| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.64, | |
| "eval_steps": 500, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0064, | |
| "grad_norm": 2.568751573562622, | |
| "learning_rate": 2e-05, | |
| "loss": 2.1237, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0128, | |
| "grad_norm": 2.210172414779663, | |
| "learning_rate": 4e-05, | |
| "loss": 1.9939, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0192, | |
| "grad_norm": 2.24556040763855, | |
| "learning_rate": 6e-05, | |
| "loss": 1.9278, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0256, | |
| "grad_norm": 1.4301820993423462, | |
| "learning_rate": 8e-05, | |
| "loss": 1.4586, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.032, | |
| "grad_norm": 1.2345417737960815, | |
| "learning_rate": 0.0001, | |
| "loss": 1.2717, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0384, | |
| "grad_norm": 1.1178274154663086, | |
| "learning_rate": 0.00012, | |
| "loss": 0.973, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0448, | |
| "grad_norm": 0.8991392850875854, | |
| "learning_rate": 0.00014, | |
| "loss": 0.6798, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0512, | |
| "grad_norm": 1.3577145338058472, | |
| "learning_rate": 0.00016, | |
| "loss": 0.7061, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0576, | |
| "grad_norm": 0.7793099284172058, | |
| "learning_rate": 0.00018, | |
| "loss": 0.4772, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.064, | |
| "grad_norm": 0.4328145682811737, | |
| "learning_rate": 0.0002, | |
| "loss": 0.4894, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0704, | |
| "grad_norm": 0.33954325318336487, | |
| "learning_rate": 0.0001999390827019096, | |
| "loss": 0.4181, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0768, | |
| "grad_norm": 0.26666146516799927, | |
| "learning_rate": 0.00019975640502598244, | |
| "loss": 0.3721, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0832, | |
| "grad_norm": 0.2060694545507431, | |
| "learning_rate": 0.00019945218953682734, | |
| "loss": 0.4174, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0896, | |
| "grad_norm": 0.2562263607978821, | |
| "learning_rate": 0.00019902680687415705, | |
| "loss": 0.4412, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": 0.22718767821788788, | |
| "learning_rate": 0.00019848077530122083, | |
| "loss": 0.3749, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1024, | |
| "grad_norm": 0.3134597837924957, | |
| "learning_rate": 0.00019781476007338058, | |
| "loss": 0.3291, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.1088, | |
| "grad_norm": 0.24586430191993713, | |
| "learning_rate": 0.00019702957262759965, | |
| "loss": 0.2975, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.1152, | |
| "grad_norm": 0.33068349957466125, | |
| "learning_rate": 0.0001961261695938319, | |
| "loss": 0.3777, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.1216, | |
| "grad_norm": 0.18534499406814575, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.302, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.128, | |
| "grad_norm": 0.21168865263462067, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 0.3034, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1344, | |
| "grad_norm": 0.1708526760339737, | |
| "learning_rate": 0.00019271838545667876, | |
| "loss": 0.331, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.1408, | |
| "grad_norm": 0.1510276347398758, | |
| "learning_rate": 0.0001913545457642601, | |
| "loss": 0.3169, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.1472, | |
| "grad_norm": 0.16446498036384583, | |
| "learning_rate": 0.0001898794046299167, | |
| "loss": 0.319, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.1536, | |
| "grad_norm": 0.29390275478363037, | |
| "learning_rate": 0.00018829475928589271, | |
| "loss": 0.2821, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.13030974566936493, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 0.256, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.1664, | |
| "grad_norm": 0.12436527013778687, | |
| "learning_rate": 0.0001848048096156426, | |
| "loss": 0.3123, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.1728, | |
| "grad_norm": 0.16936242580413818, | |
| "learning_rate": 0.00018290375725550417, | |
| "loss": 0.262, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.1792, | |
| "grad_norm": 0.14882561564445496, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.2825, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.1856, | |
| "grad_norm": 0.12063440680503845, | |
| "learning_rate": 0.00017880107536067218, | |
| "loss": 0.2362, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 0.12142356485128403, | |
| "learning_rate": 0.0001766044443118978, | |
| "loss": 0.2745, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1984, | |
| "grad_norm": 0.115916408598423, | |
| "learning_rate": 0.00017431448254773944, | |
| "loss": 0.2727, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.2048, | |
| "grad_norm": 0.11815212666988373, | |
| "learning_rate": 0.0001719339800338651, | |
| "loss": 0.2276, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.2112, | |
| "grad_norm": 0.10431115329265594, | |
| "learning_rate": 0.00016946583704589973, | |
| "loss": 0.2662, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.2176, | |
| "grad_norm": 0.0971333459019661, | |
| "learning_rate": 0.00016691306063588583, | |
| "loss": 0.251, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.224, | |
| "grad_norm": 0.11603528261184692, | |
| "learning_rate": 0.00016427876096865394, | |
| "loss": 0.2405, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.2304, | |
| "grad_norm": 0.10909801721572876, | |
| "learning_rate": 0.0001615661475325658, | |
| "loss": 0.2515, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.2368, | |
| "grad_norm": 0.08744729310274124, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.2748, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.2432, | |
| "grad_norm": 0.10513755679130554, | |
| "learning_rate": 0.0001559192903470747, | |
| "loss": 0.2292, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.2496, | |
| "grad_norm": 0.10443545877933502, | |
| "learning_rate": 0.0001529919264233205, | |
| "loss": 0.2688, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.256, | |
| "grad_norm": 0.10798890888690948, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.2578, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2624, | |
| "grad_norm": 0.10528898984193802, | |
| "learning_rate": 0.00014694715627858908, | |
| "loss": 0.266, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.2688, | |
| "grad_norm": 0.09310728311538696, | |
| "learning_rate": 0.00014383711467890774, | |
| "loss": 0.2059, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.2752, | |
| "grad_norm": 0.09954522550106049, | |
| "learning_rate": 0.00014067366430758004, | |
| "loss": 0.2992, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.2816, | |
| "grad_norm": 0.10648441314697266, | |
| "learning_rate": 0.00013746065934159123, | |
| "loss": 0.259, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 0.12005575746297836, | |
| "learning_rate": 0.00013420201433256689, | |
| "loss": 0.2654, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2944, | |
| "grad_norm": 0.09197583049535751, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.2088, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.3008, | |
| "grad_norm": 0.0905015841126442, | |
| "learning_rate": 0.0001275637355816999, | |
| "loss": 0.2659, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.3072, | |
| "grad_norm": 0.10011676698923111, | |
| "learning_rate": 0.00012419218955996676, | |
| "loss": 0.2573, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.3136, | |
| "grad_norm": 0.11158254742622375, | |
| "learning_rate": 0.00012079116908177593, | |
| "loss": 0.2734, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.10176081955432892, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 0.2386, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3264, | |
| "grad_norm": 0.1110386997461319, | |
| "learning_rate": 0.00011391731009600654, | |
| "loss": 0.2862, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.3328, | |
| "grad_norm": 0.11058996617794037, | |
| "learning_rate": 0.00011045284632676536, | |
| "loss": 0.3214, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.3392, | |
| "grad_norm": 0.09535147994756699, | |
| "learning_rate": 0.00010697564737441252, | |
| "loss": 0.267, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.3456, | |
| "grad_norm": 0.09183581173419952, | |
| "learning_rate": 0.00010348994967025012, | |
| "loss": 0.2545, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.352, | |
| "grad_norm": 0.11743531376123428, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2664, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.3584, | |
| "grad_norm": 0.1046270951628685, | |
| "learning_rate": 9.651005032974994e-05, | |
| "loss": 0.2395, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.3648, | |
| "grad_norm": 0.11286702007055283, | |
| "learning_rate": 9.302435262558747e-05, | |
| "loss": 0.2463, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.3712, | |
| "grad_norm": 0.12298794090747833, | |
| "learning_rate": 8.954715367323468e-05, | |
| "loss": 0.254, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.3776, | |
| "grad_norm": 0.1068972498178482, | |
| "learning_rate": 8.608268990399349e-05, | |
| "loss": 0.2714, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 0.11853048950433731, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 0.2367, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.3904, | |
| "grad_norm": 0.0922948494553566, | |
| "learning_rate": 7.920883091822408e-05, | |
| "loss": 0.2219, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.3968, | |
| "grad_norm": 0.1000463217496872, | |
| "learning_rate": 7.580781044003324e-05, | |
| "loss": 0.2424, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.4032, | |
| "grad_norm": 0.0933864563703537, | |
| "learning_rate": 7.243626441830009e-05, | |
| "loss": 0.2304, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.4096, | |
| "grad_norm": 0.10009445250034332, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.2183, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.416, | |
| "grad_norm": 0.10700756311416626, | |
| "learning_rate": 6.579798566743314e-05, | |
| "loss": 0.2861, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.4224, | |
| "grad_norm": 0.08552579581737518, | |
| "learning_rate": 6.25393406584088e-05, | |
| "loss": 0.2508, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.4288, | |
| "grad_norm": 0.08115057647228241, | |
| "learning_rate": 5.9326335692419995e-05, | |
| "loss": 0.28, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.4352, | |
| "grad_norm": 0.09101542085409164, | |
| "learning_rate": 5.616288532109225e-05, | |
| "loss": 0.2905, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.4416, | |
| "grad_norm": 0.08541923761367798, | |
| "learning_rate": 5.305284372141095e-05, | |
| "loss": 0.2369, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.448, | |
| "grad_norm": 0.08494503796100616, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 0.2535, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.4544, | |
| "grad_norm": 0.09222583472728729, | |
| "learning_rate": 4.700807357667952e-05, | |
| "loss": 0.2164, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.4608, | |
| "grad_norm": 0.09007549285888672, | |
| "learning_rate": 4.4080709652925336e-05, | |
| "loss": 0.2911, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.4672, | |
| "grad_norm": 0.10070206969976425, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.2687, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.4736, | |
| "grad_norm": 0.11217208206653595, | |
| "learning_rate": 3.843385246743417e-05, | |
| "loss": 0.2376, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.0864848643541336, | |
| "learning_rate": 3.5721239031346066e-05, | |
| "loss": 0.2826, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.4864, | |
| "grad_norm": 0.08239398896694183, | |
| "learning_rate": 3.308693936411421e-05, | |
| "loss": 0.2344, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.4928, | |
| "grad_norm": 0.10607580840587616, | |
| "learning_rate": 3.053416295410026e-05, | |
| "loss": 0.272, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.4992, | |
| "grad_norm": 0.11652331054210663, | |
| "learning_rate": 2.8066019966134904e-05, | |
| "loss": 0.2398, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.5056, | |
| "grad_norm": 0.08097366243600845, | |
| "learning_rate": 2.5685517452260567e-05, | |
| "loss": 0.2289, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.512, | |
| "grad_norm": 0.08761543780565262, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 0.2297, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5184, | |
| "grad_norm": 0.09609895944595337, | |
| "learning_rate": 2.119892463932781e-05, | |
| "loss": 0.2214, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.5248, | |
| "grad_norm": 0.10124152898788452, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.2442, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.5312, | |
| "grad_norm": 0.08449672907590866, | |
| "learning_rate": 1.7096242744495837e-05, | |
| "loss": 0.257, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.5376, | |
| "grad_norm": 0.10185690224170685, | |
| "learning_rate": 1.5195190384357404e-05, | |
| "loss": 0.2331, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.544, | |
| "grad_norm": 0.10394009202718735, | |
| "learning_rate": 1.339745962155613e-05, | |
| "loss": 0.2679, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.5504, | |
| "grad_norm": 0.10671041160821915, | |
| "learning_rate": 1.1705240714107302e-05, | |
| "loss": 0.2953, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.5568, | |
| "grad_norm": 0.09869690239429474, | |
| "learning_rate": 1.0120595370083318e-05, | |
| "loss": 0.2363, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.5632, | |
| "grad_norm": 0.10397417098283768, | |
| "learning_rate": 8.645454235739903e-06, | |
| "loss": 0.2221, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.5696, | |
| "grad_norm": 0.10206641256809235, | |
| "learning_rate": 7.281614543321269e-06, | |
| "loss": 0.2329, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 0.09630835801362991, | |
| "learning_rate": 6.030737921409169e-06, | |
| "loss": 0.2193, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5824, | |
| "grad_norm": 0.09056926518678665, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.2507, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.5888, | |
| "grad_norm": 0.09456932544708252, | |
| "learning_rate": 3.873830406168111e-06, | |
| "loss": 0.2413, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.5952, | |
| "grad_norm": 0.10614177584648132, | |
| "learning_rate": 2.970427372400353e-06, | |
| "loss": 0.2294, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.6016, | |
| "grad_norm": 0.11094526201486588, | |
| "learning_rate": 2.1852399266194314e-06, | |
| "loss": 0.2282, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.608, | |
| "grad_norm": 0.10783498734235764, | |
| "learning_rate": 1.5192246987791981e-06, | |
| "loss": 0.2099, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.6144, | |
| "grad_norm": 0.12005529552698135, | |
| "learning_rate": 9.731931258429638e-07, | |
| "loss": 0.2289, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.6208, | |
| "grad_norm": 0.09397509694099426, | |
| "learning_rate": 5.478104631726711e-07, | |
| "loss": 0.2478, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.6272, | |
| "grad_norm": 0.08096156269311905, | |
| "learning_rate": 2.4359497401758024e-07, | |
| "loss": 0.2515, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.6336, | |
| "grad_norm": 0.0987912192940712, | |
| "learning_rate": 6.09172980904238e-08, | |
| "loss": 0.2299, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.10416976362466812, | |
| "learning_rate": 0.0, | |
| "loss": 0.1932, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.92661873012351e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |