| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 96, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03125, | |
| "grad_norm": 0.01407550252465267, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.1816, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0625, | |
| "grad_norm": 0.01470188647122694, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.1815, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.09375, | |
| "grad_norm": 0.013899662077643845, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.1893, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 0.013171363006038193, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.1797, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.15625, | |
| "grad_norm": 0.014122307656636137, | |
| "learning_rate": 4e-05, | |
| "loss": 0.1864, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.1875, | |
| "grad_norm": 0.015493781647247856, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.1863, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.21875, | |
| "grad_norm": 0.0177690135237363, | |
| "learning_rate": 5.6e-05, | |
| "loss": 0.1806, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.019744997443696583, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.1812, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.28125, | |
| "grad_norm": 0.019100076943213144, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.1766, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3125, | |
| "grad_norm": 0.02513285782306868, | |
| "learning_rate": 8e-05, | |
| "loss": 0.1741, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.34375, | |
| "grad_norm": 0.023047357881927678, | |
| "learning_rate": 7.997331393480957e-05, | |
| "loss": 0.1615, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 0.023124231297050782, | |
| "learning_rate": 7.989329134654207e-05, | |
| "loss": 0.1413, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.40625, | |
| "grad_norm": 0.028010735700580793, | |
| "learning_rate": 7.976003900959785e-05, | |
| "loss": 0.1505, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.4375, | |
| "grad_norm": 0.02529939154236208, | |
| "learning_rate": 7.957373472300442e-05, | |
| "loss": 0.1429, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.46875, | |
| "grad_norm": 0.018177099368165615, | |
| "learning_rate": 7.933462707317864e-05, | |
| "loss": 0.1431, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.02313399335351697, | |
| "learning_rate": 7.90430351022371e-05, | |
| "loss": 0.1378, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.53125, | |
| "grad_norm": 0.01597311427695637, | |
| "learning_rate": 7.869934788229701e-05, | |
| "loss": 0.1242, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.5625, | |
| "grad_norm": 0.01859669819463546, | |
| "learning_rate": 7.830402399633624e-05, | |
| "loss": 0.125, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.59375, | |
| "grad_norm": 0.014162630783732186, | |
| "learning_rate": 7.785759092630437e-05, | |
| "loss": 0.1095, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 0.016655557233615195, | |
| "learning_rate": 7.736064434930193e-05, | |
| "loss": 0.1264, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.65625, | |
| "grad_norm": 0.015749717663925782, | |
| "learning_rate": 7.681384734276638e-05, | |
| "loss": 0.1123, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.6875, | |
| "grad_norm": 0.014072349788015267, | |
| "learning_rate": 7.621792949972588e-05, | |
| "loss": 0.1091, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.71875, | |
| "grad_norm": 0.014078522807825882, | |
| "learning_rate": 7.557368595530076e-05, | |
| "loss": 0.1038, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.0136967484140524, | |
| "learning_rate": 7.488197632575232e-05, | |
| "loss": 0.1071, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.78125, | |
| "grad_norm": 0.012490712936852387, | |
| "learning_rate": 7.414372356149387e-05, | |
| "loss": 0.0974, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.8125, | |
| "grad_norm": 0.012924580642094627, | |
| "learning_rate": 7.335991271559512e-05, | |
| "loss": 0.0914, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.84375, | |
| "grad_norm": 0.013089434696583035, | |
| "learning_rate": 7.253158962942263e-05, | |
| "loss": 0.0891, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.875, | |
| "grad_norm": 0.015993464382911503, | |
| "learning_rate": 7.165985953717017e-05, | |
| "loss": 0.0857, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.90625, | |
| "grad_norm": 0.014430703246442201, | |
| "learning_rate": 7.074588559114129e-05, | |
| "loss": 0.0815, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.9375, | |
| "grad_norm": 0.012559712427520731, | |
| "learning_rate": 6.979088730975128e-05, | |
| "loss": 0.0763, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.96875, | |
| "grad_norm": 0.012436854650590573, | |
| "learning_rate": 6.879613895031985e-05, | |
| "loss": 0.0714, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.01194365549567003, | |
| "learning_rate": 6.776296780882537e-05, | |
| "loss": 0.0688, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.03125, | |
| "grad_norm": 0.013676736512650316, | |
| "learning_rate": 6.669275244888958e-05, | |
| "loss": 0.0677, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.0625, | |
| "grad_norm": 0.012728477598631795, | |
| "learning_rate": 6.558692086235565e-05, | |
| "loss": 0.0626, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.09375, | |
| "grad_norm": 0.009181957180459447, | |
| "learning_rate": 6.444694856391398e-05, | |
| "loss": 0.0629, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.125, | |
| "grad_norm": 0.011241352188131493, | |
| "learning_rate": 6.327435662231812e-05, | |
| "loss": 0.0574, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.15625, | |
| "grad_norm": 0.011814039337883688, | |
| "learning_rate": 6.207070963081785e-05, | |
| "loss": 0.0555, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.1875, | |
| "grad_norm": 0.010712670156432241, | |
| "learning_rate": 6.083761361951722e-05, | |
| "loss": 0.0501, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.21875, | |
| "grad_norm": 0.012147779459752806, | |
| "learning_rate": 5.9576713912443424e-05, | |
| "loss": 0.0592, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.009542528460113579, | |
| "learning_rate": 5.8289692932185546e-05, | |
| "loss": 0.0526, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.28125, | |
| "grad_norm": 0.012101717968366866, | |
| "learning_rate": 5.697826795503261e-05, | |
| "loss": 0.0531, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.3125, | |
| "grad_norm": 0.011041723261196693, | |
| "learning_rate": 5.564418881960624e-05, | |
| "loss": 0.0562, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.34375, | |
| "grad_norm": 0.009312163807652834, | |
| "learning_rate": 5.428923559204531e-05, | |
| "loss": 0.0461, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.375, | |
| "grad_norm": 0.008288647576096984, | |
| "learning_rate": 5.291521619085785e-05, | |
| "loss": 0.0398, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.40625, | |
| "grad_norm": 0.0085603182551657, | |
| "learning_rate": 5.1523963974609515e-05, | |
| "loss": 0.0484, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.4375, | |
| "grad_norm": 0.008592725056521243, | |
| "learning_rate": 5.011733529566723e-05, | |
| "loss": 0.0433, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.46875, | |
| "grad_norm": 0.01107345110355633, | |
| "learning_rate": 4.869720702326229e-05, | |
| "loss": 0.0506, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.008335174087170924, | |
| "learning_rate": 4.726547403917746e-05, | |
| "loss": 0.051, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.53125, | |
| "grad_norm": 0.007423496556631077, | |
| "learning_rate": 4.582404670940021e-05, | |
| "loss": 0.0407, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.5625, | |
| "grad_norm": 0.007723363601397683, | |
| "learning_rate": 4.437484833511499e-05, | |
| "loss": 0.0491, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.59375, | |
| "grad_norm": 0.007957016981768567, | |
| "learning_rate": 4.29198125864363e-05, | |
| "loss": 0.0454, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.625, | |
| "grad_norm": 0.007979007811475326, | |
| "learning_rate": 4.1460880922306367e-05, | |
| "loss": 0.0405, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.65625, | |
| "grad_norm": 0.006641760340963778, | |
| "learning_rate": 4e-05, | |
| "loss": 0.045, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.6875, | |
| "grad_norm": 0.007772111084672135, | |
| "learning_rate": 3.853911907769365e-05, | |
| "loss": 0.0407, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.71875, | |
| "grad_norm": 0.007968572731780869, | |
| "learning_rate": 3.7080187413563696e-05, | |
| "loss": 0.0467, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 0.006331316783310332, | |
| "learning_rate": 3.5625151664885036e-05, | |
| "loss": 0.0415, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.78125, | |
| "grad_norm": 0.007565679805810291, | |
| "learning_rate": 3.417595329059982e-05, | |
| "loss": 0.0434, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.8125, | |
| "grad_norm": 0.007508804813033004, | |
| "learning_rate": 3.2734525960822545e-05, | |
| "loss": 0.0499, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.84375, | |
| "grad_norm": 0.007125337161851334, | |
| "learning_rate": 3.1302792976737726e-05, | |
| "loss": 0.0436, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 0.008750891670532251, | |
| "learning_rate": 2.988266470433277e-05, | |
| "loss": 0.0448, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.90625, | |
| "grad_norm": 0.010030347406714348, | |
| "learning_rate": 2.84760360253905e-05, | |
| "loss": 0.0512, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.9375, | |
| "grad_norm": 0.012741841327458828, | |
| "learning_rate": 2.7084783809142164e-05, | |
| "loss": 0.0559, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.96875, | |
| "grad_norm": 0.007052409575339403, | |
| "learning_rate": 2.5710764407954692e-05, | |
| "loss": 0.0446, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.0076933296048535904, | |
| "learning_rate": 2.4355811180393767e-05, | |
| "loss": 0.0432, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.03125, | |
| "grad_norm": 0.006677444020767099, | |
| "learning_rate": 2.3021732044967405e-05, | |
| "loss": 0.0388, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.0625, | |
| "grad_norm": 0.006644011597457178, | |
| "learning_rate": 2.171030706781446e-05, | |
| "loss": 0.038, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.09375, | |
| "grad_norm": 0.008273196829902876, | |
| "learning_rate": 2.042328608755659e-05, | |
| "loss": 0.0523, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.125, | |
| "grad_norm": 0.008656212729561663, | |
| "learning_rate": 1.9162386380482795e-05, | |
| "loss": 0.0428, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.15625, | |
| "grad_norm": 0.007907595953750495, | |
| "learning_rate": 1.7929290369182163e-05, | |
| "loss": 0.0421, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.1875, | |
| "grad_norm": 0.006179264059214988, | |
| "learning_rate": 1.6725643377681893e-05, | |
| "loss": 0.0411, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.21875, | |
| "grad_norm": 0.007796852540554055, | |
| "learning_rate": 1.555305143608603e-05, | |
| "loss": 0.0497, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 0.007300626672068782, | |
| "learning_rate": 1.4413079137644358e-05, | |
| "loss": 0.0444, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.28125, | |
| "grad_norm": 0.0075306917853387246, | |
| "learning_rate": 1.3307247551110427e-05, | |
| "loss": 0.0546, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.3125, | |
| "grad_norm": 0.006713948760641831, | |
| "learning_rate": 1.2237032191174642e-05, | |
| "loss": 0.0469, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.34375, | |
| "grad_norm": 0.008286233506446512, | |
| "learning_rate": 1.1203861049680174e-05, | |
| "loss": 0.04, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.375, | |
| "grad_norm": 0.0062607944469658785, | |
| "learning_rate": 1.0209112690248726e-05, | |
| "loss": 0.0307, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.40625, | |
| "grad_norm": 0.006330406280359023, | |
| "learning_rate": 9.254114408858714e-06, | |
| "loss": 0.037, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.4375, | |
| "grad_norm": 0.007069131022686873, | |
| "learning_rate": 8.34014046282984e-06, | |
| "loss": 0.0366, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.46875, | |
| "grad_norm": 0.005695617439896423, | |
| "learning_rate": 7.468410370577386e-06, | |
| "loss": 0.0388, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.007372692106164091, | |
| "learning_rate": 6.640087284404888e-06, | |
| "loss": 0.0484, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.53125, | |
| "grad_norm": 0.00751008320860994, | |
| "learning_rate": 5.856276438506143e-06, | |
| "loss": 0.0465, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.5625, | |
| "grad_norm": 0.007448804961198022, | |
| "learning_rate": 5.118023674247692e-06, | |
| "loss": 0.0476, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 2.59375, | |
| "grad_norm": 0.007958166860826373, | |
| "learning_rate": 4.426314044699247e-06, | |
| "loss": 0.0447, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 2.625, | |
| "grad_norm": 0.007892530550195764, | |
| "learning_rate": 3.7820705002741353e-06, | |
| "loss": 0.0464, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.65625, | |
| "grad_norm": 0.007452974291357488, | |
| "learning_rate": 3.1861526572336276e-06, | |
| "loss": 0.0475, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.6875, | |
| "grad_norm": 0.007568847705200401, | |
| "learning_rate": 2.6393556506980834e-06, | |
| "loss": 0.043, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 2.71875, | |
| "grad_norm": 0.006353940889517695, | |
| "learning_rate": 2.142409073695624e-06, | |
| "loss": 0.0405, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 0.00896052920211785, | |
| "learning_rate": 1.6959760036637662e-06, | |
| "loss": 0.0478, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 2.78125, | |
| "grad_norm": 0.006857728005209591, | |
| "learning_rate": 1.3006521177029918e-06, | |
| "loss": 0.0404, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 2.8125, | |
| "grad_norm": 0.0063836304247783, | |
| "learning_rate": 9.569648977629176e-07, | |
| "loss": 0.0387, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.84375, | |
| "grad_norm": 0.007680634468205475, | |
| "learning_rate": 6.653729268213571e-07, | |
| "loss": 0.0446, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 2.875, | |
| "grad_norm": 0.006757432100802972, | |
| "learning_rate": 4.2626527699558996e-07, | |
| "loss": 0.0412, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 2.90625, | |
| "grad_norm": 0.00652382645557434, | |
| "learning_rate": 2.399609904021638e-07, | |
| "loss": 0.0478, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 2.9375, | |
| "grad_norm": 0.00655177687870289, | |
| "learning_rate": 1.0670865345793425e-07, | |
| "loss": 0.0458, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 2.96875, | |
| "grad_norm": 0.006766581973263856, | |
| "learning_rate": 2.668606519042438e-08, | |
| "loss": 0.0433, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.006522064241774668, | |
| "learning_rate": 0.0, | |
| "loss": 0.041, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 96, | |
| "total_flos": 79623079919616.0, | |
| "train_loss": 0.075466869204926, | |
| "train_runtime": 1447.2969, | |
| "train_samples_per_second": 0.518, | |
| "train_steps_per_second": 0.066 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 96, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 79623079919616.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |