| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.732620320855615, | |
| "eval_steps": 12, | |
| "global_step": 138, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0213903743315508, | |
| "grad_norm": 23.385425567626953, | |
| "learning_rate": 2e-05, | |
| "loss": 2.2375, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0213903743315508, | |
| "eval_loss": 2.295367479324341, | |
| "eval_runtime": 16.9622, | |
| "eval_samples_per_second": 17.686, | |
| "eval_steps_per_second": 8.843, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0427807486631016, | |
| "grad_norm": 31.645936965942383, | |
| "learning_rate": 4e-05, | |
| "loss": 2.3841, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.06417112299465241, | |
| "grad_norm": 34.55805206298828, | |
| "learning_rate": 6e-05, | |
| "loss": 2.6477, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0855614973262032, | |
| "grad_norm": 32.25712966918945, | |
| "learning_rate": 8e-05, | |
| "loss": 1.9082, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.10695187165775401, | |
| "grad_norm": 25.384239196777344, | |
| "learning_rate": 0.0001, | |
| "loss": 1.5055, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.12834224598930483, | |
| "grad_norm": 22.698801040649414, | |
| "learning_rate": 0.00012, | |
| "loss": 1.237, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.1497326203208556, | |
| "grad_norm": 25.194936752319336, | |
| "learning_rate": 0.00014, | |
| "loss": 1.1552, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.1711229946524064, | |
| "grad_norm": 17.266931533813477, | |
| "learning_rate": 0.00016, | |
| "loss": 0.9119, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.1925133689839572, | |
| "grad_norm": 15.346185684204102, | |
| "learning_rate": 0.00018, | |
| "loss": 0.834, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.21390374331550802, | |
| "grad_norm": 26.989032745361328, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7533, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.23529411764705882, | |
| "grad_norm": 11.042716979980469, | |
| "learning_rate": 0.00019998370105646414, | |
| "loss": 0.6231, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.25668449197860965, | |
| "grad_norm": 8.389060020446777, | |
| "learning_rate": 0.0001999348095389677, | |
| "loss": 0.4795, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.25668449197860965, | |
| "eval_loss": 0.612250566482544, | |
| "eval_runtime": 17.2275, | |
| "eval_samples_per_second": 17.414, | |
| "eval_steps_per_second": 8.707, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.27807486631016043, | |
| "grad_norm": 32.6131477355957, | |
| "learning_rate": 0.00019985334138511237, | |
| "loss": 0.6842, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.2994652406417112, | |
| "grad_norm": 35.244876861572266, | |
| "learning_rate": 0.000199739323151795, | |
| "loss": 1.0486, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.32085561497326204, | |
| "grad_norm": 14.032343864440918, | |
| "learning_rate": 0.00019959279200655044, | |
| "loss": 0.9261, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.3422459893048128, | |
| "grad_norm": 14.313758850097656, | |
| "learning_rate": 0.00019941379571543596, | |
| "loss": 0.811, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 14.679015159606934, | |
| "learning_rate": 0.00019920239262746043, | |
| "loss": 0.7831, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.3850267379679144, | |
| "grad_norm": 14.768975257873535, | |
| "learning_rate": 0.00019895865165556377, | |
| "loss": 0.5663, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.40641711229946526, | |
| "grad_norm": 16.57910919189453, | |
| "learning_rate": 0.00019868265225415265, | |
| "loss": 0.7238, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.42780748663101603, | |
| "grad_norm": 11.922738075256348, | |
| "learning_rate": 0.00019837448439320027, | |
| "loss": 0.6445, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.44919786096256686, | |
| "grad_norm": 17.637170791625977, | |
| "learning_rate": 0.00019803424852891802, | |
| "loss": 1.2175, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "grad_norm": 9.833541870117188, | |
| "learning_rate": 0.00019766205557100868, | |
| "loss": 0.8675, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.4919786096256685, | |
| "grad_norm": 11.809141159057617, | |
| "learning_rate": 0.00019725802684651233, | |
| "loss": 1.1414, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.5133689839572193, | |
| "grad_norm": 18.123613357543945, | |
| "learning_rate": 0.00019682229406025635, | |
| "loss": 0.9939, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5133689839572193, | |
| "eval_loss": 0.6862085461616516, | |
| "eval_runtime": 17.2196, | |
| "eval_samples_per_second": 17.422, | |
| "eval_steps_per_second": 8.711, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5347593582887701, | |
| "grad_norm": 11.453226089477539, | |
| "learning_rate": 0.0001963549992519223, | |
| "loss": 0.8239, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.5561497326203209, | |
| "grad_norm": 7.978896617889404, | |
| "learning_rate": 0.00019585629474974415, | |
| "loss": 0.5823, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.5775401069518716, | |
| "grad_norm": 7.980404376983643, | |
| "learning_rate": 0.0001953263431208523, | |
| "loss": 0.6231, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.5989304812834224, | |
| "grad_norm": 12.718688011169434, | |
| "learning_rate": 0.00019476531711828027, | |
| "loss": 0.6814, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.6203208556149733, | |
| "grad_norm": 11.378684997558594, | |
| "learning_rate": 0.00019417339962465082, | |
| "loss": 0.8471, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.6417112299465241, | |
| "grad_norm": 10.157166481018066, | |
| "learning_rate": 0.0001935507835925601, | |
| "loss": 0.6058, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.6631016042780749, | |
| "grad_norm": 10.623138427734375, | |
| "learning_rate": 0.00019289767198167916, | |
| "loss": 0.81, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.6844919786096256, | |
| "grad_norm": 16.875181198120117, | |
| "learning_rate": 0.00019221427769259333, | |
| "loss": 0.8037, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 11.729046821594238, | |
| "learning_rate": 0.0001915008234974012, | |
| "loss": 0.5548, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 10.128629684448242, | |
| "learning_rate": 0.00019075754196709572, | |
| "loss": 0.6979, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.7486631016042781, | |
| "grad_norm": 10.893034934997559, | |
| "learning_rate": 0.0001899846753957507, | |
| "loss": 0.6173, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.7700534759358288, | |
| "grad_norm": 11.063718795776367, | |
| "learning_rate": 0.00018918247572153823, | |
| "loss": 0.6885, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.7700534759358288, | |
| "eval_loss": 0.6894535422325134, | |
| "eval_runtime": 16.9719, | |
| "eval_samples_per_second": 17.676, | |
| "eval_steps_per_second": 8.838, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.7914438502673797, | |
| "grad_norm": 13.273902893066406, | |
| "learning_rate": 0.0001883512044446023, | |
| "loss": 0.8057, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.8128342245989305, | |
| "grad_norm": 17.475479125976562, | |
| "learning_rate": 0.00018749113254181498, | |
| "loss": 0.8596, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.8342245989304813, | |
| "grad_norm": 13.216853141784668, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 0.7885, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.8556149732620321, | |
| "grad_norm": 11.209785461425781, | |
| "learning_rate": 0.00018568571761675893, | |
| "loss": 0.7028, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.8770053475935828, | |
| "grad_norm": 8.61646556854248, | |
| "learning_rate": 0.00018474096312160864, | |
| "loss": 0.5775, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.8983957219251337, | |
| "grad_norm": 10.86780071258545, | |
| "learning_rate": 0.00018376858486299647, | |
| "loss": 0.8106, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.9197860962566845, | |
| "grad_norm": 6.3408427238464355, | |
| "learning_rate": 0.00018276889981568906, | |
| "loss": 0.4089, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "grad_norm": 19.630786895751953, | |
| "learning_rate": 0.00018174223385588917, | |
| "loss": 1.1041, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.9625668449197861, | |
| "grad_norm": 10.352812767028809, | |
| "learning_rate": 0.00018068892165500704, | |
| "loss": 0.749, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.983957219251337, | |
| "grad_norm": 8.671784400939941, | |
| "learning_rate": 0.00017960930657056438, | |
| "loss": 0.6975, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.0053475935828877, | |
| "grad_norm": 10.529078483581543, | |
| "learning_rate": 0.00017850374053426723, | |
