{ "best_global_step": 3000, "best_metric": 0.92140538, "best_model_checkpoint": "/NHNHOME/data/sanghyeok/qwen-cua/runs/cosmos3_exact_full4b_8gpu/v2-20260624-130430/checkpoint-3000", "epoch": 0.7841441441441441, "eval_steps": 500, "global_step": 3400, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.00023063063063063062, "grad_norm": 44055.66015625, "learning_rate": 5.0000000000000004e-08, "loss": 27.716018676757812, "loss_alignment": 2.701171875, "loss_alignment_w": 1.3505859375, "loss_sft": 0.3752564750611782, "loss_total_v6": 1.7322510778903961, "step": 1 }, { "epoch": 0.00046126126126126124, "grad_norm": 48786.35546875, "learning_rate": 1.0000000000000001e-07, "loss": 26.97146224975586, "loss_alignment": 2.69921875, "loss_alignment_w": 1.349609375, "loss_sft": 0.33329951018095016, "loss_total_v6": 1.6857164949178696, "step": 2 }, { "epoch": 0.0006918918918918919, "grad_norm": 6249.36181640625, "learning_rate": 1.5000000000000002e-07, "loss": 26.583995819091797, "loss_alignment": 2.673828125, "loss_alignment_w": 1.3369140625, "loss_sft": 0.32513509690761566, "loss_total_v6": 1.6614998430013657, "step": 3 }, { "epoch": 0.0009225225225225225, "grad_norm": 6446.87451171875, "learning_rate": 2.0000000000000002e-07, "loss": 26.677371978759766, "loss_alignment": 2.666015625, "loss_alignment_w": 1.3330078125, "loss_sft": 0.33133721724152565, "loss_total_v6": 1.667335718870163, "step": 4 }, { "epoch": 0.0011531531531531532, "grad_norm": 5758.02880859375, "learning_rate": 2.5000000000000004e-07, "loss": 25.693408966064453, "loss_alignment": 2.533203125, "loss_alignment_w": 1.2666015625, "loss_sft": 0.3420440126210451, "loss_total_v6": 1.6058379560709, "step": 5 }, { "epoch": 0.0013837837837837837, "grad_norm": 5341.3193359375, "learning_rate": 3.0000000000000004e-07, "loss": 24.180652618408203, "loss_alignment": 2.396484375, "loss_alignment_w": 1.1982421875, "loss_sft": 0.3108512908220291, "loss_total_v6": 1.5112907737493515, "step": 6 }, { "epoch": 0.0016144144144144145, "grad_norm": 3942.52978515625, "learning_rate": 3.5000000000000004e-07, "loss": 23.897268295288086, "loss_alignment": 2.29296875, "loss_alignment_w": 1.146484375, "loss_sft": 0.3504517748951912, "loss_total_v6": 1.4935792237520218, "step": 7 }, { "epoch": 0.001845045045045045, "grad_norm": 3965.2373046875, "learning_rate": 4.0000000000000003e-07, "loss": 21.687231063842773, "loss_alignment": 2.09375, "loss_alignment_w": 1.046875, "loss_sft": 0.3079055957496166, "loss_total_v6": 1.3554520159959793, "step": 8 }, { "epoch": 0.0020756756756756755, "grad_norm": 3433.108154296875, "learning_rate": 4.5000000000000003e-07, "loss": 21.512481689453125, "loss_alignment": 2.056640625, "loss_alignment_w": 1.0283203125, "loss_sft": 0.3181324563920498, "loss_total_v6": 1.3445301800966263, "step": 9 }, { "epoch": 0.0023063063063063064, "grad_norm": 2985.976318359375, "learning_rate": 5.000000000000001e-07, "loss": 20.880176544189453, "loss_alignment": 2.0107421875, "loss_alignment_w": 1.00537109375, "loss_sft": 0.3010133430361748, "loss_total_v6": 1.3050111532211304, "step": 10 }, { "epoch": 0.002536936936936937, "grad_norm": 2623.389404296875, "learning_rate": 5.5e-07, "loss": 18.06399917602539, "loss_alignment": 1.6611328125, "loss_alignment_w": 0.83056640625, "loss_sft": 0.2983115017414093, "loss_total_v6": 1.129000037908554, "step": 11 }, { "epoch": 0.0027675675675675675, "grad_norm": 2442.382080078125, "learning_rate": 6.000000000000001e-07, "loss": 21.414426803588867, "loss_alignment": 1.9599609375, "loss_alignment_w": 0.97998046875, "loss_sft": 0.3583601675927639, "loss_total_v6": 1.3384016901254654, "step": 12 }, { "epoch": 0.0029981981981981984, "grad_norm": 1949.71435546875, "learning_rate": 6.5e-07, "loss": 18.22433090209961, "loss_alignment": 1.63671875, "loss_alignment_w": 0.818359375, "loss_sft": 0.3238656409084797, "loss_total_v6": 1.1390206515789032, "step": 13 }, { "epoch": 0.003228828828828829, "grad_norm": 1681.1497802734375, "learning_rate": 7.000000000000001e-07, "loss": 18.035442352294922, "loss_alignment": 1.5859375, "loss_alignment_w": 0.79296875, "loss_sft": 0.33427687734365463, "loss_total_v6": 1.127215176820755, "step": 14 }, { "epoch": 0.0034594594594594594, "grad_norm": 1391.765869140625, "learning_rate": 7.5e-07, "loss": 17.168598175048828, "loss_alignment": 1.5224609375, "loss_alignment_w": 0.76123046875, "loss_sft": 0.31083036586642265, "loss_total_v6": 1.0730373710393906, "step": 15 }, { "epoch": 0.00369009009009009, "grad_norm": 1172.159423828125, "learning_rate": 8.000000000000001e-07, "loss": 16.660171508789062, "loss_alignment": 1.49609375, "loss_alignment_w": 0.748046875, "loss_sft": 0.2935801465064287, "loss_total_v6": 1.0412607938051224, "step": 16 }, { "epoch": 0.003920720720720721, "grad_norm": 1009.0827026367188, "learning_rate": 8.500000000000001e-07, "loss": 16.401819229125977, "loss_alignment": 1.392578125, "loss_alignment_w": 0.6962890625, "loss_sft": 0.32946547865867615, "loss_total_v6": 1.0251136794686317, "step": 17 }, { "epoch": 0.004151351351351351, "grad_norm": 972.0714721679688, "learning_rate": 9.000000000000001e-07, "loss": 16.811511993408203, "loss_alignment": 1.451171875, "loss_alignment_w": 0.7255859375, "loss_sft": 0.326354306191206, "loss_total_v6": 1.050719566643238, "step": 18 }, { "epoch": 0.004381981981981982, "grad_norm": 848.28955078125, "learning_rate": 9.500000000000001e-07, "loss": 15.874724388122559, "loss_alignment": 1.369140625, "loss_alignment_w": 0.6845703125, "loss_sft": 0.3079967349767685, "loss_total_v6": 0.9921702966094017, "step": 19 }, { "epoch": 0.004612612612612613, "grad_norm": 698.1627197265625, "learning_rate": 1.0000000000000002e-06, "loss": 15.42516040802002, "loss_alignment": 1.30859375, "loss_alignment_w": 0.654296875, "loss_sft": 0.3082192949950695, "loss_total_v6": 0.9640725329518318, "step": 20 }, { "epoch": 0.004843243243243243, "grad_norm": 658.3645629882812, "learning_rate": 1.0500000000000001e-06, "loss": 15.91256332397461, "loss_alignment": 1.3310546875, "loss_alignment_w": 0.66552734375, "loss_sft": 0.32888578437268734, "loss_total_v6": 0.9945351779460907, "step": 21 }, { "epoch": 0.005073873873873874, "grad_norm": 600.9816284179688, "learning_rate": 1.1e-06, "loss": 15.159897804260254, "loss_alignment": 1.265625, "loss_alignment_w": 0.6328125, "loss_sft": 0.3147726859897375, "loss_total_v6": 0.9474936574697495, "step": 22 }, { "epoch": 0.005304504504504505, "grad_norm": 595.5629272460938, "learning_rate": 1.1500000000000002e-06, "loss": 15.471635818481445, "loss_alignment": 1.2900390625, "loss_alignment_w": 0.64501953125, "loss_sft": 0.322720680385828, "loss_total_v6": 0.9669772610068321, "step": 23 }, { "epoch": 0.005535135135135135, "grad_norm": 554.8722534179688, "learning_rate": 1.2000000000000002e-06, "loss": 15.321836471557617, "loss_alignment": 1.296875, "loss_alignment_w": 0.6484375, "loss_sft": 0.30954347737133503, "loss_total_v6": 0.9576147869229317, "step": 24 }, { "epoch": 0.005765765765765766, "grad_norm": 507.2185974121094, "learning_rate": 1.25e-06, "loss": 15.533111572265625, "loss_alignment": 1.35546875, "loss_alignment_w": 0.677734375, "loss_sft": 0.29384798370301723, "loss_total_v6": 0.9708194360136986, "step": 25 }, { "epoch": 0.005996396396396397, "grad_norm": 508.4649658203125, "learning_rate": 1.3e-06, "loss": 14.73175048828125, "loss_alignment": 1.263671875, "loss_alignment_w": 0.6318359375, "loss_sft": 0.2878608237951994, "loss_total_v6": 0.9207343757152557, "step": 26 }, { "epoch": 0.006227027027027027, "grad_norm": 481.900634765625, "learning_rate": 1.3500000000000002e-06, "loss": 14.474512100219727, "loss_alignment": 1.2216796875, "loss_alignment_w": 0.61083984375, "loss_sft": 0.2923828028142452, "loss_total_v6": 0.9046570211648941, "step": 27 }, { "epoch": 0.006457657657657658, "grad_norm": 449.8094787597656, "learning_rate": 1.4000000000000001e-06, "loss": 14.205368995666504, "loss_alignment": 1.1923828125, "loss_alignment_w": 0.59619140625, "loss_sft": 0.28972149454057217, "loss_total_v6": 0.8878355473279953, "step": 28 }, { "epoch": 0.006688288288288288, "grad_norm": 451.9810791015625, "learning_rate": 1.45e-06, "loss": 14.611644744873047, "loss_alignment": 1.2060546875, "loss_alignment_w": 0.60302734375, "loss_sft": 0.30778953433036804, "loss_total_v6": 0.913227766752243, "step": 29 }, { "epoch": 0.006918918918918919, "grad_norm": 432.9547424316406, "learning_rate": 1.5e-06, "loss": 14.719690322875977, "loss_alignment": 1.23046875, "loss_alignment_w": 0.615234375, "loss_sft": 0.30505143105983734, "loss_total_v6": 0.9199806451797485, "step": 30 }, { "epoch": 0.00714954954954955, "grad_norm": 386.69744873046875, "learning_rate": 1.5500000000000002e-06, "loss": 14.27001953125, "loss_alignment": 1.2275390625, "loss_alignment_w": 0.61376953125, "loss_sft": 0.27902220375835896, "loss_total_v6": 0.891876183450222, "step": 31 }, { "epoch": 0.00738018018018018, "grad_norm": 407.2129821777344, "learning_rate": 1.6000000000000001e-06, "loss": 14.543943405151367, "loss_alignment": 1.228515625, "loss_alignment_w": 0.6142578125, "loss_sft": 0.29516590759158134, "loss_total_v6": 0.908996470272541, "step": 32 }, { "epoch": 0.007610810810810811, "grad_norm": 387.7650146484375, "learning_rate": 1.6500000000000003e-06, "loss": 14.218578338623047, "loss_alignment": 1.2060546875, "loss_alignment_w": 0.60302734375, "loss_sft": 0.2846572231501341, "loss_total_v6": 0.8886611387133598, "step": 33 }, { "epoch": 0.007841441441441442, "grad_norm": 373.52508544921875, "learning_rate": 1.7000000000000002e-06, "loss": 14.587968826293945, "loss_alignment": 1.2470703125, "loss_alignment_w": 0.62353515625, "loss_sft": 0.2890368439257145, "loss_total_v6": 0.9117480143904686, "step": 34 }, { "epoch": 0.008072072072072072, "grad_norm": 347.1417236328125, "learning_rate": 1.75e-06, "loss": 14.445755004882812, "loss_alignment": 1.2626953125, "loss_alignment_w": 0.63134765625, "loss_sft": 0.2717256098985672, "loss_total_v6": 0.9028596505522728, "step": 35 }, { "epoch": 0.008302702702702702, "grad_norm": 338.967529296875, "learning_rate": 1.8000000000000001e-06, "loss": 13.968194961547852, "loss_alignment": 1.2080078125, "loss_alignment_w": 0.60400390625, "loss_sft": 0.26797071285545826, "loss_total_v6": 0.8730121925473213, "step": 36 }, { "epoch": 0.008533333333333334, "grad_norm": 310.3964538574219, "learning_rate": 1.85e-06, "loss": 14.677375793457031, "loss_alignment": 1.2275390625, "loss_alignment_w": 0.61376953125, "loss_sft": 0.30237622559070587, "loss_total_v6": 0.9173359572887421, "step": 37 }, { "epoch": 0.008763963963963964, "grad_norm": 312.062744140625, "learning_rate": 1.9000000000000002e-06, "loss": 14.907491683959961, "loss_alignment": 1.2734375, "loss_alignment_w": 0.63671875, "loss_sft": 0.29666267707943916, "loss_total_v6": 0.9317182451486588, "step": 38 }, { "epoch": 0.008994594594594594, "grad_norm": 325.9468994140625, "learning_rate": 1.9500000000000004e-06, "loss": 14.315481185913086, "loss_alignment": 1.2216796875, "loss_alignment_w": 0.61083984375, "loss_sft": 0.283511521294713, "loss_total_v6": 0.8947175964713097, "step": 39 }, { "epoch": 0.009225225225225226, "grad_norm": 296.9216003417969, "learning_rate": 2.0000000000000003e-06, "loss": 14.814508438110352, "loss_alignment": 1.255859375, "loss_alignment_w": 0.6279296875, "loss_sft": 0.29678688012063503, "loss_total_v6": 0.9259067624807358, "step": 40 }, { "epoch": 0.009455855855855856, "grad_norm": 291.2731018066406, "learning_rate": 2.05e-06, "loss": 14.067778587341309, "loss_alignment": 1.2119140625, "loss_alignment_w": 0.60595703125, "loss_sft": 0.2741031125187874, "loss_total_v6": 0.8792361691594124, "step": 41 }, { "epoch": 0.009686486486486486, "grad_norm": 262.6504821777344, "learning_rate": 2.1000000000000002e-06, "loss": 14.389019966125488, "loss_alignment": 1.2021484375, "loss_alignment_w": 0.60107421875, "loss_sft": 0.29744610004127026, "loss_total_v6": 0.8993137180805206, "step": 42 }, { "epoch": 0.009917117117117118, "grad_norm": 268.79656982421875, "learning_rate": 2.15e-06, "loss": 14.115217208862305, "loss_alignment": 1.22265625, "loss_alignment_w": 0.611328125, "loss_sft": 0.2702320832759142, "loss_total_v6": 0.8822011202573776, "step": 43 }, { "epoch": 0.010147747747747748, "grad_norm": 255.2376251220703, "learning_rate": 2.2e-06, "loss": 14.011499404907227, "loss_alignment": 1.20703125, "loss_alignment_w": 0.603515625, "loss_sft": 0.27290501073002815, "loss_total_v6": 0.8757187202572823, "step": 44 }, { "epoch": 0.010378378378378378, "grad_norm": 271.8517150878906, "learning_rate": 2.25e-06, "loss": 14.13435173034668, "loss_alignment": 1.2158203125, "loss_alignment_w": 0.60791015625, "loss_sft": 0.2768295742571354, "loss_total_v6": 0.8833969756960869, "step": 45 }, { "epoch": 0.01060900900900901, "grad_norm": 269.3031005859375, "learning_rate": 2.3000000000000004e-06, "loss": 14.212103843688965, "loss_alignment": 1.232421875, "loss_alignment_w": 0.6162109375, "loss_sft": 0.27158775366842747, "loss_total_v6": 0.8882564827799797, "step": 46 }, { "epoch": 0.01083963963963964, "grad_norm": 264.43377685546875, "learning_rate": 2.35e-06, "loss": 14.934904098510742, "loss_alignment": 1.2333984375, "loss_alignment_w": 0.61669921875, "loss_sft": 0.3156947195529938, "loss_total_v6": 0.9334315359592438, "step": 47 }, { "epoch": 0.01107027027027027, "grad_norm": 260.1601867675781, "learning_rate": 2.4000000000000003e-06, "loss": 14.420863151550293, "loss_alignment": 1.228515625, "loss_alignment_w": 0.6142578125, "loss_sft": 0.2875038683414459, "loss_total_v6": 0.9013039246201515, "step": 48 }, { "epoch": 0.011300900900900902, "grad_norm": 245.56790161132812, "learning_rate": 2.4500000000000003e-06, "loss": 13.685867309570312, "loss_alignment": 1.228515625, "loss_alignment_w": 0.6142578125, "loss_sft": 0.24178031645715237, "loss_total_v6": 0.8553667441010475, "step": 49 }, { "epoch": 0.011531531531531532, "grad_norm": 244.42799377441406, "learning_rate": 2.5e-06, "loss": 13.842784881591797, "loss_alignment": 1.203125, "loss_alignment_w": 0.6015625, "loss_sft": 0.26269599981606007, "loss_total_v6": 0.8651740401983261, "step": 50 }, { "epoch": 0.011762162162162162, "grad_norm": 236.64581298828125, "learning_rate": 2.55e-06, "loss": 13.43104362487793, "loss_alignment": 1.19921875, "loss_alignment_w": 0.599609375, "loss_sft": 0.24047170765697956, "loss_total_v6": 0.8394402116537094, "step": 51 }, { "epoch": 0.011992792792792794, "grad_norm": 247.83189392089844, "learning_rate": 2.6e-06, "loss": 15.052587509155273, "loss_alignment": 1.2646484375, "loss_alignment_w": 0.63232421875, "loss_sft": 0.3081268146634102, "loss_total_v6": 0.9407867193222046, "step": 52 }, { "epoch": 0.012223423423423424, "grad_norm": 241.6230926513672, "learning_rate": 2.6500000000000005e-06, "loss": 14.645475387573242, "loss_alignment": 1.23828125, "loss_alignment_w": 0.619140625, "loss_sft": 0.29403476044535637, "loss_total_v6": 0.9153421744704247, "step": 53 }, { "epoch": 0.012454054054054054, "grad_norm": 237.98089599609375, "learning_rate": 2.7000000000000004e-06, "loss": 13.87900161743164, "loss_alignment": 1.2041015625, "loss_alignment_w": 0.60205078125, "loss_sft": 0.2674924973398447, "loss_total_v6": 0.8674375638365746, "step": 54 }, { "epoch": 0.012684684684684684, "grad_norm": 234.13612365722656, "learning_rate": 2.7500000000000004e-06, "loss": 14.362982749938965, "loss_alignment": 1.22265625, "loss_alignment_w": 0.611328125, "loss_sft": 0.28602258674800396, "loss_total_v6": 0.8976864144206047, "step": 55 }, { "epoch": 0.012915315315315316, "grad_norm": 226.46026611328125, "learning_rate": 2.8000000000000003e-06, "loss": 13.889102935791016, "loss_alignment": 1.18359375, "loss_alignment_w": 0.591796875, "loss_sft": 0.2765772994607687, "loss_total_v6": 0.8680689632892609, "step": 56 }, { "epoch": 0.013145945945945946, "grad_norm": 221.1312713623047, "learning_rate": 2.85e-06, "loss": 13.903371810913086, "loss_alignment": 1.1669921875, "loss_alignment_w": 0.58349609375, "loss_sft": 0.28485426492989063, "loss_total_v6": 0.8689607232809067, "step": 57 }, { "epoch": 0.013376576576576576, "grad_norm": 216.47772216796875, "learning_rate": 2.9e-06, "loss": 13.62054443359375, "loss_alignment": 1.189453125, "loss_alignment_w": 0.5947265625, "loss_sft": 0.2573509160429239, "loss_total_v6": 0.8512840047478676, "step": 58 }, { "epoch": 0.013607207207207208, "grad_norm": 216.10365295410156, "learning_rate": 2.95e-06, "loss": 14.80099868774414, "loss_alignment": 1.201171875, "loss_alignment_w": 0.6005859375, "loss_sft": 0.32559041678905487, "loss_total_v6": 0.9250624254345894, "step": 59 }, { "epoch": 0.013837837837837838, "grad_norm": 231.15672302246094, "learning_rate": 3e-06, "loss": 14.262417793273926, "loss_alignment": 1.2197265625, "loss_alignment_w": 0.60986328125, "loss_sft": 0.280866414308548, "loss_total_v6": 0.8914010897278786, "step": 60 }, { "epoch": 0.014068468468468468, "grad_norm": 209.95773315429688, "learning_rate": 3.05e-06, "loss": 13.713415145874023, "loss_alignment": 1.2216796875, "loss_alignment_w": 0.61083984375, "loss_sft": 0.2485068440437317, "loss_total_v6": 0.8570883870124817, "step": 61 }, { "epoch": 0.0142990990990991, "grad_norm": 194.0113525390625, "learning_rate": 3.1000000000000004e-06, "loss": 14.165609359741211, "loss_alignment": 1.193359375, "loss_alignment_w": 0.5966796875, "loss_sft": 0.2885488513857126, "loss_total_v6": 0.8853505700826645, "step": 62 }, { "epoch": 0.01452972972972973, "grad_norm": 186.09609985351562, "learning_rate": 3.1500000000000003e-06, "loss": 13.489688873291016, "loss_alignment": 1.1962890625, "loss_alignment_w": 0.59814453125, "loss_sft": 0.24511370062828064, "loss_total_v6": 0.8431056216359138, "step": 63 }, { "epoch": 0.01476036036036036, "grad_norm": 191.33477783203125, "learning_rate": 3.2000000000000003e-06, "loss": 13.933170318603516, "loss_alignment": 1.203125, "loss_alignment_w": 0.6015625, "loss_sft": 0.2683145999908447, "loss_total_v6": 0.8708231598138809, "step": 64 }, { "epoch": 0.014990990990990992, "grad_norm": 195.3978729248047, "learning_rate": 3.2500000000000002e-06, "loss": 14.485445022583008, "loss_alignment": 1.20703125, "loss_alignment_w": 0.603515625, "loss_sft": 0.2987729627639055, "loss_total_v6": 0.9053403586149216, "step": 65 }, { "epoch": 0.015221621621621622, "grad_norm": 190.88430786132812, "learning_rate": 3.3000000000000006e-06, "loss": 13.790287017822266, "loss_alignment": 1.1669921875, "loss_alignment_w": 0.58349609375, "loss_sft": 0.2782137580215931, "loss_total_v6": 0.8618929460644722, "step": 66 }, { "epoch": 0.015452252252252252, "grad_norm": 179.0372772216797, "learning_rate": 3.3500000000000005e-06, "loss": 13.756181716918945, "loss_alignment": 1.189453125, "loss_alignment_w": 0.5947265625, "loss_sft": 0.263936135917902, "loss_total_v6": 0.8597613573074341, "step": 67 }, { "epoch": 0.015682882882882884, "grad_norm": 182.0911865234375, "learning_rate": 3.4000000000000005e-06, "loss": 13.913677215576172, "loss_alignment": 1.142578125, "loss_alignment_w": 0.5712890625, "loss_sft": 0.2977665141224861, "loss_total_v6": 0.8696048706769943, "step": 68 }, { "epoch": 0.015913513513513514, "grad_norm": 176.15818786621094, "learning_rate": 3.45e-06, "loss": 13.764228820800781, "loss_alignment": 1.220703125, "loss_alignment_w": 0.6103515625, "loss_sft": 0.24884463101625443, "loss_total_v6": 0.8602643311023712, "step": 69 }, { "epoch": 0.016144144144144144, "grad_norm": 179.27667236328125, "learning_rate": 3.5e-06, "loss": 13.890657424926758, "loss_alignment": 1.1787109375, "loss_alignment_w": 0.58935546875, "loss_sft": 0.27871908247470856, "loss_total_v6": 0.8681660890579224, "step": 70 }, { "epoch": 0.016374774774774774, "grad_norm": 180.66119384765625, "learning_rate": 3.5500000000000003e-06, "loss": 14.049020767211914, "loss_alignment": 1.1787109375, "loss_alignment_w": 0.58935546875, "loss_sft": 0.286511005833745, "loss_total_v6": 0.8780637606978416, "step": 71 }, { "epoch": 0.016605405405405404, "grad_norm": 167.7742919921875, "learning_rate": 3.6000000000000003e-06, "loss": 13.457714080810547, "loss_alignment": 1.1787109375, "loss_alignment_w": 0.58935546875, "loss_sft": 0.2517516240477562, "loss_total_v6": 0.8411071076989174, "step": 72 }, { "epoch": 0.016836036036036037, "grad_norm": 166.8986358642578, "learning_rate": 3.65e-06, "loss": 13.571273803710938, "loss_alignment": 1.1767578125, "loss_alignment_w": 0.58837890625, "loss_sft": 0.25958161801099777, "loss_total_v6": 0.8482046574354172, "step": 73 }, { "epoch": 0.017066666666666667, "grad_norm": 167.96072387695312, "learning_rate": 3.7e-06, "loss": 14.430238723754883, "loss_alignment": 1.216796875, "loss_alignment_w": 0.6083984375, "loss_sft": 0.2934609726071358, "loss_total_v6": 0.9018899649381638, "step": 74 }, { "epoch": 0.017297297297297298, "grad_norm": 154.0998077392578, "learning_rate": 3.7500000000000005e-06, "loss": 13.597441673278809, "loss_alignment": 1.1533203125, "loss_alignment_w": 0.57666015625, "loss_sft": 0.27202025055885315, "loss_total_v6": 0.8498401045799255, "step": 75 }, { "epoch": 0.017527927927927928, "grad_norm": 168.01283264160156, "learning_rate": 3.8000000000000005e-06, "loss": 13.2606201171875, "loss_alignment": 1.134765625, "loss_alignment_w": 0.5673828125, "loss_sft": 0.2617111522704363, "loss_total_v6": 0.8287887573242188, "step": 76 }, { "epoch": 0.017758558558558558, "grad_norm": 168.0078125, "learning_rate": 3.85e-06, "loss": 14.53095531463623, "loss_alignment": 1.2275390625, "loss_alignment_w": 0.61376953125, "loss_sft": 0.29395740665495396, "loss_total_v6": 0.9081847071647644, "step": 77 }, { "epoch": 0.017989189189189188, "grad_norm": 150.99842834472656, "learning_rate": 3.900000000000001e-06, "loss": 13.623941421508789, "loss_alignment": 1.2041015625, "loss_alignment_w": 0.60205078125, "loss_sft": 0.24993382580578327, "loss_total_v6": 0.8514963090419769, "step": 78 }, { "epoch": 0.01821981981981982, "grad_norm": 152.7255859375, "learning_rate": 3.95e-06, "loss": 14.18734359741211, "loss_alignment": 1.208984375, "loss_alignment_w": 0.6044921875, "loss_sft": 0.281850578263402, "loss_total_v6": 0.8867090195417404, "step": 79 }, { "epoch": 0.01845045045045045, "grad_norm": 151.4736328125, "learning_rate": 4.000000000000001e-06, "loss": 13.82949161529541, "loss_alignment": 1.189453125, "loss_alignment_w": 0.5947265625, "loss_sft": 0.2725463453680277, "loss_total_v6": 0.8643432334065437, "step": 80 }, { "epoch": 0.01868108108108108, "grad_norm": 148.77044677734375, "learning_rate": 4.05e-06, "loss": 14.316167831420898, "loss_alignment": 1.1767578125, "loss_alignment_w": 0.58837890625, "loss_sft": 0.30714454129338264, "loss_total_v6": 0.8947604969143867, "step": 81 }, { "epoch": 0.01891171171171171, "grad_norm": 147.01669311523438, "learning_rate": 4.1e-06, "loss": 14.046710014343262, "loss_alignment": 1.185546875, "loss_alignment_w": 0.5927734375, "loss_sft": 0.28526798263192177, "loss_total_v6": 0.8779193609952927, "step": 82 }, { "epoch": 0.01914234234234234, "grad_norm": 143.641357421875, "learning_rate": 4.15e-06, "loss": 14.038230895996094, "loss_alignment": 1.1611328125, "loss_alignment_w": 0.58056640625, "loss_sft": 0.29935599118471146, "loss_total_v6": 0.8773894309997559, "step": 83 }, { "epoch": 0.019372972972972972, "grad_norm": 143.36941528320312, "learning_rate": 4.2000000000000004e-06, "loss": 14.178531646728516, "loss_alignment": 1.1884765625, "loss_alignment_w": 0.59423828125, "loss_sft": 0.2921335380524397, "loss_total_v6": 0.8861581981182098, "step": 84 }, { "epoch": 0.019603603603603605, "grad_norm": 140.5781707763672, "learning_rate": 4.25e-06, "loss": 13.880905151367188, "loss_alignment": 1.17578125, "loss_alignment_w": 0.587890625, "loss_sft": 0.28097815066576004, "loss_total_v6": 0.8675565496087074, "step": 85 }, { "epoch": 0.019834234234234235, "grad_norm": 142.82272338867188, "learning_rate": 4.3e-06, "loss": 14.16914176940918, "loss_alignment": 1.1845703125, "loss_alignment_w": 0.59228515625, "loss_sft": 0.2935913782566786, "loss_total_v6": 0.8855713903903961, "step": 86 }, { "epoch": 0.020064864864864865, "grad_norm": 133.07077026367188, "learning_rate": 4.350000000000001e-06, "loss": 13.752058982849121, "loss_alignment": 1.1533203125, "loss_alignment_w": 0.57666015625, "loss_sft": 0.2822636663913727, "loss_total_v6": 0.8595036789774895, "step": 87 }, { "epoch": 0.020295495495495496, "grad_norm": 125.5561294555664, "learning_rate": 4.4e-06, "loss": 13.473275184631348, "loss_alignment": 1.146484375, "loss_alignment_w": 0.5732421875, "loss_sft": 0.27069909311831, "loss_total_v6": 0.8420796990394592, "step": 88 }, { "epoch": 0.020526126126126126, "grad_norm": 116.549560546875, "learning_rate": 4.450000000000001e-06, "loss": 13.404184341430664, "loss_alignment": 1.146484375, "loss_alignment_w": 0.5732421875, "loss_sft": 0.26561792753636837, "loss_total_v6": 0.8377614766359329, "step": 89 }, { "epoch": 0.020756756756756756, "grad_norm": 123.78921508789062, "learning_rate": 4.5e-06, "loss": 13.425397872924805, "loss_alignment": 1.177734375, "loss_alignment_w": 0.5888671875, "loss_sft": 0.2508000396192074, "loss_total_v6": 0.8390874117612839, "step": 90 }, { "epoch": 0.020987387387387386, "grad_norm": 123.4843521118164, "learning_rate": 4.5500000000000005e-06, "loss": 13.843513488769531, "loss_alignment": 1.171875, "loss_alignment_w": 0.5859375, "loss_sft": 0.27836659736931324, "loss_total_v6": 0.8652196452021599, "step": 91 }, { "epoch": 0.02121801801801802, "grad_norm": 120.42823791503906, "learning_rate": 4.600000000000001e-06, "loss": 13.525322914123535, "loss_alignment": 1.1708984375, "loss_alignment_w": 0.58544921875, "loss_sft": 0.2603717241436243, "loss_total_v6": 0.8453326672315598, "step": 92 }, { "epoch": 0.02144864864864865, "grad_norm": 117.41284942626953, "learning_rate": 4.65e-06, "loss": 13.776941299438477, "loss_alignment": 1.1591796875, "loss_alignment_w": 0.57958984375, "loss_sft": 0.2807670868933201, "loss_total_v6": 0.8610588684678078, "step": 93 }, { "epoch": 0.02167927927927928, "grad_norm": 119.99378204345703, "learning_rate": 4.7e-06, "loss": 13.85025405883789, "loss_alignment": 1.142578125, "loss_alignment_w": 0.5712890625, "loss_sft": 0.2925512585788965, "loss_total_v6": 0.8656408712267876, "step": 94 }, { "epoch": 0.02190990990990991, "grad_norm": 112.3350830078125, "learning_rate": 4.75e-06, "loss": 14.001956939697266, "loss_alignment": 1.1376953125, "loss_alignment_w": 0.56884765625, "loss_sft": 0.30554225482046604, "loss_total_v6": 0.8751223161816597, "step": 95 }, { "epoch": 0.02214054054054054, "grad_norm": 115.1137466430664, "learning_rate": 4.800000000000001e-06, "loss": 13.74767780303955, "loss_alignment": 1.1572265625, "loss_alignment_w": 0.57861328125, "loss_sft": 0.2803113963454962, "loss_total_v6": 0.8592298626899719, "step": 96 }, { "epoch": 0.02237117117117117, "grad_norm": 109.43860626220703, "learning_rate": 4.85e-06, "loss": 13.340927124023438, "loss_alignment": 1.115234375, "loss_alignment_w": 0.5576171875, "loss_sft": 0.2780981343239546, "loss_total_v6": 0.8338079750537872, "step": 97 }, { "epoch": 0.022601801801801803, "grad_norm": 110.19187927246094, "learning_rate": 4.9000000000000005e-06, "loss": 13.141258239746094, "loss_alignment": 1.1484375, "loss_alignment_w": 0.57421875, "loss_sft": 0.24768970720469952, "loss_total_v6": 0.8213286027312279, "step": 98 }, { "epoch": 0.022832432432432433, "grad_norm": 104.47296905517578, "learning_rate": 4.95e-06, "loss": 14.030513763427734, "loss_alignment": 1.134765625, "loss_alignment_w": 0.5673828125, "loss_sft": 0.3059232924133539, "loss_total_v6": 0.8769071325659752, "step": 99 }, { "epoch": 0.023063063063063063, "grad_norm": 107.31938171386719, "learning_rate": 5e-06, "loss": 14.113164901733398, "loss_alignment": 1.1259765625, "loss_alignment_w": 0.56298828125, "loss_sft": 0.31894721649587154, "loss_total_v6": 0.8820728063583374, "step": 100 }, { "epoch": 0.023293693693693693, "grad_norm": 99.52151489257812, "learning_rate": 5.050000000000001e-06, "loss": 13.808984756469727, "loss_alignment": 1.1396484375, "loss_alignment_w": 0.56982421875, "loss_sft": 0.2914520613849163, "loss_total_v6": 0.8630615472793579, "step": 101 }, { "epoch": 0.023524324324324324, "grad_norm": 96.78509521484375, "learning_rate": 5.1e-06, "loss": 13.687822341918945, "loss_alignment": 1.1328125, "loss_alignment_w": 0.56640625, "loss_sft": 0.28926576115190983, "loss_total_v6": 0.8554888740181923, "step": 102 }, { "epoch": 0.023754954954954954, "grad_norm": 98.84925842285156, "learning_rate": 5.150000000000001e-06, "loss": 13.152119636535645, "loss_alignment": 1.1357421875, "loss_alignment_w": 0.56787109375, "loss_sft": 0.25441098771989346, "loss_total_v6": 0.8220074698328972, "step": 103 }, { "epoch": 0.023985585585585587, "grad_norm": 97.75030517578125, "learning_rate": 5.2e-06, "loss": 13.754663467407227, "loss_alignment": 1.1259765625, "loss_alignment_w": 0.56298828125, "loss_sft": 0.2968001961708069, "loss_total_v6": 0.8596664294600487, "step": 104 }, { "epoch": 0.024216216216216217, "grad_norm": 96.39006805419922, "learning_rate": 5.2500000000000006e-06, "loss": 13.39680290222168, "loss_alignment": 1.1396484375, "loss_alignment_w": 0.56982421875, "loss_sft": 0.2658585198223591, "loss_total_v6": 0.8373001739382744, "step": 105 }, { "epoch": 0.024446846846846847, "grad_norm": 96.93309783935547, "learning_rate": 5.300000000000001e-06, "loss": 13.202573776245117, "loss_alignment": 1.1494140625, "loss_alignment_w": 0.57470703125, "loss_sft": 0.2507589589804411, "loss_total_v6": 0.8251608237624168, "step": 106 }, { "epoch": 0.024677477477477477, "grad_norm": 94.83592224121094, "learning_rate": 5.3500000000000004e-06, "loss": 13.460107803344727, "loss_alignment": 1.1171875, "loss_alignment_w": 0.55859375, "loss_sft": 0.2846466153860092, "loss_total_v6": 0.8412567302584648, "step": 107 }, { "epoch": 0.024908108108108108, "grad_norm": 88.53374481201172, "learning_rate": 5.400000000000001e-06, "loss": 13.57213020324707, "loss_alignment": 1.1201171875, "loss_alignment_w": 0.56005859375, "loss_sft": 0.2883521318435669, "loss_total_v6": 0.8482581228017807, "step": 108 }, { "epoch": 0.025138738738738738, "grad_norm": 86.46110534667969, "learning_rate": 5.450000000000001e-06, "loss": 13.476811408996582, "loss_alignment": 1.13671875, "loss_alignment_w": 0.568359375, "loss_sft": 0.2739718444645405, "loss_total_v6": 0.8423006981611252, "step": 109 }, { "epoch": 0.025369369369369368, "grad_norm": 93.97364044189453, "learning_rate": 5.500000000000001e-06, "loss": 13.51689624786377, "loss_alignment": 1.142578125, "loss_alignment_w": 0.5712890625, "loss_sft": 0.2743409089744091, "loss_total_v6": 0.844806008040905, "step": 110 }, { "epoch": 0.0256, "grad_norm": 91.17902374267578, "learning_rate": 5.550000000000001e-06, "loss": 13.839126586914062, "loss_alignment": 1.1015625, "loss_alignment_w": 0.55078125, "loss_sft": 0.3148508779704571, "loss_total_v6": 0.8649454414844513, "step": 111 }, { "epoch": 0.02583063063063063, "grad_norm": 87.82675170898438, "learning_rate": 5.600000000000001e-06, "loss": 13.463619232177734, "loss_alignment": 1.1416015625, "loss_alignment_w": 0.57080078125, "loss_sft": 0.27180454693734646, "loss_total_v6": 0.8414761871099472, "step": 112 }, { "epoch": 0.02606126126126126, "grad_norm": 89.50186920166016, "learning_rate": 5.65e-06, "loss": 13.221548080444336, "loss_alignment": 1.103515625, "loss_alignment_w": 0.5517578125, "loss_sft": 0.27680152654647827, "loss_total_v6": 0.8263467997312546, "step": 113 }, { "epoch": 0.02629189189189189, "grad_norm": 86.0957260131836, "learning_rate": 5.7e-06, "loss": 12.967131614685059, "loss_alignment": 1.1279296875, "loss_alignment_w": 0.56396484375, "loss_sft": 0.24529065564274788, "loss_total_v6": 0.8104456886649132, "step": 114 }, { "epoch": 0.02652252252252252, "grad_norm": 83.08650970458984, "learning_rate": 5.75e-06, "loss": 13.796067237854004, "loss_alignment": 1.1171875, "loss_alignment_w": 0.55859375, "loss_sft": 0.30382828786969185, "loss_total_v6": 0.8622541949152946, "step": 115 }, { "epoch": 0.02675315315315315, "grad_norm": 82.06464385986328, "learning_rate": 5.8e-06, "loss": 12.981435775756836, "loss_alignment": 1.10546875, "loss_alignment_w": 0.552734375, "loss_sft": 0.2592768147587776, "loss_total_v6": 0.8113397657871246, "step": 116 }, { "epoch": 0.026983783783783785, "grad_norm": 85.19735717773438, "learning_rate": 5.85e-06, "loss": 13.01391315460205, "loss_alignment": 1.095703125, "loss_alignment_w": 0.5478515625, "loss_sft": 0.2667540032416582, "loss_total_v6": 0.81336959451437, "step": 117 }, { "epoch": 0.027214414414414415, "grad_norm": 86.35409545898438, "learning_rate": 5.9e-06, "loss": 13.397953033447266, "loss_alignment": 1.138671875, "loss_alignment_w": 0.5693359375, "loss_sft": 0.2693025954067707, "loss_total_v6": 0.8373720645904541, "step": 118 }, { "epoch": 0.027445045045045045, "grad_norm": 83.04368591308594, "learning_rate": 5.950000000000001e-06, "loss": 13.182487487792969, "loss_alignment": 1.1376953125, "loss_alignment_w": 0.56884765625, "loss_sft": 0.25658367574214935, "loss_total_v6": 0.8239054456353188, "step": 119 }, { "epoch": 0.027675675675675675, "grad_norm": 78.28678894042969, "learning_rate": 6e-06, "loss": 13.190361022949219, "loss_alignment": 1.123046875, "loss_alignment_w": 0.5615234375, "loss_sft": 0.2624316122382879, "loss_total_v6": 0.8243975639343262, "step": 120 }, { "epoch": 0.027906306306306305, "grad_norm": 76.18215942382812, "learning_rate": 6.0500000000000005e-06, "loss": 12.830856323242188, "loss_alignment": 1.1171875, "loss_alignment_w": 0.55859375, "loss_sft": 0.24347207881510258, "loss_total_v6": 0.8019285127520561, "step": 121 }, { "epoch": 0.028136936936936936, "grad_norm": 74.88934326171875, "learning_rate": 6.1e-06, "loss": 13.68054485321045, "loss_alignment": 1.1142578125, "loss_alignment_w": 0.55712890625, "loss_sft": 0.2979050986468792, "loss_total_v6": 0.8550340235233307, "step": 122 }, { "epoch": 0.02836756756756757, "grad_norm": 74.35257720947266, "learning_rate": 6.15e-06, "loss": 13.168281555175781, "loss_alignment": 1.115234375, "loss_alignment_w": 0.5576171875, "loss_sft": 0.267200980335474, "loss_total_v6": 0.8230176195502281, "step": 123 }, { "epoch": 0.0285981981981982, "grad_norm": 75.08958435058594, "learning_rate": 6.200000000000001e-06, "loss": 13.980545043945312, "loss_alignment": 1.125, "loss_alignment_w": 0.5625, "loss_sft": 0.3107195422053337, "loss_total_v6": 0.8737840950489044, "step": 124 }, { "epoch": 0.02882882882882883, "grad_norm": 73.8500747680664, "learning_rate": 6.25e-06, "loss": 13.13800048828125, "loss_alignment": 1.1181640625, "loss_alignment_w": 0.55908203125, "loss_sft": 0.2621650453656912, "loss_total_v6": 0.8211249932646751, "step": 125 }, { "epoch": 0.02905945945945946, "grad_norm": 73.47252655029297, "learning_rate": 6.300000000000001e-06, "loss": 13.441648483276367, "loss_alignment": 1.1142578125, "loss_alignment_w": 0.55712890625, "loss_sft": 0.28295883536338806, "loss_total_v6": 0.8401030004024506, "step": 126 }, { "epoch": 0.02929009009009009, "grad_norm": 73.8149185180664, "learning_rate": 6.35e-06, "loss": 13.36951732635498, "loss_alignment": 1.1416015625, "loss_alignment_w": 0.57080078125, "loss_sft": 0.264458367601037, "loss_total_v6": 0.8355948105454445, "step": 127 }, { "epoch": 0.02952072072072072, "grad_norm": 70.55860137939453, "learning_rate": 6.4000000000000006e-06, "loss": 13.3710355758667, "loss_alignment": 1.1181640625, "loss_alignment_w": 0.55908203125, "loss_sft": 0.27694335021078587, "loss_total_v6": 0.8356897011399269, "step": 128 }, { "epoch": 0.029751351351351353, "grad_norm": 70.76268005371094, "learning_rate": 6.450000000000001e-06, "loss": 13.264083862304688, "loss_alignment": 1.1123046875, "loss_alignment_w": 0.55615234375, "loss_sft": 0.27205944433808327, "loss_total_v6": 0.8290052488446236, "step": 129 }, { "epoch": 0.029981981981981983, "grad_norm": 65.55786895751953, "learning_rate": 6.5000000000000004e-06, "loss": 13.351337432861328, "loss_alignment": 1.1162109375, "loss_alignment_w": 0.55810546875, "loss_sft": 0.2758648060262203, "loss_total_v6": 0.8344585299491882, "step": 130 }, { "epoch": 0.030212612612612613, "grad_norm": 69.58564758300781, "learning_rate": 6.550000000000001e-06, "loss": 13.43687915802002, "loss_alignment": 1.1298828125, "loss_alignment_w": 0.56494140625, "loss_sft": 0.27413113601505756, "loss_total_v6": 0.8398049771785736, "step": 131 }, { "epoch": 0.030443243243243243, "grad_norm": 65.12022399902344, "learning_rate": 6.600000000000001e-06, "loss": 14.033146858215332, "loss_alignment": 1.1396484375, "loss_alignment_w": 0.56982421875, "loss_sft": 0.30724750459194183, "loss_total_v6": 0.8770717158913612, "step": 132 }, { "epoch": 0.030673873873873873, "grad_norm": 66.03459930419922, "learning_rate": 6.650000000000001e-06, "loss": 12.945306777954102, "loss_alignment": 1.080078125, "loss_alignment_w": 0.5400390625, "loss_sft": 0.2695156577974558, "loss_total_v6": 0.8090816736221313, "step": 133 }, { "epoch": 0.030904504504504503, "grad_norm": 65.92438507080078, "learning_rate": 6.700000000000001e-06, "loss": 13.492109298706055, "loss_alignment": 1.109375, "loss_alignment_w": 0.5546875, "loss_sft": 0.28805045783519745, "loss_total_v6": 0.8432567790150642, "step": 134 }, { "epoch": 0.031135135135135134, "grad_norm": 64.29178619384766, "learning_rate": 6.750000000000001e-06, "loss": 12.979008674621582, "loss_alignment": 1.103515625, "loss_alignment_w": 0.5517578125, "loss_sft": 0.2584536727517843, "loss_total_v6": 0.8111880347132683, "step": 135 }, { "epoch": 0.03136576576576577, "grad_norm": 68.376708984375, "learning_rate": 6.800000000000001e-06, "loss": 13.160287857055664, "loss_alignment": 1.11328125, "loss_alignment_w": 0.556640625, "loss_sft": 0.26599937677383423, "loss_total_v6": 0.8225179612636566, "step": 136 }, { "epoch": 0.031596396396396394, "grad_norm": 62.36277770996094, "learning_rate": 6.850000000000001e-06, "loss": 13.602091789245605, "loss_alignment": 1.126953125, "loss_alignment_w": 0.5634765625, "loss_sft": 0.28744759783148766, "loss_total_v6": 0.8501307219266891, "step": 137 }, { "epoch": 0.03182702702702703, "grad_norm": 62.78669357299805, "learning_rate": 6.9e-06, "loss": 12.850295066833496, "loss_alignment": 1.11328125, "loss_alignment_w": 0.556640625, "loss_sft": 0.2483033649623394, "loss_total_v6": 0.8031434416770935, "step": 138 }, { "epoch": 0.03205765765765766, "grad_norm": 60.1368293762207, "learning_rate": 6.95e-06, "loss": 13.315153121948242, "loss_alignment": 1.0947265625, "loss_alignment_w": 0.54736328125, "loss_sft": 0.285291550680995, "loss_total_v6": 0.8321970850229263, "step": 139 }, { "epoch": 0.03228828828828829, "grad_norm": 61.84072494506836, "learning_rate": 7e-06, "loss": 13.223681449890137, "loss_alignment": 1.083984375, "loss_alignment_w": 0.5419921875, "loss_sft": 0.283648693934083, "loss_total_v6": 0.8264801055192947, "step": 140 }, { "epoch": 0.03251891891891892, "grad_norm": 65.74781799316406, "learning_rate": 7.05e-06, "loss": 13.232865333557129, "loss_alignment": 1.1103515625, "loss_alignment_w": 0.55517578125, "loss_sft": 0.2732210885733366, "loss_total_v6": 0.8270541131496429, "step": 141 }, { "epoch": 0.03274954954954955, "grad_norm": 62.495208740234375, "learning_rate": 7.100000000000001e-06, "loss": 13.33080005645752, "loss_alignment": 1.08984375, "loss_alignment_w": 0.544921875, "loss_sft": 0.28974851220846176, "loss_total_v6": 0.8331750109791756, "step": 142 }, { "epoch": 0.03298018018018018, "grad_norm": 60.50662612915039, "learning_rate": 7.15e-06, "loss": 12.816092491149902, "loss_alignment": 1.0751953125, "loss_alignment_w": 0.53759765625, "loss_sft": 0.26189751736819744, "loss_total_v6": 0.8010057881474495, "step": 143 }, { "epoch": 0.03321081081081081, "grad_norm": 60.58003234863281, "learning_rate": 7.2000000000000005e-06, "loss": 13.767581939697266, "loss_alignment": 1.1142578125, "loss_alignment_w": 0.55712890625, "loss_sft": 0.3038485236465931, "loss_total_v6": 0.8604738563299179, "step": 144 }, { "epoch": 0.03344144144144144, "grad_norm": 57.876487731933594, "learning_rate": 7.25e-06, "loss": 13.287076950073242, "loss_alignment": 1.09765625, "loss_alignment_w": 0.548828125, "loss_sft": 0.28190407156944275, "loss_total_v6": 0.8304423093795776, "step": 145 }, { "epoch": 0.033672072072072075, "grad_norm": 56.041481018066406, "learning_rate": 7.3e-06, "loss": 13.711538314819336, "loss_alignment": 1.1044921875, "loss_alignment_w": 0.55224609375, "loss_sft": 0.3062509298324585, "loss_total_v6": 0.8569711446762085, "step": 146 }, { "epoch": 0.0339027027027027, "grad_norm": 59.0147705078125, "learning_rate": 7.350000000000001e-06, "loss": 12.918313980102539, "loss_alignment": 1.1064453125, "loss_alignment_w": 0.55322265625, "loss_sft": 0.25382106378674507, "loss_total_v6": 0.8073946610093117, "step": 147 }, { "epoch": 0.034133333333333335, "grad_norm": 57.39244842529297, "learning_rate": 7.4e-06, "loss": 13.449767112731934, "loss_alignment": 1.11328125, "loss_alignment_w": 0.556640625, "loss_sft": 0.28435127809643745, "loss_total_v6": 0.8406104445457458, "step": 148 }, { "epoch": 0.03436396396396396, "grad_norm": 54.66657257080078, "learning_rate": 7.450000000000001e-06, "loss": 13.543057441711426, "loss_alignment": 1.126953125, "loss_alignment_w": 0.5634765625, "loss_sft": 0.2846430093050003, "loss_total_v6": 0.8464411050081253, "step": 149 }, { "epoch": 0.034594594594594595, "grad_norm": 52.394287109375, "learning_rate": 7.500000000000001e-06, "loss": 13.48044204711914, "loss_alignment": 1.1201171875, "loss_alignment_w": 0.56005859375, "loss_sft": 0.28280476853251457, "loss_total_v6": 0.8425276204943657, "step": 150 }, { "epoch": 0.03482522522522523, "grad_norm": 51.04662322998047, "learning_rate": 7.5500000000000006e-06, "loss": 13.871726989746094, "loss_alignment": 1.0888671875, "loss_alignment_w": 0.54443359375, "loss_sft": 0.32186271622776985, "loss_total_v6": 0.866982951760292, "step": 151 }, { "epoch": 0.035055855855855855, "grad_norm": 50.91852951049805, "learning_rate": 7.600000000000001e-06, "loss": 12.961892127990723, "loss_alignment": 1.1220703125, "loss_alignment_w": 0.56103515625, "loss_sft": 0.24981548078358173, "loss_total_v6": 0.8101182579994202, "step": 152 }, { "epoch": 0.03528648648648649, "grad_norm": 54.238731384277344, "learning_rate": 7.650000000000001e-06, "loss": 13.269562721252441, "loss_alignment": 1.1298828125, "loss_alignment_w": 0.56494140625, "loss_sft": 0.2666340693831444, "loss_total_v6": 0.8293476924300194, "step": 153 }, { "epoch": 0.035517117117117115, "grad_norm": 54.40208435058594, "learning_rate": 7.7e-06, "loss": 12.891376495361328, "loss_alignment": 1.0927734375, "loss_alignment_w": 0.54638671875, "loss_sft": 0.2590953949838877, "loss_total_v6": 0.8057110011577606, "step": 154 }, { "epoch": 0.03574774774774775, "grad_norm": 63.11353302001953, "learning_rate": 7.75e-06, "loss": 13.567313194274902, "loss_alignment": 1.1181640625, "loss_alignment_w": 0.55908203125, "loss_sft": 0.2880816347897053, "loss_total_v6": 0.8479570895433426, "step": 155 }, { "epoch": 0.035978378378378376, "grad_norm": 51.617923736572266, "learning_rate": 7.800000000000002e-06, "loss": 12.81447696685791, "loss_alignment": 1.099609375, "loss_alignment_w": 0.5498046875, "loss_sft": 0.25268705375492573, "loss_total_v6": 0.8009048178792, "step": 156 }, { "epoch": 0.03620900900900901, "grad_norm": 49.52731704711914, "learning_rate": 7.850000000000001e-06, "loss": 12.989660263061523, "loss_alignment": 1.0986328125, "loss_alignment_w": 0.54931640625, "loss_sft": 0.2627815082669258, "loss_total_v6": 0.8118537440896034, "step": 157 }, { "epoch": 0.03643963963963964, "grad_norm": 52.57988357543945, "learning_rate": 7.9e-06, "loss": 13.334933280944824, "loss_alignment": 1.0849609375, "loss_alignment_w": 0.54248046875, "loss_sft": 0.290067832916975, "loss_total_v6": 0.8334333300590515, "step": 158 }, { "epoch": 0.03667027027027027, "grad_norm": 50.10713195800781, "learning_rate": 7.950000000000002e-06, "loss": 13.198716163635254, "loss_alignment": 1.0771484375, "loss_alignment_w": 0.53857421875, "loss_sft": 0.2853689882904291, "loss_total_v6": 0.824919767677784, "step": 159 }, { "epoch": 0.0369009009009009, "grad_norm": 51.56443405151367, "learning_rate": 8.000000000000001e-06, "loss": 12.239351272583008, "loss_alignment": 1.0947265625, "loss_alignment_w": 0.54736328125, "loss_sft": 0.2172453086823225, "loss_total_v6": 0.764959491789341, "step": 160 }, { "epoch": 0.03713153153153153, "grad_norm": 51.66636276245117, "learning_rate": 8.050000000000001e-06, "loss": 12.970361709594727, "loss_alignment": 1.09375, "loss_alignment_w": 0.546875, "loss_sft": 0.2645813152194023, "loss_total_v6": 0.810647614300251, "step": 161 }, { "epoch": 0.03736216216216216, "grad_norm": 53.97978591918945, "learning_rate": 8.1e-06, "loss": 13.446555137634277, "loss_alignment": 1.09765625, "loss_alignment_w": 0.548828125, "loss_sft": 0.29123056307435036, "loss_total_v6": 0.84040966629982, "step": 162 }, { "epoch": 0.03759279279279279, "grad_norm": 50.58905792236328, "learning_rate": 8.15e-06, "loss": 12.752357482910156, "loss_alignment": 1.0732421875, "loss_alignment_w": 0.53662109375, "loss_sft": 0.2589821256697178, "loss_total_v6": 0.7970222979784012, "step": 163 }, { "epoch": 0.03782342342342342, "grad_norm": 49.50205993652344, "learning_rate": 8.2e-06, "loss": 13.142675399780273, "loss_alignment": 1.0908203125, "loss_alignment_w": 0.54541015625, "loss_sft": 0.2758239172399044, "loss_total_v6": 0.8214171826839447, "step": 164 }, { "epoch": 0.03805405405405406, "grad_norm": 49.304935455322266, "learning_rate": 8.25e-06, "loss": 12.855070114135742, "loss_alignment": 1.109375, "loss_alignment_w": 0.5546875, "loss_sft": 0.2486780844628811, "loss_total_v6": 0.8034419044852257, "step": 165 }, { "epoch": 0.03828468468468468, "grad_norm": 47.374176025390625, "learning_rate": 8.3e-06, "loss": 13.28877067565918, "loss_alignment": 1.0908203125, "loss_alignment_w": 0.54541015625, "loss_sft": 0.2860382664948702, "loss_total_v6": 0.8305481523275375, "step": 166 }, { "epoch": 0.03851531531531532, "grad_norm": 47.1614875793457, "learning_rate": 8.35e-06, "loss": 13.174293518066406, "loss_alignment": 1.09375, "loss_alignment_w": 0.546875, "loss_sft": 0.2767319865524769, "loss_total_v6": 0.8233933597803116, "step": 167 }, { "epoch": 0.038745945945945943, "grad_norm": 45.64453887939453, "learning_rate": 8.400000000000001e-06, "loss": 13.513961791992188, "loss_alignment": 1.1005859375, "loss_alignment_w": 0.55029296875, "loss_sft": 0.293749812990427, "loss_total_v6": 0.8446225896477699, "step": 168 }, { "epoch": 0.03897657657657658, "grad_norm": 49.888458251953125, "learning_rate": 8.45e-06, "loss": 13.72689151763916, "loss_alignment": 1.1123046875, "loss_alignment_w": 0.55615234375, "loss_sft": 0.3032431975007057, "loss_total_v6": 0.8579306975007057, "step": 169 }, { "epoch": 0.03920720720720721, "grad_norm": 49.19955825805664, "learning_rate": 8.5e-06, "loss": 13.695128440856934, "loss_alignment": 1.083984375, "loss_alignment_w": 0.5419921875, "loss_sft": 0.31299208104610443, "loss_total_v6": 0.8559455573558807, "step": 170 }, { "epoch": 0.03943783783783784, "grad_norm": 43.519386291503906, "learning_rate": 8.550000000000001e-06, "loss": 13.24528694152832, "loss_alignment": 1.08203125, "loss_alignment_w": 0.541015625, "loss_sft": 0.28695211187005043, "loss_total_v6": 0.8278304263949394, "step": 171 }, { "epoch": 0.03966846846846847, "grad_norm": 45.04881286621094, "learning_rate": 8.6e-06, "loss": 13.433390617370605, "loss_alignment": 1.0869140625, "loss_alignment_w": 0.54345703125, "loss_sft": 0.29629775136709213, "loss_total_v6": 0.8395869582891464, "step": 172 }, { "epoch": 0.0398990990990991, "grad_norm": 44.50508117675781, "learning_rate": 8.65e-06, "loss": 12.966285705566406, "loss_alignment": 1.08984375, "loss_alignment_w": 0.544921875, "loss_sft": 0.26504372246563435, "loss_total_v6": 0.8103928864002228, "step": 173 }, { "epoch": 0.04012972972972973, "grad_norm": 44.07991409301758, "learning_rate": 8.700000000000001e-06, "loss": 12.781091690063477, "loss_alignment": 1.083984375, "loss_alignment_w": 0.5419921875, "loss_sft": 0.25732957758009434, "loss_total_v6": 0.7988182455301285, "step": 174 }, { "epoch": 0.04036036036036036, "grad_norm": 42.37984085083008, "learning_rate": 8.750000000000001e-06, "loss": 13.332256317138672, "loss_alignment": 1.11328125, "loss_alignment_w": 0.556640625, "loss_sft": 0.27796816639602184, "loss_total_v6": 0.8332660049200058, "step": 175 }, { "epoch": 0.04059099099099099, "grad_norm": 42.43459701538086, "learning_rate": 8.8e-06, "loss": 12.9505615234375, "loss_alignment": 1.0849609375, "loss_alignment_w": 0.54248046875, "loss_sft": 0.2678909860551357, "loss_total_v6": 0.8094101250171661, "step": 176 }, { "epoch": 0.040821621621621625, "grad_norm": 43.53514099121094, "learning_rate": 8.85e-06, "loss": 12.92202377319336, "loss_alignment": 1.06640625, "loss_alignment_w": 0.533203125, "loss_sft": 0.276513759046793, "loss_total_v6": 0.8076264709234238, "step": 177 }, { "epoch": 0.04105225225225225, "grad_norm": 43.090633392333984, "learning_rate": 8.900000000000001e-06, "loss": 12.821961402893066, "loss_alignment": 1.07421875, "loss_alignment_w": 0.537109375, "loss_sft": 0.26450742222368717, "loss_total_v6": 0.8013726100325584, "step": 178 }, { "epoch": 0.041282882882882885, "grad_norm": 44.495731353759766, "learning_rate": 8.95e-06, "loss": 13.346002578735352, "loss_alignment": 1.099609375, "loss_alignment_w": 0.5498046875, "loss_sft": 0.28387803956866264, "loss_total_v6": 0.8341251909732819, "step": 179 }, { "epoch": 0.04151351351351351, "grad_norm": 45.29352569580078, "learning_rate": 9e-06, "loss": 13.257087707519531, "loss_alignment": 1.076171875, "loss_alignment_w": 0.5380859375, "loss_sft": 0.2915501967072487, "loss_total_v6": 0.8285680040717125, "step": 180 }, { "epoch": 0.041744144144144145, "grad_norm": 39.76896286010742, "learning_rate": 9.050000000000001e-06, "loss": 13.082860946655273, "loss_alignment": 1.107421875, "loss_alignment_w": 0.5537109375, "loss_sft": 0.2639068625867367, "loss_total_v6": 0.817678838968277, "step": 181 }, { "epoch": 0.04197477477477477, "grad_norm": 40.33381652832031, "learning_rate": 9.100000000000001e-06, "loss": 13.115806579589844, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.29227209091186523, "loss_total_v6": 0.8197379037737846, "step": 182 }, { "epoch": 0.042205405405405405, "grad_norm": 38.33641052246094, "learning_rate": 9.15e-06, "loss": 13.10189151763916, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.29575118236243725, "loss_total_v6": 0.8188682496547699, "step": 183 }, { "epoch": 0.04243603603603604, "grad_norm": 39.22967529296875, "learning_rate": 9.200000000000002e-06, "loss": 13.224239349365234, "loss_alignment": 1.0673828125, "loss_alignment_w": 0.53369140625, "loss_sft": 0.2914503142237663, "loss_total_v6": 0.8265149891376495, "step": 184 }, { "epoch": 0.042666666666666665, "grad_norm": 42.19475555419922, "learning_rate": 9.250000000000001e-06, "loss": 12.936641693115234, "loss_alignment": 1.0625, "loss_alignment_w": 0.53125, "loss_sft": 0.2775647845119238, "loss_total_v6": 0.8085401207208633, "step": 185 }, { "epoch": 0.0428972972972973, "grad_norm": 42.628353118896484, "learning_rate": 9.3e-06, "loss": 12.979269027709961, "loss_alignment": 1.072265625, "loss_alignment_w": 0.5361328125, "loss_sft": 0.27487311139702797, "loss_total_v6": 0.8112042993307114, "step": 186 }, { "epoch": 0.043127927927927925, "grad_norm": 39.7728385925293, "learning_rate": 9.350000000000002e-06, "loss": 12.95268440246582, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.2834349777549505, "loss_total_v6": 0.8095427677035332, "step": 187 }, { "epoch": 0.04335855855855856, "grad_norm": 38.411678314208984, "learning_rate": 9.4e-06, "loss": 12.73936653137207, "loss_alignment": 1.068359375, "loss_alignment_w": 0.5341796875, "loss_sft": 0.2611914537847042, "loss_total_v6": 0.7962104082107544, "step": 188 }, { "epoch": 0.04358918918918919, "grad_norm": 40.58721160888672, "learning_rate": 9.450000000000001e-06, "loss": 12.538961410522461, "loss_alignment": 1.0654296875, "loss_alignment_w": 0.53271484375, "loss_sft": 0.25170268677175045, "loss_total_v6": 0.7836850881576538, "step": 189 }, { "epoch": 0.04381981981981982, "grad_norm": 41.650001525878906, "learning_rate": 9.5e-06, "loss": 13.179204940795898, "loss_alignment": 1.0654296875, "loss_alignment_w": 0.53271484375, "loss_sft": 0.290710786357522, "loss_total_v6": 0.8237003087997437, "step": 190 }, { "epoch": 0.04405045045045045, "grad_norm": 41.74937438964844, "learning_rate": 9.55e-06, "loss": 13.339052200317383, "loss_alignment": 1.0849609375, "loss_alignment_w": 0.54248046875, "loss_sft": 0.29198847338557243, "loss_total_v6": 0.8336907550692558, "step": 191 }, { "epoch": 0.04428108108108108, "grad_norm": 37.8038215637207, "learning_rate": 9.600000000000001e-06, "loss": 12.887645721435547, "loss_alignment": 1.060546875, "loss_alignment_w": 0.5302734375, "loss_sft": 0.274914488196373, "loss_total_v6": 0.8054778724908829, "step": 192 }, { "epoch": 0.04451171171171171, "grad_norm": 37.681705474853516, "learning_rate": 9.65e-06, "loss": 12.719635963439941, "loss_alignment": 1.0654296875, "loss_alignment_w": 0.53271484375, "loss_sft": 0.26281170919537544, "loss_total_v6": 0.7949772402644157, "step": 193 }, { "epoch": 0.04474234234234234, "grad_norm": 37.25526809692383, "learning_rate": 9.7e-06, "loss": 12.771921157836914, "loss_alignment": 1.072265625, "loss_alignment_w": 0.5361328125, "loss_sft": 0.26247849874198437, "loss_total_v6": 0.7982451021671295, "step": 194 }, { "epoch": 0.04497297297297297, "grad_norm": 37.87118911743164, "learning_rate": 9.75e-06, "loss": 12.975030899047852, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.28728822618722916, "loss_total_v6": 0.8109393864870071, "step": 195 }, { "epoch": 0.045203603603603607, "grad_norm": 37.92841720581055, "learning_rate": 9.800000000000001e-06, "loss": 13.371946334838867, "loss_alignment": 1.06640625, "loss_alignment_w": 0.533203125, "loss_sft": 0.30225359834730625, "loss_total_v6": 0.835746631026268, "step": 196 }, { "epoch": 0.04543423423423423, "grad_norm": 41.85969924926758, "learning_rate": 9.85e-06, "loss": 13.156015396118164, "loss_alignment": 1.0654296875, "loss_alignment_w": 0.53271484375, "loss_sft": 0.2884222660213709, "loss_total_v6": 0.8222509995102882, "step": 197 }, { "epoch": 0.04566486486486487, "grad_norm": 37.94037628173828, "learning_rate": 9.9e-06, "loss": 12.727303504943848, "loss_alignment": 1.0576171875, "loss_alignment_w": 0.52880859375, "loss_sft": 0.26663256995379925, "loss_total_v6": 0.7954564616084099, "step": 198 }, { "epoch": 0.04589549549549549, "grad_norm": 42.324745178222656, "learning_rate": 9.950000000000001e-06, "loss": 13.026988983154297, "loss_alignment": 1.0751953125, "loss_alignment_w": 0.53759765625, "loss_sft": 0.27726050466299057, "loss_total_v6": 0.8141867741942406, "step": 199 }, { "epoch": 0.04612612612612613, "grad_norm": 37.096336364746094, "learning_rate": 1e-05, "loss": 14.063127517700195, "loss_alignment": 1.0859375, "loss_alignment_w": 0.54296875, "loss_sft": 0.3337794877588749, "loss_total_v6": 0.8789454698562622, "step": 200 }, { "epoch": 0.04635675675675675, "grad_norm": 38.95556640625, "learning_rate": 9.99999959444427e-06, "loss": 13.373799324035645, "loss_alignment": 1.0693359375, "loss_alignment_w": 0.53466796875, "loss_sft": 0.300340024754405, "loss_total_v6": 0.835862472653389, "step": 201 }, { "epoch": 0.04658738738738739, "grad_norm": 36.12343215942383, "learning_rate": 9.999998377777142e-06, "loss": 12.802343368530273, "loss_alignment": 1.03369140625, "loss_alignment_w": 0.516845703125, "loss_sft": 0.28143916465342045, "loss_total_v6": 0.8001464456319809, "step": 202 }, { "epoch": 0.04681801801801802, "grad_norm": 34.93511962890625, "learning_rate": 9.999996349998818e-06, "loss": 13.3242769241333, "loss_alignment": 1.0625, "loss_alignment_w": 0.53125, "loss_sft": 0.3026159331202507, "loss_total_v6": 0.8327673152089119, "step": 203 }, { "epoch": 0.04704864864864865, "grad_norm": 35.41935729980469, "learning_rate": 9.999993511109623e-06, "loss": 12.673182487487793, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.26703423634171486, "loss_total_v6": 0.7920738980174065, "step": 204 }, { "epoch": 0.04727927927927928, "grad_norm": 35.374359130859375, "learning_rate": 9.999989861110017e-06, "loss": 13.626055717468262, "loss_alignment": 1.0888671875, "loss_alignment_w": 0.54443359375, "loss_sft": 0.3086749967187643, "loss_total_v6": 0.851628452539444, "step": 205 }, { "epoch": 0.04750990990990991, "grad_norm": 34.736568450927734, "learning_rate": 9.999985400000596e-06, "loss": 12.867684364318848, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.2761235460639, "loss_total_v6": 0.8042302429676056, "step": 206 }, { "epoch": 0.04774054054054054, "grad_norm": 35.05775833129883, "learning_rate": 9.99998012778208e-06, "loss": 12.939935684204102, "loss_alignment": 1.0546875, "loss_alignment_w": 0.52734375, "loss_sft": 0.28054776787757874, "loss_total_v6": 0.8087459951639175, "step": 207 }, { "epoch": 0.047971171171171174, "grad_norm": 36.847450256347656, "learning_rate": 9.999974044455328e-06, "loss": 13.210067749023438, "loss_alignment": 1.0595703125, "loss_alignment_w": 0.52978515625, "loss_sft": 0.29566096886992455, "loss_total_v6": 0.8256292268633842, "step": 208 }, { "epoch": 0.0482018018018018, "grad_norm": 39.126346588134766, "learning_rate": 9.999967150021323e-06, "loss": 13.546093940734863, "loss_alignment": 1.064453125, "loss_alignment_w": 0.5322265625, "loss_sft": 0.31583864241838455, "loss_total_v6": 0.8466308638453484, "step": 209 }, { "epoch": 0.048432432432432435, "grad_norm": 35.87642288208008, "learning_rate": 9.999959444481187e-06, "loss": 12.38776969909668, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.2510727271437645, "loss_total_v6": 0.7742355689406395, "step": 210 }, { "epoch": 0.04866306306306306, "grad_norm": 38.345375061035156, "learning_rate": 9.999950927836166e-06, "loss": 13.329920768737793, "loss_alignment": 1.0810546875, "loss_alignment_w": 0.54052734375, "loss_sft": 0.29173823073506355, "loss_total_v6": 0.8331200629472733, "step": 211 }, { "epoch": 0.048893693693693695, "grad_norm": 34.27293014526367, "learning_rate": 9.999941600087643e-06, "loss": 13.34693717956543, "loss_alignment": 1.0751953125, "loss_alignment_w": 0.53759765625, "loss_sft": 0.29621974378824234, "loss_total_v6": 0.8341836184263229, "step": 212 }, { "epoch": 0.04912432432432432, "grad_norm": 34.32239532470703, "learning_rate": 9.999931461237135e-06, "loss": 13.099061965942383, "loss_alignment": 1.080078125, "loss_alignment_w": 0.5400390625, "loss_sft": 0.2772484980523586, "loss_total_v6": 0.8186913579702377, "step": 213 }, { "epoch": 0.049354954954954955, "grad_norm": 36.953372955322266, "learning_rate": 9.999920511286282e-06, "loss": 13.017107009887695, "loss_alignment": 1.0869140625, "loss_alignment_w": 0.54345703125, "loss_sft": 0.2696086037904024, "loss_total_v6": 0.8135691806674004, "step": 214 }, { "epoch": 0.04958558558558559, "grad_norm": 41.26488494873047, "learning_rate": 9.999908750236862e-06, "loss": 12.83399486541748, "loss_alignment": 1.0615234375, "loss_alignment_w": 0.53076171875, "loss_sft": 0.2728430312126875, "loss_total_v6": 0.8021246716380119, "step": 215 }, { "epoch": 0.049816216216216215, "grad_norm": 33.2841796875, "learning_rate": 9.999896178090783e-06, "loss": 13.22225570678711, "loss_alignment": 1.0634765625, "loss_alignment_w": 0.53173828125, "loss_sft": 0.2940423861145973, "loss_total_v6": 0.8263910189270973, "step": 216 }, { "epoch": 0.05004684684684685, "grad_norm": 34.30658721923828, "learning_rate": 9.999882794850085e-06, "loss": 12.73524284362793, "loss_alignment": 1.0634765625, "loss_alignment_w": 0.53173828125, "loss_sft": 0.265114676207304, "loss_total_v6": 0.795952670276165, "step": 217 }, { "epoch": 0.050277477477477475, "grad_norm": 33.43052291870117, "learning_rate": 9.999868600516938e-06, "loss": 12.550050735473633, "loss_alignment": 1.056640625, "loss_alignment_w": 0.5283203125, "loss_sft": 0.25569164007902145, "loss_total_v6": 0.7843781635165215, "step": 218 }, { "epoch": 0.05050810810810811, "grad_norm": 36.89902877807617, "learning_rate": 9.999853595093645e-06, "loss": 13.001619338989258, "loss_alignment": 1.0693359375, "loss_alignment_w": 0.53466796875, "loss_sft": 0.27746016904711723, "loss_total_v6": 0.8126011714339256, "step": 219 }, { "epoch": 0.050738738738738735, "grad_norm": 33.61298751831055, "learning_rate": 9.999837778582641e-06, "loss": 12.805580139160156, "loss_alignment": 1.064453125, "loss_alignment_w": 0.5322265625, "loss_sft": 0.26867150142788887, "loss_total_v6": 0.8003487437963486, "step": 220 }, { "epoch": 0.05096936936936937, "grad_norm": 36.057106018066406, "learning_rate": 9.99982115098649e-06, "loss": 12.705789566040039, "loss_alignment": 1.0654296875, "loss_alignment_w": 0.53271484375, "loss_sft": 0.260328920558095, "loss_total_v6": 0.7941118776798248, "step": 221 }, { "epoch": 0.0512, "grad_norm": 32.35150146484375, "learning_rate": 9.99980371230789e-06, "loss": 12.836806297302246, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.2752160541713238, "loss_total_v6": 0.8023003935813904, "step": 222 }, { "epoch": 0.05143063063063063, "grad_norm": 35.096412658691406, "learning_rate": 9.999785462549672e-06, "loss": 13.615804672241211, "loss_alignment": 1.06640625, "loss_alignment_w": 0.533203125, "loss_sft": 0.3176015391945839, "loss_total_v6": 0.8509877622127533, "step": 223 }, { "epoch": 0.05166126126126126, "grad_norm": 36.14472198486328, "learning_rate": 9.999766401714795e-06, "loss": 13.381385803222656, "loss_alignment": 1.0625, "loss_alignment_w": 0.53125, "loss_sft": 0.3061242289841175, "loss_total_v6": 0.8363366276025772, "step": 224 }, { "epoch": 0.05189189189189189, "grad_norm": 38.78096389770508, "learning_rate": 9.99974652980635e-06, "loss": 13.008978843688965, "loss_alignment": 1.0703125, "loss_alignment_w": 0.53515625, "loss_sft": 0.2791866287589073, "loss_total_v6": 0.8130611702799797, "step": 225 }, { "epoch": 0.05212252252252252, "grad_norm": 34.376949310302734, "learning_rate": 9.999725846827562e-06, "loss": 13.01175308227539, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.2867452912032604, "loss_total_v6": 0.8132345452904701, "step": 226 }, { "epoch": 0.052353153153153156, "grad_norm": 37.289283752441406, "learning_rate": 9.999704352781785e-06, "loss": 13.083900451660156, "loss_alignment": 1.0693359375, "loss_alignment_w": 0.53466796875, "loss_sft": 0.2818398308008909, "loss_total_v6": 0.8177437782287598, "step": 227 }, { "epoch": 0.05258378378378378, "grad_norm": 30.594432830810547, "learning_rate": 9.999682047672508e-06, "loss": 12.83252239227295, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.2808992099016905, "loss_total_v6": 0.8020326346158981, "step": 228 }, { "epoch": 0.052814414414414416, "grad_norm": 38.139564514160156, "learning_rate": 9.999658931503347e-06, "loss": 12.980740547180176, "loss_alignment": 1.060546875, "loss_alignment_w": 0.5302734375, "loss_sft": 0.2819841243326664, "loss_total_v6": 0.8112962692975998, "step": 229 }, { "epoch": 0.05304504504504504, "grad_norm": 33.7392692565918, "learning_rate": 9.999635004278054e-06, "loss": 12.479188919067383, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.2542992737144232, "loss_total_v6": 0.779949314892292, "step": 230 }, { "epoch": 0.05327567567567568, "grad_norm": 33.85866928100586, "learning_rate": 9.99961026600051e-06, "loss": 12.78300952911377, "loss_alignment": 1.04736328125, "loss_alignment_w": 0.523681640625, "loss_sft": 0.2745087593793869, "loss_total_v6": 0.7989381030201912, "step": 231 }, { "epoch": 0.0535063063063063, "grad_norm": 33.146305084228516, "learning_rate": 9.999584716674727e-06, "loss": 12.52773666381836, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.25893561728298664, "loss_total_v6": 0.7829834967851639, "step": 232 }, { "epoch": 0.05373693693693694, "grad_norm": 30.186481475830078, "learning_rate": 9.99955835630485e-06, "loss": 13.144253730773926, "loss_alignment": 1.0625, "loss_alignment_w": 0.53125, "loss_sft": 0.2901896107941866, "loss_total_v6": 0.8215158805251122, "step": 233 }, { "epoch": 0.05396756756756757, "grad_norm": 34.628273010253906, "learning_rate": 9.999531184895156e-06, "loss": 12.837987899780273, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.2807677946984768, "loss_total_v6": 0.8023742586374283, "step": 234 }, { "epoch": 0.0541981981981982, "grad_norm": 34.402442932128906, "learning_rate": 9.999503202450052e-06, "loss": 13.118330001831055, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.29409305565059185, "loss_total_v6": 0.8198956623673439, "step": 235 }, { "epoch": 0.05442882882882883, "grad_norm": 31.06708335876465, "learning_rate": 9.999474408974079e-06, "loss": 13.259977340698242, "loss_alignment": 1.076171875, "loss_alignment_w": 0.5380859375, "loss_sft": 0.2925395052880049, "loss_total_v6": 0.8287485986948013, "step": 236 }, { "epoch": 0.05465945945945946, "grad_norm": 32.61185073852539, "learning_rate": 9.999444804471906e-06, "loss": 12.88361644744873, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.28696125000715256, "loss_total_v6": 0.8052260279655457, "step": 237 }, { "epoch": 0.05489009009009009, "grad_norm": 31.397132873535156, "learning_rate": 9.999414388948337e-06, "loss": 13.096808433532715, "loss_alignment": 1.064453125, "loss_alignment_w": 0.5322265625, "loss_sft": 0.2851643320173025, "loss_total_v6": 0.8185505568981171, "step": 238 }, { "epoch": 0.05512072072072072, "grad_norm": 32.771690368652344, "learning_rate": 9.999383162408303e-06, "loss": 13.212303161621094, "loss_alignment": 1.0673828125, "loss_alignment_w": 0.53369140625, "loss_sft": 0.29255054518580437, "loss_total_v6": 0.825768917798996, "step": 239 }, { "epoch": 0.05535135135135135, "grad_norm": 33.10356903076172, "learning_rate": 9.999351124856873e-06, "loss": 13.083105087280273, "loss_alignment": 1.0771484375, "loss_alignment_w": 0.53857421875, "loss_sft": 0.2794250827282667, "loss_total_v6": 0.8176940903067589, "step": 240 }, { "epoch": 0.055581981981981984, "grad_norm": 35.634117126464844, "learning_rate": 9.999318276299246e-06, "loss": 13.526121139526367, "loss_alignment": 1.064453125, "loss_alignment_w": 0.5322265625, "loss_sft": 0.31480400264263153, "loss_total_v6": 0.8453826010227203, "step": 241 }, { "epoch": 0.05581261261261261, "grad_norm": 36.33489990234375, "learning_rate": 9.999284616740746e-06, "loss": 13.100884437561035, "loss_alignment": 1.068359375, "loss_alignment_w": 0.5341796875, "loss_sft": 0.28424414061009884, "loss_total_v6": 0.8188052847981453, "step": 242 }, { "epoch": 0.056043243243243245, "grad_norm": 31.08350944519043, "learning_rate": 9.999250146186834e-06, "loss": 13.272086143493652, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.3035806901752949, "loss_total_v6": 0.8295053765177727, "step": 243 }, { "epoch": 0.05627387387387387, "grad_norm": 33.981651306152344, "learning_rate": 9.999214864643103e-06, "loss": 12.825410842895508, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.27544985711574554, "loss_total_v6": 0.8015881404280663, "step": 244 }, { "epoch": 0.056504504504504505, "grad_norm": 38.137603759765625, "learning_rate": 9.999178772115279e-06, "loss": 13.020441055297852, "loss_alignment": 1.05859375, "loss_alignment_w": 0.529296875, "loss_sft": 0.2841602489352226, "loss_total_v6": 0.8137775510549545, "step": 245 }, { "epoch": 0.05673513513513514, "grad_norm": 33.515411376953125, "learning_rate": 9.999141868609212e-06, "loss": 12.670469284057617, "loss_alignment": 1.0654296875, "loss_alignment_w": 0.53271484375, "loss_sft": 0.2591131366789341, "loss_total_v6": 0.7919042855501175, "step": 246 }, { "epoch": 0.056965765765765765, "grad_norm": 33.36079025268555, "learning_rate": 9.999104154130894e-06, "loss": 13.43793773651123, "loss_alignment": 1.08203125, "loss_alignment_w": 0.541015625, "loss_sft": 0.2994505912065506, "loss_total_v6": 0.8398711234331131, "step": 247 }, { "epoch": 0.0571963963963964, "grad_norm": 32.37788391113281, "learning_rate": 9.999065628686439e-06, "loss": 13.377452850341797, "loss_alignment": 1.0576171875, "loss_alignment_w": 0.52880859375, "loss_sft": 0.3087470009922981, "loss_total_v6": 0.8360907658934593, "step": 248 }, { "epoch": 0.057427027027027025, "grad_norm": 32.47895050048828, "learning_rate": 9.999026292282096e-06, "loss": 12.988849639892578, "loss_alignment": 1.0634765625, "loss_alignment_w": 0.53173828125, "loss_sft": 0.2801869176328182, "loss_total_v6": 0.8118031248450279, "step": 249 }, { "epoch": 0.05765765765765766, "grad_norm": 30.74537467956543, "learning_rate": 9.998986144924253e-06, "loss": 12.838815689086914, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.2772337608039379, "loss_total_v6": 0.8024260252714157, "step": 250 }, { "epoch": 0.057888288288288285, "grad_norm": 31.551433563232422, "learning_rate": 9.998945186619413e-06, "loss": 12.469417572021484, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.2556111868470907, "loss_total_v6": 0.7793386206030846, "step": 251 }, { "epoch": 0.05811891891891892, "grad_norm": 31.18221664428711, "learning_rate": 9.998903417374228e-06, "loss": 12.973625183105469, "loss_alignment": 1.068359375, "loss_alignment_w": 0.5341796875, "loss_sft": 0.27735854499042034, "loss_total_v6": 0.810851588845253, "step": 252 }, { "epoch": 0.05834954954954955, "grad_norm": 32.96342468261719, "learning_rate": 9.998860837195472e-06, "loss": 13.233652114868164, "loss_alignment": 1.046875, "loss_alignment_w": 0.5234375, "loss_sft": 0.30253661796450615, "loss_total_v6": 0.8271032497286797, "step": 253 }, { "epoch": 0.05858018018018018, "grad_norm": 29.228397369384766, "learning_rate": 9.99881744609005e-06, "loss": 13.127246856689453, "loss_alignment": 1.04150390625, "loss_alignment_w": 0.520751953125, "loss_sft": 0.299655195325613, "loss_total_v6": 0.820452943444252, "step": 254 }, { "epoch": 0.05881081081081081, "grad_norm": 32.02198028564453, "learning_rate": 9.998773244065003e-06, "loss": 13.094779014587402, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.29413168132305145, "loss_total_v6": 0.8184236660599709, "step": 255 }, { "epoch": 0.05904144144144144, "grad_norm": 36.15755081176758, "learning_rate": 9.998728231127501e-06, "loss": 13.481880187988281, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.3172878697514534, "loss_total_v6": 0.8426174819469452, "step": 256 }, { "epoch": 0.05927207207207207, "grad_norm": 57.245765686035156, "learning_rate": 9.998682407284848e-06, "loss": 12.766216278076172, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.2757175136357546, "loss_total_v6": 0.7978885322809219, "step": 257 }, { "epoch": 0.059502702702702706, "grad_norm": 33.71204376220703, "learning_rate": 9.998635772544475e-06, "loss": 13.48333740234375, "loss_alignment": 1.03955078125, "loss_alignment_w": 0.519775390625, "loss_sft": 0.32278066873550415, "loss_total_v6": 0.8427086398005486, "step": 258 }, { "epoch": 0.05973333333333333, "grad_norm": 29.844924926757812, "learning_rate": 9.99858832691395e-06, "loss": 12.731327056884766, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.2791674379259348, "loss_total_v6": 0.7957079410552979, "step": 259 }, { "epoch": 0.059963963963963966, "grad_norm": 30.017847061157227, "learning_rate": 9.998540070400966e-06, "loss": 12.900125503540039, "loss_alignment": 1.04052734375, "loss_alignment_w": 0.520263671875, "loss_sft": 0.2861009631305933, "loss_total_v6": 0.8062578365206718, "step": 260 }, { "epoch": 0.06019459459459459, "grad_norm": 31.156593322753906, "learning_rate": 9.998491003013353e-06, "loss": 12.966547966003418, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.28459143452346325, "loss_total_v6": 0.8104092553257942, "step": 261 }, { "epoch": 0.060425225225225226, "grad_norm": 32.2390251159668, "learning_rate": 9.998441124759072e-06, "loss": 12.79544734954834, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.28363271057605743, "loss_total_v6": 0.79971544444561, "step": 262 }, { "epoch": 0.06065585585585585, "grad_norm": 35.75710678100586, "learning_rate": 9.998390435646214e-06, "loss": 12.65664291381836, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.27161575108766556, "loss_total_v6": 0.7910401821136475, "step": 263 }, { "epoch": 0.06088648648648649, "grad_norm": 35.676029205322266, "learning_rate": 9.998338935683002e-06, "loss": 13.567459106445312, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.3286180589348078, "loss_total_v6": 0.8479661643505096, "step": 264 }, { "epoch": 0.06111711711711712, "grad_norm": 33.541847229003906, "learning_rate": 9.998286624877786e-06, "loss": 12.942503929138184, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.2815017141401768, "loss_total_v6": 0.8089064881205559, "step": 265 }, { "epoch": 0.06134774774774775, "grad_norm": 30.669326782226562, "learning_rate": 9.99823350323906e-06, "loss": 12.847415924072266, "loss_alignment": 1.056640625, "loss_alignment_w": 0.5283203125, "loss_sft": 0.2741701975464821, "loss_total_v6": 0.8029635325074196, "step": 266 }, { "epoch": 0.06157837837837838, "grad_norm": 33.34056091308594, "learning_rate": 9.998179570775435e-06, "loss": 12.947603225708008, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.29063995368778706, "loss_total_v6": 0.8092251718044281, "step": 267 }, { "epoch": 0.06180900900900901, "grad_norm": 29.465856552124023, "learning_rate": 9.998124827495664e-06, "loss": 12.933792114257812, "loss_alignment": 1.0556640625, "loss_alignment_w": 0.52783203125, "loss_sft": 0.2804842311888933, "loss_total_v6": 0.8083620443940163, "step": 268 }, { "epoch": 0.06203963963963964, "grad_norm": 37.96697235107422, "learning_rate": 9.998069273408624e-06, "loss": 12.72156047821045, "loss_alignment": 1.04638671875, "loss_alignment_w": 0.523193359375, "loss_sft": 0.2733079958707094, "loss_total_v6": 0.7950975224375725, "step": 269 }, { "epoch": 0.06227027027027027, "grad_norm": 33.36376190185547, "learning_rate": 9.99801290852333e-06, "loss": 12.640301704406738, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.2658184617757797, "loss_total_v6": 0.7900188788771629, "step": 270 }, { "epoch": 0.06250090090090091, "grad_norm": 32.281211853027344, "learning_rate": 9.997955732848925e-06, "loss": 12.998958587646484, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.29940395802259445, "loss_total_v6": 0.8124349564313889, "step": 271 }, { "epoch": 0.06273153153153153, "grad_norm": 28.533870697021484, "learning_rate": 9.997897746394684e-06, "loss": 12.845548629760742, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.27518266811966896, "loss_total_v6": 0.80284683406353, "step": 272 }, { "epoch": 0.06296216216216216, "grad_norm": 28.401552200317383, "learning_rate": 9.997838949170013e-06, "loss": 12.495161056518555, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.2592495586723089, "loss_total_v6": 0.7809475809335709, "step": 273 }, { "epoch": 0.06319279279279279, "grad_norm": 28.619937896728516, "learning_rate": 9.99777934118445e-06, "loss": 12.984665870666504, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.2870665118098259, "loss_total_v6": 0.8115416094660759, "step": 274 }, { "epoch": 0.06342342342342343, "grad_norm": 30.747526168823242, "learning_rate": 9.997718922447669e-06, "loss": 13.176224708557129, "loss_alignment": 1.0693359375, "loss_alignment_w": 0.53466796875, "loss_sft": 0.29063136130571365, "loss_total_v6": 0.8235140517354012, "step": 275 }, { "epoch": 0.06365405405405405, "grad_norm": 30.006898880004883, "learning_rate": 9.997657692969465e-06, "loss": 12.968901634216309, "loss_alignment": 1.068359375, "loss_alignment_w": 0.5341796875, "loss_sft": 0.2772617172449827, "loss_total_v6": 0.8105563744902611, "step": 276 }, { "epoch": 0.06388468468468468, "grad_norm": 31.155139923095703, "learning_rate": 9.997595652759773e-06, "loss": 13.235698699951172, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.3006808217614889, "loss_total_v6": 0.8272311389446259, "step": 277 }, { "epoch": 0.06411531531531532, "grad_norm": 31.25355339050293, "learning_rate": 9.997532801828659e-06, "loss": 12.786547660827637, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.2723037153482437, "loss_total_v6": 0.7991592139005661, "step": 278 }, { "epoch": 0.06434594594594595, "grad_norm": 31.45490264892578, "learning_rate": 9.997469140186318e-06, "loss": 13.086477279663086, "loss_alignment": 1.05859375, "loss_alignment_w": 0.529296875, "loss_sft": 0.2895692642778158, "loss_total_v6": 0.8179048225283623, "step": 279 }, { "epoch": 0.06457657657657657, "grad_norm": 29.69439125061035, "learning_rate": 9.997404667843076e-06, "loss": 13.069145202636719, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.29989956319332123, "loss_total_v6": 0.8168215677142143, "step": 280 }, { "epoch": 0.0648072072072072, "grad_norm": 31.003070831298828, "learning_rate": 9.997339384809392e-06, "loss": 13.2000150680542, "loss_alignment": 1.0712890625, "loss_alignment_w": 0.53564453125, "loss_sft": 0.28808995708823204, "loss_total_v6": 0.825000949203968, "step": 281 }, { "epoch": 0.06503783783783784, "grad_norm": 33.08281707763672, "learning_rate": 9.997273291095858e-06, "loss": 12.837993621826172, "loss_alignment": 1.064453125, "loss_alignment_w": 0.5322265625, "loss_sft": 0.2697513457387686, "loss_total_v6": 0.8023746311664581, "step": 282 }, { "epoch": 0.06526846846846847, "grad_norm": 29.76360321044922, "learning_rate": 9.997206386713198e-06, "loss": 13.255952835083008, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.3056699074804783, "loss_total_v6": 0.8284970447421074, "step": 283 }, { "epoch": 0.0654990990990991, "grad_norm": 29.5471248626709, "learning_rate": 9.99713867167226e-06, "loss": 13.10443115234375, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.2940636109560728, "loss_total_v6": 0.8190269842743874, "step": 284 }, { "epoch": 0.06572972972972974, "grad_norm": 29.718515396118164, "learning_rate": 9.997070145984032e-06, "loss": 13.027894973754883, "loss_alignment": 1.04052734375, "loss_alignment_w": 0.520263671875, "loss_sft": 0.2943459916859865, "loss_total_v6": 0.8142434656620026, "step": 285 }, { "epoch": 0.06596036036036036, "grad_norm": 29.28940200805664, "learning_rate": 9.997000809659629e-06, "loss": 13.313615798950195, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.3084040582180023, "loss_total_v6": 0.8321009501814842, "step": 286 }, { "epoch": 0.06619099099099099, "grad_norm": 29.550884246826172, "learning_rate": 9.9969306627103e-06, "loss": 12.482795715332031, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.25620315596461296, "loss_total_v6": 0.7801747098565102, "step": 287 }, { "epoch": 0.06642162162162162, "grad_norm": 31.440706253051758, "learning_rate": 9.996859705147424e-06, "loss": 12.657780647277832, "loss_alignment": 1.046875, "loss_alignment_w": 0.5234375, "loss_sft": 0.26802472583949566, "loss_total_v6": 0.7911112755537033, "step": 288 }, { "epoch": 0.06665225225225226, "grad_norm": 30.88951301574707, "learning_rate": 9.996787936982513e-06, "loss": 12.623908996582031, "loss_alignment": 1.02587890625, "loss_alignment_w": 0.512939453125, "loss_sft": 0.27668051049113274, "loss_total_v6": 0.7889943569898605, "step": 289 }, { "epoch": 0.06688288288288288, "grad_norm": 28.577407836914062, "learning_rate": 9.996715358227208e-06, "loss": 13.220832824707031, "loss_alignment": 1.03564453125, "loss_alignment_w": 0.517822265625, "loss_sft": 0.30800678580999374, "loss_total_v6": 0.826302096247673, "step": 290 }, { "epoch": 0.06711351351351351, "grad_norm": 29.010494232177734, "learning_rate": 9.996641968893281e-06, "loss": 12.987323760986328, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.28842283226549625, "loss_total_v6": 0.8117077425122261, "step": 291 }, { "epoch": 0.06734414414414415, "grad_norm": 29.478254318237305, "learning_rate": 9.996567768992642e-06, "loss": 13.374385833740234, "loss_alignment": 1.05126953125, "loss_alignment_w": 0.525634765625, "loss_sft": 0.3095624968409538, "loss_total_v6": 0.8358991593122482, "step": 292 }, { "epoch": 0.06757477477477478, "grad_norm": 29.51673126220703, "learning_rate": 9.996492758537326e-06, "loss": 12.988321304321289, "loss_alignment": 1.03857421875, "loss_alignment_w": 0.519287109375, "loss_sft": 0.29126227274537086, "loss_total_v6": 0.8117700964212418, "step": 293 }, { "epoch": 0.0678054054054054, "grad_norm": 30.066686630249023, "learning_rate": 9.996416937539499e-06, "loss": 12.80174732208252, "loss_alignment": 1.0556640625, "loss_alignment_w": 0.52783203125, "loss_sft": 0.2723992299288511, "loss_total_v6": 0.8001092076301575, "step": 294 }, { "epoch": 0.06803603603603603, "grad_norm": 31.210081100463867, "learning_rate": 9.996340306011464e-06, "loss": 13.131805419921875, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.30509762465953827, "loss_total_v6": 0.820737861096859, "step": 295 }, { "epoch": 0.06826666666666667, "grad_norm": 29.669828414916992, "learning_rate": 9.996262863965651e-06, "loss": 12.623188018798828, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.274514427408576, "loss_total_v6": 0.788949228823185, "step": 296 }, { "epoch": 0.0684972972972973, "grad_norm": 28.497966766357422, "learning_rate": 9.996184611414623e-06, "loss": 13.174598693847656, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.2988914791494608, "loss_total_v6": 0.8234123587608337, "step": 297 }, { "epoch": 0.06872792792792792, "grad_norm": 31.63694190979004, "learning_rate": 9.996105548371074e-06, "loss": 13.169651985168457, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.30282433703541756, "loss_total_v6": 0.8231032565236092, "step": 298 }, { "epoch": 0.06895855855855856, "grad_norm": 32.97370147705078, "learning_rate": 9.996025674847831e-06, "loss": 12.86874771118164, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.28322432190179825, "loss_total_v6": 0.8042967095971107, "step": 299 }, { "epoch": 0.06918918918918919, "grad_norm": 31.052093505859375, "learning_rate": 9.995944990857848e-06, "loss": 13.554883003234863, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.3191651217639446, "loss_total_v6": 0.8471802249550819, "step": 300 }, { "epoch": 0.06941981981981982, "grad_norm": 33.999900817871094, "learning_rate": 9.995863496414219e-06, "loss": 12.654172897338867, "loss_alignment": 1.04248046875, "loss_alignment_w": 0.521240234375, "loss_sft": 0.2695387154817581, "loss_total_v6": 0.790885753929615, "step": 301 }, { "epoch": 0.06965045045045046, "grad_norm": 30.67936134338379, "learning_rate": 9.995781191530158e-06, "loss": 12.829086303710938, "loss_alignment": 1.0595703125, "loss_alignment_w": 0.52978515625, "loss_sft": 0.2727041095495224, "loss_total_v6": 0.8018178790807724, "step": 302 }, { "epoch": 0.06988108108108108, "grad_norm": 29.476186752319336, "learning_rate": 9.995698076219022e-06, "loss": 12.89139461517334, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.2817710842937231, "loss_total_v6": 0.805712141096592, "step": 303 }, { "epoch": 0.07011171171171171, "grad_norm": 31.571048736572266, "learning_rate": 9.995614150494293e-06, "loss": 12.794020652770996, "loss_alignment": 1.03466796875, "loss_alignment_w": 0.517333984375, "loss_sft": 0.2820329051464796, "loss_total_v6": 0.7996262982487679, "step": 304 }, { "epoch": 0.07034234234234234, "grad_norm": 29.393272399902344, "learning_rate": 9.995529414369582e-06, "loss": 13.168919563293457, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.3006270471960306, "loss_total_v6": 0.8230574429035187, "step": 305 }, { "epoch": 0.07057297297297298, "grad_norm": 31.221839904785156, "learning_rate": 9.995443867858642e-06, "loss": 12.920101165771484, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.2832600697875023, "loss_total_v6": 0.8075062856078148, "step": 306 }, { "epoch": 0.0708036036036036, "grad_norm": 30.26776123046875, "learning_rate": 9.995357510975343e-06, "loss": 12.582901000976562, "loss_alignment": 1.0556640625, "loss_alignment_w": 0.52783203125, "loss_sft": 0.25954531133174896, "loss_total_v6": 0.7864313051104546, "step": 307 }, { "epoch": 0.07103423423423423, "grad_norm": 29.96918296813965, "learning_rate": 9.995270343733697e-06, "loss": 12.972612380981445, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.29180632904171944, "loss_total_v6": 0.8107882589101791, "step": 308 }, { "epoch": 0.07126486486486487, "grad_norm": 32.443477630615234, "learning_rate": 9.995182366147846e-06, "loss": 13.297029495239258, "loss_alignment": 1.0673828125, "loss_alignment_w": 0.53369140625, "loss_sft": 0.29676254838705063, "loss_total_v6": 0.83106429874897, "step": 309 }, { "epoch": 0.0714954954954955, "grad_norm": 30.288894653320312, "learning_rate": 9.99509357823206e-06, "loss": 13.249547958374023, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.30475080758333206, "loss_total_v6": 0.8280967399477959, "step": 310 }, { "epoch": 0.07172612612612612, "grad_norm": 28.24803352355957, "learning_rate": 9.995003980000743e-06, "loss": 13.566115379333496, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.3254823088645935, "loss_total_v6": 0.8478822335600853, "step": 311 }, { "epoch": 0.07195675675675675, "grad_norm": 27.65163230895996, "learning_rate": 9.994913571468432e-06, "loss": 12.706634521484375, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2821712866425514, "loss_total_v6": 0.7941646873950958, "step": 312 }, { "epoch": 0.07218738738738739, "grad_norm": 34.42991638183594, "learning_rate": 9.99482235264979e-06, "loss": 12.712350845336914, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2789885252714157, "loss_total_v6": 0.7945219427347183, "step": 313 }, { "epoch": 0.07241801801801802, "grad_norm": 32.06767272949219, "learning_rate": 9.994730323559616e-06, "loss": 13.412830352783203, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.31631404161453247, "loss_total_v6": 0.8383019268512726, "step": 314 }, { "epoch": 0.07264864864864864, "grad_norm": 42.37416076660156, "learning_rate": 9.994637484212839e-06, "loss": 12.993596076965332, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.29027969762682915, "loss_total_v6": 0.8120997548103333, "step": 315 }, { "epoch": 0.07287927927927929, "grad_norm": 29.82160758972168, "learning_rate": 9.99454383462452e-06, "loss": 12.971562385559082, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.30706051364541054, "loss_total_v6": 0.8107226341962814, "step": 316 }, { "epoch": 0.07310990990990991, "grad_norm": 34.589229583740234, "learning_rate": 9.994449374809851e-06, "loss": 13.400375366210938, "loss_alignment": 1.0673828125, "loss_alignment_w": 0.53369140625, "loss_sft": 0.3043050840497017, "loss_total_v6": 0.8375234827399254, "step": 317 }, { "epoch": 0.07334054054054054, "grad_norm": 29.17933464050293, "learning_rate": 9.994354104784153e-06, "loss": 12.909574508666992, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.2898196093738079, "loss_total_v6": 0.806848406791687, "step": 318 }, { "epoch": 0.07357117117117117, "grad_norm": 29.207517623901367, "learning_rate": 9.994258024562887e-06, "loss": 12.537301063537598, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.26012852415442467, "loss_total_v6": 0.7835812941193581, "step": 319 }, { "epoch": 0.0738018018018018, "grad_norm": 27.765485763549805, "learning_rate": 9.994161134161635e-06, "loss": 13.236174583435059, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.3034266550093889, "loss_total_v6": 0.8272608891129494, "step": 320 }, { "epoch": 0.07403243243243243, "grad_norm": 28.6611385345459, "learning_rate": 9.994063433596113e-06, "loss": 13.113682746887207, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.30158460326492786, "loss_total_v6": 0.819605216383934, "step": 321 }, { "epoch": 0.07426306306306306, "grad_norm": 27.703889846801758, "learning_rate": 9.993964922882174e-06, "loss": 12.75886344909668, "loss_alignment": 1.03369140625, "loss_alignment_w": 0.516845703125, "loss_sft": 0.2813614420592785, "loss_total_v6": 0.7974289581179619, "step": 322 }, { "epoch": 0.0744936936936937, "grad_norm": 27.934179306030273, "learning_rate": 9.993865602035798e-06, "loss": 12.652129173278809, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.27351558953523636, "loss_total_v6": 0.7907580435276031, "step": 323 }, { "epoch": 0.07472432432432433, "grad_norm": 30.142911911010742, "learning_rate": 9.993765471073094e-06, "loss": 12.463241577148438, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.26306814700365067, "loss_total_v6": 0.7789525613188744, "step": 324 }, { "epoch": 0.07495495495495495, "grad_norm": 28.13068962097168, "learning_rate": 9.993664530010308e-06, "loss": 13.164217948913574, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.29819696955382824, "loss_total_v6": 0.8227636069059372, "step": 325 }, { "epoch": 0.07518558558558558, "grad_norm": 29.20014762878418, "learning_rate": 9.993562778863817e-06, "loss": 12.761024475097656, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.27986385114490986, "loss_total_v6": 0.7975640371441841, "step": 326 }, { "epoch": 0.07541621621621622, "grad_norm": 27.64097023010254, "learning_rate": 9.993460217650122e-06, "loss": 12.339319229125977, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.24998248927295208, "loss_total_v6": 0.7712074443697929, "step": 327 }, { "epoch": 0.07564684684684685, "grad_norm": 27.4700927734375, "learning_rate": 9.993356846385866e-06, "loss": 13.183608055114746, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.3000344783067703, "loss_total_v6": 0.8239755108952522, "step": 328 }, { "epoch": 0.07587747747747747, "grad_norm": 31.224573135375977, "learning_rate": 9.993252665087816e-06, "loss": 12.595613479614258, "loss_alignment": 1.05859375, "loss_alignment_w": 0.529296875, "loss_sft": 0.2569829560816288, "loss_total_v6": 0.7872258797287941, "step": 329 }, { "epoch": 0.07610810810810811, "grad_norm": 27.515390396118164, "learning_rate": 9.993147673772869e-06, "loss": 12.385574340820312, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2596941087394953, "loss_total_v6": 0.7740983888506889, "step": 330 }, { "epoch": 0.07633873873873874, "grad_norm": 28.548349380493164, "learning_rate": 9.993041872458064e-06, "loss": 12.910856246948242, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.28129372373223305, "loss_total_v6": 0.8069284930825233, "step": 331 }, { "epoch": 0.07656936936936937, "grad_norm": 28.016904830932617, "learning_rate": 9.992935261160559e-06, "loss": 12.900680541992188, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.2846860233694315, "loss_total_v6": 0.8062924891710281, "step": 332 }, { "epoch": 0.0768, "grad_norm": 29.0798397064209, "learning_rate": 9.99282783989765e-06, "loss": 12.314382553100586, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2616990450769663, "loss_total_v6": 0.7696488723158836, "step": 333 }, { "epoch": 0.07703063063063063, "grad_norm": 27.265390396118164, "learning_rate": 9.992719608686763e-06, "loss": 12.70334529876709, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.2779068108648062, "loss_total_v6": 0.7939590737223625, "step": 334 }, { "epoch": 0.07726126126126126, "grad_norm": 31.472450256347656, "learning_rate": 9.992610567545458e-06, "loss": 12.786520004272461, "loss_alignment": 1.03857421875, "loss_alignment_w": 0.519287109375, "loss_sft": 0.27994662895798683, "loss_total_v6": 0.7991574555635452, "step": 335 }, { "epoch": 0.07749189189189189, "grad_norm": 29.952190399169922, "learning_rate": 9.99250071649142e-06, "loss": 12.475122451782227, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.26184235513210297, "loss_total_v6": 0.7796951308846474, "step": 336 }, { "epoch": 0.07772252252252253, "grad_norm": 32.3388557434082, "learning_rate": 9.992390055542472e-06, "loss": 12.975914001464844, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.29846717417240143, "loss_total_v6": 0.8109946027398109, "step": 337 }, { "epoch": 0.07795315315315315, "grad_norm": 28.577556610107422, "learning_rate": 9.992278584716563e-06, "loss": 12.644527435302734, "loss_alignment": 1.02978515625, "loss_alignment_w": 0.514892578125, "loss_sft": 0.27511568553745747, "loss_total_v6": 0.7902829498052597, "step": 338 }, { "epoch": 0.07818378378378378, "grad_norm": 29.34807014465332, "learning_rate": 9.992166304031781e-06, "loss": 13.147138595581055, "loss_alignment": 1.0634765625, "loss_alignment_w": 0.53173828125, "loss_sft": 0.2908123917877674, "loss_total_v6": 0.8216961696743965, "step": 339 }, { "epoch": 0.07841441441441442, "grad_norm": 31.121606826782227, "learning_rate": 9.992053213506333e-06, "loss": 12.73483657836914, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.2740919981151819, "loss_total_v6": 0.7959273234009743, "step": 340 }, { "epoch": 0.07864504504504505, "grad_norm": 28.721385955810547, "learning_rate": 9.991939313158574e-06, "loss": 12.975173950195312, "loss_alignment": 1.0703125, "loss_alignment_w": 0.53515625, "loss_sft": 0.27832501381635666, "loss_total_v6": 0.8109483495354652, "step": 341 }, { "epoch": 0.07887567567567567, "grad_norm": 25.030513763427734, "learning_rate": 9.991824603006972e-06, "loss": 12.718854904174805, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.27651106007397175, "loss_total_v6": 0.7949284315109253, "step": 342 }, { "epoch": 0.0791063063063063, "grad_norm": 27.528776168823242, "learning_rate": 9.991709083070143e-06, "loss": 13.042431831359863, "loss_alignment": 1.06640625, "loss_alignment_w": 0.533203125, "loss_sft": 0.28168945014476776, "loss_total_v6": 0.8151519596576691, "step": 343 }, { "epoch": 0.07933693693693694, "grad_norm": 28.45009994506836, "learning_rate": 9.991592753366822e-06, "loss": 13.15477180480957, "loss_alignment": 1.0625, "loss_alignment_w": 0.53125, "loss_sft": 0.29081635922193527, "loss_total_v6": 0.8221732154488564, "step": 344 }, { "epoch": 0.07956756756756757, "grad_norm": 30.1481876373291, "learning_rate": 9.991475613915883e-06, "loss": 13.216221809387207, "loss_alignment": 1.046875, "loss_alignment_w": 0.5234375, "loss_sft": 0.30372075363993645, "loss_total_v6": 0.8260138630867004, "step": 345 }, { "epoch": 0.0797981981981982, "grad_norm": 37.50176239013672, "learning_rate": 9.991357664736327e-06, "loss": 12.90246868133545, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2885973509401083, "loss_total_v6": 0.806404322385788, "step": 346 }, { "epoch": 0.08002882882882884, "grad_norm": 30.19721794128418, "learning_rate": 9.991238905847289e-06, "loss": 12.861249923706055, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2882794216275215, "loss_total_v6": 0.8038280978798866, "step": 347 }, { "epoch": 0.08025945945945946, "grad_norm": 34.825889587402344, "learning_rate": 9.991119337268032e-06, "loss": 12.805431365966797, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.28265452571213245, "loss_total_v6": 0.8003394603729248, "step": 348 }, { "epoch": 0.08049009009009009, "grad_norm": 30.955007553100586, "learning_rate": 9.990998959017956e-06, "loss": 13.084518432617188, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.29805275797843933, "loss_total_v6": 0.8177824020385742, "step": 349 }, { "epoch": 0.08072072072072072, "grad_norm": 29.60041618347168, "learning_rate": 9.990877771116588e-06, "loss": 12.9472017288208, "loss_alignment": 1.0576171875, "loss_alignment_w": 0.52880859375, "loss_sft": 0.28075770288705826, "loss_total_v6": 0.8092001006007195, "step": 350 }, { "epoch": 0.08095135135135136, "grad_norm": 25.67243194580078, "learning_rate": 9.990755773583588e-06, "loss": 13.21328353881836, "loss_alignment": 1.0576171875, "loss_alignment_w": 0.52880859375, "loss_sft": 0.29708268865942955, "loss_total_v6": 0.8258302509784698, "step": 351 }, { "epoch": 0.08118198198198198, "grad_norm": 26.176042556762695, "learning_rate": 9.990632966438743e-06, "loss": 12.598169326782227, "loss_alignment": 1.03662109375, "loss_alignment_w": 0.518310546875, "loss_sft": 0.26967015862464905, "loss_total_v6": 0.7873856201767921, "step": 352 }, { "epoch": 0.08141261261261261, "grad_norm": 26.82232666015625, "learning_rate": 9.99050934970198e-06, "loss": 13.049955368041992, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.29462615586817265, "loss_total_v6": 0.8156222552061081, "step": 353 }, { "epoch": 0.08164324324324325, "grad_norm": 26.51399803161621, "learning_rate": 9.99038492339335e-06, "loss": 12.53531551361084, "loss_alignment": 1.03662109375, "loss_alignment_w": 0.518310546875, "loss_sft": 0.2651771828532219, "loss_total_v6": 0.7834572419524193, "step": 354 }, { "epoch": 0.08187387387387388, "grad_norm": 28.761171340942383, "learning_rate": 9.990259687533038e-06, "loss": 12.721296310424805, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2755803056061268, "loss_total_v6": 0.7950810268521309, "step": 355 }, { "epoch": 0.0821045045045045, "grad_norm": 31.784832000732422, "learning_rate": 9.990133642141359e-06, "loss": 13.338712692260742, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.3130396828055382, "loss_total_v6": 0.83366958796978, "step": 356 }, { "epoch": 0.08233513513513513, "grad_norm": 29.009435653686523, "learning_rate": 9.990006787238761e-06, "loss": 12.596626281738281, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.26928376965224743, "loss_total_v6": 0.7872891277074814, "step": 357 }, { "epoch": 0.08256576576576577, "grad_norm": 28.785991668701172, "learning_rate": 9.989879122845825e-06, "loss": 12.81518268585205, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2861631438136101, "loss_total_v6": 0.8009489178657532, "step": 358 }, { "epoch": 0.0827963963963964, "grad_norm": 29.82000160217285, "learning_rate": 9.989750648983258e-06, "loss": 12.472010612487793, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.266057712957263, "loss_total_v6": 0.7795007079839706, "step": 359 }, { "epoch": 0.08302702702702702, "grad_norm": 25.497276306152344, "learning_rate": 9.989621365671902e-06, "loss": 12.083662033081055, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.2542980797588825, "loss_total_v6": 0.7552288770675659, "step": 360 }, { "epoch": 0.08325765765765766, "grad_norm": 28.308208465576172, "learning_rate": 9.98949127293273e-06, "loss": 12.931331634521484, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2912099249660969, "loss_total_v6": 0.8082082271575928, "step": 361 }, { "epoch": 0.08348828828828829, "grad_norm": 27.454261779785156, "learning_rate": 9.989360370786847e-06, "loss": 12.860968589782715, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.27834358252584934, "loss_total_v6": 0.8038105368614197, "step": 362 }, { "epoch": 0.08371891891891892, "grad_norm": 27.191801071166992, "learning_rate": 9.989228659255486e-06, "loss": 13.393671989440918, "loss_alignment": 1.03857421875, "loss_alignment_w": 0.519287109375, "loss_sft": 0.31847357377409935, "loss_total_v6": 0.8371045365929604, "step": 363 }, { "epoch": 0.08394954954954954, "grad_norm": 26.7711181640625, "learning_rate": 9.989096138360014e-06, "loss": 13.419794082641602, "loss_alignment": 1.060546875, "loss_alignment_w": 0.5302734375, "loss_sft": 0.3090740516781807, "loss_total_v6": 0.8387371525168419, "step": 364 }, { "epoch": 0.08418018018018018, "grad_norm": 27.755416870117188, "learning_rate": 9.988962808121932e-06, "loss": 13.042716979980469, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.2907710261642933, "loss_total_v6": 0.8151698186993599, "step": 365 }, { "epoch": 0.08441081081081081, "grad_norm": 29.872905731201172, "learning_rate": 9.988828668562866e-06, "loss": 12.702960968017578, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.2747700475156307, "loss_total_v6": 0.7939351052045822, "step": 366 }, { "epoch": 0.08464144144144144, "grad_norm": 25.886192321777344, "learning_rate": 9.988693719704577e-06, "loss": 13.156342506408691, "loss_alignment": 1.0556640625, "loss_alignment_w": 0.52783203125, "loss_sft": 0.29588895104825497, "loss_total_v6": 0.8222714141011238, "step": 367 }, { "epoch": 0.08487207207207208, "grad_norm": 26.113426208496094, "learning_rate": 9.988557961568956e-06, "loss": 12.477398872375488, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.26126749999821186, "loss_total_v6": 0.7798374146223068, "step": 368 }, { "epoch": 0.0851027027027027, "grad_norm": 29.722061157226562, "learning_rate": 9.988421394178027e-06, "loss": 12.8818998336792, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.2853281106799841, "loss_total_v6": 0.8051187545061111, "step": 369 }, { "epoch": 0.08533333333333333, "grad_norm": 32.504249572753906, "learning_rate": 9.988284017553945e-06, "loss": 13.437504768371582, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.31741361878812313, "loss_total_v6": 0.8398440331220627, "step": 370 }, { "epoch": 0.08556396396396396, "grad_norm": 27.446762084960938, "learning_rate": 9.988145831718995e-06, "loss": 12.839937210083008, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.28059976547956467, "loss_total_v6": 0.8024960905313492, "step": 371 }, { "epoch": 0.0857945945945946, "grad_norm": 26.19923973083496, "learning_rate": 9.988006836695593e-06, "loss": 13.093706130981445, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.2933322601020336, "loss_total_v6": 0.8183566778898239, "step": 372 }, { "epoch": 0.08602522522522522, "grad_norm": 26.723669052124023, "learning_rate": 9.987867032506286e-06, "loss": 12.34341812133789, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2630254738032818, "loss_total_v6": 0.7714636027812958, "step": 373 }, { "epoch": 0.08625585585585585, "grad_norm": 29.665611267089844, "learning_rate": 9.987726419173756e-06, "loss": 12.88904857635498, "loss_alignment": 1.0546875, "loss_alignment_w": 0.52734375, "loss_sft": 0.2788168843835592, "loss_total_v6": 0.8055655658245087, "step": 374 }, { "epoch": 0.08648648648648649, "grad_norm": 29.406845092773438, "learning_rate": 9.987584996720813e-06, "loss": 13.54936695098877, "loss_alignment": 1.0546875, "loss_alignment_w": 0.52734375, "loss_sft": 0.3192780390381813, "loss_total_v6": 0.8468354120850563, "step": 375 }, { "epoch": 0.08671711711711712, "grad_norm": 27.86764144897461, "learning_rate": 9.987442765170398e-06, "loss": 13.16490364074707, "loss_alignment": 1.04833984375, "loss_alignment_w": 0.524169921875, "loss_sft": 0.2972174845635891, "loss_total_v6": 0.8228064700961113, "step": 376 }, { "epoch": 0.08694774774774774, "grad_norm": 28.90019416809082, "learning_rate": 9.987299724545585e-06, "loss": 13.393845558166504, "loss_alignment": 1.05859375, "loss_alignment_w": 0.529296875, "loss_sft": 0.30742171593010426, "loss_total_v6": 0.8371153250336647, "step": 377 }, { "epoch": 0.08717837837837838, "grad_norm": 28.72464942932129, "learning_rate": 9.987155874869579e-06, "loss": 12.7694673538208, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2821767833083868, "loss_total_v6": 0.7980917170643806, "step": 378 }, { "epoch": 0.08740900900900901, "grad_norm": 27.502193450927734, "learning_rate": 9.987011216165715e-06, "loss": 13.110069274902344, "loss_alignment": 1.0546875, "loss_alignment_w": 0.52734375, "loss_sft": 0.2924475520849228, "loss_total_v6": 0.8193793222308159, "step": 379 }, { "epoch": 0.08763963963963964, "grad_norm": 30.297622680664062, "learning_rate": 9.986865748457457e-06, "loss": 12.558738708496094, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.2653288748115301, "loss_total_v6": 0.7849211543798447, "step": 380 }, { "epoch": 0.08787027027027026, "grad_norm": 28.447235107421875, "learning_rate": 9.986719471768406e-06, "loss": 13.339821815490723, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.31931934505701065, "loss_total_v6": 0.8337388634681702, "step": 381 }, { "epoch": 0.0881009009009009, "grad_norm": 28.1668701171875, "learning_rate": 9.98657238612229e-06, "loss": 12.946065902709961, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.29174933955073357, "loss_total_v6": 0.8091290965676308, "step": 382 }, { "epoch": 0.08833153153153153, "grad_norm": 28.125003814697266, "learning_rate": 9.986424491542972e-06, "loss": 12.918296813964844, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2920431774109602, "loss_total_v6": 0.8073935285210609, "step": 383 }, { "epoch": 0.08856216216216216, "grad_norm": 29.268537521362305, "learning_rate": 9.986275788054442e-06, "loss": 12.715314865112305, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.27970778197050095, "loss_total_v6": 0.794707179069519, "step": 384 }, { "epoch": 0.0887927927927928, "grad_norm": 30.463834762573242, "learning_rate": 9.986126275680825e-06, "loss": 13.06840991973877, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.28888252563774586, "loss_total_v6": 0.8167756125330925, "step": 385 }, { "epoch": 0.08902342342342343, "grad_norm": 26.114553451538086, "learning_rate": 9.985975954446372e-06, "loss": 12.636455535888672, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.26960633136332035, "loss_total_v6": 0.7897784262895584, "step": 386 }, { "epoch": 0.08925405405405405, "grad_norm": 30.226308822631836, "learning_rate": 9.985824824375469e-06, "loss": 13.207096099853516, "loss_alignment": 1.056640625, "loss_alignment_w": 0.5283203125, "loss_sft": 0.296070359647274, "loss_total_v6": 0.8254435360431671, "step": 387 }, { "epoch": 0.08948468468468468, "grad_norm": 29.454130172729492, "learning_rate": 9.985672885492634e-06, "loss": 13.429874420166016, "loss_alignment": 1.0576171875, "loss_alignment_w": 0.52880859375, "loss_sft": 0.3096124492585659, "loss_total_v6": 0.8393671289086342, "step": 388 }, { "epoch": 0.08971531531531532, "grad_norm": 25.98806381225586, "learning_rate": 9.985520137822516e-06, "loss": 12.869606018066406, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.2834610641002655, "loss_total_v6": 0.8043503612279892, "step": 389 }, { "epoch": 0.08994594594594595, "grad_norm": 28.114887237548828, "learning_rate": 9.985366581389893e-06, "loss": 13.478965759277344, "loss_alignment": 1.02978515625, "loss_alignment_w": 0.514892578125, "loss_sft": 0.3262762799859047, "loss_total_v6": 0.8424353376030922, "step": 390 }, { "epoch": 0.09017657657657657, "grad_norm": 30.131105422973633, "learning_rate": 9.985212216219673e-06, "loss": 13.001485824584961, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.29489265009760857, "loss_total_v6": 0.8125928491353989, "step": 391 }, { "epoch": 0.09040720720720721, "grad_norm": 29.367738723754883, "learning_rate": 9.985057042336898e-06, "loss": 12.896995544433594, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.2836470678448677, "loss_total_v6": 0.806062251329422, "step": 392 }, { "epoch": 0.09063783783783784, "grad_norm": 27.027847290039062, "learning_rate": 9.984901059766746e-06, "loss": 13.608288764953613, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.33242112398147583, "loss_total_v6": 0.8505180552601814, "step": 393 }, { "epoch": 0.09086846846846847, "grad_norm": 28.213699340820312, "learning_rate": 9.984744268534515e-06, "loss": 12.708200454711914, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.27874436415731907, "loss_total_v6": 0.7942625433206558, "step": 394 }, { "epoch": 0.09109909909909909, "grad_norm": 25.949670791625977, "learning_rate": 9.984586668665641e-06, "loss": 12.924871444702148, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.2826427835971117, "loss_total_v6": 0.8078045099973679, "step": 395 }, { "epoch": 0.09132972972972973, "grad_norm": 30.192489624023438, "learning_rate": 9.984428260185691e-06, "loss": 12.76202392578125, "loss_alignment": 1.060546875, "loss_alignment_w": 0.5302734375, "loss_sft": 0.26787188090384007, "loss_total_v6": 0.7976265475153923, "step": 396 }, { "epoch": 0.09156036036036036, "grad_norm": 28.533287048339844, "learning_rate": 9.984269043120364e-06, "loss": 12.694650650024414, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.275379816070199, "loss_total_v6": 0.7934157103300095, "step": 397 }, { "epoch": 0.09179099099099099, "grad_norm": 27.48672866821289, "learning_rate": 9.984109017495485e-06, "loss": 13.203250885009766, "loss_alignment": 1.0625, "loss_alignment_w": 0.53125, "loss_sft": 0.29355642572045326, "loss_total_v6": 0.825203150510788, "step": 398 }, { "epoch": 0.09202162162162163, "grad_norm": 26.785226821899414, "learning_rate": 9.983948183337018e-06, "loss": 12.909088134765625, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.289560304954648, "loss_total_v6": 0.8068179935216904, "step": 399 }, { "epoch": 0.09225225225225225, "grad_norm": 29.571794509887695, "learning_rate": 9.983786540671052e-06, "loss": 12.88036823272705, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.2780606672167778, "loss_total_v6": 0.8050229847431183, "step": 400 }, { "epoch": 0.09248288288288288, "grad_norm": 26.007217407226562, "learning_rate": 9.983624089523806e-06, "loss": 12.797578811645508, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.27419867366552353, "loss_total_v6": 0.7998486906290054, "step": 401 }, { "epoch": 0.0927135135135135, "grad_norm": 28.924114227294922, "learning_rate": 9.983460829921637e-06, "loss": 13.078068733215332, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.29111894220113754, "loss_total_v6": 0.8173792958259583, "step": 402 }, { "epoch": 0.09294414414414415, "grad_norm": 28.096437454223633, "learning_rate": 9.983296761891029e-06, "loss": 12.885261535644531, "loss_alignment": 1.046875, "loss_alignment_w": 0.5234375, "loss_sft": 0.280655387789011, "loss_total_v6": 0.8053288385272026, "step": 403 }, { "epoch": 0.09317477477477477, "grad_norm": 28.844141006469727, "learning_rate": 9.983131885458596e-06, "loss": 12.532156944274902, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2706865705549717, "loss_total_v6": 0.7832597941160202, "step": 404 }, { "epoch": 0.0934054054054054, "grad_norm": 27.80720329284668, "learning_rate": 9.982966200651085e-06, "loss": 12.267884254455566, "loss_alignment": 1.03271484375, "loss_alignment_w": 0.516357421875, "loss_sft": 0.2516060471534729, "loss_total_v6": 0.7667427882552147, "step": 405 }, { "epoch": 0.09363603603603604, "grad_norm": 27.67591667175293, "learning_rate": 9.982799707495376e-06, "loss": 13.258272171020508, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.3060437124222517, "loss_total_v6": 0.82864198833704, "step": 406 }, { "epoch": 0.09386666666666667, "grad_norm": 27.92984390258789, "learning_rate": 9.982632406018475e-06, "loss": 12.286430358886719, "loss_alignment": 1.03076171875, "loss_alignment_w": 0.515380859375, "loss_sft": 0.2515750005841255, "loss_total_v6": 0.7679018974304199, "step": 407 }, { "epoch": 0.0940972972972973, "grad_norm": 29.370094299316406, "learning_rate": 9.982464296247523e-06, "loss": 12.741983413696289, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2803064361214638, "loss_total_v6": 0.7963739261031151, "step": 408 }, { "epoch": 0.09432792792792793, "grad_norm": 30.960575103759766, "learning_rate": 9.982295378209792e-06, "loss": 12.190655708312988, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2554462533444166, "loss_total_v6": 0.7619160041213036, "step": 409 }, { "epoch": 0.09455855855855856, "grad_norm": 26.547332763671875, "learning_rate": 9.982125651932681e-06, "loss": 12.684921264648438, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.27371885254979134, "loss_total_v6": 0.7928075790405273, "step": 410 }, { "epoch": 0.09478918918918919, "grad_norm": 28.273902893066406, "learning_rate": 9.981955117443728e-06, "loss": 13.353116989135742, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.3091638833284378, "loss_total_v6": 0.8345697745680809, "step": 411 }, { "epoch": 0.09501981981981981, "grad_norm": 25.755813598632812, "learning_rate": 9.981783774770595e-06, "loss": 12.770072937011719, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.28873020224273205, "loss_total_v6": 0.7981295958161354, "step": 412 }, { "epoch": 0.09525045045045046, "grad_norm": 25.24639129638672, "learning_rate": 9.981611623941077e-06, "loss": 12.957014083862305, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2932270560413599, "loss_total_v6": 0.8098133504390717, "step": 413 }, { "epoch": 0.09548108108108108, "grad_norm": 25.433591842651367, "learning_rate": 9.981438664983103e-06, "loss": 12.861827850341797, "loss_alignment": 1.03369140625, "loss_alignment_w": 0.516845703125, "loss_sft": 0.2877966668456793, "loss_total_v6": 0.8038642108440399, "step": 414 }, { "epoch": 0.09571171171171171, "grad_norm": 28.36766815185547, "learning_rate": 9.98126489792473e-06, "loss": 12.579294204711914, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2744413707405329, "loss_total_v6": 0.7862058728933334, "step": 415 }, { "epoch": 0.09594234234234235, "grad_norm": 29.09765625, "learning_rate": 9.981090322794145e-06, "loss": 13.650089263916016, "loss_alignment": 1.0576171875, "loss_alignment_w": 0.52880859375, "loss_sft": 0.3253137618303299, "loss_total_v6": 0.8531305491924286, "step": 416 }, { "epoch": 0.09617297297297298, "grad_norm": 27.25226402282715, "learning_rate": 9.980914939619669e-06, "loss": 12.741897583007812, "loss_alignment": 1.03271484375, "loss_alignment_w": 0.516357421875, "loss_sft": 0.2804994974285364, "loss_total_v6": 0.7963686212897301, "step": 417 }, { "epoch": 0.0964036036036036, "grad_norm": 27.119415283203125, "learning_rate": 9.980738748429755e-06, "loss": 