342342342343, "grad_norm": 37.169551849365234, "learning_rate": 1.910068590550488e-06, "loss": 9.51070499420166, "loss_alignment": 0.7900390625, "loss_alignment_w": 0.39501953125, "loss_sft": 0.1999793816357851, "loss_total_v6": 0.594419077038765, "step": 5754 }, { "epoch": 1.3272648648648648, "grad_norm": 29.56276512145996, "learning_rate": 1.9084855814708776e-06, "loss": 8.537836074829102, "loss_alignment": 0.72265625, "loss_alignment_w": 0.361328125, "loss_sft": 0.17236286774277687, "loss_total_v6": 0.5336146950721741, "step": 5755 }, { "epoch": 1.3274954954954956, "grad_norm": 29.45517921447754, "learning_rate": 1.9069030739038224e-06, "loss": 9.076329231262207, "loss_alignment": 0.7529296875, "loss_alignment_w": 0.37646484375, "loss_sft": 0.19085151702165604, "loss_total_v6": 0.5672705844044685, "step": 5756 }, { "epoch": 1.3277261261261262, "grad_norm": 31.07955551147461, "learning_rate": 1.905321068106042e-06, "loss": 9.086151123046875, "loss_alignment": 0.7919921875, "loss_alignment_w": 0.39599609375, "loss_sft": 0.17074390687048435, "loss_total_v6": 0.5678844079375267, "step": 5757 }, { "epoch": 1.3279567567567567, "grad_norm": 29.783864974975586, "learning_rate": 1.9037395643341717e-06, "loss": 9.507145881652832, "loss_alignment": 0.78564453125, "loss_alignment_w": 0.392822265625, "loss_sft": 0.2002451829612255, "loss_total_v6": 0.5941966101527214, "step": 5758 }, { "epoch": 1.3281873873873873, "grad_norm": 32.77579116821289, "learning_rate": 1.9021585628447652e-06, "loss": 9.12613296508789, "loss_alignment": 0.763671875, "loss_alignment_w": 0.3818359375, "loss_sft": 0.18831851426512003, "loss_total_v6": 0.5703833363950253, "step": 5759 }, { "epoch": 1.328418018018018, "grad_norm": 29.30274200439453, "learning_rate": 1.9005780638942982e-06, "loss": 9.657424926757812, "loss_alignment": 0.80126953125, "loss_alignment_w": 0.400634765625, "loss_sft": 0.20211505703628063, "loss_total_v6": 0.6035890877246857, "step": 5760 }, { "epoch": 1.3286486486486486, "grad_norm": 31.541156768798828, "learning_rate": 1.8989980677391646e-06, "loss": 9.174125671386719, "loss_alignment": 0.7568359375, "loss_alignment_w": 0.37841796875, "loss_sft": 0.19496486149728298, "loss_total_v6": 0.5733828321099281, "step": 5761 }, { "epoch": 1.3288792792792794, "grad_norm": 29.79270362854004, "learning_rate": 1.8974185746356727e-06, "loss": 9.013290405273438, "loss_alignment": 0.74169921875, "loss_alignment_w": 0.370849609375, "loss_sft": 0.19194696471095085, "loss_total_v6": 0.5633306428790092, "step": 5762 }, { "epoch": 1.32910990990991, "grad_norm": 29.154945373535156, "learning_rate": 1.8958395848400523e-06, "loss": 9.394766807556152, "loss_alignment": 0.77880859375, "loss_alignment_w": 0.389404296875, "loss_sft": 0.19786019064486027, "loss_total_v6": 0.5871729403734207, "step": 5763 }, { "epoch": 1.3293405405405405, "grad_norm": 30.61196517944336, "learning_rate": 1.8942610986084487e-06, "loss": 9.126838684082031, "loss_alignment": 0.7607421875, "loss_alignment_w": 0.38037109375, "loss_sft": 0.19030047953128815, "loss_total_v6": 0.5704274401068687, "step": 5764 }, { "epoch": 1.329571171171171, "grad_norm": 29.17136573791504, "learning_rate": 1.8926831161969306e-06, "loss": 9.126228332519531, "loss_alignment": 0.78564453125, "loss_alignment_w": 0.392822265625, "loss_sft": 0.1773991584777832, "loss_total_v6": 0.5703892670571804, "step": 5765 }, { "epoch": 1.3298018018018019, "grad_norm": 37.84275436401367, "learning_rate": 1.8911056378614785e-06, "loss": 9.587512969970703, "loss_alignment": 0.77197265625, "loss_alignment_w": 0.385986328125, "loss_sft": 0.2127449344843626, "loss_total_v6": 0.5992195382714272, "step": 5766 }, { "epoch": 1.3300324324324324, "grad_norm": 27.66391944885254, "learning_rate": 1.8895286638579984e-06, "loss": 9.291351318359375, "loss_alignment": 0.751953125, "loss_alignment_w": 0.3759765625, "loss_sft": 0.20467186532914639, "loss_total_v6": 0.5807094648480415, "step": 5767 }, { "epoch": 1.3302630630630632, "grad_norm": 30.81520652770996, "learning_rate": 1.887952194442309e-06, "loss": 9.369770050048828, "loss_alignment": 0.740234375, "loss_alignment_w": 0.3701171875, "loss_sft": 0.21526458114385605, "loss_total_v6": 0.5856106281280518, "step": 5768 }, { "epoch": 1.3304936936936937, "grad_norm": 33.06119155883789, "learning_rate": 1.8863762298701471e-06, "loss": 9.066927909851074, "loss_alignment": 0.74560546875, "loss_alignment_w": 0.372802734375, "loss_sft": 0.19496365264058113, "loss_total_v6": 0.5666830129921436, "step": 5769 }, { "epoch": 1.3307243243243243, "grad_norm": 37.177513122558594, "learning_rate": 1.884800770397172e-06, "loss": 9.92015266418457, "loss_alignment": 0.82666015625, "loss_alignment_w": 0.413330078125, "loss_sft": 0.2058096919208765, "loss_total_v6": 0.6200095191597939, "step": 5770 }, { "epoch": 1.3309549549549549, "grad_norm": 29.450061798095703, "learning_rate": 1.8832258162789575e-06, "loss": 9.084125518798828, "loss_alignment": 0.75390625, "loss_alignment_w": 0.376953125, "loss_sft": 0.19054529443383217, "loss_total_v6": 0.5677578151226044, "step": 5771 }, { "epoch": 1.3311855855855856, "grad_norm": 30.812145233154297, "learning_rate": 1.8816513677709935e-06, "loss": 9.3251953125, "loss_alignment": 0.7421875, "loss_alignment_w": 0.37109375, "loss_sft": 0.21211243700236082, "loss_total_v6": 0.5828247219324112, "step": 5772 }, { "epoch": 1.3314162162162162, "grad_norm": 27.73012351989746, "learning_rate": 1.8800774251286958e-06, "loss": 9.382715225219727, "loss_alignment": 0.7607421875, "loss_alignment_w": 0.38037109375, "loss_sft": 0.20665895752608776, "loss_total_v6": 0.5864196941256523, "step": 5773 }, { "epoch": 1.331646846846847, "grad_norm": 30.668804168701172, "learning_rate": 1.8785039886073874e-06, "loss": 10.328056335449219, "loss_alignment": 0.82080078125, "loss_alignment_w": 0.410400390625, "loss_sft": 0.2352557387202978, "loss_total_v6": 0.6455035507678986, "step": 5774 }, { "epoch": 1.3318774774774775, "grad_norm": 35.02747344970703, "learning_rate": 1.87693105846232e-06, "loss": 9.787345886230469, "loss_alignment": 0.8427734375, "loss_alignment_w": 0.42138671875, "loss_sft": 0.19099375046789646, "loss_total_v6": 0.6117090955376625, "step": 5775 }, { "epoch": 1.332108108108108, "grad_norm": 33.404911041259766, "learning_rate": 1.8753586349486552e-06, "loss": 9.190343856811523, "loss_alignment": 0.75439453125, "loss_alignment_w": 0.377197265625, "loss_sft": 0.1965278498828411, "loss_total_v6": 0.574396513402462, "step": 5776 }, { "epoch": 1.3323387387387386, "grad_norm": 24.81117057800293, "learning_rate": 1.873786718321476e-06, "loss": 9.451909065246582, "loss_alignment": 0.796875, "loss_alignment_w": 0.3984375, "loss_sft": 0.19163543172180653, "loss_total_v6": 0.5907443389296532, "step": 5777 }, { "epoch": 1.3325693693693694, "grad_norm": 32.64222717285156, "learning_rate": 1.8722153088357792e-06, "loss": 8.93603515625, "loss_alignment": 0.74267578125, "loss_alignment_w": 0.371337890625, "loss_sft": 0.18566893599927425, "loss_total_v6": 0.5585021935403347, "step": 5778 }, { "epoch": 1.3328, "grad_norm": 27.57991600036621, "learning_rate": 1.8706444067464858e-06, "loss": 9.491462707519531, "loss_alignment": 0.74609375, "loss_alignment_w": 0.373046875, "loss_sft": 0.219650786370039, "loss_total_v6": 0.5932164713740349, "step": 5779 }, { "epoch": 1.3330306306306308, "grad_norm": 33.161617279052734, "learning_rate": 1.8690740123084316e-06, "loss": 9.531657218933105, "loss_alignment": 0.79833984375, "loss_alignment_w": 0.399169921875, "loss_sft": 0.19567363150417805, "loss_total_v6": 0.5957285761833191, "step": 5780 }, { "epoch": 1.3332612612612613, "grad_norm": 30.756391525268555, "learning_rate": 1.8675041257763687e-06, "loss": 9.285806655883789, "loss_alignment": 0.79248046875, "loss_alignment_w": 0.396240234375, "loss_sft": 0.18490085564553738, "loss_total_v6": 0.5803628712892532, "step": 5781 }, { "epoch": 1.3334918918918919, "grad_norm": 30.038484573364258, "learning_rate": 1.8659347474049672e-06, "loss": 9.012246131896973, "loss_alignment": 0.77783203125, "loss_alignment_w": 0.388916015625, "loss_sft": 0.1744103617966175, "loss_total_v6": 0.563265360891819, "step": 5782 }, { "epoch": 1.3337225225225224, "grad_norm": 31.628793716430664, "learning_rate": 1.8643658774488132e-06, "loss": 9.546862602233887, "loss_alignment": 0.8154296875, "loss_alignment_w": 0.40771484375, "loss_sft": 0.18916245084255934, "loss_total_v6": 0.5966789349913597, "step": 5783 }, { "epoch": 1.3339531531531532, "grad_norm": 30.310911178588867, "learning_rate": 1.8627975161624167e-06, "loss": 8.791557312011719, "loss_alignment": 0.73876953125, "loss_alignment_w": 0.369384765625, "loss_sft": 0.17979765310883522, "loss_total_v6": 0.5494723618030548, "step": 5784 }, { "epoch": 1.3341837837837838, "grad_norm": 28.761924743652344, "learning_rate": 1.8612296638001975e-06, "loss": 9.06059455871582, "loss_alignment": 0.7568359375, "loss_alignment_w": 0.37841796875, "loss_sft": 0.1877318825572729, "loss_total_v6": 0.5662871859967709, "step": 5785 }, { "epoch": 1.3344144144144146, "grad_norm": 32.928680419921875, "learning_rate": 1.8596623206164988e-06, "loss": 9.23558521270752, "loss_alignment": 0.7900390625, "loss_alignment_w": 0.39501953125, "loss_sft": 0.18232663720846176, "loss_total_v6": 0.5772240832448006, "step": 5786 }, { "epoch": 1.334645045045045, "grad_norm": 28.536771774291992, "learning_rate": 1.858095486865577e-06, "loss": 9.578546524047852, "loss_alignment": 0.796875, "loss_alignment_w": 0.3984375, "loss_sft": 0.20014540571719408, "loss_total_v6": 0.5986592099070549, "step": 5787 }, { "epoch": 1.3348756756756757, "grad_norm": 31.11770248413086, "learning_rate": 1.8565291628016064e-06, "loss": 9.073875427246094, "loss_alignment": 0.77783203125, "loss_alignment_w": 0.388916015625, "loss_sft": 0.17867416702210903, "loss_total_v6": 0.5671171806752682, "step": 5788 }, { "epoch": 1.3351063063063062, "grad_norm": 27.858226776123047, "learning_rate": 1.8549633486786822e-06, "loss": 8.777166366577148, "loss_alignment": 0.7119140625, "loss_alignment_w": 0.35595703125, "loss_sft": 0.19276841916143894, "loss_total_v6": 0.54857287555933, "step": 5789 }, { "epoch": 1.335336936936937, "grad_norm": 27.372379302978516, "learning_rate": 1.8533980447508138e-06, "loss": 9.114656448364258, "loss_alignment": 0.7451171875, "loss_alignment_w": 0.37255859375, "loss_sft": 0.19677172787487507, "loss_total_v6": 0.5696660093963146, "step": 5790 }, { "epoch": 1.3355675675675676, "grad_norm": 30.357587814331055, "learning_rate": 1.8518332512719255e-06, "loss": 9.227482795715332, "loss_alignment": 0.78076171875, "loss_alignment_w": 0.390380859375, "loss_sft": 0.18561963364481926, "loss_total_v6": 0.5767176374793053, "step": 5791 }, { "epoch": 1.3357981981981981, "grad_norm": 29.960678100585938, "learning_rate": 1.8502689684958664e-06, "loss": 9.782723426818848, "loss_alignment": 0.81005859375, "loss_alignment_w": 0.405029296875, "loss_sft": 0.20637567900121212, "loss_total_v6": 0.611420214176178, "step": 5792 }, { "epoch": 1.336028828828829, "grad_norm": 29.09917449951172, "learning_rate": 1.848705196676393e-06, "loss": 9.39339542388916, "loss_alignment": 0.814453125, "loss_alignment_w": 0.4072265625, "loss_sft": 0.18147808127105236, "loss_total_v6": 0.5870872139930725, "step": 5793 }, { "epoch": 1.3362594594594595, "grad_norm": 26.613920211791992, "learning_rate": 1.847141936067189e-06, "loss": 9.456369400024414, "loss_alignment": 0.75244140625, "loss_alignment_w": 0.376220703125, "loss_sft": 0.21373425237834454, "loss_total_v6": 0.5910230502486229, "step": 5794 }, { "epoch": 1.33649009009009, "grad_norm": 30.937923431396484, "learning_rate": 1.845579186921847e-06, "loss": 9.066200256347656, "loss_alignment": 0.77197265625, "loss_alignment_w": 0.385986328125, "loss_sft": 0.18109373096376657, "loss_total_v6": 0.5666375458240509, "step": 5795 }, { "epoch": 1.3367207207207208, "grad_norm": 29.846580505371094, "learning_rate": 1.8440169494938804e-06, "loss": 8.848337173461914, "loss_alignment": 0.775390625, "loss_alignment_w": 0.3876953125, "loss_sft": 0.16526470985263586, "loss_total_v6": 0.5530210770666599, "step": 5796 }, { "epoch": 1.3369513513513513, "grad_norm": 30.294214248657227, "learning_rate": 1.8424552240367177e-06, "loss": 9.976892471313477, "loss_alignment": 0.833984375, "loss_alignment_w": 0.4169921875, "loss_sft": 0.20889816619455814, "loss_total_v6": 0.6235557645559311, "step": 5797 }, { "epoch": 1.337181981981982, "grad_norm": 31.30488395690918, "learning_rate": 1.8408940108037065e-06, "loss": 9.33218765258789, "loss_alignment": 0.7802734375, "loss_alignment_w": 0.39013671875, "loss_sft": 0.19291138648986816, "loss_total_v6": 0.5832617282867432, "step": 5798 }, { "epoch": 1.3374126126126127, "grad_norm": 30.366119384765625, "learning_rate": 1.8393333100481126e-06, "loss": 9.109750747680664, "loss_alignment": 0.7783203125, "loss_alignment_w": 0.38916015625, "loss_sft": 0.18027553148567677, "loss_total_v6": 0.5693593844771385, "step": 5799 }, { "epoch": 1.3376432432432432, "grad_norm": 28.834409713745117, "learning_rate": 1.8377731220231144e-06, "loss": 9.389060974121094, "loss_alignment": 0.7568359375, "loss_alignment_w": 0.37841796875, "loss_sft": 0.20877980440855026, "loss_total_v6": 0.5868163257837296, "step": 5800 }, { "epoch": 1.3378738738738738, "grad_norm": 35.838531494140625, "learning_rate": 1.8362134469818094e-06, "loss": 8.950702667236328, "loss_alignment": 0.73876953125, "loss_alignment_w": 0.369384765625, "loss_sft": 0.19053763709962368, "loss_total_v6": 0.5594188645482063, "step": 5801 }, { "epoch": 1.3381045045045046, "grad_norm": 31.476268768310547, "learning_rate": 1.8346542851772114e-06, "loss": 9.460148811340332, "loss_alignment": 0.80078125, "loss_alignment_w": 0.400390625, "loss_sft": 0.19060926605015993, "loss_total_v6": 0.5912593007087708, "step": 5802 }, { "epoch": 1.3383351351351351, "grad_norm": 36.450836181640625, "learning_rate": 1.8330956368622498e-06, "loss": 9.271604537963867, "loss_alignment": 0.7626953125, "loss_alignment_w": 0.38134765625, "loss_sft": 0.19824970699846745, "loss_total_v6": 0.5794752836227417, "step": 5803 }, { "epoch": 1.3385657657657657, "grad_norm": 32.3153190612793, "learning_rate": 1.8315375022897736e-06, "loss": 9.609870910644531, "loss_alignment": 0.7919921875, "loss_alignment_w": 0.39599609375, "loss_sft": 0.20582623407244682, "loss_total_v6": 0.6006168946623802, "step": 5804 }, { "epoch": 1.3387963963963965, "grad_norm": 31.726581573486328, "learning_rate": 1.8299798817125491e-06, "loss": 9.264162063598633, "loss_alignment": 0.74951171875, "loss_alignment_w": 0.374755859375, "loss_sft": 0.2039338480681181, "loss_total_v6": 0.5790101513266563, "step": 5805 }, { "epoch": 1.339027027027027, "grad_norm": 29.09587860107422, "learning_rate": 1.8284227753832545e-06, "loss": 8.9424467086792, "loss_alignment": 0.79248046875, "loss_alignment_w": 0.396240234375, "loss_sft": 0.16144194826483727, "loss_total_v6": 0.5589028932154179, "step": 5806 }, { "epoch": 1.3392576576576576, "grad_norm": 32.21702194213867, "learning_rate": 1.8268661835544882e-06, "loss": 9.35995101928711, "loss_alignment": 0.78466796875, "loss_alignment_w": 0.392333984375, "loss_sft": 0.1932122502475977, "loss_total_v6": 0.584996908903122, "step": 5807 }, { "epoch": 1.3394882882882884, "grad_norm": 30.958717346191406, "learning_rate": 1.825310106478762e-06, "loss": 8.728395462036133, "loss_alignment": 0.76318359375, "loss_alignment_w": 0.381591796875, "loss_sft": 0.1641465239226818, "loss_total_v6": 0.5455246940255165, "step": 5808 }, { "epoch": 1.339718918918919, "grad_norm": 28.428476333618164, "learning_rate": 1.8237545444085092e-06, "loss": 8.599077224731445, "loss_alignment": 0.74462890625, "loss_alignment_w": 0.372314453125, "loss_sft": 0.1641665967181325, "loss_total_v6": 0.5374423526227474, "step": 5809 }, { "epoch": 1.3399495495495495, "grad_norm": 28.632627487182617, "learning_rate": 1.8221994975960739e-06, "loss": 8.979089736938477, "loss_alignment": 0.75634765625, "loss_alignment_w": 0.378173828125, "loss_sft": 0.183248121291399, "loss_total_v6": 0.5611930899322033, "step": 5810 }, { "epoch": 1.3401801801801803, "grad_norm": 25.790760040283203, "learning_rate": 1.820644966293722e-06, "loss": 9.166089057922363, "loss_alignment": 0.7646484375, "loss_alignment_w": 0.38232421875, "loss_sft": 0.1908615343272686, "loss_total_v6": 0.5728805884718895, "step": 5811 }, { "epoch": 1.3404108108108108, "grad_norm": 29.794002532958984, "learning_rate": 1.8190909507536326e-06, "loss": 9.254951477050781, "loss_alignment": 0.76025390625, "loss_alignment_w": 0.380126953125, "loss_sft": 0.19780399091541767, "loss_total_v6": 0.578434482216835, "step": 5812 }, { "epoch": 1.3406414414414414, "grad_norm": 33.42362976074219, "learning_rate": 1.8175374512278986e-06, "loss": 9.048626899719238, "loss_alignment": 0.7568359375, "loss_alignment_w": 0.37841796875, "loss_sft": 0.1859157457947731, "loss_total_v6": 0.5655391737818718, "step": 5813 }, { "epoch": 1.3408720720720722, "grad_norm": 30.398794174194336, "learning_rate": 1.8159844679685368e-06, "loss": 9.417638778686523, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.20261605083942413, "loss_total_v6": 0.5886024013161659, "step": 5814 }, { "epoch": 1.3411027027027027, "grad_norm": 30.961843490600586, "learning_rate": 1.8144320012274735e-06, "loss": 8.809799194335938, "loss_alignment": 0.7490234375, "loss_alignment_w": 0.37451171875, "loss_sft": 0.17582604475319386, "loss_total_v6": 0.5506124272942543, "step": 5815 }, { "epoch": 1.3413333333333333, "grad_norm": 29.8526554107666, "learning_rate": 1.8128800512565514e-06, "loss": 9.491226196289062, "loss_alignment": 0.7763671875, "loss_alignment_w": 0.38818359375, "loss_sft": 0.2044534720480442, "loss_total_v6": 0.5932016521692276, "step": 5816 }, { "epoch": 1.341563963963964, "grad_norm": 34.96465301513672, "learning_rate": 1.8113286183075356e-06, "loss": 8.796062469482422, "loss_alignment": 0.77294921875, "loss_alignment_w": 0.386474609375, "loss_sft": 0.1634776657447219, "loss_total_v6": 0.5497539155185223, "step": 5817 }, { "epoch": 1.3417945945945946, "grad_norm": 33.44154739379883, "learning_rate": 1.809777702632099e-06, "loss": 9.024995803833008, "loss_alignment": 0.75732421875, "loss_alignment_w": 0.378662109375, "loss_sft": 0.1846524365246296, "loss_total_v6": 0.564062237739563, "step": 5818 }, { "epoch": 1.3420252252252252, "grad_norm": 31.310014724731445, "learning_rate": 1.8082273044818383e-06, "loss": 9.175004959106445, "loss_alignment": 0.77099609375, "loss_alignment_w": 0.385498046875, "loss_sft": 0.18707000464200974, "loss_total_v6": 0.573437824845314, "step": 5819 }, { "epoch": 1.3422558558558557, "grad_norm": 30.480234146118164, "learning_rate": 1.8066774241082612e-06, "loss": 8.704944610595703, "loss_alignment": 0.75244140625, "loss_alignment_w": 0.376220703125, "loss_sft": 0.16779254004359245, "loss_total_v6": 0.5440590158104897, "step": 5820 }, { "epoch": 1.3424864864864865, "grad_norm": 30.67875862121582, "learning_rate": 1.8051280617627925e-06, "loss": 9.322395324707031, "loss_alignment": 0.79052734375, "loss_alignment_w": 0.395263671875, "loss_sft": 0.18616530485451221, "loss_total_v6": 0.5826496854424477, "step": 5821 }, { "epoch": 1.342717117117117, "grad_norm": 32.38277816772461, "learning_rate": 1.803579217696772e-06, "loss": 9.102192878723145, "loss_alignment": 0.77294921875, "loss_alignment_w": 0.386474609375, "loss_sft": 0.18230568151921034, "loss_total_v6": 0.5688870772719383, "step": 5822 }, { "epoch": 1.3429477477477478, "grad_norm": 28.861242294311523, "learning_rate": 1.8020308921614577e-06, "loss": 8.889970779418945, "loss_alignment": 0.779296875, "loss_alignment_w": 0.3896484375, "loss_sft": 0.16634091828018427, "loss_total_v6": 0.5556231439113617, "step": 5823 }, { "epoch": 1.3431783783783784, "grad_norm": 31.010366439819336, "learning_rate": 1.8004830854080252e-06, "loss": 9.555935859680176, "loss_alignment": 0.767578125, "loss_alignment_w": 0.3837890625, "loss_sft": 0.2124193087220192, "loss_total_v6": 0.5972459763288498, "step": 5824 }, { "epoch": 1.343409009009009, "grad_norm": 31.18902015686035, "learning_rate": 1.7989357976875603e-06, "loss": 9.262174606323242, "loss_alignment": 0.767578125, "loss_alignment_w": 0.3837890625, "loss_sft": 0.1943949107080698, "loss_total_v6": 0.5788858905434608, "step": 5825 }, { "epoch": 1.3436396396396395, "grad_norm": 30.118547439575195, "learning_rate": 1.7973890292510693e-06, "loss": 9.035348892211914, "loss_alignment": 0.806640625, "loss_alignment_w": 0.4033203125, "loss_sft": 0.16111433319747448, "loss_total_v6": 0.5647092908620834, "step": 5826 }, { "epoch": 1.3438702702702703, "grad_norm": 28.38833999633789, "learning_rate": 1.795842780349471e-06, "loss": 9.765213966369629, "loss_alignment": 0.78759765625, "loss_alignment_w": 0.393798828125, "loss_sft": 0.21626760996878147, "loss_total_v6": 0.61032585054636, "step": 5827 }, { "epoch": 1.3441009009009008, "grad_norm": 30.671283721923828, "learning_rate": 1.7942970512336e-06, "loss": 8.736553192138672, "loss_alignment": 0.74853515625, "loss_alignment_w": 0.374267578125, "loss_sft": 0.17187387216836214, "loss_total_v6": 0.5460346378386021, "step": 5828 }, { "epoch": 1.3443315315315316, "grad_norm": 31.730945587158203, "learning_rate": 1.7927518421542106e-06, "loss": 8.794221878051758, "loss_alignment": 0.74072265625, "loss_alignment_w": 0.370361328125, "loss_sft": 0.17823998257517815, "loss_total_v6": 0.5496389083564281, "step": 5829 }, { "epoch": 1.3445621621621622, "grad_norm": 30.54560661315918, "learning_rate": 1.7912071533619707e-06, "loss": 9.468564987182617, "loss_alignment": 0.7822265625, "loss_alignment_w": 0.39111328125, "loss_sft": 0.20039738528430462, "loss_total_v6": 0.5917853340506554, "step": 5830 }, { "epoch": 1.3447927927927927, "grad_norm": 32.17718505859375, "learning_rate": 1.7896629851074625e-06, "loss": 9.07177734375, "loss_alignment": 0.76806640625, "loss_alignment_w": 0.384033203125, "loss_sft": 0.181503277271986, "loss_total_v6": 0.5669860616326332, "step": 5831 }, { "epoch": 1.3450234234234233, "grad_norm": 26.758623123168945, "learning_rate": 1.7881193376411822e-06, "loss": 9.122053146362305, "loss_alignment": 0.77392578125, "loss_alignment_w": 0.386962890625, "loss_sft": 0.18286026176065207, "loss_total_v6": 0.5701283365488052, "step": 5832 }, { "epoch": 1.345254054054054, "grad_norm": 29.941429138183594, "learning_rate": 1.7865762112135476e-06, "loss": 8.559992790222168, "loss_alignment": 0.791015625, "loss_alignment_w": 0.3955078125, "loss_sft": 0.139690101146698, "loss_total_v6": 0.5349995456635952, "step": 5833 }, { "epoch": 1.3454846846846846, "grad_norm": 33.85374450683594, "learning_rate": 1.7850336060748864e-06, "loss": 8.769763946533203, "loss_alignment": 0.755859375, "loss_alignment_w": 0.3779296875, "loss_sft": 0.169707540422678, "loss_total_v6": 0.5481102466583252, "step": 5834 }, { "epoch": 1.3457153153153154, "grad_norm": 28.353736877441406, "learning_rate": 1.7834915224754418e-06, "loss": 9.398691177368164, "loss_alignment": 0.7900390625, "loss_alignment_w": 0.39501953125, "loss_sft": 0.1921850685030222, "loss_total_v6": 0.5874182358384132, "step": 5835 }, { "epoch": 1.345945945945946, "grad_norm": 31.034618377685547, "learning_rate": 1.7819499606653772e-06, "loss": 9.023811340332031, "loss_alignment": 0.765625, "loss_alignment_w": 0.3828125, "loss_sft": 0.18276259303092957, "loss_total_v6": 0.5639881789684296, "step": 5836 }, { "epoch": 1.3461765765765765, "grad_norm": 30.392147064208984, "learning_rate": 1.7804089208947657e-06, "loss": 9.41877269744873, "loss_alignment": 0.78662109375, "loss_alignment_w": 0.393310546875, "loss_sft": 0.19661396648734808, "loss_total_v6": 0.5886732861399651, "step": 5837 }, { "epoch": 1.346407207207207, "grad_norm": 32.5318717956543, "learning_rate": 1.7788684034136012e-06, "loss": 9.717972755432129, "loss_alignment": 0.802734375, "loss_alignment_w": 0.4013671875, "loss_sft": 0.20652490854263306, "loss_total_v6": 0.6073732897639275, "step": 5838 }, { "epoch": 1.3466378378378379, "grad_norm": 32.63673782348633, "learning_rate": 1.7773284084717885e-06, "loss": 8.753539085388184, "loss_alignment": 0.7626953125, "loss_alignment_w": 0.38134765625, "loss_sft": 0.16567222215235233, "loss_total_v6": 0.5470961779356003, "step": 5839 }, { "epoch": 1.3468684684684684, "grad_norm": 30.146324157714844, "learning_rate": 1.7757889363191484e-06, "loss": 9.173352241516113, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.1862037591636181, "loss_total_v6": 0.5733345150947571, "step": 5840 }, { "epoch": 1.3470990990990992, "grad_norm": 29.287700653076172, "learning_rate": 1.7742499872054163e-06, "loss": 9.469451904296875, "loss_alignment": 0.77734375, "loss_alignment_w": 0.388671875, "loss_sft": 0.20272640883922577, "loss_total_v6": 0.5918407663702965, "step": 5841 }, { "epoch": 1.3473297297297298, "grad_norm": 33.77607345581055, "learning_rate": 1.7727115613802465e-06, "loss": 9.554533958435059, "loss_alignment": 0.76513671875, "loss_alignment_w": 0.382568359375, "loss_sft": 0.21478837355971336, "loss_total_v6": 0.59715835750103, "step": 5842 }, { "epoch": 1.3475603603603603, "grad_norm": 29.37690544128418, "learning_rate": 1.7711736590932066e-06, "loss": 8.928142547607422, "loss_alignment": 0.767578125, "loss_alignment_w": 0.3837890625, "loss_sft": 0.17475394532084465, "loss_total_v6": 0.5580089315772057, "step": 5843 }, { "epoch": 1.3477909909909909, "grad_norm": 29.852678298950195, "learning_rate": 1.7696362805937778e-06, "loss": 8.953460693359375, "loss_alignment": 0.75048828125, "loss_alignment_w": 0.375244140625, "loss_sft": 0.18536949157714844, "loss_total_v6": 0.5595912896096706, "step": 5844 }, { "epoch": 1.3480216216216216, "grad_norm": 27.388750076293945, "learning_rate": 1.7680994261313567e-06, "loss": 9.360191345214844, "loss_alignment": 0.76513671875, "loss_alignment_w": 0.382568359375, "loss_sft": 0.20206210762262344, "loss_total_v6": 0.5850119367241859, "step": 5845 }, { "epoch": 1.3482522522522522, "grad_norm": 27.54787826538086, "learning_rate": 1.766563095955255e-06, "loss": 9.6189546585083, "loss_alignment": 0.7568359375, "loss_alignment_w": 0.37841796875, "loss_sft": 0.22305665723979473, "loss_total_v6": 0.6011846885085106, "step": 5846 }, { "epoch": 1.348482882882883, "grad_norm": 33.305442810058594, "learning_rate": 1.7650272903146987e-06, "loss": 9.024749755859375, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.17665670067071915, "loss_total_v6": 0.564046822488308, "step": 5847 }, { "epoch": 1.3487135135135135, "grad_norm": 31.328577041625977, "learning_rate": 1.7634920094588308e-06, "loss": 9.302572250366211, "loss_alignment": 0.751953125, "loss_alignment_w": 0.3759765625, "loss_sft": 0.205586776137352, "loss_total_v6": 0.581410750746727, "step": 5848 }, { "epoch": 1.348944144144144, "grad_norm": 31.72151756286621, "learning_rate": 1.7619572536367103e-06, "loss": 8.61355209350586, "loss_alignment": 0.7646484375, "loss_alignment_w": 0.38232421875, "loss_sft": 0.15574811026453972, "loss_total_v6": 0.5383469760417938, "step": 5849 }, { "epoch": 1.3491747747747747, "grad_norm": 28.91642951965332, "learning_rate": 1.7604230230973068e-06, "loss": 8.883216857910156, "loss_alignment": 0.75, "loss_alignment_w": 0.375, "loss_sft": 0.17981962393969297, "loss_total_v6": 0.5552011094987392, "step": 5850 }, { "epoch": 1.3494054054054054, "grad_norm": 33.06352233886719, "learning_rate": 1.7588893180895073e-06, "loss": 8.882774353027344, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.1705145752057433, "loss_total_v6": 0.5551733709871769, "step": 5851 }, { "epoch": 1.349636036036036, "grad_norm": 28.390708923339844, "learning_rate": 1.7573561388621102e-06, "loss": 8.975428581237793, "loss_alignment": 0.75048828125, "loss_alignment_w": 0.375244140625, "loss_sft": 0.1858727466315031, "loss_total_v6": 0.5609643086791039, "step": 5852 }, { "epoch": 1.3498666666666668, "grad_norm": 29.74809455871582, "learning_rate": 1.7558234856638351e-06, "loss": 8.959930419921875, "loss_alignment": 0.759765625, "loss_alignment_w": 0.3798828125, "loss_sft": 0.1791972778737545, "loss_total_v6": 0.559995636343956, "step": 5853 }, { "epoch": 1.3500972972972973, "grad_norm": 31.67387580871582, "learning_rate": 1.7542913587433096e-06, "loss": 9.334718704223633, "loss_alignment": 0.7490234375, "loss_alignment_w": 0.37451171875, "loss_sft": 0.20878612995147705, "loss_total_v6": 0.5834198966622353, "step": 5854 }, { "epoch": 1.3503279279279279, "grad_norm": 29.812255859375, "learning_rate": 1.7527597583490825e-06, "loss": 9.194292068481445, "loss_alignment": 0.78662109375, "loss_alignment_w": 0.393310546875, "loss_sft": 0.18095126189291477, "loss_total_v6": 0.5746432766318321, "step": 5855 }, { "epoch": 1.3505585585585584, "grad_norm": 31.1611328125, "learning_rate": 1.7512286847296106e-06, "loss": 9.443018913269043, "loss_alignment": 0.78173828125, "loss_alignment_w": 0.390869140625, "loss_sft": 0.19950263760983944, "loss_total_v6": 0.5901886746287346, "step": 5856 }, { "epoch": 1.3507891891891892, "grad_norm": 30.84440040588379, "learning_rate": 1.7496981381332673e-06, "loss": 9.667579650878906, "loss_alignment": 0.79248046875, "loss_alignment_w": 0.396240234375, "loss_sft": 0.20807504281401634, "loss_total_v6": 0.6042237430810928, "step": 5857 }, { "epoch": 1.3510198198198198, "grad_norm": 32.91914367675781, "learning_rate": 1.7481681188083443e-06, "loss": 9.133739471435547, "loss_alignment": 0.7373046875, "loss_alignment_w": 0.36865234375, "loss_sft": 0.20315233431756496, "loss_total_v6": 0.5708586573600769, "step": 5858 }, { "epoch": 1.3512504504504506, "grad_norm": 27.910890579223633, "learning_rate": 1.7466386270030438e-06, "loss": 8.921839714050293, "loss_alignment": 0.724609375, "loss_alignment_w": 0.3623046875, "loss_sft": 0.19488306250423193, "loss_total_v6": 0.5576149933040142, "step": 5859 }, { "epoch": 1.3514810810810811, "grad_norm": 31.637195587158203, "learning_rate": 1.7451096629654813e-06, "loss": 9.406708717346191, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.20045285485684872, "loss_total_v6": 0.5879192873835564, "step": 5860 }, { "epoch": 1.3517117117117117, "grad_norm": 29.12509536743164, "learning_rate": 1.7435812269436925e-06, "loss": 9.499948501586914, "loss_alignment": 0.7802734375, "loss_alignment_w": 0.39013671875, "loss_sft": 0.2034116368740797, "loss_total_v6": 0.5937467515468597, "step": 5861 }, { "epoch": 1.3519423423423422, "grad_norm": 29.90590476989746, "learning_rate": 1.7420533191856205e-06, "loss": 8.790809631347656, "loss_alignment": 0.744140625, "loss_alignment_w": 0.3720703125, "loss_sft": 0.17784359492361546, "loss_total_v6": 0.5494256168603897, "step": 5862 }, { "epoch": 1.352172972972973, "grad_norm": 28.188642501831055, "learning_rate": 1.7405259399391288e-06, "loss": 9.080911636352539, "loss_alignment": 0.77294921875, "loss_alignment_w": 0.386474609375, "loss_sft": 0.18059408478438854, "loss_total_v6": 0.5675569698214531, "step": 5863 }, { "epoch": 1.3524036036036036, "grad_norm": 28.214763641357422, "learning_rate": 1.738999089451991e-06, "loss": 8.971885681152344, "loss_alignment": 0.75048828125, "loss_alignment_w": 0.375244140625, "loss_sft": 0.18565126322209835, "loss_total_v6": 0.5607428252696991, "step": 5864 }, { "epoch": 1.3526342342342343, "grad_norm": 29.874435424804688, "learning_rate": 1.7374727679718967e-06, "loss": 8.951457977294922, "loss_alignment": 0.73779296875, "loss_alignment_w": 0.368896484375, "loss_sft": 0.19067642651498318, "loss_total_v6": 0.5594660863280296, "step": 5865 }, { "epoch": 1.352864864864865, "grad_norm": 29.25884246826172, "learning_rate": 1.735946975746447e-06, "loss": 9.389654159545898, "loss_alignment": 0.7529296875, "loss_alignment_w": 0.37646484375, "loss_sft": 0.21145671047270298, "loss_total_v6": 0.5868534371256828, "step": 5866 }, { "epoch": 1.3530954954954955, "grad_norm": 32.23405456542969, "learning_rate": 1.734421713023161e-06, "loss": 8.936559677124023, "loss_alignment": 0.7568359375, "loss_alignment_w": 0.37841796875, "loss_sft": 0.1802696231752634, "loss_total_v6": 0.5585349872708321, "step": 5867 }, { "epoch": 1.353326126126126, "grad_norm": 31.907196044921875, "learning_rate": 1.7328969800494727e-06, "loss": 9.392057418823242, "loss_alignment": 0.7578125, "loss_alignment_w": 0.37890625, "loss_sft": 0.20786850713193417, "loss_total_v6": 0.5870036333799362, "step": 5868 }, { "epoch": 1.3535567567567568, "grad_norm": 32.824100494384766, "learning_rate": 1.7313727770727257e-06, "loss": 9.567146301269531, "loss_alignment": 0.7724609375, "loss_alignment_w": 0.38623046875, "loss_sft": 0.21180774085223675, "loss_total_v6": 0.5979466587305069, "step": 5869 }, { "epoch": 1.3537873873873874, "grad_norm": 30.012847900390625, "learning_rate": 1.7298491043401794e-06, "loss": 9.170149803161621, "loss_alignment": 0.7548828125, "loss_alignment_w": 0.37744140625, "loss_sft": 0.19581502489745617, "loss_total_v6": 0.5731343701481819, "step": 5870 }, { "epoch": 1.3540180180180181, "grad_norm": 28.391374588012695, "learning_rate": 1.7283259620990084e-06, "loss": 9.368364334106445, "loss_alignment": 0.7919921875, "loss_alignment_w": 0.39599609375, "loss_sft": 0.18996919132769108, "loss_total_v6": 0.5855227634310722, "step": 5871 }, { "epoch": 1.3542486486486487, "grad_norm": 34.15143966674805, "learning_rate": 1.7268033505962972e-06, "loss": 9.849873542785645, "loss_alignment": 0.79638671875, "loss_alignment_w": 0.398193359375, "loss_sft": 0.21775943785905838, "loss_total_v6": 0.6156171038746834, "step": 5872 }, { "epoch": 1.3544792792792792, "grad_norm": 26.72602653503418, "learning_rate": 1.7252812700790505e-06, "loss": 9.461755752563477, "loss_alignment": 0.76806640625, "loss_alignment_w": 0.384033203125, "loss_sft": 0.20657881908118725, "loss_total_v6": 0.5913597121834755, "step": 5873 }, { "epoch": 1.3547099099099098, "grad_norm": 32.83112716674805, "learning_rate": 1.7237597207941841e-06, "loss": 9.236360549926758, "loss_alignment": 0.80419921875, "loss_alignment_w": 0.402099609375, "loss_sft": 0.17471515759825706, "loss_total_v6": 0.5772725269198418, "step": 5874 }, { "epoch": 1.3549405405405406, "grad_norm": 32.2805061340332, "learning_rate": 1.7222387029885268e-06, "loss": 9.00011157989502, "loss_alignment": 0.74951171875, "loss_alignment_w": 0.374755859375, "loss_sft": 0.18756799027323723, "loss_total_v6": 0.5625069737434387, "step": 5875 }, { "epoch": 1.3551711711711711, "grad_norm": 32.62407302856445, "learning_rate": 1.7207182169088205e-06, "loss": 8.773846626281738, "loss_alignment": 0.75244140625, "loss_alignment_w": 0.376220703125, "loss_sft": 0.1720226053148508, "loss_total_v6": 0.5483653992414474, "step": 5876 }, { "epoch": 1.355401801801802, "grad_norm": 35.43008041381836, "learning_rate": 1.7191982628017212e-06, "loss": 8.978631973266602, "loss_alignment": 0.76708984375, "loss_alignment_w": 0.383544921875, "loss_sft": 0.1773754581809044, "loss_total_v6": 0.5611645132303238, "step": 5877 }, { "epoch": 1.3556324324324325, "grad_norm": 29.525789260864258, "learning_rate": 1.7176788409138018e-06, "loss": 9.280967712402344, "loss_alignment": 0.7451171875, "loss_alignment_w": 0.37255859375, "loss_sft": 0.20704415254294872, "loss_total_v6": 0.5800604969263077, "step": 5878 }, { "epoch": 1.355863063063063, "grad_norm": 28.494741439819336, "learning_rate": 1.7161599514915435e-06, "loss": 9.112411499023438, "loss_alignment": 0.7802734375, "loss_alignment_w": 0.39013671875, "loss_sft": 0.17877867072820663, "loss_total_v6": 0.569525733590126, "step": 5879 }, { "epoch": 1.3560936936936936, "grad_norm": 33.94447326660156, "learning_rate": 1.7146415947813472e-06, "loss": 9.10308837890625, "loss_alignment": 0.75537109375, "loss_alignment_w": 0.377685546875, "loss_sft": 0.19092178344726562, "loss_total_v6": 0.5689430087804794, "step": 5880 }, { "epoch": 1.3563243243243244, "grad_norm": 26.774412155151367, "learning_rate": 1.7131237710295207e-06, "loss": 8.96751880645752, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.1710198875516653, "loss_total_v6": 0.5604699552059174, "step": 5881 }, { "epoch": 1.356554954954955, "grad_norm": 23.868000030517578, "learning_rate": 1.711606480482293e-06, "loss": 8.752586364746094, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.15848680958151817, "loss_total_v6": 0.5470366254448891, "step": 5882 }, { "epoch": 1.3567855855855857, "grad_norm": 28.082468032836914, "learning_rate": 1.7100897233857998e-06, "loss": 8.95728588104248, "loss_alignment": 0.77783203125, "loss_alignment_w": 0.388916015625, "loss_sft": 0.17121954262256622, "loss_total_v6": 0.5598303824663162, "step": 5883 }, { "epoch": 1.3570162162162163, "grad_norm": 29.96059226989746, "learning_rate": 1.7085734999860937e-06, "loss": 9.035133361816406, "loss_alignment": 0.76953125, "loss_alignment_w": 0.384765625, "loss_sft": 0.1801743544638157, "loss_total_v6": 0.5646958574652672, "step": 5884 }, { "epoch": 1.3572468468468468, "grad_norm": 32.680477142333984, "learning_rate": 1.707057810529138e-06, "loss": 9.18494987487793, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.1860284060239792, "loss_total_v6": 0.5740594044327736, "step": 5885 }, { "epoch": 1.3574774774774774, "grad_norm": 28.302135467529297, "learning_rate": 1.705542655260813e-06, "loss": 9.46531867980957, "loss_alignment": 0.806640625, "loss_alignment_w": 0.4033203125, "loss_sft": 0.18801797553896904, "loss_total_v6": 0.5915824249386787, "step": 5886 }, { "epoch": 1.3577081081081082, "grad_norm": 31.8528995513916, "learning_rate": 1.7040280344269118e-06, "loss": 9.058038711547852, "loss_alignment": 0.7451171875, "loss_alignment_w": 0.37255859375, "loss_sft": 0.19343149103224277, "loss_total_v6": 0.5661274492740631, "step": 5887 }, { "epoch": 1.3579387387387387, "grad_norm": 36.212066650390625, "learning_rate": 1.7025139482731385e-06, "loss": 9.549278259277344, "loss_alignment": 0.806640625, "loss_alignment_w": 0.4033203125, "loss_sft": 0.19289921410381794, "loss_total_v6": 0.5968298465013504, "step": 5888 }, { "epoch": 1.3581693693693695, "grad_norm": 31.08193016052246, "learning_rate": 1.7010003970451117e-06, "loss": 8.951339721679688, "loss_alignment": 0.75927734375, "loss_alignment_w": 0.379638671875, "loss_sft": 0.17940810788422823, "loss_total_v6": 0.5594587624073029, "step": 5889 }, { "epoch": 1.3584, "grad_norm": 27.739273071289062, "learning_rate": 1.6994873809883627e-06, "loss": 9.122314453125, "loss_alignment": 0.7314453125, "loss_alignment_w": 0.36572265625, "loss_sft": 0.20541376620531082, "loss_total_v6": 0.570144597440958, "step": 5890 }, { "epoch": 1.3586306306306306, "grad_norm": 30.129573822021484, "learning_rate": 1.6979749003483354e-06, "loss": 8.711220741271973, "loss_alignment": 0.7783203125, "loss_alignment_w": 0.38916015625, "loss_sft": 0.15535220876336098, "loss_total_v6": 0.5444513186812401, "step": 5891 }, { "epoch": 1.3588612612612612, "grad_norm": 30.191614151000977, "learning_rate": 1.6964629553703893e-06, "loss": 9.25521469116211, "loss_alignment": 0.75634765625, "loss_alignment_w": 0.378173828125, "loss_sft": 0.19911742210388184, "loss_total_v6": 0.578450933098793, "step": 5892 }, { "epoch": 1.359091891891892, "grad_norm": 33.39608383178711, "learning_rate": 1.6949515462997972e-06, "loss": 9.169294357299805, "loss_alignment": 0.78369140625, "loss_alignment_w": 0.391845703125, "loss_sft": 0.18121989723294973, "loss_total_v6": 0.5730808898806572, "step": 5893 }, { "epoch": 1.3593225225225225, "grad_norm": 32.68357467651367, "learning_rate": 1.6934406733817417e-06, "loss": 9.621065139770508, "loss_alignment": 0.78662109375, "loss_alignment_w": 0.393310546875, "loss_sft": 0.2082653921097517, "loss_total_v6": 0.6013165339827538, "step": 5894 }, { "epoch": 1.3595531531531533, "grad_norm": 33.545711517333984, "learning_rate": 1.6919303368613194e-06, "loss": 9.348966598510742, "loss_alignment": 0.77197265625, "loss_alignment_w": 0.385986328125, "loss_sft": 0.19843086786568165, "loss_total_v6": 0.584310382604599, "step": 5895 }, { "epoch": 1.3597837837837838, "grad_norm": 26.85942268371582, "learning_rate": 1.6904205369835403e-06, "loss": 9.236462593078613, "loss_alignment": 0.74560546875, "loss_alignment_w": 0.372802734375, "loss_sft": 0.2041557254269719, "loss_total_v6": 0.5772789120674133, "step": 5896 }, { "epoch": 1.3600144144144144, "grad_norm": 32.14918518066406, "learning_rate": 1.6889112739933295e-06, "loss": 9.065315246582031, "loss_alignment": 0.77490234375, "loss_alignment_w": 0.387451171875, "loss_sft": 0.1797871943563223, "loss_total_v6": 0.5665822327136993, "step": 5897 }, { "epoch": 1.360245045045045, "grad_norm": 32.61445236206055, "learning_rate": 1.6874025481355205e-06, "loss": 8.66526985168457, "loss_alignment": 0.7255859375, "loss_alignment_w": 0.36279296875, "loss_sft": 0.17910685762763023, "loss_total_v6": 0.5415793843567371, "step": 5898 }, { "epoch": 1.3604756756756757, "grad_norm": 30.445335388183594, "learning_rate": 1.685894359654866e-06, "loss": 8.518125534057617, "loss_alignment": 0.73583984375, "loss_alignment_w": 0.367919921875, "loss_sft": 0.16478333622217178, "loss_total_v6": 0.5323828309774399, "step": 5899 }, { "epoch": 1.3607063063063063, "grad_norm": 30.57482147216797, "learning_rate": 1.6843867087960252e-06, "loss": 8.958916664123535, "loss_alignment": 0.775390625, "loss_alignment_w": 0.3876953125, "loss_sft": 0.1715503055602312, "loss_total_v6": 0.5599322617053986, "step": 5900 }, { "epoch": 1.360936936936937, "grad_norm": 27.319629669189453, "learning_rate": 1.6828795958035715e-06, "loss": 8.913080215454102, "loss_alignment": 0.7548828125, "loss_alignment_w": 0.37744140625, "loss_sft": 0.18043475970625877, "loss_total_v6": 0.5570674613118172, "step": 5901 }, { "epoch": 1.3611675675675676, "grad_norm": 29.710521697998047, "learning_rate": 1.6813730209219952e-06, "loss": 8.861777305603027, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.16897338535636663, "loss_total_v6": 0.5538610816001892, "step": 5902 }, { "epoch": 1.3613981981981982, "grad_norm": 29.426227569580078, "learning_rate": 1.6798669843956956e-06, "loss": 8.802825927734375, "loss_alignment": 0.77001953125, "loss_alignment_w": 0.385009765625, "loss_sft": 0.16499904450029135, "loss_total_v6": 0.5501766465604305, "step": 5903 }, { "epoch": 1.3616288288288287, "grad_norm": 30.287151336669922, "learning_rate": 1.6783614864689828e-06, "loss": 9.133513450622559, "loss_alignment": 0.76953125, "loss_alignment_w": 0.384765625, "loss_sft": 0.18639937974512577, "loss_total_v6": 0.5708445832133293, "step": 5904 }, { "epoch": 1.3618594594594595, "grad_norm": 31.112489700317383, "learning_rate": 1.6768565273860854e-06, "loss": 8.521637916564941, "loss_alignment": 0.73291015625, "loss_alignment_w": 0.366455078125, "loss_sft": 0.16623880807310343, "loss_total_v6": 0.5326023511588573, "step": 5905 }, { "epoch": 1.36209009009009, "grad_norm": 30.947885513305664, "learning_rate": 1.675352107391139e-06, "loss": 8.871112823486328, "loss_alignment": 0.736328125, "loss_alignment_w": 0.3681640625, "loss_sft": 0.18721129558980465, "loss_total_v6": 0.5544445663690567, "step": 5906 }, { "epoch": 1.3623207207207209, "grad_norm": 35.847816467285156, "learning_rate": 1.6738482267281963e-06, "loss": 9.286521911621094, "loss_alignment": 0.7822265625, "loss_alignment_w": 0.39111328125, "loss_sft": 0.18892817571759224, "loss_total_v6": 0.5804076641798019, "step": 5907 }, { "epoch": 1.3625513513513514, "grad_norm": 39.32270431518555, "learning_rate": 1.672344885641219e-06, "loss": 8.99570369720459, "loss_alignment": 0.74658203125, "loss_alignment_w": 0.373291015625, "loss_sft": 0.18868107348680496, "loss_total_v6": 0.5622315034270287, "step": 5908 }, { "epoch": 1.362781981981982, "grad_norm": 37.23579788208008, "learning_rate": 1.6708420843740825e-06, "loss": 8.864594459533691, "loss_alignment": 0.77294921875, "loss_alignment_w": 0.386474609375, "loss_sft": 0.16657072119414806, "loss_total_v6": 0.5540371537208557, "step": 5909 }, { "epoch": 1.3630126126126125, "grad_norm": 27.308332443237305, "learning_rate": 1.6693398231705732e-06, "loss": 9.40466022491455, "loss_alignment": 0.7841796875, "loss_alignment_w": 0.39208984375, "loss_sft": 0.19440443255007267, "loss_total_v6": 0.5877912640571594, "step": 5910 }, { "epoch": 1.3632432432432433, "grad_norm": 26.36968421936035, "learning_rate": 1.6678381022743928e-06, "loss": 8.766663551330566, "loss_alignment": 0.7275390625, "loss_alignment_w": 0.36376953125, "loss_sft": 0.18410116247832775, "loss_total_v6": 0.5479164980351925, "step": 5911 }, { "epoch": 1.3634738738738739, "grad_norm": 29.915477752685547, "learning_rate": 1.6663369219291558e-06, "loss": 8.993964195251465, "loss_alignment": 0.7802734375, "loss_alignment_w": 0.39013671875, "loss_sft": 0.17152828443795443, "loss_total_v6": 0.5621227696537971, "step": 5912 }, { "epoch": 1.3637045045045044, "grad_norm": 28.460437774658203, "learning_rate": 1.6648362823783853e-06, "loss": 9.032995223999023, "loss_alignment": 0.76416015625, "loss_alignment_w": 0.382080078125, "loss_sft": 0.1820243764668703, "loss_total_v6": 0.5645622164011002, "step": 5913 }, { "epoch": 1.3639351351351352, "grad_norm": 32.542701721191406, "learning_rate": 1.6633361838655187e-06, "loss": 9.185599327087402, "loss_alignment": 0.76611328125, "loss_alignment_w": 0.383056640625, "loss_sft": 0.19151635468006134, "loss_total_v6": 0.574099950492382, "step": 5914 }, { "epoch": 1.3641657657657658, "grad_norm": 33.648956298828125, "learning_rate": 1.6618366266339048e-06, "loss": 9.388904571533203, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.19784477725625038, "loss_total_v6": 0.5868065357208252, "step": 5915 }, { "epoch": 1.3643963963963963, "grad_norm": 28.4334659576416, "learning_rate": 1.6603376109268042e-06, "loss": 9.299188613891602, "loss_alignment": 0.77978515625, "loss_alignment_w": 0.389892578125, "loss_sft": 0.191138856112957, "loss_total_v6": 0.5811993107199669, "step": 5916 }, { "epoch": 1.364627027027027, "grad_norm": 36.5094108581543, "learning_rate": 1.6588391369873918e-06, "loss": 9.378336906433105, "loss_alignment": 0.796875, "loss_alignment_w": 0.3984375, "loss_sft": 0.18734234105795622, "loss_total_v6": 0.5861460715532303, "step": 5917 }, { "epoch": 1.3648576576576577, "grad_norm": 28.455902099609375, "learning_rate": 1.6573412050587556e-06, "loss": 9.544327735900879, "loss_alignment": 0.79296875, "loss_alignment_w": 0.396484375, "loss_sft": 0.20070747379213572, "loss_total_v6": 0.5965204909443855, "step": 5918 }, { "epoch": 1.3650882882882882, "grad_norm": 28.777463912963867, "learning_rate": 1.6558438153838917e-06, "loss": 8.960865020751953, "loss_alignment": 0.76904296875, "loss_alignment_w": 0.384521484375, "loss_sft": 0.17464754916727543, "loss_total_v6": 0.5600540414452553, "step": 5919 }, { "epoch": 1.365318918918919, "grad_norm": 28.510713577270508, "learning_rate": 1.6543469682057105e-06, "loss": 9.168886184692383, "loss_alignment": 0.77587890625, "loss_alignment_w": 0.387939453125, "loss_sft": 0.18540589604526758, "loss_total_v6": 0.5730554163455963, "step": 5920 }, { "epoch": 1.3655495495495495, "grad_norm": 30.043598175048828, "learning_rate": 1.6528506637670317e-06, "loss": 9.186976432800293, "loss_alignment": 0.75830078125, "loss_alignment_w": 0.379150390625, "loss_sft": 0.1945779174566269, "loss_total_v6": 0.5741860494017601, "step": 5921 }, { "epoch": 1.36578018018018, "grad_norm": 30.1032657623291, "learning_rate": 1.6513549023105924e-06, "loss": 8.733009338378906, "loss_alignment": 0.7587890625, "loss_alignment_w": 0.37939453125, "loss_sft": 0.16765452735126019, "loss_total_v6": 0.5458130836486816, "step": 5922 }, { "epoch": 1.3660108108108109, "grad_norm": 30.21930503845215, "learning_rate": 1.649859684079036e-06, "loss": 8.894012451171875, "loss_alignment": 0.73779296875, "loss_alignment_w": 0.368896484375, "loss_sft": 0.18642992712557316, "loss_total_v6": 0.555875726044178, "step": 5923 }, { "epoch": 1.3662414414414414, "grad_norm": 32.32978439331055, "learning_rate": 1.6483650093149228e-06, "loss": 8.787035942077637, "loss_alignment": 0.7509765625, "loss_alignment_w": 0.37548828125, "loss_sft": 0.17327423114329576, "loss_total_v6": 0.5491897575557232, "step": 5924 }, { "epoch": 1.366472072072072, "grad_norm": 29.2957763671875, "learning_rate": 1.6468708782607213e-06, "loss": 8.967180252075195, "loss_alignment": 0.751953125, "loss_alignment_w": 0.3759765625, "loss_sft": 0.18486892897635698, "loss_total_v6": 0.5604487545788288, "step": 5925 }, { "epoch": 1.3667027027027028, "grad_norm": 32.35124969482422, "learning_rate": 1.6453772911588107e-06, "loss": 9.548429489135742, "loss_alignment": 0.75146484375, "loss_alignment_w": 0.375732421875, "loss_sft": 0.22102912329137325, "loss_total_v6": 0.5967768058180809, "step": 5926 }, { "epoch": 1.3669333333333333, "grad_norm": 28.584308624267578, "learning_rate": 1.6438842482514871e-06, "loss": 9.243675231933594, "loss_alignment": 0.7587890625, "loss_alignment_w": 0.37939453125, "loss_sft": 0.19783158786594868, "loss_total_v6": 0.5777296647429466, "step": 5927 }, { "epoch": 1.367163963963964, "grad_norm": 30.266712188720703, "learning_rate": 1.6423917497809533e-06, "loss": 9.72021484375, "loss_alignment": 0.80322265625, "loss_alignment_w": 0.401611328125, "loss_sft": 0.20593265816569328, "loss_total_v6": 0.6075134351849556, "step": 5928 }, { "epoch": 1.3673945945945947, "grad_norm": 29.559022903442383, "learning_rate": 1.640899795989328e-06, "loss": 9.243247985839844, "loss_alignment": 0.771484375, "loss_alignment_w": 0.3857421875, "loss_sft": 0.19110633991658688, "loss_total_v6": 0.5777029991149902, "step": 5929 }, { "epoch": 1.3676252252252252, "grad_norm": 29.726337432861328, "learning_rate": 1.6394083871186362e-06, "loss": 9.140031814575195, "loss_alignment": 0.81982421875, "loss_alignment_w": 0.409912109375, "loss_sft": 0.16190440207719803, "loss_total_v6": 0.5712519213557243, "step": 5930 }, { "epoch": 1.3678558558558558, "grad_norm": 30.838464736938477, "learning_rate": 1.6379175234108213e-06, "loss": 9.361635208129883, "loss_alignment": 0.76708984375, "loss_alignment_w": 0.383544921875, "loss_sft": 0.20109954196959734, "loss_total_v6": 0.5851022303104401, "step": 5931 }, { "epoch": 1.3680864864864866, "grad_norm": 31.979259490966797, "learning_rate": 1.6364272051077335e-06, "loss": 9.191387176513672, "loss_alignment": 0.78515625, "loss_alignment_w": 0.392578125, "loss_sft": 0.1818377673625946, "loss_total_v6": 0.5744616836309433, "step": 5932 }, { "epoch": 1.3683171171171171, "grad_norm": 31.00572967529297, "learning_rate": 1.6349374324511347e-06, "loss": 9.406501770019531, "loss_alignment": 0.75537109375, "loss_alignment_w": 0.377685546875, "loss_sft": 0.210678618401289, "loss_total_v6": 0.5879063978791237, "step": 5933 }, { "epoch": 1.3685477477477477, "grad_norm": 28.27968978881836, "learning_rate": 1.6334482056826977e-06, "loss": 9.230131149291992, "loss_alignment": 0.755859375, "loss_alignment_w": 0.3779296875, "loss_sft": 0.19938075728714466, "loss_total_v6": 0.5768831893801689, "step": 5934 }, { "epoch": 1.3687783783783785, "grad_norm": 31.07668685913086, "learning_rate": 1.6319595250440119e-06, "loss": 9.381412506103516, "loss_alignment": 0.7880859375, "loss_alignment_w": 0.39404296875, "loss_sft": 0.19305827654898167, "loss_total_v6": 0.5863383188843727, "step": 5935 }, { "epoch": 1.369009009009009, "grad_norm": 30.80776596069336, "learning_rate": 1.6304713907765713e-06, "loss": 9.623632431030273, "loss_alignment": 0.82470703125, "loss_alignment_w": 0.412353515625, "loss_sft": 0.18829956278204918, "loss_total_v6": 0.6014770418405533, "step": 5936 }, { "epoch": 1.3692396396396396, "grad_norm": 29.73867416381836, "learning_rate": 1.628983803121787e-06, "loss": 9.23594856262207, "loss_alignment": 0.75146484375, "loss_alignment_w": 0.375732421875, "loss_sft": 0.20210941322147846, "loss_total_v6": 0.577246755361557, "step": 5937 }, { "epoch": 1.3694702702702704, "grad_norm": 32.33853530883789, "learning_rate": 1.627496762320978e-06, "loss": 9.57788372039795, "loss_alignment": 0.779296875, "loss_alignment_w": 0.3896484375, "loss_sft": 0.20849625207483768, "loss_total_v6": 0.5986177399754524, "step": 5938 }, { "epoch": 1.369700900900901, "grad_norm": 30.028459548950195, "learning_rate": 1.6260102686153755e-06, "loss": 9.336671829223633, "loss_alignment": 0.78759765625, "loss_alignment_w": 0.393798828125, "loss_sft": 0.19036877900362015, "loss_total_v6": 0.583541989326477, "step": 5939 }, { "epoch": 1.3699315315315315, "grad_norm": 28.610448837280273, "learning_rate": 1.6245243222461198e-06, "loss": 9.089004516601562, "loss_alignment": 0.7529296875, "loss_alignment_w": 0.37646484375, "loss_sft": 0.1915979441255331, "loss_total_v6": 0.5680627822875977, "step": 5940 }, { "epoch": 1.3701621621621622, "grad_norm": 34.66123962402344, "learning_rate": 1.6230389234542665e-06, "loss": 9.309706687927246, "loss_alignment": 0.78515625, "loss_alignment_w": 0.392578125, "loss_sft": 0.18944641202688217, "loss_total_v6": 0.5818566754460335, "step": 5941 }, { "epoch": 1.3703927927927928, "grad_norm": 31.96110725402832, "learning_rate": 1.6215540724807815e-06, "loss": 9.46356201171875, "loss_alignment": 0.76513671875, "loss_alignment_w": 0.382568359375, "loss_sft": 0.20923996530473232, "loss_total_v6": 0.5914726406335831, "step": 5942 }, { "epoch": 1.3706234234234234, "grad_norm": 30.015371322631836, "learning_rate": 1.6200697695665402e-06, "loss": 8.892339706420898, "loss_alignment": 0.75634765625, "loss_alignment_w": 0.378173828125, "loss_sft": 0.1774295810610056, "loss_total_v6": 0.5557712726294994, "step": 5943 }, { "epoch": 1.3708540540540541, "grad_norm": 27.032424926757812, "learning_rate": 1.6185860149523286e-06, "loss": 8.774045944213867, "loss_alignment": 0.763671875, "loss_alignment_w": 0.3818359375, "loss_sft": 0.1658247821033001, "loss_total_v6": 0.5483778864145279, "step": 5944 }, { "epoch": 1.3710846846846847, "grad_norm": 29.801170349121094, "learning_rate": 1.6171028088788432e-06, "loss": 9.661280632019043, "loss_alignment": 0.7646484375, "loss_alignment_w": 0.38232421875, "loss_sft": 0.2215668484568596, "loss_total_v6": 0.6038300395011902, "step": 5945 }, { "epoch": 1.3713153153153153, "grad_norm": 29.593841552734375, "learning_rate": 1.6156201515866971e-06, "loss": 9.367206573486328, "loss_alignment": 0.77734375, "loss_alignment_w": 0.388671875, "loss_sft": 0.19609189964830875, "loss_total_v6": 0.5854504480957985, "step": 5946 }, { "epoch": 1.3715459459459458, "grad_norm": 31.79486083984375, "learning_rate": 1.6141380433164062e-06, "loss": 8.977862358093262, "loss_alignment": 0.7626953125, "loss_alignment_w": 0.38134765625, "loss_sft": 0.18091312795877457, "loss_total_v6": 0.5611163899302483, "step": 5947 }, { "epoch": 1.3717765765765766, "grad_norm": 29.77055549621582, "learning_rate": 1.6126564843084053e-06, "loss": 8.823352813720703, "loss_alignment": 0.7646484375, "loss_alignment_w": 0.38232421875, "loss_sft": 0.16963882371783257, "loss_total_v6": 0.5514595210552216, "step": 5948 }, { "epoch": 1.3720072072072071, "grad_norm": 30.24273681640625, "learning_rate": 1.611175474803035e-06, "loss": 9.030841827392578, "loss_alignment": 0.779296875, "loss_alignment_w": 0.3896484375, "loss_sft": 0.1737873312085867, "loss_total_v6": 0.5644275844097137, "step": 5949 }, { "epoch": 1.372237837837838, "grad_norm": 30.676620483398438, "learning_rate": 1.6096950150405454e-06, "loss": 8.885329246520996, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.17171184159815311, "loss_total_v6": 0.5553330630064011, "step": 5950 }, { "epoch": 1.3724684684684685, "grad_norm": 30.756057739257812, "learning_rate": 1.608215105261104e-06, "loss": 9.552814483642578, "loss_alignment": 0.791015625, "loss_alignment_w": 0.3955078125, "loss_sft": 0.20190931670367718, "loss_total_v6": 0.5970508977770805, "step": 5951 }, { "epoch": 1.372699099099099, "grad_norm": 30.07376480102539, "learning_rate": 1.606735745704784e-06, "loss": 9.744268417358398, "loss_alignment": 0.8017578125, "loss_alignment_w": 0.40087890625, "loss_sft": 0.20844302605837584, "loss_total_v6": 0.6090167462825775, "step": 5952 }, { "epoch": 1.3729297297297296, "grad_norm": 30.07942008972168, "learning_rate": 1.6052569366115677e-06, "loss": 9.384244918823242, "loss_alignment": 0.81103515625, "loss_alignment_w": 0.405517578125, "loss_sft": 0.17991435900330544, "loss_total_v6": 0.586515299975872, "step": 5953 }, { "epoch": 1.3731603603603604, "grad_norm": 32.44475173950195, "learning_rate": 1.6037786782213549e-06, "loss": 8.849754333496094, "loss_alignment": 0.787109375, "loss_alignment_w": 0.3935546875, "loss_sft": 0.15944813657552004, "loss_total_v6": 0.5531096532940865, "step": 5954 }, { "epoch": 1.373390990990991, "grad_norm": 31.371313095092773, "learning_rate": 1.6023009707739484e-06, "loss": 9.208154678344727, "loss_alignment": 0.751953125, "loss_alignment_w": 0.3759765625, "loss_sft": 0.19994511362165213, "loss_total_v6": 0.5755097009241581, "step": 5955 }, { "epoch": 1.3736216216216217, "grad_norm": 28.735794067382812, "learning_rate": 1.6008238145090693e-06, "loss": 9.557567596435547, "loss_alignment": 0.794921875, "loss_alignment_w": 0.3974609375, "loss_sft": 0.19906307756900787, "loss_total_v6": 0.5973479598760605, "step": 5956 }, { "epoch": 1.3738522522522523, "grad_norm": 34.598663330078125, "learning_rate": 1.5993472096663432e-06, "loss": 8.795079231262207, "loss_alignment": 0.7490234375, "loss_alignment_w": 0.37451171875, "loss_sft": 0.17463144939392805, "loss_total_v6": 0.5496924668550491, "step": 5957 }, { "epoch": 1.3740828828828828, "grad_norm": 32.38496398925781, "learning_rate": 1.5978711564853077e-06, "loss": 9.199104309082031, "loss_alignment": 0.79541015625, "loss_alignment_w": 0.397705078125, "loss_sft": 0.17758991289883852, "loss_total_v6": 0.5749440267682076, "step": 5958 }, { "epoch": 1.3743135135135134, "grad_norm": 30.721227645874023, "learning_rate": 1.596395655205411e-06, "loss": 8.572702407836914, "loss_alignment": 0.72216796875, "loss_alignment_w": 0.361083984375, "loss_sft": 0.17409950867295265, "loss_total_v6": 0.5357938818633556, "step": 5959 }, { "epoch": 1.3745441441441442, "grad_norm": 33.85832595825195, "learning_rate": 1.5949207060660138e-06, "loss": 9.275077819824219, "loss_alignment": 0.76708984375, "loss_alignment_w": 0.383544921875, "loss_sft": 0.19655940309166908, "loss_total_v6": 0.5796923562884331, "step": 5960 }, { "epoch": 1.3747747747747747, "grad_norm": 27.86758041381836, "learning_rate": 1.5934463093063862e-06, "loss": 9.194524765014648, "loss_alignment": 0.75390625, "loss_alignment_w": 0.376953125, "loss_sft": 0.19836083985865116, "loss_total_v6": 0.5746578425168991, "step": 5961 }, { "epoch": 1.3750054054054055, "grad_norm": 27.77117347717285, "learning_rate": 1.5919724651657075e-06, "loss": 8.956016540527344, "loss_alignment": 0.755859375, "loss_alignment_w": 0.3779296875, "loss_sft": 0.18282846361398697, "loss_total_v6": 0.5597510561347008, "step": 5962 }, { "epoch": 1.375236036036036, "grad_norm": 30.37836456298828, "learning_rate": 1.5904991738830682e-06, "loss": 9.92164421081543, "loss_alignment": 0.76171875, "loss_alignment_w": 0.380859375, "loss_sft": 0.24011311493813992, "loss_total_v6": 0.620102733373642, "step": 5963 }, { "epoch": 1.3754666666666666, "grad_norm": 27.48906707763672, "learning_rate": 1.5890264356974688e-06, "loss": 9.135746002197266, "loss_alignment": 0.75634765625, "loss_alignment_w": 0.378173828125, "loss_sft": 0.19229156151413918, "loss_total_v6": 0.5709841623902321, "step": 5964 }, { "epoch": 1.3756972972972972, "grad_norm": 31.893918991088867, "learning_rate": 1.5875542508478187e-06, "loss": 8.963624000549316, "loss_alignment": 0.74267578125, "loss_alignment_w": 0.371337890625, "loss_sft": 0.19001775421202183, "loss_total_v6": 0.5602265000343323, "step": 5965 }, { "epoch": 1.375927927927928, "grad_norm": 30.391138076782227, "learning_rate": 1.5860826195729402e-06, "loss": 9.383268356323242, "loss_alignment": 0.75732421875, "loss_alignment_w": 0.378662109375, "loss_sft": 0.20695297420024872, "loss_total_v6": 0.5864543095231056, "step": 5966 }, { "epoch": 1.3761585585585585, "grad_norm": 29.620067596435547, "learning_rate": 1.5846115421115671e-06, "loss": 9.121949195861816, "loss_alignment": 0.74658203125, "loss_alignment_w": 0.373291015625, "loss_sft": 0.19585424661636353, "loss_total_v6": 0.5701218172907829, "step": 5967 }, { "epoch": 1.3763891891891893, "grad_norm": 34.278839111328125, "learning_rate": 1.5831410187023388e-06, "loss": 9.221940994262695, "loss_alignment": 0.7880859375, "loss_alignment_w": 0.39404296875, "loss_sft": 0.18150443583726883, "loss_total_v6": 0.576371356844902, "step": 5968 }, { "epoch": 1.3766198198198198, "grad_norm": 31.6672420501709, "learning_rate": 1.5816710495838072e-06, "loss": 8.748015403747559, "loss_alignment": 0.73193359375, "loss_alignment_w": 0.365966796875, "loss_sft": 0.181531872600317, "loss_total_v6": 0.5467509813606739, "step": 5969 }, { "epoch": 1.3768504504504504, "grad_norm": 28.83943748474121, "learning_rate": 1.580201634994432e-06, "loss": 9.004159927368164, "loss_alignment": 0.77783203125, "loss_alignment_w": 0.388916015625, "loss_sft": 0.17477477062493563, "loss_total_v6": 0.5627600029110909, "step": 5970 }, { "epoch": 1.377081081081081, "grad_norm": 29.25251579284668, "learning_rate": 1.578732775172589e-06, "loss": 8.597579956054688, "loss_alignment": 0.75048828125, "loss_alignment_w": 0.375244140625, "loss_sft": 0.16314215026795864, "loss_total_v6": 0.5373487174510956, "step": 5971 }, { "epoch": 1.3773117117117117, "grad_norm": 30.275362014770508, "learning_rate": 1.5772644703565564e-06, "loss": 9.354679107666016, "loss_alignment": 0.7802734375, "loss_alignment_w": 0.39013671875, "loss_sft": 0.19436282105743885, "loss_total_v6": 0.5846673920750618, "step": 5972 }, { "epoch": 1.3775423423423423, "grad_norm": 30.132633209228516, "learning_rate": 1.5757967207845293e-06, "loss": 8.607704162597656, "loss_alignment": 0.75, "loss_alignment_w": 0.375, "loss_sft": 0.16293572448194027, "loss_total_v6": 0.5379815101623535, "step": 5973 }, { "epoch": 1.377772972972973, "grad_norm": 31.947166442871094, "learning_rate": 1.5743295266946073e-06, "loss": 9.170411109924316, "loss_alignment": 0.7626953125, "loss_alignment_w": 0.38134765625, "loss_sft": 0.1909943539649248, "loss_total_v6": 0.5731506943702698, "step": 5974 }, { "epoch": 1.3780036036036036, "grad_norm": 30.798049926757812, "learning_rate": 1.572862888324801e-06, "loss": 9.226369857788086, "loss_alignment": 0.76416015625, "loss_alignment_w": 0.382080078125, "loss_sft": 0.19421712681651115, "loss_total_v6": 0.5766481459140778, "step": 5975 }, { "epoch": 1.3782342342342342, "grad_norm": 32.33062744140625, "learning_rate": 1.5713968059130347e-06, "loss": 8.787469863891602, "loss_alignment": 0.7353515625, "loss_alignment_w": 0.36767578125, "loss_sft": 0.18016778491437435, "loss_total_v6": 0.5492168702185154, "step": 5976 }, { "epoch": 1.3784648648648647, "grad_norm": 32.19588088989258, "learning_rate": 1.5699312796971382e-06, "loss": 9.228724479675293, "loss_alignment": 0.78466796875, "loss_alignment_w": 0.392333984375, "loss_sft": 0.18330161273479462, "loss_total_v6": 0.5767952725291252, "step": 5977 }, { "epoch": 1.3786954954954955, "grad_norm": 35.67964553833008, "learning_rate": 1.5684663099148512e-06, "loss": 9.273037910461426, "loss_alignment": 0.78955078125, "loss_alignment_w": 0.394775390625, "loss_sft": 0.18474371545016766, "loss_total_v6": 0.5795648768544197, "step": 5978 }, { "epoch": 1.378926126126126, "grad_norm": 27.575170516967773, "learning_rate": 1.5670018968038259e-06, "loss": 9.031499862670898, "loss_alignment": 0.732421875, "loss_alignment_w": 0.3662109375, "loss_sft": 0.19839508831501007, "loss_total_v6": 0.5644687116146088, "step": 5979 }, { "epoch": 1.3791567567567569, "grad_norm": 31.110071182250977, "learning_rate": 1.5655380406016236e-06, "loss": 9.018630981445312, "loss_alignment": 0.75830078125, "loss_alignment_w": 0.379150390625, "loss_sft": 0.1848954875022173, "loss_total_v6": 0.5636644288897514, "step": 5980 }, { "epoch": 1.3793873873873874, "grad_norm": 29.047752380371094, "learning_rate": 1.5640747415457136e-06, "loss": 9.005746841430664, "loss_alignment": 0.73583984375, "loss_alignment_w": 0.367919921875, "loss_sft": 0.19506136886775494, "loss_total_v6": 0.5628591924905777, "step": 5981 }, { "epoch": 1.379618018018018, "grad_norm": 30.426654815673828, "learning_rate": 1.5626119998734757e-06, "loss": 9.262907028198242, "loss_alignment": 0.7568359375, "loss_alignment_w": 0.37841796875, "loss_sft": 0.20069684088230133, "loss_total_v6": 0.5789316818118095, "step": 5982 }, { "epoch": 1.3798486486486485, "grad_norm": 33.19236373901367, "learning_rate": 1.5611498158221983e-06, "loss": 8.87515640258789, "loss_alignment": 0.76708984375, "loss_alignment_w": 0.383544921875, "loss_sft": 0.17176270578056574, "loss_total_v6": 0.5546972788870335, "step": 5983 }, { "epoch": 1.3800792792792793, "grad_norm": 28.26053237915039, "learning_rate": 1.55968818962908e-06, "loss": 9.21379566192627, "loss_alignment": 0.75048828125, "loss_alignment_w": 0.375244140625, "loss_sft": 0.20110641792416573, "loss_total_v6": 0.5758622661232948, "step": 5984 }, { "epoch": 1.3803099099099099, "grad_norm": 30.219635009765625, "learning_rate": 1.5582271215312294e-06, "loss": 9.049195289611816, "loss_alignment": 0.765625, "loss_alignment_w": 0.3828125, "loss_sft": 0.1825943924486637, "loss_total_v6": 0.5655747205018997, "step": 5985 }, { "epoch": 1.3805405405405407, "grad_norm": 29.4150447845459, "learning_rate": 1.5567666117656671e-06, "loss": 8.990970611572266, "loss_alignment": 0.7646484375, "loss_alignment_w": 0.38232421875, "loss_sft": 0.17971823550760746, "loss_total_v6": 0.5619356408715248, "step": 5986 }, { "epoch": 1.3807711711711712, "grad_norm": 31.461503982543945, "learning_rate": 1.555306660569318e-06, "loss": 9.101795196533203, "loss_alignment": 0.73828125, "loss_alignment_w": 0.369140625, "loss_sft": 0.1992179760709405, "loss_total_v6": 0.5688621401786804, "step": 5987 }, { "epoch": 1.3810018018018018, "grad_norm": 32.38654708862305, "learning_rate": 1.5538472681790185e-06, "loss": 9.038081169128418, "loss_alignment": 0.79638671875, "loss_alignment_w": 0.398193359375, "loss_sft": 0.16754118166863918, "loss_total_v6": 0.564880046993494, "step": 5988 }, { "epoch": 1.3812324324324323, "grad_norm": 29.907711029052734, "learning_rate": 1.5523884348315137e-06, "loss": 9.163755416870117, "loss_alignment": 0.75732421875, "loss_alignment_w": 0.378662109375, "loss_sft": 0.1943014208227396, "loss_total_v6": 0.5727346763014793, "step": 5989 }, { "epoch": 1.381463063063063, "grad_norm": 31.98550033569336, "learning_rate": 1.550930160763462e-06, "loss": 9.066500663757324, "loss_alignment": 0.72705078125, "loss_alignment_w": 0.363525390625, "loss_sft": 0.20174233615398407, "loss_total_v6": 0.5666562803089619, "step": 5990 }, { "epoch": 1.3816936936936937, "grad_norm": 32.57960510253906, "learning_rate": 1.5494724462114247e-06, "loss": 9.122285842895508, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.1859265547245741, "loss_total_v6": 0.5701428577303886, "step": 5991 }, { "epoch": 1.3819243243243244, "grad_norm": 33.90379333496094, "learning_rate": 1.5480152914118784e-06, "loss": 9.448440551757812, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.20676898770034313, "loss_total_v6": 0.5905275270342827, "step": 5992 }, { "epoch": 1.382154954954955, "grad_norm": 28.268115997314453, "learning_rate": 1.5465586966012047e-06, "loss": 8.826360702514648, "loss_alignment": 0.76806640625, "loss_alignment_w": 0.384033203125, "loss_sft": 0.1662867832928896, "loss_total_v6": 0.5516475290060043, "step": 5993 }, { "epoch": 1.3823855855855856, "grad_norm": 36.549163818359375, "learning_rate": 1.5451026620156944e-06, "loss": 9.22830867767334, "loss_alignment": 0.74951171875, "loss_alignment_w": 0.374755859375, "loss_sft": 0.2024711985141039, "loss_total_v6": 0.5767692774534225, "step": 5994 }, { "epoch": 1.382616216216216, "grad_norm": 32.52957534790039, "learning_rate": 1.5436471878915516e-06, "loss": 8.9302978515625, "loss_alignment": 0.72509765625, "loss_alignment_w": 0.362548828125, "loss_sft": 0.19495392590761185, "loss_total_v6": 0.5581436231732368, "step": 5995 }, { "epoch": 1.3828468468468469, "grad_norm": 29.710908889770508, "learning_rate": 1.5421922744648849e-06, "loss": 9.253976821899414, "loss_alignment": 0.7509765625, "loss_alignment_w": 0.37548828125, "loss_sft": 0.20363293960690498, "loss_total_v6": 0.5783735364675522, "step": 5996 }, { "epoch": 1.3830774774774774, "grad_norm": 32.3530158996582, "learning_rate": 1.5407379219717127e-06, "loss": 8.840791702270508, "loss_alignment": 0.7578125, "loss_alignment_w": 0.37890625, "loss_sft": 0.1745129832997918, "loss_total_v6": 0.5525494813919067, "step": 5997 }, { "epoch": 1.3833081081081082, "grad_norm": 34.648006439208984, "learning_rate": 1.5392841306479667e-06, "loss": 10.018342971801758, "loss_alignment": 0.7890625, "loss_alignment_w": 0.39453125, "loss_sft": 0.23076067864894867, "loss_total_v6": 0.6261464133858681, "step": 5998 }, { "epoch": 1.3835387387387388, "grad_norm": 36.83991622924805, "learning_rate": 1.53783090072948e-06, "loss": 9.226916313171387, "loss_alignment": 0.77392578125, "loss_alignment_w": 0.386962890625, "loss_sft": 0.19048229046165943, "loss_total_v6": 0.5766822546720505, "step": 5999 }, { "epoch": 1.3837693693693693, "grad_norm": 31.7681827545166, "learning_rate": 1.5363782324520033e-06, "loss": 8.86328125, "loss_alignment": 0.7578125, "loss_alignment_w": 0.37890625, "loss_sft": 0.17356867901980877, "loss_total_v6": 0.553955040872097, "step": 6000 }, { "epoch": 1.3837693693693693, "eval_loss": 0.8503372669219971, "eval_loss_alignment": 0.7888863620148402, "eval_loss_alignment_w": 0.3944431810074201, "eval_loss_contrast": 0.0, "eval_loss_highlevel": 0.0, "eval_loss_sft": 0.4558941211942668, "eval_loss_total_v6": 0.850337302167666, "eval_loss_wm_recon": 0.0, "eval_loss_wm_recon_w": 0.0, "eval_runtime": 334.4309, "eval_samples_per_second": 5.218, "eval_steps_per_second": 0.655, "step": 6000 }, { "epoch": 1.384, "grad_norm": 32.54508590698242, "learning_rate": 1.5349261260511894e-06, "loss": 8.9979829788208, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.17422088235616684, "loss_total_v6": 0.5623739436268806, "step": 6001 }, { "epoch": 1.3842306306306307, "grad_norm": 31.955066680908203, "learning_rate": 1.5334745817626024e-06, "loss": 8.870038986206055, "loss_alignment": 0.74267578125, "loss_alignment_w": 0.371337890625, "loss_sft": 0.1833447264507413, "loss_total_v6": 0.5543774329125881, "step": 6002 }, { "epoch": 1.3844612612612612, "grad_norm": 32.1891975402832, "learning_rate": 1.5320235998217137e-06, "loss": 9.224285125732422, "loss_alignment": 0.7822265625, "loss_alignment_w": 0.39111328125, "loss_sft": 0.18567914701998234, "loss_total_v6": 0.576517753303051, "step": 6003 }, { "epoch": 1.384691891891892, "grad_norm": 39.4611701965332, "learning_rate": 1.5305731804639067e-06, "loss": 9.01365852355957, "loss_alignment": 0.78173828125, "loss_alignment_w": 0.390869140625, "loss_sft": 0.17285071220248938, "loss_total_v6": 0.5633536502718925, "step": 6004 }, { "epoch": 1.3849225225225226, "grad_norm": 38.218833923339844, "learning_rate": 1.5291233239244728e-06, "loss": 9.514277458190918, "loss_alignment": 0.80810546875, "loss_alignment_w": 0.404052734375, "loss_sft": 0.1916729984804988, "loss_total_v6": 0.5946423560380936, "step": 6005 }, { "epoch": 1.3851531531531531, "grad_norm": 30.63805389404297, "learning_rate": 1.5276740304386106e-06, "loss": 9.125242233276367, "loss_alignment": 0.75634765625, "loss_alignment_w": 0.378173828125, "loss_sft": 0.1922606136649847, "loss_total_v6": 0.5703276544809341, "step": 6006 }, { "epoch": 1.3853837837837837, "grad_norm": 27.08920669555664, "learning_rate": 1.5262253002414262e-06, "loss": 8.8179931640625, "loss_alignment": 0.73486328125, "loss_alignment_w": 0.367431640625, "loss_sft": 0.1843032818287611, "loss_total_v6": 0.5511245653033257, "step": 6007 }, { "epoch": 1.3856144144144145, "grad_norm": 29.864946365356445, "learning_rate": 1.5247771335679372e-06, "loss": 8.956558227539062, "loss_alignment": 0.7568359375, "loss_alignment_w": 0.37841796875, "loss_sft": 0.18237400148063898, "loss_total_v6": 0.5597849041223526, "step": 6008 }, { "epoch": 1.385845045045045, "grad_norm": 29.435821533203125, "learning_rate": 1.5233295306530665e-06, "loss": 9.122265815734863, "loss_alignment": 0.7724609375, "loss_alignment_w": 0.38623046875, "loss_sft": 0.1834533680230379, "loss_total_v6": 0.5701416209340096, "step": 6009 }, { "epoch": 1.3860756756756758, "grad_norm": 36.05534744262695, "learning_rate": 1.5218824917316487e-06, "loss": 9.549673080444336, "loss_alignment": 0.806640625, "loss_alignment_w": 0.4033203125, "loss_sft": 0.19396145083010197, "loss_total_v6": 0.5968545526266098, "step": 6010 }, { "epoch": 1.3863063063063064, "grad_norm": 28.129276275634766, "learning_rate": 1.5204360170384286e-06, "loss": 9.072513580322266, "loss_alignment": 0.76025390625, "loss_alignment_w": 0.380126953125, "loss_sft": 0.18664578907191753, "loss_total_v6": 0.5670321136713028, "step": 6011 }, { "epoch": 1.386536936936937, "grad_norm": 28.508831024169922, "learning_rate": 1.5189901068080536e-06, "loss": 9.366472244262695, "loss_alignment": 0.81396484375, "loss_alignment_w": 0.406982421875, "loss_sft": 0.17869674414396286, "loss_total_v6": 0.5854045152664185, "step": 6012 }, { "epoch": 1.3867675675675675, "grad_norm": 35.61263656616211, "learning_rate": 1.5175447612750838e-06, "loss": 9.22854995727539, "loss_alignment": 0.78955078125, "loss_alignment_w": 0.394775390625, "loss_sft": 0.18231415376067162, "loss_total_v6": 0.5767843499779701, "step": 6013 }, { "epoch": 1.3869981981981983, "grad_norm": 36.4391975402832, "learning_rate": 1.5160999806739834e-06, "loss": 9.022488594055176, "loss_alignment": 0.78759765625, "loss_alignment_w": 0.393798828125, "loss_sft": 0.17028982006013393, "loss_total_v6": 0.5639055520296097, "step": 6014 }, { "epoch": 1.3872288288288288, "grad_norm": 29.823030471801758, "learning_rate": 1.514655765239133e-06, "loss": 8.843494415283203, "loss_alignment": 0.7724609375, "loss_alignment_w": 0.38623046875, "loss_sft": 0.166411641985178, "loss_total_v6": 0.5527184009552002, "step": 6015 }, { "epoch": 1.3874594594594596, "grad_norm": 29.594268798828125, "learning_rate": 1.513212115204812e-06, "loss": 9.501958847045898, "loss_alignment": 0.7939453125, "loss_alignment_w": 0.39697265625, "loss_sft": 0.19720494374632835, "loss_total_v6": 0.5938724130392075, "step": 6016 }, { "epoch": 1.3876900900900901, "grad_norm": 32.19852828979492, "learning_rate": 1.5117690308052164e-06, "loss": 8.910736083984375, "loss_alignment": 0.775390625, "loss_alignment_w": 0.3876953125, "loss_sft": 0.16882899776101112, "loss_total_v6": 0.5569210276007652, "step": 6017 }, { "epoch": 1.3879207207207207, "grad_norm": 32.269412994384766, "learning_rate": 1.5103265122744448e-06, "loss": 9.045942306518555, "loss_alignment": 0.7568359375, "loss_alignment_w": 0.37841796875, "loss_sft": 0.18666349910199642, "loss_total_v6": 0.5653713718056679, "step": 6018 }, { "epoch": 1.3881513513513513, "grad_norm": 30.24351692199707, "learning_rate": 1.5088845598465047e-06, "loss": 9.433113098144531, "loss_alignment": 0.78271484375, "loss_alignment_w": 0.391357421875, "loss_sft": 0.1979680433869362, "loss_total_v6": 0.5895695835351944, "step": 6019 }, { "epoch": 1.388381981981982, "grad_norm": 31.514476776123047, "learning_rate": 1.5074431737553158e-06, "loss": 8.94485092163086, "loss_alignment": 0.78369140625, "loss_alignment_w": 0.391845703125, "loss_sft": 0.16658184211701155, "loss_total_v6": 0.5590531527996063, "step": 6020 }, { "epoch": 1.3886126126126126, "grad_norm": 29.840946197509766, "learning_rate": 1.5060023542347013e-06, "loss": 9.446481704711914, "loss_alignment": 0.7802734375, "loss_alignment_w": 0.39013671875, "loss_sft": 0.200619300827384, "loss_total_v6": 0.590405061841011, "step": 6021 }, { "epoch": 1.3888432432432434, "grad_norm": 29.422733306884766, "learning_rate": 1.5045621015183927e-06, "loss": 8.784473419189453, "loss_alignment": 0.73681640625, "loss_alignment_w": 0.368408203125, "loss_sft": 0.17979739792644978, "loss_total_v6": 0.5490295886993408, "step": 6022 }, { "epoch": 1.389073873873874, "grad_norm": 35.21977233886719, "learning_rate": 1.5031224158400343e-06, "loss": 8.863672256469727, "loss_alignment": 0.74658203125, "loss_alignment_w": 0.373291015625, "loss_sft": 0.18091733381152153, "loss_total_v6": 0.5539794787764549, "step": 6023 }, { "epoch": 1.3893045045045045, "grad_norm": 29.65606117248535, "learning_rate": 1.5016832974331725e-06, "loss": 8.641836166381836, "loss_alignment": 0.74169921875, "loss_alignment_w": 0.370849609375, "loss_sft": 0.16824284940958023, "loss_total_v6": 0.5401147678494453, "step": 6024 }, { "epoch": 1.389535135135135, "grad_norm": 31.63178253173828, "learning_rate": 1.5002447465312675e-06, "loss": 9.255316734313965, "loss_alignment": 0.77294921875, "loss_alignment_w": 0.386474609375, "loss_sft": 0.1912960521876812, "loss_total_v6": 0.5784573070704937, "step": 6025 }, { "epoch": 1.3897657657657658, "grad_norm": 29.939090728759766, "learning_rate": 1.4988067633676818e-06, "loss": 8.77214527130127, "loss_alignment": 0.7275390625, "loss_alignment_w": 0.36376953125, "loss_sft": 0.18400124832987785, "loss_total_v6": 0.5482590869069099, "step": 6026 }, { "epoch": 1.3899963963963964, "grad_norm": 31.391054153442383, "learning_rate": 1.497369348175689e-06, "loss": 8.829662322998047, "loss_alignment": 0.73681640625, "loss_alignment_w": 0.368408203125, "loss_sft": 0.18335417099297047, "loss_total_v6": 0.5518539026379585, "step": 6027 }, { "epoch": 1.3902270270270272, "grad_norm": 25.229881286621094, "learning_rate": 1.4959325011884684e-06, "loss": 9.176412582397461, "loss_alignment": 0.74072265625, "loss_alignment_w": 0.370361328125, "loss_sft": 0.20270667783915997, "loss_total_v6": 0.5735257640480995, "step": 6028 }, { "epoch": 1.3904576576576577, "grad_norm": 29.35004997253418, "learning_rate": 1.4944962226391102e-06, "loss": 9.189030647277832, "loss_alignment": 0.74169921875, "loss_alignment_w": 0.370849609375, "loss_sft": 0.20393784157931805, "loss_total_v6": 0.5743144303560257, "step": 6029 }, { "epoch": 1.3906882882882883, "grad_norm": 30.597885131835938, "learning_rate": 1.4930605127606124e-06, "loss": 9.294997215270996, "loss_alignment": 0.80615234375, "loss_alignment_w": 0.403076171875, "loss_sft": 0.17751020472496748, "loss_total_v6": 0.5809373334050179, "step": 6030 }, { "epoch": 1.3909189189189188, "grad_norm": 29.463396072387695, "learning_rate": 1.4916253717858774e-06, "loss": 9.058958053588867, "loss_alignment": 0.74365234375, "loss_alignment_w": 0.371826171875, "loss_sft": 0.19423664174973965, "loss_total_v6": 0.5661848708987236, "step": 6031 }, { "epoch": 1.3911495495495496, "grad_norm": 27.131107330322266, "learning_rate": 1.4901907999477167e-06, "loss": 9.219053268432617, "loss_alignment": 0.75341796875, "loss_alignment_w": 0.376708984375, "loss_sft": 0.2003211062401533, "loss_total_v6": 0.5761908292770386, "step": 6032 }, { "epoch": 1.3913801801801802, "grad_norm": 29.10009002685547, "learning_rate": 1.4887567974788492e-06, "loss": 9.081740379333496, "loss_alignment": 0.7470703125, "loss_alignment_w": 0.37353515625, "loss_sft": 0.19273083750158548, "loss_total_v6": 0.5676087625324726, "step": 6033 }, { "epoch": 1.3916108108108107, "grad_norm": 32.72602462768555, "learning_rate": 1.487323364611904e-06, "loss": 9.256490707397461, "loss_alignment": 0.7666015625, "loss_alignment_w": 0.38330078125, "loss_sft": 0.19448226317763329, "loss_total_v6": 0.5785306990146637, "step": 6034 }, { "epoch": 1.3918414414414415, "grad_norm": 29.088491439819336, "learning_rate": 1.485890501579414e-06, "loss": 9.082342147827148, "loss_alignment": 0.7490234375, "loss_alignment_w": 0.37451171875, "loss_sft": 0.19365347921848297, "loss_total_v6": 0.5676464065909386, "step": 6035 }, { "epoch": 1.392072072072072, "grad_norm": 30.507017135620117, "learning_rate": 1.4844582086138231e-06, "loss": 9.086808204650879, "loss_alignment": 0.78125, "loss_alignment_w": 0.390625, "loss_sft": 0.17617136985063553, "loss_total_v6": 0.5679255351424217, "step": 6036 }, { "epoch": 1.3923027027027026, "grad_norm": 28.898412704467773, "learning_rate": 1.4830264859474814e-06, "loss": 9.322036743164062, "loss_alignment": 0.77490234375, "loss_alignment_w": 0.387451171875, "loss_sft": 0.195984847843647, "loss_total_v6": 0.5826272964477539, "step": 6037 }, { "epoch": 1.3925333333333334, "grad_norm": 31.012372970581055, "learning_rate": 1.4815953338126432e-06, "loss": 9.170026779174805, "loss_alignment": 0.75, "loss_alignment_w": 0.375, "loss_sft": 0.19736375659704208, "loss_total_v6": 0.5731266811490059, "step": 6038 }, { "epoch": 1.392763963963964, "grad_norm": 32.46270751953125, "learning_rate": 1.4801647524414775e-06, "loss": 9.179766654968262, "loss_alignment": 0.77197265625, "loss_alignment_w": 0.385986328125, "loss_sft": 0.1871082168072462, "loss_total_v6": 0.5737353935837746, "step": 6039 }, { "epoch": 1.3929945945945945, "grad_norm": 28.195016860961914, "learning_rate": 1.4787347420660541e-06, "loss": 9.293943405151367, "loss_alignment": 0.7626953125, "loss_alignment_w": 0.38134765625, "loss_sft": 0.1987761314958334, "loss_total_v6": 0.5808714777231216, "step": 6040 }, { "epoch": 1.3932252252252253, "grad_norm": 28.347448348999023, "learning_rate": 1.4773053029183514e-06, "loss": 9.008543014526367, "loss_alignment": 0.76220703125, "loss_alignment_w": 0.381103515625, "loss_sft": 0.18182361871004105, "loss_total_v6": 0.5630339533090591, "step": 6041 }, { "epoch": 1.3934558558558559, "grad_norm": 33.9819221496582, "learning_rate": 1.4758764352302596e-06, "loss": 9.404558181762695, "loss_alignment": 0.76611328125, "loss_alignment_w": 0.383056640625, "loss_sft": 0.20436202734708786, "loss_total_v6": 0.5877848491072655, "step": 6042 }, { "epoch": 1.3936864864864864, "grad_norm": 29.04290771484375, "learning_rate": 1.4744481392335692e-06, "loss": 9.124991416931152, "loss_alignment": 0.76513671875, "loss_alignment_w": 0.382568359375, "loss_sft": 0.18748419918119907, "loss_total_v6": 0.5703119486570358, "step": 6043 }, { "epoch": 1.3939171171171172, "grad_norm": 31.655366897583008, "learning_rate": 1.4730204151599848e-06, "loss": 9.962481498718262, "loss_alignment": 0.79296875, "loss_alignment_w": 0.396484375, "loss_sft": 0.22607916593551636, "loss_total_v6": 0.6226550936698914, "step": 6044 }, { "epoch": 1.3941477477477477, "grad_norm": 29.198957443237305, "learning_rate": 1.4715932632411145e-06, "loss": 9.297669410705566, "loss_alignment": 0.75390625, "loss_alignment_w": 0.376953125, "loss_sft": 0.20419704355299473, "loss_total_v6": 0.5811043679714203, "step": 6045 }, { "epoch": 1.3943783783783783, "grad_norm": 29.599868774414062, "learning_rate": 1.4701666837084726e-06, "loss": 8.742103576660156, "loss_alignment": 0.74609375, "loss_alignment_w": 0.373046875, "loss_sft": 0.17202231101691723, "loss_total_v6": 0.5463814288377762, "step": 6046 }, { "epoch": 1.394609009009009, "grad_norm": 27.471771240234375, "learning_rate": 1.468740676793482e-06, "loss": 8.516539573669434, "loss_alignment": 0.74658203125, "loss_alignment_w": 0.373291015625, "loss_sft": 0.15938946790993214, "loss_total_v6": 0.5322837345302105, "step": 6047 }, { "epoch": 1.3948396396396396, "grad_norm": 29.052154541015625, "learning_rate": 1.467315242727474e-06, "loss": 9.831517219543457, "loss_alignment": 0.81640625, "loss_alignment_w": 0.408203125, "loss_sft": 0.20693808794021606, "loss_total_v6": 0.6144698187708855, "step": 6048 }, { "epoch": 1.3950702702702702, "grad_norm": 30.13079071044922, "learning_rate": 1.4658903817416864e-06, "loss": 8.647130966186523, "loss_alignment": 0.75341796875, "loss_alignment_w": 0.376708984375, "loss_sft": 0.16291273944079876, "loss_total_v6": 0.5404456853866577, "step": 6049 }, { "epoch": 1.395300900900901, "grad_norm": 46.854190826416016, "learning_rate": 1.4644660940672628e-06, "loss": 9.377043724060059, "loss_alignment": 0.78515625, "loss_alignment_w": 0.392578125, "loss_sft": 0.193487080745399, "loss_total_v6": 0.5860652327537537, "step": 6050 }, { "epoch": 1.3955315315315315, "grad_norm": 32.258174896240234, "learning_rate": 1.4630423799352545e-06, "loss": 9.350879669189453, "loss_alignment": 0.78564453125, "loss_alignment_w": 0.392822265625, "loss_sft": 0.19214175269007683, "loss_total_v6": 0.5844299644231796, "step": 6051 }, { "epoch": 1.395762162162162, "grad_norm": 27.154598236083984, "learning_rate": 1.4616192395766189e-06, "loss": 9.03361701965332, "loss_alignment": 0.75341796875, "loss_alignment_w": 0.376708984375, "loss_sft": 0.18682395573705435, "loss_total_v6": 0.5646010488271713, "step": 6052 }, { "epoch": 1.3959927927927929, "grad_norm": 30.373876571655273, "learning_rate": 1.46019667322222e-06, "loss": 9.49513053894043, "loss_alignment": 0.7919921875, "loss_alignment_w": 0.39599609375, "loss_sft": 0.19766317307949066, "loss_total_v6": 0.5934456437826157, "step": 6053 }, { "epoch": 1.3962234234234234, "grad_norm": 33.50051498413086, "learning_rate": 1.4587746811028314e-06, "loss": 9.183238983154297, "loss_alignment": 0.771484375, "loss_alignment_w": 0.3857421875, "loss_sft": 0.188545985147357, "loss_total_v6": 0.5739524811506271, "step": 6054 }, { "epoch": 1.396454054054054, "grad_norm": 29.689653396606445, "learning_rate": 1.4573532634491333e-06, "loss": 8.673328399658203, "loss_alignment": 0.7578125, "loss_alignment_w": 0.37890625, "loss_sft": 0.16334462258964777, "loss_total_v6": 0.5420830175280571, "step": 6055 }, { "epoch": 1.3966846846846848, "grad_norm": 28.807870864868164, "learning_rate": 1.4559324204917102e-06, "loss": 9.180091857910156, "loss_alignment": 0.7841796875, "loss_alignment_w": 0.39208984375, "loss_sft": 0.18192524276673794, "loss_total_v6": 0.5737556964159012, "step": 6056 }, { "epoch": 1.3969153153153153, "grad_norm": 29.518152236938477, "learning_rate": 1.4545121524610538e-06, "loss": 8.79627799987793, "loss_alignment": 0.76708984375, "loss_alignment_w": 0.383544921875, "loss_sft": 0.16590197570621967, "loss_total_v6": 0.5497673414647579, "step": 6057 }, { "epoch": 1.3971459459459459, "grad_norm": 33.18463134765625, "learning_rate": 1.4530924595875624e-06, "loss": 8.952742576599121, "loss_alignment": 0.7578125, "loss_alignment_w": 0.37890625, "loss_sft": 0.18085381016135216, "loss_total_v6": 0.5595464184880257, "step": 6058 }, { "epoch": 1.3973765765765767, "grad_norm": 31.04057502746582, "learning_rate": 1.4516733421015439e-06, "loss": 9.226165771484375, "loss_alignment": 0.73583984375, "loss_alignment_w": 0.367919921875, "loss_sft": 0.20802884176373482, "loss_total_v6": 0.5766354016959667, "step": 6059 }, { "epoch": 1.3976072072072072, "grad_norm": 31.2918643951416, "learning_rate": 1.450254800233209e-06, "loss": 9.385558128356934, "loss_alignment": 0.78564453125, "loss_alignment_w": 0.392822265625, "loss_sft": 0.19455330725759268, "loss_total_v6": 0.5865973681211472, "step": 6060 }, { "epoch": 1.3978378378378378, "grad_norm": 32.86859130859375, "learning_rate": 1.4488368342126792e-06, "loss": 9.390239715576172, "loss_alignment": 0.755859375, "loss_alignment_w": 0.3779296875, "loss_sft": 0.20873135328292847, "loss_total_v6": 0.5868899375200272, "step": 6061 }, { "epoch": 1.3980684684684686, "grad_norm": 33.63249206542969, "learning_rate": 1.4474194442699785e-06, "loss": 8.85040283203125, "loss_alignment": 0.7470703125, "loss_alignment_w": 0.37353515625, "loss_sft": 0.17993545345962048, "loss_total_v6": 0.5531501770019531, "step": 6062 }, { "epoch": 1.398299099099099, "grad_norm": 27.851802825927734, "learning_rate": 1.4460026306350378e-06, "loss": 9.036123275756836, "loss_alignment": 0.77001953125, "loss_alignment_w": 0.385009765625, "loss_sft": 0.17985472921282053, "loss_total_v6": 0.5647576749324799, "step": 6063 }, { "epoch": 1.3985297297297297, "grad_norm": 31.505456924438477, "learning_rate": 1.444586393537699e-06, "loss": 9.230525970458984, "loss_alignment": 0.80029296875, "loss_alignment_w": 0.400146484375, "loss_sft": 0.1759832315146923, "loss_total_v6": 0.5769078731536865, "step": 6064 }, { "epoch": 1.3987603603603604, "grad_norm": 35.1982307434082, "learning_rate": 1.443170733207706e-06, "loss": 