diff --git "a/trainer_state.json" "b/trainer_state.json" new file mode 100644--- /dev/null +++ "b/trainer_state.json" @@ -0,0 +1,5451 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.8527131782945734, + "eval_steps": 500, + "global_step": 774, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.003875968992248062, + "grad_norm": 0.2644451856613159, + "learning_rate": 6.6e-06, + "loss": 2.7134, + "step": 1 + }, + { + "epoch": 0.007751937984496124, + "grad_norm": 0.3978240191936493, + "learning_rate": 1.32e-05, + "loss": 3.1698, + "step": 2 + }, + { + "epoch": 0.011627906976744186, + "grad_norm": 0.29106518626213074, + "learning_rate": 1.98e-05, + "loss": 2.8065, + "step": 3 + }, + { + "epoch": 0.015503875968992248, + "grad_norm": 0.26795685291290283, + "learning_rate": 2.64e-05, + "loss": 2.7225, + "step": 4 + }, + { + "epoch": 0.01937984496124031, + "grad_norm": 0.18800821900367737, + "learning_rate": 3.3e-05, + "loss": 2.3743, + "step": 5 + }, + { + "epoch": 0.023255813953488372, + "grad_norm": 0.2275751829147339, + "learning_rate": 3.96e-05, + "loss": 2.474, + "step": 6 + }, + { + "epoch": 0.027131782945736434, + "grad_norm": 0.37303999066352844, + "learning_rate": 4.6200000000000005e-05, + "loss": 2.9455, + "step": 7 + }, + { + "epoch": 0.031007751937984496, + "grad_norm": 0.5618475079536438, + "learning_rate": 5.28e-05, + "loss": 3.0509, + "step": 8 + }, + { + "epoch": 0.03488372093023256, + "grad_norm": 0.2064945548772812, + "learning_rate": 5.94e-05, + "loss": 2.3145, + "step": 9 + }, + { + "epoch": 0.03875968992248062, + "grad_norm": 0.1987352818250656, + "learning_rate": 6.6e-05, + "loss": 2.6002, + "step": 10 + }, + { + "epoch": 0.04263565891472868, + "grad_norm": 0.20411251485347748, + "learning_rate": 7.26e-05, + "loss": 2.4743, + "step": 11 + }, + { + "epoch": 0.046511627906976744, + "grad_norm": 0.1682678759098053, + "learning_rate": 7.92e-05, + "loss": 2.2844, + "step": 12 + }, + { + "epoch": 0.050387596899224806, + "grad_norm": 0.2902089059352875, + "learning_rate": 8.58e-05, + "loss": 2.7268, + "step": 13 + }, + { + "epoch": 0.05426356589147287, + "grad_norm": 0.34279119968414307, + "learning_rate": 9.240000000000001e-05, + "loss": 2.8172, + "step": 14 + }, + { + "epoch": 0.05813953488372093, + "grad_norm": 0.25758570432662964, + "learning_rate": 9.9e-05, + "loss": 2.5574, + "step": 15 + }, + { + "epoch": 0.06201550387596899, + "grad_norm": 0.22630520164966583, + "learning_rate": 0.0001056, + "loss": 2.4886, + "step": 16 + }, + { + "epoch": 0.06589147286821706, + "grad_norm": 0.24086053669452667, + "learning_rate": 0.0001122, + "loss": 2.1352, + "step": 17 + }, + { + "epoch": 0.06976744186046512, + "grad_norm": 0.31760409474372864, + "learning_rate": 0.0001188, + "loss": 2.3757, + "step": 18 + }, + { + "epoch": 0.07364341085271318, + "grad_norm": 0.27383461594581604, + "learning_rate": 0.0001254, + "loss": 2.2567, + "step": 19 + }, + { + "epoch": 0.07751937984496124, + "grad_norm": 0.25780704617500305, + "learning_rate": 0.000132, + "loss": 2.497, + "step": 20 + }, + { + "epoch": 0.08139534883720931, + "grad_norm": 0.20427410304546356, + "learning_rate": 0.00013859999999999998, + "loss": 2.2143, + "step": 21 + }, + { + "epoch": 0.08527131782945736, + "grad_norm": 0.2947242558002472, + "learning_rate": 0.0001452, + "loss": 2.6828, + "step": 22 + }, + { + "epoch": 0.08914728682170543, + "grad_norm": 0.23378527164459229, + "learning_rate": 0.0001518, + "loss": 2.3352, + "step": 23 + }, + { + "epoch": 0.09302325581395349, + "grad_norm": 0.23147471249103546, + "learning_rate": 0.0001584, + "loss": 2.2845, + "step": 24 + }, + { + "epoch": 0.09689922480620156, + "grad_norm": 0.21582217514514923, + "learning_rate": 0.000165, + "loss": 2.5595, + "step": 25 + }, + { + "epoch": 0.10077519379844961, + "grad_norm": 0.1932438313961029, + "learning_rate": 0.0001716, + "loss": 2.3431, + "step": 26 + }, + { + "epoch": 0.10465116279069768, + "grad_norm": 0.18576449155807495, + "learning_rate": 0.00017820000000000002, + "loss": 2.3948, + "step": 27 + }, + { + "epoch": 0.10852713178294573, + "grad_norm": 0.1673685610294342, + "learning_rate": 0.00018480000000000002, + "loss": 2.2775, + "step": 28 + }, + { + "epoch": 0.1124031007751938, + "grad_norm": 0.17178471386432648, + "learning_rate": 0.0001914, + "loss": 1.9809, + "step": 29 + }, + { + "epoch": 0.11627906976744186, + "grad_norm": 0.17103907465934753, + "learning_rate": 0.000198, + "loss": 2.2033, + "step": 30 + }, + { + "epoch": 0.12015503875968993, + "grad_norm": 0.17570005357265472, + "learning_rate": 0.00020459999999999999, + "loss": 2.1358, + "step": 31 + }, + { + "epoch": 0.12403100775193798, + "grad_norm": 0.16803690791130066, + "learning_rate": 0.0002112, + "loss": 2.164, + "step": 32 + }, + { + "epoch": 0.12790697674418605, + "grad_norm": 0.17673198878765106, + "learning_rate": 0.0002178, + "loss": 2.2587, + "step": 33 + }, + { + "epoch": 0.13178294573643412, + "grad_norm": 0.16170482337474823, + "learning_rate": 0.0002244, + "loss": 2.1818, + "step": 34 + }, + { + "epoch": 0.13565891472868216, + "grad_norm": 0.1507701873779297, + "learning_rate": 0.00023099999999999998, + "loss": 2.3218, + "step": 35 + }, + { + "epoch": 0.13953488372093023, + "grad_norm": 0.16273871064186096, + "learning_rate": 0.0002376, + "loss": 2.1462, + "step": 36 + }, + { + "epoch": 0.1434108527131783, + "grad_norm": 0.15038099884986877, + "learning_rate": 0.00024419999999999997, + "loss": 2.1007, + "step": 37 + }, + { + "epoch": 0.14728682170542637, + "grad_norm": 0.19681762158870697, + "learning_rate": 0.0002508, + "loss": 2.3083, + "step": 38 + }, + { + "epoch": 0.1511627906976744, + "grad_norm": 0.15614789724349976, + "learning_rate": 0.0002574, + "loss": 2.0723, + "step": 39 + }, + { + "epoch": 0.15503875968992248, + "grad_norm": 0.14269952476024628, + "learning_rate": 0.000264, + "loss": 2.1325, + "step": 40 + }, + { + "epoch": 0.15891472868217055, + "grad_norm": 0.16603954136371613, + "learning_rate": 0.00027059999999999996, + "loss": 2.2062, + "step": 41 + }, + { + "epoch": 0.16279069767441862, + "grad_norm": 0.16374431550502777, + "learning_rate": 0.00027719999999999996, + "loss": 2.0389, + "step": 42 + }, + { + "epoch": 0.16666666666666666, + "grad_norm": 0.13400350511074066, + "learning_rate": 0.0002838, + "loss": 1.9439, + "step": 43 + }, + { + "epoch": 0.17054263565891473, + "grad_norm": 0.13477088510990143, + "learning_rate": 0.0002904, + "loss": 2.3466, + "step": 44 + }, + { + "epoch": 0.1744186046511628, + "grad_norm": 0.14038704335689545, + "learning_rate": 0.000297, + "loss": 2.3133, + "step": 45 + }, + { + "epoch": 0.17829457364341086, + "grad_norm": 0.14085592329502106, + "learning_rate": 0.0003036, + "loss": 2.3503, + "step": 46 + }, + { + "epoch": 0.1821705426356589, + "grad_norm": 0.14309249818325043, + "learning_rate": 0.0003102, + "loss": 2.4691, + "step": 47 + }, + { + "epoch": 0.18604651162790697, + "grad_norm": 0.1526148021221161, + "learning_rate": 0.0003168, + "loss": 2.1829, + "step": 48 + }, + { + "epoch": 0.18992248062015504, + "grad_norm": 0.15030066668987274, + "learning_rate": 0.0003234, + "loss": 2.1198, + "step": 49 + }, + { + "epoch": 0.1937984496124031, + "grad_norm": 0.13567374646663666, + "learning_rate": 0.00033, + "loss": 2.3342, + "step": 50 + }, + { + "epoch": 0.19767441860465115, + "grad_norm": 0.14513419568538666, + "learning_rate": 0.00032999844662654247, + "loss": 2.1416, + "step": 51 + }, + { + "epoch": 0.20155038759689922, + "grad_norm": 0.1639217734336853, + "learning_rate": 0.00032999378653541785, + "loss": 2.06, + "step": 52 + }, + { + "epoch": 0.2054263565891473, + "grad_norm": 0.15029025077819824, + "learning_rate": 0.0003299860198143701, + "loss": 2.3455, + "step": 53 + }, + { + "epoch": 0.20930232558139536, + "grad_norm": 0.15286165475845337, + "learning_rate": 0.00032997514660963685, + "loss": 2.2418, + "step": 54 + }, + { + "epoch": 0.2131782945736434, + "grad_norm": 0.17834632098674774, + "learning_rate": 0.00032996116712594723, + "loss": 2.3385, + "step": 55 + }, + { + "epoch": 0.21705426356589147, + "grad_norm": 0.1593712419271469, + "learning_rate": 0.00032994408162651776, + "loss": 2.2248, + "step": 56 + }, + { + "epoch": 0.22093023255813954, + "grad_norm": 0.13914993405342102, + "learning_rate": 0.00032992389043304726, + "loss": 2.3005, + "step": 57 + }, + { + "epoch": 0.2248062015503876, + "grad_norm": 0.13964952528476715, + "learning_rate": 0.0003299005939257111, + "loss": 2.284, + "step": 58 + }, + { + "epoch": 0.22868217054263565, + "grad_norm": 0.151286780834198, + "learning_rate": 0.0003298741925431539, + "loss": 2.1277, + "step": 59 + }, + { + "epoch": 0.23255813953488372, + "grad_norm": 0.12762194871902466, + "learning_rate": 0.0003298446867824811, + "loss": 2.046, + "step": 60 + }, + { + "epoch": 0.2364341085271318, + "grad_norm": 0.13711339235305786, + "learning_rate": 0.00032981207719924994, + "loss": 2.1179, + "step": 61 + }, + { + "epoch": 0.24031007751937986, + "grad_norm": 0.15553320944309235, + "learning_rate": 0.0003297763644074587, + "loss": 2.3921, + "step": 62 + }, + { + "epoch": 0.2441860465116279, + "grad_norm": 0.14784659445285797, + "learning_rate": 0.0003297375490795353, + "loss": 2.072, + "step": 63 + }, + { + "epoch": 0.24806201550387597, + "grad_norm": 0.14965656399726868, + "learning_rate": 0.0003296956319463245, + "loss": 2.314, + "step": 64 + }, + { + "epoch": 0.25193798449612403, + "grad_norm": 0.14986053109169006, + "learning_rate": 0.0003296506137970745, + "loss": 2.16, + "step": 65 + }, + { + "epoch": 0.2558139534883721, + "grad_norm": 0.14208155870437622, + "learning_rate": 0.0003296024954794215, + "loss": 1.9537, + "step": 66 + }, + { + "epoch": 0.2596899224806202, + "grad_norm": 0.1486920267343521, + "learning_rate": 0.0003295512778993743, + "loss": 2.1949, + "step": 67 + }, + { + "epoch": 0.26356589147286824, + "grad_norm": 0.14816056191921234, + "learning_rate": 0.0003294969620212968, + "loss": 2.274, + "step": 68 + }, + { + "epoch": 0.26744186046511625, + "grad_norm": 0.14654242992401123, + "learning_rate": 0.00032943954886789016, + "loss": 2.3158, + "step": 69 + }, + { + "epoch": 0.2713178294573643, + "grad_norm": 0.1430758833885193, + "learning_rate": 0.0003293790395201734, + "loss": 2.2095, + "step": 70 + }, + { + "epoch": 0.2751937984496124, + "grad_norm": 0.13157042860984802, + "learning_rate": 0.00032931543511746306, + "loss": 2.3029, + "step": 71 + }, + { + "epoch": 0.27906976744186046, + "grad_norm": 0.15158364176750183, + "learning_rate": 0.0003292487368573518, + "loss": 2.4751, + "step": 72 + }, + { + "epoch": 0.28294573643410853, + "grad_norm": 0.14293590188026428, + "learning_rate": 0.0003291789459956857, + "loss": 2.2384, + "step": 73 + }, + { + "epoch": 0.2868217054263566, + "grad_norm": 0.13672257959842682, + "learning_rate": 0.00032910606384654086, + "loss": 2.389, + "step": 74 + }, + { + "epoch": 0.29069767441860467, + "grad_norm": 0.17896883189678192, + "learning_rate": 0.00032903009178219834, + "loss": 2.26, + "step": 75 + }, + { + "epoch": 0.29457364341085274, + "grad_norm": 0.1485362946987152, + "learning_rate": 0.0003289510312331187, + "loss": 1.9685, + "step": 76 + }, + { + "epoch": 0.29844961240310075, + "grad_norm": 0.1551189124584198, + "learning_rate": 0.0003288688836879147, + "loss": 2.0234, + "step": 77 + }, + { + "epoch": 0.3023255813953488, + "grad_norm": 0.14495842158794403, + "learning_rate": 0.0003287836506933235, + "loss": 2.0907, + "step": 78 + }, + { + "epoch": 0.3062015503875969, + "grad_norm": 0.16551914811134338, + "learning_rate": 0.0003286953338541776, + "loss": 2.1908, + "step": 79 + }, + { + "epoch": 0.31007751937984496, + "grad_norm": 0.14821241796016693, + "learning_rate": 0.0003286039348333743, + "loss": 1.9845, + "step": 80 + }, + { + "epoch": 0.313953488372093, + "grad_norm": 0.17918680608272552, + "learning_rate": 0.0003285094553518446, + "loss": 2.4009, + "step": 81 + }, + { + "epoch": 0.3178294573643411, + "grad_norm": 0.15694545209407806, + "learning_rate": 0.000328411897188521, + "loss": 2.1897, + "step": 82 + }, + { + "epoch": 0.32170542635658916, + "grad_norm": 0.17306634783744812, + "learning_rate": 0.0003283112621803035, + "loss": 2.2274, + "step": 83 + }, + { + "epoch": 0.32558139534883723, + "grad_norm": 0.14742298424243927, + "learning_rate": 0.0003282075522220255, + "loss": 2.1474, + "step": 84 + }, + { + "epoch": 0.32945736434108525, + "grad_norm": 0.17422489821910858, + "learning_rate": 0.0003281007692664178, + "loss": 2.3424, + "step": 85 + }, + { + "epoch": 0.3333333333333333, + "grad_norm": 0.14565351605415344, + "learning_rate": 0.0003279909153240722, + "loss": 2.1367, + "step": 86 + }, + { + "epoch": 0.3372093023255814, + "grad_norm": 0.16405850648880005, + "learning_rate": 0.00032787799246340304, + "loss": 2.0857, + "step": 87 + }, + { + "epoch": 0.34108527131782945, + "grad_norm": 0.1722440868616104, + "learning_rate": 0.0003277620028106089, + "loss": 2.0891, + "step": 88 + }, + { + "epoch": 0.3449612403100775, + "grad_norm": 0.1501605361700058, + "learning_rate": 0.00032764294854963213, + "loss": 2.1899, + "step": 89 + }, + { + "epoch": 0.3488372093023256, + "grad_norm": 0.14743047952651978, + "learning_rate": 0.0003275208319221179, + "loss": 2.165, + "step": 90 + }, + { + "epoch": 0.35271317829457366, + "grad_norm": 0.14289091527462006, + "learning_rate": 0.00032739565522737216, + "loss": 2.0733, + "step": 91 + }, + { + "epoch": 0.35658914728682173, + "grad_norm": 0.13758860528469086, + "learning_rate": 0.00032726742082231787, + "loss": 2.185, + "step": 92 + }, + { + "epoch": 0.36046511627906974, + "grad_norm": 0.13637179136276245, + "learning_rate": 0.00032713613112145105, + "loss": 2.2052, + "step": 93 + }, + { + "epoch": 0.3643410852713178, + "grad_norm": 0.14775022864341736, + "learning_rate": 0.0003270017885967952, + "loss": 2.1404, + "step": 94 + }, + { + "epoch": 0.3682170542635659, + "grad_norm": 0.1494913548231125, + "learning_rate": 0.0003268643957778546, + "loss": 1.9495, + "step": 95 + }, + { + "epoch": 0.37209302325581395, + "grad_norm": 0.1620800644159317, + "learning_rate": 0.00032672395525156715, + "loss": 2.0199, + "step": 96 + }, + { + "epoch": 0.375968992248062, + "grad_norm": 0.15806061029434204, + "learning_rate": 0.0003265804696622549, + "loss": 1.9932, + "step": 97 + }, + { + "epoch": 0.3798449612403101, + "grad_norm": 0.16337577998638153, + "learning_rate": 0.00032643394171157504, + "loss": 2.1522, + "step": 98 + }, + { + "epoch": 0.38372093023255816, + "grad_norm": 0.1602049618959427, + "learning_rate": 0.0003262843741584684, + "loss": 2.215, + "step": 99 + }, + { + "epoch": 0.3875968992248062, + "grad_norm": 0.17709355056285858, + "learning_rate": 0.00032613176981910805, + "loss": 1.8921, + "step": 100 + }, + { + "epoch": 0.39147286821705424, + "grad_norm": 0.14379487931728363, + "learning_rate": 0.0003259761315668458, + "loss": 2.0393, + "step": 101 + }, + { + "epoch": 0.3953488372093023, + "grad_norm": 0.15646255016326904, + "learning_rate": 0.0003258174623321583, + "loss": 2.2783, + "step": 102 + }, + { + "epoch": 0.3992248062015504, + "grad_norm": 0.17745360732078552, + "learning_rate": 0.00032565576510259207, + "loss": 2.0816, + "step": 103 + }, + { + "epoch": 0.40310077519379844, + "grad_norm": 0.14521199464797974, + "learning_rate": 0.0003254910429227069, + "loss": 2.061, + "step": 104 + }, + { + "epoch": 0.4069767441860465, + "grad_norm": 0.14865247905254364, + "learning_rate": 0.0003253232988940186, + "loss": 2.1272, + "step": 105 + }, + { + "epoch": 0.4108527131782946, + "grad_norm": 0.15088239312171936, + "learning_rate": 0.0003251525361749409, + "loss": 1.9842, + "step": 106 + }, + { + "epoch": 0.41472868217054265, + "grad_norm": 0.15884487330913544, + "learning_rate": 0.00032497875798072557, + "loss": 2.2848, + "step": 107 + }, + { + "epoch": 0.4186046511627907, + "grad_norm": 0.13841639459133148, + "learning_rate": 0.000324801967583402, + "loss": 2.2508, + "step": 108 + }, + { + "epoch": 0.42248062015503873, + "grad_norm": 0.15195606648921967, + "learning_rate": 0.00032462216831171607, + "loss": 2.2611, + "step": 109 + }, + { + "epoch": 0.4263565891472868, + "grad_norm": 0.17446410655975342, + "learning_rate": 0.0003244393635510664, + "loss": 2.3199, + "step": 110 + }, + { + "epoch": 0.43023255813953487, + "grad_norm": 0.1482134312391281, + "learning_rate": 0.0003242535567434419, + "loss": 2.106, + "step": 111 + }, + { + "epoch": 0.43410852713178294, + "grad_norm": 0.17814427614212036, + "learning_rate": 0.000324064751387356, + "loss": 2.3546, + "step": 112 + }, + { + "epoch": 0.437984496124031, + "grad_norm": 0.15309078991413116, + "learning_rate": 0.00032387295103778114, + "loss": 1.9944, + "step": 113 + }, + { + "epoch": 0.4418604651162791, + "grad_norm": 0.14015917479991913, + "learning_rate": 0.00032367815930608177, + "loss": 2.2504, + "step": 114 + }, + { + "epoch": 0.44573643410852715, + "grad_norm": 0.14815208315849304, + "learning_rate": 0.0003234803798599466, + "loss": 2.1902, + "step": 115 + }, + { + "epoch": 0.4496124031007752, + "grad_norm": 0.14948517084121704, + "learning_rate": 0.000323279616423319, + "loss": 2.2916, + "step": 116 + }, + { + "epoch": 0.45348837209302323, + "grad_norm": 0.18051539361476898, + "learning_rate": 0.00032307587277632753, + "loss": 2.2055, + "step": 117 + }, + { + "epoch": 0.4573643410852713, + "grad_norm": 0.17824648320674896, + "learning_rate": 0.00032286915275521423, + "loss": 1.9562, + "step": 118 + }, + { + "epoch": 0.46124031007751937, + "grad_norm": 0.15700489282608032, + "learning_rate": 0.0003226594602522626, + "loss": 2.0874, + "step": 119 + }, + { + "epoch": 0.46511627906976744, + "grad_norm": 0.15389874577522278, + "learning_rate": 0.00032244679921572447, + "loss": 2.314, + "step": 120 + }, + { + "epoch": 0.4689922480620155, + "grad_norm": 0.15873467922210693, + "learning_rate": 0.0003222311736497454, + "loss": 2.3363, + "step": 121 + }, + { + "epoch": 0.4728682170542636, + "grad_norm": 0.15486344695091248, + "learning_rate": 0.0003220125876142893, + "loss": 2.2502, + "step": 122 + }, + { + "epoch": 0.47674418604651164, + "grad_norm": 0.15193265676498413, + "learning_rate": 0.00032179104522506234, + "loss": 2.1667, + "step": 123 + }, + { + "epoch": 0.4806201550387597, + "grad_norm": 0.17307482659816742, + "learning_rate": 0.00032156655065343487, + "loss": 2.2361, + "step": 124 + }, + { + "epoch": 0.4844961240310077, + "grad_norm": 0.17050062119960785, + "learning_rate": 0.00032133910812636346, + "loss": 1.9976, + "step": 125 + }, + { + "epoch": 0.4883720930232558, + "grad_norm": 0.15616333484649658, + "learning_rate": 0.00032110872192631096, + "loss": 2.2286, + "step": 126 + }, + { + "epoch": 0.49224806201550386, + "grad_norm": 0.15690878033638, + "learning_rate": 0.00032087539639116596, + "loss": 2.2743, + "step": 127 + }, + { + "epoch": 0.49612403100775193, + "grad_norm": 0.15837594866752625, + "learning_rate": 0.000320639135914161, + "loss": 2.2109, + "step": 128 + }, + { + "epoch": 0.5, + "grad_norm": 0.16106249392032623, + "learning_rate": 0.0003203999449437902, + "loss": 2.0109, + "step": 129 + }, + { + "epoch": 0.5038759689922481, + "grad_norm": 0.15444137156009674, + "learning_rate": 0.00032015782798372504, + "loss": 1.8895, + "step": 130 + }, + { + "epoch": 0.5077519379844961, + "grad_norm": 0.16820791363716125, + "learning_rate": 0.0003199127895927299, + "loss": 2.0923, + "step": 131 + }, + { + "epoch": 0.5116279069767442, + "grad_norm": 0.17274028062820435, + "learning_rate": 0.0003196648343845761, + "loss": 2.1964, + "step": 132 + }, + { + "epoch": 0.5155038759689923, + "grad_norm": 0.17962628602981567, + "learning_rate": 0.00031941396702795505, + "loss": 2.2031, + "step": 133 + }, + { + "epoch": 0.5193798449612403, + "grad_norm": 0.15597763657569885, + "learning_rate": 0.00031916019224639017, + "loss": 1.9934, + "step": 134 + }, + { + "epoch": 0.5232558139534884, + "grad_norm": 0.15726816654205322, + "learning_rate": 0.00031890351481814844, + "loss": 2.004, + "step": 135 + }, + { + "epoch": 0.5271317829457365, + "grad_norm": 0.16479064524173737, + "learning_rate": 0.0003186439395761498, + "loss": 2.024, + "step": 136 + }, + { + "epoch": 0.5310077519379846, + "grad_norm": 0.1560487300157547, + "learning_rate": 0.00031838147140787656, + "loss": 2.2763, + "step": 137 + }, + { + "epoch": 0.5348837209302325, + "grad_norm": 0.15172263979911804, + "learning_rate": 0.0003181161152552813, + "loss": 2.2485, + "step": 138 + }, + { + "epoch": 0.5387596899224806, + "grad_norm": 0.14550605416297913, + "learning_rate": 0.00031784787611469377, + "loss": 2.319, + "step": 139 + }, + { + "epoch": 0.5426356589147286, + "grad_norm": 0.15684114396572113, + "learning_rate": 0.0003175767590367269, + "loss": 2.1815, + "step": 140 + }, + { + "epoch": 0.5465116279069767, + "grad_norm": 0.15643136203289032, + "learning_rate": 0.00031730276912618146, + "loss": 2.2636, + "step": 141 + }, + { + "epoch": 0.5503875968992248, + "grad_norm": 0.15671825408935547, + "learning_rate": 0.00031702591154195023, + "loss": 2.2284, + "step": 142 + }, + { + "epoch": 0.5542635658914729, + "grad_norm": 0.16105671226978302, + "learning_rate": 0.0003167461914969208, + "loss": 2.2209, + "step": 143 + }, + { + "epoch": 0.5581395348837209, + "grad_norm": 0.15666881203651428, + "learning_rate": 0.0003164636142578775, + "loss": 2.1917, + "step": 144 + }, + { + "epoch": 0.562015503875969, + "grad_norm": 0.16629591584205627, + "learning_rate": 0.00031617818514540164, + "loss": 2.2236, + "step": 145 + }, + { + "epoch": 0.5658914728682171, + "grad_norm": 0.15989044308662415, + "learning_rate": 0.00031588990953377227, + "loss": 2.1643, + "step": 146 + }, + { + "epoch": 0.5697674418604651, + "grad_norm": 0.15829624235630035, + "learning_rate": 0.0003155987928508643, + "loss": 2.0441, + "step": 147 + }, + { + "epoch": 0.5736434108527132, + "grad_norm": 0.16886377334594727, + "learning_rate": 0.0003153048405780466, + "loss": 2.3204, + "step": 148 + }, + { + "epoch": 0.5775193798449613, + "grad_norm": 0.14943009614944458, + "learning_rate": 0.0003150080582500786, + "loss": 2.2651, + "step": 149 + }, + { + "epoch": 0.5813953488372093, + "grad_norm": 0.14913472533226013, + "learning_rate": 0.0003147084514550064, + "loss": 2.0444, + "step": 150 + }, + { + "epoch": 0.5852713178294574, + "grad_norm": 0.17818570137023926, + "learning_rate": 0.0003144060258340569, + "loss": 2.389, + "step": 151 + }, + { + "epoch": 0.5891472868217055, + "grad_norm": 0.15725958347320557, + "learning_rate": 0.0003141007870815326, + "loss": 2.0757, + "step": 152 + }, + { + "epoch": 0.5930232558139535, + "grad_norm": 0.17349454760551453, + "learning_rate": 0.0003137927409447034, + "loss": 2.0407, + "step": 153 + }, + { + "epoch": 0.5968992248062015, + "grad_norm": 0.16550661623477936, + "learning_rate": 0.0003134818932236989, + "loss": 2.1548, + "step": 154 + }, + { + "epoch": 0.6007751937984496, + "grad_norm": 0.17092445492744446, + "learning_rate": 0.00031316824977139905, + "loss": 2.1501, + "step": 155 + }, + { + "epoch": 0.6046511627906976, + "grad_norm": 0.16574212908744812, + "learning_rate": 0.00031285181649332413, + "loss": 2.3126, + "step": 156 + }, + { + "epoch": 0.6085271317829457, + "grad_norm": 0.18889443576335907, + "learning_rate": 0.00031253259934752324, + "loss": 2.0974, + "step": 157 + }, + { + "epoch": 0.6124031007751938, + "grad_norm": 0.14138291776180267, + "learning_rate": 0.0003122106043444623, + "loss": 2.1352, + "step": 158 + }, + { + "epoch": 0.6162790697674418, + "grad_norm": 0.15668374300003052, + "learning_rate": 0.000311885837546911, + "loss": 2.0748, + "step": 159 + }, + { + "epoch": 0.6201550387596899, + "grad_norm": 0.19118763506412506, + "learning_rate": 0.0003115583050698283, + "loss": 2.126, + "step": 160 + }, + { + "epoch": 0.624031007751938, + "grad_norm": 0.19029273092746735, + "learning_rate": 0.0003112280130802476, + "loss": 2.2312, + "step": 161 + }, + { + "epoch": 0.627906976744186, + "grad_norm": 0.16116315126419067, + "learning_rate": 0.0003108949677971607, + "loss": 2.1802, + "step": 162 + }, + { + "epoch": 0.6317829457364341, + "grad_norm": 0.15257392823696136, + "learning_rate": 0.00031055917549140013, + "loss": 1.8636, + "step": 163 + }, + { + "epoch": 0.6356589147286822, + "grad_norm": 0.14367957413196564, + "learning_rate": 0.00031022064248552197, + "loss": 2.2663, + "step": 164 + }, + { + "epoch": 0.6395348837209303, + "grad_norm": 0.155740886926651, + "learning_rate": 0.0003098793751536858, + "loss": 2.0618, + "step": 165 + }, + { + "epoch": 0.6434108527131783, + "grad_norm": 0.15573202073574066, + "learning_rate": 0.0003095353799215356, + "loss": 2.147, + "step": 166 + }, + { + "epoch": 0.6472868217054264, + "grad_norm": 0.17215795814990997, + "learning_rate": 0.0003091886632660781, + "loss": 2.381, + "step": 167 + }, + { + "epoch": 0.6511627906976745, + "grad_norm": 0.16937744617462158, + "learning_rate": 0.0003088392317155612, + "loss": 2.0703, + "step": 168 + }, + { + "epoch": 0.6550387596899225, + "grad_norm": 0.16653715074062347, + "learning_rate": 0.000308487091849351, + "loss": 2.3338, + "step": 169 + }, + { + "epoch": 0.6589147286821705, + "grad_norm": 0.1645222008228302, + "learning_rate": 0.0003081322502978077, + "loss": 2.2578, + "step": 170 + }, + { + "epoch": 0.6627906976744186, + "grad_norm": 0.17217321693897247, + "learning_rate": 0.0003077747137421609, + "loss": 2.3059, + "step": 171 + }, + { + "epoch": 0.6666666666666666, + "grad_norm": 0.15005135536193848, + "learning_rate": 0.000307414488914384, + "loss": 2.3234, + "step": 172 + }, + { + "epoch": 0.6705426356589147, + "grad_norm": 0.1632111668586731, + "learning_rate": 0.0003070515825970671, + "loss": 2.3237, + "step": 173 + }, + { + "epoch": 0.6744186046511628, + "grad_norm": 0.16381341218948364, + "learning_rate": 0.0003066860016232894, + "loss": 2.2046, + "step": 174 + }, + { + "epoch": 0.6782945736434108, + "grad_norm": 0.1917879581451416, + "learning_rate": 0.00030631775287649077, + "loss": 2.2253, + "step": 175 + }, + { + "epoch": 0.6821705426356589, + "grad_norm": 0.16225776076316833, + "learning_rate": 0.00030594684329034183, + "loss": 2.1078, + "step": 176 + }, + { + "epoch": 0.686046511627907, + "grad_norm": 0.16708023846149445, + "learning_rate": 0.00030557327984861353, + "loss": 2.3477, + "step": 177 + }, + { + "epoch": 0.689922480620155, + "grad_norm": 0.1940573751926422, + "learning_rate": 0.00030519706958504573, + "loss": 2.133, + "step": 178 + }, + { + "epoch": 0.6937984496124031, + "grad_norm": 0.16478034853935242, + "learning_rate": 0.00030481821958321467, + "loss": 1.9538, + "step": 179 + }, + { + "epoch": 0.6976744186046512, + "grad_norm": 0.16942182183265686, + "learning_rate": 0.00030443673697639955, + "loss": 2.3055, + "step": 180 + }, + { + "epoch": 0.7015503875968992, + "grad_norm": 0.16306325793266296, + "learning_rate": 0.0003040526289474484, + "loss": 2.0216, + "step": 181 + }, + { + "epoch": 0.7054263565891473, + "grad_norm": 0.1586439609527588, + "learning_rate": 0.0003036659027286425, + "loss": 2.242, + "step": 182 + }, + { + "epoch": 0.7093023255813954, + "grad_norm": 0.18067839741706848, + "learning_rate": 0.00030327656560156065, + "loss": 2.201, + "step": 183 + }, + { + "epoch": 0.7131782945736435, + "grad_norm": 0.17424075305461884, + "learning_rate": 0.00030288462489694176, + "loss": 2.0878, + "step": 184 + }, + { + "epoch": 0.7170542635658915, + "grad_norm": 0.17749184370040894, + "learning_rate": 0.0003024900879945468, + "loss": 2.0539, + "step": 185 + }, + { + "epoch": 0.7209302325581395, + "grad_norm": 0.1625218391418457, + "learning_rate": 0.00030209296232302016, + "loss": 2.252, + "step": 186 + }, + { + "epoch": 0.7248062015503876, + "grad_norm": 0.1539207398891449, + "learning_rate": 0.0003016932553597493, + "loss": 2.261, + "step": 187 + }, + { + "epoch": 0.7286821705426356, + "grad_norm": 0.1533818393945694, + "learning_rate": 0.00030129097463072455, + "loss": 2.0823, + "step": 188 + }, + { + "epoch": 0.7325581395348837, + "grad_norm": 0.16169801354408264, + "learning_rate": 0.0003008861277103968, + "loss": 2.3093, + "step": 189 + }, + { + "epoch": 0.7364341085271318, + "grad_norm": 0.16680897772312164, + "learning_rate": 0.00030047872222153525, + "loss": 1.8175, + "step": 190 + }, + { + "epoch": 0.7403100775193798, + "grad_norm": 0.15020348131656647, + "learning_rate": 0.0003000687658350839, + "loss": 2.054, + "step": 191 + }, + { + "epoch": 0.7441860465116279, + "grad_norm": 0.1465124785900116, + "learning_rate": 0.0002996562662700168, + "loss": 1.9802, + "step": 192 + }, + { + "epoch": 0.748062015503876, + "grad_norm": 0.1666540950536728, + "learning_rate": 0.00029924123129319315, + "loss": 2.2759, + "step": 193 + }, + { + "epoch": 0.751937984496124, + "grad_norm": 0.20095106959342957, + "learning_rate": 0.0002988236687192107, + "loss": 2.2055, + "step": 194 + }, + { + "epoch": 0.7558139534883721, + "grad_norm": 0.15560601651668549, + "learning_rate": 0.00029840358641025877, + "loss": 2.2051, + "step": 195 + }, + { + "epoch": 0.7596899224806202, + "grad_norm": 0.16015595197677612, + "learning_rate": 0.0002979809922759702, + "loss": 2.1635, + "step": 196 + }, + { + "epoch": 0.7635658914728682, + "grad_norm": 0.15207257866859436, + "learning_rate": 0.00029755589427327237, + "loss": 2.1105, + "step": 197 + }, + { + "epoch": 0.7674418604651163, + "grad_norm": 0.18203739821910858, + "learning_rate": 0.0002971283004062375, + "loss": 2.0973, + "step": 198 + }, + { + "epoch": 0.7713178294573644, + "grad_norm": 0.165207177400589, + "learning_rate": 0.00029669821872593187, + "loss": 2.2404, + "step": 199 + }, + { + "epoch": 0.7751937984496124, + "grad_norm": 0.16326136887073517, + "learning_rate": 0.000296265657330264, + "loss": 2.0473, + "step": 200 + }, + { + "epoch": 0.7790697674418605, + "grad_norm": 0.17581120133399963, + "learning_rate": 0.0002958306243638327, + "loss": 2.1448, + "step": 201 + }, + { + "epoch": 0.7829457364341085, + "grad_norm": 0.16953004896640778, + "learning_rate": 0.0002953931280177732, + "loss": 1.9978, + "step": 202 + }, + { + "epoch": 0.7868217054263565, + "grad_norm": 0.17670084536075592, + "learning_rate": 0.0002949531765296033, + "loss": 2.3193, + "step": 203 + }, + { + "epoch": 0.7906976744186046, + "grad_norm": 0.16181717813014984, + "learning_rate": 0.000294510778183068, + "loss": 1.9897, + "step": 204 + }, + { + "epoch": 0.7945736434108527, + "grad_norm": 0.16492588818073273, + "learning_rate": 0.00029406594130798364, + "loss": 2.2115, + "step": 205 + }, + { + "epoch": 0.7984496124031008, + "grad_norm": 0.1594804972410202, + "learning_rate": 0.00029361867428008104, + "loss": 2.205, + "step": 206 + }, + { + "epoch": 0.8023255813953488, + "grad_norm": 0.1583336740732193, + "learning_rate": 0.0002931689855208479, + "loss": 1.9226, + "step": 207 + }, + { + "epoch": 0.8062015503875969, + "grad_norm": 0.18443737924098969, + "learning_rate": 0.00029271688349737, + "loss": 2.3384, + "step": 208 + }, + { + "epoch": 0.810077519379845, + "grad_norm": 0.17095071077346802, + "learning_rate": 0.00029226237672217196, + "loss": 2.0513, + "step": 209 + }, + { + "epoch": 0.813953488372093, + "grad_norm": 0.1657685935497284, + "learning_rate": 0.00029180547375305704, + "loss": 2.0806, + "step": 210 + }, + { + "epoch": 0.8178294573643411, + "grad_norm": 0.16090308129787445, + "learning_rate": 0.00029134618319294573, + "loss": 2.1466, + "step": 211 + }, + { + "epoch": 0.8217054263565892, + "grad_norm": 0.1775992214679718, + "learning_rate": 0.00029088451368971396, + "loss": 2.084, + "step": 212 + }, + { + "epoch": 0.8255813953488372, + "grad_norm": 0.16048915684223175, + "learning_rate": 0.00029042047393603034, + "loss": 1.8937, + "step": 213 + }, + { + "epoch": 0.8294573643410853, + "grad_norm": 0.17156359553337097, + "learning_rate": 0.0002899540726691922, + "loss": 2.2556, + "step": 214 + }, + { + "epoch": 0.8333333333333334, + "grad_norm": 0.15980640053749084, + "learning_rate": 0.0002894853186709613, + "loss": 1.9849, + "step": 215 + }, + { + "epoch": 0.8372093023255814, + "grad_norm": 0.15312501788139343, + "learning_rate": 0.00028901422076739854, + "loss": 2.2545, + "step": 216 + }, + { + "epoch": 0.8410852713178295, + "grad_norm": 0.14746591448783875, + "learning_rate": 0.0002885407878286976, + "loss": 2.236, + "step": 217 + }, + { + "epoch": 0.8449612403100775, + "grad_norm": 0.16490989923477173, + "learning_rate": 0.0002880650287690181, + "loss": 2.1494, + "step": 218 + }, + { + "epoch": 0.8488372093023255, + "grad_norm": 0.1589692234992981, + "learning_rate": 0.0002875869525463173, + "loss": 2.1665, + "step": 219 + }, + { + "epoch": 0.8527131782945736, + "grad_norm": 0.17528380453586578, + "learning_rate": 0.0002871065681621823, + "loss": 1.9999, + "step": 220 + }, + { + "epoch": 0.8565891472868217, + "grad_norm": 0.1483374536037445, + "learning_rate": 0.0002866238846616597, + "loss": 2.1133, + "step": 221 + }, + { + "epoch": 0.8604651162790697, + "grad_norm": 0.1600683033466339, + "learning_rate": 0.00028613891113308566, + "loss": 2.061, + "step": 222 + }, + { + "epoch": 0.8643410852713178, + "grad_norm": 0.16365991532802582, + "learning_rate": 0.000285651656707915, + "loss": 2.0138, + "step": 223 + }, + { + "epoch": 0.8682170542635659, + "grad_norm": 0.19533555209636688, + "learning_rate": 0.0002851621305605486, + "loss": 2.3357, + "step": 224 + }, + { + "epoch": 0.872093023255814, + "grad_norm": 0.17096424102783203, + "learning_rate": 0.0002846703419081614, + "loss": 2.1513, + "step": 225 + }, + { + "epoch": 0.875968992248062, + "grad_norm": 0.1619122326374054, + "learning_rate": 0.00028417630001052844, + "loss": 2.0947, + "step": 226 + }, + { + "epoch": 0.8798449612403101, + "grad_norm": 0.16593049466609955, + "learning_rate": 0.00028368001416985054, + "loss": 1.9596, + "step": 227 + }, + { + "epoch": 0.8837209302325582, + "grad_norm": 0.15227988362312317, + "learning_rate": 0.0002831814937305792, + "loss": 1.9498, + "step": 228 + }, + { + "epoch": 0.8875968992248062, + "grad_norm": 0.16401511430740356, + "learning_rate": 0.0002826807480792408, + "loss": 2.2326, + "step": 229 + }, + { + "epoch": 0.8914728682170543, + "grad_norm": 0.20172202587127686, + "learning_rate": 0.0002821777866442596, + "loss": 2.2973, + "step": 230 + }, + { + "epoch": 0.8953488372093024, + "grad_norm": 0.15986618399620056, + "learning_rate": 0.00028167261889578034, + "loss": 1.9597, + "step": 231 + }, + { + "epoch": 0.8992248062015504, + "grad_norm": 0.16535168886184692, + "learning_rate": 0.0002811652543454899, + "loss": 1.9582, + "step": 232 + }, + { + "epoch": 0.9031007751937985, + "grad_norm": 0.18099620938301086, + "learning_rate": 0.0002806557025464385, + "loss": 2.3131, + "step": 233 + }, + { + "epoch": 0.9069767441860465, + "grad_norm": 0.16411282122135162, + "learning_rate": 0.0002801439730928591, + "loss": 2.0324, + "step": 234 + }, + { + "epoch": 0.9108527131782945, + "grad_norm": 0.1549196094274521, + "learning_rate": 0.00027963007561998765, + "loss": 2.0878, + "step": 235 + }, + { + "epoch": 0.9147286821705426, + "grad_norm": 0.1656661331653595, + "learning_rate": 0.00027911401980388093, + "loss": 1.9896, + "step": 236 + }, + { + "epoch": 0.9186046511627907, + "grad_norm": 0.1635042279958725, + "learning_rate": 0.0002785958153612348, + "loss": 1.9988, + "step": 237 + }, + { + "epoch": 0.9224806201550387, + "grad_norm": 0.18084381520748138, + "learning_rate": 0.0002780754720492012, + "loss": 2.1724, + "step": 238 + }, + { + "epoch": 0.9263565891472868, + "grad_norm": 0.1592792123556137, + "learning_rate": 0.0002775529996652041, + "loss": 2.1535, + "step": 239 + }, + { + "epoch": 0.9302325581395349, + "grad_norm": 0.15081603825092316, + "learning_rate": 0.00027702840804675553, + "loss": 2.2441, + "step": 240 + }, + { + "epoch": 0.9341085271317829, + "grad_norm": 0.1784433275461197, + "learning_rate": 0.00027650170707127, + "loss": 2.2237, + "step": 241 + }, + { + "epoch": 0.937984496124031, + "grad_norm": 0.15613843500614166, + "learning_rate": 0.00027597290665587855, + "loss": 1.893, + "step": 242 + }, + { + "epoch": 0.9418604651162791, + "grad_norm": 0.15053103864192963, + "learning_rate": 0.0002754420167572421, + "loss": 2.282, + "step": 243 + }, + { + "epoch": 0.9457364341085271, + "grad_norm": 0.15055972337722778, + "learning_rate": 0.0002749090473713641, + "loss": 2.0519, + "step": 244 + }, + { + "epoch": 0.9496124031007752, + "grad_norm": 0.15554535388946533, + "learning_rate": 0.00027437400853340215, + "loss": 1.9656, + "step": 245 + }, + { + "epoch": 0.9534883720930233, + "grad_norm": 0.16338570415973663, + "learning_rate": 0.00027383691031747885, + "loss": 2.3299, + "step": 246 + }, + { + "epoch": 0.9573643410852714, + "grad_norm": 0.16029350459575653, + "learning_rate": 0.0002732977628364927, + "loss": 2.0619, + "step": 247 + }, + { + "epoch": 0.9612403100775194, + "grad_norm": 0.1667211353778839, + "learning_rate": 0.0002727565762419271, + "loss": 2.1308, + "step": 248 + }, + { + "epoch": 0.9651162790697675, + "grad_norm": 0.1782902479171753, + "learning_rate": 0.0002722133607236595, + "loss": 1.9959, + "step": 249 + }, + { + "epoch": 0.9689922480620154, + "grad_norm": 0.15829254686832428, + "learning_rate": 0.0002716681265097696, + "loss": 2.1666, + "step": 250 + }, + { + "epoch": 0.9728682170542635, + "grad_norm": 0.15275566279888153, + "learning_rate": 0.00027112088386634645, + "loss": 2.2501, + "step": 251 + }, + { + "epoch": 0.9767441860465116, + "grad_norm": 0.17710180580615997, + "learning_rate": 0.00027057164309729557, + "loss": 2.2943, + "step": 252 + }, + { + "epoch": 0.9806201550387597, + "grad_norm": 0.14490962028503418, + "learning_rate": 0.0002700204145441446, + "loss": 2.1836, + "step": 253 + }, + { + "epoch": 0.9844961240310077, + "grad_norm": 0.16613595187664032, + "learning_rate": 0.0002694672085858487, + "loss": 2.1681, + "step": 254 + }, + { + "epoch": 0.9883720930232558, + "grad_norm": 0.15662476420402527, + "learning_rate": 0.0002689120356385952, + "loss": 1.9142, + "step": 255 + }, + { + "epoch": 0.9922480620155039, + "grad_norm": 0.1657550036907196, + "learning_rate": 0.0002683549061556074, + "loss": 2.3238, + "step": 256 + }, + { + "epoch": 0.9961240310077519, + "grad_norm": 0.1951078623533249, + "learning_rate": 0.0002677958306269477, + "loss": 2.368, + "step": 257 + }, + { + "epoch": 1.0, + "grad_norm": 0.14835147559642792, + "learning_rate": 0.00026723481957932025, + "loss": 1.9971, + "step": 258 + }, + { + "epoch": 1.003875968992248, + "grad_norm": 0.1541881114244461, + "learning_rate": 0.0002666718835758724, + "loss": 2.1023, + "step": 259 + }, + { + "epoch": 1.0077519379844961, + "grad_norm": 0.1681731790304184, + "learning_rate": 0.00026610703321599643, + "loss": 2.1575, + "step": 260 + }, + { + "epoch": 1.0116279069767442, + "grad_norm": 0.1625296175479889, + "learning_rate": 0.0002655402791351292, + "loss": 2.0262, + "step": 261 + }, + { + "epoch": 1.0155038759689923, + "grad_norm": 0.1610299050807953, + "learning_rate": 0.0002649716320045527, + "loss": 2.1875, + "step": 262 + }, + { + "epoch": 1.0193798449612403, + "grad_norm": 0.16845448315143585, + "learning_rate": 0.00026440110253119235, + "loss": 2.1463, + "step": 263 + }, + { + "epoch": 1.0232558139534884, + "grad_norm": 0.15429221093654633, + "learning_rate": 0.00026382870145741594, + "loss": 2.2537, + "step": 264 + }, + { + "epoch": 1.0271317829457365, + "grad_norm": 0.16575424373149872, + "learning_rate": 0.0002632544395608312, + "loss": 2.1336, + "step": 265 + }, + { + "epoch": 1.0310077519379846, + "grad_norm": 0.15692879259586334, + "learning_rate": 0.0002626783276540828, + "loss": 1.9072, + "step": 266 + }, + { + "epoch": 1.0348837209302326, + "grad_norm": 0.20094941556453705, + "learning_rate": 0.00026210037658464886, + "loss": 2.1421, + "step": 267 + }, + { + "epoch": 1.0387596899224807, + "grad_norm": 0.19174852967262268, + "learning_rate": 0.00026152059723463664, + "loss": 1.9939, + "step": 268 + }, + { + "epoch": 1.0426356589147288, + "grad_norm": 0.17150387167930603, + "learning_rate": 0.00026093900052057774, + "loss": 2.1736, + "step": 269 + }, + { + "epoch": 1.0465116279069768, + "grad_norm": 0.15520396828651428, + "learning_rate": 0.0002603555973932225, + "loss": 2.0932, + "step": 270 + }, + { + "epoch": 1.050387596899225, + "grad_norm": 0.1705746054649353, + "learning_rate": 0.0002597703988373336, + "loss": 2.2486, + "step": 271 + }, + { + "epoch": 1.054263565891473, + "grad_norm": 0.1639339029788971, + "learning_rate": 0.0002591834158714797, + "loss": 2.2351, + "step": 272 + }, + { + "epoch": 1.058139534883721, + "grad_norm": 0.17867504060268402, + "learning_rate": 0.00025859465954782744, + "loss": 2.25, + "step": 273 + }, + { + "epoch": 1.062015503875969, + "grad_norm": 0.20789186656475067, + "learning_rate": 0.0002580041409519339, + "loss": 1.9999, + "step": 274 + }, + { + "epoch": 1.065891472868217, + "grad_norm": 0.15863926708698273, + "learning_rate": 0.0002574118712025374, + "loss": 2.0538, + "step": 275 + }, + { + "epoch": 1.069767441860465, + "grad_norm": 0.16435660421848297, + "learning_rate": 0.00025681786145134825, + "loss": 2.0152, + "step": 276 + }, + { + "epoch": 1.073643410852713, + "grad_norm": 0.1803198754787445, + "learning_rate": 0.000256222122882839, + "loss": 2.3157, + "step": 277 + }, + { + "epoch": 1.0775193798449612, + "grad_norm": 0.19417981803417206, + "learning_rate": 0.0002556246667140337, + "loss": 2.1564, + "step": 278 + }, + { + "epoch": 1.003875968992248, + "grad_norm": 0.17067064344882965, + "learning_rate": 0.0002550255041942965, + "loss": 1.8682, + "step": 279 + }, + { + "epoch": 1.0077519379844961, + "grad_norm": 0.20531415939331055, + "learning_rate": 0.0002544246466051203, + "loss": 1.799, + "step": 280 + }, + { + "epoch": 1.0116279069767442, + "grad_norm": 0.21423804759979248, + "learning_rate": 0.000253822105259914, + "loss": 1.623, + "step": 281 + }, + { + "epoch": 1.0155038759689923, + "grad_norm": 0.17934457957744598, + "learning_rate": 0.0002532178915037894, + "loss": 1.8665, + "step": 282 + }, + { + "epoch": 1.0193798449612403, + "grad_norm": 0.19024011492729187, + "learning_rate": 0.0002526120167133479, + "loss": 1.8837, + "step": 283 + }, + { + "epoch": 1.0232558139534884, + "grad_norm": 0.26798877120018005, + "learning_rate": 0.0002520044922964662, + "loss": 1.8213, + "step": 284 + }, + { + "epoch": 1.0271317829457365, + "grad_norm": 0.19613595306873322, + "learning_rate": 0.0002513953296920811, + "loss": 1.8714, + "step": 285 + }, + { + "epoch": 1.0310077519379846, + "grad_norm": 0.17628167569637299, + "learning_rate": 0.0002507845403699748, + "loss": 1.8617, + "step": 286 + }, + { + "epoch": 1.0348837209302326, + "grad_norm": 0.19648019969463348, + "learning_rate": 0.0002501721358305584, + "loss": 2.1, + "step": 287 + }, + { + "epoch": 1.0387596899224807, + "grad_norm": 0.21330788731575012, + "learning_rate": 0.0002495581276046557, + "loss": 1.8563, + "step": 288 + }, + { + "epoch": 1.0426356589147288, + "grad_norm": 0.18956711888313293, + "learning_rate": 0.00024894252725328583, + "loss": 1.769, + "step": 289 + }, + { + "epoch": 1.0465116279069768, + "grad_norm": 0.17926643788814545, + "learning_rate": 0.00024832534636744566, + "loss": 1.8329, + "step": 290 + }, + { + "epoch": 1.050387596899225, + "grad_norm": 0.1793479472398758, + "learning_rate": 0.0002477065965678917, + "loss": 1.5659, + "step": 291 + }, + { + "epoch": 1.054263565891473, + "grad_norm": 0.19575071334838867, + "learning_rate": 