13.45617389678955, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.3170545771718025, "loss_total_v6": 0.8410108685493469, "step": 418 }, { "epoch": 0.09663423423423423, "grad_norm": 26.281036376953125, "learning_rate": 9.980561749252982e-06, "loss": 13.44235610961914, "loss_alignment": 1.046875, "loss_alignment_w": 0.5234375, "loss_sft": 0.31629784032702446, "loss_total_v6": 0.8401472941040993, "step": 419 }, { "epoch": 0.09686486486486487, "grad_norm": 28.870403289794922, "learning_rate": 9.980383942118066e-06, "loss": 13.379974365234375, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.3094386551529169, "loss_total_v6": 0.8362483531236649, "step": 420 }, { "epoch": 0.0970954954954955, "grad_norm": 28.063331604003906, "learning_rate": 9.98020532705385e-06, "loss": 12.894989013671875, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.2830180823802948, "loss_total_v6": 0.8059367835521698, "step": 421 }, { "epoch": 0.09732612612612612, "grad_norm": 31.93172836303711, "learning_rate": 9.980025904089308e-06, "loss": 12.614752769470215, "loss_alignment": 1.01806640625, "loss_alignment_w": 0.509033203125, "loss_sft": 0.2789615523070097, "loss_total_v6": 0.7884220257401466, "step": 422 }, { "epoch": 0.09755675675675676, "grad_norm": 25.579242706298828, "learning_rate": 9.979845673253547e-06, "loss": 12.729191780090332, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.27755384892225266, "loss_total_v6": 0.7955744788050652, "step": 423 }, { "epoch": 0.09778738738738739, "grad_norm": 26.044300079345703, "learning_rate": 9.979664634575807e-06, "loss": 12.52851676940918, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.26942145079374313, "loss_total_v6": 0.7830322906374931, "step": 424 }, { "epoch": 0.09801801801801802, "grad_norm": 26.583250045776367, "learning_rate": 9.979482788085455e-06, "loss": 13.195645332336426, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.3099878467619419, "loss_total_v6": 0.824727825820446, "step": 425 }, { "epoch": 0.09824864864864864, "grad_norm": 28.576765060424805, "learning_rate": 9.979300133811991e-06, "loss": 13.471195220947266, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.3216555267572403, "loss_total_v6": 0.8419496864080429, "step": 426 }, { "epoch": 0.09847927927927928, "grad_norm": 30.32247543334961, "learning_rate": 9.979116671785044e-06, "loss": 13.352729797363281, "loss_alignment": 1.0595703125, "loss_alignment_w": 0.52978515625, "loss_sft": 0.30479100719094276, "loss_total_v6": 0.8345456421375275, "step": 427 }, { "epoch": 0.09870990990990991, "grad_norm": 27.605993270874023, "learning_rate": 9.978932402034376e-06, "loss": 13.783149719238281, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.3366971090435982, "loss_total_v6": 0.8614468723535538, "step": 428 }, { "epoch": 0.09894054054054054, "grad_norm": 26.439056396484375, "learning_rate": 9.97874732458988e-06, "loss": 13.129066467285156, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.2979226317256689, "loss_total_v6": 0.8205666989088058, "step": 429 }, { "epoch": 0.09917117117117118, "grad_norm": 26.337568283081055, "learning_rate": 9.97856143948158e-06, "loss": 13.00759220123291, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.30156094767153263, "loss_total_v6": 0.8129745125770569, "step": 430 }, { "epoch": 0.0994018018018018, "grad_norm": 25.58818244934082, "learning_rate": 9.978374746739631e-06, "loss": 12.788475036621094, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2885222639888525, "loss_total_v6": 0.7992797344923019, "step": 431 }, { "epoch": 0.09963243243243243, "grad_norm": 25.72158432006836, "learning_rate": 9.97818724639432e-06, "loss": 12.348318099975586, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2685960438102484, "loss_total_v6": 0.7717698663473129, "step": 432 }, { "epoch": 0.09986306306306306, "grad_norm": 28.39057159423828, "learning_rate": 9.977998938476059e-06, "loss": 12.953054428100586, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.29450548812747, "loss_total_v6": 0.809565894305706, "step": 433 }, { "epoch": 0.1000936936936937, "grad_norm": 32.02418518066406, "learning_rate": 9.9778098230154e-06, "loss": 13.183126449584961, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.30317816138267517, "loss_total_v6": 0.8239453956484795, "step": 434 }, { "epoch": 0.10032432432432432, "grad_norm": 25.98126983642578, "learning_rate": 9.977619900043021e-06, "loss": 13.614463806152344, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.33321908116340637, "loss_total_v6": 0.8509040251374245, "step": 435 }, { "epoch": 0.10055495495495495, "grad_norm": 27.318918228149414, "learning_rate": 9.977429169589731e-06, "loss": 13.410614967346191, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.31590088456869125, "loss_total_v6": 0.8381634578108788, "step": 436 }, { "epoch": 0.10078558558558559, "grad_norm": 25.041217803955078, "learning_rate": 9.977237631686472e-06, "loss": 13.072519302368164, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.2929693069308996, "loss_total_v6": 0.8170324489474297, "step": 437 }, { "epoch": 0.10101621621621622, "grad_norm": 27.89703369140625, "learning_rate": 9.977045286364314e-06, "loss": 12.705818176269531, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2826085053384304, "loss_total_v6": 0.7941136360168457, "step": 438 }, { "epoch": 0.10124684684684684, "grad_norm": 27.815284729003906, "learning_rate": 9.97685213365446e-06, "loss": 12.798667907714844, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.28394074365496635, "loss_total_v6": 0.7999167218804359, "step": 439 }, { "epoch": 0.10147747747747747, "grad_norm": 27.342487335205078, "learning_rate": 9.976658173588244e-06, "loss": 12.488390922546387, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2685157582163811, "loss_total_v6": 0.7805244475603104, "step": 440 }, { "epoch": 0.10170810810810811, "grad_norm": 27.174650192260742, "learning_rate": 9.97646340619713e-06, "loss": 13.07369327545166, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.2912727277725935, "loss_total_v6": 0.81710584461689, "step": 441 }, { "epoch": 0.10193873873873874, "grad_norm": 29.16412353515625, "learning_rate": 9.976267831512716e-06, "loss": 12.468385696411133, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.25359349697828293, "loss_total_v6": 0.7792740687727928, "step": 442 }, { "epoch": 0.10216936936936936, "grad_norm": 27.91646957397461, "learning_rate": 9.976071449566726e-06, "loss": 12.816612243652344, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2897162064909935, "loss_total_v6": 0.8010382428765297, "step": 443 }, { "epoch": 0.1024, "grad_norm": 25.56772232055664, "learning_rate": 9.975874260391019e-06, "loss": 13.094284057617188, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.30070785805583, "loss_total_v6": 0.8183927983045578, "step": 444 }, { "epoch": 0.10263063063063063, "grad_norm": 24.217477798461914, "learning_rate": 9.975676264017582e-06, "loss": 12.75079345703125, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2774238269776106, "loss_total_v6": 0.7969245761632919, "step": 445 }, { "epoch": 0.10286126126126126, "grad_norm": 25.727170944213867, "learning_rate": 9.975477460478537e-06, "loss": 12.81971263885498, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2833029739558697, "loss_total_v6": 0.8012320399284363, "step": 446 }, { "epoch": 0.1030918918918919, "grad_norm": 26.72817611694336, "learning_rate": 9.975277849806133e-06, "loss": 12.850852966308594, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2874464988708496, "loss_total_v6": 0.8031783178448677, "step": 447 }, { "epoch": 0.10332252252252253, "grad_norm": 26.387548446655273, "learning_rate": 9.975077432032749e-06, "loss": 13.071531295776367, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.29573042690753937, "loss_total_v6": 0.8169706761837006, "step": 448 }, { "epoch": 0.10355315315315315, "grad_norm": 27.98566246032715, "learning_rate": 9.9748762071909e-06, "loss": 12.904966354370117, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.28742581978440285, "loss_total_v6": 0.8065603524446487, "step": 449 }, { "epoch": 0.10378378378378378, "grad_norm": 24.956342697143555, "learning_rate": 9.974674175313228e-06, "loss": 13.104476928710938, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.2928609363734722, "loss_total_v6": 0.81902976334095, "step": 450 }, { "epoch": 0.10401441441441442, "grad_norm": 29.781761169433594, "learning_rate": 9.97447133643251e-06, "loss": 12.804171562194824, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.2788373939692974, "loss_total_v6": 0.8002607077360153, "step": 451 }, { "epoch": 0.10424504504504505, "grad_norm": 25.933818817138672, "learning_rate": 9.974267690581646e-06, "loss": 12.655235290527344, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.27297731675207615, "loss_total_v6": 0.7909521609544754, "step": 452 }, { "epoch": 0.10447567567567567, "grad_norm": 28.43085479736328, "learning_rate": 9.974063237793674e-06, "loss": 13.026575088500977, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.2930427584797144, "loss_total_v6": 0.814160943031311, "step": 453 }, { "epoch": 0.10470630630630631, "grad_norm": 26.562599182128906, "learning_rate": 9.973857978101762e-06, "loss": 12.820392608642578, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2887928616255522, "loss_total_v6": 0.8012745752930641, "step": 454 }, { "epoch": 0.10493693693693694, "grad_norm": 25.87240982055664, "learning_rate": 9.973651911539209e-06, "loss": 13.254046440124512, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.3140498735010624, "loss_total_v6": 0.8283778801560402, "step": 455 }, { "epoch": 0.10516756756756757, "grad_norm": 27.988422393798828, "learning_rate": 9.973445038139437e-06, "loss": 12.472121238708496, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.26575940288603306, "loss_total_v6": 0.779507577419281, "step": 456 }, { "epoch": 0.10539819819819819, "grad_norm": 26.76930046081543, "learning_rate": 9.973237357936014e-06, "loss": 12.973901748657227, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.29747168719768524, "loss_total_v6": 0.8108688741922379, "step": 457 }, { "epoch": 0.10562882882882883, "grad_norm": 25.67837142944336, "learning_rate": 9.973028870962625e-06, "loss": 12.70114803314209, "loss_alignment": 1.03369140625, "loss_alignment_w": 0.516845703125, "loss_sft": 0.27712863124907017, "loss_total_v6": 0.7938217595219612, "step": 458 }, { "epoch": 0.10585945945945946, "grad_norm": 26.27432632446289, "learning_rate": 9.972819577253092e-06, "loss": 13.406472206115723, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.31985336169600487, "loss_total_v6": 0.8379045128822327, "step": 459 }, { "epoch": 0.10609009009009009, "grad_norm": 24.958959579467773, "learning_rate": 9.972609476841368e-06, "loss": 13.118352890014648, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.2947963550686836, "loss_total_v6": 0.8198970630764961, "step": 460 }, { "epoch": 0.10632072072072073, "grad_norm": 26.293447494506836, "learning_rate": 9.972398569761534e-06, "loss": 12.944089889526367, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.298049733042717, "loss_total_v6": 0.809005580842495, "step": 461 }, { "epoch": 0.10655135135135135, "grad_norm": 23.977821350097656, "learning_rate": 9.972186856047808e-06, "loss": 12.236660957336426, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.25096684880554676, "loss_total_v6": 0.7647913098335266, "step": 462 }, { "epoch": 0.10678198198198198, "grad_norm": 27.333602905273438, "learning_rate": 9.971974335734532e-06, "loss": 13.123481750488281, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.3012203983962536, "loss_total_v6": 0.820217601954937, "step": 463 }, { "epoch": 0.1070126126126126, "grad_norm": 26.64301300048828, "learning_rate": 9.97176100885618e-06, "loss": 12.887081146240234, "loss_alignment": 1.02099609375, "loss_alignment_w": 0.510498046875, "loss_sft": 0.2951733972877264, "loss_total_v6": 0.8054425567388535, "step": 464 }, { "epoch": 0.10724324324324325, "grad_norm": 29.42802619934082, "learning_rate": 9.97154687544736e-06, "loss": 12.819337844848633, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.27999894693493843, "loss_total_v6": 0.8012086525559425, "step": 465 }, { "epoch": 0.10747387387387387, "grad_norm": 25.510421752929688, "learning_rate": 9.97133193554281e-06, "loss": 12.53239631652832, "loss_alignment": 1.02490234375, "loss_alignment_w": 0.512451171875, "loss_sft": 0.27099149115383625, "loss_total_v6": 0.7832747846841812, "step": 466 }, { "epoch": 0.1077045045045045, "grad_norm": 26.736331939697266, "learning_rate": 9.9711161891774e-06, "loss": 13.41908073425293, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.31699449196457863, "loss_total_v6": 0.8386925384402275, "step": 467 }, { "epoch": 0.10793513513513514, "grad_norm": 27.265300750732422, "learning_rate": 9.970899636386121e-06, "loss": 13.044757843017578, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.29356888122856617, "loss_total_v6": 0.8152974098920822, "step": 468 }, { "epoch": 0.10816576576576577, "grad_norm": 26.193187713623047, "learning_rate": 9.97068227720411e-06, "loss": 12.485492706298828, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.26976895332336426, "loss_total_v6": 0.7803433239459991, "step": 469 }, { "epoch": 0.1083963963963964, "grad_norm": 28.29349708557129, "learning_rate": 9.970464111666627e-06, "loss": 13.665536880493164, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.32849185168743134, "loss_total_v6": 0.8540960401296616, "step": 470 }, { "epoch": 0.10862702702702702, "grad_norm": 24.121339797973633, "learning_rate": 9.97024513980906e-06, "loss": 12.997257232666016, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.2943689376115799, "loss_total_v6": 0.8123285546898842, "step": 471 }, { "epoch": 0.10885765765765766, "grad_norm": 28.7268009185791, "learning_rate": 9.970025361666934e-06, "loss": 13.174856185913086, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.3144105579704046, "loss_total_v6": 0.8234284967184067, "step": 472 }, { "epoch": 0.10908828828828829, "grad_norm": 30.38288116455078, "learning_rate": 9.9698047772759e-06, "loss": 13.38265323638916, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.3095908649265766, "loss_total_v6": 0.8364158198237419, "step": 473 }, { "epoch": 0.10931891891891891, "grad_norm": 25.770614624023438, "learning_rate": 9.969583386671742e-06, "loss": 13.055570602416992, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.29816614277660847, "loss_total_v6": 0.8159731701016426, "step": 474 }, { "epoch": 0.10954954954954955, "grad_norm": 25.028594970703125, "learning_rate": 9.969361189890373e-06, "loss": 12.623464584350586, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2779496759176254, "loss_total_v6": 0.7889665439724922, "step": 475 }, { "epoch": 0.10978018018018018, "grad_norm": 30.87925910949707, "learning_rate": 9.969138186967844e-06, "loss": 12.658355712890625, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2821292821317911, "loss_total_v6": 0.7911472022533417, "step": 476 }, { "epoch": 0.11001081081081081, "grad_norm": 29.079219818115234, "learning_rate": 9.968914377940326e-06, "loss": 12.852977752685547, "loss_alignment": 1.046875, "loss_alignment_w": 0.5234375, "loss_sft": 0.28013301081955433, "loss_total_v6": 0.8033111169934273, "step": 477 }, { "epoch": 0.11024144144144143, "grad_norm": 26.944400787353516, "learning_rate": 9.968689762844127e-06, "loss": 13.220124244689941, "loss_alignment": 1.03173828125, "loss_alignment_w": 0.515869140625, "loss_sft": 0.3100071232765913, "loss_total_v6": 0.8262577280402184, "step": 478 }, { "epoch": 0.11047207207207208, "grad_norm": 29.444154739379883, "learning_rate": 9.968464341715684e-06, "loss": 12.815221786499023, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.27798688784241676, "loss_total_v6": 0.800951361656189, "step": 479 }, { "epoch": 0.1107027027027027, "grad_norm": 24.940303802490234, "learning_rate": 9.968238114591567e-06, "loss": 12.855192184448242, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.276059964671731, "loss_total_v6": 0.8034494817256927, "step": 480 }, { "epoch": 0.11093333333333333, "grad_norm": 30.57796287536621, "learning_rate": 9.968011081508471e-06, "loss": 13.033937454223633, "loss_alignment": 1.01611328125, "loss_alignment_w": 0.508056640625, "loss_sft": 0.30686961859464645, "loss_total_v6": 0.8146210759878159, "step": 481 }, { "epoch": 0.11116396396396397, "grad_norm": 24.403257369995117, "learning_rate": 9.967783242503232e-06, "loss": 13.29726791381836, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.30939651280641556, "loss_total_v6": 0.8310792446136475, "step": 482 }, { "epoch": 0.1113945945945946, "grad_norm": 27.55014991760254, "learning_rate": 9.967554597612807e-06, "loss": 13.158062934875488, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.3018711507320404, "loss_total_v6": 0.8223789557814598, "step": 483 }, { "epoch": 0.11162522522522522, "grad_norm": 25.74479866027832, "learning_rate": 9.967325146874287e-06, "loss": 12.98038101196289, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.2880041506141424, "loss_total_v6": 0.8112737983465195, "step": 484 }, { "epoch": 0.11185585585585586, "grad_norm": 26.031021118164062, "learning_rate": 9.967094890324894e-06, "loss": 13.018022537231445, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2976504825055599, "loss_total_v6": 0.8136264160275459, "step": 485 }, { "epoch": 0.11208648648648649, "grad_norm": 25.978742599487305, "learning_rate": 9.966863828001982e-06, "loss": 13.091557502746582, "loss_alignment": 1.03369140625, "loss_alignment_w": 0.516845703125, "loss_sft": 0.3018343895673752, "loss_total_v6": 0.8182223290205002, "step": 486 }, { "epoch": 0.11231711711711712, "grad_norm": 27.553199768066406, "learning_rate": 9.966631959943036e-06, "loss": 12.660459518432617, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2842291109263897, "loss_total_v6": 0.791278675198555, "step": 487 }, { "epoch": 0.11254774774774774, "grad_norm": 25.88884925842285, "learning_rate": 9.966399286185666e-06, "loss": 13.129429817199707, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.30366739444434643, "loss_total_v6": 0.8205893710255623, "step": 488 }, { "epoch": 0.11277837837837838, "grad_norm": 25.443424224853516, "learning_rate": 9.966165806767619e-06, "loss": 13.454723358154297, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.32204512506723404, "loss_total_v6": 0.8409202545881271, "step": 489 }, { "epoch": 0.11300900900900901, "grad_norm": 26.858070373535156, "learning_rate": 9.965931521726771e-06, "loss": 12.687498092651367, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.28338612616062164, "loss_total_v6": 0.7929686456918716, "step": 490 }, { "epoch": 0.11323963963963964, "grad_norm": 27.5336971282959, "learning_rate": 9.965696431101128e-06, "loss": 12.775300979614258, "loss_alignment": 1.02978515625, "loss_alignment_w": 0.514892578125, "loss_sft": 0.2827397920191288, "loss_total_v6": 0.7984563484787941, "step": 491 }, { "epoch": 0.11347027027027028, "grad_norm": 25.170961380004883, "learning_rate": 9.965460534928827e-06, "loss": 12.876227378845215, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.283905440941453, "loss_total_v6": 0.8047642335295677, "step": 492 }, { "epoch": 0.1137009009009009, "grad_norm": 24.660511016845703, "learning_rate": 9.965223833248135e-06, "loss": 13.19919490814209, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.29688875563442707, "loss_total_v6": 0.8249496817588806, "step": 493 }, { "epoch": 0.11393153153153153, "grad_norm": 25.36494255065918, "learning_rate": 9.96498632609745e-06, "loss": 13.026792526245117, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.30111299082636833, "loss_total_v6": 0.8141745179891586, "step": 494 }, { "epoch": 0.11416216216216216, "grad_norm": 26.636442184448242, "learning_rate": 9.964748013515301e-06, "loss": 12.641977310180664, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.26792207919061184, "loss_total_v6": 0.7901236042380333, "step": 495 }, { "epoch": 0.1143927927927928, "grad_norm": 24.012096405029297, "learning_rate": 9.96450889554035e-06, "loss": 12.987410545349121, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.2889317758381367, "loss_total_v6": 0.8117131292819977, "step": 496 }, { "epoch": 0.11462342342342342, "grad_norm": 25.754981994628906, "learning_rate": 9.964268972211386e-06, "loss": 13.499837875366211, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.32265225052833557, "loss_total_v6": 0.8437399119138718, "step": 497 }, { "epoch": 0.11485405405405405, "grad_norm": 26.04290199279785, "learning_rate": 9.964028243567327e-06, "loss": 13.354576110839844, "loss_alignment": 1.060546875, "loss_alignment_w": 0.5302734375, "loss_sft": 0.30461651645600796, "loss_total_v6": 0.8346610367298126, "step": 498 }, { "epoch": 0.11508468468468469, "grad_norm": 26.306217193603516, "learning_rate": 9.963786709647228e-06, "loss": 12.98816967010498, "loss_alignment": 1.02197265625, "loss_alignment_w": 0.510986328125, "loss_sft": 0.3020865134894848, "loss_total_v6": 0.8117606043815613, "step": 499 }, { "epoch": 0.11531531531531532, "grad_norm": 26.58415412902832, "learning_rate": 9.96354437049027e-06, "loss": 12.913599014282227, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.29660193249583244, "loss_total_v6": 0.8070999458432198, "step": 500 }, { "epoch": 0.11531531531531532, "eval_loss": 0.966686487197876, "eval_loss_alignment": 1.036902022688356, "eval_loss_alignment_w": 0.518451011344178, "eval_loss_contrast": 0.0, "eval_loss_highlevel": 0.0, "eval_loss_sft": 0.4482355190772716, "eval_loss_total_v6": 0.9666865320034224, "eval_loss_wm_recon": 0.0, "eval_loss_wm_recon_w": 0.0, "eval_runtime": 315.3552, "eval_samples_per_second": 5.533, "eval_steps_per_second": 0.694, "step": 500 }, { "epoch": 0.11554594594594594, "grad_norm": 26.312191009521484, "learning_rate": 9.963301226135767e-06, "loss": 13.19270133972168, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.3103684689849615, "loss_total_v6": 0.824543870985508, "step": 501 }, { "epoch": 0.11577657657657657, "grad_norm": 28.242321014404297, "learning_rate": 9.96305727662316e-06, "loss": 12.37310791015625, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.24667738564312458, "loss_total_v6": 0.7733192220330238, "step": 502 }, { "epoch": 0.11600720720720721, "grad_norm": 26.31620216369629, "learning_rate": 9.962812521992026e-06, "loss": 12.816665649414062, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2838296964764595, "loss_total_v6": 0.8010416105389595, "step": 503 }, { "epoch": 0.11623783783783784, "grad_norm": 24.813547134399414, "learning_rate": 9.962566962282068e-06, "loss": 12.359809875488281, "loss_alignment": 1.03857421875, "loss_alignment_w": 0.519287109375, "loss_sft": 0.2547116130590439, "loss_total_v6": 0.772488109767437, "step": 504 }, { "epoch": 0.11646846846846846, "grad_norm": 25.285627365112305, "learning_rate": 9.962320597533119e-06, "loss": 13.096321105957031, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.30352069810032845, "loss_total_v6": 0.818520076572895, "step": 505 }, { "epoch": 0.1166990990990991, "grad_norm": 29.984670639038086, "learning_rate": 9.962073427785149e-06, "loss": 13.618926048278809, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.3240069504827261, "loss_total_v6": 0.8511828705668449, "step": 506 }, { "epoch": 0.11692972972972973, "grad_norm": 24.860881805419922, "learning_rate": 9.961825453078254e-06, "loss": 12.95035171508789, "loss_alignment": 1.056640625, "loss_alignment_w": 0.5283203125, "loss_sft": 0.2814886327832937, "loss_total_v6": 0.8093969374895096, "step": 507 }, { "epoch": 0.11716036036036036, "grad_norm": 25.017528533935547, "learning_rate": 9.961576673452656e-06, "loss": 12.61103630065918, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.26655280590057373, "loss_total_v6": 0.7881897836923599, "step": 508 }, { "epoch": 0.11739099099099098, "grad_norm": 26.463943481445312, "learning_rate": 9.961327088948718e-06, "loss": 12.662589073181152, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.27868599258363247, "loss_total_v6": 0.7914118468761444, "step": 509 }, { "epoch": 0.11762162162162162, "grad_norm": 31.51553726196289, "learning_rate": 9.961076699606927e-06, "loss": 12.760540008544922, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.2727077305316925, "loss_total_v6": 0.7975337877869606, "step": 510 }, { "epoch": 0.11785225225225225, "grad_norm": 27.773202896118164, "learning_rate": 9.960825505467902e-06, "loss": 13.179540634155273, "loss_alignment": 1.046875, "loss_alignment_w": 0.5234375, "loss_sft": 0.2993835471570492, "loss_total_v6": 0.8237212598323822, "step": 511 }, { "epoch": 0.11808288288288288, "grad_norm": 28.212827682495117, "learning_rate": 9.960573506572391e-06, "loss": 12.647859573364258, "loss_alignment": 1.02392578125, "loss_alignment_w": 0.511962890625, "loss_sft": 0.2781316004693508, "loss_total_v6": 0.7904912084341049, "step": 512 }, { "epoch": 0.11831351351351352, "grad_norm": 28.508255004882812, "learning_rate": 9.960320702961272e-06, "loss": 13.060714721679688, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.2959852572530508, "loss_total_v6": 0.8162946850061417, "step": 513 }, { "epoch": 0.11854414414414415, "grad_norm": 25.185935974121094, "learning_rate": 9.96006709467556e-06, "loss": 12.568208694458008, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2730160728096962, "loss_total_v6": 0.7855130359530449, "step": 514 }, { "epoch": 0.11877477477477477, "grad_norm": 28.912111282348633, "learning_rate": 9.959812681756394e-06, "loss": 12.848224639892578, "loss_alignment": 1.04150390625, "loss_alignment_w": 0.520751953125, "loss_sft": 0.2825062423944473, "loss_total_v6": 0.8030140101909637, "step": 515 }, { "epoch": 0.11900540540540541, "grad_norm": 26.061908721923828, "learning_rate": 9.959557464245043e-06, "loss": 12.60865592956543, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.2688148748129606, "loss_total_v6": 0.788040965795517, "step": 516 }, { "epoch": 0.11923603603603604, "grad_norm": 24.36957359313965, "learning_rate": 9.959301442182912e-06, "loss": 12.59856128692627, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.2680008951574564, "loss_total_v6": 0.78741005808115, "step": 517 }, { "epoch": 0.11946666666666667, "grad_norm": 24.267242431640625, "learning_rate": 9.959044615611533e-06, "loss": 13.104024887084961, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.2998670004308224, "loss_total_v6": 0.8190015330910683, "step": 518 }, { "epoch": 0.11969729729729729, "grad_norm": 29.023712158203125, "learning_rate": 9.958786984572568e-06, "loss": 12.842256546020508, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.2862378805875778, "loss_total_v6": 0.8026410639286041, "step": 519 }, { "epoch": 0.11992792792792793, "grad_norm": 24.743364334106445, "learning_rate": 9.958528549107812e-06, "loss": 13.139015197753906, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.31040047109127045, "loss_total_v6": 0.8211884200572968, "step": 520 }, { "epoch": 0.12015855855855856, "grad_norm": 24.151439666748047, "learning_rate": 9.958269309259186e-06, "loss": 12.586546897888184, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.267188997939229, "loss_total_v6": 0.7866592034697533, "step": 521 }, { "epoch": 0.12038918918918919, "grad_norm": 26.88058853149414, "learning_rate": 9.958009265068747e-06, "loss": 13.035402297973633, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.2930603548884392, "loss_total_v6": 0.8147125914692879, "step": 522 }, { "epoch": 0.12061981981981983, "grad_norm": 24.492036819458008, "learning_rate": 9.95774841657868e-06, "loss": 12.96580982208252, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.28526242449879646, "loss_total_v6": 0.8103631138801575, "step": 523 }, { "epoch": 0.12085045045045045, "grad_norm": 23.380582809448242, "learning_rate": 9.9574867638313e-06, "loss": 13.132941246032715, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.30223885364830494, "loss_total_v6": 0.8208088055253029, "step": 524 }, { "epoch": 0.12108108108108108, "grad_norm": 24.47530174255371, "learning_rate": 9.957224306869053e-06, "loss": 12.846025466918945, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.2781573161482811, "loss_total_v6": 0.8028765469789505, "step": 525 }, { "epoch": 0.1213117117117117, "grad_norm": 26.518049240112305, "learning_rate": 9.956961045734514e-06, "loss": 12.53261947631836, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2667329218238592, "loss_total_v6": 0.7832886874675751, "step": 526 }, { "epoch": 0.12154234234234235, "grad_norm": 27.432859420776367, "learning_rate": 9.956696980470392e-06, "loss": 13.094131469726562, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2994623202830553, "loss_total_v6": 0.818383201956749, "step": 527 }, { "epoch": 0.12177297297297297, "grad_norm": 28.96192169189453, "learning_rate": 9.956432111119522e-06, "loss": 12.788341522216797, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.2940070368349552, "loss_total_v6": 0.799271322786808, "step": 528 }, { "epoch": 0.1220036036036036, "grad_norm": 28.569440841674805, "learning_rate": 9.956166437724875e-06, "loss": 12.724550247192383, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.29333125054836273, "loss_total_v6": 0.7952843457460403, "step": 529 }, { "epoch": 0.12223423423423424, "grad_norm": 28.267053604125977, "learning_rate": 9.955899960329546e-06, "loss": 12.675802230834961, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2793897893279791, "loss_total_v6": 0.7922376841306686, "step": 530 }, { "epoch": 0.12246486486486487, "grad_norm": 24.836353302001953, "learning_rate": 9.955632678976765e-06, "loss": 12.634208679199219, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.27288385294377804, "loss_total_v6": 0.78963802754879, "step": 531 }, { "epoch": 0.1226954954954955, "grad_norm": 23.21491241455078, "learning_rate": 9.95536459370989e-06, "loss": 12.625747680664062, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.2692117728292942, "loss_total_v6": 0.7891092449426651, "step": 532 }, { "epoch": 0.12292612612612612, "grad_norm": 24.619726181030273, "learning_rate": 9.955095704572412e-06, "loss": 12.574487686157227, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.2672745119780302, "loss_total_v6": 0.7859054729342461, "step": 533 }, { "epoch": 0.12315675675675676, "grad_norm": 26.23298454284668, "learning_rate": 9.954826011607949e-06, "loss": 13.111490249633789, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.30543002486228943, "loss_total_v6": 0.8194680958986282, "step": 534 }, { "epoch": 0.12338738738738739, "grad_norm": 26.19283103942871, "learning_rate": 9.954555514860254e-06, "loss": 13.264148712158203, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.3104241080582142, "loss_total_v6": 0.8290093317627907, "step": 535 }, { "epoch": 0.12361801801801801, "grad_norm": 25.569992065429688, "learning_rate": 9.954284214373204e-06, "loss": 12.424548149108887, "loss_alignment": 1.03369140625, "loss_alignment_w": 0.516845703125, "loss_sft": 0.25915450043976307, "loss_total_v6": 0.7765342369675636, "step": 536 }, { "epoch": 0.12384864864864865, "grad_norm": 24.669307708740234, "learning_rate": 9.954012110190812e-06, "loss": 12.920608520507812, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2873964123427868, "loss_total_v6": 0.8075380176305771, "step": 537 }, { "epoch": 0.12407927927927928, "grad_norm": 25.833316802978516, "learning_rate": 9.953739202357219e-06, "loss": 13.383655548095703, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.31496354937553406, "loss_total_v6": 0.8364784717559814, "step": 538 }, { "epoch": 0.12430990990990991, "grad_norm": 24.597332000732422, "learning_rate": 9.953465490916696e-06, "loss": 12.196725845336914, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.24624314345419407, "loss_total_v6": 0.7622953951358795, "step": 539 }, { "epoch": 0.12454054054054053, "grad_norm": 26.07408905029297, "learning_rate": 9.953190975913648e-06, "loss": 13.320520401000977, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.31367262080311775, "loss_total_v6": 0.8325325101613998, "step": 540 }, { "epoch": 0.12477117117117117, "grad_norm": 29.757699966430664, "learning_rate": 9.952915657392603e-06, "loss": 12.510936737060547, "loss_alignment": 1.02783203125, "loss_alignment_w": 0.513916015625, "loss_sft": 0.2680480685085058, "loss_total_v6": 0.7819335907697678, "step": 541 }, { "epoch": 0.12500180180180182, "grad_norm": 26.984214782714844, "learning_rate": 9.952639535398227e-06, "loss": 12.548685073852539, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2775027099996805, "loss_total_v6": 0.7842928543686867, "step": 542 }, { "epoch": 0.12523243243243243, "grad_norm": 28.05750274658203, "learning_rate": 9.952362609975313e-06, "loss": 12.937284469604492, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.2883165590465069, "loss_total_v6": 0.808580219745636, "step": 543 }, { "epoch": 0.12546306306306307, "grad_norm": 26.12648582458496, "learning_rate": 9.952084881168784e-06, "loss": 12.944934844970703, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.3001015018671751, "loss_total_v6": 0.809058390557766, "step": 544 }, { "epoch": 0.12569369369369368, "grad_norm": 25.600582122802734, "learning_rate": 9.951806349023693e-06, "loss": 12.975759506225586, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.3044999726116657, "loss_total_v6": 0.8109849467873573, "step": 545 }, { "epoch": 0.12592432432432432, "grad_norm": 25.360368728637695, "learning_rate": 9.951527013585224e-06, "loss": 13.103620529174805, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.2994908392429352, "loss_total_v6": 0.8189763128757477, "step": 546 }, { "epoch": 0.12615495495495496, "grad_norm": 25.090133666992188, "learning_rate": 9.951246874898693e-06, "loss": 13.403263092041016, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.31742504239082336, "loss_total_v6": 0.8377039581537247, "step": 547 }, { "epoch": 0.12638558558558557, "grad_norm": 26.369197845458984, "learning_rate": 9.950965933009544e-06, "loss": 12.901537895202637, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.2802840881049633, "loss_total_v6": 0.8063461259007454, "step": 548 }, { "epoch": 0.12661621621621622, "grad_norm": 27.434968948364258, "learning_rate": 9.95068418796335e-06, "loss": 12.898443222045898, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.2890476118773222, "loss_total_v6": 0.806152693927288, "step": 549 }, { "epoch": 0.12684684684684686, "grad_norm": 25.131614685058594, "learning_rate": 9.950401639805822e-06, "loss": 12.666400909423828, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2731412220746279, "loss_total_v6": 0.7916501089930534, "step": 550 }, { "epoch": 0.12707747747747747, "grad_norm": 24.35344123840332, "learning_rate": 9.95011828858279e-06, "loss": 12.707046508789062, "loss_alignment": 1.03369140625, "loss_alignment_w": 0.516845703125, "loss_sft": 0.27905370481312275, "loss_total_v6": 0.7941904291510582, "step": 551 }, { "epoch": 0.1273081081081081, "grad_norm": 28.102216720581055, "learning_rate": 9.949834134340219e-06, "loss": 12.893213272094727, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.29979860596358776, "loss_total_v6": 0.8058258146047592, "step": 552 }, { "epoch": 0.12753873873873875, "grad_norm": 26.66352653503418, "learning_rate": 9.94954917712421e-06, "loss": 12.802228927612305, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.2824543844908476, "loss_total_v6": 0.800139307975769, "step": 553 }, { "epoch": 0.12776936936936936, "grad_norm": 27.645998001098633, "learning_rate": 9.94926341698099e-06, "loss": 13.078523635864258, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2983342669904232, "loss_total_v6": 0.8174077570438385, "step": 554 }, { "epoch": 0.128, "grad_norm": 26.17543601989746, "learning_rate": 9.948976853956908e-06, "loss": 12.940515518188477, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.29303520917892456, "loss_total_v6": 0.8087822645902634, "step": 555 }, { "epoch": 0.12823063063063064, "grad_norm": 26.789915084838867, "learning_rate": 9.94868948809846e-06, "loss": 12.840702056884766, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.28546927869319916, "loss_total_v6": 0.8025438636541367, "step": 556 }, { "epoch": 0.12846126126126126, "grad_norm": 28.036903381347656, "learning_rate": 9.948401319452256e-06, "loss": 12.812328338623047, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.28453516215085983, "loss_total_v6": 0.8007705211639404, "step": 557 }, { "epoch": 0.1286918918918919, "grad_norm": 26.8886661529541, "learning_rate": 9.948112348065049e-06, "loss": 13.198465347290039, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.29746884293854237, "loss_total_v6": 0.8249040991067886, "step": 558 }, { "epoch": 0.1289225225225225, "grad_norm": 29.125978469848633, "learning_rate": 9.947822573983711e-06, "loss": 12.446516990661621, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2635335698723793, "loss_total_v6": 0.7779073342680931, "step": 559 }, { "epoch": 0.12915315315315315, "grad_norm": 26.93800926208496, "learning_rate": 9.947531997255256e-06, "loss": 12.758150100708008, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.28504003025591373, "loss_total_v6": 0.7973844334483147, "step": 560 }, { "epoch": 0.1293837837837838, "grad_norm": 27.38974952697754, "learning_rate": 9.947240617926818e-06, "loss": 12.614326477050781, "loss_alignment": 1.03173828125, "loss_alignment_w": 0.515869140625, "loss_sft": 0.27109188213944435, "loss_total_v6": 0.7883953750133514, "step": 561 }, { "epoch": 0.1296144144144144, "grad_norm": 25.97523307800293, "learning_rate": 9.946948436045664e-06, "loss": 12.780937194824219, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.27747684344649315, "loss_total_v6": 0.7988086193799973, "step": 562 }, { "epoch": 0.12984504504504504, "grad_norm": 24.13138198852539, "learning_rate": 9.946655451659197e-06, "loss": 12.899014472961426, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.2844141572713852, "loss_total_v6": 0.8061884194612503, "step": 563 }, { "epoch": 0.13007567567567568, "grad_norm": 24.302322387695312, "learning_rate": 9.946361664814942e-06, "loss": 12.890789031982422, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.28864549472928047, "loss_total_v6": 0.8056743144989014, "step": 564 }, { "epoch": 0.1303063063063063, "grad_norm": 27.038063049316406, "learning_rate": 9.94606707556056e-06, "loss": 13.345306396484375, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.31921955943107605, "loss_total_v6": 0.834081657230854, "step": 565 }, { "epoch": 0.13053693693693694, "grad_norm": 25.683603286743164, "learning_rate": 9.945771683943837e-06, "loss": 12.931060791015625, "loss_alignment": 1.02783203125, "loss_alignment_w": 0.513916015625, "loss_sft": 0.29381754621863365, "loss_total_v6": 0.808191329240799, "step": 566 }, { "epoch": 0.13076756756756758, "grad_norm": 25.053667068481445, "learning_rate": 9.945475490012695e-06, "loss": 12.360380172729492, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.2624529227614403, "loss_total_v6": 0.7725237458944321, "step": 567 }, { "epoch": 0.1309981981981982, "grad_norm": 27.635324478149414, "learning_rate": 9.945178493815181e-06, "loss": 13.057008743286133, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.30799118988215923, "loss_total_v6": 0.8160630911588669, "step": 568 }, { "epoch": 0.13122882882882883, "grad_norm": 25.083093643188477, "learning_rate": 9.944880695399476e-06, "loss": 13.095059394836426, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.29968817345798016, "loss_total_v6": 0.8184412345290184, "step": 569 }, { "epoch": 0.13145945945945947, "grad_norm": 29.043231964111328, "learning_rate": 9.94458209481389e-06, "loss": 12.919328689575195, "loss_alignment": 1.046875, "loss_alignment_w": 0.5234375, "loss_sft": 0.2845393419265747, "loss_total_v6": 0.8074580430984497, "step": 570 }, { "epoch": 0.13169009009009008, "grad_norm": 26.0061092376709, "learning_rate": 9.944282692106862e-06, "loss": 12.553863525390625, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2699374798685312, "loss_total_v6": 0.7846164405345917, "step": 571 }, { "epoch": 0.13192072072072072, "grad_norm": 25.52506446838379, "learning_rate": 9.94398248732696e-06, "loss": 12.648402214050293, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2789284624159336, "loss_total_v6": 0.7905251458287239, "step": 572 }, { "epoch": 0.13215135135135136, "grad_norm": 25.037683486938477, "learning_rate": 9.943681480522886e-06, "loss": 12.724106788635254, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2840872723609209, "loss_total_v6": 0.795256681740284, "step": 573 }, { "epoch": 0.13238198198198198, "grad_norm": 26.08819580078125, "learning_rate": 9.943379671743471e-06, "loss": 12.883283615112305, "loss_alignment": 1.046875, "loss_alignment_w": 0.5234375, "loss_sft": 0.28335460647940636, "loss_total_v6": 0.8052052110433578, "step": 574 }, { "epoch": 0.13261261261261262, "grad_norm": 26.890331268310547, "learning_rate": 9.943077061037672e-06, "loss": 13.6718111038208, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.33292754366993904, "loss_total_v6": 0.8544881939888, "step": 575 }, { "epoch": 0.13284324324324323, "grad_norm": 26.54033088684082, "learning_rate": 9.94277364845458e-06, "loss": 13.153770446777344, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.3079657256603241, "loss_total_v6": 0.822110652923584, "step": 576 }, { "epoch": 0.13307387387387387, "grad_norm": 25.017988204956055, "learning_rate": 9.942469434043418e-06, "loss": 12.857152938842773, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.2812637500464916, "loss_total_v6": 0.8035720884799957, "step": 577 }, { "epoch": 0.1333045045045045, "grad_norm": 24.802614212036133, "learning_rate": 9.942164417853534e-06, "loss": 12.930374145507812, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.29826063476502895, "loss_total_v6": 0.8081483393907547, "step": 578 }, { "epoch": 0.13353513513513512, "grad_norm": 27.501028060913086, "learning_rate": 9.94185859993441e-06, "loss": 13.037270545959473, "loss_alignment": 1.056640625, "loss_alignment_w": 0.5283203125, "loss_sft": 0.2872567381709814, "loss_total_v6": 0.8148293793201447, "step": 579 }, { "epoch": 0.13376576576576577, "grad_norm": 28.504920959472656, "learning_rate": 9.941551980335653e-06, "loss": 13.415079116821289, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.32112376764416695, "loss_total_v6": 0.8384424597024918, "step": 580 }, { "epoch": 0.1339963963963964, "grad_norm": 29.491636276245117, "learning_rate": 9.941244559107007e-06, "loss": 12.663771629333496, "loss_alignment": 1.02587890625, "loss_alignment_w": 0.512939453125, "loss_sft": 0.2787141129374504, "loss_total_v6": 0.7914857268333435, "step": 581 }, { "epoch": 0.13422702702702702, "grad_norm": 28.291955947875977, "learning_rate": 9.940936336298342e-06, "loss": 13.12311840057373, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.3026930782943964, "loss_total_v6": 0.8201949074864388, "step": 582 }, { "epoch": 0.13445765765765766, "grad_norm": 26.635475158691406, "learning_rate": 9.940627311959656e-06, "loss": 12.931357383728027, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2977118007838726, "loss_total_v6": 0.8082098364830017, "step": 583 }, { "epoch": 0.1346882882882883, "grad_norm": 26.160762786865234, "learning_rate": 9.940317486141084e-06, "loss": 12.657037734985352, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.26698651164770126, "loss_total_v6": 0.791064903140068, "step": 584 }, { "epoch": 0.1349189189189189, "grad_norm": 25.833778381347656, "learning_rate": 9.940006858892883e-06, "loss": 13.300673484802246, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.311745535582304, "loss_total_v6": 0.8312920853495598, "step": 585 }, { "epoch": 0.13514954954954955, "grad_norm": 25.695749282836914, "learning_rate": 9.939695430265446e-06, "loss": 12.763875961303711, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.27807366847991943, "loss_total_v6": 0.7977422699332237, "step": 586 }, { "epoch": 0.1353801801801802, "grad_norm": 27.13169288635254, "learning_rate": 9.93938320030929e-06, "loss": 12.938613891601562, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.30414677038788795, "loss_total_v6": 0.8086633682250977, "step": 587 }, { "epoch": 0.1356108108108108, "grad_norm": 28.525867462158203, "learning_rate": 9.939070169075072e-06, "loss": 12.987272262573242, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.29713233187794685, "loss_total_v6": 0.8117044568061829, "step": 588 }, { "epoch": 0.13584144144144145, "grad_norm": 26.484699249267578, "learning_rate": 9.938756336613568e-06, "loss": 12.292116165161133, "loss_alignment": 1.03662109375, "loss_alignment_w": 0.518310546875, "loss_sft": 0.24878701381385326, "loss_total_v6": 0.7682572156190872, "step": 589 }, { "epoch": 0.13607207207207206, "grad_norm": 25.71344566345215, "learning_rate": 9.938441702975689e-06, "loss": 12.80081558227539, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.28808810375630856, "loss_total_v6": 0.8000509887933731, "step": 590 }, { "epoch": 0.1363027027027027, "grad_norm": 26.90707778930664, "learning_rate": 9.938126268212476e-06, "loss": 12.642324447631836, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.26544130221009254, "loss_total_v6": 0.7901452705264091, "step": 591 }, { "epoch": 0.13653333333333334, "grad_norm": 27.152450561523438, "learning_rate": 9.9378100323751e-06, "loss": 13.100053787231445, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.30636320635676384, "loss_total_v6": 0.8187533095479012, "step": 592 }, { "epoch": 0.13676396396396395, "grad_norm": 23.592124938964844, "learning_rate": 9.937492995514862e-06, "loss": 12.58119010925293, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.280190359801054, "loss_total_v6": 0.7863243967294693, "step": 593 }, { "epoch": 0.1369945945945946, "grad_norm": 26.462383270263672, "learning_rate": 9.937175157683191e-06, "loss": 12.620193481445312, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.28819749504327774, "loss_total_v6": 0.788762092590332, "step": 594 }, { "epoch": 0.13722522522522523, "grad_norm": 24.87664222717285, "learning_rate": 9.93685651893165e-06, "loss": 12.692827224731445, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.28197964653372765, "loss_total_v6": 0.7933016866445541, "step": 595 }, { "epoch": 0.13745585585585585, "grad_norm": 25.76787567138672, "learning_rate": 9.936537079311925e-06, "loss": 12.698431015014648, "loss_alignment": 1.03857421875, "loss_alignment_w": 0.519287109375, "loss_sft": 0.27448692359030247, "loss_total_v6": 0.7936519607901573, "step": 596 }, { "epoch": 0.1376864864864865, "grad_norm": 28.6597900390625, "learning_rate": 9.936216838875839e-06, "loss": 13.011856079101562, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.3030481096357107, "loss_total_v6": 0.8132409751415253, "step": 597 }, { "epoch": 0.13791711711711713, "grad_norm": 40.12816619873047, "learning_rate": 9.935895797675342e-06, "loss": 13.139954566955566, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.3037910982966423, "loss_total_v6": 0.8212471529841423, "step": 598 }, { "epoch": 0.13814774774774774, "grad_norm": 25.166616439819336, "learning_rate": 9.935573955762514e-06, "loss": 13.273519515991211, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.3082326836884022, "loss_total_v6": 0.8295949921011925, "step": 599 }, { "epoch": 0.13837837837837838, "grad_norm": 29.86726951599121, "learning_rate": 9.935251313189564e-06, "loss": 13.098672866821289, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.30893191881477833, "loss_total_v6": 0.8186670318245888, "step": 600 }, { "epoch": 0.13860900900900902, "grad_norm": 28.19717788696289, "learning_rate": 9.934927870008835e-06, "loss": 13.312444686889648, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.31592974439263344, "loss_total_v6": 0.832027792930603, "step": 601 }, { "epoch": 0.13883963963963963, "grad_norm": 26.336729049682617, "learning_rate": 9.934603626272791e-06, "loss": 12.84045696258545, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.28150192461907864, "loss_total_v6": 0.8025285452604294, "step": 602 }, { "epoch": 0.13907027027027027, "grad_norm": 28.094633102416992, "learning_rate": 9.934278582034037e-06, "loss": 13.031373977661133, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.2965928167104721, "loss_total_v6": 0.8144608661532402, "step": 603 }, { "epoch": 0.13930090090090091, "grad_norm": 27.0400447845459, "learning_rate": 9.9339527373453e-06, "loss": 13.055240631103516, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.292484562844038, "loss_total_v6": 0.8159525394439697, "step": 604 }, { "epoch": 0.13953153153153153, "grad_norm": 28.267654418945312, "learning_rate": 9.933626092259439e-06, "loss": 12.378707885742188, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2597838006913662, "loss_total_v6": 0.7736692726612091, "step": 605 }, { "epoch": 0.13976216216216217, "grad_norm": 26.997207641601562, "learning_rate": 9.933298646829444e-06, "loss": 12.240303993225098, "loss_alignment": 1.02783203125, "loss_alignment_w": 0.513916015625, "loss_sft": 0.25177435763180256, "loss_total_v6": 0.765018992125988, "step": 606 }, { "epoch": 0.13999279279279278, "grad_norm": 25.53390884399414, "learning_rate": 9.932970401108433e-06, "loss": 13.379739761352539, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.316778726875782, "loss_total_v6": 0.8362336978316307, "step": 607 }, { "epoch": 0.14022342342342342, "grad_norm": 24.62921142578125, "learning_rate": 9.932641355149655e-06, "loss": 12.348782539367676, "loss_alignment": 1.03759765625, "loss_alignment_w": 0.518798828125, "loss_sft": 0.25461753085255623, "loss_total_v6": 0.7717989087104797, "step": 608 }, { "epoch": 0.14045405405405406, "grad_norm": 29.65206527709961, "learning_rate": 9.93231150900649e-06, "loss": 12.934690475463867, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2946852818131447, "loss_total_v6": 0.8084181770682335, "step": 609 }, { "epoch": 0.14068468468468467, "grad_norm": 27.445755004882812, "learning_rate": 9.931980862732445e-06, "loss": 12.987306594848633, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.29183967038989067, "loss_total_v6": 0.8117066398262978, "step": 610 }, { "epoch": 0.14091531531531531, "grad_norm": 24.42794418334961, "learning_rate": 9.931649416381158e-06, "loss": 12.759627342224121, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2802342623472214, "loss_total_v6": 0.7974767088890076, "step": 611 }, { "epoch": 0.14114594594594596, "grad_norm": 25.553932189941406, "learning_rate": 9.931317170006398e-06, "loss": 13.615676879882812, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.3283052556216717, "loss_total_v6": 0.8509798347949982, "step": 612 }, { "epoch": 0.14137657657657657, "grad_norm": 27.674116134643555, "learning_rate": 9.930984123662063e-06, "loss": 13.450263977050781, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.32567257434129715, "loss_total_v6": 0.8406414687633514, "step": 613 }, { "epoch": 0.1416072072072072, "grad_norm": 27.81513023376465, "learning_rate": 9.930650277402178e-06, "loss": 12.529117584228516, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.27402137964963913, "loss_total_v6": 0.7830698192119598, "step": 614 }, { "epoch": 0.14183783783783785, "grad_norm": 25.173355102539062, "learning_rate": 9.930315631280904e-06, "loss": 13.325647354125977, "loss_alignment": 1.02392578125, "loss_alignment_w": 0.511962890625, "loss_sft": 0.3200355414301157, "loss_total_v6": 0.8328529596328735, "step": 615 }, { "epoch": 0.14206846846846846, "grad_norm": 35.57139205932617, "learning_rate": 9.929980185352525e-06, "loss": 12.918390274047852, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.29494816809892654, "loss_total_v6": 0.8073993548750877, "step": 616 }, { "epoch": 0.1422990990990991, "grad_norm": 26.118301391601562, "learning_rate": 9.92964393967146e-06, "loss": 13.395986557006836, "loss_alignment": 1.02197265625, "loss_alignment_w": 0.510986328125, "loss_sft": 0.32600345835089684, "loss_total_v6": 0.8372491672635078, "step": 617 }, { "epoch": 0.14252972972972974, "grad_norm": 27.703407287597656, "learning_rate": 9.929306894292254e-06, "loss": 12.597076416015625, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.26877787709236145, "loss_total_v6": 0.7873173281550407, "step": 618 }, { "epoch": 0.14276036036036036, "grad_norm": 27.02568244934082, "learning_rate": 9.928969049269583e-06, "loss": 12.760261535644531, "loss_alignment": 1.02392578125, "loss_alignment_w": 0.511962890625, "loss_sft": 0.28546192497015, "loss_total_v6": 0.7975163981318474, "step": 619 }, { "epoch": 0.142990990990991, "grad_norm": 27.19855499267578, "learning_rate": 9.928630404658255e-06, "loss": 12.7402982711792, "loss_alignment": 1.01708984375, "loss_alignment_w": 0.508544921875, "loss_sft": 0.2880136240273714, "loss_total_v6": 0.7962686195969582, "step": 620 }, { "epoch": 0.1432216216216216, "grad_norm": 25.745412826538086, "learning_rate": 9.928290960513204e-06, "loss": 12.992531776428223, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.29486705735325813, "loss_total_v6": 0.8120332285761833, "step": 621 }, { "epoch": 0.14345225225225225, "grad_norm": 26.017898559570312, "learning_rate": 9.927950716889496e-06, "loss": 12.972007751464844, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2938742730766535, "loss_total_v6": 0.8107504472136497, "step": 622 }, { "epoch": 0.1436828828828829, "grad_norm": 28.519241333007812, "learning_rate": 9.927609673842327e-06, "loss": 12.966390609741211, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.3024342954158783, "loss_total_v6": 0.8103993833065033, "step": 623 }, { "epoch": 0.1439135135135135, "grad_norm": 26.313030242919922, "learning_rate": 9.92726783142702e-06, "loss": 12.725348472595215, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2900852579623461, "loss_total_v6": 0.7953342720866203, "step": 624 }, { "epoch": 0.14414414414414414, "grad_norm": 23.93934440612793, "learning_rate": 9.92692518969903e-06, "loss": 12.45691204071045, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2678758520632982, "loss_total_v6": 0.7785569950938225, "step": 625 }, { "epoch": 0.14437477477477478, "grad_norm": 25.98436737060547, "learning_rate": 9.926581748713942e-06, "loss": 13.05859088897705, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.2949063964188099, "loss_total_v6": 0.8161619007587433, "step": 626 }, { "epoch": 0.1446054054054054, "grad_norm": 26.846315383911133, "learning_rate": 9.926237508527468e-06, "loss": 13.428739547729492, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.3228929415345192, "loss_total_v6": 0.8392961844801903, "step": 627 }, { "epoch": 0.14483603603603604, "grad_norm": 33.93305587768555, "learning_rate": 9.925892469195454e-06, "loss": 13.363790512084961, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.32298414409160614, "loss_total_v6": 0.835236944258213, "step": 628 }, { "epoch": 0.14506666666666668, "grad_norm": 26.14534568786621, "learning_rate": 9.92554663077387e-06, "loss": 13.058530807495117, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.3000601977109909, "loss_total_v6": 0.8161581978201866, "step": 629 }, { "epoch": 0.1452972972972973, "grad_norm": 25.938827514648438, "learning_rate": 9.925199993318822e-06, "loss": 12.621797561645508, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2743664775043726, "loss_total_v6": 0.7888623476028442, "step": 630 }, { "epoch": 0.14552792792792793, "grad_norm": 25.684900283813477, "learning_rate": 9.92485255688654e-06, "loss": 12.74429988861084, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.2751259319484234, "loss_total_v6": 0.7965187579393387, "step": 631 }, { "epoch": 0.14575855855855857, "grad_norm": 25.37688446044922, "learning_rate": 9.924504321533387e-06, "loss": 12.474485397338867, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.26885215006768703, "loss_total_v6": 0.779655359685421, "step": 632 }, { "epoch": 0.14598918918918918, "grad_norm": 25.92481231689453, "learning_rate": 9.924155287315853e-06, "loss": 12.593707084655762, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.2657748833298683, "loss_total_v6": 0.7871066927909851, "step": 633 }, { "epoch": 0.14621981981981982, "grad_norm": 25.560678482055664, "learning_rate": 9.923805454290563e-06, "loss": 12.324453353881836, "loss_alignment": 1.01416015625, "loss_alignment_w": 0.507080078125, "loss_sft": 0.2631830330938101, "loss_total_v6": 0.7702783644199371, "step": 634 }, { "epoch": 0.14645045045045044, "grad_norm": 26.80547332763672, "learning_rate": 9.923454822514262e-06, "loss": 13.200392723083496, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.3034638911485672, "loss_total_v6": 0.8250245451927185, "step": 635 }, { "epoch": 0.14668108108108108, "grad_norm": 25.90398406982422, "learning_rate": 9.923103392043836e-06, "loss": 12.270649909973145, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2586758676916361, "loss_total_v6": 0.7669156119227409, "step": 636 }, { "epoch": 0.14691171171171172, "grad_norm": 25.74221420288086, "learning_rate": 9.92275116293629e-06, "loss": 12.52717399597168, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.277882469817996, "loss_total_v6": 0.78294837474823, "step": 637 }, { "epoch": 0.14714234234234233, "grad_norm": 28.164236068725586, "learning_rate": 9.922398135248766e-06, "loss": 13.280370712280273, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.3035033755004406, "loss_total_v6": 0.8300231695175171, "step": 638 }, { "epoch": 0.14737297297297297, "grad_norm": 23.58901596069336, "learning_rate": 9.922044309038534e-06, "loss": 12.425328254699707, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.26127845235168934, "loss_total_v6": 0.7765830233693123, "step": 639 }, { "epoch": 0.1476036036036036, "grad_norm": 24.43073844909668, "learning_rate": 9.921689684362989e-06, "loss": 12.193334579467773, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.2578567247837782, "loss_total_v6": 0.7620834037661552, "step": 640 }, { "epoch": 0.14783423423423422, "grad_norm": 26.530614852905273, "learning_rate": 9.921334261279663e-06, "loss": 12.684370040893555, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.27334877103567123, "loss_total_v6": 0.7927731871604919, "step": 641 }, { "epoch": 0.14806486486486486, "grad_norm": 28.99506378173828, "learning_rate": 9.92097803984621e-06, "loss": 12.750833511352539, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.28137840516865253, "loss_total_v6": 0.7969270944595337, "step": 642 }, { "epoch": 0.1482954954954955, "grad_norm": 28.278385162353516, "learning_rate": 9.92062102012042e-06, "loss": 12.873858451843262, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.28581733629107475, "loss_total_v6": 0.8046161234378815, "step": 643 }, { "epoch": 0.14852612612612612, "grad_norm": 25.522977828979492, "learning_rate": 9.920263202160208e-06, "loss": 12.6189603805542, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2792703155428171, "loss_total_v6": 0.7886850163340569, "step": 644 }, { "epoch": 0.14875675675675676, "grad_norm": 25.328208923339844, "learning_rate": 9.91990458602362e-06, "loss": 12.862970352172852, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.28705937042832375, "loss_total_v6": 0.8039356246590614, "step": 645 }, { "epoch": 0.1489873873873874, "grad_norm": 24.04787254333496, "learning_rate": 9.919545171768832e-06, "loss": 12.562536239624023, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.2593712117522955, "loss_total_v6": 0.7851585671305656, "step": 646 }, { "epoch": 0.149218018018018, "grad_norm": 25.412281036376953, "learning_rate": 9.919184959454148e-06, "loss": 12.762542724609375, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.2935847993940115, "loss_total_v6": 0.7976589202880859, "step": 647 }, { "epoch": 0.14944864864864865, "grad_norm": 24.56756591796875, "learning_rate": 9.918823949138004e-06, "loss": 12.700674057006836, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.28674258291721344, "loss_total_v6": 0.7937921583652496, "step": 648 }, { "epoch": 0.1496792792792793, "grad_norm": 23.611074447631836, "learning_rate": 9.918462140878965e-06, "loss": 12.58669662475586, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2724626027047634, "loss_total_v6": 0.7866685464978218, "step": 649 }, { "epoch": 0.1499099099099099, "grad_norm": 23.001985549926758, "learning_rate": 9.91809953473572e-06, "loss": 12.798919677734375, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.285879148170352, "loss_total_v6": 0.799932487308979, "step": 650 }, { "epoch": 0.15014054054054055, "grad_norm": 24.946304321289062, "learning_rate": 9.917736130767095e-06, "loss": 12.632038116455078, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.27644085697829723, "loss_total_v6": 0.7895023822784424, "step": 651 }, { "epoch": 0.15037117117117116, "grad_norm": 26.47816276550293, "learning_rate": 9.91737192903204e-06, "loss": 12.482312202453613, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.25579148903489113, "loss_total_v6": 0.7801445126533508, "step": 652 }, { "epoch": 0.1506018018018018, "grad_norm": 27.00640869140625, "learning_rate": 9.917006929589642e-06, "loss": 12.46141242980957, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.26104654371738434, "loss_total_v6": 0.7788382917642593, "step": 653 }, { "epoch": 0.15083243243243244, "grad_norm": 26.695350646972656, "learning_rate": 9.916641132499105e-06, "loss": 12.837185859680176, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2867601476609707, "loss_total_v6": 0.802324116230011, "step": 654 }, { "epoch": 0.15106306306306305, "grad_norm": 24.79532814025879, "learning_rate": 9.916274537819774e-06, "loss": 12.865705490112305, "loss_alignment": 1.01806640625, "loss_alignment_w": 0.509033203125, "loss_sft": 0.2950429394841194, "loss_total_v6": 0.8041066527366638, "step": 655 }, { "epoch": 0.1512936936936937, "grad_norm": 27.257116317749023, "learning_rate": 9.915907145611117e-06, "loss": 12.889053344726562, "loss_alignment": 1.02099609375, "loss_alignment_w": 0.510498046875, "loss_sft": 0.29657838121056557, "loss_total_v6": 0.805565819144249, "step": 656 }, { "epoch": 0.15152432432432433, "grad_norm": 28.0289363861084, "learning_rate": 9.915538955932735e-06, "loss": 12.710088729858398, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.28012880124151707, "loss_total_v6": 0.7943805009126663, "step": 657 }, { "epoch": 0.15175495495495495, "grad_norm": 27.87278938293457, "learning_rate": 9.915169968844354e-06, "loss": 12.823953628540039, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2847581896930933, "loss_total_v6": 0.8014970868825912, "step": 658 }, { "epoch": 0.1519855855855856, "grad_norm": 25.151775360107422, "learning_rate": 9.914800184405836e-06, "loss": 13.612893104553223, "loss_alignment": 1.03173828125, "loss_alignment_w": 0.515869140625, "loss_sft": 0.3356080800294876, "loss_total_v6": 0.850805826485157, "step": 659 }, { "epoch": 0.15221621621621623, "grad_norm": 24.901269912719727, "learning_rate": 9.914429602677163e-06, "loss": 13.246557235717773, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.31742706149816513, "loss_total_v6": 0.8279098272323608, "step": 660 }, { "epoch": 0.15244684684684684, "grad_norm": 24.718061447143555, "learning_rate": 9.914058223718453e-06, "loss": 13.0673828125, "loss_alignment": 1.060546875, "loss_alignment_w": 0.5302734375, "loss_sft": 0.28630065359175205, "loss_total_v6": 0.8167113959789276, "step": 661 }, { "epoch": 0.15267747747747748, "grad_norm": 28.250259399414062, "learning_rate": 9.913686047589955e-06, "loss": 13.462625503540039, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.3167863544076681, "loss_total_v6": 0.84141406416893, "step": 662 }, { "epoch": 0.15290810810810812, "grad_norm": 26.99733543395996, "learning_rate": 9.913313074352042e-06, "loss": 12.793542861938477, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2841087877750397, "loss_total_v6": 0.799596443772316, "step": 663 }, { "epoch": 0.15313873873873873, "grad_norm": 27.67563819885254, "learning_rate": 9.912939304065219e-06, "loss": 13.07211685180664, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.30368638038635254, "loss_total_v6": 0.8170072957873344, "step": 664 }, { "epoch": 0.15336936936936937, "grad_norm": 26.80008888244629, "learning_rate": 9.91256473679012e-06, "loss": 12.889068603515625, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.2888279128819704, "loss_total_v6": 0.8055667951703072, "step": 665 }, { "epoch": 0.1536, "grad_norm": 26.09194564819336, "learning_rate": 9.912189372587508e-06, "loss": 12.549737930297852, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2801624834537506, "loss_total_v6": 0.7843586206436157, "step": 666 }, { "epoch": 0.15383063063063063, "grad_norm": 26.931978225708008, "learning_rate": 9.911813211518275e-06, "loss": 13.351831436157227, "loss_alignment": 1.0576171875, "loss_alignment_w": 0.52880859375, "loss_sft": 0.3067641817033291, "loss_total_v6": 0.8344894126057625, "step": 667 }, { "epoch": 0.15406126126126127, "grad_norm": 24.806724548339844, "learning_rate": 9.911436253643445e-06, "loss": 13.260250091552734, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.31108072213828564, "loss_total_v6": 0.8287656530737877, "step": 668 }, { "epoch": 0.15429189189189188, "grad_norm": 24.159433364868164, "learning_rate": 9.911058499024164e-06, "loss": 12.799392700195312, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.29034899547696114, "loss_total_v6": 0.7999620363116264, "step": 669 }, { "epoch": 0.15452252252252252, "grad_norm": 24.812665939331055, "learning_rate": 9.910679947721718e-06, "loss": 12.940567016601562, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.28466129675507545, "loss_total_v6": 0.8087854534387589, "step": 670 }, { "epoch": 0.15475315315315316, "grad_norm": 25.970821380615234, "learning_rate": 9.910300599797514e-06, "loss": 12.930059432983398, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.29082521237432957, "loss_total_v6": 0.8081286922097206, "step": 671 }, { "epoch": 0.15498378378378377, "grad_norm": 27.05844497680664, "learning_rate": 9.909920455313087e-06, "loss": 13.162961959838867, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.3037948030978441, "loss_total_v6": 0.8226851522922516, "step": 672 }, { "epoch": 0.15521441441441441, "grad_norm": 24.561992645263672, "learning_rate": 9.90953951433011e-06, "loss": 13.201085090637207, "loss_alignment": 1.0556640625, "loss_alignment_w": 0.52783203125, "loss_sft": 0.29654916748404503, "loss_total_v6": 0.8250678479671478, "step": 673 }, { "epoch": 0.15544504504504505, "grad_norm": 25.084980010986328, "learning_rate": 9.90915777691038e-06, "loss": 12.553705215454102, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.2669673915952444, "loss_total_v6": 0.7846065387129784, "step": 674 }, { "epoch": 0.15567567567567567, "grad_norm": 26.226572036743164, "learning_rate": 9.908775243115822e-06, "loss": 12.950485229492188, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.30194377340376377, "loss_total_v6": 0.8094053193926811, "step": 675 }, { "epoch": 0.1559063063063063, "grad_norm": 26.074552536010742, "learning_rate": 9.908391913008491e-06, "loss": 13.126757621765137, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.29335324652493, "loss_total_v6": 0.820422351360321, "step": 676 }, { "epoch": 0.15613693693693695, "grad_norm": 24.526281356811523, "learning_rate": 9.90800778665057e-06, "loss": 12.72873306274414, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.2751447819173336, "loss_total_v6": 0.7955458089709282, "step": 677 }, { "epoch": 0.15636756756756756, "grad_norm": 24.50518798828125, "learning_rate": 9.907622864104376e-06, "loss": 12.576227188110352, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2790256589651108, "loss_total_v6": 0.7860142141580582, "step": 678 }, { "epoch": 0.1565981981981982, "grad_norm": 27.64592933654785, "learning_rate": 9.907237145432352e-06, "loss": 12.849087715148926, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.2872293647378683, "loss_total_v6": 0.8030679821968079, "step": 679 }, { "epoch": 0.15682882882882884, "grad_norm": 31.601991653442383, "learning_rate": 9.906850630697068e-06, "loss": 13.240705490112305, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.30586136132478714, "loss_total_v6": 0.8275441080331802, "step": 680 }, { "epoch": 0.15705945945945946, "grad_norm": 25.63794708251953, "learning_rate": 9.906463319961225e-06, "loss": 12.696516036987305, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2831715941429138, "loss_total_v6": 0.7935322672128677, "step": 681 }, { "epoch": 0.1572900900900901, "grad_norm": 30.002397537231445, "learning_rate": 9.906075213287658e-06, "loss": 13.362637519836426, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.31139167211949825, "loss_total_v6": 0.835164837539196, "step": 682 }, { "epoch": 0.1575207207207207, "grad_norm": 24.633684158325195, "learning_rate": 9.905686310739321e-06, "loss": 12.6905517578125, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.27738188207149506, "loss_total_v6": 0.7931594625115395, "step": 683 }, { "epoch": 0.15775135135135135, "grad_norm": 25.78680419921875, "learning_rate": 9.905296612379307e-06, "loss": 12.847464561462402, "loss_alignment": 1.02197265625, "loss_alignment_w": 0.510986328125, "loss_sft": 0.2921023014932871, "loss_total_v6": 0.8029665574431419, "step": 684 }, { "epoch": 0.157981981981982, "grad_norm": 23.374948501586914, "learning_rate": 9.904906118270831e-06, "loss": 12.928506851196289, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.28901926800608635, "loss_total_v6": 0.8080316707491875, "step": 685 }, { "epoch": 0.1582126126126126, "grad_norm": 25.50172233581543, "learning_rate": 9.90451482847724e-06, "loss": 12.627735137939453, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.27925414964556694, "loss_total_v6": 0.7892333939671516, "step": 686 }, { "epoch": 0.15844324324324324, "grad_norm": 25.144378662109375, "learning_rate": 9.90412274306201e-06, "loss": 12.692527770996094, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2820067200809717, "loss_total_v6": 0.7932829782366753, "step": 687 }, { "epoch": 0.15867387387387388, "grad_norm": 25.427711486816406, "learning_rate": 9.903729862088749e-06, "loss": 12.998790740966797, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.29586860351264477, "loss_total_v6": 0.8124244064092636, "step": 688 }, { "epoch": 0.1589045045045045, "grad_norm": 28.80398941040039, "learning_rate": 9.903336185621186e-06, "loss": 13.070733070373535, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.30019725300371647, "loss_total_v6": 0.8169208243489265, "step": 689 }, { "epoch": 0.15913513513513514, "grad_norm": 25.900236129760742, "learning_rate": 9.902941713723187e-06, "loss": 12.792778968811035, "loss_alignment": 1.03076171875, "loss_alignment_w": 0.515380859375, "loss_sft": 0.28294712863862514, "loss_total_v6": 0.7995486855506897, "step": 690 }, { "epoch": 0.15936576576576578, "grad_norm": 27.665361404418945, "learning_rate": 9.902546446458745e-06, "loss": 12.868614196777344, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2918372768908739, "loss_total_v6": 0.8042884320020676, "step": 691 }, { "epoch": 0.1595963963963964, "grad_norm": 26.556562423706055, "learning_rate": 9.902150383891979e-06, "loss": 12.354516983032227, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.25834811478853226, "loss_total_v6": 0.7721573188900948, "step": 692 }, { "epoch": 0.15982702702702703, "grad_norm": 24.603759765625, "learning_rate": 9.901753526087141e-06, "loss": 12.596633911132812, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2711457647383213, "loss_total_v6": 0.7872895896434784, "step": 693 }, { "epoch": 0.16005765765765767, "grad_norm": 24.424028396606445, "learning_rate": 9.901355873108611e-06, "loss": 12.359451293945312, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.2534074578434229, "loss_total_v6": 0.772465705871582, "step": 694 }, { "epoch": 0.16028828828828828, "grad_norm": 27.30283546447754, "learning_rate": 9.900957425020894e-06, "loss": 12.90626335144043, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.28849872946739197, "loss_total_v6": 0.8066414222121239, "step": 695 }, { "epoch": 0.16051891891891892, "grad_norm": 25.86164665222168, "learning_rate": 9.900558181888626e-06, "loss": 12.294149398803711, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.2472509015351534, "loss_total_v6": 0.7683843150734901, "step": 696 }, { "epoch": 0.16074954954954954, "grad_norm": 25.115148544311523, "learning_rate": 9.90015814377658e-06, "loss": 12.624631881713867, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.27173603884875774, "loss_total_v6": 0.7890395149588585, "step": 697 }, { "epoch": 0.16098018018018018, "grad_norm": 27.00659942626953, "learning_rate": 9.899757310749644e-06, "loss": 13.273676872253418, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.30999723076820374, "loss_total_v6": 0.8296047896146774, "step": 698 }, { "epoch": 0.16121081081081082, "grad_norm": 54.950286865234375, "learning_rate": 9.899355682872847e-06, "loss": 12.910636901855469, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.3009028397500515, "loss_total_v6": 0.8069148287177086, "step": 699 }, { "epoch": 0.16144144144144143, "grad_norm": 24.86806869506836, "learning_rate": 9.89895326021134e-06, "loss": 12.978174209594727, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.29030766151845455, "loss_total_v6": 0.8111359030008316, "step": 700 }, { "epoch": 0.16167207207207207, "grad_norm": 25.126781463623047, "learning_rate": 9.898550042830404e-06, "loss": 12.743075370788574, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.28858392871916294, "loss_total_v6": 0.7964422032237053, "step": 701 }, { "epoch": 0.1619027027027027, "grad_norm": 26.38187026977539, "learning_rate": 9.89814603079545e-06, "loss": 12.410459518432617, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.26019654609262943, "loss_total_v6": 0.7756537050008774, "step": 702 }, { "epoch": 0.16213333333333332, "grad_norm": 23.116649627685547, "learning_rate": 9.89774122417202e-06, "loss": 12.58230972290039, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.26765659637749195, "loss_total_v6": 0.786394365131855, "step": 703 }, { "epoch": 0.16236396396396396, "grad_norm": 26.527563095092773, "learning_rate": 9.89733562302578e-06, "loss": 12.56964111328125, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.27446363866329193, "loss_total_v6": 0.7856025844812393, "step": 704 }, { "epoch": 0.1625945945945946, "grad_norm": 23.58380699157715, "learning_rate": 9.89692922742253e-06, "loss": 12.267995834350586, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.26592570543289185, "loss_total_v6": 0.7667496725916862, "step": 705 }, { "epoch": 0.16282522522522522, "grad_norm": 25.22478485107422, "learning_rate": 9.896522037428195e-06, "loss": 13.33951187133789, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.31765200197696686, "loss_total_v6": 0.8337194919586182, "step": 706 }, { "epoch": 0.16305585585585586, "grad_norm": 25.691022872924805, "learning_rate": 9.89611405310883e-06, "loss": 12.620613098144531, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2755436636507511, "loss_total_v6": 0.7887883111834526, "step": 707 }, { "epoch": 0.1632864864864865, "grad_norm": 26.522781372070312, "learning_rate": 9.89570527453062e-06, "loss": 13.739449501037598, "loss_alignment": 1.046875, "loss_alignment_w": 0.5234375, "loss_sft": 0.33398110792040825, "loss_total_v6": 0.8587155938148499, "step": 708 }, { "epoch": 0.1635171171171171, "grad_norm": 27.207462310791016, "learning_rate": 9.895295701759878e-06, "loss": 12.762750625610352, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.2791019566357136, "loss_total_v6": 0.7976718842983246, "step": 709 }, { "epoch": 0.16374774774774775, "grad_norm": 27.504634857177734, "learning_rate": 9.894885334863044e-06, "loss": 12.86101245880127, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.30254673585295677, "loss_total_v6": 0.8038132563233376, "step": 710 }, { "epoch": 0.1639783783783784, "grad_norm": 25.526594161987305, "learning_rate": 9.89447417390669e-06, "loss": 12.81015396118164, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.28734422102570534, "loss_total_v6": 0.8006346300244331, "step": 711 }, { "epoch": 0.164209009009009, "grad_norm": 27.18994903564453, "learning_rate": 9.894062218957517e-06, "loss": 13.284619331359863, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.3138550240546465, "loss_total_v6": 0.830288715660572, "step": 712 }, { "epoch": 0.16443963963963965, "grad_norm": 34.461509704589844, "learning_rate": 9.89364947008235e-06, "loss": 12.747507095336914, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.28377972543239594, "loss_total_v6": 0.7967191934585571, "step": 713 }, { "epoch": 0.16467027027027026, "grad_norm": 24.951515197753906, "learning_rate": 9.89323592734815e-06, "loss": 12.520636558532715, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.2658161222934723, "loss_total_v6": 0.7825397774577141, "step": 714 }, { "epoch": 0.1649009009009009, "grad_norm": 26.572511672973633, "learning_rate": 9.892821590822e-06, "loss": 12.891220092773438, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.29326541535556316, "loss_total_v6": 0.8057012856006622, "step": 715 }, { "epoch": 0.16513153153153154, "grad_norm": 24.233299255371094, "learning_rate": 9.892406460571116e-06, "loss": 12.648765563964844, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.27383948117494583, "loss_total_v6": 0.7905478402972221, "step": 716 }, { "epoch": 0.16536216216216215, "grad_norm": 23.4626407623291, "learning_rate": 9.891990536662839e-06, "loss": 12.56556510925293, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.2890098821371794, "loss_total_v6": 0.7853477746248245, "step": 717 }, { "epoch": 0.1655927927927928, "grad_norm": 35.93181228637695, "learning_rate": 9.891573819164643e-06, "loss": 12.899420738220215, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.297241635620594, "loss_total_v6": 0.8062137886881828, "step": 718 }, { "epoch": 0.16582342342342343, "grad_norm": 28.088748931884766, "learning_rate": 9.891156308144131e-06, "loss": 12.819770812988281, "loss_alignment": 1.01513671875, "loss_alignment_w": 0.507568359375, "loss_sft": 0.29249238036572933, "loss_total_v6": 0.8012356460094452, "step": 719 }, { "epoch": 0.16605405405405405, "grad_norm": 26.706958770751953, "learning_rate": 9.890738003669029e-06, "loss": 12.937522888183594, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2929701842367649, "loss_total_v6": 0.808595173060894, "step": 720 }, { "epoch": 0.16628468468468469, "grad_norm": 28.350814819335938, "learning_rate": 9.890318905807196e-06, "loss": 12.99354362487793, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.29580015130341053, "loss_total_v6": 0.8120965361595154, "step": 721 }, { "epoch": 0.16651531531531533, "grad_norm": 26.401981353759766, "learning_rate": 9.88989901462662e-06, "loss": 12.789826393127441, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.27662853710353374, "loss_total_v6": 0.7993641421198845, "step": 722 }, { "epoch": 0.16674594594594594, "grad_norm": 28.78522491455078, "learning_rate": 9.889478330195419e-06, "loss": 13.031574249267578, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.29529310017824173, "loss_total_v6": 0.8144734054803848, "step": 723 }, { "epoch": 0.16697657657657658, "grad_norm": 25.434240341186523, "learning_rate": 9.889056852581833e-06, "loss": 13.141365051269531, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.3070988077670336, "loss_total_v6": 0.8213353082537651, "step": 724 }, { "epoch": 0.16720720720720722, "grad_norm": 26.063766479492188, "learning_rate": 9.888634581854235e-06, "loss": 13.324468612670898, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.30671732500195503, "loss_total_v6": 0.8327793106436729, "step": 725 }, { "epoch": 0.16743783783783783, "grad_norm": 26.956266403198242, "learning_rate": 9.88821151808113e-06, "loss": 12.816878318786621, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.28207293525338173, "loss_total_v6": 0.8010548874735832, "step": 726 }, { "epoch": 0.16766846846846847, "grad_norm": 24.96210289001465, "learning_rate": 9.887787661331148e-06, "loss": 13.025574684143066, "loss_alignment": 1.02587890625, "loss_alignment_w": 0.512939453125, "loss_sft": 0.30114371329545975, "loss_total_v6": 0.8140984252095222, "step": 727 }, { "epoch": 0.1678990990990991, "grad_norm": 23.73434066772461, "learning_rate": 9.887363011673046e-06, "loss": 12.77352237701416, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2862296774983406, "loss_total_v6": 0.7983451411128044, "step": 728 }, { "epoch": 0.16812972972972973, "grad_norm": 25.227813720703125, "learning_rate": 9.886937569175711e-06, "loss": 12.464187622070312, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.27644828148186207, "loss_total_v6": 0.7790117412805557, "step": 729 }, { "epoch": 0.16836036036036037, "grad_norm": 26.306018829345703, "learning_rate": 9.886511333908165e-06, "loss": 12.858901977539062, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2924814261496067, "loss_total_v6": 0.803681381046772, "step": 730 }, { "epoch": 0.16859099099099098, "grad_norm": 26.247892379760742, "learning_rate": 9.886084305939545e-06, "loss": 12.36287784576416, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.26610337011516094, "loss_total_v6": 0.7726798579096794, "step": 731 }, { "epoch": 0.16882162162162162, "grad_norm": 26.8717098236084, "learning_rate": 9.885656485339129e-06, "loss": 12.834117889404297, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2932517547160387, "loss_total_v6": 0.8021323829889297, "step": 732 }, { "epoch": 0.16905225225225226, "grad_norm": 24.9556827545166, "learning_rate": 9.88522787217632e-06, "loss": 12.89307975769043, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2966927345842123, "loss_total_v6": 0.8058174774050713, "step": 733 }, { "epoch": 0.16928288288288287, "grad_norm": 22.829702377319336, "learning_rate": 9.884798466520645e-06, "loss": 13.24515438079834, "loss_alignment": 1.02294921875, "loss_alignment_w": 0.511474609375, "loss_sft": 0.31695788353681564, "loss_total_v6": 0.8278221637010574, "step": 734 }, { "epoch": 0.16951351351351351, "grad_norm": 23.75434112548828, "learning_rate": 9.884368268441764e-06, "loss": 12.642912864685059, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.28157605789601803, "loss_total_v6": 0.7901820316910744, "step": 735 }, { "epoch": 0.16974414414414415, "grad_norm": 25.64857292175293, "learning_rate": 9.883937278009468e-06, "loss": 12.519207000732422, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.27195242792367935, "loss_total_v6": 0.7824504747986794, "step": 736 }, { "epoch": 0.16997477477477477, "grad_norm": 24.195571899414062, "learning_rate": 9.883505495293671e-06, "loss": 12.673824310302734, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2741696983575821, "loss_total_v6": 0.7921140491962433, "step": 737 }, { "epoch": 0.1702054054054054, "grad_norm": 22.93238067626953, "learning_rate": 9.883072920364417e-06, "loss": 13.181422233581543, "loss_alignment": 1.04052734375, "loss_alignment_w": 0.520263671875, "loss_sft": 0.30204929038882256, "loss_total_v6": 0.823838897049427, "step": 738 }, { "epoch": 0.17043603603603605, "grad_norm": 24.88545036315918, "learning_rate": 9.88263955329188e-06, "loss": 12.954841613769531, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.29113815166056156, "loss_total_v6": 0.8096775636076927, "step": 739 }, { "epoch": 0.17066666666666666, "grad_norm": 25.728748321533203, "learning_rate": 9.882205394146362e-06, "loss": 12.881635665893555, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2880887035280466, "loss_total_v6": 0.8051022216677666, "step": 740 }, { "epoch": 0.1708972972972973, "grad_norm": 23.14019203186035, "learning_rate": 9.881770442998293e-06, "loss": 13.018394470214844, "loss_alignment": 1.03271484375, "loss_alignment_w": 0.516357421875, "loss_sft": 0.2967582158744335, "loss_total_v6": 0.8136496543884277, "step": 741 }, { "epoch": 0.17112792792792791, "grad_norm": 26.10726547241211, "learning_rate": 9.881334699918231e-06, "loss": 12.699186325073242, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2768840119242668, "loss_total_v6": 0.7936991676688194, "step": 742 }, { "epoch": 0.17135855855855855, "grad_norm": 26.788999557495117, "learning_rate": 9.880898164976865e-06, "loss": 12.480337142944336, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.25766692124307156, "loss_total_v6": 0.780021071434021, "step": 743 }, { "epoch": 0.1715891891891892, "grad_norm": 23.667551040649414, "learning_rate": 9.88046083824501e-06, "loss": 12.911821365356445, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.301602503284812, "loss_total_v6": 0.8069888800382614, "step": 744 }, { "epoch": 0.1718198198198198, "grad_norm": 28.079092025756836, "learning_rate": 9.88002271979361e-06, "loss": 13.030040740966797, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2991950213909149, "loss_total_v6": 0.8143775165081024, "step": 745 }, { "epoch": 0.17205045045045045, "grad_norm": 26.46876335144043, "learning_rate": 9.879583809693737e-06, "loss": 12.716822624206543, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2828080039471388, "loss_total_v6": 0.7948014065623283, "step": 746 }, { "epoch": 0.1722810810810811, "grad_norm": 23.532888412475586, "learning_rate": 9.879144108016595e-06, "loss": 13.220942497253418, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.30149812437593937, "loss_total_v6": 0.826308898627758, "step": 747 }, { "epoch": 0.1725117117117117, "grad_norm": 24.548564910888672, "learning_rate": 9.878703614833507e-06, "loss": 12.888794898986816, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.29325113631784916, "loss_total_v6": 0.8055496886372566, "step": 748 }, { "epoch": 0.17274234234234234, "grad_norm": 25.204320907592773, "learning_rate": 9.878262330215938e-06, "loss": 13.116640090942383, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.308010246604681, "loss_total_v6": 0.8197900354862213, "step": 749 }, { "epoch": 0.17297297297297298, "grad_norm": 26.618515014648438, "learning_rate": 9.87782025423547e-06, "loss": 13.011829376220703, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.29178548231720924, "loss_total_v6": 0.8132393062114716, "step": 750 }, { "epoch": 0.1732036036036036, "grad_norm": 24.540475845336914, "learning_rate": 9.877377386963819e-06, "loss": 12.504146575927734, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.27395602874457836, "loss_total_v6": 0.781509131193161, "step": 751 }, { "epoch": 0.17343423423423424, "grad_norm": 130.2270050048828, "learning_rate": 9.876933728472826e-06, "loss": 12.869086265563965, "loss_alignment": 1.03564453125, "loss_alignment_w": 0.517822265625, "loss_sft": 0.2868923433125019, "loss_total_v6": 0.8043178766965866, "step": 752 }, { "epoch": 0.17366486486486488, "grad_norm": 24.366514205932617, "learning_rate": 9.876489278834465e-06, "loss": 13.03012752532959, "loss_alignment": 1.03369140625, "loss_alignment_w": 0.516845703125, "loss_sft": 0.2984069921076298, "loss_total_v6": 0.8143829554319382, "step": 753 }, { "epoch": 0.1738954954954955, "grad_norm": 26.863840103149414, "learning_rate": 9.876044038120835e-06, "loss": 12.915399551391602, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.3000103384256363, "loss_total_v6": 0.8072125315666199, "step": 754 }, { "epoch": 0.17412612612612613, "grad_norm": 27.497352600097656, "learning_rate": 9.875598006404164e-06, "loss": 12.43121337890625, "loss_alignment": 1.00732421875, "loss_alignment_w": 0.503662109375, "loss_sft": 0.27412791922688484, "loss_total_v6": 0.7769508138298988, "step": 755 }, { "epoch": 0.17435675675675677, "grad_norm": 25.39338493347168, "learning_rate": 9.875151183756806e-06, "loss": 12.419429779052734, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.26277138479053974, "loss_total_v6": 0.7762143462896347, "step": 756 }, { "epoch": 0.17458738738738738, "grad_norm": 23.978958129882812, "learning_rate": 9.874703570251247e-06, "loss": 12.238714218139648, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.24874532036483288, "loss_total_v6": 0.7649196311831474, "step": 757 }, { "epoch": 0.17481801801801802, "grad_norm": 24.95552635192871, "learning_rate": 9.874255165960102e-06, "loss": 12.898588180541992, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.3016146346926689, "loss_total_v6": 0.8061617389321327, "step": 758 }, { "epoch": 0.17504864864864864, "grad_norm": 24.217954635620117, "learning_rate": 9.87380597095611e-06, "loss": 12.291646957397461, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.25241982378065586, "loss_total_v6": 0.7682279273867607, "step": 759 }, { "epoch": 0.17527927927927928, "grad_norm": 24.557321548461914, "learning_rate": 9.873355985312141e-06, "loss": 13.125810623168945, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.2933093048632145, "loss_total_v6": 0.8203631639480591, "step": 760 }, { "epoch": 0.17550990990990992, "grad_norm": 24.91750717163086, "learning_rate": 9.872905209101191e-06, "loss": 13.059236526489258, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.3000737112015486, "loss_total_v6": 0.8162022978067398, "step": 761 }, { "epoch": 0.17574054054054053, "grad_norm": 26.43608856201172, "learning_rate": 9.87245364239639e-06, "loss": 12.75311279296875, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.28585433028638363, "loss_total_v6": 0.7970695272088051, "step": 762 }, { "epoch": 0.17597117117117117, "grad_norm": 24.038162231445312, "learning_rate": 9.872001285270987e-06, "loss": 12.514680862426758, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2735157962888479, "loss_total_v6": 0.7821675166487694, "step": 763 }, { "epoch": 0.1762018018018018, "grad_norm": 26.679351806640625, "learning_rate": 9.871548137798367e-06, "loss": 12.244030952453613, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2592094335705042, "loss_total_v6": 0.765251912176609, "step": 764 }, { "epoch": 0.17643243243243242, "grad_norm": 24.25432586669922, "learning_rate": 9.871094200052043e-06, "loss": 12.976471900939941, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.29277996718883514, "loss_total_v6": 0.8110294938087463, "step": 765 }, { "epoch": 0.17666306306306306, "grad_norm": 23.82530975341797, "learning_rate": 9.87063947210565e-06, "loss": 12.849987983703613, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.28946760296821594, "loss_total_v6": 0.8031242191791534, "step": 766 }, { "epoch": 0.1768936936936937, "grad_norm": 25.925724029541016, "learning_rate": 9.870183954032957e-06, "loss": 12.439226150512695, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2702341917902231, "loss_total_v6": 0.7774516046047211, "step": 767 }, { "epoch": 0.17712432432432432, "grad_norm": 24.63605308532715, "learning_rate": 9.86972764590786e-06, "loss": 12.758553504943848, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2841802351176739, "loss_total_v6": 0.7974096313118935, "step": 768 }, { "epoch": 0.17735495495495496, "grad_norm": 28.860532760620117, "learning_rate": 9.86927054780438e-06, "loss": 12.776335716247559, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.28915209509432316, "loss_total_v6": 0.7985209822654724, "step": 769 }, { "epoch": 0.1775855855855856, "grad_norm": 25.311561584472656, "learning_rate": 9.868812659796669e-06, "loss": 13.345193862915039, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.3149553220719099, "loss_total_v6": 0.8340745866298676, "step": 770 }, { "epoch": 0.1778162162162162, "grad_norm": 29.122081756591797, "learning_rate": 9.868353981959007e-06, "loss": 12.731729507446289, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.28372447565197945, "loss_total_v6": 0.7957331389188766, "step": 771 }, { "epoch": 0.17804684684684685, "grad_norm": 25.426334381103516, "learning_rate": 9.867894514365802e-06, "loss": 12.627214431762695, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.27882494777441025, "loss_total_v6": 0.789200909435749, "step": 772 }, { "epoch": 0.17827747747747746, "grad_norm": 24.307575225830078, "learning_rate": 9.86743425709159e-06, "loss": 12.894948959350586, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.30053271912038326, "loss_total_v6": 0.805934339761734, "step": 773 }, { "epoch": 0.1785081081081081, "grad_norm": 24.349458694458008, "learning_rate": 9.866973210211036e-06, "loss": 12.848241806030273, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.27620536647737026, "loss_total_v6": 0.8030150607228279, "step": 774 }, { "epoch": 0.17873873873873874, "grad_norm": 25.21619415283203, "learning_rate": 9.86651137379893e-06, "loss": 13.047924041748047, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.2913406230509281, "loss_total_v6": 0.8154952898621559, "step": 775 }, { "epoch": 0.17896936936936936, "grad_norm": 26.758432388305664, "learning_rate": 9.866048747930194e-06, "loss": 13.074543952941895, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.3020222634077072, "loss_total_v6": 0.8171589970588684, "step": 776 }, { "epoch": 0.1792, "grad_norm": 27.203420639038086, "learning_rate": 9.865585332679874e-06, "loss": 12.847286224365234, "loss_alignment": 1.02099609375, "loss_alignment_w": 0.510498046875, "loss_sft": 0.2932812999933958, "loss_total_v6": 0.8029553517699242, "step": 777 }, { "epoch": 0.17943063063063064, "grad_norm": 24.34959602355957, "learning_rate": 9.865121128123148e-06, "loss": 12.930129051208496, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2956055924296379, "loss_total_v6": 0.8081330806016922, "step": 778 }, { "epoch": 0.17966126126126125, "grad_norm": 27.562746047973633, "learning_rate": 9.864656134335321e-06, "loss": 13.057497024536133, "loss_alignment": 1.0576171875, "loss_alignment_w": 0.52880859375, "loss_sft": 0.2881852388381958, "loss_total_v6": 0.8160935416817665, "step": 779 }, { "epoch": 0.1798918918918919, "grad_norm": 26.836275100708008, "learning_rate": 9.864190351391822e-06, "loss": 13.503023147583008, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.3192349448800087, "loss_total_v6": 0.843938909471035, "step": 780 }, { "epoch": 0.18012252252252253, "grad_norm": 29.058130264282227, "learning_rate": 9.863723779368218e-06, "loss": 13.214378356933594, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.30056909658014774, "loss_total_v6": 0.8258986920118332, "step": 781 }, { "epoch": 0.18035315315315315, "grad_norm": 25.540218353271484, "learning_rate": 9.86325641834019e-06, "loss": 13.284429550170898, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.30567967519164085, "loss_total_v6": 0.83027683198452, "step": 782 }, { "epoch": 0.18058378378378379, "grad_norm": 27.155485153198242, "learning_rate": 9.862788268383557e-06, "loss": 13.239325523376465, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.30551571771502495, "loss_total_v6": 0.8274578526616096, "step": 783 }, { "epoch": 0.18081441441441443, "grad_norm": 24.92167854309082, "learning_rate": 9.862319329574264e-06, "loss": 12.75518798828125, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.2782783415168524, "loss_total_v6": 0.7971992418169975, "step": 784 }, { "epoch": 0.18104504504504504, "grad_norm": 25.286540985107422, "learning_rate": 9.861849601988384e-06, "loss": 13.235442161560059, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.30325886607170105, "loss_total_v6": 0.8272151425480843, "step": 785 }, { "epoch": 0.18127567567567568, "grad_norm": 25.82410430908203, "learning_rate": 9.861379085702117e-06, "loss": 12.560598373413086, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.27098400704562664, "loss_total_v6": 0.7850373685359955, "step": 786 }, { "epoch": 0.18150630630630632, "grad_norm": 25.587879180908203, "learning_rate": 9.86090778079179e-06, "loss": 12.719440460205078, "loss_alignment": 1.02099609375, "loss_alignment_w": 0.510498046875, "loss_sft": 0.28461961075663567, "loss_total_v6": 0.7949650511145592, "step": 787 }, { "epoch": 0.18173693693693693, "grad_norm": 25.98177146911621, "learning_rate": 9.860435687333857e-06, "loss": 12.604421615600586, "loss_alignment": 1.02392578125, "loss_alignment_w": 0.511962890625, "loss_sft": 0.27665264904499054, "loss_total_v6": 0.787776306271553, "step": 788 }, { "epoch": 0.18196756756756757, "grad_norm": 25.199718475341797, "learning_rate": 9.859962805404908e-06, "loss": 12.444589614868164, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.26554929465055466, "loss_total_v6": 0.7777868658304214, "step": 789 }, { "epoch": 0.18219819819819819, "grad_norm": 25.955196380615234, "learning_rate": 9.859489135081653e-06, "loss": 13.20944881439209, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.322432029992342, "loss_total_v6": 0.8255905583500862, "step": 790 }, { "epoch": 0.18242882882882883, "grad_norm": 25.829017639160156, "learning_rate": 9.859014676440928e-06, "loss": 12.212934494018555, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.26300322636961937, "loss_total_v6": 0.7633084356784821, "step": 791 }, { "epoch": 0.18265945945945947, "grad_norm": 24.460344314575195, "learning_rate": 9.858539429559705e-06, "loss": 12.660675048828125, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.28045847080647945, "loss_total_v6": 0.7912921980023384, "step": 792 }, { "epoch": 0.18289009009009008, "grad_norm": 26.24640464782715, "learning_rate": 9.858063394515079e-06, "loss": 12.644420623779297, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2774436492472887, "loss_total_v6": 0.790276288986206, "step": 793 }, { "epoch": 0.18312072072072072, "grad_norm": 25.099123001098633, "learning_rate": 9.85758657138427e-06, "loss": 13.030670166015625, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.2955112438648939, "loss_total_v6": 0.8144168704748154, "step": 794 }, { "epoch": 0.18335135135135136, "grad_norm": 24.91761589050293, "learning_rate": 9.857108960244633e-06, "loss": 12.459453582763672, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2685992680490017, "loss_total_v6": 0.7787158638238907, "step": 795 }, { "epoch": 0.18358198198198197, "grad_norm": 23.34810447692871, "learning_rate": 9.856630561173648e-06, "loss": 12.001541137695312, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.25589466094970703, "loss_total_v6": 0.750096321105957, "step": 796 }, { "epoch": 0.1838126126126126, "grad_norm": 24.326698303222656, "learning_rate": 9.856151374248919e-06, "loss": 12.804902076721191, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2829113285988569, "loss_total_v6": 0.8003063574433327, "step": 797 }, { "epoch": 0.18404324324324325, "grad_norm": 23.786527633666992, "learning_rate": 9.85567139954818e-06, "loss": 13.156200408935547, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.30752257630228996, "loss_total_v6": 0.8222625404596329, "step": 798 }, { "epoch": 0.18427387387387387, "grad_norm": 26.919523239135742, "learning_rate": 9.855190637149299e-06, "loss": 12.578397750854492, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.27615537494421005, "loss_total_v6": 0.7861498594284058, "step": 799 }, { "epoch": 0.1845045045045045, "grad_norm": 27.966108322143555, "learning_rate": 9.854709087130261e-06, "loss": 13.008110046386719, "loss_alignment": 1.0517578125, "loss_alignment_w": 0.52587890625, "loss_sft": 0.28825712390244007, "loss_total_v6": 0.8130068555474281, "step": 800 }, { "epoch": 0.18473513513513515, "grad_norm": 24.662837982177734, "learning_rate": 9.854226749569187e-06, "loss": 12.618715286254883, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.27942286618053913, "loss_total_v6": 0.788669690489769, "step": 801 }, { "epoch": 0.18496576576576576, "grad_norm": 25.433988571166992, "learning_rate": 9.85374362454432e-06, "loss": 13.191226959228516, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.30820112489163876, "loss_total_v6": 0.8244517296552658, "step": 802 }, { "epoch": 0.1851963963963964, "grad_norm": 26.322032928466797, "learning_rate": 9.853259712134038e-06, "loss": 12.900972366333008, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.299139179289341, "loss_total_v6": 0.8063108026981354, "step": 803 }, { "epoch": 0.185427027027027, "grad_norm": 24.09431266784668, "learning_rate": 9.85277501241684e-06, "loss": 12.856502532958984, "loss_alignment": 1.01513671875, "loss_alignment_w": 0.507568359375, "loss_sft": 0.2962834760546684, "loss_total_v6": 0.8035314083099365, "step": 804 }, { "epoch": 0.18565765765765765, "grad_norm": 23.73737144470215, "learning_rate": 9.852289525471353e-06, "loss": 12.863628387451172, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2897555623203516, "loss_total_v6": 0.8039767816662788, "step": 805 }, { "epoch": 0.1858882882882883, "grad_norm": 29.819664001464844, "learning_rate": 9.851803251376337e-06, "loss": 13.367655754089355, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.3133074454963207, "loss_total_v6": 0.8354784771800041, "step": 806 }, { "epoch": 0.1861189189189189, "grad_norm": 26.24610137939453, "learning_rate": 9.851316190210676e-06, "loss": 12.968624114990234, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.29027529433369637, "loss_total_v6": 0.8105389848351479, "step": 807 }, { "epoch": 0.18634954954954955, "grad_norm": 25.527429580688477, "learning_rate": 9.850828342053382e-06, "loss": 12.918888092041016, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.29142401553690434, "loss_total_v6": 0.8074304908514023, "step": 808 }, { "epoch": 0.1865801801801802, "grad_norm": 25.64033317565918, "learning_rate": 9.850339706983594e-06, "loss": 12.822704315185547, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2929809093475342, "loss_total_v6": 0.8014190346002579, "step": 809 }, { "epoch": 0.1868108108108108, "grad_norm": 25.374250411987305, "learning_rate": 9.84985028508058e-06, "loss": 12.712275505065918, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.2759320419281721, "loss_total_v6": 0.7945172414183617, "step": 810 }, { "epoch": 0.18704144144144144, "grad_norm": 25.804048538208008, "learning_rate": 9.849360076423736e-06, "loss": 12.619983673095703, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.27219320833683014, "loss_total_v6": 0.7887489795684814, "step": 811 }, { "epoch": 0.18727207207207208, "grad_norm": 24.680395126342773, "learning_rate": 9.848869081092581e-06, "loss": 12.925713539123535, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.30445445515215397, "loss_total_v6": 0.8078571259975433, "step": 812 }, { "epoch": 0.1875027027027027, "grad_norm": 23.267370223999023, "learning_rate": 9.848377299166771e-06, "loss": 13.088725090026855, "loss_alignment": 1.03173828125, "loss_alignment_w": 0.515869140625, "loss_sft": 0.3024050612002611, "loss_total_v6": 0.8180453255772591, "step": 813 }, { "epoch": 0.18773333333333334, "grad_norm": 25.326946258544922, "learning_rate": 9.847884730726081e-06, "loss": 13.014900207519531, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2973484992980957, "loss_total_v6": 0.8134312704205513, "step": 814 }, { "epoch": 0.18796396396396398, "grad_norm": 26.476362228393555, "learning_rate": 9.847391375850415e-06, "loss": 12.792768478393555, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.28276341781020164, "loss_total_v6": 0.7995480820536613, "step": 815 }, { "epoch": 0.1881945945945946, "grad_norm": 23.626073837280273, "learning_rate": 9.84689723461981e-06, "loss": 12.529796600341797, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.2857216037809849, "loss_total_v6": 0.7831123173236847, "step": 816 }, { "epoch": 0.18842522522522523, "grad_norm": 27.497501373291016, "learning_rate": 9.846402307114423e-06, "loss": 12.85400676727295, "loss_alignment": 1.01806640625, "loss_alignment_w": 0.509033203125, "loss_sft": 0.292801134288311, "loss_total_v6": 0.8033754229545593, "step": 817 }, { "epoch": 0.18865585585585587, "grad_norm": 24.83193588256836, "learning_rate": 9.845906593414545e-06, "loss": 13.10726547241211, "loss_alignment": 1.03759765625, "loss_alignment_w": 0.518798828125, "loss_sft": 0.30029840394854546, "loss_total_v6": 0.8192040622234344, "step": 818 }, { "epoch": 0.18888648648648648, "grad_norm": 28.052265167236328, "learning_rate": 9.845410093600591e-06, "loss": 13.158634185791016, "loss_alignment": 1.03662109375, "loss_alignment_w": 0.518310546875, "loss_sft": 0.3037684094160795, "loss_total_v6": 0.8224146142601967, "step": 819 }, { "epoch": 0.18911711711711712, "grad_norm": 26.540891647338867, "learning_rate": 9.844912807753105e-06, "loss": 12.359521865844727, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2661071661859751, "loss_total_v6": 0.7724701017141342, "step": 820 }, { "epoch": 0.18934774774774774, "grad_norm": 22.331825256347656, "learning_rate": 9.844414735952754e-06, "loss": 13.29101276397705, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.31799302995204926, "loss_total_v6": 0.8306883051991463, "step": 821 }, { "epoch": 0.18957837837837838, "grad_norm": 23.499143600463867, "learning_rate": 9.843915878280342e-06, "loss": 13.220477104187012, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.3210918288677931, "loss_total_v6": 0.8262798115611076, "step": 822 }, { "epoch": 0.18980900900900902, "grad_norm": 23.97972869873047, "learning_rate": 9.843416234816791e-06, "loss": 12.956291198730469, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.30148265324532986, "loss_total_v6": 0.8097681701183319, "step": 823 }, { "epoch": 0.19003963963963963, "grad_norm": 24.293617248535156, "learning_rate": 9.842915805643156e-06, "loss": 13.029071807861328, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2994091436266899, "loss_total_v6": 0.8143169730901718, "step": 824 }, { "epoch": 0.19027027027027027, "grad_norm": 24.57569694519043, "learning_rate": 9.842414590840618e-06, "loss": 12.445659637451172, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.2782809715718031, "loss_total_v6": 0.7778537049889565, "step": 825 }, { "epoch": 0.1905009009009009, "grad_norm": 23.32499885559082, "learning_rate": 9.841912590490484e-06, "loss": 12.730192184448242, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2868937700986862, "loss_total_v6": 0.7956370487809181, "step": 826 }, { "epoch": 0.19073153153153152, "grad_norm": 22.829147338867188, "learning_rate": 9.84140980467419e-06, "loss": 12.625785827636719, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.27130462415516376, "loss_total_v6": 0.7891116216778755, "step": 827 }, { "epoch": 0.19096216216216216, "grad_norm": 28.346220016479492, "learning_rate": 9.840906233473298e-06, "loss": 13.19163703918457, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.30169595032930374, "loss_total_v6": 0.8244773298501968, "step": 828 }, { "epoch": 0.1911927927927928, "grad_norm": 23.122394561767578, "learning_rate": 9.840401876969501e-06, "loss": 12.817073822021484, "loss_alignment": 1.03076171875, "loss_alignment_w": 0.515380859375, "loss_sft": 0.2862050011754036, "loss_total_v6": 0.8010670766234398, "step": 829 }, { "epoch": 0.19142342342342342, "grad_norm": 26.536354064941406, "learning_rate": 9.839896735244615e-06, "loss": 12.076944351196289, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.25464119762182236, "loss_total_v6": 0.7548090517520905, "step": 830 }, { "epoch": 0.19165405405405406, "grad_norm": 25.552995681762695, "learning_rate": 9.839390808380586e-06, "loss": 12.66198444366455, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2847517170011997, "loss_total_v6": 0.7913740277290344, "step": 831 }, { "epoch": 0.1918846846846847, "grad_norm": 25.999210357666016, "learning_rate": 9.838884096459486e-06, "loss": 13.062463760375977, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.29400404915213585, "loss_total_v6": 0.8164039701223373, "step": 832 }, { "epoch": 0.1921153153153153, "grad_norm": 26.37415313720703, "learning_rate": 9.838376599563517e-06, "loss": 12.79658317565918, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.2809113170951605, "loss_total_v6": 0.7997864112257957, "step": 833 }, { "epoch": 0.19234594594594595, "grad_norm": 27.987621307373047, "learning_rate": 9.837868317775002e-06, "loss": 13.342815399169922, "loss_alignment": 1.046875, "loss_alignment_w": 0.5234375, "loss_sft": 0.3103206120431423, "loss_total_v6": 0.8339259326457977, "step": 834 }, { "epoch": 0.19257657657657656, "grad_norm": 27.31919288635254, "learning_rate": 9.8373592511764e-06, "loss": 12.805718421936035, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.2981906197965145, "loss_total_v6": 0.8003573790192604, "step": 835 }, { "epoch": 0.1928072072072072, "grad_norm": 24.72922706604004, "learning_rate": 9.836849399850291e-06, "loss": 12.3326416015625, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2591323684900999, "loss_total_v6": 0.7707900851964951, "step": 836 }, { "epoch": 0.19303783783783784, "grad_norm": 24.19696807861328, "learning_rate": 9.836338763879386e-06, "loss": 12.81197738647461, "loss_alignment": 1.02392578125, "loss_alignment_w": 0.511962890625, "loss_sft": 0.2886637020856142, "loss_total_v6": 0.8007486090064049, "step": 837 }, { "epoch": 0.19326846846846846, "grad_norm": 26.532209396362305, "learning_rate": 9.835827343346518e-06, "loss": 13.146507263183594, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.3018202893435955, "loss_total_v6": 0.8216566890478134, "step": 838 }, { "epoch": 0.1934990990990991, "grad_norm": 25.21561622619629, "learning_rate": 9.835315138334653e-06, "loss": 12.732148170471191, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.27781492844223976, "loss_total_v6": 0.7957592681050301, "step": 839 }, { "epoch": 0.19372972972972974, "grad_norm": 25.734649658203125, "learning_rate": 9.834802148926883e-06, "loss": 12.93410873413086, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2974869906902313, "loss_total_v6": 0.8083817809820175, "step": 840 }, { "epoch": 0.19396036036036035, "grad_norm": 24.570375442504883, "learning_rate": 9.834288375206424e-06, "loss": 12.622756958007812, "loss_alignment": 1.01416015625, "loss_alignment_w": 0.507080078125, "loss_sft": 0.28123185969889164, "loss_total_v6": 0.7889223024249077, "step": 841 }, { "epoch": 0.194190990990991, "grad_norm": 26.045785903930664, "learning_rate": 9.833773817256624e-06, "loss": 12.846722602844238, "loss_alignment": 1.01513671875, "loss_alignment_w": 0.507568359375, "loss_sft": 0.29623679630458355, "loss_total_v6": 0.8029201552271843, "step": 842 }, { "epoch": 0.19442162162162163, "grad_norm": 25.69863510131836, "learning_rate": 9.833258475160955e-06, "loss": 13.035877227783203, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.3081505261361599, "loss_total_v6": 0.8147422969341278, "step": 843 }, { "epoch": 0.19465225225225224, "grad_norm": 30.272192001342773, "learning_rate": 9.832742349003014e-06, "loss": 13.019092559814453, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.29751897417008877, "loss_total_v6": 0.8136932775378227, "step": 844 }, { "epoch": 0.19488288288288289, "grad_norm": 25.61075782775879, "learning_rate": 9.832225438866532e-06, "loss": 12.994476318359375, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.29408841021358967, "loss_total_v6": 0.8121547996997833, "step": 845 }, { "epoch": 0.19511351351351353, "grad_norm": 23.888967514038086, "learning_rate": 9.831707744835363e-06, "loss": 13.290983200073242, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.30412090942263603, "loss_total_v6": 0.8306864947080612, "step": 846 }, { "epoch": 0.19534414414414414, "grad_norm": 26.001628875732422, "learning_rate": 9.831189266993486e-06, "loss": 12.958112716674805, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2969273626804352, "loss_total_v6": 0.8098820745944977, "step": 847 }, { "epoch": 0.19557477477477478, "grad_norm": 23.0902042388916, "learning_rate": 9.830670005425012e-06, "loss": 12.747451782226562, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.28165532648563385, "loss_total_v6": 0.7967157289385796, "step": 848 }, { "epoch": 0.1958054054054054, "grad_norm": 23.727441787719727, "learning_rate": 9.830149960214177e-06, "loss": 12.439176559448242, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.26579083502292633, "loss_total_v6": 0.7774485424160957, "step": 849 }, { "epoch": 0.19603603603603603, "grad_norm": 23.630077362060547, "learning_rate": 9.829629131445342e-06, "loss": 12.810001373291016, "loss_alignment": 1.04833984375, "loss_alignment_w": 0.524169921875, "loss_sft": 0.27730970084667206, "loss_total_v6": 0.8006251305341721, "step": 850 }, { "epoch": 0.19626666666666667, "grad_norm": 26.49712371826172, "learning_rate": 9.829107519202999e-06, "loss": 12.450311660766602, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.26912648417055607, "loss_total_v6": 0.7781444415450096, "step": 851 }, { "epoch": 0.19649729729729729, "grad_norm": 23.78496742248535, "learning_rate": 9.828585123571763e-06, "loss": 13.455516815185547, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.33199767395853996, "loss_total_v6": 0.8409698233008385, "step": 852 }, { "epoch": 0.19672792792792793, "grad_norm": 24.557058334350586, "learning_rate": 9.828061944636381e-06, "loss": 12.765837669372559, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.2752818185836077, "loss_total_v6": 0.7978648468852043, "step": 853 }, { "epoch": 0.19695855855855857, "grad_norm": 25.068559646606445, "learning_rate": 9.827537982481721e-06, "loss": 12.914548873901367, "loss_alignment": 1.03564453125, "loss_alignment_w": 0.517822265625, "loss_sft": 0.28863509371876717, "loss_total_v6": 0.8071592897176743, "step": 854 }, { "epoch": 0.19718918918918918, "grad_norm": 23.0399227142334, "learning_rate": 9.827013237192781e-06, "loss": 13.05763053894043, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.31350792199373245, "loss_total_v6": 0.8161019161343575, "step": 855 }, { "epoch": 0.19741981981981982, "grad_norm": 24.853599548339844, "learning_rate": 9.826487708854693e-06, "loss": 13.211982727050781, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.3001904785633087, "loss_total_v6": 0.825748935341835, "step": 856 }, { "epoch": 0.19765045045045046, "grad_norm": 26.840181350708008, "learning_rate": 9.825961397552702e-06, "loss": 12.714981079101562, "loss_alignment": 1.01904296875, "loss_alignment_w": 0.509521484375, "loss_sft": 0.2848902028053999, "loss_total_v6": 0.7946863025426865, "step": 857 }, { "epoch": 0.19788108108108107, "grad_norm": 26.424854278564453, "learning_rate": 9.82543430337219e-06, "loss": 12.861289024353027, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.2881446033716202, "loss_total_v6": 0.8038306087255478, "step": 858 }, { "epoch": 0.1981117117117117, "grad_norm": 25.910812377929688, "learning_rate": 9.824906426398665e-06, "loss": 12.784424781799316, "loss_alignment": 1.02099609375, "loss_alignment_w": 0.510498046875, "loss_sft": 0.28807077184319496, "loss_total_v6": 0.7990265786647797, "step": 859 }, { "epoch": 0.19834234234234235, "grad_norm": 26.549110412597656, "learning_rate": 9.824377766717758e-06, "loss": 13.04052448272705, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.3086088076233864, "loss_total_v6": 0.8150327727198601, "step": 860 }, { "epoch": 0.19857297297297297, "grad_norm": 30.0708065032959, "learning_rate": 9.82384832441523e-06, "loss": 12.522346496582031, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2705312352627516, "loss_total_v6": 0.7826466932892799, "step": 861 }, { "epoch": 0.1988036036036036, "grad_norm": 25.405672073364258, "learning_rate": 9.82331809957697e-06, "loss": 12.817218780517578, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2888996358960867, "loss_total_v6": 0.8010761886835098, "step": 862 }, { "epoch": 0.19903423423423425, "grad_norm": 27.778182983398438, "learning_rate": 9.822787092288991e-06, "loss": 12.719718933105469, "loss_alignment": 1.01708984375, "loss_alignment_w": 0.508544921875, "loss_sft": 0.28507942520081997, "loss_total_v6": 0.7949823811650276, "step": 863 }, { "epoch": 0.19926486486486486, "grad_norm": 30.10167121887207, "learning_rate": 9.822255302637434e-06, "loss": 12.726430892944336, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.29099216125905514, "loss_total_v6": 0.7954019457101822, "step": 864 }, { "epoch": 0.1994954954954955, "grad_norm": 26.63839340209961, "learning_rate": 9.821722730708567e-06, "loss": 13.133904457092285, "loss_alignment": 1.05078125, "loss_alignment_w": 0.525390625, "loss_sft": 0.2957987915724516, "loss_total_v6": 0.820869006216526, "step": 865 }, { "epoch": 0.1997261261261261, "grad_norm": 29.474308013916016, "learning_rate": 9.821189376588787e-06, "loss": 12.753986358642578, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.28702286072075367, "loss_total_v6": 0.7971241921186447, "step": 866 }, { "epoch": 0.19995675675675675, "grad_norm": 26.261476516723633, "learning_rate": 9.820655240364612e-06, "loss": 12.331867218017578, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2681629955768585, "loss_total_v6": 0.7707417234778404, "step": 867 }, { "epoch": 0.2001873873873874, "grad_norm": 24.17524528503418, "learning_rate": 9.820120322122695e-06, "loss": 12.547246932983398, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2765430025756359, "loss_total_v6": 0.784202940762043, "step": 868 }, { "epoch": 0.200418018018018, "grad_norm": 23.9962215423584, "learning_rate": 9.81958462194981e-06, "loss": 12.621373176574707, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.27639990858733654, "loss_total_v6": 0.7888358384370804, "step": 869 }, { "epoch": 0.20064864864864865, "grad_norm": 24.55247688293457, "learning_rate": 9.819048139932858e-06, "loss": 12.941123962402344, "loss_alignment": 1.02685546875, "loss_alignment_w": 0.513427734375, "loss_sft": 0.2968268617987633, "loss_total_v6": 0.8088202700018883, "step": 870 }, { "epoch": 0.2008792792792793, "grad_norm": 24.668916702270508, "learning_rate": 9.818510876158871e-06, "loss": 12.516768455505371, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.27584356628358364, "loss_total_v6": 0.7822980359196663, "step": 871 }, { "epoch": 0.2011099099099099, "grad_norm": 26.446008682250977, "learning_rate": 9.817972830715003e-06, "loss": 13.165813446044922, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.3035457208752632, "loss_total_v6": 0.82286337018013, "step": 872 }, { "epoch": 0.20134054054054054, "grad_norm": 23.393293380737305, "learning_rate": 9.817434003688537e-06, "loss": 12.537313461303711, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.278531476855278, "loss_total_v6": 0.7835820987820625, "step": 873 }, { "epoch": 0.20157117117117118, "grad_norm": 23.68467903137207, "learning_rate": 9.816894395166885e-06, "loss": 12.934778213500977, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2979560848325491, "loss_total_v6": 0.8084236606955528, "step": 874 }, { "epoch": 0.2018018018018018, "grad_norm": 23.871726989746094, "learning_rate": 9.816354005237583e-06, "loss": 12.920633316040039, "loss_alignment": 1.02294921875, "loss_alignment_w": 0.511474609375, "loss_sft": 0.2953782659024, "loss_total_v6": 0.8075395449995995, "step": 875 }, { "epoch": 0.20203243243243243, "grad_norm": 24.501943588256836, "learning_rate": 9.815812833988292e-06, "loss": 13.178450584411621, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.30683796666562557, "loss_total_v6": 0.8236531391739845, "step": 876 }, { "epoch": 0.20226306306306308, "grad_norm": 24.3010311126709, "learning_rate": 9.815270881506804e-06, "loss": 12.6051025390625, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2774887103587389, "loss_total_v6": 0.7878188863396645, "step": 877 }, { "epoch": 0.2024936936936937, "grad_norm": 25.546350479125977, "learning_rate": 9.814728147881034e-06, "loss": 12.512014389038086, "loss_alignment": 1.01220703125, "loss_alignment_w": 0.506103515625, "loss_sft": 0.2759278677403927, "loss_total_v6": 0.782000869512558, "step": 878 }, { "epoch": 0.20272432432432433, "grad_norm": 26.833423614501953, "learning_rate": 9.814184633199027e-06, "loss": 13.034786224365234, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.30161260813474655, "loss_total_v6": 0.8146741464734077, "step": 879 }, { "epoch": 0.20295495495495494, "grad_norm": 22.741718292236328, "learning_rate": 9.813640337548955e-06, "loss": 12.413747787475586, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.2742112837731838, "loss_total_v6": 0.7758592143654823, "step": 880 }, { "epoch": 0.20318558558558558, "grad_norm": 23.902732849121094, "learning_rate": 9.81309526101911e-06, "loss": 12.81964111328125, "loss_alignment": 1.02685546875, "loss_alignment_w": 0.513427734375, "loss_sft": 0.2876167632639408, "loss_total_v6": 0.8012275919318199, "step": 881 }, { "epoch": 0.20341621621621622, "grad_norm": 25.661598205566406, "learning_rate": 9.812549403697919e-06, "loss": 13.0083646774292, "loss_alignment": 1.02490234375, "loss_alignment_w": 0.512451171875, "loss_sft": 0.30162446945905685, "loss_total_v6": 0.8130227774381638, "step": 882 }, { "epoch": 0.20364684684684684, "grad_norm": 25.570436477661133, "learning_rate": 9.812002765673934e-06, "loss": 12.835664749145508, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.28480353206396103, "loss_total_v6": 0.802229069173336, "step": 883 }, { "epoch": 0.20387747747747748, "grad_norm": 28.65538215637207, "learning_rate": 9.811455347035827e-06, "loss": 12.96902084350586, "loss_alignment": 1.01708984375, "loss_alignment_w": 0.508544921875, "loss_sft": 0.3033616431057453, "loss_total_v6": 0.8105638027191162, "step": 884 }, { "epoch": 0.20410810810810812, "grad_norm": 46.240543365478516, "learning_rate": 9.810907147872407e-06, "loss": 13.235782623291016, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.31705888733267784, "loss_total_v6": 0.8272364661097527, "step": 885 }, { "epoch": 0.20433873873873873, "grad_norm": 26.752864837646484, "learning_rate": 9.810358168272597e-06, "loss": 12.880470275878906, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2958283517509699, "loss_total_v6": 0.8050293922424316, "step": 886 }, { "epoch": 0.20456936936936937, "grad_norm": 26.158546447753906, "learning_rate": 9.80980840832546e-06, "loss": 12.826783180236816, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2886429764330387, "loss_total_v6": 0.801673948764801, "step": 887 }, { "epoch": 0.2048, "grad_norm": 26.65227699279785, "learning_rate": 9.809257868120176e-06, "loss": 12.78530216217041, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.27877191081643105, "loss_total_v6": 0.79908137768507, "step": 888 }, { "epoch": 0.20503063063063062, "grad_norm": 29.08559799194336, "learning_rate": 9.808706547746057e-06, "loss": 13.077878952026367, "loss_alignment": 1.02783203125, "loss_alignment_w": 0.513916015625, "loss_sft": 0.3023680280894041, "loss_total_v6": 0.817367397248745, "step": 889 }, { "epoch": 0.20526126126126126, "grad_norm": 25.747751235961914, "learning_rate": 9.808154447292539e-06, "loss": 12.335296630859375, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2655544299632311, "loss_total_v6": 0.7709560543298721, "step": 890 }, { "epoch": 0.2054918918918919, "grad_norm": 25.002765655517578, "learning_rate": 9.807601566849183e-06, "loss": 13.226334571838379, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.3107157461345196, "loss_total_v6": 0.8266459181904793, "step": 891 }, { "epoch": 0.20572252252252252, "grad_norm": 22.918474197387695, "learning_rate": 9.807047906505683e-06, "loss": 12.17115592956543, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.25935452431440353, "loss_total_v6": 0.7606972828507423, "step": 892 }, { "epoch": 0.20595315315315316, "grad_norm": 29.61777687072754, "learning_rate": 9.806493466351849e-06, "loss": 12.70079517364502, "loss_alignment": 1.02392578125, "loss_alignment_w": 0.511962890625, "loss_sft": 0.2827675621956587, "loss_total_v6": 0.7937996834516525, "step": 893 }, { "epoch": 0.2061837837837838, "grad_norm": 26.866430282592773, "learning_rate": 9.805938246477627e-06, "loss": 12.541398048400879, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.2738580796867609, "loss_total_v6": 0.783837340772152, "step": 894 }, { "epoch": 0.2064144144144144, "grad_norm": 22.740840911865234, "learning_rate": 9.805382246973088e-06, "loss": 12.952519416809082, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2898029088973999, "loss_total_v6": 0.80953249335289, "step": 895 }, { "epoch": 0.20664504504504505, "grad_norm": 25.405670166015625, "learning_rate": 9.804825467928423e-06, "loss": 13.036293983459473, "loss_alignment": 1.02392578125, "loss_alignment_w": 0.511962890625, "loss_sft": 0.30304959416389465, "loss_total_v6": 0.8147683590650558, "step": 896 }, { "epoch": 0.20687567567567566, "grad_norm": 27.97117805480957, "learning_rate": 9.804267909433956e-06, "loss": 12.50732421875, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2683105580508709, "loss_total_v6": 0.7817077785730362, "step": 897 }, { "epoch": 0.2071063063063063, "grad_norm": 22.751007080078125, "learning_rate": 9.803709571580135e-06, "loss": 12.699535369873047, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.27608182840049267, "loss_total_v6": 0.7937209755182266, "step": 898 }, { "epoch": 0.20733693693693694, "grad_norm": 25.410850524902344, "learning_rate": 9.803150454457538e-06, "loss": 13.090825080871582, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.3067782577127218, "loss_total_v6": 0.8181765675544739, "step": 899 }, { "epoch": 0.20756756756756756, "grad_norm": 24.274200439453125, "learning_rate": 9.802590558156863e-06, "loss": 12.604161262512207, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.272547060623765, "loss_total_v6": 0.7877600491046906, "step": 900 }, { "epoch": 0.2077981981981982, "grad_norm": 26.605676651000977, "learning_rate": 9.802029882768938e-06, "loss": 13.139263153076172, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.30567050725221634, "loss_total_v6": 0.8212039545178413, "step": 901 }, { "epoch": 0.20802882882882884, "grad_norm": 25.69849967956543, "learning_rate": 9.801468428384716e-06, "loss": 12.693059921264648, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2855342626571655, "loss_total_v6": 0.7933162599802017, "step": 902 }, { "epoch": 0.20825945945945945, "grad_norm": 26.741180419921875, "learning_rate": 9.80090619509528e-06, "loss": 13.10641098022461, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.3061654753983021, "loss_total_v6": 0.8191507086157799, "step": 903 }, { "epoch": 0.2084900900900901, "grad_norm": 25.728349685668945, "learning_rate": 9.800343182991837e-06, "loss": 12.887385368347168, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.28608292154967785, "loss_total_v6": 0.8054616004228592, "step": 904 }, { "epoch": 0.20872072072072073, "grad_norm": 26.33112144470215, "learning_rate": 9.799779392165716e-06, "loss": 13.250600814819336, "loss_alignment": 1.02197265625, "loss_alignment_w": 0.510986328125, "loss_sft": 0.31790865026414394, "loss_total_v6": 0.8281625434756279, "step": 905 }, { "epoch": 0.20895135135135134, "grad_norm": 23.70078468322754, "learning_rate": 9.799214822708381e-06, "loss": 12.835328102111816, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2878037113696337, "loss_total_v6": 0.8022080287337303, "step": 906 }, { "epoch": 0.20918198198198198, "grad_norm": 24.036422729492188, "learning_rate": 9.798649474711415e-06, "loss": 12.725160598754883, "loss_alignment": 1.01416015625, "loss_alignment_w": 0.507080078125, "loss_sft": 0.28767793625593185, "loss_total_v6": 0.7953225746750832, "step": 907 }, { "epoch": 0.20941261261261263, "grad_norm": 24.85350227355957, "learning_rate": 9.798083348266531e-06, "loss": 12.457895278930664, "loss_alignment": 1.01318359375, "loss_alignment_w": 0.506591796875, "loss_sft": 0.27207241766154766, "loss_total_v6": 0.7786184400320053, "step": 908 }, { "epoch": 0.20964324324324324, "grad_norm": 29.387653350830078, "learning_rate": 9.797516443465566e-06, "loss": 13.06512451171875, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.3054009098559618, "loss_total_v6": 0.8165703043341637, "step": 909 }, { "epoch": 0.20987387387387388, "grad_norm": 23.262348175048828, "learning_rate": 9.796948760400488e-06, "loss": 12.507550239562988, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.26566960848867893, "loss_total_v6": 0.7817218750715256, "step": 910 }, { "epoch": 0.2101045045045045, "grad_norm": 25.555500030517578, "learning_rate": 9.796380299163386e-06, "loss": 12.987896919250488, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.3020542673766613, "loss_total_v6": 0.8117435872554779, "step": 911 }, { "epoch": 0.21033513513513513, "grad_norm": 25.63285255432129, "learning_rate": 9.795811059846476e-06, "loss": 13.0618314743042, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.30055634677410126, "loss_total_v6": 0.8163644522428513, "step": 912 }, { "epoch": 0.21056576576576577, "grad_norm": 24.98735237121582, "learning_rate": 9.7952410425421e-06, "loss": 12.984095573425293, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.302335437387228, "loss_total_v6": 0.8115059956908226, "step": 913 }, { "epoch": 0.21079639639639638, "grad_norm": 24.538742065429688, "learning_rate": 9.794670247342732e-06, "loss": 12.54768180847168, "loss_alignment": 1.01611328125, "loss_alignment_w": 0.508056640625, "loss_sft": 0.27759247832000256, "loss_total_v6": 0.7842300683259964, "step": 914 }, { "epoch": 0.21102702702702703, "grad_norm": 25.60331153869629, "learning_rate": 9.794098674340966e-06, "loss": 13.194136619567871, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.3069180827587843, "loss_total_v6": 0.8246335238218307, "step": 915 }, { "epoch": 0.21125765765765767, "grad_norm": 24.970378875732422, "learning_rate": 9.793526323629522e-06, "loss": 13.00385856628418, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.3073396123945713, "loss_total_v6": 0.8127411901950836, "step": 916 }, { "epoch": 0.21148828828828828, "grad_norm": 24.50788688659668, "learning_rate": 9.79295319530125e-06, "loss": 12.470345497131348, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.277885977178812, "loss_total_v6": 0.7793965935707092, "step": 917 }, { "epoch": 0.21171891891891892, "grad_norm": 25.59886360168457, "learning_rate": 9.792379289449122e-06, "loss": 13.099166870117188, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.30282881669700146, "loss_total_v6": 0.818697951734066, "step": 918 }, { "epoch": 0.21194954954954956, "grad_norm": 25.802169799804688, "learning_rate": 9.791804606166241e-06, "loss": 12.962381362915039, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.2906022109091282, "loss_total_v6": 0.8101487755775452, "step": 919 }, { "epoch": 0.21218018018018017, "grad_norm": 25.047855377197266, "learning_rate": 9.791229145545832e-06, "loss": 12.91757583618164, "loss_alignment": 1.02783203125, "loss_alignment_w": 0.513916015625, "loss_sft": 0.2945005316287279, "loss_total_v6": 0.8073484525084496, "step": 920 }, { "epoch": 0.2124108108108108, "grad_norm": 24.10572624206543, "learning_rate": 9.790652907681248e-06, "loss": 12.327584266662598, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2605252843350172, "loss_total_v6": 0.7704740166664124, "step": 921 }, { "epoch": 0.21264144144144145, "grad_norm": 23.06370735168457, "learning_rate": 9.790075892665966e-06, "loss": 12.740816116333008, "loss_alignment": 1.02880859375, "loss_alignment_w": 0.514404296875, "loss_sft": 0.28269019164144993, "loss_total_v6": 0.7963010221719742, "step": 922 }, { "epoch": 0.21287207207207207, "grad_norm": 26.505701065063477, "learning_rate": 9.789498100593593e-06, "loss": 12.729235649108887, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.2772361822426319, "loss_total_v6": 0.795577235519886, "step": 923 }, { "epoch": 0.2131027027027027, "grad_norm": 28.232824325561523, "learning_rate": 9.78891953155786e-06, "loss": 12.85052490234375, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.29485699348151684, "loss_total_v6": 0.8031578063964844, "step": 924 }, { "epoch": 0.21333333333333335, "grad_norm": 26.27642059326172, "learning_rate": 9.78834018565262e-06, "loss": 13.496548652648926, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.3207376301288605, "loss_total_v6": 0.8435342684388161, "step": 925 }, { "epoch": 0.21356396396396396, "grad_norm": 26.452829360961914, "learning_rate": 9.78776006297186e-06, "loss": 12.517385482788086, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.26858842372894287, "loss_total_v6": 0.7823366150259972, "step": 926 }, { "epoch": 0.2137945945945946, "grad_norm": 23.65169906616211, "learning_rate": 9.787179163609688e-06, "loss": 12.514568328857422, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.288828594610095, "loss_total_v6": 0.7821604982018471, "step": 927 }, { "epoch": 0.2140252252252252, "grad_norm": 23.230907440185547, "learning_rate": 9.786597487660336e-06, "loss": 12.917976379394531, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.29243518598377705, "loss_total_v6": 0.8073735162615776, "step": 928 }, { "epoch": 0.21425585585585585, "grad_norm": 25.51637077331543, "learning_rate": 9.78601503521817e-06, "loss": 12.609697341918945, "loss_alignment": 1.01611328125, "loss_alignment_w": 0.508056640625, "loss_sft": 0.28201786056160927, "loss_total_v6": 0.7881061062216759, "step": 929 }, { "epoch": 0.2144864864864865, "grad_norm": 23.575599670410156, "learning_rate": 9.785431806377672e-06, "loss": 12.793508529663086, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2859835112467408, "loss_total_v6": 0.7995943278074265, "step": 930 }, { "epoch": 0.2147171171171171, "grad_norm": 24.558460235595703, "learning_rate": 9.784847801233454e-06, "loss": 12.567123413085938, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.2816915325820446, "loss_total_v6": 0.7854452058672905, "step": 931 }, { "epoch": 0.21494774774774775, "grad_norm": 24.908252716064453, "learning_rate": 9.78426301988026e-06, "loss": 12.439376831054688, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2649183087050915, "loss_total_v6": 0.7774610072374344, "step": 932 }, { "epoch": 0.2151783783783784, "grad_norm": 26.093994140625, "learning_rate": 9.783677462412951e-06, "loss": 12.974735260009766, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2946702819317579, "loss_total_v6": 0.8109209090471268, "step": 933 }, { "epoch": 0.215409009009009, "grad_norm": 22.521886825561523, "learning_rate": 9.783091128926519e-06, "loss": 12.530837059020996, "loss_alignment": 1.02294921875, "loss_alignment_w": 0.511474609375, "loss_sft": 0.27106184139847755, "loss_total_v6": 0.7831773310899734, "step": 934 }, { "epoch": 0.21563963963963964, "grad_norm": 24.243099212646484, "learning_rate": 9.782504019516079e-06, "loss": 12.932093620300293, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.2895027585327625, "loss_total_v6": 0.8082558289170265, "step": 935 }, { "epoch": 0.21587027027027028, "grad_norm": 26.05591583251953, "learning_rate": 9.781916134276873e-06, "loss": 12.934974670410156, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.2917885687202215, "loss_total_v6": 0.8084359094500542, "step": 936 }, { "epoch": 0.2161009009009009, "grad_norm": 24.463565826416016, "learning_rate": 9.78132747330427e-06, "loss": 12.753751754760742, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2875575013458729, "loss_total_v6": 0.7971095219254494, "step": 937 }, { "epoch": 0.21633153153153153, "grad_norm": 21.89810562133789, "learning_rate": 9.780738036693764e-06, "loss": 12.864544868469238, "loss_alignment": 1.01708984375, "loss_alignment_w": 0.508544921875, "loss_sft": 0.29532129131257534, "loss_total_v6": 0.8040340542793274, "step": 938 }, { "epoch": 0.21656216216216218, "grad_norm": 24.23625373840332, "learning_rate": 9.780147824540973e-06, "loss": 12.298789978027344, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2573218047618866, "loss_total_v6": 0.7686743438243866, "step": 939 }, { "epoch": 0.2167927927927928, "grad_norm": 24.639909744262695, "learning_rate": 9.779556836941646e-06, "loss": 13.22480297088623, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.3173643462359905, "loss_total_v6": 0.8265501782298088, "step": 940 }, { "epoch": 0.21702342342342343, "grad_norm": 27.98533821105957, "learning_rate": 9.778965073991652e-06, "loss": 12.305400848388672, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.26847718097269535, "loss_total_v6": 0.7690875381231308, "step": 941 }, { "epoch": 0.21725405405405404, "grad_norm": 23.402040481567383, "learning_rate": 9.778372535786987e-06, "loss": 12.680309295654297, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.28128885850310326, "loss_total_v6": 0.7925193086266518, "step": 942 }, { "epoch": 0.21748468468468468, "grad_norm": 24.548503875732422, "learning_rate": 9.777779222423778e-06, "loss": 12.972795486450195, "loss_alignment": 1.02783203125, "loss_alignment_w": 0.513916015625, "loss_sft": 0.2959986738860607, "loss_total_v6": 0.8107997179031372, "step": 943 }, { "epoch": 0.21771531531531532, "grad_norm": 25.160917282104492, "learning_rate": 9.777185133998268e-06, "loss": 13.322732925415039, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.31461965665221214, "loss_total_v6": 0.8326708227396011, "step": 944 }, { "epoch": 0.21794594594594593, "grad_norm": 25.41672134399414, "learning_rate": 9.776590270606837e-06, "loss": 12.837905883789062, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.28463839925825596, "loss_total_v6": 0.8023691475391388, "step": 945 }, { "epoch": 0.21817657657657658, "grad_norm": 23.704256057739258, "learning_rate": 9.775994632345982e-06, "loss": 12.833993911743164, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.29608217254281044, "loss_total_v6": 0.8021246492862701, "step": 946 }, { "epoch": 0.21840720720720722, "grad_norm": 22.918384552001953, "learning_rate": 9.775398219312328e-06, "loss": 13.226282119750977, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.3047310281544924, "loss_total_v6": 0.8266426473855972, "step": 947 }, { "epoch": 0.21863783783783783, "grad_norm": 23.541645050048828, "learning_rate": 9.77480103160263e-06, "loss": 12.717704772949219, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2809558380395174, "loss_total_v6": 0.794856570661068, "step": 948 }, { "epoch": 0.21886846846846847, "grad_norm": 25.39034652709961, "learning_rate": 9.774203069313763e-06, "loss": 12.945231437683105, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2939555197954178, "loss_total_v6": 0.8090770021080971, "step": 949 }, { "epoch": 0.2190990990990991, "grad_norm": 25.69516372680664, "learning_rate": 9.77360433254273e-06, "loss": 12.376961708068848, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2593236658722162, "loss_total_v6": 0.7735600993037224, "step": 950 }, { "epoch": 0.21932972972972972, "grad_norm": 26.478105545043945, "learning_rate": 9.773004821386659e-06, "loss": 12.303717613220215, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.26826515048742294, "loss_total_v6": 0.7689823433756828, "step": 951 }, { "epoch": 0.21956036036036036, "grad_norm": 27.707563400268555, "learning_rate": 9.772404535942802e-06, "loss": 12.095935821533203, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.253356259316206, "loss_total_v6": 0.7559960037469864, "step": 952 }, { "epoch": 0.219790990990991, "grad_norm": 24.728647232055664, "learning_rate": 9.771803476308544e-06, "loss": 12.94943904876709, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.29028166085481644, "loss_total_v6": 0.8093399330973625, "step": 953 }, { "epoch": 0.22002162162162162, "grad_norm": 25.09752082824707, "learning_rate": 9.771201642581384e-06, "loss": 12.812981605529785, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.28272970020771027, "loss_total_v6": 0.8008113503456116, "step": 954 }, { "epoch": 0.22025225225225226, "grad_norm": 24.87118148803711, "learning_rate": 9.77059903485896e-06, "loss": 12.99169921875, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.3051147162914276, "loss_total_v6": 0.8119811564683914, "step": 955 }, { "epoch": 0.22048288288288287, "grad_norm": 27.820619583129883, "learning_rate": 9.769995653239023e-06, "loss": 13.341773986816406, "loss_alignment": 1.02880859375, "loss_alignment_w": 0.514404296875, "loss_sft": 0.31934974156320095, "loss_total_v6": 0.8338608741760254, "step": 956 }, { "epoch": 0.2207135135135135, "grad_norm": 24.487655639648438, "learning_rate": 9.769391497819456e-06, "loss": 13.224498748779297, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.31913067027926445, "loss_total_v6": 0.8265311792492867, "step": 957 }, { "epoch": 0.22094414414414415, "grad_norm": 26.253530502319336, "learning_rate": 9.768786568698267e-06, "loss": 12.437110900878906, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.26468510553240776, "loss_total_v6": 0.7773194089531898, "step": 958 }, { "epoch": 0.22117477477477476, "grad_norm": 26.22098731994629, "learning_rate": 9.768180865973591e-06, "loss": 13.077695846557617, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.29265195690095425, "loss_total_v6": 0.8173559382557869, "step": 959 }, { "epoch": 0.2214054054054054, "grad_norm": 23.68361473083496, "learning_rate": 9.767574389743683e-06, "loss": 12.99252986907959, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.2955230697989464, "loss_total_v6": 0.8120331093668938, "step": 960 }, { "epoch": 0.22163603603603604, "grad_norm": 26.636272430419922, "learning_rate": 9.766967140106928e-06, "loss": 13.060405731201172, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.30432770401239395, "loss_total_v6": 0.8162753507494926, "step": 961 }, { "epoch": 0.22186666666666666, "grad_norm": 24.54336929321289, "learning_rate": 9.766359117161836e-06, "loss": 12.336923599243164, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.26510685309767723, "loss_total_v6": 0.7710577547550201, "step": 962 }, { "epoch": 0.2220972972972973, "grad_norm": 26.843374252319336, "learning_rate": 9.765750321007042e-06, "loss": 13.003762245178223, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.3021455369889736, "loss_total_v6": 0.8127351328730583, "step": 963 }, { "epoch": 0.22232792792792794, "grad_norm": 24.136333465576172, "learning_rate": 9.765140751741307e-06, "loss": 12.630385398864746, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2838601525872946, "loss_total_v6": 0.7893990650773048, "step": 964 }, { "epoch": 0.22255855855855855, "grad_norm": 25.54161262512207, "learning_rate": 9.764530409463516e-06, "loss": 13.00772762298584, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.29395526088774204, "loss_total_v6": 0.8129829615354538, "step": 965 }, { "epoch": 0.2227891891891892, "grad_norm": 23.711448669433594, "learning_rate": 9.763919294272678e-06, "loss": 12.692632675170898, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2780154272913933, "loss_total_v6": 0.793289490044117, "step": 966 }, { "epoch": 0.22301981981981983, "grad_norm": 28.6368350982666, "learning_rate": 9.763307406267933e-06, "loss": 13.045052528381348, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.30411578342318535, "loss_total_v6": 0.8153157457709312, "step": 967 }, { "epoch": 0.22325045045045044, "grad_norm": 24.12069320678711, "learning_rate": 9.76269474554854e-06, "loss": 12.569719314575195, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.2676478289067745, "loss_total_v6": 0.7856074199080467, "step": 968 }, { "epoch": 0.22348108108108108, "grad_norm": 25.277986526489258, "learning_rate": 9.762081312213889e-06, "loss": 12.648415565490723, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.27650310657918453, "loss_total_v6": 0.7905259504914284, "step": 969 }, { "epoch": 0.22371171171171172, "grad_norm": 25.99574089050293, "learning_rate": 9.761467106363488e-06, "loss": 12.435340881347656, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.26657344214618206, "loss_total_v6": 0.7772088199853897, "step": 970 }, { "epoch": 0.22394234234234234, "grad_norm": 22.75480079650879, "learning_rate": 9.760852128096981e-06, "loss": 12.75729751586914, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2842543516308069, "loss_total_v6": 0.7973311245441437, "step": 971 }, { "epoch": 0.22417297297297298, "grad_norm": 24.195377349853516, "learning_rate": 9.760236377514128e-06, "loss": 12.673910140991211, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2756399158388376, "loss_total_v6": 0.7921194136142731, "step": 972 }, { "epoch": 0.2244036036036036, "grad_norm": 26.131534576416016, "learning_rate": 9.759619854714817e-06, "loss": 12.629568099975586, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.27958243526518345, "loss_total_v6": 0.7893480584025383, "step": 973 }, { "epoch": 0.22463423423423423, "grad_norm": 22.69232940673828, "learning_rate": 9.759002559799062e-06, "loss": 12.923133850097656, "loss_alignment": 1.01416015625, "loss_alignment_w": 0.507080078125, "loss_sft": 0.30014270916581154, "loss_total_v6": 0.8076958134770393, "step": 974 }, { "epoch": 0.22486486486486487, "grad_norm": 23.365894317626953, "learning_rate": 9.758384492867004e-06, "loss": 12.588360786437988, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.2812030892819166, "loss_total_v6": 0.7867725491523743, "step": 975 }, { "epoch": 0.22509549549549548, "grad_norm": 24.818815231323242, "learning_rate": 9.757765654018905e-06, "loss": 12.906594276428223, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.29758312553167343, "loss_total_v6": 0.8066621273756027, "step": 976 }, { "epoch": 0.22532612612612613, "grad_norm": 20.804399490356445, "learning_rate": 9.757146043355154e-06, "loss": 12.665998458862305, "loss_alignment": 1.0595703125, "loss_alignment_w": 0.52978515625, "loss_sft": 0.26289258524775505, "loss_total_v6": 0.7916248813271523, "step": 977 }, { "epoch": 0.22555675675675677, "grad_norm": 22.535703659057617, "learning_rate": 9.756525660976266e-06, "loss": 12.550384521484375, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.28172871097922325, "loss_total_v6": 0.784399002790451, "step": 978 }, { "epoch": 0.22578738738738738, "grad_norm": 25.18543243408203, "learning_rate": 9.755904506982884e-06, "loss": 12.362136840820312, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2609148435294628, "loss_total_v6": 0.7726335749030113, "step": 979 }, { "epoch": 0.22601801801801802, "grad_norm": 24.872398376464844, "learning_rate": 9.755282581475769e-06, "loss": 13.106431007385254, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.2973776161670685, "loss_total_v6": 0.8191519305109978, "step": 980 }, { "epoch": 0.22624864864864866, "grad_norm": 23.583662033081055, "learning_rate": 9.754659884555813e-06, "loss": 12.687078475952148, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2840160597115755, "loss_total_v6": 0.7929424121975899, "step": 981 }, { "epoch": 0.22647927927927927, "grad_norm": 25.759902954101562, "learning_rate": 9.75403641632403e-06, "loss": 12.918306350708008, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.2838956192135811, "loss_total_v6": 0.8073941469192505, "step": 982 }, { "epoch": 0.2267099099099099, "grad_norm": 24.90462303161621, "learning_rate": 9.753412176881564e-06, "loss": 12.682065963745117, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2806509658694267, "loss_total_v6": 0.7926291152834892, "step": 983 }, { "epoch": 0.22694054054054055, "grad_norm": 28.425098419189453, "learning_rate": 9.752787166329677e-06, "loss": 12.718433380126953, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.2748520355671644, "loss_total_v6": 0.7949020937085152, "step": 984 }, { "epoch": 0.22717117117117117, "grad_norm": 24.729251861572266, "learning_rate": 9.75216138476976e-06, "loss": 13.054716110229492, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.3005388379096985, "loss_total_v6": 0.8159197121858597, "step": 985 }, { "epoch": 0.2274018018018018, "grad_norm": 26.23525619506836, "learning_rate": 9.751534832303332e-06, "loss": 12.818678855895996, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.28418438136577606, "loss_total_v6": 0.8011674284934998, "step": 986 }, { "epoch": 0.22763243243243242, "grad_norm": 26.228769302368164, "learning_rate": 9.750907509032028e-06, "loss": 13.14309024810791, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.30903778225183487, "loss_total_v6": 0.821443147957325, "step": 987 }, { "epoch": 0.22786306306306306, "grad_norm": 24.840171813964844, "learning_rate": 9.750279415057618e-06, "loss": 12.915996551513672, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.30154304578900337, "loss_total_v6": 0.8072498217225075, "step": 988 }, { "epoch": 0.2280936936936937, "grad_norm": 23.328935623168945, "learning_rate": 9.749650550481994e-06, "loss": 12.88155746459961, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.28985381685197353, "loss_total_v6": 0.8050973564386368, "step": 989 }, { "epoch": 0.2283243243243243, "grad_norm": 25.63646125793457, "learning_rate": 9.749020915407165e-06, "loss": 12.94575023651123, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.3000304289162159, "loss_total_v6": 0.8091094046831131, "step": 990 }, { "epoch": 0.22855495495495495, "grad_norm": 23.223812103271484, "learning_rate": 9.748390509935282e-06, "loss": 13.029241561889648, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.2951168268918991, "loss_total_v6": 0.8143276199698448, "step": 991 }, { "epoch": 0.2287855855855856, "grad_norm": 27.37331771850586, "learning_rate": 9.747759334168602e-06, "loss": 12.783979415893555, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.29919707402586937, "loss_total_v6": 0.7989987209439278, "step": 992 }, { "epoch": 0.2290162162162162, "grad_norm": 28.05451202392578, "learning_rate": 9.74712738820952e-06, "loss": 12.765542984008789, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.29821265302598476, "loss_total_v6": 0.7978463992476463, "step": 993 }, { "epoch": 0.22924684684684685, "grad_norm": 23.579980850219727, "learning_rate": 9.746494672160552e-06, "loss": 12.270483016967773, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2557052820920944, "loss_total_v6": 0.7669052183628082, "step": 994 }, { "epoch": 0.2294774774774775, "grad_norm": 27.910303115844727, "learning_rate": 9.745861186124336e-06, "loss": 13.367338180541992, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.3210085555911064, "loss_total_v6": 0.8354586213827133, "step": 995 }, { "epoch": 0.2297081081081081, "grad_norm": 24.275423049926758, "learning_rate": 9.74522693020364e-06, "loss": 12.715767860412598, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.2756772320717573, "loss_total_v6": 0.7947354912757874, "step": 996 }, { "epoch": 0.22993873873873874, "grad_norm": 28.27248191833496, "learning_rate": 9.744591904501353e-06, "loss": 13.438344955444336, "loss_alignment": 1.046875, "loss_alignment_w": 0.5234375, "loss_sft": 0.3164285682141781, "loss_total_v6": 0.8398965895175934, "step": 997 }, { "epoch": 0.23016936936936938, "grad_norm": 24.634275436401367, "learning_rate": 9.743956109120491e-06, "loss": 13.570280075073242, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.3321054857224226, "loss_total_v6": 0.8481424748897552, "step": 998 }, { "epoch": 0.2304, "grad_norm": 27.007423400878906, "learning_rate": 9.743319544164195e-06, "loss": 13.20191764831543, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.3102426193654537, "loss_total_v6": 0.8251199126243591, "step": 999 }, { "epoch": 0.23063063063063063, "grad_norm": 25.101545333862305, "learning_rate": 9.742682209735727e-06, "loss": 12.472810745239258, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.27973377890884876, "loss_total_v6": 0.7795506343245506, "step": 1000 }, { "epoch": 0.23063063063063063, "eval_loss": 0.923358678817749, "eval_loss_alignment": 1.0293615332477168, "eval_loss_alignment_w": 0.5146807666238584, "eval_loss_contrast": 0.0, "eval_loss_highlevel": 0.0, "eval_loss_sft": 0.4086779152812887, "eval_loss_total_v6": 0.9233586843631584, "eval_loss_wm_recon": 0.0, "eval_loss_wm_recon_w": 0.0, "eval_runtime": 222.2938, "eval_samples_per_second": 7.85, "eval_steps_per_second": 0.985, "step": 1000 }, { "epoch": 0.23086126126126127, "grad_norm": 24.51674461364746, "learning_rate": 9.742044105938481e-06, "loss": 12.692214965820312, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.29005912505090237, "loss_total_v6": 0.7932634577155113, "step": 1001 }, { "epoch": 0.2310918918918919, "grad_norm": 23.823518753051758, "learning_rate": 9.74140523287597e-06, "loss": 13.117696762084961, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.3052533380687237, "loss_total_v6": 0.8198559954762459, "step": 1002 }, { "epoch": 0.23132252252252253, "grad_norm": 27.40005874633789, "learning_rate": 9.740765590651833e-06, "loss": 12.97912311553955, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.3001173511147499, "loss_total_v6": 0.8111952394247055, "step": 1003 }, { "epoch": 0.23155315315315314, "grad_norm": 24.511945724487305, "learning_rate": 9.740125179369833e-06, "loss": 12.696215629577637, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2809860222041607, "loss_total_v6": 0.7935134842991829, "step": 1004 }, { "epoch": 0.23178378378378378, "grad_norm": 22.943281173706055, "learning_rate": 9.739483999133861e-06, "loss": 11.883405685424805, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.2473972849547863, "loss_total_v6": 0.7427128329873085, "step": 1005 }, { "epoch": 0.23201441441441442, "grad_norm": 24.177043914794922, "learning_rate": 9.73884205004793e-06, "loss": 12.465997695922852, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.279964167624712, "loss_total_v6": 0.7791248932480812, "step": 1006 }, { "epoch": 0.23224504504504503, "grad_norm": 32.84754943847656, "learning_rate": 9.738199332216178e-06, "loss": 12.726789474487305, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.27917380444705486, "loss_total_v6": 0.7954243719577789, "step": 1007 }, { "epoch": 0.23247567567567567, "grad_norm": 23.900157928466797, "learning_rate": 9.737555845742869e-06, "loss": 13.047876358032227, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.3013626392930746, "loss_total_v6": 0.8154922872781754, "step": 1008 }, { "epoch": 0.23270630630630632, "grad_norm": 26.650104522705078, "learning_rate": 9.73691159073239e-06, "loss": 12.436132431030273, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.27048336900770664, "loss_total_v6": 0.7772582471370697, "step": 1009 }, { "epoch": 0.23293693693693693, "grad_norm": 25.345184326171875, "learning_rate": 9.736266567289255e-06, "loss": 12.759965896606445, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.2920199129730463, "loss_total_v6": 0.797497846186161, "step": 1010 }, { "epoch": 0.23316756756756757, "grad_norm": 26.66474723815918, "learning_rate": 9.735620775518098e-06, "loss": 12.760942459106445, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2867403905838728, "loss_total_v6": 0.7975588664412498, "step": 1011 }, { "epoch": 0.2333981981981982, "grad_norm": 26.187828063964844, "learning_rate": 9.734974215523684e-06, "loss": 12.447161674499512, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.2767116613686085, "loss_total_v6": 0.7779475897550583, "step": 1012 }, { "epoch": 0.23362882882882882, "grad_norm": 28.650806427001953, "learning_rate": 9.734326887410902e-06, "loss": 13.264442443847656, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.311617411673069, "loss_total_v6": 0.8290276974439621, "step": 1013 }, { "epoch": 0.23385945945945946, "grad_norm": 24.767412185668945, "learning_rate": 9.733678791284755e-06, "loss": 12.67329216003418, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2840088624507189, "loss_total_v6": 0.792080782353878, "step": 1014 }, { "epoch": 0.2340900900900901, "grad_norm": 24.6458740234375, "learning_rate": 9.733029927250385e-06, "loss": 12.748557090759277, "loss_alignment": 1.01513671875, "loss_alignment_w": 0.507568359375, "loss_sft": 0.28605788573622704, "loss_total_v6": 0.7967848107218742, "step": 1015 }, { "epoch": 0.23432072072072072, "grad_norm": 24.60102081298828, "learning_rate": 9.73238029541305e-06, "loss": 12.822996139526367, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2839507218450308, "loss_total_v6": 0.8014372810721397, "step": 1016 }, { "epoch": 0.23455135135135136, "grad_norm": 26.33922576904297, "learning_rate": 9.731729895878135e-06, "loss": 12.77472972869873, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.281178193166852, "loss_total_v6": 0.7984206303954124, "step": 1017 }, { "epoch": 0.23478198198198197, "grad_norm": 25.813053131103516, "learning_rate": 9.73107872875115e-06, "loss": 12.776124954223633, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.28091441467404366, "loss_total_v6": 0.7985077872872353, "step": 1018 }, { "epoch": 0.2350126126126126, "grad_norm": 25.43490982055664, "learning_rate": 9.730426794137727e-06, "loss": 12.888826370239258, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.3053838051855564, "loss_total_v6": 0.8055516928434372, "step": 1019 }, { "epoch": 0.23524324324324325, "grad_norm": 25.20985221862793, "learning_rate": 9.729774092143627e-06, "loss": 12.744729995727539, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2876802496612072, "loss_total_v6": 0.79654560983181, "step": 1020 }, { "epoch": 0.23547387387387386, "grad_norm": 23.436992645263672, "learning_rate": 9.72912062287473e-06, "loss": 12.838212966918945, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.29012022353708744, "loss_total_v6": 0.8023882731795311, "step": 1021 }, { "epoch": 0.2357045045045045, "grad_norm": 23.871240615844727, "learning_rate": 9.728466386437046e-06, "loss": 13.079828262329102, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.31135515682399273, "loss_total_v6": 0.8174892291426659, "step": 1022 }, { "epoch": 0.23593513513513514, "grad_norm": 24.47878646850586, "learning_rate": 9.727811382936704e-06, "loss": 12.772592544555664, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2934652790427208, "loss_total_v6": 0.7982870414853096, "step": 1023 }, { "epoch": 0.23616576576576576, "grad_norm": 23.78693962097168, "learning_rate": 9.727155612479964e-06, "loss": 12.201717376708984, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.25136165507137775, "loss_total_v6": 0.7626073807477951, "step": 1024 }, { "epoch": 0.2363963963963964, "grad_norm": 22.445899963378906, "learning_rate": 9.726499075173201e-06, "loss": 12.6097993850708, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.2670248467475176, "loss_total_v6": 0.7881124839186668, "step": 1025 }, { "epoch": 0.23662702702702704, "grad_norm": 25.390214920043945, "learning_rate": 9.725841771122927e-06, "loss": 13.384756088256836, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.32362309470772743, "loss_total_v6": 0.8365472927689552, "step": 1026 }, { "epoch": 0.23685765765765765, "grad_norm": 24.34918212890625, "learning_rate": 9.725183700435764e-06, "loss": 12.514373779296875, "loss_alignment": 1.00732421875, "loss_alignment_w": 0.503662109375, "loss_sft": 0.27924923226237297, "loss_total_v6": 0.7821483761072159, "step": 1027 }, { "epoch": 0.2370882882882883, "grad_norm": 25.86640739440918, "learning_rate": 9.72452486321847e-06, "loss": 13.119049072265625, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.31344037130475044, "loss_total_v6": 0.8199405893683434, "step": 1028 }, { "epoch": 0.23731891891891893, "grad_norm": 27.092586517333984, "learning_rate": 9.723865259577923e-06, "loss": 12.266823768615723, "loss_alignment": 1.01513671875, "loss_alignment_w": 0.507568359375, "loss_sft": 0.2606339640915394, "loss_total_v6": 0.7666764631867409, "step": 1029 }, { "epoch": 0.23754954954954954, "grad_norm": 24.465669631958008, "learning_rate": 9.723204889621125e-06, "loss": 12.701040267944336, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.28916111774742603, "loss_total_v6": 0.7938150763511658, "step": 1030 }, { "epoch": 0.23778018018018018, "grad_norm": 22.022802352905273, "learning_rate": 9.722543753455202e-06, "loss": 12.967504501342773, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2978195007890463, "loss_total_v6": 0.8104690462350845, "step": 1031 }, { "epoch": 0.23801081081081082, "grad_norm": 23.314695358276367, "learning_rate": 9.721881851187406e-06, "loss": 12.763378143310547, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.28352043963968754, "loss_total_v6": 0.797711119055748, "step": 1032 }, { "epoch": 0.23824144144144144, "grad_norm": 25.34224510192871, "learning_rate": 9.721219182925113e-06, "loss": 12.615077018737793, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.28154534101486206, "loss_total_v6": 0.7884423211216927, "step": 1033 }, { "epoch": 0.23847207207207208, "grad_norm": 24.416730880737305, "learning_rate": 9.72055574877582e-06, "loss": 13.000125885009766, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.30045340955257416, "loss_total_v6": 0.8125078603625298, "step": 1034 }, { "epoch": 0.2387027027027027, "grad_norm": 24.28573226928711, "learning_rate": 9.71989154884715e-06, "loss": 13.15919017791748, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.3028724044561386, "loss_total_v6": 0.8224494233727455, "step": 1035 }, { "epoch": 0.23893333333333333, "grad_norm": 24.27040672302246, "learning_rate": 9.719226583246856e-06, "loss": 12.736177444458008, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.27289403788745403, "loss_total_v6": 0.7960111051797867, "step": 1036 }, { "epoch": 0.23916396396396397, "grad_norm": 24.581897735595703, "learning_rate": 9.718560852082806e-06, "loss": 12.663301467895508, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2839642260223627, "loss_total_v6": 0.791456326842308, "step": 1037 }, { "epoch": 0.23939459459459458, "grad_norm": 22.542882919311523, "learning_rate": 9.717894355463e-06, "loss": 12.627939224243164, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.27844300866127014, "loss_total_v6": 0.7892462313175201, "step": 1038 }, { "epoch": 0.23962522522522522, "grad_norm": 24.89057159423828, "learning_rate": 9.717227093495554e-06, "loss": 13.320127487182617, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.317432276904583, "loss_total_v6": 0.8325079679489136, "step": 1039 }, { "epoch": 0.23985585585585587, "grad_norm": 24.903535842895508, "learning_rate": 9.716559066288716e-06, "loss": 13.469083786010742, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.3266352489590645, "loss_total_v6": 0.8418177589774132, "step": 1040 }, { "epoch": 0.24008648648648648, "grad_norm": 24.12944221496582, "learning_rate": 9.715890273950854e-06, "loss": 12.679156303405762, "loss_alignment": 1.02392578125, "loss_alignment_w": 0.511962890625, "loss_sft": 0.2798130474984646, "loss_total_v6": 0.7924472913146019, "step": 1041 }, { "epoch": 0.24031711711711712, "grad_norm": 22.177276611328125, "learning_rate": 9.715220716590464e-06, "loss": 13.323118209838867, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.3125838302075863, "loss_total_v6": 0.8326948955655098, "step": 1042 }, { "epoch": 0.24054774774774776, "grad_norm": 23.13252067565918, "learning_rate": 9.714550394316156e-06, "loss": 12.970134735107422, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.30364493280649185, "loss_total_v6": 0.8106334581971169, "step": 1043 }, { "epoch": 0.24077837837837837, "grad_norm": 26.20540428161621, "learning_rate": 9.713879307236678e-06, "loss": 12.858427047729492, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.29503040574491024, "loss_total_v6": 0.8036516383290291, "step": 1044 }, { "epoch": 0.241009009009009, "grad_norm": 24.126188278198242, "learning_rate": 9.713207455460893e-06, "loss": 13.08230972290039, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.29658724553883076, "loss_total_v6": 0.8176443949341774, "step": 1045 }, { "epoch": 0.24123963963963965, "grad_norm": 24.038829803466797, "learning_rate": 9.712534839097789e-06, "loss": 12.99352741241455, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.2898634523153305, "loss_total_v6": 0.8120954632759094, "step": 1046 }, { "epoch": 0.24147027027027027, "grad_norm": 24.51917266845703, "learning_rate": 9.711861458256482e-06, "loss": 13.004039764404297, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.30089638382196426, "loss_total_v6": 0.812752477824688, "step": 1047 }, { "epoch": 0.2417009009009009, "grad_norm": 24.509342193603516, "learning_rate": 9.711187313046207e-06, "loss": 12.527959823608398, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.2643359862267971, "loss_total_v6": 0.7829975038766861, "step": 1048 }, { "epoch": 0.24193153153153152, "grad_norm": 24.6774845123291, "learning_rate": 9.710512403576326e-06, "loss": 12.55313491821289, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2760412786155939, "loss_total_v6": 0.7845709323883057, "step": 1049 }, { "epoch": 0.24216216216216216, "grad_norm": 24.336191177368164, "learning_rate": 9.709836729956326e-06, "loss": 12.575183868408203, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.27972342632710934, "loss_total_v6": 0.7859490066766739, "step": 1050 }, { "epoch": 0.2423927927927928, "grad_norm": 23.28298568725586, "learning_rate": 9.709160292295814e-06, "loss": 12.64154052734375, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.27740098163485527, "loss_total_v6": 0.7900962829589844, "step": 1051 }, { "epoch": 0.2426234234234234, "grad_norm": 24.210073471069336, "learning_rate": 9.708483090704524e-06, "loss": 12.840803146362305, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2870167884975672, "loss_total_v6": 0.802550233900547, "step": 1052 }, { "epoch": 0.24285405405405405, "grad_norm": 26.36126136779785, "learning_rate": 9.707805125292313e-06, "loss": 13.028890609741211, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.3031973410397768, "loss_total_v6": 0.8143057003617287, "step": 1053 }, { "epoch": 0.2430846846846847, "grad_norm": 27.248199462890625, "learning_rate": 9.707126396169165e-06, "loss": 12.962287902832031, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.29740188270807266, "loss_total_v6": 0.8101430088281631, "step": 1054 }, { "epoch": 0.2433153153153153, "grad_norm": 23.994035720825195, "learning_rate": 9.706446903445179e-06, "loss": 12.52621078491211, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.2808587234467268, "loss_total_v6": 0.7828881293535233, "step": 1055 }, { "epoch": 0.24354594594594595, "grad_norm": 24.877910614013672, "learning_rate": 9.70576664723059e-06, "loss": 12.446990966796875, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.2790813148021698, "loss_total_v6": 0.7779368981719017, "step": 1056 }, { "epoch": 0.2437765765765766, "grad_norm": 25.302776336669922, "learning_rate": 9.705085627635746e-06, "loss": 12.98890495300293, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.295143973082304, "loss_total_v6": 0.8118065670132637, "step": 1057 }, { "epoch": 0.2440072072072072, "grad_norm": 24.73955535888672, "learning_rate": 9.704403844771128e-06, "loss": 12.828285217285156, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.29882287234067917, "loss_total_v6": 0.8017678111791611, "step": 1058 }, { "epoch": 0.24423783783783784, "grad_norm": 25.71407127380371, "learning_rate": 9.703721298747332e-06, "loss": 13.093423843383789, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.30384310707449913, "loss_total_v6": 0.818338967859745, "step": 1059 }, { "epoch": 0.24446846846846848, "grad_norm": 29.73530387878418, "learning_rate": 9.703037989675088e-06, "loss": 13.023932456970215, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.30587813444435596, "loss_total_v6": 0.8139958009123802, "step": 1060 }, { "epoch": 0.2446990990990991, "grad_norm": 24.20162010192871, "learning_rate": 9.702353917665238e-06, "loss": 12.614871978759766, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.27510853484272957, "loss_total_v6": 0.7884294614195824, "step": 1061 }, { "epoch": 0.24492972972972973, "grad_norm": 24.598854064941406, "learning_rate": 9.701669082828754e-06, "loss": 12.882652282714844, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.29329442605376244, "loss_total_v6": 0.8051657527685165, "step": 1062 }, { "epoch": 0.24516036036036035, "grad_norm": 26.067102432250977, "learning_rate": 9.700983485276735e-06, "loss": 13.161767959594727, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.3129364736378193, "loss_total_v6": 0.822610504925251, "step": 1063 }, { "epoch": 0.245390990990991, "grad_norm": 22.569910049438477, "learning_rate": 9.7002971251204e-06, "loss": 12.788557052612305, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2887715045362711, "loss_total_v6": 0.7992847934365273, "step": 1064 }, { "epoch": 0.24562162162162163, "grad_norm": 29.791698455810547, "learning_rate": 9.699610002471088e-06, "loss": 12.548663139343262, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2711536157876253, "loss_total_v6": 0.7842914536595345, "step": 1065 }, { "epoch": 0.24585225225225224, "grad_norm": 23.823829650878906, "learning_rate": 9.69892211744027e-06, "loss": 12.993852615356445, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2988712266087532, "loss_total_v6": 0.8121158331632614, "step": 1066 }, { "epoch": 0.24608288288288288, "grad_norm": 27.141674041748047, "learning_rate": 9.698233470139534e-06, "loss": 13.045393943786621, "loss_alignment": 1.01611328125, "loss_alignment_w": 0.508056640625, "loss_sft": 0.30692951940000057, "loss_total_v6": 0.8153371214866638, "step": 1067 }, { "epoch": 0.24631351351351352, "grad_norm": 26.05982780456543, "learning_rate": 9.697544060680594e-06, "loss": 12.521785736083984, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.2881200034171343, "loss_total_v6": 0.7826115936040878, "step": 1068 }, { "epoch": 0.24654414414414413, "grad_norm": 23.76301383972168, "learning_rate": 9.696853889175288e-06, "loss": 12.224626541137695, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2609873916953802, "loss_total_v6": 0.7640391066670418, "step": 1069 }, { "epoch": 0.24677477477477477, "grad_norm": 23.199520111083984, "learning_rate": 9.696162955735577e-06, "loss": 12.275738716125488, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.25656776409596205, "loss_total_v6": 0.767233669757843, "step": 1070 }, { "epoch": 0.24700540540540541, "grad_norm": 22.39129066467285, "learning_rate": 9.695471260473546e-06, "loss": 13.253756523132324, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.31073586642742157, "loss_total_v6": 0.8283597752451897, "step": 1071 }, { "epoch": 0.24723603603603603, "grad_norm": 24.555591583251953, "learning_rate": 9.694778803501404e-06, "loss": 12.010652542114258, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.25048269145190716, "loss_total_v6": 0.7506658136844635, "step": 1072 }, { "epoch": 0.24746666666666667, "grad_norm": 24.73032569885254, "learning_rate": 9.694085584931483e-06, "loss": 13.060358047485352, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.30121200159192085, "loss_total_v6": 0.816272422671318, "step": 1073 }, { "epoch": 0.2476972972972973, "grad_norm": 23.85005760192871, "learning_rate": 9.693391604876236e-06, "loss": 12.773170471191406, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.28542946837842464, "loss_total_v6": 0.7983231544494629, "step": 1074 }, { "epoch": 0.24792792792792792, "grad_norm": 24.889249801635742, "learning_rate": 9.692696863448246e-06, "loss": 12.787275314331055, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.27672846801579, "loss_total_v6": 0.7992046698927879, "step": 1075 }, { "epoch": 0.24815855855855856, "grad_norm": 23.103485107421875, "learning_rate": 9.692001360760212e-06, "loss": 12.862387657165527, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.28874722868204117, "loss_total_v6": 0.8038992285728455, "step": 1076 }, { "epoch": 0.2483891891891892, "grad_norm": 24.709999084472656, "learning_rate": 9.69130509692496e-06, "loss": 13.105886459350586, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.3023027181625366, "loss_total_v6": 0.819117896258831, "step": 1077 }, { "epoch": 0.24861981981981982, "grad_norm": 26.476102828979492, "learning_rate": 9.690608072055443e-06, "loss": 12.809348106384277, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.29573197662830353, "loss_total_v6": 0.8005842640995979, "step": 1078 }, { "epoch": 0.24885045045045046, "grad_norm": 26.531314849853516, "learning_rate": 9.689910286264731e-06, "loss": 13.391551971435547, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.31548757664859295, "loss_total_v6": 0.8369719535112381, "step": 1079 }, { "epoch": 0.24908108108108107, "grad_norm": 22.792438507080078, "learning_rate": 9.689211739666023e-06, "loss": 12.302303314208984, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.2524144649505615, "loss_total_v6": 0.7688939496874809, "step": 1080 }, { "epoch": 0.2493117117117117, "grad_norm": 26.267431259155273, "learning_rate": 9.688512432372637e-06, "loss": 13.215476036071777, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.3149503692984581, "loss_total_v6": 0.8259672522544861, "step": 1081 }, { "epoch": 0.24954234234234235, "grad_norm": 28.130401611328125, "learning_rate": 9.687812364498015e-06, "loss": 12.553899765014648, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.28059045784175396, "loss_total_v6": 0.7846187725663185, "step": 1082 }, { "epoch": 0.24977297297297296, "grad_norm": 23.886762619018555, "learning_rate": 9.687111536155726e-06, "loss": 12.736980438232422, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.27668258361518383, "loss_total_v6": 0.796061284840107, "step": 1083 }, { "epoch": 0.25000360360360363, "grad_norm": 28.1365909576416, "learning_rate": 9.68640994745946e-06, "loss": 12.322441101074219, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.26647524535655975, "loss_total_v6": 0.7701526135206223, "step": 1084 }, { "epoch": 0.25023423423423424, "grad_norm": 24.034196853637695, "learning_rate": 9.685707598523027e-06, "loss": 12.922435760498047, "loss_alignment": 1.02587890625, "loss_alignment_w": 0.512939453125, "loss_sft": 0.29497218504548073, "loss_total_v6": 0.8076522126793861, "step": 1085 }, { "epoch": 0.25046486486486486, "grad_norm": 23.909849166870117, "learning_rate": 9.685004489460366e-06, "loss": 12.6168212890625, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.273765554651618, "loss_total_v6": 0.7885512933135033, "step": 1086 }, { "epoch": 0.25069549549549547, "grad_norm": 24.521013259887695, "learning_rate": 9.684300620385538e-06, "loss": 12.196929931640625, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.25429726019501686, "loss_total_v6": 0.7623081281781197, "step": 1087 }, { "epoch": 0.25092612612612614, "grad_norm": 22.380029678344727, "learning_rate": 9.683595991412725e-06, "loss": 12.52566146850586, "loss_alignment": 1.02587890625, "loss_alignment_w": 0.512939453125, "loss_sft": 0.2702500708401203, "loss_total_v6": 0.7828538119792938, "step": 1088 }, { "epoch": 0.25115675675675675, "grad_norm": 25.366987228393555, "learning_rate": 9.682890602656233e-06, "loss": 12.562848091125488, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.27182659320533276, "loss_total_v6": 0.7851780280470848, "step": 1089 }, { "epoch": 0.25138738738738736, "grad_norm": 27.59398651123047, "learning_rate": 9.682184454230495e-06, "loss": 13.221534729003906, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.30907291173934937, "loss_total_v6": 0.82634586840868, "step": 1090 }, { "epoch": 0.25161801801801803, "grad_norm": 26.299434661865234, "learning_rate": 9.68147754625006e-06, "loss": 12.665227890014648, "loss_alignment": 0.9912109375, "loss_alignment_w": 0.49560546875, "loss_sft": 0.29569664038717747, "loss_total_v6": 0.7915767729282379, "step": 1091 }, { "epoch": 0.25184864864864864, "grad_norm": 24.312929153442383, "learning_rate": 9.680769878829606e-06, "loss": 12.234210968017578, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.26059466414153576, "loss_total_v6": 0.7646382302045822, "step": 1092 }, { "epoch": 0.25207927927927926, "grad_norm": 26.011398315429688, "learning_rate": 9.680061452083933e-06, "loss": 12.625181198120117, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.279445543885231, "loss_total_v6": 0.7890738546848297, "step": 1093 }, { "epoch": 0.2523099099099099, "grad_norm": 25.36072540283203, "learning_rate": 9.67935226612796e-06, "loss": 12.150004386901855, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.25037264078855515, "loss_total_v6": 0.7593752890825272, "step": 1094 }, { "epoch": 0.25254054054054054, "grad_norm": 22.10685920715332, "learning_rate": 9.678642321076738e-06, "loss": 12.972715377807617, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2920263782143593, "loss_total_v6": 0.8107947036623955, "step": 1095 }, { "epoch": 0.25277117117117115, "grad_norm": 25.33932876586914, "learning_rate": 9.677931617045432e-06, "loss": 13.012516021728516, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.29689435474574566, "loss_total_v6": 0.8132822513580322, "step": 1096 }, { "epoch": 0.2530018018018018, "grad_norm": 29.402061462402344, "learning_rate": 9.677220154149338e-06, "loss": 12.786957740783691, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2856655530631542, "loss_total_v6": 0.7991848438978195, "step": 1097 }, { "epoch": 0.25323243243243243, "grad_norm": 23.069480895996094, "learning_rate": 9.676507932503868e-06, "loss": 12.493978500366211, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2755941115319729, "loss_total_v6": 0.7808736562728882, "step": 1098 }, { "epoch": 0.25346306306306304, "grad_norm": 24.900386810302734, "learning_rate": 9.675794952224558e-06, "loss": 12.920269012451172, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.29271584562957287, "loss_total_v6": 0.8075168430805206, "step": 1099 }, { "epoch": 0.2536936936936937, "grad_norm": 22.385908126831055, "learning_rate": 9.675081213427076e-06, "loss": 12.721911430358887, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.286223566159606, "loss_total_v6": 0.7951194569468498, "step": 1100 }, { "epoch": 0.2539243243243243, "grad_norm": 24.470388412475586, "learning_rate": 9.6743667162272e-06, "loss": 13.15005874633789, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.31672120839357376, "loss_total_v6": 0.8218786865472794, "step": 1101 }, { "epoch": 0.25415495495495494, "grad_norm": 24.390792846679688, "learning_rate": 9.67365146074084e-06, "loss": 13.124813079833984, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.31300707906484604, "loss_total_v6": 0.8203008249402046, "step": 1102 }, { "epoch": 0.2543855855855856, "grad_norm": 25.76036834716797, "learning_rate": 9.672935447084028e-06, "loss": 12.776143074035645, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2867291923612356, "loss_total_v6": 0.7985089644789696, "step": 1103 }, { "epoch": 0.2546162162162162, "grad_norm": 25.8612117767334, "learning_rate": 9.672218675372914e-06, "loss": 12.500904083251953, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2698776889592409, "loss_total_v6": 0.7813065201044083, "step": 1104 }, { "epoch": 0.25484684684684683, "grad_norm": 22.032424926757812, "learning_rate": 9.671501145723777e-06, "loss": 12.643732070922852, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.274684589356184, "loss_total_v6": 0.7902332842350006, "step": 1105 }, { "epoch": 0.2550774774774775, "grad_norm": 24.612428665161133, "learning_rate": 9.670782858253015e-06, "loss": 12.475208282470703, "loss_alignment": 1.02880859375, "loss_alignment_w": 0.514404296875, "loss_sft": 0.2663491014391184, "loss_total_v6": 0.7797005325555801, "step": 1106 }, { "epoch": 0.2553081081081081, "grad_norm": 23.40472984313965, "learning_rate": 9.670063813077151e-06, "loss": 12.974767684936523, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2987922504544258, "loss_total_v6": 0.8109229728579521, "step": 1107 }, { "epoch": 0.2555387387387387, "grad_norm": 21.244417190551758, "learning_rate": 9.669344010312829e-06, "loss": 12.357698440551758, "loss_alignment": 1.01513671875, "loss_alignment_w": 0.507568359375, "loss_sft": 0.26603902876377106, "loss_total_v6": 0.7723561450839043, "step": 1108 }, { "epoch": 0.2557693693693694, "grad_norm": 22.406587600708008, "learning_rate": 9.668623450076818e-06, "loss": 13.188515663146973, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.3136468678712845, "loss_total_v6": 0.8242822214961052, "step": 1109 }, { "epoch": 0.256, "grad_norm": 24.78691864013672, "learning_rate": 9.667902132486009e-06, "loss": 12.476913452148438, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.26404469460248947, "loss_total_v6": 0.7798070386052132, "step": 1110 }, { "epoch": 0.2562306306306306, "grad_norm": 24.60459327697754, "learning_rate": 9.667180057657416e-06, "loss": 12.901510238647461, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.29004794731736183, "loss_total_v6": 0.8063443675637245, "step": 1111 }, { "epoch": 0.2564612612612613, "grad_norm": 31.529130935668945, "learning_rate": 9.666457225708175e-06, "loss": 12.523772239685059, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2785395812243223, "loss_total_v6": 0.7827357575297356, "step": 1112 }, { "epoch": 0.2566918918918919, "grad_norm": 22.04999542236328, "learning_rate": 9.665733636755546e-06, "loss": 12.381412506103516, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2609750684350729, "loss_total_v6": 0.7738382145762444, "step": 1113 }, { "epoch": 0.2569225225225225, "grad_norm": 28.324010848999023, "learning_rate": 9.66500929091691e-06, "loss": 12.831327438354492, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.2856921087950468, "loss_total_v6": 0.8019579872488976, "step": 1114 }, { "epoch": 0.2571531531531532, "grad_norm": 22.10662269592285, "learning_rate": 9.66428418830977e-06, "loss": 12.93429183959961, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.2885415144264698, "loss_total_v6": 0.8083931878209114, "step": 1115 }, { "epoch": 0.2573837837837838, "grad_norm": 23.63983726501465, "learning_rate": 9.663558329051761e-06, "loss": 12.986368179321289, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.29487861320376396, "loss_total_v6": 0.8116480261087418, "step": 1116 }, { "epoch": 0.2576144144144144, "grad_norm": 26.306034088134766, "learning_rate": 9.662831713260629e-06, "loss": 12.458158493041992, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.270379900932312, "loss_total_v6": 0.7786348760128021, "step": 1117 }, { "epoch": 0.257845045045045, "grad_norm": 25.347806930541992, "learning_rate": 9.662104341054246e-06, "loss": 12.892435073852539, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2974306605756283, "loss_total_v6": 0.8057771772146225, "step": 1118 }, { "epoch": 0.2580756756756757, "grad_norm": 27.59033203125, "learning_rate": 9.66137621255061e-06, "loss": 13.35174560546875, "loss_alignment": 1.0498046875, "loss_alignment_w": 0.52490234375, "loss_sft": 0.31028367206454277, "loss_total_v6": 0.8344841301441193, "step": 1119 }, { "epoch": 0.2583063063063063, "grad_norm": 26.708744049072266, "learning_rate": 9.66064732786784e-06, "loss": 13.21506118774414, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.3136884644627571, "loss_total_v6": 0.8259412944316864, "step": 1120 }, { "epoch": 0.2585369369369369, "grad_norm": 22.068693161010742, "learning_rate": 9.659917687124176e-06, "loss": 12.602836608886719, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2771792858839035, "loss_total_v6": 0.7876773253083229, "step": 1121 }, { "epoch": 0.2587675675675676, "grad_norm": 24.555713653564453, "learning_rate": 9.659187290437982e-06, "loss": 12.970504760742188, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.2927274890244007, "loss_total_v6": 0.8106565326452255, "step": 1122 }, { "epoch": 0.2589981981981982, "grad_norm": 23.331867218017578, "learning_rate": 9.658456137927745e-06, "loss": 12.308717727661133, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.26244362629950047, "loss_total_v6": 0.7692948207259178, "step": 1123 }, { "epoch": 0.2592288288288288, "grad_norm": 23.335784912109375, "learning_rate": 9.657724229712076e-06, "loss": 12.26659870147705, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2540433965623379, "loss_total_v6": 0.7666624188423157, "step": 1124 }, { "epoch": 0.2594594594594595, "grad_norm": 23.844697952270508, "learning_rate": 9.656991565909703e-06, "loss": 12.891532897949219, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.29511591978371143, "loss_total_v6": 0.805720791220665, "step": 1125 }, { "epoch": 0.2596900900900901, "grad_norm": 22.768455505371094, "learning_rate": 9.656258146639485e-06, "loss": 12.650022506713867, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.28279861621558666, "loss_total_v6": 0.7906263694167137, "step": 1126 }, { "epoch": 0.2599207207207207, "grad_norm": 26.504945755004883, "learning_rate": 9.655523972020396e-06, "loss": 12.729177474975586, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2861740943044424, "loss_total_v6": 0.7955735474824905, "step": 1127 }, { "epoch": 0.26015135135135137, "grad_norm": 21.57057762145996, "learning_rate": 9.654789042171535e-06, "loss": 12.668363571166992, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.27761257626116276, "loss_total_v6": 0.7917727082967758, "step": 1128 }, { "epoch": 0.260381981981982, "grad_norm": 26.608642578125, "learning_rate": 9.654053357212125e-06, "loss": 12.391706466674805, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.2701939642429352, "loss_total_v6": 0.7744816839694977, "step": 1129 }, { "epoch": 0.2606126126126126, "grad_norm": 23.09879493713379, "learning_rate": 9.653316917261511e-06, "loss": 12.605541229248047, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.2820480056107044, "loss_total_v6": 0.7878463566303253, "step": 1130 }, { "epoch": 0.26084324324324326, "grad_norm": 24.833181381225586, "learning_rate": 9.65257972243916e-06, "loss": 13.038583755493164, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.3018652517348528, "loss_total_v6": 0.814911499619484, "step": 1131 }, { "epoch": 0.2610738738738739, "grad_norm": 26.140960693359375, "learning_rate": 9.65184177286466e-06, "loss": 12.839055061340332, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.28956254571676254, "loss_total_v6": 0.8024409487843513, "step": 1132 }, { "epoch": 0.2613045045045045, "grad_norm": 25.85650634765625, "learning_rate": 9.651103068657726e-06, "loss": 12.801596641540527, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.27962251380085945, "loss_total_v6": 0.8000998273491859, "step": 1133 }, { "epoch": 0.26153513513513515, "grad_norm": 25.839441299438477, "learning_rate": 9.650363609938189e-06, "loss": 13.479567527770996, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.33810902386903763, "loss_total_v6": 0.8424730077385902, "step": 1134 }, { "epoch": 0.26176576576576577, "grad_norm": 25.00703239440918, "learning_rate": 9.649623396826006e-06, "loss": 13.095144271850586, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.2996476925909519, "loss_total_v6": 0.8184465318918228, "step": 1135 }, { "epoch": 0.2619963963963964, "grad_norm": 24.901247024536133, "learning_rate": 9.648882429441258e-06, "loss": 13.06010627746582, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.298678457736969, "loss_total_v6": 0.8162566125392914, "step": 1136 }, { "epoch": 0.26222702702702705, "grad_norm": 24.54839515686035, "learning_rate": 9.648140707904146e-06, "loss": 12.782679557800293, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.285932257771492, "loss_total_v6": 0.7989174649119377, "step": 1137 }, { "epoch": 0.26245765765765766, "grad_norm": 25.047399520874023, "learning_rate": 9.64739823233499e-06, "loss": 12.516798973083496, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.2618226446211338, "loss_total_v6": 0.7822999432682991, "step": 1138 }, { "epoch": 0.2626882882882883, "grad_norm": 25.608917236328125, "learning_rate": 9.646655002854242e-06, "loss": 12.615335464477539, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.28127159364521503, "loss_total_v6": 0.7884584963321686, "step": 1139 }, { "epoch": 0.26291891891891894, "grad_norm": 24.54959487915039, "learning_rate": 9.645911019582467e-06, "loss": 13.449810028076172, "loss_alignment": 1.04248046875, "loss_alignment_w": 0.521240234375, "loss_sft": 0.31867101043462753, "loss_total_v6": 0.8406131565570831, "step": 1140 }, { "epoch": 0.26314954954954956, "grad_norm": 23.967653274536133, "learning_rate": 9.645166282640355e-06, "loss": 12.719622611999512, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2894374504685402, "loss_total_v6": 0.7949763834476471, "step": 1141 }, { "epoch": 0.26338018018018017, "grad_norm": 25.07622528076172, "learning_rate": 9.644420792148722e-06, "loss": 13.199472427368164, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.30668701604008675, "loss_total_v6": 0.824967049062252, "step": 1142 }, { "epoch": 0.26361081081081084, "grad_norm": 24.628631591796875, "learning_rate": 9.643674548228502e-06, "loss": 12.906000137329102, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.28684971295297146, "loss_total_v6": 0.8066250681877136, "step": 1143 }, { "epoch": 0.26384144144144145, "grad_norm": 24.123703002929688, "learning_rate": 9.642927551000749e-06, "loss": 12.746610641479492, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.2801836710423231, "loss_total_v6": 0.7966631501913071, "step": 1144 }, { "epoch": 0.26407207207207206, "grad_norm": 26.168100357055664, "learning_rate": 9.642179800586648e-06, "loss": 13.01645565032959, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.3046936597675085, "loss_total_v6": 0.8135284930467606, "step": 1145 }, { "epoch": 0.26430270270270273, "grad_norm": 26.18060874938965, "learning_rate": 9.641431297107496e-06, "loss": 12.865802764892578, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2911122180521488, "loss_total_v6": 0.8041126951575279, "step": 1146 }, { "epoch": 0.26453333333333334, "grad_norm": 24.833890914916992, "learning_rate": 9.640682040684722e-06, "loss": 12.750753402709961, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.2923902589827776, "loss_total_v6": 0.7969221323728561, "step": 1147 }, { "epoch": 0.26476396396396396, "grad_norm": 29.48882293701172, "learning_rate": 9.639932031439866e-06, "loss": 13.503936767578125, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.3306141346693039, "loss_total_v6": 0.8439960777759552, "step": 1148 }, { "epoch": 0.26499459459459457, "grad_norm": 24.87557601928711, "learning_rate": 9.6391812694946e-06, "loss": 12.530982971191406, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.27317668683826923, "loss_total_v6": 0.7831864506006241, "step": 1149 }, { "epoch": 0.26522522522522524, "grad_norm": 24.65150260925293, "learning_rate": 9.638429754970715e-06, "loss": 12.746885299682617, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.28633488342165947, "loss_total_v6": 0.7966803312301636, "step": 1150 }, { "epoch": 0.26545585585585585, "grad_norm": 26.550949096679688, "learning_rate": 9.637677487990122e-06, "loss": 13.294382095336914, "loss_alignment": 1.0546875, "loss_alignment_w": 0.52734375, "loss_sft": 0.30241070687770844, "loss_total_v6": 0.8308988660573959, "step": 1151 }, { "epoch": 0.26568648648648646, "grad_norm": 23.20376205444336, "learning_rate": 9.636924468674856e-06, "loss": 12.911039352416992, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2927951402962208, "loss_total_v6": 0.8069400042295456, "step": 1152 }, { "epoch": 0.26591711711711713, "grad_norm": 24.516067504882812, "learning_rate": 9.636170697147071e-06, "loss": 12.498064994812012, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.26503104344010353, "loss_total_v6": 0.7811290621757507, "step": 1153 }, { "epoch": 0.26614774774774774, "grad_norm": 25.139698028564453, "learning_rate": 9.63541617352905e-06, "loss": 12.46711540222168, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.26920021139085293, "loss_total_v6": 0.7791947349905968, "step": 1154 }, { "epoch": 0.26637837837837836, "grad_norm": 24.481584548950195, "learning_rate": 9.634660897943192e-06, "loss": 13.293172836303711, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.3140080776065588, "loss_total_v6": 0.8308232575654984, "step": 1155 }, { "epoch": 0.266609009009009, "grad_norm": 23.804311752319336, "learning_rate": 9.633904870512015e-06, "loss": 12.470104217529297, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.264916243031621, "loss_total_v6": 0.7793815433979034, "step": 1156 }, { "epoch": 0.26683963963963964, "grad_norm": 25.393327713012695, "learning_rate": 9.63314809135817e-06, "loss": 12.831330299377441, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.28610425628721714, "loss_total_v6": 0.8019581511616707, "step": 1157 }, { "epoch": 0.26707027027027025, "grad_norm": 26.180187225341797, "learning_rate": 9.632390560604421e-06, "loss": 13.248826026916504, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.3119383491575718, "loss_total_v6": 0.8280516192317009, "step": 1158 }, { "epoch": 0.2673009009009009, "grad_norm": 23.172996520996094, "learning_rate": 9.631632278373655e-06, "loss": 13.175981521606445, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.30285362526774406, "loss_total_v6": 0.8234988301992416, "step": 1159 }, { "epoch": 0.26753153153153153, "grad_norm": 24.96244239807129, "learning_rate": 9.630873244788884e-06, "loss": 12.739282608032227, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.2785813082009554, "loss_total_v6": 0.796205148100853, "step": 1160 }, { "epoch": 0.26776216216216214, "grad_norm": 27.19485855102539, "learning_rate": 9.630113459973238e-06, "loss": 13.038525581359863, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.30489808693528175, "loss_total_v6": 0.8149078413844109, "step": 1161 }, { "epoch": 0.2679927927927928, "grad_norm": 24.73175048828125, "learning_rate": 9.629352924049975e-06, "loss": 13.659106254577637, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.3370315618813038, "loss_total_v6": 0.8536941483616829, "step": 1162 }, { "epoch": 0.2682234234234234, "grad_norm": 23.89459228515625, "learning_rate": 9.628591637142467e-06, "loss": 12.444732666015625, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2709446419030428, "loss_total_v6": 0.7777957990765572, "step": 1163 }, { "epoch": 0.26845405405405404, "grad_norm": 26.10181999206543, "learning_rate": 9.627829599374214e-06, "loss": 12.673437118530273, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.27846369706094265, "loss_total_v6": 0.7920897901058197, "step": 1164 }, { "epoch": 0.2686846846846847, "grad_norm": 25.841161727905273, "learning_rate": 9.627066810868833e-06, "loss": 13.589006423950195, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.3359003998339176, "loss_total_v6": 0.849312886595726, "step": 1165 }, { "epoch": 0.2689153153153153, "grad_norm": 23.996984481811523, "learning_rate": 9.62630327175007e-06, "loss": 12.594524383544922, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.27821617759764194, "loss_total_v6": 0.7871577963232994, "step": 1166 }, { "epoch": 0.26914594594594593, "grad_norm": 23.851945877075195, "learning_rate": 9.625538982141782e-06, "loss": 12.739860534667969, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.28621624782681465, "loss_total_v6": 0.796241283416748, "step": 1167 }, { "epoch": 0.2693765765765766, "grad_norm": 24.998268127441406, "learning_rate": 9.624773942167958e-06, "loss": 12.522932052612305, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.28755077347159386, "loss_total_v6": 0.7826832458376884, "step": 1168 }, { "epoch": 0.2696072072072072, "grad_norm": 28.453163146972656, "learning_rate": 9.624008151952705e-06, "loss": 12.804160118103027, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.293698713183403, "loss_total_v6": 0.8002600148320198, "step": 1169 }, { "epoch": 0.2698378378378378, "grad_norm": 27.761693954467773, "learning_rate": 9.623241611620248e-06, "loss": 12.846978187561035, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2948489859700203, "loss_total_v6": 0.8029361441731453, "step": 1170 }, { "epoch": 0.2700684684684685, "grad_norm": 24.63437843322754, "learning_rate": 9.622474321294939e-06, "loss": 12.796026229858398, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.27874029986560345, "loss_total_v6": 0.7997516170144081, "step": 1171 }, { "epoch": 0.2702990990990991, "grad_norm": 28.420482635498047, "learning_rate": 9.62170628110125e-06, "loss": 13.300704956054688, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.3126630373299122, "loss_total_v6": 0.8312940374016762, "step": 1172 }, { "epoch": 0.2705297297297297, "grad_norm": 26.78115463256836, "learning_rate": 9.620937491163772e-06, "loss": 13.015251159667969, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.3066783621907234, "loss_total_v6": 0.8134532496333122, "step": 1173 }, { "epoch": 0.2707603603603604, "grad_norm": 26.15261459350586, "learning_rate": 9.620167951607221e-06, "loss": 12.320856094360352, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.25163616612553596, "loss_total_v6": 0.770053505897522, "step": 1174 }, { "epoch": 0.270990990990991, "grad_norm": 26.677379608154297, "learning_rate": 9.619397662556434e-06, "loss": 13.415703773498535, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.32311589270830154, "loss_total_v6": 0.8384815081954002, "step": 1175 }, { "epoch": 0.2712216216216216, "grad_norm": 26.387805938720703, "learning_rate": 9.61862662413637e-06, "loss": 12.981380462646484, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.300533065572381, "loss_total_v6": 0.8113362565636635, "step": 1176 }, { "epoch": 0.2714522522522523, "grad_norm": 23.5004940032959, "learning_rate": 9.617854836472107e-06, "loss": 13.190299034118652, "loss_alignment": 1.046875, "loss_alignment_w": 0.5234375, "loss_sft": 0.30030008032917976, "loss_total_v6": 0.8243937119841576, "step": 1177 }, { "epoch": 0.2716828828828829, "grad_norm": 26.14524269104004, "learning_rate": 9.617082299688846e-06, "loss": 12.542251586914062, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2750406116247177, "loss_total_v6": 0.7838907167315483, "step": 1178 }, { "epoch": 0.2719135135135135, "grad_norm": 23.623332977294922, "learning_rate": 9.61630901391191e-06, "loss": 12.642358779907227, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.284272750839591, "loss_total_v6": 0.7901473864912987, "step": 1179 }, { "epoch": 0.2721441441441441, "grad_norm": 28.380924224853516, "learning_rate": 9.615534979266745e-06, "loss": 13.242812156677246, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.3074578568339348, "loss_total_v6": 0.8276757523417473, "step": 1180 }, { "epoch": 0.2723747747747748, "grad_norm": 24.87741470336914, "learning_rate": 9.614760195878914e-06, "loss": 12.725868225097656, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.28560114838182926, "loss_total_v6": 0.7953667715191841, "step": 1181 }, { "epoch": 0.2726054054054054, "grad_norm": 23.9757080078125, "learning_rate": 9.613984663874106e-06, "loss": 12.930523872375488, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.28841287828981876, "loss_total_v6": 0.8081577271223068, "step": 1182 }, { "epoch": 0.272836036036036, "grad_norm": 24.369783401489258, "learning_rate": 9.613208383378128e-06, "loss": 12.708629608154297, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.29074934124946594, "loss_total_v6": 0.7942893579602242, "step": 1183 }, { "epoch": 0.2730666666666667, "grad_norm": 28.7420711517334, "learning_rate": 9.612431354516912e-06, "loss": 12.948149681091309, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2966555692255497, "loss_total_v6": 0.8092593625187874, "step": 1184 }, { "epoch": 0.2732972972972973, "grad_norm": 24.971664428710938, "learning_rate": 9.611653577416508e-06, "loss": 12.837041854858398, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.28638490848243237, "loss_total_v6": 0.8023150786757469, "step": 1185 }, { "epoch": 0.2735279279279279, "grad_norm": 21.631866455078125, "learning_rate": 9.610875052203087e-06, "loss": 12.786937713623047, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.28783107176423073, "loss_total_v6": 0.7991836294531822, "step": 1186 }, { "epoch": 0.2737585585585586, "grad_norm": 23.802648544311523, "learning_rate": 9.610095779002948e-06, "loss": 12.11536979675293, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.24437797255814075, "loss_total_v6": 0.7572106271982193, "step": 1187 }, { "epoch": 0.2739891891891892, "grad_norm": 25.190725326538086, "learning_rate": 9.609315757942504e-06, "loss": 12.889793395996094, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.29233696311712265, "loss_total_v6": 0.805612102150917, "step": 1188 }, { "epoch": 0.2742198198198198, "grad_norm": 23.89413070678711, "learning_rate": 9.60853498914829e-06, "loss": 12.956853866577148, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.29787104949355125, "loss_total_v6": 0.8098033741116524, "step": 1189 }, { "epoch": 0.27445045045045047, "grad_norm": 24.2908878326416, "learning_rate": 9.607753472746967e-06, "loss": 12.645339965820312, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2870226074010134, "loss_total_v6": 0.7903337478637695, "step": 1190 }, { "epoch": 0.2746810810810811, "grad_norm": 23.899507522583008, "learning_rate": 9.60697120886531e-06, "loss": 12.854933738708496, "loss_alignment": 1.04150390625, "loss_alignment_w": 0.520751953125, "loss_sft": 0.28289503790438175, "loss_total_v6": 0.8034333512187004, "step": 1191 }, { "epoch": 0.2749117117117117, "grad_norm": 23.948837280273438, "learning_rate": 9.606188197630224e-06, "loss": 13.011219024658203, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.31156856566667557, "loss_total_v6": 0.8132012337446213, "step": 1192 }, { "epoch": 0.27514234234234236, "grad_norm": 24.449100494384766, "learning_rate": 9.60540443916873e-06, "loss": 12.851941108703613, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2994468826800585, "loss_total_v6": 0.8032463267445564, "step": 1193 }, { "epoch": 0.275372972972973, "grad_norm": 24.55183219909668, "learning_rate": 9.604619933607971e-06, "loss": 12.968141555786133, "loss_alignment": 1.02197265625, "loss_alignment_w": 0.510986328125, "loss_sft": 0.29917153529822826, "loss_total_v6": 0.8105088248848915, "step": 1194 }, { "epoch": 0.2756036036036036, "grad_norm": 22.713802337646484, "learning_rate": 9.60383468107521e-06, "loss": 12.386503219604492, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.26335326209664345, "loss_total_v6": 0.7741564884781837, "step": 1195 }, { "epoch": 0.27583423423423425, "grad_norm": 22.73587989807129, "learning_rate": 9.603048681697834e-06, "loss": 12.307640075683594, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.26240688003599644, "loss_total_v6": 0.7692275643348694, "step": 1196 }, { "epoch": 0.27606486486486487, "grad_norm": 21.621646881103516, "learning_rate": 9.602261935603348e-06, "loss": 13.006136894226074, "loss_alignment": 1.01513671875, "loss_alignment_w": 0.507568359375, "loss_sft": 0.3052542060613632, "loss_total_v6": 0.812883585691452, "step": 1197 }, { "epoch": 0.2762954954954955, "grad_norm": 26.532697677612305, "learning_rate": 9.601474442919383e-06, "loss": 12.739477157592773, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2853530365973711, "loss_total_v6": 0.7962172850966454, "step": 1198 }, { "epoch": 0.27652612612612615, "grad_norm": 23.816913604736328, "learning_rate": 9.600686203773684e-06, "loss": 13.601625442504883, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.34550875797867775, "loss_total_v6": 0.8501016348600388, "step": 1199 }, { "epoch": 0.27675675675675676, "grad_norm": 25.501590728759766, "learning_rate": 9.599897218294122e-06, "loss": 12.821215629577637, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2908126451075077, "loss_total_v6": 0.8013259693980217, "step": 1200 }, { "epoch": 0.2769873873873874, "grad_norm": 27.51442527770996, "learning_rate": 9.59910748660869e-06, "loss": 13.007308959960938, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.2940511852502823, "loss_total_v6": 0.8129567950963974, "step": 1201 }, { "epoch": 0.27721801801801804, "grad_norm": 24.814329147338867, "learning_rate": 9.598317008845496e-06, "loss": 12.784549713134766, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.28537772223353386, "loss_total_v6": 0.7990343421697617, "step": 1202 }, { "epoch": 0.27744864864864865, "grad_norm": 24.432443618774414, "learning_rate": 9.597525785132778e-06, "loss": 13.285202026367188, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.31733992137014866, "loss_total_v6": 0.8303251266479492, "step": 1203 }, { "epoch": 0.27767927927927927, "grad_norm": 23.285751342773438, "learning_rate": 9.596733815598888e-06, "loss": 13.021656036376953, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.2958633527159691, "loss_total_v6": 0.8138534650206566, "step": 1204 }, { "epoch": 0.27790990990990994, "grad_norm": 23.567550659179688, "learning_rate": 9.595941100372301e-06, "loss": 12.604490280151367, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2789457682520151, "loss_total_v6": 0.7877806127071381, "step": 1205 }, { "epoch": 0.27814054054054055, "grad_norm": 25.091371536254883, "learning_rate": 9.595147639581613e-06, "loss": 12.886829376220703, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.2814857978373766, "loss_total_v6": 0.805426836013794, "step": 1206 }, { "epoch": 0.27837117117117116, "grad_norm": 27.11408805847168, "learning_rate": 9.594353433355541e-06, "loss": 13.266374588012695, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.31527817621827126, "loss_total_v6": 0.8291483968496323, "step": 1207 }, { "epoch": 0.27860180180180183, "grad_norm": 27.08864402770996, "learning_rate": 9.593558481822923e-06, "loss": 12.698482513427734, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.28358437679708004, "loss_total_v6": 0.7936551794409752, "step": 1208 }, { "epoch": 0.27883243243243244, "grad_norm": 27.030651092529297, "learning_rate": 9.592762785112716e-06, "loss": 12.450089454650879, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2655878756195307, "loss_total_v6": 0.7781305909156799, "step": 1209 }, { "epoch": 0.27906306306306305, "grad_norm": 25.866249084472656, "learning_rate": 9.591966343354005e-06, "loss": 12.985499382019043, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.30475782230496407, "loss_total_v6": 0.811593733727932, "step": 1210 }, { "epoch": 0.27929369369369367, "grad_norm": 21.57199478149414, "learning_rate": 9.591169156675985e-06, "loss": 13.040770530700684, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2976836636662483, "loss_total_v6": 0.8150481730699539, "step": 1211 }, { "epoch": 0.27952432432432434, "grad_norm": 24.39735221862793, "learning_rate": 9.590371225207981e-06, "loss": 12.2890625, "loss_alignment": 1.01220703125, "loss_alignment_w": 0.506103515625, "loss_sft": 0.26202392764389515, "loss_total_v6": 0.7680664360523224, "step": 1212 }, { "epoch": 0.27975495495495495, "grad_norm": 23.000045776367188, "learning_rate": 9.589572549079434e-06, "loss": 13.065366744995117, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.30816254019737244, "loss_total_v6": 0.8165854513645172, "step": 1213 }, { "epoch": 0.27998558558558556, "grad_norm": 26.250572204589844, "learning_rate": 9.588773128419907e-06, "loss": 13.60793685913086, "loss_alignment": 1.048828125, "loss_alignment_w": 0.5244140625, "loss_sft": 0.3266007900238037, "loss_total_v6": 0.8504960760474205, "step": 1214 }, { "epoch": 0.28021621621621623, "grad_norm": 25.056283950805664, "learning_rate": 9.587972963359084e-06, "loss": 13.286609649658203, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.3169396109879017, "loss_total_v6": 0.8304131180047989, "step": 1215 }, { "epoch": 0.28044684684684684, "grad_norm": 25.946317672729492, "learning_rate": 9.587172054026769e-06, "loss": 12.802812576293945, "loss_alignment": 1.01611328125, "loss_alignment_w": 0.508056640625, "loss_sft": 0.29185971431434155, "loss_total_v6": 0.800175778567791, "step": 1216 }, { "epoch": 0.28067747747747746, "grad_norm": 23.326202392578125, "learning_rate": 9.586370400552888e-06, "loss": 13.072839736938477, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.30653916485607624, "loss_total_v6": 0.8170524835586548, "step": 1217 }, { "epoch": 0.2809081081081081, "grad_norm": 24.488292694091797, "learning_rate": 9.585568003067487e-06, "loss": 13.260452270507812, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.32479577511548996, "loss_total_v6": 0.8287783041596413, "step": 1218 }, { "epoch": 0.28113873873873874, "grad_norm": 26.847946166992188, "learning_rate": 9.584764861700732e-06, "loss": 13.149012565612793, "loss_alignment": 1.0537109375, "loss_alignment_w": 0.52685546875, "loss_sft": 0.2961021885275841, "loss_total_v6": 0.8218132629990578, "step": 1219 }, { "epoch": 0.28136936936936935, "grad_norm": 26.47580337524414, "learning_rate": 9.583960976582914e-06, "loss": 12.415088653564453, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2652008142322302, "loss_total_v6": 0.7759430408477783, "step": 1220 }, { "epoch": 0.2816, "grad_norm": 25.326486587524414, "learning_rate": 9.583156347844435e-06, "loss": 12.786873817443848, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2875524275004864, "loss_total_v6": 0.7991796433925629, "step": 1221 }, { "epoch": 0.28183063063063063, "grad_norm": 25.293540954589844, "learning_rate": 9.582350975615828e-06, "loss": 12.57264518737793, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.26908197067677975, "loss_total_v6": 0.7857903465628624, "step": 1222 }, { "epoch": 0.28206126126126124, "grad_norm": 25.47283935546875, "learning_rate": 9.581544860027742e-06, "loss": 12.620347023010254, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2823324501514435, "loss_total_v6": 0.7887716591358185, "step": 1223 }, { "epoch": 0.2822918918918919, "grad_norm": 24.38860511779785, "learning_rate": 9.580738001210946e-06, "loss": 13.094470977783203, "loss_alignment": 1.02490234375, "loss_alignment_w": 0.512451171875, "loss_sft": 0.30657885037362576, "loss_total_v6": 0.8184044361114502, "step": 1224 }, { "epoch": 0.2825225225225225, "grad_norm": 26.526323318481445, "learning_rate": 9.57993039929633e-06, "loss": 12.94643497467041, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.29248955473303795, "loss_total_v6": 0.8091521561145782, "step": 1225 }, { "epoch": 0.28275315315315314, "grad_norm": 26.989410400390625, "learning_rate": 9.579122054414907e-06, "loss": 12.712360382080078, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.28791545145213604, "loss_total_v6": 0.7945225164294243, "step": 1226 }, { "epoch": 0.2829837837837838, "grad_norm": 22.770021438598633, "learning_rate": 9.578312966697807e-06, "loss": 12.773187637329102, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.29611166939139366, "loss_total_v6": 0.7983241975307465, "step": 1227 }, { "epoch": 0.2832144144144144, "grad_norm": 24.692121505737305, "learning_rate": 9.57750313627628e-06, "loss": 12.716264724731445, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2837497089058161, "loss_total_v6": 0.7947665154933929, "step": 1228 }, { "epoch": 0.28344504504504503, "grad_norm": 24.731340408325195, "learning_rate": 9.576692563281705e-06, "loss": 12.969952583312988, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.28770333155989647, "loss_total_v6": 0.81062201410532, "step": 1229 }, { "epoch": 0.2836756756756757, "grad_norm": 24.037097930908203, "learning_rate": 9.575881247845568e-06, "loss": 13.133169174194336, "loss_alignment": 1.01806640625, "loss_alignment_w": 0.509033203125, "loss_sft": 0.3106454983353615, "loss_total_v6": 0.8208231180906296, "step": 1230 }, { "epoch": 0.2839063063063063, "grad_norm": 21.361804962158203, "learning_rate": 9.575069190099487e-06, "loss": 12.497401237487793, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.2591607179492712, "loss_total_v6": 0.7810875847935677, "step": 1231 }, { "epoch": 0.2841369369369369, "grad_norm": 25.631839752197266, "learning_rate": 9.574256390175192e-06, "loss": 12.99005126953125, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.30038839019834995, "loss_total_v6": 0.8118782341480255, "step": 1232 }, { "epoch": 0.2843675675675676, "grad_norm": 22.44606590270996, "learning_rate": 9.573442848204542e-06, "loss": 13.33050537109375, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.3219871111214161, "loss_total_v6": 0.8331565409898758, "step": 1233 }, { "epoch": 0.2845981981981982, "grad_norm": 26.762157440185547, "learning_rate": 9.572628564319508e-06, "loss": 12.684236526489258, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.28215997852385044, "loss_total_v6": 0.7927648052573204, "step": 1234 }, { "epoch": 0.2848288288288288, "grad_norm": 24.830198287963867, "learning_rate": 9.571813538652184e-06, "loss": 12.651044845581055, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2752789109945297, "loss_total_v6": 0.7906902804970741, "step": 1235 }, { "epoch": 0.2850594594594595, "grad_norm": 24.43252182006836, "learning_rate": 9.57099777133479e-06, "loss": 13.314455032348633, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.31805430725216866, "loss_total_v6": 0.8321534395217896, "step": 1236 }, { "epoch": 0.2852900900900901, "grad_norm": 23.985607147216797, "learning_rate": 9.570181262499659e-06, "loss": 12.550152778625488, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2690647169947624, "loss_total_v6": 0.7843845635652542, "step": 1237 }, { "epoch": 0.2855207207207207, "grad_norm": 25.959638595581055, "learning_rate": 9.569364012279247e-06, "loss": 12.691640853881836, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2821038607507944, "loss_total_v6": 0.7932275235652924, "step": 1238 }, { "epoch": 0.2857513513513513, "grad_norm": 28.156736373901367, "learning_rate": 9.56854602080613e-06, "loss": 12.493424415588379, "loss_alignment": 1.01513671875, "loss_alignment_w": 0.507568359375, "loss_sft": 0.2749338448047638, "loss_total_v6": 0.7808390185236931, "step": 1239 }, { "epoch": 0.285981981981982, "grad_norm": 24.900985717773438, "learning_rate": 9.567727288213005e-06, "loss": 12.678149223327637, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2862350884824991, "loss_total_v6": 0.7923843562602997, "step": 1240 }, { "epoch": 0.2862126126126126, "grad_norm": 28.92009925842285, "learning_rate": 9.56690781463269e-06, "loss": 12.776102066040039, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.28680294193327427, "loss_total_v6": 0.7985064014792442, "step": 1241 }, { "epoch": 0.2864432432432432, "grad_norm": 26.44086265563965, "learning_rate": 9.566087600198118e-06, "loss": 12.57429027557373, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.28645772114396095, "loss_total_v6": 0.785893127322197, "step": 1242 }, { "epoch": 0.2866738738738739, "grad_norm": 25.839763641357422, "learning_rate": 9.565266645042349e-06, "loss": 12.655038833618164, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.27716122940182686, "loss_total_v6": 0.7909399047493935, "step": 1243 }, { "epoch": 0.2869045045045045, "grad_norm": 23.781171798706055, "learning_rate": 9.564444949298559e-06, "loss": 12.595135688781738, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2735088914632797, "loss_total_v6": 0.787196010351181, "step": 1244 }, { "epoch": 0.2871351351351351, "grad_norm": 23.225528717041016, "learning_rate": 9.563622513100046e-06, "loss": 11.944534301757812, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.24429034441709518, "loss_total_v6": 0.7465333938598633, "step": 1245 }, { "epoch": 0.2873657657657658, "grad_norm": 25.49283218383789, "learning_rate": 9.562799336580227e-06, "loss": 12.548213005065918, "loss_alignment": 1.01708984375, "loss_alignment_w": 0.508544921875, "loss_sft": 0.27544375881552696, "loss_total_v6": 0.7842633128166199, "step": 1246 }, { "epoch": 0.2875963963963964, "grad_norm": 25.920886993408203, "learning_rate": 9.56197541987264e-06, "loss": 12.805896759033203, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2895500585436821, "loss_total_v6": 0.8003685623407364, "step": 1247 }, { "epoch": 0.287827027027027, "grad_norm": 27.99931526184082, "learning_rate": 9.561150763110945e-06, "loss": 12.573173522949219, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2795520517975092, "loss_total_v6": 0.785823404788971, "step": 1248 }, { "epoch": 0.2880576576576577, "grad_norm": 24.314598083496094, "learning_rate": 9.560325366428915e-06, "loss": 12.28512191772461, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.25828330032527447, "loss_total_v6": 0.7678200751543045, "step": 1249 }, { "epoch": 0.2882882882882883, "grad_norm": 27.72245979309082, "learning_rate": 9.55949922996045e-06, "loss": 13.065019607543945, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.3017016872763634, "loss_total_v6": 0.8165637329220772, "step": 1250 }, { "epoch": 0.2885189189189189, "grad_norm": 23.9322452545166, "learning_rate": 9.55867235383957e-06, "loss": 13.018402099609375, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.30331989005208015, "loss_total_v6": 0.813650093972683, "step": 1251 }, { "epoch": 0.28874954954954957, "grad_norm": 24.035696029663086, "learning_rate": 9.557844738200408e-06, "loss": 12.356815338134766, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2590410467237234, "loss_total_v6": 0.7723009288311005, "step": 1252 }, { "epoch": 0.2889801801801802, "grad_norm": 24.601057052612305, "learning_rate": 9.557016383177226e-06, "loss": 12.821462631225586, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2909197062253952, "loss_total_v6": 0.8013414517045021, "step": 1253 }, { "epoch": 0.2892108108108108, "grad_norm": 24.182191848754883, "learning_rate": 9.556187288904401e-06, "loss": 12.490381240844727, "loss_alignment": 1.01220703125, "loss_alignment_w": 0.506103515625, "loss_sft": 0.27466731145977974, "loss_total_v6": 0.7806487753987312, "step": 1254 }, { "epoch": 0.28944144144144146, "grad_norm": 22.4071102142334, "learning_rate": 9.555357455516427e-06, "loss": 12.742963790893555, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.28956871666014194, "loss_total_v6": 0.7964351922273636, "step": 1255 }, { "epoch": 0.2896720720720721, "grad_norm": 24.25261688232422, "learning_rate": 9.554526883147926e-06, "loss": 12.5204439163208, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.27756863459944725, "loss_total_v6": 0.7825277522206306, "step": 1256 }, { "epoch": 0.2899027027027027, "grad_norm": 27.315547943115234, "learning_rate": 9.553695571933632e-06, "loss": 13.222748756408691, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.3161984197795391, "loss_total_v6": 0.8264218047261238, "step": 1257 }, { "epoch": 0.29013333333333335, "grad_norm": 28.444677352905273, "learning_rate": 9.552863522008404e-06, "loss": 12.632949829101562, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2836083862930536, "loss_total_v6": 0.7895593121647835, "step": 1258 }, { "epoch": 0.29036396396396397, "grad_norm": 24.529136657714844, "learning_rate": 9.55203073350722e-06, "loss": 12.072677612304688, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.2542371600866318, "loss_total_v6": 0.7545423656702042, "step": 1259 }, { "epoch": 0.2905945945945946, "grad_norm": 23.725889205932617, "learning_rate": 9.551197206565174e-06, "loss": 13.161599159240723, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.31330733746290207, "loss_total_v6": 0.8225999400019646, "step": 1260 }, { "epoch": 0.29082522522522525, "grad_norm": 25.325958251953125, "learning_rate": 9.550362941317484e-06, "loss": 13.050385475158691, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.3015041686594486, "loss_total_v6": 0.8156490698456764, "step": 1261 }, { "epoch": 0.29105585585585586, "grad_norm": 26.179637908935547, "learning_rate": 9.549527937899487e-06, "loss": 12.729767799377441, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.2964649982750416, "loss_total_v6": 0.7956104949116707, "step": 1262 }, { "epoch": 0.2912864864864865, "grad_norm": 25.321500778198242, "learning_rate": 9.548692196446638e-06, "loss": 12.802457809448242, "loss_alignment": 1.01416015625, "loss_alignment_w": 0.507080078125, "loss_sft": 0.2933481689542532, "loss_total_v6": 0.8001535907387733, "step": 1263 }, { "epoch": 0.29151711711711714, "grad_norm": 23.848419189453125, "learning_rate": 9.547855717094516e-06, "loss": 12.728681564331055, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2877453211694956, "loss_total_v6": 0.7955425754189491, "step": 1264 }, { "epoch": 0.29174774774774775, "grad_norm": 23.046144485473633, "learning_rate": 9.547018499978811e-06, "loss": 12.122171401977539, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.26199973188340664, "loss_total_v6": 0.7576357051730156, "step": 1265 }, { "epoch": 0.29197837837837837, "grad_norm": 24.44838523864746, "learning_rate": 9.546180545235344e-06, "loss": 12.488012313842773, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2688735853880644, "loss_total_v6": 0.7805007919669151, "step": 1266 }, { "epoch": 0.29220900900900904, "grad_norm": 24.528589248657227, "learning_rate": 9.545341853000046e-06, "loss": 12.566903114318848, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.27874808944761753, "loss_total_v6": 0.785431444644928, "step": 1267 }, { "epoch": 0.29243963963963965, "grad_norm": 25.502647399902344, "learning_rate": 9.544502423408974e-06, "loss": 12.89615249633789, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2916205320507288, "loss_total_v6": 0.8060095459222794, "step": 1268 }, { "epoch": 0.29267027027027026, "grad_norm": 25.374731063842773, "learning_rate": 9.5436622565983e-06, "loss": 12.382574081420898, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.2733616065233946, "loss_total_v6": 0.7739109098911285, "step": 1269 }, { "epoch": 0.2929009009009009, "grad_norm": 22.788278579711914, "learning_rate": 9.54282135270432e-06, "loss": 12.430469512939453, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2661163844168186, "loss_total_v6": 0.7769043296575546, "step": 1270 }, { "epoch": 0.29313153153153154, "grad_norm": 25.191625595092773, "learning_rate": 9.541979711863445e-06, "loss": 12.943092346191406, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.3026414178311825, "loss_total_v6": 0.8089433014392853, "step": 1271 }, { "epoch": 0.29336216216216215, "grad_norm": 27.897613525390625, "learning_rate": 9.541137334212212e-06, "loss": 13.021003723144531, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.30351308174431324, "loss_total_v6": 0.8138127475976944, "step": 1272 }, { "epoch": 0.29359279279279277, "grad_norm": 24.14040184020996, "learning_rate": 9.540294219887267e-06, "loss": 12.332098007202148, "loss_alignment": 0.99267578125, "loss_alignment_w": 0.496337890625, "loss_sft": 0.2741130795329809, "loss_total_v6": 0.7707561179995537, "step": 1273 }, { "epoch": 0.29382342342342344, "grad_norm": 25.569076538085938, "learning_rate": 9.539450369025388e-06, "loss": 13.079924583435059, "loss_alignment": 1.01904296875, "loss_alignment_w": 0.509521484375, "loss_sft": 0.30951494351029396, "loss_total_v6": 0.8174952939152718, "step": 1274 }, { "epoch": 0.29405405405405405, "grad_norm": 23.60920524597168, "learning_rate": 9.538605781763464e-06, "loss": 13.079202651977539, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.30089438520371914, "loss_total_v6": 0.8174501359462738, "step": 1275 }, { "epoch": 0.29428468468468466, "grad_norm": 23.54745101928711, "learning_rate": 9.537760458238505e-06, "loss": 12.204919815063477, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2533013056963682, "loss_total_v6": 0.7628075480461121, "step": 1276 }, { "epoch": 0.29451531531531533, "grad_norm": 23.833255767822266, "learning_rate": 9.536914398587642e-06, "loss": 12.80191421508789, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.290216650813818, "loss_total_v6": 0.8001196160912514, "step": 1277 }, { "epoch": 0.29474594594594594, "grad_norm": 24.491518020629883, "learning_rate": 9.536067602948126e-06, "loss": 12.389581680297852, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.27032052539289, "loss_total_v6": 0.7743488326668739, "step": 1278 }, { "epoch": 0.29497657657657655, "grad_norm": 25.20941162109375, "learning_rate": 9.535220071457325e-06, "loss": 12.402242660522461, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.261590339243412, "loss_total_v6": 0.7751401364803314, "step": 1279 }, { "epoch": 0.2952072072072072, "grad_norm": 23.61174964904785, "learning_rate": 9.534371804252727e-06, "loss": 12.505334854125977, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.26530230417847633, "loss_total_v6": 0.7815834283828735, "step": 1280 }, { "epoch": 0.29543783783783784, "grad_norm": 24.648176193237305, "learning_rate": 9.53352280147194e-06, "loss": 12.908470153808594, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2926954831928015, "loss_total_v6": 0.8067793622612953, "step": 1281 }, { "epoch": 0.29566846846846845, "grad_norm": 26.161701202392578, "learning_rate": 9.532673063252696e-06, "loss": 12.802321434020996, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.29990096390247345, "loss_total_v6": 0.8001450970768929, "step": 1282 }, { "epoch": 0.2958990990990991, "grad_norm": 25.08116912841797, "learning_rate": 9.531822589732831e-06, "loss": 13.094524383544922, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.30124158784747124, "loss_total_v6": 0.8184077367186546, "step": 1283 }, { "epoch": 0.29612972972972973, "grad_norm": 25.70285987854004, "learning_rate": 9.53097138105032e-06, "loss": 12.619434356689453, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2757904175668955, "loss_total_v6": 0.7887146174907684, "step": 1284 }, { "epoch": 0.29636036036036034, "grad_norm": 24.30356788635254, "learning_rate": 9.530119437343245e-06, "loss": 13.205535888671875, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.3046245835721493, "loss_total_v6": 0.8253460228443146, "step": 1285 }, { "epoch": 0.296590990990991, "grad_norm": 25.49668312072754, "learning_rate": 9.52926675874981e-06, "loss": 12.424379348754883, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2652322072535753, "loss_total_v6": 0.7765237241983414, "step": 1286 }, { "epoch": 0.2968216216216216, "grad_norm": 24.536840438842773, "learning_rate": 9.528413345408339e-06, "loss": 12.292791366577148, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2544292286038399, "loss_total_v6": 0.7682994604110718, "step": 1287 }, { "epoch": 0.29705225225225224, "grad_norm": 26.561519622802734, "learning_rate": 9.527559197457273e-06, "loss": 12.404959678649902, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.26690233685076237, "loss_total_v6": 0.7753099575638771, "step": 1288 }, { "epoch": 0.2972828828828829, "grad_norm": 26.550613403320312, "learning_rate": 9.526704315035176e-06, "loss": 12.482955932617188, "loss_alignment": 1.01904296875, "loss_alignment_w": 0.509521484375, "loss_sft": 0.270709041506052, "loss_total_v6": 0.780184768140316, "step": 1289 }, { "epoch": 0.2975135135135135, "grad_norm": 25.261350631713867, "learning_rate": 9.525848698280727e-06, "loss": 12.658427238464355, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.2851092256605625, "loss_total_v6": 0.7911516800522804, "step": 1290 }, { "epoch": 0.29774414414414413, "grad_norm": 25.395177841186523, "learning_rate": 9.524992347332729e-06, "loss": 12.677912712097168, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2813069447875023, "loss_total_v6": 0.792369544506073, "step": 1291 }, { "epoch": 0.2979747747747748, "grad_norm": 25.626066207885742, "learning_rate": 9.524135262330098e-06, "loss": 12.533084869384766, "loss_alignment": 1.01611328125, "loss_alignment_w": 0.508056640625, "loss_sft": 0.27603939175605774, "loss_total_v6": 0.783317819237709, "step": 1292 }, { "epoch": 0.2982054054054054, "grad_norm": 23.862098693847656, "learning_rate": 9.523277443411876e-06, "loss": 12.913885116577148, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.30550042167305946, "loss_total_v6": 0.8071178421378136, "step": 1293 }, { "epoch": 0.298436036036036, "grad_norm": 22.139991760253906, "learning_rate": 9.522418890717216e-06, "loss": 12.465408325195312, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.276829669252038, "loss_total_v6": 0.7790879756212234, "step": 1294 }, { "epoch": 0.2986666666666667, "grad_norm": 23.89369773864746, "learning_rate": 9.521559604385396e-06, "loss": 12.952310562133789, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.3006998300552368, "loss_total_v6": 0.8095193952322006, "step": 1295 }, { "epoch": 0.2988972972972973, "grad_norm": 22.01059341430664, "learning_rate": 9.520699584555813e-06, "loss": 13.037362098693848, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.3115392215549946, "loss_total_v6": 0.8148351237177849, "step": 1296 }, { "epoch": 0.2991279279279279, "grad_norm": 25.67611312866211, "learning_rate": 9.51983883136798e-06, "loss": 12.897891998291016, "loss_alignment": 1.01611328125, "loss_alignment_w": 0.508056640625, "loss_sft": 0.2980157919228077, "loss_total_v6": 0.8061182275414467, "step": 1297 }, { "epoch": 0.2993585585585586, "grad_norm": 24.00016975402832, "learning_rate": 9.51897734496153e-06, "loss": 13.23574161529541, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.31812434270977974, "loss_total_v6": 0.8272338137030602, "step": 1298 }, { "epoch": 0.2995891891891892, "grad_norm": 24.00811195373535, "learning_rate": 9.518115125476219e-06, "loss": 12.463720321655273, "loss_alignment": 1.02783203125, "loss_alignment_w": 0.513916015625, "loss_sft": 0.26494445092976093, "loss_total_v6": 0.7789825424551964, "step": 1299 }, { "epoch": 0.2998198198198198, "grad_norm": 24.63102149963379, "learning_rate": 9.517252173051912e-06, "loss": 13.271197319030762, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.31242102570831776, "loss_total_v6": 0.8294498324394226, "step": 1300 }, { "epoch": 0.3000504504504504, "grad_norm": 22.69171142578125, "learning_rate": 9.516388487828604e-06, "loss": 12.404380798339844, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.2731223478913307, "loss_total_v6": 0.7752738296985626, "step": 1301 }, { "epoch": 0.3002810810810811, "grad_norm": 25.810806274414062, "learning_rate": 9.5155240699464e-06, "loss": 12.77044677734375, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2947349473834038, "loss_total_v6": 0.7981529086828232, "step": 1302 }, { "epoch": 0.3005117117117117, "grad_norm": 24.814687728881836, "learning_rate": 9.514658919545533e-06, "loss": 12.817876815795898, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2948765214532614, "loss_total_v6": 0.8011173382401466, "step": 1303 }, { "epoch": 0.3007423423423423, "grad_norm": 26.636728286743164, "learning_rate": 9.513793036766345e-06, "loss": 12.51790714263916, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.2815452441573143, "loss_total_v6": 0.7823692262172699, "step": 1304 }, { "epoch": 0.300972972972973, "grad_norm": 23.68447494506836, "learning_rate": 9.512926421749305e-06, "loss": 13.23094367980957, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.3152761831879616, "loss_total_v6": 0.8269339352846146, "step": 1305 }, { "epoch": 0.3012036036036036, "grad_norm": 24.36134147644043, "learning_rate": 9.512059074634995e-06, "loss": 13.076333999633789, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.30449929274618626, "loss_total_v6": 0.8172708749771118, "step": 1306 }, { "epoch": 0.3014342342342342, "grad_norm": 26.265722274780273, "learning_rate": 9.511190995564117e-06, "loss": 12.656953811645508, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2813550475984812, "loss_total_v6": 0.7910596206784248, "step": 1307 }, { "epoch": 0.3016648648648649, "grad_norm": 24.623003005981445, "learning_rate": 9.510322184677494e-06, "loss": 12.60630989074707, "loss_alignment": 1.0478515625, "loss_alignment_w": 0.52392578125, "loss_sft": 0.26419743709266186, "loss_total_v6": 0.7878943607211113, "step": 1308 }, { "epoch": 0.3018954954954955, "grad_norm": 25.77446174621582, "learning_rate": 9.509452642116068e-06, "loss": 13.323514938354492, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.31622491031885147, "loss_total_v6": 0.8327196761965752, "step": 1309 }, { "epoch": 0.3021261261261261, "grad_norm": 26.47726058959961, "learning_rate": 9.508582368020897e-06, "loss": 12.734515190124512, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.289559543132782, "loss_total_v6": 0.7959072068333626, "step": 1310 }, { "epoch": 0.3023567567567568, "grad_norm": 25.50837516784668, "learning_rate": 9.507711362533161e-06, "loss": 12.64854621887207, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2858801670372486, "loss_total_v6": 0.790534108877182, "step": 1311 }, { "epoch": 0.3025873873873874, "grad_norm": 25.207048416137695, "learning_rate": 9.506839625794152e-06, "loss": 13.258289337158203, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.3138268459588289, "loss_total_v6": 0.8286430910229683, "step": 1312 }, { "epoch": 0.302818018018018, "grad_norm": 24.80887222290039, "learning_rate": 9.505967157945288e-06, "loss": 12.635404586791992, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2867830619215965, "loss_total_v6": 0.7897127494215965, "step": 1313 }, { "epoch": 0.30304864864864867, "grad_norm": 26.074886322021484, "learning_rate": 9.505093959128102e-06, "loss": 12.946928024291992, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2955874763429165, "loss_total_v6": 0.8091830387711525, "step": 1314 }, { "epoch": 0.3032792792792793, "grad_norm": 21.79716682434082, "learning_rate": 9.504220029484247e-06, "loss": 12.381659507751465, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2584423590451479, "loss_total_v6": 0.7738537192344666, "step": 1315 }, { "epoch": 0.3035099099099099, "grad_norm": 23.6180419921875, "learning_rate": 9.503345369155494e-06, "loss": 12.38829231262207, "loss_alignment": 0.9873046875, "loss_alignment_w": 0.49365234375, "loss_sft": 0.27925786934792995, "loss_total_v6": 0.7742682695388794, "step": 1316 }, { "epoch": 0.30374054054054056, "grad_norm": 24.7965087890625, "learning_rate": 9.502469978283733e-06, "loss": 12.824020385742188, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2877683565020561, "loss_total_v6": 0.8015012666583061, "step": 1317 }, { "epoch": 0.3039711711711712, "grad_norm": 24.417253494262695, "learning_rate": 9.501593857010968e-06, "loss": 13.203972816467285, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.3104930445551872, "loss_total_v6": 0.8252483010292053, "step": 1318 }, { "epoch": 0.3042018018018018, "grad_norm": 23.857975006103516, "learning_rate": 9.50071700547933e-06, "loss": 12.453150749206543, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2728440426290035, "loss_total_v6": 0.7783219218254089, "step": 1319 }, { "epoch": 0.30443243243243245, "grad_norm": 23.831241607666016, "learning_rate": 9.499839423831062e-06, "loss": 12.556783676147461, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2817167043685913, "loss_total_v6": 0.7847989723086357, "step": 1320 }, { "epoch": 0.30466306306306307, "grad_norm": 26.466297149658203, "learning_rate": 9.498961112208526e-06, "loss": 12.567760467529297, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.26837994903326035, "loss_total_v6": 0.7854850590229034, "step": 1321 }, { "epoch": 0.3048936936936937, "grad_norm": 22.061723709106445, "learning_rate": 9.498082070754206e-06, "loss": 12.605134010314941, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2793064434081316, "loss_total_v6": 0.7878208607435226, "step": 1322 }, { "epoch": 0.30512432432432435, "grad_norm": 25.07915496826172, "learning_rate": 9.497202299610702e-06, "loss": 12.77527141571045, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.28133408911526203, "loss_total_v6": 0.7984544709324837, "step": 1323 }, { "epoch": 0.30535495495495496, "grad_norm": 23.993824005126953, "learning_rate": 9.49632179892073e-06, "loss": 13.218404769897461, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.3232664167881012, "loss_total_v6": 0.8261503502726555, "step": 1324 }, { "epoch": 0.3055855855855856, "grad_norm": 23.153940200805664, "learning_rate": 9.49544056882713e-06, "loss": 13.015107154846191, "loss_alignment": 1.02685546875, "loss_alignment_w": 0.513427734375, "loss_sft": 0.3001232333481312, "loss_total_v6": 0.8134441897273064, "step": 1325 }, { "epoch": 0.30581621621621624, "grad_norm": 26.5958251953125, "learning_rate": 9.494558609472856e-06, "loss": 13.504666328430176, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.325517401099205, "loss_total_v6": 0.8440416157245636, "step": 1326 }, { "epoch": 0.30604684684684685, "grad_norm": 23.71669578552246, "learning_rate": 9.49367592100098e-06, "loss": 12.613167762756348, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.27724510431289673, "loss_total_v6": 0.7883229628205299, "step": 1327 }, { "epoch": 0.30627747747747747, "grad_norm": 26.205284118652344, "learning_rate": 9.492792503554695e-06, "loss": 13.09041976928711, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.302495738491416, "loss_total_v6": 0.8181512504816055, "step": 1328 }, { "epoch": 0.30650810810810813, "grad_norm": 23.19664192199707, "learning_rate": 9.491908357277313e-06, "loss": 12.86247444152832, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2944289315491915, "loss_total_v6": 0.8039046376943588, "step": 1329 }, { "epoch": 0.30673873873873875, "grad_norm": 22.890716552734375, "learning_rate": 9.491023482312258e-06, "loss": 12.579023361206055, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.28031432442367077, "loss_total_v6": 0.7861889451742172, "step": 1330 }, { "epoch": 0.30696936936936936, "grad_norm": 24.12767791748047, "learning_rate": 9.490137878803078e-06, "loss": 12.880626678466797, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2920691855251789, "loss_total_v6": 0.8050391599535942, "step": 1331 }, { "epoch": 0.3072, "grad_norm": 23.374431610107422, "learning_rate": 9.489251546893441e-06, "loss": 13.435665130615234, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.3215101081877947, "loss_total_v6": 0.8397291153669357, "step": 1332 }, { "epoch": 0.30743063063063064, "grad_norm": 26.290870666503906, "learning_rate": 9.488364486727126e-06, "loss": 12.830852508544922, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.3009669203311205, "loss_total_v6": 0.8019282594323158, "step": 1333 }, { "epoch": 0.30766126126126125, "grad_norm": 24.669816970825195, "learning_rate": 9.487476698448033e-06, "loss": 12.565765380859375, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2724819052964449, "loss_total_v6": 0.7853603288531303, "step": 1334 }, { "epoch": 0.30789189189189187, "grad_norm": 24.608537673950195, "learning_rate": 9.486588182200184e-06, "loss": 12.634767532348633, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.27545178309082985, "loss_total_v6": 0.7896729856729507, "step": 1335 }, { "epoch": 0.30812252252252254, "grad_norm": 24.219144821166992, "learning_rate": 9.485698938127716e-06, "loss": 12.774715423583984, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2847936861217022, "loss_total_v6": 0.7984197363257408, "step": 1336 }, { "epoch": 0.30835315315315315, "grad_norm": 23.612756729125977, "learning_rate": 9.484808966374883e-06, "loss": 13.179401397705078, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.30819443613290787, "loss_total_v6": 0.8237126246094704, "step": 1337 }, { "epoch": 0.30858378378378376, "grad_norm": 22.831298828125, "learning_rate": 9.483918267086058e-06, "loss": 12.656156539916992, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.28049653209745884, "loss_total_v6": 0.7910098060965538, "step": 1338 }, { "epoch": 0.30881441441441443, "grad_norm": 23.47891616821289, "learning_rate": 9.483026840405734e-06, "loss": 12.543663024902344, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2714514434337616, "loss_total_v6": 0.7839789092540741, "step": 1339 }, { "epoch": 0.30904504504504504, "grad_norm": 25.003398895263672, "learning_rate": 9.48213468647852e-06, "loss": 12.184915542602539, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.2628389187157154, "loss_total_v6": 0.7615571916103363, "step": 1340 }, { "epoch": 0.30927567567567565, "grad_norm": 23.781089782714844, "learning_rate": 9.48124180544914e-06, "loss": 12.458513259887695, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.27395738288760185, "loss_total_v6": 0.7786570936441422, "step": 1341 }, { "epoch": 0.3095063063063063, "grad_norm": 25.52983856201172, "learning_rate": 9.480348197462443e-06, "loss": 13.03332805633545, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.3054429367184639, "loss_total_v6": 0.8145829811692238, "step": 1342 }, { "epoch": 0.30973693693693694, "grad_norm": 23.35460662841797, "learning_rate": 9.479453862663391e-06, "loss": 12.509260177612305, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.26888926699757576, "loss_total_v6": 0.7818287163972855, "step": 1343 }, { "epoch": 0.30996756756756755, "grad_norm": 29.829055786132812, "learning_rate": 9.478558801197065e-06, "loss": 12.807511329650879, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2928705532103777, "loss_total_v6": 0.8004694730043411, "step": 1344 }, { "epoch": 0.3101981981981982, "grad_norm": 24.921796798706055, "learning_rate": 9.477663013208663e-06, "loss": 13.208821296691895, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.32420854829251766, "loss_total_v6": 0.8255513086915016, "step": 1345 }, { "epoch": 0.31042882882882883, "grad_norm": 26.45315933227539, "learning_rate": 9.476766498843504e-06, "loss": 12.91246509552002, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2913887891918421, "loss_total_v6": 0.8070290610194206, "step": 1346 }, { "epoch": 0.31065945945945944, "grad_norm": 25.620826721191406, "learning_rate": 9.475869258247019e-06, "loss": 12.504735946655273, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.28073724173009396, "loss_total_v6": 0.7815459668636322, "step": 1347 }, { "epoch": 0.3108900900900901, "grad_norm": 25.971664428710938, "learning_rate": 9.474971291564764e-06, "loss": 12.36679744720459, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2643646728247404, "loss_total_v6": 0.7729248479008675, "step": 1348 }, { "epoch": 0.3111207207207207, "grad_norm": 25.472305297851562, "learning_rate": 9.474072598942408e-06, "loss": 12.717235565185547, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2870757896453142, "loss_total_v6": 0.7948272451758385, "step": 1349 }, { "epoch": 0.31135135135135134, "grad_norm": 25.761737823486328, "learning_rate": 9.473173180525737e-06, "loss": 12.588495254516602, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2714611943811178, "loss_total_v6": 0.7867809906601906, "step": 1350 }, { "epoch": 0.311581981981982, "grad_norm": 27.404651641845703, "learning_rate": 9.472273036460659e-06, "loss": 12.612202644348145, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2734921835362911, "loss_total_v6": 0.7882626801729202, "step": 1351 }, { "epoch": 0.3118126126126126, "grad_norm": 23.84276008605957, "learning_rate": 9.4713721668932e-06, "loss": 12.351744651794434, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.2552603669464588, "loss_total_v6": 0.7719840109348297, "step": 1352 }, { "epoch": 0.31204324324324323, "grad_norm": 22.44334602355957, "learning_rate": 9.470470571969494e-06, "loss": 13.070899963378906, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.30325939133763313, "loss_total_v6": 0.8169312626123428, "step": 1353 }, { "epoch": 0.3122738738738739, "grad_norm": 26.5201416015625, "learning_rate": 9.469568251835806e-06, "loss": 13.171692848205566, "loss_alignment": 1.03564453125, "loss_alignment_w": 0.517822265625, "loss_sft": 0.3048592023551464, "loss_total_v6": 0.8232308030128479, "step": 1354 }, { "epoch": 0.3125045045045045, "grad_norm": 21.3040714263916, "learning_rate": 9.46866520663851e-06, "loss": 12.471529006958008, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.26424227468669415, "loss_total_v6": 0.7794705554842949, "step": 1355 }, { "epoch": 0.3127351351351351, "grad_norm": 24.883392333984375, "learning_rate": 9.4677614365241e-06, "loss": 12.660056114196777, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2832121178507805, "loss_total_v6": 0.7912534773349762, "step": 1356 }, { "epoch": 0.3129657657657658, "grad_norm": 24.057323455810547, "learning_rate": 9.46685694163919e-06, "loss": 12.707952499389648, "loss_alignment": 1.0439453125, "loss_alignment_w": 0.52197265625, "loss_sft": 0.2739986181259155, "loss_total_v6": 0.7942470461130142, "step": 1357 }, { "epoch": 0.3131963963963964, "grad_norm": 24.16899299621582, "learning_rate": 9.465951722130504e-06, "loss": 12.853246688842773, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.28198088705539703, "loss_total_v6": 0.8033279404044151, "step": 1358 }, { "epoch": 0.313427027027027, "grad_norm": 25.17423439025879, "learning_rate": 9.465045778144894e-06, "loss": 13.085212707519531, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.3019566610455513, "loss_total_v6": 0.8178258165717125, "step": 1359 }, { "epoch": 0.3136576576576577, "grad_norm": 24.511680603027344, "learning_rate": 9.46413910982932e-06, "loss": 12.587544441223145, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.27208830788731575, "loss_total_v6": 0.7867215275764465, "step": 1360 }, { "epoch": 0.3138882882882883, "grad_norm": 25.806005477905273, "learning_rate": 9.463231717330869e-06, "loss": 13.227322578430176, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.31350881420075893, "loss_total_v6": 0.8267076388001442, "step": 1361 }, { "epoch": 0.3141189189189189, "grad_norm": 26.142044067382812, "learning_rate": 9.462323600796736e-06, "loss": 13.146330833435059, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.31479448452591896, "loss_total_v6": 0.82164566218853, "step": 1362 }, { "epoch": 0.3143495495495495, "grad_norm": 27.580856323242188, "learning_rate": 9.461414760374242e-06, "loss": 12.85869026184082, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2913694977760315, "loss_total_v6": 0.8036680892109871, "step": 1363 }, { "epoch": 0.3145801801801802, "grad_norm": 24.50928497314453, "learning_rate": 9.460505196210814e-06, "loss": 12.950852394104004, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.29972369596362114, "loss_total_v6": 0.809428296983242, "step": 1364 }, { "epoch": 0.3148108108108108, "grad_norm": 23.622644424438477, "learning_rate": 9.45959490845401e-06, "loss": 13.20009994506836, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.32219860330224037, "loss_total_v6": 0.8250062018632889, "step": 1365 }, { "epoch": 0.3150414414414414, "grad_norm": 25.84944725036621, "learning_rate": 9.458683897251498e-06, "loss": 12.49705696105957, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.2757407631725073, "loss_total_v6": 0.7810660898685455, "step": 1366 }, { "epoch": 0.3152720720720721, "grad_norm": 24.245210647583008, "learning_rate": 9.457772162751061e-06, "loss": 12.581974029541016, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.28249758668243885, "loss_total_v6": 0.7863733693957329, "step": 1367 }, { "epoch": 0.3155027027027027, "grad_norm": 23.154678344726562, "learning_rate": 9.456859705100605e-06, "loss": 12.787439346313477, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.28774039819836617, "loss_total_v6": 0.7992150038480759, "step": 1368 }, { "epoch": 0.3157333333333333, "grad_norm": 22.522579193115234, "learning_rate": 9.455946524448153e-06, "loss": 12.733711242675781, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2823834288865328, "loss_total_v6": 0.7958569526672363, "step": 1369 }, { "epoch": 0.315963963963964, "grad_norm": 26.117847442626953, "learning_rate": 9.45503262094184e-06, "loss": 12.430655479431152, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.27091928012669086, "loss_total_v6": 0.7769159898161888, "step": 1370 }, { "epoch": 0.3161945945945946, "grad_norm": 28.430950164794922, "learning_rate": 9.454117994729923e-06, "loss": 12.857101440429688, "loss_alignment": 1.0458984375, "loss_alignment_w": 0.52294921875, "loss_sft": 0.2789258863776922, "loss_total_v6": 0.8035688623785973, "step": 1371 }, { "epoch": 0.3164252252252252, "grad_norm": 24.98025131225586, "learning_rate": 9.453202645960775e-06, "loss": 13.334758758544922, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.3204372301697731, "loss_total_v6": 0.8334224373102188, "step": 1372 }, { "epoch": 0.3166558558558559, "grad_norm": 23.297561645507812, "learning_rate": 9.452286574782883e-06, "loss": 13.249396324157715, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.3162769079208374, "loss_total_v6": 0.8280872404575348, "step": 1373 }, { "epoch": 0.3168864864864865, "grad_norm": 22.1562442779541, "learning_rate": 9.451369781344859e-06, "loss": 12.601882934570312, "loss_alignment": 1.02587890625, "loss_alignment_w": 0.512939453125, "loss_sft": 0.2750292494893074, "loss_total_v6": 0.7876177281141281, "step": 1374 }, { "epoch": 0.3171171171171171, "grad_norm": 23.77452850341797, "learning_rate": 9.450452265795423e-06, "loss": 13.035186767578125, "loss_alignment": 1.02294921875, "loss_alignment_w": 0.511474609375, "loss_sft": 0.3046283256262541, "loss_total_v6": 0.8146991580724716, "step": 1375 }, { "epoch": 0.31734774774774777, "grad_norm": 23.681251525878906, "learning_rate": 9.44953402828342e-06, "loss": 12.85354232788086, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.29614426009356976, "loss_total_v6": 0.8033464252948761, "step": 1376 }, { "epoch": 0.3175783783783784, "grad_norm": 27.172304153442383, "learning_rate": 9.448615068957807e-06, "loss": 13.164260864257812, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.3091860394924879, "loss_total_v6": 0.8227663412690163, "step": 1377 }, { "epoch": 0.317809009009009, "grad_norm": 24.40294075012207, "learning_rate": 9.447695387967657e-06, "loss": 12.918039321899414, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.2913862932473421, "loss_total_v6": 0.807377465069294, "step": 1378 }, { "epoch": 0.31803963963963966, "grad_norm": 25.390350341796875, "learning_rate": 9.446774985462168e-06, "loss": 12.805377006530762, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.2954990305006504, "loss_total_v6": 0.8003360778093338, "step": 1379 }, { "epoch": 0.3182702702702703, "grad_norm": 25.430994033813477, "learning_rate": 9.445853861590647e-06, "loss": 12.592558860778809, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.28540220856666565, "loss_total_v6": 0.7870349287986755, "step": 1380 }, { "epoch": 0.3185009009009009, "grad_norm": 22.168170928955078, "learning_rate": 9.44493201650252e-06, "loss": 12.942811965942383, "loss_alignment": 1.02490234375, "loss_alignment_w": 0.512451171875, "loss_sft": 0.2975884545594454, "loss_total_v6": 0.8089257106184959, "step": 1381 }, { "epoch": 0.31873153153153155, "grad_norm": 26.74012565612793, "learning_rate": 9.444009450347334e-06, "loss": 12.383636474609375, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2599239069968462, "loss_total_v6": 0.7739772647619247, "step": 1382 }, { "epoch": 0.31896216216216217, "grad_norm": 27.392004013061523, "learning_rate": 9.443086163274745e-06, "loss": 12.902128219604492, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.296464866027236, "loss_total_v6": 0.8063830435276031, "step": 1383 }, { "epoch": 0.3191927927927928, "grad_norm": 26.8750057220459, "learning_rate": 9.442162155434534e-06, "loss": 12.940376281738281, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.3042416535317898, "loss_total_v6": 0.8087735399603844, "step": 1384 }, { "epoch": 0.31942342342342345, "grad_norm": 24.513803482055664, "learning_rate": 9.441237426976596e-06, "loss": 12.490474700927734, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2730405144393444, "loss_total_v6": 0.780654639005661, "step": 1385 }, { "epoch": 0.31965405405405406, "grad_norm": 26.901948928833008, "learning_rate": 9.440311978050942e-06, "loss": 12.870630264282227, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.291703874245286, "loss_total_v6": 0.8044144287705421, "step": 1386 }, { "epoch": 0.3198846846846847, "grad_norm": 26.552751541137695, "learning_rate": 9.439385808807698e-06, "loss": 12.42088508605957, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.2748863063752651, "loss_total_v6": 0.7763053551316261, "step": 1387 }, { "epoch": 0.32011531531531534, "grad_norm": 26.403467178344727, "learning_rate": 9.438458919397113e-06, "loss": 12.654956817626953, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.27976540848612785, "loss_total_v6": 0.7909348458051682, "step": 1388 }, { "epoch": 0.32034594594594595, "grad_norm": 25.800952911376953, "learning_rate": 9.437531309969548e-06, "loss": 12.811107635498047, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.3013656623661518, "loss_total_v6": 0.8006942644715309, "step": 1389 }, { "epoch": 0.32057657657657657, "grad_norm": 23.789199829101562, "learning_rate": 9.43660298067548e-06, "loss": 12.825723648071289, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2903161831200123, "loss_total_v6": 0.8016077056527138, "step": 1390 }, { "epoch": 0.32080720720720723, "grad_norm": 22.454349517822266, "learning_rate": 9.435673931665508e-06, "loss": 13.056510925292969, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.30611368641257286, "loss_total_v6": 0.8160318955779076, "step": 1391 }, { "epoch": 0.32103783783783785, "grad_norm": 25.980533599853516, "learning_rate": 9.434744163090341e-06, "loss": 12.737001419067383, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.28419122472405434, "loss_total_v6": 0.7960625812411308, "step": 1392 }, { "epoch": 0.32126846846846846, "grad_norm": 25.546092987060547, "learning_rate": 9.433813675100809e-06, "loss": 12.765203475952148, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2859691046178341, "loss_total_v6": 0.7978251725435257, "step": 1393 }, { "epoch": 0.3214990990990991, "grad_norm": 25.169456481933594, "learning_rate": 9.43288246784786e-06, "loss": 12.375356674194336, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.27403962053358555, "loss_total_v6": 0.7734597623348236, "step": 1394 }, { "epoch": 0.32172972972972974, "grad_norm": 24.89133071899414, "learning_rate": 9.431950541482554e-06, "loss": 13.047504425048828, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.3051387779414654, "loss_total_v6": 0.8154689818620682, "step": 1395 }, { "epoch": 0.32196036036036035, "grad_norm": 24.335283279418945, "learning_rate": 9.431017896156074e-06, "loss": 13.125232696533203, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.3075096234679222, "loss_total_v6": 0.8203270137310028, "step": 1396 }, { "epoch": 0.32219099099099097, "grad_norm": 24.34799575805664, "learning_rate": 9.430084532019711e-06, "loss": 12.5639009475708, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.277431296184659, "loss_total_v6": 0.7852438166737556, "step": 1397 }, { "epoch": 0.32242162162162163, "grad_norm": 24.186128616333008, "learning_rate": 9.429150449224882e-06, "loss": 12.431947708129883, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.26524750515818596, "loss_total_v6": 0.7769967466592789, "step": 1398 }, { "epoch": 0.32265225225225225, "grad_norm": 25.860336303710938, "learning_rate": 9.428215647923113e-06, "loss": 12.683849334716797, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.2906043194234371, "loss_total_v6": 0.7927405685186386, "step": 1399 }, { "epoch": 0.32288288288288286, "grad_norm": 27.198793411254883, "learning_rate": 9.427280128266049e-06, "loss": 12.515374183654785, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.27737387269735336, "loss_total_v6": 0.7822109088301659, "step": 1400 }, { "epoch": 0.32311351351351353, "grad_norm": 23.592992782592773, "learning_rate": 9.426343890405456e-06, "loss": 12.557035446166992, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2709902860224247, "loss_total_v6": 0.7848147079348564, "step": 1401 }, { "epoch": 0.32334414414414414, "grad_norm": 24.985544204711914, "learning_rate": 9.425406934493208e-06, "loss": 12.935677528381348, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.29746295884251595, "loss_total_v6": 0.8084798157215118, "step": 1402 }, { "epoch": 0.32357477477477475, "grad_norm": 25.48567008972168, "learning_rate": 9.424469260681304e-06, "loss": 13.37214183807373, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.32001177966594696, "loss_total_v6": 0.8357588648796082, "step": 1403 }, { "epoch": 0.3238054054054054, "grad_norm": 25.22052001953125, "learning_rate": 9.423530869121855e-06, "loss": 13.241998672485352, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.3180118389427662, "loss_total_v6": 0.8276248872280121, "step": 1404 }, { "epoch": 0.32403603603603603, "grad_norm": 25.10825538635254, "learning_rate": 9.422591759967085e-06, "loss": 13.078874588012695, "loss_alignment": 1.03271484375, "loss_alignment_w": 0.516357421875, "loss_sft": 0.30159099586308, "loss_total_v6": 0.8174296468496323, "step": 1405 }, { "epoch": 0.32426666666666665, "grad_norm": 23.89163589477539, "learning_rate": 9.421651933369344e-06, "loss": 12.657594680786133, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2771988920867443, "loss_total_v6": 0.7910996451973915, "step": 1406 }, { "epoch": 0.3244972972972973, "grad_norm": 25.73100471496582, "learning_rate": 9.42071138948109e-06, "loss": 13.066267967224121, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.3057622779160738, "loss_total_v6": 0.8166417703032494, "step": 1407 }, { "epoch": 0.32472792792792793, "grad_norm": 22.904254913330078, "learning_rate": 9.4197701284549e-06, "loss": 12.513298988342285, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2677379082888365, "loss_total_v6": 0.7820811718702316, "step": 1408 }, { "epoch": 0.32495855855855854, "grad_norm": 25.387741088867188, "learning_rate": 9.418828150443469e-06, "loss": 12.472827911376953, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.269023172557354, "loss_total_v6": 0.7795517519116402, "step": 1409 }, { "epoch": 0.3251891891891892, "grad_norm": 25.73078727722168, "learning_rate": 9.417885455599604e-06, "loss": 13.257633209228516, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.3167460113763809, "loss_total_v6": 0.8286020457744598, "step": 1410 }, { "epoch": 0.3254198198198198, "grad_norm": 26.21857452392578, "learning_rate": 9.416942044076233e-06, "loss": 12.695039749145508, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.28272825106978416, "loss_total_v6": 0.7934399396181107, "step": 1411 }, { "epoch": 0.32565045045045043, "grad_norm": 22.25879669189453, "learning_rate": 9.415997916026401e-06, "loss": 12.759807586669922, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.28895826265215874, "loss_total_v6": 0.7974879443645477, "step": 1412 }, { "epoch": 0.3258810810810811, "grad_norm": 24.226919174194336, "learning_rate": 9.415053071603262e-06, "loss": 12.564733505249023, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.27577437460422516, "loss_total_v6": 0.7852958515286446, "step": 1413 }, { "epoch": 0.3261117117117117, "grad_norm": 24.817852020263672, "learning_rate": 9.414107510960092e-06, "loss": 12.435956954956055, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.26685608737170696, "loss_total_v6": 0.7772473245859146, "step": 1414 }, { "epoch": 0.32634234234234233, "grad_norm": 22.879806518554688, "learning_rate": 9.413161234250284e-06, "loss": 13.059414863586426, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.30985056050121784, "loss_total_v6": 0.8162134513258934, "step": 1415 }, { "epoch": 0.326572972972973, "grad_norm": 23.811161041259766, "learning_rate": 9.412214241627344e-06, "loss": 12.920503616333008, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.30148895829916, "loss_total_v6": 0.80753143876791, "step": 1416 }, { "epoch": 0.3268036036036036, "grad_norm": 23.272212982177734, "learning_rate": 9.411266533244894e-06, "loss": 12.532968521118164, "loss_alignment": 0.97900390625, "loss_alignment_w": 0.489501953125, "loss_sft": 0.294373108074069, "loss_total_v6": 0.7833104953169823, "step": 1417 }, { "epoch": 0.3270342342342342, "grad_norm": 24.511497497558594, "learning_rate": 9.410318109256677e-06, "loss": 12.811179161071777, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.2784055806696415, "loss_total_v6": 0.8006986826658249, "step": 1418 }, { "epoch": 0.3272648648648649, "grad_norm": 24.89691925048828, "learning_rate": 9.409368969816545e-06, "loss": 12.37376594543457, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.2699118945747614, "loss_total_v6": 0.77336036413908, "step": 1419 }, { "epoch": 0.3274954954954955, "grad_norm": 24.886714935302734, "learning_rate": 9.40841911507847e-06, "loss": 12.679794311523438, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.28435415402054787, "loss_total_v6": 0.7924871072173119, "step": 1420 }, { "epoch": 0.3277261261261261, "grad_norm": 23.165573120117188, "learning_rate": 9.407468545196541e-06, "loss": 12.496376991271973, "loss_alignment": 1.01708984375, "loss_alignment_w": 0.508544921875, "loss_sft": 0.2745386194437742, "loss_total_v6": 0.7810235843062401, "step": 1421 }, { "epoch": 0.3279567567567568, "grad_norm": 26.77147674560547, "learning_rate": 9.406517260324962e-06, "loss": 12.33230209350586, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.26661851815879345, "loss_total_v6": 0.7707688957452774, "step": 1422 }, { "epoch": 0.3281873873873874, "grad_norm": 25.778776168823242, "learning_rate": 9.405565260618051e-06, "loss": 12.93847370147705, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.30470260977745056, "loss_total_v6": 0.8086546063423157, "step": 1423 }, { "epoch": 0.328418018018018, "grad_norm": 27.239959716796875, "learning_rate": 9.404612546230244e-06, "loss": 12.336356163024902, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.2601121962070465, "loss_total_v6": 0.771022230386734, "step": 1424 }, { "epoch": 0.3286486486486486, "grad_norm": 23.43678855895996, "learning_rate": 9.403659117316093e-06, "loss": 12.864624977111816, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.29155739583075047, "loss_total_v6": 0.8040390536189079, "step": 1425 }, { "epoch": 0.3288792792792793, "grad_norm": 23.545053482055664, "learning_rate": 9.402704974030266e-06, "loss": 12.712575912475586, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.29314747266471386, "loss_total_v6": 0.7945360094308853, "step": 1426 }, { "epoch": 0.3291099099099099, "grad_norm": 25.689058303833008, "learning_rate": 9.401750116527544e-06, "loss": 12.48377513885498, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.2636649180203676, "loss_total_v6": 0.7802359461784363, "step": 1427 }, { "epoch": 0.3293405405405405, "grad_norm": 27.36534881591797, "learning_rate": 9.400794544962828e-06, "loss": 12.805744171142578, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2883045319467783, "loss_total_v6": 0.8003589883446693, "step": 1428 }, { "epoch": 0.3295711711711712, "grad_norm": 26.58898162841797, "learning_rate": 9.399838259491134e-06, "loss": 12.309513092041016, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.26424811594188213, "loss_total_v6": 0.7693445459008217, "step": 1429 }, { "epoch": 0.3298018018018018, "grad_norm": 24.839580535888672, "learning_rate": 9.398881260267589e-06, "loss": 12.491518020629883, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2710763607174158, "loss_total_v6": 0.7807198911905289, "step": 1430 }, { "epoch": 0.3300324324324324, "grad_norm": 26.394062042236328, "learning_rate": 9.397923547447445e-06, "loss": 12.915607452392578, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2955524493008852, "loss_total_v6": 0.8072254434227943, "step": 1431 }, { "epoch": 0.3302630630630631, "grad_norm": 24.704174041748047, "learning_rate": 9.396965121186058e-06, "loss": 13.129903793334961, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.31015150249004364, "loss_total_v6": 0.820619024336338, "step": 1432 }, { "epoch": 0.3304936936936937, "grad_norm": 24.14802360534668, "learning_rate": 9.396005981638913e-06, "loss": 12.565366744995117, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.27123637311160564, "loss_total_v6": 0.785335473716259, "step": 1433 }, { "epoch": 0.3307243243243243, "grad_norm": 28.85645294189453, "learning_rate": 9.3950461289616e-06, "loss": 12.662227630615234, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.27736641466617584, "loss_total_v6": 0.7913892269134521, "step": 1434 }, { "epoch": 0.33095495495495497, "grad_norm": 25.687780380249023, "learning_rate": 9.394085563309827e-06, "loss": 13.162622451782227, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.30580298602581024, "loss_total_v6": 0.8226639479398727, "step": 1435 }, { "epoch": 0.3311855855855856, "grad_norm": 38.137054443359375, "learning_rate": 9.393124284839424e-06, "loss": 13.068300247192383, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.3025628700852394, "loss_total_v6": 0.8167687803506851, "step": 1436 }, { "epoch": 0.3314162162162162, "grad_norm": 27.067033767700195, "learning_rate": 9.392162293706325e-06, "loss": 12.856430053710938, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.29386798851192, "loss_total_v6": 0.8035268187522888, "step": 1437 }, { "epoch": 0.33164684684684687, "grad_norm": 26.442108154296875, "learning_rate": 9.391199590066593e-06, "loss": 12.615437507629395, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.27747853472828865, "loss_total_v6": 0.7884648516774178, "step": 1438 }, { "epoch": 0.3318774774774775, "grad_norm": 25.342878341674805, "learning_rate": 9.390236174076396e-06, "loss": 12.704004287719727, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.28446354903280735, "loss_total_v6": 0.7940002754330635, "step": 1439 }, { "epoch": 0.3321081081081081, "grad_norm": 26.01539421081543, "learning_rate": 9.389272045892023e-06, "loss": 12.789740562438965, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.29264491237699986, "loss_total_v6": 0.7993587926030159, "step": 1440 }, { "epoch": 0.33233873873873876, "grad_norm": 21.984638214111328, "learning_rate": 9.388307205669878e-06, "loss": 12.372140884399414, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.2569319363683462, "loss_total_v6": 0.7732588276267052, "step": 1441 }, { "epoch": 0.33256936936936937, "grad_norm": 23.216827392578125, "learning_rate": 9.387341653566478e-06, "loss": 12.638542175292969, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2741007972508669, "loss_total_v6": 0.7899088934063911, "step": 1442 }, { "epoch": 0.3328, "grad_norm": 24.780860900878906, "learning_rate": 9.386375389738455e-06, "loss": 12.837329864501953, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.29943390376865864, "loss_total_v6": 0.8023330569267273, "step": 1443 }, { "epoch": 0.33303063063063065, "grad_norm": 23.806615829467773, "learning_rate": 9.385408414342565e-06, "loss": 12.917972564697266, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.3007203955203295, "loss_total_v6": 0.8073732703924179, "step": 1444 }, { "epoch": 0.33326126126126127, "grad_norm": 24.13672637939453, "learning_rate": 9.384440727535666e-06, "loss": 12.466960906982422, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.26711534708738327, "loss_total_v6": 0.7791850492358208, "step": 1445 }, { "epoch": 0.3334918918918919, "grad_norm": 22.28155517578125, "learning_rate": 9.383472329474743e-06, "loss": 13.09046745300293, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.30347529239952564, "loss_total_v6": 0.8181542381644249, "step": 1446 }, { "epoch": 0.33372252252252255, "grad_norm": 22.87697410583496, "learning_rate": 9.38250322031689e-06, "loss": 12.231142044067383, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.24395380355417728, "loss_total_v6": 0.7644463703036308, "step": 1447 }, { "epoch": 0.33395315315315316, "grad_norm": 24.757488250732422, "learning_rate": 9.381533400219319e-06, "loss": 12.766444206237793, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.29480520635843277, "loss_total_v6": 0.7979027703404427, "step": 1448 }, { "epoch": 0.33418378378378377, "grad_norm": 23.155874252319336, "learning_rate": 9.380562869339354e-06, "loss": 12.924514770507812, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.30347919277846813, "loss_total_v6": 0.8077821582555771, "step": 1449 }, { "epoch": 0.33441441441441444, "grad_norm": 24.44606590270996, "learning_rate": 9.37959162783444e-06, "loss": 13.162282943725586, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.31016100384294987, "loss_total_v6": 0.8226426765322685, "step": 1450 }, { "epoch": 0.33464504504504505, "grad_norm": 25.584646224975586, "learning_rate": 9.378619675862131e-06, "loss": 13.259062767028809, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.31739993765950203, "loss_total_v6": 0.8286914378404617, "step": 1451 }, { "epoch": 0.33487567567567567, "grad_norm": 24.91716957092285, "learning_rate": 9.377647013580102e-06, "loss": 12.789227485656738, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.28888968378305435, "loss_total_v6": 0.7993267253041267, "step": 1452 }, { "epoch": 0.3351063063063063, "grad_norm": 22.14035987854004, "learning_rate": 9.376673641146139e-06, "loss": 12.866532325744629, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.28616816736757755, "loss_total_v6": 0.8041582554578781, "step": 1453 }, { "epoch": 0.33533693693693695, "grad_norm": 23.33866310119629, "learning_rate": 9.375699558718146e-06, "loss": 12.461668014526367, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.27986135706305504, "loss_total_v6": 0.7788542658090591, "step": 1454 }, { "epoch": 0.33556756756756756, "grad_norm": 23.957420349121094, "learning_rate": 9.374724766454138e-06, "loss": 12.299007415771484, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.26623125933110714, "loss_total_v6": 0.7686879336833954, "step": 1455 }, { "epoch": 0.3357981981981982, "grad_norm": 25.07638931274414, "learning_rate": 9.37374926451225e-06, "loss": 12.490851402282715, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.26416824385523796, "loss_total_v6": 0.7806782424449921, "step": 1456 }, { "epoch": 0.33602882882882884, "grad_norm": 24.536008834838867, "learning_rate": 9.372773053050733e-06, "loss": 12.974759101867676, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.30323198810219765, "loss_total_v6": 0.8109224289655685, "step": 1457 }, { "epoch": 0.33625945945945945, "grad_norm": 25.373680114746094, "learning_rate": 9.371796132227943e-06, "loss": 12.622715950012207, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.279444083571434, "loss_total_v6": 0.7889197543263435, "step": 1458 }, { "epoch": 0.33649009009009007, "grad_norm": 24.071060180664062, "learning_rate": 9.370818502202366e-06, "loss": 12.398965835571289, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.25523631647229195, "loss_total_v6": 0.7749353870749474, "step": 1459 }, { "epoch": 0.33672072072072073, "grad_norm": 24.394155502319336, "learning_rate": 9.36984016313259e-06, "loss": 13.21689510345459, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.31186531111598015, "loss_total_v6": 0.8260559514164925, "step": 1460 }, { "epoch": 0.33695135135135135, "grad_norm": 25.389869689941406, "learning_rate": 9.368861115177327e-06, "loss": 12.609861373901367, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2713621612638235, "loss_total_v6": 0.7881163582205772, "step": 1461 }, { "epoch": 0.33718198198198196, "grad_norm": 25.53763771057129, "learning_rate": 9.367881358495399e-06, "loss": 12.704744338989258, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2848759889602661, "loss_total_v6": 0.7940465062856674, "step": 1462 }, { "epoch": 0.33741261261261263, "grad_norm": 27.523155212402344, "learning_rate": 9.366900893245744e-06, "loss": 13.245172500610352, "loss_alignment": 1.01708984375, "loss_alignment_w": 0.508544921875, "loss_sft": 0.32062119245529175, "loss_total_v6": 0.8278233259916306, "step": 1463 }, { "epoch": 0.33764324324324324, "grad_norm": 24.7232666015625, "learning_rate": 9.365919719587413e-06, "loss": 12.62982177734375, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2802390977740288, "loss_total_v6": 0.7893638461828232, "step": 1464 }, { "epoch": 0.33787387387387385, "grad_norm": 23.364686965942383, "learning_rate": 9.36493783767958e-06, "loss": 12.293405532836914, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2626310084015131, "loss_total_v6": 0.7683378159999847, "step": 1465 }, { "epoch": 0.3381045045045045, "grad_norm": 25.646907806396484, "learning_rate": 9.363955247681521e-06, "loss": 12.697792053222656, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.28265614062547684, "loss_total_v6": 0.793611966073513, "step": 1466 }, { "epoch": 0.33833513513513513, "grad_norm": 22.813594818115234, "learning_rate": 9.362971949752642e-06, "loss": 12.901481628417969, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2908549774438143, "loss_total_v6": 0.8063426539301872, "step": 1467 }, { "epoch": 0.33856576576576575, "grad_norm": 24.67946434020996, "learning_rate": 9.36198794405245e-06, "loss": 12.582574844360352, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.2833133451640606, "loss_total_v6": 0.7864108979701996, "step": 1468 }, { "epoch": 0.3387963963963964, "grad_norm": 24.1263427734375, "learning_rate": 9.361003230740574e-06, "loss": 12.86676025390625, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2914162054657936, "loss_total_v6": 0.8041725382208824, "step": 1469 }, { "epoch": 0.33902702702702703, "grad_norm": 22.516063690185547, "learning_rate": 9.360017809976758e-06, "loss": 12.516189575195312, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2683610450476408, "loss_total_v6": 0.7822618260979652, "step": 1470 }, { "epoch": 0.33925765765765764, "grad_norm": 23.015365600585938, "learning_rate": 9.359031681920858e-06, "loss": 12.818123817443848, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.29840145260095596, "loss_total_v6": 0.8011327534914017, "step": 1471 }, { "epoch": 0.3394882882882883, "grad_norm": 21.475666046142578, "learning_rate": 9.358044846732848e-06, "loss": 12.936027526855469, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2937922850251198, "loss_total_v6": 0.8085017278790474, "step": 1472 }, { "epoch": 0.3397189189189189, "grad_norm": 28.085060119628906, "learning_rate": 9.35705730457281e-06, "loss": 12.660536766052246, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2747277710586786, "loss_total_v6": 0.7912835329771042, "step": 1473 }, { "epoch": 0.33994954954954953, "grad_norm": 23.13882064819336, "learning_rate": 9.356069055600949e-06, "loss": 12.521856307983398, "loss_alignment": 1.01708984375, "loss_alignment_w": 0.508544921875, "loss_sft": 0.273445475846529, "loss_total_v6": 0.7826159968972206, "step": 1474 }, { "epoch": 0.3401801801801802, "grad_norm": 24.363075256347656, "learning_rate": 9.355080099977579e-06, "loss": 12.890697479248047, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.2862441670149565, "loss_total_v6": 0.8056685999035835, "step": 1475 }, { "epoch": 0.3404108108108108, "grad_norm": 25.07550048828125, "learning_rate": 9.354090437863133e-06, "loss": 12.756586074829102, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.28628501296043396, "loss_total_v6": 0.7972865849733353, "step": 1476 }, { "epoch": 0.34064144144144143, "grad_norm": 24.39599609375, "learning_rate": 9.353100069418153e-06, "loss": 12.527074813842773, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2798141483217478, "loss_total_v6": 0.7829421758651733, "step": 1477 }, { "epoch": 0.3408720720720721, "grad_norm": 24.06687355041504, "learning_rate": 9.352108994803303e-06, "loss": 12.564358711242676, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.278695834800601, "loss_total_v6": 0.7852724194526672, "step": 1478 }, { "epoch": 0.3411027027027027, "grad_norm": 26.251436233520508, "learning_rate": 9.351117214179354e-06, "loss": 12.562060356140137, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.2858153823763132, "loss_total_v6": 0.785128727555275, "step": 1479 }, { "epoch": 0.3413333333333333, "grad_norm": 26.71209144592285, "learning_rate": 9.350124727707197e-06, "loss": 13.172410011291504, "loss_alignment": 1.00732421875, "loss_alignment_w": 0.503662109375, "loss_sft": 0.3189269080758095, "loss_total_v6": 0.8232756331562996, "step": 1480 }, { "epoch": 0.341563963963964, "grad_norm": 23.378440856933594, "learning_rate": 9.349131535547834e-06, "loss": 12.689213752746582, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.28495819866657257, "loss_total_v6": 0.7930758446455002, "step": 1481 }, { "epoch": 0.3417945945945946, "grad_norm": 23.29081916809082, "learning_rate": 9.348137637862385e-06, "loss": 12.727815628051758, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.29081931710243225, "loss_total_v6": 0.7954884842038155, "step": 1482 }, { "epoch": 0.3420252252252252, "grad_norm": 24.935670852661133, "learning_rate": 9.347143034812078e-06, "loss": 13.19704818725586, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.30986188352108, "loss_total_v6": 0.8248155191540718, "step": 1483 }, { "epoch": 0.34225585585585583, "grad_norm": 24.562042236328125, "learning_rate": 9.346147726558266e-06, "loss": 13.248601913452148, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.31573908776044846, "loss_total_v6": 0.8280376493930817, "step": 1484 }, { "epoch": 0.3424864864864865, "grad_norm": 23.30052375793457, "learning_rate": 9.345151713262404e-06, "loss": 12.379745483398438, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2629766147583723, "loss_total_v6": 0.77373406291008, "step": 1485 }, { "epoch": 0.3427171171171171, "grad_norm": 22.530668258666992, "learning_rate": 9.344154995086072e-06, "loss": 12.808365821838379, "loss_alignment": 1.01806640625, "loss_alignment_w": 0.509033203125, "loss_sft": 0.2920084372162819, "loss_total_v6": 0.8005228489637375, "step": 1486 }, { "epoch": 0.3429477477477477, "grad_norm": 24.284690856933594, "learning_rate": 9.343157572190957e-06, "loss": 12.865569114685059, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2923945505172014, "loss_total_v6": 0.804098054766655, "step": 1487 }, { "epoch": 0.3431783783783784, "grad_norm": 30.92436408996582, "learning_rate": 9.342159444738865e-06, "loss": 13.439945220947266, "loss_alignment": 1.03466796875, "loss_alignment_w": 0.517333984375, "loss_sft": 0.3237918019294739, "loss_total_v6": 0.8399966061115265, "step": 1488 }, { "epoch": 0.343409009009009, "grad_norm": 22.817602157592773, "learning_rate": 9.341160612891716e-06, "loss": 13.066478729248047, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.307026581838727, "loss_total_v6": 0.8166549205780029, "step": 1489 }, { "epoch": 0.3436396396396396, "grad_norm": 23.252893447875977, "learning_rate": 9.340161076811539e-06, "loss": 13.322981834411621, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.31787004694342613, "loss_total_v6": 0.8326863795518875, "step": 1490 }, { "epoch": 0.3438702702702703, "grad_norm": 23.260278701782227, "learning_rate": 9.33916083666048e-06, "loss": 13.103103637695312, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.31143671087920666, "loss_total_v6": 0.8189440071582794, "step": 1491 }, { "epoch": 0.3441009009009009, "grad_norm": 25.680278778076172, "learning_rate": 9.338159892600809e-06, "loss": 12.266340255737305, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.2682789135724306, "loss_total_v6": 0.7666462436318398, "step": 1492 }, { "epoch": 0.3443315315315315, "grad_norm": 25.969070434570312, "learning_rate": 9.337158244794891e-06, "loss": 12.57398796081543, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.266770226880908, "loss_total_v6": 0.785874217748642, "step": 1493 }, { "epoch": 0.3445621621621622, "grad_norm": 22.26404571533203, "learning_rate": 9.336155893405224e-06, "loss": 12.557518005371094, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.27196649089455605, "loss_total_v6": 0.784844882786274, "step": 1494 }, { "epoch": 0.3447927927927928, "grad_norm": 21.58669662475586, "learning_rate": 9.335152838594404e-06, "loss": 12.560004234313965, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.27930875681340694, "loss_total_v6": 0.7850002646446228, "step": 1495 }, { "epoch": 0.3450234234234234, "grad_norm": 24.618301391601562, "learning_rate": 9.334149080525154e-06, "loss": 12.778480529785156, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2899880688637495, "loss_total_v6": 0.7986550554633141, "step": 1496 }, { "epoch": 0.34525405405405407, "grad_norm": 25.080791473388672, "learning_rate": 9.333144619360306e-06, "loss": 13.188030242919922, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.3183162212371826, "loss_total_v6": 0.8242519050836563, "step": 1497 }, { "epoch": 0.3454846846846847, "grad_norm": 23.54523277282715, "learning_rate": 9.332139455262803e-06, "loss": 12.804327011108398, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.28777350299060345, "loss_total_v6": 0.8002704381942749, "step": 1498 }, { "epoch": 0.3457153153153153, "grad_norm": 24.93089485168457, "learning_rate": 9.331133588395709e-06, "loss": 12.607963562011719, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.28349632024765015, "loss_total_v6": 0.7879976779222488, "step": 1499 }, { "epoch": 0.34594594594594597, "grad_norm": 26.429443359375, "learning_rate": 9.330127018922195e-06, "loss": 13.060152053833008, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.3000698760151863, "loss_total_v6": 0.8162594512104988, "step": 1500 }, { "epoch": 0.34594594594594597, "eval_loss": 0.9457743167877197, "eval_loss_alignment": 1.0265366367009132, "eval_loss_alignment_w": 0.5132683183504566, "eval_loss_contrast": 0.0, "eval_loss_highlevel": 0.0, "eval_loss_sft": 0.4325060201341023, "eval_loss_total_v6": 0.9457743292733959, "eval_loss_wm_recon": 0.0, "eval_loss_wm_recon_w": 0.0, "eval_runtime": 298.2556, "eval_samples_per_second": 5.851, "eval_steps_per_second": 0.734, "step": 1500 }, { "epoch": 0.3461765765765766, "grad_norm": 25.64781951904297, "learning_rate": 9.329119747005548e-06, "loss": 13.022125244140625, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.30338482931256294, "loss_total_v6": 0.813882865011692, "step": 1501 }, { "epoch": 0.3464072072072072, "grad_norm": 27.887680053710938, "learning_rate": 9.328111772809173e-06, "loss": 12.686065673828125, "loss_alignment": 1.01220703125, "loss_alignment_w": 0.506103515625, "loss_sft": 0.2866229582577944, "loss_total_v6": 0.7928790748119354, "step": 1502 }, { "epoch": 0.34663783783783786, "grad_norm": 23.664491653442383, "learning_rate": 9.327103096496584e-06, "loss": 12.518967628479004, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.2783308643847704, "loss_total_v6": 0.7824354767799377, "step": 1503 }, { "epoch": 0.34686846846846847, "grad_norm": 23.906625747680664, "learning_rate": 9.326093718231413e-06, "loss": 12.549375534057617, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.282367579638958, "loss_total_v6": 0.7843359410762787, "step": 1504 }, { "epoch": 0.3470990990990991, "grad_norm": 23.387958526611328, "learning_rate": 9.325083638177401e-06, "loss": 12.321707725524902, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.25963920913636684, "loss_total_v6": 0.770106740295887, "step": 1505 }, { "epoch": 0.34732972972972975, "grad_norm": 24.073993682861328, "learning_rate": 9.324072856498404e-06, "loss": 12.585527420043945, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2841692976653576, "loss_total_v6": 0.7865954414010048, "step": 1506 }, { "epoch": 0.34756036036036037, "grad_norm": 22.14650535583496, "learning_rate": 9.3230613733584e-06, "loss": 12.394091606140137, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2588683497160673, "loss_total_v6": 0.7746306955814362, "step": 1507 }, { "epoch": 0.347790990990991, "grad_norm": 25.185667037963867, "learning_rate": 9.322049188921467e-06, "loss": 12.944402694702148, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2993663605302572, "loss_total_v6": 0.8090251758694649, "step": 1508 }, { "epoch": 0.34802162162162165, "grad_norm": 21.971546173095703, "learning_rate": 9.321036303351809e-06, "loss": 12.816094398498535, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.29595522955060005, "loss_total_v6": 0.8010058999061584, "step": 1509 }, { "epoch": 0.34825225225225226, "grad_norm": 24.682466506958008, "learning_rate": 9.320022716813735e-06, "loss": 12.76348876953125, "loss_alignment": 1.01611328125, "loss_alignment_w": 0.508056640625, "loss_sft": 0.29069897904992104, "loss_total_v6": 0.7977180555462837, "step": 1510 }, { "epoch": 0.34848288288288287, "grad_norm": 23.6947078704834, "learning_rate": 9.319008429471675e-06, "loss": 12.42659854888916, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.27705912105739117, "loss_total_v6": 0.7766624093055725, "step": 1511 }, { "epoch": 0.34871351351351354, "grad_norm": 23.434947967529297, "learning_rate": 9.317993441490163e-06, "loss": 12.505496978759766, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.2653582412749529, "loss_total_v6": 0.7815935835242271, "step": 1512 }, { "epoch": 0.34894414414414415, "grad_norm": 26.404159545898438, "learning_rate": 9.316977753033858e-06, "loss": 12.67478084564209, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.29142611660063267, "loss_total_v6": 0.7921737879514694, "step": 1513 }, { "epoch": 0.34917477477477477, "grad_norm": 25.39919090270996, "learning_rate": 9.315961364267527e-06, "loss": 12.445831298828125, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.26260566897690296, "loss_total_v6": 0.7778644561767578, "step": 1514 }, { "epoch": 0.3494054054054054, "grad_norm": 24.522762298583984, "learning_rate": 9.314944275356049e-06, "loss": 12.899341583251953, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2906907144933939, "loss_total_v6": 0.8062088936567307, "step": 1515 }, { "epoch": 0.34963603603603605, "grad_norm": 24.28596305847168, "learning_rate": 9.31392648646442e-06, "loss": 12.495769500732422, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2694347146898508, "loss_total_v6": 0.7809855937957764, "step": 1516 }, { "epoch": 0.34986666666666666, "grad_norm": 27.075803756713867, "learning_rate": 9.312907997757747e-06, "loss": 13.418278694152832, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.3247568979859352, "loss_total_v6": 0.8386424034833908, "step": 1517 }, { "epoch": 0.35009729729729727, "grad_norm": 27.348541259765625, "learning_rate": 9.31188880940125e-06, "loss": 13.654169082641602, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.3390728421509266, "loss_total_v6": 0.8533855825662613, "step": 1518 }, { "epoch": 0.35032792792792794, "grad_norm": 24.0833683013916, "learning_rate": 9.310868921560268e-06, "loss": 12.759966850280762, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.28774756379425526, "loss_total_v6": 0.7974979355931282, "step": 1519 }, { "epoch": 0.35055855855855855, "grad_norm": 25.678098678588867, "learning_rate": 9.309848334400247e-06, "loss": 13.30223274230957, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.3262472990900278, "loss_total_v6": 0.8313895240426064, "step": 1520 }, { "epoch": 0.35078918918918917, "grad_norm": 26.553226470947266, "learning_rate": 9.30882704808675e-06, "loss": 12.722452163696289, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.28724921494722366, "loss_total_v6": 0.7951532751321793, "step": 1521 }, { "epoch": 0.35101981981981983, "grad_norm": 26.39850425720215, "learning_rate": 9.307805062785451e-06, "loss": 12.591093063354492, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2802904564887285, "loss_total_v6": 0.7869432866573334, "step": 1522 }, { "epoch": 0.35125045045045045, "grad_norm": 31.796049118041992, "learning_rate": 9.306782378662138e-06, "loss": 11.976300239562988, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.24404796585440636, "loss_total_v6": 0.7485187649726868, "step": 1523 }, { "epoch": 0.35148108108108106, "grad_norm": 29.659276962280273, "learning_rate": 9.305758995882716e-06, "loss": 13.178213119506836, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.30636539310216904, "loss_total_v6": 0.8236383125185966, "step": 1524 }, { "epoch": 0.3517117117117117, "grad_norm": 26.628549575805664, "learning_rate": 9.3047349146132e-06, "loss": 12.818774223327637, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.28460230492055416, "loss_total_v6": 0.8011733517050743, "step": 1525 }, { "epoch": 0.35194234234234234, "grad_norm": 23.338123321533203, "learning_rate": 9.30371013501972e-06, "loss": 12.890740394592285, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.29367783293128014, "loss_total_v6": 0.8056712672114372, "step": 1526 }, { "epoch": 0.35217297297297295, "grad_norm": 24.515331268310547, "learning_rate": 9.302684657268513e-06, "loss": 12.50686264038086, "loss_alignment": 1.01806640625, "loss_alignment_w": 0.509033203125, "loss_sft": 0.272355817258358, "loss_total_v6": 0.7816789373755455, "step": 1527 }, { "epoch": 0.3524036036036036, "grad_norm": 29.069398880004883, "learning_rate": 9.30165848152594e-06, "loss": 12.803466796875, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.2940826043486595, "loss_total_v6": 0.8002166599035263, "step": 1528 }, { "epoch": 0.35263423423423423, "grad_norm": 27.07203483581543, "learning_rate": 9.300631607958464e-06, "loss": 12.811439514160156, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.2936805970966816, "loss_total_v6": 0.800714910030365, "step": 1529 }, { "epoch": 0.35286486486486485, "grad_norm": 25.588367462158203, "learning_rate": 9.299604036732672e-06, "loss": 13.181032180786133, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.3134384900331497, "loss_total_v6": 0.8238145038485527, "step": 1530 }, { "epoch": 0.3530954954954955, "grad_norm": 28.166086196899414, "learning_rate": 9.298575768015257e-06, "loss": 12.05003547668457, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.24681008979678154, "loss_total_v6": 0.7531271874904633, "step": 1531 }, { "epoch": 0.3533261261261261, "grad_norm": 25.540267944335938, "learning_rate": 9.297546801973027e-06, "loss": 13.266037940979004, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.31981947645545006, "loss_total_v6": 0.8291273415088654, "step": 1532 }, { "epoch": 0.35355675675675674, "grad_norm": 23.811458587646484, "learning_rate": 9.296517138772902e-06, "loss": 12.820304870605469, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.29914806969463825, "loss_total_v6": 0.8012690320611, "step": 1533 }, { "epoch": 0.3537873873873874, "grad_norm": 29.63978385925293, "learning_rate": 9.295486778581919e-06, "loss": 13.080289840698242, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.31045326218008995, "loss_total_v6": 0.8175180777907372, "step": 1534 }, { "epoch": 0.354018018018018, "grad_norm": 25.710206985473633, "learning_rate": 9.294455721567224e-06, "loss": 12.937625885009766, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.30368828773498535, "loss_total_v6": 0.8086016178131104, "step": 1535 }, { "epoch": 0.35424864864864863, "grad_norm": 23.225833892822266, "learning_rate": 9.293423967896076e-06, "loss": 12.201339721679688, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.26418593525886536, "loss_total_v6": 0.762583777308464, "step": 1536 }, { "epoch": 0.3544792792792793, "grad_norm": 22.03575897216797, "learning_rate": 9.292391517735851e-06, "loss": 13.135698318481445, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.31422150880098343, "loss_total_v6": 0.8209811523556709, "step": 1537 }, { "epoch": 0.3547099099099099, "grad_norm": 24.387287139892578, "learning_rate": 9.291358371254033e-06, "loss": 13.006954193115234, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.3120495639741421, "loss_total_v6": 0.8129345774650574, "step": 1538 }, { "epoch": 0.35494054054054053, "grad_norm": 26.28998374938965, "learning_rate": 9.290324528618225e-06, "loss": 13.154339790344238, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.3096493110060692, "loss_total_v6": 0.8221462517976761, "step": 1539 }, { "epoch": 0.3551711711711712, "grad_norm": 26.298032760620117, "learning_rate": 9.289289989996133e-06, "loss": 12.686573028564453, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.2916901148855686, "loss_total_v6": 0.7929107993841171, "step": 1540 }, { "epoch": 0.3554018018018018, "grad_norm": 23.435958862304688, "learning_rate": 9.288254755555591e-06, "loss": 12.18691635131836, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.25594499707221985, "loss_total_v6": 0.7616823092103004, "step": 1541 }, { "epoch": 0.3556324324324324, "grad_norm": 26.122468948364258, "learning_rate": 9.287218825464528e-06, "loss": 12.525368690490723, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2692552302032709, "loss_total_v6": 0.7828355133533478, "step": 1542 }, { "epoch": 0.3558630630630631, "grad_norm": 23.77718734741211, "learning_rate": 9.286182199891e-06, "loss": 12.80996322631836, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.3002260010689497, "loss_total_v6": 0.8006227314472198, "step": 1543 }, { "epoch": 0.3560936936936937, "grad_norm": 25.32520866394043, "learning_rate": 9.285144879003173e-06, "loss": 12.693148612976074, "loss_alignment": 1.046875, "loss_alignment_w": 0.5234375, "loss_sft": 0.2706319186836481, "loss_total_v6": 0.793321780860424, "step": 1544 }, { "epoch": 0.3563243243243243, "grad_norm": 25.856874465942383, "learning_rate": 9.284106862969318e-06, "loss": 12.784512519836426, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2891138233244419, "loss_total_v6": 0.7990320399403572, "step": 1545 }, { "epoch": 0.35655495495495493, "grad_norm": 23.935344696044922, "learning_rate": 9.283068151957828e-06, "loss": 12.603863716125488, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.26494484208524227, "loss_total_v6": 0.7877414748072624, "step": 1546 }, { "epoch": 0.3567855855855856, "grad_norm": 24.52143096923828, "learning_rate": 9.282028746137203e-06, "loss": 13.090494155883789, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.30300396122038364, "loss_total_v6": 0.8181559145450592, "step": 1547 }, { "epoch": 0.3570162162162162, "grad_norm": 22.575349807739258, "learning_rate": 9.28098864567606e-06, "loss": 13.20495891571045, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.31255353800952435, "loss_total_v6": 0.8253099024295807, "step": 1548 }, { "epoch": 0.3572468468468468, "grad_norm": 25.313291549682617, "learning_rate": 9.279947850743125e-06, "loss": 12.82470989227295, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.29722611233592033, "loss_total_v6": 0.8015443533658981, "step": 1549 }, { "epoch": 0.3574774774774775, "grad_norm": 25.17112922668457, "learning_rate": 9.278906361507238e-06, "loss": 12.614381790161133, "loss_alignment": 0.9853515625, "loss_alignment_w": 0.49267578125, "loss_sft": 0.2972183842211962, "loss_total_v6": 0.788398839533329, "step": 1550 }, { "epoch": 0.3577081081081081, "grad_norm": 25.334049224853516, "learning_rate": 9.277864178137352e-06, "loss": 12.893842697143555, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.30198944732546806, "loss_total_v6": 0.8058651760220528, "step": 1551 }, { "epoch": 0.3579387387387387, "grad_norm": 24.742656707763672, "learning_rate": 9.276821300802535e-06, "loss": 12.871049880981445, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.29411036521196365, "loss_total_v6": 0.8044406026601791, "step": 1552 }, { "epoch": 0.3581693693693694, "grad_norm": 24.58173942565918, "learning_rate": 9.275777729671959e-06, "loss": 13.026375770568848, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2995152659714222, "loss_total_v6": 0.814148485660553, "step": 1553 }, { "epoch": 0.3584, "grad_norm": 25.917091369628906, "learning_rate": 9.27473346491492e-06, "loss": 13.040616989135742, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.3089045360684395, "loss_total_v6": 0.8150385394692421, "step": 1554 }, { "epoch": 0.3586306306306306, "grad_norm": 25.24124526977539, "learning_rate": 9.27368850670082e-06, "loss": 12.765094757080078, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.28889198787510395, "loss_total_v6": 0.7978183999657631, "step": 1555 }, { "epoch": 0.3588612612612613, "grad_norm": 23.143718719482422, "learning_rate": 9.272642855199171e-06, "loss": 12.327269554138184, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.2707594931125641, "loss_total_v6": 0.7704543471336365, "step": 1556 }, { "epoch": 0.3590918918918919, "grad_norm": 24.264774322509766, "learning_rate": 9.271596510579606e-06, "loss": 13.090597152709961, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.2930617108941078, "loss_total_v6": 0.8181623816490173, "step": 1557 }, { "epoch": 0.3593225225225225, "grad_norm": 22.560405731201172, "learning_rate": 9.27054947301186e-06, "loss": 13.075128555297852, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.3037525601685047, "loss_total_v6": 0.8171955347061157, "step": 1558 }, { "epoch": 0.35955315315315317, "grad_norm": 27.32748794555664, "learning_rate": 9.269501742665792e-06, "loss": 12.59998607635498, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2835318502038717, "loss_total_v6": 0.7874991223216057, "step": 1559 }, { "epoch": 0.3597837837837838, "grad_norm": 24.135713577270508, "learning_rate": 9.268453319711362e-06, "loss": 13.180501937866211, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.3090108372271061, "loss_total_v6": 0.8237813413143158, "step": 1560 }, { "epoch": 0.3600144144144144, "grad_norm": 23.280611038208008, "learning_rate": 9.267404204318651e-06, "loss": 12.87541389465332, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.3059646151959896, "loss_total_v6": 0.8047133982181549, "step": 1561 }, { "epoch": 0.36024504504504506, "grad_norm": 23.673097610473633, "learning_rate": 9.266354396657846e-06, "loss": 12.981626510620117, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2986715994775295, "loss_total_v6": 0.8113516569137573, "step": 1562 }, { "epoch": 0.3604756756756757, "grad_norm": 24.588279724121094, "learning_rate": 9.265303896899251e-06, "loss": 12.71211051940918, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.28004157915711403, "loss_total_v6": 0.7945069149136543, "step": 1563 }, { "epoch": 0.3607063063063063, "grad_norm": 25.379425048828125, "learning_rate": 9.264252705213281e-06, "loss": 12.669761657714844, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.2938284296542406, "loss_total_v6": 0.7918600887060165, "step": 1564 }, { "epoch": 0.36093693693693696, "grad_norm": 28.271387100219727, "learning_rate": 9.263200821770462e-06, "loss": 12.94688606262207, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.3018714040517807, "loss_total_v6": 0.8091803714632988, "step": 1565 }, { "epoch": 0.36116756756756757, "grad_norm": 27.076745986938477, "learning_rate": 9.262148246741433e-06, "loss": 12.778383255004883, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2885629218071699, "loss_total_v6": 0.7986489906907082, "step": 1566 }, { "epoch": 0.3613981981981982, "grad_norm": 23.26421356201172, "learning_rate": 9.261094980296944e-06, "loss": 12.462769508361816, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2668991684913635, "loss_total_v6": 0.7789231017231941, "step": 1567 }, { "epoch": 0.36162882882882885, "grad_norm": 25.75769805908203, "learning_rate": 9.26004102260786e-06, "loss": 12.470186233520508, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.2851086389273405, "loss_total_v6": 0.7793865874409676, "step": 1568 }, { "epoch": 0.36185945945945946, "grad_norm": 26.76333236694336, "learning_rate": 9.258986373845156e-06, "loss": 13.000285148620605, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.30444591119885445, "loss_total_v6": 0.812517836689949, "step": 1569 }, { "epoch": 0.3620900900900901, "grad_norm": 26.40073013305664, "learning_rate": 9.257931034179919e-06, "loss": 12.261959075927734, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2596128396689892, "loss_total_v6": 0.7663724720478058, "step": 1570 }, { "epoch": 0.36232072072072075, "grad_norm": 23.65530014038086, "learning_rate": 9.256875003783348e-06, "loss": 12.710062026977539, "loss_alignment": 1.01416015625, "loss_alignment_w": 0.507080078125, "loss_sft": 0.28734463080763817, "loss_total_v6": 0.7943789213895798, "step": 1571 }, { "epoch": 0.36255135135135136, "grad_norm": 23.684322357177734, "learning_rate": 9.255818282826755e-06, "loss": 12.918924331665039, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.30137501657009125, "loss_total_v6": 0.807432733476162, "step": 1572 }, { "epoch": 0.36278198198198197, "grad_norm": 25.604522705078125, "learning_rate": 9.254760871481566e-06, "loss": 12.24679183959961, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.24650355242192745, "loss_total_v6": 0.7654244378209114, "step": 1573 }, { "epoch": 0.36301261261261264, "grad_norm": 27.121829986572266, "learning_rate": 9.253702769919315e-06, "loss": 13.182064056396484, "loss_alignment": 1.02392578125, "loss_alignment_w": 0.511962890625, "loss_sft": 0.31064960919320583, "loss_total_v6": 0.8238789662718773, "step": 1574 }, { "epoch": 0.36324324324324325, "grad_norm": 21.915508270263672, "learning_rate": 9.252643978311649e-06, "loss": 12.32547378540039, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.27467566542327404, "loss_total_v6": 0.7703421339392662, "step": 1575 }, { "epoch": 0.36347387387387387, "grad_norm": 27.881900787353516, "learning_rate": 9.251584496830329e-06, "loss": 12.570650100708008, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.28041660226881504, "loss_total_v6": 0.7856656312942505, "step": 1576 }, { "epoch": 0.3637045045045045, "grad_norm": 25.496030807495117, "learning_rate": 9.250524325647224e-06, "loss": 12.832395553588867, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2867660168558359, "loss_total_v6": 0.8020247742533684, "step": 1577 }, { "epoch": 0.36393513513513515, "grad_norm": 22.31851577758789, "learning_rate": 9.24946346493432e-06, "loss": 12.219213485717773, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.25157015584409237, "loss_total_v6": 0.7637008875608444, "step": 1578 }, { "epoch": 0.36416576576576576, "grad_norm": 23.862966537475586, "learning_rate": 9.248401914863712e-06, "loss": 12.737422943115234, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.29164865985512733, "loss_total_v6": 0.7960889637470245, "step": 1579 }, { "epoch": 0.36439639639639637, "grad_norm": 25.147361755371094, "learning_rate": 9.247339675607606e-06, "loss": 12.573333740234375, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.28413973189890385, "loss_total_v6": 0.7858334109187126, "step": 1580 }, { "epoch": 0.36462702702702704, "grad_norm": 24.464643478393555, "learning_rate": 9.24627674733832e-06, "loss": 12.60179615020752, "loss_alignment": 1.044921875, "loss_alignment_w": 0.5224609375, "loss_sft": 0.2649987693876028, "loss_total_v6": 0.7876122891902924, "step": 1581 }, { "epoch": 0.36485765765765765, "grad_norm": 25.387691497802734, "learning_rate": 9.245213130228288e-06, "loss": 12.853246688842773, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.29156339541077614, "loss_total_v6": 0.8033279180526733, "step": 1582 }, { "epoch": 0.36508828828828827, "grad_norm": 25.421533584594727, "learning_rate": 9.24414882445005e-06, "loss": 12.237524032592773, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.2634872496128082, "loss_total_v6": 0.7648452818393707, "step": 1583 }, { "epoch": 0.36531891891891893, "grad_norm": 24.322587966918945, "learning_rate": 9.243083830176261e-06, "loss": 13.173285484313965, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.3127865009009838, "loss_total_v6": 0.8233303353190422, "step": 1584 }, { "epoch": 0.36554954954954955, "grad_norm": 24.01921844482422, "learning_rate": 9.242018147579685e-06, "loss": 12.895197868347168, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.29575700499117374, "loss_total_v6": 0.8059498444199562, "step": 1585 }, { "epoch": 0.36578018018018016, "grad_norm": 22.328113555908203, "learning_rate": 9.240951776833202e-06, "loss": 12.8544921875, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.30290221609175205, "loss_total_v6": 0.8034057766199112, "step": 1586 }, { "epoch": 0.3660108108108108, "grad_norm": 25.09893226623535, "learning_rate": 9.2398847181098e-06, "loss": 12.568233489990234, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.2813184857368469, "loss_total_v6": 0.7855146452784538, "step": 1587 }, { "epoch": 0.36624144144144144, "grad_norm": 24.42222785949707, "learning_rate": 9.238816971582578e-06, "loss": 12.944245338439941, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.29213907569646835, "loss_total_v6": 0.8090153187513351, "step": 1588 }, { "epoch": 0.36647207207207205, "grad_norm": 21.729249954223633, "learning_rate": 9.23774853742475e-06, "loss": 12.525229454040527, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.27672331407666206, "loss_total_v6": 0.7828268483281136, "step": 1589 }, { "epoch": 0.3667027027027027, "grad_norm": 24.480775833129883, "learning_rate": 9.236679415809641e-06, "loss": 12.924280166625977, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.29836804792284966, "loss_total_v6": 0.80776746571064, "step": 1590 }, { "epoch": 0.36693333333333333, "grad_norm": 24.575952529907227, "learning_rate": 9.235609606910687e-06, "loss": 12.411237716674805, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.26880541630089283, "loss_total_v6": 0.7757023796439171, "step": 1591 }, { "epoch": 0.36716396396396395, "grad_norm": 23.29962730407715, "learning_rate": 9.23453911090143e-06, "loss": 12.57681655883789, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.27684997767210007, "loss_total_v6": 0.786051020026207, "step": 1592 }, { "epoch": 0.3673945945945946, "grad_norm": 25.32788848876953, "learning_rate": 9.233467927955533e-06, "loss": 12.192338943481445, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.26556118950247765, "loss_total_v6": 0.7620211839675903, "step": 1593 }, { "epoch": 0.3676252252252252, "grad_norm": 26.118330001831055, "learning_rate": 9.232396058246763e-06, "loss": 12.810201644897461, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.29213847033679485, "loss_total_v6": 0.8006376251578331, "step": 1594 }, { "epoch": 0.36785585585585584, "grad_norm": 27.213321685791016, "learning_rate": 9.231323501949003e-06, "loss": 13.101934432983398, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.3163074068725109, "loss_total_v6": 0.8188709244132042, "step": 1595 }, { "epoch": 0.3680864864864865, "grad_norm": 27.23192596435547, "learning_rate": 9.230250259236245e-06, "loss": 13.061328887939453, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.3066742364317179, "loss_total_v6": 0.8163330480456352, "step": 1596 }, { "epoch": 0.3683171171171171, "grad_norm": 24.596426010131836, "learning_rate": 9.22917633028259e-06, "loss": 12.580720901489258, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.27582747116684914, "loss_total_v6": 0.786295011639595, "step": 1597 }, { "epoch": 0.36854774774774773, "grad_norm": 23.906024932861328, "learning_rate": 9.22810171526226e-06, "loss": 13.063982009887695, "loss_alignment": 1.02685546875, "loss_alignment_w": 0.513427734375, "loss_sft": 0.3034220952540636, "loss_total_v6": 0.8164988681674004, "step": 1598 }, { "epoch": 0.3687783783783784, "grad_norm": 21.32941246032715, "learning_rate": 9.227026414349574e-06, "loss": 12.480144500732422, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.27413436584174633, "loss_total_v6": 0.7800090238451958, "step": 1599 }, { "epoch": 0.369009009009009, "grad_norm": 26.018314361572266, "learning_rate": 9.225950427718974e-06, "loss": 12.738334655761719, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2849765121936798, "loss_total_v6": 0.7961459532380104, "step": 1600 }, { "epoch": 0.3692396396396396, "grad_norm": 23.147226333618164, "learning_rate": 9.22487375554501e-06, "loss": 12.618764877319336, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.28839810751378536, "loss_total_v6": 0.7886727824807167, "step": 1601 }, { "epoch": 0.3694702702702703, "grad_norm": 23.88650131225586, "learning_rate": 9.223796398002337e-06, "loss": 13.12941837310791, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.3167281970381737, "loss_total_v6": 0.8205886483192444, "step": 1602 }, { "epoch": 0.3697009009009009, "grad_norm": 22.370311737060547, "learning_rate": 9.222718355265733e-06, "loss": 12.574009895324707, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.2834799848496914, "loss_total_v6": 0.7858756110072136, "step": 1603 }, { "epoch": 0.3699315315315315, "grad_norm": 22.198474884033203, "learning_rate": 9.221639627510076e-06, "loss": 12.94155502319336, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.30494093522429466, "loss_total_v6": 0.8088472336530685, "step": 1604 }, { "epoch": 0.3701621621621622, "grad_norm": 24.505098342895508, "learning_rate": 9.220560214910362e-06, "loss": 12.360735893249512, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.27042498253285885, "loss_total_v6": 0.7725459784269333, "step": 1605 }, { "epoch": 0.3703927927927928, "grad_norm": 24.355266571044922, "learning_rate": 9.219480117641693e-06, "loss": 12.489590644836426, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2727106139063835, "loss_total_v6": 0.780599407851696, "step": 1606 }, { "epoch": 0.3706234234234234, "grad_norm": 27.445497512817383, "learning_rate": 9.21839933587929e-06, "loss": 13.262011528015137, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.32333677262067795, "loss_total_v6": 0.8288757279515266, "step": 1607 }, { "epoch": 0.370854054054054, "grad_norm": 32.24522399902344, "learning_rate": 9.217317869798473e-06, "loss": 12.478391647338867, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.2784346006810665, "loss_total_v6": 0.779899463057518, "step": 1608 }, { "epoch": 0.3710846846846847, "grad_norm": 25.61947250366211, "learning_rate": 9.216235719574684e-06, "loss": 13.098031997680664, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.31446130760014057, "loss_total_v6": 0.8186269626021385, "step": 1609 }, { "epoch": 0.3713153153153153, "grad_norm": 25.827943801879883, "learning_rate": 9.215152885383473e-06, "loss": 13.458131790161133, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.3266984224319458, "loss_total_v6": 0.8411332294344902, "step": 1610 }, { "epoch": 0.3715459459459459, "grad_norm": 23.483036041259766, "learning_rate": 9.214069367400497e-06, "loss": 13.094600677490234, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.3119427263736725, "loss_total_v6": 0.8184124752879143, "step": 1611 }, { "epoch": 0.3717765765765766, "grad_norm": 26.817445755004883, "learning_rate": 9.21298516580153e-06, "loss": 12.924503326416016, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.31182507425546646, "loss_total_v6": 0.8077814653515816, "step": 1612 }, { "epoch": 0.3720072072072072, "grad_norm": 27.732484817504883, "learning_rate": 9.21190028076245e-06, "loss": 12.853565216064453, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2864105850458145, "loss_total_v6": 0.8033478483557701, "step": 1613 }, { "epoch": 0.3722378378378378, "grad_norm": 24.740720748901367, "learning_rate": 9.210814712459251e-06, "loss": 12.657400131225586, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2784836869686842, "loss_total_v6": 0.7910875082015991, "step": 1614 }, { "epoch": 0.3724684684684685, "grad_norm": 24.651723861694336, "learning_rate": 9.209728461068037e-06, "loss": 13.102968215942383, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.30768976733088493, "loss_total_v6": 0.8189355283975601, "step": 1615 }, { "epoch": 0.3726990990990991, "grad_norm": 24.980823516845703, "learning_rate": 9.208641526765024e-06, "loss": 12.697738647460938, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.29461570642888546, "loss_total_v6": 0.7936086058616638, "step": 1616 }, { "epoch": 0.3729297297297297, "grad_norm": 24.045608520507812, "learning_rate": 9.207553909726532e-06, "loss": 12.887399673461914, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2930723447352648, "loss_total_v6": 0.8054625019431114, "step": 1617 }, { "epoch": 0.3731603603603604, "grad_norm": 23.55909538269043, "learning_rate": 9.206465610129e-06, "loss": 12.764580726623535, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2901111040264368, "loss_total_v6": 0.797786295413971, "step": 1618 }, { "epoch": 0.373390990990991, "grad_norm": 26.14499855041504, "learning_rate": 9.205376628148974e-06, "loss": 12.731877326965332, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.28507643006742, "loss_total_v6": 0.7957423180341721, "step": 1619 }, { "epoch": 0.3736216216216216, "grad_norm": 26.22720718383789, "learning_rate": 9.204286963963112e-06, "loss": 12.84950065612793, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.3013542853295803, "loss_total_v6": 0.8030937984585762, "step": 1620 }, { "epoch": 0.37385225225225227, "grad_norm": 21.734085083007812, "learning_rate": 9.20319661774818e-06, "loss": 12.754526138305664, "loss_alignment": 1.052734375, "loss_alignment_w": 0.5263671875, "loss_sft": 0.2704855166375637, "loss_total_v6": 0.7971578910946846, "step": 1621 }, { "epoch": 0.3740828828828829, "grad_norm": 26.51085090637207, "learning_rate": 9.202105589681058e-06, "loss": 12.818632125854492, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.2850665058940649, "loss_total_v6": 0.8011645451188087, "step": 1622 }, { "epoch": 0.3743135135135135, "grad_norm": 25.8449764251709, "learning_rate": 9.201013879938733e-06, "loss": 12.41108226776123, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2677886076271534, "loss_total_v6": 0.7756926342844963, "step": 1623 }, { "epoch": 0.37454414414414416, "grad_norm": 22.864633560180664, "learning_rate": 9.199921488698307e-06, "loss": 13.095551490783691, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.3067685179412365, "loss_total_v6": 0.8184719979763031, "step": 1624 }, { "epoch": 0.3747747747747748, "grad_norm": 24.30397605895996, "learning_rate": 9.198828416136991e-06, "loss": 13.293764114379883, "loss_alignment": 1.02587890625, "loss_alignment_w": 0.512939453125, "loss_sft": 0.31851593032479286, "loss_total_v6": 0.8308602645993233, "step": 1625 }, { "epoch": 0.3750054054054054, "grad_norm": 42.08715057373047, "learning_rate": 9.197734662432101e-06, "loss": 12.95974349975586, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.304170411080122, "loss_total_v6": 0.8099840059876442, "step": 1626 }, { "epoch": 0.37523603603603606, "grad_norm": 24.510845184326172, "learning_rate": 9.196640227761073e-06, "loss": 12.87796401977539, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.30040189251303673, "loss_total_v6": 0.8048727437853813, "step": 1627 }, { "epoch": 0.37546666666666667, "grad_norm": 25.389745712280273, "learning_rate": 9.195545112301448e-06, "loss": 13.11095142364502, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.2976907007396221, "loss_total_v6": 0.8194344639778137, "step": 1628 }, { "epoch": 0.3756972972972973, "grad_norm": 22.53106117248535, "learning_rate": 9.194449316230876e-06, "loss": 12.753710746765137, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.29129330813884735, "loss_total_v6": 0.7971068993210793, "step": 1629 }, { "epoch": 0.37592792792792795, "grad_norm": 22.888545989990234, "learning_rate": 9.193352839727122e-06, "loss": 12.788016319274902, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.28728812374174595, "loss_total_v6": 0.7992510199546814, "step": 1630 }, { "epoch": 0.37615855855855856, "grad_norm": 22.984840393066406, "learning_rate": 9.192255682968056e-06, "loss": 13.136162757873535, "loss_alignment": 1.01806640625, "loss_alignment_w": 0.509033203125, "loss_sft": 0.31133610755205154, "loss_total_v6": 0.8210101798176765, "step": 1631 }, { "epoch": 0.3763891891891892, "grad_norm": 25.930410385131836, "learning_rate": 9.191157846131662e-06, "loss": 12.605010986328125, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.27472112886607647, "loss_total_v6": 0.7878132089972496, "step": 1632 }, { "epoch": 0.37661981981981985, "grad_norm": 23.53544044494629, "learning_rate": 9.190059329396038e-06, "loss": 12.427654266357422, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2679241020232439, "loss_total_v6": 0.7767284214496613, "step": 1633 }, { "epoch": 0.37685045045045046, "grad_norm": 25.875375747680664, "learning_rate": 9.18896013293938e-06, "loss": 12.384199142456055, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2678936757147312, "loss_total_v6": 0.7740124613046646, "step": 1634 }, { "epoch": 0.37708108108108107, "grad_norm": 24.79372215270996, "learning_rate": 9.18786025694001e-06, "loss": 13.294615745544434, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.3188285008072853, "loss_total_v6": 0.8309134691953659, "step": 1635 }, { "epoch": 0.37731171171171174, "grad_norm": 23.761947631835938, "learning_rate": 9.18675970157635e-06, "loss": 13.007783889770508, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2995893023908138, "loss_total_v6": 0.8129865303635597, "step": 1636 }, { "epoch": 0.37754234234234235, "grad_norm": 26.98575210571289, "learning_rate": 9.18565846702693e-06, "loss": 12.346644401550293, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.26783532835543156, "loss_total_v6": 0.7716652899980545, "step": 1637 }, { "epoch": 0.37777297297297296, "grad_norm": 22.91792106628418, "learning_rate": 9.1845565534704e-06, "loss": 12.326035499572754, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2639227546751499, "loss_total_v6": 0.7703772112727165, "step": 1638 }, { "epoch": 0.3780036036036036, "grad_norm": 24.795541763305664, "learning_rate": 9.183453961085513e-06, "loss": 13.404621124267578, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.32800791412591934, "loss_total_v6": 0.837788812816143, "step": 1639 }, { "epoch": 0.37823423423423425, "grad_norm": 24.7641544342041, "learning_rate": 9.182350690051134e-06, "loss": 11.995353698730469, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2440333440899849, "loss_total_v6": 0.7497096285223961, "step": 1640 }, { "epoch": 0.37846486486486486, "grad_norm": 26.23995018005371, "learning_rate": 9.18124674054624e-06, "loss": 12.550704956054688, "loss_alignment": 1.01513671875, "loss_alignment_w": 0.507568359375, "loss_sft": 0.27714063227176666, "loss_total_v6": 0.7844190299510956, "step": 1641 }, { "epoch": 0.37869549549549547, "grad_norm": 27.0361270904541, "learning_rate": 9.180142112749915e-06, "loss": 13.05233383178711, "loss_alignment": 1.02099609375, "loss_alignment_w": 0.510498046875, "loss_sft": 0.3047388233244419, "loss_total_v6": 0.8157709091901779, "step": 1642 }, { "epoch": 0.37892612612612614, "grad_norm": 24.203569412231445, "learning_rate": 9.179036806841352e-06, "loss": 12.331274032592773, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.2654250580817461, "loss_total_v6": 0.7707046121358871, "step": 1643 }, { "epoch": 0.37915675675675675, "grad_norm": 24.431869506835938, "learning_rate": 9.17793082299986e-06, "loss": 12.53573226928711, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2730004508048296, "loss_total_v6": 0.7834832519292831, "step": 1644 }, { "epoch": 0.37938738738738736, "grad_norm": 28.25771141052246, "learning_rate": 9.176824161404851e-06, "loss": 12.271366119384766, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2589494548738003, "loss_total_v6": 0.7669603377580643, "step": 1645 }, { "epoch": 0.37961801801801803, "grad_norm": 25.316360473632812, "learning_rate": 9.175716822235854e-06, "loss": 12.558526992797852, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.27805672585964203, "loss_total_v6": 0.7849079445004463, "step": 1646 }, { "epoch": 0.37984864864864865, "grad_norm": 26.25047492980957, "learning_rate": 9.1746088056725e-06, "loss": 12.556578636169434, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2759513258934021, "loss_total_v6": 0.784786157310009, "step": 1647 }, { "epoch": 0.38007927927927926, "grad_norm": 23.13949966430664, "learning_rate": 9.173500111894536e-06, "loss": 13.034173965454102, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.29792752861976624, "loss_total_v6": 0.8146358728408813, "step": 1648 }, { "epoch": 0.3803099099099099, "grad_norm": 26.554595947265625, "learning_rate": 9.172390741081817e-06, "loss": 13.112382888793945, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.3070421889424324, "loss_total_v6": 0.8195239007472992, "step": 1649 }, { "epoch": 0.38054054054054054, "grad_norm": 25.83377456665039, "learning_rate": 9.171280693414307e-06, "loss": 12.73359203338623, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.2932402901351452, "loss_total_v6": 0.7958495318889618, "step": 1650 }, { "epoch": 0.38077117117117115, "grad_norm": 22.604169845581055, "learning_rate": 9.170169969072081e-06, "loss": 12.51407241821289, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2717230189591646, "loss_total_v6": 0.7821295261383057, "step": 1651 }, { "epoch": 0.3810018018018018, "grad_norm": 26.18514633178711, "learning_rate": 9.169058568235324e-06, "loss": 13.281999588012695, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.319962615147233, "loss_total_v6": 0.8301249518990517, "step": 1652 }, { "epoch": 0.38123243243243243, "grad_norm": 25.620229721069336, "learning_rate": 9.167946491084327e-06, "loss": 12.703168869018555, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.28961450420320034, "loss_total_v6": 0.7939480096101761, "step": 1653 }, { "epoch": 0.38146306306306305, "grad_norm": 22.780860900878906, "learning_rate": 9.166833737799497e-06, "loss": 12.334854125976562, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.25841611064970493, "loss_total_v6": 0.7709283530712128, "step": 1654 }, { "epoch": 0.3816936936936937, "grad_norm": 27.091663360595703, "learning_rate": 9.165720308561347e-06, "loss": 12.986225128173828, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.2923367340117693, "loss_total_v6": 0.8116391003131866, "step": 1655 }, { "epoch": 0.3819243243243243, "grad_norm": 24.618757247924805, "learning_rate": 9.164606203550498e-06, "loss": 12.837128639221191, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.28918271511793137, "loss_total_v6": 0.8023205474019051, "step": 1656 }, { "epoch": 0.38215495495495494, "grad_norm": 24.71617889404297, "learning_rate": 9.163491422947684e-06, "loss": 12.325742721557617, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.26240914687514305, "loss_total_v6": 0.7703589424490929, "step": 1657 }, { "epoch": 0.3823855855855856, "grad_norm": 25.093591690063477, "learning_rate": 9.162375966933748e-06, "loss": 13.000920295715332, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.3016322124749422, "loss_total_v6": 0.8125575259327888, "step": 1658 }, { "epoch": 0.3826162162162162, "grad_norm": 23.94870948791504, "learning_rate": 9.16125983568964e-06, "loss": 12.43557357788086, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2691361363977194, "loss_total_v6": 0.7772233262658119, "step": 1659 }, { "epoch": 0.38284684684684683, "grad_norm": 24.58452033996582, "learning_rate": 9.160143029396422e-06, "loss": 12.775236129760742, "loss_alignment": 1.02978515625, "loss_alignment_w": 0.514892578125, "loss_sft": 0.2838648483157158, "loss_total_v6": 0.7984522730112076, "step": 1660 }, { "epoch": 0.3830774774774775, "grad_norm": 36.23276138305664, "learning_rate": 9.159025548235265e-06, "loss": 13.060323715209961, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.31649911776185036, "loss_total_v6": 0.8162702322006226, "step": 1661 }, { "epoch": 0.3833081081081081, "grad_norm": 27.291841506958008, "learning_rate": 9.15790739238745e-06, "loss": 12.79770278930664, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.29704878106713295, "loss_total_v6": 0.7998564317822456, "step": 1662 }, { "epoch": 0.3835387387387387, "grad_norm": 24.592294692993164, "learning_rate": 9.15678856203437e-06, "loss": 12.912514686584473, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.309351546689868, "loss_total_v6": 0.8070322051644325, "step": 1663 }, { "epoch": 0.3837693693693694, "grad_norm": 23.156566619873047, "learning_rate": 9.155669057357515e-06, "loss": 13.076980590820312, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.3072252497076988, "loss_total_v6": 0.8173112794756889, "step": 1664 }, { "epoch": 0.384, "grad_norm": 24.18606948852539, "learning_rate": 9.1545488785385e-06, "loss": 12.626747131347656, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.29504629597067833, "loss_total_v6": 0.7891716882586479, "step": 1665 }, { "epoch": 0.3842306306306306, "grad_norm": 21.884546279907227, "learning_rate": 9.153428025759046e-06, "loss": 13.078831672668457, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.31339864060282707, "loss_total_v6": 0.8174269795417786, "step": 1666 }, { "epoch": 0.3844612612612613, "grad_norm": 24.628511428833008, "learning_rate": 9.152306499200973e-06, "loss": 12.497492790222168, "loss_alignment": 1.01416015625, "loss_alignment_w": 0.507080078125, "loss_sft": 0.2739979699254036, "loss_total_v6": 0.7810932993888855, "step": 1667 }, { "epoch": 0.3846918918918919, "grad_norm": 23.40794563293457, "learning_rate": 9.151184299046221e-06, "loss": 12.878889083862305, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.2890614625066519, "loss_total_v6": 0.8049305900931358, "step": 1668 }, { "epoch": 0.3849225225225225, "grad_norm": 25.5513858795166, "learning_rate": 9.150061425476839e-06, "loss": 13.17354679107666, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.3139625322073698, "loss_total_v6": 0.8233466818928719, "step": 1669 }, { "epoch": 0.3851531531531531, "grad_norm": 23.56546974182129, "learning_rate": 9.148937878674975e-06, "loss": 12.79606819152832, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.30030358768999577, "loss_total_v6": 0.7997542843222618, "step": 1670 }, { "epoch": 0.3853837837837838, "grad_norm": 23.146602630615234, "learning_rate": 9.147813658822902e-06, "loss": 12.29452133178711, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.27356510423123837, "loss_total_v6": 0.7684076055884361, "step": 1671 }, { "epoch": 0.3856144144144144, "grad_norm": 23.020750045776367, "learning_rate": 9.146688766102985e-06, "loss": 12.728883743286133, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.29366314224898815, "loss_total_v6": 0.7955552488565445, "step": 1672 }, { "epoch": 0.385845045045045, "grad_norm": 25.013595581054688, "learning_rate": 9.145563200697713e-06, "loss": 12.68641471862793, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2782830558717251, "loss_total_v6": 0.7929009348154068, "step": 1673 }, { "epoch": 0.3860756756756757, "grad_norm": 26.048076629638672, "learning_rate": 9.144436962789675e-06, "loss": 13.206296920776367, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.31118763610720634, "loss_total_v6": 0.8253935500979424, "step": 1674 }, { "epoch": 0.3863063063063063, "grad_norm": 24.438989639282227, "learning_rate": 9.14331005256157e-06, "loss": 12.398046493530273, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.2844604328274727, "loss_total_v6": 0.7748778834939003, "step": 1675 }, { "epoch": 0.3865369369369369, "grad_norm": 23.58249855041504, "learning_rate": 9.142182470196213e-06, "loss": 12.472480773925781, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2681164611130953, "loss_total_v6": 0.7795300409197807, "step": 1676 }, { "epoch": 0.3867675675675676, "grad_norm": 24.582733154296875, "learning_rate": 9.14105421587652e-06, "loss": 13.311370849609375, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.32010457292199135, "loss_total_v6": 0.8319606482982635, "step": 1677 }, { "epoch": 0.3869981981981982, "grad_norm": 21.43752670288086, "learning_rate": 9.139925289785518e-06, "loss": 12.491003036499023, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.27543868497014046, "loss_total_v6": 0.780687689781189, "step": 1678 }, { "epoch": 0.3872288288288288, "grad_norm": 26.621686935424805, "learning_rate": 9.138795692106346e-06, "loss": 13.530841827392578, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.33478283137083054, "loss_total_v6": 0.8456776216626167, "step": 1679 }, { "epoch": 0.3874594594594595, "grad_norm": 25.41166114807129, "learning_rate": 9.13766542302225e-06, "loss": 12.901811599731445, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.29095189087092876, "loss_total_v6": 0.8063632473349571, "step": 1680 }, { "epoch": 0.3876900900900901, "grad_norm": 23.153642654418945, "learning_rate": 9.136534482716583e-06, "loss": 12.705618858337402, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.28979822620749474, "loss_total_v6": 0.7941011935472488, "step": 1681 }, { "epoch": 0.3879207207207207, "grad_norm": 24.596406936645508, "learning_rate": 9.13540287137281e-06, "loss": 12.566207885742188, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2796507403254509, "loss_total_v6": 0.7853880301117897, "step": 1682 }, { "epoch": 0.38815135135135137, "grad_norm": 22.672698974609375, "learning_rate": 9.134270589174503e-06, "loss": 12.513725280761719, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.2666659653186798, "loss_total_v6": 0.7821078523993492, "step": 1683 }, { "epoch": 0.388381981981982, "grad_norm": 24.549081802368164, "learning_rate": 9.133137636305345e-06, "loss": 12.870677947998047, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2931869626045227, "loss_total_v6": 0.8044174164533615, "step": 1684 }, { "epoch": 0.3886126126126126, "grad_norm": 24.907543182373047, "learning_rate": 9.132004012949124e-06, "loss": 12.58584213256836, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2762239445000887, "loss_total_v6": 0.786615177989006, "step": 1685 }, { "epoch": 0.38884324324324326, "grad_norm": 26.244922637939453, "learning_rate": 9.13086971928974e-06, "loss": 12.480643272399902, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.27715628035366535, "loss_total_v6": 0.7800401970744133, "step": 1686 }, { "epoch": 0.3890738738738739, "grad_norm": 24.036033630371094, "learning_rate": 9.1297347555112e-06, "loss": 13.239130020141602, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.3115764893591404, "loss_total_v6": 0.8274456262588501, "step": 1687 }, { "epoch": 0.3893045045045045, "grad_norm": 24.083389282226562, "learning_rate": 9.12859912179762e-06, "loss": 12.76795768737793, "loss_alignment": 1.01611328125, "loss_alignment_w": 0.508056640625, "loss_sft": 0.29158859699964523, "loss_total_v6": 0.7979973256587982, "step": 1688 }, { "epoch": 0.38953513513513516, "grad_norm": 25.836200714111328, "learning_rate": 9.127462818333231e-06, "loss": 12.831364631652832, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.3006175570189953, "loss_total_v6": 0.8019602969288826, "step": 1689 }, { "epoch": 0.38976576576576577, "grad_norm": 24.74970817565918, "learning_rate": 9.126325845302357e-06, "loss": 12.941826820373535, "loss_alignment": 1.03369140625, "loss_alignment_w": 0.516845703125, "loss_sft": 0.29217105731368065, "loss_total_v6": 0.8088641613721848, "step": 1690 }, { "epoch": 0.3899963963963964, "grad_norm": 25.049062728881836, "learning_rate": 9.125188202889449e-06, "loss": 12.54755687713623, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2699095718562603, "loss_total_v6": 0.7842223048210144, "step": 1691 }, { "epoch": 0.39022702702702705, "grad_norm": 25.53455352783203, "learning_rate": 9.124049891279052e-06, "loss": 13.111598014831543, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.3044755160808563, "loss_total_v6": 0.8194748908281326, "step": 1692 }, { "epoch": 0.39045765765765766, "grad_norm": 22.567251205444336, "learning_rate": 9.122910910655827e-06, "loss": 12.779550552368164, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.28802546486258507, "loss_total_v6": 0.7987218946218491, "step": 1693 }, { "epoch": 0.3906882882882883, "grad_norm": 25.12813377380371, "learning_rate": 9.121771261204544e-06, "loss": 13.275018692016602, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.3176799453794956, "loss_total_v6": 0.8296886309981346, "step": 1694 }, { "epoch": 0.39091891891891895, "grad_norm": 26.359619140625, "learning_rate": 9.120630943110078e-06, "loss": 12.658805847167969, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.28900855779647827, "loss_total_v6": 0.7911753356456757, "step": 1695 }, { "epoch": 0.39114954954954956, "grad_norm": 23.466323852539062, "learning_rate": 9.119489956557416e-06, "loss": 12.745012283325195, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.2926570028066635, "loss_total_v6": 0.7965632453560829, "step": 1696 }, { "epoch": 0.39138018018018017, "grad_norm": 25.459304809570312, "learning_rate": 9.118348301731649e-06, "loss": 12.966828346252441, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.2906514275819063, "loss_total_v6": 0.8104267865419388, "step": 1697 }, { "epoch": 0.3916108108108108, "grad_norm": 24.944494247436523, "learning_rate": 9.117205978817979e-06, "loss": 12.655340194702148, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.28635068237781525, "loss_total_v6": 0.7909587994217873, "step": 1698 }, { "epoch": 0.39184144144144145, "grad_norm": 24.93559455871582, "learning_rate": 9.116062988001717e-06, "loss": 12.778385162353516, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2865641266107559, "loss_total_v6": 0.7986490726470947, "step": 1699 }, { "epoch": 0.39207207207207206, "grad_norm": 26.331586837768555, "learning_rate": 9.114919329468283e-06, "loss": 12.81273078918457, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2887718342244625, "loss_total_v6": 0.8007957264780998, "step": 1700 }, { "epoch": 0.3923027027027027, "grad_norm": 21.946630477905273, "learning_rate": 9.113775003403201e-06, "loss": 12.97708797454834, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.3001274447888136, "loss_total_v6": 0.8110679760575294, "step": 1701 }, { "epoch": 0.39253333333333335, "grad_norm": 23.437334060668945, "learning_rate": 9.112630009992109e-06, "loss": 13.01358413696289, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.3042394630610943, "loss_total_v6": 0.8133490011096001, "step": 1702 }, { "epoch": 0.39276396396396396, "grad_norm": 22.879135131835938, "learning_rate": 9.111484349420747e-06, "loss": 13.22825813293457, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.31307899579405785, "loss_total_v6": 0.826766163110733, "step": 1703 }, { "epoch": 0.39299459459459457, "grad_norm": 27.03520965576172, "learning_rate": 9.11033802187497e-06, "loss": 13.259028434753418, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.31445289961993694, "loss_total_v6": 0.8286892920732498, "step": 1704 }, { "epoch": 0.39322522522522524, "grad_norm": 25.856679916381836, "learning_rate": 9.109191027540739e-06, "loss": 12.866586685180664, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2917715609073639, "loss_total_v6": 0.8041616603732109, "step": 1705 }, { "epoch": 0.39345585585585585, "grad_norm": 21.65005874633789, "learning_rate": 9.108043366604116e-06, "loss": 12.839784622192383, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.29302604869008064, "loss_total_v6": 0.8024865165352821, "step": 1706 }, { "epoch": 0.39368648648648646, "grad_norm": 25.575651168823242, "learning_rate": 9.106895039251282e-06, "loss": 12.833575248718262, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.28697701171040535, "loss_total_v6": 0.802098460495472, "step": 1707 }, { "epoch": 0.39391711711711713, "grad_norm": 22.132692337036133, "learning_rate": 9.10574604566852e-06, "loss": 13.193302154541016, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.3216211758553982, "loss_total_v6": 0.8245814144611359, "step": 1708 }, { "epoch": 0.39414774774774775, "grad_norm": 23.633522033691406, "learning_rate": 9.104596386042223e-06, "loss": 12.621018409729004, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.28138261660933495, "loss_total_v6": 0.7888136506080627, "step": 1709 }, { "epoch": 0.39437837837837836, "grad_norm": 24.665990829467773, "learning_rate": 9.103446060558892e-06, "loss": 12.896598815917969, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.3012156207114458, "loss_total_v6": 0.8060374036431313, "step": 1710 }, { "epoch": 0.394609009009009, "grad_norm": 24.08192253112793, "learning_rate": 9.102295069405133e-06, "loss": 13.433019638061523, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.331842802464962, "loss_total_v6": 0.8395637422800064, "step": 1711 }, { "epoch": 0.39483963963963964, "grad_norm": 24.041324615478516, "learning_rate": 9.101143412767665e-06, "loss": 12.460721969604492, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.27256949804723263, "loss_total_v6": 0.7787950932979584, "step": 1712 }, { "epoch": 0.39507027027027025, "grad_norm": 23.33445167541504, "learning_rate": 9.09999109083331e-06, "loss": 12.568073272705078, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.28617597557604313, "loss_total_v6": 0.785504586994648, "step": 1713 }, { "epoch": 0.3953009009009009, "grad_norm": 25.68356704711914, "learning_rate": 9.098838103789004e-06, "loss": 12.57643985748291, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2795729748904705, "loss_total_v6": 0.7860274687409401, "step": 1714 }, { "epoch": 0.39553153153153153, "grad_norm": 23.276615142822266, "learning_rate": 9.097684451821783e-06, "loss": 12.495004653930664, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2687917537987232, "loss_total_v6": 0.7809377536177635, "step": 1715 }, { "epoch": 0.39576216216216215, "grad_norm": 24.5363826751709, "learning_rate": 9.096530135118798e-06, "loss": 13.01905345916748, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.3061530012637377, "loss_total_v6": 0.8136908560991287, "step": 1716 }, { "epoch": 0.3959927927927928, "grad_norm": 23.109371185302734, "learning_rate": 9.095375153867302e-06, "loss": 12.488946914672852, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.273341765627265, "loss_total_v6": 0.7805591970682144, "step": 1717 }, { "epoch": 0.3962234234234234, "grad_norm": 25.873323440551758, "learning_rate": 9.09421950825466e-06, "loss": 12.947318077087402, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.29730552062392235, "loss_total_v6": 0.8092073947191238, "step": 1718 }, { "epoch": 0.39645405405405404, "grad_norm": 22.947540283203125, "learning_rate": 9.093063198468346e-06, "loss": 12.849874496459961, "loss_alignment": 1.02099609375, "loss_alignment_w": 0.510498046875, "loss_sft": 0.2926953174173832, "loss_total_v6": 0.803117111325264, "step": 1719 }, { "epoch": 0.3966846846846847, "grad_norm": 24.503572463989258, "learning_rate": 9.091906224695935e-06, "loss": 13.28397274017334, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.3220238536596298, "loss_total_v6": 0.8302483111619949, "step": 1720 }, { "epoch": 0.3969153153153153, "grad_norm": 24.752368927001953, "learning_rate": 9.090748587125118e-06, "loss": 12.383594512939453, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.2721894308924675, "loss_total_v6": 0.7739747017621994, "step": 1721 }, { "epoch": 0.39714594594594593, "grad_norm": 24.018905639648438, "learning_rate": 9.089590285943687e-06, "loss": 13.077006340026855, "loss_alignment": 1.0390625, "loss_alignment_w": 0.51953125, "loss_sft": 0.29764432832598686, "loss_total_v6": 0.8173128738999367, "step": 1722 }, { "epoch": 0.3973765765765766, "grad_norm": 22.668102264404297, "learning_rate": 9.088431321339546e-06, "loss": 12.648837089538574, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2808934971690178, "loss_total_v6": 0.7905523329973221, "step": 1723 }, { "epoch": 0.3976072072072072, "grad_norm": 24.853939056396484, "learning_rate": 9.087271693500701e-06, "loss": 12.590825080871582, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.27647428028285503, "loss_total_v6": 0.7869265526533127, "step": 1724 }, { "epoch": 0.3978378378378378, "grad_norm": 23.586421966552734, "learning_rate": 9.086111402615274e-06, "loss": 13.071873664855957, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.30180960707366467, "loss_total_v6": 0.8169921040534973, "step": 1725 }, { "epoch": 0.3980684684684685, "grad_norm": 23.331871032714844, "learning_rate": 9.084950448871489e-06, "loss": 12.763641357421875, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.28872491978108883, "loss_total_v6": 0.7977276146411896, "step": 1726 }, { "epoch": 0.3982990990990991, "grad_norm": 23.58505630493164, "learning_rate": 9.083788832457677e-06, "loss": 12.615703582763672, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.287169236689806, "loss_total_v6": 0.7884815037250519, "step": 1727 }, { "epoch": 0.3985297297297297, "grad_norm": 25.292972564697266, "learning_rate": 9.08262655356228e-06, "loss": 12.774019241333008, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.29010598361492157, "loss_total_v6": 0.7983762472867966, "step": 1728 }, { "epoch": 0.39876036036036033, "grad_norm": 24.927528381347656, "learning_rate": 9.081463612373845e-06, "loss": 12.866571426391602, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2942120395600796, "loss_total_v6": 0.8041607588529587, "step": 1729 }, { "epoch": 0.398990990990991, "grad_norm": 24.467138290405273, "learning_rate": 9.080300009081025e-06, "loss": 12.69411849975586, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.2810075283050537, "loss_total_v6": 0.7933823689818382, "step": 1730 }, { "epoch": 0.3992216216216216, "grad_norm": 24.283735275268555, "learning_rate": 9.079135743872584e-06, "loss": 12.66758918762207, "loss_alignment": 1.01611328125, "loss_alignment_w": 0.508056640625, "loss_sft": 0.28334725089371204, "loss_total_v6": 0.7917243465781212, "step": 1731 }, { "epoch": 0.3994522522522522, "grad_norm": 23.31682777404785, "learning_rate": 9.077970816937394e-06, "loss": 12.60755443572998, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.279091514647007, "loss_total_v6": 0.7879721373319626, "step": 1732 }, { "epoch": 0.3996828828828829, "grad_norm": 24.19922637939453, "learning_rate": 9.076805228464426e-06, "loss": 12.821550369262695, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2909557167440653, "loss_total_v6": 0.8013469204306602, "step": 1733 }, { "epoch": 0.3999135135135135, "grad_norm": 23.73583221435547, "learning_rate": 9.07563897864277e-06, "loss": 12.499451637268066, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.27247243747115135, "loss_total_v6": 0.7812157198786736, "step": 1734 }, { "epoch": 0.4001441441441441, "grad_norm": 23.69680404663086, "learning_rate": 9.07447206766162e-06, "loss": 12.464895248413086, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.27444777451455593, "loss_total_v6": 0.7790559455752373, "step": 1735 }, { "epoch": 0.4003747747747748, "grad_norm": 24.02540397644043, "learning_rate": 9.073304495710267e-06, "loss": 13.242355346679688, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.3121442124247551, "loss_total_v6": 0.8276471644639969, "step": 1736 }, { "epoch": 0.4006054054054054, "grad_norm": 23.952880859375, "learning_rate": 9.07213626297812e-06, "loss": 12.483534812927246, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.277565885335207, "loss_total_v6": 0.7802208960056305, "step": 1737 }, { "epoch": 0.400836036036036, "grad_norm": 23.57798957824707, "learning_rate": 9.070967369654697e-06, "loss": 12.60978889465332, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.28565509617328644, "loss_total_v6": 0.7881117910146713, "step": 1738 }, { "epoch": 0.4010666666666667, "grad_norm": 25.953208923339844, "learning_rate": 9.069797815929614e-06, "loss": 13.375659942626953, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.330851785838604, "loss_total_v6": 0.8359787315130234, "step": 1739 }, { "epoch": 0.4012972972972973, "grad_norm": 23.4453067779541, "learning_rate": 9.0686276019926e-06, "loss": 12.564805030822754, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.2835913263261318, "loss_total_v6": 0.7853003069758415, "step": 1740 }, { "epoch": 0.4015279279279279, "grad_norm": 24.319324493408203, "learning_rate": 9.067456728033487e-06, "loss": 13.114164352416992, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.31070889718830585, "loss_total_v6": 0.819635272026062, "step": 1741 }, { "epoch": 0.4017585585585586, "grad_norm": 24.75042724609375, "learning_rate": 9.066285194242221e-06, "loss": 12.887287139892578, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.29779545217752457, "loss_total_v6": 0.8054554015398026, "step": 1742 }, { "epoch": 0.4019891891891892, "grad_norm": 22.786367416381836, "learning_rate": 9.06511300080885e-06, "loss": 12.583126068115234, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.28369883261621, "loss_total_v6": 0.7864453941583633, "step": 1743 }, { "epoch": 0.4022198198198198, "grad_norm": 24.677734375, "learning_rate": 9.063940147923529e-06, "loss": 12.608448028564453, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.28602910973131657, "loss_total_v6": 0.7880280092358589, "step": 1744 }, { "epoch": 0.40245045045045047, "grad_norm": 24.310752868652344, "learning_rate": 9.062766635776523e-06, "loss": 12.662445068359375, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.29309654608368874, "loss_total_v6": 0.7914028242230415, "step": 1745 }, { "epoch": 0.4026810810810811, "grad_norm": 23.958494186401367, "learning_rate": 9.061592464558198e-06, "loss": 12.602132797241211, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.27733357436954975, "loss_total_v6": 0.7876332551240921, "step": 1746 }, { "epoch": 0.4029117117117117, "grad_norm": 22.88248634338379, "learning_rate": 9.060417634459032e-06, "loss": 12.39274787902832, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.2615920379757881, "loss_total_v6": 0.7745467126369476, "step": 1747 }, { "epoch": 0.40314234234234236, "grad_norm": 24.590530395507812, "learning_rate": 9.05924214566961e-06, "loss": 12.928417205810547, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.3011137805879116, "loss_total_v6": 0.8080260455608368, "step": 1748 }, { "epoch": 0.403372972972973, "grad_norm": 23.988113403320312, "learning_rate": 9.058065998380623e-06, "loss": 12.387982368469238, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2705868352204561, "loss_total_v6": 0.7742489203810692, "step": 1749 }, { "epoch": 0.4036036036036036, "grad_norm": 24.051509857177734, "learning_rate": 9.056889192782865e-06, "loss": 12.872766494750977, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.29897846281528473, "loss_total_v6": 0.804547943174839, "step": 1750 }, { "epoch": 0.40383423423423426, "grad_norm": 22.694425582885742, "learning_rate": 9.055711729067245e-06, "loss": 12.857328414916992, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.29319174587726593, "loss_total_v6": 0.8035830035805702, "step": 1751 }, { "epoch": 0.40406486486486487, "grad_norm": 27.176082611083984, "learning_rate": 9.05453360742477e-06, "loss": 12.75010871887207, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.2908698096871376, "loss_total_v6": 0.7968817949295044, "step": 1752 }, { "epoch": 0.4042954954954955, "grad_norm": 28.856369018554688, "learning_rate": 9.053354828046558e-06, "loss": 12.58834457397461, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.27682289481163025, "loss_total_v6": 0.7867715805768967, "step": 1753 }, { "epoch": 0.40452612612612615, "grad_norm": 23.73065948486328, "learning_rate": 9.052175391123834e-06, "loss": 12.607421875, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2794951666146517, "loss_total_v6": 0.7879638224840164, "step": 1754 }, { "epoch": 0.40475675675675676, "grad_norm": 22.172101974487305, "learning_rate": 9.05099529684793e-06, "loss": 12.291701316833496, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.26177689619362354, "loss_total_v6": 0.7682313695549965, "step": 1755 }, { "epoch": 0.4049873873873874, "grad_norm": 27.199630737304688, "learning_rate": 9.049814545410282e-06, "loss": 13.105436325073242, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.30930889770388603, "loss_total_v6": 0.819089762866497, "step": 1756 }, { "epoch": 0.40521801801801804, "grad_norm": 23.39617347717285, "learning_rate": 9.048633137002435e-06, "loss": 12.887672424316406, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2989030256867409, "loss_total_v6": 0.8054795414209366, "step": 1757 }, { "epoch": 0.40544864864864866, "grad_norm": 21.639123916625977, "learning_rate": 9.047451071816038e-06, "loss": 12.906012535095215, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2942661475390196, "loss_total_v6": 0.8066257759928703, "step": 1758 }, { "epoch": 0.40567927927927927, "grad_norm": 25.77152442932129, "learning_rate": 9.046268350042851e-06, "loss": 12.729490280151367, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.29237352311611176, "loss_total_v6": 0.7955931127071381, "step": 1759 }, { "epoch": 0.4059099099099099, "grad_norm": 22.619905471801758, "learning_rate": 9.045084971874738e-06, "loss": 12.726442337036133, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.28275314159691334, "loss_total_v6": 0.7954026982188225, "step": 1760 }, { "epoch": 0.40614054054054055, "grad_norm": 23.65019416809082, "learning_rate": 9.043900937503667e-06, "loss": 12.472593307495117, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2746543139219284, "loss_total_v6": 0.779537096619606, "step": 1761 }, { "epoch": 0.40637117117117116, "grad_norm": 25.680763244628906, "learning_rate": 9.042716247121716e-06, "loss": 12.84130859375, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2946777604520321, "loss_total_v6": 0.8025818094611168, "step": 1762 }, { "epoch": 0.4066018018018018, "grad_norm": 24.87892723083496, "learning_rate": 9.041530900921067e-06, "loss": 12.690303802490234, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.29058045335114, "loss_total_v6": 0.7931439578533173, "step": 1763 }, { "epoch": 0.40683243243243244, "grad_norm": 23.824182510375977, "learning_rate": 9.04034489909401e-06, "loss": 12.738828659057617, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2805213239043951, "loss_total_v6": 0.7961768358945847, "step": 1764 }, { "epoch": 0.40706306306306306, "grad_norm": 23.533655166625977, "learning_rate": 9.039158241832944e-06, "loss": 13.198043823242188, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.3155393283814192, "loss_total_v6": 0.8248777538537979, "step": 1765 }, { "epoch": 0.40729369369369367, "grad_norm": 25.68902587890625, "learning_rate": 9.037970929330369e-06, "loss": 12.924301147460938, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.301878958940506, "loss_total_v6": 0.807768851518631, "step": 1766 }, { "epoch": 0.40752432432432434, "grad_norm": 23.91891860961914, "learning_rate": 9.036782961778892e-06, "loss": 13.016048431396484, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.30529382452368736, "loss_total_v6": 0.8135030567646027, "step": 1767 }, { "epoch": 0.40775495495495495, "grad_norm": 23.2220516204834, "learning_rate": 9.035594339371229e-06, "loss": 13.300285339355469, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.32260084338486195, "loss_total_v6": 0.8312678560614586, "step": 1768 }, { "epoch": 0.40798558558558556, "grad_norm": 26.37877082824707, "learning_rate": 9.034405062300201e-06, "loss": 12.302358627319336, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.2699655294418335, "loss_total_v6": 0.7688973844051361, "step": 1769 }, { "epoch": 0.40821621621621623, "grad_norm": 23.895708084106445, "learning_rate": 9.033215130758735e-06, "loss": 12.709322929382324, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.28534524887800217, "loss_total_v6": 0.7943326756358147, "step": 1770 }, { "epoch": 0.40844684684684684, "grad_norm": 27.095203399658203, "learning_rate": 9.032024544939866e-06, "loss": 13.413932800292969, "loss_alignment": 1.0400390625, "loss_alignment_w": 0.52001953125, "loss_sft": 0.3196634780615568, "loss_total_v6": 0.8383707702159882, "step": 1771 }, { "epoch": 0.40867747747747746, "grad_norm": 21.262941360473633, "learning_rate": 9.030833305036732e-06, "loss": 12.781030654907227, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.30540619418025017, "loss_total_v6": 0.7988144010305405, "step": 1772 }, { "epoch": 0.4089081081081081, "grad_norm": 23.72272300720215, "learning_rate": 9.02964141124258e-06, "loss": 13.324678421020508, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.32137882709503174, "loss_total_v6": 0.8327924013137817, "step": 1773 }, { "epoch": 0.40913873873873874, "grad_norm": 22.952367782592773, "learning_rate": 9.02844886375076e-06, "loss": 12.494232177734375, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2674160245805979, "loss_total_v6": 0.780889518558979, "step": 1774 }, { "epoch": 0.40936936936936935, "grad_norm": 25.841840744018555, "learning_rate": 9.02725566275473e-06, "loss": 13.300788879394531, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.32717943377792835, "loss_total_v6": 0.831299290060997, "step": 1775 }, { "epoch": 0.4096, "grad_norm": 23.918798446655273, "learning_rate": 9.026061808448055e-06, "loss": 12.464700698852539, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.27591576986014843, "loss_total_v6": 0.7790438234806061, "step": 1776 }, { "epoch": 0.40983063063063063, "grad_norm": 23.396347045898438, "learning_rate": 9.024867301024405e-06, "loss": 12.466585159301758, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2707540076225996, "loss_total_v6": 0.7791616022586823, "step": 1777 }, { "epoch": 0.41006126126126125, "grad_norm": 26.926021575927734, "learning_rate": 9.023672140677556e-06, "loss": 12.860030174255371, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2943677753210068, "loss_total_v6": 0.8037519305944443, "step": 1778 }, { "epoch": 0.4102918918918919, "grad_norm": 25.953630447387695, "learning_rate": 9.022476327601385e-06, "loss": 13.469301223754883, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.3302193358540535, "loss_total_v6": 0.8418313190340996, "step": 1779 }, { "epoch": 0.4105225225225225, "grad_norm": 24.79774284362793, "learning_rate": 9.021279861989884e-06, "loss": 12.98973274230957, "loss_alignment": 1.01708984375, "loss_alignment_w": 0.508544921875, "loss_sft": 0.30285557359457016, "loss_total_v6": 0.811858281493187, "step": 1780 }, { "epoch": 0.41075315315315314, "grad_norm": 26.96802520751953, "learning_rate": 9.020082744037147e-06, "loss": 13.198619842529297, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.3029257655143738, "loss_total_v6": 0.8249136805534363, "step": 1781 }, { "epoch": 0.4109837837837838, "grad_norm": 26.756406784057617, "learning_rate": 9.018884973937372e-06, "loss": 13.254863739013672, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.3099963627755642, "loss_total_v6": 0.8284289762377739, "step": 1782 }, { "epoch": 0.4112144144144144, "grad_norm": 23.672821044921875, "learning_rate": 9.017686551884864e-06, "loss": 12.678428649902344, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.28538271598517895, "loss_total_v6": 0.7924017608165741, "step": 1783 }, { "epoch": 0.41144504504504503, "grad_norm": 21.106143951416016, "learning_rate": 9.016487478074032e-06, "loss": 12.551695823669434, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.2855185866355896, "loss_total_v6": 0.7844809964299202, "step": 1784 }, { "epoch": 0.4116756756756757, "grad_norm": 22.794864654541016, "learning_rate": 9.015287752699395e-06, "loss": 12.331390380859375, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.27033045142889023, "loss_total_v6": 0.7707119137048721, "step": 1785 }, { "epoch": 0.4119063063063063, "grad_norm": 21.608909606933594, "learning_rate": 9.014087375955574e-06, "loss": 12.163592338562012, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.26043814420700073, "loss_total_v6": 0.7602245211601257, "step": 1786 }, { "epoch": 0.4121369369369369, "grad_norm": 24.746212005615234, "learning_rate": 9.012886348037296e-06, "loss": 13.187829971313477, "loss_alignment": 1.033203125, "loss_alignment_w": 0.5166015625, "loss_sft": 0.3063255175948143, "loss_total_v6": 0.8242393434047699, "step": 1787 }, { "epoch": 0.4123675675675676, "grad_norm": 23.86135482788086, "learning_rate": 9.011684669139398e-06, "loss": 12.691028594970703, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.29297560080885887, "loss_total_v6": 0.7931892350316048, "step": 1788 }, { "epoch": 0.4125981981981982, "grad_norm": 25.41025733947754, "learning_rate": 9.010482339456814e-06, "loss": 12.479445457458496, "loss_alignment": 0.99267578125, "loss_alignment_w": 0.496337890625, "loss_sft": 0.2828949596732855, "loss_total_v6": 0.7799653112888336, "step": 1789 }, { "epoch": 0.4128288288288288, "grad_norm": 24.088970184326172, "learning_rate": 9.009279359184594e-06, "loss": 12.805244445800781, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.29767267033457756, "loss_total_v6": 0.8003277257084846, "step": 1790 }, { "epoch": 0.41305945945945943, "grad_norm": 23.587657928466797, "learning_rate": 9.008075728517884e-06, "loss": 12.802820205688477, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.29092943854629993, "loss_total_v6": 0.8001762628555298, "step": 1791 }, { "epoch": 0.4132900900900901, "grad_norm": 22.76079559326172, "learning_rate": 9.006871447651941e-06, "loss": 12.78744125366211, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.29366088658571243, "loss_total_v6": 0.7992151081562042, "step": 1792 }, { "epoch": 0.4135207207207207, "grad_norm": 24.295434951782227, "learning_rate": 9.005666516782127e-06, "loss": 12.55905532836914, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.2871229387819767, "loss_total_v6": 0.7849409505724907, "step": 1793 }, { "epoch": 0.4137513513513513, "grad_norm": 27.754531860351562, "learning_rate": 9.004460936103909e-06, "loss": 12.392984390258789, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.2716776020824909, "loss_total_v6": 0.7745615243911743, "step": 1794 }, { "epoch": 0.413981981981982, "grad_norm": 23.892873764038086, "learning_rate": 9.003254705812858e-06, "loss": 12.428691864013672, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2724903356283903, "loss_total_v6": 0.7767933011054993, "step": 1795 }, { "epoch": 0.4142126126126126, "grad_norm": 23.16175651550293, "learning_rate": 9.002047826104652e-06, "loss": 12.785326957702637, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.28835603035986423, "loss_total_v6": 0.7990829274058342, "step": 1796 }, { "epoch": 0.4144432432432432, "grad_norm": 25.506874084472656, "learning_rate": 9.000840297175072e-06, "loss": 12.548449516296387, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.2894508093595505, "loss_total_v6": 0.7842780575156212, "step": 1797 }, { "epoch": 0.4146738738738739, "grad_norm": 23.066875457763672, "learning_rate": 8.999632119220008e-06, "loss": 13.226821899414062, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.31625461392104626, "loss_total_v6": 0.8266763612627983, "step": 1798 }, { "epoch": 0.4149045045045045, "grad_norm": 22.289621353149414, "learning_rate": 8.998423292435455e-06, "loss": 12.430950164794922, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2660396248102188, "loss_total_v6": 0.776934415102005, "step": 1799 }, { "epoch": 0.4151351351351351, "grad_norm": 23.232555389404297, "learning_rate": 8.997213817017508e-06, "loss": 12.630836486816406, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.28324748761951923, "loss_total_v6": 0.789427287876606, "step": 1800 }, { "epoch": 0.4153657657657658, "grad_norm": 24.737899780273438, "learning_rate": 8.996003693162372e-06, "loss": 12.676935195922852, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.28831062465906143, "loss_total_v6": 0.7923084422945976, "step": 1801 }, { "epoch": 0.4155963963963964, "grad_norm": 23.214704513549805, "learning_rate": 8.994792921066356e-06, "loss": 12.62813949584961, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.29185271821916103, "loss_total_v6": 0.7892586812376976, "step": 1802 }, { "epoch": 0.415827027027027, "grad_norm": 25.70131492614746, "learning_rate": 8.993581500925877e-06, "loss": 12.541271209716797, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.2812964767217636, "loss_total_v6": 0.7838294431567192, "step": 1803 }, { "epoch": 0.4160576576576577, "grad_norm": 23.228227615356445, "learning_rate": 8.992369432937451e-06, "loss": 11.951251029968262, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.24504579603672028, "loss_total_v6": 0.7469531670212746, "step": 1804 }, { "epoch": 0.4162882882882883, "grad_norm": 24.77968406677246, "learning_rate": 8.991156717297702e-06, "loss": 12.74376106262207, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.28497995994985104, "loss_total_v6": 0.7964850962162018, "step": 1805 }, { "epoch": 0.4165189189189189, "grad_norm": 23.932666778564453, "learning_rate": 8.989943354203364e-06, "loss": 13.279911041259766, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.32302122935652733, "loss_total_v6": 0.8299944624304771, "step": 1806 }, { "epoch": 0.41674954954954957, "grad_norm": 24.326475143432617, "learning_rate": 8.988729343851268e-06, "loss": 12.506000518798828, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.27088285610079765, "loss_total_v6": 0.781625047326088, "step": 1807 }, { "epoch": 0.4169801801801802, "grad_norm": 25.281139373779297, "learning_rate": 8.987514686438353e-06, "loss": 12.798527717590332, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.29073746502399445, "loss_total_v6": 0.7999079748988152, "step": 1808 }, { "epoch": 0.4172108108108108, "grad_norm": 26.659215927124023, "learning_rate": 8.986299382161665e-06, "loss": 12.967809677124023, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.3068106733262539, "loss_total_v6": 0.8104880601167679, "step": 1809 }, { "epoch": 0.41744144144144146, "grad_norm": 26.41988182067871, "learning_rate": 8.985083431218354e-06, "loss": 12.23311710357666, "loss_alignment": 1.00732421875, "loss_alignment_w": 0.503662109375, "loss_sft": 0.26139603182673454, "loss_total_v6": 0.7645698487758636, "step": 1810 }, { "epoch": 0.4176720720720721, "grad_norm": 24.10870361328125, "learning_rate": 8.983866833805673e-06, "loss": 12.528121948242188, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.2795438654720783, "loss_total_v6": 0.7830075994133949, "step": 1811 }, { "epoch": 0.4179027027027027, "grad_norm": 25.102697372436523, "learning_rate": 8.982649590120982e-06, "loss": 12.620807647705078, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.28014879301190376, "loss_total_v6": 0.7888005077838898, "step": 1812 }, { "epoch": 0.41813333333333336, "grad_norm": 23.810855865478516, "learning_rate": 8.981431700361745e-06, "loss": 12.696599960327148, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.2946971822530031, "loss_total_v6": 0.7935375273227692, "step": 1813 }, { "epoch": 0.41836396396396397, "grad_norm": 24.648221969604492, "learning_rate": 8.980213164725531e-06, "loss": 12.490921974182129, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.27523526549339294, "loss_total_v6": 0.7806826308369637, "step": 1814 }, { "epoch": 0.4185945945945946, "grad_norm": 27.555330276489258, "learning_rate": 8.978993983410011e-06, "loss": 12.6165132522583, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.2913854271173477, "loss_total_v6": 0.7885320633649826, "step": 1815 }, { "epoch": 0.41882522522522525, "grad_norm": 24.75269889831543, "learning_rate": 8.977774156612968e-06, "loss": 12.629167556762695, "loss_alignment": 1.01220703125, "loss_alignment_w": 0.506103515625, "loss_sft": 0.2837077472358942, "loss_total_v6": 0.7893229722976685, "step": 1816 }, { "epoch": 0.41905585585585586, "grad_norm": 22.348772048950195, "learning_rate": 8.976553684532281e-06, "loss": 12.082160949707031, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.2542805466800928, "loss_total_v6": 0.7551350295543671, "step": 1817 }, { "epoch": 0.4192864864864865, "grad_norm": 24.79530906677246, "learning_rate": 8.975332567365942e-06, "loss": 12.693198204040527, "loss_alignment": 1.00732421875, "loss_alignment_w": 0.503662109375, "loss_sft": 0.28986112028360367, "loss_total_v6": 0.7933249026536942, "step": 1818 }, { "epoch": 0.41951711711711714, "grad_norm": 21.90372085571289, "learning_rate": 8.97411080531204e-06, "loss": 12.079626083374023, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.2566398214548826, "loss_total_v6": 0.7549766004085541, "step": 1819 }, { "epoch": 0.41974774774774776, "grad_norm": 22.754512786865234, "learning_rate": 8.972888398568772e-06, "loss": 12.704084396362305, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.29144180193543434, "loss_total_v6": 0.7940052822232246, "step": 1820 }, { "epoch": 0.41997837837837837, "grad_norm": 21.931049346923828, "learning_rate": 8.971665347334442e-06, "loss": 12.648494720458984, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2808874174952507, "loss_total_v6": 0.7905309721827507, "step": 1821 }, { "epoch": 0.420209009009009, "grad_norm": 26.383913040161133, "learning_rate": 8.970441651807454e-06, "loss": 12.913214683532715, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.3050770051777363, "loss_total_v6": 0.8070758953690529, "step": 1822 }, { "epoch": 0.42043963963963965, "grad_norm": 23.966320037841797, "learning_rate": 8.969217312186319e-06, "loss": 12.651792526245117, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.272716399282217, "loss_total_v6": 0.7907370328903198, "step": 1823 }, { "epoch": 0.42067027027027026, "grad_norm": 27.729175567626953, "learning_rate": 8.967992328669655e-06, "loss": 12.42154312133789, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.26706913113594055, "loss_total_v6": 0.7763464897871017, "step": 1824 }, { "epoch": 0.4209009009009009, "grad_norm": 21.216691970825195, "learning_rate": 8.966766701456177e-06, "loss": 12.96243953704834, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.31051869317889214, "loss_total_v6": 0.8101524785161018, "step": 1825 }, { "epoch": 0.42113153153153154, "grad_norm": 24.692811965942383, "learning_rate": 8.965540430744713e-06, "loss": 13.020204544067383, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.30253230407834053, "loss_total_v6": 0.8137627840042114, "step": 1826 }, { "epoch": 0.42136216216216216, "grad_norm": 24.886747360229492, "learning_rate": 8.964313516734188e-06, "loss": 12.995363235473633, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.3077699691057205, "loss_total_v6": 0.8122102618217468, "step": 1827 }, { "epoch": 0.42159279279279277, "grad_norm": 25.788137435913086, "learning_rate": 8.963085959623638e-06, "loss": 13.219366073608398, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.30765571258962154, "loss_total_v6": 0.8262103945016861, "step": 1828 }, { "epoch": 0.42182342342342344, "grad_norm": 27.700260162353516, "learning_rate": 8.961857759612198e-06, "loss": 12.47222900390625, "loss_alignment": 1.01806640625, "loss_alignment_w": 0.509033203125, "loss_sft": 0.2703132312744856, "loss_total_v6": 0.7795142754912376, "step": 1829 }, { "epoch": 0.42205405405405405, "grad_norm": 27.536083221435547, "learning_rate": 8.96062891689911e-06, "loss": 13.54158878326416, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.3365989550948143, "loss_total_v6": 0.8463493064045906, "step": 1830 }, { "epoch": 0.42228468468468466, "grad_norm": 24.631916046142578, "learning_rate": 8.959399431683721e-06, "loss": 12.833385467529297, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2945486791431904, "loss_total_v6": 0.8020865470170975, "step": 1831 }, { "epoch": 0.42251531531531533, "grad_norm": 25.29170799255371, "learning_rate": 8.95816930416548e-06, "loss": 12.206954956054688, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.257334653288126, "loss_total_v6": 0.7629346549510956, "step": 1832 }, { "epoch": 0.42274594594594594, "grad_norm": 25.049375534057617, "learning_rate": 8.956938534543939e-06, "loss": 13.105989456176758, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.3125020172446966, "loss_total_v6": 0.819124348461628, "step": 1833 }, { "epoch": 0.42297657657657656, "grad_norm": 23.64631462097168, "learning_rate": 8.95570712301876e-06, "loss": 12.736686706542969, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.28871868923306465, "loss_total_v6": 0.7960429340600967, "step": 1834 }, { "epoch": 0.4232072072072072, "grad_norm": 21.96673011779785, "learning_rate": 8.954475069789703e-06, "loss": 12.81940746307373, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.29350726306438446, "loss_total_v6": 0.8012129664421082, "step": 1835 }, { "epoch": 0.42343783783783784, "grad_norm": 27.69236946105957, "learning_rate": 8.953242375056634e-06, "loss": 12.770261764526367, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.29196153581142426, "loss_total_v6": 0.7981413453817368, "step": 1836 }, { "epoch": 0.42366846846846845, "grad_norm": 23.883642196655273, "learning_rate": 8.952009039019526e-06, "loss": 12.58774471282959, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.28243102319538593, "loss_total_v6": 0.7867340445518494, "step": 1837 }, { "epoch": 0.4238990990990991, "grad_norm": 27.4970760345459, "learning_rate": 8.950775061878453e-06, "loss": 12.730464935302734, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.28515602089464664, "loss_total_v6": 0.7956540882587433, "step": 1838 }, { "epoch": 0.42412972972972973, "grad_norm": 23.731042861938477, "learning_rate": 8.949540443833591e-06, "loss": 12.622963905334473, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.2748361602425575, "loss_total_v6": 0.7889352738857269, "step": 1839 }, { "epoch": 0.42436036036036034, "grad_norm": 25.279621124267578, "learning_rate": 8.948305185085226e-06, "loss": 12.483007431030273, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.2779296524822712, "loss_total_v6": 0.7801879495382309, "step": 1840 }, { "epoch": 0.424590990990991, "grad_norm": 25.740671157836914, "learning_rate": 8.947069285833741e-06, "loss": 12.814416885375977, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.29470596835017204, "loss_total_v6": 0.8009010329842567, "step": 1841 }, { "epoch": 0.4248216216216216, "grad_norm": 24.41035270690918, "learning_rate": 8.945832746279631e-06, "loss": 12.784601211547852, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2852284386754036, "loss_total_v6": 0.7990376427769661, "step": 1842 }, { "epoch": 0.42505225225225224, "grad_norm": 27.872642517089844, "learning_rate": 8.944595566623483e-06, "loss": 12.94649887084961, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.29589636251330376, "loss_total_v6": 0.809156209230423, "step": 1843 }, { "epoch": 0.4252828828828829, "grad_norm": 25.839921951293945, "learning_rate": 8.943357747066004e-06, "loss": 12.900293350219727, "loss_alignment": 1.02685546875, "loss_alignment_w": 0.513427734375, "loss_sft": 0.29241336323320866, "loss_total_v6": 0.8062683045864105, "step": 1844 }, { "epoch": 0.4255135135135135, "grad_norm": 28.418210983276367, "learning_rate": 8.942119287807991e-06, "loss": 12.557586669921875, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.28549005649983883, "loss_total_v6": 0.7848491966724396, "step": 1845 }, { "epoch": 0.42574414414414413, "grad_norm": 23.047861099243164, "learning_rate": 8.940880189050349e-06, "loss": 12.288816452026367, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.2679747547954321, "loss_total_v6": 0.7680510208010674, "step": 1846 }, { "epoch": 0.4259747747747748, "grad_norm": 25.675405502319336, "learning_rate": 8.93964045099409e-06, "loss": 12.446413040161133, "loss_alignment": 0.99267578125, "loss_alignment_w": 0.496337890625, "loss_sft": 0.2810288816690445, "loss_total_v6": 0.7779008150100708, "step": 1847 }, { "epoch": 0.4262054054054054, "grad_norm": 24.604581832885742, "learning_rate": 8.938400073840325e-06, "loss": 12.847639083862305, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2889394089579582, "loss_total_v6": 0.8029774650931358, "step": 1848 }, { "epoch": 0.426436036036036, "grad_norm": 25.26367950439453, "learning_rate": 8.937159057790272e-06, "loss": 12.482664108276367, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.268051041290164, "loss_total_v6": 0.7801665365695953, "step": 1849 }, { "epoch": 0.4266666666666667, "grad_norm": 22.667613983154297, "learning_rate": 8.935917403045251e-06, "loss": 12.929550170898438, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.30278682336211205, "loss_total_v6": 0.80809685587883, "step": 1850 }, { "epoch": 0.4268972972972973, "grad_norm": 23.30341339111328, "learning_rate": 8.934675109806688e-06, "loss": 12.795257568359375, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.29181476682424545, "loss_total_v6": 0.799703560769558, "step": 1851 }, { "epoch": 0.4271279279279279, "grad_norm": 28.948516845703125, "learning_rate": 8.933432178276108e-06, "loss": 12.522903442382812, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.27215287648141384, "loss_total_v6": 0.7826814278960228, "step": 1852 }, { "epoch": 0.42735855855855853, "grad_norm": 24.132633209228516, "learning_rate": 8.932188608655141e-06, "loss": 12.738065719604492, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.2967852298170328, "loss_total_v6": 0.7961291149258614, "step": 1853 }, { "epoch": 0.4275891891891892, "grad_norm": 23.317148208618164, "learning_rate": 8.930944401145527e-06, "loss": 13.06513786315918, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.3052186220884323, "loss_total_v6": 0.8165711238980293, "step": 1854 }, { "epoch": 0.4278198198198198, "grad_norm": 25.09851837158203, "learning_rate": 8.9296995559491e-06, "loss": 12.420570373535156, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.2736458908766508, "loss_total_v6": 0.7762856706976891, "step": 1855 }, { "epoch": 0.4280504504504504, "grad_norm": 25.919208526611328, "learning_rate": 8.928454073267801e-06, "loss": 12.606891632080078, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.27389270812273026, "loss_total_v6": 0.7879307940602303, "step": 1856 }, { "epoch": 0.4282810810810811, "grad_norm": 25.039531707763672, "learning_rate": 8.927207953303678e-06, "loss": 12.315202713012695, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.2707530315965414, "loss_total_v6": 0.7697001695632935, "step": 1857 }, { "epoch": 0.4285117117117117, "grad_norm": 24.674083709716797, "learning_rate": 8.92596119625888e-06, "loss": 12.45578670501709, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.27972265891730785, "loss_total_v6": 0.7784866839647293, "step": 1858 }, { "epoch": 0.4287423423423423, "grad_norm": 22.62765884399414, "learning_rate": 8.924713802335656e-06, "loss": 12.400874137878418, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.27850308269262314, "loss_total_v6": 0.7750545963644981, "step": 1859 }, { "epoch": 0.428972972972973, "grad_norm": 24.952028274536133, "learning_rate": 8.923465771736361e-06, "loss": 13.198301315307617, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.30618657171726227, "loss_total_v6": 0.8248938396573067, "step": 1860 }, { "epoch": 0.4292036036036036, "grad_norm": 28.84981346130371, "learning_rate": 8.922217104663454e-06, "loss": 12.777173042297363, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.29117274656891823, "loss_total_v6": 0.7985732927918434, "step": 1861 }, { "epoch": 0.4294342342342342, "grad_norm": 23.25343132019043, "learning_rate": 8.920967801319498e-06, "loss": 12.87352466583252, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.29327322728931904, "loss_total_v6": 0.8045952692627907, "step": 1862 }, { "epoch": 0.4296648648648649, "grad_norm": 25.334976196289062, "learning_rate": 8.919717861907158e-06, "loss": 12.603734970092773, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.28047030232846737, "loss_total_v6": 0.7877334654331207, "step": 1863 }, { "epoch": 0.4298954954954955, "grad_norm": 21.28816795349121, "learning_rate": 8.9184672866292e-06, "loss": 12.422967910766602, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.26860770396888256, "loss_total_v6": 0.7764354720711708, "step": 1864 }, { "epoch": 0.4301261261261261, "grad_norm": 25.595216751098633, "learning_rate": 8.917216075688496e-06, "loss": 12.541993141174316, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.2810058929026127, "loss_total_v6": 0.7838745638728142, "step": 1865 }, { "epoch": 0.4303567567567568, "grad_norm": 23.450353622436523, "learning_rate": 8.915964229288021e-06, "loss": 13.36800765991211, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.3363244738429785, "loss_total_v6": 0.8355005159974098, "step": 1866 }, { "epoch": 0.4305873873873874, "grad_norm": 21.862934112548828, "learning_rate": 8.91471174763085e-06, "loss": 12.505392074584961, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.27298106625676155, "loss_total_v6": 0.7815870195627213, "step": 1867 }, { "epoch": 0.430818018018018, "grad_norm": 22.562551498413086, "learning_rate": 8.913458630920169e-06, "loss": 12.615583419799805, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.282278910279274, "loss_total_v6": 0.7884739711880684, "step": 1868 }, { "epoch": 0.43104864864864867, "grad_norm": 27.038223266601562, "learning_rate": 8.912204879359255e-06, "loss": 12.288936614990234, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2588269840925932, "loss_total_v6": 0.7680585607886314, "step": 1869 }, { "epoch": 0.4312792792792793, "grad_norm": 23.967145919799805, "learning_rate": 8.910950493151498e-06, "loss": 12.36834716796875, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.2695884685963392, "loss_total_v6": 0.7730216905474663, "step": 1870 }, { "epoch": 0.4315099099099099, "grad_norm": 22.795520782470703, "learning_rate": 8.909695472500387e-06, "loss": 12.997623443603516, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.3060190621763468, "loss_total_v6": 0.8123514652252197, "step": 1871 }, { "epoch": 0.43174054054054056, "grad_norm": 27.123756408691406, "learning_rate": 8.908439817609514e-06, "loss": 12.725412368774414, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2896924912929535, "loss_total_v6": 0.7953382357954979, "step": 1872 }, { "epoch": 0.4319711711711712, "grad_norm": 21.356483459472656, "learning_rate": 8.907183528682575e-06, "loss": 12.500494003295898, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2755893524736166, "loss_total_v6": 0.7812808603048325, "step": 1873 }, { "epoch": 0.4322018018018018, "grad_norm": 25.8114013671875, "learning_rate": 8.905926605923365e-06, "loss": 12.850631713867188, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.286990150809288, "loss_total_v6": 0.803164467215538, "step": 1874 }, { "epoch": 0.43243243243243246, "grad_norm": 25.6119384765625, "learning_rate": 8.90466904953579e-06, "loss": 12.144754409790039, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.2613359708338976, "loss_total_v6": 0.7590471282601357, "step": 1875 }, { "epoch": 0.43266306306306307, "grad_norm": 22.720373153686523, "learning_rate": 8.903410859723848e-06, "loss": 12.477066993713379, "loss_alignment": 1.01513671875, "loss_alignment_w": 0.507568359375, "loss_sft": 0.2741556614637375, "loss_total_v6": 0.7798166871070862, "step": 1876 }, { "epoch": 0.4328936936936937, "grad_norm": 25.291095733642578, "learning_rate": 8.902152036691649e-06, "loss": 12.818214416503906, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.2951417211443186, "loss_total_v6": 0.8011384308338165, "step": 1877 }, { "epoch": 0.43312432432432435, "grad_norm": 25.599552154541016, "learning_rate": 8.900892580643404e-06, "loss": 12.805999755859375, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.29512593150138855, "loss_total_v6": 0.8003749698400497, "step": 1878 }, { "epoch": 0.43335495495495496, "grad_norm": 27.96613883972168, "learning_rate": 8.899632491783419e-06, "loss": 12.456555366516113, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.2781990207731724, "loss_total_v6": 0.7785347178578377, "step": 1879 }, { "epoch": 0.4335855855855856, "grad_norm": 24.391887664794922, "learning_rate": 8.898371770316113e-06, "loss": 12.747206687927246, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.28624808229506016, "loss_total_v6": 0.7967003881931305, "step": 1880 }, { "epoch": 0.43381621621621624, "grad_norm": 23.208402633666992, "learning_rate": 8.897110416446002e-06, "loss": 12.55550479888916, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.27744060195982456, "loss_total_v6": 0.7847190424799919, "step": 1881 }, { "epoch": 0.43404684684684686, "grad_norm": 24.503005981445312, "learning_rate": 8.895848430377705e-06, "loss": 12.910614013671875, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.29771227575838566, "loss_total_v6": 0.806913323700428, "step": 1882 }, { "epoch": 0.43427747747747747, "grad_norm": 25.370498657226562, "learning_rate": 8.894585812315945e-06, "loss": 12.962471008300781, "loss_alignment": 1.01220703125, "loss_alignment_w": 0.506103515625, "loss_sft": 0.3040051516145468, "loss_total_v6": 0.8101544305682182, "step": 1883 }, { "epoch": 0.4345081081081081, "grad_norm": 24.7705020904541, "learning_rate": 8.893322562465547e-06, "loss": 12.848125457763672, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.29369998164474964, "loss_total_v6": 0.8030078113079071, "step": 1884 }, { "epoch": 0.43473873873873875, "grad_norm": 24.829374313354492, "learning_rate": 8.89205868103144e-06, "loss": 13.303524017333984, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.3172490932047367, "loss_total_v6": 0.8314702585339546, "step": 1885 }, { "epoch": 0.43496936936936936, "grad_norm": 25.74825096130371, "learning_rate": 8.89079416821865e-06, "loss": 13.054033279418945, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.3085223324596882, "loss_total_v6": 0.8158770874142647, "step": 1886 }, { "epoch": 0.4352, "grad_norm": 28.362964630126953, "learning_rate": 8.889529024232311e-06, "loss": 12.791454315185547, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.28645008243620396, "loss_total_v6": 0.7994658425450325, "step": 1887 }, { "epoch": 0.43543063063063064, "grad_norm": 22.67374610900879, "learning_rate": 8.888263249277656e-06, "loss": 12.568618774414062, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2849588394165039, "loss_total_v6": 0.7855386808514595, "step": 1888 }, { "epoch": 0.43566126126126126, "grad_norm": 23.3035831451416, "learning_rate": 8.886996843560025e-06, "loss": 12.914990425109863, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.29624634981155396, "loss_total_v6": 0.8071868941187859, "step": 1889 }, { "epoch": 0.43589189189189187, "grad_norm": 23.584148406982422, "learning_rate": 8.885729807284855e-06, "loss": 12.487537384033203, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.27513047866523266, "loss_total_v6": 0.7804710268974304, "step": 1890 }, { "epoch": 0.43612252252252254, "grad_norm": 24.546091079711914, "learning_rate": 8.884462140657689e-06, "loss": 13.23662281036377, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.3180726245045662, "loss_total_v6": 0.8272889032959938, "step": 1891 }, { "epoch": 0.43635315315315315, "grad_norm": 23.698713302612305, "learning_rate": 8.883193843884169e-06, "loss": 13.22851848602295, "loss_alignment": 1.01806640625, "loss_alignment_w": 0.509033203125, "loss_sft": 0.3188172969967127, "loss_total_v6": 0.8267823830246925, "step": 1892 }, { "epoch": 0.43658378378378376, "grad_norm": 24.32585906982422, "learning_rate": 8.881924917170041e-06, "loss": 12.489950180053711, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2736180815845728, "loss_total_v6": 0.780621849000454, "step": 1893 }, { "epoch": 0.43681441441441443, "grad_norm": 22.87884521484375, "learning_rate": 8.880655360721157e-06, "loss": 12.836421966552734, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.28933689557015896, "loss_total_v6": 0.8022763729095459, "step": 1894 }, { "epoch": 0.43704504504504504, "grad_norm": 22.283615112304688, "learning_rate": 8.879385174743462e-06, "loss": 12.80717945098877, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2894013524055481, "loss_total_v6": 0.8004487007856369, "step": 1895 }, { "epoch": 0.43727567567567566, "grad_norm": 22.24859619140625, "learning_rate": 8.878114359443012e-06, "loss": 12.467384338378906, "loss_alignment": 0.990234375, "loss_alignment_w": 0.4951171875, "loss_sft": 0.28476578183472157, "loss_total_v6": 0.7792115658521652, "step": 1896 }, { "epoch": 0.4375063063063063, "grad_norm": 23.17738914489746, "learning_rate": 8.87684291502596e-06, "loss": 12.855979919433594, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.29278712160885334, "loss_total_v6": 0.8034988045692444, "step": 1897 }, { "epoch": 0.43773693693693694, "grad_norm": 25.867576599121094, "learning_rate": 8.875570841698564e-06, "loss": 12.687551498413086, "loss_alignment": 1.02294921875, "loss_alignment_w": 0.511474609375, "loss_sft": 0.2830079570412636, "loss_total_v6": 0.7929719388484955, "step": 1898 }, { "epoch": 0.43796756756756755, "grad_norm": 25.540180206298828, "learning_rate": 8.874298139667182e-06, "loss": 12.09090805053711, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.2588708642870188, "loss_total_v6": 0.7556817680597305, "step": 1899 }, { "epoch": 0.4381981981981982, "grad_norm": 23.15928840637207, "learning_rate": 8.873024809138272e-06, "loss": 12.792181015014648, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.29325526393949986, "loss_total_v6": 0.7995113655924797, "step": 1900 }, { "epoch": 0.43842882882882883, "grad_norm": 25.083419799804688, "learning_rate": 8.871750850318402e-06, "loss": 12.977887153625488, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.302435714751482, "loss_total_v6": 0.811117947101593, "step": 1901 }, { "epoch": 0.43865945945945944, "grad_norm": 26.048107147216797, "learning_rate": 8.870476263414235e-06, "loss": 12.92796516418457, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.30233683064579964, "loss_total_v6": 0.8079978376626968, "step": 1902 }, { "epoch": 0.4388900900900901, "grad_norm": 25.32650375366211, "learning_rate": 8.869201048632531e-06, "loss": 13.312932968139648, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.32320817932486534, "loss_total_v6": 0.8320582509040833, "step": 1903 }, { "epoch": 0.4391207207207207, "grad_norm": 26.195148468017578, "learning_rate": 8.867925206180166e-06, "loss": 12.14289665222168, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.25687109865248203, "loss_total_v6": 0.7589310705661774, "step": 1904 }, { "epoch": 0.43935135135135134, "grad_norm": 22.332929611206055, "learning_rate": 8.866648736264108e-06, "loss": 12.499284744262695, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.27851977199316025, "loss_total_v6": 0.781205303966999, "step": 1905 }, { "epoch": 0.439581981981982, "grad_norm": 25.180768966674805, "learning_rate": 8.865371639091426e-06, "loss": 12.865677833557129, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.3027621228247881, "loss_total_v6": 0.8041048720479012, "step": 1906 }, { "epoch": 0.4398126126126126, "grad_norm": 25.677221298217773, "learning_rate": 8.864093914869298e-06, "loss": 12.720223426818848, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.28790334425866604, "loss_total_v6": 0.7950139418244362, "step": 1907 }, { "epoch": 0.44004324324324323, "grad_norm": 23.52196502685547, "learning_rate": 8.862815563804996e-06, "loss": 12.713056564331055, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2862652614712715, "loss_total_v6": 0.7945660501718521, "step": 1908 }, { "epoch": 0.4402738738738739, "grad_norm": 25.59962272644043, "learning_rate": 8.8615365861059e-06, "loss": 12.32994556427002, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2669747658073902, "loss_total_v6": 0.7706216052174568, "step": 1909 }, { "epoch": 0.4405045045045045, "grad_norm": 25.125272750854492, "learning_rate": 8.860256981979485e-06, "loss": 12.996633529663086, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.30699480697512627, "loss_total_v6": 0.8122896179556847, "step": 1910 }, { "epoch": 0.4407351351351351, "grad_norm": 26.55634307861328, "learning_rate": 8.858976751633334e-06, "loss": 12.579577445983887, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.2813102211803198, "loss_total_v6": 0.7862235531210899, "step": 1911 }, { "epoch": 0.44096576576576574, "grad_norm": 21.861984252929688, "learning_rate": 8.857695895275127e-06, "loss": 12.556795120239258, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.2733708508312702, "loss_total_v6": 0.7847996652126312, "step": 1912 }, { "epoch": 0.4411963963963964, "grad_norm": 24.22409439086914, "learning_rate": 8.856414413112651e-06, "loss": 13.056330680847168, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.30501906387507915, "loss_total_v6": 0.8160206377506256, "step": 1913 }, { "epoch": 0.441427027027027, "grad_norm": 23.588775634765625, "learning_rate": 8.855132305353787e-06, "loss": 12.96541976928711, "loss_alignment": 1.03125, "loss_alignment_w": 0.515625, "loss_sft": 0.29567504674196243, "loss_total_v6": 0.8103387132287025, "step": 1914 }, { "epoch": 0.44165765765765763, "grad_norm": 24.904823303222656, "learning_rate": 8.853849572206523e-06, "loss": 12.661788940429688, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2815808802843094, "loss_total_v6": 0.7913617864251137, "step": 1915 }, { "epoch": 0.4418882882882883, "grad_norm": 24.116928100585938, "learning_rate": 8.852566213878947e-06, "loss": 13.025081634521484, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.3068349305540323, "loss_total_v6": 0.8140676394104958, "step": 1916 }, { "epoch": 0.4421189189189189, "grad_norm": 21.963653564453125, "learning_rate": 8.851282230579248e-06, "loss": 12.960253715515137, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.30224914848804474, "loss_total_v6": 0.8100158497691154, "step": 1917 }, { "epoch": 0.4423495495495495, "grad_norm": 23.451984405517578, "learning_rate": 8.849997622515718e-06, "loss": 12.828311920166016, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.2840693350881338, "loss_total_v6": 0.801769532263279, "step": 1918 }, { "epoch": 0.4425801801801802, "grad_norm": 26.777685165405273, "learning_rate": 8.848712389896748e-06, "loss": 12.979571342468262, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.30663033574819565, "loss_total_v6": 0.8112232461571693, "step": 1919 }, { "epoch": 0.4428108108108108, "grad_norm": 22.139404296875, "learning_rate": 8.84742653293083e-06, "loss": 13.036989212036133, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.302543718367815, "loss_total_v6": 0.8148118406534195, "step": 1920 }, { "epoch": 0.4430414414414414, "grad_norm": 24.76690101623535, "learning_rate": 8.846140051826562e-06, "loss": 12.573484420776367, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.2834929507225752, "loss_total_v6": 0.7858428061008453, "step": 1921 }, { "epoch": 0.4432720720720721, "grad_norm": 24.397594451904297, "learning_rate": 8.844852946792636e-06, "loss": 12.543691635131836, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.28401124849915504, "loss_total_v6": 0.7839807122945786, "step": 1922 }, { "epoch": 0.4435027027027027, "grad_norm": 24.025957107543945, "learning_rate": 8.843565218037853e-06, "loss": 12.38602066040039, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2623159773647785, "loss_total_v6": 0.774126261472702, "step": 1923 }, { "epoch": 0.4437333333333333, "grad_norm": 23.760831832885742, "learning_rate": 8.84227686577111e-06, "loss": 13.209511756896973, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.3119531497359276, "loss_total_v6": 0.8255944922566414, "step": 1924 }, { "epoch": 0.443963963963964, "grad_norm": 22.87009620666504, "learning_rate": 8.840987890201404e-06, "loss": 12.678236961364746, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2788857836276293, "loss_total_v6": 0.7923897802829742, "step": 1925 }, { "epoch": 0.4441945945945946, "grad_norm": 25.192211151123047, "learning_rate": 8.839698291537836e-06, "loss": 12.46753215789795, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.2792818211019039, "loss_total_v6": 0.7792207598686218, "step": 1926 }, { "epoch": 0.4444252252252252, "grad_norm": 22.975074768066406, "learning_rate": 8.838408069989613e-06, "loss": 12.883893013000488, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.299887552857399, "loss_total_v6": 0.8052433580160141, "step": 1927 }, { "epoch": 0.4446558558558559, "grad_norm": 26.544822692871094, "learning_rate": 8.837117225766033e-06, "loss": 12.447901725769043, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2711578942835331, "loss_total_v6": 0.777993842959404, "step": 1928 }, { "epoch": 0.4448864864864865, "grad_norm": 23.515567779541016, "learning_rate": 8.8358257590765e-06, "loss": 12.207267761230469, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.263763003051281, "loss_total_v6": 0.7629542797803879, "step": 1929 }, { "epoch": 0.4451171171171171, "grad_norm": 24.393510818481445, "learning_rate": 8.834533670130518e-06, "loss": 13.035989761352539, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.3075318969786167, "loss_total_v6": 0.8147493228316307, "step": 1930 }, { "epoch": 0.44534774774774777, "grad_norm": 23.26384162902832, "learning_rate": 8.833240959137695e-06, "loss": 13.27086067199707, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.32855902053415775, "loss_total_v6": 0.8294287845492363, "step": 1931 }, { "epoch": 0.4455783783783784, "grad_norm": 23.38255500793457, "learning_rate": 8.831947626307735e-06, "loss": 13.213739395141602, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.316016785800457, "loss_total_v6": 0.8258586898446083, "step": 1932 }, { "epoch": 0.445809009009009, "grad_norm": 26.31210708618164, "learning_rate": 8.830653671850448e-06, "loss": 13.105459213256836, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.31265199929475784, "loss_total_v6": 0.8190912082791328, "step": 1933 }, { "epoch": 0.44603963963963966, "grad_norm": 23.41739845275879, "learning_rate": 8.829359095975741e-06, "loss": 12.721333503723145, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.27871069125831127, "loss_total_v6": 0.7950833514332771, "step": 1934 }, { "epoch": 0.4462702702702703, "grad_norm": 26.071752548217773, "learning_rate": 8.828063898893623e-06, "loss": 13.179298400878906, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.32237862050533295, "loss_total_v6": 0.8237061351537704, "step": 1935 }, { "epoch": 0.4465009009009009, "grad_norm": 26.349462509155273, "learning_rate": 8.826768080814205e-06, "loss": 12.438308715820312, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.26889518089592457, "loss_total_v6": 0.7773943170905113, "step": 1936 }, { "epoch": 0.44673153153153156, "grad_norm": 21.30394172668457, "learning_rate": 8.825471641947696e-06, "loss": 12.999267578125, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.30108644254505634, "loss_total_v6": 0.8124542236328125, "step": 1937 }, { "epoch": 0.44696216216216217, "grad_norm": 26.810436248779297, "learning_rate": 8.824174582504409e-06, "loss": 12.877426147460938, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.3014516420662403, "loss_total_v6": 0.804839126765728, "step": 1938 }, { "epoch": 0.4471927927927928, "grad_norm": 22.4686279296875, "learning_rate": 8.822876902694753e-06, "loss": 12.805143356323242, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.28796184808015823, "loss_total_v6": 0.8003214672207832, "step": 1939 }, { "epoch": 0.44742342342342345, "grad_norm": 24.553054809570312, "learning_rate": 8.821578602729242e-06, "loss": 12.857595443725586, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.29815235175192356, "loss_total_v6": 0.8035997450351715, "step": 1940 }, { "epoch": 0.44765405405405406, "grad_norm": 23.00379753112793, "learning_rate": 8.82027968281849e-06, "loss": 12.556156158447266, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.276458952575922, "loss_total_v6": 0.7847597301006317, "step": 1941 }, { "epoch": 0.4478846846846847, "grad_norm": 24.169233322143555, "learning_rate": 8.818980143173212e-06, "loss": 12.998505592346191, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.30879021994769573, "loss_total_v6": 0.8124065771698952, "step": 1942 }, { "epoch": 0.4481153153153153, "grad_norm": 25.057235717773438, "learning_rate": 8.817679984004219e-06, "loss": 12.963376998901367, "loss_alignment": 1.02099609375, "loss_alignment_w": 0.510498046875, "loss_sft": 0.299575736746192, "loss_total_v6": 0.8102110847830772, "step": 1943 }, { "epoch": 0.44834594594594596, "grad_norm": 24.22496223449707, "learning_rate": 8.816379205522428e-06, "loss": 12.298449516296387, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2611457630991936, "loss_total_v6": 0.7686530873179436, "step": 1944 }, { "epoch": 0.44857657657657657, "grad_norm": 26.07494354248047, "learning_rate": 8.815077807938855e-06, "loss": 12.82107162475586, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.2999894432723522, "loss_total_v6": 0.8013169541954994, "step": 1945 }, { "epoch": 0.4488072072072072, "grad_norm": 26.717493057250977, "learning_rate": 8.813775791464611e-06, "loss": 12.693987846374512, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.2912227511405945, "loss_total_v6": 0.7933742478489876, "step": 1946 }, { "epoch": 0.44903783783783785, "grad_norm": 23.03387451171875, "learning_rate": 8.812473156310918e-06, "loss": 13.053108215332031, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.31114999763667583, "loss_total_v6": 0.8158192113041878, "step": 1947 }, { "epoch": 0.44926846846846846, "grad_norm": 25.04765510559082, "learning_rate": 8.81116990268909e-06, "loss": 13.25977897644043, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.31088342517614365, "loss_total_v6": 0.8287362158298492, "step": 1948 }, { "epoch": 0.4494990990990991, "grad_norm": 22.730819702148438, "learning_rate": 8.809866030810544e-06, "loss": 12.29753303527832, "loss_alignment": 0.98291015625, "loss_alignment_w": 0.491455078125, "loss_sft": 0.2772475238889456, "loss_total_v6": 0.7685958072543144, "step": 1949 }, { "epoch": 0.44972972972972974, "grad_norm": 24.009389877319336, "learning_rate": 8.808561540886796e-06, "loss": 13.455302238464355, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.3278796523809433, "loss_total_v6": 0.840956412255764, "step": 1950 }, { "epoch": 0.44996036036036036, "grad_norm": 23.63783836364746, "learning_rate": 8.807256433129465e-06, "loss": 12.736822128295898, "loss_alignment": 1.04296875, "loss_alignment_w": 0.521484375, "loss_sft": 0.27639808133244514, "loss_total_v6": 0.7960513979196548, "step": 1951 }, { "epoch": 0.45019099099099097, "grad_norm": 22.752262115478516, "learning_rate": 8.805950707750268e-06, "loss": 12.61110782623291, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.282823134213686, "loss_total_v6": 0.7881942316889763, "step": 1952 }, { "epoch": 0.45042162162162164, "grad_norm": 26.961734771728516, "learning_rate": 8.80464436496102e-06, "loss": 12.858516693115234, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.30367251485586166, "loss_total_v6": 0.803657278418541, "step": 1953 }, { "epoch": 0.45065225225225225, "grad_norm": 25.02182388305664, "learning_rate": 8.803337404973644e-06, "loss": 12.807804107666016, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.2959406338632107, "loss_total_v6": 0.8004877492785454, "step": 1954 }, { "epoch": 0.45088288288288286, "grad_norm": 22.806856155395508, "learning_rate": 8.802029828000157e-06, "loss": 12.50324821472168, "loss_alignment": 1.01318359375, "loss_alignment_w": 0.506591796875, "loss_sft": 0.27464757673442364, "loss_total_v6": 0.781453013420105, "step": 1955 }, { "epoch": 0.45111351351351353, "grad_norm": 28.95684814453125, "learning_rate": 8.800721634252671e-06, "loss": 12.615621566772461, "loss_alignment": 0.98583984375, "loss_alignment_w": 0.492919921875, "loss_sft": 0.2962735965847969, "loss_total_v6": 0.7884763553738594, "step": 1956 }, { "epoch": 0.45134414414414414, "grad_norm": 23.488788604736328, "learning_rate": 8.799412823943412e-06, "loss": 12.827332496643066, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2952080052345991, "loss_total_v6": 0.8017082661390305, "step": 1957 }, { "epoch": 0.45157477477477476, "grad_norm": 24.709491729736328, "learning_rate": 8.798103397284695e-06, "loss": 12.849170684814453, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2934906668961048, "loss_total_v6": 0.8030731901526451, "step": 1958 }, { "epoch": 0.4518054054054054, "grad_norm": 23.74071502685547, "learning_rate": 8.796793354488938e-06, "loss": 12.611099243164062, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2793740890920162, "loss_total_v6": 0.7881937026977539, "step": 1959 }, { "epoch": 0.45203603603603604, "grad_norm": 24.82657241821289, "learning_rate": 8.795482695768658e-06, "loss": 12.328049659729004, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.26685623452067375, "loss_total_v6": 0.770503081381321, "step": 1960 }, { "epoch": 0.45226666666666665, "grad_norm": 26.480297088623047, "learning_rate": 8.794171421336475e-06, "loss": 13.182058334350586, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.3165086582303047, "loss_total_v6": 0.8238786458969116, "step": 1961 }, { "epoch": 0.4524972972972973, "grad_norm": 24.8834228515625, "learning_rate": 8.792859531405106e-06, "loss": 12.911723136901855, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.29871243983507156, "loss_total_v6": 0.8069827109575272, "step": 1962 }, { "epoch": 0.45272792792792793, "grad_norm": 22.931791305541992, "learning_rate": 8.791547026187373e-06, "loss": 12.95737075805664, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.30257250368595123, "loss_total_v6": 0.8098356947302818, "step": 1963 }, { "epoch": 0.45295855855855854, "grad_norm": 23.77233123779297, "learning_rate": 8.790233905896187e-06, "loss": 12.990652084350586, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.30944383703172207, "loss_total_v6": 0.8119157552719116, "step": 1964 }, { "epoch": 0.4531891891891892, "grad_norm": 23.86072540283203, "learning_rate": 8.788920170744567e-06, "loss": 12.439432144165039, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.266157740727067, "loss_total_v6": 0.7774644792079926, "step": 1965 }, { "epoch": 0.4534198198198198, "grad_norm": 24.293010711669922, "learning_rate": 8.787605820945633e-06, "loss": 12.679208755493164, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2886511143296957, "loss_total_v6": 0.7924505472183228, "step": 1966 }, { "epoch": 0.45365045045045044, "grad_norm": 26.345338821411133, "learning_rate": 8.786290856712598e-06, "loss": 12.84569263458252, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.29800349846482277, "loss_total_v6": 0.8028557896614075, "step": 1967 }, { "epoch": 0.4538810810810811, "grad_norm": 23.616840362548828, "learning_rate": 8.784975278258783e-06, "loss": 12.092303276062012, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.25885121896862984, "loss_total_v6": 0.7557689547538757, "step": 1968 }, { "epoch": 0.4541117117117117, "grad_norm": 28.71210479736328, "learning_rate": 8.783659085797602e-06, "loss": 12.474447250366211, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.26276147924363613, "loss_total_v6": 0.7796529307961464, "step": 1969 }, { "epoch": 0.45434234234234233, "grad_norm": 23.403573989868164, "learning_rate": 8.782342279542569e-06, "loss": 12.986390113830566, "loss_alignment": 1.02294921875, "loss_alignment_w": 0.511474609375, "loss_sft": 0.3002358339726925, "loss_total_v6": 0.8116493597626686, "step": 1970 }, { "epoch": 0.454572972972973, "grad_norm": 23.82845687866211, "learning_rate": 8.781024859707301e-06, "loss": 12.83920955657959, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2886413652449846, "loss_total_v6": 0.8024505823850632, "step": 1971 }, { "epoch": 0.4548036036036036, "grad_norm": 24.172325134277344, "learning_rate": 8.779706826505513e-06, "loss": 13.194490432739258, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.310648150742054, "loss_total_v6": 0.8246557042002678, "step": 1972 }, { "epoch": 0.4550342342342342, "grad_norm": 22.44249153137207, "learning_rate": 8.778388180151019e-06, "loss": 12.51583480834961, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.2815835475921631, "loss_total_v6": 0.7822397127747536, "step": 1973 }, { "epoch": 0.45526486486486484, "grad_norm": 24.638893127441406, "learning_rate": 8.777068920857734e-06, "loss": 12.546916961669922, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.28195448964834213, "loss_total_v6": 0.7841822803020477, "step": 1974 }, { "epoch": 0.4554954954954955, "grad_norm": 26.124387741088867, "learning_rate": 8.77574904883967e-06, "loss": 13.249028205871582, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.3209994751960039, "loss_total_v6": 0.8280643001198769, "step": 1975 }, { "epoch": 0.4557261261261261, "grad_norm": 23.153932571411133, "learning_rate": 8.774428564310939e-06, "loss": 12.985321998596191, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.30698976293206215, "loss_total_v6": 0.811582624912262, "step": 1976 }, { "epoch": 0.45595675675675673, "grad_norm": 23.067201614379883, "learning_rate": 8.773107467485752e-06, "loss": 12.672994613647461, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.28562298230826855, "loss_total_v6": 0.7920621857047081, "step": 1977 }, { "epoch": 0.4561873873873874, "grad_norm": 22.017545700073242, "learning_rate": 8.771785758578427e-06, "loss": 12.526851654052734, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.274139229208231, "loss_total_v6": 0.7829282730817795, "step": 1978 }, { "epoch": 0.456418018018018, "grad_norm": 27.02994155883789, "learning_rate": 8.770463437803367e-06, "loss": 13.224542617797852, "loss_alignment": 1.02783203125, "loss_alignment_w": 0.513916015625, "loss_sft": 0.3138386234641075, "loss_total_v6": 0.8265339508652687, "step": 1979 }, { "epoch": 0.4566486486486486, "grad_norm": 24.05189323425293, "learning_rate": 8.769140505375084e-06, "loss": 12.627525329589844, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2775168716907501, "loss_total_v6": 0.7892203778028488, "step": 1980 }, { "epoch": 0.4568792792792793, "grad_norm": 23.06142234802246, "learning_rate": 8.767816961508191e-06, "loss": 12.550983428955078, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.27061202749609947, "loss_total_v6": 0.7844364792108536, "step": 1981 }, { "epoch": 0.4571099099099099, "grad_norm": 24.420982360839844, "learning_rate": 8.76649280641739e-06, "loss": 12.338653564453125, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2604541480541229, "loss_total_v6": 0.7711658254265785, "step": 1982 }, { "epoch": 0.4573405405405405, "grad_norm": 25.01079559326172, "learning_rate": 8.765168040317494e-06, "loss": 12.60763931274414, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.2913496848195791, "loss_total_v6": 0.78797747194767, "step": 1983 }, { "epoch": 0.4575711711711712, "grad_norm": 21.726810455322266, "learning_rate": 8.763842663423408e-06, "loss": 12.468658447265625, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.2764529939740896, "loss_total_v6": 0.7792911306023598, "step": 1984 }, { "epoch": 0.4578018018018018, "grad_norm": 23.72804832458496, "learning_rate": 8.762516675950134e-06, "loss": 13.074047088623047, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.3058364558964968, "loss_total_v6": 0.8171279728412628, "step": 1985 }, { "epoch": 0.4580324324324324, "grad_norm": 24.954097747802734, "learning_rate": 8.761190078112783e-06, "loss": 12.219305038452148, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.2674144636839628, "loss_total_v6": 0.7637065798044205, "step": 1986 }, { "epoch": 0.4582630630630631, "grad_norm": 29.044692993164062, "learning_rate": 8.759862870126552e-06, "loss": 13.151185989379883, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.319492407143116, "loss_total_v6": 0.8219490945339203, "step": 1987 }, { "epoch": 0.4584936936936937, "grad_norm": 29.2136173248291, "learning_rate": 8.758535052206749e-06, "loss": 12.953434944152832, "loss_alignment": 1.01220703125, "loss_alignment_w": 0.506103515625, "loss_sft": 0.3013652265071869, "loss_total_v6": 0.8095897138118744, "step": 1988 }, { "epoch": 0.4587243243243243, "grad_norm": 26.143449783325195, "learning_rate": 8.757206624568774e-06, "loss": 12.868413925170898, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.30412331223487854, "loss_total_v6": 0.8042758703231812, "step": 1989 }, { "epoch": 0.458954954954955, "grad_norm": 22.515914916992188, "learning_rate": 8.755877587428127e-06, "loss": 12.578622817993164, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.28236449882388115, "loss_total_v6": 0.7861639112234116, "step": 1990 }, { "epoch": 0.4591855855855856, "grad_norm": 26.006305694580078, "learning_rate": 8.754547941000412e-06, "loss": 12.381248474121094, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.27088306099176407, "loss_total_v6": 0.7738279849290848, "step": 1991 }, { "epoch": 0.4594162162162162, "grad_norm": 26.803592681884766, "learning_rate": 8.753217685501317e-06, "loss": 12.748730659484863, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2825439777225256, "loss_total_v6": 0.7967956811189651, "step": 1992 }, { "epoch": 0.45964684684684687, "grad_norm": 29.00254249572754, "learning_rate": 8.75188682114665e-06, "loss": 12.689112663269043, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2871338352560997, "loss_total_v6": 0.7930695414543152, "step": 1993 }, { "epoch": 0.4598774774774775, "grad_norm": 25.6502742767334, "learning_rate": 8.750555348152299e-06, "loss": 13.634099960327148, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.3429454639554024, "loss_total_v6": 0.8521312549710274, "step": 1994 }, { "epoch": 0.4601081081081081, "grad_norm": 24.329063415527344, "learning_rate": 8.749223266734262e-06, "loss": 12.911590576171875, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.30119131319224834, "loss_total_v6": 0.806974396109581, "step": 1995 }, { "epoch": 0.46033873873873876, "grad_norm": 22.85418701171875, "learning_rate": 8.747890577108633e-06, "loss": 13.041754722595215, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.31124918535351753, "loss_total_v6": 0.8151096627116203, "step": 1996 }, { "epoch": 0.4605693693693694, "grad_norm": 24.008424758911133, "learning_rate": 8.746557279491603e-06, "loss": 12.802875518798828, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.29743317514657974, "loss_total_v6": 0.8001797497272491, "step": 1997 }, { "epoch": 0.4608, "grad_norm": 23.412113189697266, "learning_rate": 8.745223374099464e-06, "loss": 12.602890014648438, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.280432740226388, "loss_total_v6": 0.7876806482672691, "step": 1998 }, { "epoch": 0.46103063063063066, "grad_norm": 24.29045867919922, "learning_rate": 8.743888861148602e-06, "loss": 12.891863822937012, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.3014232777059078, "loss_total_v6": 0.8057415187358856, "step": 1999 }, { "epoch": 0.46126126126126127, "grad_norm": 22.814817428588867, "learning_rate": 8.742553740855507e-06, "loss": 12.709677696228027, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2825292870402336, "loss_total_v6": 0.7943548485636711, "step": 2000 }, { "epoch": 0.46126126126126127, "eval_loss": 0.9547907710075378, "eval_loss_alignment": 1.0103988299086757, "eval_loss_alignment_w": 0.5051994149543378, "eval_loss_contrast": 0.0, "eval_loss_highlevel": 0.0, "eval_loss_sft": 0.4495913638099688, "eval_loss_total_v6": 0.9547907791045158, "eval_loss_wm_recon": 0.0, "eval_loss_wm_recon_w": 0.0, "eval_runtime": 273.1222, "eval_samples_per_second": 6.389, "eval_steps_per_second": 0.802, "step": 2000 }, { "epoch": 0.4614918918918919, "grad_norm": 29.21103858947754, "learning_rate": 8.741218013436764e-06, "loss": 13.126705169677734, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.3186947964131832, "loss_total_v6": 0.8204190507531166, "step": 2001 }, { "epoch": 0.46172252252252255, "grad_norm": 21.05173110961914, "learning_rate": 8.739881679109059e-06, "loss": 12.415545463562012, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.2654277887195349, "loss_total_v6": 0.7759716212749481, "step": 2002 }, { "epoch": 0.46195315315315316, "grad_norm": 23.760663986206055, "learning_rate": 8.738544738089177e-06, "loss": 12.840696334838867, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.2890547215938568, "loss_total_v6": 0.802543506026268, "step": 2003 }, { "epoch": 0.4621837837837838, "grad_norm": 24.10474967956543, "learning_rate": 8.737207190593995e-06, "loss": 12.563920974731445, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.28260526433587074, "loss_total_v6": 0.7852450236678123, "step": 2004 }, { "epoch": 0.4624144144144144, "grad_norm": 25.40556526184082, "learning_rate": 8.735869036840496e-06, "loss": 13.158617973327637, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.3118392676115036, "loss_total_v6": 0.8224136084318161, "step": 2005 }, { "epoch": 0.46264504504504506, "grad_norm": 21.81351661682129, "learning_rate": 8.734530277045759e-06, "loss": 12.986310005187988, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.307646531611681, "loss_total_v6": 0.8116443455219269, "step": 2006 }, { "epoch": 0.46287567567567567, "grad_norm": 23.745405197143555, "learning_rate": 8.733190911426957e-06, "loss": 12.151180267333984, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.25409292057156563, "loss_total_v6": 0.7594487518072128, "step": 2007 }, { "epoch": 0.4631063063063063, "grad_norm": 24.72746467590332, "learning_rate": 8.731850940201369e-06, "loss": 12.550132751464844, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2752738557755947, "loss_total_v6": 0.7843833342194557, "step": 2008 }, { "epoch": 0.46333693693693695, "grad_norm": 25.087366104125977, "learning_rate": 8.730510363586368e-06, "loss": 12.595877647399902, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.28416003845632076, "loss_total_v6": 0.7872423380613327, "step": 2009 }, { "epoch": 0.46356756756756756, "grad_norm": 23.818675994873047, "learning_rate": 8.729169181799423e-06, "loss": 12.148138046264648, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.25873987935483456, "loss_total_v6": 0.7592586651444435, "step": 2010 }, { "epoch": 0.4637981981981982, "grad_norm": 22.8364315032959, "learning_rate": 8.727827395058103e-06, "loss": 12.194857597351074, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.2639028411358595, "loss_total_v6": 0.7621785625815392, "step": 2011 }, { "epoch": 0.46402882882882884, "grad_norm": 23.763425827026367, "learning_rate": 8.72648500358008e-06, "loss": 12.5006103515625, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.28444679640233517, "loss_total_v6": 0.7812882140278816, "step": 2012 }, { "epoch": 0.46425945945945946, "grad_norm": 20.82619285583496, "learning_rate": 8.725142007583115e-06, "loss": 12.159200668334961, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2552961353212595, "loss_total_v6": 0.7599500566720963, "step": 2013 }, { "epoch": 0.46449009009009007, "grad_norm": 23.4376277923584, "learning_rate": 8.723798407285078e-06, "loss": 12.711176872253418, "loss_alignment": 1.0380859375, "loss_alignment_w": 0.51904296875, "loss_sft": 0.2746579069644213, "loss_total_v6": 0.7944485396146774, "step": 2014 }, { "epoch": 0.46472072072072074, "grad_norm": 23.9061279296875, "learning_rate": 8.722454202903923e-06, "loss": 12.820090293884277, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.2984633482992649, "loss_total_v6": 0.8012556880712509, "step": 2015 }, { "epoch": 0.46495135135135135, "grad_norm": 23.585926055908203, "learning_rate": 8.721109394657717e-06, "loss": 12.616327285766602, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2812572792172432, "loss_total_v6": 0.788520447909832, "step": 2016 }, { "epoch": 0.46518198198198196, "grad_norm": 26.497175216674805, "learning_rate": 8.719763982764613e-06, "loss": 12.508371353149414, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.28119340538978577, "loss_total_v6": 0.7817732393741608, "step": 2017 }, { "epoch": 0.46541261261261263, "grad_norm": 28.435447692871094, "learning_rate": 8.718417967442868e-06, "loss": 12.582606315612793, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.2895867060869932, "loss_total_v6": 0.7864128649234772, "step": 2018 }, { "epoch": 0.46564324324324324, "grad_norm": 24.622833251953125, "learning_rate": 8.717071348910837e-06, "loss": 12.953336715698242, "loss_alignment": 1.01318359375, "loss_alignment_w": 0.506591796875, "loss_sft": 0.3025797437876463, "loss_total_v6": 0.8095835298299789, "step": 2019 }, { "epoch": 0.46587387387387386, "grad_norm": 23.319612503051758, "learning_rate": 8.715724127386971e-06, "loss": 12.613643646240234, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.276160990819335, "loss_total_v6": 0.7883527502417564, "step": 2020 }, { "epoch": 0.4661045045045045, "grad_norm": 25.27623176574707, "learning_rate": 8.71437630308982e-06, "loss": 12.245157241821289, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.26564279571175575, "loss_total_v6": 0.7653223723173141, "step": 2021 }, { "epoch": 0.46633513513513514, "grad_norm": 26.373905181884766, "learning_rate": 8.71302787623803e-06, "loss": 13.441608428955078, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.33021286129951477, "loss_total_v6": 0.8401005864143372, "step": 2022 }, { "epoch": 0.46656576576576575, "grad_norm": 25.815176010131836, "learning_rate": 8.711678847050347e-06, "loss": 13.135499954223633, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.3139191847294569, "loss_total_v6": 0.820968747138977, "step": 2023 }, { "epoch": 0.4667963963963964, "grad_norm": 25.97149658203125, "learning_rate": 8.710329215745612e-06, "loss": 12.821609497070312, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.30200669541954994, "loss_total_v6": 0.8013505563139915, "step": 2024 }, { "epoch": 0.46702702702702703, "grad_norm": 24.200292587280273, "learning_rate": 8.708978982542766e-06, "loss": 12.523053169250488, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2752140033990145, "loss_total_v6": 0.7826908156275749, "step": 2025 }, { "epoch": 0.46725765765765764, "grad_norm": 23.381847381591797, "learning_rate": 8.707628147660847e-06, "loss": 12.83433723449707, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.29939952678978443, "loss_total_v6": 0.8021460995078087, "step": 2026 }, { "epoch": 0.4674882882882883, "grad_norm": 25.724613189697266, "learning_rate": 8.70627671131899e-06, "loss": 12.565966606140137, "loss_alignment": 1.00732421875, "loss_alignment_w": 0.503662109375, "loss_sft": 0.2807495202869177, "loss_total_v6": 0.7853729203343391, "step": 2027 }, { "epoch": 0.4677189189189189, "grad_norm": 25.80101203918457, "learning_rate": 8.70492467373643e-06, "loss": 12.699689865112305, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.2933796364814043, "loss_total_v6": 0.7937305867671967, "step": 2028 }, { "epoch": 0.46794954954954954, "grad_norm": 27.42498779296875, "learning_rate": 8.703572035132497e-06, "loss": 12.919857025146484, "loss_alignment": 1.0263671875, "loss_alignment_w": 0.51318359375, "loss_sft": 0.29331568256020546, "loss_total_v6": 0.8074910640716553, "step": 2029 }, { "epoch": 0.4681801801801802, "grad_norm": 21.428119659423828, "learning_rate": 8.702218795726619e-06, "loss": 12.197562217712402, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.26344627141952515, "loss_total_v6": 0.7623476535081863, "step": 2030 }, { "epoch": 0.4684108108108108, "grad_norm": 25.839521408081055, "learning_rate": 8.700864955738318e-06, "loss": 12.414196014404297, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.2724845688790083, "loss_total_v6": 0.7758872583508492, "step": 2031 }, { "epoch": 0.46864144144144143, "grad_norm": 23.077579498291016, "learning_rate": 8.699510515387222e-06, "loss": 12.684438705444336, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.28525481931865215, "loss_total_v6": 0.792777419090271, "step": 2032 }, { "epoch": 0.4688720720720721, "grad_norm": 23.595733642578125, "learning_rate": 8.69815547489305e-06, "loss": 12.7847900390625, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.29621127620339394, "loss_total_v6": 0.799049399793148, "step": 2033 }, { "epoch": 0.4691027027027027, "grad_norm": 26.999000549316406, "learning_rate": 8.696799834475618e-06, "loss": 13.067485809326172, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.31421541050076485, "loss_total_v6": 0.8167178556323051, "step": 2034 }, { "epoch": 0.4693333333333333, "grad_norm": 22.851089477539062, "learning_rate": 8.695443594354843e-06, "loss": 12.861278533935547, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2972228843718767, "loss_total_v6": 0.8038299530744553, "step": 2035 }, { "epoch": 0.46956396396396394, "grad_norm": 22.819377899169922, "learning_rate": 8.694086754750738e-06, "loss": 12.56634521484375, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.27843857556581497, "loss_total_v6": 0.7853965759277344, "step": 2036 }, { "epoch": 0.4697945945945946, "grad_norm": 23.032058715820312, "learning_rate": 8.692729315883409e-06, "loss": 12.784433364868164, "loss_alignment": 1.01513671875, "loss_alignment_w": 0.507568359375, "loss_sft": 0.29054320976138115, "loss_total_v6": 0.7990270853042603, "step": 2037 }, { "epoch": 0.4700252252252252, "grad_norm": 22.328439712524414, "learning_rate": 8.691371277973066e-06, "loss": 12.357097625732422, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.27062484808266163, "loss_total_v6": 0.7723185792565346, "step": 2038 }, { "epoch": 0.47025585585585583, "grad_norm": 26.803224563598633, "learning_rate": 8.690012641240014e-06, "loss": 12.836010932922363, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.28691561706364155, "loss_total_v6": 0.8022506907582283, "step": 2039 }, { "epoch": 0.4704864864864865, "grad_norm": 22.202472686767578, "learning_rate": 8.688653405904653e-06, "loss": 12.371309280395508, "loss_alignment": 0.9736328125, "loss_alignment_w": 0.48681640625, "loss_sft": 0.28643619641661644, "loss_total_v6": 0.7732068449258804, "step": 2040 }, { "epoch": 0.4707171171171171, "grad_norm": 22.443603515625, "learning_rate": 8.687293572187478e-06, "loss": 12.556344985961914, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.2875181809067726, "loss_total_v6": 0.7847715839743614, "step": 2041 }, { "epoch": 0.4709477477477477, "grad_norm": 22.99461555480957, "learning_rate": 8.685933140309088e-06, "loss": 13.120118141174316, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.31411751359701157, "loss_total_v6": 0.820007398724556, "step": 2042 }, { "epoch": 0.4711783783783784, "grad_norm": 25.028623580932617, "learning_rate": 8.684572110490174e-06, "loss": 12.664749145507812, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2782716304063797, "loss_total_v6": 0.7915467917919159, "step": 2043 }, { "epoch": 0.471409009009009, "grad_norm": 26.448692321777344, "learning_rate": 8.683210482951528e-06, "loss": 13.55910873413086, "loss_alignment": 1.00732421875, "loss_alignment_w": 0.503662109375, "loss_sft": 0.3425767794251442, "loss_total_v6": 0.8474443480372429, "step": 2044 }, { "epoch": 0.4716396396396396, "grad_norm": 23.411651611328125, "learning_rate": 8.681848257914031e-06, "loss": 12.31869888305664, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.26921677961945534, "loss_total_v6": 0.7699186727404594, "step": 2045 }, { "epoch": 0.4718702702702703, "grad_norm": 24.437904357910156, "learning_rate": 8.680485435598674e-06, "loss": 12.993217468261719, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.30752899684011936, "loss_total_v6": 0.8120760694146156, "step": 2046 }, { "epoch": 0.4721009009009009, "grad_norm": 25.62885284423828, "learning_rate": 8.679122016226529e-06, "loss": 12.474369049072266, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.27092004008591175, "loss_total_v6": 0.7796480506658554, "step": 2047 }, { "epoch": 0.4723315315315315, "grad_norm": 23.019664764404297, "learning_rate": 8.677758000018777e-06, "loss": 12.516317367553711, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.2794012166559696, "loss_total_v6": 0.7822698801755905, "step": 2048 }, { "epoch": 0.4725621621621622, "grad_norm": 23.32268714904785, "learning_rate": 8.676393387196692e-06, "loss": 12.625051498413086, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.28842486441135406, "loss_total_v6": 0.7890657037496567, "step": 2049 }, { "epoch": 0.4727927927927928, "grad_norm": 24.8573055267334, "learning_rate": 8.675028177981643e-06, "loss": 13.068016052246094, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.31202080473303795, "loss_total_v6": 0.8167509958148003, "step": 2050 }, { "epoch": 0.4730234234234234, "grad_norm": 25.13443946838379, "learning_rate": 8.6736623725951e-06, "loss": 13.17574691772461, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.31208582036197186, "loss_total_v6": 0.8234841600060463, "step": 2051 }, { "epoch": 0.4732540540540541, "grad_norm": 22.8553409576416, "learning_rate": 8.672295971258624e-06, "loss": 12.844535827636719, "loss_alignment": 1.02294921875, "loss_alignment_w": 0.511474609375, "loss_sft": 0.29156829603016376, "loss_total_v6": 0.8027835041284561, "step": 2052 }, { "epoch": 0.4734846846846847, "grad_norm": 25.64182472229004, "learning_rate": 8.670928974193879e-06, "loss": 12.493919372558594, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.2830825597047806, "loss_total_v6": 0.7808700203895569, "step": 2053 }, { "epoch": 0.4737153153153153, "grad_norm": 23.810405731201172, "learning_rate": 8.66956138162262e-06, "loss": 13.56873607635498, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.3412558436393738, "loss_total_v6": 0.8480460122227669, "step": 2054 }, { "epoch": 0.47394594594594597, "grad_norm": 21.81785774230957, "learning_rate": 8.668193193766703e-06, "loss": 12.563077926635742, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.28278143890202045, "loss_total_v6": 0.7851923331618309, "step": 2055 }, { "epoch": 0.4741765765765766, "grad_norm": 29.162662506103516, "learning_rate": 8.666824410848075e-06, "loss": 12.58963394165039, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.28633330576121807, "loss_total_v6": 0.7868520990014076, "step": 2056 }, { "epoch": 0.4744072072072072, "grad_norm": 22.23171043395996, "learning_rate": 8.665455033088787e-06, "loss": 13.557640075683594, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.3487563505768776, "loss_total_v6": 0.8473525419831276, "step": 2057 }, { "epoch": 0.47463783783783786, "grad_norm": 25.048612594604492, "learning_rate": 8.66408506071098e-06, "loss": 12.446233749389648, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.2769741117954254, "loss_total_v6": 0.7778896391391754, "step": 2058 }, { "epoch": 0.4748684684684685, "grad_norm": 25.063154220581055, "learning_rate": 8.662714493936895e-06, "loss": 12.302321434020996, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.26500413194298744, "loss_total_v6": 0.768895111978054, "step": 2059 }, { "epoch": 0.4750990990990991, "grad_norm": 26.81671142578125, "learning_rate": 8.661343332988869e-06, "loss": 13.10356616973877, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.31134346313774586, "loss_total_v6": 0.8189728632569313, "step": 2060 }, { "epoch": 0.47532972972972976, "grad_norm": 24.10758399963379, "learning_rate": 8.659971578089333e-06, "loss": 12.656228065490723, "loss_alignment": 1.01708984375, "loss_alignment_w": 0.508544921875, "loss_sft": 0.28146228566765785, "loss_total_v6": 0.7910142838954926, "step": 2061 }, { "epoch": 0.47556036036036037, "grad_norm": 25.031261444091797, "learning_rate": 8.658599229460818e-06, "loss": 12.387310028076172, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.2701786160469055, "loss_total_v6": 0.7742068916559219, "step": 2062 }, { "epoch": 0.475790990990991, "grad_norm": 22.385526657104492, "learning_rate": 8.657226287325951e-06, "loss": 12.697381973266602, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.2948375232517719, "loss_total_v6": 0.7935863360762596, "step": 2063 }, { "epoch": 0.47602162162162165, "grad_norm": 25.962039947509766, "learning_rate": 8.65585275190745e-06, "loss": 13.333497047424316, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.32092295214533806, "loss_total_v6": 0.8333435654640198, "step": 2064 }, { "epoch": 0.47625225225225226, "grad_norm": 24.53900909423828, "learning_rate": 8.654478623428136e-06, "loss": 12.79334831237793, "loss_alignment": 1.01318359375, "loss_alignment_w": 0.506591796875, "loss_sft": 0.29215325973927975, "loss_total_v6": 0.7995842769742012, "step": 2065 }, { "epoch": 0.4764828828828829, "grad_norm": 24.93050765991211, "learning_rate": 8.653103902110923e-06, "loss": 12.379027366638184, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.26200096867978573, "loss_total_v6": 0.7736892178654671, "step": 2066 }, { "epoch": 0.4767135135135135, "grad_norm": 26.150238037109375, "learning_rate": 8.65172858817882e-06, "loss": 12.360219955444336, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.27530610747635365, "loss_total_v6": 0.7725137621164322, "step": 2067 }, { "epoch": 0.47694414414414416, "grad_norm": 25.28207015991211, "learning_rate": 8.650352681854934e-06, "loss": 12.622923851013184, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2817458603531122, "loss_total_v6": 0.7889327332377434, "step": 2068 }, { "epoch": 0.47717477477477477, "grad_norm": 24.89661979675293, "learning_rate": 8.648976183362468e-06, "loss": 12.800675392150879, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.29412182979285717, "loss_total_v6": 0.8000422492623329, "step": 2069 }, { "epoch": 0.4774054054054054, "grad_norm": 24.384857177734375, "learning_rate": 8.64759909292472e-06, "loss": 12.746822357177734, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.2862698696553707, "loss_total_v6": 0.7966763749718666, "step": 2070 }, { "epoch": 0.47763603603603605, "grad_norm": 26.277023315429688, "learning_rate": 8.646221410765088e-06, "loss": 12.536359786987305, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2722919788211584, "loss_total_v6": 0.7835224643349648, "step": 2071 }, { "epoch": 0.47786666666666666, "grad_norm": 25.71500015258789, "learning_rate": 8.644843137107058e-06, "loss": 12.374773979187012, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2716075275093317, "loss_total_v6": 0.7734233662486076, "step": 2072 }, { "epoch": 0.4780972972972973, "grad_norm": 23.114192962646484, "learning_rate": 8.64346427217422e-06, "loss": 13.12549877166748, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.30899110808968544, "loss_total_v6": 0.8203436508774757, "step": 2073 }, { "epoch": 0.47832792792792794, "grad_norm": 26.17015838623047, "learning_rate": 8.642084816190254e-06, "loss": 12.280778884887695, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2646342683583498, "loss_total_v6": 0.7675486952066422, "step": 2074 }, { "epoch": 0.47855855855855856, "grad_norm": 24.215341567993164, "learning_rate": 8.640704769378943e-06, "loss": 12.451886177062988, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.27548104524612427, "loss_total_v6": 0.7782428711652756, "step": 2075 }, { "epoch": 0.47878918918918917, "grad_norm": 23.966907501220703, "learning_rate": 8.639324131964156e-06, "loss": 12.771890640258789, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.30034887231886387, "loss_total_v6": 0.7982431799173355, "step": 2076 }, { "epoch": 0.47901981981981984, "grad_norm": 26.103866577148438, "learning_rate": 8.637942904169867e-06, "loss": 12.518712043762207, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.2828467544168234, "loss_total_v6": 0.7824194952845573, "step": 2077 }, { "epoch": 0.47925045045045045, "grad_norm": 24.370044708251953, "learning_rate": 8.63656108622014e-06, "loss": 12.331171035766602, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.2731549069285393, "loss_total_v6": 0.7706982493400574, "step": 2078 }, { "epoch": 0.47948108108108106, "grad_norm": 24.935117721557617, "learning_rate": 8.635178678339137e-06, "loss": 12.444055557250977, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2736488599330187, "loss_total_v6": 0.7777534648776054, "step": 2079 }, { "epoch": 0.47971171171171173, "grad_norm": 23.883968353271484, "learning_rate": 8.633795680751116e-06, "loss": 12.863277435302734, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.29971288330852985, "loss_total_v6": 0.8039548173546791, "step": 2080 }, { "epoch": 0.47994234234234234, "grad_norm": 23.33243179321289, "learning_rate": 8.63241209368043e-06, "loss": 12.854652404785156, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.29259724728763103, "loss_total_v6": 0.8034157380461693, "step": 2081 }, { "epoch": 0.48017297297297296, "grad_norm": 24.150442123413086, "learning_rate": 8.631027917351528e-06, "loss": 12.881828308105469, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.3063655123114586, "loss_total_v6": 0.8051142692565918, "step": 2082 }, { "epoch": 0.4804036036036036, "grad_norm": 23.28131103515625, "learning_rate": 8.629643151988951e-06, "loss": 12.836199760437012, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.3026897609233856, "loss_total_v6": 0.8022624775767326, "step": 2083 }, { "epoch": 0.48063423423423424, "grad_norm": 23.277706146240234, "learning_rate": 8.628257797817344e-06, "loss": 13.314546585083008, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.3257809355854988, "loss_total_v6": 0.8321591317653656, "step": 2084 }, { "epoch": 0.48086486486486485, "grad_norm": 29.074071884155273, "learning_rate": 8.626871855061438e-06, "loss": 12.95718765258789, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.30883246660232544, "loss_total_v6": 0.8098242729902267, "step": 2085 }, { "epoch": 0.4810954954954955, "grad_norm": 21.612768173217773, "learning_rate": 8.625485323946068e-06, "loss": 12.930524826049805, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.3061588928103447, "loss_total_v6": 0.8081578239798546, "step": 2086 }, { "epoch": 0.48132612612612613, "grad_norm": 24.06807518005371, "learning_rate": 8.624098204696156e-06, "loss": 12.238964080810547, "loss_alignment": 0.990234375, "loss_alignment_w": 0.4951171875, "loss_sft": 0.2692840825766325, "loss_total_v6": 0.7649352848529816, "step": 2087 }, { "epoch": 0.48155675675675674, "grad_norm": 22.0113525390625, "learning_rate": 8.622710497536727e-06, "loss": 12.674590110778809, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2859973441809416, "loss_total_v6": 0.7921618819236755, "step": 2088 }, { "epoch": 0.4817873873873874, "grad_norm": 23.29907989501953, "learning_rate": 8.621322202692894e-06, "loss": 12.297046661376953, "loss_alignment": 0.9873046875, "loss_alignment_w": 0.49365234375, "loss_sft": 0.27517250552773476, "loss_total_v6": 0.768565446138382, "step": 2089 }, { "epoch": 0.482018018018018, "grad_norm": 23.456485748291016, "learning_rate": 8.619933320389872e-06, "loss": 12.394508361816406, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.27581645362079144, "loss_total_v6": 0.774656780064106, "step": 2090 }, { "epoch": 0.48224864864864864, "grad_norm": 25.406110763549805, "learning_rate": 8.618543850852969e-06, "loss": 12.480062484741211, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.27876797318458557, "loss_total_v6": 0.7800039350986481, "step": 2091 }, { "epoch": 0.4824792792792793, "grad_norm": 25.179256439208984, "learning_rate": 8.617153794307588e-06, "loss": 13.045106887817383, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.31063467636704445, "loss_total_v6": 0.8153191581368446, "step": 2092 }, { "epoch": 0.4827099099099099, "grad_norm": 25.559213638305664, "learning_rate": 8.615763150979227e-06, "loss": 12.399478912353516, "loss_alignment": 0.9912109375, "loss_alignment_w": 0.49560546875, "loss_sft": 0.2797586750239134, "loss_total_v6": 0.7749674096703529, "step": 2093 }, { "epoch": 0.48294054054054053, "grad_norm": 25.599363327026367, "learning_rate": 8.614371921093479e-06, "loss": 13.429500579833984, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.3321416452527046, "loss_total_v6": 0.839343786239624, "step": 2094 }, { "epoch": 0.4831711711711712, "grad_norm": 24.88795280456543, "learning_rate": 8.612980104876032e-06, "loss": 12.802312850952148, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.30019035190343857, "loss_total_v6": 0.8001445531845093, "step": 2095 }, { "epoch": 0.4834018018018018, "grad_norm": 23.662918090820312, "learning_rate": 8.61158770255267e-06, "loss": 12.74144172668457, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.2961418032646179, "loss_total_v6": 0.7963401451706886, "step": 2096 }, { "epoch": 0.4836324324324324, "grad_norm": 22.19641876220703, "learning_rate": 8.610194714349271e-06, "loss": 12.557443618774414, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.28119335882365704, "loss_total_v6": 0.7848402261734009, "step": 2097 }, { "epoch": 0.48386306306306304, "grad_norm": 24.230554580688477, "learning_rate": 8.608801140491811e-06, "loss": 12.398504257202148, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.264072785153985, "loss_total_v6": 0.7749065309762955, "step": 2098 }, { "epoch": 0.4840936936936937, "grad_norm": 21.971513748168945, "learning_rate": 8.607406981206357e-06, "loss": 12.08409309387207, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.2536383271217346, "loss_total_v6": 0.755255788564682, "step": 2099 }, { "epoch": 0.4843243243243243, "grad_norm": 27.28352165222168, "learning_rate": 8.606012236719073e-06, "loss": 12.889622688293457, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.30041344463825226, "loss_total_v6": 0.8056014403700829, "step": 2100 }, { "epoch": 0.48455495495495493, "grad_norm": 24.951196670532227, "learning_rate": 8.604616907256218e-06, "loss": 12.582977294921875, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.27433586679399014, "loss_total_v6": 0.7864360958337784, "step": 2101 }, { "epoch": 0.4847855855855856, "grad_norm": 24.458271026611328, "learning_rate": 8.603220993044145e-06, "loss": 12.438529968261719, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.27664520405232906, "loss_total_v6": 0.7774081453680992, "step": 2102 }, { "epoch": 0.4850162162162162, "grad_norm": 26.07750129699707, "learning_rate": 8.6018244943093e-06, "loss": 12.483482360839844, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.27646402455866337, "loss_total_v6": 0.7802176773548126, "step": 2103 }, { "epoch": 0.4852468468468468, "grad_norm": 24.017446517944336, "learning_rate": 8.600427411278234e-06, "loss": 12.505821228027344, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.27511352486908436, "loss_total_v6": 0.781613789498806, "step": 2104 }, { "epoch": 0.4854774774774775, "grad_norm": 25.660154342651367, "learning_rate": 8.599029744177577e-06, "loss": 13.038536071777344, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.3076453059911728, "loss_total_v6": 0.8149084746837616, "step": 2105 }, { "epoch": 0.4857081081081081, "grad_norm": 22.05361557006836, "learning_rate": 8.597631493234065e-06, "loss": 12.1312894821167, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.2603570893406868, "loss_total_v6": 0.7582056075334549, "step": 2106 }, { "epoch": 0.4859387387387387, "grad_norm": 23.829999923706055, "learning_rate": 8.596232658674524e-06, "loss": 12.491750717163086, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2740357629954815, "loss_total_v6": 0.7807343751192093, "step": 2107 }, { "epoch": 0.4861693693693694, "grad_norm": 23.799169540405273, "learning_rate": 8.594833240725877e-06, "loss": 12.525150299072266, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.27717618085443974, "loss_total_v6": 0.7828219011425972, "step": 2108 }, { "epoch": 0.4864, "grad_norm": 22.979476928710938, "learning_rate": 8.593433239615141e-06, "loss": 12.417987823486328, "loss_alignment": 0.990234375, "loss_alignment_w": 0.4951171875, "loss_sft": 0.2799083683639765, "loss_total_v6": 0.7761242091655731, "step": 2109 }, { "epoch": 0.4866306306306306, "grad_norm": 23.993972778320312, "learning_rate": 8.59203265556943e-06, "loss": 12.534406661987305, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.27787673100829124, "loss_total_v6": 0.7834003940224648, "step": 2110 }, { "epoch": 0.4868612612612613, "grad_norm": 23.603261947631836, "learning_rate": 8.590631488815945e-06, "loss": 12.435187339782715, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2715076319873333, "loss_total_v6": 0.7771991640329361, "step": 2111 }, { "epoch": 0.4870918918918919, "grad_norm": 29.660152435302734, "learning_rate": 8.58922973958199e-06, "loss": 12.5266695022583, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.2868841215968132, "loss_total_v6": 0.782916821539402, "step": 2112 }, { "epoch": 0.4873225225225225, "grad_norm": 27.318410873413086, "learning_rate": 8.587827408094958e-06, "loss": 12.436385154724121, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.2792729679495096, "loss_total_v6": 0.7772740721702576, "step": 2113 }, { "epoch": 0.4875531531531532, "grad_norm": 27.79838752746582, "learning_rate": 8.58642449458234e-06, "loss": 12.80748462677002, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.30324488691985607, "loss_total_v6": 0.8004677817225456, "step": 2114 }, { "epoch": 0.4877837837837838, "grad_norm": 27.449905395507812, "learning_rate": 8.585020999271718e-06, "loss": 13.134441375732422, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.3188731446862221, "loss_total_v6": 0.8209025710821152, "step": 2115 }, { "epoch": 0.4880144144144144, "grad_norm": 25.499204635620117, "learning_rate": 8.583616922390772e-06, "loss": 12.25340747833252, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.26560914888978004, "loss_total_v6": 0.7658379673957825, "step": 2116 }, { "epoch": 0.48824504504504507, "grad_norm": 25.94791030883789, "learning_rate": 8.582212264167274e-06, "loss": 12.3132905960083, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2634618915617466, "loss_total_v6": 0.7695806324481964, "step": 2117 }, { "epoch": 0.4884756756756757, "grad_norm": 25.42261505126953, "learning_rate": 8.58080702482909e-06, "loss": 13.269627571105957, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.3273528218269348, "loss_total_v6": 0.8293517082929611, "step": 2118 }, { "epoch": 0.4887063063063063, "grad_norm": 25.574373245239258, "learning_rate": 8.579401204604184e-06, "loss": 12.728363990783691, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.2932186685502529, "loss_total_v6": 0.7955227270722389, "step": 2119 }, { "epoch": 0.48893693693693696, "grad_norm": 25.96900177001953, "learning_rate": 8.577994803720605e-06, "loss": 12.415865898132324, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.27551862969994545, "loss_total_v6": 0.7759916186332703, "step": 2120 }, { "epoch": 0.4891675675675676, "grad_norm": 24.54798698425293, "learning_rate": 8.576587822406511e-06, "loss": 12.91342544555664, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.30405256897211075, "loss_total_v6": 0.8070890679955482, "step": 2121 }, { "epoch": 0.4893981981981982, "grad_norm": 23.36564826965332, "learning_rate": 8.57518026089014e-06, "loss": 12.647958755493164, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.2850652877241373, "loss_total_v6": 0.790497399866581, "step": 2122 }, { "epoch": 0.48962882882882885, "grad_norm": 23.2702579498291, "learning_rate": 8.573772119399832e-06, "loss": 12.90478515625, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.30090333707630634, "loss_total_v6": 0.806549072265625, "step": 2123 }, { "epoch": 0.48985945945945947, "grad_norm": 28.6075439453125, "learning_rate": 8.572363398164017e-06, "loss": 12.99770736694336, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.30060745030641556, "loss_total_v6": 0.81235671043396, "step": 2124 }, { "epoch": 0.4900900900900901, "grad_norm": 22.306028366088867, "learning_rate": 8.570954097411224e-06, "loss": 12.342235565185547, "loss_alignment": 0.98583984375, "loss_alignment_w": 0.492919921875, "loss_sft": 0.27715749479830265, "loss_total_v6": 0.7713896781206131, "step": 2125 }, { "epoch": 0.4903207207207207, "grad_norm": 22.941936492919922, "learning_rate": 8.569544217370071e-06, "loss": 13.494582176208496, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.3306398279964924, "loss_total_v6": 0.8434114158153534, "step": 2126 }, { "epoch": 0.49055135135135136, "grad_norm": 24.300811767578125, "learning_rate": 8.568133758269272e-06, "loss": 12.22207260131836, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.27080702409148216, "loss_total_v6": 0.7638795524835587, "step": 2127 }, { "epoch": 0.490781981981982, "grad_norm": 25.49948501586914, "learning_rate": 8.566722720337634e-06, "loss": 12.53105354309082, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2751342114061117, "loss_total_v6": 0.7831908464431763, "step": 2128 }, { "epoch": 0.4910126126126126, "grad_norm": 24.735774993896484, "learning_rate": 8.565311103804063e-06, "loss": 13.502716064453125, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.3313160464167595, "loss_total_v6": 0.8439197838306427, "step": 2129 }, { "epoch": 0.49124324324324325, "grad_norm": 22.621463775634766, "learning_rate": 8.56389890889755e-06, "loss": 12.082479476928711, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.25275934860110283, "loss_total_v6": 0.7551549896597862, "step": 2130 }, { "epoch": 0.49147387387387387, "grad_norm": 30.504552841186523, "learning_rate": 8.562486135847186e-06, "loss": 12.368783950805664, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.26361909694969654, "loss_total_v6": 0.7730489820241928, "step": 2131 }, { "epoch": 0.4917045045045045, "grad_norm": 23.95705795288086, "learning_rate": 8.561072784882156e-06, "loss": 12.590988159179688, "loss_alignment": 1.01904296875, "loss_alignment_w": 0.509521484375, "loss_sft": 0.2764386963099241, "loss_total_v6": 0.7869367375969887, "step": 2132 }, { "epoch": 0.49193513513513515, "grad_norm": 23.587596893310547, "learning_rate": 8.559658856231734e-06, "loss": 12.938674926757812, "loss_alignment": 0.98583984375, "loss_alignment_w": 0.492919921875, "loss_sft": 0.3156404010951519, "loss_total_v6": 0.8086671531200409, "step": 2133 }, { "epoch": 0.49216576576576576, "grad_norm": 26.777551651000977, "learning_rate": 8.558244350125293e-06, "loss": 13.187599182128906, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.3221802990883589, "loss_total_v6": 0.8242249488830566, "step": 2134 }, { "epoch": 0.4923963963963964, "grad_norm": 27.300018310546875, "learning_rate": 8.556829266792297e-06, "loss": 12.742532730102539, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.2942110449075699, "loss_total_v6": 0.7964083030819893, "step": 2135 }, { "epoch": 0.49262702702702704, "grad_norm": 25.012441635131836, "learning_rate": 8.555413606462302e-06, "loss": 12.859235763549805, "loss_alignment": 1.01708984375, "loss_alignment_w": 0.508544921875, "loss_sft": 0.2946080453693867, "loss_total_v6": 0.8037022799253464, "step": 2136 }, { "epoch": 0.49285765765765766, "grad_norm": 23.7838191986084, "learning_rate": 8.553997369364964e-06, "loss": 12.8817138671875, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.3007888253778219, "loss_total_v6": 0.8051070794463158, "step": 2137 }, { "epoch": 0.49308828828828827, "grad_norm": 23.56551170349121, "learning_rate": 8.552580555730023e-06, "loss": 12.172934532165527, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.26224272698163986, "loss_total_v6": 0.7608084157109261, "step": 2138 }, { "epoch": 0.49331891891891894, "grad_norm": 23.95273208618164, "learning_rate": 8.551163165787322e-06, "loss": 12.816003799438477, "loss_alignment": 1.02978515625, "loss_alignment_w": 0.514892578125, "loss_sft": 0.2866112105548382, "loss_total_v6": 0.801000252366066, "step": 2139 }, { "epoch": 0.49354954954954955, "grad_norm": 23.85032844543457, "learning_rate": 8.549745199766792e-06, "loss": 12.451204299926758, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2762013375759125, "loss_total_v6": 0.7782002314925194, "step": 2140 }, { "epoch": 0.49378018018018016, "grad_norm": 24.045473098754883, "learning_rate": 8.548326657898458e-06, "loss": 12.809173583984375, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.306203868240118, "loss_total_v6": 0.8005733713507652, "step": 2141 }, { "epoch": 0.49401081081081083, "grad_norm": 26.043235778808594, "learning_rate": 8.54690754041244e-06, "loss": 12.749902725219727, "loss_alignment": 0.9833984375, "loss_alignment_w": 0.49169921875, "loss_sft": 0.3033539168536663, "loss_total_v6": 0.7968689203262329, "step": 2142 }, { "epoch": 0.49424144144144144, "grad_norm": 25.49140167236328, "learning_rate": 8.545487847538949e-06, "loss": 12.636407852172852, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.29192696139216423, "loss_total_v6": 0.789775475859642, "step": 2143 }, { "epoch": 0.49447207207207206, "grad_norm": 25.14011001586914, "learning_rate": 8.544067579508292e-06, "loss": 12.312053680419922, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.27673598751425743, "loss_total_v6": 0.7695033252239227, "step": 2144 }, { "epoch": 0.4947027027027027, "grad_norm": 23.08805274963379, "learning_rate": 8.542646736550868e-06, "loss": 13.186467170715332, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.3212855104357004, "loss_total_v6": 0.8241541907191277, "step": 2145 }, { "epoch": 0.49493333333333334, "grad_norm": 27.058788299560547, "learning_rate": 8.541225318897168e-06, "loss": 12.694853782653809, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.28210633993148804, "loss_total_v6": 0.793428361415863, "step": 2146 }, { "epoch": 0.49516396396396395, "grad_norm": 24.67147445678711, "learning_rate": 8.539803326777782e-06, "loss": 12.570159912109375, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.28690148517489433, "loss_total_v6": 0.7856349870562553, "step": 2147 }, { "epoch": 0.4953945945945946, "grad_norm": 23.616884231567383, "learning_rate": 8.538380760423384e-06, "loss": 12.548870086669922, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2740657478570938, "loss_total_v6": 0.7843043655157089, "step": 2148 }, { "epoch": 0.49562522522522523, "grad_norm": 25.35395622253418, "learning_rate": 8.536957620064747e-06, "loss": 12.911059379577637, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2990676276385784, "loss_total_v6": 0.8069411888718605, "step": 2149 }, { "epoch": 0.49585585585585584, "grad_norm": 23.04271125793457, "learning_rate": 8.535533905932739e-06, "loss": 12.686527252197266, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2839510589838028, "loss_total_v6": 0.7929079756140709, "step": 2150 }, { "epoch": 0.4960864864864865, "grad_norm": 24.73019027709961, "learning_rate": 8.534109618258315e-06, "loss": 12.948862075805664, "loss_alignment": 0.990234375, "loss_alignment_w": 0.4951171875, "loss_sft": 0.3132254183292389, "loss_total_v6": 0.8093039095401764, "step": 2151 }, { "epoch": 0.4963171171171171, "grad_norm": 24.18784523010254, "learning_rate": 8.532684757272527e-06, "loss": 13.15050983428955, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.3136976398527622, "loss_total_v6": 0.8219068720936775, "step": 2152 }, { "epoch": 0.49654774774774774, "grad_norm": 24.121932983398438, "learning_rate": 8.53125932320652e-06, "loss": 13.067144393920898, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.32020607218146324, "loss_total_v6": 0.8166965469717979, "step": 2153 }, { "epoch": 0.4967783783783784, "grad_norm": 23.88463592529297, "learning_rate": 8.52983331629153e-06, "loss": 12.735177993774414, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.28773942962288857, "loss_total_v6": 0.7959486544132233, "step": 2154 }, { "epoch": 0.497009009009009, "grad_norm": 22.501110076904297, "learning_rate": 8.528406736758888e-06, "loss": 12.904955863952637, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.29824370704591274, "loss_total_v6": 0.8065597265958786, "step": 2155 }, { "epoch": 0.49723963963963963, "grad_norm": 25.745004653930664, "learning_rate": 8.526979584840016e-06, "loss": 12.950504302978516, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.30037326738238335, "loss_total_v6": 0.809406504034996, "step": 2156 }, { "epoch": 0.49747027027027024, "grad_norm": 20.709148406982422, "learning_rate": 8.52555186076643e-06, "loss": 12.997188568115234, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.3053815010935068, "loss_total_v6": 0.8123242482542992, "step": 2157 }, { "epoch": 0.4977009009009009, "grad_norm": 27.040237426757812, "learning_rate": 8.524123564769741e-06, "loss": 12.933942794799805, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.3001622185111046, "loss_total_v6": 0.8083714246749878, "step": 2158 }, { "epoch": 0.4979315315315315, "grad_norm": 21.866403579711914, "learning_rate": 8.522694697081648e-06, "loss": 12.805601119995117, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.2884481940418482, "loss_total_v6": 0.8003500774502754, "step": 2159 }, { "epoch": 0.49816216216216214, "grad_norm": 23.205829620361328, "learning_rate": 8.521265257933948e-06, "loss": 12.961668014526367, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.30563340336084366, "loss_total_v6": 0.8101042285561562, "step": 2160 }, { "epoch": 0.4983927927927928, "grad_norm": 25.728727340698242, "learning_rate": 8.519835247558523e-06, "loss": 13.129888534545898, "loss_alignment": 1.01904296875, "loss_alignment_w": 0.509521484375, "loss_sft": 0.3116763923317194, "loss_total_v6": 0.8206180408596992, "step": 2161 }, { "epoch": 0.4986234234234234, "grad_norm": 25.382579803466797, "learning_rate": 8.518404666187356e-06, "loss": 12.198185920715332, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.26760513335466385, "loss_total_v6": 0.7623866274952888, "step": 2162 }, { "epoch": 0.49885405405405403, "grad_norm": 24.426776885986328, "learning_rate": 8.51697351405252e-06, "loss": 12.87137222290039, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.3000204749405384, "loss_total_v6": 0.8044607788324356, "step": 2163 }, { "epoch": 0.4990846846846847, "grad_norm": 25.5844669342041, "learning_rate": 8.515541791386176e-06, "loss": 12.686948776245117, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.29380403086543083, "loss_total_v6": 0.7929342687129974, "step": 2164 }, { "epoch": 0.4993153153153153, "grad_norm": 23.47614860534668, "learning_rate": 8.514109498420586e-06, "loss": 12.779654502868652, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.3055948317050934, "loss_total_v6": 0.7987283691763878, "step": 2165 }, { "epoch": 0.4995459459459459, "grad_norm": 20.758806228637695, "learning_rate": 8.512676635388098e-06, "loss": 12.465691566467285, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.2847667634487152, "loss_total_v6": 0.7791057452559471, "step": 2166 }, { "epoch": 0.4997765765765766, "grad_norm": 24.161211013793945, "learning_rate": 8.511243202521151e-06, "loss": 12.9974365234375, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.31571197137236595, "loss_total_v6": 0.8123397901654243, "step": 2167 }, { "epoch": 0.5000072072072073, "grad_norm": 23.623897552490234, "learning_rate": 8.509809200052286e-06, "loss": 12.72524642944336, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.286004776135087, "loss_total_v6": 0.7953278720378876, "step": 2168 }, { "epoch": 0.5002378378378378, "grad_norm": 24.285587310791016, "learning_rate": 8.508374628214124e-06, "loss": 12.776601791381836, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.29183897748589516, "loss_total_v6": 0.7985376045107841, "step": 2169 }, { "epoch": 0.5004684684684685, "grad_norm": 25.735387802124023, "learning_rate": 8.506939487239389e-06, "loss": 12.36719036102295, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2670442182570696, "loss_total_v6": 0.7729493752121925, "step": 2170 }, { "epoch": 0.5006990990990992, "grad_norm": 24.450420379638672, "learning_rate": 8.505503777360889e-06, "loss": 13.212244033813477, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.3130394145846367, "loss_total_v6": 0.8257652446627617, "step": 2171 }, { "epoch": 0.5009297297297297, "grad_norm": 23.574899673461914, "learning_rate": 8.504067498811533e-06, "loss": 12.635356903076172, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2866580355912447, "loss_total_v6": 0.7897097989916801, "step": 2172 }, { "epoch": 0.5011603603603604, "grad_norm": 25.38331413269043, "learning_rate": 8.502630651824312e-06, "loss": 12.761110305786133, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.2975388467311859, "loss_total_v6": 0.7975693643093109, "step": 2173 }, { "epoch": 0.5013909909909909, "grad_norm": 26.139144897460938, "learning_rate": 8.50119323663232e-06, "loss": 12.782060623168945, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.29361449740827084, "loss_total_v6": 0.7988787740468979, "step": 2174 }, { "epoch": 0.5016216216216216, "grad_norm": 24.083311080932617, "learning_rate": 8.499755253468732e-06, "loss": 12.509650230407715, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.278206292539835, "loss_total_v6": 0.781853124499321, "step": 2175 }, { "epoch": 0.5018522522522523, "grad_norm": 23.08713722229004, "learning_rate": 8.498316702566828e-06, "loss": 12.866462707519531, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.3013004958629608, "loss_total_v6": 0.8041539117693901, "step": 2176 }, { "epoch": 0.5020828828828828, "grad_norm": 25.777538299560547, "learning_rate": 8.496877584159967e-06, "loss": 12.457837104797363, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.26814728043973446, "loss_total_v6": 0.778614804148674, "step": 2177 }, { "epoch": 0.5023135135135135, "grad_norm": 25.918373107910156, "learning_rate": 8.49543789848161e-06, "loss": 12.822479248046875, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.28068347461521626, "loss_total_v6": 0.8014049232006073, "step": 2178 }, { "epoch": 0.5025441441441442, "grad_norm": 25.032201766967773, "learning_rate": 8.493997645765302e-06, "loss": 12.575567245483398, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2777178976684809, "loss_total_v6": 0.7859729453921318, "step": 2179 }, { "epoch": 0.5027747747747747, "grad_norm": 24.03710174560547, "learning_rate": 8.492556826244687e-06, "loss": 13.086706161499023, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.31141892820596695, "loss_total_v6": 0.8179191499948502, "step": 2180 }, { "epoch": 0.5030054054054054, "grad_norm": 25.294065475463867, "learning_rate": 8.491115440153495e-06, "loss": 12.677684783935547, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.29046323522925377, "loss_total_v6": 0.7923553287982941, "step": 2181 }, { "epoch": 0.5032360360360361, "grad_norm": 23.492767333984375, "learning_rate": 8.489673487725556e-06, "loss": 13.235152244567871, "loss_alignment": 1.0234375, "loss_alignment_w": 0.51171875, "loss_sft": 0.31445593014359474, "loss_total_v6": 0.8271970003843307, "step": 2182 }, { "epoch": 0.5034666666666666, "grad_norm": 24.626869201660156, "learning_rate": 8.488230969194785e-06, "loss": 12.617500305175781, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.28442813642323017, "loss_total_v6": 0.7885937914252281, "step": 2183 }, { "epoch": 0.5036972972972973, "grad_norm": 26.25412368774414, "learning_rate": 8.48678788479519e-06, "loss": 12.684549331665039, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.2828050684183836, "loss_total_v6": 0.7927843183279037, "step": 2184 }, { "epoch": 0.503927927927928, "grad_norm": 28.12506866455078, "learning_rate": 8.485344234760869e-06, "loss": 12.809282302856445, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.29459868371486664, "loss_total_v6": 0.8005801364779472, "step": 2185 }, { "epoch": 0.5041585585585585, "grad_norm": 23.725006103515625, "learning_rate": 8.483900019326017e-06, "loss": 12.859851837158203, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.301039956510067, "loss_total_v6": 0.8037407621741295, "step": 2186 }, { "epoch": 0.5043891891891892, "grad_norm": 22.90800666809082, "learning_rate": 8.482455238724919e-06, "loss": 12.86115837097168, "loss_alignment": 1.01220703125, "loss_alignment_w": 0.506103515625, "loss_sft": 0.2973069306463003, "loss_total_v6": 0.8038224056363106, "step": 2187 }, { "epoch": 0.5046198198198198, "grad_norm": 24.910036087036133, "learning_rate": 8.481009893191947e-06, "loss": 12.838973999023438, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2992468345910311, "loss_total_v6": 0.8024358823895454, "step": 2188 }, { "epoch": 0.5048504504504504, "grad_norm": 23.850341796875, "learning_rate": 8.479563982961572e-06, "loss": 12.567146301269531, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2803196907043457, "loss_total_v6": 0.7854466438293457, "step": 2189 }, { "epoch": 0.5050810810810811, "grad_norm": 32.034324645996094, "learning_rate": 8.478117508268352e-06, "loss": 13.12693977355957, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.30639570392668247, "loss_total_v6": 0.8204337656497955, "step": 2190 }, { "epoch": 0.5053117117117117, "grad_norm": 26.801267623901367, "learning_rate": 8.476670469346935e-06, "loss": 12.653997421264648, "loss_alignment": 0.99267578125, "loss_alignment_w": 0.496337890625, "loss_sft": 0.2944301590323448, "loss_total_v6": 0.7908748537302017, "step": 2191 }, { "epoch": 0.5055423423423423, "grad_norm": 23.221546173095703, "learning_rate": 8.475222866432065e-06, "loss": 12.466602325439453, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.268588300794363, "loss_total_v6": 0.7791626453399658, "step": 2192 }, { "epoch": 0.505772972972973, "grad_norm": 24.137842178344727, "learning_rate": 8.473774699758574e-06, "loss": 13.432289123535156, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.32818080857396126, "loss_total_v6": 0.8395180702209473, "step": 2193 }, { "epoch": 0.5060036036036036, "grad_norm": 24.91803741455078, "learning_rate": 8.47232596956139e-06, "loss": 13.301107406616211, "loss_alignment": 1.0322265625, "loss_alignment_w": 0.51611328125, "loss_sft": 0.315953616052866, "loss_total_v6": 0.831319198012352, "step": 2194 }, { "epoch": 0.5062342342342342, "grad_norm": 24.987323760986328, "learning_rate": 8.470876676075528e-06, "loss": 12.986751556396484, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.3093526177108288, "loss_total_v6": 0.8116719499230385, "step": 2195 }, { "epoch": 0.5064648648648649, "grad_norm": 24.095136642456055, "learning_rate": 8.469426819536093e-06, "loss": 12.825185775756836, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.30296266078948975, "loss_total_v6": 0.8015741184353828, "step": 2196 }, { "epoch": 0.5066954954954955, "grad_norm": 25.040695190429688, "learning_rate": 8.467976400178289e-06, "loss": 12.694870948791504, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.29257489927113056, "loss_total_v6": 0.7934294119477272, "step": 2197 }, { "epoch": 0.5069261261261261, "grad_norm": 23.34026527404785, "learning_rate": 8.4665254182374e-06, "loss": 12.402410507202148, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.27463183365762234, "loss_total_v6": 0.7751506417989731, "step": 2198 }, { "epoch": 0.5071567567567568, "grad_norm": 25.105676651000977, "learning_rate": 8.465073873948812e-06, "loss": 12.668716430664062, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2844552956521511, "loss_total_v6": 0.7917947769165039, "step": 2199 }, { "epoch": 0.5073873873873874, "grad_norm": 26.27223014831543, "learning_rate": 8.463621767547998e-06, "loss": 12.74365520477295, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.2949525713920593, "loss_total_v6": 0.7964784651994705, "step": 2200 }, { "epoch": 0.507618018018018, "grad_norm": 22.022499084472656, "learning_rate": 8.46216909927052e-06, "loss": 12.333252906799316, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.2716370280832052, "loss_total_v6": 0.7708283141255379, "step": 2201 }, { "epoch": 0.5078486486486486, "grad_norm": 24.0716495513916, "learning_rate": 8.460715869352035e-06, "loss": 12.479835510253906, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.28288893960416317, "loss_total_v6": 0.7799897566437721, "step": 2202 }, { "epoch": 0.5080792792792793, "grad_norm": 23.747112274169922, "learning_rate": 8.459262078028287e-06, "loss": 13.202109336853027, "loss_alignment": 1.03515625, "loss_alignment_w": 0.517578125, "loss_sft": 0.30689762346446514, "loss_total_v6": 0.8251318633556366, "step": 2203 }, { "epoch": 0.5083099099099099, "grad_norm": 27.249305725097656, "learning_rate": 8.457807725535117e-06, "loss": 12.325017929077148, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.26948962174355984, "loss_total_v6": 0.77031359821558, "step": 2204 }, { "epoch": 0.5085405405405405, "grad_norm": 21.66231346130371, "learning_rate": 8.45635281210845e-06, "loss": 12.350582122802734, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.2738492786884308, "loss_total_v6": 0.7719114348292351, "step": 2205 }, { "epoch": 0.5087711711711712, "grad_norm": 23.35287094116211, "learning_rate": 8.454897337984307e-06, "loss": 12.260429382324219, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.26024957932531834, "loss_total_v6": 0.7662767842411995, "step": 2206 }, { "epoch": 0.5090018018018018, "grad_norm": 23.326269149780273, "learning_rate": 8.453441303398797e-06, "loss": 12.091409683227539, "loss_alignment": 0.97021484375, "loss_alignment_w": 0.485107421875, "loss_sft": 0.27088035456836224, "loss_total_v6": 0.7557131201028824, "step": 2207 }, { "epoch": 0.5092324324324324, "grad_norm": 25.10110092163086, "learning_rate": 8.451984708588122e-06, "loss": 12.817747116088867, "loss_alignment": 1.01318359375, "loss_alignment_w": 0.506591796875, "loss_sft": 0.2938002273440361, "loss_total_v6": 0.801109179854393, "step": 2208 }, { "epoch": 0.5094630630630631, "grad_norm": 22.486059188842773, "learning_rate": 8.450527553788576e-06, "loss": 12.82321548461914, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.29723960533738136, "loss_total_v6": 0.8014510348439217, "step": 2209 }, { "epoch": 0.5096936936936937, "grad_norm": 24.58180046081543, "learning_rate": 8.44906983923654e-06, "loss": 12.586365699768066, "loss_alignment": 1.00830078125, "loss_alignment_w": 0.504150390625, "loss_sft": 0.2833672296255827, "loss_total_v6": 0.7866478711366653, "step": 2210 }, { "epoch": 0.5099243243243243, "grad_norm": 24.304393768310547, "learning_rate": 8.447611565168487e-06, "loss": 12.511503219604492, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.2829454727470875, "loss_total_v6": 0.7819689139723778, "step": 2211 }, { "epoch": 0.510154954954955, "grad_norm": 24.205886840820312, "learning_rate": 8.446152731820984e-06, "loss": 12.301738739013672, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.26214477606117725, "loss_total_v6": 0.7688586413860321, "step": 2212 }, { "epoch": 0.5103855855855856, "grad_norm": 23.823747634887695, "learning_rate": 8.444693339430683e-06, "loss": 12.782157897949219, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.2969774901866913, "loss_total_v6": 0.7988848760724068, "step": 2213 }, { "epoch": 0.5106162162162162, "grad_norm": 23.866626739501953, "learning_rate": 8.443233388234334e-06, "loss": 12.964883804321289, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.31300608068704605, "loss_total_v6": 0.8103052526712418, "step": 2214 }, { "epoch": 0.5108468468468469, "grad_norm": 27.646034240722656, "learning_rate": 8.44177287846877e-06, "loss": 12.507259368896484, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.2809712812304497, "loss_total_v6": 0.7817037180066109, "step": 2215 }, { "epoch": 0.5110774774774774, "grad_norm": 23.028472900390625, "learning_rate": 8.440311810370921e-06, "loss": 12.470283508300781, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.27905699610710144, "loss_total_v6": 0.7793926894664764, "step": 2216 }, { "epoch": 0.5113081081081081, "grad_norm": 24.909515380859375, "learning_rate": 8.438850184177805e-06, "loss": 13.423933029174805, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.3250187821686268, "loss_total_v6": 0.8389958068728447, "step": 2217 }, { "epoch": 0.5115387387387388, "grad_norm": 23.859811782836914, "learning_rate": 8.437388000126527e-06, "loss": 12.938063621520996, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.30272382870316505, "loss_total_v6": 0.8086289763450623, "step": 2218 }, { "epoch": 0.5117693693693693, "grad_norm": 23.309329986572266, "learning_rate": 8.435925258454287e-06, "loss": 12.41860580444336, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.2743775974959135, "loss_total_v6": 0.77616286277771, "step": 2219 }, { "epoch": 0.512, "grad_norm": 22.943696975708008, "learning_rate": 8.434461959398377e-06, "loss": 12.205509185791016, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.26052503660321236, "loss_total_v6": 0.7628443613648415, "step": 2220 }, { "epoch": 0.5122306306306307, "grad_norm": 25.883169174194336, "learning_rate": 8.432998103196175e-06, "loss": 12.853480339050293, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2988106720149517, "loss_total_v6": 0.8033425137400627, "step": 2221 }, { "epoch": 0.5124612612612612, "grad_norm": 22.73973274230957, "learning_rate": 8.43153369008515e-06, "loss": 12.655136108398438, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.285910589620471, "loss_total_v6": 0.7909459844231606, "step": 2222 }, { "epoch": 0.5126918918918919, "grad_norm": 25.918811798095703, "learning_rate": 8.430068720302863e-06, "loss": 13.249269485473633, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.32165535911917686, "loss_total_v6": 0.8280793279409409, "step": 2223 }, { "epoch": 0.5129225225225226, "grad_norm": 24.248281478881836, "learning_rate": 8.428603194086965e-06, "loss": 12.72816276550293, "loss_alignment": 0.984375, "loss_alignment_w": 0.4921875, "loss_sft": 0.3027580678462982, "loss_total_v6": 0.7955101951956749, "step": 2224 }, { "epoch": 0.5131531531531531, "grad_norm": 25.303945541381836, "learning_rate": 8.4271371116752e-06, "loss": 12.95062255859375, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.30804067477583885, "loss_total_v6": 0.8094139620661736, "step": 2225 }, { "epoch": 0.5133837837837838, "grad_norm": 21.84794807434082, "learning_rate": 8.425670473305395e-06, "loss": 12.173418045043945, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.2611132822930813, "loss_total_v6": 0.7608386427164078, "step": 2226 }, { "epoch": 0.5136144144144145, "grad_norm": 23.128299713134766, "learning_rate": 8.424203279215473e-06, "loss": 13.07494068145752, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.31425410881638527, "loss_total_v6": 0.817183792591095, "step": 2227 }, { "epoch": 0.513845045045045, "grad_norm": 24.752883911132812, "learning_rate": 8.422735529643445e-06, "loss": 12.966575622558594, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.3106550946831703, "loss_total_v6": 0.8104109615087509, "step": 2228 }, { "epoch": 0.5140756756756757, "grad_norm": 21.507471084594727, "learning_rate": 8.421267224827411e-06, "loss": 12.882755279541016, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.30247134529054165, "loss_total_v6": 0.8051721528172493, "step": 2229 }, { "epoch": 0.5143063063063064, "grad_norm": 22.218427658081055, "learning_rate": 8.419798365005569e-06, "loss": 12.62319278717041, "loss_alignment": 0.98388671875, "loss_alignment_w": 0.491943359375, "loss_sft": 0.2967925760895014, "loss_total_v6": 0.7889495566487312, "step": 2230 }, { "epoch": 0.5145369369369369, "grad_norm": 24.068941116333008, "learning_rate": 8.418328950416195e-06, "loss": 12.580782890319824, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.2909376174211502, "loss_total_v6": 0.786298967897892, "step": 2231 }, { "epoch": 0.5147675675675676, "grad_norm": 22.051496505737305, "learning_rate": 8.416858981297663e-06, "loss": 12.889866828918457, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.31014858186244965, "loss_total_v6": 0.8056166842579842, "step": 2232 }, { "epoch": 0.5149981981981981, "grad_norm": 23.475418090820312, "learning_rate": 8.415388457888433e-06, "loss": 12.53075122833252, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2756341192871332, "loss_total_v6": 0.7831719666719437, "step": 2233 }, { "epoch": 0.5152288288288288, "grad_norm": 27.006919860839844, "learning_rate": 8.41391738042706e-06, "loss": 13.08095932006836, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.3079927060753107, "loss_total_v6": 0.8175599426031113, "step": 2234 }, { "epoch": 0.5154594594594595, "grad_norm": 64.86107635498047, "learning_rate": 8.412445749152182e-06, "loss": 12.6987886428833, "loss_alignment": 0.98486328125, "loss_alignment_w": 0.492431640625, "loss_sft": 0.30089167691767216, "loss_total_v6": 0.7936742901802063, "step": 2235 }, { "epoch": 0.51569009009009, "grad_norm": 24.971418380737305, "learning_rate": 8.410973564302534e-06, "loss": 13.532402992248535, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.33791688084602356, "loss_total_v6": 0.8457751795649529, "step": 2236 }, { "epoch": 0.5159207207207207, "grad_norm": 22.493671417236328, "learning_rate": 8.409500826116934e-06, "loss": 12.679817199707031, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.2922444250434637, "loss_total_v6": 0.7924885749816895, "step": 2237 }, { "epoch": 0.5161513513513514, "grad_norm": 27.399211883544922, "learning_rate": 8.408027534834293e-06, "loss": 13.340414047241211, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.32358305901288986, "loss_total_v6": 0.8337759301066399, "step": 2238 }, { "epoch": 0.5163819819819819, "grad_norm": 24.77648162841797, "learning_rate": 8.406553690693615e-06, "loss": 12.887992858886719, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.3078646920621395, "loss_total_v6": 0.8054995685815811, "step": 2239 }, { "epoch": 0.5166126126126126, "grad_norm": 24.479829788208008, "learning_rate": 8.405079293933986e-06, "loss": 12.873871803283691, "loss_alignment": 0.98583984375, "loss_alignment_w": 0.492919921875, "loss_sft": 0.3111629541963339, "loss_total_v6": 0.8046169653534889, "step": 2240 }, { "epoch": 0.5168432432432433, "grad_norm": 25.78459930419922, "learning_rate": 8.40360434479459e-06, "loss": 12.796710014343262, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.30591318011283875, "loss_total_v6": 0.7997943833470345, "step": 2241 }, { "epoch": 0.5170738738738738, "grad_norm": 28.033885955810547, "learning_rate": 8.402128843514695e-06, "loss": 13.0794677734375, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.31909947097301483, "loss_total_v6": 0.8174667581915855, "step": 2242 }, { "epoch": 0.5173045045045045, "grad_norm": 21.93509864807129, "learning_rate": 8.400652790333659e-06, "loss": 12.334074974060059, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.25888629630208015, "loss_total_v6": 0.7708796858787537, "step": 2243 }, { "epoch": 0.5175351351351352, "grad_norm": 24.309534072875977, "learning_rate": 8.399176185490932e-06, "loss": 13.00098705291748, "loss_alignment": 1.0341796875, "loss_alignment_w": 0.51708984375, "loss_sft": 0.2949835918843746, "loss_total_v6": 0.8125617131590843, "step": 2244 }, { "epoch": 0.5177657657657657, "grad_norm": 24.72577476501465, "learning_rate": 8.397699029226052e-06, "loss": 12.822352409362793, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.30611198768019676, "loss_total_v6": 0.8013970106840134, "step": 2245 }, { "epoch": 0.5179963963963964, "grad_norm": 25.897266387939453, "learning_rate": 8.396221321778645e-06, "loss": 12.968648910522461, "loss_alignment": 0.9853515625, "loss_alignment_w": 0.49267578125, "loss_sft": 0.3181241750717163, "loss_total_v6": 0.8105405271053314, "step": 2246 }, { "epoch": 0.518227027027027, "grad_norm": 27.783103942871094, "learning_rate": 8.394743063388433e-06, "loss": 12.951757431030273, "loss_alignment": 1.0205078125, "loss_alignment_w": 0.51025390625, "loss_sft": 0.29967341385781765, "loss_total_v6": 0.8094848319888115, "step": 2247 }, { "epoch": 0.5184576576576576, "grad_norm": 26.929241180419922, "learning_rate": 8.393264254295218e-06, "loss": 12.48271369934082, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.26634511910378933, "loss_total_v6": 0.7801695540547371, "step": 2248 }, { "epoch": 0.5186882882882883, "grad_norm": 23.681875228881836, "learning_rate": 8.391784894738898e-06, "loss": 12.788202285766602, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2910382319241762, "loss_total_v6": 0.7992626875638962, "step": 2249 }, { "epoch": 0.518918918918919, "grad_norm": 25.289113998413086, "learning_rate": 8.390304984959455e-06, "loss": 12.349178314208984, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2693822253495455, "loss_total_v6": 0.7718236520886421, "step": 2250 }, { "epoch": 0.5191495495495495, "grad_norm": 26.121299743652344, "learning_rate": 8.388824525196967e-06, "loss": 12.785629272460938, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.29345611296594143, "loss_total_v6": 0.7991018816828728, "step": 2251 }, { "epoch": 0.5193801801801802, "grad_norm": 23.507078170776367, "learning_rate": 8.387343515691594e-06, "loss": 13.01129150390625, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.30766673386096954, "loss_total_v6": 0.8132056966423988, "step": 2252 }, { "epoch": 0.5196108108108108, "grad_norm": 22.796974182128906, "learning_rate": 8.385861956683594e-06, "loss": 13.07017707824707, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.3093024082481861, "loss_total_v6": 0.8168860375881195, "step": 2253 }, { "epoch": 0.5198414414414414, "grad_norm": 24.222986221313477, "learning_rate": 8.384379848413304e-06, "loss": 12.551855087280273, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.2840941995382309, "loss_total_v6": 0.7844908982515335, "step": 2254 }, { "epoch": 0.5200720720720721, "grad_norm": 21.711122512817383, "learning_rate": 8.382897191121157e-06, "loss": 12.596688270568848, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.2867742795497179, "loss_total_v6": 0.7872930392622948, "step": 2255 }, { "epoch": 0.5203027027027027, "grad_norm": 25.454845428466797, "learning_rate": 8.381413985047672e-06, "loss": 13.087929725646973, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.3099083751440048, "loss_total_v6": 0.8179955631494522, "step": 2256 }, { "epoch": 0.5205333333333333, "grad_norm": 22.607301712036133, "learning_rate": 8.379930230433461e-06, "loss": 13.141716003417969, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.31539106369018555, "loss_total_v6": 0.821357250213623, "step": 2257 }, { "epoch": 0.520763963963964, "grad_norm": 26.15815544128418, "learning_rate": 8.378445927519218e-06, "loss": 12.625642776489258, "loss_alignment": 0.9873046875, "loss_alignment_w": 0.49365234375, "loss_sft": 0.2949468083679676, "loss_total_v6": 0.7891026809811592, "step": 2258 }, { "epoch": 0.5209945945945946, "grad_norm": 22.168367385864258, "learning_rate": 8.376961076545733e-06, "loss": 12.500455856323242, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.2796000447124243, "loss_total_v6": 0.781278483569622, "step": 2259 }, { "epoch": 0.5212252252252252, "grad_norm": 24.762723922729492, "learning_rate": 8.375475677753882e-06, "loss": 12.254913330078125, "loss_alignment": 1.02099609375, "loss_alignment_w": 0.510498046875, "loss_sft": 0.25580024532973766, "loss_total_v6": 0.7659320682287216, "step": 2260 }, { "epoch": 0.5214558558558559, "grad_norm": 23.922590255737305, "learning_rate": 8.373989731384627e-06, "loss": 12.292818069458008, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.2675687037408352, "loss_total_v6": 0.7683011367917061, "step": 2261 }, { "epoch": 0.5216864864864865, "grad_norm": 24.12310028076172, "learning_rate": 8.372503237679024e-06, "loss": 13.347980499267578, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.33311958238482475, "loss_total_v6": 0.8342487439513206, "step": 2262 }, { "epoch": 0.5219171171171171, "grad_norm": 27.024192810058594, "learning_rate": 8.371016196878215e-06, "loss": 12.639775276184082, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.28760555759072304, "loss_total_v6": 0.7899859398603439, "step": 2263 }, { "epoch": 0.5221477477477477, "grad_norm": 22.174549102783203, "learning_rate": 8.36952860922343e-06, "loss": 12.838090896606445, "loss_alignment": 1.00732421875, "loss_alignment_w": 0.503662109375, "loss_sft": 0.2992678489536047, "loss_total_v6": 0.8023806288838387, "step": 2264 }, { "epoch": 0.5223783783783784, "grad_norm": 25.14573860168457, "learning_rate": 8.36804047495599e-06, "loss": 12.42478084564209, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.2834610491991043, "loss_total_v6": 0.7765488103032112, "step": 2265 }, { "epoch": 0.522609009009009, "grad_norm": 23.238391876220703, "learning_rate": 8.366551794317304e-06, "loss": 13.059846878051758, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.31035052984952927, "loss_total_v6": 0.8162404149770737, "step": 2266 }, { "epoch": 0.5228396396396396, "grad_norm": 23.82570457458496, "learning_rate": 8.365062567548868e-06, "loss": 13.052799224853516, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.31155798211693764, "loss_total_v6": 0.8157999068498611, "step": 2267 }, { "epoch": 0.5230702702702703, "grad_norm": 25.09827423095703, "learning_rate": 8.363572794892268e-06, "loss": 13.436161041259766, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.33684559166431427, "loss_total_v6": 0.8397600278258324, "step": 2268 }, { "epoch": 0.5233009009009009, "grad_norm": 27.853252410888672, "learning_rate": 8.36208247658918e-06, "loss": 12.5537109375, "loss_alignment": 0.982421875, "loss_alignment_w": 0.4912109375, "loss_sft": 0.2933502569794655, "loss_total_v6": 0.7846069633960724, "step": 2269 }, { "epoch": 0.5235315315315315, "grad_norm": 22.76788902282715, "learning_rate": 8.360591612881363e-06, "loss": 12.668261528015137, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.28985900431871414, "loss_total_v6": 0.791766345500946, "step": 2270 }, { "epoch": 0.5237621621621622, "grad_norm": 26.321231842041016, "learning_rate": 8.359100204010673e-06, "loss": 12.953981399536133, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.3130570240318775, "loss_total_v6": 0.8096238449215889, "step": 2271 }, { "epoch": 0.5239927927927928, "grad_norm": 22.913625717163086, "learning_rate": 8.357608250219046e-06, "loss": 12.818392753601074, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.30262964591383934, "loss_total_v6": 0.8011495247483253, "step": 2272 }, { "epoch": 0.5242234234234234, "grad_norm": 23.810955047607422, "learning_rate": 8.356115751748514e-06, "loss": 12.778879165649414, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2914167959243059, "loss_total_v6": 0.798679992556572, "step": 2273 }, { "epoch": 0.5244540540540541, "grad_norm": 22.546499252319336, "learning_rate": 8.354622708841192e-06, "loss": 12.043601989746094, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.2579588871449232, "loss_total_v6": 0.7527251318097115, "step": 2274 }, { "epoch": 0.5246846846846847, "grad_norm": 25.795612335205078, "learning_rate": 8.35312912173928e-06, "loss": 12.251504898071289, "loss_alignment": 0.99267578125, "loss_alignment_w": 0.496337890625, "loss_sft": 0.26904548332095146, "loss_total_v6": 0.765719048678875, "step": 2275 }, { "epoch": 0.5249153153153153, "grad_norm": 25.984098434448242, "learning_rate": 8.35163499068508e-06, "loss": 12.830476760864258, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2940007448196411, "loss_total_v6": 0.8019048124551773, "step": 2276 }, { "epoch": 0.525145945945946, "grad_norm": 23.142202377319336, "learning_rate": 8.350140315920964e-06, "loss": 12.415116310119629, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.27032952196896076, "loss_total_v6": 0.7759447619318962, "step": 2277 }, { "epoch": 0.5253765765765765, "grad_norm": 23.896045684814453, "learning_rate": 8.348645097689409e-06, "loss": 11.938642501831055, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.23963440023362637, "loss_total_v6": 0.7461651638150215, "step": 2278 }, { "epoch": 0.5256072072072072, "grad_norm": 27.85228729248047, "learning_rate": 8.34714933623297e-06, "loss": 12.629989624023438, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.285437585785985, "loss_total_v6": 0.7893743515014648, "step": 2279 }, { "epoch": 0.5258378378378379, "grad_norm": 25.347599029541016, "learning_rate": 8.345653031794292e-06, "loss": 12.833819389343262, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.30009957030415535, "loss_total_v6": 0.8021137416362762, "step": 2280 }, { "epoch": 0.5260684684684684, "grad_norm": 24.56403923034668, "learning_rate": 8.34415618461611e-06, "loss": 13.034079551696777, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.3071379065513611, "loss_total_v6": 0.8146299794316292, "step": 2281 }, { "epoch": 0.5262990990990991, "grad_norm": 21.55256462097168, "learning_rate": 8.342658794941245e-06, "loss": 12.752894401550293, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2915627174079418, "loss_total_v6": 0.7970558851957321, "step": 2282 }, { "epoch": 0.5265297297297298, "grad_norm": 23.56252670288086, "learning_rate": 8.34116086301261e-06, "loss": 12.341014862060547, "loss_alignment": 1.02294921875, "loss_alignment_w": 0.511474609375, "loss_sft": 0.26037289574742317, "loss_total_v6": 0.7713134363293648, "step": 2283 }, { "epoch": 0.5267603603603603, "grad_norm": 24.596813201904297, "learning_rate": 8.339662389073197e-06, "loss": 12.781441688537598, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.29258400574326515, "loss_total_v6": 0.7988401129841805, "step": 2284 }, { "epoch": 0.526990990990991, "grad_norm": 21.6434383392334, "learning_rate": 8.338163373366099e-06, "loss": 12.904107093811035, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2989993579685688, "loss_total_v6": 0.8065066859126091, "step": 2285 }, { "epoch": 0.5272216216216217, "grad_norm": 26.869836807250977, "learning_rate": 8.336663816134483e-06, "loss": 12.485641479492188, "loss_alignment": 1.02734375, "loss_alignment_w": 0.513671875, "loss_sft": 0.26703169755637646, "loss_total_v6": 0.7803526297211647, "step": 2286 }, { "epoch": 0.5274522522522522, "grad_norm": 24.116357803344727, "learning_rate": 8.335163717621615e-06, "loss": 12.58117389678955, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.2877119053155184, "loss_total_v6": 0.7863233685493469, "step": 2287 }, { "epoch": 0.5276828828828829, "grad_norm": 26.16998291015625, "learning_rate": 8.333663078070845e-06, "loss": 12.272825241088867, "loss_alignment": 1.02001953125, "loss_alignment_w": 0.510009765625, "loss_sft": 0.25580589659512043, "loss_total_v6": 0.7670516222715378, "step": 2288 }, { "epoch": 0.5279135135135136, "grad_norm": 22.33031463623047, "learning_rate": 8.332161897725607e-06, "loss": 13.324150085449219, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.327571390196681, "loss_total_v6": 0.8327593877911568, "step": 2289 }, { "epoch": 0.5281441441441441, "grad_norm": 23.71630859375, "learning_rate": 8.330660176829427e-06, "loss": 12.524555206298828, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2768337205052376, "loss_total_v6": 0.7827846854925156, "step": 2290 }, { "epoch": 0.5283747747747748, "grad_norm": 22.240615844726562, "learning_rate": 8.32915791562592e-06, "loss": 12.900971412658691, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.30025298707187176, "loss_total_v6": 0.8063107281923294, "step": 2291 }, { "epoch": 0.5286054054054055, "grad_norm": 22.67416000366211, "learning_rate": 8.327655114358782e-06, "loss": 12.757698059082031, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2876057103276253, "loss_total_v6": 0.7973560690879822, "step": 2292 }, { "epoch": 0.528836036036036, "grad_norm": 23.82894515991211, "learning_rate": 8.326151773271805e-06, "loss": 12.522087097167969, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.28571268916130066, "loss_total_v6": 0.7826304137706757, "step": 2293 }, { "epoch": 0.5290666666666667, "grad_norm": 23.163427352905273, "learning_rate": 8.324647892608861e-06, "loss": 12.112550735473633, "loss_alignment": 0.962890625, "loss_alignment_w": 0.4814453125, "loss_sft": 0.2760773077607155, "loss_total_v6": 0.7570343911647797, "step": 2294 }, { "epoch": 0.5292972972972972, "grad_norm": 21.767393112182617, "learning_rate": 8.323143472613916e-06, "loss": 12.405998229980469, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.27352854795753956, "loss_total_v6": 0.7753748521208763, "step": 2295 }, { "epoch": 0.5295279279279279, "grad_norm": 25.39256477355957, "learning_rate": 8.321638513531018e-06, "loss": 12.72439193725586, "loss_alignment": 0.99267578125, "loss_alignment_w": 0.496337890625, "loss_sft": 0.29957746900618076, "loss_total_v6": 0.7952745109796524, "step": 2296 }, { "epoch": 0.5297585585585586, "grad_norm": 24.081838607788086, "learning_rate": 8.320133015604306e-06, "loss": 12.87074089050293, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.29381644912064075, "loss_total_v6": 0.8044213131070137, "step": 2297 }, { "epoch": 0.5299891891891891, "grad_norm": 24.992389678955078, "learning_rate": 8.318626979078006e-06, "loss": 12.617973327636719, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.291888777166605, "loss_total_v6": 0.7886233553290367, "step": 2298 }, { "epoch": 0.5302198198198198, "grad_norm": 24.21412467956543, "learning_rate": 8.31712040419643e-06, "loss": 13.194607734680176, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.3243120536208153, "loss_total_v6": 0.824662983417511, "step": 2299 }, { "epoch": 0.5304504504504505, "grad_norm": 24.64269256591797, "learning_rate": 8.315613291203977e-06, "loss": 12.276653289794922, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.26285047829151154, "loss_total_v6": 0.7672908082604408, "step": 2300 }, { "epoch": 0.530681081081081, "grad_norm": 25.42557144165039, "learning_rate": 8.314105640345135e-06, "loss": 12.929786682128906, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.3075471092015505, "loss_total_v6": 0.8081116527318954, "step": 2301 }, { "epoch": 0.5309117117117117, "grad_norm": 23.349586486816406, "learning_rate": 8.31259745186448e-06, "loss": 12.76199722290039, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.30053926445543766, "loss_total_v6": 0.7976248264312744, "step": 2302 }, { "epoch": 0.5311423423423424, "grad_norm": 20.75191307067871, "learning_rate": 8.311088726006674e-06, "loss": 12.457172393798828, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2703487556427717, "loss_total_v6": 0.778573252260685, "step": 2303 }, { "epoch": 0.5313729729729729, "grad_norm": 28.809417724609375, "learning_rate": 8.309579463016461e-06, "loss": 12.900236129760742, "loss_alignment": 0.98486328125, "loss_alignment_w": 0.492431640625, "loss_sft": 0.3122004419565201, "loss_total_v6": 0.8062647357583046, "step": 2304 }, { "epoch": 0.5316036036036036, "grad_norm": 23.23138427734375, "learning_rate": 8.308069663138682e-06, "loss": 13.079744338989258, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.32045946456491947, "loss_total_v6": 0.8174840211868286, "step": 2305 }, { "epoch": 0.5318342342342343, "grad_norm": 22.201580047607422, "learning_rate": 8.30655932661826e-06, "loss": 12.749052047729492, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.2948473170399666, "loss_total_v6": 0.7968157082796097, "step": 2306 }, { "epoch": 0.5320648648648648, "grad_norm": 23.440702438354492, "learning_rate": 8.305048453700204e-06, "loss": 12.466951370239258, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.2830754593014717, "loss_total_v6": 0.7791844829916954, "step": 2307 }, { "epoch": 0.5322954954954955, "grad_norm": 22.176342010498047, "learning_rate": 8.303537044629611e-06, "loss": 12.654996871948242, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.28268231824040413, "loss_total_v6": 0.7909373417496681, "step": 2308 }, { "epoch": 0.5325261261261262, "grad_norm": 24.000131607055664, "learning_rate": 8.302025099651666e-06, "loss": 12.952747344970703, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.30934836342930794, "loss_total_v6": 0.8095467165112495, "step": 2309 }, { "epoch": 0.5327567567567567, "grad_norm": 28.600805282592773, "learning_rate": 8.30051261901164e-06, "loss": 12.256975173950195, "loss_alignment": 0.990234375, "loss_alignment_w": 0.4951171875, "loss_sft": 0.2704859711229801, "loss_total_v6": 0.7660609260201454, "step": 2310 }, { "epoch": 0.5329873873873874, "grad_norm": 24.60181999206543, "learning_rate": 8.29899960295489e-06, "loss": 12.700092315673828, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.28740813583135605, "loss_total_v6": 0.7937558144330978, "step": 2311 }, { "epoch": 0.533218018018018, "grad_norm": 23.183921813964844, "learning_rate": 8.297486051726864e-06, "loss": 12.514416694641113, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.27853470854461193, "loss_total_v6": 0.7821510508656502, "step": 2312 }, { "epoch": 0.5334486486486486, "grad_norm": 30.104820251464844, "learning_rate": 8.295971965573088e-06, "loss": 12.951135635375977, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2973762918263674, "loss_total_v6": 0.8094459623098373, "step": 2313 }, { "epoch": 0.5336792792792793, "grad_norm": 23.2142391204834, "learning_rate": 8.294457344739188e-06, "loss": 12.978567123413086, "loss_alignment": 1.02197265625, "loss_alignment_w": 0.510986328125, "loss_sft": 0.2997468039393425, "loss_total_v6": 0.8111604005098343, "step": 2314 }, { "epoch": 0.5339099099099099, "grad_norm": 24.41550064086914, "learning_rate": 8.292942189470863e-06, "loss": 12.588329315185547, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.2875182908028364, "loss_total_v6": 0.7867706120014191, "step": 2315 }, { "epoch": 0.5341405405405405, "grad_norm": 22.88343620300293, "learning_rate": 8.291426500013909e-06, "loss": 12.329879760742188, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.2703886292874813, "loss_total_v6": 0.7706175222992897, "step": 2316 }, { "epoch": 0.5343711711711712, "grad_norm": 22.393531799316406, "learning_rate": 8.289910276614202e-06, "loss": 12.357624053955078, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.2739841416478157, "loss_total_v6": 0.77235147356987, "step": 2317 }, { "epoch": 0.5346018018018018, "grad_norm": 23.55630111694336, "learning_rate": 8.288393519517708e-06, "loss": 12.201805114746094, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.2582487724721432, "loss_total_v6": 0.7626127824187279, "step": 2318 }, { "epoch": 0.5348324324324324, "grad_norm": 20.611513137817383, "learning_rate": 8.28687622897048e-06, "loss": 12.59176254272461, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.2831704318523407, "loss_total_v6": 0.7869851365685463, "step": 2319 }, { "epoch": 0.5350630630630631, "grad_norm": 23.493810653686523, "learning_rate": 8.285358405218655e-06, "loss": 12.71807861328125, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.28406147845089436, "loss_total_v6": 0.7948799431324005, "step": 2320 }, { "epoch": 0.5352936936936937, "grad_norm": 28.466127395629883, "learning_rate": 8.283840048508458e-06, "loss": 12.91750717163086, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.3025377355515957, "loss_total_v6": 0.8073442429304123, "step": 2321 }, { "epoch": 0.5355243243243243, "grad_norm": 24.33927345275879, "learning_rate": 8.2823211590862e-06, "loss": 12.152158737182617, "loss_alignment": 1.00732421875, "loss_alignment_w": 0.503662109375, "loss_sft": 0.25709899328649044, "loss_total_v6": 0.7595098912715912, "step": 2322 }, { "epoch": 0.535754954954955, "grad_norm": 24.668994903564453, "learning_rate": 8.280801737198281e-06, "loss": 13.375463485717773, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.32499533891677856, "loss_total_v6": 0.835966445505619, "step": 2323 }, { "epoch": 0.5359855855855856, "grad_norm": 23.93448257446289, "learning_rate": 8.279281783091182e-06, "loss": 12.97872543334961, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.3058907687664032, "loss_total_v6": 0.8111703470349312, "step": 2324 }, { "epoch": 0.5362162162162162, "grad_norm": 24.175291061401367, "learning_rate": 8.277761297011475e-06, "loss": 12.807433128356934, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.296405715867877, "loss_total_v6": 0.8004645556211472, "step": 2325 }, { "epoch": 0.5364468468468468, "grad_norm": 24.729787826538086, "learning_rate": 8.276240279205817e-06, "loss": 12.416873931884766, "loss_alignment": 0.9912109375, "loss_alignment_w": 0.49560546875, "loss_sft": 0.2796099241822958, "loss_total_v6": 0.7760546430945396, "step": 2326 }, { "epoch": 0.5366774774774775, "grad_norm": 25.017616271972656, "learning_rate": 8.27471872992095e-06, "loss": 13.24254035949707, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.3301612511277199, "loss_total_v6": 0.8276588022708893, "step": 2327 }, { "epoch": 0.5369081081081081, "grad_norm": 20.916765213012695, "learning_rate": 8.273196649403703e-06, "loss": 12.371260643005371, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.26284307800233364, "loss_total_v6": 0.7732037976384163, "step": 2328 }, { "epoch": 0.5371387387387387, "grad_norm": 26.008827209472656, "learning_rate": 8.271674037900994e-06, "loss": 12.491316795349121, "loss_alignment": 0.982421875, "loss_alignment_w": 0.4912109375, "loss_sft": 0.2896336819976568, "loss_total_v6": 0.7807072922587395, "step": 2329 }, { "epoch": 0.5373693693693694, "grad_norm": 24.259225845336914, "learning_rate": 8.270150895659824e-06, "loss": 12.967889785766602, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.30284844152629375, "loss_total_v6": 0.8104930892586708, "step": 2330 }, { "epoch": 0.5376, "grad_norm": 22.03498649597168, "learning_rate": 8.268627222927276e-06, "loss": 12.678125381469727, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.2981353662908077, "loss_total_v6": 0.7923828065395355, "step": 2331 }, { "epoch": 0.5378306306306306, "grad_norm": 26.93153190612793, "learning_rate": 8.267103019950529e-06, "loss": 13.071014404296875, "loss_alignment": 0.9755859375, "loss_alignment_w": 0.48779296875, "loss_sft": 0.3279247246682644, "loss_total_v6": 0.8169383704662323, "step": 2332 }, { "epoch": 0.5380612612612613, "grad_norm": 24.45736312866211, "learning_rate": 8.265578286976839e-06, "loss": 12.49305534362793, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2742699310183525, "loss_total_v6": 0.7808159664273262, "step": 2333 }, { "epoch": 0.5382918918918919, "grad_norm": 24.538347244262695, "learning_rate": 8.264053024253554e-06, "loss": 12.31425666809082, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.2670165169984102, "loss_total_v6": 0.7696410343050957, "step": 2334 }, { "epoch": 0.5385225225225225, "grad_norm": 22.948198318481445, "learning_rate": 8.262527232028106e-06, "loss": 12.83397388458252, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.3031304366886616, "loss_total_v6": 0.8021233454346657, "step": 2335 }, { "epoch": 0.5387531531531532, "grad_norm": 22.08570098876953, "learning_rate": 8.26100091054801e-06, "loss": 12.328447341918945, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.27234376035630703, "loss_total_v6": 0.7705279588699341, "step": 2336 }, { "epoch": 0.5389837837837838, "grad_norm": 24.135421752929688, "learning_rate": 8.259474060060872e-06, "loss": 12.338140487670898, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.27009621262550354, "loss_total_v6": 0.7711338028311729, "step": 2337 }, { "epoch": 0.5392144144144144, "grad_norm": 23.829872131347656, "learning_rate": 8.25794668081438e-06, "loss": 12.719537734985352, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.29263648204505444, "loss_total_v6": 0.7949711158871651, "step": 2338 }, { "epoch": 0.5394450450450451, "grad_norm": 24.467880249023438, "learning_rate": 8.256418773056309e-06, "loss": 12.903789520263672, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2980792336165905, "loss_total_v6": 0.8064868301153183, "step": 2339 }, { "epoch": 0.5396756756756756, "grad_norm": 24.752840042114258, "learning_rate": 8.25489033703452e-06, "loss": 12.699121475219727, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.28292238153517246, "loss_total_v6": 0.7936950847506523, "step": 2340 }, { "epoch": 0.5399063063063063, "grad_norm": 23.21992301940918, "learning_rate": 8.253361372996958e-06, "loss": 12.799829483032227, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.293855344876647, "loss_total_v6": 0.7999893724918365, "step": 2341 }, { "epoch": 0.540136936936937, "grad_norm": 27.107633590698242, "learning_rate": 8.251831881191656e-06, "loss": 12.530646324157715, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.27684825100004673, "loss_total_v6": 0.7831653952598572, "step": 2342 }, { "epoch": 0.5403675675675675, "grad_norm": 22.99418830871582, "learning_rate": 8.250301861866734e-06, "loss": 12.551326751708984, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.2805974595248699, "loss_total_v6": 0.7844579070806503, "step": 2343 }, { "epoch": 0.5405981981981982, "grad_norm": 23.67787742614746, "learning_rate": 8.248771315270391e-06, "loss": 12.481971740722656, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.28000115789473057, "loss_total_v6": 0.7801232039928436, "step": 2344 }, { "epoch": 0.5408288288288289, "grad_norm": 24.44037437438965, "learning_rate": 8.247240241650918e-06, "loss": 13.227231979370117, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.32311615720391273, "loss_total_v6": 0.8267019689083099, "step": 2345 }, { "epoch": 0.5410594594594594, "grad_norm": 23.63628387451172, "learning_rate": 8.24570864125669e-06, "loss": 12.688274383544922, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.2827937714755535, "loss_total_v6": 0.7930171564221382, "step": 2346 }, { "epoch": 0.5412900900900901, "grad_norm": 22.955097198486328, "learning_rate": 8.244176514336167e-06, "loss": 12.462217330932617, "loss_alignment": 1.01123046875, "loss_alignment_w": 0.505615234375, "loss_sft": 0.27505859173834324, "loss_total_v6": 0.7788885682821274, "step": 2347 }, { "epoch": 0.5415207207207208, "grad_norm": 22.378189086914062, "learning_rate": 8.242643861137891e-06, "loss": 12.756548881530762, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.29389690794050694, "loss_total_v6": 0.7972843423485756, "step": 2348 }, { "epoch": 0.5417513513513513, "grad_norm": 22.586074829101562, "learning_rate": 8.241110681910495e-06, "loss": 12.763374328613281, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.28754851780831814, "loss_total_v6": 0.7977108657360077, "step": 2349 }, { "epoch": 0.541981981981982, "grad_norm": 27.22332000732422, "learning_rate": 8.239576976902694e-06, "loss": 12.632740020751953, "loss_alignment": 0.9833984375, "loss_alignment_w": 0.49169921875, "loss_sft": 0.29651952162384987, "loss_total_v6": 0.7895462214946747, "step": 2350 }, { "epoch": 0.5422126126126127, "grad_norm": 24.96503448486328, "learning_rate": 8.238042746363291e-06, "loss": 12.191535949707031, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.2616200875490904, "loss_total_v6": 0.761971041560173, "step": 2351 }, { "epoch": 0.5424432432432432, "grad_norm": 26.166213989257812, "learning_rate": 8.23650799054117e-06, "loss": 12.380226135253906, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.27823495492339134, "loss_total_v6": 0.7737641334533691, "step": 2352 }, { "epoch": 0.5426738738738739, "grad_norm": 23.007095336914062, "learning_rate": 8.234972709685303e-06, "loss": 12.563213348388672, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.2919757440686226, "loss_total_v6": 0.7852008268237114, "step": 2353 }, { "epoch": 0.5429045045045046, "grad_norm": 21.742084503173828, "learning_rate": 8.233436904044748e-06, "loss": 12.18192195892334, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.26144641637802124, "loss_total_v6": 0.7613701373338699, "step": 2354 }, { "epoch": 0.5431351351351351, "grad_norm": 23.17499351501465, "learning_rate": 8.231900573868645e-06, "loss": 12.643000602722168, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2870594374835491, "loss_total_v6": 0.7901875153183937, "step": 2355 }, { "epoch": 0.5433657657657658, "grad_norm": 25.1129150390625, "learning_rate": 8.230363719406223e-06, "loss": 12.605388641357422, "loss_alignment": 1.021484375, "loss_alignment_w": 0.5107421875, "loss_sft": 0.2769573051482439, "loss_total_v6": 0.7878368273377419, "step": 2356 }, { "epoch": 0.5435963963963963, "grad_norm": 20.355228424072266, "learning_rate": 8.228826340906795e-06, "loss": 12.369133949279785, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2640986815094948, "loss_total_v6": 0.7730708494782448, "step": 2357 }, { "epoch": 0.543827027027027, "grad_norm": 24.364194869995117, "learning_rate": 8.227288438619754e-06, "loss": 13.208642959594727, "loss_alignment": 1.0419921875, "loss_alignment_w": 0.52099609375, "loss_sft": 0.30460512451827526, "loss_total_v6": 0.8255401700735092, "step": 2358 }, { "epoch": 0.5440576576576577, "grad_norm": 23.955379486083984, "learning_rate": 8.225750012794585e-06, "loss": 12.686933517456055, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2897290140390396, "loss_total_v6": 0.7929333597421646, "step": 2359 }, { "epoch": 0.5442882882882882, "grad_norm": 24.963346481323242, "learning_rate": 8.224211063680854e-06, "loss": 12.935588836669922, "loss_alignment": 1.01318359375, "loss_alignment_w": 0.506591796875, "loss_sft": 0.30133309587836266, "loss_total_v6": 0.8084742501378059, "step": 2360 }, { "epoch": 0.5445189189189189, "grad_norm": 22.407432556152344, "learning_rate": 8.222671591528212e-06, "loss": 13.018898010253906, "loss_alignment": 0.990234375, "loss_alignment_w": 0.4951171875, "loss_sft": 0.31912846863269806, "loss_total_v6": 0.8136810883879662, "step": 2361 }, { "epoch": 0.5447495495495496, "grad_norm": 22.957645416259766, "learning_rate": 8.2211315965864e-06, "loss": 12.996467590332031, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.31438497081398964, "loss_total_v6": 0.8122792616486549, "step": 2362 }, { "epoch": 0.5449801801801801, "grad_norm": 24.798046112060547, "learning_rate": 8.219591079105234e-06, "loss": 12.824224472045898, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.3075106739997864, "loss_total_v6": 0.8015139847993851, "step": 2363 }, { "epoch": 0.5452108108108108, "grad_norm": 23.233234405517578, "learning_rate": 8.218050039334624e-06, "loss": 13.054723739624023, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.31845320388674736, "loss_total_v6": 0.8159202113747597, "step": 2364 }, { "epoch": 0.5454414414414415, "grad_norm": 23.011962890625, "learning_rate": 8.216508477524558e-06, "loss": 12.280548095703125, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.25847051478922367, "loss_total_v6": 0.7675342485308647, "step": 2365 }, { "epoch": 0.545672072072072, "grad_norm": 24.519956588745117, "learning_rate": 8.214966393925115e-06, "loss": 12.715812683105469, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.29203740879893303, "loss_total_v6": 0.794738233089447, "step": 2366 }, { "epoch": 0.5459027027027027, "grad_norm": 23.70586395263672, "learning_rate": 8.213423788786453e-06, "loss": 12.70609188079834, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2851891126483679, "loss_total_v6": 0.7941307500004768, "step": 2367 }, { "epoch": 0.5461333333333334, "grad_norm": 24.311195373535156, "learning_rate": 8.211880662358818e-06, "loss": 12.728696823120117, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2807730920612812, "loss_total_v6": 0.7955435812473297, "step": 2368 }, { "epoch": 0.5463639639639639, "grad_norm": 22.571720123291016, "learning_rate": 8.21033701489254e-06, "loss": 12.706727981567383, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.2955742720514536, "loss_total_v6": 0.7941705137491226, "step": 2369 }, { "epoch": 0.5465945945945946, "grad_norm": 23.853025436401367, "learning_rate": 8.20879284663803e-06, "loss": 12.352188110351562, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.2671136800199747, "loss_total_v6": 0.7720117792487144, "step": 2370 }, { "epoch": 0.5468252252252253, "grad_norm": 24.93474769592285, "learning_rate": 8.20724815784579e-06, "loss": 12.56427001953125, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2785376850515604, "loss_total_v6": 0.7852668389678001, "step": 2371 }, { "epoch": 0.5470558558558558, "grad_norm": 26.622692108154297, "learning_rate": 8.2057029487664e-06, "loss": 12.925710678100586, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.3031114209443331, "loss_total_v6": 0.8078568950295448, "step": 2372 }, { "epoch": 0.5472864864864865, "grad_norm": 22.055002212524414, "learning_rate": 8.204157219650533e-06, "loss": 12.852153778076172, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.31267430633306503, "loss_total_v6": 0.8032596260309219, "step": 2373 }, { "epoch": 0.5475171171171171, "grad_norm": 23.135345458984375, "learning_rate": 8.202610970748934e-06, "loss": 12.366353988647461, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.276513434946537, "loss_total_v6": 0.7728971019387245, "step": 2374 }, { "epoch": 0.5477477477477477, "grad_norm": 22.948442459106445, "learning_rate": 8.20106420231244e-06, "loss": 12.990885734558105, "loss_alignment": 1.01953125, "loss_alignment_w": 0.509765625, "loss_sft": 0.3035075217485428, "loss_total_v6": 0.8119303658604622, "step": 2375 }, { "epoch": 0.5479783783783784, "grad_norm": 26.093082427978516, "learning_rate": 8.199516914591975e-06, "loss": 13.130887031555176, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.3181932792067528, "loss_total_v6": 0.8206804543733597, "step": 2376 }, { "epoch": 0.548209009009009, "grad_norm": 23.473424911499023, "learning_rate": 8.197969107838541e-06, "loss": 12.335997581481934, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.27450938895344734, "loss_total_v6": 0.7709998562932014, "step": 2377 }, { "epoch": 0.5484396396396396, "grad_norm": 22.7973690032959, "learning_rate": 8.19642078230323e-06, "loss": 12.664751052856445, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2874118313193321, "loss_total_v6": 0.7915469631552696, "step": 2378 }, { "epoch": 0.5486702702702703, "grad_norm": 26.0163516998291, "learning_rate": 8.19487193823721e-06, "loss": 13.248428344726562, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.33042238280177116, "loss_total_v6": 0.8280267640948296, "step": 2379 }, { "epoch": 0.5489009009009009, "grad_norm": 27.719518661499023, "learning_rate": 8.19332257589174e-06, "loss": 12.742931365966797, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.29200819320976734, "loss_total_v6": 0.7964332401752472, "step": 2380 }, { "epoch": 0.5491315315315315, "grad_norm": 23.968481063842773, "learning_rate": 8.191772695518164e-06, "loss": 13.054704666137695, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.3120128605514765, "loss_total_v6": 0.8159190788865089, "step": 2381 }, { "epoch": 0.5493621621621622, "grad_norm": 22.740663528442383, "learning_rate": 8.190222297367901e-06, "loss": 12.588537216186523, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2834266237914562, "loss_total_v6": 0.7867835760116577, "step": 2382 }, { "epoch": 0.5495927927927928, "grad_norm": 24.966005325317383, "learning_rate": 8.188671381692466e-06, "loss": 12.716075897216797, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.2965095490217209, "loss_total_v6": 0.7947547435760498, "step": 2383 }, { "epoch": 0.5498234234234234, "grad_norm": 28.036115646362305, "learning_rate": 8.18711994874345e-06, "loss": 12.891624450683594, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.30906818993389606, "loss_total_v6": 0.8057264909148216, "step": 2384 }, { "epoch": 0.550054054054054, "grad_norm": 24.25111198425293, "learning_rate": 8.185567998772528e-06, "loss": 12.736454010009766, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.28986387699842453, "loss_total_v6": 0.7960284352302551, "step": 2385 }, { "epoch": 0.5502846846846847, "grad_norm": 26.408681869506836, "learning_rate": 8.184015532031465e-06, "loss": 13.641181945800781, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.35315365716814995, "loss_total_v6": 0.8525738343596458, "step": 2386 }, { "epoch": 0.5505153153153153, "grad_norm": 26.404653549194336, "learning_rate": 8.1824625487721e-06, "loss": 12.548988342285156, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.27930696681141853, "loss_total_v6": 0.7843118384480476, "step": 2387 }, { "epoch": 0.550745945945946, "grad_norm": 25.81954574584961, "learning_rate": 8.18090904924637e-06, "loss": 12.446556091308594, "loss_alignment": 0.9873046875, "loss_alignment_w": 0.49365234375, "loss_sft": 0.2840437777340412, "loss_total_v6": 0.7779097259044647, "step": 2388 }, { "epoch": 0.5509765765765766, "grad_norm": 21.9201717376709, "learning_rate": 8.179355033706279e-06, "loss": 12.944705963134766, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.31198908761143684, "loss_total_v6": 0.8090441226959229, "step": 2389 }, { "epoch": 0.5512072072072072, "grad_norm": 23.537967681884766, "learning_rate": 8.177800502403928e-06, "loss": 12.365514755249023, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.2841972764581442, "loss_total_v6": 0.772844709455967, "step": 2390 }, { "epoch": 0.5514378378378378, "grad_norm": 20.59288787841797, "learning_rate": 8.176245455591493e-06, "loss": 13.368730545043945, "loss_alignment": 1.025390625, "loss_alignment_w": 0.5126953125, "loss_sft": 0.32391849160194397, "loss_total_v6": 0.8355456739664078, "step": 2391 }, { "epoch": 0.5516684684684685, "grad_norm": 24.623382568359375, "learning_rate": 8.174689893521239e-06, "loss": 13.38542366027832, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.3307753801345825, "loss_total_v6": 0.8365889936685562, "step": 2392 }, { "epoch": 0.5518990990990991, "grad_norm": 21.71136474609375, "learning_rate": 8.173133816445513e-06, "loss": 12.595514297485352, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.29059186950325966, "loss_total_v6": 0.7872196659445763, "step": 2393 }, { "epoch": 0.5521297297297297, "grad_norm": 21.386564254760742, "learning_rate": 8.171577224616746e-06, "loss": 12.106033325195312, "loss_alignment": 0.9892578125, "loss_alignment_w": 0.49462890625, "loss_sft": 0.2631883881986141, "loss_total_v6": 0.75662712007761, "step": 2394 }, { "epoch": 0.5523603603603604, "grad_norm": 22.765287399291992, "learning_rate": 8.17002011828745e-06, "loss": 12.31547737121582, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.2766448799520731, "loss_total_v6": 0.769717387855053, "step": 2395 }, { "epoch": 0.552590990990991, "grad_norm": 22.500633239746094, "learning_rate": 8.168462497710227e-06, "loss": 12.942296028137207, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.3096106518059969, "loss_total_v6": 0.808893509209156, "step": 2396 }, { "epoch": 0.5528216216216216, "grad_norm": 24.4906005859375, "learning_rate": 8.166904363137752e-06, "loss": 12.901505470275879, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.30750374868512154, "loss_total_v6": 0.8063440918922424, "step": 2397 }, { "epoch": 0.5530522522522523, "grad_norm": 23.145328521728516, "learning_rate": 8.165345714822791e-06, "loss": 13.062658309936523, "loss_alignment": 1.02783203125, "loss_alignment_w": 0.513916015625, "loss_sft": 0.30248487927019596, "loss_total_v6": 0.8164161369204521, "step": 2398 }, { "epoch": 0.5532828828828829, "grad_norm": 24.502714157104492, "learning_rate": 8.163786553018192e-06, "loss": 12.956220626831055, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.3061322420835495, "loss_total_v6": 0.8097638338804245, "step": 2399 }, { "epoch": 0.5535135135135135, "grad_norm": 23.342687606811523, "learning_rate": 8.162226877976886e-06, "loss": 13.583108901977539, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.35106536000967026, "loss_total_v6": 0.8489443734288216, "step": 2400 }, { "epoch": 0.5537441441441442, "grad_norm": 23.84000587463379, "learning_rate": 8.160666689951888e-06, "loss": 12.78958511352539, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.29232997074723244, "loss_total_v6": 0.7993490248918533, "step": 2401 }, { "epoch": 0.5539747747747747, "grad_norm": 23.586027145385742, "learning_rate": 8.159105989196294e-06, "loss": 13.188050270080566, "loss_alignment": 1.041015625, "loss_alignment_w": 0.5205078125, "loss_sft": 0.3045388050377369, "loss_total_v6": 0.8242531567811966, "step": 2402 }, { "epoch": 0.5542054054054054, "grad_norm": 26.51287841796875, "learning_rate": 8.157544775963285e-06, "loss": 12.671040534973145, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.29294710606336594, "loss_total_v6": 0.7919400185346603, "step": 2403 }, { "epoch": 0.5544360360360361, "grad_norm": 22.836103439331055, "learning_rate": 8.155983050506123e-06, "loss": 12.839213371276855, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.29784268140792847, "loss_total_v6": 0.8024508208036423, "step": 2404 }, { "epoch": 0.5546666666666666, "grad_norm": 21.74692153930664, "learning_rate": 8.154420813078155e-06, "loss": 12.44512939453125, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.27620313316583633, "loss_total_v6": 0.7778205722570419, "step": 2405 }, { "epoch": 0.5548972972972973, "grad_norm": 21.808364868164062, "learning_rate": 8.152858063932811e-06, "loss": 12.212997436523438, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.2644872833043337, "loss_total_v6": 0.7633123323321342, "step": 2406 }, { "epoch": 0.555127927927928, "grad_norm": 23.5383358001709, "learning_rate": 8.151294803323607e-06, "loss": 13.44573974609375, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.3426475264132023, "loss_total_v6": 0.8403586894273758, "step": 2407 }, { "epoch": 0.5553585585585585, "grad_norm": 22.870227813720703, "learning_rate": 8.149731031504136e-06, "loss": 12.331586837768555, "loss_alignment": 0.982421875, "loss_alignment_w": 0.4912109375, "loss_sft": 0.2800014764070511, "loss_total_v6": 0.7707241475582123, "step": 2408 }, { "epoch": 0.5555891891891892, "grad_norm": 22.884418487548828, "learning_rate": 8.148166748728075e-06, "loss": 12.300712585449219, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.27074764110147953, "loss_total_v6": 0.7687944918870926, "step": 2409 }, { "epoch": 0.5558198198198199, "grad_norm": 22.816068649291992, "learning_rate": 8.146601955249187e-06, "loss": 12.720104217529297, "loss_alignment": 0.990234375, "loss_alignment_w": 0.4951171875, "loss_sft": 0.2991416361182928, "loss_total_v6": 0.7950064912438393, "step": 2410 }, { "epoch": 0.5560504504504504, "grad_norm": 23.9337215423584, "learning_rate": 8.145036651321319e-06, "loss": 13.019331932067871, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.30638405680656433, "loss_total_v6": 0.8137082755565643, "step": 2411 }, { "epoch": 0.5562810810810811, "grad_norm": 26.111305236816406, "learning_rate": 8.143470837198394e-06, "loss": 12.371543884277344, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2701697889715433, "loss_total_v6": 0.773221530020237, "step": 2412 }, { "epoch": 0.5565117117117118, "grad_norm": 24.033580780029297, "learning_rate": 8.141904513134425e-06, "loss": 12.60104751586914, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.28791642002761364, "loss_total_v6": 0.7875654920935631, "step": 2413 }, { "epoch": 0.5567423423423423, "grad_norm": 23.70138168334961, "learning_rate": 8.140337679383502e-06, "loss": 13.194230079650879, "loss_alignment": 1.0087890625, "loss_alignment_w": 0.50439453125, "loss_sft": 0.32087042182683945, "loss_total_v6": 0.8246393501758575, "step": 2414 }, { "epoch": 0.556972972972973, "grad_norm": 26.886808395385742, "learning_rate": 8.138770336199803e-06, "loss": 12.659930229187012, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.29676929116249084, "loss_total_v6": 0.7912456095218658, "step": 2415 }, { "epoch": 0.5572036036036037, "grad_norm": 21.537109375, "learning_rate": 8.137202483837584e-06, "loss": 12.235082626342773, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.2621749136596918, "loss_total_v6": 0.7646926119923592, "step": 2416 }, { "epoch": 0.5574342342342342, "grad_norm": 25.026905059814453, "learning_rate": 8.135634122551188e-06, "loss": 12.692002296447754, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.28838259913027287, "loss_total_v6": 0.7932501584291458, "step": 2417 }, { "epoch": 0.5576648648648649, "grad_norm": 20.645496368408203, "learning_rate": 8.134065252595035e-06, "loss": 12.346173286437988, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2679126765578985, "loss_total_v6": 0.7716358080506325, "step": 2418 }, { "epoch": 0.5578954954954954, "grad_norm": 24.85247039794922, "learning_rate": 8.132495874223633e-06, "loss": 12.871200561523438, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.305060300976038, "loss_total_v6": 0.8044499829411507, "step": 2419 }, { "epoch": 0.5581261261261261, "grad_norm": 24.954496383666992, "learning_rate": 8.13092598769157e-06, "loss": 12.894838333129883, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2979622669517994, "loss_total_v6": 0.8059273734688759, "step": 2420 }, { "epoch": 0.5583567567567568, "grad_norm": 23.780366897583008, "learning_rate": 8.129355593253513e-06, "loss": 12.845954895019531, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.3044438399374485, "loss_total_v6": 0.8028721809387207, "step": 2421 }, { "epoch": 0.5585873873873873, "grad_norm": 23.633663177490234, "learning_rate": 8.127784691164223e-06, "loss": 12.336387634277344, "loss_alignment": 0.9775390625, "loss_alignment_w": 0.48876953125, "loss_sft": 0.2818427234888077, "loss_total_v6": 0.7710242345929146, "step": 2422 }, { "epoch": 0.558818018018018, "grad_norm": 23.81949806213379, "learning_rate": 8.126213281678527e-06, "loss": 12.76591682434082, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.29669489338994026, "loss_total_v6": 0.7978698089718819, "step": 2423 }, { "epoch": 0.5590486486486487, "grad_norm": 22.51995849609375, "learning_rate": 8.124641365051346e-06, "loss": 12.830486297607422, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.30056265369057655, "loss_total_v6": 0.8019054159522057, "step": 2424 }, { "epoch": 0.5592792792792792, "grad_norm": 24.90481948852539, "learning_rate": 8.123068941537681e-06, "loss": 12.812013626098633, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.29669199138879776, "loss_total_v6": 0.8007508218288422, "step": 2425 }, { "epoch": 0.5595099099099099, "grad_norm": 23.246593475341797, "learning_rate": 8.121496011392613e-06, "loss": 12.668779373168945, "loss_alignment": 0.9892578125, "loss_alignment_w": 0.49462890625, "loss_sft": 0.2977954000234604, "loss_total_v6": 0.7917986884713173, "step": 2426 }, { "epoch": 0.5597405405405406, "grad_norm": 22.355426788330078, "learning_rate": 8.119922574871305e-06, "loss": 12.74209976196289, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2891485393047333, "loss_total_v6": 0.7963811904191971, "step": 2427 }, { "epoch": 0.5599711711711711, "grad_norm": 21.630538940429688, "learning_rate": 8.118348632229006e-06, "loss": 12.307453155517578, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.2715504467487335, "loss_total_v6": 0.7692158594727516, "step": 2428 }, { "epoch": 0.5602018018018018, "grad_norm": 23.61184310913086, "learning_rate": 8.116774183721045e-06, "loss": 12.874787330627441, "loss_alignment": 0.990234375, "loss_alignment_w": 0.4951171875, "loss_sft": 0.30969434417784214, "loss_total_v6": 0.8046742156147957, "step": 2429 }, { "epoch": 0.5604324324324325, "grad_norm": 24.191097259521484, "learning_rate": 8.115199229602831e-06, "loss": 12.963296890258789, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.31673677265644073, "loss_total_v6": 0.8102060332894325, "step": 2430 }, { "epoch": 0.560663063063063, "grad_norm": 25.316574096679688, "learning_rate": 8.113623770129853e-06, "loss": 12.660987854003906, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.28974013216793537, "loss_total_v6": 0.7913117706775665, "step": 2431 }, { "epoch": 0.5608936936936937, "grad_norm": 26.71382713317871, "learning_rate": 8.112047805557693e-06, "loss": 13.821608543395996, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.3487290684133768, "loss_total_v6": 0.8638505563139915, "step": 2432 }, { "epoch": 0.5611243243243244, "grad_norm": 23.42742156982422, "learning_rate": 8.110471336142003e-06, "loss": 12.363115310668945, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.27779106982052326, "loss_total_v6": 0.7726946473121643, "step": 2433 }, { "epoch": 0.5613549549549549, "grad_norm": 23.54366111755371, "learning_rate": 8.108894362138521e-06, "loss": 12.488092422485352, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.2806278094649315, "loss_total_v6": 0.7805057391524315, "step": 2434 }, { "epoch": 0.5615855855855856, "grad_norm": 24.53725814819336, "learning_rate": 8.107316883803071e-06, "loss": 12.807557106018066, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.29716117680072784, "loss_total_v6": 0.8004723116755486, "step": 2435 }, { "epoch": 0.5618162162162162, "grad_norm": 23.874935150146484, "learning_rate": 8.105738901391553e-06, "loss": 13.536003112792969, "loss_alignment": 1.01318359375, "loss_alignment_w": 0.506591796875, "loss_sft": 0.3387217242270708, "loss_total_v6": 0.8460001796483994, "step": 2436 }, { "epoch": 0.5620468468468468, "grad_norm": 24.85726547241211, "learning_rate": 8.10416041515995e-06, "loss": 12.409910202026367, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.2788847666233778, "loss_total_v6": 0.7756193652749062, "step": 2437 }, { "epoch": 0.5622774774774775, "grad_norm": 21.218713760375977, "learning_rate": 8.10258142536433e-06, "loss": 13.047785758972168, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.3091541565954685, "loss_total_v6": 0.8154865875840187, "step": 2438 }, { "epoch": 0.5625081081081081, "grad_norm": 23.375490188598633, "learning_rate": 8.101001932260837e-06, "loss": 12.658605575561523, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.28818734362721443, "loss_total_v6": 0.7911628261208534, "step": 2439 }, { "epoch": 0.5627387387387387, "grad_norm": 26.181167602539062, "learning_rate": 8.099421936105702e-06, "loss": 12.61422348022461, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.2882058918476105, "loss_total_v6": 0.7883890047669411, "step": 2440 }, { "epoch": 0.5629693693693694, "grad_norm": 24.471162796020508, "learning_rate": 8.097841437155237e-06, "loss": 12.221620559692383, "loss_alignment": 0.990234375, "loss_alignment_w": 0.4951171875, "loss_sft": 0.268825626000762, "loss_total_v6": 0.7638512402772903, "step": 2441 }, { "epoch": 0.5632, "grad_norm": 24.015806198120117, "learning_rate": 8.096260435665831e-06, "loss": 12.818286895751953, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.2974654845893383, "loss_total_v6": 0.8011428862810135, "step": 2442 }, { "epoch": 0.5634306306306306, "grad_norm": 23.48256492614746, "learning_rate": 8.09467893189396e-06, "loss": 12.557727813720703, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.27721327170729637, "loss_total_v6": 0.7848579287528992, "step": 2443 }, { "epoch": 0.5636612612612613, "grad_norm": 24.86937713623047, "learning_rate": 8.093096926096178e-06, "loss": 13.348846435546875, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.32465935125947, "loss_total_v6": 0.8343029320240021, "step": 2444 }, { "epoch": 0.5638918918918919, "grad_norm": 25.099082946777344, "learning_rate": 8.091514418529124e-06, "loss": 12.610883712768555, "loss_alignment": 0.9853515625, "loss_alignment_w": 0.49267578125, "loss_sft": 0.29574862867593765, "loss_total_v6": 0.7881802767515182, "step": 2445 }, { "epoch": 0.5641225225225225, "grad_norm": 27.541208267211914, "learning_rate": 8.089931409449511e-06, "loss": 12.612003326416016, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.2928430363535881, "loss_total_v6": 0.7882501631975174, "step": 2446 }, { "epoch": 0.5643531531531532, "grad_norm": 24.242238998413086, "learning_rate": 8.088347899114146e-06, "loss": 12.560413360595703, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.2833016086369753, "loss_total_v6": 0.7850258648395538, "step": 2447 }, { "epoch": 0.5645837837837838, "grad_norm": 23.016376495361328, "learning_rate": 8.0867638877799e-06, "loss": 12.4415283203125, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.2790146470069885, "loss_total_v6": 0.7775955572724342, "step": 2448 }, { "epoch": 0.5648144144144144, "grad_norm": 23.330530166625977, "learning_rate": 8.085179375703745e-06, "loss": 12.541315078735352, "loss_alignment": 0.9873046875, "loss_alignment_w": 0.49365234375, "loss_sft": 0.28891339898109436, "loss_total_v6": 0.7838322073221207, "step": 2449 }, { "epoch": 0.565045045045045, "grad_norm": 23.98782730102539, "learning_rate": 8.083594363142717e-06, "loss": 12.869901657104492, "loss_alignment": 1.0224609375, "loss_alignment_w": 0.51123046875, "loss_sft": 0.29188720509409904, "loss_total_v6": 0.8043689057230949, "step": 2450 }, { "epoch": 0.5652756756756757, "grad_norm": 23.471874237060547, "learning_rate": 8.082008850353942e-06, "loss": 12.889713287353516, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.30694988928735256, "loss_total_v6": 0.8056070953607559, "step": 2451 }, { "epoch": 0.5655063063063063, "grad_norm": 24.935890197753906, "learning_rate": 8.080422837594627e-06, "loss": 12.366073608398438, "loss_alignment": 0.97607421875, "loss_alignment_w": 0.488037109375, "loss_sft": 0.28487301617860794, "loss_total_v6": 0.7728796154260635, "step": 2452 }, { "epoch": 0.5657369369369369, "grad_norm": 23.197528839111328, "learning_rate": 8.078836325122057e-06, "loss": 12.985588073730469, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.3085475340485573, "loss_total_v6": 0.8115992844104767, "step": 2453 }, { "epoch": 0.5659675675675676, "grad_norm": 22.399072647094727, "learning_rate": 8.077249313193602e-06, "loss": 12.28750991821289, "loss_alignment": 0.97412109375, "loss_alignment_w": 0.487060546875, "loss_sft": 0.28107667714357376, "loss_total_v6": 0.7679693847894669, "step": 2454 }, { "epoch": 0.5661981981981982, "grad_norm": 23.78049087524414, "learning_rate": 8.075661802066707e-06, "loss": 12.675451278686523, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.2869208734482527, "loss_total_v6": 0.7922157123684883, "step": 2455 }, { "epoch": 0.5664288288288288, "grad_norm": 23.581592559814453, "learning_rate": 8.074073791998907e-06, "loss": 13.014517784118652, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.30835666321218014, "loss_total_v6": 0.813407339155674, "step": 2456 }, { "epoch": 0.5666594594594595, "grad_norm": 24.7822208404541, "learning_rate": 8.072485283247807e-06, "loss": 12.1512451171875, "loss_alignment": 0.98583984375, "loss_alignment_w": 0.492919921875, "loss_sft": 0.26633454859256744, "loss_total_v6": 0.7594528570771217, "step": 2457 }, { "epoch": 0.5668900900900901, "grad_norm": 25.305828094482422, "learning_rate": 8.070896276071102e-06, "loss": 13.132360458374023, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.32232896611094475, "loss_total_v6": 0.8207725584506989, "step": 2458 }, { "epoch": 0.5671207207207207, "grad_norm": 25.291250228881836, "learning_rate": 8.069306770726563e-06, "loss": 12.819845199584961, "loss_alignment": 0.9873046875, "loss_alignment_w": 0.49365234375, "loss_sft": 0.30876287817955017, "loss_total_v6": 0.8012402877211571, "step": 2459 }, { "epoch": 0.5673513513513514, "grad_norm": 24.713510513305664, "learning_rate": 8.067716767472045e-06, "loss": 13.11775016784668, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.32133954390883446, "loss_total_v6": 0.8198594301939011, "step": 2460 }, { "epoch": 0.567581981981982, "grad_norm": 22.93992805480957, "learning_rate": 8.06612626656548e-06, "loss": 12.317445755004883, "loss_alignment": 0.99267578125, "loss_alignment_w": 0.496337890625, "loss_sft": 0.2747079227119684, "loss_total_v6": 0.769840382039547, "step": 2461 }, { "epoch": 0.5678126126126126, "grad_norm": 23.31140899658203, "learning_rate": 8.064535268264883e-06, "loss": 12.673672676086426, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.2988031730055809, "loss_total_v6": 0.7921045497059822, "step": 2462 }, { "epoch": 0.5680432432432433, "grad_norm": 21.654788970947266, "learning_rate": 8.06294377282835e-06, "loss": 12.22848892211914, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.2696974612772465, "loss_total_v6": 0.7642805725336075, "step": 2463 }, { "epoch": 0.5682738738738738, "grad_norm": 25.072385787963867, "learning_rate": 8.061351780514057e-06, "loss": 12.721916198730469, "loss_alignment": 0.9873046875, "loss_alignment_w": 0.49365234375, "loss_sft": 0.3007656019181013, "loss_total_v6": 0.7951197996735573, "step": 2464 }, { "epoch": 0.5685045045045045, "grad_norm": 23.692148208618164, "learning_rate": 8.05975929158026e-06, "loss": 12.742815017700195, "loss_alignment": 0.984375, "loss_alignment_w": 0.4921875, "loss_sft": 0.30333816073834896, "loss_total_v6": 0.796425960958004, "step": 2465 }, { "epoch": 0.5687351351351352, "grad_norm": 25.277759552001953, "learning_rate": 8.058166306285297e-06, "loss": 13.003606796264648, "loss_alignment": 1.0078125, "loss_alignment_w": 0.50390625, "loss_sft": 0.30817833356559277, "loss_total_v6": 0.8127254545688629, "step": 2466 }, { "epoch": 0.5689657657657657, "grad_norm": 21.49736976623535, "learning_rate": 8.056572824887584e-06, "loss": 12.750539779663086, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.29815991409122944, "loss_total_v6": 0.7969087287783623, "step": 2467 }, { "epoch": 0.5691963963963964, "grad_norm": 21.347991943359375, "learning_rate": 8.054978847645623e-06, "loss": 12.25428581237793, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.2721489518880844, "loss_total_v6": 0.7658928409218788, "step": 2468 }, { "epoch": 0.5694270270270271, "grad_norm": 21.301515579223633, "learning_rate": 8.053384374817989e-06, "loss": 12.518896102905273, "loss_alignment": 0.9892578125, "loss_alignment_w": 0.49462890625, "loss_sft": 0.2872985992580652, "loss_total_v6": 0.7824310436844826, "step": 2469 }, { "epoch": 0.5696576576576576, "grad_norm": 22.478757858276367, "learning_rate": 8.051789406663342e-06, "loss": 12.956611633300781, "loss_alignment": 1.0126953125, "loss_alignment_w": 0.50634765625, "loss_sft": 0.3037763424217701, "loss_total_v6": 0.809788279235363, "step": 2470 }, { "epoch": 0.5698882882882883, "grad_norm": 22.52838134765625, "learning_rate": 8.05019394344042e-06, "loss": 12.87199592590332, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.30372150614857674, "loss_total_v6": 0.8044997230172157, "step": 2471 }, { "epoch": 0.570118918918919, "grad_norm": 21.46530532836914, "learning_rate": 8.048597985408047e-06, "loss": 12.893501281738281, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.29937404207885265, "loss_total_v6": 0.8058437928557396, "step": 2472 }, { "epoch": 0.5703495495495495, "grad_norm": 22.230224609375, "learning_rate": 8.047001532825119e-06, "loss": 12.41061782836914, "loss_alignment": 0.9755859375, "loss_alignment_w": 0.48779296875, "loss_sft": 0.2884962987154722, "loss_total_v6": 0.7756636440753937, "step": 2473 }, { "epoch": 0.5705801801801802, "grad_norm": 22.45844841003418, "learning_rate": 8.045404585950618e-06, "loss": 12.689592361450195, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.291070107370615, "loss_total_v6": 0.793099544942379, "step": 2474 }, { "epoch": 0.5708108108108109, "grad_norm": 23.754793167114258, "learning_rate": 8.043807145043604e-06, "loss": 12.347822189331055, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.266428804025054, "loss_total_v6": 0.7717388570308685, "step": 2475 }, { "epoch": 0.5710414414414414, "grad_norm": 21.93956184387207, "learning_rate": 8.042209210363216e-06, "loss": 12.631210327148438, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.29091552272439003, "loss_total_v6": 0.7894506901502609, "step": 2476 }, { "epoch": 0.5712720720720721, "grad_norm": 24.198699951171875, "learning_rate": 8.040610782168678e-06, "loss": 13.262566566467285, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.3230510242283344, "loss_total_v6": 0.8289104104042053, "step": 2477 }, { "epoch": 0.5715027027027026, "grad_norm": 23.924823760986328, "learning_rate": 8.039011860719288e-06, "loss": 12.553665161132812, "loss_alignment": 0.98486328125, "loss_alignment_w": 0.492431640625, "loss_sft": 0.29144001193344593, "loss_total_v6": 0.7846040949225426, "step": 2478 }, { "epoch": 0.5717333333333333, "grad_norm": 21.059722900390625, "learning_rate": 8.037412446274428e-06, "loss": 12.680669784545898, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2900394145399332, "loss_total_v6": 0.7925418466329575, "step": 2479 }, { "epoch": 0.571963963963964, "grad_norm": 24.713369369506836, "learning_rate": 8.035812539093557e-06, "loss": 12.791976928710938, "loss_alignment": 1.0302734375, "loss_alignment_w": 0.51513671875, "loss_sft": 0.2849263995885849, "loss_total_v6": 0.799498550593853, "step": 2480 }, { "epoch": 0.5721945945945945, "grad_norm": 25.238887786865234, "learning_rate": 8.034212139436217e-06, "loss": 12.43309211730957, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.2744131963700056, "loss_total_v6": 0.7770682275295258, "step": 2481 }, { "epoch": 0.5724252252252252, "grad_norm": 19.884843826293945, "learning_rate": 8.032611247562028e-06, "loss": 12.886201858520508, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.30839353427290916, "loss_total_v6": 0.8053876012563705, "step": 2482 }, { "epoch": 0.5726558558558559, "grad_norm": 24.401891708374023, "learning_rate": 8.031009863730692e-06, "loss": 13.306112289428711, "loss_alignment": 1.0361328125, "loss_alignment_w": 0.51806640625, "loss_sft": 0.3131994344294071, "loss_total_v6": 0.831632025539875, "step": 2483 }, { "epoch": 0.5728864864864864, "grad_norm": 23.06884002685547, "learning_rate": 8.029407988201986e-06, "loss": 12.932275772094727, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.30135506205260754, "loss_total_v6": 0.808267280459404, "step": 2484 }, { "epoch": 0.5731171171171171, "grad_norm": 22.972307205200195, "learning_rate": 8.027805621235771e-06, "loss": 12.066020011901855, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.2535464186221361, "loss_total_v6": 0.7541262805461884, "step": 2485 }, { "epoch": 0.5733477477477478, "grad_norm": 24.214868545532227, "learning_rate": 8.02620276309199e-06, "loss": 12.308175086975098, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.26926095969974995, "loss_total_v6": 0.7692609503865242, "step": 2486 }, { "epoch": 0.5735783783783783, "grad_norm": 25.244844436645508, "learning_rate": 8.024599414030657e-06, "loss": 12.993839263916016, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.3163416348397732, "loss_total_v6": 0.8121149465441704, "step": 2487 }, { "epoch": 0.573809009009009, "grad_norm": 25.700428009033203, "learning_rate": 8.022995574311876e-06, "loss": 13.120428085327148, "loss_alignment": 1.0244140625, "loss_alignment_w": 0.51220703125, "loss_sft": 0.3094371668994427, "loss_total_v6": 0.8200267627835274, "step": 2488 }, { "epoch": 0.5740396396396397, "grad_norm": 25.023645401000977, "learning_rate": 8.021391244195822e-06, "loss": 12.61750602722168, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.2900895569473505, "loss_total_v6": 0.7885941788554192, "step": 2489 }, { "epoch": 0.5742702702702702, "grad_norm": 23.712820053100586, "learning_rate": 8.019786423942752e-06, "loss": 12.377668380737305, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.2678974326699972, "loss_total_v6": 0.773604229092598, "step": 2490 }, { "epoch": 0.5745009009009009, "grad_norm": 24.678627014160156, "learning_rate": 8.018181113813009e-06, "loss": 12.913908004760742, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.3055322878062725, "loss_total_v6": 0.8071192502975464, "step": 2491 }, { "epoch": 0.5747315315315316, "grad_norm": 29.689056396484375, "learning_rate": 8.016575314067005e-06, "loss": 12.476507186889648, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.27622645534574986, "loss_total_v6": 0.779781736433506, "step": 2492 }, { "epoch": 0.5749621621621621, "grad_norm": 23.86417007446289, "learning_rate": 8.01496902496524e-06, "loss": 12.811566352844238, "loss_alignment": 1.015625, "loss_alignment_w": 0.5078125, "loss_sft": 0.2943142168223858, "loss_total_v6": 0.8007228821516037, "step": 2493 }, { "epoch": 0.5751927927927928, "grad_norm": 23.404577255249023, "learning_rate": 8.013362246768288e-06, "loss": 12.888612747192383, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.30434813909232616, "loss_total_v6": 0.8055383488535881, "step": 2494 }, { "epoch": 0.5754234234234235, "grad_norm": 24.581012725830078, "learning_rate": 8.011754979736804e-06, "loss": 12.746068000793457, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.2970259953290224, "loss_total_v6": 0.7966292575001717, "step": 2495 }, { "epoch": 0.575654054054054, "grad_norm": 26.591588973999023, "learning_rate": 8.010147224131523e-06, "loss": 12.759364128112793, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.3006646391004324, "loss_total_v6": 0.7974602803587914, "step": 2496 }, { "epoch": 0.5758846846846847, "grad_norm": 22.599592208862305, "learning_rate": 8.00853898021326e-06, "loss": 12.788381576538086, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.2996095307171345, "loss_total_v6": 0.7992738261818886, "step": 2497 }, { "epoch": 0.5761153153153153, "grad_norm": 23.194372177124023, "learning_rate": 8.006930248242905e-06, "loss": 12.97047233581543, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.30587850511074066, "loss_total_v6": 0.8106545507907867, "step": 2498 }, { "epoch": 0.5763459459459459, "grad_norm": 23.526185989379883, "learning_rate": 8.005321028481434e-06, "loss": 12.909704208374023, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.31334152445197105, "loss_total_v6": 0.8068565353751183, "step": 2499 }, { "epoch": 0.5765765765765766, "grad_norm": 24.365047454833984, "learning_rate": 8.003711321189895e-06, "loss": 12.718360900878906, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.2960876580327749, "loss_total_v6": 0.7948975041508675, "step": 2500 }, { "epoch": 0.5765765765765766, "eval_loss": 0.9512087106704712, "eval_loss_alignment": 1.0058192422945205, "eval_loss_alignment_w": 0.5029096211472602, "eval_loss_contrast": 0.0, "eval_loss_highlevel": 0.0, "eval_loss_sft": 0.4482990710073274, "eval_loss_total_v6": 0.9512086888630641, "eval_loss_wm_recon": 0.0, "eval_loss_wm_recon_w": 0.0, "eval_runtime": 275.428, "eval_samples_per_second": 6.336, "eval_steps_per_second": 0.795, "step": 2500 }, { "epoch": 0.5768072072072072, "grad_norm": 22.07613182067871, "learning_rate": 8.002101126629422e-06, "loss": 12.695425033569336, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.2900003455579281, "loss_total_v6": 0.7934641093015671, "step": 2501 }, { "epoch": 0.5770378378378378, "grad_norm": 22.313623428344727, "learning_rate": 8.00049044506122e-06, "loss": 12.852815628051758, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.29892172291874886, "loss_total_v6": 0.8033009991049767, "step": 2502 }, { "epoch": 0.5772684684684685, "grad_norm": 27.743337631225586, "learning_rate": 7.99887927674658e-06, "loss": 13.241842269897461, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.32399873808026314, "loss_total_v6": 0.8276150971651077, "step": 2503 }, { "epoch": 0.5774990990990991, "grad_norm": 24.174421310424805, "learning_rate": 7.997267621946871e-06, "loss": 12.194377899169922, "loss_alignment": 0.97705078125, "loss_alignment_w": 0.488525390625, "loss_sft": 0.27447773702442646, "loss_total_v6": 0.7621486335992813, "step": 2504 }, { "epoch": 0.5777297297297297, "grad_norm": 23.49244499206543, "learning_rate": 7.995655480923536e-06, "loss": 12.502009391784668, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.2877690326422453, "loss_total_v6": 0.7813756018877029, "step": 2505 }, { "epoch": 0.5779603603603604, "grad_norm": 23.564329147338867, "learning_rate": 7.994042853938102e-06, "loss": 12.102621078491211, "loss_alignment": 0.97705078125, "loss_alignment_w": 0.488525390625, "loss_sft": 0.2686360627412796, "loss_total_v6": 0.7564137876033783, "step": 2506 }, { "epoch": 0.578190990990991, "grad_norm": 23.108007431030273, "learning_rate": 7.992429741252174e-06, "loss": 12.809503555297852, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.3001972772181034, "loss_total_v6": 0.8005940318107605, "step": 2507 }, { "epoch": 0.5784216216216216, "grad_norm": 24.069026947021484, "learning_rate": 7.990816143127431e-06, "loss": 12.565972328186035, "loss_alignment": 0.9912109375, "loss_alignment_w": 0.49560546875, "loss_sft": 0.28993567638099194, "loss_total_v6": 0.7853732854127884, "step": 2508 }, { "epoch": 0.5786522522522523, "grad_norm": 25.162128448486328, "learning_rate": 7.989202059825639e-06, "loss": 12.620599746704102, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.28201259300112724, "loss_total_v6": 0.7887875139713287, "step": 2509 }, { "epoch": 0.5788828828828829, "grad_norm": 26.90616798400879, "learning_rate": 7.987587491608636e-06, "loss": 12.986029624938965, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.3160671293735504, "loss_total_v6": 0.8116268366575241, "step": 2510 }, { "epoch": 0.5791135135135135, "grad_norm": 23.63249969482422, "learning_rate": 7.985972438738343e-06, "loss": 12.317607879638672, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.2705828882753849, "loss_total_v6": 0.7698504701256752, "step": 2511 }, { "epoch": 0.5793441441441441, "grad_norm": 22.936555862426758, "learning_rate": 7.984356901476755e-06, "loss": 12.319486618041992, "loss_alignment": 1.029296875, "loss_alignment_w": 0.5146484375, "loss_sft": 0.2561586797237396, "loss_total_v6": 0.7699678838253021, "step": 2512 }, { "epoch": 0.5795747747747748, "grad_norm": 20.77228355407715, "learning_rate": 7.982740880085948e-06, "loss": 12.77545166015625, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.30014416947960854, "loss_total_v6": 0.7984656989574432, "step": 2513 }, { "epoch": 0.5798054054054054, "grad_norm": 22.952171325683594, "learning_rate": 7.981124374828079e-06, "loss": 12.968711853027344, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.3023657612502575, "loss_total_v6": 0.810544453561306, "step": 2514 }, { "epoch": 0.580036036036036, "grad_norm": 23.44701385498047, "learning_rate": 7.97950738596538e-06, "loss": 12.906532287597656, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2967247348278761, "loss_total_v6": 0.8066582381725311, "step": 2515 }, { "epoch": 0.5802666666666667, "grad_norm": 25.248340606689453, "learning_rate": 7.977889913760163e-06, "loss": 12.80632495880127, "loss_alignment": 1.0185546875, "loss_alignment_w": 0.50927734375, "loss_sft": 0.2914689350873232, "loss_total_v6": 0.8003953397274017, "step": 2516 }, { "epoch": 0.5804972972972973, "grad_norm": 24.787525177001953, "learning_rate": 7.976271958474816e-06, "loss": 12.154583930969238, "loss_alignment": 0.9853515625, "loss_alignment_w": 0.49267578125, "loss_sft": 0.26755031384527683, "loss_total_v6": 0.7596615105867386, "step": 2517 }, { "epoch": 0.5807279279279279, "grad_norm": 26.35141372680664, "learning_rate": 7.974653520371813e-06, "loss": 13.351390838623047, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.3283279202878475, "loss_total_v6": 0.8344619646668434, "step": 2518 }, { "epoch": 0.5809585585585586, "grad_norm": 21.719648361206055, "learning_rate": 7.973034599713693e-06, "loss": 13.11036205291748, "loss_alignment": 0.99267578125, "loss_alignment_w": 0.496337890625, "loss_sft": 0.32090830616652966, "loss_total_v6": 0.8193976432085037, "step": 2519 }, { "epoch": 0.5811891891891892, "grad_norm": 24.986408233642578, "learning_rate": 7.971415196763088e-06, "loss": 12.311662673950195, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.2702418304979801, "loss_total_v6": 0.7694789320230484, "step": 2520 }, { "epoch": 0.5814198198198198, "grad_norm": 23.620201110839844, "learning_rate": 7.969795311782698e-06, "loss": 12.308221817016602, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.273841455578804, "loss_total_v6": 0.7692638263106346, "step": 2521 }, { "epoch": 0.5816504504504505, "grad_norm": 24.4238224029541, "learning_rate": 7.968174945035301e-06, "loss": 12.634696960449219, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.27926203794777393, "loss_total_v6": 0.7896685525774956, "step": 2522 }, { "epoch": 0.581881081081081, "grad_norm": 28.542665481567383, "learning_rate": 7.966554096783763e-06, "loss": 13.143948554992676, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.3168428875505924, "loss_total_v6": 0.8214967995882034, "step": 2523 }, { "epoch": 0.5821117117117117, "grad_norm": 24.71877670288086, "learning_rate": 7.964932767291019e-06, "loss": 13.080839157104492, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.31659113615751266, "loss_total_v6": 0.817552462220192, "step": 2524 }, { "epoch": 0.5823423423423424, "grad_norm": 24.15839195251465, "learning_rate": 7.963310956820085e-06, "loss": 12.72781753540039, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.293535428121686, "loss_total_v6": 0.7954885810613632, "step": 2525 }, { "epoch": 0.582572972972973, "grad_norm": 26.292186737060547, "learning_rate": 7.961688665634054e-06, "loss": 12.345283508300781, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.26846737414598465, "loss_total_v6": 0.7715801671147346, "step": 2526 }, { "epoch": 0.5828036036036036, "grad_norm": 23.699970245361328, "learning_rate": 7.960065893996099e-06, "loss": 13.02673053741455, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.3163679242134094, "loss_total_v6": 0.8141706585884094, "step": 2527 }, { "epoch": 0.5830342342342343, "grad_norm": 23.155794143676758, "learning_rate": 7.958442642169469e-06, "loss": 12.569684982299805, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.2910374850034714, "loss_total_v6": 0.7856053411960602, "step": 2528 }, { "epoch": 0.5832648648648648, "grad_norm": 26.474750518798828, "learning_rate": 7.956818910417489e-06, "loss": 12.294807434082031, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2609487120062113, "loss_total_v6": 0.7684255167841911, "step": 2529 }, { "epoch": 0.5834954954954955, "grad_norm": 25.281335830688477, "learning_rate": 7.95519469900357e-06, "loss": 12.367498397827148, "loss_alignment": 0.96875, "loss_alignment_w": 0.484375, "loss_sft": 0.2882121279835701, "loss_total_v6": 0.7729686051607132, "step": 2530 }, { "epoch": 0.5837261261261262, "grad_norm": 27.338754653930664, "learning_rate": 7.953570008191192e-06, "loss": 12.361821174621582, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.272583294659853, "loss_total_v6": 0.7726138234138489, "step": 2531 }, { "epoch": 0.5839567567567567, "grad_norm": 24.15619659423828, "learning_rate": 7.951944838243916e-06, "loss": 12.564117431640625, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.28853799775242805, "loss_total_v6": 0.7852573469281197, "step": 2532 }, { "epoch": 0.5841873873873874, "grad_norm": 23.347612380981445, "learning_rate": 7.950319189425384e-06, "loss": 12.563008308410645, "loss_alignment": 0.9912109375, "loss_alignment_w": 0.49560546875, "loss_sft": 0.2897961847484112, "loss_total_v6": 0.7851880192756653, "step": 2533 }, { "epoch": 0.5844180180180181, "grad_norm": 23.24466896057129, "learning_rate": 7.948693061999307e-06, "loss": 12.170167922973633, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.26104746386408806, "loss_total_v6": 0.7606354728341103, "step": 2534 }, { "epoch": 0.5846486486486486, "grad_norm": 23.543977737426758, "learning_rate": 7.947066456229485e-06, "loss": 12.16127872467041, "loss_alignment": 1.01171875, "loss_alignment_w": 0.505859375, "loss_sft": 0.2540832180529833, "loss_total_v6": 0.7600799202919006, "step": 2535 }, { "epoch": 0.5848792792792793, "grad_norm": 29.588401794433594, "learning_rate": 7.945439372379782e-06, "loss": 12.623966217041016, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.28934881277382374, "loss_total_v6": 0.7889978662133217, "step": 2536 }, { "epoch": 0.58510990990991, "grad_norm": 22.74664306640625, "learning_rate": 7.943811810714155e-06, "loss": 12.359127044677734, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.2750394530594349, "loss_total_v6": 0.7724454551935196, "step": 2537 }, { "epoch": 0.5853405405405405, "grad_norm": 23.858983993530273, "learning_rate": 7.942183771496628e-06, "loss": 12.620896339416504, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.2918730266392231, "loss_total_v6": 0.7888059988617897, "step": 2538 }, { "epoch": 0.5855711711711712, "grad_norm": 24.3796443939209, "learning_rate": 7.940555254991305e-06, "loss": 12.571961402893066, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.28977594152092934, "loss_total_v6": 0.7857476100325584, "step": 2539 }, { "epoch": 0.5858018018018017, "grad_norm": 24.751522064208984, "learning_rate": 7.938926261462366e-06, "loss": 12.445175170898438, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.2780523709952831, "loss_total_v6": 0.7778234854340553, "step": 2540 }, { "epoch": 0.5860324324324324, "grad_norm": 26.295562744140625, "learning_rate": 7.937296791174072e-06, "loss": 12.867053985595703, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.3015358578413725, "loss_total_v6": 0.8041909039020538, "step": 2541 }, { "epoch": 0.5862630630630631, "grad_norm": 22.926454544067383, "learning_rate": 7.935666844390761e-06, "loss": 12.249946594238281, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.26246304251253605, "loss_total_v6": 0.7656216397881508, "step": 2542 }, { "epoch": 0.5864936936936936, "grad_norm": 24.087831497192383, "learning_rate": 7.934036421376845e-06, "loss": 12.752349853515625, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.30210305750370026, "loss_total_v6": 0.797021858394146, "step": 2543 }, { "epoch": 0.5867243243243243, "grad_norm": 22.574098587036133, "learning_rate": 7.932405522396814e-06, "loss": 12.350940704345703, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.27065213210880756, "loss_total_v6": 0.7719338238239288, "step": 2544 }, { "epoch": 0.586954954954955, "grad_norm": 26.13311767578125, "learning_rate": 7.930774147715235e-06, "loss": 12.686296463012695, "loss_alignment": 0.98583984375, "loss_alignment_w": 0.492919921875, "loss_sft": 0.298752861097455, "loss_total_v6": 0.7928934842348099, "step": 2545 }, { "epoch": 0.5871855855855855, "grad_norm": 24.536808013916016, "learning_rate": 7.929142297596757e-06, "loss": 12.844447135925293, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2941261827945709, "loss_total_v6": 0.8027779161930084, "step": 2546 }, { "epoch": 0.5874162162162162, "grad_norm": 22.163103103637695, "learning_rate": 7.927509972306103e-06, "loss": 12.759025573730469, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.29147283360362053, "loss_total_v6": 0.7974390462040901, "step": 2547 }, { "epoch": 0.5876468468468469, "grad_norm": 23.885658264160156, "learning_rate": 7.925877172108067e-06, "loss": 12.867090225219727, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.3035980612039566, "loss_total_v6": 0.8041931465268135, "step": 2548 }, { "epoch": 0.5878774774774774, "grad_norm": 22.407432556152344, "learning_rate": 7.92424389726753e-06, "loss": 12.662355422973633, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.2891236450523138, "loss_total_v6": 0.791397213935852, "step": 2549 }, { "epoch": 0.5881081081081081, "grad_norm": 23.99271011352539, "learning_rate": 7.922610148049445e-06, "loss": 12.980376243591309, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.3124026618897915, "loss_total_v6": 0.8112735077738762, "step": 2550 }, { "epoch": 0.5883387387387388, "grad_norm": 24.949199676513672, "learning_rate": 7.92097592471884e-06, "loss": 13.056689262390137, "loss_alignment": 0.96484375, "loss_alignment_w": 0.482421875, "loss_sft": 0.33328548446297646, "loss_total_v6": 0.816043071448803, "step": 2551 }, { "epoch": 0.5885693693693693, "grad_norm": 23.410903930664062, "learning_rate": 7.919341227540828e-06, "loss": 12.5101318359375, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.28227998688817024, "loss_total_v6": 0.7818832546472549, "step": 2552 }, { "epoch": 0.5888, "grad_norm": 23.330825805664062, "learning_rate": 7.917706056780588e-06, "loss": 13.206002235412598, "loss_alignment": 0.9892578125, "loss_alignment_w": 0.49462890625, "loss_sft": 0.32967811450362206, "loss_total_v6": 0.8253751322627068, "step": 2553 }, { "epoch": 0.5890306306306307, "grad_norm": 24.782394409179688, "learning_rate": 7.916070412703383e-06, "loss": 12.929595947265625, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.30925944074988365, "loss_total_v6": 0.8080997467041016, "step": 2554 }, { "epoch": 0.5892612612612612, "grad_norm": 21.778919219970703, "learning_rate": 7.914434295574552e-06, "loss": 13.184329986572266, "loss_alignment": 0.97802734375, "loss_alignment_w": 0.489013671875, "loss_sft": 0.3359530232846737, "loss_total_v6": 0.8240206316113472, "step": 2555 }, { "epoch": 0.5894918918918919, "grad_norm": 24.984281539916992, "learning_rate": 7.912797705659508e-06, "loss": 12.669254302978516, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.293644193559885, "loss_total_v6": 0.791828416287899, "step": 2556 }, { "epoch": 0.5897225225225226, "grad_norm": 22.439022064208984, "learning_rate": 7.911160643223742e-06, "loss": 12.933856964111328, "loss_alignment": 0.974609375, "loss_alignment_w": 0.4873046875, "loss_sft": 0.3220378905534744, "loss_total_v6": 0.8083660081028938, "step": 2557 }, { "epoch": 0.5899531531531531, "grad_norm": 25.829914093017578, "learning_rate": 7.909523108532824e-06, "loss": 12.486011505126953, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.28043672628700733, "loss_total_v6": 0.7803757041692734, "step": 2558 }, { "epoch": 0.5901837837837838, "grad_norm": 21.44267463684082, "learning_rate": 7.907885101852397e-06, "loss": 12.514973640441895, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2746480293571949, "loss_total_v6": 0.7821858748793602, "step": 2559 }, { "epoch": 0.5904144144144144, "grad_norm": 24.182893753051758, "learning_rate": 7.906246623448184e-06, "loss": 13.102055549621582, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.321411419659853, "loss_total_v6": 0.8188784569501877, "step": 2560 }, { "epoch": 0.590645045045045, "grad_norm": 68.76834869384766, "learning_rate": 7.90460767358598e-06, "loss": 12.848637580871582, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.30900601483881474, "loss_total_v6": 0.8030398413538933, "step": 2561 }, { "epoch": 0.5908756756756757, "grad_norm": 22.505952835083008, "learning_rate": 7.902968252531661e-06, "loss": 12.76508903503418, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.3029145337641239, "loss_total_v6": 0.7978180944919586, "step": 2562 }, { "epoch": 0.5911063063063063, "grad_norm": 24.86307144165039, "learning_rate": 7.90132836055118e-06, "loss": 13.095085144042969, "loss_alignment": 1.037109375, "loss_alignment_w": 0.5185546875, "loss_sft": 0.30001017823815346, "loss_total_v6": 0.8184428215026855, "step": 2563 }, { "epoch": 0.5913369369369369, "grad_norm": 29.86951446533203, "learning_rate": 7.899687997910558e-06, "loss": 12.728178977966309, "loss_alignment": 1.017578125, "loss_alignment_w": 0.5087890625, "loss_sft": 0.2864474207162857, "loss_total_v6": 0.7955111637711525, "step": 2564 }, { "epoch": 0.5915675675675676, "grad_norm": 23.343902587890625, "learning_rate": 7.898047164875903e-06, "loss": 12.736032485961914, "loss_alignment": 0.98388671875, "loss_alignment_w": 0.491943359375, "loss_sft": 0.30431812442839146, "loss_total_v6": 0.7960020676255226, "step": 2565 }, { "epoch": 0.5917981981981982, "grad_norm": 24.025157928466797, "learning_rate": 7.896405861713393e-06, "loss": 12.700026512145996, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.2958268281072378, "loss_total_v6": 0.7937516644597054, "step": 2566 }, { "epoch": 0.5920288288288288, "grad_norm": 23.92403793334961, "learning_rate": 7.894764088689287e-06, "loss": 12.523111343383789, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.28834026865661144, "loss_total_v6": 0.7826944962143898, "step": 2567 }, { "epoch": 0.5922594594594595, "grad_norm": 22.976951599121094, "learning_rate": 7.893121846069913e-06, "loss": 12.94906997680664, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.3087523430585861, "loss_total_v6": 0.8093168959021568, "step": 2568 }, { "epoch": 0.5924900900900901, "grad_norm": 23.117033004760742, "learning_rate": 7.891479134121682e-06, "loss": 13.106952667236328, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.3162853177636862, "loss_total_v6": 0.8191844895482063, "step": 2569 }, { "epoch": 0.5927207207207207, "grad_norm": 22.4005184173584, "learning_rate": 7.889835953111075e-06, "loss": 12.871673583984375, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.30534932762384415, "loss_total_v6": 0.8044795542955399, "step": 2570 }, { "epoch": 0.5929513513513514, "grad_norm": 24.456300735473633, "learning_rate": 7.888192303304658e-06, "loss": 12.761178970336914, "loss_alignment": 1.00439453125, "loss_alignment_w": 0.502197265625, "loss_sft": 0.29557475820183754, "loss_total_v6": 0.7975736558437347, "step": 2571 }, { "epoch": 0.593181981981982, "grad_norm": 22.404624938964844, "learning_rate": 7.886548184969063e-06, "loss": 13.035314559936523, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.31456982158124447, "loss_total_v6": 0.8147071450948715, "step": 2572 }, { "epoch": 0.5934126126126126, "grad_norm": 23.62592124938965, "learning_rate": 7.884903598371004e-06, "loss": 12.698486328125, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.2899016756564379, "loss_total_v6": 0.7936553582549095, "step": 2573 }, { "epoch": 0.5936432432432432, "grad_norm": 24.677326202392578, "learning_rate": 7.883258543777271e-06, "loss": 12.70596694946289, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.29804443195462227, "loss_total_v6": 0.7941229045391083, "step": 2574 }, { "epoch": 0.5938738738738739, "grad_norm": 27.510711669921875, "learning_rate": 7.881613021454728e-06, "loss": 12.75152587890625, "loss_alignment": 1.0107421875, "loss_alignment_w": 0.50537109375, "loss_sft": 0.29098893515765667, "loss_total_v6": 0.7969703376293182, "step": 2575 }, { "epoch": 0.5941045045045045, "grad_norm": 25.321680068969727, "learning_rate": 7.879967031670313e-06, "loss": 12.954123497009277, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.3140882756561041, "loss_total_v6": 0.809632733464241, "step": 2576 }, { "epoch": 0.5943351351351351, "grad_norm": 25.26012420654297, "learning_rate": 7.878320574691047e-06, "loss": 13.385083198547363, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.3343551568686962, "loss_total_v6": 0.8365676999092102, "step": 2577 }, { "epoch": 0.5945657657657658, "grad_norm": 24.22256088256836, "learning_rate": 7.876673650784017e-06, "loss": 12.827192306518555, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.30700958147644997, "loss_total_v6": 0.8016995266079903, "step": 2578 }, { "epoch": 0.5947963963963964, "grad_norm": 21.33861541748047, "learning_rate": 7.875026260216395e-06, "loss": 12.524858474731445, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.27821074053645134, "loss_total_v6": 0.7828036323189735, "step": 2579 }, { "epoch": 0.595027027027027, "grad_norm": 23.853090286254883, "learning_rate": 7.87337840325542e-06, "loss": 13.37112808227539, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.33767911046743393, "loss_total_v6": 0.8356954678893089, "step": 2580 }, { "epoch": 0.5952576576576577, "grad_norm": 22.157047271728516, "learning_rate": 7.871730080168414e-06, "loss": 12.89200210571289, "loss_alignment": 1.0283203125, "loss_alignment_w": 0.51416015625, "loss_sft": 0.2911321744322777, "loss_total_v6": 0.8057501092553139, "step": 2581 }, { "epoch": 0.5954882882882883, "grad_norm": 24.184404373168945, "learning_rate": 7.870081291222772e-06, "loss": 13.183205604553223, "loss_alignment": 0.98291015625, "loss_alignment_w": 0.491455078125, "loss_sft": 0.3324342407286167, "loss_total_v6": 0.8239503502845764, "step": 2582 }, { "epoch": 0.5957189189189189, "grad_norm": 22.760021209716797, "learning_rate": 7.868432036685964e-06, "loss": 12.159557342529297, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.2639396693557501, "loss_total_v6": 0.7599723488092422, "step": 2583 }, { "epoch": 0.5959495495495496, "grad_norm": 22.584739685058594, "learning_rate": 7.866782316825535e-06, "loss": 12.406511306762695, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.275406951084733, "loss_total_v6": 0.7754069492220879, "step": 2584 }, { "epoch": 0.5961801801801802, "grad_norm": 22.648197174072266, "learning_rate": 7.865132131909106e-06, "loss": 12.980838775634766, "loss_alignment": 0.9912109375, "loss_alignment_w": 0.49560546875, "loss_sft": 0.3156817276030779, "loss_total_v6": 0.8113024309277534, "step": 2585 }, { "epoch": 0.5964108108108108, "grad_norm": 21.802629470825195, "learning_rate": 7.863481482204378e-06, "loss": 12.572959899902344, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.28164437785744667, "loss_total_v6": 0.7858100160956383, "step": 2586 }, { "epoch": 0.5966414414414415, "grad_norm": 23.12298011779785, "learning_rate": 7.861830367979117e-06, "loss": 12.459531784057617, "loss_alignment": 0.974609375, "loss_alignment_w": 0.4873046875, "loss_sft": 0.29059202782809734, "loss_total_v6": 0.7787206992506981, "step": 2587 }, { "epoch": 0.596872072072072, "grad_norm": 24.132892608642578, "learning_rate": 7.860178789501173e-06, "loss": 13.199705123901367, "loss_alignment": 0.98388671875, "loss_alignment_w": 0.491943359375, "loss_sft": 0.33245838060975075, "loss_total_v6": 0.8249815702438354, "step": 2588 }, { "epoch": 0.5971027027027027, "grad_norm": 23.610153198242188, "learning_rate": 7.85852674703847e-06, "loss": 12.86036491394043, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.30032436177134514, "loss_total_v6": 0.8037728443741798, "step": 2589 }, { "epoch": 0.5973333333333334, "grad_norm": 22.86330795288086, "learning_rate": 7.856874240859006e-06, "loss": 12.556303024291992, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.2884005382657051, "loss_total_v6": 0.7847689390182495, "step": 2590 }, { "epoch": 0.5975639639639639, "grad_norm": 23.284420013427734, "learning_rate": 7.855221271230852e-06, "loss": 13.077293395996094, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.3148741628974676, "loss_total_v6": 0.8173308297991753, "step": 2591 }, { "epoch": 0.5977945945945946, "grad_norm": 26.074731826782227, "learning_rate": 7.85356783842216e-06, "loss": 12.867624282836914, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.3027464337646961, "loss_total_v6": 0.8042265474796295, "step": 2592 }, { "epoch": 0.5980252252252253, "grad_norm": 27.7381591796875, "learning_rate": 7.85191394270115e-06, "loss": 12.451876640319824, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.284193217754364, "loss_total_v6": 0.778242290019989, "step": 2593 }, { "epoch": 0.5982558558558558, "grad_norm": 21.708951950073242, "learning_rate": 7.850259584336124e-06, "loss": 12.374923706054688, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.26891606487333775, "loss_total_v6": 0.7734326794743538, "step": 2594 }, { "epoch": 0.5984864864864865, "grad_norm": 26.621828079223633, "learning_rate": 7.848604763595452e-06, "loss": 12.80783462524414, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.3014052379876375, "loss_total_v6": 0.8004896864295006, "step": 2595 }, { "epoch": 0.5987171171171172, "grad_norm": 22.344501495361328, "learning_rate": 7.846949480747588e-06, "loss": 12.552001953125, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2821655850857496, "loss_total_v6": 0.7845001667737961, "step": 2596 }, { "epoch": 0.5989477477477477, "grad_norm": 22.37714958190918, "learning_rate": 7.845293736061052e-06, "loss": 13.369173049926758, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.33564964681863785, "loss_total_v6": 0.8355733305215836, "step": 2597 }, { "epoch": 0.5991783783783784, "grad_norm": 29.263635635375977, "learning_rate": 7.843637529804445e-06, "loss": 12.55005168914795, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.2827150486409664, "loss_total_v6": 0.7843782603740692, "step": 2598 }, { "epoch": 0.5994090090090091, "grad_norm": 27.277345657348633, "learning_rate": 7.841980862246437e-06, "loss": 12.511739730834961, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.27963392063975334, "loss_total_v6": 0.7819837778806686, "step": 2599 }, { "epoch": 0.5996396396396396, "grad_norm": 23.0993709564209, "learning_rate": 7.84032373365578e-06, "loss": 12.766565322875977, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.29942088946700096, "loss_total_v6": 0.797910287976265, "step": 2600 }, { "epoch": 0.5998702702702703, "grad_norm": 31.724124908447266, "learning_rate": 7.838666144301294e-06, "loss": 12.398269653320312, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.279042249545455, "loss_total_v6": 0.7748918682336807, "step": 2601 }, { "epoch": 0.6001009009009008, "grad_norm": 26.188386917114258, "learning_rate": 7.837008094451882e-06, "loss": 12.666433334350586, "loss_alignment": 1.00732421875, "loss_alignment_w": 0.503662109375, "loss_sft": 0.28878344036638737, "loss_total_v6": 0.7916520684957504, "step": 2602 }, { "epoch": 0.6003315315315315, "grad_norm": 26.67953109741211, "learning_rate": 7.835349584376511e-06, "loss": 12.390049934387207, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.27961187437176704, "loss_total_v6": 0.7743781208992004, "step": 2603 }, { "epoch": 0.6005621621621622, "grad_norm": 23.779922485351562, "learning_rate": 7.833690614344232e-06, "loss": 12.612071990966797, "loss_alignment": 0.9833984375, "loss_alignment_w": 0.49169921875, "loss_sft": 0.296601053327322, "loss_total_v6": 0.7882544994354248, "step": 2604 }, { "epoch": 0.6007927927927927, "grad_norm": 25.632219314575195, "learning_rate": 7.832031184624165e-06, "loss": 13.22723388671875, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.32363510876893997, "loss_total_v6": 0.8267021179199219, "step": 2605 }, { "epoch": 0.6010234234234234, "grad_norm": 22.882097244262695, "learning_rate": 7.830371295485507e-06, "loss": 12.26905632019043, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.2694100011140108, "loss_total_v6": 0.7668160051107407, "step": 2606 }, { "epoch": 0.6012540540540541, "grad_norm": 22.032625198364258, "learning_rate": 7.82871094719753e-06, "loss": 13.034734725952148, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.3136028051376343, "loss_total_v6": 0.8146709278225899, "step": 2607 }, { "epoch": 0.6014846846846846, "grad_norm": 24.87152671813965, "learning_rate": 7.827050140029578e-06, "loss": 12.39961051940918, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.28033143654465675, "loss_total_v6": 0.7749756276607513, "step": 2608 }, { "epoch": 0.6017153153153153, "grad_norm": 21.8125, "learning_rate": 7.825388874251072e-06, "loss": 12.986810684204102, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.3111415598541498, "loss_total_v6": 0.8116756305098534, "step": 2609 }, { "epoch": 0.601945945945946, "grad_norm": 23.906858444213867, "learning_rate": 7.823727150131506e-06, "loss": 12.686556816101074, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.29765529930591583, "loss_total_v6": 0.792909786105156, "step": 2610 }, { "epoch": 0.6021765765765765, "grad_norm": 24.822935104370117, "learning_rate": 7.822064967940449e-06, "loss": 12.7205810546875, "loss_alignment": 0.9697265625, "loss_alignment_w": 0.48486328125, "loss_sft": 0.3104629721492529, "loss_total_v6": 0.7950363382697105, "step": 2611 }, { "epoch": 0.6024072072072072, "grad_norm": 23.461631774902344, "learning_rate": 7.820402327947543e-06, "loss": 12.067148208618164, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.25993405655026436, "loss_total_v6": 0.7541967555880547, "step": 2612 }, { "epoch": 0.6026378378378379, "grad_norm": 23.079713821411133, "learning_rate": 7.818739230422505e-06, "loss": 12.428555488586426, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.2783106006681919, "loss_total_v6": 0.7767847254872322, "step": 2613 }, { "epoch": 0.6028684684684684, "grad_norm": 25.945003509521484, "learning_rate": 7.817075675635129e-06, "loss": 12.658210754394531, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.29501391761004925, "loss_total_v6": 0.7911381870508194, "step": 2614 }, { "epoch": 0.6030990990990991, "grad_norm": 26.446699142456055, "learning_rate": 7.815411663855279e-06, "loss": 12.906122207641602, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.3125683218240738, "loss_total_v6": 0.8066326379776001, "step": 2615 }, { "epoch": 0.6033297297297298, "grad_norm": 24.908876419067383, "learning_rate": 7.813747195352896e-06, "loss": 12.629817962646484, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.2915456872433424, "loss_total_v6": 0.7893636897206306, "step": 2616 }, { "epoch": 0.6035603603603603, "grad_norm": 21.88838005065918, "learning_rate": 7.812082270397989e-06, "loss": 12.816024780273438, "loss_alignment": 0.9833984375, "loss_alignment_w": 0.49169921875, "loss_sft": 0.30945487692952156, "loss_total_v6": 0.8010014966130257, "step": 2617 }, { "epoch": 0.603790990990991, "grad_norm": 24.336627960205078, "learning_rate": 7.810416889260653e-06, "loss": 12.62192153930664, "loss_alignment": 0.9775390625, "loss_alignment_w": 0.48876953125, "loss_sft": 0.3012144938111305, "loss_total_v6": 0.7888701260089874, "step": 2618 }, { "epoch": 0.6040216216216217, "grad_norm": 20.538639068603516, "learning_rate": 7.808751052211048e-06, "loss": 12.804357528686523, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.29538953118026257, "loss_total_v6": 0.8002723529934883, "step": 2619 }, { "epoch": 0.6042522522522522, "grad_norm": 24.053211212158203, "learning_rate": 7.807084759519405e-06, "loss": 12.200155258178711, "loss_alignment": 0.9833984375, "loss_alignment_w": 0.49169921875, "loss_sft": 0.27045949548482895, "loss_total_v6": 0.7625096440315247, "step": 2620 }, { "epoch": 0.6044828828828829, "grad_norm": 23.448978424072266, "learning_rate": 7.80541801145604e-06, "loss": 12.637533187866211, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.29033415392041206, "loss_total_v6": 0.7898458614945412, "step": 2621 }, { "epoch": 0.6047135135135135, "grad_norm": 24.853445053100586, "learning_rate": 7.803750808291332e-06, "loss": 12.417407989501953, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.2834580447524786, "loss_total_v6": 0.7760880067944527, "step": 2622 }, { "epoch": 0.6049441441441441, "grad_norm": 23.429399490356445, "learning_rate": 7.802083150295744e-06, "loss": 12.410099029541016, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.2803156152367592, "loss_total_v6": 0.7756311669945717, "step": 2623 }, { "epoch": 0.6051747747747748, "grad_norm": 23.43737030029297, "learning_rate": 7.800415037739802e-06, "loss": 12.876165390014648, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.3075221572071314, "loss_total_v6": 0.8047602847218513, "step": 2624 }, { "epoch": 0.6054054054054054, "grad_norm": 26.162260055541992, "learning_rate": 7.798746470894113e-06, "loss": 13.498861312866211, "loss_alignment": 1.009765625, "loss_alignment_w": 0.5048828125, "loss_sft": 0.3391316533088684, "loss_total_v6": 0.843678779900074, "step": 2625 }, { "epoch": 0.605636036036036, "grad_norm": 25.452268600463867, "learning_rate": 7.797077450029353e-06, "loss": 12.164484977722168, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.2615620195865631, "loss_total_v6": 0.7602802813053131, "step": 2626 }, { "epoch": 0.6058666666666667, "grad_norm": 21.70323944091797, "learning_rate": 7.79540797541628e-06, "loss": 12.184950828552246, "loss_alignment": 1.001953125, "loss_alignment_w": 0.5009765625, "loss_sft": 0.2600335329771042, "loss_total_v6": 0.7615594267845154, "step": 2627 }, { "epoch": 0.6060972972972973, "grad_norm": 29.149606704711914, "learning_rate": 7.793738047325718e-06, "loss": 12.323923110961914, "loss_alignment": 0.9912109375, "loss_alignment_w": 0.49560546875, "loss_sft": 0.2750670202076435, "loss_total_v6": 0.7702452167868614, "step": 2628 }, { "epoch": 0.6063279279279279, "grad_norm": 23.999906539916992, "learning_rate": 7.792067666028565e-06, "loss": 12.44401741027832, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.27926171012222767, "loss_total_v6": 0.777751050889492, "step": 2629 }, { "epoch": 0.6065585585585586, "grad_norm": 24.944917678833008, "learning_rate": 7.790396831795792e-06, "loss": 13.236316680908203, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.3323052264750004, "loss_total_v6": 0.8272698447108269, "step": 2630 }, { "epoch": 0.6067891891891892, "grad_norm": 24.67766571044922, "learning_rate": 7.788725544898452e-06, "loss": 12.399871826171875, "loss_alignment": 0.98046875, "loss_alignment_w": 0.490234375, "loss_sft": 0.28396411798894405, "loss_total_v6": 0.774991974234581, "step": 2631 }, { "epoch": 0.6070198198198198, "grad_norm": 23.61012840270996, "learning_rate": 7.78705380560766e-06, "loss": 12.869437217712402, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.302249388769269, "loss_total_v6": 0.8043398186564445, "step": 2632 }, { "epoch": 0.6072504504504505, "grad_norm": 22.647857666015625, "learning_rate": 7.78538161419461e-06, "loss": 12.598982810974121, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.28629201650619507, "loss_total_v6": 0.7874364331364632, "step": 2633 }, { "epoch": 0.6074810810810811, "grad_norm": 23.320743560791016, "learning_rate": 7.78370897093057e-06, "loss": 12.61659049987793, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.2951287589967251, "loss_total_v6": 0.7885369285941124, "step": 2634 }, { "epoch": 0.6077117117117117, "grad_norm": 24.223812103271484, "learning_rate": 7.78203587608688e-06, "loss": 13.146671295166016, "loss_alignment": 1.01025390625, "loss_alignment_w": 0.505126953125, "loss_sft": 0.3158075623214245, "loss_total_v6": 0.8216669410467148, "step": 2635 }, { "epoch": 0.6079423423423423, "grad_norm": 26.03194808959961, "learning_rate": 7.780362329934953e-06, "loss": 12.68989372253418, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.2988404184579849, "loss_total_v6": 0.7931183651089668, "step": 2636 }, { "epoch": 0.608172972972973, "grad_norm": 23.337661743164062, "learning_rate": 7.778688332746274e-06, "loss": 12.8802490234375, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.31137846037745476, "loss_total_v6": 0.8050155863165855, "step": 2637 }, { "epoch": 0.6084036036036036, "grad_norm": 21.480276107788086, "learning_rate": 7.777013884792403e-06, "loss": 12.302560806274414, "loss_alignment": 0.96484375, "loss_alignment_w": 0.482421875, "loss_sft": 0.28723590821027756, "loss_total_v6": 0.7689101025462151, "step": 2638 }, { "epoch": 0.6086342342342342, "grad_norm": 23.49728012084961, "learning_rate": 7.775338986344975e-06, "loss": 13.13333511352539, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.32527371868491173, "loss_total_v6": 0.8208334073424339, "step": 2639 }, { "epoch": 0.6088648648648649, "grad_norm": 22.85935401916504, "learning_rate": 7.773663637675695e-06, "loss": 13.087018013000488, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.31561924517154694, "loss_total_v6": 0.8179386183619499, "step": 2640 }, { "epoch": 0.6090954954954955, "grad_norm": 24.34484100341797, "learning_rate": 7.77198783905634e-06, "loss": 12.439516067504883, "loss_alignment": 0.990234375, "loss_alignment_w": 0.4951171875, "loss_sft": 0.2820015847682953, "loss_total_v6": 0.7774697169661522, "step": 2641 }, { "epoch": 0.6093261261261261, "grad_norm": 22.34261703491211, "learning_rate": 7.770311590758766e-06, "loss": 12.570768356323242, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.2910899221897125, "loss_total_v6": 0.7856730446219444, "step": 2642 }, { "epoch": 0.6095567567567568, "grad_norm": 23.589948654174805, "learning_rate": 7.768634893054894e-06, "loss": 12.701957702636719, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.2970309853553772, "loss_total_v6": 0.7938723787665367, "step": 2643 }, { "epoch": 0.6097873873873874, "grad_norm": 21.548057556152344, "learning_rate": 7.76695774621672e-06, "loss": 13.065171241760254, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.32159334421157837, "loss_total_v6": 0.8165731951594353, "step": 2644 }, { "epoch": 0.610018018018018, "grad_norm": 25.00515365600586, "learning_rate": 7.765280150516322e-06, "loss": 12.589879035949707, "loss_alignment": 0.9833984375, "loss_alignment_w": 0.49169921875, "loss_sft": 0.29486302845180035, "loss_total_v6": 0.7868674248456955, "step": 2645 }, { "epoch": 0.6102486486486487, "grad_norm": 23.75655174255371, "learning_rate": 7.763602106225836e-06, "loss": 12.769782066345215, "loss_alignment": 0.9912109375, "loss_alignment_w": 0.49560546875, "loss_sft": 0.3008884247392416, "loss_total_v6": 0.7981113418936729, "step": 2646 }, { "epoch": 0.6104792792792793, "grad_norm": 22.950929641723633, "learning_rate": 7.761923613617481e-06, "loss": 12.224685668945312, "loss_alignment": 0.990234375, "loss_alignment_w": 0.4951171875, "loss_sft": 0.2702989783138037, "loss_total_v6": 0.7640428692102432, "step": 2647 }, { "epoch": 0.6107099099099099, "grad_norm": 24.96103858947754, "learning_rate": 7.760244672963548e-06, "loss": 12.33489990234375, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.27439497970044613, "loss_total_v6": 0.7709312364459038, "step": 2648 }, { "epoch": 0.6109405405405406, "grad_norm": 26.34832000732422, "learning_rate": 7.758565284536395e-06, "loss": 12.47083854675293, "loss_alignment": 0.99267578125, "loss_alignment_w": 0.496337890625, "loss_sft": 0.28405080549418926, "loss_total_v6": 0.7794274017214775, "step": 2649 }, { "epoch": 0.6111711711711711, "grad_norm": 24.896516799926758, "learning_rate": 7.75688544860846e-06, "loss": 13.205644607543945, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.3285570926964283, "loss_total_v6": 0.8253527879714966, "step": 2650 }, { "epoch": 0.6114018018018018, "grad_norm": 23.19784164428711, "learning_rate": 7.755205165452245e-06, "loss": 11.944812774658203, "loss_alignment": 0.98193359375, "loss_alignment_w": 0.490966796875, "loss_sft": 0.2564232014119625, "loss_total_v6": 0.7465507835149765, "step": 2651 }, { "epoch": 0.6116324324324325, "grad_norm": 23.066577911376953, "learning_rate": 7.753524435340334e-06, "loss": 12.42551040649414, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.27479384653270245, "loss_total_v6": 0.776594378054142, "step": 2652 }, { "epoch": 0.611863063063063, "grad_norm": 23.633861541748047, "learning_rate": 7.751843258545377e-06, "loss": 12.2981538772583, "loss_alignment": 0.97314453125, "loss_alignment_w": 0.486572265625, "loss_sft": 0.28174193762242794, "loss_total_v6": 0.7686346247792244, "step": 2653 }, { "epoch": 0.6120936936936937, "grad_norm": 23.03799819946289, "learning_rate": 7.750161635340097e-06, "loss": 12.644453048706055, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.2887066211551428, "loss_total_v6": 0.7902782782912254, "step": 2654 }, { "epoch": 0.6123243243243244, "grad_norm": 25.280399322509766, "learning_rate": 7.748479565997293e-06, "loss": 12.70376205444336, "loss_alignment": 0.990234375, "loss_alignment_w": 0.4951171875, "loss_sft": 0.29725051671266556, "loss_total_v6": 0.7939851582050323, "step": 2655 }, { "epoch": 0.6125549549549549, "grad_norm": 26.939756393432617, "learning_rate": 7.746797050789834e-06, "loss": 12.897537231445312, "loss_alignment": 0.95849609375, "loss_alignment_w": 0.479248046875, "loss_sft": 0.32652759179472923, "loss_total_v6": 0.8060961067676544, "step": 2656 }, { "epoch": 0.6127855855855856, "grad_norm": 27.600679397583008, "learning_rate": 7.74511408999066e-06, "loss": 12.770624160766602, "loss_alignment": 1.0166015625, "loss_alignment_w": 0.50830078125, "loss_sft": 0.2898480035364628, "loss_total_v6": 0.7981640249490738, "step": 2657 }, { "epoch": 0.6130162162162163, "grad_norm": 24.52500343322754, "learning_rate": 7.743430683872785e-06, "loss": 12.586685180664062, "loss_alignment": 0.955078125, "loss_alignment_w": 0.4775390625, "loss_sft": 0.3098764717578888, "loss_total_v6": 0.7866678461432457, "step": 2658 }, { "epoch": 0.6132468468468468, "grad_norm": 22.646331787109375, "learning_rate": 7.741746832709296e-06, "loss": 12.46410846710205, "loss_alignment": 0.97705078125, "loss_alignment_w": 0.488525390625, "loss_sft": 0.2897184304893017, "loss_total_v6": 0.77900680154562, "step": 2659 }, { "epoch": 0.6134774774774775, "grad_norm": 24.074989318847656, "learning_rate": 7.740062536773352e-06, "loss": 12.961747169494629, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.30656910315155983, "loss_total_v6": 0.8101091682910919, "step": 2660 }, { "epoch": 0.6137081081081082, "grad_norm": 23.711362838745117, "learning_rate": 7.738377796338179e-06, "loss": 12.236188888549805, "loss_alignment": 0.97802734375, "loss_alignment_w": 0.489013671875, "loss_sft": 0.27445114590227604, "loss_total_v6": 0.7647617980837822, "step": 2661 }, { "epoch": 0.6139387387387387, "grad_norm": 23.68733024597168, "learning_rate": 7.736692611677084e-06, "loss": 12.595080375671387, "loss_alignment": 0.97900390625, "loss_alignment_w": 0.489501953125, "loss_sft": 0.2965156063437462, "loss_total_v6": 0.7871925160288811, "step": 2662 }, { "epoch": 0.6141693693693694, "grad_norm": 21.80451774597168, "learning_rate": 7.735006983063443e-06, "loss": 12.798696517944336, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.29521886073052883, "loss_total_v6": 0.7999185547232628, "step": 2663 }, { "epoch": 0.6144, "grad_norm": 22.93754768371582, "learning_rate": 7.733320910770695e-06, "loss": 12.640811920166016, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.288906367495656, "loss_total_v6": 0.790050782263279, "step": 2664 }, { "epoch": 0.6146306306306306, "grad_norm": 23.053884506225586, "learning_rate": 7.731634395072366e-06, "loss": 12.09417724609375, "loss_alignment": 0.98388671875, "loss_alignment_w": 0.491943359375, "loss_sft": 0.26359170489013195, "loss_total_v6": 0.7558860629796982, "step": 2665 }, { "epoch": 0.6148612612612613, "grad_norm": 23.21442222595215, "learning_rate": 7.729947436242041e-06, "loss": 12.534377098083496, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.28107925131917, "loss_total_v6": 0.7833985835313797, "step": 2666 }, { "epoch": 0.6150918918918918, "grad_norm": 24.004985809326172, "learning_rate": 7.728260034553387e-06, "loss": 13.037137031555176, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.3133256770670414, "loss_total_v6": 0.8148210644721985, "step": 2667 }, { "epoch": 0.6153225225225225, "grad_norm": 21.672712326049805, "learning_rate": 7.726572190280134e-06, "loss": 12.41335678100586, "loss_alignment": 0.98486328125, "loss_alignment_w": 0.492431640625, "loss_sft": 0.28271649591624737, "loss_total_v6": 0.7758347764611244, "step": 2668 }, { "epoch": 0.6155531531531532, "grad_norm": 23.649023056030273, "learning_rate": 7.724883903696092e-06, "loss": 12.581341743469238, "loss_alignment": 1.013671875, "loss_alignment_w": 0.5068359375, "loss_sft": 0.2790248915553093, "loss_total_v6": 0.7863338515162468, "step": 2669 }, { "epoch": 0.6157837837837837, "grad_norm": 24.86736488342285, "learning_rate": 7.723195175075136e-06, "loss": 12.600702285766602, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.28772706165909767, "loss_total_v6": 0.7875439450144768, "step": 2670 }, { "epoch": 0.6160144144144144, "grad_norm": 23.098072052001953, "learning_rate": 7.721506004691216e-06, "loss": 12.718385696411133, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.29604351334273815, "loss_total_v6": 0.7948991134762764, "step": 2671 }, { "epoch": 0.6162450450450451, "grad_norm": 24.61551856994629, "learning_rate": 7.719816392818354e-06, "loss": 12.56968879699707, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.2893439084291458, "loss_total_v6": 0.7856055274605751, "step": 2672 }, { "epoch": 0.6164756756756756, "grad_norm": 21.551536560058594, "learning_rate": 7.718126339730642e-06, "loss": 12.83061695098877, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2981141321361065, "loss_total_v6": 0.8019135817885399, "step": 2673 }, { "epoch": 0.6167063063063063, "grad_norm": 20.90448570251465, "learning_rate": 7.716435845702242e-06, "loss": 12.511935234069824, "loss_alignment": 0.98291015625, "loss_alignment_w": 0.491455078125, "loss_sft": 0.28954908810555935, "loss_total_v6": 0.781995952129364, "step": 2674 }, { "epoch": 0.616936936936937, "grad_norm": 24.248409271240234, "learning_rate": 7.714744911007395e-06, "loss": 12.484354019165039, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.2913195323199034, "loss_total_v6": 0.7802721485495567, "step": 2675 }, { "epoch": 0.6171675675675675, "grad_norm": 23.268461227416992, "learning_rate": 7.713053535920403e-06, "loss": 12.525460243225098, "loss_alignment": 0.9873046875, "loss_alignment_w": 0.49365234375, "loss_sft": 0.28805980645120144, "loss_total_v6": 0.7828412801027298, "step": 2676 }, { "epoch": 0.6173981981981982, "grad_norm": 27.120935440063477, "learning_rate": 7.71136172071565e-06, "loss": 12.862691879272461, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.2981962226331234, "loss_total_v6": 0.8039182722568512, "step": 2677 }, { "epoch": 0.6176288288288289, "grad_norm": 21.88187599182129, "learning_rate": 7.709669465667578e-06, "loss": 12.619444847106934, "loss_alignment": 0.986328125, "loss_alignment_w": 0.4931640625, "loss_sft": 0.29558177664875984, "loss_total_v6": 0.7887153178453445, "step": 2678 }, { "epoch": 0.6178594594594594, "grad_norm": 23.087770462036133, "learning_rate": 7.707976771050718e-06, "loss": 12.434793472290039, "loss_alignment": 0.99267578125, "loss_alignment_w": 0.496337890625, "loss_sft": 0.28068406507372856, "loss_total_v6": 0.7771745696663857, "step": 2679 }, { "epoch": 0.6180900900900901, "grad_norm": 22.908803939819336, "learning_rate": 7.706283637139658e-06, "loss": 12.395007133483887, "loss_alignment": 0.97607421875, "loss_alignment_w": 0.488037109375, "loss_sft": 0.28703230060636997, "loss_total_v6": 0.7746879309415817, "step": 2680 }, { "epoch": 0.6183207207207208, "grad_norm": 22.144495010375977, "learning_rate": 7.704590064209062e-06, "loss": 12.702707290649414, "loss_alignment": 1.005859375, "loss_alignment_w": 0.5029296875, "loss_sft": 0.2921949680894613, "loss_total_v6": 0.7939192429184914, "step": 2681 }, { "epoch": 0.6185513513513513, "grad_norm": 24.55200958251953, "learning_rate": 7.702896052533666e-06, "loss": 12.835886001586914, "loss_alignment": 0.97265625, "loss_alignment_w": 0.486328125, "loss_sft": 0.3149992246180773, "loss_total_v6": 0.8022428974509239, "step": 2682 }, { "epoch": 0.618781981981982, "grad_norm": 21.078386306762695, "learning_rate": 7.701201602388276e-06, "loss": 13.199417114257812, "loss_alignment": 1.0146484375, "loss_alignment_w": 0.50732421875, "loss_sft": 0.3175324946641922, "loss_total_v6": 0.8249635323882103, "step": 2683 }, { "epoch": 0.6190126126126126, "grad_norm": 24.02169418334961, "learning_rate": 7.69950671404777e-06, "loss": 12.652273178100586, "loss_alignment": 0.98583984375, "loss_alignment_w": 0.492919921875, "loss_sft": 0.2990373820066452, "loss_total_v6": 0.790767103433609, "step": 2684 }, { "epoch": 0.6192432432432432, "grad_norm": 25.424652099609375, "learning_rate": 7.697811387787097e-06, "loss": 12.462459564208984, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.27936152182519436, "loss_total_v6": 0.7789037451148033, "step": 2685 }, { "epoch": 0.6194738738738739, "grad_norm": 24.284996032714844, "learning_rate": 7.696115623881275e-06, "loss": 12.575632095336914, "loss_alignment": 0.978515625, "loss_alignment_w": 0.4892578125, "loss_sft": 0.29691758193075657, "loss_total_v6": 0.7859770208597183, "step": 2686 }, { "epoch": 0.6197045045045045, "grad_norm": 25.952407836914062, "learning_rate": 7.694419422605399e-06, "loss": 12.675971984863281, "loss_alignment": 0.97314453125, "loss_alignment_w": 0.486572265625, "loss_sft": 0.3054165877401829, "loss_total_v6": 0.7922482565045357, "step": 2687 }, { "epoch": 0.6199351351351351, "grad_norm": 25.109289169311523, "learning_rate": 7.692722784234624e-06, "loss": 12.391657829284668, "loss_alignment": 0.978515625, "loss_alignment_w": 0.4892578125, "loss_sft": 0.2856327947229147, "loss_total_v6": 0.7744786292314529, "step": 2688 }, { "epoch": 0.6201657657657658, "grad_norm": 23.58951187133789, "learning_rate": 7.691025709044189e-06, "loss": 12.600648880004883, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.2970620598644018, "loss_total_v6": 0.7875405848026276, "step": 2689 }, { "epoch": 0.6203963963963964, "grad_norm": 24.530216217041016, "learning_rate": 7.689328197309394e-06, "loss": 12.909404754638672, "loss_alignment": 0.982421875, "loss_alignment_w": 0.4912109375, "loss_sft": 0.3168628551065922, "loss_total_v6": 0.8068378195166588, "step": 2690 }, { "epoch": 0.620627027027027, "grad_norm": 23.837831497192383, "learning_rate": 7.687630249305613e-06, "loss": 12.331794738769531, "loss_alignment": 0.9873046875, "loss_alignment_w": 0.49365234375, "loss_sft": 0.2748417295515537, "loss_total_v6": 0.7707371413707733, "step": 2691 }, { "epoch": 0.6208576576576577, "grad_norm": 23.932584762573242, "learning_rate": 7.685931865308293e-06, "loss": 12.804098129272461, "loss_alignment": 1.00341796875, "loss_alignment_w": 0.501708984375, "loss_sft": 0.2971281334757805, "loss_total_v6": 0.800256185233593, "step": 2692 }, { "epoch": 0.6210882882882883, "grad_norm": 21.20093536376953, "learning_rate": 7.684233045592946e-06, "loss": 12.047022819519043, "loss_alignment": 0.97119140625, "loss_alignment_w": 0.485595703125, "loss_sft": 0.26697697304189205, "loss_total_v6": 0.7529389038681984, "step": 2693 }, { "epoch": 0.6213189189189189, "grad_norm": 22.42945098876953, "learning_rate": 7.682533790435165e-06, "loss": 12.522970199584961, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.27934401482343674, "loss_total_v6": 0.7826856523752213, "step": 2694 }, { "epoch": 0.6215495495495496, "grad_norm": 23.09042739868164, "learning_rate": 7.680834100110601e-06, "loss": 12.998003005981445, "loss_alignment": 0.99609375, "loss_alignment_w": 0.498046875, "loss_sft": 0.3139315992593765, "loss_total_v6": 0.8123751878738403, "step": 2695 }, { "epoch": 0.6217801801801802, "grad_norm": 25.252723693847656, "learning_rate": 7.679133974894984e-06, "loss": 12.56060791015625, "loss_alignment": 0.97900390625, "loss_alignment_w": 0.489501953125, "loss_sft": 0.2958412542939186, "loss_total_v6": 0.7850380465388298, "step": 2696 }, { "epoch": 0.6220108108108108, "grad_norm": 23.754060745239258, "learning_rate": 7.677433415064112e-06, "loss": 12.75544261932373, "loss_alignment": 1.00634765625, "loss_alignment_w": 0.503173828125, "loss_sft": 0.29388878121972084, "loss_total_v6": 0.7972152084112167, "step": 2697 }, { "epoch": 0.6222414414414414, "grad_norm": 22.116893768310547, "learning_rate": 7.675732420893854e-06, "loss": 12.949836730957031, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.31243180483579636, "loss_total_v6": 0.8093647956848145, "step": 2698 }, { "epoch": 0.6224720720720721, "grad_norm": 24.881425857543945, "learning_rate": 7.674030992660149e-06, "loss": 12.52409839630127, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.2890885416418314, "loss_total_v6": 0.7827561572194099, "step": 2699 }, { "epoch": 0.6227027027027027, "grad_norm": 22.422039031982422, "learning_rate": 7.672329130639007e-06, "loss": 12.822248458862305, "loss_alignment": 1.00927734375, "loss_alignment_w": 0.504638671875, "loss_sft": 0.29569895192980766, "loss_total_v6": 0.8013905063271523, "step": 2700 }, { "epoch": 0.6229333333333333, "grad_norm": 22.862112045288086, "learning_rate": 7.670626835106507e-06, "loss": 12.460817337036133, "loss_alignment": 0.98388671875, "loss_alignment_w": 0.491943359375, "loss_sft": 0.2861558683216572, "loss_total_v6": 0.7788011357188225, "step": 2701 }, { "epoch": 0.623163963963964, "grad_norm": 23.9091739654541, "learning_rate": 7.6689241063388e-06, "loss": 12.917387008666992, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.30739767849445343, "loss_total_v6": 0.8073366656899452, "step": 2702 }, { "epoch": 0.6233945945945946, "grad_norm": 25.218582153320312, "learning_rate": 7.667220944612103e-06, "loss": 12.959379196166992, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.3126925565302372, "loss_total_v6": 0.8099612295627594, "step": 2703 }, { "epoch": 0.6236252252252252, "grad_norm": 25.563247680664062, "learning_rate": 7.665517350202715e-06, "loss": 12.296753883361816, "loss_alignment": 0.97216796875, "loss_alignment_w": 0.486083984375, "loss_sft": 0.28253943659365177, "loss_total_v6": 0.7685470953583717, "step": 2704 }, { "epoch": 0.6238558558558559, "grad_norm": 23.414220809936523, "learning_rate": 7.663813323386988e-06, "loss": 12.346723556518555, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.2723721042275429, "loss_total_v6": 0.7716702073812485, "step": 2705 }, { "epoch": 0.6240864864864865, "grad_norm": 23.198902130126953, "learning_rate": 7.66210886444136e-06, "loss": 12.544166564941406, "loss_alignment": 1.00390625, "loss_alignment_w": 0.501953125, "loss_sft": 0.28173677809536457, "loss_total_v6": 0.7840103730559349, "step": 2706 }, { "epoch": 0.6243171171171171, "grad_norm": 21.755569458007812, "learning_rate": 7.66040397364233e-06, "loss": 12.823076248168945, "loss_alignment": 1.0009765625, "loss_alignment_w": 0.50048828125, "loss_sft": 0.30077087320387363, "loss_total_v6": 0.8014422506093979, "step": 2707 }, { "epoch": 0.6245477477477478, "grad_norm": 24.98353385925293, "learning_rate": 7.658698651266467e-06, "loss": 12.870694160461426, "loss_alignment": 0.99267578125, "loss_alignment_w": 0.496337890625, "loss_sft": 0.3082788474857807, "loss_total_v6": 0.8044183850288391, "step": 2708 }, { "epoch": 0.6247783783783784, "grad_norm": 25.09702491760254, "learning_rate": 7.656992897590416e-06, "loss": 12.4851655960083, "loss_alignment": 0.97216796875, "loss_alignment_w": 0.486083984375, "loss_sft": 0.2936437875032425, "loss_total_v6": 0.7803228199481964, "step": 2709 }, { "epoch": 0.625009009009009, "grad_norm": 22.745275497436523, "learning_rate": 7.655286712890883e-06, "loss": 12.331567764282227, "loss_alignment": 0.9775390625, "loss_alignment_w": 0.48876953125, "loss_sft": 0.28209070675075054, "loss_total_v6": 0.770722933113575, "step": 2710 }, { "epoch": 0.6252396396396397, "grad_norm": 20.582975387573242, "learning_rate": 7.653580097444655e-06, "loss": 12.026132583618164, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.2621923368424177, "loss_total_v6": 0.7516332566738129, "step": 2711 }, { "epoch": 0.6254702702702702, "grad_norm": 25.22572135925293, "learning_rate": 7.651873051528582e-06, "loss": 12.887457847595215, "loss_alignment": 0.99267578125, "loss_alignment_w": 0.496337890625, "loss_sft": 0.3074039947241545, "loss_total_v6": 0.8054661229252815, "step": 2712 }, { "epoch": 0.6257009009009009, "grad_norm": 22.69854736328125, "learning_rate": 7.650165575419582e-06, "loss": 12.681550979614258, "loss_alignment": 0.9853515625, "loss_alignment_w": 0.49267578125, "loss_sft": 0.2992344759404659, "loss_total_v6": 0.7925969138741493, "step": 2713 }, { "epoch": 0.6259315315315316, "grad_norm": 24.32147979736328, "learning_rate": 7.648457669394645e-06, "loss": 12.522701263427734, "loss_alignment": 0.97802734375, "loss_alignment_w": 0.489013671875, "loss_sft": 0.2940824441611767, "loss_total_v6": 0.7826688438653946, "step": 2714 }, { "epoch": 0.6261621621621621, "grad_norm": 24.676084518432617, "learning_rate": 7.646749333730837e-06, "loss": 12.744757652282715, "loss_alignment": 0.97900390625, "loss_alignment_w": 0.489501953125, "loss_sft": 0.3059772774577141, "loss_total_v6": 0.7965473458170891, "step": 2715 }, { "epoch": 0.6263927927927928, "grad_norm": 25.03060531616211, "learning_rate": 7.645040568705281e-06, "loss": 12.59445571899414, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.2925398461520672, "loss_total_v6": 0.7871534824371338, "step": 2716 }, { "epoch": 0.6266234234234235, "grad_norm": 23.023130416870117, "learning_rate": 7.643331374595184e-06, "loss": 12.124227523803711, "loss_alignment": 0.9912109375, "loss_alignment_w": 0.49560546875, "loss_sft": 0.26263178512454033, "loss_total_v6": 0.7577642202377319, "step": 2717 }, { "epoch": 0.626854054054054, "grad_norm": 22.908626556396484, "learning_rate": 7.64162175167781e-06, "loss": 12.03890609741211, "loss_alignment": 0.95361328125, "loss_alignment_w": 0.476806640625, "loss_sft": 0.275228263810277, "loss_total_v6": 0.7524316385388374, "step": 2718 }, { "epoch": 0.6270846846846847, "grad_norm": 22.349414825439453, "learning_rate": 7.639911700230502e-06, "loss": 12.526901245117188, "loss_alignment": 0.9833984375, "loss_alignment_w": 0.49169921875, "loss_sft": 0.29124736227095127, "loss_total_v6": 0.7829313278198242, "step": 2719 }, { "epoch": 0.6273153153153154, "grad_norm": 24.666414260864258, "learning_rate": 7.638201220530664e-06, "loss": 13.015632629394531, "loss_alignment": 0.98193359375, "loss_alignment_w": 0.490966796875, "loss_sft": 0.3215184509754181, "loss_total_v6": 0.8134770691394806, "step": 2720 }, { "epoch": 0.6275459459459459, "grad_norm": 24.756025314331055, "learning_rate": 7.636490312855778e-06, "loss": 13.388176918029785, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.3379054516553879, "loss_total_v6": 0.8367610424757004, "step": 2721 }, { "epoch": 0.6277765765765766, "grad_norm": 23.728755950927734, "learning_rate": 7.634778977483389e-06, "loss": 12.265174865722656, "loss_alignment": 0.98046875, "loss_alignment_w": 0.490234375, "loss_sft": 0.2764611020684242, "loss_total_v6": 0.7665733993053436, "step": 2722 }, { "epoch": 0.6280072072072072, "grad_norm": 21.09589195251465, "learning_rate": 7.633067214691115e-06, "loss": 11.898651123046875, "loss_alignment": 0.97314453125, "loss_alignment_w": 0.486572265625, "loss_sft": 0.25665097311139107, "loss_total_v6": 0.7436657249927521, "step": 2723 }, { "epoch": 0.6282378378378378, "grad_norm": 23.937559127807617, "learning_rate": 7.63135502475664e-06, "loss": 12.186063766479492, "loss_alignment": 0.97607421875, "loss_alignment_w": 0.488037109375, "loss_sft": 0.2717303317040205, "loss_total_v6": 0.7616290226578712, "step": 2724 }, { "epoch": 0.6284684684684685, "grad_norm": 23.60840606689453, "learning_rate": 7.62964240795772e-06, "loss": 12.732843399047852, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.3023182414472103, "loss_total_v6": 0.7958027124404907, "step": 2725 }, { "epoch": 0.628699099099099, "grad_norm": 24.158830642700195, "learning_rate": 7.627929364572184e-06, "loss": 12.439308166503906, "loss_alignment": 0.98291015625, "loss_alignment_w": 0.491455078125, "loss_sft": 0.28650519624352455, "loss_total_v6": 0.777456745505333, "step": 2726 }, { "epoch": 0.6289297297297297, "grad_norm": 23.78984260559082, "learning_rate": 7.6262158948779185e-06, "loss": 12.726400375366211, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.29836025834083557, "loss_total_v6": 0.7954000681638718, "step": 2727 }, { "epoch": 0.6291603603603604, "grad_norm": 24.763694763183594, "learning_rate": 7.6245019991528934e-06, "loss": 13.066898345947266, "loss_alignment": 0.9970703125, "loss_alignment_w": 0.49853515625, "loss_sft": 0.31858850084245205, "loss_total_v6": 0.8166811689734459, "step": 2728 }, { "epoch": 0.6293909909909909, "grad_norm": 22.86802864074707, "learning_rate": 7.622787677675136e-06, "loss": 12.530311584472656, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.2880883887410164, "loss_total_v6": 0.7831445336341858, "step": 2729 }, { "epoch": 0.6296216216216216, "grad_norm": 22.561044692993164, "learning_rate": 7.621072930722749e-06, "loss": 12.476984977722168, "loss_alignment": 0.9677734375, "loss_alignment_w": 0.48388671875, "loss_sft": 0.29368182457983494, "loss_total_v6": 0.7798115909099579, "step": 2730 }, { "epoch": 0.6298522522522523, "grad_norm": 22.409881591796875, "learning_rate": 7.619357758573904e-06, "loss": 12.792448997497559, "loss_alignment": 0.9990234375, "loss_alignment_w": 0.49951171875, "loss_sft": 0.2997263986617327, "loss_total_v6": 0.799528032541275, "step": 2731 }, { "epoch": 0.6300828828828828, "grad_norm": 24.290224075317383, "learning_rate": 7.617642161506837e-06, "loss": 12.180788040161133, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.2606431860476732, "loss_total_v6": 0.7612992823123932, "step": 2732 }, { "epoch": 0.6303135135135135, "grad_norm": 22.946306228637695, "learning_rate": 7.615926139799859e-06, "loss": 13.274510383605957, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.3328765295445919, "loss_total_v6": 0.8296569064259529, "step": 2733 }, { "epoch": 0.6305441441441442, "grad_norm": 22.34003448486328, "learning_rate": 7.614209693731347e-06, "loss": 12.717184066772461, "loss_alignment": 0.99267578125, "loss_alignment_w": 0.496337890625, "loss_sft": 0.29831823520362377, "loss_total_v6": 0.794823981821537, "step": 2734 }, { "epoch": 0.6307747747747747, "grad_norm": 23.960142135620117, "learning_rate": 7.612492823579744e-06, "loss": 13.007827758789062, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.3178110048174858, "loss_total_v6": 0.8129892200231552, "step": 2735 }, { "epoch": 0.6310054054054054, "grad_norm": 21.6503963470459, "learning_rate": 7.610775529623568e-06, "loss": 12.269769668579102, "loss_alignment": 0.96337890625, "loss_alignment_w": 0.481689453125, "loss_sft": 0.2841335106641054, "loss_total_v6": 0.7668605893850327, "step": 2736 }, { "epoch": 0.6312360360360361, "grad_norm": 23.96177101135254, "learning_rate": 7.609057812141399e-06, "loss": 12.663660049438477, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.301290113478899, "loss_total_v6": 0.791478730738163, "step": 2737 }, { "epoch": 0.6314666666666666, "grad_norm": 25.03627586364746, "learning_rate": 7.607339671411893e-06, "loss": 12.363182067871094, "loss_alignment": 0.97705078125, "loss_alignment_w": 0.488525390625, "loss_sft": 0.28499746322631836, "loss_total_v6": 0.7726988717913628, "step": 2738 }, { "epoch": 0.6316972972972973, "grad_norm": 22.642070770263672, "learning_rate": 7.605621107713768e-06, "loss": 12.724287033081055, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.2935742698609829, "loss_total_v6": 0.7952679842710495, "step": 2739 }, { "epoch": 0.631927927927928, "grad_norm": 22.981462478637695, "learning_rate": 7.603902121325813e-06, "loss": 12.323018074035645, "loss_alignment": 0.9736328125, "loss_alignment_w": 0.48681640625, "loss_sft": 0.28390624560415745, "loss_total_v6": 0.770188607275486, "step": 2740 }, { "epoch": 0.6321585585585585, "grad_norm": 23.176631927490234, "learning_rate": 7.6021827125268865e-06, "loss": 12.250786781311035, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.27296786196529865, "loss_total_v6": 0.7656741440296173, "step": 2741 }, { "epoch": 0.6323891891891892, "grad_norm": 24.772953033447266, "learning_rate": 7.6004628815959155e-06, "loss": 12.668383598327637, "loss_alignment": 0.97607421875, "loss_alignment_w": 0.488037109375, "loss_sft": 0.3052627258002758, "loss_total_v6": 0.7917739450931549, "step": 2742 }, { "epoch": 0.6326198198198199, "grad_norm": 22.61684226989746, "learning_rate": 7.598742628811894e-06, "loss": 12.416543006896973, "loss_alignment": 0.97119140625, "loss_alignment_w": 0.485595703125, "loss_sft": 0.289583757519722, "loss_total_v6": 0.7760339379310608, "step": 2743 }, { "epoch": 0.6328504504504504, "grad_norm": 23.591672897338867, "learning_rate": 7.597021954453887e-06, "loss": 12.52444076538086, "loss_alignment": 0.98193359375, "loss_alignment_w": 0.490966796875, "loss_sft": 0.2913224417716265, "loss_total_v6": 0.7827775478363037, "step": 2744 }, { "epoch": 0.6330810810810811, "grad_norm": 24.105236053466797, "learning_rate": 7.595300858801025e-06, "loss": 12.44068717956543, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.2902077902108431, "loss_total_v6": 0.7775429859757423, "step": 2745 }, { "epoch": 0.6333117117117117, "grad_norm": 22.42656135559082, "learning_rate": 7.5935793421325076e-06, "loss": 12.387552261352539, "loss_alignment": 0.99267578125, "loss_alignment_w": 0.496337890625, "loss_sft": 0.2775789685547352, "loss_total_v6": 0.7742220312356949, "step": 2746 }, { "epoch": 0.6335423423423423, "grad_norm": 23.524024963378906, "learning_rate": 7.591857404727605e-06, "loss": 12.418298721313477, "loss_alignment": 0.97021484375, "loss_alignment_w": 0.485107421875, "loss_sft": 0.2917839400470257, "loss_total_v6": 0.7761436626315117, "step": 2747 }, { "epoch": 0.633772972972973, "grad_norm": 24.26080894470215, "learning_rate": 7.590135046865652e-06, "loss": 12.622241973876953, "loss_alignment": 0.97607421875, "loss_alignment_w": 0.488037109375, "loss_sft": 0.3010514099150896, "loss_total_v6": 0.7888901382684708, "step": 2748 }, { "epoch": 0.6340036036036036, "grad_norm": 24.564706802368164, "learning_rate": 7.588412268826053e-06, "loss": 12.729217529296875, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.2934550791978836, "loss_total_v6": 0.7955760508775711, "step": 2749 }, { "epoch": 0.6342342342342342, "grad_norm": 23.07603645324707, "learning_rate": 7.586689070888284e-06, "loss": 12.488346099853516, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.292606582865119, "loss_total_v6": 0.7805216088891029, "step": 2750 }, { "epoch": 0.6344648648648649, "grad_norm": 21.203996658325195, "learning_rate": 7.5849654533318855e-06, "loss": 12.688582420349121, "loss_alignment": 0.97802734375, "loss_alignment_w": 0.489013671875, "loss_sft": 0.30388540774583817, "loss_total_v6": 0.7930364012718201, "step": 2751 }, { "epoch": 0.6346954954954955, "grad_norm": 23.52547264099121, "learning_rate": 7.583241416436462e-06, "loss": 12.658292770385742, "loss_alignment": 0.990234375, "loss_alignment_w": 0.4951171875, "loss_sft": 0.29643817245960236, "loss_total_v6": 0.7911433354020119, "step": 2752 }, { "epoch": 0.6349261261261261, "grad_norm": 25.34828758239746, "learning_rate": 7.5815169604816965e-06, "loss": 13.067955017089844, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.32120272889733315, "loss_total_v6": 0.8167471587657928, "step": 2753 }, { "epoch": 0.6351567567567568, "grad_norm": 24.302745819091797, "learning_rate": 7.57979208574733e-06, "loss": 12.611742973327637, "loss_alignment": 0.9814453125, "loss_alignment_w": 0.49072265625, "loss_sft": 0.2969466857612133, "loss_total_v6": 0.7882339134812355, "step": 2754 }, { "epoch": 0.6353873873873874, "grad_norm": 22.791601181030273, "learning_rate": 7.578066792513179e-06, "loss": 12.900141716003418, "loss_alignment": 0.98583984375, "loss_alignment_w": 0.492919921875, "loss_sft": 0.3132016360759735, "loss_total_v6": 0.806258887052536, "step": 2755 }, { "epoch": 0.635618018018018, "grad_norm": 22.296161651611328, "learning_rate": 7.576341081059125e-06, "loss": 12.747843742370605, "loss_alignment": 0.966796875, "loss_alignment_w": 0.4833984375, "loss_sft": 0.3137843422591686, "loss_total_v6": 0.7967402562499046, "step": 2756 }, { "epoch": 0.6358486486486487, "grad_norm": 24.56186294555664, "learning_rate": 7.5746149516651114e-06, "loss": 12.77966594696045, "loss_alignment": 0.9853515625, "loss_alignment_w": 0.49267578125, "loss_sft": 0.3063432741910219, "loss_total_v6": 0.7987291291356087, "step": 2757 }, { "epoch": 0.6360792792792793, "grad_norm": 22.838436126708984, "learning_rate": 7.57288840461116e-06, "loss": 12.422172546386719, "loss_alignment": 0.96484375, "loss_alignment_w": 0.482421875, "loss_sft": 0.29402491077780724, "loss_total_v6": 0.7763857915997505, "step": 2758 }, { "epoch": 0.6363099099099099, "grad_norm": 23.34752082824707, "learning_rate": 7.571161440177353e-06, "loss": 12.946605682373047, "loss_alignment": 0.99658203125, "loss_alignment_w": 0.498291015625, "loss_sft": 0.31027680076658726, "loss_total_v6": 0.809162899851799, "step": 2759 }, { "epoch": 0.6365405405405405, "grad_norm": 25.631845474243164, "learning_rate": 7.5694340586438446e-06, "loss": 12.938982009887695, "loss_alignment": 1.0068359375, "loss_alignment_w": 0.50341796875, "loss_sft": 0.3060007840394974, "loss_total_v6": 0.8086863309144974, "step": 2760 }, { "epoch": 0.6367711711711712, "grad_norm": 24.49873161315918, "learning_rate": 7.5677062602908515e-06, "loss": 13.02805233001709, "loss_alignment": 0.98388671875, "loss_alignment_w": 0.491943359375, "loss_sft": 0.3219436760991812, "loss_total_v6": 0.8142532333731651, "step": 2761 }, { "epoch": 0.6370018018018018, "grad_norm": 24.78513526916504, "learning_rate": 7.565978045398663e-06, "loss": 12.518179893493652, "loss_alignment": 0.96923828125, "loss_alignment_w": 0.484619140625, "loss_sft": 0.29654646664857864, "loss_total_v6": 0.7823862656950951, "step": 2762 }, { "epoch": 0.6372324324324324, "grad_norm": 25.040630340576172, "learning_rate": 7.5642494142476346e-06, "loss": 12.87378215789795, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.3109895698726177, "loss_total_v6": 0.8046113774180412, "step": 2763 }, { "epoch": 0.6374630630630631, "grad_norm": 24.13980484008789, "learning_rate": 7.562520367118187e-06, "loss": 12.824411392211914, "loss_alignment": 1.00048828125, "loss_alignment_w": 0.500244140625, "loss_sft": 0.3021818473935127, "loss_total_v6": 0.801525741815567, "step": 2764 }, { "epoch": 0.6376936936936937, "grad_norm": 24.8851261138916, "learning_rate": 7.5607909042908115e-06, "loss": 12.951745986938477, "loss_alignment": 1.0, "loss_alignment_w": 0.5, "loss_sft": 0.3102928288280964, "loss_total_v6": 0.8094840869307518, "step": 2765 }, { "epoch": 0.6379243243243243, "grad_norm": 24.448198318481445, "learning_rate": 7.559061026046066e-06, "loss": 12.597731590270996, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.2982072439044714, "loss_total_v6": 0.7873582169413567, "step": 2766 }, { "epoch": 0.638154954954955, "grad_norm": 28.133825302124023, "learning_rate": 7.557330732664574e-06, "loss": 12.829980850219727, "loss_alignment": 0.9912109375, "loss_alignment_w": 0.49560546875, "loss_sft": 0.3076873905956745, "loss_total_v6": 0.8018737882375717, "step": 2767 }, { "epoch": 0.6383855855855856, "grad_norm": 21.627927780151367, "learning_rate": 7.555600024427028e-06, "loss": 12.59195613861084, "loss_alignment": 0.96826171875, "loss_alignment_w": 0.484130859375, "loss_sft": 0.3029121942818165, "loss_total_v6": 0.7869972810149193, "step": 2768 }, { "epoch": 0.6386162162162162, "grad_norm": 22.816776275634766, "learning_rate": 7.553868901614187e-06, "loss": 13.01413345336914, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.3141004703938961, "loss_total_v6": 0.8133833110332489, "step": 2769 }, { "epoch": 0.6388468468468469, "grad_norm": 22.07181167602539, "learning_rate": 7.5521373645068794e-06, "loss": 12.133879661560059, "loss_alignment": 0.96435546875, "loss_alignment_w": 0.482177734375, "loss_sft": 0.2752131726592779, "loss_total_v6": 0.7583674937486649, "step": 2770 }, { "epoch": 0.6390774774774775, "grad_norm": 22.0741024017334, "learning_rate": 7.550405413385999e-06, "loss": 12.356282234191895, "loss_alignment": 0.9892578125, "loss_alignment_w": 0.49462890625, "loss_sft": 0.27751667238771915, "loss_total_v6": 0.7722676694393158, "step": 2771 }, { "epoch": 0.6393081081081081, "grad_norm": 22.13582992553711, "learning_rate": 7.548673048532504e-06, "loss": 12.544803619384766, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.28867364674806595, "loss_total_v6": 0.7840502485632896, "step": 2772 }, { "epoch": 0.6395387387387388, "grad_norm": 24.347997665405273, "learning_rate": 7.546940270227425e-06, "loss": 12.781914710998535, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.30280646681785583, "loss_total_v6": 0.7988696917891502, "step": 2773 }, { "epoch": 0.6397693693693693, "grad_norm": 22.555072784423828, "learning_rate": 7.545207078751858e-06, "loss": 12.67037296295166, "loss_alignment": 0.984375, "loss_alignment_w": 0.4921875, "loss_sft": 0.2992835193872452, "loss_total_v6": 0.7918982654809952, "step": 2774 }, { "epoch": 0.64, "grad_norm": 20.73798942565918, "learning_rate": 7.543473474386962e-06, "loss": 12.587366104125977, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.29093706980347633, "loss_total_v6": 0.7867103889584541, "step": 2775 }, { "epoch": 0.6402306306306307, "grad_norm": 26.122041702270508, "learning_rate": 7.54173945741397e-06, "loss": 12.94763469696045, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.3107682764530182, "loss_total_v6": 0.8092271238565445, "step": 2776 }, { "epoch": 0.6404612612612612, "grad_norm": 23.420528411865234, "learning_rate": 7.5400050281141765e-06, "loss": 12.768460273742676, "loss_alignment": 0.98388671875, "loss_alignment_w": 0.491943359375, "loss_sft": 0.30603962019085884, "loss_total_v6": 0.7980287596583366, "step": 2777 }, { "epoch": 0.6406918918918919, "grad_norm": 22.963947296142578, "learning_rate": 7.538270186768945e-06, "loss": 12.516061782836914, "loss_alignment": 0.96875, "loss_alignment_w": 0.484375, "loss_sft": 0.29713119752705097, "loss_total_v6": 0.7822538837790489, "step": 2778 }, { "epoch": 0.6409225225225226, "grad_norm": 21.334341049194336, "learning_rate": 7.536534933659705e-06, "loss": 12.69189167022705, "loss_alignment": 0.98095703125, "loss_alignment_w": 0.490478515625, "loss_sft": 0.30252060666680336, "loss_total_v6": 0.7932432368397713, "step": 2779 }, { "epoch": 0.6411531531531531, "grad_norm": 23.998308181762695, "learning_rate": 7.534799269067952e-06, "loss": 12.303504943847656, "loss_alignment": 0.96337890625, "loss_alignment_w": 0.481689453125, "loss_sft": 0.28584523499011993, "loss_total_v6": 0.7689690217375755, "step": 2780 }, { "epoch": 0.6413837837837838, "grad_norm": 24.599435806274414, "learning_rate": 7.5330631932752515e-06, "loss": 12.520318984985352, "loss_alignment": 0.97119140625, "loss_alignment_w": 0.485595703125, "loss_sft": 0.29626812413334846, "loss_total_v6": 0.7825199291110039, "step": 2781 }, { "epoch": 0.6416144144144145, "grad_norm": 23.934417724609375, "learning_rate": 7.531326706563236e-06, "loss": 12.312753677368164, "loss_alignment": 0.9775390625, "loss_alignment_w": 0.48876953125, "loss_sft": 0.2823492493480444, "loss_total_v6": 0.7695471420884132, "step": 2782 }, { "epoch": 0.641845045045045, "grad_norm": 24.222965240478516, "learning_rate": 7.529589809213595e-06, "loss": 12.312005996704102, "loss_alignment": 0.9580078125, "loss_alignment_w": 0.47900390625, "loss_sft": 0.29000818356871605, "loss_total_v6": 0.769500382244587, "step": 2783 }, { "epoch": 0.6420756756756757, "grad_norm": 25.975425720214844, "learning_rate": 7.527852501508099e-06, "loss": 12.279583930969238, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.2732875905930996, "loss_total_v6": 0.7674739882349968, "step": 2784 }, { "epoch": 0.6423063063063063, "grad_norm": 20.28019905090332, "learning_rate": 7.526114783728575e-06, "loss": 12.390422821044922, "loss_alignment": 0.9736328125, "loss_alignment_w": 0.48681640625, "loss_sft": 0.2865016646683216, "loss_total_v6": 0.7744014412164688, "step": 2785 }, { "epoch": 0.6425369369369369, "grad_norm": 21.91490936279297, "learning_rate": 7.52437665615692e-06, "loss": 12.352672576904297, "loss_alignment": 1.00244140625, "loss_alignment_w": 0.501220703125, "loss_sft": 0.2710501402616501, "loss_total_v6": 0.7720420062541962, "step": 2786 }, { "epoch": 0.6427675675675676, "grad_norm": 23.231338500976562, "learning_rate": 7.522638119075097e-06, "loss": 12.781753540039062, "loss_alignment": 0.970703125, "loss_alignment_w": 0.4853515625, "loss_sft": 0.3127756603062153, "loss_total_v6": 0.7988596484065056, "step": 2787 }, { "epoch": 0.6429981981981981, "grad_norm": 21.863018035888672, "learning_rate": 7.520899172765136e-06, "loss": 12.76069450378418, "loss_alignment": 0.9775390625, "loss_alignment_w": 0.48876953125, "loss_sft": 0.3090943116694689, "loss_total_v6": 0.7975434064865112, "step": 2788 }, { "epoch": 0.6432288288288288, "grad_norm": 24.586105346679688, "learning_rate": 7.519159817509132e-06, "loss": 12.608613967895508, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.29482850432395935, "loss_total_v6": 0.7880383729934692, "step": 2789 }, { "epoch": 0.6434594594594595, "grad_norm": 21.552486419677734, "learning_rate": 7.517420053589246e-06, "loss": 12.195682525634766, "loss_alignment": 0.97705078125, "loss_alignment_w": 0.488525390625, "loss_sft": 0.273445351049304, "loss_total_v6": 0.7622301280498505, "step": 2790 }, { "epoch": 0.64369009009009, "grad_norm": 25.471588134765625, "learning_rate": 7.51567988128771e-06, "loss": 12.34603500366211, "loss_alignment": 0.96240234375, "loss_alignment_w": 0.481201171875, "loss_sft": 0.2901361286640167, "loss_total_v6": 0.7716272100806236, "step": 2791 }, { "epoch": 0.6439207207207207, "grad_norm": 24.978160858154297, "learning_rate": 7.513939300886816e-06, "loss": 12.579718589782715, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.2902149986475706, "loss_total_v6": 0.7862324267625809, "step": 2792 }, { "epoch": 0.6441513513513514, "grad_norm": 23.015470504760742, "learning_rate": 7.512198312668926e-06, "loss": 12.978107452392578, "loss_alignment": 1.00146484375, "loss_alignment_w": 0.500732421875, "loss_sft": 0.31134527549147606, "loss_total_v6": 0.8111316710710526, "step": 2793 }, { "epoch": 0.6443819819819819, "grad_norm": 24.21462059020996, "learning_rate": 7.510456916916466e-06, "loss": 12.418258666992188, "loss_alignment": 0.9765625, "loss_alignment_w": 0.48828125, "loss_sft": 0.28789035603404045, "loss_total_v6": 0.7761411145329475, "step": 2794 }, { "epoch": 0.6446126126126126, "grad_norm": 23.700849533081055, "learning_rate": 7.508715113911933e-06, "loss": 12.595989227294922, "loss_alignment": 0.9912109375, "loss_alignment_w": 0.49560546875, "loss_sft": 0.2912776432931423, "loss_total_v6": 0.787249319255352, "step": 2795 }, { "epoch": 0.6448432432432433, "grad_norm": 26.858633041381836, "learning_rate": 7.506972903937879e-06, "loss": 12.276239395141602, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.2796550989151001, "loss_total_v6": 0.7672649770975113, "step": 2796 }, { "epoch": 0.6450738738738738, "grad_norm": 24.086416244506836, "learning_rate": 7.505230287276937e-06, "loss": 12.619622230529785, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.29290734604001045, "loss_total_v6": 0.7887264117598534, "step": 2797 }, { "epoch": 0.6453045045045045, "grad_norm": 24.305906295776367, "learning_rate": 7.503487264211793e-06, "loss": 12.40182113647461, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.27848606184124947, "loss_total_v6": 0.7751138508319855, "step": 2798 }, { "epoch": 0.6455351351351352, "grad_norm": 23.788564682006836, "learning_rate": 7.5017438350252056e-06, "loss": 12.93414306640625, "loss_alignment": 0.97216796875, "loss_alignment_w": 0.486083984375, "loss_sft": 0.32243724912405014, "loss_total_v6": 0.8083839192986488, "step": 2799 }, { "epoch": 0.6457657657657657, "grad_norm": 28.425338745117188, "learning_rate": 7.500000000000001e-06, "loss": 13.378416061401367, "loss_alignment": 0.978515625, "loss_alignment_w": 0.4892578125, "loss_sft": 0.34640491381287575, "loss_total_v6": 0.8361510187387466, "step": 2800 }, { "epoch": 0.6459963963963964, "grad_norm": 23.827192306518555, "learning_rate": 7.4982557594190645e-06, "loss": 12.68177604675293, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.2988060526549816, "loss_total_v6": 0.7926109805703163, "step": 2801 }, { "epoch": 0.6462270270270271, "grad_norm": 27.883378982543945, "learning_rate": 7.49651111356535e-06, "loss": 12.477624893188477, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.28314740397036076, "loss_total_v6": 0.7798515036702156, "step": 2802 }, { "epoch": 0.6464576576576576, "grad_norm": 25.391212463378906, "learning_rate": 7.494766062721881e-06, "loss": 12.647626876831055, "loss_alignment": 0.970703125, "loss_alignment_w": 0.4853515625, "loss_sft": 0.30477415956556797, "loss_total_v6": 0.7904766499996185, "step": 2803 }, { "epoch": 0.6466882882882883, "grad_norm": 25.972532272338867, "learning_rate": 7.4930206071717435e-06, "loss": 12.625452041625977, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.29603347554802895, "loss_total_v6": 0.7890907526016235, "step": 2804 }, { "epoch": 0.646918918918919, "grad_norm": 23.94711685180664, "learning_rate": 7.491274747198086e-06, "loss": 12.738348007202148, "loss_alignment": 0.98583984375, "loss_alignment_w": 0.492919921875, "loss_sft": 0.3033641707152128, "loss_total_v6": 0.7961467504501343, "step": 2805 }, { "epoch": 0.6471495495495495, "grad_norm": 25.139001846313477, "learning_rate": 7.48952848308413e-06, "loss": 12.53288459777832, "loss_alignment": 0.97119140625, "loss_alignment_w": 0.485595703125, "loss_sft": 0.2986556086689234, "loss_total_v6": 0.783305287361145, "step": 2806 }, { "epoch": 0.6473801801801802, "grad_norm": 21.295385360717773, "learning_rate": 7.4877818151131555e-06, "loss": 12.765754699707031, "loss_alignment": 0.99853515625, "loss_alignment_w": 0.499267578125, "loss_sft": 0.2995533440262079, "loss_total_v6": 0.7978596314787865, "step": 2807 }, { "epoch": 0.6476108108108108, "grad_norm": 23.673547744750977, "learning_rate": 7.486034743568512e-06, "loss": 12.572795867919922, "loss_alignment": 0.98681640625, "loss_alignment_w": 0.493408203125, "loss_sft": 0.29152184166014194, "loss_total_v6": 0.7857997864484787, "step": 2808 }, { "epoch": 0.6478414414414414, "grad_norm": 22.615867614746094, "learning_rate": 7.484287268733614e-06, "loss": 12.627544403076172, "loss_alignment": 0.95654296875, "loss_alignment_w": 0.478271484375, "loss_sft": 0.31072110310196877, "loss_total_v6": 0.7892214730381966, "step": 2809 }, { "epoch": 0.6480720720720721, "grad_norm": 23.514698028564453, "learning_rate": 7.482539390891941e-06, "loss": 12.358543395996094, "loss_alignment": 0.97900390625, "loss_alignment_w": 0.489501953125, "loss_sft": 0.28276968374848366, "loss_total_v6": 0.7724089846014977, "step": 2810 }, { "epoch": 0.6483027027027027, "grad_norm": 24.622358322143555, "learning_rate": 7.480791110327035e-06, "loss": 13.202417373657227, "loss_alignment": 0.970703125, "loss_alignment_w": 0.4853515625, "loss_sft": 0.34092873334884644, "loss_total_v6": 0.8251511007547379, "step": 2811 }, { "epoch": 0.6485333333333333, "grad_norm": 23.113483428955078, "learning_rate": 7.479042427322509e-06, "loss": 12.867546081542969, "loss_alignment": 1.0029296875, "loss_alignment_w": 0.50146484375, "loss_sft": 0.3031230289489031, "loss_total_v6": 0.8042216524481773, "step": 2812 }, { "epoch": 0.648763963963964, "grad_norm": 23.22216033935547, "learning_rate": 7.477293342162038e-06, "loss": 12.795930862426758, "loss_alignment": 0.97900390625, "loss_alignment_w": 0.489501953125, "loss_sft": 0.3108235541731119, "loss_total_v6": 0.7997456789016724, "step": 2813 }, { "epoch": 0.6489945945945946, "grad_norm": 25.610492706298828, "learning_rate": 7.475543855129363e-06, "loss": 12.412644386291504, "loss_alignment": 0.96630859375, "loss_alignment_w": 0.483154296875, "loss_sft": 0.2920103594660759, "loss_total_v6": 0.7757902964949608, "step": 2814 }, { "epoch": 0.6492252252252252, "grad_norm": 23.321744918823242, "learning_rate": 7.473793966508288e-06, "loss": 12.665427207946777, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.30141579173505306, "loss_total_v6": 0.7915891632437706, "step": 2815 }, { "epoch": 0.6494558558558559, "grad_norm": 22.964496612548828, "learning_rate": 7.472043676582685e-06, "loss": 12.589620590209961, "loss_alignment": 1.00537109375, "loss_alignment_w": 0.502685546875, "loss_sft": 0.2835554424673319, "loss_total_v6": 0.7868513315916061, "step": 2816 }, { "epoch": 0.6496864864864865, "grad_norm": 23.583805084228516, "learning_rate": 7.47029298563649e-06, "loss": 12.41411018371582, "loss_alignment": 0.97314453125, "loss_alignment_w": 0.486572265625, "loss_sft": 0.28898923471570015, "loss_total_v6": 0.7758819460868835, "step": 2817 }, { "epoch": 0.6499171171171171, "grad_norm": 23.547771453857422, "learning_rate": 7.468541893953703e-06, "loss": 12.228622436523438, "loss_alignment": 0.9716796875, "loss_alignment_w": 0.48583984375, "loss_sft": 0.2780218329280615, "loss_total_v6": 0.7642889320850372, "step": 2818 }, { "epoch": 0.6501477477477478, "grad_norm": 22.226350784301758, "learning_rate": 7.4667904018183935e-06, "loss": 12.719135284423828, "loss_alignment": 0.98046875, "loss_alignment_w": 0.490234375, "loss_sft": 0.3042843136936426, "loss_total_v6": 0.7949459403753281, "step": 2819 }, { "epoch": 0.6503783783783784, "grad_norm": 24.684511184692383, "learning_rate": 7.465038509514688e-06, "loss": 12.573474884033203, "loss_alignment": 0.96875, "loss_alignment_w": 0.484375, "loss_sft": 0.30145194195210934, "loss_total_v6": 0.7858421877026558, "step": 2820 }, { "epoch": 0.650609009009009, "grad_norm": 23.421741485595703, "learning_rate": 7.463286217326786e-06, "loss": 12.479167938232422, "loss_alignment": 0.962890625, "loss_alignment_w": 0.4814453125, "loss_sft": 0.29737358540296555, "loss_total_v6": 0.7799480259418488, "step": 2821 }, { "epoch": 0.6508396396396396, "grad_norm": 26.08751678466797, "learning_rate": 7.461533525538947e-06, "loss": 13.0355224609375, "loss_alignment": 0.97607421875, "loss_alignment_w": 0.488037109375, "loss_sft": 0.32674407586455345, "loss_total_v6": 0.8147201612591743, "step": 2822 }, { "epoch": 0.6510702702702703, "grad_norm": 25.70287322998047, "learning_rate": 7.459780434435496e-06, "loss": 12.60519027709961, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.2989328298717737, "loss_total_v6": 0.7878243997693062, "step": 2823 }, { "epoch": 0.6513009009009009, "grad_norm": 22.63545799255371, "learning_rate": 7.4580269443008245e-06, "loss": 12.523271560668945, "loss_alignment": 0.998046875, "loss_alignment_w": 0.4990234375, "loss_sft": 0.2836810499429703, "loss_total_v6": 0.7827044948935509, "step": 2824 }, { "epoch": 0.6515315315315315, "grad_norm": 24.374473571777344, "learning_rate": 7.4562730554193875e-06, "loss": 12.853164672851562, "loss_alignment": 0.9951171875, "loss_alignment_w": 0.49755859375, "loss_sft": 0.30568788945674896, "loss_total_v6": 0.8033227771520615, "step": 2825 }, { "epoch": 0.6517621621621622, "grad_norm": 23.76016616821289, "learning_rate": 7.454518768075705e-06, "loss": 13.055727005004883, "loss_alignment": 0.994140625, "loss_alignment_w": 0.4970703125, "loss_sft": 0.3194466456770897, "loss_total_v6": 0.8159829080104828, "step": 2826 }, { "epoch": 0.6519927927927928, "grad_norm": 24.4549560546875, "learning_rate": 7.452764082554361e-06, "loss": 11.834482192993164, "loss_alignment": 0.97021484375, "loss_alignment_w": 0.485107421875, "loss_sft": 0.25450197234749794, "loss_total_v6": 0.7396551594138145, "step": 2827 }, { "epoch": 0.6522234234234234, "grad_norm": 24.394113540649414, "learning_rate": 7.451008999140005e-06, "loss": 12.416647911071777, "loss_alignment": 0.98046875, "loss_alignment_w": 0.490234375, "loss_sft": 0.2857603542506695, "loss_total_v6": 0.7760405018925667, "step": 2828 }, { "epoch": 0.6524540540540541, "grad_norm": 46.17145919799805, "learning_rate": 7.449253518117351e-06, "loss": 12.637354850769043, "loss_alignment": 0.974609375, "loss_alignment_w": 0.4873046875, "loss_sft": 0.301828121766448, "loss_total_v6": 0.789834700524807, "step": 2829 }, { "epoch": 0.6526846846846847, "grad_norm": 24.80842399597168, "learning_rate": 7.447497639771175e-06, "loss": 12.444424629211426, "loss_alignment": 0.97021484375, "loss_alignment_w": 0.485107421875, "loss_sft": 0.292241882532835, "loss_total_v6": 0.7777765467762947, "step": 2830 }, { "epoch": 0.6529153153153153, "grad_norm": 25.821138381958008, "learning_rate": 7.445741364386323e-06, "loss": 12.62314224243164, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.29411909729242325, "loss_total_v6": 0.7889463603496552, "step": 2831 }, { "epoch": 0.653145945945946, "grad_norm": 24.71358299255371, "learning_rate": 7.443984692247701e-06, "loss": 12.183578491210938, "loss_alignment": 0.98095703125, "loss_alignment_w": 0.490478515625, "loss_sft": 0.2708273120224476, "loss_total_v6": 0.7614736557006836, "step": 2832 }, { "epoch": 0.6533765765765766, "grad_norm": 22.98895263671875, "learning_rate": 7.442227623640277e-06, "loss": 12.16218376159668, "loss_alignment": 0.97216796875, "loss_alignment_w": 0.486083984375, "loss_sft": 0.27496808022260666, "loss_total_v6": 0.7601365223526955, "step": 2833 }, { "epoch": 0.6536072072072072, "grad_norm": 21.21369171142578, "learning_rate": 7.440470158849091e-06, "loss": 12.292594909667969, "loss_alignment": 0.96923828125, "loss_alignment_w": 0.484619140625, "loss_sft": 0.2828593607991934, "loss_total_v6": 0.7682871893048286, "step": 2834 }, { "epoch": 0.6538378378378379, "grad_norm": 22.17862892150879, "learning_rate": 7.438712298159242e-06, "loss": 12.168207168579102, "loss_alignment": 0.966796875, "loss_alignment_w": 0.4833984375, "loss_sft": 0.27786215394735336, "loss_total_v6": 0.7605129256844521, "step": 2835 }, { "epoch": 0.6540684684684684, "grad_norm": 26.057395935058594, "learning_rate": 7.436954041855893e-06, "loss": 12.711956024169922, "loss_alignment": 0.9580078125, "loss_alignment_w": 0.47900390625, "loss_sft": 0.3156611993908882, "loss_total_v6": 0.7944972217082977, "step": 2836 }, { "epoch": 0.6542990990990991, "grad_norm": 25.046823501586914, "learning_rate": 7.435195390224272e-06, "loss": 12.309433937072754, "loss_alignment": 0.96630859375, "loss_alignment_w": 0.483154296875, "loss_sft": 0.2858801446855068, "loss_total_v6": 0.7693396210670471, "step": 2837 }, { "epoch": 0.6545297297297298, "grad_norm": 25.048383712768555, "learning_rate": 7.433436343549674e-06, "loss": 12.78824234008789, "loss_alignment": 0.9775390625, "loss_alignment_w": 0.48876953125, "loss_sft": 0.31000738590955734, "loss_total_v6": 0.799265168607235, "step": 2838 }, { "epoch": 0.6547603603603603, "grad_norm": 23.80265998840332, "learning_rate": 7.431676902117453e-06, "loss": 12.517087936401367, "loss_alignment": 0.970703125, "loss_alignment_w": 0.4853515625, "loss_sft": 0.2970121931284666, "loss_total_v6": 0.7823179736733437, "step": 2839 }, { "epoch": 0.654990990990991, "grad_norm": 23.07223892211914, "learning_rate": 7.42991706621303e-06, "loss": 12.218183517456055, "loss_alignment": 0.96630859375, "loss_alignment_w": 0.483154296875, "loss_sft": 0.28083314560353756, "loss_total_v6": 0.7636364474892616, "step": 2840 }, { "epoch": 0.6552216216216217, "grad_norm": 24.25544548034668, "learning_rate": 7.428156836121892e-06, "loss": 12.320140838623047, "loss_alignment": 0.966796875, "loss_alignment_w": 0.4833984375, "loss_sft": 0.2863967474550009, "loss_total_v6": 0.770008809864521, "step": 2841 }, { "epoch": 0.6554522522522522, "grad_norm": 27.89842414855957, "learning_rate": 7.426396212129585e-06, "loss": 12.621163368225098, "loss_alignment": 0.96875, "loss_alignment_w": 0.484375, "loss_sft": 0.30469181574881077, "loss_total_v6": 0.7888226732611656, "step": 2842 }, { "epoch": 0.6556828828828829, "grad_norm": 23.354644775390625, "learning_rate": 7.424635194521722e-06, "loss": 12.044425964355469, "loss_alignment": 0.9638671875, "loss_alignment_w": 0.48193359375, "loss_sft": 0.2705684080719948, "loss_total_v6": 0.7527766451239586, "step": 2843 }, { "epoch": 0.6559135135135136, "grad_norm": 22.45591926574707, "learning_rate": 7.422873783583982e-06, "loss": 12.348669052124023, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.28125226125121117, "loss_total_v6": 0.7717918232083321, "step": 2844 }, { "epoch": 0.6561441441441441, "grad_norm": 22.72498321533203, "learning_rate": 7.421111979602099e-06, "loss": 12.263797760009766, "loss_alignment": 0.9619140625, "loss_alignment_w": 0.48095703125, "loss_sft": 0.28435535356402397, "loss_total_v6": 0.7664873376488686, "step": 2845 }, { "epoch": 0.6563747747747748, "grad_norm": 24.49428367614746, "learning_rate": 7.419349782861882e-06, "loss": 12.576702117919922, "loss_alignment": 0.98095703125, "loss_alignment_w": 0.490478515625, "loss_sft": 0.29583999887108803, "loss_total_v6": 0.7860438898205757, "step": 2846 }, { "epoch": 0.6566054054054054, "grad_norm": 24.490034103393555, "learning_rate": 7.417587193649197e-06, "loss": 12.409656524658203, "loss_alignment": 0.9609375, "loss_alignment_w": 0.48046875, "loss_sft": 0.2958214320242405, "loss_total_v6": 0.7756035700440407, "step": 2847 }, { "epoch": 0.656836036036036, "grad_norm": 23.980844497680664, "learning_rate": 7.415824212249977e-06, "loss": 12.61572265625, "loss_alignment": 0.99365234375, "loss_alignment_w": 0.496826171875, "loss_sft": 0.29229734279215336, "loss_total_v6": 0.7884826809167862, "step": 2848 }, { "epoch": 0.6570666666666667, "grad_norm": 25.083724975585938, "learning_rate": 7.414060838950214e-06, "loss": 12.873167037963867, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.31450635753571987, "loss_total_v6": 0.8045728951692581, "step": 2849 }, { "epoch": 0.6572972972972972, "grad_norm": 25.947765350341797, "learning_rate": 7.412297074035968e-06, "loss": 12.848249435424805, "loss_alignment": 0.9912109375, "loss_alignment_w": 0.49560546875, "loss_sft": 0.3060673959553242, "loss_total_v6": 0.8030156046152115, "step": 2850 }, { "epoch": 0.6575279279279279, "grad_norm": 24.26760482788086, "learning_rate": 7.41053291779336e-06, "loss": 12.067444801330566, "loss_alignment": 0.96435546875, "loss_alignment_w": 0.482177734375, "loss_sft": 0.27205283381044865, "loss_total_v6": 0.754215307533741, "step": 2851 }, { "epoch": 0.6577585585585586, "grad_norm": 24.240102767944336, "learning_rate": 7.408768370508577e-06, "loss": 12.692139625549316, "loss_alignment": 0.9716796875, "loss_alignment_w": 0.48583984375, "loss_sft": 0.30601505376398563, "loss_total_v6": 0.7932587042450905, "step": 2852 }, { "epoch": 0.6579891891891891, "grad_norm": 25.114307403564453, "learning_rate": 7.407003432467867e-06, "loss": 12.149616241455078, "loss_alignment": 0.95751953125, "loss_alignment_w": 0.478759765625, "loss_sft": 0.2799808457493782, "loss_total_v6": 0.7593509703874588, "step": 2853 }, { "epoch": 0.6582198198198198, "grad_norm": 22.717275619506836, "learning_rate": 7.405238103957544e-06, "loss": 12.03740119934082, "loss_alignment": 0.962890625, "loss_alignment_w": 0.4814453125, "loss_sft": 0.27055661007761955, "loss_total_v6": 0.7523376196622849, "step": 2854 }, { "epoch": 0.6584504504504505, "grad_norm": 25.757923126220703, "learning_rate": 7.403472385263979e-06, "loss": 12.746423721313477, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.30432668700814247, "loss_total_v6": 0.7966515198349953, "step": 2855 }, { "epoch": 0.658681081081081, "grad_norm": 25.08664894104004, "learning_rate": 7.401706276673616e-06, "loss": 12.413225173950195, "loss_alignment": 0.95361328125, "loss_alignment_w": 0.476806640625, "loss_sft": 0.2988521195948124, "loss_total_v6": 0.7758266106247902, "step": 2856 }, { "epoch": 0.6589117117117117, "grad_norm": 24.6656494140625, "learning_rate": 7.399939778472955e-06, "loss": 13.195109367370605, "loss_alignment": 0.9677734375, "loss_alignment_w": 0.48388671875, "loss_sft": 0.33931222185492516, "loss_total_v6": 0.8246943205595016, "step": 2857 }, { "epoch": 0.6591423423423424, "grad_norm": 25.32415771484375, "learning_rate": 7.39817289094856e-06, "loss": 12.835906982421875, "loss_alignment": 0.99560546875, "loss_alignment_w": 0.497802734375, "loss_sft": 0.30561638250947, "loss_total_v6": 0.802244171500206, "step": 2858 }, { "epoch": 0.6593729729729729, "grad_norm": 36.450706481933594, "learning_rate": 7.396405614387065e-06, "loss": 12.419232368469238, "loss_alignment": 0.962890625, "loss_alignment_w": 0.4814453125, "loss_sft": 0.2954891584813595, "loss_total_v6": 0.7762020453810692, "step": 2859 }, { "epoch": 0.6596036036036036, "grad_norm": 24.84856605529785, "learning_rate": 7.3946379490751545e-06, "loss": 12.799190521240234, "loss_alignment": 0.9677734375, "loss_alignment_w": 0.48388671875, "loss_sft": 0.31557444855570793, "loss_total_v6": 0.7999494150280952, "step": 2860 }, { "epoch": 0.6598342342342343, "grad_norm": 22.367143630981445, "learning_rate": 7.392869895299587e-06, "loss": 12.929604530334473, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.3202004637569189, "loss_total_v6": 0.8081002458930016, "step": 2861 }, { "epoch": 0.6600648648648648, "grad_norm": 25.047447204589844, "learning_rate": 7.3911014533471805e-06, "loss": 12.170001983642578, "loss_alignment": 0.94921875, "loss_alignment_w": 0.474609375, "loss_sft": 0.28624467737972736, "loss_total_v6": 0.7606251686811447, "step": 2862 }, { "epoch": 0.6602954954954955, "grad_norm": 22.754072189331055, "learning_rate": 7.389332623504815e-06, "loss": 12.902843475341797, "loss_alignment": 0.98291015625, "loss_alignment_w": 0.491455078125, "loss_sft": 0.3148810714483261, "loss_total_v6": 0.8064277097582817, "step": 2863 }, { "epoch": 0.6605261261261262, "grad_norm": 24.701740264892578, "learning_rate": 7.387563406059433e-06, "loss": 12.306869506835938, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.27842618338763714, "loss_total_v6": 0.7691793441772461, "step": 2864 }, { "epoch": 0.6607567567567567, "grad_norm": 23.415952682495117, "learning_rate": 7.3857938012980425e-06, "loss": 13.128018379211426, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.3206384629011154, "loss_total_v6": 0.8205011561512947, "step": 2865 }, { "epoch": 0.6609873873873874, "grad_norm": 22.208812713623047, "learning_rate": 7.3840238095077135e-06, "loss": 12.352967262268066, "loss_alignment": 1.0048828125, "loss_alignment_w": 0.50244140625, "loss_sft": 0.2698937691748142, "loss_total_v6": 0.7720604911446571, "step": 2866 }, { "epoch": 0.661218018018018, "grad_norm": 26.077787399291992, "learning_rate": 7.3822534309755756e-06, "loss": 12.868295669555664, "loss_alignment": 0.96728515625, "loss_alignment_w": 0.483642578125, "loss_sft": 0.3199087083339691, "loss_total_v6": 0.804268442094326, "step": 2867 }, { "epoch": 0.6614486486486486, "grad_norm": 22.843204498291016, "learning_rate": 7.380482665988826e-06, "loss": 12.793249130249023, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.3128074575215578, "loss_total_v6": 0.799578070640564, "step": 2868 }, { "epoch": 0.6616792792792793, "grad_norm": 21.977054595947266, "learning_rate": 7.378711514834722e-06, "loss": 12.293265342712402, "loss_alignment": 0.98095703125, "loss_alignment_w": 0.490478515625, "loss_sft": 0.2772248890250921, "loss_total_v6": 0.7683290541172028, "step": 2869 }, { "epoch": 0.6619099099099099, "grad_norm": 25.32178497314453, "learning_rate": 7.376939977800581e-06, "loss": 12.581277847290039, "loss_alignment": 0.97021484375, "loss_alignment_w": 0.485107421875, "loss_sft": 0.30088671669363976, "loss_total_v6": 0.786329872906208, "step": 2870 }, { "epoch": 0.6621405405405405, "grad_norm": 24.962162017822266, "learning_rate": 7.375168055173789e-06, "loss": 12.17145824432373, "loss_alignment": 0.96728515625, "loss_alignment_w": 0.483642578125, "loss_sft": 0.2764021847397089, "loss_total_v6": 0.7607161328196526, "step": 2871 }, { "epoch": 0.6623711711711712, "grad_norm": 26.340436935424805, "learning_rate": 7.373395747241792e-06, "loss": 12.17719841003418, "loss_alignment": 0.953125, "loss_alignment_w": 0.4765625, "loss_sft": 0.28382569923996925, "loss_total_v6": 0.7610748708248138, "step": 2872 }, { "epoch": 0.6626018018018018, "grad_norm": 23.794666290283203, "learning_rate": 7.371623054292095e-06, "loss": 12.154869079589844, "loss_alignment": 0.9833984375, "loss_alignment_w": 0.49169921875, "loss_sft": 0.2695822548121214, "loss_total_v6": 0.7596793174743652, "step": 2873 }, { "epoch": 0.6628324324324324, "grad_norm": 22.429157257080078, "learning_rate": 7.369849976612268e-06, "loss": 12.656888961791992, "loss_alignment": 0.990234375, "loss_alignment_w": 0.4951171875, "loss_sft": 0.29595366306602955, "loss_total_v6": 0.7910556048154831, "step": 2874 }, { "epoch": 0.6630630630630631, "grad_norm": 24.30463409423828, "learning_rate": 7.368076514489947e-06, "loss": 12.789676666259766, "loss_alignment": 0.95751953125, "loss_alignment_w": 0.478759765625, "loss_sft": 0.31937428563833237, "loss_total_v6": 0.7993547767400742, "step": 2875 }, { "epoch": 0.6632936936936937, "grad_norm": 24.61914825439453, "learning_rate": 7.3663026682128255e-06, "loss": 12.468828201293945, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.2916156277060509, "loss_total_v6": 0.7793017700314522, "step": 2876 }, { "epoch": 0.6635243243243243, "grad_norm": 27.908693313598633, "learning_rate": 7.364528438068659e-06, "loss": 12.021188735961914, "loss_alignment": 0.98193359375, "loss_alignment_w": 0.490966796875, "loss_sft": 0.26043378189206123, "loss_total_v6": 0.7513242959976196, "step": 2877 }, { "epoch": 0.663754954954955, "grad_norm": 23.909751892089844, "learning_rate": 7.362753824345271e-06, "loss": 12.301496505737305, "loss_alignment": 0.9453125, "loss_alignment_w": 0.47265625, "loss_sft": 0.29557691514492035, "loss_total_v6": 0.768843524158001, "step": 2878 }, { "epoch": 0.6639855855855856, "grad_norm": 24.12335968017578, "learning_rate": 7.36097882733054e-06, "loss": 12.754412651062012, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.3107463475316763, "loss_total_v6": 0.7971507832407951, "step": 2879 }, { "epoch": 0.6642162162162162, "grad_norm": 23.65043830871582, "learning_rate": 7.35920344731241e-06, "loss": 12.48719596862793, "loss_alignment": 0.9677734375, "loss_alignment_w": 0.48388671875, "loss_sft": 0.2953728921711445, "loss_total_v6": 0.7804497852921486, "step": 2880 }, { "epoch": 0.6644468468468469, "grad_norm": 25.21615982055664, "learning_rate": 7.357427684578893e-06, "loss": 12.32551383972168, "loss_alignment": 0.9931640625, "loss_alignment_w": 0.49658203125, "loss_sft": 0.27388468757271767, "loss_total_v6": 0.7703446373343468, "step": 2881 }, { "epoch": 0.6646774774774775, "grad_norm": 22.530763626098633, "learning_rate": 7.355651539418049e-06, "loss": 12.76544189453125, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.3077277895063162, "loss_total_v6": 0.7978401109576225, "step": 2882 }, { "epoch": 0.6649081081081081, "grad_norm": 24.915395736694336, "learning_rate": 7.353875012118013e-06, "loss": 12.798894882202148, "loss_alignment": 0.97265625, "loss_alignment_w": 0.486328125, "loss_sft": 0.3141826447099447, "loss_total_v6": 0.7999309375882149, "step": 2883 }, { "epoch": 0.6651387387387387, "grad_norm": 27.00406265258789, "learning_rate": 7.352098102966978e-06, "loss": 12.754199981689453, "loss_alignment": 0.9765625, "loss_alignment_w": 0.48828125, "loss_sft": 0.309725983068347, "loss_total_v6": 0.7971374616026878, "step": 2884 }, { "epoch": 0.6653693693693694, "grad_norm": 28.854354858398438, "learning_rate": 7.350320812253196e-06, "loss": 12.668655395507812, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.2947359252721071, "loss_total_v6": 0.7917909845709801, "step": 2885 }, { "epoch": 0.6656, "grad_norm": 26.986595153808594, "learning_rate": 7.348543140264984e-06, "loss": 12.338479995727539, "loss_alignment": 0.96337890625, "loss_alignment_w": 0.481689453125, "loss_sft": 0.2883669249713421, "loss_total_v6": 0.771155022084713, "step": 2886 }, { "epoch": 0.6658306306306306, "grad_norm": 23.573421478271484, "learning_rate": 7.346765087290719e-06, "loss": 12.082195281982422, "loss_alignment": 0.962890625, "loss_alignment_w": 0.4814453125, "loss_sft": 0.2738291621208191, "loss_total_v6": 0.755137175321579, "step": 2887 }, { "epoch": 0.6660612612612613, "grad_norm": 23.896141052246094, "learning_rate": 7.344986653618844e-06, "loss": 12.366188049316406, "loss_alignment": 0.9775390625, "loss_alignment_w": 0.48876953125, "loss_sft": 0.28428502567112446, "loss_total_v6": 0.7728867083787918, "step": 2888 }, { "epoch": 0.6662918918918919, "grad_norm": 23.973434448242188, "learning_rate": 7.343207839537858e-06, "loss": 12.881837844848633, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.3163300231099129, "loss_total_v6": 0.805114820599556, "step": 2889 }, { "epoch": 0.6665225225225225, "grad_norm": 23.98604393005371, "learning_rate": 7.341428645336324e-06, "loss": 12.258932113647461, "loss_alignment": 0.974609375, "loss_alignment_w": 0.4873046875, "loss_sft": 0.27932109870016575, "loss_total_v6": 0.7661832720041275, "step": 2890 }, { "epoch": 0.6667531531531532, "grad_norm": 24.259714126586914, "learning_rate": 7.3396490713028674e-06, "loss": 12.584646224975586, "loss_alignment": 0.98046875, "loss_alignment_w": 0.490234375, "loss_sft": 0.2949632257223129, "loss_total_v6": 0.7865403667092323, "step": 2891 }, { "epoch": 0.6669837837837838, "grad_norm": 23.13737678527832, "learning_rate": 7.337869117726176e-06, "loss": 12.27007007598877, "loss_alignment": 0.96630859375, "loss_alignment_w": 0.483154296875, "loss_sft": 0.2842744328081608, "loss_total_v6": 0.7668794021010399, "step": 2892 }, { "epoch": 0.6672144144144144, "grad_norm": 23.071901321411133, "learning_rate": 7.336088784894997e-06, "loss": 12.552639961242676, "loss_alignment": 0.97216796875, "loss_alignment_w": 0.486083984375, "loss_sft": 0.29960047639906406, "loss_total_v6": 0.7845400273799896, "step": 2893 }, { "epoch": 0.6674450450450451, "grad_norm": 22.079492568969727, "learning_rate": 7.33430807309814e-06, "loss": 11.695676803588867, "loss_alignment": 0.95654296875, "loss_alignment_w": 0.478271484375, "loss_sft": 0.25267775543034077, "loss_total_v6": 0.7309797778725624, "step": 2894 }, { "epoch": 0.6676756756756757, "grad_norm": 24.65489959716797, "learning_rate": 7.332526982624476e-06, "loss": 12.431428909301758, "loss_alignment": 0.9755859375, "loss_alignment_w": 0.48779296875, "loss_sft": 0.28877459466457367, "loss_total_v6": 0.7769642993807793, "step": 2895 }, { "epoch": 0.6679063063063063, "grad_norm": 22.308589935302734, "learning_rate": 7.330745513762937e-06, "loss": 12.758691787719727, "loss_alignment": 0.958984375, "loss_alignment_w": 0.4794921875, "loss_sft": 0.317224133759737, "loss_total_v6": 0.7974182441830635, "step": 2896 }, { "epoch": 0.668136936936937, "grad_norm": 23.28189468383789, "learning_rate": 7.32896366680252e-06, "loss": 12.731426239013672, "loss_alignment": 0.96435546875, "loss_alignment_w": 0.482177734375, "loss_sft": 0.31330755539238453, "loss_total_v6": 0.7957141995429993, "step": 2897 }, { "epoch": 0.6683675675675675, "grad_norm": 24.73931884765625, "learning_rate": 7.327181442032277e-06, "loss": 12.561880111694336, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.29447111673653126, "loss_total_v6": 0.7851174771785736, "step": 2898 }, { "epoch": 0.6685981981981982, "grad_norm": 25.18741798400879, "learning_rate": 7.325398839741325e-06, "loss": 12.781929016113281, "loss_alignment": 0.974609375, "loss_alignment_w": 0.4873046875, "loss_sft": 0.31123014725744724, "loss_total_v6": 0.7988705560564995, "step": 2899 }, { "epoch": 0.6688288288288289, "grad_norm": 22.992774963378906, "learning_rate": 7.323615860218844e-06, "loss": 12.207280158996582, "loss_alignment": 0.9677734375, "loss_alignment_w": 0.48388671875, "loss_sft": 0.2792361155152321, "loss_total_v6": 0.7629550248384476, "step": 2900 }, { "epoch": 0.6690594594594594, "grad_norm": 25.62847900390625, "learning_rate": 7.321832503754071e-06, "loss": 12.507904052734375, "loss_alignment": 0.9873046875, "loss_alignment_w": 0.49365234375, "loss_sft": 0.2891597989946604, "loss_total_v6": 0.781744047999382, "step": 2901 }, { "epoch": 0.6692900900900901, "grad_norm": 25.269817352294922, "learning_rate": 7.3200487706363054e-06, "loss": 12.489738464355469, "loss_alignment": 0.97021484375, "loss_alignment_w": 0.485107421875, "loss_sft": 0.2968288026750088, "loss_total_v6": 0.7806086912751198, "step": 2902 }, { "epoch": 0.6695207207207208, "grad_norm": 23.19713592529297, "learning_rate": 7.3182646611549125e-06, "loss": 12.341083526611328, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.27305724658071995, "loss_total_v6": 0.7713177353143692, "step": 2903 }, { "epoch": 0.6697513513513513, "grad_norm": 22.443981170654297, "learning_rate": 7.31648017559931e-06, "loss": 12.255722999572754, "loss_alignment": 0.96826171875, "loss_alignment_w": 0.484130859375, "loss_sft": 0.28308783285319805, "loss_total_v6": 0.7659827023744583, "step": 2904 }, { "epoch": 0.669981981981982, "grad_norm": 31.763269424438477, "learning_rate": 7.314695314258983e-06, "loss": 12.015974044799805, "loss_alignment": 0.939453125, "loss_alignment_w": 0.4697265625, "loss_sft": 0.28169899992644787, "loss_total_v6": 0.7509983330965042, "step": 2905 }, { "epoch": 0.6702126126126126, "grad_norm": 23.797883987426758, "learning_rate": 7.312910077423477e-06, "loss": 12.476656913757324, "loss_alignment": 0.98095703125, "loss_alignment_w": 0.490478515625, "loss_sft": 0.2895109038800001, "loss_total_v6": 0.7797910422086716, "step": 2906 }, { "epoch": 0.6704432432432432, "grad_norm": 25.968978881835938, "learning_rate": 7.311124465382395e-06, "loss": 12.894265174865723, "loss_alignment": 0.9853515625, "loss_alignment_w": 0.49267578125, "loss_sft": 0.31355152651667595, "loss_total_v6": 0.8058915734291077, "step": 2907 }, { "epoch": 0.6706738738738739, "grad_norm": 24.375459671020508, "learning_rate": 7.309338478425405e-06, "loss": 12.728482246398926, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.3011148292571306, "loss_total_v6": 0.7955301403999329, "step": 2908 }, { "epoch": 0.6709045045045045, "grad_norm": 24.938533782958984, "learning_rate": 7.307552116842233e-06, "loss": 12.71177864074707, "loss_alignment": 0.9814453125, "loss_alignment_w": 0.49072265625, "loss_sft": 0.3049079664051533, "loss_total_v6": 0.7944861948490143, "step": 2909 }, { "epoch": 0.6711351351351351, "grad_norm": 23.708881378173828, "learning_rate": 7.305765380922668e-06, "loss": 12.464508056640625, "loss_alignment": 0.9921875, "loss_alignment_w": 0.49609375, "loss_sft": 0.28396032378077507, "loss_total_v6": 0.7790317088365555, "step": 2910 }, { "epoch": 0.6713657657657658, "grad_norm": 24.227195739746094, "learning_rate": 7.303978270956557e-06, "loss": 12.071555137634277, "loss_alignment": 0.9716796875, "loss_alignment_w": 0.48583984375, "loss_sft": 0.2675947304815054, "loss_total_v6": 0.75447216629982, "step": 2911 }, { "epoch": 0.6715963963963963, "grad_norm": 26.686153411865234, "learning_rate": 7.302190787233808e-06, "loss": 12.235845565795898, "loss_alignment": 0.97021484375, "loss_alignment_w": 0.485107421875, "loss_sft": 0.2802737336605787, "loss_total_v6": 0.7647402957081795, "step": 2912 }, { "epoch": 0.671827027027027, "grad_norm": 24.136388778686523, "learning_rate": 7.300402930044394e-06, "loss": 12.151859283447266, "loss_alignment": 0.974609375, "loss_alignment_w": 0.4873046875, "loss_sft": 0.27226278372108936, "loss_total_v6": 0.7594911754131317, "step": 2913 }, { "epoch": 0.6720576576576577, "grad_norm": 23.612211227416992, "learning_rate": 7.298614699678341e-06, "loss": 12.537408828735352, "loss_alignment": 0.9765625, "loss_alignment_w": 0.48828125, "loss_sft": 0.29616132751107216, "loss_total_v6": 0.7835880815982819, "step": 2914 }, { "epoch": 0.6722882882882882, "grad_norm": 23.769916534423828, "learning_rate": 7.296826096425743e-06, "loss": 12.421648979187012, "loss_alignment": 0.95751953125, "loss_alignment_w": 0.478759765625, "loss_sft": 0.29673884622752666, "loss_total_v6": 0.77635308355093, "step": 2915 }, { "epoch": 0.6725189189189189, "grad_norm": 23.952478408813477, "learning_rate": 7.29503712057675e-06, "loss": 12.858907699584961, "loss_alignment": 0.97265625, "loss_alignment_w": 0.486328125, "loss_sft": 0.31829955615103245, "loss_total_v6": 0.8036816939711571, "step": 2916 }, { "epoch": 0.6727495495495496, "grad_norm": 23.195289611816406, "learning_rate": 7.293247772421577e-06, "loss": 12.863603591918945, "loss_alignment": 0.98193359375, "loss_alignment_w": 0.490966796875, "loss_sft": 0.3125964645296335, "loss_total_v6": 0.8039752617478371, "step": 2917 }, { "epoch": 0.6729801801801801, "grad_norm": 27.740819931030273, "learning_rate": 7.291458052250488e-06, "loss": 12.047163963317871, "loss_alignment": 0.9697265625, "loss_alignment_w": 0.48486328125, "loss_sft": 0.26776405423879623, "loss_total_v6": 0.7529477700591087, "step": 2918 }, { "epoch": 0.6732108108108108, "grad_norm": 23.739089965820312, "learning_rate": 7.289667960353826e-06, "loss": 11.752435684204102, "loss_alignment": 0.9677734375, "loss_alignment_w": 0.48388671875, "loss_sft": 0.25076255202293396, "loss_total_v6": 0.7345272079110146, "step": 2919 }, { "epoch": 0.6734414414414415, "grad_norm": 22.757299423217773, "learning_rate": 7.287877497021978e-06, "loss": 12.854207992553711, "loss_alignment": 0.98291015625, "loss_alignment_w": 0.491455078125, "loss_sft": 0.31124632246792316, "loss_total_v6": 0.8033880293369293, "step": 2920 }, { "epoch": 0.673672072072072, "grad_norm": 27.201066970825195, "learning_rate": 7.286086662545395e-06, "loss": 12.402887344360352, "loss_alignment": 0.982421875, "loss_alignment_w": 0.4912109375, "loss_sft": 0.2841831650584936, "loss_total_v6": 0.7751804739236832, "step": 2921 }, { "epoch": 0.6739027027027027, "grad_norm": 22.8150634765625, "learning_rate": 7.284295457214596e-06, "loss": 12.265596389770508, "loss_alignment": 0.9638671875, "loss_alignment_w": 0.48193359375, "loss_sft": 0.28390319272875786, "loss_total_v6": 0.7665997371077538, "step": 2922 }, { "epoch": 0.6741333333333334, "grad_norm": 24.688457489013672, "learning_rate": 7.282503881320149e-06, "loss": 12.412263870239258, "loss_alignment": 0.97119140625, "loss_alignment_w": 0.485595703125, "loss_sft": 0.2898655813187361, "loss_total_v6": 0.7757664695382118, "step": 2923 }, { "epoch": 0.6743639639639639, "grad_norm": 24.28192901611328, "learning_rate": 7.2807119351526915e-06, "loss": 12.416942596435547, "loss_alignment": 0.9638671875, "loss_alignment_w": 0.48193359375, "loss_sft": 0.2934997230768204, "loss_total_v6": 0.7760588824748993, "step": 2924 }, { "epoch": 0.6745945945945946, "grad_norm": 23.575794219970703, "learning_rate": 7.2789196190029155e-06, "loss": 12.420074462890625, "loss_alignment": 0.9638671875, "loss_alignment_w": 0.48193359375, "loss_sft": 0.29284095019102097, "loss_total_v6": 0.7762546688318253, "step": 2925 }, { "epoch": 0.6748252252252253, "grad_norm": 25.128686904907227, "learning_rate": 7.2771269331615744e-06, "loss": 12.520009994506836, "loss_alignment": 0.96533203125, "loss_alignment_w": 0.482666015625, "loss_sft": 0.30017028003931046, "loss_total_v6": 0.7825005799531937, "step": 2926 }, { "epoch": 0.6750558558558558, "grad_norm": 24.077213287353516, "learning_rate": 7.275333877919483e-06, "loss": 12.537382125854492, "loss_alignment": 0.94677734375, "loss_alignment_w": 0.473388671875, "loss_sft": 0.31149471551179886, "loss_total_v6": 0.783586397767067, "step": 2927 }, { "epoch": 0.6752864864864865, "grad_norm": 22.775012969970703, "learning_rate": 7.273540453567512e-06, "loss": 12.048942565917969, "loss_alignment": 0.96826171875, "loss_alignment_w": 0.484130859375, "loss_sft": 0.2682872097939253, "loss_total_v6": 0.753058910369873, "step": 2928 }, { "epoch": 0.6755171171171171, "grad_norm": 23.73981285095215, "learning_rate": 7.271746660396599e-06, "loss": 11.918807983398438, "loss_alignment": 0.9716796875, "loss_alignment_w": 0.48583984375, "loss_sft": 0.2596807237714529, "loss_total_v6": 0.7449254617094994, "step": 2929 }, { "epoch": 0.6757477477477477, "grad_norm": 24.47640609741211, "learning_rate": 7.269952498697734e-06, "loss": 13.263053894042969, "loss_alignment": 0.9638671875, "loss_alignment_w": 0.48193359375, "loss_sft": 0.3465952165424824, "loss_total_v6": 0.8289408311247826, "step": 2930 }, { "epoch": 0.6759783783783784, "grad_norm": 25.6655216217041, "learning_rate": 7.2681579687619715e-06, "loss": 12.272584915161133, "loss_alignment": 0.98828125, "loss_alignment_w": 0.494140625, "loss_sft": 0.27239236421883106, "loss_total_v6": 0.767036534845829, "step": 2931 }, { "epoch": 0.676209009009009, "grad_norm": 22.77598762512207, "learning_rate": 7.266363070880424e-06, "loss": 12.737041473388672, "loss_alignment": 0.97216796875, "loss_alignment_w": 0.486083984375, "loss_sft": 0.3091571554541588, "loss_total_v6": 0.796065129339695, "step": 2932 }, { "epoch": 0.6764396396396396, "grad_norm": 24.811338424682617, "learning_rate": 7.2645678053442645e-06, "loss": 12.740395545959473, "loss_alignment": 0.95849609375, "loss_alignment_w": 0.479248046875, "loss_sft": 0.31606535613536835, "loss_total_v6": 0.7962746992707253, "step": 2933 }, { "epoch": 0.6766702702702703, "grad_norm": 23.621747970581055, "learning_rate": 7.262772172444723e-06, "loss": 12.55820369720459, "loss_alignment": 0.96142578125, "loss_alignment_w": 0.480712890625, "loss_sft": 0.3046478144824505, "loss_total_v6": 0.7848877087235451, "step": 2934 }, { "epoch": 0.6769009009009009, "grad_norm": 24.548072814941406, "learning_rate": 7.260976172473094e-06, "loss": 12.273098945617676, "loss_alignment": 0.958984375, "loss_alignment_w": 0.4794921875, "loss_sft": 0.28785115107893944, "loss_total_v6": 0.7670686841011047, "step": 2935 }, { "epoch": 0.6771315315315315, "grad_norm": 24.078792572021484, "learning_rate": 7.259179805720726e-06, "loss": 12.341171264648438, "loss_alignment": 0.958984375, "loss_alignment_w": 0.4794921875, "loss_sft": 0.2922430448234081, "loss_total_v6": 0.7713232114911079, "step": 2936 }, { "epoch": 0.6773621621621622, "grad_norm": 25.242691040039062, "learning_rate": 7.257383072479032e-06, "loss": 12.528083801269531, "loss_alignment": 0.97607421875, "loss_alignment_w": 0.488037109375, "loss_sft": 0.29434250481426716, "loss_total_v6": 0.7830052077770233, "step": 2937 }, { "epoch": 0.6775927927927928, "grad_norm": 25.267934799194336, "learning_rate": 7.25558597303948e-06, "loss": 12.731524467468262, "loss_alignment": 0.98876953125, "loss_alignment_w": 0.494384765625, "loss_sft": 0.30055730789899826, "loss_total_v6": 0.7957202792167664, "step": 2938 }, { "epoch": 0.6778234234234234, "grad_norm": 24.456417083740234, "learning_rate": 7.253788507693602e-06, "loss": 12.473621368408203, "loss_alignment": 0.94970703125, "loss_alignment_w": 0.474853515625, "loss_sft": 0.30512930639088154, "loss_total_v6": 0.7796013504266739, "step": 2939 }, { "epoch": 0.6780540540540541, "grad_norm": 24.964448928833008, "learning_rate": 7.251990676732985e-06, "loss": 12.385955810546875, "loss_alignment": 0.95458984375, "loss_alignment_w": 0.477294921875, "loss_sft": 0.2967204786837101, "loss_total_v6": 0.7741222381591797, "step": 2940 }, { "epoch": 0.6782846846846847, "grad_norm": 25.35713005065918, "learning_rate": 7.2501924804492784e-06, "loss": 12.35898494720459, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.28150028735399246, "loss_total_v6": 0.7724365517497063, "step": 2941 }, { "epoch": 0.6785153153153153, "grad_norm": 23.7077693939209, "learning_rate": 7.248393919134189e-06, "loss": 12.250101089477539, "loss_alignment": 0.966796875, "loss_alignment_w": 0.4833984375, "loss_sft": 0.282614354044199, "loss_total_v6": 0.7656312957406044, "step": 2942 }, { "epoch": 0.678745945945946, "grad_norm": 24.510683059692383, "learning_rate": 7.246594993079483e-06, "loss": 12.3048095703125, "loss_alignment": 0.9638671875, "loss_alignment_w": 0.48193359375, "loss_sft": 0.2876663338392973, "loss_total_v6": 0.7690505906939507, "step": 2943 }, { "epoch": 0.6789765765765766, "grad_norm": 27.003477096557617, "learning_rate": 7.24479570257699e-06, "loss": 11.76080322265625, "loss_alignment": 0.94140625, "loss_alignment_w": 0.470703125, "loss_sft": 0.2644539251923561, "loss_total_v6": 0.735050231218338, "step": 2944 }, { "epoch": 0.6792072072072072, "grad_norm": 24.341548919677734, "learning_rate": 7.242996047918589e-06, "loss": 12.163661003112793, "loss_alignment": 0.96044921875, "loss_alignment_w": 0.480224609375, "loss_sft": 0.27852407470345497, "loss_total_v6": 0.760228805243969, "step": 2945 }, { "epoch": 0.6794378378378378, "grad_norm": 25.65957260131836, "learning_rate": 7.2411960293962294e-06, "loss": 12.408575057983398, "loss_alignment": 0.98779296875, "loss_alignment_w": 0.493896484375, "loss_sft": 0.28194460831582546, "loss_total_v6": 0.7755359560251236, "step": 2946 }, { "epoch": 0.6796684684684685, "grad_norm": 24.966629028320312, "learning_rate": 7.239395647301912e-06, "loss": 12.493430137634277, "loss_alignment": 0.95849609375, "loss_alignment_w": 0.479248046875, "loss_sft": 0.3026289641857147, "loss_total_v6": 0.7808394134044647, "step": 2947 }, { "epoch": 0.6798990990990991, "grad_norm": 24.20103645324707, "learning_rate": 7.237594901927699e-06, "loss": 12.64678955078125, "loss_alignment": 0.95458984375, "loss_alignment_w": 0.477294921875, "loss_sft": 0.3126411195844412, "loss_total_v6": 0.7904243171215057, "step": 2948 }, { "epoch": 0.6801297297297297, "grad_norm": 26.16387367248535, "learning_rate": 7.2357937935657105e-06, "loss": 11.919319152832031, "loss_alignment": 0.9482421875, "loss_alignment_w": 0.47412109375, "loss_sft": 0.27083632722496986, "loss_total_v6": 0.7449574545025826, "step": 2949 }, { "epoch": 0.6803603603603604, "grad_norm": 25.060699462890625, "learning_rate": 7.2339923225081296e-06, "loss": 12.379541397094727, "loss_alignment": 0.9599609375, "loss_alignment_w": 0.47998046875, "loss_sft": 0.2943206634372473, "loss_total_v6": 0.7737212926149368, "step": 2950 }, { "epoch": 0.680590990990991, "grad_norm": 26.05463218688965, "learning_rate": 7.232190489047193e-06, "loss": 12.812103271484375, "loss_alignment": 0.9677734375, "loss_alignment_w": 0.48388671875, "loss_sft": 0.31691549718379974, "loss_total_v6": 0.800756424665451, "step": 2951 }, { "epoch": 0.6808216216216216, "grad_norm": 26.3049259185791, "learning_rate": 7.2303882934751965e-06, "loss": 12.336515426635742, "loss_alignment": 0.962890625, "loss_alignment_w": 0.4814453125, "loss_sft": 0.28999892622232437, "loss_total_v6": 0.7710322514176369, "step": 2952 }, { "epoch": 0.6810522522522523, "grad_norm": 31.085721969604492, "learning_rate": 7.228585736084499e-06, "loss": 12.024388313293457, "loss_alignment": 0.95654296875, "loss_alignment_w": 0.478271484375, "loss_sft": 0.2730391751974821, "loss_total_v6": 0.7515242621302605, "step": 2953 }, { "epoch": 0.6812828828828829, "grad_norm": 25.679704666137695, "learning_rate": 7.226782817167515e-06, "loss": 12.508596420288086, "loss_alignment": 0.97119140625, "loss_alignment_w": 0.485595703125, "loss_sft": 0.29559648036956787, "loss_total_v6": 0.781787283718586, "step": 2954 }, { "epoch": 0.6815135135135135, "grad_norm": 23.858211517333984, "learning_rate": 7.224979537016716e-06, "loss": 12.3841552734375, "loss_alignment": 0.9599609375, "loss_alignment_w": 0.47998046875, "loss_sft": 0.2938766349107027, "loss_total_v6": 0.7740097194910049, "step": 2955 }, { "epoch": 0.6817441441441442, "grad_norm": 25.51239585876465, "learning_rate": 7.223175895924638e-06, "loss": 11.831805229187012, "loss_alignment": 0.95556640625, "loss_alignment_w": 0.477783203125, "loss_sft": 0.26095689833164215, "loss_total_v6": 0.7394877895712852, "step": 2956 }, { "epoch": 0.6819747747747748, "grad_norm": 22.945331573486328, "learning_rate": 7.2213718941838704e-06, "loss": 12.231874465942383, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.2755395006388426, "loss_total_v6": 0.7644921541213989, "step": 2957 }, { "epoch": 0.6822054054054054, "grad_norm": 23.383670806884766, "learning_rate": 7.21956753208706e-06, "loss": 12.435657501220703, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.2783272322267294, "loss_total_v6": 0.7772286012768745, "step": 2958 }, { "epoch": 0.6824360360360361, "grad_norm": 25.188720703125, "learning_rate": 7.217762809926917e-06, "loss": 13.059224128723145, "loss_alignment": 0.9716796875, "loss_alignment_w": 0.48583984375, "loss_sft": 0.33014808036386967, "loss_total_v6": 0.8162015154957771, "step": 2959 }, { "epoch": 0.6826666666666666, "grad_norm": 25.055328369140625, "learning_rate": 7.215957727996208e-06, "loss": 12.328994750976562, "loss_alignment": 0.97119140625, "loss_alignment_w": 0.485595703125, "loss_sft": 0.2859277706593275, "loss_total_v6": 0.7705621868371964, "step": 2960 }, { "epoch": 0.6828972972972973, "grad_norm": 25.202011108398438, "learning_rate": 7.214152286587755e-06, "loss": 12.165658950805664, "loss_alignment": 0.94384765625, "loss_alignment_w": 0.471923828125, "loss_sft": 0.288521409034729, "loss_total_v6": 0.7603536918759346, "step": 2961 }, { "epoch": 0.683127927927928, "grad_norm": 27.525049209594727, "learning_rate": 7.212346485994443e-06, "loss": 12.450613975524902, "loss_alignment": 0.94970703125, "loss_alignment_w": 0.474853515625, "loss_sft": 0.30340137146413326, "loss_total_v6": 0.7781633585691452, "step": 2962 }, { "epoch": 0.6833585585585585, "grad_norm": 24.301815032958984, "learning_rate": 7.210540326509213e-06, "loss": 12.295331001281738, "loss_alignment": 0.9775390625, "loss_alignment_w": 0.48876953125, "loss_sft": 0.2806499097496271, "loss_total_v6": 0.7684581652283669, "step": 2963 }, { "epoch": 0.6835891891891892, "grad_norm": 23.33169174194336, "learning_rate": 7.208733808425064e-06, "loss": 12.496688842773438, "loss_alignment": 0.98486328125, "loss_alignment_w": 0.492431640625, "loss_sft": 0.28815366327762604, "loss_total_v6": 0.7810430750250816, "step": 2964 }, { "epoch": 0.6838198198198199, "grad_norm": 24.96976661682129, "learning_rate": 7.206926932035052e-06, "loss": 12.597990036010742, "loss_alignment": 0.970703125, "loss_alignment_w": 0.4853515625, "loss_sft": 0.3024348206818104, "loss_total_v6": 0.787374384701252, "step": 2965 }, { "epoch": 0.6840504504504504, "grad_norm": 24.763568878173828, "learning_rate": 7.205119697632297e-06, "loss": 12.487759590148926, "loss_alignment": 0.953125, "loss_alignment_w": 0.4765625, "loss_sft": 0.30407504737377167, "loss_total_v6": 0.780484989285469, "step": 2966 }, { "epoch": 0.6842810810810811, "grad_norm": 26.097694396972656, "learning_rate": 7.2033121055099675e-06, "loss": 12.314029693603516, "loss_alignment": 0.9541015625, "loss_alignment_w": 0.47705078125, "loss_sft": 0.2926218379288912, "loss_total_v6": 0.7696268334984779, "step": 2967 }, { "epoch": 0.6845117117117117, "grad_norm": 22.811817169189453, "learning_rate": 7.201504155961296e-06, "loss": 12.249113082885742, "loss_alignment": 0.96337890625, "loss_alignment_w": 0.481689453125, "loss_sft": 0.28557380847632885, "loss_total_v6": 0.7655695751309395, "step": 2968 }, { "epoch": 0.6847423423423423, "grad_norm": 23.322154998779297, "learning_rate": 7.199695849279576e-06, "loss": 12.26995849609375, "loss_alignment": 0.94775390625, "loss_alignment_w": 0.473876953125, "loss_sft": 0.2934074494987726, "loss_total_v6": 0.7668724283576012, "step": 2969 }, { "epoch": 0.684972972972973, "grad_norm": 26.337261199951172, "learning_rate": 7.197887185758153e-06, "loss": 12.627945899963379, "loss_alignment": 0.95556640625, "loss_alignment_w": 0.477783203125, "loss_sft": 0.3116160184144974, "loss_total_v6": 0.7892466485500336, "step": 2970 }, { "epoch": 0.6852036036036036, "grad_norm": 22.3640193939209, "learning_rate": 7.196078165690431e-06, "loss": 12.610620498657227, "loss_alignment": 0.99072265625, "loss_alignment_w": 0.495361328125, "loss_sft": 0.2926346007734537, "loss_total_v6": 0.7881638184189796, "step": 2971 }, { "epoch": 0.6854342342342342, "grad_norm": 23.998355865478516, "learning_rate": 7.194268789369875e-06, "loss": 12.846067428588867, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.3138197734951973, "loss_total_v6": 0.8028792217373848, "step": 2972 }, { "epoch": 0.6856648648648649, "grad_norm": 24.55845832824707, "learning_rate": 7.192459057090007e-06, "loss": 12.439680099487305, "loss_alignment": 0.9521484375, "loss_alignment_w": 0.47607421875, "loss_sft": 0.30155837908387184, "loss_total_v6": 0.7774800211191177, "step": 2973 }, { "epoch": 0.6858954954954954, "grad_norm": 21.832622528076172, "learning_rate": 7.190648969144403e-06, "loss": 12.46784782409668, "loss_alignment": 0.95751953125, "loss_alignment_w": 0.478759765625, "loss_sft": 0.3001145217567682, "loss_total_v6": 0.7792404666543007, "step": 2974 }, { "epoch": 0.6861261261261261, "grad_norm": 31.454660415649414, "learning_rate": 7.188838525826702e-06, "loss": 12.729355812072754, "loss_alignment": 0.99462890625, "loss_alignment_w": 0.497314453125, "loss_sft": 0.29851443134248257, "loss_total_v6": 0.7955847159028053, "step": 2975 }, { "epoch": 0.6863567567567568, "grad_norm": 25.316083908081055, "learning_rate": 7.1870277274306e-06, "loss": 12.261096000671387, "loss_alignment": 0.9482421875, "loss_alignment_w": 0.47412109375, "loss_sft": 0.29236522875726223, "loss_total_v6": 0.7663184925913811, "step": 2976 }, { "epoch": 0.6865873873873873, "grad_norm": 22.360151290893555, "learning_rate": 7.185216574249844e-06, "loss": 12.873533248901367, "loss_alignment": 0.97998046875, "loss_alignment_w": 0.489990234375, "loss_sft": 0.3163451123982668, "loss_total_v6": 0.8045958429574966, "step": 2977 }, { "epoch": 0.686818018018018, "grad_norm": 23.586620330810547, "learning_rate": 7.1834050665782474e-06, "loss": 13.451428413391113, "loss_alignment": 0.99755859375, "loss_alignment_w": 0.498779296875, "loss_sft": 0.3426521848887205, "loss_total_v6": 0.8407143205404282, "step": 2978 }, { "epoch": 0.6870486486486487, "grad_norm": 27.036197662353516, "learning_rate": 7.181593204709676e-06, "loss": 12.477331161499023, "loss_alignment": 0.97802734375, "loss_alignment_w": 0.489013671875, "loss_sft": 0.2916282042860985, "loss_total_v6": 0.7798331677913666, "step": 2979 }, { "epoch": 0.6872792792792792, "grad_norm": 24.880714416503906, "learning_rate": 7.179780988938051e-06, "loss": 12.435365676879883, "loss_alignment": 0.9541015625, "loss_alignment_w": 0.47705078125, "loss_sft": 0.30060208961367607, "loss_total_v6": 0.7772103324532509, "step": 2980 }, { "epoch": 0.6875099099099099, "grad_norm": 23.809627532958984, "learning_rate": 7.177968419557359e-06, "loss": 12.539772033691406, "loss_alignment": 0.982421875, "loss_alignment_w": 0.4912109375, "loss_sft": 0.29287574999034405, "loss_total_v6": 0.7837357223033905, "step": 2981 }, { "epoch": 0.6877405405405406, "grad_norm": 25.389781951904297, "learning_rate": 7.176155496861639e-06, "loss": 12.912806510925293, "loss_alignment": 0.99951171875, "loss_alignment_w": 0.499755859375, "loss_sft": 0.3065468519926071, "loss_total_v6": 0.8070503920316696, "step": 2982 }, { "epoch": 0.6879711711711711, "grad_norm": 23.358062744140625, "learning_rate": 7.174342221144983e-06, "loss": 12.260498046875, "loss_alignment": 0.96826171875, "loss_alignment_w": 0.484130859375, "loss_sft": 0.28169249929487705, "loss_total_v6": 0.7662811204791069, "step": 2983 }, { "epoch": 0.6882018018018018, "grad_norm": 26.209535598754883, "learning_rate": 7.17252859270155e-06, "loss": 12.51815414428711, "loss_alignment": 0.9716796875, "loss_alignment_w": 0.48583984375, "loss_sft": 0.2968499790877104, "loss_total_v6": 0.7823846489191055, "step": 2984 }, { "epoch": 0.6884324324324325, "grad_norm": 24.376474380493164, "learning_rate": 7.170714611825547e-06, "loss": 12.012666702270508, "loss_alignment": 0.947265625, "loss_alignment_w": 0.4736328125, "loss_sft": 0.2774639278650284, "loss_total_v6": 0.750791609287262, "step": 2985 }, { "epoch": 0.688663063063063, "grad_norm": 23.38321304321289, "learning_rate": 7.168900278811243e-06, "loss": 12.41131591796875, "loss_alignment": 0.97265625, "loss_alignment_w": 0.486328125, "loss_sft": 0.29139323718845844, "loss_total_v6": 0.7757072225213051, "step": 2986 }, { "epoch": 0.6888936936936937, "grad_norm": 24.06397247314453, "learning_rate": 7.167085593952964e-06, "loss": 12.78561782836914, "loss_alignment": 0.95166015625, "loss_alignment_w": 0.475830078125, "loss_sft": 0.324339110404253, "loss_total_v6": 0.7991010546684265, "step": 2987 }, { "epoch": 0.6891243243243244, "grad_norm": 26.289113998413086, "learning_rate": 7.165270557545094e-06, "loss": 12.69450855255127, "loss_alignment": 0.96533203125, "loss_alignment_w": 0.482666015625, "loss_sft": 0.31029830127954483, "loss_total_v6": 0.7934067994356155, "step": 2988 }, { "epoch": 0.6893549549549549, "grad_norm": 26.015653610229492, "learning_rate": 7.163455169882068e-06, "loss": 12.34361743927002, "loss_alignment": 0.96142578125, "loss_alignment_w": 0.480712890625, "loss_sft": 0.2896645702421665, "loss_total_v6": 0.7714761048555374, "step": 2989 }, { "epoch": 0.6895855855855856, "grad_norm": 22.394201278686523, "learning_rate": 7.161639431258387e-06, "loss": 12.459516525268555, "loss_alignment": 0.97119140625, "loss_alignment_w": 0.485595703125, "loss_sft": 0.2921932712197304, "loss_total_v6": 0.7787197977304459, "step": 2990 }, { "epoch": 0.6898162162162162, "grad_norm": 22.470422744750977, "learning_rate": 7.159823341968601e-06, "loss": 12.568863868713379, "loss_alignment": 0.97607421875, "loss_alignment_w": 0.488037109375, "loss_sft": 0.29768473468720913, "loss_total_v6": 0.7855539992451668, "step": 2991 }, { "epoch": 0.6900468468468468, "grad_norm": 23.827356338500977, "learning_rate": 7.158006902307322e-06, "loss": 12.632058143615723, "loss_alignment": 0.96875, "loss_alignment_w": 0.484375, "loss_sft": 0.3051438834518194, "loss_total_v6": 0.7895036414265633, "step": 2992 }, { "epoch": 0.6902774774774775, "grad_norm": 24.096126556396484, "learning_rate": 7.156190112569215e-06, "loss": 12.652124404907227, "loss_alignment": 0.96923828125, "loss_alignment_w": 0.484619140625, "loss_sft": 0.30610810965299606, "loss_total_v6": 0.7907577753067017, "step": 2993 }, { "epoch": 0.6905081081081081, "grad_norm": 22.11733055114746, "learning_rate": 7.154372973049007e-06, "loss": 12.363816261291504, "loss_alignment": 0.96826171875, "loss_alignment_w": 0.484130859375, "loss_sft": 0.28866869397461414, "loss_total_v6": 0.7727385237812996, "step": 2994 }, { "epoch": 0.6907387387387387, "grad_norm": 21.745317459106445, "learning_rate": 7.1525554840414765e-06, "loss": 12.561665534973145, "loss_alignment": 0.94921875, "loss_alignment_w": 0.474609375, "loss_sft": 0.3097470849752426, "loss_total_v6": 0.7851041257381439, "step": 2995 }, { "epoch": 0.6909693693693694, "grad_norm": 23.405906677246094, "learning_rate": 7.15073764584146e-06, "loss": 12.246532440185547, "loss_alignment": 0.95703125, "loss_alignment_w": 0.478515625, "loss_sft": 0.28651116229593754, "loss_total_v6": 0.7654082626104355, "step": 2996 }, { "epoch": 0.6912, "grad_norm": 24.343975067138672, "learning_rate": 7.148919458743853e-06, "loss": 12.745601654052734, "loss_alignment": 0.970703125, "loss_alignment_w": 0.4853515625, "loss_sft": 0.31117225252091885, "loss_total_v6": 0.7966001257300377, "step": 2997 }, { "epoch": 0.6914306306306306, "grad_norm": 24.214204788208008, "learning_rate": 7.147100923043606e-06, "loss": 12.8651123046875, "loss_alignment": 0.96142578125, "loss_alignment_w": 0.480712890625, "loss_sft": 0.32247161492705345, "loss_total_v6": 0.804069496691227, "step": 2998 }, { "epoch": 0.6916612612612613, "grad_norm": 27.159881591796875, "learning_rate": 7.1452820390357225e-06, "loss": 12.10367202758789, "loss_alignment": 0.97705078125, "loss_alignment_w": 0.488525390625, "loss_sft": 0.26754208467900753, "loss_total_v6": 0.7564794793725014, "step": 2999 }, { "epoch": 0.6918918918918919, "grad_norm": 24.014711380004883, "learning_rate": 7.143462807015271e-06, "loss": 12.478660583496094, "loss_alignment": 0.9619140625, "loss_alignment_w": 0.48095703125, "loss_sft": 0.29844045266509056, "loss_total_v6": 0.7799162492156029, "step": 3000 }, { "epoch": 0.6918918918918919, "eval_loss": 0.9214053750038147, "eval_loss_alignment": 0.9706629922945206, "eval_loss_alignment_w": 0.4853314961472603, "eval_loss_contrast": 0.0, "eval_loss_highlevel": 0.0, "eval_loss_sft": 0.436073868922447, "eval_loss_total_v6": 0.9214053672470458, "eval_loss_wm_recon": 0.0, "eval_loss_wm_recon_w": 0.0, "eval_runtime": 297.8439, "eval_samples_per_second": 5.859, "eval_steps_per_second": 0.735, "step": 3000 }, { "epoch": 0.6921225225225225, "grad_norm": 22.54041290283203, "learning_rate": 7.141643227277371e-06, "loss": 12.53954029083252, "loss_alignment": 0.93701171875, "loss_alignment_w": 0.468505859375, "loss_sft": 0.31387266144156456, "loss_total_v6": 0.7837212979793549, "step": 3001 }, { "epoch": 0.6923531531531532, "grad_norm": 25.360082626342773, "learning_rate": 7.139823300117194e-06, "loss": 11.743752479553223, "loss_alignment": 0.9541015625, "loss_alignment_w": 0.47705078125, "loss_sft": 0.25624709762632847, "loss_total_v6": 0.7339845150709152, "step": 3002 }, { "epoch": 0.6925837837837838, "grad_norm": 23.94280433654785, "learning_rate": 7.1380030258299775e-06, "loss": 12.059821128845215, "loss_alignment": 0.95263671875, "loss_alignment_w": 0.476318359375, "loss_sft": 0.2763370852917433, "loss_total_v6": 0.7537388354539871, "step": 3003 }, { "epoch": 0.6928144144144144, "grad_norm": 24.003660202026367, "learning_rate": 7.1361824047110084e-06, "loss": 13.100013732910156, "loss_alignment": 0.96240234375, "loss_alignment_w": 0.481201171875, "loss_sft": 0.33751918748021126, "loss_total_v6": 0.8187508657574654, "step": 3004 }, { "epoch": 0.693045045045045, "grad_norm": 21.819059371948242, "learning_rate": 7.134361437055633e-06, "loss": 12.385110855102539, "loss_alignment": 0.970703125, "loss_alignment_w": 0.4853515625, "loss_sft": 0.28836692310869694, "loss_total_v6": 0.7740694209933281, "step": 3005 }, { "epoch": 0.6932756756756757, "grad_norm": 25.64296531677246, "learning_rate": 7.132540123159253e-06, "loss": 12.288347244262695, "loss_alignment": 0.958984375, "loss_alignment_w": 0.4794921875, "loss_sft": 0.28833117708563805, "loss_total_v6": 0.768021710216999, "step": 3006 }, { "epoch": 0.6935063063063063, "grad_norm": 23.741832733154297, "learning_rate": 7.1307184633173256e-06, "loss": 12.441186904907227, "loss_alignment": 0.951171875, "loss_alignment_w": 0.4755859375, "loss_sft": 0.30241551250219345, "loss_total_v6": 0.777574211359024, "step": 3007 }, { "epoch": 0.6937369369369369, "grad_norm": 22.5711727142334, "learning_rate": 7.128896457825364e-06, "loss": 12.118429183959961, "loss_alignment": 0.962890625, "loss_alignment_w": 0.4814453125, "loss_sft": 0.2752088364213705, "loss_total_v6": 0.7574018239974976, "step": 3008 }, { "epoch": 0.6939675675675676, "grad_norm": 34.925750732421875, "learning_rate": 7.127074106978939e-06, "loss": 12.468620300292969, "loss_alignment": 0.947265625, "loss_alignment_w": 0.4736328125, "loss_sft": 0.30556444451212883, "loss_total_v6": 0.7792887836694717, "step": 3009 }, { "epoch": 0.6941981981981982, "grad_norm": 24.288606643676758, "learning_rate": 7.125251411073677e-06, "loss": 12.18994140625, "loss_alignment": 0.94677734375, "loss_alignment_w": 0.473388671875, "loss_sft": 0.28775029443204403, "loss_total_v6": 0.761871375143528, "step": 3010 }, { "epoch": 0.6944288288288288, "grad_norm": 25.90190315246582, "learning_rate": 7.123428370405258e-06, "loss": 12.55066967010498, "loss_alignment": 0.9794921875, "loss_alignment_w": 0.48974609375, "loss_sft": 0.29418250918388367, "loss_total_v6": 0.7844168692827225, "step": 3011 }, { "epoch": 0.6946594594594595, "grad_norm": 25.416118621826172, "learning_rate": 7.121604985269423e-06, "loss": 12.408397674560547, "loss_alignment": 0.95947265625, "loss_alignment_w": 0.479736328125, "loss_sft": 0.2960631903260946, "loss_total_v6": 0.7755248621106148, "step": 3012 }, { "epoch": 0.6948900900900901, "grad_norm": 24.998559951782227, "learning_rate": 7.119781255961963e-06, "loss": 11.990535736083984, "loss_alignment": 0.94091796875, "loss_alignment_w": 0.470458984375, "loss_sft": 0.2789494637399912, "loss_total_v6": 0.7494084686040878, "step": 3013 }, { "epoch": 0.6951207207207207, "grad_norm": 29.231412887573242, "learning_rate": 7.1179571827787275e-06, "loss": 12.391523361206055, "loss_alignment": 0.9755859375, "loss_alignment_w": 0.48779296875, "loss_sft": 0.2864788994193077, "loss_total_v6": 0.7744702473282814, "step": 3014 }, { "epoch": 0.6953513513513514, "grad_norm": 23.31269645690918, "learning_rate": 7.116132766015622e-06, "loss": 11.991302490234375, "loss_alignment": 0.951171875, "loss_alignment_w": 0.4755859375, "loss_sft": 0.27217676490545273, "loss_total_v6": 0.749456450343132, "step": 3015 }, { "epoch": 0.695581981981982, "grad_norm": 22.458040237426758, "learning_rate": 7.11430800596861e-06, "loss": 12.966591835021973, "loss_alignment": 0.96337890625, "loss_alignment_w": 0.481689453125, "loss_sft": 0.3292413204908371, "loss_total_v6": 0.8104119747877121, "step": 3016 }, { "epoch": 0.6958126126126126, "grad_norm": 25.208341598510742, "learning_rate": 7.1124829029337066e-06, "loss": 12.012802124023438, "loss_alignment": 0.9375, "loss_alignment_w": 0.46875, "loss_sft": 0.28163812682032585, "loss_total_v6": 0.7508001327514648, "step": 3017 }, { "epoch": 0.6960432432432433, "grad_norm": 23.69672393798828, "learning_rate": 7.110657457206982e-06, "loss": 13.029825210571289, "loss_alignment": 0.9736328125, "loss_alignment_w": 0.48681640625, "loss_sft": 0.32815802097320557, "loss_total_v6": 0.814364068210125, "step": 3018 }, { "epoch": 0.6962738738738738, "grad_norm": 26.11501693725586, "learning_rate": 7.108831669084569e-06, "loss": 12.25575065612793, "loss_alignment": 0.9404296875, "loss_alignment_w": 0.47021484375, "loss_sft": 0.2956475615501404, "loss_total_v6": 0.7659844607114792, "step": 3019 }, { "epoch": 0.6965045045045045, "grad_norm": 27.157649993896484, "learning_rate": 7.107005538862647e-06, "loss": 12.544737815856934, "loss_alignment": 0.93310546875, "loss_alignment_w": 0.466552734375, "loss_sft": 0.31734079867601395, "loss_total_v6": 0.7840461283922195, "step": 3020 }, { "epoch": 0.6967351351351352, "grad_norm": 28.503877639770508, "learning_rate": 7.105179066837456e-06, "loss": 11.997400283813477, "loss_alignment": 0.9619140625, "loss_alignment_w": 0.48095703125, "loss_sft": 0.26852953247725964, "loss_total_v6": 0.7498375326395035, "step": 3021 }, { "epoch": 0.6969657657657657, "grad_norm": 23.00932502746582, "learning_rate": 7.103352253305291e-06, "loss": 12.028735160827637, "loss_alignment": 0.97021484375, "loss_alignment_w": 0.485107421875, "loss_sft": 0.26902315951883793, "loss_total_v6": 0.7517959624528885, "step": 3022 }, { "epoch": 0.6971963963963964, "grad_norm": 21.971519470214844, "learning_rate": 7.101525098562503e-06, "loss": 12.200943946838379, "loss_alignment": 0.97021484375, "loss_alignment_w": 0.485107421875, "loss_sft": 0.27768045105040073, "loss_total_v6": 0.7625589817762375, "step": 3023 }, { "epoch": 0.6974270270270271, "grad_norm": 23.386924743652344, "learning_rate": 7.099697602905494e-06, "loss": 12.313176155090332, "loss_alignment": 0.923828125, "loss_alignment_w": 0.4619140625, "loss_sft": 0.30808666348457336, "loss_total_v6": 0.7695734947919846, "step": 3024 }, { "epoch": 0.6976576576576576, "grad_norm": 22.976747512817383, "learning_rate": 7.09786976663073e-06, "loss": 11.77659797668457, "loss_alignment": 0.9736328125, "loss_alignment_w": 0.48681640625, "loss_sft": 0.24932780675590038, "loss_total_v6": 0.736037403345108, "step": 3025 }, { "epoch": 0.6978882882882883, "grad_norm": 24.209556579589844, "learning_rate": 7.0960415900347205e-06, "loss": 12.45799446105957, "loss_alignment": 0.95556640625, "loss_alignment_w": 0.477783203125, "loss_sft": 0.2990256678313017, "loss_total_v6": 0.7786246612668037, "step": 3026 }, { "epoch": 0.698118918918919, "grad_norm": 23.786823272705078, "learning_rate": 7.094213073414039e-06, "loss": 12.192468643188477, "loss_alignment": 0.958984375, "loss_alignment_w": 0.4794921875, "loss_sft": 0.282247232273221, "loss_total_v6": 0.7620293274521828, "step": 3027 }, { "epoch": 0.6983495495495495, "grad_norm": 24.952213287353516, "learning_rate": 7.092384217065315e-06, "loss": 12.075170516967773, "loss_alignment": 0.9306640625, "loss_alignment_w": 0.46533203125, "loss_sft": 0.289121987298131, "loss_total_v6": 0.7546981647610664, "step": 3028 }, { "epoch": 0.6985801801801802, "grad_norm": 23.14744758605957, "learning_rate": 7.090555021285224e-06, "loss": 12.640600204467773, "loss_alignment": 0.96923828125, "loss_alignment_w": 0.484619140625, "loss_sft": 0.3066543899476528, "loss_total_v6": 0.7900375574827194, "step": 3029 }, { "epoch": 0.6988108108108108, "grad_norm": 28.22916603088379, "learning_rate": 7.088725486370506e-06, "loss": 12.149904251098633, "loss_alignment": 0.9345703125, "loss_alignment_w": 0.46728515625, "loss_sft": 0.29148873314261436, "loss_total_v6": 0.7593690231442451, "step": 3030 }, { "epoch": 0.6990414414414414, "grad_norm": 23.997272491455078, "learning_rate": 7.086895612617951e-06, "loss": 12.346378326416016, "loss_alignment": 0.97216796875, "loss_alignment_w": 0.486083984375, "loss_sft": 0.2846338413655758, "loss_total_v6": 0.7716486304998398, "step": 3031 }, { "epoch": 0.6992720720720721, "grad_norm": 23.962568283081055, "learning_rate": 7.085065400324407e-06, "loss": 12.108587265014648, "loss_alignment": 0.935546875, "loss_alignment_w": 0.4677734375, "loss_sft": 0.28834180533885956, "loss_total_v6": 0.756786659359932, "step": 3032 }, { "epoch": 0.6995027027027027, "grad_norm": 24.542644500732422, "learning_rate": 7.083234849786773e-06, "loss": 12.168968200683594, "loss_alignment": 0.9462890625, "loss_alignment_w": 0.47314453125, "loss_sft": 0.2875991351902485, "loss_total_v6": 0.7605605646967888, "step": 3033 }, { "epoch": 0.6997333333333333, "grad_norm": 26.24909019470215, "learning_rate": 7.081403961302007e-06, "loss": 12.644652366638184, "loss_alignment": 0.95556640625, "loss_alignment_w": 0.477783203125, "loss_sft": 0.31217185594141483, "loss_total_v6": 0.7902907580137253, "step": 3034 }, { "epoch": 0.699963963963964, "grad_norm": 24.52322006225586, "learning_rate": 7.079572735167119e-06, "loss": 13.263384819030762, "loss_alignment": 0.9833984375, "loss_alignment_w": 0.49169921875, "loss_sft": 0.3361331671476364, "loss_total_v6": 0.828961543738842, "step": 3035 }, { "epoch": 0.7001945945945945, "grad_norm": 23.93297576904297, "learning_rate": 7.0777411716791734e-06, "loss": 12.466584205627441, "loss_alignment": 0.9638671875, "loss_alignment_w": 0.48193359375, "loss_sft": 0.29733477532863617, "loss_total_v6": 0.7791615501046181, "step": 3036 }, { "epoch": 0.7004252252252252, "grad_norm": 35.47918701171875, "learning_rate": 7.075909271135293e-06, "loss": 12.186305046081543, "loss_alignment": 0.97412109375, "loss_alignment_w": 0.487060546875, "loss_sft": 0.27397317066788673, "loss_total_v6": 0.7616440802812576, "step": 3037 }, { "epoch": 0.7006558558558559, "grad_norm": 26.886138916015625, "learning_rate": 7.074077033832653e-06, "loss": 12.15757942199707, "loss_alignment": 0.9462890625, "loss_alignment_w": 0.47314453125, "loss_sft": 0.2857581824064255, "loss_total_v6": 0.7598487362265587, "step": 3038 }, { "epoch": 0.7008864864864864, "grad_norm": 27.21884536743164, "learning_rate": 7.072244460068481e-06, "loss": 12.417257308959961, "loss_alignment": 0.95458984375, "loss_alignment_w": 0.477294921875, "loss_sft": 0.2997296452522278, "loss_total_v6": 0.776078537106514, "step": 3039 }, { "epoch": 0.7011171171171171, "grad_norm": 26.91739273071289, "learning_rate": 7.07041155014006e-06, "loss": 13.009422302246094, "loss_alignment": 0.98388671875, "loss_alignment_w": 0.491943359375, "loss_sft": 0.3194822929799557, "loss_total_v6": 0.8130888640880585, "step": 3040 }, { "epoch": 0.7013477477477478, "grad_norm": 22.471851348876953, "learning_rate": 7.068578304344734e-06, "loss": 12.197074890136719, "loss_alignment": 0.9775390625, "loss_alignment_w": 0.48876953125, "loss_sft": 0.2737917881458998, "loss_total_v6": 0.7623171806335449, "step": 3041 }, { "epoch": 0.7015783783783783, "grad_norm": 21.93410873413086, "learning_rate": 7.066744722979893e-06, "loss": 12.328256607055664, "loss_alignment": 0.953125, "loss_alignment_w": 0.4765625, "loss_sft": 0.2952048033475876, "loss_total_v6": 0.7705160826444626, "step": 3042 }, { "epoch": 0.701809009009009, "grad_norm": 25.211376190185547, "learning_rate": 7.0649108063429835e-06, "loss": 12.72784423828125, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.3074684180319309, "loss_total_v6": 0.7954902499914169, "step": 3043 }, { "epoch": 0.7020396396396397, "grad_norm": 24.664175033569336, "learning_rate": 7.0630765547315125e-06, "loss": 11.881482124328613, "loss_alignment": 0.9580078125, "loss_alignment_w": 0.47900390625, "loss_sft": 0.2632072661072016, "loss_total_v6": 0.7425926625728607, "step": 3044 }, { "epoch": 0.7022702702702702, "grad_norm": 23.078596115112305, "learning_rate": 7.061241968443033e-06, "loss": 12.54932975769043, "loss_alignment": 0.94873046875, "loss_alignment_w": 0.474365234375, "loss_sft": 0.3094949256628752, "loss_total_v6": 0.7843331694602966, "step": 3045 }, { "epoch": 0.7025009009009009, "grad_norm": 23.44065284729004, "learning_rate": 7.059407047775154e-06, "loss": 12.621601104736328, "loss_alignment": 0.98046875, "loss_alignment_w": 0.490234375, "loss_sft": 0.29876826889812946, "loss_total_v6": 0.7888500764966011, "step": 3046 }, { "epoch": 0.7027315315315316, "grad_norm": 24.060657501220703, "learning_rate": 7.057571793025545e-06, "loss": 12.104084014892578, "loss_alignment": 0.9501953125, "loss_alignment_w": 0.47509765625, "loss_sft": 0.28226208314299583, "loss_total_v6": 0.7565052658319473, "step": 3047 }, { "epoch": 0.7029621621621621, "grad_norm": 24.68221092224121, "learning_rate": 7.055736204491922e-06, "loss": 12.813457489013672, "loss_alignment": 0.96337890625, "loss_alignment_w": 0.481689453125, "loss_sft": 0.3192279450595379, "loss_total_v6": 0.8008411079645157, "step": 3048 }, { "epoch": 0.7031927927927928, "grad_norm": 25.33338165283203, "learning_rate": 7.053900282472059e-06, "loss": 12.14004898071289, "loss_alignment": 0.955078125, "loss_alignment_w": 0.4775390625, "loss_sft": 0.28090887144207954, "loss_total_v6": 0.7587530761957169, "step": 3049 }, { "epoch": 0.7034234234234235, "grad_norm": 25.68490219116211, "learning_rate": 7.052064027263785e-06, "loss": 12.102859497070312, "loss_alignment": 0.93115234375, "loss_alignment_w": 0.465576171875, "loss_sft": 0.29053214006125927, "loss_total_v6": 0.7564287483692169, "step": 3050 }, { "epoch": 0.703654054054054, "grad_norm": 22.490951538085938, "learning_rate": 7.050227439164979e-06, "loss": 12.77453899383545, "loss_alignment": 0.96240234375, "loss_alignment_w": 0.481201171875, "loss_sft": 0.3172838054597378, "loss_total_v6": 0.7984086871147156, "step": 3051 }, { "epoch": 0.7038846846846847, "grad_norm": 23.341812133789062, "learning_rate": 7.048390518473579e-06, "loss": 12.536994934082031, "loss_alignment": 0.94775390625, "loss_alignment_w": 0.473876953125, "loss_sft": 0.30922747775912285, "loss_total_v6": 0.7835622057318687, "step": 3052 }, { "epoch": 0.7041153153153153, "grad_norm": 24.941608428955078, "learning_rate": 7.046553265487573e-06, "loss": 12.45206356048584, "loss_alignment": 0.99169921875, "loss_alignment_w": 0.495849609375, "loss_sft": 0.2836708463728428, "loss_total_v6": 0.7782539948821068, "step": 3053 }, { "epoch": 0.7043459459459459, "grad_norm": 29.050458908081055, "learning_rate": 7.044715680505005e-06, "loss": 12.170053482055664, "loss_alignment": 0.94921875, "loss_alignment_w": 0.474609375, "loss_sft": 0.2859426848590374, "loss_total_v6": 0.7606283500790596, "step": 3054 }, { "epoch": 0.7045765765765766, "grad_norm": 23.816070556640625, "learning_rate": 7.042877763823973e-06, "loss": 13.113420486450195, "loss_alignment": 0.970703125, "loss_alignment_w": 0.4853515625, "loss_sft": 0.3355647474527359, "loss_total_v6": 0.8195887580513954, "step": 3055 }, { "epoch": 0.7048072072072072, "grad_norm": 26.154611587524414, "learning_rate": 7.041039515742626e-06, "loss": 12.420248985290527, "loss_alignment": 0.9580078125, "loss_alignment_w": 0.47900390625, "loss_sft": 0.29666654393076897, "loss_total_v6": 0.7762655690312386, "step": 3056 }, { "epoch": 0.7050378378378378, "grad_norm": 22.896907806396484, "learning_rate": 7.039200936559171e-06, "loss": 12.668392181396484, "loss_alignment": 0.955078125, "loss_alignment_w": 0.4775390625, "loss_sft": 0.3128468692302704, "loss_total_v6": 0.7917745038866997, "step": 3057 }, { "epoch": 0.7052684684684685, "grad_norm": 26.020540237426758, "learning_rate": 7.037362026571863e-06, "loss": 12.16384220123291, "loss_alignment": 0.9716796875, "loss_alignment_w": 0.48583984375, "loss_sft": 0.2749953866004944, "loss_total_v6": 0.7602401524782181, "step": 3058 }, { "epoch": 0.7054990990990991, "grad_norm": 26.301301956176758, "learning_rate": 7.035522786079019e-06, "loss": 12.769454002380371, "loss_alignment": 0.97802734375, "loss_alignment_w": 0.489013671875, "loss_sft": 0.3076886162161827, "loss_total_v6": 0.7980908527970314, "step": 3059 }, { "epoch": 0.7057297297297297, "grad_norm": 24.29932975769043, "learning_rate": 7.033683215379002e-06, "loss": 12.438023567199707, "loss_alignment": 0.95703125, "loss_alignment_w": 0.478515625, "loss_sft": 0.2995779775083065, "loss_total_v6": 0.7773764580488205, "step": 3060 }, { "epoch": 0.7059603603603604, "grad_norm": 26.496929168701172, "learning_rate": 7.031843314770233e-06, "loss": 12.082785606384277, "loss_alignment": 0.951171875, "loss_alignment_w": 0.4755859375, "loss_sft": 0.2798170428723097, "loss_total_v6": 0.7551740929484367, "step": 3061 }, { "epoch": 0.706190990990991, "grad_norm": 24.364093780517578, "learning_rate": 7.0300030845511815e-06, "loss": 12.182538986206055, "loss_alignment": 0.955078125, "loss_alignment_w": 0.4775390625, "loss_sft": 0.2840374633669853, "loss_total_v6": 0.7614086866378784, "step": 3062 }, { "epoch": 0.7064216216216216, "grad_norm": 25.60946273803711, "learning_rate": 7.0281625250203764e-06, "loss": 12.334436416625977, "loss_alignment": 0.955078125, "loss_alignment_w": 0.4775390625, "loss_sft": 0.29435503855347633, "loss_total_v6": 0.7709022834897041, "step": 3063 }, { "epoch": 0.7066522522522523, "grad_norm": 24.12169647216797, "learning_rate": 7.026321636476398e-06, "loss": 12.251949310302734, "loss_alignment": 0.94384765625, "loss_alignment_w": 0.471923828125, "loss_sft": 0.2939297817647457, "loss_total_v6": 0.7657467946410179, "step": 3064 }, { "epoch": 0.7068828828828829, "grad_norm": 26.316221237182617, "learning_rate": 7.024480419217878e-06, "loss": 12.145557403564453, "loss_alignment": 0.94189453125, "loss_alignment_w": 0.470947265625, "loss_sft": 0.2876617833971977, "loss_total_v6": 0.7590973451733589, "step": 3065 }, { "epoch": 0.7071135135135135, "grad_norm": 25.46406364440918, "learning_rate": 7.022638873543504e-06, "loss": 12.381366729736328, "loss_alignment": 0.96728515625, "loss_alignment_w": 0.483642578125, "loss_sft": 0.29141352139413357, "loss_total_v6": 0.7738354057073593, "step": 3066 }, { "epoch": 0.7073441441441441, "grad_norm": 23.624649047851562, "learning_rate": 7.020796999752016e-06, "loss": 12.687975883483887, "loss_alignment": 0.953125, "loss_alignment_w": 0.4765625, "loss_sft": 0.31619186140596867, "loss_total_v6": 0.7929984927177429, "step": 3067 }, { "epoch": 0.7075747747747748, "grad_norm": 25.98607635498047, "learning_rate": 7.018954798142205e-06, "loss": 12.012181282043457, "loss_alignment": 0.9482421875, "loss_alignment_w": 0.47412109375, "loss_sft": 0.2761061489582062, "loss_total_v6": 0.7507613226771355, "step": 3068 }, { "epoch": 0.7078054054054054, "grad_norm": 26.05266571044922, "learning_rate": 7.0171122690129186e-06, "loss": 12.778249740600586, "loss_alignment": 0.970703125, "loss_alignment_w": 0.4853515625, "loss_sft": 0.312098853290081, "loss_total_v6": 0.7986405938863754, "step": 3069 }, { "epoch": 0.708036036036036, "grad_norm": 30.473981857299805, "learning_rate": 7.015269412663057e-06, "loss": 12.466367721557617, "loss_alignment": 0.95458984375, "loss_alignment_w": 0.477294921875, "loss_sft": 0.30145633965730667, "loss_total_v6": 0.7791480049490929, "step": 3070 }, { "epoch": 0.7082666666666667, "grad_norm": 23.494510650634766, "learning_rate": 7.0134262293915704e-06, "loss": 12.356466293334961, "loss_alignment": 0.96875, "loss_alignment_w": 0.484375, "loss_sft": 0.2885754927992821, "loss_total_v6": 0.7722790837287903, "step": 3071 }, { "epoch": 0.7084972972972973, "grad_norm": 26.288206100463867, "learning_rate": 7.011582719497466e-06, "loss": 12.120580673217773, "loss_alignment": 0.94873046875, "loss_alignment_w": 0.474365234375, "loss_sft": 0.28385772183537483, "loss_total_v6": 0.7575362846255302, "step": 3072 }, { "epoch": 0.7087279279279279, "grad_norm": 25.660015106201172, "learning_rate": 7.009738883279802e-06, "loss": 12.69607925415039, "loss_alignment": 0.95849609375, "loss_alignment_w": 0.479248046875, "loss_sft": 0.3143942765891552, "loss_total_v6": 0.7935049533843994, "step": 3073 }, { "epoch": 0.7089585585585586, "grad_norm": 22.149206161499023, "learning_rate": 7.007894721037688e-06, "loss": 12.412250518798828, "loss_alignment": 0.96484375, "loss_alignment_w": 0.482421875, "loss_sft": 0.2934048380702734, "loss_total_v6": 0.7757656574249268, "step": 3074 }, { "epoch": 0.7091891891891892, "grad_norm": 26.74291229248047, "learning_rate": 7.006050233070289e-06, "loss": 12.45438003540039, "loss_alignment": 0.97607421875, "loss_alignment_w": 0.488037109375, "loss_sft": 0.28978175669908524, "loss_total_v6": 0.7783987149596214, "step": 3075 }, { "epoch": 0.7094198198198198, "grad_norm": 25.568126678466797, "learning_rate": 7.004205419676826e-06, "loss": 12.81646728515625, "loss_alignment": 0.9365234375, "loss_alignment_w": 0.46826171875, "loss_sft": 0.33189767971634865, "loss_total_v6": 0.8010291680693626, "step": 3076 }, { "epoch": 0.7096504504504505, "grad_norm": 29.88728141784668, "learning_rate": 7.0023602811565625e-06, "loss": 12.677652359008789, "loss_alignment": 0.95654296875, "loss_alignment_w": 0.478271484375, "loss_sft": 0.3139597438275814, "loss_total_v6": 0.7923532798886299, "step": 3077 }, { "epoch": 0.7098810810810811, "grad_norm": 24.400209426879883, "learning_rate": 7.000514817808824e-06, "loss": 11.960512161254883, "loss_alignment": 0.96240234375, "loss_alignment_w": 0.481201171875, "loss_sft": 0.26712432503700256, "loss_total_v6": 0.7475320175290108, "step": 3078 }, { "epoch": 0.7101117117117117, "grad_norm": 23.928255081176758, "learning_rate": 6.998669029932989e-06, "loss": 11.912702560424805, "loss_alignment": 0.95361328125, "loss_alignment_w": 0.476806640625, "loss_sft": 0.26892749033868313, "loss_total_v6": 0.7445439323782921, "step": 3079 }, { "epoch": 0.7103423423423424, "grad_norm": 25.296218872070312, "learning_rate": 6.9968229178284775e-06, "loss": 11.792428970336914, "loss_alignment": 0.9423828125, "loss_alignment_w": 0.47119140625, "loss_sft": 0.2647062409669161, "loss_total_v6": 0.7370268031954765, "step": 3080 }, { "epoch": 0.710572972972973, "grad_norm": 24.565818786621094, "learning_rate": 6.994976481794778e-06, "loss": 12.041034698486328, "loss_alignment": 0.92529296875, "loss_alignment_w": 0.462646484375, "loss_sft": 0.29026908800005913, "loss_total_v6": 0.7525646388530731, "step": 3081 }, { "epoch": 0.7108036036036036, "grad_norm": 26.76308250427246, "learning_rate": 6.993129722131421e-06, "loss": 12.09178638458252, "loss_alignment": 0.9384765625, "loss_alignment_w": 0.46923828125, "loss_sft": 0.2878716439008713, "loss_total_v6": 0.7557366341352463, "step": 3082 }, { "epoch": 0.7110342342342343, "grad_norm": 25.40316390991211, "learning_rate": 6.991282639137988e-06, "loss": 12.413901329040527, "loss_alignment": 0.95654296875, "loss_alignment_w": 0.478271484375, "loss_sft": 0.2985129188746214, "loss_total_v6": 0.7758688479661942, "step": 3083 }, { "epoch": 0.7112648648648648, "grad_norm": 25.614477157592773, "learning_rate": 6.989435233114124e-06, "loss": 12.187110900878906, "loss_alignment": 0.95166015625, "loss_alignment_w": 0.475830078125, "loss_sft": 0.2863525655120611, "loss_total_v6": 0.7616943642497063, "step": 3084 }, { "epoch": 0.7114954954954955, "grad_norm": 25.645654678344727, "learning_rate": 6.987587504359514e-06, "loss": 12.632817268371582, "loss_alignment": 0.95751953125, "loss_alignment_w": 0.478759765625, "loss_sft": 0.3107912763953209, "loss_total_v6": 0.7895510420203209, "step": 3085 }, { "epoch": 0.7117261261261262, "grad_norm": 23.099502563476562, "learning_rate": 6.985739453173903e-06, "loss": 12.320451736450195, "loss_alignment": 0.947265625, "loss_alignment_w": 0.4736328125, "loss_sft": 0.296151302754879, "loss_total_v6": 0.770028218626976, "step": 3086 }, { "epoch": 0.7119567567567567, "grad_norm": 25.6939697265625, "learning_rate": 6.983891079857088e-06, "loss": 12.39040756225586, "loss_alignment": 0.958984375, "loss_alignment_w": 0.4794921875, "loss_sft": 0.2947710007429123, "loss_total_v6": 0.774400494992733, "step": 3087 }, { "epoch": 0.7121873873873874, "grad_norm": 26.326162338256836, "learning_rate": 6.982042384708911e-06, "loss": 12.672578811645508, "loss_alignment": 0.96826171875, "loss_alignment_w": 0.484130859375, "loss_sft": 0.3079663757234812, "loss_total_v6": 0.792036198079586, "step": 3088 }, { "epoch": 0.7124180180180181, "grad_norm": 23.367952346801758, "learning_rate": 6.980193368029276e-06, "loss": 12.04609489440918, "loss_alignment": 0.96533203125, "loss_alignment_w": 0.482666015625, "loss_sft": 0.2700318042188883, "loss_total_v6": 0.7528809383511543, "step": 3089 }, { "epoch": 0.7126486486486486, "grad_norm": 25.91721534729004, "learning_rate": 6.978344030118134e-06, "loss": 12.740022659301758, "loss_alignment": 0.943359375, "loss_alignment_w": 0.4716796875, "loss_sft": 0.325563482940197, "loss_total_v6": 0.7962513864040375, "step": 3090 }, { "epoch": 0.7128792792792793, "grad_norm": 26.254070281982422, "learning_rate": 6.976494371275488e-06, "loss": 12.829338073730469, "loss_alignment": 0.9775390625, "loss_alignment_w": 0.48876953125, "loss_sft": 0.3123622089624405, "loss_total_v6": 0.8018336445093155, "step": 3091 }, { "epoch": 0.7131099099099099, "grad_norm": 24.5595645904541, "learning_rate": 6.974644391801395e-06, "loss": 12.650681495666504, "loss_alignment": 0.95703125, "loss_alignment_w": 0.478515625, "loss_sft": 0.31308279372751713, "loss_total_v6": 0.7906676232814789, "step": 3092 }, { "epoch": 0.7133405405405405, "grad_norm": 27.374326705932617, "learning_rate": 6.972794091995961e-06, "loss": 12.048885345458984, "loss_alignment": 0.962890625, "loss_alignment_w": 0.4814453125, "loss_sft": 0.2711065113544464, "loss_total_v6": 0.7530553489923477, "step": 3093 }, { "epoch": 0.7135711711711712, "grad_norm": 23.902109146118164, "learning_rate": 6.970943472159349e-06, "loss": 12.007793426513672, "loss_alignment": 0.9462890625, "loss_alignment_w": 0.47314453125, "loss_sft": 0.2758777067065239, "loss_total_v6": 0.7504870668053627, "step": 3094 }, { "epoch": 0.7138018018018017, "grad_norm": 24.246910095214844, "learning_rate": 6.969092532591767e-06, "loss": 12.468276977539062, "loss_alignment": 0.96142578125, "loss_alignment_w": 0.480712890625, "loss_sft": 0.29899696819484234, "loss_total_v6": 0.7792673408985138, "step": 3095 }, { "epoch": 0.7140324324324324, "grad_norm": 23.61089515686035, "learning_rate": 6.9672412735934805e-06, "loss": 11.851634979248047, "loss_alignment": 0.93115234375, "loss_alignment_w": 0.465576171875, "loss_sft": 0.27394556626677513, "loss_total_v6": 0.7407271713018417, "step": 3096 }, { "epoch": 0.7142630630630631, "grad_norm": 24.260066986083984, "learning_rate": 6.965389695464805e-06, "loss": 12.46514892578125, "loss_alignment": 0.9765625, "loss_alignment_w": 0.48828125, "loss_sft": 0.2917823661118746, "loss_total_v6": 0.7790717855095863, "step": 3097 }, { "epoch": 0.7144936936936936, "grad_norm": 23.098237991333008, "learning_rate": 6.963537798506108e-06, "loss": 12.264252662658691, "loss_alignment": 0.94580078125, "loss_alignment_w": 0.472900390625, "loss_sft": 0.29266938380897045, "loss_total_v6": 0.76651581376791, "step": 3098 }, { "epoch": 0.7147243243243243, "grad_norm": 25.661401748657227, "learning_rate": 6.961685583017808e-06, "loss": 12.587533950805664, "loss_alignment": 0.98291015625, "loss_alignment_w": 0.491455078125, "loss_sft": 0.2956319600343704, "loss_total_v6": 0.7867208495736122, "step": 3099 }, { "epoch": 0.714954954954955, "grad_norm": 25.89670181274414, "learning_rate": 6.959833049300376e-06, "loss": 12.34244155883789, "loss_alignment": 0.96142578125, "loss_alignment_w": 0.480712890625, "loss_sft": 0.29140690714120865, "loss_total_v6": 0.7714026048779488, "step": 3100 }, { "epoch": 0.7151855855855855, "grad_norm": 25.332311630249023, "learning_rate": 6.957980197654335e-06, "loss": 12.666966438293457, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.30341940373182297, "loss_total_v6": 0.7916854098439217, "step": 3101 }, { "epoch": 0.7154162162162162, "grad_norm": 22.70188331604004, "learning_rate": 6.9561270283802555e-06, "loss": 12.414480209350586, "loss_alignment": 0.9482421875, "loss_alignment_w": 0.47412109375, "loss_sft": 0.30170765705406666, "loss_total_v6": 0.7759050577878952, "step": 3102 }, { "epoch": 0.7156468468468469, "grad_norm": 24.544437408447266, "learning_rate": 6.954273541778767e-06, "loss": 11.972982406616211, "loss_alignment": 0.94287109375, "loss_alignment_w": 0.471435546875, "loss_sft": 0.27715050987899303, "loss_total_v6": 0.7483114153146744, "step": 3103 }, { "epoch": 0.7158774774774774, "grad_norm": 27.658517837524414, "learning_rate": 6.952419738150546e-06, "loss": 12.354795455932617, "loss_alignment": 0.939453125, "loss_alignment_w": 0.4697265625, "loss_sft": 0.3021887484937906, "loss_total_v6": 0.7721746936440468, "step": 3104 }, { "epoch": 0.7161081081081081, "grad_norm": 25.280719757080078, "learning_rate": 6.950565617796317e-06, "loss": 12.007410049438477, "loss_alignment": 0.9208984375, "loss_alignment_w": 0.46044921875, "loss_sft": 0.28911361284554005, "loss_total_v6": 0.7504631280899048, "step": 3105 }, { "epoch": 0.7163387387387388, "grad_norm": 24.2636661529541, "learning_rate": 6.948711181016863e-06, "loss": 11.975454330444336, "loss_alignment": 0.951171875, "loss_alignment_w": 0.4755859375, "loss_sft": 0.27223908342421055, "loss_total_v6": 0.748465895652771, "step": 3106 }, { "epoch": 0.7165693693693693, "grad_norm": 24.52878761291504, "learning_rate": 6.946856428113014e-06, "loss": 12.421202659606934, "loss_alignment": 0.94287109375, "loss_alignment_w": 0.471435546875, "loss_sft": 0.30394357815384865, "loss_total_v6": 0.7763251811265945, "step": 3107 }, { "epoch": 0.7168, "grad_norm": 24.607032775878906, "learning_rate": 6.9450013593856515e-06, "loss": 11.562482833862305, "loss_alignment": 0.953125, "loss_alignment_w": 0.4765625, "loss_sft": 0.24645891040563583, "loss_total_v6": 0.7226551845669746, "step": 3108 }, { "epoch": 0.7170306306306307, "grad_norm": 26.33462905883789, "learning_rate": 6.943145975135711e-06, "loss": 12.577021598815918, "loss_alignment": 0.95458984375, "loss_alignment_w": 0.477294921875, "loss_sft": 0.3087384067475796, "loss_total_v6": 0.7860638573765755, "step": 3109 }, { "epoch": 0.7172612612612612, "grad_norm": 25.46951675415039, "learning_rate": 6.941290275664175e-06, "loss": 12.053072929382324, "loss_alignment": 0.96630859375, "loss_alignment_w": 0.483154296875, "loss_sft": 0.26964398846030235, "loss_total_v6": 0.7533170878887177, "step": 3110 }, { "epoch": 0.7174918918918919, "grad_norm": 23.520296096801758, "learning_rate": 6.939434261272081e-06, "loss": 12.672480583190918, "loss_alignment": 0.94287109375, "loss_alignment_w": 0.471435546875, "loss_sft": 0.3203503452241421, "loss_total_v6": 0.792030043900013, "step": 3111 }, { "epoch": 0.7177225225225226, "grad_norm": 27.82611083984375, "learning_rate": 6.9375779322605154e-06, "loss": 12.445902824401855, "loss_alignment": 0.9560546875, "loss_alignment_w": 0.47802734375, "loss_sft": 0.29941438138484955, "loss_total_v6": 0.7778689414262772, "step": 3112 }, { "epoch": 0.7179531531531531, "grad_norm": 23.63677978515625, "learning_rate": 6.935721288930617e-06, "loss": 12.343719482421875, "loss_alignment": 0.9677734375, "loss_alignment_w": 0.48388671875, "loss_sft": 0.28603939339518547, "loss_total_v6": 0.7714825049042702, "step": 3113 }, { "epoch": 0.7181837837837838, "grad_norm": 22.64321517944336, "learning_rate": 6.933864331583572e-06, "loss": 12.556836128234863, "loss_alignment": 0.94580078125, "loss_alignment_w": 0.472900390625, "loss_sft": 0.31087950989603996, "loss_total_v6": 0.7848022505640984, "step": 3114 }, { "epoch": 0.7184144144144144, "grad_norm": 24.576934814453125, "learning_rate": 6.932007060520622e-06, "loss": 11.49879264831543, "loss_alignment": 0.94873046875, "loss_alignment_w": 0.474365234375, "loss_sft": 0.24530117772519588, "loss_total_v6": 0.7186745628714561, "step": 3115 }, { "epoch": 0.718645045045045, "grad_norm": 24.555274963378906, "learning_rate": 6.930149476043059e-06, "loss": 11.93179702758789, "loss_alignment": 0.93896484375, "loss_alignment_w": 0.469482421875, "loss_sft": 0.27678897976875305, "loss_total_v6": 0.7457373216748238, "step": 3116 }, { "epoch": 0.7188756756756757, "grad_norm": 25.507596969604492, "learning_rate": 6.928291578452222e-06, "loss": 12.71314811706543, "loss_alignment": 0.951171875, "loss_alignment_w": 0.4755859375, "loss_sft": 0.31787196546792984, "loss_total_v6": 0.7945717796683311, "step": 3117 }, { "epoch": 0.7191063063063063, "grad_norm": 24.37274742126465, "learning_rate": 6.926433368049506e-06, "loss": 12.049811363220215, "loss_alignment": 0.9609375, "loss_alignment_w": 0.48046875, "loss_sft": 0.27294961363077164, "loss_total_v6": 0.7531131878495216, "step": 3118 }, { "epoch": 0.7193369369369369, "grad_norm": 23.65010643005371, "learning_rate": 6.924574845136352e-06, "loss": 12.412521362304688, "loss_alignment": 0.962890625, "loss_alignment_w": 0.4814453125, "loss_sft": 0.29334539361298084, "loss_total_v6": 0.7757825404405594, "step": 3119 }, { "epoch": 0.7195675675675676, "grad_norm": 22.723403930664062, "learning_rate": 6.922716010014256e-06, "loss": 12.195015907287598, "loss_alignment": 0.95458984375, "loss_alignment_w": 0.477294921875, "loss_sft": 0.28471049666404724, "loss_total_v6": 0.7621885016560555, "step": 3120 }, { "epoch": 0.7197981981981982, "grad_norm": 24.56511688232422, "learning_rate": 6.920856862984759e-06, "loss": 12.660492897033691, "loss_alignment": 0.98974609375, "loss_alignment_w": 0.494873046875, "loss_sft": 0.29604156501591206, "loss_total_v6": 0.7912808135151863, "step": 3121 }, { "epoch": 0.7200288288288288, "grad_norm": 23.111648559570312, "learning_rate": 6.91899740434946e-06, "loss": 12.210702896118164, "loss_alignment": 0.92822265625, "loss_alignment_w": 0.464111328125, "loss_sft": 0.2988135162740946, "loss_total_v6": 0.7631689831614494, "step": 3122 }, { "epoch": 0.7202594594594595, "grad_norm": 30.434850692749023, "learning_rate": 6.917137634410004e-06, "loss": 11.900548934936523, "loss_alignment": 0.9326171875, "loss_alignment_w": 0.46630859375, "loss_sft": 0.27862011454999447, "loss_total_v6": 0.7437842860817909, "step": 3123 }, { "epoch": 0.7204900900900901, "grad_norm": 25.112154006958008, "learning_rate": 6.915277553468083e-06, "loss": 12.309947967529297, "loss_alignment": 0.93994140625, "loss_alignment_w": 0.469970703125, "loss_sft": 0.29985881224274635, "loss_total_v6": 0.7693717330694199, "step": 3124 }, { "epoch": 0.7207207207207207, "grad_norm": 24.370853424072266, "learning_rate": 6.913417161825449e-06, "loss": 12.137537002563477, "loss_alignment": 0.966796875, "loss_alignment_w": 0.4833984375, "loss_sft": 0.27483142353594303, "loss_total_v6": 0.7585960775613785, "step": 3125 }, { "epoch": 0.7209513513513514, "grad_norm": 25.13949203491211, "learning_rate": 6.911556459783898e-06, "loss": 12.01905632019043, "loss_alignment": 0.9306640625, "loss_alignment_w": 0.46533203125, "loss_sft": 0.2856911476701498, "loss_total_v6": 0.7511910274624825, "step": 3126 }, { "epoch": 0.721181981981982, "grad_norm": 25.031774520874023, "learning_rate": 6.9096954476452724e-06, "loss": 12.368987083435059, "loss_alignment": 0.9677734375, "loss_alignment_w": 0.48388671875, "loss_sft": 0.28917499259114265, "loss_total_v6": 0.7730616852641106, "step": 3127 }, { "epoch": 0.7214126126126126, "grad_norm": 23.500940322875977, "learning_rate": 6.9078341257114765e-06, "loss": 12.30607795715332, "loss_alignment": 0.95263671875, "loss_alignment_w": 0.476318359375, "loss_sft": 0.29210966266691685, "loss_total_v6": 0.7691299244761467, "step": 3128 }, { "epoch": 0.7216432432432432, "grad_norm": 22.69459342956543, "learning_rate": 6.905972494284456e-06, "loss": 12.500852584838867, "loss_alignment": 0.9375, "loss_alignment_w": 0.46875, "loss_sft": 0.3128126785159111, "loss_total_v6": 0.7813032865524292, "step": 3129 }, { "epoch": 0.7218738738738739, "grad_norm": 23.05376625061035, "learning_rate": 6.904110553666206e-06, "loss": 12.154472351074219, "loss_alignment": 0.94482421875, "loss_alignment_w": 0.472412109375, "loss_sft": 0.2858080919831991, "loss_total_v6": 0.7596545517444611, "step": 3130 }, { "epoch": 0.7221045045045045, "grad_norm": 24.667705535888672, "learning_rate": 6.9022483041587785e-06, "loss": 12.899511337280273, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.3184264749288559, "loss_total_v6": 0.8062194138765335, "step": 3131 }, { "epoch": 0.7223351351351351, "grad_norm": 25.27851676940918, "learning_rate": 6.900385746064268e-06, "loss": 12.026577949523926, "loss_alignment": 0.9677734375, "loss_alignment_w": 0.48388671875, "loss_sft": 0.26865946874022484, "loss_total_v6": 0.7516611516475677, "step": 3132 }, { "epoch": 0.7225657657657658, "grad_norm": 26.743724822998047, "learning_rate": 6.8985228796848284e-06, "loss": 12.628307342529297, "loss_alignment": 0.96142578125, "loss_alignment_w": 0.480712890625, "loss_sft": 0.30902934074401855, "loss_total_v6": 0.789269208908081, "step": 3133 }, { "epoch": 0.7227963963963964, "grad_norm": 27.089536666870117, "learning_rate": 6.896659705322654e-06, "loss": 12.887877464294434, "loss_alignment": 0.94580078125, "loss_alignment_w": 0.472900390625, "loss_sft": 0.3323631025850773, "loss_total_v6": 0.8054923489689827, "step": 3134 }, { "epoch": 0.723027027027027, "grad_norm": 23.340396881103516, "learning_rate": 6.894796223279994e-06, "loss": 12.313036918640137, "loss_alignment": 0.95361328125, "loss_alignment_w": 0.476806640625, "loss_sft": 0.2924682889133692, "loss_total_v6": 0.7695648148655891, "step": 3135 }, { "epoch": 0.7232576576576577, "grad_norm": 23.923540115356445, "learning_rate": 6.892932433859147e-06, "loss": 12.099367141723633, "loss_alignment": 0.94287109375, "loss_alignment_w": 0.471435546875, "loss_sft": 0.28376781567931175, "loss_total_v6": 0.756210446357727, "step": 3136 }, { "epoch": 0.7234882882882883, "grad_norm": 25.86600685119629, "learning_rate": 6.891068337362462e-06, "loss": 12.376245498657227, "loss_alignment": 0.95849609375, "loss_alignment_w": 0.479248046875, "loss_sft": 0.29409938119351864, "loss_total_v6": 0.7735153064131737, "step": 3137 }, { "epoch": 0.7237189189189189, "grad_norm": 26.10675811767578, "learning_rate": 6.889203934092337e-06, "loss": 12.134180068969727, "loss_alignment": 0.95703125, "loss_alignment_w": 0.478515625, "loss_sft": 0.2797943279147148, "loss_total_v6": 0.7583862543106079, "step": 3138 }, { "epoch": 0.7239495495495496, "grad_norm": 26.32684326171875, "learning_rate": 6.887339224351217e-06, "loss": 12.270305633544922, "loss_alignment": 0.9609375, "loss_alignment_w": 0.48046875, "loss_sft": 0.28596759401261806, "loss_total_v6": 0.7668941095471382, "step": 3139 }, { "epoch": 0.7241801801801802, "grad_norm": 23.26316261291504, "learning_rate": 6.885474208441602e-06, "loss": 11.86265754699707, "loss_alignment": 0.94287109375, "loss_alignment_w": 0.471435546875, "loss_sft": 0.26946179009974003, "loss_total_v6": 0.7414161115884781, "step": 3140 }, { "epoch": 0.7244108108108108, "grad_norm": 25.78865623474121, "learning_rate": 6.88360888666604e-06, "loss": 11.737129211425781, "loss_alignment": 0.9423828125, "loss_alignment_w": 0.47119140625, "loss_sft": 0.26143317483365536, "loss_total_v6": 0.7335706204175949, "step": 3141 }, { "epoch": 0.7246414414414415, "grad_norm": 25.256546020507812, "learning_rate": 6.8817432593271275e-06, "loss": 12.440337181091309, "loss_alignment": 0.93603515625, "loss_alignment_w": 0.468017578125, "loss_sft": 0.3094424344599247, "loss_total_v6": 0.7775210589170456, "step": 3142 }, { "epoch": 0.724872072072072, "grad_norm": 25.3456974029541, "learning_rate": 6.8798773267275065e-06, "loss": 12.070053100585938, "loss_alignment": 0.94775390625, "loss_alignment_w": 0.473876953125, "loss_sft": 0.28062343783676624, "loss_total_v6": 0.7543783187866211, "step": 3143 }, { "epoch": 0.7251027027027027, "grad_norm": 25.340017318725586, "learning_rate": 6.878011089169878e-06, "loss": 12.048255920410156, "loss_alignment": 0.9423828125, "loss_alignment_w": 0.47119140625, "loss_sft": 0.28159570321440697, "loss_total_v6": 0.7530159652233124, "step": 3144 }, { "epoch": 0.7253333333333334, "grad_norm": 23.623062133789062, "learning_rate": 6.876144546956987e-06, "loss": 12.2508544921875, "loss_alignment": 0.9443359375, "loss_alignment_w": 0.47216796875, "loss_sft": 0.2928847763687372, "loss_total_v6": 0.7656783685088158, "step": 3145 }, { "epoch": 0.7255639639639639, "grad_norm": 26.594581604003906, "learning_rate": 6.874277700391623e-06, "loss": 12.360376358032227, "loss_alignment": 0.94970703125, "loss_alignment_w": 0.474853515625, "loss_sft": 0.29803624376654625, "loss_total_v6": 0.7725235521793365, "step": 3146 }, { "epoch": 0.7257945945945946, "grad_norm": 25.898109436035156, "learning_rate": 6.872410549776636e-06, "loss": 11.802318572998047, "loss_alignment": 0.91552734375, "loss_alignment_w": 0.457763671875, "loss_sft": 0.28020166978240013, "loss_total_v6": 0.7376449480652809, "step": 3147 }, { "epoch": 0.7260252252252253, "grad_norm": 26.215740203857422, "learning_rate": 6.870543095414919e-06, "loss": 12.747753143310547, "loss_alignment": 0.94580078125, "loss_alignment_w": 0.472900390625, "loss_sft": 0.32362050563097, "loss_total_v6": 0.7967345491051674, "step": 3148 }, { "epoch": 0.7262558558558558, "grad_norm": 24.97307777404785, "learning_rate": 6.8686753376094095e-06, "loss": 11.730022430419922, "loss_alignment": 0.9287109375, "loss_alignment_w": 0.46435546875, "loss_sft": 0.2679927386343479, "loss_total_v6": 0.7331264019012451, "step": 3149 }, { "epoch": 0.7264864864864865, "grad_norm": 25.80947494506836, "learning_rate": 6.8668072766631054e-06, "loss": 11.612720489501953, "loss_alignment": 0.9443359375, "loss_alignment_w": 0.47216796875, "loss_sft": 0.2542679738253355, "loss_total_v6": 0.7257950752973557, "step": 3150 }, { "epoch": 0.7267171171171171, "grad_norm": 24.329248428344727, "learning_rate": 6.8649389128790455e-06, "loss": 12.577815055847168, "loss_alignment": 0.953125, "loss_alignment_w": 0.4765625, "loss_sft": 0.30985610373318195, "loss_total_v6": 0.786113440990448, "step": 3151 }, { "epoch": 0.7269477477477477, "grad_norm": 25.83603858947754, "learning_rate": 6.863070246560319e-06, "loss": 12.329157829284668, "loss_alignment": 0.97412109375, "loss_alignment_w": 0.487060546875, "loss_sft": 0.282886216416955, "loss_total_v6": 0.7705723643302917, "step": 3152 }, { "epoch": 0.7271783783783784, "grad_norm": 24.114770889282227, "learning_rate": 6.861201278010068e-06, "loss": 12.147615432739258, "loss_alignment": 0.94677734375, "loss_alignment_w": 0.473388671875, "loss_sft": 0.2865087389945984, "loss_total_v6": 0.7592259868979454, "step": 3153 }, { "epoch": 0.727409009009009, "grad_norm": 26.421279907226562, "learning_rate": 6.859332007531477e-06, "loss": 12.473795890808105, "loss_alignment": 0.95556640625, "loss_alignment_w": 0.477783203125, "loss_sft": 0.3015391491353512, "loss_total_v6": 0.7796122506260872, "step": 3154 }, { "epoch": 0.7276396396396396, "grad_norm": 24.327585220336914, "learning_rate": 6.8574624354277866e-06, "loss": 12.242427825927734, "loss_alignment": 0.94970703125, "loss_alignment_w": 0.474853515625, "loss_sft": 0.2918851375579834, "loss_total_v6": 0.7651517316699028, "step": 3155 }, { "epoch": 0.7278702702702703, "grad_norm": 25.133867263793945, "learning_rate": 6.855592562002281e-06, "loss": 12.13492488861084, "loss_alignment": 0.95361328125, "loss_alignment_w": 0.476806640625, "loss_sft": 0.2811531163752079, "loss_total_v6": 0.7584327980875969, "step": 3156 }, { "epoch": 0.7281009009009008, "grad_norm": 24.55325698852539, "learning_rate": 6.853722387558298e-06, "loss": 12.654487609863281, "loss_alignment": 0.98486328125, "loss_alignment_w": 0.492431640625, "loss_sft": 0.2989468649029732, "loss_total_v6": 0.7909055054187775, "step": 3157 }, { "epoch": 0.7283315315315315, "grad_norm": 24.19072151184082, "learning_rate": 6.851851912399219e-06, "loss": 12.263727188110352, "loss_alignment": 0.9375, "loss_alignment_w": 0.46875, "loss_sft": 0.29826703667640686, "loss_total_v6": 0.7664829865098, "step": 3158 }, { "epoch": 0.7285621621621622, "grad_norm": 24.10113525390625, "learning_rate": 6.849981136828478e-06, "loss": 12.130891799926758, "loss_alignment": 0.9404296875, "loss_alignment_w": 0.47021484375, "loss_sft": 0.2883931566029787, "loss_total_v6": 0.758180744946003, "step": 3159 }, { "epoch": 0.7287927927927927, "grad_norm": 26.16681671142578, "learning_rate": 6.848110061149555e-06, "loss": 12.342342376708984, "loss_alignment": 0.95556640625, "loss_alignment_w": 0.477783203125, "loss_sft": 0.29400993697345257, "loss_total_v6": 0.7713964134454727, "step": 3160 }, { "epoch": 0.7290234234234234, "grad_norm": 22.45236587524414, "learning_rate": 6.8462386856659824e-06, "loss": 11.854530334472656, "loss_alignment": 0.9345703125, "loss_alignment_w": 0.46728515625, "loss_sft": 0.2739892266690731, "loss_total_v6": 0.7409081757068634, "step": 3161 }, { "epoch": 0.7292540540540541, "grad_norm": 25.357295989990234, "learning_rate": 6.844367010681338e-06, "loss": 12.13377571105957, "loss_alignment": 0.95654296875, "loss_alignment_w": 0.478271484375, "loss_sft": 0.27960119768977165, "loss_total_v6": 0.7583609893918037, "step": 3162 }, { "epoch": 0.7294846846846846, "grad_norm": 25.026384353637695, "learning_rate": 6.842495036499248e-06, "loss": 11.78144645690918, "loss_alignment": 0.935546875, "loss_alignment_w": 0.4677734375, "loss_sft": 0.2685669958591461, "loss_total_v6": 0.7363404333591461, "step": 3163 }, { "epoch": 0.7297153153153153, "grad_norm": 26.453964233398438, "learning_rate": 6.840622763423391e-06, "loss": 12.801790237426758, "loss_alignment": 0.95947265625, "loss_alignment_w": 0.479736328125, "loss_sft": 0.3206044267863035, "loss_total_v6": 0.8001118525862694, "step": 3164 }, { "epoch": 0.729945945945946, "grad_norm": 25.65868377685547, "learning_rate": 6.838750191757488e-06, "loss": 12.170520782470703, "loss_alignment": 0.96875, "loss_alignment_w": 0.484375, "loss_sft": 0.27551959827542305, "loss_total_v6": 0.760657548904419, "step": 3165 }, { "epoch": 0.7301765765765765, "grad_norm": 26.274555206298828, "learning_rate": 6.836877321805316e-06, "loss": 12.823089599609375, "loss_alignment": 0.94873046875, "loss_alignment_w": 0.474365234375, "loss_sft": 0.3277187645435333, "loss_total_v6": 0.8014431223273277, "step": 3166 }, { "epoch": 0.7304072072072072, "grad_norm": 25.40155029296875, "learning_rate": 6.835004153870694e-06, "loss": 12.264942169189453, "loss_alignment": 0.97509765625, "loss_alignment_w": 0.487548828125, "loss_sft": 0.27994086034595966, "loss_total_v6": 0.7665589302778244, "step": 3167 }, { "epoch": 0.7306378378378379, "grad_norm": 26.4873046875, "learning_rate": 6.833130688257489e-06, "loss": 12.34309196472168, "loss_alignment": 0.9365234375, "loss_alignment_w": 0.46826171875, "loss_sft": 0.30486005917191505, "loss_total_v6": 0.7714432999491692, "step": 3168 }, { "epoch": 0.7308684684684684, "grad_norm": 25.603099822998047, "learning_rate": 6.831256925269624e-06, "loss": 12.136655807495117, "loss_alignment": 0.9423828125, "loss_alignment_w": 0.47119140625, "loss_sft": 0.2892264649271965, "loss_total_v6": 0.7585410550236702, "step": 3169 }, { "epoch": 0.7310990990990991, "grad_norm": 26.31768798828125, "learning_rate": 6.829382865211063e-06, "loss": 12.447410583496094, "loss_alignment": 0.93408203125, "loss_alignment_w": 0.467041015625, "loss_sft": 0.31035755947232246, "loss_total_v6": 0.7779631465673447, "step": 3170 }, { "epoch": 0.7313297297297298, "grad_norm": 31.196901321411133, "learning_rate": 6.827508508385817e-06, "loss": 12.258234024047852, "loss_alignment": 0.94091796875, "loss_alignment_w": 0.470458984375, "loss_sft": 0.29545176960527897, "loss_total_v6": 0.7661396488547325, "step": 3171 }, { "epoch": 0.7315603603603603, "grad_norm": 24.92073631286621, "learning_rate": 6.825633855097954e-06, "loss": 12.519811630249023, "loss_alignment": 0.95751953125, "loss_alignment_w": 0.478759765625, "loss_sft": 0.3034232333302498, "loss_total_v6": 0.7824882119894028, "step": 3172 }, { "epoch": 0.731790990990991, "grad_norm": 22.734201431274414, "learning_rate": 6.823758905651582e-06, "loss": 12.370670318603516, "loss_alignment": 0.9384765625, "loss_alignment_w": 0.46923828125, "loss_sft": 0.30307411774992943, "loss_total_v6": 0.7731668725609779, "step": 3173 }, { "epoch": 0.7320216216216217, "grad_norm": 24.41073226928711, "learning_rate": 6.821883660350858e-06, "loss": 12.738788604736328, "loss_alignment": 0.94970703125, "loss_alignment_w": 0.474853515625, "loss_sft": 0.3211376704275608, "loss_total_v6": 0.7961742654442787, "step": 3174 }, { "epoch": 0.7322522522522522, "grad_norm": 24.704376220703125, "learning_rate": 6.820008119499992e-06, "loss": 12.083452224731445, "loss_alignment": 0.9345703125, "loss_alignment_w": 0.46728515625, "loss_sft": 0.2889987789094448, "loss_total_v6": 0.7552157938480377, "step": 3175 }, { "epoch": 0.7324828828828829, "grad_norm": 26.26121711730957, "learning_rate": 6.818132283403236e-06, "loss": 12.481993675231934, "loss_alignment": 0.96142578125, "loss_alignment_w": 0.480712890625, "loss_sft": 0.3005866166204214, "loss_total_v6": 0.7801245748996735, "step": 3176 }, { "epoch": 0.7327135135135135, "grad_norm": 24.23988151550293, "learning_rate": 6.816256152364893e-06, "loss": 12.985971450805664, "loss_alignment": 0.96630859375, "loss_alignment_w": 0.483154296875, "loss_sft": 0.32941489666700363, "loss_total_v6": 0.8116231709718704, "step": 3177 }, { "epoch": 0.7329441441441441, "grad_norm": 29.47722053527832, "learning_rate": 6.814379726689313e-06, "loss": 12.451605796813965, "loss_alignment": 0.9296875, "loss_alignment_w": 0.46484375, "loss_sft": 0.31298488192260265, "loss_total_v6": 0.778225339949131, "step": 3178 }, { "epoch": 0.7331747747747748, "grad_norm": 24.244596481323242, "learning_rate": 6.812503006680895e-06, "loss": 11.74451732635498, "loss_alignment": 0.95458984375, "loss_alignment_w": 0.477294921875, "loss_sft": 0.256371196359396, "loss_total_v6": 0.7340323254466057, "step": 3179 }, { "epoch": 0.7334054054054054, "grad_norm": 24.30312156677246, "learning_rate": 6.810625992644085e-06, "loss": 12.723489761352539, "loss_alignment": 0.9599609375, "loss_alignment_w": 0.47998046875, "loss_sft": 0.31516134552657604, "loss_total_v6": 0.7952181100845337, "step": 3180 }, { "epoch": 0.733636036036036, "grad_norm": 23.864755630493164, "learning_rate": 6.808748684883374e-06, "loss": 12.557075500488281, "loss_alignment": 0.96044921875, "loss_alignment_w": 0.480224609375, "loss_sft": 0.30537085607647896, "loss_total_v6": 0.78481724858284, "step": 3181 }, { "epoch": 0.7338666666666667, "grad_norm": 24.887327194213867, "learning_rate": 6.806871083703307e-06, "loss": 12.290371894836426, "loss_alignment": 0.93505859375, "loss_alignment_w": 0.467529296875, "loss_sft": 0.3005884326994419, "loss_total_v6": 0.7681482583284378, "step": 3182 }, { "epoch": 0.7340972972972973, "grad_norm": 23.977205276489258, "learning_rate": 6.80499318940847e-06, "loss": 12.539334297180176, "loss_alignment": 0.943359375, "loss_alignment_w": 0.4716796875, "loss_sft": 0.3125627711415291, "loss_total_v6": 0.783708393573761, "step": 3183 }, { "epoch": 0.7343279279279279, "grad_norm": 31.532678604125977, "learning_rate": 6.8031150023035e-06, "loss": 12.008476257324219, "loss_alignment": 0.91796875, "loss_alignment_w": 0.458984375, "loss_sft": 0.292109951376915, "loss_total_v6": 0.7505297511816025, "step": 3184 }, { "epoch": 0.7345585585585586, "grad_norm": 23.44512939453125, "learning_rate": 6.8012365226930825e-06, "loss": 12.927427291870117, "loss_alignment": 0.96240234375, "loss_alignment_w": 0.481201171875, "loss_sft": 0.32731233909726143, "loss_total_v6": 0.8079642057418823, "step": 3185 }, { "epoch": 0.7347891891891892, "grad_norm": 25.9437313079834, "learning_rate": 6.799357750881944e-06, "loss": 11.78310775756836, "loss_alignment": 0.92822265625, "loss_alignment_w": 0.464111328125, "loss_sft": 0.27269907109439373, "loss_total_v6": 0.7364442050457001, "step": 3186 }, { "epoch": 0.7350198198198198, "grad_norm": 28.489181518554688, "learning_rate": 6.7974786871748674e-06, "loss": 12.637033462524414, "loss_alignment": 0.95947265625, "loss_alignment_w": 0.479736328125, "loss_sft": 0.31030716374516487, "loss_total_v6": 0.7898145690560341, "step": 3187 }, { "epoch": 0.7352504504504505, "grad_norm": 24.771570205688477, "learning_rate": 6.795599331876679e-06, "loss": 11.393564224243164, "loss_alignment": 0.947265625, "loss_alignment_w": 0.4736328125, "loss_sft": 0.23829715698957443, "loss_total_v6": 0.7120978087186813, "step": 3188 }, { "epoch": 0.7354810810810811, "grad_norm": 26.454975128173828, "learning_rate": 6.79371968529225e-06, "loss": 11.940434455871582, "loss_alignment": 0.9365234375, "loss_alignment_w": 0.46826171875, "loss_sft": 0.2774661034345627, "loss_total_v6": 0.7462771534919739, "step": 3189 }, { "epoch": 0.7357117117117117, "grad_norm": 28.454933166503906, "learning_rate": 6.7918397477265e-06, "loss": 12.452850341796875, "loss_alignment": 0.94677734375, "loss_alignment_w": 0.473388671875, "loss_sft": 0.30480767600238323, "loss_total_v6": 0.7783031612634659, "step": 3190 }, { "epoch": 0.7359423423423423, "grad_norm": 24.22615623474121, "learning_rate": 6.7899595194844025e-06, "loss": 11.915952682495117, "loss_alignment": 0.9453125, "loss_alignment_w": 0.47265625, "loss_sft": 0.2712973318994045, "loss_total_v6": 0.7447470128536224, "step": 3191 }, { "epoch": 0.736172972972973, "grad_norm": 27.020631790161133, "learning_rate": 6.788079000870966e-06, "loss": 12.538559913635254, "loss_alignment": 0.94873046875, "loss_alignment_w": 0.474365234375, "loss_sft": 0.3099050931632519, "loss_total_v6": 0.7836599946022034, "step": 3192 }, { "epoch": 0.7364036036036036, "grad_norm": 26.627893447875977, "learning_rate": 6.7861981921912555e-06, "loss": 12.505351066589355, "loss_alignment": 0.955078125, "loss_alignment_w": 0.4775390625, "loss_sft": 0.3034045398235321, "loss_total_v6": 0.7815844640135765, "step": 3193 }, { "epoch": 0.7366342342342342, "grad_norm": 31.92746925354004, "learning_rate": 6.784317093750381e-06, "loss": 12.380181312561035, "loss_alignment": 0.9560546875, "loss_alignment_w": 0.47802734375, "loss_sft": 0.29465066082775593, "loss_total_v6": 0.7737613692879677, "step": 3194 }, { "epoch": 0.7368648648648649, "grad_norm": 25.650997161865234, "learning_rate": 6.782435705853495e-06, "loss": 12.829675674438477, "loss_alignment": 0.94873046875, "loss_alignment_w": 0.474365234375, "loss_sft": 0.3261772058904171, "loss_total_v6": 0.8018547296524048, "step": 3195 }, { "epoch": 0.7370954954954955, "grad_norm": 25.693723678588867, "learning_rate": 6.780554028805803e-06, "loss": 12.386273384094238, "loss_alignment": 0.95361328125, "loss_alignment_w": 0.476806640625, "loss_sft": 0.2976253628730774, "loss_total_v6": 0.7741420641541481, "step": 3196 }, { "epoch": 0.7373261261261261, "grad_norm": 27.615236282348633, "learning_rate": 6.7786720629125565e-06, "loss": 12.485456466674805, "loss_alignment": 0.95166015625, "loss_alignment_w": 0.475830078125, "loss_sft": 0.3039311468601227, "loss_total_v6": 0.7803410664200783, "step": 3197 }, { "epoch": 0.7375567567567568, "grad_norm": 24.591291427612305, "learning_rate": 6.776789808479049e-06, "loss": 12.116286277770996, "loss_alignment": 0.9462890625, "loss_alignment_w": 0.47314453125, "loss_sft": 0.28462688624858856, "loss_total_v6": 0.7572678700089455, "step": 3198 }, { "epoch": 0.7377873873873874, "grad_norm": 26.819250106811523, "learning_rate": 6.774907265810626e-06, "loss": 12.286468505859375, "loss_alignment": 0.951171875, "loss_alignment_w": 0.4755859375, "loss_sft": 0.29338644444942474, "loss_total_v6": 0.7679042518138885, "step": 3199 }, { "epoch": 0.738018018018018, "grad_norm": 26.596113204956055, "learning_rate": 6.773024435212678e-06, "loss": 11.936399459838867, "loss_alignment": 0.90673828125, "loss_alignment_w": 0.453369140625, "loss_sft": 0.29201496578752995, "loss_total_v6": 0.7460249736905098, "step": 3200 }, { "epoch": 0.7382486486486487, "grad_norm": 25.357908248901367, "learning_rate": 6.771141316990643e-06, "loss": 11.907037734985352, "loss_alignment": 0.94482421875, "loss_alignment_w": 0.472412109375, "loss_sft": 0.27229660749435425, "loss_total_v6": 0.7441898882389069, "step": 3201 }, { "epoch": 0.7384792792792793, "grad_norm": 24.374826431274414, "learning_rate": 6.7692579114500015e-06, "loss": 12.001604080200195, "loss_alignment": 0.947265625, "loss_alignment_w": 0.4736328125, "loss_sft": 0.27692522294819355, "loss_total_v6": 0.7501002550125122, "step": 3202 }, { "epoch": 0.7387099099099099, "grad_norm": 25.654630661010742, "learning_rate": 6.767374218896286e-06, "loss": 11.741874694824219, "loss_alignment": 0.9150390625, "loss_alignment_w": 0.45751953125, "loss_sft": 0.27587463706731796, "loss_total_v6": 0.7338671684265137, "step": 3203 }, { "epoch": 0.7389405405405406, "grad_norm": 23.7484073638916, "learning_rate": 6.765490239635075e-06, "loss": 12.105451583862305, "loss_alignment": 0.95166015625, "loss_alignment_w": 0.475830078125, "loss_sft": 0.28027237206697464, "loss_total_v6": 0.7565907314419746, "step": 3204 }, { "epoch": 0.7391711711711711, "grad_norm": 24.323596954345703, "learning_rate": 6.7636059739719876e-06, "loss": 12.00876235961914, "loss_alignment": 0.90966796875, "loss_alignment_w": 0.454833984375, "loss_sft": 0.2956678941845894, "loss_total_v6": 0.7505476474761963, "step": 3205 }, { "epoch": 0.7394018018018018, "grad_norm": 26.177898406982422, "learning_rate": 6.761721422212697e-06, "loss": 12.21371078491211, "loss_alignment": 0.9453125, "loss_alignment_w": 0.47265625, "loss_sft": 0.29005973413586617, "loss_total_v6": 0.7633568644523621, "step": 3206 }, { "epoch": 0.7396324324324325, "grad_norm": 23.618572235107422, "learning_rate": 6.75983658466292e-06, "loss": 12.520929336547852, "loss_alignment": 0.939453125, "loss_alignment_w": 0.4697265625, "loss_sft": 0.3121754042804241, "loss_total_v6": 0.7825581133365631, "step": 3207 }, { "epoch": 0.739863063063063, "grad_norm": 24.745941162109375, "learning_rate": 6.757951461628416e-06, "loss": 12.163570404052734, "loss_alignment": 0.95361328125, "loss_alignment_w": 0.476806640625, "loss_sft": 0.2832181714475155, "loss_total_v6": 0.7602231577038765, "step": 3208 }, { "epoch": 0.7400936936936937, "grad_norm": 27.52906036376953, "learning_rate": 6.756066053414998e-06, "loss": 11.8135986328125, "loss_alignment": 0.9482421875, "loss_alignment_w": 0.47412109375, "loss_sft": 0.26401522383093834, "loss_total_v6": 0.7383499518036842, "step": 3209 }, { "epoch": 0.7403243243243244, "grad_norm": 25.963220596313477, "learning_rate": 6.75418036032852e-06, "loss": 12.970258712768555, "loss_alignment": 0.94482421875, "loss_alignment_w": 0.472412109375, "loss_sft": 0.33870214223861694, "loss_total_v6": 0.8106411918997765, "step": 3210 }, { "epoch": 0.7405549549549549, "grad_norm": 27.003324508666992, "learning_rate": 6.75229438267488e-06, "loss": 12.439934730529785, "loss_alignment": 0.95751953125, "loss_alignment_w": 0.478759765625, "loss_sft": 0.29795799404382706, "loss_total_v6": 0.7774959281086922, "step": 3211 }, { "epoch": 0.7407855855855856, "grad_norm": 24.818910598754883, "learning_rate": 6.7504081207600295e-06, "loss": 12.132070541381836, "loss_alignment": 0.95263671875, "loss_alignment_w": 0.476318359375, "loss_sft": 0.28228697180747986, "loss_total_v6": 0.7582544088363647, "step": 3212 }, { "epoch": 0.7410162162162162, "grad_norm": 25.475704193115234, "learning_rate": 6.748521574889961e-06, "loss": 12.201499938964844, "loss_alignment": 0.9326171875, "loss_alignment_w": 0.46630859375, "loss_sft": 0.29373691603541374, "loss_total_v6": 0.7625937312841415, "step": 3213 }, { "epoch": 0.7412468468468468, "grad_norm": 25.018770217895508, "learning_rate": 6.746634745370715e-06, "loss": 11.864032745361328, "loss_alignment": 0.92236328125, "loss_alignment_w": 0.461181640625, "loss_sft": 0.28009155951440334, "loss_total_v6": 0.7415020391345024, "step": 3214 }, { "epoch": 0.7414774774774775, "grad_norm": 25.54918098449707, "learning_rate": 6.7447476325083764e-06, "loss": 12.199568748474121, "loss_alignment": 0.9599609375, "loss_alignment_w": 0.47998046875, "loss_sft": 0.2830113470554352, "loss_total_v6": 0.7624730542302132, "step": 3215 }, { "epoch": 0.741708108108108, "grad_norm": 24.88933563232422, "learning_rate": 6.7428602366090764e-06, "loss": 12.167320251464844, "loss_alignment": 0.95703125, "loss_alignment_w": 0.478515625, "loss_sft": 0.2814383804798126, "loss_total_v6": 0.7604575529694557, "step": 3216 }, { "epoch": 0.7419387387387387, "grad_norm": 26.286449432373047, "learning_rate": 6.740972557978995e-06, "loss": 12.17575454711914, "loss_alignment": 0.9365234375, "loss_alignment_w": 0.46826171875, "loss_sft": 0.29321120493113995, "loss_total_v6": 0.7609846442937851, "step": 3217 }, { "epoch": 0.7421693693693694, "grad_norm": 26.999757766723633, "learning_rate": 6.739084596924353e-06, "loss": 11.560463905334473, "loss_alignment": 0.931640625, "loss_alignment_w": 0.4658203125, "loss_sft": 0.25638823956251144, "loss_total_v6": 0.7225290015339851, "step": 3218 }, { "epoch": 0.7424, "grad_norm": 24.199981689453125, "learning_rate": 6.737196353751423e-06, "loss": 12.473945617675781, "loss_alignment": 0.92626953125, "loss_alignment_w": 0.463134765625, "loss_sft": 0.3165173213928938, "loss_total_v6": 0.7796215862035751, "step": 3219 }, { "epoch": 0.7426306306306306, "grad_norm": 23.525623321533203, "learning_rate": 6.735307828766515e-06, "loss": 11.694913864135742, "loss_alignment": 0.93505859375, "loss_alignment_w": 0.467529296875, "loss_sft": 0.2630365788936615, "loss_total_v6": 0.7309321165084839, "step": 3220 }, { "epoch": 0.7428612612612613, "grad_norm": 25.9883975982666, "learning_rate": 6.733419022275995e-06, "loss": 12.39391803741455, "loss_alignment": 0.92626953125, "loss_alignment_w": 0.463134765625, "loss_sft": 0.31117987632751465, "loss_total_v6": 0.7746198400855064, "step": 3221 }, { "epoch": 0.7430918918918918, "grad_norm": 26.870969772338867, "learning_rate": 6.731529934586267e-06, "loss": 12.365766525268555, "loss_alignment": 0.95849609375, "loss_alignment_w": 0.479248046875, "loss_sft": 0.2939938101917505, "loss_total_v6": 0.7728604078292847, "step": 3222 }, { "epoch": 0.7433225225225225, "grad_norm": 25.20780372619629, "learning_rate": 6.729640566003785e-06, "loss": 12.025927543640137, "loss_alignment": 0.93798828125, "loss_alignment_w": 0.468994140625, "loss_sft": 0.2815734390169382, "loss_total_v6": 0.7516204416751862, "step": 3223 }, { "epoch": 0.7435531531531532, "grad_norm": 24.055625915527344, "learning_rate": 6.7277509168350445e-06, "loss": 12.268239974975586, "loss_alignment": 0.94482421875, "loss_alignment_w": 0.472412109375, "loss_sft": 0.2935441490262747, "loss_total_v6": 0.7667649611830711, "step": 3224 }, { "epoch": 0.7437837837837837, "grad_norm": 31.99447250366211, "learning_rate": 6.7258609873865895e-06, "loss": 12.138716697692871, "loss_alignment": 0.9267578125, "loss_alignment_w": 0.46337890625, "loss_sft": 0.29561130702495575, "loss_total_v6": 0.7586698085069656, "step": 3225 }, { "epoch": 0.7440144144144144, "grad_norm": 27.180198669433594, "learning_rate": 6.72397077796501e-06, "loss": 12.149343490600586, "loss_alignment": 0.89404296875, "loss_alignment_w": 0.447021484375, "loss_sft": 0.31167162023484707, "loss_total_v6": 0.7593339532613754, "step": 3226 }, { "epoch": 0.7442450450450451, "grad_norm": 25.877403259277344, "learning_rate": 6.722080288876936e-06, "loss": 12.87858772277832, "loss_alignment": 0.9599609375, "loss_alignment_w": 0.47998046875, "loss_sft": 0.3255873769521713, "loss_total_v6": 0.8049117177724838, "step": 3227 }, { "epoch": 0.7444756756756756, "grad_norm": 24.51775550842285, "learning_rate": 6.720189520429052e-06, "loss": 12.060554504394531, "loss_alignment": 0.95166015625, "loss_alignment_w": 0.475830078125, "loss_sft": 0.27887003868818283, "loss_total_v6": 0.753784604370594, "step": 3228 }, { "epoch": 0.7447063063063063, "grad_norm": 25.27400779724121, "learning_rate": 6.718298472928082e-06, "loss": 11.79014778137207, "loss_alignment": 0.92529296875, "loss_alignment_w": 0.462646484375, "loss_sft": 0.27489390037953854, "loss_total_v6": 0.7368842586874962, "step": 3229 }, { "epoch": 0.744936936936937, "grad_norm": 27.007661819458008, "learning_rate": 6.716407146680793e-06, "loss": 12.528303146362305, "loss_alignment": 0.94970703125, "loss_alignment_w": 0.474853515625, "loss_sft": 0.30827223509550095, "loss_total_v6": 0.7830189391970634, "step": 3230 }, { "epoch": 0.7451675675675675, "grad_norm": 24.838211059570312, "learning_rate": 6.714515541994002e-06, "loss": 12.591413497924805, "loss_alignment": 0.94384765625, "loss_alignment_w": 0.471923828125, "loss_sft": 0.3148868791759014, "loss_total_v6": 0.7869632989168167, "step": 3231 }, { "epoch": 0.7453981981981982, "grad_norm": 22.970134735107422, "learning_rate": 6.712623659174569e-06, "loss": 12.65234375, "loss_alignment": 0.94921875, "loss_alignment_w": 0.474609375, "loss_sft": 0.3161468394100666, "loss_total_v6": 0.7907714918255806, "step": 3232 }, { "epoch": 0.7456288288288289, "grad_norm": 24.818899154663086, "learning_rate": 6.7107314985294e-06, "loss": 12.191532135009766, "loss_alignment": 0.9208984375, "loss_alignment_w": 0.46044921875, "loss_sft": 0.30120112374424934, "loss_total_v6": 0.7619707509875298, "step": 3233 }, { "epoch": 0.7458594594594594, "grad_norm": 23.75156021118164, "learning_rate": 6.708839060365449e-06, "loss": 12.354751586914062, "loss_alignment": 0.9443359375, "loss_alignment_w": 0.47216796875, "loss_sft": 0.30024815164506435, "loss_total_v6": 0.7721719816327095, "step": 3234 }, { "epoch": 0.7460900900900901, "grad_norm": 25.796649932861328, "learning_rate": 6.706946344989704e-06, "loss": 12.400472640991211, "loss_alignment": 0.96044921875, "loss_alignment_w": 0.480224609375, "loss_sft": 0.2951253894716501, "loss_total_v6": 0.7750295549631119, "step": 3235 }, { "epoch": 0.7463207207207208, "grad_norm": 25.687623977661133, "learning_rate": 6.705053352709212e-06, "loss": 12.405288696289062, "loss_alignment": 0.94189453125, "loss_alignment_w": 0.470947265625, "loss_sft": 0.3030404821038246, "loss_total_v6": 0.775330550968647, "step": 3236 }, { "epoch": 0.7465513513513513, "grad_norm": 25.37873649597168, "learning_rate": 6.703160083831056e-06, "loss": 11.925952911376953, "loss_alignment": 0.93310546875, "loss_alignment_w": 0.466552734375, "loss_sft": 0.2794144097715616, "loss_total_v6": 0.745372049510479, "step": 3237 }, { "epoch": 0.746781981981982, "grad_norm": 24.1702938079834, "learning_rate": 6.701266538662368e-06, "loss": 11.373611450195312, "loss_alignment": 0.9130859375, "loss_alignment_w": 0.45654296875, "loss_sft": 0.25551318004727364, "loss_total_v6": 0.7108507081866264, "step": 3238 }, { "epoch": 0.7470126126126126, "grad_norm": 27.024301528930664, "learning_rate": 6.699372717510321e-06, "loss": 12.5531005859375, "loss_alignment": 0.93896484375, "loss_alignment_w": 0.469482421875, "loss_sft": 0.31488801166415215, "loss_total_v6": 0.7845687940716743, "step": 3239 }, { "epoch": 0.7472432432432432, "grad_norm": 25.59103775024414, "learning_rate": 6.697478620682137e-06, "loss": 12.589470863342285, "loss_alignment": 0.9521484375, "loss_alignment_w": 0.47607421875, "loss_sft": 0.3096843399107456, "loss_total_v6": 0.7868419289588928, "step": 3240 }, { "epoch": 0.7474738738738739, "grad_norm": 23.46809196472168, "learning_rate": 6.69558424848508e-06, "loss": 11.525700569152832, "loss_alignment": 0.91845703125, "loss_alignment_w": 0.459228515625, "loss_sft": 0.26018174178898335, "loss_total_v6": 0.7203563004732132, "step": 3241 }, { "epoch": 0.7477045045045045, "grad_norm": 24.56229591369629, "learning_rate": 6.693689601226458e-06, "loss": 12.491231918334961, "loss_alignment": 0.9453125, "loss_alignment_w": 0.47265625, "loss_sft": 0.30755744874477386, "loss_total_v6": 0.7807020097970963, "step": 3242 }, { "epoch": 0.7479351351351351, "grad_norm": 24.825634002685547, "learning_rate": 6.691794679213627e-06, "loss": 12.37324047088623, "loss_alignment": 0.94970703125, "loss_alignment_w": 0.474853515625, "loss_sft": 0.2993742600083351, "loss_total_v6": 0.7733275294303894, "step": 3243 }, { "epoch": 0.7481657657657658, "grad_norm": 23.373905181884766, "learning_rate": 6.689899482753984e-06, "loss": 11.995612144470215, "loss_alignment": 0.962890625, "loss_alignment_w": 0.4814453125, "loss_sft": 0.26799052581191063, "loss_total_v6": 0.7497257515788078, "step": 3244 }, { "epoch": 0.7483963963963964, "grad_norm": 25.649242401123047, "learning_rate": 6.688004012154975e-06, "loss": 12.426469802856445, "loss_alignment": 0.95361328125, "loss_alignment_w": 0.476806640625, "loss_sft": 0.29986295476555824, "loss_total_v6": 0.7766543254256248, "step": 3245 }, { "epoch": 0.748627027027027, "grad_norm": 25.190845489501953, "learning_rate": 6.686108267724085e-06, "loss": 11.927513122558594, "loss_alignment": 0.91845703125, "loss_alignment_w": 0.459228515625, "loss_sft": 0.28685134276747704, "loss_total_v6": 0.7454695254564285, "step": 3246 }, { "epoch": 0.7488576576576577, "grad_norm": 22.4102725982666, "learning_rate": 6.684212249768845e-06, "loss": 12.25119400024414, "loss_alignment": 0.9462890625, "loss_alignment_w": 0.47314453125, "loss_sft": 0.2933027632534504, "loss_total_v6": 0.7656995952129364, "step": 3247 }, { "epoch": 0.7490882882882883, "grad_norm": 25.043785095214844, "learning_rate": 6.6823159585968355e-06, "loss": 12.611003875732422, "loss_alignment": 0.93798828125, "loss_alignment_w": 0.468994140625, "loss_sft": 0.32026174291968346, "loss_total_v6": 0.7881877571344376, "step": 3248 }, { "epoch": 0.7493189189189189, "grad_norm": 23.712785720825195, "learning_rate": 6.680419394515672e-06, "loss": 11.902573585510254, "loss_alignment": 0.92724609375, "loss_alignment_w": 0.463623046875, "loss_sft": 0.2804556265473366, "loss_total_v6": 0.7439108267426491, "step": 3249 }, { "epoch": 0.7495495495495496, "grad_norm": 24.59575080871582, "learning_rate": 6.678522557833025e-06, "loss": 11.895355224609375, "loss_alignment": 0.9619140625, "loss_alignment_w": 0.48095703125, "loss_sft": 0.2620143909007311, "loss_total_v6": 0.743459664285183, "step": 3250 }, { "epoch": 0.7497801801801802, "grad_norm": 25.232830047607422, "learning_rate": 6.6766254488566e-06, "loss": 12.107919692993164, "loss_alignment": 0.94873046875, "loss_alignment_w": 0.474365234375, "loss_sft": 0.28308167681097984, "loss_total_v6": 0.7567450180649757, "step": 3251 }, { "epoch": 0.7500108108108108, "grad_norm": 26.147502899169922, "learning_rate": 6.674728067894149e-06, "loss": 11.95089340209961, "loss_alignment": 0.92431640625, "loss_alignment_w": 0.462158203125, "loss_sft": 0.28446751087903976, "loss_total_v6": 0.7469308748841286, "step": 3252 }, { "epoch": 0.7502414414414414, "grad_norm": 23.208208084106445, "learning_rate": 6.672830415253475e-06, "loss": 11.8381986618042, "loss_alignment": 0.9580078125, "loss_alignment_w": 0.47900390625, "loss_sft": 0.2599832247942686, "loss_total_v6": 0.7398874014616013, "step": 3253 }, { "epoch": 0.7504720720720721, "grad_norm": 25.16766929626465, "learning_rate": 6.670932491242414e-06, "loss": 11.780546188354492, "loss_alignment": 0.91064453125, "loss_alignment_w": 0.455322265625, "loss_sft": 0.2818621341139078, "loss_total_v6": 0.7362841293215752, "step": 3254 }, { "epoch": 0.7507027027027027, "grad_norm": 26.880905151367188, "learning_rate": 6.669034296168855e-06, "loss": 12.65074348449707, "loss_alignment": 0.95166015625, "loss_alignment_w": 0.475830078125, "loss_sft": 0.3152991198003292, "loss_total_v6": 0.7906714603304863, "step": 3255 }, { "epoch": 0.7509333333333333, "grad_norm": 25.305774688720703, "learning_rate": 6.667135830340727e-06, "loss": 11.728452682495117, "loss_alignment": 0.93115234375, "loss_alignment_w": 0.465576171875, "loss_sft": 0.2678336026147008, "loss_total_v6": 0.7330282554030418, "step": 3256 }, { "epoch": 0.751163963963964, "grad_norm": 25.019065856933594, "learning_rate": 6.665237094066002e-06, "loss": 12.351593971252441, "loss_alignment": 0.93359375, "loss_alignment_w": 0.466796875, "loss_sft": 0.3050251975655556, "loss_total_v6": 0.7719746306538582, "step": 3257 }, { "epoch": 0.7513945945945946, "grad_norm": 24.59293556213379, "learning_rate": 6.663338087652697e-06, "loss": 12.383551597595215, "loss_alignment": 0.94775390625, "loss_alignment_w": 0.473876953125, "loss_sft": 0.30169717967510223, "loss_total_v6": 0.7739719599485397, "step": 3258 }, { "epoch": 0.7516252252252252, "grad_norm": 25.030616760253906, "learning_rate": 6.661438811408876e-06, "loss": 11.830633163452148, "loss_alignment": 0.95654296875, "loss_alignment_w": 0.478271484375, "loss_sft": 0.26077692210674286, "loss_total_v6": 0.7394146099686623, "step": 3259 }, { "epoch": 0.7518558558558559, "grad_norm": 26.937082290649414, "learning_rate": 6.659539265642643e-06, "loss": 12.914186477661133, "loss_alignment": 0.95849609375, "loss_alignment_w": 0.479248046875, "loss_sft": 0.3277207836508751, "loss_total_v6": 0.8071366772055626, "step": 3260 }, { "epoch": 0.7520864864864865, "grad_norm": 25.523828506469727, "learning_rate": 6.657639450662145e-06, "loss": 12.395015716552734, "loss_alignment": 0.9072265625, "loss_alignment_w": 0.45361328125, "loss_sft": 0.3203122764825821, "loss_total_v6": 0.7746885120868683, "step": 3261 }, { "epoch": 0.7523171171171171, "grad_norm": 26.664306640625, "learning_rate": 6.655739366775575e-06, "loss": 12.291730880737305, "loss_alignment": 0.91748046875, "loss_alignment_w": 0.458740234375, "loss_sft": 0.30949294194579124, "loss_total_v6": 0.768233172595501, "step": 3262 }, { "epoch": 0.7525477477477478, "grad_norm": 25.0908145904541, "learning_rate": 6.653839014291171e-06, "loss": 12.182523727416992, "loss_alignment": 0.9130859375, "loss_alignment_w": 0.45654296875, "loss_sft": 0.304437555372715, "loss_total_v6": 0.7614077627658844, "step": 3263 }, { "epoch": 0.7527783783783784, "grad_norm": 29.847251892089844, "learning_rate": 6.65193839351721e-06, "loss": 12.036979675292969, "loss_alignment": 0.9189453125, "loss_alignment_w": 0.45947265625, "loss_sft": 0.2918467093259096, "loss_total_v6": 0.7523111775517464, "step": 3264 }, { "epoch": 0.753009009009009, "grad_norm": 25.406164169311523, "learning_rate": 6.650037504762015e-06, "loss": 12.697540283203125, "loss_alignment": 0.93310546875, "loss_alignment_w": 0.466552734375, "loss_sft": 0.3263111859560013, "loss_total_v6": 0.7935963124036789, "step": 3265 }, { "epoch": 0.7532396396396397, "grad_norm": 24.82821273803711, "learning_rate": 6.648136348333955e-06, "loss": 12.224527359008789, "loss_alignment": 0.92626953125, "loss_alignment_w": 0.463134765625, "loss_sft": 0.30103555135428905, "loss_total_v6": 0.7640329599380493, "step": 3266 }, { "epoch": 0.7534702702702702, "grad_norm": 24.08056640625, "learning_rate": 6.646234924541438e-06, "loss": 11.962459564208984, "loss_alignment": 0.92431640625, "loss_alignment_w": 0.462158203125, "loss_sft": 0.28525138832628727, "loss_total_v6": 0.7476537153124809, "step": 3267 }, { "epoch": 0.7537009009009009, "grad_norm": 25.946720123291016, "learning_rate": 6.644333233692917e-06, "loss": 11.528709411621094, "loss_alignment": 0.92626953125, "loss_alignment_w": 0.463134765625, "loss_sft": 0.2571959737688303, "loss_total_v6": 0.7205443382263184, "step": 3268 }, { "epoch": 0.7539315315315316, "grad_norm": 23.121524810791016, "learning_rate": 6.64243127609689e-06, "loss": 12.151339530944824, "loss_alignment": 0.94677734375, "loss_alignment_w": 0.473388671875, "loss_sft": 0.2859784886240959, "loss_total_v6": 0.7594587132334709, "step": 3269 }, { "epoch": 0.7541621621621621, "grad_norm": 26.192996978759766, "learning_rate": 6.640529052061897e-06, "loss": 12.153291702270508, "loss_alignment": 0.91796875, "loss_alignment_w": 0.458984375, "loss_sft": 0.30132879689335823, "loss_total_v6": 0.7595807388424873, "step": 3270 }, { "epoch": 0.7543927927927928, "grad_norm": 23.617280960083008, "learning_rate": 6.638626561896519e-06, "loss": 11.915847778320312, "loss_alignment": 0.9404296875, "loss_alignment_w": 0.47021484375, "loss_sft": 0.27557849511504173, "loss_total_v6": 0.7447405159473419, "step": 3271 }, { "epoch": 0.7546234234234235, "grad_norm": 37.4443359375, "learning_rate": 6.636723805909384e-06, "loss": 12.489377975463867, "loss_alignment": 0.9482421875, "loss_alignment_w": 0.47412109375, "loss_sft": 0.3071974888443947, "loss_total_v6": 0.7805861234664917, "step": 3272 }, { "epoch": 0.754854054054054, "grad_norm": 27.581506729125977, "learning_rate": 6.634820784409163e-06, "loss": 12.233322143554688, "loss_alignment": 0.953125, "loss_alignment_w": 0.4765625, "loss_sft": 0.2884626295417547, "loss_total_v6": 0.7645826116204262, "step": 3273 }, { "epoch": 0.7550846846846847, "grad_norm": 28.445049285888672, "learning_rate": 6.6329174977045644e-06, "loss": 12.371036529541016, "loss_alignment": 0.9580078125, "loss_alignment_w": 0.47900390625, "loss_sft": 0.2957422863692045, "loss_total_v6": 0.7731897681951523, "step": 3274 }, { "epoch": 0.7553153153153153, "grad_norm": 23.865964889526367, "learning_rate": 6.631013946104348e-06, "loss": 12.34591293334961, "loss_alignment": 0.9501953125, "loss_alignment_w": 0.47509765625, "loss_sft": 0.29713230580091476, "loss_total_v6": 0.7716196030378342, "step": 3275 }, { "epoch": 0.7555459459459459, "grad_norm": 26.04497718811035, "learning_rate": 6.629110129917309e-06, "loss": 12.379222869873047, "loss_alignment": 0.9560546875, "loss_alignment_w": 0.47802734375, "loss_sft": 0.2970473989844322, "loss_total_v6": 0.7737014666199684, "step": 3276 }, { "epoch": 0.7557765765765766, "grad_norm": 27.501399993896484, "learning_rate": 6.62720604945229e-06, "loss": 11.899053573608398, "loss_alignment": 0.90576171875, "loss_alignment_w": 0.452880859375, "loss_sft": 0.2911609187722206, "loss_total_v6": 0.7436908334493637, "step": 3277 }, { "epoch": 0.7560072072072072, "grad_norm": 23.41965675354004, "learning_rate": 6.625301705018178e-06, "loss": 11.97484302520752, "loss_alignment": 0.9453125, "loss_alignment_w": 0.47265625, "loss_sft": 0.27595449052751064, "loss_total_v6": 0.7484276741743088, "step": 3278 }, { "epoch": 0.7562378378378378, "grad_norm": 25.746706008911133, "learning_rate": 6.623397096923895e-06, "loss": 11.894767761230469, "loss_alignment": 0.92919921875, "loss_alignment_w": 0.464599609375, "loss_sft": 0.27879285626113415, "loss_total_v6": 0.7434229776263237, "step": 3279 }, { "epoch": 0.7564684684684685, "grad_norm": 26.081823348999023, "learning_rate": 6.6214922254784145e-06, "loss": 12.14565658569336, "loss_alignment": 0.943359375, "loss_alignment_w": 0.4716796875, "loss_sft": 0.28670668601989746, "loss_total_v6": 0.7591035142540932, "step": 3280 }, { "epoch": 0.756699099099099, "grad_norm": 23.475481033325195, "learning_rate": 6.619587090990748e-06, "loss": 12.549571990966797, "loss_alignment": 0.96337890625, "loss_alignment_w": 0.481689453125, "loss_sft": 0.3025366887450218, "loss_total_v6": 0.7843482196331024, "step": 3281 }, { "epoch": 0.7569297297297297, "grad_norm": 23.265029907226562, "learning_rate": 6.617681693769951e-06, "loss": 12.32142162322998, "loss_alignment": 0.93505859375, "loss_alignment_w": 0.467529296875, "loss_sft": 0.30289526656270027, "loss_total_v6": 0.7700888589024544, "step": 3282 }, { "epoch": 0.7571603603603604, "grad_norm": 29.18403434753418, "learning_rate": 6.615776034125121e-06, "loss": 12.334030151367188, "loss_alignment": 0.94140625, "loss_alignment_w": 0.470703125, "loss_sft": 0.3010129816830158, "loss_total_v6": 0.7708768844604492, "step": 3283 }, { "epoch": 0.7573909909909909, "grad_norm": 27.261287689208984, "learning_rate": 6.613870112365399e-06, "loss": 12.623832702636719, "loss_alignment": 0.96728515625, "loss_alignment_w": 0.483642578125, "loss_sft": 0.3062319979071617, "loss_total_v6": 0.7889895364642143, "step": 3284 }, { "epoch": 0.7576216216216216, "grad_norm": 24.910099029541016, "learning_rate": 6.611963928799968e-06, "loss": 12.123236656188965, "loss_alignment": 0.9365234375, "loss_alignment_w": 0.46826171875, "loss_sft": 0.28922696225345135, "loss_total_v6": 0.7577022984623909, "step": 3285 }, { "epoch": 0.7578522522522523, "grad_norm": 25.936716079711914, "learning_rate": 6.6100574837380525e-06, "loss": 12.137060165405273, "loss_alignment": 0.92041015625, "loss_alignment_w": 0.460205078125, "loss_sft": 0.2982086017727852, "loss_total_v6": 0.7585662677884102, "step": 3286 }, { "epoch": 0.7580828828828828, "grad_norm": 24.195653915405273, "learning_rate": 6.608150777488922e-06, "loss": 12.023857116699219, "loss_alignment": 0.9189453125, "loss_alignment_w": 0.45947265625, "loss_sft": 0.2924303635954857, "loss_total_v6": 0.7514910325407982, "step": 3287 }, { "epoch": 0.7583135135135135, "grad_norm": 26.382177352905273, "learning_rate": 6.606243810361885e-06, "loss": 12.250910758972168, "loss_alignment": 0.93505859375, "loss_alignment_w": 0.467529296875, "loss_sft": 0.2989918664097786, "loss_total_v6": 0.7656819075345993, "step": 3288 }, { "epoch": 0.7585441441441442, "grad_norm": 25.74786376953125, "learning_rate": 6.604336582666298e-06, "loss": 12.178155899047852, "loss_alignment": 0.94140625, "loss_alignment_w": 0.470703125, "loss_sft": 0.2907978128641844, "loss_total_v6": 0.7611347436904907, "step": 3289 }, { "epoch": 0.7587747747747747, "grad_norm": 27.87723731994629, "learning_rate": 6.602429094711549e-06, "loss": 12.742487907409668, "loss_alignment": 0.91552734375, "loss_alignment_w": 0.457763671875, "loss_sft": 0.33851975202560425, "loss_total_v6": 0.7964054867625237, "step": 3290 }, { "epoch": 0.7590054054054054, "grad_norm": 27.865093231201172, "learning_rate": 6.600521346807081e-06, "loss": 12.215639114379883, "loss_alignment": 0.935546875, "loss_alignment_w": 0.4677734375, "loss_sft": 0.2959023602306843, "loss_total_v6": 0.7634774073958397, "step": 3291 }, { "epoch": 0.7592360360360361, "grad_norm": 31.61392593383789, "learning_rate": 6.598613339262369e-06, "loss": 12.256237030029297, "loss_alignment": 0.92919921875, "loss_alignment_w": 0.464599609375, "loss_sft": 0.3017204534262419, "loss_total_v6": 0.7660148590803146, "step": 3292 }, { "epoch": 0.7594666666666666, "grad_norm": 26.459228515625, "learning_rate": 6.596705072386938e-06, "loss": 12.035226821899414, "loss_alignment": 0.93115234375, "loss_alignment_w": 0.465576171875, "loss_sft": 0.28641192242503166, "loss_total_v6": 0.7522016912698746, "step": 3293 }, { "epoch": 0.7596972972972973, "grad_norm": 24.68377685546875, "learning_rate": 6.594796546490351e-06, "loss": 11.747329711914062, "loss_alignment": 0.91259765625, "loss_alignment_w": 0.456298828125, "loss_sft": 0.2763376273214817, "loss_total_v6": 0.7342081069946289, "step": 3294 }, { "epoch": 0.759927927927928, "grad_norm": 29.565959930419922, "learning_rate": 6.592887761882212e-06, "loss": 11.993834495544434, "loss_alignment": 0.95751953125, "loss_alignment_w": 0.478759765625, "loss_sft": 0.27054969407618046, "loss_total_v6": 0.7496146485209465, "step": 3295 }, { "epoch": 0.7601585585585585, "grad_norm": 24.38688087463379, "learning_rate": 6.590978718872166e-06, "loss": 12.36219596862793, "loss_alignment": 0.9091796875, "loss_alignment_w": 0.45458984375, "loss_sft": 0.3182304874062538, "loss_total_v6": 0.7726372331380844, "step": 3296 }, { "epoch": 0.7603891891891892, "grad_norm": 27.124719619750977, "learning_rate": 6.5890694177699085e-06, "loss": 11.947919845581055, "loss_alignment": 0.92333984375, "loss_alignment_w": 0.461669921875, "loss_sft": 0.2854260206222534, "loss_total_v6": 0.7467449903488159, "step": 3297 }, { "epoch": 0.7606198198198199, "grad_norm": 27.2324275970459, "learning_rate": 6.587159858885166e-06, "loss": 11.766342163085938, "loss_alignment": 0.88916015625, "loss_alignment_w": 0.444580078125, "loss_sft": 0.2901601307094097, "loss_total_v6": 0.7353963628411293, "step": 3298 }, { "epoch": 0.7608504504504504, "grad_norm": 25.355932235717773, "learning_rate": 6.585250042527712e-06, "loss": 11.926448822021484, "loss_alignment": 0.96044921875, "loss_alignment_w": 0.480224609375, "loss_sft": 0.26461380906403065, "loss_total_v6": 0.7454029992222786, "step": 3299 }, { "epoch": 0.7610810810810811, "grad_norm": 27.043289184570312, "learning_rate": 6.583339969007364e-06, "loss": 12.432210922241211, "loss_alignment": 0.9404296875, "loss_alignment_w": 0.47021484375, "loss_sft": 0.3065390009433031, "loss_total_v6": 0.7770132347941399, "step": 3300 }, { "epoch": 0.7613117117117117, "grad_norm": 26.695354461669922, "learning_rate": 6.581429638633975e-06, "loss": 12.224342346191406, "loss_alignment": 0.92822265625, "loss_alignment_w": 0.464111328125, "loss_sft": 0.29925398528575897, "loss_total_v6": 0.7640214487910271, "step": 3301 }, { "epoch": 0.7615423423423423, "grad_norm": 24.73065757751465, "learning_rate": 6.5795190517174446e-06, "loss": 11.970205307006836, "loss_alignment": 0.92626953125, "loss_alignment_w": 0.463134765625, "loss_sft": 0.28506410494446754, "loss_total_v6": 0.7481378465890884, "step": 3302 }, { "epoch": 0.761772972972973, "grad_norm": 24.686153411865234, "learning_rate": 6.577608208567714e-06, "loss": 12.267032623291016, "loss_alignment": 0.92724609375, "loss_alignment_w": 0.463623046875, "loss_sft": 0.3028986044228077, "loss_total_v6": 0.7666895017027855, "step": 3303 }, { "epoch": 0.7620036036036036, "grad_norm": 26.818695068359375, "learning_rate": 6.575697109494763e-06, "loss": 11.772871017456055, "loss_alignment": 0.9189453125, "loss_alignment_w": 0.45947265625, "loss_sft": 0.2766064628958702, "loss_total_v6": 0.7358044534921646, "step": 3304 }, { "epoch": 0.7622342342342342, "grad_norm": 25.360876083374023, "learning_rate": 6.573785754808615e-06, "loss": 12.077983856201172, "loss_alignment": 0.9072265625, "loss_alignment_w": 0.45361328125, "loss_sft": 0.3003756944090128, "loss_total_v6": 0.7548739910125732, "step": 3305 }, { "epoch": 0.7624648648648649, "grad_norm": 24.086368560791016, "learning_rate": 6.571874144819333e-06, "loss": 12.173271179199219, "loss_alignment": 0.94287109375, "loss_alignment_w": 0.471435546875, "loss_sft": 0.2886462192982435, "loss_total_v6": 0.7608294859528542, "step": 3306 }, { "epoch": 0.7626954954954955, "grad_norm": 24.369272232055664, "learning_rate": 6.569962279837025e-06, "loss": 12.440983772277832, "loss_alignment": 0.92724609375, "loss_alignment_w": 0.463623046875, "loss_sft": 0.3132365718483925, "loss_total_v6": 0.7775615081191063, "step": 3307 }, { "epoch": 0.7629261261261261, "grad_norm": 25.722938537597656, "learning_rate": 6.568050160171837e-06, "loss": 12.066120147705078, "loss_alignment": 0.9365234375, "loss_alignment_w": 0.46826171875, "loss_sft": 0.28664898686110973, "loss_total_v6": 0.7541324943304062, "step": 3308 }, { "epoch": 0.7631567567567568, "grad_norm": 25.01263999938965, "learning_rate": 6.566137786133957e-06, "loss": 12.188444137573242, "loss_alignment": 0.95361328125, "loss_alignment_w": 0.476806640625, "loss_sft": 0.28407085686922073, "loss_total_v6": 0.761777751147747, "step": 3309 }, { "epoch": 0.7633873873873874, "grad_norm": 26.119136810302734, "learning_rate": 6.564225158033616e-06, "loss": 12.423465728759766, "loss_alignment": 0.923828125, "loss_alignment_w": 0.4619140625, "loss_sft": 0.31351498141884804, "loss_total_v6": 0.776466615498066, "step": 3310 }, { "epoch": 0.763618018018018, "grad_norm": 26.567344665527344, "learning_rate": 6.562312276181083e-06, "loss": 12.342053413391113, "loss_alignment": 0.93310546875, "loss_alignment_w": 0.466552734375, "loss_sft": 0.3057868666946888, "loss_total_v6": 0.7713783010840416, "step": 3311 }, { "epoch": 0.7638486486486487, "grad_norm": 24.520055770874023, "learning_rate": 6.560399140886673e-06, "loss": 12.313713073730469, "loss_alignment": 0.93994140625, "loss_alignment_w": 0.469970703125, "loss_sft": 0.3000178150832653, "loss_total_v6": 0.7696070522069931, "step": 3312 }, { "epoch": 0.7640792792792793, "grad_norm": 26.38159942626953, "learning_rate": 6.558485752460735e-06, "loss": 12.24026870727539, "loss_alignment": 0.92626953125, "loss_alignment_w": 0.463134765625, "loss_sft": 0.3031027838587761, "loss_total_v6": 0.7650168389081955, "step": 3313 }, { "epoch": 0.7643099099099099, "grad_norm": 26.058412551879883, "learning_rate": 6.556572111213667e-06, "loss": 12.223355293273926, "loss_alignment": 0.912109375, "loss_alignment_w": 0.4560546875, "loss_sft": 0.30741674825549126, "loss_total_v6": 0.7639596983790398, "step": 3314 }, { "epoch": 0.7645405405405405, "grad_norm": 26.723880767822266, "learning_rate": 6.554658217455902e-06, "loss": 12.270100593566895, "loss_alignment": 0.9580078125, "loss_alignment_w": 0.47900390625, "loss_sft": 0.28821307234466076, "loss_total_v6": 0.7668812721967697, "step": 3315 }, { "epoch": 0.7647711711711712, "grad_norm": 24.641090393066406, "learning_rate": 6.552744071497918e-06, "loss": 11.593441009521484, "loss_alignment": 0.921875, "loss_alignment_w": 0.4609375, "loss_sft": 0.26337792351841927, "loss_total_v6": 0.7245901077985764, "step": 3316 }, { "epoch": 0.7650018018018018, "grad_norm": 24.786861419677734, "learning_rate": 6.550829673650231e-06, "loss": 12.212435722351074, "loss_alignment": 0.8974609375, "loss_alignment_w": 0.44873046875, "loss_sft": 0.31331077590584755, "loss_total_v6": 0.7632772400975227, "step": 3317 }, { "epoch": 0.7652324324324324, "grad_norm": 26.20187759399414, "learning_rate": 6.548915024223399e-06, "loss": 12.80499267578125, "loss_alignment": 0.9287109375, "loss_alignment_w": 0.46435546875, "loss_sft": 0.3356666825711727, "loss_total_v6": 0.8003120794892311, "step": 3318 }, { "epoch": 0.7654630630630631, "grad_norm": 37.12614440917969, "learning_rate": 6.547000123528022e-06, "loss": 12.140254974365234, "loss_alignment": 0.93505859375, "loss_alignment_w": 0.467529296875, "loss_sft": 0.29230472445487976, "loss_total_v6": 0.7587659060955048, "step": 3319 }, { "epoch": 0.7656936936936937, "grad_norm": 27.423147201538086, "learning_rate": 6.545084971874738e-06, "loss": 11.561027526855469, "loss_alignment": 0.92138671875, "loss_alignment_w": 0.460693359375, "loss_sft": 0.26162674836814404, "loss_total_v6": 0.7225642278790474, "step": 3320 }, { "epoch": 0.7659243243243243, "grad_norm": 28.562091827392578, "learning_rate": 6.543169569574228e-06, "loss": 12.793184280395508, "loss_alignment": 0.94091796875, "loss_alignment_w": 0.470458984375, "loss_sft": 0.32949646562337875, "loss_total_v6": 0.7995739951729774, "step": 3321 }, { "epoch": 0.766154954954955, "grad_norm": 25.630203247070312, "learning_rate": 6.541253916937214e-06, "loss": 12.124002456665039, "loss_alignment": 0.94384765625, "loss_alignment_w": 0.471923828125, "loss_sft": 0.2859636154025793, "loss_total_v6": 0.7577501237392426, "step": 3322 }, { "epoch": 0.7663855855855856, "grad_norm": 23.614660263061523, "learning_rate": 6.539338014274454e-06, "loss": 11.717673301696777, "loss_alignment": 0.92138671875, "loss_alignment_w": 0.460693359375, "loss_sft": 0.2724394388496876, "loss_total_v6": 0.7323545962572098, "step": 3323 }, { "epoch": 0.7666162162162162, "grad_norm": 26.56806182861328, "learning_rate": 6.537421861896754e-06, "loss": 11.917200088500977, "loss_alignment": 0.93603515625, "loss_alignment_w": 0.468017578125, "loss_sft": 0.27653276547789574, "loss_total_v6": 0.7448249980807304, "step": 3324 }, { "epoch": 0.7668468468468469, "grad_norm": 26.463741302490234, "learning_rate": 6.5355054601149545e-06, "loss": 12.253780364990234, "loss_alignment": 0.9443359375, "loss_alignment_w": 0.47216796875, "loss_sft": 0.2947613727301359, "loss_total_v6": 0.765861265361309, "step": 3325 }, { "epoch": 0.7670774774774775, "grad_norm": 25.637605667114258, "learning_rate": 6.5335888092399415e-06, "loss": 12.234248161315918, "loss_alignment": 0.955078125, "loss_alignment_w": 0.4775390625, "loss_sft": 0.2871014419943094, "loss_total_v6": 0.7646405175328255, "step": 3326 }, { "epoch": 0.7673081081081081, "grad_norm": 43.171321868896484, "learning_rate": 6.531671909582634e-06, "loss": 13.035324096679688, "loss_alignment": 0.95556640625, "loss_alignment_w": 0.477783203125, "loss_sft": 0.33764174953103065, "loss_total_v6": 0.8147077932953835, "step": 3327 }, { "epoch": 0.7675387387387388, "grad_norm": 26.772064208984375, "learning_rate": 6.529754761453999e-06, "loss": 11.956830978393555, "loss_alignment": 0.94287109375, "loss_alignment_w": 0.471435546875, "loss_sft": 0.27638516016304493, "loss_total_v6": 0.7473019137978554, "step": 3328 }, { "epoch": 0.7677693693693693, "grad_norm": 25.89600944519043, "learning_rate": 6.52783736516504e-06, "loss": 12.138450622558594, "loss_alignment": 0.93115234375, "loss_alignment_w": 0.465576171875, "loss_sft": 0.29393143951892853, "loss_total_v6": 0.7586531341075897, "step": 3329 }, { "epoch": 0.768, "grad_norm": 23.727497100830078, "learning_rate": 6.5259197210268e-06, "loss": 12.55551528930664, "loss_alignment": 0.9658203125, "loss_alignment_w": 0.48291015625, "loss_sft": 0.3024962358176708, "loss_total_v6": 0.7847197204828262, "step": 3330 }, { "epoch": 0.7682306306306307, "grad_norm": 26.55047035217285, "learning_rate": 6.524001829350366e-06, "loss": 12.364139556884766, "loss_alignment": 0.900390625, "loss_alignment_w": 0.4501953125, "loss_sft": 0.32236501201987267, "loss_total_v6": 0.7727586776018143, "step": 3331 }, { "epoch": 0.7684612612612612, "grad_norm": 25.295812606811523, "learning_rate": 6.522083690446863e-06, "loss": 12.205554962158203, "loss_alignment": 0.90576171875, "loss_alignment_w": 0.452880859375, "loss_sft": 0.3096306249499321, "loss_total_v6": 0.7628471553325653, "step": 3332 }, { "epoch": 0.7686918918918919, "grad_norm": 26.93442153930664, "learning_rate": 6.520165304627452e-06, "loss": 12.686786651611328, "loss_alignment": 0.96337890625, "loss_alignment_w": 0.481689453125, "loss_sft": 0.3111584708094597, "loss_total_v6": 0.7929242178797722, "step": 3333 }, { "epoch": 0.7689225225225226, "grad_norm": 25.26239013671875, "learning_rate": 6.5182466722033426e-06, "loss": 12.556869506835938, "loss_alignment": 0.95703125, "loss_alignment_w": 0.478515625, "loss_sft": 0.3060751222074032, "loss_total_v6": 0.7848043665289879, "step": 3334 }, { "epoch": 0.7691531531531531, "grad_norm": 25.357341766357422, "learning_rate": 6.516327793485776e-06, "loss": 11.443840026855469, "loss_alignment": 0.91552734375, "loss_alignment_w": 0.457763671875, "loss_sft": 0.2587733119726181, "loss_total_v6": 0.7152399867773056, "step": 3335 }, { "epoch": 0.7693837837837838, "grad_norm": 26.43051528930664, "learning_rate": 6.5144086687860385e-06, "loss": 12.469364166259766, "loss_alignment": 0.90625, "loss_alignment_w": 0.453125, "loss_sft": 0.3269273526966572, "loss_total_v6": 0.7793352156877518, "step": 3336 }, { "epoch": 0.7696144144144144, "grad_norm": 25.085670471191406, "learning_rate": 6.512489298415457e-06, "loss": 12.0423583984375, "loss_alignment": 0.9150390625, "loss_alignment_w": 0.45751953125, "loss_sft": 0.29611966386437416, "loss_total_v6": 0.7526473551988602, "step": 3337 }, { "epoch": 0.769845045045045, "grad_norm": 25.557388305664062, "learning_rate": 6.5105696826853935e-06, "loss": 12.29660415649414, "loss_alignment": 0.923828125, "loss_alignment_w": 0.4619140625, "loss_sft": 0.3061659149825573, "loss_total_v6": 0.7685377672314644, "step": 3338 }, { "epoch": 0.7700756756756757, "grad_norm": 25.001216888427734, "learning_rate": 6.508649821907254e-06, "loss": 12.164055824279785, "loss_alignment": 0.91650390625, "loss_alignment_w": 0.458251953125, "loss_sft": 0.30122334882616997, "loss_total_v6": 0.7602534964680672, "step": 3339 }, { "epoch": 0.7703063063063063, "grad_norm": 24.90357780456543, "learning_rate": 6.50672971639248e-06, "loss": 12.134023666381836, "loss_alignment": 0.9267578125, "loss_alignment_w": 0.46337890625, "loss_sft": 0.29534859023988247, "loss_total_v6": 0.7583765089511871, "step": 3340 }, { "epoch": 0.7705369369369369, "grad_norm": 25.30909538269043, "learning_rate": 6.50480936645256e-06, "loss": 12.387313842773438, "loss_alignment": 0.95703125, "loss_alignment_w": 0.478515625, "loss_sft": 0.29645441472530365, "loss_total_v6": 0.7742071002721786, "step": 3341 }, { "epoch": 0.7707675675675676, "grad_norm": 31.684246063232422, "learning_rate": 6.502888772399013e-06, "loss": 11.561660766601562, "loss_alignment": 0.916015625, "loss_alignment_w": 0.4580078125, "loss_sft": 0.26415350660681725, "loss_total_v6": 0.722603864967823, "step": 3342 }, { "epoch": 0.7709981981981981, "grad_norm": 24.621671676635742, "learning_rate": 6.5009679345434056e-06, "loss": 11.688657760620117, "loss_alignment": 0.91015625, "loss_alignment_w": 0.455078125, "loss_sft": 0.2747305817902088, "loss_total_v6": 0.7305411174893379, "step": 3343 }, { "epoch": 0.7712288288288288, "grad_norm": 24.38562774658203, "learning_rate": 6.499046853197338e-06, "loss": 12.196533203125, "loss_alignment": 0.9130859375, "loss_alignment_w": 0.45654296875, "loss_sft": 0.3052673973143101, "loss_total_v6": 0.7622833549976349, "step": 3344 }, { "epoch": 0.7714594594594595, "grad_norm": 27.138599395751953, "learning_rate": 6.497125528672453e-06, "loss": 11.618693351745605, "loss_alignment": 0.9052734375, "loss_alignment_w": 0.45263671875, "loss_sft": 0.27305855602025986, "loss_total_v6": 0.7261683195829391, "step": 3345 }, { "epoch": 0.77169009009009, "grad_norm": 26.181013107299805, "learning_rate": 6.495203961280434e-06, "loss": 12.451849937438965, "loss_alignment": 0.970703125, "loss_alignment_w": 0.4853515625, "loss_sft": 0.29250763170421124, "loss_total_v6": 0.7782406359910965, "step": 3346 }, { "epoch": 0.7719207207207207, "grad_norm": 26.475873947143555, "learning_rate": 6.493282151333e-06, "loss": 12.300088882446289, "loss_alignment": 0.9248046875, "loss_alignment_w": 0.46240234375, "loss_sft": 0.3050715234130621, "loss_total_v6": 0.7687555775046349, "step": 3347 }, { "epoch": 0.7721513513513514, "grad_norm": 25.281293869018555, "learning_rate": 6.491360099141913e-06, "loss": 12.33978271484375, "loss_alignment": 0.9375, "loss_alignment_w": 0.46875, "loss_sft": 0.3030815124511719, "loss_total_v6": 0.7712364196777344, "step": 3348 }, { "epoch": 0.7723819819819819, "grad_norm": 25.537179946899414, "learning_rate": 6.489437805018972e-06, "loss": 11.995424270629883, "loss_alignment": 0.9306640625, "loss_alignment_w": 0.46533203125, "loss_sft": 0.2854348383843899, "loss_total_v6": 0.7497140243649483, "step": 3349 }, { "epoch": 0.7726126126126126, "grad_norm": 24.298494338989258, "learning_rate": 6.487515269276015e-06, "loss": 11.667007446289062, "loss_alignment": 0.90234375, "loss_alignment_w": 0.451171875, "loss_sft": 0.2770090326666832, "loss_total_v6": 0.7291879877448082, "step": 3350 }, { "epoch": 0.7728432432432433, "grad_norm": 25.804244995117188, "learning_rate": 6.485592492224923e-06, "loss": 12.275154113769531, "loss_alignment": 0.94091796875, "loss_alignment_w": 0.470458984375, "loss_sft": 0.29769950546324253, "loss_total_v6": 0.7671971470117569, "step": 3351 }, { "epoch": 0.7730738738738738, "grad_norm": 25.59315299987793, "learning_rate": 6.483669474177609e-06, "loss": 12.141600608825684, "loss_alignment": 0.9375, "loss_alignment_w": 0.46875, "loss_sft": 0.2903900146484375, "loss_total_v6": 0.7588500753045082, "step": 3352 }, { "epoch": 0.7733045045045045, "grad_norm": 26.676027297973633, "learning_rate": 6.481746215446033e-06, "loss": 11.823293685913086, "loss_alignment": 0.92626953125, "loss_alignment_w": 0.463134765625, "loss_sft": 0.27516499534249306, "loss_total_v6": 0.7389558851718903, "step": 3353 }, { "epoch": 0.7735351351351352, "grad_norm": 27.252878189086914, "learning_rate": 6.47982271634219e-06, "loss": 12.042898178100586, "loss_alignment": 0.91650390625, "loss_alignment_w": 0.458251953125, "loss_sft": 0.2948869466781616, "loss_total_v6": 0.7526811510324478, "step": 3354 }, { "epoch": 0.7737657657657657, "grad_norm": 26.80589485168457, "learning_rate": 6.477898977178112e-06, "loss": 12.667320251464844, "loss_alignment": 0.94091796875, "loss_alignment_w": 0.470458984375, "loss_sft": 0.3216300345957279, "loss_total_v6": 0.7917075231671333, "step": 3355 }, { "epoch": 0.7739963963963964, "grad_norm": 25.997434616088867, "learning_rate": 6.475974998265875e-06, "loss": 12.193828582763672, "loss_alignment": 0.92724609375, "loss_alignment_w": 0.463623046875, "loss_sft": 0.29864384047687054, "loss_total_v6": 0.7621143013238907, "step": 3356 }, { "epoch": 0.7742270270270271, "grad_norm": 25.95760154724121, "learning_rate": 6.47405077991759e-06, "loss": 12.114627838134766, "loss_alignment": 0.9296875, "loss_alignment_w": 0.46484375, "loss_sft": 0.29357172548770905, "loss_total_v6": 0.757164254784584, "step": 3357 }, { "epoch": 0.7744576576576576, "grad_norm": 24.833086013793945, "learning_rate": 6.472126322445409e-06, "loss": 12.132160186767578, "loss_alignment": 0.93017578125, "loss_alignment_w": 0.465087890625, "loss_sft": 0.29373670741915703, "loss_total_v6": 0.758260004222393, "step": 3358 }, { "epoch": 0.7746882882882883, "grad_norm": 23.272733688354492, "learning_rate": 6.47020162616152e-06, "loss": 12.232651710510254, "loss_alignment": 0.91650390625, "loss_alignment_w": 0.458251953125, "loss_sft": 0.30621251091361046, "loss_total_v6": 0.7645407393574715, "step": 3359 }, { "epoch": 0.774918918918919, "grad_norm": 25.52269172668457, "learning_rate": 6.468276691378155e-06, "loss": 12.349090576171875, "loss_alignment": 0.92431640625, "loss_alignment_w": 0.462158203125, "loss_sft": 0.30918690375983715, "loss_total_v6": 0.7718181237578392, "step": 3360 }, { "epoch": 0.7751495495495495, "grad_norm": 26.282808303833008, "learning_rate": 6.466351518407578e-06, "loss": 11.945606231689453, "loss_alignment": 0.92041015625, "loss_alignment_w": 0.460205078125, "loss_sft": 0.2871887292712927, "loss_total_v6": 0.7466003522276878, "step": 3361 }, { "epoch": 0.7753801801801802, "grad_norm": 25.436824798583984, "learning_rate": 6.464426107562095e-06, "loss": 11.944072723388672, "loss_alignment": 0.91015625, "loss_alignment_w": 0.455078125, "loss_sft": 0.29176211915910244, "loss_total_v6": 0.7465045526623726, "step": 3362 }, { "epoch": 0.7756108108108108, "grad_norm": 26.24846076965332, "learning_rate": 6.462500459154054e-06, "loss": 12.169607162475586, "loss_alignment": 0.89306640625, "loss_alignment_w": 0.446533203125, "loss_sft": 0.3156541511416435, "loss_total_v6": 0.7606004402041435, "step": 3363 }, { "epoch": 0.7758414414414414, "grad_norm": 27.022233963012695, "learning_rate": 6.460574573495835e-06, "loss": 12.167574882507324, "loss_alignment": 0.896484375, "loss_alignment_w": 0.4482421875, "loss_sft": 0.31186502799391747, "loss_total_v6": 0.7604734376072884, "step": 3364 }, { "epoch": 0.7760720720720721, "grad_norm": 27.15389633178711, "learning_rate": 6.45864845089986e-06, "loss": 12.175212860107422, "loss_alignment": 0.94384765625, "loss_alignment_w": 0.471923828125, "loss_sft": 0.2886607814580202, "loss_total_v6": 0.7609507963061333, "step": 3365 }, { "epoch": 0.7763027027027027, "grad_norm": 28.078540802001953, "learning_rate": 6.456722091678591e-06, "loss": 12.674043655395508, "loss_alignment": 0.93359375, "loss_alignment_w": 0.466796875, "loss_sft": 0.32586484029889107, "loss_total_v6": 0.792127713561058, "step": 3366 }, { "epoch": 0.7765333333333333, "grad_norm": 26.44945526123047, "learning_rate": 6.4547954961445235e-06, "loss": 11.957322120666504, "loss_alignment": 0.91552734375, "loss_alignment_w": 0.457763671875, "loss_sft": 0.2899657152593136, "loss_total_v6": 0.7473326548933983, "step": 3367 }, { "epoch": 0.776763963963964, "grad_norm": 24.647401809692383, "learning_rate": 6.452868664610197e-06, "loss": 12.110271453857422, "loss_alignment": 0.94287109375, "loss_alignment_w": 0.471435546875, "loss_sft": 0.2850901987403631, "loss_total_v6": 0.7568919807672501, "step": 3368 }, { "epoch": 0.7769945945945946, "grad_norm": 24.66259765625, "learning_rate": 6.450941597388184e-06, "loss": 11.537353515625, "loss_alignment": 0.89892578125, "loss_alignment_w": 0.449462890625, "loss_sft": 0.2713317349553108, "loss_total_v6": 0.7210845574736595, "step": 3369 }, { "epoch": 0.7772252252252252, "grad_norm": 26.016944885253906, "learning_rate": 6.449014294791101e-06, "loss": 12.486196517944336, "loss_alignment": 0.927734375, "loss_alignment_w": 0.4638671875, "loss_sft": 0.3166421726346016, "loss_total_v6": 0.7803872898221016, "step": 3370 }, { "epoch": 0.7774558558558559, "grad_norm": 25.546144485473633, "learning_rate": 6.447086757131596e-06, "loss": 12.246442794799805, "loss_alignment": 0.9443359375, "loss_alignment_w": 0.47216796875, "loss_sft": 0.29222765378654003, "loss_total_v6": 0.7654026672244072, "step": 3371 }, { "epoch": 0.7776864864864865, "grad_norm": 29.790807723999023, "learning_rate": 6.445158984722358e-06, "loss": 13.005769729614258, "loss_alignment": 0.93359375, "loss_alignment_w": 0.466796875, "loss_sft": 0.34491930343210697, "loss_total_v6": 0.8128606006503105, "step": 3372 }, { "epoch": 0.7779171171171171, "grad_norm": 30.515684127807617, "learning_rate": 6.4432309778761205e-06, "loss": 12.242172241210938, "loss_alignment": 0.94140625, "loss_alignment_w": 0.470703125, "loss_sft": 0.2940053418278694, "loss_total_v6": 0.7651357129216194, "step": 3373 }, { "epoch": 0.7781477477477478, "grad_norm": 25.447172164916992, "learning_rate": 6.441302736905641e-06, "loss": 12.028966903686523, "loss_alignment": 0.91552734375, "loss_alignment_w": 0.457763671875, "loss_sft": 0.2941840887069702, "loss_total_v6": 0.7518104314804077, "step": 3374 }, { "epoch": 0.7783783783783784, "grad_norm": 26.565330505371094, "learning_rate": 6.439374262123731e-06, "loss": 12.288990020751953, "loss_alignment": 0.93603515625, "loss_alignment_w": 0.468017578125, "loss_sft": 0.2987778317183256, "loss_total_v6": 0.7680618837475777, "step": 3375 }, { "epoch": 0.778609009009009, "grad_norm": 24.544544219970703, "learning_rate": 6.437445553843229e-06, "loss": 12.135671615600586, "loss_alignment": 0.93310546875, "loss_alignment_w": 0.466552734375, "loss_sft": 0.29250663705170155, "loss_total_v6": 0.7584795132279396, "step": 3376 }, { "epoch": 0.7788396396396396, "grad_norm": 23.602941513061523, "learning_rate": 6.435516612377011e-06, "loss": 12.152116775512695, "loss_alignment": 0.9423828125, "loss_alignment_w": 0.47119140625, "loss_sft": 0.2883006762713194, "loss_total_v6": 0.7595073133707047, "step": 3377 }, { "epoch": 0.7790702702702703, "grad_norm": 24.85007095336914, "learning_rate": 6.4335874380380005e-06, "loss": 12.034168243408203, "loss_alignment": 0.9248046875, "loss_alignment_w": 0.46240234375, "loss_sft": 0.29038936272263527, "loss_total_v6": 0.7521355375647545, "step": 3378 }, { "epoch": 0.7793009009009009, "grad_norm": 26.150724411010742, "learning_rate": 6.431658031139148e-06, "loss": 12.061202049255371, "loss_alignment": 0.92529296875, "loss_alignment_w": 0.462646484375, "loss_sft": 0.2918042354285717, "loss_total_v6": 0.7538251355290413, "step": 3379 }, { "epoch": 0.7795315315315315, "grad_norm": 25.290212631225586, "learning_rate": 6.429728391993446e-06, "loss": 12.333911895751953, "loss_alignment": 0.94189453125, "loss_alignment_w": 0.470947265625, "loss_sft": 0.29867101460695267, "loss_total_v6": 0.7708694860339165, "step": 3380 }, { "epoch": 0.7797621621621622, "grad_norm": 35.02584457397461, "learning_rate": 6.427798520913929e-06, "loss": 11.55433464050293, "loss_alignment": 0.916015625, "loss_alignment_w": 0.4580078125, "loss_sft": 0.2633751183748245, "loss_total_v6": 0.7221458777785301, "step": 3381 }, { "epoch": 0.7799927927927928, "grad_norm": 27.12017822265625, "learning_rate": 6.425868418213663e-06, "loss": 12.11390209197998, "loss_alignment": 0.9150390625, "loss_alignment_w": 0.45751953125, "loss_sft": 0.29763095267117023, "loss_total_v6": 0.7571188881993294, "step": 3382 }, { "epoch": 0.7802234234234234, "grad_norm": 26.710081100463867, "learning_rate": 6.423938084205753e-06, "loss": 12.065710067749023, "loss_alignment": 0.9404296875, "loss_alignment_w": 0.47021484375, "loss_sft": 0.28370892629027367, "loss_total_v6": 0.7541068643331528, "step": 3383 }, { "epoch": 0.7804540540540541, "grad_norm": 27.81574821472168, "learning_rate": 6.4220075192033425e-06, "loss": 12.069536209106445, "loss_alignment": 0.91552734375, "loss_alignment_w": 0.457763671875, "loss_sft": 0.29748261347413063, "loss_total_v6": 0.7543460056185722, "step": 3384 }, { "epoch": 0.7806846846846847, "grad_norm": 24.329225540161133, "learning_rate": 6.420076723519615e-06, "loss": 11.983146667480469, "loss_alignment": 0.94677734375, "loss_alignment_w": 0.473388671875, "loss_sft": 0.27577166073024273, "loss_total_v6": 0.7489467188715935, "step": 3385 }, { "epoch": 0.7809153153153153, "grad_norm": 25.675148010253906, "learning_rate": 6.418145697467784e-06, "loss": 11.82009506225586, "loss_alignment": 0.92138671875, "loss_alignment_w": 0.460693359375, "loss_sft": 0.27762009017169476, "loss_total_v6": 0.7387559562921524, "step": 3386 }, { "epoch": 0.781145945945946, "grad_norm": 24.239089965820312, "learning_rate": 6.416214441361106e-06, "loss": 11.376155853271484, "loss_alignment": 0.92529296875, "loss_alignment_w": 0.462646484375, "loss_sft": 0.24795121513307095, "loss_total_v6": 0.7110096961259842, "step": 3387 }, { "epoch": 0.7813765765765766, "grad_norm": 25.419099807739258, "learning_rate": 6.414282955512876e-06, "loss": 12.30314826965332, "loss_alignment": 0.93505859375, "loss_alignment_w": 0.467529296875, "loss_sft": 0.3026076927781105, "loss_total_v6": 0.7689467966556549, "step": 3388 }, { "epoch": 0.7816072072072072, "grad_norm": 26.477279663085938, "learning_rate": 6.412351240236423e-06, "loss": 11.818531036376953, "loss_alignment": 0.9306640625, "loss_alignment_w": 0.46533203125, "loss_sft": 0.2729141619056463, "loss_total_v6": 0.7386581599712372, "step": 3389 }, { "epoch": 0.7818378378378379, "grad_norm": 25.354602813720703, "learning_rate": 6.410419295845114e-06, "loss": 12.34187126159668, "loss_alignment": 0.95166015625, "loss_alignment_w": 0.475830078125, "loss_sft": 0.2957962844520807, "loss_total_v6": 0.7713669613003731, "step": 3390 }, { "epoch": 0.7820684684684684, "grad_norm": 27.322919845581055, "learning_rate": 6.408487122652353e-06, "loss": 12.534379959106445, "loss_alignment": 0.9208984375, "loss_alignment_w": 0.46044921875, "loss_sft": 0.32247647270560265, "loss_total_v6": 0.7833987176418304, "step": 3391 }, { "epoch": 0.7822990990990991, "grad_norm": 27.467519760131836, "learning_rate": 6.406554720971583e-06, "loss": 12.447897911071777, "loss_alignment": 0.9267578125, "loss_alignment_w": 0.46337890625, "loss_sft": 0.3156828507781029, "loss_total_v6": 0.7779936417937279, "step": 3392 }, { "epoch": 0.7825297297297298, "grad_norm": 25.148197174072266, "learning_rate": 6.404622091116282e-06, "loss": 11.465353012084961, "loss_alignment": 0.92333984375, "loss_alignment_w": 0.461669921875, "loss_sft": 0.2554639633744955, "loss_total_v6": 0.7165845632553101, "step": 3393 }, { "epoch": 0.7827603603603603, "grad_norm": 27.64503288269043, "learning_rate": 6.402689233399965e-06, "loss": 11.680834770202637, "loss_alignment": 0.9140625, "loss_alignment_w": 0.45703125, "loss_sft": 0.2727767862379551, "loss_total_v6": 0.7300521805882454, "step": 3394 }, { "epoch": 0.782990990990991, "grad_norm": 24.02297019958496, "learning_rate": 6.400756148136185e-06, "loss": 11.646190643310547, "loss_alignment": 0.93017578125, "loss_alignment_w": 0.465087890625, "loss_sft": 0.2628905288875103, "loss_total_v6": 0.7278868705034256, "step": 3395 }, { "epoch": 0.7832216216216216, "grad_norm": 26.732385635375977, "learning_rate": 6.398822835638531e-06, "loss": 12.11442756652832, "loss_alignment": 0.91455078125, "loss_alignment_w": 0.457275390625, "loss_sft": 0.2993727829307318, "loss_total_v6": 0.757151685655117, "step": 3396 }, { "epoch": 0.7834522522522522, "grad_norm": 27.107345581054688, "learning_rate": 6.396889296220631e-06, "loss": 12.042805671691895, "loss_alignment": 0.90185546875, "loss_alignment_w": 0.450927734375, "loss_sft": 0.3023885302245617, "loss_total_v6": 0.7526753693819046, "step": 3397 }, { "epoch": 0.7836828828828829, "grad_norm": 25.435768127441406, "learning_rate": 6.3949555301961474e-06, "loss": 12.499950408935547, "loss_alignment": 0.94287109375, "loss_alignment_w": 0.471435546875, "loss_sft": 0.3102080784738064, "loss_total_v6": 0.7812468856573105, "step": 3398 }, { "epoch": 0.7839135135135135, "grad_norm": 24.15265655517578, "learning_rate": 6.393021537878778e-06, "loss": 12.375082015991211, "loss_alignment": 0.916015625, "loss_alignment_w": 0.4580078125, "loss_sft": 0.315190676599741, "loss_total_v6": 0.7734426409006119, "step": 3399 }, { "epoch": 0.7841441441441441, "grad_norm": 25.82203483581543, "learning_rate": 6.391087319582264e-06, "loss": 12.300570487976074, "loss_alignment": 0.93603515625, "loss_alignment_w": 0.468017578125, "loss_sft": 0.30111901089549065, "loss_total_v6": 0.768785648047924, "step": 3400 } ], "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": 7.65690951402552e+19, "train_batch_size": 1, "trial_name": null, "trial_params": null }