8.915447235107422, "loss_alignment": 0.7578125, "loss_alignment_w": 0.37890625, "loss_sft": 0.1781260846182704, "loss_total_v6": 0.5572154596447945, "step": 6065 }, { "epoch": 1.398990990990991, "grad_norm": 36.57386016845703, "learning_rate": 1.4417556498747081e-06, "loss": 8.754222869873047, "loss_alignment": 0.70947265625, "loss_alignment_w": 0.354736328125, "loss_sft": 0.19186860416084528, "loss_total_v6": 0.5471389703452587, "step": 6066 }, { "epoch": 1.3992216216216216, "grad_norm": 26.65218162536621, "learning_rate": 1.440341143768268e-06, "loss": 9.204294204711914, "loss_alignment": 0.7509765625, "loss_alignment_w": 0.37548828125, "loss_sft": 0.1987729836255312, "loss_total_v6": 0.5752683579921722, "step": 6067 }, { "epoch": 1.3994522522522521, "grad_norm": 30.214601516723633, "learning_rate": 1.4389272151178452e-06, "loss": 9.339010238647461, "loss_alignment": 0.775390625, "loss_alignment_w": 0.3876953125, "loss_sft": 0.19568768609315157, "loss_total_v6": 0.5836881846189499, "step": 6068 }, { "epoch": 1.399682882882883, "grad_norm": 30.649253845214844, "learning_rate": 1.4375138641528146e-06, "loss": 9.538951873779297, "loss_alignment": 0.8251953125, "loss_alignment_w": 0.41259765625, "loss_sft": 0.1839378122240305, "loss_total_v6": 0.5961844995617867, "step": 6069 }, { "epoch": 1.3999135135135135, "grad_norm": 30.200294494628906, "learning_rate": 1.4361010911024515e-06, "loss": 9.066429138183594, "loss_alignment": 0.7578125, "loss_alignment_w": 0.37890625, "loss_sft": 0.18649437464773655, "loss_total_v6": 0.5666518285870552, "step": 6070 }, { "epoch": 1.4001441441441442, "grad_norm": 31.192546844482422, "learning_rate": 1.4346888961959392e-06, "loss": 9.578608512878418, "loss_alignment": 0.806640625, "loss_alignment_w": 0.4033203125, "loss_sft": 0.19602933339774609, "loss_total_v6": 0.5986630171537399, "step": 6071 }, { "epoch": 1.4003747747747748, "grad_norm": 29.136032104492188, "learning_rate": 1.4332772796623655e-06, "loss": 8.8494291305542, "loss_alignment": 0.76513671875, "loss_alignment_w": 0.382568359375, "loss_sft": 0.17131441831588745, "loss_total_v6": 0.5530893318355083, "step": 6072 }, { "epoch": 1.4006054054054053, "grad_norm": 32.373172760009766, "learning_rate": 1.4318662417307284e-06, "loss": 9.056461334228516, "loss_alignment": 0.77978515625, "loss_alignment_w": 0.389892578125, "loss_sft": 0.1758157955482602, "loss_total_v6": 0.5660288445651531, "step": 6073 }, { "epoch": 1.400836036036036, "grad_norm": 30.29098892211914, "learning_rate": 1.4304557826299297e-06, "loss": 9.269762992858887, "loss_alignment": 0.7705078125, "loss_alignment_w": 0.38525390625, "loss_sft": 0.19415205158293247, "loss_total_v6": 0.5793601721525192, "step": 6074 }, { "epoch": 1.4010666666666667, "grad_norm": 28.555248260498047, "learning_rate": 1.4290459025887771e-06, "loss": 9.71567440032959, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.21925967186689377, "loss_total_v6": 0.6072296500205994, "step": 6075 }, { "epoch": 1.4012972972972972, "grad_norm": 31.61930274963379, "learning_rate": 1.4276366018359845e-06, "loss": 9.313578605651855, "loss_alignment": 0.78076171875, "loss_alignment_w": 0.390380859375, "loss_sft": 0.19142789766192436, "loss_total_v6": 0.5820986852049828, "step": 6076 }, { "epoch": 1.401527927927928, "grad_norm": 32.24472427368164, "learning_rate": 1.4262278806001696e-06, "loss": 8.758206367492676, "loss_alignment": 0.75, "loss_alignment_w": 0.375, "loss_sft": 0.17258627153933048, "loss_total_v6": 0.5473879054188728, "step": 6077 }, { "epoch": 1.4017585585585586, "grad_norm": 30.6589412689209, "learning_rate": 1.4248197391098617e-06, "loss": 9.211894989013672, "loss_alignment": 0.77392578125, "loss_alignment_w": 0.386962890625, "loss_sft": 0.18926887959241867, "loss_total_v6": 0.5757434815168381, "step": 6078 }, { "epoch": 1.4019891891891891, "grad_norm": 33.69215774536133, "learning_rate": 1.4234121775934895e-06, "loss": 9.208122253417969, "loss_alignment": 0.78759765625, "loss_alignment_w": 0.393798828125, "loss_sft": 0.18135784566402435, "loss_total_v6": 0.575507640838623, "step": 6079 }, { "epoch": 1.4022198198198197, "grad_norm": 36.212467193603516, "learning_rate": 1.4220051962793952e-06, "loss": 9.26637077331543, "loss_alignment": 0.76708984375, "loss_alignment_w": 0.383544921875, "loss_sft": 0.19459620118141174, "loss_total_v6": 0.5791481956839561, "step": 6080 }, { "epoch": 1.4024504504504505, "grad_norm": 32.691322326660156, "learning_rate": 1.4205987953958195e-06, "loss": 8.725851058959961, "loss_alignment": 0.76806640625, "loss_alignment_w": 0.384033203125, "loss_sft": 0.160981516353786, "loss_total_v6": 0.5453656576573849, "step": 6081 }, { "epoch": 1.402681081081081, "grad_norm": 30.3294734954834, "learning_rate": 1.4191929751709104e-06, "loss": 9.648765563964844, "loss_alignment": 0.8134765625, "loss_alignment_w": 0.40673828125, "loss_sft": 0.19707250781357288, "loss_total_v6": 0.6030478700995445, "step": 6082 }, { "epoch": 1.4029117117117118, "grad_norm": 32.245216369628906, "learning_rate": 1.417787735832728e-06, "loss": 9.484176635742188, "loss_alignment": 0.80126953125, "loss_alignment_w": 0.400634765625, "loss_sft": 0.19299601949751377, "loss_total_v6": 0.5927610397338867, "step": 6083 }, { "epoch": 1.4031423423423424, "grad_norm": 30.356948852539062, "learning_rate": 1.4163830776092302e-06, "loss": 9.261494636535645, "loss_alignment": 0.779296875, "loss_alignment_w": 0.3896484375, "loss_sft": 0.19035462103784084, "loss_total_v6": 0.578843392431736, "step": 6084 }, { "epoch": 1.403372972972973, "grad_norm": 28.351661682128906, "learning_rate": 1.4149790007282832e-06, "loss": 9.227165222167969, "loss_alignment": 0.77490234375, "loss_alignment_w": 0.387451171875, "loss_sft": 0.18909407034516335, "loss_total_v6": 0.5766978189349174, "step": 6085 }, { "epoch": 1.4036036036036035, "grad_norm": 32.679534912109375, "learning_rate": 1.4135755054176619e-06, "loss": 8.636276245117188, "loss_alignment": 0.74365234375, "loss_alignment_w": 0.371826171875, "loss_sft": 0.16774266120046377, "loss_total_v6": 0.5397672057151794, "step": 6086 }, { "epoch": 1.4038342342342343, "grad_norm": 32.608360290527344, "learning_rate": 1.4121725919050428e-06, "loss": 8.893465042114258, "loss_alignment": 0.78125, "loss_alignment_w": 0.390625, "loss_sft": 0.16526235733181238, "loss_total_v6": 0.5558415800333023, "step": 6087 }, { "epoch": 1.4040648648648648, "grad_norm": 29.026138305664062, "learning_rate": 1.4107702604180118e-06, "loss": 9.662040710449219, "loss_alignment": 0.76416015625, "loss_alignment_w": 0.382080078125, "loss_sft": 0.22292664274573326, "loss_total_v6": 0.6038775593042374, "step": 6088 }, { "epoch": 1.4042954954954956, "grad_norm": 27.708911895751953, "learning_rate": 1.4093685111840567e-06, "loss": 8.767440795898438, "loss_alignment": 0.7509765625, "loss_alignment_w": 0.37548828125, "loss_sft": 0.1723241452127695, "loss_total_v6": 0.5479650273919106, "step": 6089 }, { "epoch": 1.4045261261261262, "grad_norm": 31.13570213317871, "learning_rate": 1.4079673444305731e-06, "loss": 9.082992553710938, "loss_alignment": 0.75146484375, "loss_alignment_w": 0.375732421875, "loss_sft": 0.19219872914254665, "loss_total_v6": 0.56768698990345, "step": 6090 }, { "epoch": 1.4047567567567567, "grad_norm": 30.822418212890625, "learning_rate": 1.4065667603848593e-06, "loss": 8.690671920776367, "loss_alignment": 0.740234375, "loss_alignment_w": 0.3701171875, "loss_sft": 0.17378219589591026, "loss_total_v6": 0.5431669689714909, "step": 6091 }, { "epoch": 1.4049873873873873, "grad_norm": 31.62602424621582, "learning_rate": 1.405166759274123e-06, "loss": 9.482318878173828, "loss_alignment": 0.80029296875, "loss_alignment_w": 0.400146484375, "loss_sft": 0.19460411183536053, "loss_total_v6": 0.5926448777318001, "step": 6092 }, { "epoch": 1.405218018018018, "grad_norm": 46.70426559448242, "learning_rate": 1.4037673413254777e-06, "loss": 8.829767227172852, "loss_alignment": 0.7353515625, "loss_alignment_w": 0.36767578125, "loss_sft": 0.18366583064198494, "loss_total_v6": 0.551860399544239, "step": 6093 }, { "epoch": 1.4054486486486486, "grad_norm": 34.5512580871582, "learning_rate": 1.4023685067659376e-06, "loss": 9.474538803100586, "loss_alignment": 0.783203125, "loss_alignment_w": 0.3916015625, "loss_sft": 0.20048078522086143, "loss_total_v6": 0.5921586453914642, "step": 6094 }, { "epoch": 1.4056792792792794, "grad_norm": 26.09296226501465, "learning_rate": 1.4009702558224258e-06, "loss": 8.825788497924805, "loss_alignment": 0.74462890625, "loss_alignment_w": 0.372314453125, "loss_sft": 0.1807469241321087, "loss_total_v6": 0.5516117811203003, "step": 6095 }, { "epoch": 1.40590990990991, "grad_norm": 29.34990119934082, "learning_rate": 1.3995725887217693e-06, "loss": 9.195892333984375, "loss_alignment": 0.7666015625, "loss_alignment_w": 0.38330078125, "loss_sft": 0.19190027378499508, "loss_total_v6": 0.5747433006763458, "step": 6096 }, { "epoch": 1.4061405405405405, "grad_norm": 26.666364669799805, "learning_rate": 1.3981755056906992e-06, "loss": 9.277892112731934, "loss_alignment": 0.763671875, "loss_alignment_w": 0.3818359375, "loss_sft": 0.19806279987096786, "loss_total_v6": 0.5798682235181332, "step": 6097 }, { "epoch": 1.406371171171171, "grad_norm": 27.969572067260742, "learning_rate": 1.3967790069558557e-06, "loss": 9.032184600830078, "loss_alignment": 0.767578125, "loss_alignment_w": 0.3837890625, "loss_sft": 0.18176009878516197, "loss_total_v6": 0.5645115450024605, "step": 6098 }, { "epoch": 1.4066018018018018, "grad_norm": 32.84252166748047, "learning_rate": 1.395383092743783e-06, "loss": 9.009279251098633, "loss_alignment": 0.76953125, "loss_alignment_w": 0.384765625, "loss_sft": 0.17724620923399925, "loss_total_v6": 0.563079945743084, "step": 6099 }, { "epoch": 1.4068324324324324, "grad_norm": 29.37404441833496, "learning_rate": 1.3939877632809279e-06, "loss": 9.170547485351562, "loss_alignment": 0.78125, "loss_alignment_w": 0.390625, "loss_sft": 0.1832666276022792, "loss_total_v6": 0.5731592178344727, "step": 6100 }, { "epoch": 1.4070630630630632, "grad_norm": 32.336830139160156, "learning_rate": 1.3925930187936443e-06, "loss": 9.073328971862793, "loss_alignment": 0.744140625, "loss_alignment_w": 0.3720703125, "loss_sft": 0.1946770492941141, "loss_total_v6": 0.5670830644667149, "step": 6101 }, { "epoch": 1.4072936936936937, "grad_norm": 31.414485931396484, "learning_rate": 1.3911988595081894e-06, "loss": 9.13290786743164, "loss_alignment": 0.74169921875, "loss_alignment_w": 0.370849609375, "loss_sft": 0.20100998133420944, "loss_total_v6": 0.5708067342638969, "step": 6102 }, { "epoch": 1.4075243243243243, "grad_norm": 36.439083099365234, "learning_rate": 1.3898052856507293e-06, "loss": 8.773717880249023, "loss_alignment": 0.74951171875, "loss_alignment_w": 0.374755859375, "loss_sft": 0.173952404409647, "loss_total_v6": 0.5483573228120804, "step": 6103 }, { "epoch": 1.4077549549549548, "grad_norm": 29.22920036315918, "learning_rate": 1.3884122974473307e-06, "loss": 9.109687805175781, "loss_alignment": 0.76220703125, "loss_alignment_w": 0.381103515625, "loss_sft": 0.18799255788326263, "loss_total_v6": 0.5693554729223251, "step": 6104 }, { "epoch": 1.4079855855855856, "grad_norm": 30.998149871826172, "learning_rate": 1.3870198951239695e-06, "loss": 9.454715728759766, "loss_alignment": 0.74462890625, "loss_alignment_w": 0.372314453125, "loss_sft": 0.2192766983062029, "loss_total_v6": 0.5909197703003883, "step": 6105 }, { "epoch": 1.4082162162162162, "grad_norm": 31.382125854492188, "learning_rate": 1.385628078906523e-06, "loss": 9.185441970825195, "loss_alignment": 0.7705078125, "loss_alignment_w": 0.38525390625, "loss_sft": 0.18904981762170792, "loss_total_v6": 0.5740901120007038, "step": 6106 }, { "epoch": 1.408446846846847, "grad_norm": 28.866718292236328, "learning_rate": 1.3842368490207737e-06, "loss": 8.91541862487793, "loss_alignment": 0.7861328125, "loss_alignment_w": 0.39306640625, "loss_sft": 0.16361325699836016, "loss_total_v6": 0.5572137124836445, "step": 6107 }, { "epoch": 1.4086774774774775, "grad_norm": 27.5968074798584, "learning_rate": 1.382846205692413e-06, "loss": 8.884528160095215, "loss_alignment": 0.7666015625, "loss_alignment_w": 0.38330078125, "loss_sft": 0.17176859825849533, "loss_total_v6": 0.5552830025553703, "step": 6108 }, { "epoch": 1.408908108108108, "grad_norm": 33.517696380615234, "learning_rate": 1.3814561491470323e-06, "loss": 8.865914344787598, "loss_alignment": 0.734375, "loss_alignment_w": 0.3671875, "loss_sft": 0.187710365280509, "loss_total_v6": 0.5541196763515472, "step": 6109 }, { "epoch": 1.4091387387387386, "grad_norm": 30.35405731201172, "learning_rate": 1.3800666796101291e-06, "loss": 8.863112449645996, "loss_alignment": 0.75830078125, "loss_alignment_w": 0.379150390625, "loss_sft": 0.17453476786613464, "loss_total_v6": 0.5539445467293262, "step": 6110 }, { "epoch": 1.4093693693693694, "grad_norm": 30.79891586303711, "learning_rate": 1.3786777973071086e-06, "loss": 9.532670021057129, "loss_alignment": 0.82421875, "loss_alignment_w": 0.412109375, "loss_sft": 0.18409449979662895, "loss_total_v6": 0.5957919135689735, "step": 6111 }, { "epoch": 1.4096, "grad_norm": 27.38064956665039, "learning_rate": 1.3772895024632753e-06, "loss": 9.114680290222168, "loss_alignment": 0.7568359375, "loss_alignment_w": 0.37841796875, "loss_sft": 0.1912953071296215, "loss_total_v6": 0.569667499512434, "step": 6112 }, { "epoch": 1.4098306306306307, "grad_norm": 30.114721298217773, "learning_rate": 1.3759017953038456e-06, "loss": 9.062426567077637, "loss_alignment": 0.7939453125, "loss_alignment_w": 0.39697265625, "loss_sft": 0.16894069779664278, "loss_total_v6": 0.5664016380906105, "step": 6113 }, { "epoch": 1.4100612612612613, "grad_norm": 31.56633949279785, "learning_rate": 1.3745146760539346e-06, "loss": 8.719343185424805, "loss_alignment": 0.75048828125, "loss_alignment_w": 0.375244140625, "loss_sft": 0.16992838494479656, "loss_total_v6": 0.5449589118361473, "step": 6114 }, { "epoch": 1.4102918918918919, "grad_norm": 33.241485595703125, "learning_rate": 1.373128144938563e-06, "loss": 8.58505630493164, "loss_alignment": 0.74560546875, "loss_alignment_w": 0.372802734375, "loss_sft": 0.16477037407457829, "loss_total_v6": 0.5365660190582275, "step": 6115 }, { "epoch": 1.4105225225225224, "grad_norm": 26.814743041992188, "learning_rate": 1.371742202182657e-06, "loss": 8.703010559082031, "loss_alignment": 0.7021484375, "loss_alignment_w": 0.35107421875, "loss_sft": 0.19356583803892136, "loss_total_v6": 0.5439381673932076, "step": 6116 }, { "epoch": 1.4107531531531532, "grad_norm": 33.104469299316406, "learning_rate": 1.370356848011048e-06, "loss": 8.975475311279297, "loss_alignment": 0.74755859375, "loss_alignment_w": 0.373779296875, "loss_sft": 0.18816445022821426, "loss_total_v6": 0.5609671846032143, "step": 6117 }, { "epoch": 1.4109837837837838, "grad_norm": 26.55149269104004, "learning_rate": 1.368972082648473e-06, "loss": 8.807714462280273, "loss_alignment": 0.77392578125, "loss_alignment_w": 0.386962890625, "loss_sft": 0.1633208654820919, "loss_total_v6": 0.5504821315407753, "step": 6118 }, { "epoch": 1.4112144144144145, "grad_norm": 29.836435317993164, "learning_rate": 1.3675879063195707e-06, "loss": 8.641022682189941, "loss_alignment": 0.7275390625, "loss_alignment_w": 0.36376953125, "loss_sft": 0.17609596252441406, "loss_total_v6": 0.540063887834549, "step": 6119 }, { "epoch": 1.411445045045045, "grad_norm": 30.442747116088867, "learning_rate": 1.366204319248885e-06, "loss": 9.605011940002441, "loss_alignment": 0.791015625, "loss_alignment_w": 0.3955078125, "loss_sft": 0.205171637237072, "loss_total_v6": 0.6003132462501526, "step": 6120 }, { "epoch": 1.4116756756756756, "grad_norm": 35.4991569519043, "learning_rate": 1.3648213216608646e-06, "loss": 9.120882034301758, "loss_alignment": 0.75439453125, "loss_alignment_w": 0.377197265625, "loss_sft": 0.19137777760624886, "loss_total_v6": 0.5700551196932793, "step": 6121 }, { "epoch": 1.4119063063063062, "grad_norm": 31.634498596191406, "learning_rate": 1.3634389137798609e-06, "loss": 9.036981582641602, "loss_alignment": 0.78173828125, "loss_alignment_w": 0.390869140625, "loss_sft": 0.17433892376720905, "loss_total_v6": 0.5648113265633583, "step": 6122 }, { "epoch": 1.412136936936937, "grad_norm": 33.5152702331543, "learning_rate": 1.3620570958301333e-06, "loss": 8.859838485717773, "loss_alignment": 0.78515625, "loss_alignment_w": 0.392578125, "loss_sft": 0.16154319513589144, "loss_total_v6": 0.553739856928587, "step": 6123 }, { "epoch": 1.4123675675675675, "grad_norm": 32.98740005493164, "learning_rate": 1.3606758680358446e-06, "loss": 8.687929153442383, "loss_alignment": 0.73095703125, "loss_alignment_w": 0.365478515625, "loss_sft": 0.17768491990864277, "loss_total_v6": 0.542995598167181, "step": 6124 }, { "epoch": 1.4125981981981983, "grad_norm": 30.59202766418457, "learning_rate": 1.3592952306210589e-06, "loss": 9.870912551879883, "loss_alignment": 0.78369140625, "loss_alignment_w": 0.391845703125, "loss_sft": 0.22462855465710163, "loss_total_v6": 0.6169320195913315, "step": 6125 }, { "epoch": 1.4128288288288289, "grad_norm": 28.38738250732422, "learning_rate": 1.3579151838097454e-06, "loss": 8.71632194519043, "loss_alignment": 0.71875, "loss_alignment_w": 0.359375, "loss_sft": 0.18458639457821846, "loss_total_v6": 0.5447701215744019, "step": 6126 }, { "epoch": 1.4130594594594594, "grad_norm": 30.582197189331055, "learning_rate": 1.3565357278257813e-06, "loss": 9.454190254211426, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.20677738264203072, "loss_total_v6": 0.5908868834376335, "step": 6127 }, { "epoch": 1.41329009009009, "grad_norm": 35.600791931152344, "learning_rate": 1.3551568628929434e-06, "loss": 9.581356048583984, "loss_alignment": 0.75634765625, "loss_alignment_w": 0.378173828125, "loss_sft": 0.22145435772836208, "loss_total_v6": 0.598834715783596, "step": 6128 }, { "epoch": 1.4135207207207208, "grad_norm": 30.02948760986328, "learning_rate": 1.3537785892349132e-06, "loss": 9.071765899658203, "loss_alignment": 0.7724609375, "loss_alignment_w": 0.38623046875, "loss_sft": 0.18127365224063396, "loss_total_v6": 0.5669853463768959, "step": 6129 }, { "epoch": 1.4137513513513513, "grad_norm": 29.828189849853516, "learning_rate": 1.3524009070752802e-06, "loss": 8.734413146972656, "loss_alignment": 0.74658203125, "loss_alignment_w": 0.373291015625, "loss_sft": 0.17294543702155352, "loss_total_v6": 0.5459007769823074, "step": 6130 }, { "epoch": 1.413981981981982, "grad_norm": 35.051395416259766, "learning_rate": 1.3510238166375323e-06, "loss": 9.259599685668945, "loss_alignment": 0.72998046875, "loss_alignment_w": 0.364990234375, "loss_sft": 0.2125904057174921, "loss_total_v6": 0.5787250250577927, "step": 6131 }, { "epoch": 1.4142126126126127, "grad_norm": 32.63114929199219, "learning_rate": 1.349647318145067e-06, "loss": 9.220853805541992, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.18900473415851593, "loss_total_v6": 0.5763033255934715, "step": 6132 }, { "epoch": 1.4144432432432432, "grad_norm": 33.68939971923828, "learning_rate": 1.3482714118211815e-06, "loss": 8.883806228637695, "loss_alignment": 0.7646484375, "loss_alignment_w": 0.38232421875, "loss_sft": 0.1723490757867694, "loss_total_v6": 0.5552378818392754, "step": 6133 }, { "epoch": 1.4146738738738738, "grad_norm": 31.52144432067871, "learning_rate": 1.3468960978890793e-06, "loss": 9.616323471069336, "loss_alignment": 0.78466796875, "loss_alignment_w": 0.392333984375, "loss_sft": 0.20839632581919432, "loss_total_v6": 0.6010202169418335, "step": 6134 }, { "epoch": 1.4149045045045046, "grad_norm": 31.12336540222168, "learning_rate": 1.3455213765718645e-06, "loss": 8.90892219543457, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.1699667926877737, "loss_total_v6": 0.556807603687048, "step": 6135 }, { "epoch": 1.4151351351351351, "grad_norm": 33.36836242675781, "learning_rate": 1.3441472480925494e-06, "loss": 8.628560066223145, "loss_alignment": 0.7431640625, "loss_alignment_w": 0.37158203125, "loss_sft": 0.1689999420195818, "loss_total_v6": 0.5392849966883659, "step": 6136 }, { "epoch": 1.415365765765766, "grad_norm": 33.503631591796875, "learning_rate": 1.3427737126740498e-06, "loss": 9.364768981933594, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.2012190818786621, "loss_total_v6": 0.5852980613708496, "step": 6137 }, { "epoch": 1.4155963963963965, "grad_norm": 27.658926010131836, "learning_rate": 1.3414007705391818e-06, "loss": 8.930213928222656, "loss_alignment": 0.72705078125, "loss_alignment_w": 0.363525390625, "loss_sft": 0.19513171166181564, "loss_total_v6": 0.5581383258104324, "step": 6138 }, { "epoch": 1.415827027027027, "grad_norm": 28.92208480834961, "learning_rate": 1.3400284219106686e-06, "loss": 9.487297058105469, "loss_alignment": 0.77099609375, "loss_alignment_w": 0.385498046875, "loss_sft": 0.2082209587097168, "loss_total_v6": 0.5929560661315918, "step": 6139 }, { "epoch": 1.4160576576576576, "grad_norm": 26.964092254638672, "learning_rate": 1.3386566670111339e-06, "loss": 8.630109786987305, "loss_alignment": 0.73046875, "loss_alignment_w": 0.365234375, "loss_sft": 0.1729419995099306, "loss_total_v6": 0.5393818356096745, "step": 6140 }, { "epoch": 1.4162882882882883, "grad_norm": 27.92633628845215, "learning_rate": 1.3372855060631067e-06, "loss": 8.643522262573242, "loss_alignment": 0.73291015625, "loss_alignment_w": 0.366455078125, "loss_sft": 0.17382607981562614, "loss_total_v6": 0.5402201376855373, "step": 6141 }, { "epoch": 1.416518918918919, "grad_norm": 35.74359893798828, "learning_rate": 1.3359149392890215e-06, "loss": 9.151230812072754, "loss_alignment": 0.78369140625, "loss_alignment_w": 0.391845703125, "loss_sft": 0.1806555539369583, "loss_total_v6": 0.5719519406557083, "step": 6142 }, { "epoch": 1.4167495495495497, "grad_norm": 29.571714401245117, "learning_rate": 1.334544966911215e-06, "loss": 9.113702774047852, "loss_alignment": 0.76904296875, "loss_alignment_w": 0.384521484375, "loss_sft": 0.18502395413815975, "loss_total_v6": 0.5696064606308937, "step": 6143 }, { "epoch": 1.4169801801801802, "grad_norm": 29.375003814697266, "learning_rate": 1.3331755891519267e-06, "loss": 9.383333206176758, "loss_alignment": 0.78076171875, "loss_alignment_w": 0.390380859375, "loss_sft": 0.19700824841856956, "loss_total_v6": 0.586458332836628, "step": 6144 }, { "epoch": 1.4172108108108108, "grad_norm": 29.039880752563477, "learning_rate": 1.3318068062333e-06, "loss": 9.019999504089355, "loss_alignment": 0.7607421875, "loss_alignment_w": 0.38037109375, "loss_sft": 0.1817766847088933, "loss_total_v6": 0.5637499578297138, "step": 6145 }, { "epoch": 1.4174414414414414, "grad_norm": 31.163387298583984, "learning_rate": 1.330438618377381e-06, "loss": 8.95388126373291, "loss_alignment": 0.76904296875, "loss_alignment_w": 0.384521484375, "loss_sft": 0.175263955257833, "loss_total_v6": 0.5596175864338875, "step": 6146 }, { "epoch": 1.4176720720720721, "grad_norm": 32.171722412109375, "learning_rate": 1.3290710258061222e-06, "loss": 9.487037658691406, "loss_alignment": 0.779296875, "loss_alignment_w": 0.3896484375, "loss_sft": 0.2027268186211586, "loss_total_v6": 0.5929398164153099, "step": 6147 }, { "epoch": 1.4179027027027027, "grad_norm": 28.610294342041016, "learning_rate": 1.3277040287413756e-06, "loss": 8.857418060302734, "loss_alignment": 0.7470703125, "loss_alignment_w": 0.37353515625, "loss_sft": 0.18084688857197762, "loss_total_v6": 0.5535885840654373, "step": 6148 }, { "epoch": 1.4181333333333335, "grad_norm": 25.45015525817871, "learning_rate": 1.3263376274049011e-06, "loss": 8.532903671264648, "loss_alignment": 0.7255859375, "loss_alignment_w": 0.36279296875, "loss_sft": 0.17100177332758904, "loss_total_v6": 0.533306460827589, "step": 6149 }, { "epoch": 1.418363963963964, "grad_norm": 30.059600830078125, "learning_rate": 1.3249718220183583e-06, "loss": 9.03917407989502, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.18131189234554768, "loss_total_v6": 0.5649483911693096, "step": 6150 }, { "epoch": 1.4185945945945946, "grad_norm": 34.94578170776367, "learning_rate": 1.3236066128033093e-06, "loss": 8.674837112426758, "loss_alignment": 0.765625, "loss_alignment_w": 0.3828125, "loss_sft": 0.15771686658263206, "loss_total_v6": 0.5421773344278336, "step": 6151 }, { "epoch": 1.4188252252252251, "grad_norm": 34.03462600708008, "learning_rate": 1.3222419999812248e-06, "loss": 9.01728343963623, "loss_alignment": 0.75, "loss_alignment_w": 0.375, "loss_sft": 0.18867179192602634, "loss_total_v6": 0.5635802447795868, "step": 6152 }, { "epoch": 1.419055855855856, "grad_norm": 33.05644226074219, "learning_rate": 1.3208779837734736e-06, "loss": 9.165182113647461, "loss_alignment": 0.7958984375, "loss_alignment_w": 0.39794921875, "loss_sft": 0.1766141951084137, "loss_total_v6": 0.5728238970041275, "step": 6153 }, { "epoch": 1.4192864864864865, "grad_norm": 29.561784744262695, "learning_rate": 1.3195145644013286e-06, "loss": 8.710806846618652, "loss_alignment": 0.71484375, "loss_alignment_w": 0.357421875, "loss_sft": 0.18729343824088573, "loss_total_v6": 0.5444253981113434, "step": 6154 }, { "epoch": 1.4195171171171173, "grad_norm": 30.890060424804688, "learning_rate": 1.3181517420859696e-06, "loss": 9.03477668762207, "loss_alignment": 0.7646484375, "loss_alignment_w": 0.38232421875, "loss_sft": 0.18317325972020626, "loss_total_v6": 0.5646735280752182, "step": 6155 }, { "epoch": 1.4197477477477478, "grad_norm": 27.873369216918945, "learning_rate": 1.316789517048473e-06, "loss": 9.012413024902344, "loss_alignment": 0.73388671875, "loss_alignment_w": 0.366943359375, "loss_sft": 0.19665289111435413, "loss_total_v6": 0.5632758252322674, "step": 6156 }, { "epoch": 1.4199783783783784, "grad_norm": 30.753662109375, "learning_rate": 1.3154278895098266e-06, "loss": 9.142276763916016, "loss_alignment": 0.775390625, "loss_alignment_w": 0.3876953125, "loss_sft": 0.18462778721004725, "loss_total_v6": 0.5713923163712025, "step": 6157 }, { "epoch": 1.420209009009009, "grad_norm": 30.996990203857422, "learning_rate": 1.3140668596909133e-06, "loss": 9.062177658081055, "loss_alignment": 0.755859375, "loss_alignment_w": 0.3779296875, "loss_sft": 0.18839536048471928, "loss_total_v6": 0.5663860738277435, "step": 6158 }, { "epoch": 1.4204396396396397, "grad_norm": 31.067731857299805, "learning_rate": 1.3127064278125245e-06, "loss": 8.998881340026855, "loss_alignment": 0.75244140625, "loss_alignment_w": 0.376220703125, "loss_sft": 0.18671290203928947, "loss_total_v6": 0.5624300688505173, "step": 6159 }, { "epoch": 1.4206702702702703, "grad_norm": 36.4854736328125, "learning_rate": 1.3113465940953495e-06, "loss": 9.003429412841797, "loss_alignment": 0.7626953125, "loss_alignment_w": 0.38134765625, "loss_sft": 0.18211443163454533, "loss_total_v6": 0.5627143792808056, "step": 6160 }, { "epoch": 1.4209009009009008, "grad_norm": 29.80785369873047, "learning_rate": 1.309987358759986e-06, "loss": 8.879704475402832, "loss_alignment": 0.77001953125, "loss_alignment_w": 0.385009765625, "loss_sft": 0.17078044638037682, "loss_total_v6": 0.5549814961850643, "step": 6161 }, { "epoch": 1.4211315315315316, "grad_norm": 32.700157165527344, "learning_rate": 1.308628722026934e-06, "loss": 9.14309310913086, "loss_alignment": 0.75048828125, "loss_alignment_w": 0.375244140625, "loss_sft": 0.19622972421348095, "loss_total_v6": 0.5714433640241623, "step": 6162 }, { "epoch": 1.4213621621621622, "grad_norm": 31.301998138427734, "learning_rate": 1.307270684116592e-06, "loss": 8.905041694641113, "loss_alignment": 0.73193359375, "loss_alignment_w": 0.365966796875, "loss_sft": 0.18986587971448898, "loss_total_v6": 0.5565651096403599, "step": 6163 }, { "epoch": 1.4215927927927927, "grad_norm": 29.963754653930664, "learning_rate": 1.305913245249265e-06, "loss": 9.23783016204834, "loss_alignment": 0.7587890625, "loss_alignment_w": 0.37939453125, "loss_sft": 0.19880910404026508, "loss_total_v6": 0.5773643925786018, "step": 6164 }, { "epoch": 1.4218234234234235, "grad_norm": 28.751523971557617, "learning_rate": 1.3045564056451576e-06, "loss": 8.920906066894531, "loss_alignment": 0.80029296875, "loss_alignment_w": 0.400146484375, "loss_sft": 0.15679978299885988, "loss_total_v6": 0.5575566031038761, "step": 6165 }, { "epoch": 1.422054054054054, "grad_norm": 32.83769226074219, "learning_rate": 1.3032001655243836e-06, "loss": 9.409472465515137, "loss_alignment": 0.7666015625, "loss_alignment_w": 0.38330078125, "loss_sft": 0.20416565611958504, "loss_total_v6": 0.588092029094696, "step": 6166 }, { "epoch": 1.4222846846846846, "grad_norm": 28.158241271972656, "learning_rate": 1.301844525106951e-06, "loss": 8.464855194091797, "loss_alignment": 0.72412109375, "loss_alignment_w": 0.362060546875, "loss_sft": 0.16746588051319122, "loss_total_v6": 0.5290533937513828, "step": 6167 }, { "epoch": 1.4225153153153154, "grad_norm": 32.795494079589844, "learning_rate": 1.300489484612779e-06, "loss": 8.635965347290039, "loss_alignment": 0.755859375, "loss_alignment_w": 0.3779296875, "loss_sft": 0.16160458698868752, "loss_total_v6": 0.5397478714585304, "step": 6168 }, { "epoch": 1.422745945945946, "grad_norm": 35.1377067565918, "learning_rate": 1.2991350442616835e-06, "loss": 9.285613059997559, "loss_alignment": 0.75830078125, "loss_alignment_w": 0.379150390625, "loss_sft": 0.20179550722241402, "loss_total_v6": 0.5803507938981056, "step": 6169 }, { "epoch": 1.4229765765765765, "grad_norm": 30.153244018554688, "learning_rate": 1.297781204273385e-06, "loss": 9.224559783935547, "loss_alignment": 0.78759765625, "loss_alignment_w": 0.393798828125, "loss_sft": 0.18261411599814892, "loss_total_v6": 0.5765350088477135, "step": 6170 }, { "epoch": 1.4232072072072073, "grad_norm": 34.559043884277344, "learning_rate": 1.2964279648675042e-06, "loss": 8.837445259094238, "loss_alignment": 0.73779296875, "loss_alignment_w": 0.368896484375, "loss_sft": 0.18303181044757366, "loss_total_v6": 0.5523403026163578, "step": 6171 }, { "epoch": 1.4234378378378378, "grad_norm": 34.262725830078125, "learning_rate": 1.2950753262635712e-06, "loss": 8.688922882080078, "loss_alignment": 0.73974609375, "loss_alignment_w": 0.369873046875, "loss_sft": 0.17292527854442596, "loss_total_v6": 0.5430577211081982, "step": 6172 }, { "epoch": 1.4236684684684684, "grad_norm": 33.80079650878906, "learning_rate": 1.2937232886810098e-06, "loss": 8.934235572814941, "loss_alignment": 0.72314453125, "loss_alignment_w": 0.361572265625, "loss_sft": 0.1981449741870165, "loss_total_v6": 0.5583897158503532, "step": 6173 }, { "epoch": 1.4238990990990992, "grad_norm": 31.40361213684082, "learning_rate": 1.2923718523391544e-06, "loss": 9.433198928833008, "loss_alignment": 0.77490234375, "loss_alignment_w": 0.387451171875, "loss_sft": 0.20194068923592567, "loss_total_v6": 0.5895749777555466, "step": 6174 }, { "epoch": 1.4241297297297297, "grad_norm": 28.232267379760742, "learning_rate": 1.2910210174572346e-06, "loss": 8.776494979858398, "loss_alignment": 0.75341796875, "loss_alignment_w": 0.376708984375, "loss_sft": 0.17130319122225046, "loss_total_v6": 0.5485309772193432, "step": 6175 }, { "epoch": 1.4243603603603603, "grad_norm": 30.878751754760742, "learning_rate": 1.2896707842543898e-06, "loss": 8.860054016113281, "loss_alignment": 0.7509765625, "loss_alignment_w": 0.37548828125, "loss_sft": 0.17806668020784855, "loss_total_v6": 0.5537533201277256, "step": 6176 }, { "epoch": 1.424590990990991, "grad_norm": 30.690046310424805, "learning_rate": 1.2883211529496553e-06, "loss": 8.925137519836426, "loss_alignment": 0.728515625, "loss_alignment_w": 0.3642578125, "loss_sft": 0.19310554210096598, "loss_total_v6": 0.5578210987150669, "step": 6177 }, { "epoch": 1.4248216216216216, "grad_norm": 32.4571418762207, "learning_rate": 1.2869721237619704e-06, "loss": 9.146203994750977, "loss_alignment": 0.76416015625, "loss_alignment_w": 0.382080078125, "loss_sft": 0.19042744860053062, "loss_total_v6": 0.5716377682983875, "step": 6178 }, { "epoch": 1.4250522522522522, "grad_norm": 28.240875244140625, "learning_rate": 1.285623696910181e-06, "loss": 8.904949188232422, "loss_alignment": 0.7490234375, "loss_alignment_w": 0.37451171875, "loss_sft": 0.18204758875072002, "loss_total_v6": 0.5565593019127846, "step": 6179 }, { "epoch": 1.425282882882883, "grad_norm": 29.156824111938477, "learning_rate": 1.2842758726130283e-06, "loss": 8.967144012451172, "loss_alignment": 0.73828125, "loss_alignment_w": 0.369140625, "loss_sft": 0.1909548733383417, "loss_total_v6": 0.5604464411735535, "step": 6180 }, { "epoch": 1.4255135135135135, "grad_norm": 30.718751907348633, "learning_rate": 1.282928651089163e-06, "loss": 9.064618110656738, "loss_alignment": 0.7412109375, "loss_alignment_w": 0.37060546875, "loss_sft": 0.19634516723453999, "loss_total_v6": 0.5665386356413364, "step": 6181 }, { "epoch": 1.425744144144144, "grad_norm": 39.8773307800293, "learning_rate": 1.2815820325571327e-06, "loss": 9.250572204589844, "loss_alignment": 0.7490234375, "loss_alignment_w": 0.37451171875, "loss_sft": 0.2031301986426115, "loss_total_v6": 0.5781607329845428, "step": 6182 }, { "epoch": 1.4259747747747749, "grad_norm": 30.134113311767578, "learning_rate": 1.2802360172353895e-06, "loss": 9.368257522583008, "loss_alignment": 0.76123046875, "loss_alignment_w": 0.380615234375, "loss_sft": 0.2047177804633975, "loss_total_v6": 0.5855161175131798, "step": 6183 }, { "epoch": 1.4262054054054054, "grad_norm": 31.60428237915039, "learning_rate": 1.278890605342285e-06, "loss": 9.390594482421875, "loss_alignment": 0.8076171875, "loss_alignment_w": 0.40380859375, "loss_sft": 0.18362238444387913, "loss_total_v6": 0.5869121849536896, "step": 6184 }, { "epoch": 1.426436036036036, "grad_norm": 29.568355560302734, "learning_rate": 1.2775457970960765e-06, "loss": 9.272668838500977, "loss_alignment": 0.771484375, "loss_alignment_w": 0.3857421875, "loss_sft": 0.1945930365473032, "loss_total_v6": 0.5795417726039886, "step": 6185 }, { "epoch": 1.4266666666666667, "grad_norm": 32.239463806152344, "learning_rate": 1.2762015927149236e-06, "loss": 8.812994956970215, "loss_alignment": 0.7431640625, "loss_alignment_w": 0.37158203125, "loss_sft": 0.17898599617183208, "loss_total_v6": 0.5508121810853481, "step": 6186 }, { "epoch": 1.4268972972972973, "grad_norm": 37.84839630126953, "learning_rate": 1.2748579924168841e-06, "loss": 8.875646591186523, "loss_alignment": 0.7890625, "loss_alignment_w": 0.39453125, "loss_sft": 0.16083753108978271, "loss_total_v6": 0.5547279305756092, "step": 6187 }, { "epoch": 1.4271279279279279, "grad_norm": 31.209209442138672, "learning_rate": 1.2735149964199212e-06, "loss": 9.434467315673828, "loss_alignment": 0.7890625, "loss_alignment_w": 0.39453125, "loss_sft": 0.19553494453430176, "loss_total_v6": 0.5896542221307755, "step": 6188 }, { "epoch": 1.4273585585585584, "grad_norm": 34.508583068847656, "learning_rate": 1.2721726049418981e-06, "loss": 9.08961296081543, "loss_alignment": 0.75341796875, "loss_alignment_w": 0.376708984375, "loss_sft": 0.191803814843297, "loss_total_v6": 0.5681008100509644, "step": 6189 }, { "epoch": 1.4275891891891892, "grad_norm": 26.481664657592773, "learning_rate": 1.2708308182005784e-06, "loss": 8.658374786376953, "loss_alignment": 0.75, "loss_alignment_w": 0.375, "loss_sft": 0.16540072858333588, "loss_total_v6": 0.5411483906209469, "step": 6190 }, { "epoch": 1.4278198198198198, "grad_norm": 29.422605514526367, "learning_rate": 1.2694896364136328e-06, "loss": 9.43799877166748, "loss_alignment": 0.7705078125, "loss_alignment_w": 0.38525390625, "loss_sft": 0.20483462139964104, "loss_total_v6": 0.5898749232292175, "step": 6191 }, { "epoch": 1.4280504504504505, "grad_norm": 32.353519439697266, "learning_rate": 1.2681490597986313e-06, "loss": 9.384830474853516, "loss_alignment": 0.78564453125, "loss_alignment_w": 0.392822265625, "loss_sft": 0.19295144639909267, "loss_total_v6": 0.5865519270300865, "step": 6192 }, { "epoch": 1.428281081081081, "grad_norm": 32.110694885253906, "learning_rate": 1.2668090885730439e-06, "loss": 8.540899276733398, "loss_alignment": 0.73291015625, "loss_alignment_w": 0.366455078125, "loss_sft": 0.1672900952398777, "loss_total_v6": 0.533806212246418, "step": 6193 }, { "epoch": 1.4285117117117117, "grad_norm": 31.38689422607422, "learning_rate": 1.2654697229542439e-06, "loss": 8.538125038146973, "loss_alignment": 0.720703125, "loss_alignment_w": 0.3603515625, "loss_sft": 0.17351015843451023, "loss_total_v6": 0.5336328335106373, "step": 6194 }, { "epoch": 1.4287423423423422, "grad_norm": 35.69779586791992, "learning_rate": 1.264130963159505e-06, "loss": 8.952571868896484, "loss_alignment": 0.78466796875, "loss_alignment_w": 0.392333984375, "loss_sft": 0.1664693122729659, "loss_total_v6": 0.5595357082784176, "step": 6195 }, { "epoch": 1.428972972972973, "grad_norm": 35.41346740722656, "learning_rate": 1.2627928094060065e-06, "loss": 9.247257232666016, "loss_alignment": 0.759765625, "loss_alignment_w": 0.3798828125, "loss_sft": 0.19840647093951702, "loss_total_v6": 0.5779535993933678, "step": 6196 }, { "epoch": 1.4292036036036035, "grad_norm": 34.36933898925781, "learning_rate": 1.2614552619108244e-06, "loss": 9.09854507446289, "loss_alignment": 0.7626953125, "loss_alignment_w": 0.38134765625, "loss_sft": 0.18763182684779167, "loss_total_v6": 0.5686590448021889, "step": 6197 }, { "epoch": 1.4294342342342343, "grad_norm": 31.702211380004883, "learning_rate": 1.2601183208909412e-06, "loss": 9.12845230102539, "loss_alignment": 0.73779296875, "loss_alignment_w": 0.368896484375, "loss_sft": 0.20201323740184307, "loss_total_v6": 0.5705282688140869, "step": 6198 }, { "epoch": 1.4296648648648649, "grad_norm": 33.63267135620117, "learning_rate": 1.2587819865632373e-06, "loss": 8.805154800415039, "loss_alignment": 0.759765625, "loss_alignment_w": 0.3798828125, "loss_sft": 0.17072932329028845, "loss_total_v6": 0.5503222122788429, "step": 6199 }, { "epoch": 1.4298954954954954, "grad_norm": 37.05552291870117, "learning_rate": 1.257446259144494e-06, "loss": 9.224456787109375, "loss_alignment": 0.7724609375, "loss_alignment_w": 0.38623046875, "loss_sft": 0.1899013500660658, "loss_total_v6": 0.5765285491943359, "step": 6200 }, { "epoch": 1.430126126126126, "grad_norm": 33.43770217895508, "learning_rate": 1.2561111388513997e-06, "loss": 9.789414405822754, "loss_alignment": 0.7578125, "loss_alignment_w": 0.37890625, "loss_sft": 0.23325257562100887, "loss_total_v6": 0.6118383854627609, "step": 6201 }, { "epoch": 1.4303567567567568, "grad_norm": 30.812856674194336, "learning_rate": 1.254776625900539e-06, "loss": 8.735140800476074, "loss_alignment": 0.7275390625, "loss_alignment_w": 0.36376953125, "loss_sft": 0.1823598351329565, "loss_total_v6": 0.5459462702274323, "step": 6202 }, { "epoch": 1.4305873873873873, "grad_norm": 25.00602149963379, "learning_rate": 1.2534427205083978e-06, "loss": 8.670907020568848, "loss_alignment": 0.73876953125, "loss_alignment_w": 0.369384765625, "loss_sft": 0.17245537415146828, "loss_total_v6": 0.5419316925108433, "step": 6203 }, { "epoch": 1.4308180180180181, "grad_norm": 30.993896484375, "learning_rate": 1.2521094228913683e-06, "loss": 8.991447448730469, "loss_alignment": 0.7353515625, "loss_alignment_w": 0.36767578125, "loss_sft": 0.1947626918554306, "loss_total_v6": 0.5619654580950737, "step": 6204 }, { "epoch": 1.4310486486486487, "grad_norm": 30.548131942749023, "learning_rate": 1.2507767332657379e-06, "loss": 9.438592910766602, "loss_alignment": 0.78857421875, "loss_alignment_w": 0.394287109375, "loss_sft": 0.1957927718758583, "loss_total_v6": 0.5899120345711708, "step": 6205 }, { "epoch": 1.4312792792792792, "grad_norm": 32.709476470947266, "learning_rate": 1.2494446518477022e-06, "loss": 8.937654495239258, "loss_alignment": 0.7783203125, "loss_alignment_w": 0.38916015625, "loss_sft": 0.16973319370299578, "loss_total_v6": 0.5586034283041954, "step": 6206 }, { "epoch": 1.4315099099099098, "grad_norm": 28.508148193359375, "learning_rate": 1.248113178853353e-06, "loss": 8.620562553405762, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.15443154610693455, "loss_total_v6": 0.5387851595878601, "step": 6207 }, { "epoch": 1.4317405405405406, "grad_norm": 33.339202880859375, "learning_rate": 1.2467823144986846e-06, "loss": 9.55276870727539, "loss_alignment": 0.7548828125, "loss_alignment_w": 0.37744140625, "loss_sft": 0.21942356787621975, "loss_total_v6": 0.5970480814576149, "step": 6208 }, { "epoch": 1.4319711711711711, "grad_norm": 33.280635833740234, "learning_rate": 1.2454520589995911e-06, "loss": 9.00080680847168, "loss_alignment": 0.744140625, "loss_alignment_w": 0.3720703125, "loss_sft": 0.18996132910251617, "loss_total_v6": 0.5625504404306412, "step": 6209 }, { "epoch": 1.432201801801802, "grad_norm": 30.286962509155273, "learning_rate": 1.2441224125718714e-06, "loss": 9.114596366882324, "loss_alignment": 0.77099609375, "loss_alignment_w": 0.385498046875, "loss_sft": 0.18428626749664545, "loss_total_v6": 0.5696622505784035, "step": 6210 }, { "epoch": 1.4324324324324325, "grad_norm": 29.431289672851562, "learning_rate": 1.2427933754312255e-06, "loss": 9.124016761779785, "loss_alignment": 0.7646484375, "loss_alignment_w": 0.38232421875, "loss_sft": 0.18882711417973042, "loss_total_v6": 0.5702510625123978, "step": 6211 }, { "epoch": 1.432663063063063, "grad_norm": 33.80103302001953, "learning_rate": 1.2414649477932511e-06, "loss": 8.600017547607422, "loss_alignment": 0.71728515625, "loss_alignment_w": 0.358642578125, "loss_sft": 0.17859912477433681, "loss_total_v6": 0.5375010892748833, "step": 6212 }, { "epoch": 1.4328936936936936, "grad_norm": 37.50033187866211, "learning_rate": 1.2401371298734488e-06, "loss": 9.395054817199707, "loss_alignment": 0.7705078125, "loss_alignment_w": 0.38525390625, "loss_sft": 0.2012808844447136, "loss_total_v6": 0.5871908962726593, "step": 6213 }, { "epoch": 1.4331243243243244, "grad_norm": 33.913963317871094, "learning_rate": 1.2388099218872202e-06, "loss": 9.136907577514648, "loss_alignment": 0.7783203125, "loss_alignment_w": 0.38916015625, "loss_sft": 0.18262901902198792, "loss_total_v6": 0.5710567533969879, "step": 6214 }, { "epoch": 1.433354954954955, "grad_norm": 30.077905654907227, "learning_rate": 1.2374833240498668e-06, "loss": 8.876016616821289, "loss_alignment": 0.79150390625, "loss_alignment_w": 0.395751953125, "loss_sft": 0.15861757658421993, "loss_total_v6": 0.5547510012984276, "step": 6215 }, { "epoch": 1.4335855855855857, "grad_norm": 30.01618003845215, "learning_rate": 1.236157336576594e-06, "loss": 9.302438735961914, "loss_alignment": 0.74560546875, "loss_alignment_w": 0.372802734375, "loss_sft": 0.20911851897835732, "loss_total_v6": 0.5814024582505226, "step": 6216 }, { "epoch": 1.4338162162162162, "grad_norm": 29.826093673706055, "learning_rate": 1.234831959682507e-06, "loss": 9.07815170288086, "loss_alignment": 0.7890625, "loss_alignment_w": 0.39453125, "loss_sft": 0.1731736697256565, "loss_total_v6": 0.5673845037817955, "step": 6217 }, { "epoch": 1.4340468468468468, "grad_norm": 37.407470703125, "learning_rate": 1.233507193582611e-06, "loss": 8.342090606689453, "loss_alignment": 0.712890625, "loss_alignment_w": 0.3564453125, "loss_sft": 0.16517953108996153, "loss_total_v6": 0.5213807001709938, "step": 6218 }, { "epoch": 1.4342774774774774, "grad_norm": 31.82499885559082, "learning_rate": 1.2321830384918116e-06, "loss": 8.833819389343262, "loss_alignment": 0.7529296875, "loss_alignment_w": 0.37646484375, "loss_sft": 0.17670173570513725, "loss_total_v6": 0.5521137118339539, "step": 6219 }, { "epoch": 1.4345081081081081, "grad_norm": 32.886940002441406, "learning_rate": 1.2308594946249163e-06, "loss": 8.905258178710938, "loss_alignment": 0.75, "loss_alignment_w": 0.375, "loss_sft": 0.18144132383167744, "loss_total_v6": 0.556578665971756, "step": 6220 }, { "epoch": 1.4347387387387387, "grad_norm": 32.34313201904297, "learning_rate": 1.2295365621966348e-06, "loss": 9.162729263305664, "loss_alignment": 0.7734375, "loss_alignment_w": 0.38671875, "loss_sft": 0.18511254340410233, "loss_total_v6": 0.5726705268025398, "step": 6221 }, { "epoch": 1.4349693693693695, "grad_norm": 