| "loss": 0.5922, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.0267379679144386, | |
| "grad_norm": 9.373391151428223, | |
| "learning_rate": 0.00017737258393728364, | |
| "loss": 0.729, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.0267379679144386, | |
| "eval_loss": 0.6926581859588623, | |
| "eval_runtime": 17.8687, | |
| "eval_samples_per_second": 16.789, | |
| "eval_steps_per_second": 8.395, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.0481283422459893, | |
| "grad_norm": 18.76338005065918, | |
| "learning_rate": 0.00017621620551276366, | |
| "loss": 0.9321, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.0695187165775402, | |
| "grad_norm": 7.443489074707031, | |
| "learning_rate": 0.00017503498221564025, | |
| "loss": 0.5063, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 8.333600044250488, | |
| "learning_rate": 0.00017382929909974987, | |
| "loss": 0.4377, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.1122994652406417, | |
| "grad_norm": 14.056548118591309, | |
| "learning_rate": 0.0001725995491923131, | |
| "loss": 1.2233, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.0213903743315509, | |
| "grad_norm": 13.999143600463867, | |
| "learning_rate": 0.00017134613336581599, | |
| "loss": 0.3174, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.0427807486631016, | |
| "grad_norm": 10.154765129089355, | |
| "learning_rate": 0.00017006946020733425, | |
| "loss": 0.162, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.0641711229946524, | |
| "grad_norm": 4.229214191436768, | |
| "learning_rate": 0.00016876994588534234, | |
| "loss": 0.1357, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.085561497326203, | |
| "grad_norm": 4.834641456604004, | |
| "learning_rate": 0.0001674480140140514, | |
| "loss": 0.1686, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.106951871657754, | |
| "grad_norm": 8.32540225982666, | |
| "learning_rate": 0.00016610409551532005, | |
| "loss": 0.2455, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.1283422459893049, | |
| "grad_norm": 5.407528400421143, | |
| "learning_rate": 0.00016473862847818277, | |
| "loss": 0.1116, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.1497326203208555, | |
| "grad_norm": 5.010147571563721, | |
| "learning_rate": 0.0001633520580160424, | |
| "loss": 0.1156, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.1711229946524064, | |
| "grad_norm": 6.022050380706787, | |
| "learning_rate": 0.0001619448361215723, | |
| "loss": 0.225, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.1711229946524064, | |
| "eval_loss": 0.7360826134681702, | |
| "eval_runtime": 17.9845, | |
| "eval_samples_per_second": 16.681, | |
| "eval_steps_per_second": 8.34, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.192513368983957, | |
| "grad_norm": 5.873435974121094, | |
| "learning_rate": 0.00016051742151937655, | |
| "loss": 0.0572, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.213903743315508, | |
| "grad_norm": 10.556258201599121, | |
| "learning_rate": 0.0001590702795164551, | |
| "loss": 0.1261, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.2352941176470589, | |
| "grad_norm": 17.438634872436523, | |
| "learning_rate": 0.00015760388185052398, | |
| "loss": 0.382, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.2566844919786098, | |
| "grad_norm": 8.371882438659668, | |
| "learning_rate": 0.00015611870653623825, | |
| "loss": 0.228, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.2780748663101604, | |
| "grad_norm": 9.435052871704102, | |
| "learning_rate": 0.0001546152377093697, | |
| "loss": 0.1978, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.299465240641711, | |
| "grad_norm": 8.652084350585938, | |
| "learning_rate": 0.0001530939654689887, | |
| "loss": 0.1587, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.320855614973262, | |
| "grad_norm": 8.71427059173584, | |
| "learning_rate": 0.00015155538571770218, | |
| "loss": 0.0637, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.3422459893048129, | |
| "grad_norm": 5.761900901794434, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.2027, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.3636363636363638, | |
| "grad_norm": 8.064814567565918, | |
| "learning_rate": 0.00014842831533876195, | |
| "loss": 0.1595, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.3850267379679144, | |
| "grad_norm": 7.962220668792725, | |
| "learning_rate": 0.00014684084406997903, | |
| "loss": 0.3658, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.4064171122994653, | |
| "grad_norm": 4.9641242027282715, | |
| "learning_rate": 0.00014523810367574272, | |
| "loss": 0.1368, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 1.427807486631016, | |
| "grad_norm": 7.027231216430664, | |
| "learning_rate": 0.00014362061661555675, | |
| "loss": 0.1688, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.427807486631016, | |
| "eval_loss": 0.8078603148460388, | |
| "eval_runtime": 17.7522, | |
| "eval_samples_per_second": 16.899, | |
| "eval_steps_per_second": 8.45, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.4491978609625669, | |
| "grad_norm": 3.6481807231903076, | |
| "learning_rate": 0.00014198891015602646, | |
| "loss": 0.0844, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 1.4705882352941178, | |
| "grad_norm": 7.879701614379883, | |
| "learning_rate": 0.00014034351619898088, | |
| "loss": 0.3332, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 1.4919786096256684, | |
| "grad_norm": 5.169126987457275, | |
| "learning_rate": 0.00013868497110808395, | |
| "loss": 0.1301, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.5133689839572193, | |
| "grad_norm": 5.538336753845215, | |
| "learning_rate": 0.00013701381553399145, | |
| "loss": 0.145, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.53475935828877, | |
| "grad_norm": 25.778339385986328, | |
| "learning_rate": 0.00013533059423811026, | |
| "loss": 0.2704, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 1.5561497326203209, | |
| "grad_norm": 5.043937683105469, | |
| "learning_rate": 0.0001336358559150175, | |
| "loss": 0.1458, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.5775401069518717, | |
| "grad_norm": 7.204008102416992, | |
| "learning_rate": 0.000131930153013598, | |
| "loss": 0.1012, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 1.5989304812834224, | |
| "grad_norm": 8.367019653320312, | |
| "learning_rate": 0.00013021404155695725, | |
| "loss": 0.1889, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.6203208556149733, | |
| "grad_norm": 5.196023464202881, | |
| "learning_rate": 0.00012848808096117, | |
| "loss": 0.0881, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.641711229946524, | |
| "grad_norm": 8.067776679992676, | |
| "learning_rate": 0.00012675283385292212, | |
| "loss": 0.2948, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 1.6631016042780749, | |
| "grad_norm": 8.772676467895508, | |
| "learning_rate": 0.0001250088658861063, | |
| "loss": 0.154, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 1.6844919786096257, | |
| "grad_norm": 16.28201675415039, | |
| "learning_rate": 0.00012325674555743106, | |
| "loss": 0.3085, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.6844919786096257, | |
| "eval_loss": 0.780529797077179, | |
| "eval_runtime": 17.4078, | |
| "eval_samples_per_second": 17.234, | |
| "eval_steps_per_second": 8.617, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.7058823529411766, | |
| "grad_norm": 8.888002395629883, | |
| "learning_rate": 0.00012149704402110243, | |
| "loss": 0.2488, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.7272727272727273, | |
| "grad_norm": 6.574310779571533, | |
| "learning_rate": 0.00011973033490264001, | |
| "loss": 0.1698, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 1.748663101604278, | |
| "grad_norm": 10.671408653259277, | |
| "learning_rate": 0.00011795719411188718, | |
| "loss": 0.1778, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.7700534759358288, | |
| "grad_norm": 6.570982933044434, | |
| "learning_rate": 0.0001161781996552765, | |
| "loss": 0.115, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.7914438502673797, | |
| "grad_norm": 11.171520233154297, | |
| "learning_rate": 0.0001143939314474119, | |
| "loss": 0.2032, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 1.8128342245989306, | |
| "grad_norm": 4.231898307800293, | |
| "learning_rate": 0.00011260497112202895, | |
| "loss": 0.0596, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.8342245989304813, | |
| "grad_norm": 5.200542449951172, | |
| "learning_rate": 0.00011081190184239419, | |
| "loss": 0.1294, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 1.855614973262032, | |
| "grad_norm": 6.871387958526611, | |
| "learning_rate": 0.00010901530811120655, | |
| "loss": 0.283, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 1.8770053475935828, | |
| "grad_norm": 4.964606285095215, | |
| "learning_rate": 0.00010721577558006164, | |
| "loss": 0.2673, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.8983957219251337, | |
| "grad_norm": 8.798148155212402, | |
| "learning_rate": 0.00010541389085854176, | |
| "loss": 0.2055, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 1.9197860962566846, | |
| "grad_norm": 5.59722375869751, | |
| "learning_rate": 0.00010361024132299364, | |
| "loss": 0.209, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.9411764705882353, | |
| "grad_norm": 8.256745338439941, | |
| "learning_rate": 0.00010180541492505604, | |
| "loss": 0.1079, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.9411764705882353, | |
| "eval_loss": 0.7684900760650635, | |
| "eval_runtime": 17.746, | |
| "eval_samples_per_second": 16.905, | |
| "eval_steps_per_second": 8.453, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.962566844919786, | |
| "grad_norm": 4.481995582580566, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1824, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 1.9839572192513368, | |
| "grad_norm": 3.852792263031006, | |
| "learning_rate": 9.819458507494394e-05, | |
| "loss": 0.124, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 2.0053475935828877, | |
| "grad_norm": 29.360998153686523, | |
| "learning_rate": 9.638975867700638e-05, | |
| "loss": 0.1844, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.0267379679144386, | |
| "grad_norm": 4.80811071395874, | |
| "learning_rate": 9.458610914145826e-05, | |
| "loss": 0.1347, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.048128342245989, | |
| "grad_norm": 17.53403091430664, | |
| "learning_rate": 9.27842244199384e-05, | |
| "loss": 0.3378, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 2.06951871657754, | |
| "grad_norm": 7.34214973449707, | |
| "learning_rate": 9.098469188879349e-05, | |
| "loss": 0.2045, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 2.090909090909091, | |
| "grad_norm": 4.968944072723389, | |
| "learning_rate": 8.918809815760585e-05, | |
| "loss": 0.2284, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 2.0053475935828877, | |
| "grad_norm": 4.202042579650879, | |
| "learning_rate": 8.739502887797107e-05, | |
| "loss": 0.1655, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 2.0267379679144386, | |
| "grad_norm": 3.561790704727173, | |
| "learning_rate": 8.560606855258808e-05, | |
| "loss": 0.0442, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.0481283422459895, | |
| "grad_norm": 3.8292624950408936, | |
| "learning_rate": 8.382180034472353e-05, | |
| "loss": 0.0821, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 2.06951871657754, | |
| "grad_norm": 2.1650640964508057, | |
| "learning_rate": 8.204280588811283e-05, | |
| "loss": 0.0384, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 2.090909090909091, | |
| "grad_norm": 1.6922334432601929, | |
| "learning_rate": 8.026966509736001e-05, | |
| "loss": 0.0342, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 2.090909090909091, | |
| "eval_loss": 0.7716657519340515, | |
| "eval_runtime": 17.9156, | |
| "eval_samples_per_second": 16.745, | |
| "eval_steps_per_second": 8.373, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 2.1122994652406417, | |
| "grad_norm": 1.0654356479644775, | |
| "learning_rate": 7.85029559788976e-05, | |
| "loss": 0.0184, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 2.1336898395721926, | |
| "grad_norm": 3.1057019233703613, | |
| "learning_rate": 7.674325444256899e-05, | |
| "loss": 0.0417, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.1550802139037435, | |
| "grad_norm": 0.19042205810546875, | |
| "learning_rate": 7.499113411389371e-05, | |
| "loss": 0.0026, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 2.176470588235294, | |
| "grad_norm": 1.5116851329803467, | |
| "learning_rate": 7.324716614707793e-05, | |
| "loss": 0.0089, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 2.197860962566845, | |
| "grad_norm": 2.5151679515838623, | |
| "learning_rate": 7.151191903883001e-05, | |
| "loss": 0.0357, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 2.2192513368983957, | |
| "grad_norm": 2.838503837585449, | |
| "learning_rate": 6.978595844304271e-05, | |
| "loss": 0.0366, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 2.2406417112299466, | |
| "grad_norm": 3.835000514984131, | |
| "learning_rate": 6.806984698640202e-05, | |
| "loss": 0.1412, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 2.2620320855614975, | |
| "grad_norm": 3.4443538188934326, | |
| "learning_rate": 6.636414408498249e-05, | |
| "loss": 0.0707, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 2.283422459893048, | |