0.00024708628950492116, + "loss": 1.7525, + "step": 292 + }, + { + "epoch": 1.058139534883721, + "grad_norm": 0.20665033161640167, + "learning_rate": 0.00024646443685815247, + "loss": 1.7979, + "step": 293 + }, + { + "epoch": 1.062015503875969, + "grad_norm": 0.2565125823020935, + "learning_rate": 0.0002458410503363055, + "loss": 1.8589, + "step": 294 + }, + { + "epoch": 1.0658914728682172, + "grad_norm": 0.18845273554325104, + "learning_rate": 0.0002452161416769813, + "loss": 1.9473, + "step": 295 + }, + { + "epoch": 1.069767441860465, + "grad_norm": 0.18714243173599243, + "learning_rate": 0.0002445897226464408, + "loss": 1.7174, + "step": 296 + }, + { + "epoch": 1.073643410852713, + "grad_norm": 0.20173239707946777, + "learning_rate": 0.0002439618050393833, + "loss": 1.7722, + "step": 297 + }, + { + "epoch": 1.0775193798449612, + "grad_norm": 0.1991068571805954, + "learning_rate": 0.0002433324006787244, + "loss": 1.8291, + "step": 298 + }, + { + "epoch": 1.0813953488372092, + "grad_norm": 0.20716890692710876, + "learning_rate": 0.00024270152141537367, + "loss": 1.8379, + "step": 299 + }, + { + "epoch": 1.0852713178294573, + "grad_norm": 0.18439684808254242, + "learning_rate": 0.00024206917912801104, + "loss": 1.7841, + "step": 300 + }, + { + "epoch": 1.0891472868217054, + "grad_norm": 0.1679239273071289, + "learning_rate": 0.00024143538572286356, + "loss": 1.8962, + "step": 301 + }, + { + "epoch": 1.0930232558139534, + "grad_norm": 0.1798442304134369, + "learning_rate": 0.00024080015313348086, + "loss": 1.9051, + "step": 302 + }, + { + "epoch": 1.0968992248062015, + "grad_norm": 0.18058885633945465, + "learning_rate": 0.00024016349332051082, + "loss": 1.9452, + "step": 303 + }, + { + "epoch": 1.1007751937984496, + "grad_norm": 0.17939876019954681, + "learning_rate": 0.00023952541827147416, + "loss": 1.7577, + "step": 304 + }, + { + "epoch": 1.1046511627906976, + "grad_norm": 0.1852078139781952, + "learning_rate": 0.00023888594000053855, + "loss": 1.9129, + "step": 305 + }, + { + "epoch": 1.1085271317829457, + "grad_norm": 0.18097913265228271, + "learning_rate": 0.0002382450705482929, + "loss": 1.8608, + "step": 306 + }, + { + "epoch": 1.1124031007751938, + "grad_norm": 0.21096573770046234, + "learning_rate": 0.00023760282198152007, + "loss": 2.0269, + "step": 307 + }, + { + "epoch": 1.1162790697674418, + "grad_norm": 0.18521122634410858, + "learning_rate": 0.0002369592063929702, + "loss": 1.7748, + "step": 308 + }, + { + "epoch": 1.12015503875969, + "grad_norm": 0.20764781534671783, + "learning_rate": 0.00023631423590113259, + "loss": 1.5756, + "step": 309 + }, + { + "epoch": 1.124031007751938, + "grad_norm": 0.17593301832675934, + "learning_rate": 0.00023566792265000772, + "loss": 1.7376, + "step": 310 + }, + { + "epoch": 1.127906976744186, + "grad_norm": 0.1941826045513153, + "learning_rate": 0.00023502027880887874, + "loss": 2.0077, + "step": 311 + }, + { + "epoch": 1.1317829457364341, + "grad_norm": 0.1869935393333435, + "learning_rate": 0.00023437131657208193, + "loss": 1.7087, + "step": 312 + }, + { + "epoch": 1.1356589147286822, + "grad_norm": 0.18788675963878632, + "learning_rate": 0.0002337210481587775, + "loss": 1.5735, + "step": 313 + }, + { + "epoch": 1.1395348837209303, + "grad_norm": 0.2027387022972107, + "learning_rate": 0.0002330694858127193, + "loss": 1.9059, + "step": 314 + }, + { + "epoch": 1.1434108527131783, + "grad_norm": 0.18780571222305298, + "learning_rate": 0.0002324166418020244, + "loss": 1.7282, + "step": 315 + }, + { + "epoch": 1.1472868217054264, + "grad_norm": 0.1976962685585022, + "learning_rate": 0.00023176252841894187, + "loss": 2.018, + "step": 316 + }, + { + "epoch": 1.1511627906976745, + "grad_norm": 0.19255152344703674, + "learning_rate": 0.00023110715797962177, + "loss": 1.729, + "step": 317 + }, + { + "epoch": 1.1550387596899225, + "grad_norm": 0.18749909102916718, + "learning_rate": 0.00023045054282388273, + "loss": 1.7549, + "step": 318 + }, + { + "epoch": 1.1589147286821706, + "grad_norm": 0.22557330131530762, + "learning_rate": 0.00022979269531497995, + "loss": 2.0677, + "step": 319 + }, + { + "epoch": 1.1627906976744187, + "grad_norm": 0.1746273934841156, + "learning_rate": 0.0002291336278393724, + "loss": 1.7589, + "step": 320 + }, + { + "epoch": 1.1666666666666667, + "grad_norm": 0.17809656262397766, + "learning_rate": 0.00022847335280648943, + "loss": 1.7851, + "step": 321 + }, + { + "epoch": 1.1705426356589148, + "grad_norm": 0.19532962143421173, + "learning_rate": 0.0002278118826484972, + "loss": 1.8515, + "step": 322 + }, + { + "epoch": 1.1744186046511629, + "grad_norm": 0.20081496238708496, + "learning_rate": 0.00022714922982006467, + "loss": 1.7351, + "step": 323 + }, + { + "epoch": 1.178294573643411, + "grad_norm": 0.22169405221939087, + "learning_rate": 0.000226485406798129, + "loss": 1.8978, + "step": 324 + }, + { + "epoch": 1.1821705426356588, + "grad_norm": 0.1990557610988617, + "learning_rate": 0.00022582042608166063, + "loss": 1.7837, + "step": 325 + }, + { + "epoch": 1.1860465116279069, + "grad_norm": 0.19927580654621124, + "learning_rate": 0.00022515430019142793, + "loss": 1.7193, + "step": 326 + }, + { + "epoch": 1.189922480620155, + "grad_norm": 0.19535136222839355, + "learning_rate": 0.00022448704166976166, + "loss": 2.1294, + "step": 327 + }, + { + "epoch": 1.193798449612403, + "grad_norm": 0.2005845010280609, + "learning_rate": 0.00022381866308031839, + "loss": 1.9261, + "step": 328 + }, + { + "epoch": 1.197674418604651, + "grad_norm": 0.20202966034412384, + "learning_rate": 0.00022314917700784438, + "loss": 1.8773, + "step": 329 + }, + { + "epoch": 1.2015503875968991, + "grad_norm": 0.23102688789367676, + "learning_rate": 0.00022247859605793835, + "loss": 1.7446, + "step": 330 + }, + { + "epoch": 1.2054263565891472, + "grad_norm": 0.19356843829154968, + "learning_rate": 0.00022180693285681419, + "loss": 1.7991, + "step": 331 + }, + { + "epoch": 1.2093023255813953, + "grad_norm": 0.22885212302207947, + "learning_rate": 0.0002211342000510633, + "loss": 1.6866, + "step": 332 + }, + { + "epoch": 1.2131782945736433, + "grad_norm": 0.18752378225326538, + "learning_rate": 0.00022046041030741645, + "loss": 1.8063, + "step": 333 + }, + { + "epoch": 1.2170542635658914, + "grad_norm": 0.18102799355983734, + "learning_rate": 0.00021978557631250505, + "loss": 2.0372, + "step": 334 + }, + { + "epoch": 1.2209302325581395, + "grad_norm": 0.19595623016357422, + "learning_rate": 0.0002191097107726228, + "loss": 1.9375, + "step": 335 + }, + { + "epoch": 1.2248062015503876, + "grad_norm": 0.1951194554567337, + "learning_rate": 0.00021843282641348586, + "loss": 1.9015, + "step": 336 + }, + { + "epoch": 1.2286821705426356, + "grad_norm": 0.19051988422870636, + "learning_rate": 0.00021775493597999359, + "loss": 1.978, + "step": 337 + }, + { + "epoch": 1.2325581395348837, + "grad_norm": 0.20707543194293976, + "learning_rate": 0.00021707605223598853, + "loss": 1.648, + "step": 338 + }, + { + "epoch": 1.2364341085271318, + "grad_norm": 0.1884542554616928, + "learning_rate": 0.00021639618796401595, + "loss": 1.8844, + "step": 339 + }, + { + "epoch": 1.2403100775193798, + "grad_norm": 0.19425931572914124, + "learning_rate": 0.00021571535596508337, + "loss": 1.8605, + "step": 340 + }, + { + "epoch": 1.244186046511628, + "grad_norm": 0.20685045421123505, + "learning_rate": 0.0002150335690584193, + "loss": 1.865, + "step": 341 + }, + { + "epoch": 1.248062015503876, + "grad_norm": 0.1994500607252121, + "learning_rate": 0.00021435084008123202, + "loss": 2.01, + "step": 342 + }, + { + "epoch": 1.251937984496124, + "grad_norm": 0.18918661773204803, + "learning_rate": 0.00021366718188846795, + "loss": 1.9086, + "step": 343 + }, + { + "epoch": 1.255813953488372, + "grad_norm": 0.18662329018115997, + "learning_rate": 0.0002129826073525693, + "loss": 1.7579, + "step": 344 + }, + { + "epoch": 1.2596899224806202, + "grad_norm": 0.21391060948371887, + "learning_rate": 0.00021229712936323206, + "loss": 1.8439, + "step": 345 + }, + { + "epoch": 1.2635658914728682, + "grad_norm": 0.19121584296226501, + "learning_rate": 0.000211610760827163, + "loss": 1.7058, + "step": 346 + }, + { + "epoch": 1.2674418604651163, + "grad_norm": 0.19377876818180084, + "learning_rate": 0.00021092351466783695, + "loss": 1.7458, + "step": 347 + }, + { + "epoch": 1.2713178294573644, + "grad_norm": 0.181095689535141, + "learning_rate": 0.00021023540382525313, + "loss": 1.8254, + "step": 348 + }, + { + "epoch": 1.2751937984496124, + "grad_norm": 0.18665671348571777, + "learning_rate": 0.00020954644125569186, + "loss": 1.8718, + "step": 349 + }, + { + "epoch": 1.2790697674418605, + "grad_norm": 0.19250045716762543, + "learning_rate": 0.00020885663993147022, + "loss": 1.9262, + "step": 350 + }, + { + "epoch": 1.2829457364341086, + "grad_norm": 0.17288216948509216, + "learning_rate": 0.00020816601284069818, + "loss": 1.8269, + "step": 351 + }, + { + "epoch": 1.2868217054263567, + "grad_norm": 0.18719065189361572, + "learning_rate": 0.00020747457298703388, + "loss": 1.6828, + "step": 352 + }, + { + "epoch": 1.2906976744186047, + "grad_norm": 0.214838445186615, + "learning_rate": 0.00020678233338943861, + "loss": 1.6086, + "step": 353 + }, + { + "epoch": 1.2945736434108528, + "grad_norm": 0.20970992743968964, + "learning_rate": 0.0002060893070819321, + "loss": 1.8411, + "step": 354 + }, + { + "epoch": 1.2984496124031009, + "grad_norm": 0.19377633929252625, + "learning_rate": 0.00020539550711334666, + "loss": 1.7452, + "step": 355 + }, + { + "epoch": 1.302325581395349, + "grad_norm": 0.21270965039730072, + "learning_rate": 0.0002047009465470818, + "loss": 1.8104, + "step": 356 + }, + { + "epoch": 1.306201550387597, + "grad_norm": 0.20388975739479065, + "learning_rate": 0.00020400563846085804, + "loss": 1.8727, + "step": 357 + }, + { + "epoch": 1.310077519379845, + "grad_norm": 0.20084132254123688, + "learning_rate": 0.00020330959594647087, + "loss": 1.833, + "step": 358 + }, + { + "epoch": 1.3139534883720931, + "grad_norm": 0.21278367936611176, + "learning_rate": 0.00020261283210954418, + "loss": 1.8254, + "step": 359 + }, + { + "epoch": 1.3178294573643412, + "grad_norm": 0.20003218948841095, + "learning_rate": 0.00020191536006928338, + "loss": 1.6369, + "step": 360 + }, + { + "epoch": 1.3217054263565893, + "grad_norm": 0.1956341415643692, + "learning_rate": 0.00020121719295822856, + "loss": 1.56, + "step": 361 + }, + { + "epoch": 1.3255813953488373, + "grad_norm": 0.2056959867477417, + "learning_rate": 0.0002005183439220071, + "loss": 1.9608, + "step": 362 + }, + { + "epoch": 1.3294573643410852, + "grad_norm": 0.20560531318187714, + "learning_rate": 0.00019981882611908616, + "loss": 1.5359, + "step": 363 + }, + { + "epoch": 1.3333333333333333, + "grad_norm": 0.20920397341251373, + "learning_rate": 0.000199118652720525, + "loss": 1.7916, + "step": 364 + }, + { + "epoch": 1.3372093023255813, + "grad_norm": 0.20703092217445374, + "learning_rate": 0.00019841783690972695, + "loss": 1.7696, + "step": 365 + }, + { + "epoch": 1.3410852713178294, + "grad_norm": 0.2005213499069214, + "learning_rate": 0.00019771639188219118, + "loss": 1.755, + "step": 366 + }, + { + "epoch": 1.3449612403100775, + "grad_norm": 0.19476981461048126, + "learning_rate": 0.00019701433084526416, + "loss": 1.9473, + "step": 367 + }, + { + "epoch": 1.3488372093023255, + "grad_norm": 0.1986457258462906, + "learning_rate": 0.00019631166701789115, + "loss": 1.5943, + "step": 368 + }, + { + "epoch": 1.3527131782945736, + "grad_norm": 0.1900886595249176, + "learning_rate": 0.0001956084136303671, + "loss": 1.829, + "step": 369 + }, + { + "epoch": 1.3565891472868217, + "grad_norm": 0.19796966016292572, + "learning_rate": 0.00019490458392408777, + "loss": 1.6652, + "step": 370 + }, + { + "epoch": 1.3604651162790697, + "grad_norm": 0.19012434780597687, + "learning_rate": 0.00019420019115130022, + "loss": 2.0372, + "step": 371 + }, + { + "epoch": 1.3643410852713178, + "grad_norm": 0.20295971632003784, + "learning_rate": 0.0001934952485748534, + "loss": 1.7472, + "step": 372 + }, + { + "epoch": 1.3682170542635659, + "grad_norm": 0.24137261509895325, + "learning_rate": 0.00019278976946794838, + "loss": 1.7415, + "step": 373 + }, + { + "epoch": 1.372093023255814, + "grad_norm": 0.19063839316368103, + "learning_rate": 0.00019208376711388837, + "loss": 1.6088, + "step": 374 + }, + { + "epoch": 1.375968992248062, + "grad_norm": 0.21088172495365143, + "learning_rate": 0.00019137725480582884, + "loss": 2.0018, + "step": 375 + }, + { + "epoch": 1.37984496124031, + "grad_norm": 0.18817083537578583, + "learning_rate": 0.00019067024584652688, + "loss": 1.7041, + "step": 376 + }, + { + "epoch": 1.3837209302325582, + "grad_norm": 0.19199052453041077, + "learning_rate": 0.000189962753548091, + "loss": 1.7152, + "step": 377 + }, + { + "epoch": 1.3875968992248062, + "grad_norm": 0.19253146648406982, + "learning_rate": 0.00018925479123173042, + "loss": 1.8655, + "step": 378 + }, + { + "epoch": 1.3914728682170543, + "grad_norm": 0.19239702820777893, + "learning_rate": 0.00018854637222750418, + "loss": 1.9605, + "step": 379 + }, + { + "epoch": 1.3953488372093024, + "grad_norm": 0.19195358455181122, + "learning_rate": 0.00018783750987407015, + "loss": 1.7392, + "step": 380 + }, + { + "epoch": 1.3992248062015504, + "grad_norm": 0.22022220492362976, + "learning_rate": 0.00018712821751843398, + "loss": 1.7857, + "step": 381 + }, + { + "epoch": 1.4031007751937985, + "grad_norm": 0.20397743582725525, + "learning_rate": 0.0001864185085156978, + "loss": 1.8367, + "step": 382 + }, + { + "epoch": 1.4069767441860466, + "grad_norm": 0.21266423165798187, + "learning_rate": 0.00018570839622880848, + "loss": 1.7252, + "step": 383 + }, + { + "epoch": 1.4108527131782946, + "grad_norm": 0.2537650465965271, + "learning_rate": 0.0001849978940283065, + "loss": 1.6755, + "step": 384 + }, + { + "epoch": 1.4147286821705427, + "grad_norm": 0.2007053792476654, + "learning_rate": 0.00018428701529207367, + "loss": 1.8301, + "step": 385 + }, + { + "epoch": 1.4186046511627908, + "grad_norm": 0.2502276301383972, + "learning_rate": 0.0001835757734050817, + "loss": 1.8444, + "step": 386 + }, + { + "epoch": 1.4224806201550386, + "grad_norm": 0.21285481750965118, + "learning_rate": 0.00018286418175913986, + "loss": 1.8921, + "step": 387 + }, + { + "epoch": 1.4263565891472867, + "grad_norm": 0.192243754863739, + "learning_rate": 0.0001821522537526431, + "loss": 1.7928, + "step": 388 + }, + { + "epoch": 1.4302325581395348, + "grad_norm": 0.18463876843452454, + "learning_rate": 0.0001814400027903194, + "loss": 1.7785, + "step": 389 + }, + { + "epoch": 1.4341085271317828, + "grad_norm": 0.19326354563236237, + "learning_rate": 0.00018072744228297784, + "loss": 1.8843, + "step": 390 + }, + { + "epoch": 1.437984496124031, + "grad_norm": 0.1868479996919632, + "learning_rate": 0.00018001458564725572, + "loss": 1.5857, + "step": 391 + }, + { + "epoch": 1.441860465116279, + "grad_norm": 0.23313790559768677, + "learning_rate": 0.00017930144630536607, + "loss": 1.8934, + "step": 392 + }, + { + "epoch": 1.445736434108527, + "grad_norm": 0.2015128880739212, + "learning_rate": 0.00017858803768484497, + "loss": 1.6785, + "step": 393 + }, + { + "epoch": 1.449612403100775, + "grad_norm": 0.20901648700237274, + "learning_rate": 0.00017787437321829862, + "loss": 1.6072, + "step": 394 + }, + { + "epoch": 1.4534883720930232, + "grad_norm": 0.21417102217674255, + "learning_rate": 0.0001771604663431506, + "loss": 1.8706, + "step": 395 + }, + { + "epoch": 1.4573643410852712, + "grad_norm": 0.26056888699531555, + "learning_rate": 0.00017644633050138862, + "loss": 1.8553, + "step": 396 + }, + { + "epoch": 1.4612403100775193, + "grad_norm": 0.19383443892002106, + "learning_rate": 0.00017573197913931166, + "loss": 1.7848, + "step": 397 + }, + { + "epoch": 1.4651162790697674, + "grad_norm": 0.2180427759885788, + "learning_rate": 0.00017501742570727666, + "loss": 1.8704, + "step": 398 + }, + { + "epoch": 1.4689922480620154, + "grad_norm": 0.2105623483657837, + "learning_rate": 0.00017430268365944518, + "loss": 1.6322, + "step": 399 + }, + { + "epoch": 1.4728682170542635, + "grad_norm": 0.19639936089515686, + "learning_rate": 0.0001735877664535303, + "loss": 1.7703, + "step": 400 + }, + { + "epoch": 1.4767441860465116, + "grad_norm": 0.18290045857429504, + "learning_rate": 0.00017287268755054305, + "loss": 1.7469, + "step": 401 + }, + { + "epoch": 1.4806201550387597, + "grad_norm": 0.20623917877674103, + "learning_rate": 0.00017215746041453913, + "loss": 1.7563, + "step": 402 + }, + { + "epoch": 1.4844961240310077, + "grad_norm": 0.1958610713481903, + "learning_rate": 0.00017144209851236504, + "loss": 1.8197, + "step": 403 + }, + { + "epoch": 1.4883720930232558, + "grad_norm": 0.1950949728488922, + "learning_rate": 0.00017072661531340502, + "loss": 1.8786, + "step": 404 + }, + { + "epoch": 1.4922480620155039, + "grad_norm": 0.21500743925571442, + "learning_rate": 0.0001700110242893271, + "loss": 1.7809, + "step": 405 + }, + { + "epoch": 1.496124031007752, + "grad_norm": 0.2010657638311386, + "learning_rate": 0.00016929533891382945, + "loss": 1.6953, + "step": 406 + }, + { + "epoch": 1.5, + "grad_norm": 0.19953016936779022, + "learning_rate": 0.00016857957266238688, + "loss": 1.466, + "step": 407 + }, + { + "epoch": 1.503875968992248, + "grad_norm": 0.19881819188594818, + "learning_rate": 0.00016786373901199684, + "loss": 1.9139, + "step": 408 + }, + { + "epoch": 1.5077519379844961, + "grad_norm": 0.1984422504901886, + "learning_rate": 0.000167147851440926, + "loss": 1.9282, + "step": 409 + }, + { + "epoch": 1.5116279069767442, + "grad_norm": 0.19584600627422333, + "learning_rate": 0.00016643192342845602, + "loss": 1.7367, + "step": 410 + }, + { + "epoch": 1.5155038759689923, + "grad_norm": 0.1895333230495453, + "learning_rate": 0.00016571596845463044, + "loss": 1.8496, + "step": 411 + }, + { + "epoch": 1.5193798449612403, + "grad_norm": 0.1868915855884552, + "learning_rate": 0.000165, + "loss": 1.7451, + "step": 412 + }, + { + "epoch": 1.5232558139534884, + "grad_norm": 0.192902609705925, + "learning_rate": 0.00016428403154536962, + "loss": 1.996, + "step": 413 + }, + { + "epoch": 1.5271317829457365, + "grad_norm": 0.22422343492507935, + "learning_rate": 0.00016356807657154395, + "loss": 1.8168, + "step": 414 + }, + { + "epoch": 1.5310077519379846, + "grad_norm": 0.1989205926656723, + "learning_rate": 0.00016285214855907406, + "loss": 1.8789, + "step": 415 + }, + { + "epoch": 1.5348837209302326, + "grad_norm": 0.19881393015384674, + "learning_rate": 0.0001621362609880032, + "loss": 1.732, + "step": 416 + }, + { + "epoch": 1.5387596899224807, + "grad_norm": 0.197114959359169, + "learning_rate": 0.00016142042733761317, + "loss": 1.7816, + "step": 417 + }, + { + "epoch": 1.5426356589147288, + "grad_norm": 0.20128700137138367, + "learning_rate": 0.00016070466108617055, + "loss": 1.8697, + "step": 418 + }, + { + "epoch": 1.5465116279069768, + "grad_norm": 0.2489478439092636, + "learning_rate": 0.00015998897571067292, + "loss": 1.7559, + "step": 419 + }, + { + "epoch": 1.550387596899225, + "grad_norm": 0.2148464322090149, + "learning_rate": 0.00015927338468659497, + "loss": 1.6178, + "step": 420 + }, + { + "epoch": 1.554263565891473, + "grad_norm": 0.18926125764846802, + "learning_rate": 0.00015855790148763498, + "loss": 1.8989, + "step": 421 + }, + { + "epoch": 1.558139534883721, + "grad_norm": 0.2206331193447113, + "learning_rate": 0.00015784253958546092, + "loss": 1.4578, + "step": 422 + }, + { + "epoch": 1.562015503875969, + "grad_norm": 0.20707029104232788, + "learning_rate": 0.00015712731244945697, + "loss": 1.8484, + "step": 423 + }, + { + "epoch": 1.5658914728682172, + "grad_norm": 0.2055550217628479, + "learning_rate": 0.00015641223354646977, + "loss": 1.7355, + "step": 424 + }, + { + "epoch": 1.5697674418604652, + "grad_norm": 0.22906459867954254, + "learning_rate": 0.00015569731634055482, + "loss": 1.8945, + "step": 425 + }, + { + "epoch": 1.5736434108527133, + "grad_norm": 0.1993347406387329, + "learning_rate": 0.00015498257429272336, + "loss": 1.6523, + "step": 426 + }, + { + "epoch": 1.5775193798449614, + "grad_norm": 0.20685455203056335, + "learning_rate": 0.00015426802086068833, + "loss": 1.7329, + "step": 427 + }, + { + "epoch": 1.5813953488372094, + "grad_norm": 0.2264314740896225, + "learning_rate": 0.00015355366949861135, + "loss": 1.7889, + "step": 428 + }, + { + "epoch": 1.5852713178294575, + "grad_norm": 0.23605959117412567, + "learning_rate": 0.00015283953365684941, + "loss": 1.8501, + "step": 429 + }, + { + "epoch": 1.5891472868217056, + "grad_norm": 0.1887684017419815, + "learning_rate": 0.0001521256267817014, + "loss": 1.7191, + "step": 430 + }, + { + "epoch": 1.5930232558139537, + "grad_norm": 0.1952105611562729, + "learning_rate": 0.00015141196231515508, + "loss": 1.7227, + "step": 431 + }, + { + "epoch": 1.5968992248062015, + "grad_norm": 0.19459757208824158, + "learning_rate": 0.00015069855369463392, + "loss": 1.4825, + "step": 432 + }, + { + "epoch": 1.6007751937984496, + "grad_norm": 0.20146332681179047, + "learning_rate": 0.0001499854143527443, + "loss": 1.8256, + "step": 433 + }, + { + "epoch": 1.6046511627906976, + "grad_norm": 0.2057608962059021, + "learning_rate": 0.00014927255771702219, + "loss": 1.6188, + "step": 434 + }, + { + "epoch": 1.6085271317829457, + "grad_norm": 0.2795553207397461, + "learning_rate": 0.00014855999720968062, + "loss": 1.5248, + "step": 435 + }, + { + "epoch": 1.6124031007751938, + "grad_norm": 0.20762935280799866, + "learning_rate": 0.00014784774624735695, + "loss": 1.812, + "step": 436 + }, + { + "epoch": 1.6162790697674418, + "grad_norm": 0.20336000621318817, + "learning_rate": 0.00014713581824086014, + "loss": 2.0534, + "step": 437 + }, + { + "epoch": 1.62015503875969, + "grad_norm": 0.20982185006141663, + "learning_rate": 0.00014642422659491837, + "loss": 1.708, + "step": 438 + }, + { + "epoch": 1.624031007751938, + "grad_norm": 0.19309498369693756, + "learning_rate": 0.00014571298470792635, + "loss": 1.7285, + "step": 439 + }, + { + "epoch": 1.627906976744186, + "grad_norm": 0.2094258964061737, + "learning_rate": 0.00014500210597169353, + "loss": 1.9404, + "step": 440 + }, + { + "epoch": 1.6317829457364341, + "grad_norm": 0.20995444059371948, + "learning_rate": 0.00014429160377119152, + "loss": 1.8059, + "step": 441 + }, + { + "epoch": 1.6356589147286822, + "grad_norm": 0.19962961971759796, + "learning_rate": 0.00014358149148430225, + "loss": 1.5697, + "step": 442 + }, + { + "epoch": 1.6395348837209303, + "grad_norm": 0.2032284438610077, + "learning_rate": 0.000142871782481566, + "loss": 1.8137, + "step": 443 + }, + { + "epoch": 1.6434108527131783, + "grad_norm": 0.19651709496974945, + "learning_rate": 0.00014216249012592987, + "loss": 1.7584, + "step": 444 + }, + { + "epoch": 1.6472868217054264, + "grad_norm": 0.20097319781780243, + "learning_rate": 0.00014145362777249587, + "loss": 1.875, + "step": 445 + }, + { + "epoch": 1.6511627906976745, + "grad_norm": 0.1974637806415558, + "learning_rate": 0.00014074520876826955, + "loss": 1.9157, + "step": 446 + }, + { + "epoch": 1.6550387596899225, + "grad_norm": 0.1912589967250824, + "learning_rate": 0.00014003724645190898, + "loss": 1.8198, + "step": 447 + }, + { + "epoch": 1.6589147286821704, + "grad_norm": 0.22384630143642426, + "learning_rate": 0.00013932975415347314, + "loss": 2.0134, + "step": 448 + }, + { + "epoch": 1.6627906976744184, + "grad_norm": 0.22261658310890198, + "learning_rate": 0.00013862274519417118, + "loss": 1.9937, + "step": 449 + }, + { + "epoch": 1.6666666666666665, + "grad_norm": 0.20569778978824615, + "learning_rate": 0.0001379162328861116, + "loss": 1.7494, + "step": 450 + }, + { + "epoch": 1.6705426356589146, + "grad_norm": 0.19439737498760223, + "learning_rate": 0.00013721023053205164, + "loss": 1.8848, + "step": 451 + }, + { + "epoch": 1.6744186046511627, + "grad_norm": 0.1926531344652176, + "learning_rate": 0.0001365047514251466, + "loss": 1.8646, + "step": 452 + }, + { + "epoch": 1.6782945736434107, + "grad_norm": 0.2144368290901184, + "learning_rate": 0.00013579980884869977, + "loss": 1.8047, + "step": 453 + }, + { + "epoch": 1.6821705426356588, + "grad_norm": 0.20255987346172333, + "learning_rate": 0.00013509541607591226, + "loss": 1.8394, + "step": 454 + }, + { + "epoch": 1.6860465116279069, + "grad_norm": 0.1870235800743103, + "learning_rate": 0.00013439158636963292, + "loss": 1.8102, + "step": 455 + }, + { + "epoch": 1.689922480620155, + "grad_norm": 0.20020346343517303, + "learning_rate": 0.0001336883329821089, + "loss": 1.8804, + "step": 456 + }, + { + "epoch": 1.693798449612403, + "grad_norm": 0.1906907856464386, + "learning_rate": 0.00013298566915473581, + "loss": 1.9384, + "step": 457 + }, + { + "epoch": 1.697674418604651, + "grad_norm": 0.2018539309501648, + "learning_rate": 0.00013228360811780882, + "loss": 1.8863, + "step": 458 + }, + { + "epoch": 1.7015503875968991, + "grad_norm": 0.1981591433286667, + "learning_rate": 0.00013158216309027305, + "loss": 1.9656, + "step": 459 + }, + { + "epoch": 1.7054263565891472, + "grad_norm": 0.2033264935016632, + "learning_rate": 0.000130881347279475, + "loss": 1.6587, + "step": 460 + }, + { + "epoch": 1.7093023255813953, + "grad_norm": 0.1924613118171692, + "learning_rate": 0.00013018117388091386, + "loss": 1.6004, + "step": 461 + }, + { + "epoch": 1.7131782945736433, + "grad_norm": 0.19790911674499512, + "learning_rate": 0.00012948165607799296, + "loss": 1.8981, + "step": 462 + }, + { + "epoch": 1.7170542635658914, + "grad_norm": 0.20132358372211456, + "learning_rate": 0.0001287828070417715, + "loss": 1.9168, + "step": 463 + }, + { + "epoch": 1.7209302325581395, + "grad_norm": 0.19085679948329926, + "learning_rate": 0.00012808463993071661, + "loss": 1.6915, + "step": 464 + }, + { + "epoch": 1.7248062015503876, + "grad_norm": 0.2047281414270401, + "learning_rate": 0.00012738716789045582, + "loss": 1.6943, + "step": 465 + }, + { + "epoch": 1.7286821705426356, + "grad_norm": 0.21793071925640106, + "learning_rate": 0.00012669040405352916, + "loss": 1.9838, + "step": 466 + }, + { + "epoch": 1.7325581395348837, + "grad_norm": 0.21391068398952484, + "learning_rate": 0.00012599436153914198, + "loss": 1.8181, + "step": 467 + }, + { + "epoch": 1.7364341085271318, + "grad_norm": 0.21556377410888672, + "learning_rate": 0.00012529905345291825, + "loss": 1.8422, + "step": 468 + }, + { + "epoch": 1.7403100775193798, + "grad_norm": 0.2581706941127777, + "learning_rate": 0.00012460449288665337, + "loss": 1.9742, + "step": 469 + }, + { + "epoch": 1.744186046511628, + "grad_norm": 0.19443494081497192, + "learning_rate": 0.00012391069291806792, + "loss": 1.6538, + "step": 470 + }, + { + "epoch": 1.748062015503876, + "grad_norm": 0.1977704018354416, + "learning_rate": 0.00012321766661056138, + "loss": 1.5853, + "step": 471 + }, + { + "epoch": 1.751937984496124, + "grad_norm": 0.21871508657932281, + "learning_rate": 0.00012252542701296612, + "loss": 1.8603, + "step": 472 + }, + { + "epoch": 1.755813953488372, + "grad_norm": 0.21219857037067413, + "learning_rate": 0.00012183398715930184, + "loss": 2.0966, + "step": 473 + }, + { + "epoch": 1.7596899224806202, + "grad_norm": 0.2329765260219574, + "learning_rate": 0.00012114336006852986, + "loss": 1.62, + "step": 474 + }, + { + "epoch": 1.7635658914728682, + "grad_norm": 0.2001282125711441, + "learning_rate": 0.00012045355874430818, + "loss": 1.6761, + "step": 475 + }, + { + "epoch": 1.7674418604651163, + "grad_norm": 0.21139495074748993, + "learning_rate": 0.0001197645961747469, + "loss": 1.6687, + "step": 476 + }, + { + "epoch": 1.7713178294573644, + "grad_norm": 0.20851844549179077, + "learning_rate": 0.00011907648533216309, + "loss": 1.7873, + "step": 477 + }, + { + "epoch": 1.7751937984496124, + "grad_norm": 0.20665644109249115, + "learning_rate": 0.00011838923917283697, + "loss": 2.0133, + "step": 478 + }, + { + "epoch": 1.7790697674418605, + "grad_norm": 0.21509003639221191, + "learning_rate": 0.00011770287063676794, + "loss": 1.8063, + "step": 479 + }, + { + "epoch": 1.7829457364341086, + "grad_norm": 0.20424823462963104, + "learning_rate": 0.0001170173926474307, + "loss": 1.7912, + "step": 480 + }, + { + "epoch": 1.7868217054263567, + "grad_norm": 0.1964096873998642, + "learning_rate": 0.00011633281811153209, + "loss": 1.7892, + "step": 481 + }, + { + "epoch": 1.7906976744186047, + "grad_norm": 0.19199204444885254, + "learning_rate": 0.00011564915991876793, + "loss": 1.5681, + "step": 482 + }, + { + "epoch": 1.7945736434108528, + "grad_norm": 0.20553924143314362, + "learning_rate": 0.00011496643094158072, + "loss": 1.7004, + "step": 483 + }, + { + "epoch": 1.7984496124031009, + "grad_norm": 0.20590077340602875, + "learning_rate": 0.00011428464403491667, + "loss": 1.9121, + "step": 484 + }, + { + "epoch": 1.802325581395349, + "grad_norm": 0.2159814089536667, + "learning_rate": 0.00011360381203598404, + "loss": 1.7843, + "step": 485 + }, + { + "epoch": 1.806201550387597, + "grad_norm": 0.19363482296466827, + "learning_rate": 0.00011292394776401152, + "loss": 2.0043, + "step": 486 + }, + { + "epoch": 1.810077519379845, + "grad_norm": 0.20251871645450592, + "learning_rate": 0.00011224506402000645, + "loss": 1.7657, + "step": 487 + }, + { + "epoch": 1.8139534883720931, + "grad_norm": 0.2011270970106125, + "learning_rate": 0.0001115671735865142, + "loss": 1.9705, + "step": 488 + }, + { + "epoch": 1.8178294573643412, + "grad_norm": 0.20194721221923828, + "learning_rate": 0.00011089028922737721, + "loss": 1.7546, + "step": 489 + }, + { + "epoch": 1.8217054263565893, + "grad_norm": 0.19558173418045044, + "learning_rate": 0.00011021442368749496, + "loss": 1.7589, + "step": 490 + }, + { + "epoch": 1.8255813953488373, + "grad_norm": 0.1979272961616516, + "learning_rate": 0.00010953958969258363, + "loss": 1.9451, + "step": 491 + }, + { + "epoch": 1.8294573643410854, + "grad_norm": 0.1853906363248825, + "learning_rate": 0.00010886579994893669, + "loss": 1.6267, + "step": 492 + }, + { + "epoch": 1.8333333333333335, + "grad_norm": 0.22289901971817017, + "learning_rate": 0.0001081930671431858, + "loss": 1.8587, + "step": 493 + }, + { + "epoch": 1.8372093023255816, + "grad_norm": 0.1921209841966629, + "learning_rate": 0.00010752140394206169, + "loss": 1.543, + "step": 494 + }, + { + "epoch": 1.8410852713178296, + "grad_norm": 0.1961943507194519, + "learning_rate": 0.00010685082299215565, + "loss": 1.7071, + "step": 495 + }, + { + "epoch": 1.8449612403100775, + "grad_norm": 0.19070090353488922, + "learning_rate": 0.00010618133691968159, + "loss": 1.7627, + "step": 496 + }, + { + "epoch": 1.8488372093023255, + "grad_norm": 0.1972433626651764, + "learning_rate": 0.00010551295833023834, + "loss": 1.9585, + "step": 497 + }, + { + "epoch": 1.8527131782945736, + "grad_norm": 0.19546210765838623, + "learning_rate": 0.00010484569980857207, + "loss": 1.7804, + "step": 498 + }, + { + "epoch": 1.8565891472868217, + "grad_norm": 0.18603426218032837, + "learning_rate": 0.00010417957391833937, + "loss": 1.7544, + "step": 499 + }, + { + "epoch": 1.8604651162790697, + "grad_norm": 0.21424385905265808, + "learning_rate": 0.00010351459320187102, + "loss": 2.0136, + "step": 500 + }, + { + "epoch": 1.8643410852713178, + "grad_norm": 0.19600240886211395, + "learning_rate": 0.00010285077017993536, + "loss": 2.1243, + "step": 501 + }, + { + "epoch": 1.8682170542635659, + "grad_norm": 0.19789910316467285, + "learning_rate": 0.00010218811735150283, + "loss": 1.772, + "step": 502 + }, + { + "epoch": 1.872093023255814, + "grad_norm": 0.24993658065795898, + "learning_rate": 0.0001015266471935106, + "loss": 1.7537, + "step": 503 + }, + { + "epoch": 1.875968992248062, + "grad_norm": 0.2263271063566208, + "learning_rate": 0.0001008663721606276, + "loss": 1.6373, + "step": 504 + }, + { + "epoch": 1.87984496124031, + "grad_norm": 0.19725754857063293, + "learning_rate": 0.00010020730468502006, + "loss": 1.5511, + "step": 505 + }, + { + "epoch": 1.8837209302325582, + "grad_norm": 0.18990269303321838, + "learning_rate": 9.954945717611734e-05, + "loss": 1.9, + "step": 506 + }, + { + "epoch": 1.8875968992248062, + "grad_norm": 0.20127974450588226, + "learning_rate": 9.889284202037826e-05, + "loss": 1.8697, + "step": 507 + }, + { + "epoch": 1.8914728682170543, + "grad_norm": 0.19391153752803802, + "learning_rate": 9.823747158105813e-05, + "loss": 1.64, + "step": 508 + }, + { + "epoch": 1.8953488372093024, + "grad_norm": 0.19547265768051147, + "learning_rate": 9.758335819797565e-05, + "loss": 1.4242, + "step": 509 + }, + { + "epoch": 1.8992248062015504, + "grad_norm": 0.19835124909877777, + "learning_rate": 9.693051418728067e-05, + "loss": 1.6888, + "step": 510 + }, + { + "epoch": 1.9031007751937985, + "grad_norm": 0.2248801290988922, + "learning_rate": 9.627895184122254e-05, + "loss": 1.8677, + "step": 511 + }, + { + "epoch": 1.9069767441860463, + "grad_norm": 0.20312006771564484, + "learning_rate": 9.562868342791809e-05, + "loss": 1.7185, + "step": 512 + }, + { + "epoch": 1.9108527131782944, + "grad_norm": 0.2256229966878891, + "learning_rate": 9.497972119112128e-05, + "loss": 1.7775, + "step": 513 + }, + { + "epoch": 1.9147286821705425, + "grad_norm": 0.2004292607307434, + "learning_rate": 9.433207734999223e-05, + "loss": 1.7049, + "step": 514 + }, + { + "epoch": 1.9186046511627906, + "grad_norm": 0.19018398225307465, + "learning_rate": 9.368576409886741e-05, + "loss": 1.5486, + "step": 515 + }, + { + "epoch": 1.9224806201550386, + "grad_norm": 0.2169780731201172, + "learning_rate": 9.304079360702984e-05, + "loss": 1.7943, + "step": 516 + }, + { + "epoch": 1.9263565891472867, + "grad_norm": 0.18505652248859406, + "learning_rate": 9.239717801847988e-05, + "loss": 1.4213, + "step": 517 + }, + { + "epoch": 1.9302325581395348, + "grad_norm": 0.19745871424674988, + "learning_rate": 9.175492945170716e-05, + "loss": 1.7484, + "step": 518 + }, + { + "epoch": 1.9341085271317828, + "grad_norm": 0.19631022214889526, + "learning_rate": 9.111405999946145e-05, + "loss": 1.6813, + "step": 519 + }, + { + "epoch": 1.937984496124031, + "grad_norm": 0.20178832113742828, + "learning_rate": 9.047458172852592e-05, + "loss": 1.797, + "step": 520 + }, + { + "epoch": 1.941860465116279, + "grad_norm": 0.19533181190490723, + "learning_rate": 8.983650667948915e-05, + "loss": 1.7296, + "step": 521 + }, + { + "epoch": 1.945736434108527, + "grad_norm": 0.19685396552085876, + "learning_rate": 8.919984686651917e-05, + "loss": 1.9964, + "step": 522 + }, + { + "epoch": 1.949612403100775, + "grad_norm": 0.21287135779857635, + "learning_rate": 8.856461427713651e-05, + "loss": 1.7688, + "step": 523 + }, + { + "epoch": 1.9534883720930232, + "grad_norm": 0.19684074819087982, + "learning_rate": 8.793082087198896e-05, + "loss": 1.7588, + "step": 524 + }, + { + "epoch": 1.9573643410852712, + "grad_norm": 0.20424464344978333, + "learning_rate": 8.729847858462635e-05, + "loss": 1.804, + "step": 525 + }, + { + "epoch": 1.9612403100775193, + "grad_norm": 0.2178419530391693, + "learning_rate": 8.666759932127563e-05, + "loss": 1.7219, + "step": 526 + }, + { + "epoch": 1.9651162790697674, + "grad_norm": 0.20854228734970093, + "learning_rate": 8.603819496061677e-05, + "loss": 1.6547, + "step": 527 + }, + { + "epoch": 1.9689922480620154, + "grad_norm": 0.19146962463855743, + "learning_rate": 8.54102773535592e-05, + "loss": 1.7676, + "step": 528 + }, + { + "epoch": 1.9728682170542635, + "grad_norm": 0.20440424978733063, + "learning_rate": 8.478385832301868e-05, + "loss": 1.863, + "step": 529 + }, + { + "epoch": 1.9767441860465116, + "grad_norm": 0.19717960059642792, + "learning_rate": 8.415894966369449e-05, + "loss": 1.9114, + "step": 530 + }, + { + "epoch": 1.9806201550387597, + "grad_norm": 0.19504162669181824, + "learning_rate": 8.353556314184756e-05, + "loss": 1.8955, + "step": 531 + }, + { + "epoch": 1.9844961240310077, + "grad_norm": 0.21640577912330627, + "learning_rate": 8.291371049507882e-05, + "loss": 1.5879, + "step": 532 + }, + { + "epoch": 1.9883720930232558, + "grad_norm": 0.20818272233009338, + "learning_rate": 8.229340343210828e-05, + "loss": 1.6215, + "step": 533 + }, + { + "epoch": 1.9922480620155039, + "grad_norm": 0.199616938829422, + "learning_rate": 8.167465363255434e-05, + "loss": 1.8259, + "step": 534 + }, + { + "epoch": 1.996124031007752, + "grad_norm": 0.21627697348594666, + "learning_rate": 8.105747274671419e-05, + "loss": 2.042, + "step": 535 + }, + { + "epoch": 2.0, + "grad_norm": 0.2383459061384201, + "learning_rate": 8.044187239534429e-05, + "loss": 1.7217, + "step": 536 + }, + { + "epoch": 2.003875968992248, + "grad_norm": 0.2019493281841278, + "learning_rate": 7.982786416944163e-05, + "loss": 1.6845, + "step": 537 + }, + { + "epoch": 2.007751937984496, + "grad_norm": 0.2237274944782257, + "learning_rate": 7.92154596300252e-05, + "loss": 1.4392, + "step": 538 + }, + { + "epoch": 2.011627906976744, + "grad_norm": 0.22651292383670807, + "learning_rate": 7.860467030791895e-05, + "loss": 1.5396, + "step": 539 + }, + { + "epoch": 2.0155038759689923, + "grad_norm": 0.2347649782896042, + "learning_rate": 7.799550770353385e-05, + "loss": 1.7393, + "step": 540 + }, + { + "epoch": 2.0193798449612403, + "grad_norm": 0.22472216188907623, + "learning_rate": 7.738798328665212e-05, + "loss": 1.7223, + "step": 541 + }, + { + "epoch": 2.0232558139534884, + "grad_norm": 0.20871466398239136, + "learning_rate": 7.678210849621058e-05, + "loss": 1.6426, + "step": 542 + }, + { + "epoch": 2.0271317829457365, + "grad_norm": 0.20463484525680542, + "learning_rate": 7.617789474008606e-05, + "loss": 1.6959, + "step": 543 + }, + { + "epoch": 2.0310077519379846, + "grad_norm": 0.21245327591896057, + "learning_rate": 7.55753533948797e-05, + "loss": 1.4701, + "step": 544 + }, + { + "epoch": 2.0348837209302326, + "grad_norm": 0.20299085974693298, + "learning_rate": 7.497449580570352e-05, + "loss": 1.6918, + "step": 545 + }, + { + "epoch": 2.0387596899224807, + "grad_norm": 0.20907790958881378, + "learning_rate": 7.437533328596635e-05, + "loss": 2.0656, + "step": 546 + }, + { + "epoch": 2.0426356589147288, + "grad_norm": 0.20104455947875977, + "learning_rate": 7.377787711716099e-05, + "loss": 1.7928, + "step": 547 + }, + { + "epoch": 2.046511627906977, + "grad_norm": 0.20086194574832916, + "learning_rate": 7.318213854865179e-05, + "loss": 1.6931, + "step": 548 + }, + { + "epoch": 2.050387596899225, + "grad_norm": 0.19403457641601562, + "learning_rate": 7.258812879746258e-05, + "loss": 1.648, + "step": 549 + }, + { + "epoch": 2.054263565891473, + "grad_norm": 0.19223634898662567, + "learning_rate": 7.19958590480661e-05, + "loss": 1.7832, + "step": 550 + }, + { + "epoch": 2.058139534883721, + "grad_norm": 0.2628048360347748, + "learning_rate": 7.140534045217254e-05, + "loss": 1.8131, + "step": 551 + }, + { + "epoch": 2.062015503875969, + "grad_norm": 0.19258858263492584, + "learning_rate": 7.08165841285204e-05, + "loss": 2.0739, + "step": 552 + }, + { + "epoch": 2.065891472868217, + "grad_norm": 0.19849108159542084, + "learning_rate": 7.02296011626664e-05, + "loss": 1.5161, + "step": 553 + }, + { + "epoch": 2.0697674418604652, + "grad_norm": 0.2288181036710739, + "learning_rate": 6.964440260677757e-05, + "loss": 1.8076, + "step": 554 + }, + { + "epoch": 2.003875968992248, + "grad_norm": 0.2025528997182846, + "learning_rate": 6.906099947942226e-05, + "loss": 1.3829, + "step": 555 + }, + { + "epoch": 2.007751937984496, + "grad_norm": 0.2283174991607666, + "learning_rate": 6.847940276536336e-05, + "loss": 1.3538, + "step": 556 + }, + { + "epoch": 2.011627906976744, + "grad_norm": 0.22030316293239594, + "learning_rate": 6.789962341535114e-05, + "loss": 1.4349, + "step": 557 + }, + { + "epoch": 2.0155038759689923, + "grad_norm": 0.2190942019224167, + "learning_rate": 6.732167234591726e-05, + "loss": 1.3267, + "step": 558 + }, + { + "epoch": 2.0193798449612403, + "grad_norm": 0.24970054626464844, + "learning_rate": 6.674556043916883e-05, + "loss": 1.4094, + "step": 559 + }, + { + "epoch": 2.0232558139534884, + "grad_norm": 0.24456344544887543, + "learning_rate": 6.617129854258406e-05, + "loss": 1.129, + "step": 560 + }, + { + "epoch": 2.0271317829457365, + "grad_norm": 0.27006039023399353, + "learning_rate": 6.559889746880764e-05, + "loss": 1.1812, + "step": 561 + }, + { + "epoch": 2.0310077519379846, + "grad_norm": 0.27538713812828064, + "learning_rate": 6.502836799544728e-05, + "loss": 1.1389, + "step": 562 + }, + { + "epoch": 2.0348837209302326, + "grad_norm": 0.3555956482887268, + "learning_rate": 6.445972086487074e-05, + "loss": 1.5623, + "step": 563 + }, + { + "epoch": 2.0387596899224807, + "grad_norm": 0.35483232140541077, + "learning_rate": 6.389296678400357e-05, + "loss": 1.1459, + "step": 564 + }, + { + "epoch": 2.0426356589147288, + "grad_norm": 0.36105233430862427, + "learning_rate": 6.332811642412762e-05, + "loss": 1.2068, + "step": 565 + }, + { + "epoch": 2.046511627906977, + "grad_norm": 0.25484374165534973, + "learning_rate": 6.276518042067982e-05, + "loss": 1.3551, + "step": 566 + }, + { + "epoch": 2.050387596899225, + "grad_norm": 0.24508428573608398, + "learning_rate": 6.22041693730523e-05, + "loss": 1.3885, + "step": 567 + }, + { + "epoch": 2.054263565891473, + "grad_norm": 0.24084091186523438, + "learning_rate": 6.164509384439258e-05, + "loss": 1.2559, + "step": 568 + }, + { + "epoch": 2.058139534883721, + "grad_norm": 0.24399569630622864, + "learning_rate": 6.10879643614048e-05, + "loss": 1.2449, + "step": 569 + }, + { + "epoch": 2.062015503875969, + "grad_norm": 0.22874587774276733, + "learning_rate": 6.053279141415124e-05, + "loss": 1.2807, + "step": 570 + }, + { + "epoch": 2.065891472868217, + "grad_norm": 0.23427356779575348, + "learning_rate": 5.997958545585541e-05, + "loss": 1.223, + "step": 571 + }, + { + "epoch": 2.0697674418604652, + "grad_norm": 0.23317009210586548, + "learning_rate": 5.9428356902704406e-05, + "loss": 1.3503, + "step": 572 + }, + { + "epoch": 2.0736434108527133, + "grad_norm": 0.2481788992881775, + "learning_rate": 5.887911613365358e-05, + "loss": 1.5248, + "step": 573 + }, + { + "epoch": 2.0775193798449614, + "grad_norm": 0.2363855540752411, + "learning_rate": 5.833187349023042e-05, + "loss": 1.3902, + "step": 574 + }, + { + "epoch": 2.0813953488372094, + "grad_norm": 0.21973107755184174, + "learning_rate": 5.778663927634054e-05, + "loss": 1.1951, + "step": 575 + }, + { + "epoch": 2.0852713178294575, + "grad_norm": 0.24858590960502625, + "learning_rate": 5.724342375807296e-05, + "loss": 1.306, + "step": 576 + }, + { + "epoch": 2.0891472868217056, + "grad_norm": 0.2570536136627197, + "learning_rate": 5.670223716350734e-05, + "loss": 1.2423, + "step": 577 + }, + { + "epoch": 2.0930232558139537, + "grad_norm": 0.2506595849990845, + "learning_rate": 5.616308968252116e-05, + "loss": 1.5181, + "step": 578 + }, + { + "epoch": 2.0968992248062017, + "grad_norm": 0.2839462459087372, + "learning_rate": 5.562599146659789e-05, + "loss": 1.379, + "step": 579 + }, + { + "epoch": 2.10077519379845, + "grad_norm": 0.245258629322052, + "learning_rate": 5.509095262863592e-05, + "loss": 1.3119, + "step": 580 + }, + { + "epoch": 2.104651162790698, + "grad_norm": 0.25935980677604675, + "learning_rate": 5.455798324275785e-05, + "loss": 1.2769, + "step": 581 + }, + { + "epoch": 2.108527131782946, + "grad_norm": 0.2609214186668396, + "learning_rate": 5.4027093344121484e-05, + "loss": 1.3932, + "step": 582 + }, + { + "epoch": 2.112403100775194, + "grad_norm": 0.26635345816612244, + "learning_rate": 5.3498292928730014e-05, + "loss": 1.3211, + "step": 583 + }, + { + "epoch": 2.116279069767442, + "grad_norm": 0.28587472438812256, + "learning_rate": 5.2971591953244504e-05, + "loss": 1.1748, + "step": 584 + }, + { + "epoch": 2.12015503875969, + "grad_norm": 0.26677605509757996, + "learning_rate": 5.244700033479588e-05, + "loss": 1.2278, + "step": 585 + }, + { + "epoch": 2.124031007751938, + "grad_norm": 0.26765909790992737, + "learning_rate": 5.192452795079885e-05, + "loss": 1.2706, + "step": 586 + }, + { + "epoch": 2.1279069767441863, + "grad_norm": 0.28276336193084717, + "learning_rate": 5.140418463876519e-05, + "loss": 1.522, + "step": 587 + }, + { + "epoch": 2.1317829457364343, + "grad_norm": 0.2587178945541382, + "learning_rate": 5.088598019611909e-05, + "loss": 1.2271, + "step": 588 + }, + { + "epoch": 2.135658914728682, + "grad_norm": 0.24723570048809052, + "learning_rate": 5.036992438001236e-05, + "loss": 1.1214, + "step": 589 + }, + { + "epoch": 2.13953488372093, + "grad_norm": 0.2772403359413147, + "learning_rate": 4.985602690714091e-05, + "loss": 1.2235, + "step": 590 + }, + { + "epoch": 2.143410852713178, + "grad_norm": 0.24144676327705383, + "learning_rate": 4.934429745356153e-05, + "loss": 1.1351, + "step": 591 + }, + { + "epoch": 2.147286821705426, + "grad_norm": 0.24926505982875824, + "learning_rate": 4.883474565451004e-05, + "loss": 1.3695, + "step": 592 + }, + { + "epoch": 2.1511627906976742, + "grad_norm": 0.2533572316169739, + "learning_rate": 4.832738110421967e-05, + "loss": 1.2713, + "step": 593 + }, + { + "epoch": 2.1550387596899223, + "grad_norm": 0.2442951202392578, + "learning_rate": 4.7822213355740394e-05, + "loss": 1.4233, + "step": 594 + }, + { + "epoch": 2.1589147286821704, + "grad_norm": 0.2528780698776245, + "learning_rate": 4.7319251920759175e-05, + "loss": 1.4764, + "step": 595 + }, + { + "epoch": 2.1627906976744184, + "grad_norm": 0.26493072509765625, + "learning_rate": 4.681850626942078e-05, + "loss": 1.16, + "step": 596 + }, + { + "epoch": 2.1666666666666665, + "grad_norm": 0.23339848220348358, + "learning_rate": 4.6319985830149517e-05, + "loss": 1.2937, + "step": 597 + }, + { + "epoch": 2.1705426356589146, + "grad_norm": 0.25760483741760254, + "learning_rate": 4.582369998947161e-05, + "loss": 1.381, + "step": 598 + }, + { + "epoch": 2.1744186046511627, + "grad_norm": 0.28579002618789673, + "learning_rate": 4.532965809183861e-05, + "loss": 1.3233, + "step": 599 + }, + { + "epoch": 2.1782945736434107, + "grad_norm": 0.25108829140663147, + "learning_rate": 4.4837869439451415e-05, + "loss": 1.2818, + "step": 600 + }, + { + "epoch": 2.182170542635659, + "grad_norm": 0.23311974108219147, + "learning_rate": 4.434834329208507e-05, + "loss": 1.096, + "step": 601 + }, + { + "epoch": 2.186046511627907, + "grad_norm": 0.2580869495868683, + "learning_rate": 4.3861088866914274e-05, + "loss": 1.0973, + "step": 602 + }, + { + "epoch": 2.189922480620155, + "grad_norm": 0.24833805859088898, + "learning_rate": 4.337611533834032e-05, + "loss": 1.3511, + "step": 603 + }, + { + "epoch": 2.193798449612403, + "grad_norm": 0.26399341225624084, + "learning_rate": 4.289343183781769e-05, + "loss": 1.218, + "step": 604 + }, + { + "epoch": 2.197674418604651, + "grad_norm": 0.2563190162181854, + "learning_rate": 4.241304745368273e-05, + "loss": 1.2484, + "step": 605 + }, + { + "epoch": 2.201550387596899, + "grad_norm": 0.27118462324142456, + "learning_rate": 4.193497123098196e-05, + "loss": 1.2956, + "step": 606 + }, + { + "epoch": 2.205426356589147, + "grad_norm": 0.25910139083862305, + "learning_rate": 4.14592121713024e-05, + "loss": 1.3615, + "step": 607 + }, + { + "epoch": 2.2093023255813953, + "grad_norm": 0.2624959945678711, + "learning_rate": 4.098577923260146e-05, + "loss": 1.5, + "step": 608 + }, + { + "epoch": 2.2131782945736433, + "grad_norm": 0.26731380820274353, + "learning_rate": 4.051468132903872e-05, + "loss": 1.2989, + "step": 609 + }, + { + "epoch": 2.2170542635658914, + "grad_norm": 0.2650803327560425, + "learning_rate": 4.004592733080782e-05, + "loss": 1.5101, + "step": 610 + }, + { + "epoch": 2.2209302325581395, + "grad_norm": 0.23880143463611603, + "learning_rate": 3.957952606396964e-05, + "loss": 1.1257, + "step": 611 + }, + { + "epoch": 2.2248062015503876, + "grad_norm": 0.25078123807907104, + "learning_rate": 3.911548631028602e-05, + "loss": 1.4061, + "step": 612 + }, + { + "epoch": 2.2286821705426356, + "grad_norm": 0.2630424201488495, + "learning_rate": 3.865381680705422e-05, + "loss": 1.284, + "step": 613 + }, + { + "epoch": 2.2325581395348837, + "grad_norm": 0.2509412467479706, + "learning_rate": 3.8194526246942975e-05, + "loss": 1.4054, + "step": 614 + }, + { + "epoch": 2.2364341085271318, + "grad_norm": 0.24037672579288483, + "learning_rate": 3.7737623277828025e-05, + "loss": 1.3992, + "step": 615 + }, + { + "epoch": 2.24031007751938, + "grad_norm": 0.26423701643943787, + "learning_rate": 3.728311650263008e-05, + "loss": 1.461, + "step": 616 + }, + { + "epoch": 2.244186046511628, + "grad_norm": 0.25176694989204407, + "learning_rate": 3.683101447915211e-05, + "loss": 1.402, + "step": 617 + }, + { + "epoch": 2.248062015503876, + "grad_norm": 0.25743353366851807, + "learning_rate": 3.6381325719918976e-05, + "loss": 1.3649, + "step": 618 + }, + { + "epoch": 2.251937984496124, + "grad_norm": 0.24827177822589874, + "learning_rate": 3.593405869201637e-05, + "loss": 1.4252, + "step": 619 + }, + { + "epoch": 2.255813953488372, + "grad_norm": 0.2707473039627075, + "learning_rate": 3.548922181693199e-05, + "loss": 1.3534, + "step": 620 + }, + { + "epoch": 2.25968992248062, + "grad_norm": 0.2554126977920532, + "learning_rate": 3.504682347039667e-05, + "loss": 1.2142, + "step": 621 + }, + { + "epoch": 2.2635658914728682, + "grad_norm": 0.2582671642303467, + "learning_rate": 3.460687198222681e-05, + "loss": 1.4812, + "step": 622 + }, + { + "epoch": 2.2674418604651163, + "grad_norm": 0.24931831657886505, + "learning_rate": 3.416937563616733e-05, + "loss": 1.24, + "step": 623 + }, + { + "epoch": 2.2713178294573644, + "grad_norm": 0.254974901676178, + "learning_rate": 3.373434266973601e-05, + "loss": 1.3225, + "step": 624 + }, + { + "epoch": 2.2751937984496124, + "grad_norm": 0.25716301798820496, + "learning_rate": 3.330178127406817e-05, + "loss": 1.3147, + "step": 625 + }, + { + "epoch": 2.2790697674418605, + "grad_norm": 0.25187844038009644, + "learning_rate": 3.2871699593762476e-05, + "loss": 1.2468, + "step": 626 + }, + { + "epoch": 2.2829457364341086, + "grad_norm": 0.2536908686161041, + "learning_rate": 3.24441057267276e-05, + "loss": 1.3098, + "step": 627 + }, + { + "epoch": 2.2868217054263567, + "grad_norm": 0.28247135877609253, + "learning_rate": 3.201900772402978e-05, + "loss": 1.3243, + "step": 628 + }, + { + "epoch": 2.2906976744186047, + "grad_norm": 0.27947402000427246, + "learning_rate": 3.159641358974126e-05, + "loss": 1.272, + "step": 629 + }, + { + "epoch": 2.294573643410853, + "grad_norm": 0.27251169085502625, + "learning_rate": 3.117633128078931e-05, + "loss": 1.209, + "step": 630 + }, + { + "epoch": 2.298449612403101, + "grad_norm": 0.2831606864929199, + "learning_rate": 3.0758768706806846e-05, + "loss": 1.3171, + "step": 631 + }, + { + "epoch": 2.302325581395349, + "grad_norm": 0.33359742164611816, + "learning_rate": 3.0343733729983185e-05, + "loss": 1.2489, + "step": 632 + }, + { + "epoch": 2.306201550387597, + "grad_norm": 0.3156187832355499, + "learning_rate": 2.9931234164916155e-05, + "loss": 1.3697, + "step": 633 + }, + { + "epoch": 2.310077519379845, + "grad_norm": 0.24116098880767822, + "learning_rate": 2.952127777846472e-05, + "loss": 1.2184, + "step": 634 + }, + { + "epoch": 2.313953488372093, + "grad_norm": 0.30879032611846924, + "learning_rate": 2.911387228960322e-05, + "loss": 1.362, + "step": 635 + }, + { + "epoch": 2.317829457364341, + "grad_norm": 0.25808271765708923, + "learning_rate": 2.8709025369275442e-05, + "loss": 1.2858, + "step": 636 + }, + { + "epoch": 2.3217054263565893, + "grad_norm": 0.2646956443786621, + "learning_rate": 2.8306744640250702e-05, + "loss": 1.3307, + "step": 637 + }, + { + "epoch": 2.3255813953488373, + "grad_norm": 0.25518307089805603, + "learning_rate": 2.790703767697985e-05, + "loss": 1.2729, + "step": 638 + }, + { + "epoch": 2.3294573643410854, + "grad_norm": 0.2450859695672989, + "learning_rate": 2.7509912005453224e-05, + "loss": 1.067, + "step": 639 + }, + { + "epoch": 2.3333333333333335, + "grad_norm": 0.26299986243247986, + "learning_rate": 2.7115375103058288e-05, + "loss": 1.3176, + "step": 640 + }, + { + "epoch": 2.3372093023255816, + "grad_norm": 0.2573457956314087, + "learning_rate": 2.672343439843936e-05, + "loss": 1.2987, + "step": 641 + }, + { + "epoch": 2.3410852713178296, + "grad_norm": 0.25773999094963074, + "learning_rate": 2.6334097271357512e-05, + "loss": 1.3351, + "step": 642 + }, + { + "epoch": 2.3449612403100777, + "grad_norm": 0.2760131359100342, + "learning_rate": 2.5947371052551607e-05, + "loss": 1.212, + "step": 643 + }, + { + "epoch": 2.3488372093023258, + "grad_norm": 0.2653418779373169, + "learning_rate": 2.556326302360044e-05, + "loss": 1.2889, + "step": 644 + }, + { + "epoch": 2.352713178294574, + "grad_norm": 0.26841187477111816, + "learning_rate": 2.5181780416785284e-05, + "loss": 1.2536, + "step": 645 + }, + { + "epoch": 2.356589147286822, + "grad_norm": 0.25338879227638245, + "learning_rate": 2.4802930414954242e-05, + "loss": 1.2338, + "step": 646 + }, + { + "epoch": 2.3604651162790695, + "grad_norm": 0.27084293961524963, + "learning_rate": 2.4426720151386478e-05, + "loss": 1.5218, + "step": 647 + }, + { + "epoch": 2.3643410852713176, + "grad_norm": 0.27807363867759705, + "learning_rate": 2.4053156709658235e-05, + "loss": 1.355, + "step": 648 + }, + { + "epoch": 2.3682170542635657, + "grad_norm": 0.25639334321022034, + "learning_rate": 2.3682247123509232e-05, + "loss": 1.1381, + "step": 649 + }, + { + "epoch": 2.3720930232558137, + "grad_norm": 0.27499181032180786, + "learning_rate": 2.3313998376710624e-05, + "loss": 1.3216, + "step": 650 + }, + { + "epoch": 2.375968992248062, + "grad_norm": 0.2780856490135193, + "learning_rate": 2.294841740293295e-05, + "loss": 1.165, + "step": 651 + }, + { + "epoch": 2.37984496124031, + "grad_norm": 0.24923618137836456, + "learning_rate": 2.2585511085615996e-05, + "loss": 1.1451, + "step": 652 + }, + { + "epoch": 2.383720930232558, + "grad_norm": 0.2662268877029419, + "learning_rate": 2.2225286257839073e-05, + "loss": 1.4445, + "step": 653 + }, + { + "epoch": 2.387596899224806, + "grad_norm": 0.2631353735923767, + "learning_rate": 2.1867749702192334e-05, + "loss": 1.3777, + "step": 654 + }, + { + "epoch": 2.391472868217054, + "grad_norm": 0.2878567576408386, + "learning_rate": 2.151290815064901e-05, + "loss": 1.2377, + "step": 655 + }, + { + "epoch": 2.395348837209302, + "grad_norm": 0.2559579908847809, + "learning_rate": 2.1160768284438775e-05, + "loss": 1.3982, + "step": 656 + }, + { + "epoch": 2.39922480620155, + "grad_norm": 0.2648860514163971, + "learning_rate": 2.0811336733921926e-05, + "loss": 1.237, + "step": 657 + }, + { + "epoch": 2.4031007751937983, + "grad_norm": 0.2823598384857178, + "learning_rate": 2.046462007846444e-05, + "loss": 1.2538, + "step": 658 + }, + { + "epoch": 2.4069767441860463, + "grad_norm": 0.2508406639099121, + "learning_rate": 2.0120624846314217e-05, + "loss": 1.3027, + "step": 659 + }, + { + "epoch": 2.4108527131782944, + "grad_norm": 0.26299890875816345, + "learning_rate": 1.9779357514478066e-05, + "loss": 1.4021, + "step": 660 + }, + { + "epoch": 2.4147286821705425, + "grad_norm": 0.2663334608078003, + "learning_rate": 1.944082450859986e-05, + "loss": 1.4515, + "step": 661 + }, + { + "epoch": 2.4186046511627906, + "grad_norm": 0.25878193974494934, + "learning_rate": 1.910503220283934e-05, + "loss": 1.3473, + "step": 662 + }, + { + "epoch": 2.4224806201550386, + "grad_norm": 0.25262004137039185, + "learning_rate": 1.8771986919752367e-05, + "loss": 1.3389, + "step": 663 + }, + { + "epoch": 2.4263565891472867, + "grad_norm": 0.26410895586013794, + "learning_rate": 1.844169493017171e-05, + "loss": 1.3108, + "step": 664 + }, + { + "epoch": 2.4302325581395348, + "grad_norm": 0.24397780001163483, + "learning_rate": 1.8114162453089036e-05, + "loss": 1.2601, + "step": 665 + }, + { + "epoch": 2.434108527131783, + "grad_norm": 0.24081014096736908, + "learning_rate": 1.778939565553765e-05, + "loss": 1.2056, + "step": 666 + }, + { + "epoch": 2.437984496124031, + "grad_norm": 0.271356999874115, + "learning_rate": 1.7467400652476762e-05, + "loss": 1.4752, + "step": 667 + }, + { + "epoch": 2.441860465116279, + "grad_norm": 0.2456904500722885, + "learning_rate": 1.7148183506675864e-05, + "loss": 1.4085, + "step": 668 + }, + { + "epoch": 2.445736434108527, + "grad_norm": 0.2549789249897003, + "learning_rate": 1.6831750228600952e-05, + "loss": 1.4168, + "step": 669 + }, + { + "epoch": 2.449612403100775, + "grad_norm": 0.2592543363571167, + "learning_rate": 1.6518106776301112e-05, + "loss": 1.3086, + "step": 670 + }, + { + "epoch": 2.453488372093023, + "grad_norm": 0.254226952791214, + "learning_rate": 1.620725905529663e-05, + "loss": 1.2994, + "step": 671 + }, + { + "epoch": 2.4573643410852712, + "grad_norm": 0.2479204684495926, + "learning_rate": 1.589921291846741e-05, + "loss": 1.2711, + "step": 672 + }, + { + "epoch": 2.4612403100775193, + "grad_norm": 0.2645493447780609, + "learning_rate": 1.5593974165943074e-05, + "loss": 1.3929, + "step": 673 + }, + { + "epoch": 2.4651162790697674, + "grad_norm": 0.2703893780708313, + "learning_rate": 1.5291548544993655e-05, + "loss": 1.4931, + "step": 674 + }, + { + "epoch": 2.4689922480620154, + "grad_norm": 0.2592923939228058, + "learning_rate": 1.4991941749921369e-05, + "loss": 1.2048, + "step": 675 + }, + { + "epoch": 2.4728682170542635, + "grad_norm": 0.2813677489757538, + "learning_rate": 1.4695159421953419e-05, + "loss": 1.2728, + "step": 676 + }, + { + "epoch": 2.4767441860465116, + "grad_norm": 0.2619397044181824, + "learning_rate": 1.4401207149135698e-05, + "loss": 1.1384, + "step": 677 + }, + { + "epoch": 2.4806201550387597, + "grad_norm": 0.27888745069503784, + "learning_rate": 1.411009046622775e-05, + "loss": 1.3301, + "step": 678 + }, + { + "epoch": 2.4844961240310077, + "grad_norm": 0.2559435963630676, + "learning_rate": 1.3821814854598378e-05, + "loss": 1.2627, + "step": 679 + }, + { + "epoch": 2.488372093023256, + "grad_norm": 0.25446653366088867, + "learning_rate": 1.3536385742122538e-05, + "loss": 1.128, + "step": 680 + }, + { + "epoch": 2.492248062015504, + "grad_norm": 0.2939082682132721, + "learning_rate": 1.3253808503079129e-05, + "loss": 1.2637, + "step": 681 + }, + { + "epoch": 2.496124031007752, + "grad_norm": 0.24750369787216187, + "learning_rate": 1.2974088458049774e-05, + "loss": 1.2881, + "step": 682 + }, + { + "epoch": 2.5, + "grad_norm": 0.25020653009414673, + "learning_rate": 1.2697230873818587e-05, + "loss": 1.3799, + "step": 683 + }, + { + "epoch": 2.503875968992248, + "grad_norm": 0.25006988644599915, + "learning_rate": 1.2423240963273123e-05, + "loss": 1.2513, + "step": 684 + }, + { + "epoch": 2.507751937984496, + "grad_norm": 0.27393513917922974, + "learning_rate": 1.2152123885306202e-05, + "loss": 1.2718, + "step": 685 + }, + { + "epoch": 2.511627906976744, + "grad_norm": 0.2598724961280823, + "learning_rate": 1.1883884744718704e-05, + "loss": 1.4166, + "step": 686 + }, + { + "epoch": 2.5155038759689923, + "grad_norm": 0.31768569350242615, + "learning_rate": 1.1618528592123451e-05, + "loss": 1.2255, + "step": 687 + }, + { + "epoch": 2.5193798449612403, + "grad_norm": 0.2585943341255188, + "learning_rate": 1.135606042385021e-05, + "loss": 1.2544, + "step": 688 + }, + { + "epoch": 2.5232558139534884, + "grad_norm": 0.2472546398639679, + "learning_rate": 1.1096485181851552e-05, + "loss": 1.1121, + "step": 689 + }, + { + "epoch": 2.5271317829457365, + "grad_norm": 0.26214876770973206, + "learning_rate": 1.0839807753609787e-05, + "loss": 1.3611, + "step": 690 + }, + { + "epoch": 2.5310077519379846, + "grad_norm": 0.26527321338653564, + "learning_rate": 1.0586032972044979e-05, + "loss": 0.9816, + "step": 691 + }, + { + "epoch": 2.5348837209302326, + "grad_norm": 0.2589716911315918, + "learning_rate": 1.0335165615423909e-05, + "loss": 1.1664, + "step": 692 + }, + { + "epoch": 2.5387596899224807, + "grad_norm": 0.26980432868003845, + "learning_rate": 1.0087210407270105e-05, + "loss": 1.5304, + "step": 693 + }, + { + "epoch": 2.5426356589147288, + "grad_norm": 0.26864027976989746, + "learning_rate": 9.842172016274983e-06, + "loss": 1.4349, + "step": 694 + }, + { + "epoch": 2.546511627906977, + "grad_norm": 0.24509486556053162, + "learning_rate": 9.600055056209803e-06, + "loss": 1.3234, + "step": 695 + }, + { + "epoch": 2.550387596899225, + "grad_norm": 0.24848882853984833, + "learning_rate": 9.360864085838973e-06, + "loss": 1.3879, + "step": 696 + }, + { + "epoch": 2.554263565891473, + "grad_norm": 0.27096468210220337, + "learning_rate": 9.124603608834071e-06, + "loss": 1.4016, + "step": 697 + }, + { + "epoch": 2.558139534883721, + "grad_norm": 0.24919958412647247, + "learning_rate": 8.891278073688985e-06, + "loss": 1.3114, + "step": 698 + }, + { + "epoch": 2.562015503875969, + "grad_norm": 0.2680855691432953, + "learning_rate": 8.660891873636498e-06, + "loss": 1.2027, + "step": 699 + }, + { + "epoch": 2.565891472868217, + "grad_norm": 0.2762061655521393, + "learning_rate": 8.433449346565108e-06, + "loss": 1.4543, + "step": 700 + }, + { + "epoch": 2.5697674418604652, + "grad_norm": 0.252043753862381, + "learning_rate": 8.208954774937692e-06, + "loss": 1.2538, + "step": 701 + }, + { + "epoch": 2.5736434108527133, + "grad_norm": 0.27675777673721313, + "learning_rate": 7.987412385710668e-06, + "loss": 1.2478, + "step": 702 + }, + { + "epoch": 2.5775193798449614, + "grad_norm": 0.25470390915870667, + "learning_rate": 7.768826350254634e-06, + "loss": 1.3053, + "step": 703 + }, + { + "epoch": 2.5813953488372094, + "grad_norm": 0.25610047578811646, + "learning_rate": 7.553200784275528e-06, + "loss": 1.366, + "step": 704 + }, + { + "epoch": 2.5852713178294575, + "grad_norm": 0.25024861097335815, + "learning_rate": 7.3405397477373775e-06, + "loss": 1.3907, + "step": 705 + }, + { + "epoch": 2.5891472868217056, + "grad_norm": 0.2611881196498871, + "learning_rate": 7.130847244785779e-06, + "loss": 1.2982, + "step": 706 + }, + { + "epoch": 2.5930232558139537, + "grad_norm": 0.2670976519584656, + "learning_rate": 6.9241272236724684e-06, + "loss": 1.4931, + "step": 707 + }, + { + "epoch": 2.5968992248062017, + "grad_norm": 0.25510528683662415, + "learning_rate": 6.720383576680977e-06, + "loss": 1.3591, + "step": 708 + }, + { + "epoch": 2.60077519379845, + "grad_norm": 0.3071907162666321, + "learning_rate": 6.519620140053416e-06, + "loss": 1.4327, + "step": 709 + }, + { + "epoch": 2.604651162790698, + "grad_norm": 0.27962440252304077, + "learning_rate": 6.321840693918205e-06, + "loss": 1.1472, + "step": 710 + }, + { + "epoch": 2.608527131782946, + "grad_norm": 0.2648572623729706, + "learning_rate": 6.127048962218875e-06, + "loss": 1.5193, + "step": 711 + }, + { + "epoch": 2.612403100775194, + "grad_norm": 0.2580854892730713, + "learning_rate": 5.9352486126439925e-06, + "loss": 1.4547, + "step": 712 + }, + { + "epoch": 2.616279069767442, + "grad_norm": 0.26917338371276855, + "learning_rate": 5.746443256558062e-06, + "loss": 1.4236, + "step": 713 + }, + { + "epoch": 2.62015503875969, + "grad_norm": 0.2621963620185852, + "learning_rate": 5.560636448933566e-06, + "loss": 1.4411, + "step": 714 + }, + { + "epoch": 2.624031007751938, + "grad_norm": 0.297036737203598, + "learning_rate": 5.377831688283975e-06, + "loss": 1.4621, + "step": 715 + }, + { + "epoch": 2.6279069767441863, + "grad_norm": 0.2709357440471649, + "learning_rate": 5.198032416597942e-06, + "loss": 1.3117, + "step": 716 + }, + { + "epoch": 2.6317829457364343, + "grad_norm": 0.2600153982639313, + "learning_rate": 5.021242019274458e-06, + "loss": 1.2619, + "step": 717 + }, + { + "epoch": 2.6356589147286824, + "grad_norm": 0.28544628620147705, + "learning_rate": 4.847463825059103e-06, + "loss": 1.2316, + "step": 718 + }, + { + "epoch": 2.6395348837209305, + "grad_norm": 0.25166937708854675, + "learning_rate": 4.6767011059813755e-06, + "loss": 1.2125, + "step": 719 + }, + { + "epoch": 2.6434108527131785, + "grad_norm": 0.2691721022129059, + "learning_rate": 4.5089570772931255e-06, + "loss": 1.2869, + "step": 720 + }, + { + "epoch": 2.6472868217054266, + "grad_norm": 0.28785765171051025, + "learning_rate": 4.344234897407919e-06, + "loss": 1.3122, + "step": 721 + }, + { + "epoch": 2.6511627906976747, + "grad_norm": 0.24798186123371124, + "learning_rate": 4.182537667841703e-06, + "loss": 1.1715, + "step": 722 + }, + { + "epoch": 2.6550387596899228, + "grad_norm": 0.28408151865005493, + "learning_rate": 4.023868433154223e-06, + "loss": 1.3831, + "step": 723 + }, + { + "epoch": 2.6589147286821704, + "grad_norm": 0.2505200207233429, + "learning_rate": 3.868230180891944e-06, + "loss": 1.3047, + "step": 724 + }, + { + "epoch": 2.6627906976744184, + "grad_norm": 0.24836868047714233, + "learning_rate": 3.7156258415315505e-06, + "loss": 1.2543, + "step": 725 + }, + { + "epoch": 2.6666666666666665, + "grad_norm": 0.2672933340072632, + "learning_rate": 3.566058288424942e-06, + "loss": 1.4565, + "step": 726 + }, + { + "epoch": 2.6705426356589146, + "grad_norm": 0.24122579395771027, + "learning_rate": 3.4195303377450594e-06, + "loss": 1.2132, + "step": 727 + }, + { + "epoch": 2.6744186046511627, + "grad_norm": 0.31489935517311096, + "learning_rate": 3.2760447484328548e-06, + "loss": 1.2152, + "step": 728 + }, + { + "epoch": 2.6782945736434107, + "grad_norm": 0.24762488901615143, + "learning_rate": 3.1356042221453766e-06, + "loss": 1.1724, + "step": 729 + }, + { + "epoch": 2.682170542635659, + "grad_norm": 0.2507333755493164, + "learning_rate": 2.998211403204843e-06, + "loss": 1.4529, + "step": 730 + }, + { + "epoch": 2.686046511627907, + "grad_norm": 0.27383190393447876, + "learning_rate": 2.863868878548979e-06, + "loss": 1.3259, + "step": 731 + }, + { + "epoch": 2.689922480620155, + "grad_norm": 0.24372157454490662, + "learning_rate": 2.7325791776821445e-06, + "loss": 1.2469, + "step": 732 + }, + { + "epoch": 2.693798449612403, + "grad_norm": 0.25599318742752075, + "learning_rate": 2.60434477262785e-06, + "loss": 1.3733, + "step": 733 + }, + { + "epoch": 2.697674418604651, + "grad_norm": 0.2459976226091385, + "learning_rate": 2.4791680778820455e-06, + "loss": 1.3466, + "step": 734 + }, + { + "epoch": 2.701550387596899, + "grad_norm": 0.2694660723209381, + "learning_rate": 2.357051450367873e-06, + "loss": 1.5102, + "step": 735 + }, + { + "epoch": 2.705426356589147, + "grad_norm": 0.27480435371398926, + "learning_rate": 2.2379971893911144e-06, + "loss": 1.4782, + "step": 736 + }, + { + "epoch": 2.7093023255813953, + "grad_norm": 0.253065288066864, + "learning_rate": 2.122007536596961e-06, + "loss": 1.1246, + "step": 737 + }, + { + "epoch": 2.7131782945736433, + "grad_norm": 0.269283652305603, + "learning_rate": 2.0090846759278064e-06, + "loss": 1.5083, + "step": 738 + }, + { + "epoch": 2.7170542635658914, + "grad_norm": 0.2605230510234833, + "learning_rate": 1.8992307335821763e-06, + "loss": 1.2764, + "step": 739 + }, + { + "epoch": 2.7209302325581395, + "grad_norm": 0.2701490819454193, + "learning_rate": 1.7924477779745368e-06, + "loss": 1.2148, + "step": 740 + }, + { + "epoch": 2.7248062015503876, + "grad_norm": 0.2751643657684326, + "learning_rate": 1.688737819696533e-06, + "loss": 1.3495, + "step": 741 + }, + { + "epoch": 2.7286821705426356, + "grad_norm": 0.25778815150260925, + "learning_rate": 1.5881028114790319e-06, + "loss": 1.3935, + "step": 742 + }, + { + "epoch": 2.7325581395348837, + "grad_norm": 0.26106202602386475, + "learning_rate": 1.4905446481553752e-06, + "loss": 1.3074, + "step": 743 + }, + { + "epoch": 2.7364341085271318, + "grad_norm": 0.30311137437820435, + "learning_rate": 1.3960651666257135e-06, + "loss": 1.5208, + "step": 744 + }, + { + "epoch": 2.74031007751938, + "grad_norm": 0.2914187014102936, + "learning_rate": 1.304666145822383e-06, + "loss": 1.2835, + "step": 745 + }, + { + "epoch": 2.744186046511628, + "grad_norm": 0.2645190358161926, + "learning_rate": 1.2163493066764564e-06, + "loss": 1.4204, + "step": 746 + }, + { + "epoch": 2.748062015503876, + "grad_norm": 0.3073311448097229, + "learning_rate": 1.1311163120853002e-06, + "loss": 1.4899, + "step": 747 + }, + { + "epoch": 2.751937984496124, + "grad_norm": 0.25737103819847107, + "learning_rate": 1.0489687668813048e-06, + "loss": 1.2717, + "step": 748 + }, + { + "epoch": 2.755813953488372, + "grad_norm": 0.27693504095077515, + "learning_rate": 9.6990821780164e-07, + "loss": 1.4203, + "step": 749 + }, + { + "epoch": 2.75968992248062, + "grad_norm": 0.3176767826080322, + "learning_rate": 8.939361534591472e-07, + "loss": 1.4588, + "step": 750 + }, + { + "epoch": 2.7635658914728682, + "grad_norm": 0.34483450651168823, + "learning_rate": 8.21054004314275e-07, + "loss": 1.2022, + "step": 751 + }, + { + "epoch": 2.7674418604651163, + "grad_norm": 0.33260253071784973, + "learning_rate": 7.512631426481869e-07, + "loss": 1.2149, + "step": 752 + }, + { + "epoch": 2.7713178294573644, + "grad_norm": 0.277261346578598, + "learning_rate": 6.845648825369143e-07, + "loss": 1.311, + "step": 753 + }, + { + "epoch": 2.7751937984496124, + "grad_norm": 0.32710519433021545, + "learning_rate": 6.209604798265894e-07, + "loss": 1.5075, + "step": 754 + }, + { + "epoch": 2.7790697674418605, + "grad_norm": 0.2529277205467224, + "learning_rate": 5.604511321098504e-07, + "loss": 1.4405, + "step": 755 + }, + { + "epoch": 2.7829457364341086, + "grad_norm": 0.273580938577652, + "learning_rate": 5.030379787032185e-07, + "loss": 1.3051, + "step": 756 + }, + { + "epoch": 2.7868217054263567, + "grad_norm": 0.2600751519203186, + "learning_rate": 4.487221006257197e-07, + "loss": 1.3619, + "step": 757 + }, + { + "epoch": 2.7906976744186047, + "grad_norm": 0.2721461057662964, + "learning_rate": 3.9750452057847775e-07, + "loss": 1.2126, + "step": 758 + }, + { + "epoch": 2.794573643410853, + "grad_norm": 0.24929043650627136, + "learning_rate": 3.493862029254979e-07, + "loss": 1.4246, + "step": 759 + }, + { + "epoch": 2.798449612403101, + "grad_norm": 0.25558316707611084, + "learning_rate": 3.043680536754767e-07, + "loss": 1.4952, + "step": 760 + }, + { + "epoch": 2.802325581395349, + "grad_norm": 0.3264126181602478, + "learning_rate": 2.624509204647285e-07, + "loss": 1.4849, + "step": 761 + }, + { + "epoch": 2.806201550387597, + "grad_norm": 0.25953492522239685, + "learning_rate": 2.236355925413036e-07, + "loss": 1.2316, + "step": 762 + }, + { + "epoch": 2.810077519379845, + "grad_norm": 0.25779008865356445, + "learning_rate": 1.8792280075005829e-07, + "loss": 1.2583, + "step": 763 + }, + { + "epoch": 2.813953488372093, + "grad_norm": 0.263167142868042, + "learning_rate": 1.5531321751887928e-07, + "loss": 1.507, + "step": 764 + }, + { + "epoch": 2.817829457364341, + "grad_norm": 0.2614395320415497, + "learning_rate": 1.2580745684609872e-07, + "loss": 1.2681, + "step": 765 + }, + { + "epoch": 2.8217054263565893, + "grad_norm": 0.2607451379299164, + "learning_rate": 9.940607428888027e-08, + "loss": 1.3721, + "step": 766 + }, + { + "epoch": 2.8255813953488373, + "grad_norm": 0.27197694778442383, + "learning_rate": 7.610956695275895e-08, + "loss": 1.3152, + "step": 767 + }, + { + "epoch": 2.8294573643410854, + "grad_norm": 0.25525933504104614, + "learning_rate": 5.591837348228046e-08, + "loss": 1.3361, + "step": 768 + }, + { + "epoch": 2.8333333333333335, + "grad_norm": 0.2669144868850708, + "learning_rate": 3.883287405277602e-08, + "loss": 1.2546, + "step": 769 + }, + { + "epoch": 2.8372093023255816, + "grad_norm": 0.2557834982872009, + "learning_rate": 2.4853390363163142e-08, + "loss": 1.1581, + "step": 770 + }, + { + "epoch": 2.8410852713178296, + "grad_norm": 0.26671895384788513, + "learning_rate": 1.398018562993708e-08, + "loss": 1.2647, + "step": 771 + }, + { + "epoch": 2.8449612403100772, + "grad_norm": 0.26321831345558167, + "learning_rate": 6.213464582133232e-09, + "loss": 1.2884, + "step": 772 + }, + { + "epoch": 2.8488372093023253, + "grad_norm": 0.2515983581542969, + "learning_rate": 1.5533734575534641e-09, + "loss": 1.412, + "step": 773 + }, + { + "epoch": 2.8527131782945734, + "grad_norm": 0.29071158170700073, + "learning_rate": 0.0, + "loss": 1.3587, + "step": 774 + } + ], + "logging_steps": 1, + "max_steps": 774, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 258, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 4.6656805872756326e+17, + "train_batch_size": 2, + "trial_name": null, + "trial_params": null +}