32.05421829223633, "learning_rate": 1.2282142414215747e-06, "loss": 9.357336044311523, "loss_alignment": 0.77734375, "loss_alignment_w": 0.388671875, "loss_sft": 0.19632946141064167, "loss_total_v6": 0.5848335064947605, "step": 6222 }, { "epoch": 1.4352, "grad_norm": 30.93949317932129, "learning_rate": 1.2268925325142478e-06, "loss": 9.413446426391602, "loss_alignment": 0.79248046875, "loss_alignment_w": 0.396240234375, "loss_sft": 0.19351927749812603, "loss_total_v6": 0.5883404612541199, "step": 6223 }, { "epoch": 1.4354306306306306, "grad_norm": 31.621183395385742, "learning_rate": 1.225571435689062e-06, "loss": 9.31019401550293, "loss_alignment": 0.77880859375, "loss_alignment_w": 0.389404296875, "loss_sft": 0.19298639334738255, "loss_total_v6": 0.5818871483206749, "step": 6224 }, { "epoch": 1.4356612612612611, "grad_norm": 30.48735237121582, "learning_rate": 1.2242509511603318e-06, "loss": 8.949642181396484, "loss_alignment": 0.77294921875, "loss_alignment_w": 0.386474609375, "loss_sft": 0.17232869286090136, "loss_total_v6": 0.5593526437878609, "step": 6225 }, { "epoch": 1.435891891891892, "grad_norm": 33.12965393066406, "learning_rate": 1.2229310791422678e-06, "loss": 9.389331817626953, "loss_alignment": 0.771484375, "loss_alignment_w": 0.3857421875, "loss_sft": 0.20078590605407953, "loss_total_v6": 0.5868332609534264, "step": 6226 }, { "epoch": 1.4361225225225225, "grad_norm": 35.753929138183594, "learning_rate": 1.2216118198489823e-06, "loss": 9.876080513000488, "loss_alignment": 0.81298828125, "loss_alignment_w": 0.406494140625, "loss_sft": 0.21088296733796597, "loss_total_v6": 0.6172550320625305, "step": 6227 }, { "epoch": 1.4363531531531533, "grad_norm": 31.42167091369629, "learning_rate": 1.2202931734944878e-06, "loss": 9.487518310546875, "loss_alignment": 0.77880859375, "loss_alignment_w": 0.389404296875, "loss_sft": 0.2033519558608532, "loss_total_v6": 0.5929698720574379, "step": 6228 }, { "epoch": 1.4365837837837838, "grad_norm": 30.116287231445312, "learning_rate": 1.2189751402926991e-06, "loss": 8.873269081115723, "loss_alignment": 0.765625, "loss_alignment_w": 0.3828125, "loss_sft": 0.17198044806718826, "loss_total_v6": 0.5545793250203133, "step": 6229 }, { "epoch": 1.4368144144144144, "grad_norm": 26.53211212158203, "learning_rate": 1.2176577204574318e-06, "loss": 9.361441612243652, "loss_alignment": 0.765625, "loss_alignment_w": 0.3828125, "loss_sft": 0.2032236736267805, "loss_total_v6": 0.5850901156663895, "step": 6230 }, { "epoch": 1.437045045045045, "grad_norm": 28.069942474365234, "learning_rate": 1.2163409142024e-06, "loss": 8.676342964172363, "loss_alignment": 0.720703125, "loss_alignment_w": 0.3603515625, "loss_sft": 0.18216403014957905, "loss_total_v6": 0.5422714650630951, "step": 6231 }, { "epoch": 1.4372756756756757, "grad_norm": 29.702594757080078, "learning_rate": 1.2150247217412186e-06, "loss": 9.303956985473633, "loss_alignment": 0.76318359375, "loss_alignment_w": 0.381591796875, "loss_sft": 0.19851691462099552, "loss_total_v6": 0.5814972892403603, "step": 6232 }, { "epoch": 1.4375063063063063, "grad_norm": 32.8925666809082, "learning_rate": 1.213709143287403e-06, "loss": 8.917701721191406, "loss_alignment": 0.7451171875, "loss_alignment_w": 0.37255859375, "loss_sft": 0.1845535822212696, "loss_total_v6": 0.5573563203215599, "step": 6233 }, { "epoch": 1.437736936936937, "grad_norm": 33.30370330810547, "learning_rate": 1.2123941790543691e-06, "loss": 8.89742660522461, "loss_alignment": 0.74169921875, "loss_alignment_w": 0.370849609375, "loss_sft": 0.18610930256545544, "loss_total_v6": 0.5560891665518284, "step": 6234 }, { "epoch": 1.4379675675675676, "grad_norm": 29.883535385131836, "learning_rate": 1.2110798292554333e-06, "loss": 9.796424865722656, "loss_alignment": 0.8203125, "loss_alignment_w": 0.41015625, "loss_sft": 0.20183039270341396, "loss_total_v6": 0.6122765466570854, "step": 6235 }, { "epoch": 1.4381981981981982, "grad_norm": 26.5616455078125, "learning_rate": 1.2097660941038148e-06, "loss": 9.180988311767578, "loss_alignment": 0.7490234375, "loss_alignment_w": 0.37451171875, "loss_sft": 0.1995442397892475, "loss_total_v6": 0.5738118253648281, "step": 6236 }, { "epoch": 1.4384288288288287, "grad_norm": 29.501169204711914, "learning_rate": 1.2084529738126294e-06, "loss": 9.287425994873047, "loss_alignment": 0.76025390625, "loss_alignment_w": 0.380126953125, "loss_sft": 0.20015409961342812, "loss_total_v6": 0.5804641470313072, "step": 6237 }, { "epoch": 1.4386594594594595, "grad_norm": 29.48787498474121, "learning_rate": 1.207140468594894e-06, "loss": 9.405415534973145, "loss_alignment": 0.7890625, "loss_alignment_w": 0.39453125, "loss_sft": 0.19295623525977135, "loss_total_v6": 0.5878384560346603, "step": 6238 }, { "epoch": 1.43889009009009, "grad_norm": 27.412208557128906, "learning_rate": 1.2058285786635248e-06, "loss": 9.061468124389648, "loss_alignment": 0.77978515625, "loss_alignment_w": 0.389892578125, "loss_sft": 0.1766017209738493, "loss_total_v6": 0.56634172052145, "step": 6239 }, { "epoch": 1.4391207207207208, "grad_norm": 32.57073211669922, "learning_rate": 1.2045173042313429e-06, "loss": 9.381210327148438, "loss_alignment": 0.75146484375, "loss_alignment_w": 0.375732421875, "loss_sft": 0.20981504023075104, "loss_total_v6": 0.5863256603479385, "step": 6240 }, { "epoch": 1.4393513513513514, "grad_norm": 33.08819580078125, "learning_rate": 1.203206645511063e-06, "loss": 9.044547080993652, "loss_alignment": 0.75244140625, "loss_alignment_w": 0.376220703125, "loss_sft": 0.18946018442511559, "loss_total_v6": 0.5652841478586197, "step": 6241 }, { "epoch": 1.439581981981982, "grad_norm": 27.59758186340332, "learning_rate": 1.2018966027153062e-06, "loss": 8.532064437866211, "loss_alignment": 0.734375, "loss_alignment_w": 0.3671875, "loss_sft": 0.1655782349407673, "loss_total_v6": 0.5332539975643158, "step": 6242 }, { "epoch": 1.4398126126126125, "grad_norm": 34.115848541259766, "learning_rate": 1.200587176056589e-06, "loss": 8.903294563293457, "loss_alignment": 0.7578125, "loss_alignment_w": 0.37890625, "loss_sft": 0.17713768314570189, "loss_total_v6": 0.5564558953046799, "step": 6243 }, { "epoch": 1.4400432432432433, "grad_norm": 34.909141540527344, "learning_rate": 1.199278365747329e-06, "loss": 8.806360244750977, "loss_alignment": 0.7216796875, "loss_alignment_w": 0.36083984375, "loss_sft": 0.18905416317284107, "loss_total_v6": 0.5503975301980972, "step": 6244 }, { "epoch": 1.4402738738738738, "grad_norm": 30.225664138793945, "learning_rate": 1.1979701719998454e-06, "loss": 8.861053466796875, "loss_alignment": 0.75927734375, "loss_alignment_w": 0.379638671875, "loss_sft": 0.1745433323085308, "loss_total_v6": 0.5538158193230629, "step": 6245 }, { "epoch": 1.4405045045045046, "grad_norm": 32.191368103027344, "learning_rate": 1.1966625950263572e-06, "loss": 9.140294075012207, "loss_alignment": 0.7607421875, "loss_alignment_w": 0.38037109375, "loss_sft": 0.1916754674166441, "loss_total_v6": 0.5712683647871017, "step": 6246 }, { "epoch": 1.4407351351351352, "grad_norm": 35.35086441040039, "learning_rate": 1.1953556350389795e-06, "loss": 9.157062530517578, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.18794752284884453, "loss_total_v6": 0.5723164081573486, "step": 6247 }, { "epoch": 1.4409657657657657, "grad_norm": 27.403932571411133, "learning_rate": 1.1940492922497337e-06, "loss": 9.235360145568848, "loss_alignment": 0.75927734375, "loss_alignment_w": 0.379638671875, "loss_sft": 0.19764763116836548, "loss_total_v6": 0.5772100165486336, "step": 6248 }, { "epoch": 1.4411963963963963, "grad_norm": 33.86410903930664, "learning_rate": 1.1927435668705357e-06, "loss": 9.352705001831055, "loss_alignment": 0.75, "loss_alignment_w": 0.375, "loss_sft": 0.20829286240041256, "loss_total_v6": 0.5845441073179245, "step": 6249 }, { "epoch": 1.441427027027027, "grad_norm": 29.592674255371094, "learning_rate": 1.1914384591132045e-06, "loss": 8.919734954833984, "loss_alignment": 0.75048828125, "loss_alignment_w": 0.375244140625, "loss_sft": 0.18321583606302738, "loss_total_v6": 0.5574834197759628, "step": 6250 }, { "epoch": 1.4416576576576576, "grad_norm": 30.21247100830078, "learning_rate": 1.1901339691894575e-06, "loss": 9.194636344909668, "loss_alignment": 0.7724609375, "loss_alignment_w": 0.38623046875, "loss_sft": 0.18777816370129585, "loss_total_v6": 0.5746647641062737, "step": 6251 }, { "epoch": 1.4418882882882884, "grad_norm": 28.95271873474121, "learning_rate": 1.1888300973109112e-06, "loss": 9.433436393737793, "loss_alignment": 0.771484375, "loss_alignment_w": 0.3857421875, "loss_sft": 0.20367972366511822, "loss_total_v6": 0.5895897522568703, "step": 6252 }, { "epoch": 1.442118918918919, "grad_norm": 30.30364990234375, "learning_rate": 1.1875268436890824e-06, "loss": 8.809093475341797, "loss_alignment": 0.7421875, "loss_alignment_w": 0.37109375, "loss_sft": 0.18019171990454197, "loss_total_v6": 0.550568301230669, "step": 6253 }, { "epoch": 1.4423495495495495, "grad_norm": 32.858070373535156, "learning_rate": 1.186224208535388e-06, "loss": 8.880935668945312, "loss_alignment": 0.71044921875, "loss_alignment_w": 0.355224609375, "loss_sft": 0.19913194328546524, "loss_total_v6": 0.5550584681332111, "step": 6254 }, { "epoch": 1.44258018018018, "grad_norm": 31.738065719604492, "learning_rate": 1.1849221920611464e-06, "loss": 9.00433349609375, "loss_alignment": 0.74609375, "loss_alignment_w": 0.373046875, "loss_sft": 0.18929672800004482, "loss_total_v6": 0.56277085095644, "step": 6255 }, { "epoch": 1.4428108108108109, "grad_norm": 35.19448471069336, "learning_rate": 1.1836207944775729e-06, "loss": 9.089146614074707, "loss_alignment": 0.75927734375, "loss_alignment_w": 0.379638671875, "loss_sft": 0.18895177356898785, "loss_total_v6": 0.5680716559290886, "step": 6256 }, { "epoch": 1.4430414414414414, "grad_norm": 28.218955993652344, "learning_rate": 1.1823200159957827e-06, "loss": 9.292259216308594, "loss_alignment": 0.75634765625, "loss_alignment_w": 0.378173828125, "loss_sft": 0.20268385857343674, "loss_total_v6": 0.5807661488652229, "step": 6257 }, { "epoch": 1.4432720720720722, "grad_norm": 30.469871520996094, "learning_rate": 1.1810198568267906e-06, "loss": 9.234752655029297, "loss_alignment": 0.75830078125, "loss_alignment_w": 0.379150390625, "loss_sft": 0.19869300723075867, "loss_total_v6": 0.5771720185875893, "step": 6258 }, { "epoch": 1.4435027027027028, "grad_norm": 30.740985870361328, "learning_rate": 1.1797203171815103e-06, "loss": 8.979669570922852, "loss_alignment": 0.73974609375, "loss_alignment_w": 0.369873046875, "loss_sft": 0.1915394039824605, "loss_total_v6": 0.5612293481826782, "step": 6259 }, { "epoch": 1.4437333333333333, "grad_norm": 31.641796112060547, "learning_rate": 1.1784213972707581e-06, "loss": 8.942825317382812, "loss_alignment": 0.7529296875, "loss_alignment_w": 0.37646484375, "loss_sft": 0.18133255932480097, "loss_total_v6": 0.5589265674352646, "step": 6260 }, { "epoch": 1.4439639639639639, "grad_norm": 27.94162368774414, "learning_rate": 1.1771230973052484e-06, "loss": 8.88869857788086, "loss_alignment": 0.73486328125, "loss_alignment_w": 0.367431640625, "loss_sft": 0.18817304261028767, "loss_total_v6": 0.5555436536669731, "step": 6261 }, { "epoch": 1.4441945945945946, "grad_norm": 28.98949432373047, "learning_rate": 1.1758254174955935e-06, "loss": 9.043537139892578, "loss_alignment": 0.7587890625, "loss_alignment_w": 0.37939453125, "loss_sft": 0.18599438294768333, "loss_total_v6": 0.5652210563421249, "step": 6262 }, { "epoch": 1.4444252252252252, "grad_norm": 30.59282684326172, "learning_rate": 1.1745283580523059e-06, "loss": 8.951107025146484, "loss_alignment": 0.7646484375, "loss_alignment_w": 0.38232421875, "loss_sft": 0.17690638359636068, "loss_total_v6": 0.5594441965222359, "step": 6263 }, { "epoch": 1.444655855855856, "grad_norm": 27.679292678833008, "learning_rate": 1.1732319191857955e-06, "loss": 9.301401138305664, "loss_alignment": 0.78515625, "loss_alignment_w": 0.392578125, "loss_sft": 0.1888204626739025, "loss_total_v6": 0.5813375562429428, "step": 6264 }, { "epoch": 1.4448864864864865, "grad_norm": 44.871768951416016, "learning_rate": 1.1719361011063773e-06, "loss": 9.262012481689453, "loss_alignment": 0.7763671875, "loss_alignment_w": 0.38818359375, "loss_sft": 0.19104315154254436, "loss_total_v6": 0.578875795006752, "step": 6265 }, { "epoch": 1.445117117117117, "grad_norm": 31.820829391479492, "learning_rate": 1.1706409040242588e-06, "loss": 9.01907730102539, "loss_alignment": 0.7529296875, "loss_alignment_w": 0.37646484375, "loss_sft": 0.18776154704391956, "loss_total_v6": 0.5636923462152481, "step": 6266 }, { "epoch": 1.4453477477477477, "grad_norm": 30.952041625976562, "learning_rate": 1.1693463281495527e-06, "loss": 9.449295997619629, "loss_alignment": 0.7861328125, "loss_alignment_w": 0.39306640625, "loss_sft": 0.1974230445921421, "loss_total_v6": 0.5905809998512268, "step": 6267 }, { "epoch": 1.4455783783783784, "grad_norm": 28.71096420288086, "learning_rate": 1.168052373692266e-06, "loss": 9.08074951171875, "loss_alignment": 0.7353515625, "loss_alignment_w": 0.36767578125, "loss_sft": 0.20083234459161758, "loss_total_v6": 0.5675468221306801, "step": 6268 }, { "epoch": 1.445809009009009, "grad_norm": 33.8764762878418, "learning_rate": 1.1667590408623065e-06, "loss": 8.842552185058594, "loss_alignment": 0.77001953125, "loss_alignment_w": 0.385009765625, "loss_sft": 0.16659687459468842, "loss_total_v6": 0.5526594892144203, "step": 6269 }, { "epoch": 1.4460396396396398, "grad_norm": 33.17107391357422, "learning_rate": 1.1654663298694834e-06, "loss": 9.533500671386719, "loss_alignment": 0.771484375, "loss_alignment_w": 0.3857421875, "loss_sft": 0.21025422401726246, "loss_total_v6": 0.5958437994122505, "step": 6270 }, { "epoch": 1.4462702702702703, "grad_norm": 31.608442306518555, "learning_rate": 1.164174240923503e-06, "loss": 8.862136840820312, "loss_alignment": 0.76611328125, "loss_alignment_w": 0.383056640625, "loss_sft": 0.1714372830465436, "loss_total_v6": 0.5538835600018501, "step": 6271 }, { "epoch": 1.4465009009009009, "grad_norm": 28.022165298461914, "learning_rate": 1.1628827742339688e-06, "loss": 9.643732070922852, "loss_alignment": 0.80126953125, "loss_alignment_w": 0.400634765625, "loss_sft": 0.20220530778169632, "loss_total_v6": 0.60273327678442, "step": 6272 }, { "epoch": 1.4467315315315314, "grad_norm": 33.18366622924805, "learning_rate": 1.1615919300103874e-06, "loss": 9.139265060424805, "loss_alignment": 0.7607421875, "loss_alignment_w": 0.38037109375, "loss_sft": 0.19049726705998182, "loss_total_v6": 0.5712040513753891, "step": 6273 }, { "epoch": 1.4469621621621622, "grad_norm": 35.67607879638672, "learning_rate": 1.1603017084621632e-06, "loss": 8.916508674621582, "loss_alignment": 0.74365234375, "loss_alignment_w": 0.371826171875, "loss_sft": 0.1855929270386696, "loss_total_v6": 0.5572817847132683, "step": 6274 }, { "epoch": 1.4471927927927928, "grad_norm": 39.222198486328125, "learning_rate": 1.159012109798598e-06, "loss": 8.623029708862305, "loss_alignment": 0.72119140625, "loss_alignment_w": 0.360595703125, "loss_sft": 0.17825214751064777, "loss_total_v6": 0.5389393866062164, "step": 6275 }, { "epoch": 1.4474234234234236, "grad_norm": 30.31977653503418, "learning_rate": 1.157723134228893e-06, "loss": 9.291229248046875, "loss_alignment": 0.77392578125, "loss_alignment_w": 0.386962890625, "loss_sft": 0.19358638674020767, "loss_total_v6": 0.5807018801569939, "step": 6276 }, { "epoch": 1.4476540540540541, "grad_norm": 31.52369499206543, "learning_rate": 1.156434781962149e-06, "loss": 8.887389183044434, "loss_alignment": 0.75927734375, "loss_alignment_w": 0.379638671875, "loss_sft": 0.17550271563231945, "loss_total_v6": 0.5554618164896965, "step": 6277 }, { "epoch": 1.4478846846846847, "grad_norm": 29.025409698486328, "learning_rate": 1.1551470532073643e-06, "loss": 9.354110717773438, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.19661614298820496, "loss_total_v6": 0.5846318602561951, "step": 6278 }, { "epoch": 1.4481153153153152, "grad_norm": 29.143747329711914, "learning_rate": 1.1538599481734385e-06, "loss": 9.270474433898926, "loss_alignment": 0.76220703125, "loss_alignment_w": 0.381103515625, "loss_sft": 0.19708039984107018, "loss_total_v6": 0.5794046297669411, "step": 6279 }, { "epoch": 1.448345945945946, "grad_norm": 28.88372802734375, "learning_rate": 1.1525734670691702e-06, "loss": 9.030699729919434, "loss_alignment": 0.75830078125, "loss_alignment_w": 0.379150390625, "loss_sft": 0.18447486590594053, "loss_total_v6": 0.564418725669384, "step": 6280 }, { "epoch": 1.4485765765765766, "grad_norm": 34.33720397949219, "learning_rate": 1.1512876101032538e-06, "loss": 9.340335845947266, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.1993105560541153, "loss_total_v6": 0.5837709978222847, "step": 6281 }, { "epoch": 1.4488072072072071, "grad_norm": 30.389812469482422, "learning_rate": 1.150002377484284e-06, "loss": 8.883548736572266, "loss_alignment": 0.71875, "loss_alignment_w": 0.359375, "loss_sft": 0.19619769230484962, "loss_total_v6": 0.5552217364311218, "step": 6282 }, { "epoch": 1.449037837837838, "grad_norm": 30.654062271118164, "learning_rate": 1.1487177694207525e-06, "loss": 8.348703384399414, "loss_alignment": 0.73193359375, "loss_alignment_w": 0.365966796875, "loss_sft": 0.15605599898844957, "loss_total_v6": 0.5217939205467701, "step": 6283 }, { "epoch": 1.4492684684684685, "grad_norm": 33.60588073730469, "learning_rate": 1.1474337861210543e-06, "loss": 9.074441909790039, "loss_alignment": 0.75537109375, "loss_alignment_w": 0.377685546875, "loss_sft": 0.1908555682748556, "loss_total_v6": 0.5671525746583939, "step": 6284 }, { "epoch": 1.449499099099099, "grad_norm": 32.8346061706543, "learning_rate": 1.1461504277934776e-06, "loss": 9.132001876831055, "loss_alignment": 0.771484375, "loss_alignment_w": 0.3857421875, "loss_sft": 0.18554197996854782, "loss_total_v6": 0.5707500949501991, "step": 6285 }, { "epoch": 1.4497297297297298, "grad_norm": 31.636295318603516, "learning_rate": 1.1448676946462145e-06, "loss": 8.98447036743164, "loss_alignment": 0.767578125, "loss_alignment_w": 0.3837890625, "loss_sft": 0.17836601845920086, "loss_total_v6": 0.5615294575691223, "step": 6286 }, { "epoch": 1.4499603603603604, "grad_norm": 31.83011817932129, "learning_rate": 1.1435855868873513e-06, "loss": 9.12756061553955, "loss_alignment": 0.7529296875, "loss_alignment_w": 0.37646484375, "loss_sft": 0.19405345804989338, "loss_total_v6": 0.5704725235700607, "step": 6287 }, { "epoch": 1.450190990990991, "grad_norm": 31.15192413330078, "learning_rate": 1.142304104724873e-06, "loss": 9.52149772644043, "loss_alignment": 0.78173828125, "loss_alignment_w": 0.390869140625, "loss_sft": 0.20434660278260708, "loss_total_v6": 0.5950936526060104, "step": 6288 }, { "epoch": 1.4504216216216217, "grad_norm": 44.05287170410156, "learning_rate": 1.1410232483666679e-06, "loss": 9.647663116455078, "loss_alignment": 0.7841796875, "loss_alignment_w": 0.39208984375, "loss_sft": 0.211453627794981, "loss_total_v6": 0.6029789075255394, "step": 6289 }, { "epoch": 1.4506522522522523, "grad_norm": 32.102264404296875, "learning_rate": 1.139743018020517e-06, "loss": 8.967330932617188, "loss_alignment": 0.7666015625, "loss_alignment_w": 0.38330078125, "loss_sft": 0.17611974105238914, "loss_total_v6": 0.5604581460356712, "step": 6290 }, { "epoch": 1.4508828828828828, "grad_norm": 31.7083740234375, "learning_rate": 1.1384634138941014e-06, "loss": 8.814531326293945, "loss_alignment": 0.75390625, "loss_alignment_w": 0.376953125, "loss_sft": 0.17461120896041393, "loss_total_v6": 0.5509081929922104, "step": 6291 }, { "epoch": 1.4511135135135136, "grad_norm": 33.78828811645508, "learning_rate": 1.1371844361950045e-06, "loss": 9.27971363067627, "loss_alignment": 0.77490234375, "loss_alignment_w": 0.387451171875, "loss_sft": 0.1928971093147993, "loss_total_v6": 0.579982079565525, "step": 6292 }, { "epoch": 1.4513441441441441, "grad_norm": 25.342374801635742, "learning_rate": 1.1359060851307024e-06, "loss": 8.707742691040039, "loss_alignment": 0.7470703125, "loss_alignment_w": 0.37353515625, "loss_sft": 0.17088184598833323, "loss_total_v6": 0.5442339181900024, "step": 6293 }, { "epoch": 1.4515747747747747, "grad_norm": 33.595863342285156, "learning_rate": 1.134628360908574e-06, "loss": 9.380895614624023, "loss_alignment": 0.78955078125, "loss_alignment_w": 0.394775390625, "loss_sft": 0.19124064035713673, "loss_total_v6": 0.5863059461116791, "step": 6294 }, { "epoch": 1.4518054054054055, "grad_norm": 31.138975143432617, "learning_rate": 1.133351263735894e-06, "loss": 8.795724868774414, "loss_alignment": 0.7548828125, "loss_alignment_w": 0.37744140625, "loss_sft": 0.17323741875588894, "loss_total_v6": 0.5497327856719494, "step": 6295 }, { "epoch": 1.452036036036036, "grad_norm": 28.826391220092773, "learning_rate": 1.1320747938198356e-06, "loss": 9.210542678833008, "loss_alignment": 0.80419921875, "loss_alignment_w": 0.402099609375, "loss_sft": 0.1755735008046031, "loss_total_v6": 0.575658954679966, "step": 6296 }, { "epoch": 1.4522666666666666, "grad_norm": 29.09354019165039, "learning_rate": 1.1307989513674695e-06, "loss": 9.069656372070312, "loss_alignment": 0.74267578125, "loss_alignment_w": 0.371337890625, "loss_sft": 0.1953325066715479, "loss_total_v6": 0.5668535158038139, "step": 6297 }, { "epoch": 1.4524972972972974, "grad_norm": 31.798688888549805, "learning_rate": 1.129523736585767e-06, "loss": 9.306827545166016, "loss_alignment": 0.76416015625, "loss_alignment_w": 0.382080078125, "loss_sft": 0.20080207102000713, "loss_total_v6": 0.5816766992211342, "step": 6298 }, { "epoch": 1.452727927927928, "grad_norm": 33.07534408569336, "learning_rate": 1.1282491496815978e-06, "loss": 9.40402603149414, "loss_alignment": 0.79833984375, "loss_alignment_w": 0.399169921875, "loss_sft": 0.18881058041006327, "loss_total_v6": 0.5877516195178032, "step": 6299 }, { "epoch": 1.4529585585585585, "grad_norm": 31.63080406188965, "learning_rate": 1.1269751908617277e-06, "loss": 8.843154907226562, "loss_alignment": 0.755859375, "loss_alignment_w": 0.3779296875, "loss_sft": 0.17459961492568254, "loss_total_v6": 0.5526971444487572, "step": 6300 }, { "epoch": 1.4531891891891893, "grad_norm": 30.152799606323242, "learning_rate": 1.12570186033282e-06, "loss": 8.873156547546387, "loss_alignment": 0.7509765625, "loss_alignment_w": 0.37548828125, "loss_sft": 0.17842783220112324, "loss_total_v6": 0.5545722581446171, "step": 6301 }, { "epoch": 1.4534198198198198, "grad_norm": 30.45302963256836, "learning_rate": 1.1244291583014377e-06, "loss": 8.861457824707031, "loss_alignment": 0.76025390625, "loss_alignment_w": 0.380126953125, "loss_sft": 0.1728444416075945, "loss_total_v6": 0.5538411736488342, "step": 6302 }, { "epoch": 1.4536504504504504, "grad_norm": 30.57218360900879, "learning_rate": 1.1231570849740403e-06, "loss": 8.813064575195312, "loss_alignment": 0.72216796875, "loss_alignment_w": 0.361083984375, "loss_sft": 0.1897173449397087, "loss_total_v6": 0.5508165918290615, "step": 6303 }, { "epoch": 1.4538810810810812, "grad_norm": 42.532615661621094, "learning_rate": 1.1218856405569883e-06, "loss": 9.181068420410156, "loss_alignment": 0.74755859375, "loss_alignment_w": 0.373779296875, "loss_sft": 0.20031209848821163, "loss_total_v6": 0.5738167390227318, "step": 6304 }, { "epoch": 1.4541117117117117, "grad_norm": 29.793598175048828, "learning_rate": 1.1206148252565385e-06, "loss": 9.037949562072754, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.17824464477598667, "loss_total_v6": 0.5648718476295471, "step": 6305 }, { "epoch": 1.4543423423423423, "grad_norm": 30.563629150390625, "learning_rate": 1.1193446392788448e-06, "loss": 8.76371955871582, "loss_alignment": 0.720703125, "loss_alignment_w": 0.3603515625, "loss_sft": 0.18645013682544231, "loss_total_v6": 0.5477324873209, "step": 6306 }, { "epoch": 1.454572972972973, "grad_norm": 32.22815704345703, "learning_rate": 1.1180750828299597e-06, "loss": 8.894201278686523, "loss_alignment": 0.744140625, "loss_alignment_w": 0.3720703125, "loss_sft": 0.18345104902982712, "loss_total_v6": 0.5558875799179077, "step": 6307 }, { "epoch": 1.4548036036036036, "grad_norm": 30.617300033569336, "learning_rate": 1.1168061561158321e-06, "loss": 9.245755195617676, "loss_alignment": 0.77294921875, "loss_alignment_w": 0.386474609375, "loss_sft": 0.19143087975680828, "loss_total_v6": 0.5778596922755241, "step": 6308 }, { "epoch": 1.4550342342342342, "grad_norm": 29.11826515197754, "learning_rate": 1.1155378593423127e-06, "loss": 9.438143730163574, "loss_alignment": 0.7861328125, "loss_alignment_w": 0.39306640625, "loss_sft": 0.1966954693198204, "loss_total_v6": 0.5898839607834816, "step": 6309 }, { "epoch": 1.4552648648648647, "grad_norm": 34.50224685668945, "learning_rate": 1.1142701927151456e-06, "loss": 9.20007038116455, "loss_alignment": 0.77490234375, "loss_alignment_w": 0.387451171875, "loss_sft": 0.18772105313837528, "loss_total_v6": 0.5750043913722038, "step": 6310 }, { "epoch": 1.4554954954954955, "grad_norm": 31.112686157226562, "learning_rate": 1.1130031564399762e-06, "loss": 9.55176067352295, "loss_alignment": 0.78125, "loss_alignment_w": 0.390625, "loss_sft": 0.20547500625252724, "loss_total_v6": 0.5969850420951843, "step": 6311 }, { "epoch": 1.455726126126126, "grad_norm": 32.07720184326172, "learning_rate": 1.1117367507223452e-06, "loss": 9.310301780700684, "loss_alignment": 0.75341796875, "loss_alignment_w": 0.376708984375, "loss_sft": 0.2053526882082224, "loss_total_v6": 0.5818938389420509, "step": 6312 }, { "epoch": 1.4559567567567568, "grad_norm": 32.239723205566406, "learning_rate": 1.1104709757676908e-06, "loss": 9.258005142211914, "loss_alignment": 0.7607421875, "loss_alignment_w": 0.38037109375, "loss_sft": 0.19881877303123474, "loss_total_v6": 0.5786252990365028, "step": 6313 }, { "epoch": 1.4561873873873874, "grad_norm": 29.361587524414062, "learning_rate": 1.109205831781352e-06, "loss": 9.013845443725586, "loss_alignment": 0.76611328125, "loss_alignment_w": 0.383056640625, "loss_sft": 0.1801865790039301, "loss_total_v6": 0.5633652880787849, "step": 6314 }, { "epoch": 1.456418018018018, "grad_norm": 30.554996490478516, "learning_rate": 1.1079413189685629e-06, "loss": 8.527324676513672, "loss_alignment": 0.71728515625, "loss_alignment_w": 0.358642578125, "loss_sft": 0.1742541529238224, "loss_total_v6": 0.5329577811062336, "step": 6315 }, { "epoch": 1.4566486486486485, "grad_norm": 29.164186477661133, "learning_rate": 1.106677437534453e-06, "loss": 9.00307846069336, "loss_alignment": 0.76953125, "loss_alignment_w": 0.384765625, "loss_sft": 0.1781556848436594, "loss_total_v6": 0.5626924186944962, "step": 6316 }, { "epoch": 1.4568792792792793, "grad_norm": 30.617076873779297, "learning_rate": 1.1054141876840557e-06, "loss": 8.812485694885254, "loss_alignment": 0.74853515625, "loss_alignment_w": 0.374267578125, "loss_sft": 0.1763449590653181, "loss_total_v6": 0.5507803745567799, "step": 6317 }, { "epoch": 1.4571099099099099, "grad_norm": 29.136486053466797, "learning_rate": 1.1041515696222953e-06, "loss": 9.246654510498047, "loss_alignment": 0.76806640625, "loss_alignment_w": 0.384033203125, "loss_sft": 0.19502713717520237, "loss_total_v6": 0.5779159218072891, "step": 6318 }, { "epoch": 1.4573405405405406, "grad_norm": 27.858190536499023, "learning_rate": 1.1028895835539994e-06, "loss": 8.847345352172852, "loss_alignment": 0.7646484375, "loss_alignment_w": 0.38232421875, "loss_sft": 0.1708942512050271, "loss_total_v6": 0.5529590845108032, "step": 6319 }, { "epoch": 1.4575711711711712, "grad_norm": 32.5302848815918, "learning_rate": 1.1016282296838887e-06, "loss": 9.660381317138672, "loss_alignment": 0.78466796875, "loss_alignment_w": 0.392333984375, "loss_sft": 0.21174507588148117, "loss_total_v6": 0.6037738770246506, "step": 6320 }, { "epoch": 1.4578018018018017, "grad_norm": 30.0101318359375, "learning_rate": 1.1003675082165833e-06, "loss": 9.16413402557373, "loss_alignment": 0.76171875, "loss_alignment_w": 0.380859375, "loss_sft": 0.1928297858685255, "loss_total_v6": 0.5727583691477776, "step": 6321 }, { "epoch": 1.4580324324324323, "grad_norm": 34.851375579833984, "learning_rate": 1.0991074193565987e-06, "loss": 9.31153678894043, "loss_alignment": 0.78564453125, "loss_alignment_w": 0.392822265625, "loss_sft": 0.1896065790206194, "loss_total_v6": 0.5819710865616798, "step": 6322 }, { "epoch": 1.458263063063063, "grad_norm": 30.54810333251953, "learning_rate": 1.097847963308351e-06, "loss": 8.88930606842041, "loss_alignment": 0.7236328125, "loss_alignment_w": 0.36181640625, "loss_sft": 0.1924834791570902, "loss_total_v6": 0.5555816255509853, "step": 6323 }, { "epoch": 1.4584936936936936, "grad_norm": 34.58028793334961, "learning_rate": 1.096589140276153e-06, "loss": 8.473812103271484, "loss_alignment": 0.740234375, "loss_alignment_w": 0.3701171875, "loss_sft": 0.15897726453840733, "loss_total_v6": 0.5296132639050484, "step": 6324 }, { "epoch": 1.4587243243243244, "grad_norm": 33.46931838989258, "learning_rate": 1.095330950464213e-06, "loss": 8.866788864135742, "loss_alignment": 0.73388671875, "loss_alignment_w": 0.366943359375, "loss_sft": 0.18694103881716728, "loss_total_v6": 0.5541743114590645, "step": 6325 }, { "epoch": 1.458954954954955, "grad_norm": 42.30652618408203, "learning_rate": 1.0940733940766369e-06, "loss": 9.2659330368042, "loss_alignment": 0.75146484375, "loss_alignment_w": 0.375732421875, "loss_sft": 0.2024881336838007, "loss_total_v6": 0.5791208073496819, "step": 6326 }, { "epoch": 1.4591855855855855, "grad_norm": 34.924007415771484, "learning_rate": 1.0928164713174272e-06, "loss": 9.221063613891602, "loss_alignment": 0.7607421875, "loss_alignment_w": 0.38037109375, "loss_sft": 0.19541134871542454, "loss_total_v6": 0.5763165056705475, "step": 6327 }, { "epoch": 1.459416216216216, "grad_norm": 31.0493221282959, "learning_rate": 1.0915601823904875e-06, "loss": 9.153436660766602, "loss_alignment": 0.74267578125, "loss_alignment_w": 0.371337890625, "loss_sft": 0.2000346351414919, "loss_total_v6": 0.5720897391438484, "step": 6328 }, { "epoch": 1.4596468468468469, "grad_norm": 28.700847625732422, "learning_rate": 1.0903045274996133e-06, "loss": 8.911001205444336, "loss_alignment": 0.736328125, "loss_alignment_w": 0.3681640625, "loss_sft": 0.18784273974597454, "loss_total_v6": 0.5569375902414322, "step": 6329 }, { "epoch": 1.4598774774774774, "grad_norm": 32.67145919799805, "learning_rate": 1.0890495068485024e-06, "loss": 8.824068069458008, "loss_alignment": 0.755859375, "loss_alignment_w": 0.3779296875, "loss_sft": 0.17358984891325235, "loss_total_v6": 0.5515042655169964, "step": 6330 }, { "epoch": 1.4601081081081082, "grad_norm": 28.672054290771484, "learning_rate": 1.0877951206407462e-06, "loss": 8.884801864624023, "loss_alignment": 0.77001953125, "loss_alignment_w": 0.385009765625, "loss_sft": 0.17039712239056826, "loss_total_v6": 0.5553000792860985, "step": 6331 }, { "epoch": 1.4603387387387388, "grad_norm": 29.805044174194336, "learning_rate": 1.0865413690798321e-06, "loss": 9.28123950958252, "loss_alignment": 0.76953125, "loss_alignment_w": 0.384765625, "loss_sft": 0.19589166529476643, "loss_total_v6": 0.5800774544477463, "step": 6332 }, { "epoch": 1.4605693693693693, "grad_norm": 29.77666473388672, "learning_rate": 1.0852882523691498e-06, "loss": 8.88191032409668, "loss_alignment": 0.75, "loss_alignment_w": 0.375, "loss_sft": 0.18121794797480106, "loss_total_v6": 0.5551193431019783, "step": 6333 }, { "epoch": 1.4607999999999999, "grad_norm": 34.93225860595703, "learning_rate": 1.0840357707119808e-06, "loss": 9.266852378845215, "loss_alignment": 0.73974609375, "loss_alignment_w": 0.369873046875, "loss_sft": 0.20848125033080578, "loss_total_v6": 0.5791782885789871, "step": 6334 }, { "epoch": 1.4610306306306307, "grad_norm": 29.667695999145508, "learning_rate": 1.0827839243115046e-06, "loss": 8.765541076660156, "loss_alignment": 0.74609375, "loss_alignment_w": 0.373046875, "loss_sft": 0.17400601133704185, "loss_total_v6": 0.5478463247418404, "step": 6335 }, { "epoch": 1.4612612612612612, "grad_norm": 35.778656005859375, "learning_rate": 1.0815327133708015e-06, "loss": 9.271269798278809, "loss_alignment": 0.78369140625, "loss_alignment_w": 0.391845703125, "loss_sft": 0.18712038174271584, "loss_total_v6": 0.5794543698430061, "step": 6336 }, { "epoch": 1.461491891891892, "grad_norm": 30.156641006469727, "learning_rate": 1.080282138092843e-06, "loss": 8.656723976135254, "loss_alignment": 0.72705078125, "loss_alignment_w": 0.363525390625, "loss_sft": 0.17796237766742706, "loss_total_v6": 0.5410452634096146, "step": 6337 }, { "epoch": 1.4617225225225225, "grad_norm": 29.18646812438965, "learning_rate": 1.0790321986805025e-06, "loss": 8.321104049682617, "loss_alignment": 0.70947265625, "loss_alignment_w": 0.354736328125, "loss_sft": 0.16382207162678242, "loss_total_v6": 0.5200690105557442, "step": 6338 }, { "epoch": 1.461953153153153, "grad_norm": 30.319040298461914, "learning_rate": 1.0777828953365471e-06, "loss": 9.224384307861328, "loss_alignment": 0.7412109375, "loss_alignment_w": 0.37060546875, "loss_sft": 0.2058117762207985, "loss_total_v6": 0.5765240490436554, "step": 6339 }, { "epoch": 1.4621837837837837, "grad_norm": 30.30957794189453, "learning_rate": 1.0765342282636416e-06, "loss": 8.69576644897461, "loss_alignment": 0.73388671875, "loss_alignment_w": 0.366943359375, "loss_sft": 0.1766641065478325, "loss_total_v6": 0.5434853807091713, "step": 6340 }, { "epoch": 1.4624144144144144, "grad_norm": 25.268545150756836, "learning_rate": 1.0752861976643465e-06, "loss": 8.705554962158203, "loss_alignment": 0.7626953125, "loss_alignment_w": 0.38134765625, "loss_sft": 0.1620934223756194, "loss_total_v6": 0.5440971925854683, "step": 6341 }, { "epoch": 1.462645045045045, "grad_norm": 30.010347366333008, "learning_rate": 1.0740388037411208e-06, "loss": 8.818771362304688, "loss_alignment": 0.77880859375, "loss_alignment_w": 0.389404296875, "loss_sft": 0.16213508415967226, "loss_total_v6": 0.5511731803417206, "step": 6342 }, { "epoch": 1.4628756756756758, "grad_norm": 27.88242530822754, "learning_rate": 1.0727920466963221e-06, "loss": 8.600163459777832, "loss_alignment": 0.7568359375, "loss_alignment_w": 0.37841796875, "loss_sft": 0.15825300477445126, "loss_total_v6": 0.5375101789832115, "step": 6343 }, { "epoch": 1.4631063063063063, "grad_norm": 27.137664794921875, "learning_rate": 1.0715459267321998e-06, "loss": 9.271597862243652, "loss_alignment": 0.74658203125, "loss_alignment_w": 0.373291015625, "loss_sft": 0.20468847267329693, "loss_total_v6": 0.5794748589396477, "step": 6344 }, { "epoch": 1.463336936936937, "grad_norm": 31.67778968811035, "learning_rate": 1.0703004440509023e-06, "loss": 9.04772663116455, "loss_alignment": 0.75732421875, "loss_alignment_w": 0.378662109375, "loss_sft": 0.18663767352700233, "loss_total_v6": 0.5654828920960426, "step": 6345 }, { "epoch": 1.4635675675675675, "grad_norm": 35.48328399658203, "learning_rate": 1.069055598854476e-06, "loss": 9.170829772949219, "loss_alignment": 0.763671875, "loss_alignment_w": 0.3818359375, "loss_sft": 0.18972345627844334, "loss_total_v6": 0.5731768310070038, "step": 6346 }, { "epoch": 1.4637981981981982, "grad_norm": 29.339895248413086, "learning_rate": 1.0678113913448596e-06, "loss": 9.215768814086914, "loss_alignment": 0.7470703125, "loss_alignment_w": 0.37353515625, "loss_sft": 0.2024808879941702, "loss_total_v6": 0.5759855285286903, "step": 6347 }, { "epoch": 1.4640288288288288, "grad_norm": 29.890958786010742, "learning_rate": 1.0665678217238933e-06, "loss": 9.173837661743164, "loss_alignment": 0.7265625, "loss_alignment_w": 0.36328125, "loss_sft": 0.20964113995432854, "loss_total_v6": 0.5733648911118507, "step": 6348 }, { "epoch": 1.4642594594594596, "grad_norm": 27.700428009033203, "learning_rate": 1.065324890193314e-06, "loss": 9.03053092956543, "loss_alignment": 0.79296875, "loss_alignment_w": 0.396484375, "loss_sft": 0.16897670086473227, "loss_total_v6": 0.5644082129001617, "step": 6349 }, { "epoch": 1.4644900900900901, "grad_norm": 32.15394973754883, "learning_rate": 1.0640825969547498e-06, "loss": 9.428600311279297, "loss_alignment": 0.79736328125, "loss_alignment_w": 0.398681640625, "loss_sft": 0.19080421887338161, "loss_total_v6": 0.5892875082790852, "step": 6350 }, { "epoch": 1.4647207207207207, "grad_norm": 27.017398834228516, "learning_rate": 1.0628409422097297e-06, "loss": 8.798852920532227, "loss_alignment": 0.75927734375, "loss_alignment_w": 0.379638671875, "loss_sft": 0.1716171447187662, "loss_total_v6": 0.549928292632103, "step": 6351 }, { "epoch": 1.4649513513513512, "grad_norm": 28.062286376953125, "learning_rate": 1.061599926159676e-06, "loss": 9.555619239807129, "loss_alignment": 0.775390625, "loss_alignment_w": 0.3876953125, "loss_sft": 0.20892053749412298, "loss_total_v6": 0.597226195037365, "step": 6352 }, { "epoch": 1.465181981981982, "grad_norm": 36.30968475341797, "learning_rate": 1.0603595490059116e-06, "loss": 8.961711883544922, "loss_alignment": 0.7529296875, "loss_alignment_w": 0.37646484375, "loss_sft": 0.18356584012508392, "loss_total_v6": 0.5601069778203964, "step": 6353 }, { "epoch": 1.4654126126126126, "grad_norm": 33.194366455078125, "learning_rate": 1.0591198109496514e-06, "loss": 8.802964210510254, "loss_alignment": 0.72607421875, "loss_alignment_w": 0.363037109375, "loss_sft": 0.18772798590362072, "loss_total_v6": 0.5501852706074715, "step": 6354 }, { "epoch": 1.4656432432432434, "grad_norm": 34.452938079833984, "learning_rate": 1.0578807121920103e-06, "loss": 9.2531099319458, "loss_alignment": 0.77783203125, "loss_alignment_w": 0.388916015625, "loss_sft": 0.18944909237325191, "loss_total_v6": 0.5783193409442902, "step": 6355 }, { "epoch": 1.465873873873874, "grad_norm": 29.47919464111328, "learning_rate": 1.0566422529339971e-06, "loss": 9.221111297607422, "loss_alignment": 0.79345703125, "loss_alignment_w": 0.396728515625, "loss_sft": 0.18117785453796387, "loss_total_v6": 0.5763194784522057, "step": 6356 }, { "epoch": 1.4661045045045045, "grad_norm": 33.334590911865234, "learning_rate": 1.0554044333765163e-06, "loss": 9.328380584716797, "loss_alignment": 0.75048828125, "loss_alignment_w": 0.375244140625, "loss_sft": 0.207733865827322, "loss_total_v6": 0.5830237716436386, "step": 6357 }, { "epoch": 1.466335135135135, "grad_norm": 33.70010757446289, "learning_rate": 1.0541672537203718e-06, "loss": 9.004018783569336, "loss_alignment": 0.744140625, "loss_alignment_w": 0.3720703125, "loss_sft": 0.189826431684196, "loss_total_v6": 0.5627512149512768, "step": 6358 }, { "epoch": 1.4665657657657658, "grad_norm": 30.37147331237793, "learning_rate": 1.0529307141662604e-06, "loss": 9.267938613891602, "loss_alignment": 0.77734375, "loss_alignment_w": 0.388671875, "loss_sft": 0.1907116435468197, "loss_total_v6": 0.5792461857199669, "step": 6359 }, { "epoch": 1.4667963963963964, "grad_norm": 35.76924133300781, "learning_rate": 1.0516948149147755e-06, "loss": 9.25826644897461, "loss_alignment": 0.76611328125, "loss_alignment_w": 0.383056640625, "loss_sft": 0.19555448926985264, "loss_total_v6": 0.5786416307091713, "step": 6360 }, { "epoch": 1.4670270270270271, "grad_norm": 26.56786346435547, "learning_rate": 1.05045955616641e-06, "loss": 8.86978530883789, "loss_alignment": 0.74951171875, "loss_alignment_w": 0.374755859375, "loss_sft": 0.1798498947173357, "loss_total_v6": 0.5543616153299809, "step": 6361 }, { "epoch": 1.4672576576576577, "grad_norm": 29.7852725982666, "learning_rate": 1.049224938121548e-06, "loss": 9.411998748779297, "loss_alignment": 0.77001953125, "loss_alignment_w": 0.385009765625, "loss_sft": 0.20400312263518572, "loss_total_v6": 0.5882499441504478, "step": 6362 }, { "epoch": 1.4674882882882883, "grad_norm": 28.424158096313477, "learning_rate": 1.0479909609804745e-06, "loss": 8.835899353027344, "loss_alignment": 0.78515625, "loss_alignment_w": 0.392578125, "loss_sft": 0.15879579354077578, "loss_total_v6": 0.5522436983883381, "step": 6363 }, { "epoch": 1.4677189189189188, "grad_norm": 28.000207901000977, "learning_rate": 1.0467576249433663e-06, "loss": 9.435853004455566, "loss_alignment": 0.7802734375, "loss_alignment_w": 0.39013671875, "loss_sft": 0.2001534029841423, "loss_total_v6": 0.5897407978773117, "step": 6364 }, { "epoch": 1.4679495495495496, "grad_norm": 31.113866806030273, "learning_rate": 1.0455249302102994e-06, "loss": 8.475005149841309, "loss_alignment": 0.73681640625, "loss_alignment_w": 0.368408203125, "loss_sft": 0.16108123399317265, "loss_total_v6": 0.5296878106892109, "step": 6365 }, { "epoch": 1.4681801801801801, "grad_norm": 30.176305770874023, "learning_rate": 1.044292876981241e-06, "loss": 9.277507781982422, "loss_alignment": 0.77880859375, "loss_alignment_w": 0.389404296875, "loss_sft": 0.19019579701125622, "loss_total_v6": 0.579844243824482, "step": 6366 }, { "epoch": 1.468410810810811, "grad_norm": 33.64862060546875, "learning_rate": 1.0430614654560606e-06, "loss": 9.28587532043457, "loss_alignment": 0.78125, "loss_alignment_w": 0.390625, "loss_sft": 0.19047460705041885, "loss_total_v6": 0.5803671926259995, "step": 6367 }, { "epoch": 1.4686414414414415, "grad_norm": 32.578125, "learning_rate": 1.0418306958345214e-06, "loss": 9.294239044189453, "loss_alignment": 0.7451171875, "loss_alignment_w": 0.37255859375, "loss_sft": 0.2079041376709938, "loss_total_v6": 0.5808899775147438, "step": 6368 }, { "epoch": 1.468872072072072, "grad_norm": 29.171422958374023, "learning_rate": 1.0406005683162802e-06, "loss": 8.869138717651367, "loss_alignment": 0.75146484375, "loss_alignment_w": 0.375732421875, "loss_sft": 0.17878705263137817, "loss_total_v6": 0.5543211214244366, "step": 6369 }, { "epoch": 1.4691027027027026, "grad_norm": 32.27128982543945, "learning_rate": 1.0393710831008913e-06, "loss": 8.939876556396484, "loss_alignment": 0.75537109375, "loss_alignment_w": 0.377685546875, "loss_sft": 0.18110246770083904, "loss_total_v6": 0.5587422549724579, "step": 6370 }, { "epoch": 1.4693333333333334, "grad_norm": 27.916534423828125, "learning_rate": 1.0381422403878032e-06, "loss": 8.637312889099121, "loss_alignment": 0.74169921875, "loss_alignment_w": 0.370849609375, "loss_sft": 0.16924187168478966, "loss_total_v6": 0.5398320555686951, "step": 6371 }, { "epoch": 1.469563963963964, "grad_norm": 33.599029541015625, "learning_rate": 1.036914040376364e-06, "loss": 9.095940589904785, "loss_alignment": 0.7509765625, "loss_alignment_w": 0.37548828125, "loss_sft": 0.19267231598496437, "loss_total_v6": 0.5684962794184685, "step": 6372 }, { "epoch": 1.4697945945945947, "grad_norm": 29.399959564208984, "learning_rate": 1.0356864832658126e-06, "loss": 9.140998840332031, "loss_alignment": 0.74267578125, "loss_alignment_w": 0.371337890625, "loss_sft": 0.20023397356271744, "loss_total_v6": 0.571312453597784, "step": 6373 }, { "epoch": 1.4700252252252253, "grad_norm": 27.581180572509766, "learning_rate": 1.0344595692552894e-06, "loss": 9.079885482788086, "loss_alignment": 0.765625, "loss_alignment_w": 0.3828125, "loss_sft": 0.18469557538628578, "loss_total_v6": 0.567492812871933, "step": 6374 }, { "epoch": 1.4702558558558558, "grad_norm": 36.227657318115234, "learning_rate": 1.0332332985438248e-06, "loss": 8.922441482543945, "loss_alignment": 0.73193359375, "loss_alignment_w": 0.365966796875, "loss_sft": 0.19246402569115162, "loss_total_v6": 0.557652622461319, "step": 6375 }, { "epoch": 1.4704864864864864, "grad_norm": 29.815933227539062, "learning_rate": 1.0320076713303467e-06, "loss": 9.030479431152344, "loss_alignment": 0.75341796875, "loss_alignment_w": 0.376708984375, "loss_sft": 0.18871831335127354, "loss_total_v6": 0.5644049644470215, "step": 6376 }, { "epoch": 1.4707171171171172, "grad_norm": 30.67189598083496, "learning_rate": 1.0307826878136818e-06, "loss": 9.155938148498535, "loss_alignment": 0.73974609375, "loss_alignment_w": 0.369873046875, "loss_sft": 0.202571464702487, "loss_total_v6": 0.5722461491823196, "step": 6377 }, { "epoch": 1.4709477477477477, "grad_norm": 29.257028579711914, "learning_rate": 1.0295583481925481e-06, "loss": 8.903623580932617, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.1681402251124382, "loss_total_v6": 0.5564764216542244, "step": 6378 }, { "epoch": 1.4711783783783785, "grad_norm": 29.19443702697754, "learning_rate": 1.0283346526655596e-06, "loss": 9.407480239868164, "loss_alignment": 0.76708984375, "loss_alignment_w": 0.383544921875, "loss_sft": 0.20451414585113525, "loss_total_v6": 0.5879675224423409, "step": 6379 }, { "epoch": 1.471409009009009, "grad_norm": 31.50822639465332, "learning_rate": 1.0271116014312293e-06, "loss": 8.709331512451172, "loss_alignment": 0.72509765625, "loss_alignment_w": 0.362548828125, "loss_sft": 0.18193697184324265, "loss_total_v6": 0.5443332232534885, "step": 6380 }, { "epoch": 1.4716396396396396, "grad_norm": 30.917280197143555, "learning_rate": 1.0258891946879612e-06, "loss": 8.718995094299316, "loss_alignment": 0.734375, "loss_alignment_w": 0.3671875, "loss_sft": 0.17794807814061642, "loss_total_v6": 0.5449372045695782, "step": 6381 }, { "epoch": 1.4718702702702702, "grad_norm": 32.71045684814453, "learning_rate": 1.0246674326340594e-06, "loss": 8.305105209350586, "loss_alignment": 0.70458984375, "loss_alignment_w": 0.352294921875, "loss_sft": 0.16703355871140957, "loss_total_v6": 0.5190690867602825, "step": 6382 }, { "epoch": 1.472100900900901, "grad_norm": 31.496606826782227, "learning_rate": 1.0234463154677193e-06, "loss": 9.537544250488281, "loss_alignment": 0.771484375, "loss_alignment_w": 0.3857421875, "loss_sft": 0.21053746156394482, "loss_total_v6": 0.5960965231060982, "step": 6383 }, { "epoch": 1.4723315315315315, "grad_norm": 33.196861267089844, "learning_rate": 1.022225843387034e-06, "loss": 8.866432189941406, "loss_alignment": 0.7548828125, "loss_alignment_w": 0.37744140625, "loss_sft": 0.17660379596054554, "loss_total_v6": 0.5541520118713379, "step": 6384 }, { "epoch": 1.4725621621621623, "grad_norm": 25.69391632080078, "learning_rate": 1.0210060165899894e-06, "loss": 8.969860076904297, "loss_alignment": 0.78125, "loss_alignment_w": 0.390625, "loss_sft": 0.17046421952545643, "loss_total_v6": 0.5606162026524544, "step": 6385 }, { "epoch": 1.4727927927927928, "grad_norm": 29.83038330078125, "learning_rate": 1.01978683527447e-06, "loss": 9.129743576049805, "loss_alignment": 0.74853515625, "loss_alignment_w": 0.374267578125, "loss_sft": 0.19501384161412716, "loss_total_v6": 0.5706089362502098, "step": 6386 }, { "epoch": 1.4730234234234234, "grad_norm": 32.9056510925293, "learning_rate": 1.0185682996382556e-06, "loss": 8.812355041503906, "loss_alignment": 0.755859375, "loss_alignment_w": 0.3779296875, "loss_sft": 0.17391063645482063, "loss_total_v6": 0.5507721975445747, "step": 6387 }, { "epoch": 1.473254054054054, "grad_norm": 32.25996780395508, "learning_rate": 1.0173504098790188e-06, "loss": 8.593790054321289, "loss_alignment": 0.7294921875, "loss_alignment_w": 0.36474609375, "loss_sft": 0.1723657799884677, "loss_total_v6": 0.5371118560433388, "step": 6388 }, { "epoch": 1.4734846846846847, "grad_norm": 31.927398681640625, "learning_rate": 1.0161331661943286e-06, "loss": 9.126859664916992, "loss_alignment": 0.75830078125, "loss_alignment_w": 0.379150390625, "loss_sft": 0.19068321958184242, "loss_total_v6": 0.570428691804409, "step": 6389 }, { "epoch": 1.4737153153153153, "grad_norm": 35.86326217651367, "learning_rate": 1.0149165687816481e-06, "loss": 9.184452056884766, "loss_alignment": 0.779296875, "loss_alignment_w": 0.3896484375, "loss_sft": 0.1828996855765581, "loss_total_v6": 0.5740282237529755, "step": 6390 }, { "epoch": 1.473945945945946, "grad_norm": 29.767444610595703, "learning_rate": 1.0137006178383358e-06, "loss": 9.112274169921875, "loss_alignment": 0.74462890625, "loss_alignment_w": 0.372314453125, "loss_sft": 0.19697380252182484, "loss_total_v6": 0.5695171356201172, "step": 6391 }, { "epoch": 1.4741765765765766, "grad_norm": 32.04582214355469, "learning_rate": 1.0124853135616475e-06, "loss": 8.684587478637695, "loss_alignment": 0.7255859375, "loss_alignment_w": 0.36279296875, "loss_sft": 0.1779643315821886, "loss_total_v6": 0.5427867211401463, "step": 6392 }, { "epoch": 1.4744072072072072, "grad_norm": 32.87609100341797, "learning_rate": 1.0112706561487335e-06, "loss": 8.759052276611328, "loss_alignment": 0.73193359375, "loss_alignment_w": 0.365966796875, "loss_sft": 0.18248106446117163, "loss_total_v6": 0.5474407933652401, "step": 6393 }, { "epoch": 1.4746378378378378, "grad_norm": 35.49005889892578, "learning_rate": 1.0100566457966377e-06, "loss": 9.049729347229004, "loss_alignment": 0.7734375, "loss_alignment_w": 0.38671875, "loss_sft": 0.1794081525877118, "loss_total_v6": 0.5656080916523933, "step": 6394 }, { "epoch": 1.4748684684684685, "grad_norm": 34.33622741699219, "learning_rate": 1.0088432827022986e-06, "loss": 9.526103973388672, "loss_alignment": 0.76708984375, "loss_alignment_w": 0.383544921875, "loss_sft": 0.21258431300520897, "loss_total_v6": 0.595381535589695, "step": 6395 }, { "epoch": 1.475099099099099, "grad_norm": 32.4424934387207, "learning_rate": 1.0076305670625508e-06, "loss": 8.660466194152832, "loss_alignment": 0.77880859375, "loss_alignment_w": 0.389404296875, "loss_sft": 0.1513865478336811, "loss_total_v6": 0.5412791222333908, "step": 6396 }, { "epoch": 1.4753297297297299, "grad_norm": 33.90140151977539, "learning_rate": 1.0064184990741254e-06, "loss": 9.720002174377441, "loss_alignment": 0.8037109375, "loss_alignment_w": 0.40185546875, "loss_sft": 0.20550730731338263, "loss_total_v6": 0.6075001060962677, "step": 6397 }, { "epoch": 1.4755603603603604, "grad_norm": 33.25774002075195, "learning_rate": 1.0052070789336444e-06, "loss": 8.726430892944336, "loss_alignment": 0.73095703125, "loss_alignment_w": 0.365478515625, "loss_sft": 0.18071682937443256, "loss_total_v6": 0.5454019084572792, "step": 6398 }, { "epoch": 1.475790990990991, "grad_norm": 31.767942428588867, "learning_rate": 1.0039963068376302e-06, "loss": 8.96531867980957, "loss_alignment": 0.73388671875, "loss_alignment_w": 0.366943359375, "loss_sft": 0.1931296493858099, "loss_total_v6": 0.5603324100375175, "step": 6399 }, { "epoch": 1.4760216216216215, "grad_norm": 34.18801498413086, "learning_rate": 1.0027861829824953e-06, "loss": 9.659892082214355, "loss_alignment": 0.86376953125, "loss_alignment_w": 0.431884765625, "loss_sft": 0.17187374737113714, "loss_total_v6": 0.6037432476878166, "step": 6400 }, { "epoch": 1.4762522522522523, "grad_norm": 33.87697982788086, "learning_rate": 1.0015767075645472e-06, "loss": 9.371932983398438, "loss_alignment": 0.8017578125, "loss_alignment_w": 0.40087890625, "loss_sft": 0.18523306120187044, "loss_total_v6": 0.5857458114624023, "step": 6401 }, { "epoch": 1.4764828828828829, "grad_norm": 28.541425704956055, "learning_rate": 1.0003678807799932e-06, "loss": 9.044172286987305, "loss_alignment": 0.7724609375, "loss_alignment_w": 0.38623046875, "loss_sft": 0.1792286131531, "loss_total_v6": 0.5652607306838036, "step": 6402 }, { "epoch": 1.4767135135135134, "grad_norm": 29.821212768554688, "learning_rate": 9.9915970282493e-07, "loss": 8.308292388916016, "loss_alignment": 0.70947265625, "loss_alignment_w": 0.354736328125, "loss_sft": 0.16442518681287766, "loss_total_v6": 0.5192683041095734, "step": 6403 }, { "epoch": 1.4769441441441442, "grad_norm": 29.905893325805664, "learning_rate": 9.9795217389535e-07, "loss": 9.365732192993164, "loss_alignment": 0.814453125, "loss_alignment_w": 0.4072265625, "loss_sft": 0.17865046672523022, "loss_total_v6": 0.585358239710331, "step": 6404 }, { "epoch": 1.4771747747747748, "grad_norm": 31.364437103271484, "learning_rate": 9.967452941871442e-07, "loss": 8.813652038574219, "loss_alignment": 0.7548828125, "loss_alignment_w": 0.37744140625, "loss_sft": 0.17325925640761852, "loss_total_v6": 0.5508532300591469, "step": 6405 }, { "epoch": 1.4774054054054053, "grad_norm": 32.71952438354492, "learning_rate": 9.955390638960916e-07, "loss": 9.421865463256836, "loss_alignment": 0.783203125, "loss_alignment_w": 0.3916015625, "loss_sft": 0.1970514226704836, "loss_total_v6": 0.5888665989041328, "step": 6406 }, { "epoch": 1.477636036036036, "grad_norm": 34.267478942871094, "learning_rate": 9.943334832178736e-07, "loss": 9.51297378540039, "loss_alignment": 0.7880859375, "loss_alignment_w": 0.39404296875, "loss_sft": 0.19984649121761322, "loss_total_v6": 0.5945608764886856, "step": 6407 }, { "epoch": 1.4778666666666667, "grad_norm": 32.780574798583984, "learning_rate": 9.931285523480605e-07, "loss": 9.108963012695312, "loss_alignment": 0.76220703125, "loss_alignment_w": 0.381103515625, "loss_sft": 0.18921378441154957, "loss_total_v6": 0.5693101957440376, "step": 6408 }, { "epoch": 1.4780972972972972, "grad_norm": 30.940061569213867, "learning_rate": 9.919242714821187e-07, "loss": 8.709556579589844, "loss_alignment": 0.73046875, "loss_alignment_w": 0.365234375, "loss_sft": 0.179509655572474, "loss_total_v6": 0.544347308576107, "step": 6409 }, { "epoch": 1.478327927927928, "grad_norm": 31.714555740356445, "learning_rate": 9.90720640815408e-07, "loss": 9.419595718383789, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.204599991440773, "loss_total_v6": 0.5887247398495674, "step": 6410 }, { "epoch": 1.4785585585585586, "grad_norm": 29.040573120117188, "learning_rate": 9.89517660543186e-07, "loss": 8.842439651489258, "loss_alignment": 0.77734375, "loss_alignment_w": 0.388671875, "loss_sft": 0.1650029420852661, "loss_total_v6": 0.5526524558663368, "step": 6411 }, { "epoch": 1.4787891891891891, "grad_norm": 31.27401351928711, "learning_rate": 9.883153308606035e-07, "loss": 9.253622055053711, "loss_alignment": 0.74267578125, "loss_alignment_w": 0.371337890625, "loss_sft": 0.20750180445611477, "loss_total_v6": 0.5783514082431793, "step": 6412 }, { "epoch": 1.47901981981982, "grad_norm": 31.502872467041016, "learning_rate": 9.87113651962705e-07, "loss": 8.620369911193848, "loss_alignment": 0.72314453125, "loss_alignment_w": 0.361572265625, "loss_sft": 0.17695672996342182, "loss_total_v6": 0.5387731119990349, "step": 6413 }, { "epoch": 1.4792504504504504, "grad_norm": 33.100582122802734, "learning_rate": 9.859126240444284e-07, "loss": 9.479161262512207, "loss_alignment": 0.802734375, "loss_alignment_w": 0.4013671875, "loss_sft": 0.19077516719698906, "loss_total_v6": 0.5924475714564323, "step": 6414 }, { "epoch": 1.479481081081081, "grad_norm": 28.792268753051758, "learning_rate": 9.847122473006066e-07, "loss": 9.219661712646484, "loss_alignment": 0.7626953125, "loss_alignment_w": 0.38134765625, "loss_sft": 0.1948506459593773, "loss_total_v6": 0.5762288421392441, "step": 6415 }, { "epoch": 1.4797117117117118, "grad_norm": 33.13207244873047, "learning_rate": 9.835125219259694e-07, "loss": 9.441530227661133, "loss_alignment": 0.76171875, "loss_alignment_w": 0.380859375, "loss_sft": 0.20876321382820606, "loss_total_v6": 0.590095616877079, "step": 6416 }, { "epoch": 1.4799423423423423, "grad_norm": 32.175418853759766, "learning_rate": 9.823134481151376e-07, "loss": 9.237143516540527, "loss_alignment": 0.7626953125, "loss_alignment_w": 0.38134765625, "loss_sft": 0.19615690410137177, "loss_total_v6": 0.577321469783783, "step": 6417 }, { "epoch": 1.480172972972973, "grad_norm": 32.65604019165039, "learning_rate": 9.81115026062629e-07, "loss": 9.209622383117676, "loss_alignment": 0.7373046875, "loss_alignment_w": 0.36865234375, "loss_sft": 0.2074067648500204, "loss_total_v6": 0.5756013691425323, "step": 6418 }, { "epoch": 1.4804036036036037, "grad_norm": 28.95018768310547, "learning_rate": 9.799172559628538e-07, "loss": 8.591608047485352, "loss_alignment": 0.7490234375, "loss_alignment_w": 0.37451171875, "loss_sft": 0.16116677969694138, "loss_total_v6": 0.536975484341383, "step": 6419 }, { "epoch": 1.4806342342342342, "grad_norm": 32.2278938293457, "learning_rate": 9.787201380101157e-07, "loss": 9.043556213378906, "loss_alignment": 0.7587890625, "loss_alignment_w": 0.37939453125, "loss_sft": 0.1861481722444296, "loss_total_v6": 0.5652222707867622, "step": 6420 }, { "epoch": 1.4808648648648648, "grad_norm": 31.637073516845703, "learning_rate": 9.775236723986159e-07, "loss": 8.677627563476562, "loss_alignment": 0.73681640625, "loss_alignment_w": 0.368408203125, "loss_sft": 0.17258549109101295, "loss_total_v6": 0.5423517152667046, "step": 6421 }, { "epoch": 1.4810954954954956, "grad_norm": 37.07212448120117, "learning_rate": 9.763278593224462e-07, "loss": 9.51228141784668, "loss_alignment": 0.771484375, "loss_alignment_w": 0.3857421875, "loss_sft": 0.20848551392555237, "loss_total_v6": 0.5945176184177399, "step": 6422 }, { "epoch": 1.4813261261261261, "grad_norm": 31.04726219177246, "learning_rate": 9.75132698975596e-07, "loss": 9.24907112121582, "loss_alignment": 0.74609375, "loss_alignment_w": 0.373046875, "loss_sft": 0.2041350193321705, "loss_total_v6": 0.5780669152736664, "step": 6423 }, { "epoch": 1.4815567567567567, "grad_norm": 33.8231201171875, "learning_rate": 9.739381915519458e-07, "loss": 9.112106323242188, "loss_alignment": 0.7568359375, "loss_alignment_w": 0.37841796875, "loss_sft": 0.19235512241721153, "loss_total_v6": 0.5695066079497337, "step": 6424 }, { "epoch": 1.4817873873873875, "grad_norm": 35.44392013549805, "learning_rate": 9.7274433724527e-07, "loss": 9.435193061828613, "loss_alignment": 0.7626953125, "loss_alignment_w": 0.38134765625, "loss_sft": 0.20792469196021557, "loss_total_v6": 0.5896995887160301, "step": 6425 }, { "epoch": 1.482018018018018, "grad_norm": 29.222105026245117, "learning_rate": 9.715511362492413e-07, "loss": 9.296150207519531, "loss_alignment": 0.7919921875, "loss_alignment_w": 0.39599609375, "loss_sft": 0.1856694184243679, "loss_total_v6": 0.5810093805193901, "step": 6426 }, { "epoch": 1.4822486486486486, "grad_norm": 31.481077194213867, "learning_rate": 9.70358588757422e-07, "loss": 9.220240592956543, "loss_alignment": 0.76220703125, "loss_alignment_w": 0.381103515625, "loss_sft": 0.1944290865212679, "loss_total_v6": 0.5762650296092033, "step": 6427 }, { "epoch": 1.4824792792792794, "grad_norm": 29.459060668945312, "learning_rate": 9.691666949632684e-07, "loss": 9.223432540893555, "loss_alignment": 0.7548828125, "loss_alignment_w": 0.37744140625, "loss_sft": 0.1993283536285162, "loss_total_v6": 0.5764645859599113, "step": 6428 }, { "epoch": 1.48270990990991, "grad_norm": 28.281675338745117, "learning_rate": 9.679754550601344e-07, "loss": 8.700834274291992, "loss_alignment": 0.740234375, "loss_alignment_w": 0.3701171875, "loss_sft": 0.17467675730586052, "loss_total_v6": 0.5438021160662174, "step": 6429 }, { "epoch": 1.4829405405405405, "grad_norm": 30.236942291259766, "learning_rate": 9.667848692412645e-07, "loss": 8.731956481933594, "loss_alignment": 0.74951171875, "loss_alignment_w": 0.374755859375, "loss_sft": 0.1703352825716138, "loss_total_v6": 0.545747272670269, "step": 6430 }, { "epoch": 1.4831711711711713, "grad_norm": 31.2298641204834, "learning_rate": 9.655949376997996e-07, "loss": 8.92129135131836, "loss_alignment": 0.74658203125, "loss_alignment_w": 0.373291015625, "loss_sft": 0.1848848294466734, "loss_total_v6": 0.557580754160881, "step": 6431 }, { "epoch": 1.4834018018018018, "grad_norm": 28.3850040435791, "learning_rate": 9.644056606287727e-07, "loss": 9.234649658203125, "loss_alignment": 0.76171875, "loss_alignment_w": 0.380859375, "loss_sft": 0.19645880535244942, "loss_total_v6": 0.5771655961871147, "step": 6432 }, { "epoch": 1.4836324324324324, "grad_norm": 31.908802032470703, "learning_rate": 9.632170382211102e-07, "loss": 8.867191314697266, "loss_alignment": 0.7880859375, "loss_alignment_w": 0.39404296875, "loss_sft": 0.1602785624563694, "loss_total_v6": 0.5541994571685791, "step": 6433 }, { "epoch": 1.4838630630630631, "grad_norm": 31.389217376708984, "learning_rate": 9.620290706696323e-07, "loss": 9.190145492553711, "loss_alignment": 0.76904296875, "loss_alignment_w": 0.384521484375, "loss_sft": 0.1896642129868269, "loss_total_v6": 0.5743840485811234, "step": 6434 }, { "epoch": 1.4840936936936937, "grad_norm": 28.816722869873047, "learning_rate": 9.608417581670555e-07, "loss": 9.249405860900879, "loss_alignment": 0.7421875, "loss_alignment_w": 0.37109375, "loss_sft": 0.20833691582083702, "loss_total_v6": 0.5780878812074661, "step": 6435 }, { "epoch": 1.4843243243243243, "grad_norm": 35.78972244262695, "learning_rate": 9.596551009059884e-07, "loss": 9.141616821289062, "loss_alignment": 0.78662109375, "loss_alignment_w": 0.393310546875, "loss_sft": 0.17726234160363674, "loss_total_v6": 0.5713510811328888, "step": 6436 }, { "epoch": 1.4845549549549548, "grad_norm": 34.69001770019531, "learning_rate": 9.584690990789336e-07, "loss": 9.069156646728516, "loss_alignment": 0.7958984375, "loss_alignment_w": 0.39794921875, "loss_sft": 0.16864421498030424, "loss_total_v6": 0.5668223202228546, "step": 6437 }, { "epoch": 1.4847855855855856, "grad_norm": 30.58083152770996, "learning_rate": 9.572837528782858e-07, "loss": 9.139291763305664, "loss_alignment": 0.748046875, "loss_alignment_w": 0.3740234375, "loss_sft": 0.19797573424875736, "loss_total_v6": 0.5712057128548622, "step": 6438 }, { "epoch": 1.4850162162162162, "grad_norm": 30.812759399414062, "learning_rate": 9.56099062496335e-07, "loss": 8.896221160888672, "loss_alignment": 0.72216796875, "loss_alignment_w": 0.361083984375, "loss_sft": 0.19373963680118322, "loss_total_v6": 0.5560138076543808, "step": 6439 }, { "epoch": 1.485246846846847, "grad_norm": 35.397369384765625, "learning_rate": 9.549150281252633e-07, "loss": 8.913174629211426, "loss_alignment": 0.78955078125, "loss_alignment_w": 0.394775390625, "loss_sft": 0.16162661649286747, "loss_total_v6": 0.5570734068751335, "step": 6440 }, { "epoch": 1.4854774774774775, "grad_norm": 30.064058303833008, "learning_rate": 9.537316499571487e-07, "loss": 8.507488250732422, "loss_alignment": 0.73876953125, "loss_alignment_w": 0.369384765625, "loss_sft": 0.16224167123436928, "loss_total_v6": 0.5317179895937443, "step": 6441 }, { "epoch": 1.485708108108108, "grad_norm": 33.37664794921875, "learning_rate": 9.525489281839622e-07, "loss": 9.133127212524414, "loss_alignment": 0.7666015625, "loss_alignment_w": 0.38330078125, "loss_sft": 0.18719923682510853, "loss_total_v6": 0.5708204433321953, "step": 6442 }, { "epoch": 1.4859387387387386, "grad_norm": 33.064979553222656, "learning_rate": 9.51366862997567e-07, "loss": 9.179513931274414, "loss_alignment": 0.7666015625, "loss_alignment_w": 0.38330078125, "loss_sft": 0.1900220401585102, "loss_total_v6": 0.5737195536494255, "step": 6443 }, { "epoch": 1.4861693693693694, "grad_norm": 33.36014175415039, "learning_rate": 9.501854545897204e-07, "loss": 9.178977012634277, "loss_alignment": 0.74072265625, "loss_alignment_w": 0.370361328125, "loss_sft": 0.20385880023241043, "loss_total_v6": 0.5736860632896423, "step": 6444 }, { "epoch": 1.4864, "grad_norm": 31.928783416748047, "learning_rate": 9.490047031520716e-07, "loss": 9.47870922088623, "loss_alignment": 0.76025390625, "loss_alignment_w": 0.380126953125, "loss_sft": 0.21247546933591366, "loss_total_v6": 0.5924193263053894, "step": 6445 }, { "epoch": 1.4866306306306307, "grad_norm": 27.587129592895508, "learning_rate": 9.478246088761672e-07, "loss": 8.915602684020996, "loss_alignment": 0.7470703125, "loss_alignment_w": 0.37353515625, "loss_sft": 0.18336958438158035, "loss_total_v6": 0.5572251677513123, "step": 6446 }, { "epoch": 1.4868612612612613, "grad_norm": 40.96836471557617, "learning_rate": 9.466451719534431e-07, "loss": 9.293944358825684, "loss_alignment": 0.79150390625, "loss_alignment_w": 0.395751953125, "loss_sft": 0.18482965789735317, "loss_total_v6": 0.5808715224266052, "step": 6447 }, { "epoch": 1.4870918918918918, "grad_norm": 33.43051528930664, "learning_rate": 9.454663925752317e-07, "loss": 8.837648391723633, "loss_alignment": 0.736328125, "loss_alignment_w": 0.3681640625, "loss_sft": 0.1838532155379653, "loss_total_v6": 0.5523529686033726, "step": 6448 }, { "epoch": 1.4873225225225224, "grad_norm": 27.61272430419922, "learning_rate": 9.442882709327573e-07, "loss": 9.237194061279297, "loss_alignment": 0.78759765625, "loss_alignment_w": 0.393798828125, "loss_sft": 0.1824271623045206, "loss_total_v6": 0.5773246213793755, "step": 6449 }, { "epoch": 1.4875531531531532, "grad_norm": 31.50642967224121, "learning_rate": 9.431108072171346e-07, "loss": 8.920604705810547, "loss_alignment": 0.79052734375, "loss_alignment_w": 0.395263671875, "loss_sft": 0.16254876554012299, "loss_total_v6": 0.5575377866625786, "step": 6450 }, { "epoch": 1.4877837837837837, "grad_norm": 31.72177505493164, "learning_rate": 9.419340016193785e-07, "loss": 9.218097686767578, "loss_alignment": 0.76513671875, "loss_alignment_w": 0.382568359375, "loss_sft": 0.19330330565571785, "loss_total_v6": 0.5761310756206512, "step": 6451 }, { "epoch": 1.4880144144144145, "grad_norm": 30.141252517700195, "learning_rate": 9.407578543303913e-07, "loss": 9.20806884765625, "loss_alignment": 0.7509765625, "loss_alignment_w": 0.37548828125, "loss_sft": 0.19989398308098316, "loss_total_v6": 0.5755043402314186, "step": 6452 }, { "epoch": 1.488245045045045, "grad_norm": 29.663801193237305, "learning_rate": 9.395823655409686e-07, "loss": 8.650735855102539, "loss_alignment": 0.73388671875, "loss_alignment_w": 0.366943359375, "loss_sft": 0.17323934100568295, "loss_total_v6": 0.5406709797680378, "step": 6453 }, { "epoch": 1.4884756756756756, "grad_norm": 32.245548248291016, "learning_rate": 9.384075354418043e-07, "loss": 8.699777603149414, "loss_alignment": 0.72607421875, "loss_alignment_w": 0.363037109375, "loss_sft": 0.18152295891195536, "loss_total_v6": 0.5437360927462578, "step": 6454 }, { "epoch": 1.4887063063063062, "grad_norm": 28.723880767822266, "learning_rate": 9.372333642234787e-07, "loss": 9.118978500366211, "loss_alignment": 0.744140625, "loss_alignment_w": 0.3720703125, "loss_sft": 0.19778959639370441, "loss_total_v6": 0.5699362084269524, "step": 6455 }, { "epoch": 1.488936936936937, "grad_norm": 34.922767639160156, "learning_rate": 9.360598520764713e-07, "loss": 9.22769546508789, "loss_alignment": 0.78271484375, "loss_alignment_w": 0.391357421875, "loss_sft": 0.18554138764739037, "loss_total_v6": 0.5767309665679932, "step": 6456 }, { "epoch": 1.4891675675675675, "grad_norm": 33.139198303222656, "learning_rate": 9.348869991911508e-07, "loss": 8.901956558227539, "loss_alignment": 0.73046875, "loss_alignment_w": 0.365234375, "loss_sft": 0.19000870920717716, "loss_total_v6": 0.55637226998806, "step": 6457 }, { "epoch": 1.4893981981981983, "grad_norm": 30.191905975341797, "learning_rate": 9.3371480575778e-07, "loss": 9.066614151000977, "loss_alignment": 0.73095703125, "loss_alignment_w": 0.365478515625, "loss_sft": 0.20058975275605917, "loss_total_v6": 0.566663358360529, "step": 6458 }, { "epoch": 1.4896288288288289, "grad_norm": 35.29780197143555, "learning_rate": 9.32543271966514e-07, "loss": 9.685487747192383, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.2167321853339672, "loss_total_v6": 0.6053429991006851, "step": 6459 }, { "epoch": 1.4898594594594594, "grad_norm": 35.232765197753906, "learning_rate": 9.313723980074018e-07, "loss": 9.465288162231445, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.20382415130734444, "loss_total_v6": 0.5915804989635944, "step": 6460 }, { "epoch": 1.49009009009009, "grad_norm": 27.921018600463867, "learning_rate": 9.302021840703879e-07, "loss": 9.373970031738281, "loss_alignment": 0.7705078125, "loss_alignment_w": 0.38525390625, "loss_sft": 0.20193150639533997, "loss_total_v6": 0.5858731493353844, "step": 6461 }, { "epoch": 1.4903207207207207, "grad_norm": 30.8655948638916, "learning_rate": 9.290326303453045e-07, "loss": 9.770263671875, "loss_alignment": 0.8076171875, "loss_alignment_w": 0.40380859375, "loss_sft": 0.20751952938735485, "loss_total_v6": 0.6106414794921875, "step": 6462 }, { "epoch": 1.4905513513513513, "grad_norm": 28.950227737426758, "learning_rate": 9.278637370218807e-07, "loss": 8.481290817260742, "loss_alignment": 0.7353515625, "loss_alignment_w": 0.36767578125, "loss_sft": 0.16199289169162512, "loss_total_v6": 0.5300806611776352, "step": 6463 }, { "epoch": 1.490781981981982, "grad_norm": 32.09400177001953, "learning_rate": 9.266955042897357e-07, "loss": 8.719075202941895, "loss_alignment": 0.73583984375, "loss_alignment_w": 0.367919921875, "loss_sft": 0.1761678010225296, "loss_total_v6": 0.5449422188103199, "step": 6464 }, { "epoch": 1.4910126126126126, "grad_norm": 31.657337188720703, "learning_rate": 9.255279323383825e-07, "loss": 8.924477577209473, "loss_alignment": 0.7021484375, "loss_alignment_w": 0.35107421875, "loss_sft": 0.20620210841298103, "loss_total_v6": 0.557779848575592, "step": 6465 }, { "epoch": 1.4912432432432432, "grad_norm": 35.32645034790039, "learning_rate": 9.243610213572285e-07, "loss": 8.873773574829102, "loss_alignment": 0.775390625, "loss_alignment_w": 0.3876953125, "loss_sft": 0.16619840078055859, "loss_total_v6": 0.5546108596026897, "step": 6466 }, { "epoch": 1.4914738738738738, "grad_norm": 32.156646728515625, "learning_rate": 9.231947715355738e-07, "loss": 8.706478118896484, "loss_alignment": 0.76220703125, "loss_alignment_w": 0.381103515625, "loss_sft": 0.16338708717375994, "loss_total_v6": 0.5441549085080624, "step": 6467 }, { "epoch": 1.4917045045045045, "grad_norm": 31.31900978088379, "learning_rate": 9.220291830626082e-07, "loss": 8.805353164672852, "loss_alignment": 0.75244140625, "loss_alignment_w": 0.376220703125, "loss_sft": 0.17441903799772263, "loss_total_v6": 0.5503345504403114, "step": 6468 }, { "epoch": 1.491935135135135, "grad_norm": 32.066925048828125, "learning_rate": 9.208642561274162e-07, "loss": 8.707918167114258, "loss_alignment": 0.74755859375, "loss_alignment_w": 0.373779296875, "loss_sft": 0.1703893020749092, "loss_total_v6": 0.5442448891699314, "step": 6469 }, { "epoch": 1.4921657657657659, "grad_norm": 32.994544982910156, "learning_rate": 9.196999909189764e-07, "loss": 9.520235061645508, "loss_alignment": 0.802734375, "loss_alignment_w": 0.4013671875, "loss_sft": 0.19283880852162838, "loss_total_v6": 0.5950146690011024, "step": 6470 }, { "epoch": 1.4923963963963964, "grad_norm": 30.704206466674805, "learning_rate": 9.185363876261572e-07, "loss": 9.432180404663086, "loss_alignment": 0.78076171875, "loss_alignment_w": 0.390380859375, "loss_sft": 0.19920673593878746, "loss_total_v6": 0.5895112827420235, "step": 6471 }, { "epoch": 1.492627027027027, "grad_norm": 35.95928192138672, "learning_rate": 9.173734464377204e-07, "loss": 9.489654541015625, "loss_alignment": 0.80126953125, "loss_alignment_w": 0.400634765625, "loss_sft": 0.19205662049353123, "loss_total_v6": 0.5931033566594124, "step": 6472 }, { "epoch": 1.4928576576576575, "grad_norm": 34.19660568237305, "learning_rate": 9.162111675423235e-07, "loss": 8.840023040771484, "loss_alignment": 0.7548828125, "loss_alignment_w": 0.37744140625, "loss_sft": 0.1752889472991228, "loss_total_v6": 0.552501454949379, "step": 6473 }, { "epoch": 1.4930882882882883, "grad_norm": 28.939058303833008, "learning_rate": 9.150495511285112e-07, "loss": 8.71083927154541, "loss_alignment": 0.7041015625, "loss_alignment_w": 0.35205078125, "loss_sft": 0.19260550290346146, "loss_total_v6": 0.5444274172186852, "step": 6474 }, { "epoch": 1.4933189189189189, "grad_norm": 33.897369384765625, "learning_rate": 9.138885973847262e-07, "loss": 9.261472702026367, "loss_alignment": 0.75732421875, "loss_alignment_w": 0.378662109375, "loss_sft": 0.2003477942198515, "loss_total_v6": 0.5788420513272285, "step": 6475 }, { "epoch": 1.4935495495495497, "grad_norm": 29.831464767456055, "learning_rate": 9.127283064992998e-07, "loss": 9.01078987121582, "loss_alignment": 0.75634765625, "loss_alignment_w": 0.378173828125, "loss_sft": 0.18416130356490612, "loss_total_v6": 0.5631743669509888, "step": 6476 }, { "epoch": 1.4937801801801802, "grad_norm": 28.785240173339844, "learning_rate": 9.115686786604572e-07, "loss": 9.122071266174316, "loss_alignment": 0.76611328125, "loss_alignment_w": 0.383056640625, "loss_sft": 0.18606576696038246, "loss_total_v6": 0.5701294839382172, "step": 6477 }, { "epoch": 1.4940108108108108, "grad_norm": 30.708513259887695, "learning_rate": 9.104097140563139e-07, "loss": 9.131622314453125, "loss_alignment": 0.76220703125, "loss_alignment_w": 0.381103515625, "loss_sft": 0.19044680893421173, "loss_total_v6": 0.5707263723015785, "step": 6478 }, { "epoch": 1.4942414414414413, "grad_norm": 29.19276237487793, "learning_rate": 9.09251412874882e-07, "loss": 9.552977561950684, "loss_alignment": 0.76806640625, "loss_alignment_w": 0.384033203125, "loss_sft": 0.21305841486901045, "loss_total_v6": 0.5970610901713371, "step": 6479 }, { "epoch": 1.494472072072072, "grad_norm": 36.39649200439453, "learning_rate": 9.080937753040647e-07, "loss": 9.175735473632812, "loss_alignment": 0.77392578125, "loss_alignment_w": 0.386962890625, "loss_sft": 0.18803118355572224, "loss_total_v6": 0.5734834671020508, "step": 6480 }, { "epoch": 1.4947027027027027, "grad_norm": 30.834651947021484, "learning_rate": 9.069368015316554e-07, "loss": 8.770552635192871, "loss_alignment": 0.734375, "loss_alignment_w": 0.3671875, "loss_sft": 0.18039220478385687, "loss_total_v6": 0.5481595322489738, "step": 6481 }, { "epoch": 1.4949333333333334, "grad_norm": 29.253862380981445, "learning_rate": 9.057804917453411e-07, "loss": 9.096517562866211, "loss_alignment": 0.74755859375, "loss_alignment_w": 0.373779296875, "loss_sft": 0.19467675872147083, "loss_total_v6": 0.5685323625802994, "step": 6482 }, { "epoch": 1.495163963963964, "grad_norm": 29.08411979675293, "learning_rate": 9.046248461327006e-07, "loss": 8.719385147094727, "loss_alignment": 0.7548828125, "loss_alignment_w": 0.37744140625, "loss_sft": 0.16599428746849298, "loss_total_v6": 0.5449615940451622, "step": 6483 }, { "epoch": 1.4953945945945946, "grad_norm": 34.04157638549805, "learning_rate": 9.034698648812045e-07, "loss": 9.19178295135498, "loss_alignment": 0.75830078125, "loss_alignment_w": 0.379150390625, "loss_sft": 0.19481726735830307, "loss_total_v6": 0.5744864642620087, "step": 6484 }, { "epoch": 1.4956252252252251, "grad_norm": 30.292984008789062, "learning_rate": 9.023155481782175e-07, "loss": 9.281723976135254, "loss_alignment": 0.7607421875, "loss_alignment_w": 0.38037109375, "loss_sft": 0.19949252158403397, "loss_total_v6": 0.5801077410578728, "step": 6485 }, { "epoch": 1.495855855855856, "grad_norm": 30.482816696166992, "learning_rate": 9.01161896210997e-07, "loss": 9.686089515686035, "loss_alignment": 0.7578125, "loss_alignment_w": 0.37890625, "loss_sft": 0.22619968093931675, "loss_total_v6": 0.6053805723786354, "step": 6486 }, { "epoch": 1.4960864864864865, "grad_norm": 28.770401000976562, "learning_rate": 9.000089091666897e-07, "loss": 8.95336627960205, "loss_alignment": 0.75146484375, "loss_alignment_w": 0.375732421875, "loss_sft": 0.1833036821335554, "loss_total_v6": 0.5595853999257088, "step": 6487 }, { "epoch": 1.4963171171171172, "grad_norm": 31.990493774414062, "learning_rate": 8.988565872323362e-07, "loss": 8.807392120361328, "loss_alignment": 0.73193359375, "loss_alignment_w": 0.365966796875, "loss_sft": 0.1844799816608429, "loss_total_v6": 0.5504620559513569, "step": 6488 }, { "epoch": 1.4965477477477478, "grad_norm": 31.333934783935547, "learning_rate": 8.977049305948671e-07, "loss": 9.05459213256836, "loss_alignment": 0.7333984375, "loss_alignment_w": 0.36669921875, "loss_sft": 0.1998231541365385, "loss_total_v6": 0.5659119859337807, "step": 6489 }, { "epoch": 1.4967783783783783, "grad_norm": 26.455007553100586, "learning_rate": 8.965539394411088e-07, "loss": 8.803716659545898, "loss_alignment": 0.72900390625, "loss_alignment_w": 0.364501953125, "loss_sft": 0.18518108781427145, "loss_total_v6": 0.5502323321998119, "step": 6490 }, { "epoch": 1.497009009009009, "grad_norm": 26.795534133911133, "learning_rate": 8.954036139577765e-07, "loss": 9.129389762878418, "loss_alignment": 0.74755859375, "loss_alignment_w": 0.373779296875, "loss_sft": 0.1972958417609334, "loss_total_v6": 0.5705868415534496, "step": 6491 }, { "epoch": 1.4972396396396397, "grad_norm": 28.238323211669922, "learning_rate": 8.942539543314799e-07, "loss": 9.100963592529297, "loss_alignment": 0.75, "loss_alignment_w": 0.375, "loss_sft": 0.19335239753127098, "loss_total_v6": 0.5688101649284363, "step": 6492 }, { "epoch": 1.4974702702702702, "grad_norm": 28.217117309570312, "learning_rate": 8.931049607487191e-07, "loss": 9.205909729003906, "loss_alignment": 0.7548828125, "loss_alignment_w": 0.37744140625, "loss_sft": 0.198050020262599, "loss_total_v6": 0.575369343161583, "step": 6493 }, { "epoch": 1.497700900900901, "grad_norm": 29.01219940185547, "learning_rate": 8.919566333958846e-07, "loss": 9.291345596313477, "loss_alignment": 0.7822265625, "loss_alignment_w": 0.39111328125, "loss_sft": 0.1895042285323143, "loss_total_v6": 0.5807090699672699, "step": 6494 }, { "epoch": 1.4979315315315316, "grad_norm": 33.417388916015625, "learning_rate": 8.908089724592639e-07, "loss": 9.297149658203125, "loss_alignment": 0.7529296875, "loss_alignment_w": 0.37646484375, "loss_sft": 0.20489693619310856, "loss_total_v6": 0.5810718685388565, "step": 6495 }, { "epoch": 1.4981621621621621, "grad_norm": 32.52310562133789, "learning_rate": 8.896619781250309e-07, "loss": 9.403383255004883, "loss_alignment": 0.76220703125, "loss_alignment_w": 0.381103515625, "loss_sft": 0.207294549793005, "loss_total_v6": 0.5877114236354828, "step": 6496 }, { "epoch": 1.4983927927927927, "grad_norm": 31.88824462890625, "learning_rate": 8.885156505792531e-07, "loss": 9.773664474487305, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.22376910038292408, "loss_total_v6": 0.6108540520071983, "step": 6497 }, { "epoch": 1.4986234234234235, "grad_norm": 32.536827087402344, "learning_rate": 8.87369990007893e-07, "loss": 9.271934509277344, "loss_alignment": 0.763671875, "loss_alignment_w": 0.3818359375, "loss_sft": 0.19764474034309387, "loss_total_v6": 0.5794959217309952, "step": 6498 }, { "epoch": 1.498854054054054, "grad_norm": 33.46738815307617, "learning_rate": 8.862249965967995e-07, "loss": 8.939168930053711, "loss_alignment": 0.7451171875, "loss_alignment_w": 0.37255859375, "loss_sft": 0.1860173847526312, "loss_total_v6": 0.5586980655789375, "step": 6499 }, { "epoch": 1.4990846846846848, "grad_norm": 35.462364196777344, "learning_rate": 8.850806705317183e-07, "loss": 9.778459548950195, "loss_alignment": 0.84619140625, "loss_alignment_w": 0.423095703125, "loss_sft": 0.18775282427668571, "loss_total_v6": 0.6111536994576454, "step": 6500 }, { "epoch": 1.4990846846846848, "eval_loss": 0.8608778715133667, "eval_loss_alignment": 0.7771453160673516, "eval_loss_alignment_w": 0.3885726580336758, "eval_loss_contrast": 0.0, "eval_loss_highlevel": 0.0, "eval_loss_sft": 0.4723052413428211, "eval_loss_total_v6": 0.8608779017919819, "eval_loss_wm_recon": 0.0, "eval_loss_wm_recon_w": 0.0, "eval_runtime": 341.5424, "eval_samples_per_second": 5.109, "eval_steps_per_second": 0.641, "step": 6500 }, { "epoch": 1.4993153153153154, "grad_norm": 30.506555557250977, "learning_rate": 8.839370119982837e-07, "loss": 9.563909530639648, "loss_alignment": 0.77880859375, "loss_alignment_w": 0.389404296875, "loss_sft": 0.20815694332122803, "loss_total_v6": 0.5977443307638168, "step": 6501 }, { "epoch": 1.499545945945946, "grad_norm": 35.151912689208984, "learning_rate": 8.827940211820224e-07, "loss": 8.8345365524292, "loss_alignment": 0.77099609375, "loss_alignment_w": 0.385498046875, "loss_sft": 0.16612642910331488, "loss_total_v6": 0.552158534526825, "step": 6502 }, { "epoch": 1.4997765765765765, "grad_norm": 32.97955322265625, "learning_rate": 8.816516982683521e-07, "loss": 8.828876495361328, "loss_alignment": 0.7666015625, "loss_alignment_w": 0.38330078125, "loss_sft": 0.1676495373249054, "loss_total_v6": 0.5518048033118248, "step": 6503 }, { "epoch": 1.5000072072072073, "grad_norm": 32.530643463134766, "learning_rate": 8.805100434425845e-07, "loss": 8.644362449645996, "loss_alignment": 0.7666015625, "loss_alignment_w": 0.38330078125, "loss_sft": 0.1577195255085826, "loss_total_v6": 0.5402726382017136, "step": 6504 }, { "epoch": 1.5002378378378378, "grad_norm": 30.994760513305664, "learning_rate": 8.793690568899216e-07, "loss": 8.556787490844727, "loss_alignment": 0.73046875, "loss_alignment_w": 0.365234375, "loss_sft": 0.16892400570213795, "loss_total_v6": 0.5347992442548275, "step": 6505 }, { "epoch": 1.5004684684684686, "grad_norm": 32.55122756958008, "learning_rate": 8.782287387954563e-07, "loss": 8.875969886779785, "loss_alignment": 0.7421875, "loss_alignment_w": 0.37109375, "loss_sft": 0.18327287770807743, "loss_total_v6": 0.5547481030225754, "step": 6506 }, { "epoch": 1.5006990990990992, "grad_norm": 33.92348861694336, "learning_rate": 8.770890893441741e-07, "loss": 9.236518859863281, "loss_alignment": 0.71484375, "loss_alignment_w": 0.357421875, "loss_sft": 0.2193112149834633, "loss_total_v6": 0.5772824212908745, "step": 6507 }, { "epoch": 1.5009297297297297, "grad_norm": 33.98495864868164, "learning_rate": 8.75950108720951e-07, "loss": 9.373058319091797, "loss_alignment": 0.77783203125, "loss_alignment_w": 0.388916015625, "loss_sft": 0.19688484631478786, "loss_total_v6": 0.5858161151409149, "step": 6508 }, { "epoch": 1.5011603603603603, "grad_norm": 28.962711334228516, "learning_rate": 8.748117971105536e-07, "loss": 9.006978988647461, "loss_alignment": 0.740234375, "loss_alignment_w": 0.3701171875, "loss_sft": 0.19321571849286556, "loss_total_v6": 0.562936183065176, "step": 6509 }, { "epoch": 1.5013909909909908, "grad_norm": 31.500757217407227, "learning_rate": 8.73674154697643e-07, "loss": 9.325607299804688, "loss_alignment": 0.783203125, "loss_alignment_w": 0.3916015625, "loss_sft": 0.19182870909571648, "loss_total_v6": 0.5828504264354706, "step": 6510 }, { "epoch": 1.5016216216216216, "grad_norm": 28.7293701171875, "learning_rate": 8.72537181666771e-07, "loss": 8.835297584533691, "loss_alignment": 0.71435546875, "loss_alignment_w": 0.357177734375, "loss_sft": 0.1945706307888031, "loss_total_v6": 0.5522061139345169, "step": 6511 }, { "epoch": 1.5018522522522524, "grad_norm": 28.793161392211914, "learning_rate": 8.714008782023797e-07, "loss": 9.53242015838623, "loss_alignment": 0.76611328125, "loss_alignment_w": 0.383056640625, "loss_sft": 0.21320791728794575, "loss_total_v6": 0.5957762598991394, "step": 6512 }, { "epoch": 1.502082882882883, "grad_norm": 31.50783348083496, "learning_rate": 8.702652444888021e-07, "loss": 9.052387237548828, "loss_alignment": 0.748046875, "loss_alignment_w": 0.3740234375, "loss_sft": 0.19208645075559616, "loss_total_v6": 0.5657742023468018, "step": 6513 }, { "epoch": 1.5023135135135135, "grad_norm": 30.018268585205078, "learning_rate": 8.691302807102619e-07, "loss": 9.570878982543945, "loss_alignment": 0.78466796875, "loss_alignment_w": 0.392333984375, "loss_sft": 0.20639530755579472, "loss_total_v6": 0.5981799513101578, "step": 6514 }, { "epoch": 1.502544144144144, "grad_norm": 31.070980072021484, "learning_rate": 8.67995987050878e-07, "loss": 8.620676040649414, "loss_alignment": 0.7236328125, "loss_alignment_w": 0.36181640625, "loss_sft": 0.1779524702578783, "loss_total_v6": 0.5387923046946526, "step": 6515 }, { "epoch": 1.5027747747747746, "grad_norm": 30.780126571655273, "learning_rate": 8.668623636946566e-07, "loss": 9.152111053466797, "loss_alignment": 0.77587890625, "loss_alignment_w": 0.387939453125, "loss_sft": 0.1846625618636608, "loss_total_v6": 0.5720069035887718, "step": 6516 }, { "epoch": 1.5030054054054054, "grad_norm": 31.953866958618164, "learning_rate": 8.657294108254982e-07, "loss": 9.207876205444336, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.18892611470073462, "loss_total_v6": 0.5754922740161419, "step": 6517 }, { "epoch": 1.5032360360360362, "grad_norm": 33.76603698730469, "learning_rate": 8.645971286271903e-07, "loss": 8.8561372756958, "loss_alignment": 0.73974609375, "loss_alignment_w": 0.369873046875, "loss_sft": 0.1830251608043909, "loss_total_v6": 0.5535085797309875, "step": 6518 }, { "epoch": 1.5034666666666667, "grad_norm": 32.261474609375, "learning_rate": 8.634655172834178e-07, "loss": 9.088579177856445, "loss_alignment": 0.7861328125, "loss_alignment_w": 0.39306640625, "loss_sft": 0.17567165568470955, "loss_total_v6": 0.5680361613631248, "step": 6519 }, { "epoch": 1.5036972972972973, "grad_norm": 32.16005325317383, "learning_rate": 8.623345769777514e-07, "loss": 9.251556396484375, "loss_alignment": 0.748046875, "loss_alignment_w": 0.3740234375, "loss_sft": 0.2048701774328947, "loss_total_v6": 0.5782222524285316, "step": 6520 }, { "epoch": 1.5039279279279278, "grad_norm": 34.97526931762695, "learning_rate": 8.612043078936555e-07, "loss": 9.382179260253906, "loss_alignment": 0.77490234375, "loss_alignment_w": 0.387451171875, "loss_sft": 0.20007946714758873, "loss_total_v6": 0.5863862335681915, "step": 6521 }, { "epoch": 1.5041585585585584, "grad_norm": 35.886444091796875, "learning_rate": 8.600747102144824e-07, "loss": 9.062749862670898, "loss_alignment": 0.759765625, "loss_alignment_w": 0.3798828125, "loss_sft": 0.1857608873397112, "loss_total_v6": 0.566421877592802, "step": 6522 }, { "epoch": 1.5043891891891892, "grad_norm": 31.681489944458008, "learning_rate": 8.589457841234805e-07, "loss": 8.93216323852539, "loss_alignment": 0.7275390625, "loss_alignment_w": 0.36376953125, "loss_sft": 0.19409395288676023, "loss_total_v6": 0.5582602173089981, "step": 6523 }, { "epoch": 1.50461981981982, "grad_norm": 33.04971694946289, "learning_rate": 8.578175298037872e-07, "loss": 8.937178611755371, "loss_alignment": 0.7509765625, "loss_alignment_w": 0.37548828125, "loss_sft": 0.1827191784977913, "loss_total_v6": 0.5585736781358719, "step": 6524 }, { "epoch": 1.5048504504504505, "grad_norm": 35.14860153198242, "learning_rate": 8.566899474384299e-07, "loss": 8.534676551818848, "loss_alignment": 0.76611328125, "loss_alignment_w": 0.383056640625, "loss_sft": 0.15025383792817593, "loss_total_v6": 0.5334172882139683, "step": 6525 }, { "epoch": 1.505081081081081, "grad_norm": 31.9490909576416, "learning_rate": 8.555630372103274e-07, "loss": 9.107110977172852, "loss_alignment": 0.7763671875, "loss_alignment_w": 0.38818359375, "loss_sft": 0.18162118643522263, "loss_total_v6": 0.5691944286227226, "step": 6526 }, { "epoch": 1.5053117117117116, "grad_norm": 31.197282791137695, "learning_rate": 8.544367993022895e-07, "loss": 9.01144027709961, "loss_alignment": 0.76953125, "loss_alignment_w": 0.384765625, "loss_sft": 0.17875462118536234, "loss_total_v6": 0.5632150620222092, "step": 6527 }, { "epoch": 1.5055423423423422, "grad_norm": 32.65122604370117, "learning_rate": 8.533112338970157e-07, "loss": 8.955690383911133, "loss_alignment": 0.75732421875, "loss_alignment_w": 0.378662109375, "loss_sft": 0.18172465451061726, "loss_total_v6": 0.559730626642704, "step": 6528 }, { "epoch": 1.505772972972973, "grad_norm": 31.892662048339844, "learning_rate": 8.521863411770997e-07, "loss": 9.194742202758789, "loss_alignment": 0.75048828125, "loss_alignment_w": 0.375244140625, "loss_sft": 0.20037334971129894, "loss_total_v6": 0.5746714547276497, "step": 6529 }, { "epoch": 1.5060036036036037, "grad_norm": 32.165279388427734, "learning_rate": 8.510621213250248e-07, "loss": 8.720541000366211, "loss_alignment": 0.73046875, "loss_alignment_w": 0.365234375, "loss_sft": 0.18047085218131542, "loss_total_v6": 0.5450338385999203, "step": 6530 }, { "epoch": 1.5062342342342343, "grad_norm": 34.47011947631836, "learning_rate": 8.499385745231631e-07, "loss": 9.144723892211914, "loss_alignment": 0.77978515625, "loss_alignment_w": 0.389892578125, "loss_sft": 0.1819731406867504, "loss_total_v6": 0.5715452805161476, "step": 6531 }, { "epoch": 1.5064648648648649, "grad_norm": 29.95462417602539, "learning_rate": 8.488157009537796e-07, "loss": 8.796209335327148, "loss_alignment": 0.73486328125, "loss_alignment_w": 0.367431640625, "loss_sft": 0.1821025488898158, "loss_total_v6": 0.549763061106205, "step": 6532 }, { "epoch": 1.5066954954954954, "grad_norm": 31.78359031677246, "learning_rate": 8.476935007990278e-07, "loss": 9.0140962600708, "loss_alignment": 0.71826171875, "loss_alignment_w": 0.359130859375, "loss_sft": 0.20463162288069725, "loss_total_v6": 0.563381016254425, "step": 6533 }, { "epoch": 1.506926126126126, "grad_norm": 30.23809051513672, "learning_rate": 8.465719742409567e-07, "loss": 9.102316856384277, "loss_alignment": 0.75, "loss_alignment_w": 0.375, "loss_sft": 0.19526806939393282, "loss_total_v6": 0.568894799798727, "step": 6534 }, { "epoch": 1.5071567567567568, "grad_norm": 32.88969421386719, "learning_rate": 8.45451121461499e-07, "loss": 9.015384674072266, "loss_alignment": 0.75439453125, "loss_alignment_w": 0.377197265625, "loss_sft": 0.18566915299743414, "loss_total_v6": 0.5634615309536457, "step": 6535 }, { "epoch": 1.5073873873873875, "grad_norm": 31.81912612915039, "learning_rate": 8.443309426424862e-07, "loss": 9.31138801574707, "loss_alignment": 0.78076171875, "loss_alignment_w": 0.390380859375, "loss_sft": 0.1919776387512684, "loss_total_v6": 0.5819617509841919, "step": 6536 }, { "epoch": 1.507618018018018, "grad_norm": 32.970584869384766, "learning_rate": 8.432114379656337e-07, "loss": 9.575286865234375, "loss_alignment": 0.8017578125, "loss_alignment_w": 0.40087890625, "loss_sft": 0.1973171029239893, "loss_total_v6": 0.5984554067254066, "step": 6537 }, { "epoch": 1.5078486486486486, "grad_norm": 33.45290756225586, "learning_rate": 8.420926076125496e-07, "loss": 9.332558631896973, "loss_alignment": 0.77001953125, "loss_alignment_w": 0.385009765625, "loss_sft": 0.19845825619995594, "loss_total_v6": 0.5832849144935608, "step": 6538 }, { "epoch": 1.5080792792792792, "grad_norm": 31.183406829833984, "learning_rate": 8.409744517647351e-07, "loss": 9.416258811950684, "loss_alignment": 0.76953125, "loss_alignment_w": 0.384765625, "loss_sft": 0.20404046773910522, "loss_total_v6": 0.5885161608457565, "step": 6539 }, { "epoch": 1.5083099099099098, "grad_norm": 31.02783966064453, "learning_rate": 8.398569706035791e-07, "loss": 9.072824478149414, "loss_alignment": 0.77197265625, "loss_alignment_w": 0.385986328125, "loss_sft": 0.18169083818793297, "loss_total_v6": 0.5670515596866608, "step": 6540 }, { "epoch": 1.5085405405405405, "grad_norm": 29.63292121887207, "learning_rate": 8.387401643103609e-07, "loss": 9.009647369384766, "loss_alignment": 0.73193359375, "loss_alignment_w": 0.365966796875, "loss_sft": 0.19684623926877975, "loss_total_v6": 0.5631029605865479, "step": 6541 }, { "epoch": 1.5087711711711713, "grad_norm": 28.78763198852539, "learning_rate": 8.376240330662538e-07, "loss": 8.650192260742188, "loss_alignment": 0.74462890625, "loss_alignment_w": 0.372314453125, "loss_sft": 0.16838360577821732, "loss_total_v6": 0.5406370311975479, "step": 6542 }, { "epoch": 1.5090018018018019, "grad_norm": 36.961326599121094, "learning_rate": 8.365085770523163e-07, "loss": 9.001350402832031, "loss_alignment": 0.7529296875, "loss_alignment_w": 0.37646484375, "loss_sft": 0.18610429111868143, "loss_total_v6": 0.5625843740999699, "step": 6543 }, { "epoch": 1.5092324324324324, "grad_norm": 34.4307975769043, "learning_rate": 8.353937964495029e-07, "loss": 9.686386108398438, "loss_alignment": 0.78515625, "loss_alignment_w": 0.392578125, "loss_sft": 0.21225648559629917, "loss_total_v6": 0.6053991764783859, "step": 6544 }, { "epoch": 1.509463063063063, "grad_norm": 29.061302185058594, "learning_rate": 8.342796914386548e-07, "loss": 9.13833236694336, "loss_alignment": 0.77490234375, "loss_alignment_w": 0.387451171875, "loss_sft": 0.1844880860298872, "loss_total_v6": 0.5711457952857018, "step": 6545 }, { "epoch": 1.5096936936936936, "grad_norm": 30.945049285888672, "learning_rate": 8.331662622005037e-07, "loss": 8.981180191040039, "loss_alignment": 0.7412109375, "loss_alignment_w": 0.37060546875, "loss_sft": 0.19142018258571625, "loss_total_v6": 0.5613237470388412, "step": 6546 }, { "epoch": 1.5099243243243243, "grad_norm": 30.27975845336914, "learning_rate": 8.320535089156734e-07, "loss": 9.200510025024414, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.18701612576842308, "loss_total_v6": 0.575031865388155, "step": 6547 }, { "epoch": 1.510154954954955, "grad_norm": 30.09566307067871, "learning_rate": 8.309414317646769e-07, "loss": 8.960346221923828, "loss_alignment": 0.720703125, "loss_alignment_w": 0.3603515625, "loss_sft": 0.19962435588240623, "loss_total_v6": 0.5600216872990131, "step": 6548 }, { "epoch": 1.5103855855855857, "grad_norm": 31.39507293701172, "learning_rate": 8.298300309279195e-07, "loss": 8.914546012878418, "loss_alignment": 0.7138671875, "loss_alignment_w": 0.35693359375, "loss_sft": 0.20042387396097183, "loss_total_v6": 0.5571591183543205, "step": 6549 }, { "epoch": 1.5106162162162162, "grad_norm": 29.692359924316406, "learning_rate": 8.287193065856936e-07, "loss": 9.187705993652344, "loss_alignment": 0.7470703125, "loss_alignment_w": 0.37353515625, "loss_sft": 0.20010138023644686, "loss_total_v6": 0.5742316246032715, "step": 6550 }, { "epoch": 1.5108468468468468, "grad_norm": 32.2934455871582, "learning_rate": 8.276092589181844e-07, "loss": 8.587715148925781, "loss_alignment": 0.73095703125, "loss_alignment_w": 0.365478515625, "loss_sft": 0.17096382472664118, "loss_total_v6": 0.5367322452366352, "step": 6551 }, { "epoch": 1.5110774774774773, "grad_norm": 32.198814392089844, "learning_rate": 8.264998881054659e-07, "loss": 8.893256187438965, "loss_alignment": 0.74755859375, "loss_alignment_w": 0.373779296875, "loss_sft": 0.18226280994713306, "loss_total_v6": 0.5558285042643547, "step": 6552 }, { "epoch": 1.5113081081081081, "grad_norm": 30.184123992919922, "learning_rate": 8.253911943275017e-07, "loss": 8.885568618774414, "loss_alignment": 0.74267578125, "loss_alignment_w": 0.371337890625, "loss_sft": 0.18375073187053204, "loss_total_v6": 0.5553480312228203, "step": 6553 }, { "epoch": 1.511538738738739, "grad_norm": 34.00137710571289, "learning_rate": 8.242831777641474e-07, "loss": 8.906880378723145, "loss_alignment": 0.7236328125, "loss_alignment_w": 0.36181640625, "loss_sft": 0.19443637318909168, "loss_total_v6": 0.5566800087690353, "step": 6554 }, { "epoch": 1.5117693693693695, "grad_norm": 29.97522735595703, "learning_rate": 8.231758385951494e-07, "loss": 9.304353713989258, "loss_alignment": 0.7578125, "loss_alignment_w": 0.37890625, "loss_sft": 0.20386701822280884, "loss_total_v6": 0.581522062420845, "step": 6555 }, { "epoch": 1.512, "grad_norm": 32.20949172973633, "learning_rate": 8.220691770001421e-07, "loss": 8.627670288085938, "loss_alignment": 0.7080078125, "loss_alignment_w": 0.35400390625, "loss_sft": 0.18530178256332874, "loss_total_v6": 0.5392293892800808, "step": 6556 }, { "epoch": 1.5122306306306306, "grad_norm": 31.738283157348633, "learning_rate": 8.209631931586499e-07, "loss": 8.589284896850586, "loss_alignment": 0.74072265625, "loss_alignment_w": 0.370361328125, "loss_sft": 0.1665605753660202, "loss_total_v6": 0.5368303433060646, "step": 6557 }, { "epoch": 1.5124612612612611, "grad_norm": 29.42032814025879, "learning_rate": 8.198578872500873e-07, "loss": 10.071670532226562, "loss_alignment": 0.81494140625, "loss_alignment_w": 0.407470703125, "loss_sft": 0.22159669920802116, "loss_total_v6": 0.6294794082641602, "step": 6558 }, { "epoch": 1.512691891891892, "grad_norm": 33.078636169433594, "learning_rate": 8.187532594537618e-07, "loss": 8.665055274963379, "loss_alignment": 0.74072265625, "loss_alignment_w": 0.370361328125, "loss_sft": 0.17016702890396118, "loss_total_v6": 0.5415659472346306, "step": 6559 }, { "epoch": 1.5129225225225227, "grad_norm": 32.27989959716797, "learning_rate": 8.176493099488664e-07, "loss": 9.251708984375, "loss_alignment": 0.76416015625, "loss_alignment_w": 0.382080078125, "loss_sft": 0.1972961686551571, "loss_total_v6": 0.5782318264245987, "step": 6560 }, { "epoch": 1.5131531531531532, "grad_norm": 32.64252471923828, "learning_rate": 8.165460389144886e-07, "loss": 8.941682815551758, "loss_alignment": 0.7412109375, "loss_alignment_w": 0.37060546875, "loss_sft": 0.18803608510643244, "loss_total_v6": 0.5588551834225655, "step": 6561 }, { "epoch": 1.5133837837837838, "grad_norm": 35.87751007080078, "learning_rate": 8.154434465296007e-07, "loss": 9.05364990234375, "loss_alignment": 0.75439453125, "loss_alignment_w": 0.377197265625, "loss_sft": 0.18864056654274464, "loss_total_v6": 0.565853089094162, "step": 6562 }, { "epoch": 1.5136144144144144, "grad_norm": 31.269521713256836, "learning_rate": 8.143415329730714e-07, "loss": 9.142858505249023, "loss_alignment": 0.77490234375, "loss_alignment_w": 0.387451171875, "loss_sft": 0.18428264372050762, "loss_total_v6": 0.5714286267757416, "step": 6563 }, { "epoch": 1.513845045045045, "grad_norm": 30.15747833251953, "learning_rate": 8.132402984236531e-07, "loss": 8.784017562866211, "loss_alignment": 0.744140625, "loss_alignment_w": 0.3720703125, "loss_sft": 0.17694608308374882, "loss_total_v6": 0.5490011386573315, "step": 6564 }, { "epoch": 1.5140756756756757, "grad_norm": 31.37479019165039, "learning_rate": 8.121397430599919e-07, "loss": 8.904924392700195, "loss_alignment": 0.74365234375, "loss_alignment_w": 0.371826171875, "loss_sft": 0.18433484621345997, "loss_total_v6": 0.5565577447414398, "step": 6565 }, { "epoch": 1.5143063063063065, "grad_norm": 33.24531173706055, "learning_rate": 8.110398670606201e-07, "loss": 9.046041488647461, "loss_alignment": 0.7724609375, "loss_alignment_w": 0.38623046875, "loss_sft": 0.1791165815666318, "loss_total_v6": 0.565377589315176, "step": 6566 }, { "epoch": 1.514536936936937, "grad_norm": 31.364694595336914, "learning_rate": 8.099406706039636e-07, "loss": 9.124034881591797, "loss_alignment": 0.78564453125, "loss_alignment_w": 0.392822265625, "loss_sft": 0.17776559479534626, "loss_total_v6": 0.5702521689236164, "step": 6567 }, { "epoch": 1.5147675675675676, "grad_norm": 28.999536514282227, "learning_rate": 8.088421538683377e-07, "loss": 9.148280143737793, "loss_alignment": 0.80712890625, "loss_alignment_w": 0.403564453125, "loss_sft": 0.16914908960461617, "loss_total_v6": 0.5717675015330315, "step": 6568 }, { "epoch": 1.5149981981981981, "grad_norm": 32.118408203125, "learning_rate": 8.077443170319455e-07, "loss": 8.984247207641602, "loss_alignment": 0.76904296875, "loss_alignment_w": 0.384521484375, "loss_sft": 0.1766429990530014, "loss_total_v6": 0.5615154504776001, "step": 6569 }, { "epoch": 1.5152288288288287, "grad_norm": 33.0762939453125, "learning_rate": 8.066471602728804e-07, "loss": 8.956629753112793, "loss_alignment": 0.783203125, "loss_alignment_w": 0.3916015625, "loss_sft": 0.16956106573343277, "loss_total_v6": 0.559789352118969, "step": 6570 }, { "epoch": 1.5154594594594595, "grad_norm": 29.266345977783203, "learning_rate": 8.055506837691257e-07, "loss": 9.233617782592773, "loss_alignment": 0.73388671875, "loss_alignment_w": 0.366943359375, "loss_sft": 0.21017306484282017, "loss_total_v6": 0.5771011486649513, "step": 6571 }, { "epoch": 1.51569009009009, "grad_norm": 30.440101623535156, "learning_rate": 8.044548876985531e-07, "loss": 9.040544509887695, "loss_alignment": 0.75244140625, "loss_alignment_w": 0.376220703125, "loss_sft": 0.18876757100224495, "loss_total_v6": 0.5650340542197227, "step": 6572 }, { "epoch": 1.5159207207207208, "grad_norm": 31.282623291015625, "learning_rate": 8.033597722389264e-07, "loss": 8.987981796264648, "loss_alignment": 0.75244140625, "loss_alignment_w": 0.376220703125, "loss_sft": 0.1858790647238493, "loss_total_v6": 0.5617488101124763, "step": 6573 }, { "epoch": 1.5161513513513514, "grad_norm": 31.060148239135742, "learning_rate": 8.02265337567899e-07, "loss": 8.920112609863281, "loss_alignment": 0.7587890625, "loss_alignment_w": 0.37939453125, "loss_sft": 0.1783719202503562, "loss_total_v6": 0.5575070269405842, "step": 6574 }, { "epoch": 1.516381981981982, "grad_norm": 31.589187622070312, "learning_rate": 8.011715838630107e-07, "loss": 9.004328727722168, "loss_alignment": 0.76806640625, "loss_alignment_w": 0.384033203125, "loss_sft": 0.17803548090159893, "loss_total_v6": 0.5627705827355385, "step": 6575 }, { "epoch": 1.5166126126126125, "grad_norm": 30.667095184326172, "learning_rate": 8.00078511301694e-07, "loss": 8.682830810546875, "loss_alignment": 0.7421875, "loss_alignment_w": 0.37109375, "loss_sft": 0.17140006367117167, "loss_total_v6": 0.5426769405603409, "step": 6576 }, { "epoch": 1.5168432432432433, "grad_norm": 29.39139747619629, "learning_rate": 7.989861200612675e-07, "loss": 8.912077903747559, "loss_alignment": 0.71826171875, "loss_alignment_w": 0.359130859375, "loss_sft": 0.19721789099276066, "loss_total_v6": 0.5570048838853836, "step": 6577 }, { "epoch": 1.5170738738738738, "grad_norm": 30.860239028930664, "learning_rate": 7.97894410318944e-07, "loss": 9.478836059570312, "loss_alignment": 0.77099609375, "loss_alignment_w": 0.385498046875, "loss_sft": 0.20483870804309845, "loss_total_v6": 0.592427209019661, "step": 6578 }, { "epoch": 1.5173045045045046, "grad_norm": 30.283912658691406, "learning_rate": 7.968033822518212e-07, "loss": 8.901113510131836, "loss_alignment": 0.75390625, "loss_alignment_w": 0.376953125, "loss_sft": 0.1795647982507944, "loss_total_v6": 0.5563195645809174, "step": 6579 }, { "epoch": 1.5175351351351352, "grad_norm": 31.4973087310791, "learning_rate": 7.957130360368898e-07, "loss": 8.899088859558105, "loss_alignment": 0.73828125, "loss_alignment_w": 0.369140625, "loss_sft": 0.18639631383121014, "loss_total_v6": 0.5561930909752846, "step": 6580 }, { "epoch": 1.5177657657657657, "grad_norm": 30.261415481567383, "learning_rate": 7.946233718510277e-07, "loss": 8.787729263305664, "loss_alignment": 0.7548828125, "loss_alignment_w": 0.37744140625, "loss_sft": 0.171608604490757, "loss_total_v6": 0.5492331236600876, "step": 6581 }, { "epoch": 1.5179963963963963, "grad_norm": 30.74416160583496, "learning_rate": 7.935343898710007e-07, "loss": 9.317588806152344, "loss_alignment": 0.779296875, "loss_alignment_w": 0.3896484375, "loss_sft": 0.19192268699407578, "loss_total_v6": 0.5823493227362633, "step": 6582 }, { "epoch": 1.518227027027027, "grad_norm": 29.26449203491211, "learning_rate": 7.924460902734698e-07, "loss": 8.898258209228516, "loss_alignment": 0.75341796875, "loss_alignment_w": 0.376708984375, "loss_sft": 0.17966100294142962, "loss_total_v6": 0.5561411082744598, "step": 6583 }, { "epoch": 1.5184576576576576, "grad_norm": 32.8100471496582, "learning_rate": 7.913584732349788e-07, "loss": 9.008529663085938, "loss_alignment": 0.72509765625, "loss_alignment_w": 0.362548828125, "loss_sft": 0.2005147747695446, "loss_total_v6": 0.5630330815911293, "step": 6584 }, { "epoch": 1.5186882882882884, "grad_norm": 26.770885467529297, "learning_rate": 7.902715389319632e-07, "loss": 8.689788818359375, "loss_alignment": 0.7529296875, "loss_alignment_w": 0.37646484375, "loss_sft": 0.1664181463420391, "loss_total_v6": 0.5431118682026863, "step": 6585 }, { "epoch": 1.518918918918919, "grad_norm": 34.82318115234375, "learning_rate": 7.891852875407496e-07, "loss": 9.080764770507812, "loss_alignment": 0.7373046875, "loss_alignment_w": 0.36865234375, "loss_sft": 0.19917011633515358, "loss_total_v6": 0.5675477981567383, "step": 6586 }, { "epoch": 1.5191495495495495, "grad_norm": 34.27235412597656, "learning_rate": 7.880997192375506e-07, "loss": 9.080534934997559, "loss_alignment": 0.7724609375, "loss_alignment_w": 0.38623046875, "loss_sft": 0.18276783637702465, "loss_total_v6": 0.5675334706902504, "step": 6587 }, { "epoch": 1.51938018018018, "grad_norm": 28.949644088745117, "learning_rate": 7.870148341984712e-07, "loss": 9.026998519897461, "loss_alignment": 0.7841796875, "loss_alignment_w": 0.39208984375, "loss_sft": 0.17337932996451855, "loss_total_v6": 0.5641874149441719, "step": 6588 }, { "epoch": 1.5196108108108108, "grad_norm": 27.874433517456055, "learning_rate": 7.859306325995037e-07, "loss": 9.338348388671875, "loss_alignment": 0.798828125, "loss_alignment_w": 0.3994140625, "loss_sft": 0.1834392286837101, "loss_total_v6": 0.5836467295885086, "step": 6589 }, { "epoch": 1.5198414414414414, "grad_norm": 36.72683334350586, "learning_rate": 7.848471146165287e-07, "loss": 9.581991195678711, "loss_alignment": 0.771484375, "loss_alignment_w": 0.3857421875, "loss_sft": 0.21348323859274387, "loss_total_v6": 0.5988744869828224, "step": 6590 }, { "epoch": 1.5200720720720722, "grad_norm": 28.48818588256836, "learning_rate": 7.837642804253159e-07, "loss": 9.485002517700195, "loss_alignment": 0.78759765625, "loss_alignment_w": 0.393798828125, "loss_sft": 0.19904438965022564, "loss_total_v6": 0.5928126871585846, "step": 6591 }, { "epoch": 1.5203027027027027, "grad_norm": 34.4343147277832, "learning_rate": 7.826821302015275e-07, "loss": 9.0187349319458, "loss_alignment": 0.74462890625, "loss_alignment_w": 0.372314453125, "loss_sft": 0.19079190026968718, "loss_total_v6": 0.5636709444224834, "step": 6592 }, { "epoch": 1.5205333333333333, "grad_norm": 33.59428787231445, "learning_rate": 7.816006641207124e-07, "loss": 8.677821159362793, "loss_alignment": 0.7333984375, "loss_alignment_w": 0.36669921875, "loss_sft": 0.17703790590167046, "loss_total_v6": 0.5423638373613358, "step": 6593 }, { "epoch": 1.5207639639639638, "grad_norm": 32.26648712158203, "learning_rate": 7.805198823583076e-07, "loss": 9.09707260131836, "loss_alignment": 0.72900390625, "loss_alignment_w": 0.364501953125, "loss_sft": 0.20275282114744186, "loss_total_v6": 0.5685670301318169, "step": 6594 }, { "epoch": 1.5209945945945946, "grad_norm": 26.859708786010742, "learning_rate": 7.794397850896401e-07, "loss": 9.186321258544922, "loss_alignment": 0.755859375, "loss_alignment_w": 0.3779296875, "loss_sft": 0.1958338748663664, "loss_total_v6": 0.5741450488567352, "step": 6595 }, { "epoch": 1.5212252252252252, "grad_norm": 32.27123260498047, "learning_rate": 7.783603724899258e-07, "loss": 9.011383056640625, "loss_alignment": 0.734375, "loss_alignment_w": 0.3671875, "loss_sft": 0.19652744755148888, "loss_total_v6": 0.5632114112377167, "step": 6596 }, { "epoch": 1.521455855855856, "grad_norm": 31.237985610961914, "learning_rate": 7.772816447342685e-07, "loss": 9.240903854370117, "loss_alignment": 0.77392578125, "loss_alignment_w": 0.386962890625, "loss_sft": 0.1908529493957758, "loss_total_v6": 0.5775564536452293, "step": 6597 }, { "epoch": 1.5216864864864865, "grad_norm": 31.685810089111328, "learning_rate": 7.762036019976626e-07, "loss": 8.873705863952637, "loss_alignment": 0.74072265625, "loss_alignment_w": 0.370361328125, "loss_sft": 0.18487090431153774, "loss_total_v6": 0.5546066462993622, "step": 6598 }, { "epoch": 1.521917117117117, "grad_norm": 40.62203598022461, "learning_rate": 7.751262444549917e-07, "loss": 9.524394035339355, "loss_alignment": 0.76318359375, "loss_alignment_w": 0.381591796875, "loss_sft": 0.21317926049232483, "loss_total_v6": 0.5952746272087097, "step": 6599 }, { "epoch": 1.5221477477477476, "grad_norm": 33.5678825378418, "learning_rate": 7.740495722810271e-07, "loss": 9.094056129455566, "loss_alignment": 0.77490234375, "loss_alignment_w": 0.387451171875, "loss_sft": 0.18086625635623932, "loss_total_v6": 0.5683784931898117, "step": 6600 }, { "epoch": 1.5223783783783784, "grad_norm": 31.77696990966797, "learning_rate": 7.729735856504278e-07, "loss": 9.275996208190918, "loss_alignment": 0.75634765625, "loss_alignment_w": 0.378173828125, "loss_sft": 0.2015149313956499, "loss_total_v6": 0.5797497928142548, "step": 6601 }, { "epoch": 1.522609009009009, "grad_norm": 32.1407470703125, "learning_rate": 7.718982847377421e-07, "loss": 8.69454574584961, "loss_alignment": 0.7412109375, "loss_alignment_w": 0.37060546875, "loss_sft": 0.17352078016847372, "loss_total_v6": 0.5434090904891491, "step": 6602 }, { "epoch": 1.5228396396396398, "grad_norm": 31.664953231811523, "learning_rate": 7.708236697174104e-07, "loss": 8.789419174194336, "loss_alignment": 0.75537109375, "loss_alignment_w": 0.377685546875, "loss_sft": 0.17140896245837212, "loss_total_v6": 0.549338661134243, "step": 6603 }, { "epoch": 1.5230702702702703, "grad_norm": 36.149940490722656, "learning_rate": 7.697497407637566e-07, "loss": 8.944100379943848, "loss_alignment": 0.7490234375, "loss_alignment_w": 0.37451171875, "loss_sft": 0.18347224034368992, "loss_total_v6": 0.5590062960982323, "step": 6604 }, { "epoch": 1.5233009009009009, "grad_norm": 36.50287628173828, "learning_rate": 7.686764980509986e-07, "loss": 8.849323272705078, "loss_alignment": 0.7548828125, "loss_alignment_w": 0.37744140625, "loss_sft": 0.17515302076935768, "loss_total_v6": 0.5530826821923256, "step": 6605 }, { "epoch": 1.5235315315315314, "grad_norm": 33.03681564331055, "learning_rate": 7.676039417532383e-07, "loss": 8.877182006835938, "loss_alignment": 0.73486328125, "loss_alignment_w": 0.367431640625, "loss_sft": 0.18815522454679012, "loss_total_v6": 0.5548239201307297, "step": 6606 }, { "epoch": 1.5237621621621622, "grad_norm": 29.749908447265625, "learning_rate": 7.66532072044468e-07, "loss": 8.738547325134277, "loss_alignment": 0.7255859375, "loss_alignment_w": 0.36279296875, "loss_sft": 0.18484636023640633, "loss_total_v6": 0.5461592152714729, "step": 6607 }, { "epoch": 1.5239927927927928, "grad_norm": 32.01963424682617, "learning_rate": 7.654608890985709e-07, "loss": 9.011587142944336, "loss_alignment": 0.74267578125, "loss_alignment_w": 0.371337890625, "loss_sft": 0.19353429414331913, "loss_total_v6": 0.5632242374122143, "step": 6608 }, { "epoch": 1.5242234234234235, "grad_norm": 30.693695068359375, "learning_rate": 7.643903930893154e-07, "loss": 9.796775817871094, "loss_alignment": 0.77734375, "loss_alignment_w": 0.388671875, "loss_sft": 0.22335194237530231, "loss_total_v6": 0.6122984737157822, "step": 6609 }, { "epoch": 1.524454054054054, "grad_norm": 31.981834411621094, "learning_rate": 7.633205841903585e-07, "loss": 8.908673286437988, "loss_alignment": 0.7265625, "loss_alignment_w": 0.36328125, "loss_sft": 0.19267156347632408, "loss_total_v6": 0.5567920543253422, "step": 6610 }, { "epoch": 1.5246846846846847, "grad_norm": 33.05351257324219, "learning_rate": 7.622514625752487e-07, "loss": 9.159961700439453, "loss_alignment": 0.73828125, "loss_alignment_w": 0.369140625, "loss_sft": 0.20372319966554642, "loss_total_v6": 0.5724976062774658, "step": 6611 }, { "epoch": 1.5249153153153152, "grad_norm": 28.312463760375977, "learning_rate": 7.61183028417422e-07, "loss": 8.862564086914062, "loss_alignment": 0.7294921875, "loss_alignment_w": 0.36474609375, "loss_sft": 0.1890268437564373, "loss_total_v6": 0.5539102703332901, "step": 6612 }, { "epoch": 1.525145945945946, "grad_norm": 33.25980758666992, "learning_rate": 7.601152818902013e-07, "loss": 9.548637390136719, "loss_alignment": 0.76611328125, "loss_alignment_w": 0.383056640625, "loss_sft": 0.214114710688591, "loss_total_v6": 0.5967898741364479, "step": 6613 }, { "epoch": 1.5253765765765765, "grad_norm": 28.5249080657959, "learning_rate": 7.590482231667995e-07, "loss": 9.115303039550781, "loss_alignment": 0.73876953125, "loss_alignment_w": 0.369384765625, "loss_sft": 0.2000775709748268, "loss_total_v6": 0.569706454873085, "step": 6614 }, { "epoch": 1.5256072072072073, "grad_norm": 28.224946975708008, "learning_rate": 7.579818524203164e-07, "loss": 9.056312561035156, "loss_alignment": 0.74951171875, "loss_alignment_w": 0.374755859375, "loss_sft": 0.19179773144423962, "loss_total_v6": 0.5660195089876652, "step": 6615 }, { "epoch": 1.5258378378378379, "grad_norm": 30.2745304107666, "learning_rate": 7.569161698237404e-07, "loss": 8.841137886047363, "loss_alignment": 0.724609375, "loss_alignment_w": 0.3623046875, "loss_sft": 0.19159391336143017, "loss_total_v6": 0.5525710955262184, "step": 6616 }, { "epoch": 1.5260684684684684, "grad_norm": 32.081207275390625, "learning_rate": 7.558511755499498e-07, "loss": 8.95630931854248, "loss_alignment": 0.77099609375, "loss_alignment_w": 0.385498046875, "loss_sft": 0.1743933390825987, "loss_total_v6": 0.5597693249583244, "step": 6617 }, { "epoch": 1.526299099099099, "grad_norm": 29.710124969482422, "learning_rate": 7.54786869771712e-07, "loss": 8.757102966308594, "loss_alignment": 0.73583984375, "loss_alignment_w": 0.367919921875, "loss_sft": 0.18036028370261192, "loss_total_v6": 0.547318909317255, "step": 6618 }, { "epoch": 1.5265297297297298, "grad_norm": 27.762731552124023, "learning_rate": 7.537232526616801e-07, "loss": 9.232438087463379, "loss_alignment": 0.755859375, "loss_alignment_w": 0.3779296875, "loss_sft": 0.19894511997699738, "loss_total_v6": 0.5770273804664612, "step": 6619 }, { "epoch": 1.5267603603603603, "grad_norm": 26.295682907104492, "learning_rate": 7.526603243923958e-07, "loss": 8.978368759155273, "loss_alignment": 0.755859375, "loss_alignment_w": 0.3779296875, "loss_sft": 0.18273009732365608, "loss_total_v6": 0.5611480548977852, "step": 6620 }, { "epoch": 1.5269909909909911, "grad_norm": 35.05677795410156, "learning_rate": 7.515980851362892e-07, "loss": 8.669907569885254, "loss_alignment": 0.7197265625, "loss_alignment_w": 0.35986328125, "loss_sft": 0.18168550729751587, "loss_total_v6": 0.5418692231178284, "step": 6621 }, { "epoch": 1.5272216216216217, "grad_norm": 30.451168060302734, "learning_rate": 7.505365350656813e-07, "loss": 9.192075729370117, "loss_alignment": 0.7626953125, "loss_alignment_w": 0.38134765625, "loss_sft": 0.19268405623733997, "loss_total_v6": 0.5745047181844711, "step": 6622 }, { "epoch": 1.5274522522522522, "grad_norm": 30.564613342285156, "learning_rate": 7.494756743527765e-07, "loss": 9.139583587646484, "loss_alignment": 0.7587890625, "loss_alignment_w": 0.37939453125, "loss_sft": 0.19102077186107635, "loss_total_v6": 0.571224007755518, "step": 6623 }, { "epoch": 1.5276828828828828, "grad_norm": 30.016677856445312, "learning_rate": 7.484155031696727e-07, "loss": 9.171891212463379, "loss_alignment": 0.7900390625, "loss_alignment_w": 0.39501953125, "loss_sft": 0.1776896007359028, "loss_total_v6": 0.5732431933283806, "step": 6624 }, { "epoch": 1.5279135135135136, "grad_norm": 33.46433639526367, "learning_rate": 7.473560216883524e-07, "loss": 8.89771842956543, "loss_alignment": 0.7294921875, "loss_alignment_w": 0.36474609375, "loss_sft": 0.19044583477079868, "loss_total_v6": 0.5561074465513229, "step": 6625 }, { "epoch": 1.5281441441441441, "grad_norm": 31.06566619873047, "learning_rate": 7.462972300806859e-07, "loss": 9.165258407592773, "loss_alignment": 0.77734375, "loss_alignment_w": 0.388671875, "loss_sft": 0.18496542423963547, "loss_total_v6": 0.5728285983204842, "step": 6626 }, { "epoch": 1.528374774774775, "grad_norm": 33.20802688598633, "learning_rate": 7.452391285184346e-07, "loss": 8.832195281982422, "loss_alignment": 0.74365234375, "loss_alignment_w": 0.371826171875, "loss_sft": 0.1804149467498064, "loss_total_v6": 0.552012238651514, "step": 6627 }, { "epoch": 1.5286054054054055, "grad_norm": 29.510530471801758, "learning_rate": 7.441817171732457e-07, "loss": 9.270472526550293, "loss_alignment": 0.7470703125, "loss_alignment_w": 0.37353515625, "loss_sft": 0.20608298107981682, "loss_total_v6": 0.5794045329093933, "step": 6628 }, { "epoch": 1.528836036036036, "grad_norm": 29.54072380065918, "learning_rate": 7.431249962166531e-07, "loss": 9.527753829956055, "loss_alignment": 0.77490234375, "loss_alignment_w": 0.387451171875, "loss_sft": 0.20844541862607002, "loss_total_v6": 0.5954846143722534, "step": 6629 }, { "epoch": 1.5290666666666666, "grad_norm": 33.78466796875, "learning_rate": 7.420689658200836e-07, "loss": 8.690860748291016, "loss_alignment": 0.75634765625, "loss_alignment_w": 0.378173828125, "loss_sft": 0.16555426269769669, "loss_total_v6": 0.5431787744164467, "step": 6630 }, { "epoch": 1.5292972972972971, "grad_norm": 32.78797149658203, "learning_rate": 7.410136261548456e-07, "loss": 8.877582550048828, "loss_alignment": 0.74609375, "loss_alignment_w": 0.373046875, "loss_sft": 0.18167992681264877, "loss_total_v6": 0.55484888702631, "step": 6631 }, { "epoch": 1.529527927927928, "grad_norm": 33.463539123535156, "learning_rate": 7.399589773921412e-07, "loss": 9.618528366088867, "loss_alignment": 0.77734375, "loss_alignment_w": 0.388671875, "loss_sft": 0.21265394426882267, "loss_total_v6": 0.6011579856276512, "step": 6632 }, { "epoch": 1.5297585585585587, "grad_norm": 35.20152282714844, "learning_rate": 7.389050197030573e-07, "loss": 8.98340892791748, "loss_alignment": 0.74169921875, "loss_alignment_w": 0.370849609375, "loss_sft": 0.19154421240091324, "loss_total_v6": 0.5614630468189716, "step": 6633 }, { "epoch": 1.5299891891891892, "grad_norm": 37.52211380004883, "learning_rate": 7.378517532585694e-07, "loss": 9.240653991699219, "loss_alignment": 0.78369140625, "loss_alignment_w": 0.391845703125, "loss_sft": 0.18555784597992897, "loss_total_v6": 0.5775408744812012, "step": 6634 }, { "epoch": 1.5302198198198198, "grad_norm": 31.989286422729492, "learning_rate": 7.367991782295392e-07, "loss": 8.855203628540039, "loss_alignment": 0.7548828125, "loss_alignment_w": 0.37744140625, "loss_sft": 0.17623769119381905, "loss_total_v6": 0.5534502640366554, "step": 6635 }, { "epoch": 1.5304504504504504, "grad_norm": 32.78994369506836, "learning_rate": 7.357472947867189e-07, "loss": 9.575008392333984, "loss_alignment": 0.7861328125, "loss_alignment_w": 0.39306640625, "loss_sft": 0.20550897531211376, "loss_total_v6": 0.5984380319714546, "step": 6636 }, { "epoch": 1.530681081081081, "grad_norm": 36.452003479003906, "learning_rate": 7.346961031007494e-07, "loss": 8.863927841186523, "loss_alignment": 0.74560546875, "loss_alignment_w": 0.372802734375, "loss_sft": 0.18177255056798458, "loss_total_v6": 0.5539954602718353, "step": 6637 }, { "epoch": 1.5309117117117117, "grad_norm": 30.13498306274414, "learning_rate": 7.336456033421551e-07, "loss": 8.767845153808594, "loss_alignment": 0.73681640625, "loss_alignment_w": 0.368408203125, "loss_sft": 0.179444775916636, "loss_total_v6": 0.5479903034865856, "step": 6638 }, { "epoch": 1.5311423423423425, "grad_norm": 32.166038513183594, "learning_rate": 7.325957956813512e-07, "loss": 9.18891716003418, "loss_alignment": 0.7724609375, "loss_alignment_w": 0.38623046875, "loss_sft": 0.18859560042619705, "loss_total_v6": 0.5743072852492332, "step": 6639 }, { "epoch": 1.531372972972973, "grad_norm": 33.08745574951172, "learning_rate": 7.315466802886401e-07, "loss": 9.086057662963867, "loss_alignment": 0.77197265625, "loss_alignment_w": 0.385986328125, "loss_sft": 0.1824263734742999, "loss_total_v6": 0.5678786262869835, "step": 6640 }, { "epoch": 1.5316036036036036, "grad_norm": 29.235300064086914, "learning_rate": 7.3049825733421e-07, "loss": 8.406017303466797, "loss_alignment": 0.712890625, "loss_alignment_w": 0.3564453125, "loss_sft": 0.16905284859240055, "loss_total_v6": 0.5253760926425457, "step": 6641 }, { "epoch": 1.5318342342342341, "grad_norm": 28.372201919555664, "learning_rate": 7.294505269881397e-07, "loss": 9.049383163452148, "loss_alignment": 0.75537109375, "loss_alignment_w": 0.377685546875, "loss_sft": 0.18935048580169678, "loss_total_v6": 0.5655864290893078, "step": 6642 }, { "epoch": 1.5320648648648647, "grad_norm": 31.12255096435547, "learning_rate": 7.284034894203962e-07, "loss": 8.627631187438965, "loss_alignment": 0.70654296875, "loss_alignment_w": 0.353271484375, "loss_sft": 0.18594019301235676, "loss_total_v6": 0.5392269566655159, "step": 6643 }, { "epoch": 1.5322954954954955, "grad_norm": 28.72942543029785, "learning_rate": 7.273571448008305e-07, "loss": 9.147491455078125, "loss_alignment": 0.77783203125, "loss_alignment_w": 0.388916015625, "loss_sft": 0.18306163139641285, "loss_total_v6": 0.5717182531952858, "step": 6644 }, { "epoch": 1.5325261261261263, "grad_norm": 29.180742263793945, "learning_rate": 7.263114932991833e-07, "loss": 9.09991455078125, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.18111036345362663, "loss_total_v6": 0.5687446743249893, "step": 6645 }, { "epoch": 1.5327567567567568, "grad_norm": 34.941009521484375, "learning_rate": 7.252665350850808e-07, "loss": 9.249235153198242, "loss_alignment": 0.7646484375, "loss_alignment_w": 0.38232421875, "loss_sft": 0.19544782489538193, "loss_total_v6": 0.5780772045254707, "step": 6646 }, { "epoch": 1.5329873873873874, "grad_norm": 31.615694046020508, "learning_rate": 7.242222703280422e-07, "loss": 8.983583450317383, "loss_alignment": 0.75048828125, "loss_alignment_w": 0.375244140625, "loss_sft": 0.1864587590098381, "loss_total_v6": 0.5614740028977394, "step": 6647 }, { "epoch": 1.533218018018018, "grad_norm": 31.702558517456055, "learning_rate": 7.23178699197467e-07, "loss": 9.027052879333496, "loss_alignment": 0.75634765625, "loss_alignment_w": 0.378173828125, "loss_sft": 0.1861848421394825, "loss_total_v6": 0.5641908347606659, "step": 6648 }, { "epoch": 1.5334486486486485, "grad_norm": 31.8967227935791, "learning_rate": 7.221358218626484e-07, "loss": 9.350205421447754, "loss_alignment": 0.7373046875, "loss_alignment_w": 0.36865234375, "loss_sft": 0.21631530672311783, "loss_total_v6": 0.5843878239393234, "step": 6649 }, { "epoch": 1.5336792792792793, "grad_norm": 29.778743743896484, "learning_rate": 7.210936384927631e-07, "loss": 8.746015548706055, "loss_alignment": 0.74169921875, "loss_alignment_w": 0.370849609375, "loss_sft": 0.17516603134572506, "loss_total_v6": 0.5466260090470314, "step": 6650 }, { "epoch": 1.53390990990991, "grad_norm": 33.94708251953125, "learning_rate": 7.200521492568757e-07, "loss": 9.063369750976562, "loss_alignment": 0.75, "loss_alignment_w": 0.375, "loss_sft": 0.1917047956958413, "loss_total_v6": 0.5664606504142284, "step": 6651 }, { "epoch": 1.5341405405405406, "grad_norm": 30.950855255126953, "learning_rate": 7.190113543239408e-07, "loss": 8.793758392333984, "loss_alignment": 0.736328125, "loss_alignment_w": 0.3681640625, "loss_sft": 0.18115593865513802, "loss_total_v6": 0.5496098957955837, "step": 6652 }, { "epoch": 1.5343711711711712, "grad_norm": 28.895029067993164, "learning_rate": 7.179712538627981e-07, "loss": 8.78280258178711, "loss_alignment": 0.74609375, "loss_alignment_w": 0.373046875, "loss_sft": 0.17717529274523258, "loss_total_v6": 0.5489251837134361, "step": 6653 }, { "epoch": 1.5346018018018017, "grad_norm": 33.57350158691406, "learning_rate": 7.169318480421728e-07, "loss": 8.817241668701172, "loss_alignment": 0.73681640625, "loss_alignment_w": 0.368408203125, "loss_sft": 0.18358496017754078, "loss_total_v6": 0.5510776117444038, "step": 6654 }, { "epoch": 1.5348324324324323, "grad_norm": 30.251554489135742, "learning_rate": 7.158931370306832e-07, "loss": 9.510971069335938, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.20976156741380692, "loss_total_v6": 0.5944356843829155, "step": 6655 }, { "epoch": 1.535063063063063, "grad_norm": 29.712751388549805, "learning_rate": 7.148551209968279e-07, "loss": 9.035102844238281, "loss_alignment": 0.7734375, "loss_alignment_w": 0.38671875, "loss_sft": 0.17834144923835993, "loss_total_v6": 0.5646939873695374, "step": 6656 }, { "epoch": 1.5352936936936938, "grad_norm": 31.52184295654297, "learning_rate": 7.138178001089996e-07, "loss": 9.239123344421387, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.19338146410882473, "loss_total_v6": 0.5774451866745949, "step": 6657 }, { "epoch": 1.5355243243243244, "grad_norm": 27.6643123626709, "learning_rate": 7.127811745354729e-07, "loss": 9.31109619140625, "loss_alignment": 0.7763671875, "loss_alignment_w": 0.38818359375, "loss_sft": 0.19452284462749958, "loss_total_v6": 0.5819434970617294, "step": 6658 }, { "epoch": 1.535754954954955, "grad_norm": 32.160526275634766, "learning_rate": 7.117452444444112e-07, "loss": 8.984115600585938, "loss_alignment": 0.72265625, "loss_alignment_w": 0.361328125, "loss_sft": 0.2004231959581375, "loss_total_v6": 0.5615071915090084, "step": 6659 }, { "epoch": 1.5359855855855855, "grad_norm": 32.452171325683594, "learning_rate": 7.107100100038672e-07, "loss": 9.153366088867188, "loss_alignment": 0.72314453125, "loss_alignment_w": 0.361572265625, "loss_sft": 0.21065038815140724, "loss_total_v6": 0.5720853321254253, "step": 6660 }, { "epoch": 1.536216216216216, "grad_norm": 32.2254524230957, "learning_rate": 7.096754713817771e-07, "loss": 9.266663551330566, "loss_alignment": 0.72314453125, "loss_alignment_w": 0.361572265625, "loss_sft": 0.21750265173614025, "loss_total_v6": 0.5791664645075798, "step": 6661 }, { "epoch": 1.5364468468468468, "grad_norm": 31.02385902404785, "learning_rate": 7.086416287459679e-07, "loss": 9.182761192321777, "loss_alignment": 0.7724609375, "loss_alignment_w": 0.38623046875, "loss_sft": 0.18779895827174187, "loss_total_v6": 0.5739226043224335, "step": 6662 }, { "epoch": 1.5366774774774776, "grad_norm": 33.18397903442383, "learning_rate": 7.07608482264151e-07, "loss": 8.92133903503418, "loss_alignment": 0.736328125, "loss_alignment_w": 0.3681640625, "loss_sft": 0.18784793373197317, "loss_total_v6": 0.5575836449861526, "step": 6663 }, { "epoch": 1.5369081081081082, "grad_norm": 29.1661319732666, "learning_rate": 7.065760321039261e-07, "loss": 9.23757266998291, "loss_alignment": 0.77099609375, "loss_alignment_w": 0.385498046875, "loss_sft": 0.19075163267552853, "loss_total_v6": 0.5773482918739319, "step": 6664 }, { "epoch": 1.5371387387387387, "grad_norm": 30.392942428588867, "learning_rate": 7.055442784327782e-07, "loss": 8.646873474121094, "loss_alignment": 0.73291015625, "loss_alignment_w": 0.366455078125, "loss_sft": 0.17340996954590082, "loss_total_v6": 0.5404296331107616, "step": 6665 }, { "epoch": 1.5373693693693693, "grad_norm": 30.740924835205078, "learning_rate": 7.045132214180817e-07, "loss": 9.31120491027832, "loss_alignment": 0.783203125, "loss_alignment_w": 0.3916015625, "loss_sft": 0.1908675581216812, "loss_total_v6": 0.5819503143429756, "step": 6666 }, { "epoch": 1.5375999999999999, "grad_norm": 28.427467346191406, "learning_rate": 7.034828612270983e-07, "loss": 8.823211669921875, "loss_alignment": 0.74951171875, "loss_alignment_w": 0.374755859375, "loss_sft": 0.1770458184182644, "loss_total_v6": 0.551450714468956, "step": 6667 }, { "epoch": 1.5378306306306306, "grad_norm": 36.88418960571289, "learning_rate": 7.024531980269744e-07, "loss": 9.072965621948242, "loss_alignment": 0.74658203125, "loss_alignment_w": 0.373291015625, "loss_sft": 0.19387616775929928, "loss_total_v6": 0.5670603588223457, "step": 6668 }, { "epoch": 1.5380612612612614, "grad_norm": 32.702919006347656, "learning_rate": 7.014242319847447e-07, "loss": 8.728052139282227, "loss_alignment": 0.755859375, "loss_alignment_w": 0.3779296875, "loss_sft": 0.1678939936682582, "loss_total_v6": 0.5455032661557198, "step": 6669 }, { "epoch": 1.538291891891892, "grad_norm": 31.222150802612305, "learning_rate": 7.003959632673285e-07, "loss": 8.696564674377441, "loss_alignment": 0.72998046875, "loss_alignment_w": 0.364990234375, "loss_sft": 0.17772110551595688, "loss_total_v6": 0.5435353368520737, "step": 6670 }, { "epoch": 1.5385225225225225, "grad_norm": 29.343881607055664, "learning_rate": 6.993683920415367e-07, "loss": 9.182247161865234, "loss_alignment": 0.75244140625, "loss_alignment_w": 0.376220703125, "loss_sft": 0.19765442609786987, "loss_total_v6": 0.573890395462513, "step": 6671 }, { "epoch": 1.538753153153153, "grad_norm": 32.2764892578125, "learning_rate": 6.983415184740616e-07, "loss": 8.627466201782227, "loss_alignment": 0.74267578125, "loss_alignment_w": 0.371337890625, "loss_sft": 0.16853489726781845, "loss_total_v6": 0.5392166301608086, "step": 6672 }, { "epoch": 1.5389837837837836, "grad_norm": 34.4785041809082, "learning_rate": 6.973153427314877e-07, "loss": 9.084254264831543, "loss_alignment": 0.76171875, "loss_alignment_w": 0.380859375, "loss_sft": 0.18699807114899158, "loss_total_v6": 0.5677659064531326, "step": 6673 }, { "epoch": 1.5392144144144144, "grad_norm": 29.275768280029297, "learning_rate": 6.962898649802824e-07, "loss": 9.262975692749023, "loss_alignment": 0.78369140625, "loss_alignment_w": 0.391845703125, "loss_sft": 0.18698347359895706, "loss_total_v6": 0.578935980796814, "step": 6674 }, { "epoch": 1.5394450450450452, "grad_norm": 32.647891998291016, "learning_rate": 6.952650853867993e-07, "loss": 9.108304977416992, "loss_alignment": 0.75, "loss_alignment_w": 0.375, "loss_sft": 0.1942233070731163, "loss_total_v6": 0.5692690908908844, "step": 6675 }, { "epoch": 1.5396756756756758, "grad_norm": 34.8470458984375, "learning_rate": 6.942410041172837e-07, "loss": 9.407978057861328, "loss_alignment": 0.76025390625, "loss_alignment_w": 0.380126953125, "loss_sft": 0.20796327479183674, "loss_total_v6": 0.5879986733198166, "step": 6676 }, { "epoch": 1.5399063063063063, "grad_norm": 31.648448944091797, "learning_rate": 6.932176213378628e-07, "loss": 8.704914093017578, "loss_alignment": 0.70654296875, "loss_alignment_w": 0.353271484375, "loss_sft": 0.18980913702398539, "loss_total_v6": 0.5440571680665016, "step": 6677 }, { "epoch": 1.5401369369369369, "grad_norm": 29.644855499267578, "learning_rate": 6.921949372145503e-07, "loss": 8.686124801635742, "loss_alignment": 0.7412109375, "loss_alignment_w": 0.37060546875, "loss_sft": 0.17247576266527176, "loss_total_v6": 0.5428828671574593, "step": 6678 }, { "epoch": 1.5403675675675674, "grad_norm": 29.273855209350586, "learning_rate": 6.91172951913252e-07, "loss": 8.846220016479492, "loss_alignment": 0.728515625, "loss_alignment_w": 0.3642578125, "loss_sft": 0.18986688740551472, "loss_total_v6": 0.5528887286782265, "step": 6679 }, { "epoch": 1.5405981981981982, "grad_norm": 34.762123107910156, "learning_rate": 6.901516655997536e-07, "loss": 8.570338249206543, "loss_alignment": 0.73486328125, "loss_alignment_w": 0.367431640625, "loss_sft": 0.16891640983521938, "loss_total_v6": 0.5356461368501186, "step": 6680 }, { "epoch": 1.540828828828829, "grad_norm": 33.037071228027344, "learning_rate": 6.891310784397326e-07, "loss": 9.485301971435547, "loss_alignment": 0.78515625, "loss_alignment_w": 0.392578125, "loss_sft": 0.20043632201850414, "loss_total_v6": 0.5928313359618187, "step": 6681 }, { "epoch": 1.5410594594594595, "grad_norm": 30.672054290771484, "learning_rate": 6.881111905987503e-07, "loss": 8.756519317626953, "loss_alignment": 0.75927734375, "loss_alignment_w": 0.379638671875, "loss_sft": 0.16686558723449707, "loss_total_v6": 0.5472824573516846, "step": 6682 }, { "epoch": 1.54129009009009, "grad_norm": 34.72328567504883, "learning_rate": 6.870920022422551e-07, "loss": 8.880126953125, "loss_alignment": 0.73779296875, "loss_alignment_w": 0.368896484375, "loss_sft": 0.1858367584645748, "loss_total_v6": 0.5550079047679901, "step": 6683 }, { "epoch": 1.5415207207207207, "grad_norm": 28.163829803466797, "learning_rate": 6.860735135355812e-07, "loss": 8.975197792053223, "loss_alignment": 0.73828125, "loss_alignment_w": 0.369140625, "loss_sft": 0.1905733235180378, "loss_total_v6": 0.5609498843550682, "step": 6684 }, { "epoch": 1.5417513513513512, "grad_norm": 31.632966995239258, "learning_rate": 6.850557246439505e-07, "loss": 9.12186050415039, "loss_alignment": 0.75830078125, "loss_alignment_w": 0.379150390625, "loss_sft": 0.19087430741637945, "loss_total_v6": 0.5701162442564964, "step": 6685 }, { "epoch": 1.541981981981982, "grad_norm": 33.1104736328125, "learning_rate": 6.84038635732473e-07, "loss": 8.973015785217285, "loss_alignment": 0.72900390625, "loss_alignment_w": 0.364501953125, "loss_sft": 0.1966929966583848, "loss_total_v6": 0.5608134791254997, "step": 6686 }, { "epoch": 1.5422126126126128, "grad_norm": 31.40317726135254, "learning_rate": 6.830222469661419e-07, "loss": 9.414052963256836, "loss_alignment": 0.7705078125, "loss_alignment_w": 0.38525390625, "loss_sft": 0.20346012525260448, "loss_total_v6": 0.5883783176541328, "step": 6687 }, { "epoch": 1.5424432432432433, "grad_norm": 32.11691665649414, "learning_rate": 6.820065585098379e-07, "loss": 9.514124870300293, "loss_alignment": 0.7724609375, "loss_alignment_w": 0.38623046875, "loss_sft": 0.20870756916701794, "loss_total_v6": 0.5946328490972519, "step": 6688 }, { "epoch": 1.542673873873874, "grad_norm": 30.763530731201172, "learning_rate": 6.809915705283282e-07, "loss": 8.968500137329102, "loss_alignment": 0.74853515625, "loss_alignment_w": 0.374267578125, "loss_sft": 0.18730123154819012, "loss_total_v6": 0.5605312138795853, "step": 6689 }, { "epoch": 1.5429045045045044, "grad_norm": 30.47683334350586, "learning_rate": 6.799772831862661e-07, "loss": 9.424296379089355, "loss_alignment": 0.77099609375, "loss_alignment_w": 0.385498046875, "loss_sft": 0.20388666354119778, "loss_total_v6": 0.5890185236930847, "step": 6690 }, { "epoch": 1.543135135135135, "grad_norm": 29.28960609436035, "learning_rate": 6.789636966481916e-07, "loss": 9.478033065795898, "loss_alignment": 0.77392578125, "loss_alignment_w": 0.386962890625, "loss_sft": 0.20568879880011082, "loss_total_v6": 0.5923770517110825, "step": 6691 }, { "epoch": 1.5433657657657658, "grad_norm": 33.242210388183594, "learning_rate": 6.779508110785332e-07, "loss": 8.976720809936523, "loss_alignment": 0.767578125, "loss_alignment_w": 0.3837890625, "loss_sft": 0.17635569721460342, "loss_total_v6": 0.5610450282692909, "step": 6692 }, { "epoch": 1.5435963963963963, "grad_norm": 31.145729064941406, "learning_rate": 6.769386266416017e-07, "loss": 9.390630722045898, "loss_alignment": 0.75927734375, "loss_alignment_w": 0.379638671875, "loss_sft": 0.20782502088695765, "loss_total_v6": 0.5869143828749657, "step": 6693 }, { "epoch": 1.5438270270270271, "grad_norm": 34.69143295288086, "learning_rate": 6.759271435015962e-07, "loss": 8.96845531463623, "loss_alignment": 0.74072265625, "loss_alignment_w": 0.370361328125, "loss_sft": 0.19024344347417355, "loss_total_v6": 0.5605284720659256, "step": 6694 }, { "epoch": 1.5440576576576577, "grad_norm": 29.626462936401367, "learning_rate": 6.749163618226006e-07, "loss": 9.475631713867188, "loss_alignment": 0.74853515625, "loss_alignment_w": 0.374267578125, "loss_sft": 0.21820353716611862, "loss_total_v6": 0.5922269932925701, "step": 6695 }, { "epoch": 1.5442882882882882, "grad_norm": 33.544681549072266, "learning_rate": 6.739062817685893e-07, "loss": 9.013540267944336, "loss_alignment": 0.78662109375, "loss_alignment_w": 0.393310546875, "loss_sft": 0.17055456712841988, "loss_total_v6": 0.5633463263511658, "step": 6696 }, { "epoch": 1.5445189189189188, "grad_norm": 34.324222564697266, "learning_rate": 6.728969035034161e-07, "loss": 9.260759353637695, "loss_alignment": 0.751953125, "loss_alignment_w": 0.3759765625, "loss_sft": 0.20131024159491062, "loss_total_v6": 0.5787974521517754, "step": 6697 }, { "epoch": 1.5447495495495496, "grad_norm": 32.287628173828125, "learning_rate": 6.718882271908278e-07, "loss": 8.864950180053711, "loss_alignment": 0.748046875, "loss_alignment_w": 0.3740234375, "loss_sft": 0.17886104620993137, "loss_total_v6": 0.5540594197809696, "step": 6698 }, { "epoch": 1.5449801801801801, "grad_norm": 34.393585205078125, "learning_rate": 6.70880252994453e-07, "loss": 8.955489158630371, "loss_alignment": 0.7265625, "loss_alignment_w": 0.36328125, "loss_sft": 0.19742867164313793, "loss_total_v6": 0.5597181022167206, "step": 6699 }, { "epoch": 1.545210810810811, "grad_norm": 30.31242561340332, "learning_rate": 6.698729810778065e-07, "loss": 9.0693359375, "loss_alignment": 0.744140625, "loss_alignment_w": 0.3720703125, "loss_sft": 0.1951293908059597, "loss_total_v6": 0.56683349609375, "step": 6700 }, { "epoch": 1.5454414414414415, "grad_norm": 36.971038818359375, "learning_rate": 6.688664116042926e-07, "loss": 9.088623046875, "loss_alignment": 0.77783203125, "loss_alignment_w": 0.388916015625, "loss_sft": 0.17895509861409664, "loss_total_v6": 0.5680389627814293, "step": 6701 }, { "epoch": 1.545672072072072, "grad_norm": 33.26077651977539, "learning_rate": 6.678605447371978e-07, "loss": 9.331003189086914, "loss_alignment": 0.796875, "loss_alignment_w": 0.3984375, "loss_sft": 0.18478072993457317, "loss_total_v6": 0.5831877142190933, "step": 6702 }, { "epoch": 1.5459027027027026, "grad_norm": 32.23991775512695, "learning_rate": 6.668553806396949e-07, "loss": 9.057645797729492, "loss_alignment": 0.7431640625, "loss_alignment_w": 0.37158203125, "loss_sft": 0.1949937790632248, "loss_total_v6": 0.5661028102040291, "step": 6703 }, { "epoch": 1.5461333333333334, "grad_norm": 36.11677169799805, "learning_rate": 6.658509194748463e-07, "loss": 8.868216514587402, "loss_alignment": 0.74609375, "loss_alignment_w": 0.373046875, "loss_sft": 0.1810335274785757, "loss_total_v6": 0.5542635172605515, "step": 6704 }, { "epoch": 1.546363963963964, "grad_norm": 33.16103744506836, "learning_rate": 6.648471614055963e-07, "loss": 8.973897933959961, "loss_alignment": 0.73681640625, "loss_alignment_w": 0.368408203125, "loss_sft": 0.19215518422424793, "loss_total_v6": 0.5608685687184334, "step": 6705 }, { "epoch": 1.5465945945945947, "grad_norm": 30.438129425048828, "learning_rate": 6.638441065947782e-07, "loss": 8.791759490966797, "loss_alignment": 0.71142578125, "loss_alignment_w": 0.355712890625, "loss_sft": 0.1943977139890194, "loss_total_v6": 0.5494849979877472, "step": 6706 }, { "epoch": 1.5468252252252253, "grad_norm": 35.15589904785156, "learning_rate": 6.628417552051098e-07, "loss": 8.918169021606445, "loss_alignment": 0.71630859375, "loss_alignment_w": 0.358154296875, "loss_sft": 0.19909393042325974, "loss_total_v6": 0.5573855340480804, "step": 6707 }, { "epoch": 1.5470558558558558, "grad_norm": 34.679935455322266, "learning_rate": 6.618401073991937e-07, "loss": 8.786418914794922, "loss_alignment": 0.7197265625, "loss_alignment_w": 0.35986328125, "loss_sft": 0.1882808767259121, "loss_total_v6": 0.549151211977005, "step": 6708 }, { "epoch": 1.5472864864864864, "grad_norm": 34.34612274169922, "learning_rate": 6.608391633395195e-07, "loss": 8.687314987182617, "loss_alignment": 0.71630859375, "loss_alignment_w": 0.358154296875, "loss_sft": 0.18480290099978447, "loss_total_v6": 0.5429572016000748, "step": 6709 }, { "epoch": 1.5475171171171171, "grad_norm": 32.120452880859375, "learning_rate": 6.598389231884628e-07, "loss": 8.939109802246094, "loss_alignment": 0.7236328125, "loss_alignment_w": 0.36181640625, "loss_sft": 0.1976714301854372, "loss_total_v6": 0.5586943998932838, "step": 6710 }, { "epoch": 1.5477477477477477, "grad_norm": 27.4657039642334, "learning_rate": 6.58839387108286e-07, "loss": 8.97068977355957, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.17646710108965635, "loss_total_v6": 0.5606681369245052, "step": 6711 }, { "epoch": 1.5479783783783785, "grad_norm": 31.51811408996582, "learning_rate": 6.578405552611361e-07, "loss": 8.688414573669434, "loss_alignment": 0.73388671875, "loss_alignment_w": 0.366943359375, "loss_sft": 0.1758842021226883, "loss_total_v6": 0.5430259108543396, "step": 6712 }, { "epoch": 1.548209009009009, "grad_norm": 32.79857635498047, "learning_rate": 6.568424278090446e-07, "loss": 8.987045288085938, "loss_alignment": 0.728515625, "loss_alignment_w": 0.3642578125, "loss_sft": 0.1980276145040989, "loss_total_v6": 0.5616903230547905, "step": 6713 }, { "epoch": 1.5484396396396396, "grad_norm": 31.512773513793945, "learning_rate": 6.558450049139297e-07, "loss": 9.480085372924805, "loss_alignment": 0.81982421875, "loss_alignment_w": 0.409912109375, "loss_sft": 0.18334086425602436, "loss_total_v6": 0.5925052836537361, "step": 6714 }, { "epoch": 1.5486702702702702, "grad_norm": 36.320987701416016, "learning_rate": 6.548482867375971e-07, "loss": 9.286087036132812, "loss_alignment": 0.77001953125, "loss_alignment_w": 0.385009765625, "loss_sft": 0.19576741009950638, "loss_total_v6": 0.5803804323077202, "step": 6715 }, { "epoch": 1.548900900900901, "grad_norm": 30.715351104736328, "learning_rate": 6.538522734417357e-07, "loss": 9.058067321777344, "loss_alignment": 0.7529296875, "loss_alignment_w": 0.37646484375, "loss_sft": 0.190885073505342, "loss_total_v6": 0.5661292225122452, "step": 6716 }, { "epoch": 1.5491315315315315, "grad_norm": 27.45846176147461, "learning_rate": 6.528569651879224e-07, "loss": 9.357621192932129, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.19797997176647186, "loss_total_v6": 0.5848513171076775, "step": 6717 }, { "epoch": 1.5493621621621623, "grad_norm": 29.66644859313965, "learning_rate": 6.518623621376174e-07, "loss": 9.014017105102539, "loss_alignment": 0.744140625, "loss_alignment_w": 0.3720703125, "loss_sft": 0.19063438568264246, "loss_total_v6": 0.5633761025965214, "step": 6718 }, { "epoch": 1.5495927927927928, "grad_norm": 33.04835891723633, "learning_rate": 6.508684644521662e-07, "loss": 8.699825286865234, "loss_alignment": 0.712890625, "loss_alignment_w": 0.3564453125, "loss_sft": 0.18619515281170607, "loss_total_v6": 0.5437390878796577, "step": 6719 }, { "epoch": 1.5498234234234234, "grad_norm": 31.45037269592285, "learning_rate": 6.498752722928042e-07, "loss": 9.212671279907227, "loss_alignment": 0.77490234375, "loss_alignment_w": 0.387451171875, "loss_sft": 0.1879592351615429, "loss_total_v6": 0.5757918953895569, "step": 6720 }, { "epoch": 1.550054054054054, "grad_norm": 29.79319190979004, "learning_rate": 6.48882785820647e-07, "loss": 8.42845630645752, "loss_alignment": 0.72412109375, "loss_alignment_w": 0.362060546875, "loss_sft": 0.16406184621155262, "loss_total_v6": 0.5267785266041756, "step": 6721 }, { "epoch": 1.5502846846846847, "grad_norm": 30.964752197265625, "learning_rate": 6.47891005196698e-07, "loss": 8.492607116699219, "loss_alignment": 0.72412109375, "loss_alignment_w": 0.362060546875, "loss_sft": 0.16836119908839464, "loss_total_v6": 0.5307879708707333, "step": 6722 }, { "epoch": 1.5505153153153153, "grad_norm": 30.57882308959961, "learning_rate": 6.468999305818474e-07, "loss": 8.692130088806152, "loss_alignment": 0.7353515625, "loss_alignment_w": 0.36767578125, "loss_sft": 0.17538396921008825, "loss_total_v6": 0.5432581380009651, "step": 6723 }, { "epoch": 1.550745945945946, "grad_norm": 30.554393768310547, "learning_rate": 6.459095621368682e-07, "loss": 8.640487670898438, "loss_alignment": 0.68603515625, "loss_alignment_w": 0.343017578125, "loss_sft": 0.19638723880052567, "loss_total_v6": 0.5400304421782494, "step": 6724 }, { "epoch": 1.5509765765765766, "grad_norm": 52.6024055480957, "learning_rate": 6.449199000224221e-07, "loss": 9.212154388427734, "loss_alignment": 0.77734375, "loss_alignment_w": 0.388671875, "loss_sft": 0.18630957044661045, "loss_total_v6": 0.5757596641778946, "step": 6725 }, { "epoch": 1.5512072072072072, "grad_norm": 28.076738357543945, "learning_rate": 6.439309443990532e-07, "loss": 8.850743293762207, "loss_alignment": 0.73681640625, "loss_alignment_w": 0.368408203125, "loss_sft": 0.18557201139628887, "loss_total_v6": 0.5531714595854282, "step": 6726 }, { "epoch": 1.5514378378378377, "grad_norm": 33.403594970703125, "learning_rate": 6.429426954271923e-07, "loss": 9.214912414550781, "loss_alignment": 0.759765625, "loss_alignment_w": 0.3798828125, "loss_sft": 0.19636959582567215, "loss_total_v6": 0.575931966304779, "step": 6727 }, { "epoch": 1.5516684684684685, "grad_norm": 33.02140426635742, "learning_rate": 6.419551532671542e-07, "loss": 8.629035949707031, "loss_alignment": 0.72216796875, "loss_alignment_w": 0.361083984375, "loss_sft": 0.1781086577102542, "loss_total_v6": 0.5393146984279156, "step": 6728 }, { "epoch": 1.551899099099099, "grad_norm": 41.15239715576172, "learning_rate": 6.409683180791415e-07, "loss": 9.191975593566895, "loss_alignment": 0.7548828125, "loss_alignment_w": 0.37744140625, "loss_sft": 0.19673660956323147, "loss_total_v6": 0.5744984447956085, "step": 6729 }, { "epoch": 1.5521297297297298, "grad_norm": 30.7864990234375, "learning_rate": 6.399821900232417e-07, "loss": 9.133369445800781, "loss_alignment": 0.7373046875, "loss_alignment_w": 0.36865234375, "loss_sft": 0.20265624672174454, "loss_total_v6": 0.5708355903625488, "step": 6730 }, { "epoch": 1.5523603603603604, "grad_norm": 33.44023132324219, "learning_rate": 6.389967692594256e-07, "loss": 9.077831268310547, "loss_alignment": 0.7587890625, "loss_alignment_w": 0.37939453125, "loss_sft": 0.18816835805773735, "loss_total_v6": 0.5673645175993443, "step": 6731 }, { "epoch": 1.552590990990991, "grad_norm": 33.28827667236328, "learning_rate": 6.380120559475505e-07, "loss": 8.88812255859375, "loss_alignment": 0.73046875, "loss_alignment_w": 0.365234375, "loss_sft": 0.19053268246352673, "loss_total_v6": 0.55550766736269, "step": 6732 }, { "epoch": 1.5528216216216215, "grad_norm": 36.73570251464844, "learning_rate": 6.370280502473597e-07, "loss": 9.348432540893555, "loss_alignment": 0.75830078125, "loss_alignment_w": 0.379150390625, "loss_sft": 0.2054775981232524, "loss_total_v6": 0.5842770412564278, "step": 6733 }, { "epoch": 1.5530522522522523, "grad_norm": 41.39202880859375, "learning_rate": 6.360447523184782e-07, "loss": 9.100542068481445, "loss_alignment": 0.765625, "loss_alignment_w": 0.3828125, "loss_sft": 0.18650541827082634, "loss_total_v6": 0.5687838718295097, "step": 6734 }, { "epoch": 1.5532828828828829, "grad_norm": 31.709228515625, "learning_rate": 6.350621623204212e-07, "loss": 8.779434204101562, "loss_alignment": 0.71533203125, "loss_alignment_w": 0.357666015625, "loss_sft": 0.19083502516150475, "loss_total_v6": 0.5487146750092506, "step": 6735 }, { "epoch": 1.5535135135135136, "grad_norm": 34.12590026855469, "learning_rate": 6.340802804125872e-07, "loss": 9.662273406982422, "loss_alignment": 0.77392578125, "loss_alignment_w": 0.386962890625, "loss_sft": 0.21699021942913532, "loss_total_v6": 0.603892095386982, "step": 6736 }, { "epoch": 1.5537441441441442, "grad_norm": 33.597164154052734, "learning_rate": 6.330991067542586e-07, "loss": 9.418190002441406, "loss_alignment": 0.72900390625, "loss_alignment_w": 0.364501953125, "loss_sft": 0.22497415635734797, "loss_total_v6": 0.5886368677020073, "step": 6737 }, { "epoch": 1.5539747747747747, "grad_norm": 39.887107849121094, "learning_rate": 6.321186415046038e-07, "loss": 9.070304870605469, "loss_alignment": 0.76220703125, "loss_alignment_w": 0.381103515625, "loss_sft": 0.1853327341377735, "loss_total_v6": 0.5668940171599388, "step": 6738 }, { "epoch": 1.5542054054054053, "grad_norm": 29.368587493896484, "learning_rate": 6.311388848226741e-07, "loss": 9.108123779296875, "loss_alignment": 0.77978515625, "loss_alignment_w": 0.389892578125, "loss_sft": 0.1799449659883976, "loss_total_v6": 0.5692577213048935, "step": 6739 }, { "epoch": 1.554436036036036, "grad_norm": 32.16875457763672, "learning_rate": 6.301598368674106e-07, "loss": 9.375564575195312, "loss_alignment": 0.755859375, "loss_alignment_w": 0.3779296875, "loss_sft": 0.20860771276056767, "loss_total_v6": 0.5859728083014488, "step": 6740 }, { "epoch": 1.5546666666666666, "grad_norm": 30.849027633666992, "learning_rate": 6.291814977976352e-07, "loss": 8.735713958740234, "loss_alignment": 0.72900390625, "loss_alignment_w": 0.364501953125, "loss_sft": 0.18183108419179916, "loss_total_v6": 0.5459820777177811, "step": 6741 }, { "epoch": 1.5548972972972974, "grad_norm": 27.78065299987793, "learning_rate": 6.28203867772057e-07, "loss": 8.703621864318848, "loss_alignment": 0.73974609375, "loss_alignment_w": 0.369873046875, "loss_sft": 0.1743932580575347, "loss_total_v6": 0.5439763888716698, "step": 6742 }, { "epoch": 1.555127927927928, "grad_norm": 33.52186584472656, "learning_rate": 6.272269469492697e-07, "loss": 9.194671630859375, "loss_alignment": 0.7900390625, "loss_alignment_w": 0.39501953125, "loss_sft": 0.18024257477372885, "loss_total_v6": 0.5746670290827751, "step": 6743 }, { "epoch": 1.5553585585585585, "grad_norm": 33.238887786865234, "learning_rate": 6.262507354877495e-07, "loss": 9.03441047668457, "loss_alignment": 0.767578125, "loss_alignment_w": 0.3837890625, "loss_sft": 0.1809683870524168, "loss_total_v6": 0.5646506249904633, "step": 6744 }, { "epoch": 1.555589189189189, "grad_norm": 31.607717514038086, "learning_rate": 6.252752335458629e-07, "loss": 9.163681983947754, "loss_alignment": 0.7705078125, "loss_alignment_w": 0.38525390625, "loss_sft": 0.18677431717514992, "loss_total_v6": 0.5727301351726055, "step": 6745 }, { "epoch": 1.5558198198198199, "grad_norm": 27.41339874267578, "learning_rate": 6.243004412818559e-07, "loss": 8.733739852905273, "loss_alignment": 0.7294921875, "loss_alignment_w": 0.36474609375, "loss_sft": 0.1815093420445919, "loss_total_v6": 0.5458587147295475, "step": 6746 }, { "epoch": 1.5560504504504504, "grad_norm": 28.581018447875977, "learning_rate": 6.233263588538612e-07, "loss": 8.525564193725586, "loss_alignment": 0.708984375, "loss_alignment_w": 0.3544921875, "loss_sft": 0.1783250654116273, "loss_total_v6": 0.5328477546572685, "step": 6747 }, { "epoch": 1.5562810810810812, "grad_norm": 31.63334846496582, "learning_rate": 6.223529864198985e-07, "loss": 9.00898551940918, "loss_alignment": 0.76416015625, "loss_alignment_w": 0.382080078125, "loss_sft": 0.18101196456700563, "loss_total_v6": 0.5630615502595901, "step": 6748 }, { "epoch": 1.5565117117117118, "grad_norm": 32.62205123901367, "learning_rate": 6.213803241378686e-07, "loss": 8.913040161132812, "loss_alignment": 0.732421875, "loss_alignment_w": 0.3662109375, "loss_sft": 0.19102183729410172, "loss_total_v6": 0.5570649579167366, "step": 6749 }, { "epoch": 1.5567423423423423, "grad_norm": 31.491153717041016, "learning_rate": 6.204083721655607e-07, "loss": 9.777088165283203, "loss_alignment": 0.79638671875, "loss_alignment_w": 0.398193359375, "loss_sft": 0.21253899298608303, "loss_total_v6": 0.6110680475831032, "step": 6750 }, { "epoch": 1.5569729729729729, "grad_norm": 35.98027038574219, "learning_rate": 6.194371306606467e-07, "loss": 8.72642707824707, "loss_alignment": 0.74609375, "loss_alignment_w": 0.373046875, "loss_sft": 0.17256843484938145, "loss_total_v6": 0.5454016849398613, "step": 6751 }, { "epoch": 1.5572036036036037, "grad_norm": 33.93722152709961, "learning_rate": 6.184665997806832e-07, "loss": 9.059216499328613, "loss_alignment": 0.7802734375, "loss_alignment_w": 0.39013671875, "loss_sft": 0.17603381909430027, "loss_total_v6": 0.5662010535597801, "step": 6752 }, { "epoch": 1.5574342342342342, "grad_norm": 34.665748596191406, "learning_rate": 6.174967796831105e-07, "loss": 8.879596710205078, "loss_alignment": 0.75341796875, "loss_alignment_w": 0.376708984375, "loss_sft": 0.17831157334148884, "loss_total_v6": 0.5549747720360756, "step": 6753 }, { "epoch": 1.557664864864865, "grad_norm": 33.45756530761719, "learning_rate": 6.165276705252565e-07, "loss": 8.568313598632812, "loss_alignment": 0.7421875, "loss_alignment_w": 0.37109375, "loss_sft": 0.1637849621474743, "loss_total_v6": 0.5355195924639702, "step": 6754 }, { "epoch": 1.5578954954954956, "grad_norm": 31.864099502563477, "learning_rate": 6.155592724643339e-07, "loss": 9.396108627319336, "loss_alignment": 0.8017578125, "loss_alignment_w": 0.40087890625, "loss_sft": 0.18553862161934376, "loss_total_v6": 0.5872567594051361, "step": 6755 }, { "epoch": 1.558126126126126, "grad_norm": 35.012874603271484, "learning_rate": 6.145915856574364e-07, "loss": 8.727169036865234, "loss_alignment": 0.73974609375, "loss_alignment_w": 0.369873046875, "loss_sft": 0.17501043528318405, "loss_total_v6": 0.5454480797052383, "step": 6756 }, { "epoch": 1.5583567567567567, "grad_norm": 30.579191207885742, "learning_rate": 6.136246102615451e-07, "loss": 9.153769493103027, "loss_alignment": 0.7490234375, "loss_alignment_w": 0.37451171875, "loss_sft": 0.1981024369597435, "loss_total_v6": 0.5721106007695198, "step": 6757 }, { "epoch": 1.5585873873873872, "grad_norm": 31.320953369140625, "learning_rate": 6.126583464335239e-07, "loss": 9.148384094238281, "loss_alignment": 0.74365234375, "loss_alignment_w": 0.371826171875, "loss_sft": 0.1997189074754715, "loss_total_v6": 0.571773961186409, "step": 6758 }, { "epoch": 1.558818018018018, "grad_norm": 31.322799682617188, "learning_rate": 6.116927943301237e-07, "loss": 8.87562084197998, "loss_alignment": 0.72705078125, "loss_alignment_w": 0.363525390625, "loss_sft": 0.19147556647658348, "loss_total_v6": 0.5547262877225876, "step": 6759 }, { "epoch": 1.5590486486486488, "grad_norm": 29.784879684448242, "learning_rate": 6.107279541079769e-07, "loss": 8.738433837890625, "loss_alignment": 0.73046875, "loss_alignment_w": 0.365234375, "loss_sft": 0.18064312264323235, "loss_total_v6": 0.5461521446704865, "step": 6760 }, { "epoch": 1.5592792792792793, "grad_norm": 32.0879020690918, "learning_rate": 6.09763825923605e-07, "loss": 9.166622161865234, "loss_alignment": 0.77294921875, "loss_alignment_w": 0.386474609375, "loss_sft": 0.18675976060330868, "loss_total_v6": 0.5729139149188995, "step": 6761 }, { "epoch": 1.55950990990991, "grad_norm": 38.38193130493164, "learning_rate": 6.088004099334083e-07, "loss": 9.025489807128906, "loss_alignment": 0.7509765625, "loss_alignment_w": 0.37548828125, "loss_sft": 0.18767407722771168, "loss_total_v6": 0.5640931501984596, "step": 6762 }, { "epoch": 1.5597405405405405, "grad_norm": 34.527095794677734, "learning_rate": 6.078377062936752e-07, "loss": 8.9862060546875, "loss_alignment": 0.76416015625, "loss_alignment_w": 0.382080078125, "loss_sft": 0.17836763709783554, "loss_total_v6": 0.5616379007697105, "step": 6763 }, { "epoch": 1.559971171171171, "grad_norm": 27.32448959350586, "learning_rate": 6.06875715160578e-07, "loss": 9.451460838317871, "loss_alignment": 0.77294921875, "loss_alignment_w": 0.386474609375, "loss_sft": 0.20460790395736694, "loss_total_v6": 0.5907162949442863, "step": 6764 }, { "epoch": 1.5602018018018018, "grad_norm": 33.09762191772461, "learning_rate": 6.059144366901737e-07, "loss": 9.082849502563477, "loss_alignment": 0.76806640625, "loss_alignment_w": 0.384033203125, "loss_sft": 0.1832482237368822, "loss_total_v6": 0.5676781386137009, "step": 6765 }, { "epoch": 1.5604324324324326, "grad_norm": 34.66225051879883, "learning_rate": 6.049538710384012e-07, "loss": 9.161212921142578, "loss_alignment": 0.7294921875, "loss_alignment_w": 0.36474609375, "loss_sft": 0.20811966620385647, "loss_total_v6": 0.5725758448243141, "step": 6766 }, { "epoch": 1.5606630630630631, "grad_norm": 32.50428771972656, "learning_rate": 6.039940183610877e-07, "loss": 8.891857147216797, "loss_alignment": 0.744140625, "loss_alignment_w": 0.3720703125, "loss_sft": 0.1833655796945095, "loss_total_v6": 0.5557410791516304, "step": 6767 }, { "epoch": 1.5608936936936937, "grad_norm": 30.744651794433594, "learning_rate": 6.030348788139406e-07, "loss": 8.670389175415039, "loss_alignment": 0.74365234375, "loss_alignment_w": 0.371826171875, "loss_sft": 0.17004266940057278, "loss_total_v6": 0.5418993607163429, "step": 6768 }, { "epoch": 1.5611243243243242, "grad_norm": 34.00824737548828, "learning_rate": 6.020764525525563e-07, "loss": 8.82815170288086, "loss_alignment": 0.71435546875, "loss_alignment_w": 0.357177734375, "loss_sft": 0.1939561516046524, "loss_total_v6": 0.5517594963312149, "step": 6769 }, { "epoch": 1.5613549549549548, "grad_norm": 32.69343566894531, "learning_rate": 6.011187397324114e-07, "loss": 9.566238403320312, "loss_alignment": 0.7451171875, "loss_alignment_w": 0.37255859375, "loss_sft": 0.22487354651093483, "loss_total_v6": 0.5978898927569389, "step": 6770 }, { "epoch": 1.5615855855855856, "grad_norm": 30.894071578979492, "learning_rate": 6.001617405088683e-07, "loss": 9.081609725952148, "loss_alignment": 0.76904296875, "loss_alignment_w": 0.384521484375, "loss_sft": 0.18390314560383558, "loss_total_v6": 0.5676006525754929, "step": 6771 }, { "epoch": 1.5618162162162164, "grad_norm": 36.304771423339844, "learning_rate": 5.992054550371723e-07, "loss": 8.732975959777832, "loss_alignment": 0.70361328125, "loss_alignment_w": 0.351806640625, "loss_sft": 0.19412639178335667, "loss_total_v6": 0.5458109639585018, "step": 6772 }, { "epoch": 1.562046846846847, "grad_norm": 34.68620681762695, "learning_rate": 5.98249883472456e-07, "loss": 8.779886245727539, "loss_alignment": 0.7802734375, "loss_alignment_w": 0.39013671875, "loss_sft": 0.15910965763032436, "loss_total_v6": 0.5487428680062294, "step": 6773 }, { "epoch": 1.5622774774774775, "grad_norm": 31.06509780883789, "learning_rate": 5.972950259697352e-07, "loss": 9.103036880493164, "loss_alignment": 0.74365234375, "loss_alignment_w": 0.371826171875, "loss_sft": 0.19689996913075447, "loss_total_v6": 0.5689397528767586, "step": 6774 }, { "epoch": 1.562508108108108, "grad_norm": 29.976167678833008, "learning_rate": 5.963408826839079e-07, "loss": 8.969285011291504, "loss_alignment": 0.72998046875, "loss_alignment_w": 0.364990234375, "loss_sft": 0.19436939992010593, "loss_total_v6": 0.560580313205719, "step": 6775 }, { "epoch": 1.5627387387387386, "grad_norm": 28.599822998046875, "learning_rate": 5.953874537697573e-07, "loss": 8.867985725402832, "loss_alignment": 0.73193359375, "loss_alignment_w": 0.365966796875, "loss_sft": 0.18832803145051003, "loss_total_v6": 0.5542490780353546, "step": 6776 }, { "epoch": 1.5629693693693694, "grad_norm": 32.601383209228516, "learning_rate": 5.944347393819511e-07, "loss": 8.618677139282227, "loss_alignment": 0.72119140625, "loss_alignment_w": 0.360595703125, "loss_sft": 0.1777054462581873, "loss_total_v6": 0.5386673584580421, "step": 6777 }, { "epoch": 1.5632000000000001, "grad_norm": 31.361867904663086, "learning_rate": 5.934827396750392e-07, "loss": 9.27615737915039, "loss_alignment": 0.744140625, "loss_alignment_w": 0.3720703125, "loss_sft": 0.20741485804319382, "loss_total_v6": 0.579759806394577, "step": 6778 }, { "epoch": 1.5634306306306307, "grad_norm": 33.57398223876953, "learning_rate": 5.925314548034588e-07, "loss": 9.010234832763672, "loss_alignment": 0.7919921875, "loss_alignment_w": 0.39599609375, "loss_sft": 0.1657092534005642, "loss_total_v6": 0.5631396770477295, "step": 6779 }, { "epoch": 1.5636612612612613, "grad_norm": 31.563718795776367, "learning_rate": 5.915808849215304e-07, "loss": 8.835506439208984, "loss_alignment": 0.77490234375, "loss_alignment_w": 0.387451171875, "loss_sft": 0.1642033662647009, "loss_total_v6": 0.5522191300988197, "step": 6780 }, { "epoch": 1.5638918918918918, "grad_norm": 30.48030662536621, "learning_rate": 5.906310301834567e-07, "loss": 9.313752174377441, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.1940327174961567, "loss_total_v6": 0.5821095034480095, "step": 6781 }, { "epoch": 1.5641225225225224, "grad_norm": 37.54851150512695, "learning_rate": 5.896818907433255e-07, "loss": 9.558712005615234, "loss_alignment": 0.78125, "loss_alignment_w": 0.390625, "loss_sft": 0.20690138079226017, "loss_total_v6": 0.5974195525050163, "step": 6782 }, { "epoch": 1.5643531531531532, "grad_norm": 31.150562286376953, "learning_rate": 5.887334667551065e-07, "loss": 8.750641822814941, "loss_alignment": 0.7646484375, "loss_alignment_w": 0.38232421875, "loss_sft": 0.164835037663579, "loss_total_v6": 0.5469151213765144, "step": 6783 }, { "epoch": 1.564583783783784, "grad_norm": 36.62806701660156, "learning_rate": 5.877857583726576e-07, "loss": 9.260448455810547, "loss_alignment": 0.76806640625, "loss_alignment_w": 0.384033203125, "loss_sft": 0.19527893513441086, "loss_total_v6": 0.5787780731916428, "step": 6784 }, { "epoch": 1.5648144144144145, "grad_norm": 28.06264877319336, "learning_rate": 5.868387657497171e-07, "loss": 9.248589515686035, "loss_alignment": 0.75341796875, "loss_alignment_w": 0.376708984375, "loss_sft": 0.20212134718894958, "loss_total_v6": 0.5780368745326996, "step": 6785 }, { "epoch": 1.565045045045045, "grad_norm": 28.268165588378906, "learning_rate": 5.858924890399093e-07, "loss": 8.680610656738281, "loss_alignment": 0.76513671875, "loss_alignment_w": 0.382568359375, "loss_sft": 0.1603207541629672, "loss_total_v6": 0.5425381809473038, "step": 6786 }, { "epoch": 1.5652756756756756, "grad_norm": 31.03791046142578, "learning_rate": 5.849469283967407e-07, "loss": 8.266188621520996, "loss_alignment": 0.68994140625, "loss_alignment_w": 0.344970703125, "loss_sft": 0.17183390352874994, "loss_total_v6": 0.5166367627680302, "step": 6787 }, { "epoch": 1.5655063063063062, "grad_norm": 32.77495193481445, "learning_rate": 5.84002083973601e-07, "loss": 8.529788970947266, "loss_alignment": 0.74267578125, "loss_alignment_w": 0.371337890625, "loss_sft": 0.16229271609336138, "loss_total_v6": 0.5331118069589138, "step": 6788 }, { "epoch": 1.565736936936937, "grad_norm": 27.780948638916016, "learning_rate": 5.830579559237676e-07, "loss": 9.005289077758789, "loss_alignment": 0.7412109375, "loss_alignment_w": 0.37060546875, "loss_sft": 0.1922556273639202, "loss_total_v6": 0.5628305599093437, "step": 6789 }, { "epoch": 1.5659675675675677, "grad_norm": 33.25018310546875, "learning_rate": 5.821145444003978e-07, "loss": 8.938970565795898, "loss_alignment": 0.76611328125, "loss_alignment_w": 0.383056640625, "loss_sft": 0.1760257799178362, "loss_total_v6": 0.5586856864392757, "step": 6790 }, { "epoch": 1.5661981981981983, "grad_norm": 36.95413589477539, "learning_rate": 5.811718495565327e-07, "loss": 8.542431831359863, "loss_alignment": 0.74658203125, "loss_alignment_w": 0.373291015625, "loss_sft": 0.16056520398706198, "loss_total_v6": 0.5339020006358624, "step": 6791 }, { "epoch": 1.5664288288288288, "grad_norm": 28.098711013793945, "learning_rate": 5.802298715451016e-07, "loss": 8.494604110717773, "loss_alignment": 0.724609375, "loss_alignment_w": 0.3623046875, "loss_sft": 0.16879119165241718, "loss_total_v6": 0.5309127792716026, "step": 6792 }, { "epoch": 1.5666594594594594, "grad_norm": 29.67628288269043, "learning_rate": 5.792886105189105e-07, "loss": 8.856030464172363, "loss_alignment": 0.70947265625, "loss_alignment_w": 0.354736328125, "loss_sft": 0.1970413364470005, "loss_total_v6": 0.5535019189119339, "step": 6793 }, { "epoch": 1.56689009009009, "grad_norm": 35.36674118041992, "learning_rate": 5.783480666306568e-07, "loss": 9.000516891479492, "loss_alignment": 0.779296875, "loss_alignment_w": 0.3896484375, "loss_sft": 0.1731890207156539, "loss_total_v6": 0.5625322833657265, "step": 6794 }, { "epoch": 1.5671207207207207, "grad_norm": 32.83179473876953, "learning_rate": 5.774082400329151e-07, "loss": 8.965080261230469, "loss_alignment": 0.7470703125, "loss_alignment_w": 0.37353515625, "loss_sft": 0.186675526201725, "loss_total_v6": 0.5603175051510334, "step": 6795 }, { "epoch": 1.5673513513513515, "grad_norm": 31.909090042114258, "learning_rate": 5.764691308781473e-07, "loss": 9.094919204711914, "loss_alignment": 0.7451171875, "loss_alignment_w": 0.37255859375, "loss_sft": 0.19566023163497448, "loss_total_v6": 0.5684324279427528, "step": 6796 }, { "epoch": 1.567581981981982, "grad_norm": 32.91122817993164, "learning_rate": 5.75530739318696e-07, "loss": 8.950567245483398, "loss_alignment": 0.7333984375, "loss_alignment_w": 0.36669921875, "loss_sft": 0.19222295843064785, "loss_total_v6": 0.5594104304909706, "step": 6797 }, { "epoch": 1.5678126126126126, "grad_norm": 32.46985626220703, "learning_rate": 5.74593065506791e-07, "loss": 9.40429401397705, "loss_alignment": 0.830078125, "loss_alignment_w": 0.4150390625, "loss_sft": 0.17240891326218843, "loss_total_v6": 0.5877683945000172, "step": 6798 }, { "epoch": 1.5680432432432432, "grad_norm": 28.488727569580078, "learning_rate": 5.736561095945447e-07, "loss": 9.012136459350586, "loss_alignment": 0.7529296875, "loss_alignment_w": 0.37646484375, "loss_sft": 0.18720568530261517, "loss_total_v6": 0.5632585361599922, "step": 6799 }, { "epoch": 1.5682738738738737, "grad_norm": 30.705570220947266, "learning_rate": 5.727198717339511e-07, "loss": 8.393941879272461, "loss_alignment": 0.69580078125, "loss_alignment_w": 0.347900390625, "loss_sft": 0.17685829661786556, "loss_total_v6": 0.5246213376522064, "step": 6800 }, { "epoch": 1.5685045045045045, "grad_norm": 31.967653274536133, "learning_rate": 5.717843520768884e-07, "loss": 9.3623046875, "loss_alignment": 0.76904296875, "loss_alignment_w": 0.384521484375, "loss_sft": 0.2004242241382599, "loss_total_v6": 0.5851440504193306, "step": 6801 }, { "epoch": 1.5687351351351353, "grad_norm": 34.090328216552734, "learning_rate": 5.708495507751194e-07, "loss": 9.345890998840332, "loss_alignment": 0.75537109375, "loss_alignment_w": 0.377685546875, "loss_sft": 0.20595961064100266, "loss_total_v6": 0.5841181874275208, "step": 6802 }, { "epoch": 1.5689657657657659, "grad_norm": 32.5278205871582, "learning_rate": 5.699154679802887e-07, "loss": 9.470925331115723, "loss_alignment": 0.787109375, "loss_alignment_w": 0.3935546875, "loss_sft": 0.19723374396562576, "loss_total_v6": 0.5919328257441521, "step": 6803 }, { "epoch": 1.5691963963963964, "grad_norm": 28.2291259765625, "learning_rate": 5.689821038439264e-07, "loss": 9.514135360717773, "loss_alignment": 0.75927734375, "loss_alignment_w": 0.379638671875, "loss_sft": 0.21578818000853062, "loss_total_v6": 0.5946334078907967, "step": 6804 }, { "epoch": 1.569427027027027, "grad_norm": 32.974971771240234, "learning_rate": 5.680494585174451e-07, "loss": 9.285446166992188, "loss_alignment": 0.7734375, "loss_alignment_w": 0.38671875, "loss_sft": 0.19371325336396694, "loss_total_v6": 0.5803404301404953, "step": 6805 }, { "epoch": 1.5696576576576575, "grad_norm": 42.75745391845703, "learning_rate": 5.671175321521411e-07, "loss": 9.17768383026123, "loss_alignment": 0.76123046875, "loss_alignment_w": 0.380615234375, "loss_sft": 0.1930967941880226, "loss_total_v6": 0.5736052170395851, "step": 6806 }, { "epoch": 1.5698882882882883, "grad_norm": 29.347103118896484, "learning_rate": 5.661863248991911e-07, "loss": 9.073469161987305, "loss_alignment": 0.73876953125, "loss_alignment_w": 0.369384765625, "loss_sft": 0.19717302732169628, "loss_total_v6": 0.5670918375253677, "step": 6807 }, { "epoch": 1.570118918918919, "grad_norm": 35.087406158447266, "learning_rate": 5.652558369096606e-07, "loss": 8.9941987991333, "loss_alignment": 0.736328125, "loss_alignment_w": 0.3681640625, "loss_sft": 0.194202221930027, "loss_total_v6": 0.5621374174952507, "step": 6808 }, { "epoch": 1.5703495495495496, "grad_norm": 27.242752075195312, "learning_rate": 5.643260683344947e-07, "loss": 8.86585807800293, "loss_alignment": 0.74462890625, "loss_alignment_w": 0.372314453125, "loss_sft": 0.1824730597436428, "loss_total_v6": 0.5541161224246025, "step": 6809 }, { "epoch": 1.5705801801801802, "grad_norm": 32.64211654663086, "learning_rate": 5.633970193245208e-07, "loss": 8.80722427368164, "loss_alignment": 0.72216796875, "loss_alignment_w": 0.361083984375, "loss_sft": 0.18945910595357418, "loss_total_v6": 0.5504515469074249, "step": 6810 }, { "epoch": 1.5708108108108108, "grad_norm": 34.109161376953125, "learning_rate": 5.624686900304537e-07, "loss": 9.296302795410156, "loss_alignment": 0.74755859375, "loss_alignment_w": 0.373779296875, "loss_sft": 0.20693443901836872, "loss_total_v6": 0.581018939614296, "step": 6811 }, { "epoch": 1.5710414414414413, "grad_norm": 31.216646194458008, "learning_rate": 5.615410806028875e-07, "loss": 8.696046829223633, "loss_alignment": 0.7255859375, "loss_alignment_w": 0.36279296875, "loss_sft": 0.18084727972745895, "loss_total_v6": 0.5435029193758965, "step": 6812 }, { "epoch": 1.571272072072072, "grad_norm": 27.2130126953125, "learning_rate": 5.606141911923024e-07, "loss": 9.212228775024414, "loss_alignment": 0.748046875, "loss_alignment_w": 0.3740234375, "loss_sft": 0.20116103626787663, "loss_total_v6": 0.5757642984390259, "step": 6813 }, { "epoch": 1.5715027027027026, "grad_norm": 29.588726043701172, "learning_rate": 5.596880219490603e-07, "loss": 8.89400863647461, "loss_alignment": 0.74755859375, "loss_alignment_w": 0.373779296875, "loss_sft": 0.18229463137686253, "loss_total_v6": 0.5558755770325661, "step": 6814 }, { "epoch": 1.5717333333333334, "grad_norm": 28.264911651611328, "learning_rate": 5.587625730234059e-07, "loss": 9.103522300720215, "loss_alignment": 0.74755859375, "loss_alignment_w": 0.373779296875, "loss_sft": 0.1954197082668543, "loss_total_v6": 0.5689701549708843, "step": 6815 }, { "epoch": 1.571963963963964, "grad_norm": 34.20136260986328, "learning_rate": 5.578378445654664e-07, "loss": 8.727527618408203, "loss_alignment": 0.72998046875, "loss_alignment_w": 0.364990234375, "loss_sft": 0.18019030801951885, "loss_total_v6": 0.5454704463481903, "step": 6816 }, { "epoch": 1.5721945945945945, "grad_norm": 28.998035430908203, "learning_rate": 5.569138367252553e-07, "loss": 9.512945175170898, "loss_alignment": 0.78662109375, "loss_alignment_w": 0.393310546875, "loss_sft": 0.2017062958329916, "loss_total_v6": 0.5945590734481812, "step": 6817 }, { "epoch": 1.572425225225225, "grad_norm": 35.10688400268555, "learning_rate": 5.55990549652668e-07, "loss": 9.316139221191406, "loss_alignment": 0.76904296875, "loss_alignment_w": 0.384521484375, "loss_sft": 0.19753891602158546, "loss_total_v6": 0.5822587460279465, "step": 6818 }, { "epoch": 1.5726558558558559, "grad_norm": 28.235260009765625, "learning_rate": 5.55067983497481e-07, "loss": 8.931527137756348, "loss_alignment": 0.72802734375, "loss_alignment_w": 0.364013671875, "loss_sft": 0.19391686655580997, "loss_total_v6": 0.5582204461097717, "step": 6819 }, { "epoch": 1.5728864864864864, "grad_norm": 33.324607849121094, "learning_rate": 5.541461384093549e-07, "loss": 8.797710418701172, "loss_alignment": 0.7294921875, "loss_alignment_w": 0.36474609375, "loss_sft": 0.18427160009741783, "loss_total_v6": 0.549856923520565, "step": 6820 }, { "epoch": 1.5731171171171172, "grad_norm": 33.04621505737305, "learning_rate": 5.532250145378337e-07, "loss": 9.08901309967041, "loss_alignment": 0.740234375, "loss_alignment_w": 0.3701171875, "loss_sft": 0.19754942692816257, "loss_total_v6": 0.568063348531723, "step": 6821 }, { "epoch": 1.5733477477477478, "grad_norm": 30.40576934814453, "learning_rate": 5.523046120323433e-07, "loss": 9.054450988769531, "loss_alignment": 0.751953125, "loss_alignment_w": 0.3759765625, "loss_sft": 0.19026234932243824, "loss_total_v6": 0.5659032389521599, "step": 6822 }, { "epoch": 1.5735783783783783, "grad_norm": 33.161983489990234, "learning_rate": 5.513849310421948e-07, "loss": 8.967297554016113, "loss_alignment": 0.74169921875, "loss_alignment_w": 0.370849609375, "loss_sft": 0.18908771127462387, "loss_total_v6": 0.5604561045765877, "step": 6823 }, { "epoch": 1.5738090090090089, "grad_norm": 36.79682922363281, "learning_rate": 5.504659717165812e-07, "loss": 9.075555801391602, "loss_alignment": 0.75, "loss_alignment_w": 0.375, "loss_sft": 0.19153556413948536, "loss_total_v6": 0.5672222003340721, "step": 6824 }, { "epoch": 1.5740396396396397, "grad_norm": 32.168731689453125, "learning_rate": 5.49547734204578e-07, "loss": 9.0801362991333, "loss_alignment": 0.75439453125, "loss_alignment_w": 0.377197265625, "loss_sft": 0.18938048370182514, "loss_total_v6": 0.5675085186958313, "step": 6825 }, { "epoch": 1.5742702702702702, "grad_norm": 31.611133575439453, "learning_rate": 5.486302186551434e-07, "loss": 9.022997856140137, "loss_alignment": 0.74365234375, "loss_alignment_w": 0.371826171875, "loss_sft": 0.1909820456057787, "loss_total_v6": 0.563937358558178, "step": 6826 }, { "epoch": 1.574500900900901, "grad_norm": 32.82326126098633, "learning_rate": 5.477134252171179e-07, "loss": 9.118095397949219, "loss_alignment": 0.75537109375, "loss_alignment_w": 0.377685546875, "loss_sft": 0.19259212911128998, "loss_total_v6": 0.5698809400200844, "step": 6827 }, { "epoch": 1.5747315315315316, "grad_norm": 33.94868850708008, "learning_rate": 5.467973540392274e-07, "loss": 8.825139045715332, "loss_alignment": 0.74072265625, "loss_alignment_w": 0.370361328125, "loss_sft": 0.18070630729198456, "loss_total_v6": 0.5515711829066277, "step": 6828 }, { "epoch": 1.5749621621621621, "grad_norm": 32.35527801513672, "learning_rate": 5.45882005270078e-07, "loss": 8.968117713928223, "loss_alignment": 0.72314453125, "loss_alignment_w": 0.361572265625, "loss_sft": 0.19934706203639507, "loss_total_v6": 0.5605073422193527, "step": 6829 }, { "epoch": 1.5751927927927927, "grad_norm": 33.57963562011719, "learning_rate": 5.449673790581611e-07, "loss": 8.692678451538086, "loss_alignment": 0.76513671875, "loss_alignment_w": 0.382568359375, "loss_sft": 0.16090720519423485, "loss_total_v6": 0.5432924330234528, "step": 6830 }, { "epoch": 1.5754234234234235, "grad_norm": 33.63493728637695, "learning_rate": 5.440534755518484e-07, "loss": 9.209590911865234, "loss_alignment": 0.7412109375, "loss_alignment_w": 0.37060546875, "loss_sft": 0.20470405742526054, "loss_total_v6": 0.5755994394421577, "step": 6831 }, { "epoch": 1.575654054054054, "grad_norm": 29.463735580444336, "learning_rate": 5.431402948993947e-07, "loss": 9.018199920654297, "loss_alignment": 0.7470703125, "loss_alignment_w": 0.37353515625, "loss_sft": 0.1906211068853736, "loss_total_v6": 0.56363745033741, "step": 6832 }, { "epoch": 1.5758846846846848, "grad_norm": 31.795513153076172, "learning_rate": 5.422278372489397e-07, "loss": 8.828706741333008, "loss_alignment": 0.74853515625, "loss_alignment_w": 0.374267578125, "loss_sft": 0.17676364071667194, "loss_total_v6": 0.5517941638827324, "step": 6833 }, { "epoch": 1.5761153153153153, "grad_norm": 37.021827697753906, "learning_rate": 5.413161027485042e-07, "loss": 8.94381332397461, "loss_alignment": 0.76318359375, "loss_alignment_w": 0.381591796875, "loss_sft": 0.17797642014920712, "loss_total_v6": 0.5589883551001549, "step": 6834 }, { "epoch": 1.576345945945946, "grad_norm": 36.0457763671875, "learning_rate": 5.404050915459908e-07, "loss": 8.980070114135742, "loss_alignment": 0.7451171875, "loss_alignment_w": 0.37255859375, "loss_sft": 0.18958081305027008, "loss_total_v6": 0.561254408210516, "step": 6835 }, { "epoch": 1.5765765765765765, "grad_norm": 36.74065017700195, "learning_rate": 5.394948037891867e-07, "loss": 9.306589126586914, "loss_alignment": 0.7734375, "loss_alignment_w": 0.38671875, "loss_sft": 0.1956906821578741, "loss_total_v6": 0.581661768257618, "step": 6836 }, { "epoch": 1.5768072072072072, "grad_norm": 32.267208099365234, "learning_rate": 5.385852396257601e-07, "loss": 9.464944839477539, "loss_alignment": 0.80419921875, "loss_alignment_w": 0.402099609375, "loss_sft": 0.18970360606908798, "loss_total_v6": 0.591559074819088, "step": 6837 }, { "epoch": 1.5770378378378378, "grad_norm": 35.87898254394531, "learning_rate": 5.376763992032641e-07, "loss": 9.003662109375, "loss_alignment": 0.75244140625, "loss_alignment_w": 0.376220703125, "loss_sft": 0.18734733387827873, "loss_total_v6": 0.5627288222312927, "step": 6838 }, { "epoch": 1.5772684684684686, "grad_norm": 32.419071197509766, "learning_rate": 5.367682826691322e-07, "loss": 9.192901611328125, "loss_alignment": 0.7470703125, "loss_alignment_w": 0.37353515625, "loss_sft": 0.20100592449307442, "loss_total_v6": 0.5745563432574272, "step": 6839 }, { "epoch": 1.5774990990990991, "grad_norm": 32.74751281738281, "learning_rate": 5.358608901706802e-07, "loss": 9.287452697753906, "loss_alignment": 0.7646484375, "loss_alignment_w": 0.38232421875, "loss_sft": 0.19824836403131485, "loss_total_v6": 0.5804657824337482, "step": 6840 }, { "epoch": 1.5777297297297297, "grad_norm": 32.08269119262695, "learning_rate": 5.349542218551074e-07, "loss": 9.287788391113281, "loss_alignment": 0.7568359375, "loss_alignment_w": 0.37841796875, "loss_sft": 0.20185517892241478, "loss_total_v6": 0.5804867595434189, "step": 6841 }, { "epoch": 1.5779603603603602, "grad_norm": 28.082927703857422, "learning_rate": 5.340482778694967e-07, "loss": 8.734922409057617, "loss_alignment": 0.73046875, "loss_alignment_w": 0.365234375, "loss_sft": 0.17970646731555462, "loss_total_v6": 0.5459326282143593, "step": 6842 }, { "epoch": 1.578190990990991, "grad_norm": 27.533828735351562, "learning_rate": 5.331430583608122e-07, "loss": 8.709277153015137, "loss_alignment": 0.74169921875, "loss_alignment_w": 0.370849609375, "loss_sft": 0.17293088044971228, "loss_total_v6": 0.544329822063446, "step": 6843 }, { "epoch": 1.5784216216216216, "grad_norm": 27.20947265625, "learning_rate": 5.322385634759009e-07, "loss": 9.168848037719727, "loss_alignment": 0.7509765625, "loss_alignment_w": 0.37548828125, "loss_sft": 0.19844973273575306, "loss_total_v6": 0.5730530172586441, "step": 6844 }, { "epoch": 1.5786522522522524, "grad_norm": 33.323577880859375, "learning_rate": 5.313347933614915e-07, "loss": 8.874773979187012, "loss_alignment": 0.7373046875, "loss_alignment_w": 0.36865234375, "loss_sft": 0.184983448125422, "loss_total_v6": 0.5546733997762203, "step": 6845 }, { "epoch": 1.578882882882883, "grad_norm": 39.227378845214844, "learning_rate": 5.304317481641958e-07, "loss": 9.301027297973633, "loss_alignment": 0.75, "loss_alignment_w": 0.375, "loss_sft": 0.20610061567276716, "loss_total_v6": 0.5813142359256744, "step": 6846 }, { "epoch": 1.5791135135135135, "grad_norm": 35.1466064453125, "learning_rate": 5.295294280305063e-07, "loss": 9.375744819641113, "loss_alignment": 0.77099609375, "loss_alignment_w": 0.385498046875, "loss_sft": 0.20030291378498077, "loss_total_v6": 0.5859840512275696, "step": 6847 }, { "epoch": 1.579344144144144, "grad_norm": 28.868972778320312, "learning_rate": 5.286278331068018e-07, "loss": 9.204238891601562, "loss_alignment": 0.74072265625, "loss_alignment_w": 0.370361328125, "loss_sft": 0.20490362495183945, "loss_total_v6": 0.5752649530768394, "step": 6848 }, { "epoch": 1.5795747747747748, "grad_norm": 32.94902420043945, "learning_rate": 5.277269635393406e-07, "loss": 8.75919246673584, "loss_alignment": 0.759765625, "loss_alignment_w": 0.3798828125, "loss_sft": 0.16808548476547003, "loss_total_v6": 0.5474495068192482, "step": 6849 }, { "epoch": 1.5798054054054054, "grad_norm": 37.463653564453125, "learning_rate": 5.268268194742638e-07, "loss": 9.021188735961914, "loss_alignment": 0.72412109375, "loss_alignment_w": 0.362060546875, "loss_sft": 0.2015806222334504, "loss_total_v6": 0.5638242624700069, "step": 6850 }, { "epoch": 1.5800360360360362, "grad_norm": 28.537988662719727, "learning_rate": 5.259274010575943e-07, "loss": 9.284500122070312, "loss_alignment": 0.78173828125, "loss_alignment_w": 0.390869140625, "loss_sft": 0.1895189518108964, "loss_total_v6": 0.5802812725305557, "step": 6851 }, { "epoch": 1.5802666666666667, "grad_norm": 33.13558578491211, "learning_rate": 5.250287084352373e-07, "loss": 9.510379791259766, "loss_alignment": 0.7666015625, "loss_alignment_w": 0.38330078125, "loss_sft": 0.21163202077150345, "loss_total_v6": 0.5943987518548965, "step": 6852 }, { "epoch": 1.5804972972972973, "grad_norm": 29.42376136779785, "learning_rate": 5.241307417529823e-07, "loss": 9.687223434448242, "loss_alignment": 0.779296875, "loss_alignment_w": 0.3896484375, "loss_sft": 0.2160318810492754, "loss_total_v6": 0.6054514274001122, "step": 6853 }, { "epoch": 1.5807279279279278, "grad_norm": 26.733291625976562, "learning_rate": 5.232335011564976e-07, "loss": 8.919450759887695, "loss_alignment": 0.75048828125, "loss_alignment_w": 0.375244140625, "loss_sft": 0.18194682151079178, "loss_total_v6": 0.5574656501412392, "step": 6854 }, { "epoch": 1.5809585585585586, "grad_norm": 36.878028869628906, "learning_rate": 5.223369867913374e-07, "loss": 9.102285385131836, "loss_alignment": 0.78466796875, "loss_alignment_w": 0.392333984375, "loss_sft": 0.17604001611471176, "loss_total_v6": 0.5688927993178368, "step": 6855 }, { "epoch": 1.5811891891891892, "grad_norm": 31.607192993164062, "learning_rate": 5.214411988029355e-07, "loss": 8.868429183959961, "loss_alignment": 0.712890625, "loss_alignment_w": 0.3564453125, "loss_sft": 0.19679385051131248, "loss_total_v6": 0.5542767718434334, "step": 6856 }, { "epoch": 1.58141981981982, "grad_norm": 32.124717712402344, "learning_rate": 5.205461373366094e-07, "loss": 9.10238265991211, "loss_alignment": 0.75244140625, "loss_alignment_w": 0.376220703125, "loss_sft": 0.19351750053465366, "loss_total_v6": 0.5688989758491516, "step": 6857 }, { "epoch": 1.5816504504504505, "grad_norm": 29.25907325744629, "learning_rate": 5.196518025375574e-07, "loss": 8.93304443359375, "loss_alignment": 0.748046875, "loss_alignment_w": 0.3740234375, "loss_sft": 0.18389511108398438, "loss_total_v6": 0.5583152696490288, "step": 6858 }, { "epoch": 1.581881081081081, "grad_norm": 32.00300979614258, "learning_rate": 5.187581945508608e-07, "loss": 8.852926254272461, "loss_alignment": 0.74267578125, "loss_alignment_w": 0.371337890625, "loss_sft": 0.18196993507444859, "loss_total_v6": 0.5533078387379646, "step": 6859 }, { "epoch": 1.5821117117117116, "grad_norm": 36.903541564941406, "learning_rate": 5.178653135214811e-07, "loss": 9.885547637939453, "loss_alignment": 0.77978515625, "loss_alignment_w": 0.389892578125, "loss_sft": 0.22879336774349213, "loss_total_v6": 0.617846742272377, "step": 6860 }, { "epoch": 1.5823423423423424, "grad_norm": 34.03038787841797, "learning_rate": 5.169731595942656e-07, "loss": 9.484061241149902, "loss_alignment": 0.76904296875, "loss_alignment_w": 0.384521484375, "loss_sft": 0.20794241316616535, "loss_total_v6": 0.5927538126707077, "step": 6861 }, { "epoch": 1.582572972972973, "grad_norm": 30.084392547607422, "learning_rate": 5.160817329139417e-07, "loss": 8.778196334838867, "loss_alignment": 0.72802734375, "loss_alignment_w": 0.364013671875, "loss_sft": 0.1846083728596568, "loss_total_v6": 0.5486373230814934, "step": 6862 }, { "epoch": 1.5828036036036037, "grad_norm": 29.229990005493164, "learning_rate": 5.151910336251176e-07, "loss": 9.343749046325684, "loss_alignment": 0.76904296875, "loss_alignment_w": 0.384521484375, "loss_sft": 0.19889825955033302, "loss_total_v6": 0.5839843079447746, "step": 6863 }, { "epoch": 1.5830342342342343, "grad_norm": 30.662267684936523, "learning_rate": 5.143010618722849e-07, "loss": 9.056788444519043, "loss_alignment": 0.76025390625, "loss_alignment_w": 0.380126953125, "loss_sft": 0.18468637019395828, "loss_total_v6": 0.5660492777824402, "step": 6864 }, { "epoch": 1.5832648648648648, "grad_norm": 28.4180965423584, "learning_rate": 5.134118177998171e-07, "loss": 9.53292465209961, "loss_alignment": 0.75439453125, "loss_alignment_w": 0.377197265625, "loss_sft": 0.21943453140556812, "loss_total_v6": 0.595807820558548, "step": 6865 }, { "epoch": 1.5834954954954954, "grad_norm": 30.528017044067383, "learning_rate": 5.125233015519676e-07, "loss": 9.295510292053223, "loss_alignment": 0.7255859375, "loss_alignment_w": 0.36279296875, "loss_sft": 0.21819170005619526, "loss_total_v6": 0.5809694081544876, "step": 6866 }, { "epoch": 1.5837261261261262, "grad_norm": 33.66737747192383, "learning_rate": 5.116355132728757e-07, "loss": 8.54614543914795, "loss_alignment": 0.712890625, "loss_alignment_w": 0.3564453125, "loss_sft": 0.17756665125489235, "loss_total_v6": 0.5341340526938438, "step": 6867 }, { "epoch": 1.5839567567567567, "grad_norm": 36.45005416870117, "learning_rate": 5.107484531065604e-07, "loss": 8.888370513916016, "loss_alignment": 0.7373046875, "loss_alignment_w": 0.36865234375, "loss_sft": 0.18662660010159016, "loss_total_v6": 0.5555231124162674, "step": 6868 }, { "epoch": 1.5841873873873875, "grad_norm": 31.302581787109375, "learning_rate": 5.098621211969224e-07, "loss": 8.893352508544922, "loss_alignment": 0.7255859375, "loss_alignment_w": 0.36279296875, "loss_sft": 0.1934383250772953, "loss_total_v6": 0.5558345653116703, "step": 6869 }, { "epoch": 1.584418018018018, "grad_norm": 30.38475227355957, "learning_rate": 5.089765176877437e-07, "loss": 8.565311431884766, "loss_alignment": 0.73583984375, "loss_alignment_w": 0.367919921875, "loss_sft": 0.1678698044270277, "loss_total_v6": 0.5353319682180882, "step": 6870 }, { "epoch": 1.5846486486486486, "grad_norm": 32.90543746948242, "learning_rate": 5.08091642722689e-07, "loss": 9.11431884765625, "loss_alignment": 0.7861328125, "loss_alignment_w": 0.39306640625, "loss_sft": 0.17807386070489883, "loss_total_v6": 0.5696449056267738, "step": 6871 }, { "epoch": 1.5848792792792792, "grad_norm": 28.830717086791992, "learning_rate": 5.072074964453055e-07, "loss": 9.112594604492188, "loss_alignment": 0.74560546875, "loss_alignment_w": 0.372802734375, "loss_sft": 0.19601723924279213, "loss_total_v6": 0.5695371553301811, "step": 6872 }, { "epoch": 1.58510990990991, "grad_norm": 28.349281311035156, "learning_rate": 5.063240789990203e-07, "loss": 8.694259643554688, "loss_alignment": 0.7412109375, "loss_alignment_w": 0.37060546875, "loss_sft": 0.1723127830773592, "loss_total_v6": 0.543391264975071, "step": 6873 }, { "epoch": 1.5853405405405405, "grad_norm": 33.528263092041016, "learning_rate": 5.054413905271449e-07, "loss": 9.182586669921875, "loss_alignment": 0.76171875, "loss_alignment_w": 0.380859375, "loss_sft": 0.1942424364387989, "loss_total_v6": 0.5739116221666336, "step": 6874 }, { "epoch": 1.5855711711711713, "grad_norm": 30.5664119720459, "learning_rate": 5.045594311728708e-07, "loss": 9.127582550048828, "loss_alignment": 0.76171875, "loss_alignment_w": 0.380859375, "loss_sft": 0.18946199864149094, "loss_total_v6": 0.5704739540815353, "step": 6875 }, { "epoch": 1.5858018018018019, "grad_norm": 29.172569274902344, "learning_rate": 5.036782010792695e-07, "loss": 9.425358772277832, "loss_alignment": 0.7841796875, "loss_alignment_w": 0.39208984375, "loss_sft": 0.196964580565691, "loss_total_v6": 0.5890849307179451, "step": 6876 }, { "epoch": 1.5860324324324324, "grad_norm": 33.51467514038086, "learning_rate": 5.027977003892992e-07, "loss": 8.680377960205078, "loss_alignment": 0.7197265625, "loss_alignment_w": 0.35986328125, "loss_sft": 0.18159227818250656, "loss_total_v6": 0.5425236783921719, "step": 6877 }, { "epoch": 1.586263063063063, "grad_norm": 36.54745101928711, "learning_rate": 5.019179292457948e-07, "loss": 9.490013122558594, "loss_alignment": 0.75634765625, "loss_alignment_w": 0.378173828125, "loss_sft": 0.21536397747695446, "loss_total_v6": 0.5931258201599121, "step": 6878 }, { "epoch": 1.5864936936936935, "grad_norm": 29.75149917602539, "learning_rate": 5.010388877914746e-07, "loss": 9.112894058227539, "loss_alignment": 0.77197265625, "loss_alignment_w": 0.385986328125, "loss_sft": 0.18343224935233593, "loss_total_v6": 0.5695559307932854, "step": 6879 }, { "epoch": 1.5867243243243243, "grad_norm": 29.670827865600586, "learning_rate": 5.001605761689399e-07, "loss": 8.994438171386719, "loss_alignment": 0.7333984375, "loss_alignment_w": 0.36669921875, "loss_sft": 0.19581936858594418, "loss_total_v6": 0.5621523819863796, "step": 6880 }, { "epoch": 1.586954954954955, "grad_norm": 33.14450454711914, "learning_rate": 4.992829945206706e-07, "loss": 9.45561408996582, "loss_alignment": 0.787109375, "loss_alignment_w": 0.3935546875, "loss_sft": 0.19642935506999493, "loss_total_v6": 0.5909758731722832, "step": 6881 }, { "epoch": 1.5871855855855856, "grad_norm": 31.681983947753906, "learning_rate": 4.984061429890324e-07, "loss": 8.505810737609863, "loss_alignment": 0.7353515625, "loss_alignment_w": 0.36767578125, "loss_sft": 0.16430359985679388, "loss_total_v6": 0.5316131822764874, "step": 6882 }, { "epoch": 1.5874162162162162, "grad_norm": 34.17319107055664, "learning_rate": 4.975300217162693e-07, "loss": 9.35988998413086, "loss_alignment": 0.7587890625, "loss_alignment_w": 0.37939453125, "loss_sft": 0.20665142685174942, "loss_total_v6": 0.5849930942058563, "step": 6883 }, { "epoch": 1.5876468468468468, "grad_norm": 30.786266326904297, "learning_rate": 4.966546308445075e-07, "loss": 8.906763076782227, "loss_alignment": 0.7158203125, "loss_alignment_w": 0.35791015625, "loss_sft": 0.19920501299202442, "loss_total_v6": 0.556672677397728, "step": 6884 }, { "epoch": 1.5878774774774773, "grad_norm": 37.559654235839844, "learning_rate": 4.957799705157534e-07, "loss": 9.568645477294922, "loss_alignment": 0.77197265625, "loss_alignment_w": 0.385986328125, "loss_sft": 0.21144365705549717, "loss_total_v6": 0.5980403274297714, "step": 6885 }, { "epoch": 1.588108108108108, "grad_norm": 30.095149993896484, "learning_rate": 4.949060408718975e-07, "loss": 9.434940338134766, "loss_alignment": 0.7724609375, "loss_alignment_w": 0.38623046875, "loss_sft": 0.20340755954384804, "loss_total_v6": 0.5896838009357452, "step": 6886 }, { "epoch": 1.5883387387387389, "grad_norm": 34.85954284667969, "learning_rate": 4.940328420547125e-07, "loss": 8.534456253051758, "loss_alignment": 0.73193359375, "loss_alignment_w": 0.365966796875, "loss_sft": 0.16784874070435762, "loss_total_v6": 0.5334035605192184, "step": 6887 }, { "epoch": 1.5885693693693694, "grad_norm": 36.66889190673828, "learning_rate": 4.931603742058494e-07, "loss": 9.06793212890625, "loss_alignment": 0.75341796875, "loss_alignment_w": 0.376708984375, "loss_sft": 0.19050983898341656, "loss_total_v6": 0.566745787858963, "step": 6888 }, { "epoch": 1.5888, "grad_norm": 32.23127746582031, "learning_rate": 4.922886374668412e-07, "loss": 8.915945053100586, "loss_alignment": 0.740234375, "loss_alignment_w": 0.3701171875, "loss_sft": 0.1863054558634758, "loss_total_v6": 0.5572466216981411, "step": 6889 }, { "epoch": 1.5890306306306305, "grad_norm": 33.900753021240234, "learning_rate": 4.914176319791037e-07, "loss": 9.48993968963623, "loss_alignment": 0.75732421875, "loss_alignment_w": 0.378662109375, "loss_sft": 0.2145507000386715, "loss_total_v6": 0.5931212604045868, "step": 6890 }, { "epoch": 1.589261261261261, "grad_norm": 32.66768264770508, "learning_rate": 4.905473578839326e-07, "loss": 9.414896011352539, "loss_alignment": 0.74365234375, "loss_alignment_w": 0.371826171875, "loss_sft": 0.21729142777621746, "loss_total_v6": 0.5884309634566307, "step": 6891 }, { "epoch": 1.5894918918918919, "grad_norm": 33.245662689208984, "learning_rate": 4.896778153225062e-07, "loss": 8.74117660522461, "loss_alignment": 0.76806640625, "loss_alignment_w": 0.384033203125, "loss_sft": 0.16270231176167727, "loss_total_v6": 0.5463235154747963, "step": 6892 }, { "epoch": 1.5897225225225227, "grad_norm": 30.629732131958008, "learning_rate": 4.888090044358851e-07, "loss": 9.516914367675781, "loss_alignment": 0.7890625, "loss_alignment_w": 0.39453125, "loss_sft": 0.19899419881403446, "loss_total_v6": 0.5948071926832199, "step": 6893 }, { "epoch": 1.5899531531531532, "grad_norm": 33.69466781616211, "learning_rate": 4.87940925365008e-07, "loss": 9.323277473449707, "loss_alignment": 0.73583984375, "loss_alignment_w": 0.367919921875, "loss_sft": 0.21464761905372143, "loss_total_v6": 0.5827048644423485, "step": 6894 }, { "epoch": 1.5901837837837838, "grad_norm": 28.56545066833496, "learning_rate": 4.87073578250698e-07, "loss": 9.021478652954102, "loss_alignment": 0.7353515625, "loss_alignment_w": 0.36767578125, "loss_sft": 0.19615134224295616, "loss_total_v6": 0.5638423934578896, "step": 6895 }, { "epoch": 1.5904144144144143, "grad_norm": 34.484344482421875, "learning_rate": 4.862069632336558e-07, "loss": 8.939657211303711, "loss_alignment": 0.74462890625, "loss_alignment_w": 0.372314453125, "loss_sft": 0.18662775680422783, "loss_total_v6": 0.5587285831570625, "step": 6896 }, { "epoch": 1.590645045045045, "grad_norm": 30.487171173095703, "learning_rate": 4.853410804544689e-07, "loss": 8.97221851348877, "loss_alignment": 0.77197265625, "loss_alignment_w": 0.385986328125, "loss_sft": 0.17409068066626787, "loss_total_v6": 0.5607636719942093, "step": 6897 }, { "epoch": 1.5908756756756757, "grad_norm": 32.03640365600586, "learning_rate": 4.844759300536e-07, "loss": 9.591193199157715, "loss_alignment": 0.77294921875, "loss_alignment_w": 0.386474609375, "loss_sft": 0.21271554939448833, "loss_total_v6": 0.5994495674967766, "step": 6898 }, { "epoch": 1.5911063063063065, "grad_norm": 31.79030418395996, "learning_rate": 4.83611512171398e-07, "loss": 9.432323455810547, "loss_alignment": 0.7841796875, "loss_alignment_w": 0.39208984375, "loss_sft": 0.1988951861858368, "loss_total_v6": 0.5895202085375786, "step": 6899 }, { "epoch": 1.591336936936937, "grad_norm": 25.311321258544922, "learning_rate": 4.827478269480895e-07, "loss": 8.830169677734375, "loss_alignment": 0.72021484375, "loss_alignment_w": 0.360107421875, "loss_sft": 0.19173247180879116, "loss_total_v6": 0.5518856644630432, "step": 6900 }, { "epoch": 1.5915675675675676, "grad_norm": 35.16579055786133, "learning_rate": 4.818848745237825e-07, "loss": 9.164678573608398, "loss_alignment": 0.7333984375, "loss_alignment_w": 0.36669921875, "loss_sft": 0.20667303912341595, "loss_total_v6": 0.5727924071252346, "step": 6901 }, { "epoch": 1.5917981981981981, "grad_norm": 31.58246421813965, "learning_rate": 4.8102265503847e-07, "loss": 9.44212818145752, "loss_alignment": 0.77197265625, "loss_alignment_w": 0.385986328125, "loss_sft": 0.20428399369120598, "loss_total_v6": 0.590133011341095, "step": 6902 }, { "epoch": 1.5920288288288287, "grad_norm": 35.81277847290039, "learning_rate": 4.801611686320201e-07, "loss": 8.803655624389648, "loss_alignment": 0.74658203125, "loss_alignment_w": 0.373291015625, "loss_sft": 0.17840228602290154, "loss_total_v6": 0.5502284690737724, "step": 6903 }, { "epoch": 1.5922594594594595, "grad_norm": 31.4847354888916, "learning_rate": 4.793004154441877e-07, "loss": 8.838401794433594, "loss_alignment": 0.73046875, "loss_alignment_w": 0.365234375, "loss_sft": 0.18650958128273487, "loss_total_v6": 0.5524001121520996, "step": 6904 }, { "epoch": 1.5924900900900902, "grad_norm": 30.17256736755371, "learning_rate": 4.784403956146039e-07, "loss": 9.207447052001953, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.18787692673504353, "loss_total_v6": 0.5754654332995415, "step": 6905 }, { "epoch": 1.5927207207207208, "grad_norm": 31.609926223754883, "learning_rate": 4.775811092827853e-07, "loss": 8.843843460083008, "loss_alignment": 0.759765625, "loss_alignment_w": 0.3798828125, "loss_sft": 0.17320833168923855, "loss_total_v6": 0.5527401864528656, "step": 6906 }, { "epoch": 1.5929513513513514, "grad_norm": 28.223615646362305, "learning_rate": 4.7672255658812616e-07, "loss": 9.086660385131836, "loss_alignment": 0.73828125, "loss_alignment_w": 0.369140625, "loss_sft": 0.1985162552446127, "loss_total_v6": 0.5679162815213203, "step": 6907 }, { "epoch": 1.593181981981982, "grad_norm": 30.647470474243164, "learning_rate": 4.758647376699033e-07, "loss": 8.90025520324707, "loss_alignment": 0.7470703125, "loss_alignment_w": 0.37353515625, "loss_sft": 0.18192213028669357, "loss_total_v6": 0.5562659911811352, "step": 6908 }, { "epoch": 1.5934126126126125, "grad_norm": 30.55972671508789, "learning_rate": 4.75007652667272e-07, "loss": 9.128121376037598, "loss_alignment": 0.7548828125, "loss_alignment_w": 0.37744140625, "loss_sft": 0.1932187918573618, "loss_total_v6": 0.5705075934529305, "step": 6909 }, { "epoch": 1.5936432432432432, "grad_norm": 33.803707122802734, "learning_rate": 4.741513017192734e-07, "loss": 9.184059143066406, "loss_alignment": 0.763671875, "loss_alignment_w": 0.3818359375, "loss_sft": 0.1927017867565155, "loss_total_v6": 0.5740036778151989, "step": 6910 }, { "epoch": 1.593873873873874, "grad_norm": 35.388648986816406, "learning_rate": 4.7329568496482516e-07, "loss": 8.593965530395508, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.1477032620459795, "loss_total_v6": 0.5371228232979774, "step": 6911 }, { "epoch": 1.5941045045045046, "grad_norm": 34.42433166503906, "learning_rate": 4.7244080254272795e-07, "loss": 8.906169891357422, "loss_alignment": 0.75390625, "loss_alignment_w": 0.376953125, "loss_sft": 0.17988082952797413, "loss_total_v6": 0.5566355809569359, "step": 6912 }, { "epoch": 1.5943351351351351, "grad_norm": 33.31684112548828, "learning_rate": 4.715866545916631e-07, "loss": 9.230735778808594, "loss_alignment": 0.740234375, "loss_alignment_w": 0.3701171875, "loss_sft": 0.20579668134450912, "loss_total_v6": 0.5769209563732147, "step": 6913 }, { "epoch": 1.5945657657657657, "grad_norm": 32.062774658203125, "learning_rate": 4.707332412501919e-07, "loss": 9.13139820098877, "loss_alignment": 0.7197265625, "loss_alignment_w": 0.35986328125, "loss_sft": 0.21066603623330593, "loss_total_v6": 0.5707123950123787, "step": 6914 }, { "epoch": 1.5947963963963963, "grad_norm": 33.581268310546875, "learning_rate": 4.69880562656756e-07, "loss": 9.33803939819336, "loss_alignment": 0.74658203125, "loss_alignment_w": 0.373291015625, "loss_sft": 0.2101533431559801, "loss_total_v6": 0.583627462387085, "step": 6915 }, { "epoch": 1.595027027027027, "grad_norm": 35.104976654052734, "learning_rate": 4.6902861894967955e-07, "loss": 8.801262855529785, "loss_alignment": 0.73291015625, "loss_alignment_w": 0.366455078125, "loss_sft": 0.18344076070934534, "loss_total_v6": 0.550078921020031, "step": 6916 }, { "epoch": 1.5952576576576578, "grad_norm": 31.5927677154541, "learning_rate": 4.681774102671688e-07, "loss": 8.932957649230957, "loss_alignment": 0.712890625, "loss_alignment_w": 0.3564453125, "loss_sft": 0.20140678621828556, "loss_total_v6": 0.5583098456263542, "step": 6917 }, { "epoch": 1.5954882882882884, "grad_norm": 28.8625545501709, "learning_rate": 4.6732693674730656e-07, "loss": 9.630409240722656, "loss_alignment": 0.78466796875, "loss_alignment_w": 0.392333984375, "loss_sft": 0.21061945520341396, "loss_total_v6": 0.6019005700945854, "step": 6918 }, { "epoch": 1.595718918918919, "grad_norm": 32.1186637878418, "learning_rate": 4.6647719852805975e-07, "loss": 8.741238594055176, "loss_alignment": 0.740234375, "loss_alignment_w": 0.3701171875, "loss_sft": 0.17692737746983767, "loss_total_v6": 0.5463274121284485, "step": 6919 }, { "epoch": 1.5959495495495495, "grad_norm": 31.474885940551758, "learning_rate": 4.6562819574727304e-07, "loss": 8.877128601074219, "loss_alignment": 0.6904296875, "loss_alignment_w": 0.34521484375, "loss_sft": 0.20948362164199352, "loss_total_v6": 0.5548205226659775, "step": 6920 }, { "epoch": 1.59618018018018, "grad_norm": 32.12015914916992, "learning_rate": 4.647799285426757e-07, "loss": 8.774173736572266, "loss_alignment": 0.73046875, "loss_alignment_w": 0.365234375, "loss_sft": 0.18399071507155895, "loss_total_v6": 0.5483858622610569, "step": 6921 }, { "epoch": 1.5964108108108108, "grad_norm": 31.395965576171875, "learning_rate": 4.639323970518739e-07, "loss": 9.060712814331055, "loss_alignment": 0.7646484375, "loss_alignment_w": 0.38232421875, "loss_sft": 0.18340579327195883, "loss_total_v6": 0.5662945955991745, "step": 6922 }, { "epoch": 1.5966414414414416, "grad_norm": 33.59931182861328, "learning_rate": 4.63085601412358e-07, "loss": 9.552175521850586, "loss_alignment": 0.79150390625, "loss_alignment_w": 0.395751953125, "loss_sft": 0.20194563828408718, "loss_total_v6": 0.5970109701156616, "step": 6923 }, { "epoch": 1.5968720720720722, "grad_norm": 29.678495407104492, "learning_rate": 4.622395417614961e-07, "loss": 9.143617630004883, "loss_alignment": 0.7587890625, "loss_alignment_w": 0.37939453125, "loss_sft": 0.19340914860367775, "loss_total_v6": 0.5714761540293694, "step": 6924 }, { "epoch": 1.5971027027027027, "grad_norm": 32.32847595214844, "learning_rate": 4.6139421823653716e-07, "loss": 8.853937149047852, "loss_alignment": 0.73095703125, "loss_alignment_w": 0.365478515625, "loss_sft": 0.18842660635709763, "loss_total_v6": 0.5533710345625877, "step": 6925 }, { "epoch": 1.5973333333333333, "grad_norm": 32.07052993774414, "learning_rate": 4.605496309746127e-07, "loss": 8.687447547912598, "loss_alignment": 0.7265625, "loss_alignment_w": 0.36328125, "loss_sft": 0.178860267624259, "loss_total_v6": 0.5429654866456985, "step": 6926 }, { "epoch": 1.5975639639639638, "grad_norm": 32.01847457885742, "learning_rate": 4.597057801127336e-07, "loss": 9.502273559570312, "loss_alignment": 0.771484375, "loss_alignment_w": 0.3857421875, "loss_sft": 0.20680715516209602, "loss_total_v6": 0.5938920974731445, "step": 6927 }, { "epoch": 1.5977945945945946, "grad_norm": 31.94378089904785, "learning_rate": 4.588626657877898e-07, "loss": 9.123895645141602, "loss_alignment": 0.74755859375, "loss_alignment_w": 0.373779296875, "loss_sft": 0.19593014381825924, "loss_total_v6": 0.570243488997221, "step": 6928 }, { "epoch": 1.5980252252252254, "grad_norm": 32.27718734741211, "learning_rate": 4.580202881365553e-07, "loss": 8.805569648742676, "loss_alignment": 0.7587890625, "loss_alignment_w": 0.37939453125, "loss_sft": 0.17081623431295156, "loss_total_v6": 0.5503481067717075, "step": 6929 }, { "epoch": 1.598255855855856, "grad_norm": 34.1374626159668, "learning_rate": 4.571786472956807e-07, "loss": 8.591238975524902, "loss_alignment": 0.72412109375, "loss_alignment_w": 0.362060546875, "loss_sft": 0.17447988130152225, "loss_total_v6": 0.5369524210691452, "step": 6930 }, { "epoch": 1.5984864864864865, "grad_norm": 29.01218032836914, "learning_rate": 4.563377434017008e-07, "loss": 8.975811004638672, "loss_alignment": 0.73046875, "loss_alignment_w": 0.365234375, "loss_sft": 0.19663883559405804, "loss_total_v6": 0.5609882026910782, "step": 6931 }, { "epoch": 1.598717117117117, "grad_norm": 32.278656005859375, "learning_rate": 4.55497576591028e-07, "loss": 8.82658576965332, "loss_alignment": 0.74609375, "loss_alignment_w": 0.373046875, "loss_sft": 0.17878257296979427, "loss_total_v6": 0.5516616106033325, "step": 6932 }, { "epoch": 1.5989477477477476, "grad_norm": 31.27260971069336, "learning_rate": 4.546581469999556e-07, "loss": 8.94141674041748, "loss_alignment": 0.73583984375, "loss_alignment_w": 0.367919921875, "loss_sft": 0.19104071706533432, "loss_total_v6": 0.5588385723531246, "step": 6933 }, { "epoch": 1.5991783783783784, "grad_norm": 30.476274490356445, "learning_rate": 4.538194547646574e-07, "loss": 9.556489944458008, "loss_alignment": 0.779296875, "loss_alignment_w": 0.3896484375, "loss_sft": 0.20746437087655067, "loss_total_v6": 0.5972806289792061, "step": 6934 }, { "epoch": 1.5994090090090092, "grad_norm": 34.567237854003906, "learning_rate": 4.52981500021189e-07, "loss": 8.684137344360352, "loss_alignment": 0.7373046875, "loss_alignment_w": 0.36865234375, "loss_sft": 0.17384681850671768, "loss_total_v6": 0.5427585542201996, "step": 6935 }, { "epoch": 1.5996396396396397, "grad_norm": 32.24525451660156, "learning_rate": 4.521442829054856e-07, "loss": 9.165236473083496, "loss_alignment": 0.73193359375, "loss_alignment_w": 0.365966796875, "loss_sft": 0.2062348574399948, "loss_total_v6": 0.5728272646665573, "step": 6936 }, { "epoch": 1.5998702702702703, "grad_norm": 28.681684494018555, "learning_rate": 4.5130780355336245e-07, "loss": 8.96729850769043, "loss_alignment": 0.7470703125, "loss_alignment_w": 0.37353515625, "loss_sft": 0.187622906640172, "loss_total_v6": 0.5604561492800713, "step": 6937 }, { "epoch": 1.6001009009009008, "grad_norm": 28.875911712646484, "learning_rate": 4.5047206210051433e-07, "loss": 8.772830963134766, "loss_alignment": 0.716796875, "loss_alignment_w": 0.3583984375, "loss_sft": 0.19083428476005793, "loss_total_v6": 0.5483019314706326, "step": 6938 }, { "epoch": 1.6003315315315314, "grad_norm": 30.3718204498291, "learning_rate": 4.496370586825177e-07, "loss": 9.158912658691406, "loss_alignment": 0.76220703125, "loss_alignment_w": 0.381103515625, "loss_sft": 0.19207621552050114, "loss_total_v6": 0.5724320337176323, "step": 6939 }, { "epoch": 1.6005621621621622, "grad_norm": 38.544944763183594, "learning_rate": 4.4880279343482713e-07, "loss": 9.303640365600586, "loss_alignment": 0.771484375, "loss_alignment_w": 0.3857421875, "loss_sft": 0.19559797272086143, "loss_total_v6": 0.5814775079488754, "step": 6940 }, { "epoch": 1.6007927927927927, "grad_norm": 31.57826805114746, "learning_rate": 4.4796926649278094e-07, "loss": 9.311444282531738, "loss_alignment": 0.76416015625, "loss_alignment_w": 0.382080078125, "loss_sft": 0.20006834901869297, "loss_total_v6": 0.581965297460556, "step": 6941 }, { "epoch": 1.6010234234234235, "grad_norm": 33.05231475830078, "learning_rate": 4.47136477991596e-07, "loss": 9.246170043945312, "loss_alignment": 0.767578125, "loss_alignment_w": 0.3837890625, "loss_sft": 0.19423388876020908, "loss_total_v6": 0.5778856128454208, "step": 6942 }, { "epoch": 1.601254054054054, "grad_norm": 36.42112731933594, "learning_rate": 4.463044280663686e-07, "loss": 8.665623664855957, "loss_alignment": 0.71923828125, "loss_alignment_w": 0.359619140625, "loss_sft": 0.1809295266866684, "loss_total_v6": 0.5416015051305294, "step": 6943 }, { "epoch": 1.6014846846846846, "grad_norm": 28.913042068481445, "learning_rate": 4.454731168520754e-07, "loss": 9.252148628234863, "loss_alignment": 0.775390625, "loss_alignment_w": 0.3876953125, "loss_sft": 0.1900604162365198, "loss_total_v6": 0.5782592818140984, "step": 6944 }, { "epoch": 1.6017153153153152, "grad_norm": 33.186927795410156, "learning_rate": 4.446425444835728e-07, "loss": 9.081541061401367, "loss_alignment": 0.732421875, "loss_alignment_w": 0.3662109375, "loss_sft": 0.20150744356215, "loss_total_v6": 0.5675963014364243, "step": 6945 }, { "epoch": 1.601945945945946, "grad_norm": 33.05127716064453, "learning_rate": 4.438127110956003e-07, "loss": 9.13900375366211, "loss_alignment": 0.7685546875, "loss_alignment_w": 0.38427734375, "loss_sft": 0.18709352612495422, "loss_total_v6": 0.5711877644062042, "step": 6946 }, { "epoch": 1.6021765765765765, "grad_norm": 31.858266830444336, "learning_rate": 4.4298361682277355e-07, "loss": 9.03931713104248, "loss_alignment": 0.720703125, "loss_alignment_w": 0.3603515625, "loss_sft": 0.20498725771903992, "loss_total_v6": 0.564957357943058, "step": 6947 }, { "epoch": 1.6024072072072073, "grad_norm": 33.573551177978516, "learning_rate": 4.421552617995917e-07, "loss": 9.220375061035156, "loss_alignment": 0.7607421875, "loss_alignment_w": 0.38037109375, "loss_sft": 0.19672628492116928, "loss_total_v6": 0.5762733966112137, "step": 6948 }, { "epoch": 1.6026378378378379, "grad_norm": 30.383729934692383, "learning_rate": 4.4132764616043157e-07, "loss": 9.005083084106445, "loss_alignment": 0.74462890625, "loss_alignment_w": 0.372314453125, "loss_sft": 0.19044220633804798, "loss_total_v6": 0.5628176964819431, "step": 6949 }, { "epoch": 1.6028684684684684, "grad_norm": 32.39057540893555, "learning_rate": 4.405007700395497e-07, "loss": 8.752198219299316, "loss_alignment": 0.73583984375, "loss_alignment_w": 0.367919921875, "loss_sft": 0.1804962493479252, "loss_total_v6": 0.5470123775303364, "step": 6950 }, { "epoch": 1.603099099099099, "grad_norm": 31.936609268188477, "learning_rate": 4.3967463357108607e-07, "loss": 9.124889373779297, "loss_alignment": 0.76123046875, "loss_alignment_w": 0.380615234375, "loss_sft": 0.1895835231989622, "loss_total_v6": 0.5703055635094643, "step": 6951 }, { "epoch": 1.6033297297297298, "grad_norm": 31.43744468688965, "learning_rate": 4.3884923688905676e-07, "loss": 8.66007137298584, "loss_alignment": 0.708984375, "loss_alignment_w": 0.3544921875, "loss_sft": 0.1876930445432663, "loss_total_v6": 0.5412544645369053, "step": 6952 }, { "epoch": 1.6035603603603603, "grad_norm": 31.85852813720703, "learning_rate": 4.380245801273597e-07, "loss": 9.131352424621582, "loss_alignment": 0.76171875, "loss_alignment_w": 0.380859375, "loss_sft": 0.18985016830265522, "loss_total_v6": 0.5707095339894295, "step": 6953 }, { "epoch": 1.603790990990991, "grad_norm": 33.7512092590332, "learning_rate": 4.3720066341977285e-07, "loss": 8.609725952148438, "loss_alignment": 0.71337890625, "loss_alignment_w": 0.356689453125, "loss_sft": 0.18100645393133163, "loss_total_v6": 0.5381078831851482, "step": 6954 }, { "epoch": 1.6040216216216217, "grad_norm": 34.96730422973633, "learning_rate": 4.3637748689995486e-07, "loss": 9.036844253540039, "loss_alignment": 0.74951171875, "loss_alignment_w": 0.374755859375, "loss_sft": 0.1906877737492323, "loss_total_v6": 0.5648027658462524, "step": 6955 }, { "epoch": 1.6042522522522522, "grad_norm": 27.489763259887695, "learning_rate": 4.3555505070144276e-07, "loss": 8.758420944213867, "loss_alignment": 0.736328125, "loss_alignment_w": 0.3681640625, "loss_sft": 0.17983235232532024, "loss_total_v6": 0.5474013388156891, "step": 6956 }, { "epoch": 1.6044828828828828, "grad_norm": 33.957672119140625, "learning_rate": 4.3473335495765325e-07, "loss": 8.977773666381836, "loss_alignment": 0.748046875, "loss_alignment_w": 0.3740234375, "loss_sft": 0.1867974866181612, "loss_total_v6": 0.5611108541488647, "step": 6957 }, { "epoch": 1.6047135135135135, "grad_norm": 28.861108779907227, "learning_rate": 4.339123998018846e-07, "loss": 8.726572036743164, "loss_alignment": 0.72119140625, "loss_alignment_w": 0.360595703125, "loss_sft": 0.1841436643153429, "loss_total_v6": 0.5454107634723186, "step": 6958 }, { "epoch": 1.604944144144144, "grad_norm": 37.13908004760742, "learning_rate": 4.3309218536731256e-07, "loss": 9.090337753295898, "loss_alignment": 0.76416015625, "loss_alignment_w": 0.382080078125, "loss_sft": 0.18663058057427406, "loss_total_v6": 0.5681460946798325, "step": 6959 }, { "epoch": 1.6051747747747749, "grad_norm": 31.511259078979492, "learning_rate": 4.322727117869951e-07, "loss": 8.67930793762207, "loss_alignment": 0.7265625, "loss_alignment_w": 0.36328125, "loss_sft": 0.17960271053016186, "loss_total_v6": 0.5424567200243473, "step": 6960 }, { "epoch": 1.6054054054054054, "grad_norm": 28.573320388793945, "learning_rate": 4.3145397919387043e-07, "loss": 8.853788375854492, "loss_alignment": 0.7265625, "loss_alignment_w": 0.36328125, "loss_sft": 0.19032466039061546, "loss_total_v6": 0.5533617921173573, "step": 6961 }, { "epoch": 1.605636036036036, "grad_norm": 30.784273147583008, "learning_rate": 4.306359877207539e-07, "loss": 9.159045219421387, "loss_alignment": 0.76123046875, "loss_alignment_w": 0.380615234375, "loss_sft": 0.191580967977643, "loss_total_v6": 0.5724403411149979, "step": 6962 }, { "epoch": 1.6058666666666666, "grad_norm": 29.118101119995117, "learning_rate": 4.298187375003421e-07, "loss": 9.162696838378906, "loss_alignment": 0.75, "loss_alignment_w": 0.375, "loss_sft": 0.1974549014121294, "loss_total_v6": 0.5726685225963593, "step": 6963 }, { "epoch": 1.6060972972972973, "grad_norm": 31.685827255249023, "learning_rate": 4.2900222866521015e-07, "loss": 8.514875411987305, "loss_alignment": 0.720703125, "loss_alignment_w": 0.3603515625, "loss_sft": 0.17126360908150673, "loss_total_v6": 0.5321797393262386, "step": 6964 }, { "epoch": 1.606327927927928, "grad_norm": 33.7999153137207, "learning_rate": 4.281864613478159e-07, "loss": 9.225188255310059, "loss_alignment": 0.7744140625, "loss_alignment_w": 0.38720703125, "loss_sft": 0.18909253738820553, "loss_total_v6": 0.5765742287039757, "step": 6965 }, { "epoch": 1.6065585585585587, "grad_norm": 30.923994064331055, "learning_rate": 4.273714356804937e-07, "loss": 9.212738037109375, "loss_alignment": 0.75048828125, "loss_alignment_w": 0.375244140625, "loss_sft": 0.2003841381520033, "loss_total_v6": 0.5757961049675941, "step": 6966 }, { "epoch": 1.6067891891891892, "grad_norm": 35.49785614013672, "learning_rate": 4.265571517954598e-07, "loss": 9.29350757598877, "loss_alignment": 0.77587890625, "loss_alignment_w": 0.387939453125, "loss_sft": 0.1926606521010399, "loss_total_v6": 0.5808442309498787, "step": 6967 }, { "epoch": 1.6070198198198198, "grad_norm": 33.204341888427734, "learning_rate": 4.257436098248091e-07, "loss": 8.709576606750488, "loss_alignment": 0.7333984375, "loss_alignment_w": 0.36669921875, "loss_sft": 0.17848853301256895, "loss_total_v6": 0.5443485267460346, "step": 6968 }, { "epoch": 1.6072504504504503, "grad_norm": 25.17047882080078, "learning_rate": 4.2493080990051503e-07, "loss": 9.298053741455078, "loss_alignment": 0.755859375, "loss_alignment_w": 0.3779296875, "loss_sft": 0.20289349742233753, "loss_total_v6": 0.5811283737421036, "step": 6969 }, { "epoch": 1.6074810810810811, "grad_norm": 34.00276184082031, "learning_rate": 4.241187521544332e-07, "loss": 8.7531156539917, "loss_alignment": 0.724609375, "loss_alignment_w": 0.3623046875, "loss_sft": 0.18528381548821926, "loss_total_v6": 0.5470697246491909, "step": 6970 }, { "epoch": 1.6077117117117117, "grad_norm": 32.0533332824707, "learning_rate": 4.2330743671829766e-07, "loss": 8.888460159301758, "loss_alignment": 0.74755859375, "loss_alignment_w": 0.373779296875, "loss_sft": 0.18188679590821266, "loss_total_v6": 0.5555287599563599, "step": 6971 }, { "epoch": 1.6079423423423425, "grad_norm": 28.760318756103516, "learning_rate": 4.224968637237198e-07, "loss": 8.8143949508667, "loss_alignment": 0.77099609375, "loss_alignment_w": 0.385498046875, "loss_sft": 0.16679018177092075, "loss_total_v6": 0.5508996695280075, "step": 6972 }, { "epoch": 1.608172972972973, "grad_norm": 29.1726131439209, "learning_rate": 4.2168703330219494e-07, "loss": 8.715615272521973, "loss_alignment": 0.7490234375, "loss_alignment_w": 0.37451171875, "loss_sft": 0.16961912252008915, "loss_total_v6": 0.5447259359061718, "step": 6973 }, { "epoch": 1.6084036036036036, "grad_norm": 41.965782165527344, "learning_rate": 4.208779455850942e-07, "loss": 9.352434158325195, "loss_alignment": 0.759765625, "loss_alignment_w": 0.3798828125, "loss_sft": 0.20394239388406277, "loss_total_v6": 0.5845271274447441, "step": 6974 }, { "epoch": 1.6086342342342341, "grad_norm": 29.784807205200195, "learning_rate": 4.200696007036703e-07, "loss": 8.873357772827148, "loss_alignment": 0.75830078125, "loss_alignment_w": 0.379150390625, "loss_sft": 0.17422899045050144, "loss_total_v6": 0.5545848309993744, "step": 6975 }, { "epoch": 1.608864864864865, "grad_norm": 33.22885513305664, "learning_rate": 4.192619987890556e-07, "loss": 9.431635856628418, "loss_alignment": 0.8212890625, "loss_alignment_w": 0.41064453125, "loss_sft": 0.17816131561994553, "loss_total_v6": 0.5894772335886955, "step": 6976 }, { "epoch": 1.6090954954954955, "grad_norm": 32.43887710571289, "learning_rate": 4.184551399722597e-07, "loss": 8.977192878723145, "loss_alignment": 0.71728515625, "loss_alignment_w": 0.358642578125, "loss_sft": 0.20287447050213814, "loss_total_v6": 0.5610745511949062, "step": 6977 }, { "epoch": 1.6093261261261262, "grad_norm": 36.784934997558594, "learning_rate": 4.176490243841724e-07, "loss": 9.460468292236328, "loss_alignment": 0.77685546875, "loss_alignment_w": 0.388427734375, "loss_sft": 0.2027142345905304, "loss_total_v6": 0.5912793055176735, "step": 6978 }, { "epoch": 1.6095567567567568, "grad_norm": 30.645875930786133, "learning_rate": 4.168436521555652e-07, "loss": 8.652052879333496, "loss_alignment": 0.73291015625, "loss_alignment_w": 0.366455078125, "loss_sft": 0.17376417107880116, "loss_total_v6": 0.5407533049583435, "step": 6979 }, { "epoch": 1.6097873873873874, "grad_norm": 30.530643463134766, "learning_rate": 4.1603902341708804e-07, "loss": 9.32318115234375, "loss_alignment": 0.77294921875, "loss_alignment_w": 0.386474609375, "loss_sft": 0.196392054669559, "loss_total_v6": 0.5826988220214844, "step": 6980 }, { "epoch": 1.610018018018018, "grad_norm": 30.734418869018555, "learning_rate": 4.152351382992681e-07, "loss": 9.358307838439941, "loss_alignment": 0.7568359375, "loss_alignment_w": 0.37841796875, "loss_sft": 0.20580486208200455, "loss_total_v6": 0.5848942250013351, "step": 6981 }, { "epoch": 1.6102486486486487, "grad_norm": 32.13405227661133, "learning_rate": 4.144319969325145e-07, "loss": 9.317039489746094, "loss_alignment": 0.79150390625, "loss_alignment_w": 0.395751953125, "loss_sft": 0.1873870100826025, "loss_total_v6": 0.5823149904608727, "step": 6982 }, { "epoch": 1.6104792792792793, "grad_norm": 31.393678665161133, "learning_rate": 4.13629599447114e-07, "loss": 9.044090270996094, "loss_alignment": 0.71630859375, "loss_alignment_w": 0.358154296875, "loss_sft": 0.20569754391908646, "loss_total_v6": 0.5652556642889977, "step": 6983 }, { "epoch": 1.61070990990991, "grad_norm": 29.027528762817383, "learning_rate": 4.1282794597323263e-07, "loss": 8.500948905944824, "loss_alignment": 0.7392578125, "loss_alignment_w": 0.36962890625, "loss_sft": 0.16146676987409592, "loss_total_v6": 0.5313093140721321, "step": 6984 }, { "epoch": 1.6109405405405406, "grad_norm": 34.17704772949219, "learning_rate": 4.120270366409174e-07, "loss": 9.351678848266602, "loss_alignment": 0.8046875, "loss_alignment_w": 0.40234375, "loss_sft": 0.18234973680227995, "loss_total_v6": 0.584479883313179, "step": 6985 }, { "epoch": 1.6111711711711711, "grad_norm": 35.127376556396484, "learning_rate": 4.112268715800943e-07, "loss": 9.024822235107422, "loss_alignment": 0.77197265625, "loss_alignment_w": 0.385986328125, "loss_sft": 0.17830922827124596, "loss_total_v6": 0.5640513971447945, "step": 6986 }, { "epoch": 1.6114018018018017, "grad_norm": 34.13936996459961, "learning_rate": 4.1042745092056744e-07, "loss": 9.05674934387207, "loss_alignment": 0.77392578125, "loss_alignment_w": 0.386962890625, "loss_sft": 0.1793891042470932, "loss_total_v6": 0.566046804189682, "step": 6987 }, { "epoch": 1.6116324324324325, "grad_norm": 33.216243743896484, "learning_rate": 4.096287747920202e-07, "loss": 9.306434631347656, "loss_alignment": 0.77197265625, "loss_alignment_w": 0.385986328125, "loss_sft": 0.19656607881188393, "loss_total_v6": 0.5816521644592285, "step": 6988 }, { "epoch": 1.611863063063063, "grad_norm": 31.28342628479004, "learning_rate": 4.088308433240157e-07, "loss": 9.194367408752441, "loss_alignment": 0.74658203125, "loss_alignment_w": 0.373291015625, "loss_sft": 0.20216566137969494, "loss_total_v6": 0.5746479630470276, "step": 6989 }, { "epoch": 1.6120936936936938, "grad_norm": 28.18913459777832, "learning_rate": 4.0803365664599694e-07, "loss": 8.827881813049316, "loss_alignment": 0.75537109375, "loss_alignment_w": 0.377685546875, "loss_sft": 0.17489626351743937, "loss_total_v6": 0.5517426021397114, "step": 6990 }, { "epoch": 1.6123243243243244, "grad_norm": 30.03456687927246, "learning_rate": 4.0723721488728384e-07, "loss": 9.013411521911621, "loss_alignment": 0.73046875, "loss_alignment_w": 0.365234375, "loss_sft": 0.19865316711366177, "loss_total_v6": 0.5633382052183151, "step": 6991 }, { "epoch": 1.612554954954955, "grad_norm": 32.38825607299805, "learning_rate": 4.064415181770787e-07, "loss": 9.231709480285645, "loss_alignment": 0.755859375, "loss_alignment_w": 0.3779296875, "loss_sft": 0.19873172044754028, "loss_total_v6": 0.5769818350672722, "step": 6992 }, { "epoch": 1.6127855855855855, "grad_norm": 30.336063385009766, "learning_rate": 4.0564656664446144e-07, "loss": 9.160135269165039, "loss_alignment": 0.76416015625, "loss_alignment_w": 0.382080078125, "loss_sft": 0.18966546934098005, "loss_total_v6": 0.5725084915757179, "step": 6993 }, { "epoch": 1.6130162162162163, "grad_norm": 32.12842559814453, "learning_rate": 4.048523604183885e-07, "loss": 8.596883773803711, "loss_alignment": 0.7353515625, "loss_alignment_w": 0.36767578125, "loss_sft": 0.16976676136255264, "loss_total_v6": 0.5373052358627319, "step": 6994 }, { "epoch": 1.6132468468468468, "grad_norm": 36.41461944580078, "learning_rate": 4.040588996277006e-07, "loss": 9.016511917114258, "loss_alignment": 0.76904296875, "loss_alignment_w": 0.384521484375, "loss_sft": 0.17934618704020977, "loss_total_v6": 0.5635319799184799, "step": 6995 }, { "epoch": 1.6134774774774776, "grad_norm": 35.70296096801758, "learning_rate": 4.032661844011132e-07, "loss": 8.85542106628418, "loss_alignment": 0.7373046875, "loss_alignment_w": 0.36865234375, "loss_sft": 0.1851471597328782, "loss_total_v6": 0.5534637831151485, "step": 6996 }, { "epoch": 1.6137081081081082, "grad_norm": 32.69444274902344, "learning_rate": 4.0247421486722226e-07, "loss": 9.01541519165039, "loss_alignment": 0.76806640625, "loss_alignment_w": 0.384033203125, "loss_sft": 0.18066617660224438, "loss_total_v6": 0.5634634122252464, "step": 6997 }, { "epoch": 1.6139387387387387, "grad_norm": 34.30086898803711, "learning_rate": 4.016829911545045e-07, "loss": 8.738080978393555, "loss_alignment": 0.744140625, "loss_alignment_w": 0.3720703125, "loss_sft": 0.17364774830639362, "loss_total_v6": 0.5461300387978554, "step": 6998 }, { "epoch": 1.6141693693693693, "grad_norm": 28.278383255004883, "learning_rate": 4.0089251339131164e-07, "loss": 9.409034729003906, "loss_alignment": 0.75, "loss_alignment_w": 0.375, "loss_sft": 0.211630342528224, "loss_total_v6": 0.588064655661583, "step": 6999 }, { "epoch": 1.6143999999999998, "grad_norm": 32.64834213256836, "learning_rate": 4.001027817058789e-07, "loss": 9.496053695678711, "loss_alignment": 0.72265625, "loss_alignment_w": 0.361328125, "loss_sft": 0.23299918696284294, "loss_total_v6": 0.5935033485293388, "step": 7000 }, { "epoch": 1.6143999999999998, "eval_loss": 0.8671430349349976, "eval_loss_alignment": 0.7691410709189498, "eval_loss_alignment_w": 0.3845705354594749, "eval_loss_contrast": 0.0, "eval_loss_highlevel": 0.0, "eval_loss_sft": 0.4825724924502035, "eval_loss_total_v6": 0.8671430295596928, "eval_loss_wm_recon": 0.0, "eval_loss_wm_recon_w": 0.0, "eval_runtime": 321.4171, "eval_samples_per_second": 5.429, "eval_steps_per_second": 0.681, "step": 7000 } ], "logging_steps": 1, "max_steps": 8000, "num_input_tokens_seen": 0, "num_train_epochs": 2, "save_steps": 200, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 1.4764858926218346e+20, "train_batch_size": 1, "trial_name": null, "trial_params": null }