| "grad_norm": 2.701524496078491, | |
| "learning_rate": 6.466940576188977e-05, | |
| "loss": 0.0497, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 2.304812834224599, | |
| "grad_norm": 2.612593412399292, | |
| "learning_rate": 6.298618446600856e-05, | |
| "loss": 0.052, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 2.3262032085561497, | |
| "grad_norm": 4.986962795257568, | |
| "learning_rate": 6.13150288919161e-05, | |
| "loss": 0.1255, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 2.3475935828877006, | |
| "grad_norm": 1.8598374128341675, | |
| "learning_rate": 5.965648380101916e-05, | |
| "loss": 0.0309, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.3475935828877006, | |
| "eval_loss": 0.785007119178772, | |
| "eval_runtime": 17.7923, | |
| "eval_samples_per_second": 16.861, | |
| "eval_steps_per_second": 8.431, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.3689839572192515, | |
| "grad_norm": 1.5813214778900146, | |
| "learning_rate": 5.801108984397354e-05, | |
| "loss": 0.0201, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 2.3903743315508024, | |
| "grad_norm": 0.13843385875225067, | |
| "learning_rate": 5.6379383384443255e-05, | |
| "loss": 0.0018, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.411764705882353, | |
| "grad_norm": 4.4155707359313965, | |
| "learning_rate": 5.476189632425732e-05, | |
| "loss": 0.0326, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 2.4331550802139037, | |
| "grad_norm": 3.5101325511932373, | |
| "learning_rate": 5.3159155930021e-05, | |
| "loss": 0.0259, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 2.4545454545454546, | |
| "grad_norm": 5.201532363891602, | |
| "learning_rate": 5.1571684661238075e-05, | |
| "loss": 0.0761, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 2.4759358288770055, | |
| "grad_norm": 2.48543119430542, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 0.0587, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 2.497326203208556, | |
| "grad_norm": 7.39755916595459, | |
| "learning_rate": 4.844461428229782e-05, | |
| "loss": 0.0391, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 2.518716577540107, | |
| "grad_norm": 4.151485443115234, | |
| "learning_rate": 4.6906034531011346e-05, | |
| "loss": 0.0982, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 2.5401069518716577, | |
| "grad_norm": 4.144845485687256, | |
| "learning_rate": 4.53847622906303e-05, | |
| "loss": 0.0707, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 2.5614973262032086, | |
| "grad_norm": 7.3682732582092285, | |
| "learning_rate": 4.388129346376178e-05, | |
| "loss": 0.0455, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.5828877005347595, | |
| "grad_norm": 4.947929382324219, | |
| "learning_rate": 4.239611814947605e-05, | |
| "loss": 0.033, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 2.6042780748663104, | |
| "grad_norm": 3.0208606719970703, | |
| "learning_rate": 4.092972048354491e-05, | |
| "loss": 0.0373, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 2.6042780748663104, | |
| "eval_loss": 0.776565432548523, | |
| "eval_runtime": 17.3019, | |
| "eval_samples_per_second": 17.339, | |
| "eval_steps_per_second": 8.67, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 2.625668449197861, | |
| "grad_norm": 7.514610290527344, | |
| "learning_rate": 3.948257848062351e-05, | |
| "loss": 0.0323, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 2.6470588235294117, | |
| "grad_norm": 1.8352607488632202, | |
| "learning_rate": 3.80551638784277e-05, | |
| "loss": 0.043, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 2.6684491978609626, | |
| "grad_norm": 3.525506019592285, | |
| "learning_rate": 3.664794198395764e-05, | |
| "loss": 0.0643, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.6898395721925135, | |
| "grad_norm": 5.074891567230225, | |
| "learning_rate": 3.5261371521817244e-05, | |
| "loss": 0.0658, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 2.711229946524064, | |
| "grad_norm": 3.6220922470092773, | |
| "learning_rate": 3.3895904484679984e-05, | |
| "loss": 0.1535, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 2.732620320855615, | |
| "grad_norm": 3.9044840335845947, | |
| "learning_rate": 3.2551985985948616e-05, | |
| "loss": 0.0572, | |
| "step": 138 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 184, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 46, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3759516646572032e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |