| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.008, | |
| "eval_steps": 500, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 8e-05, | |
| "grad_norm": 0.1861909031867981, | |
| "learning_rate": 4e-05, | |
| "loss": 0.6283, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00016, | |
| "grad_norm": 0.16733917593955994, | |
| "learning_rate": 8e-05, | |
| "loss": 0.6772, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00024, | |
| "grad_norm": 0.20496870577335358, | |
| "learning_rate": 0.00012, | |
| "loss": 0.8642, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00032, | |
| "grad_norm": 0.19442355632781982, | |
| "learning_rate": 0.00016, | |
| "loss": 0.7268, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0004, | |
| "grad_norm": 0.1718096137046814, | |
| "learning_rate": 0.0002, | |
| "loss": 0.6486, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00048, | |
| "grad_norm": 0.1942625492811203, | |
| "learning_rate": 0.00019555555555555556, | |
| "loss": 0.7806, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00056, | |
| "grad_norm": 0.17527641355991364, | |
| "learning_rate": 0.00019111111111111114, | |
| "loss": 0.5136, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.00064, | |
| "grad_norm": 0.2261490821838379, | |
| "learning_rate": 0.0001866666666666667, | |
| "loss": 0.8569, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00072, | |
| "grad_norm": 0.17914293706417084, | |
| "learning_rate": 0.00018222222222222224, | |
| "loss": 0.7845, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0008, | |
| "grad_norm": 0.21953529119491577, | |
| "learning_rate": 0.00017777777777777779, | |
| "loss": 0.6735, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00088, | |
| "grad_norm": 0.1893182247877121, | |
| "learning_rate": 0.00017333333333333334, | |
| "loss": 0.8183, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.00096, | |
| "grad_norm": 0.20547300577163696, | |
| "learning_rate": 0.00016888888888888889, | |
| "loss": 1.0212, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.00104, | |
| "grad_norm": 0.2539304196834564, | |
| "learning_rate": 0.00016444444444444444, | |
| "loss": 0.8543, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.00112, | |
| "grad_norm": 0.17561326920986176, | |
| "learning_rate": 0.00016, | |
| "loss": 0.5858, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0012, | |
| "grad_norm": 0.19194333255290985, | |
| "learning_rate": 0.00015555555555555556, | |
| "loss": 0.8022, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00128, | |
| "grad_norm": 0.22850346565246582, | |
| "learning_rate": 0.0001511111111111111, | |
| "loss": 0.5618, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.00136, | |
| "grad_norm": 0.2263827621936798, | |
| "learning_rate": 0.00014666666666666666, | |
| "loss": 0.9284, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.00144, | |
| "grad_norm": 0.2525479197502136, | |
| "learning_rate": 0.00014222222222222224, | |
| "loss": 0.7528, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.00152, | |
| "grad_norm": 0.19500723481178284, | |
| "learning_rate": 0.0001377777777777778, | |
| "loss": 0.7256, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0016, | |
| "grad_norm": 0.24944712221622467, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 0.8527, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00168, | |
| "grad_norm": 0.20906241238117218, | |
| "learning_rate": 0.00012888888888888892, | |
| "loss": 0.8557, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.00176, | |
| "grad_norm": 0.196335569024086, | |
| "learning_rate": 0.00012444444444444444, | |
| "loss": 0.8134, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.00184, | |
| "grad_norm": 0.15679122507572174, | |
| "learning_rate": 0.00012, | |
| "loss": 1.0036, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.00192, | |
| "grad_norm": 0.2803596556186676, | |
| "learning_rate": 0.00011555555555555555, | |
| "loss": 0.8113, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.002, | |
| "grad_norm": 0.22238129377365112, | |
| "learning_rate": 0.00011111111111111112, | |
| "loss": 0.5934, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.00208, | |
| "grad_norm": 0.23300746083259583, | |
| "learning_rate": 0.00010666666666666667, | |
| "loss": 0.7823, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.00216, | |
| "grad_norm": 0.2822686731815338, | |
| "learning_rate": 0.00010222222222222222, | |
| "loss": 0.7634, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.00224, | |
| "grad_norm": 0.25708135962486267, | |
| "learning_rate": 9.777777777777778e-05, | |
| "loss": 0.7359, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.00232, | |
| "grad_norm": 0.2508324980735779, | |
| "learning_rate": 9.333333333333334e-05, | |
| "loss": 1.0442, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0024, | |
| "grad_norm": 0.2428746372461319, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 1.0047, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.00248, | |
| "grad_norm": 0.2691624164581299, | |
| "learning_rate": 8.444444444444444e-05, | |
| "loss": 0.633, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.00256, | |
| "grad_norm": 0.1821339726448059, | |
| "learning_rate": 8e-05, | |
| "loss": 0.5192, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.00264, | |
| "grad_norm": 0.2801157534122467, | |
| "learning_rate": 7.555555555555556e-05, | |
| "loss": 0.5962, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.00272, | |
| "grad_norm": 0.22572503983974457, | |
| "learning_rate": 7.111111111111112e-05, | |
| "loss": 0.5493, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.0028, | |
| "grad_norm": 0.25089773535728455, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.7187, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.00288, | |
| "grad_norm": 0.2668441832065582, | |
| "learning_rate": 6.222222222222222e-05, | |
| "loss": 0.9809, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.00296, | |
| "grad_norm": 0.281890869140625, | |
| "learning_rate": 5.7777777777777776e-05, | |
| "loss": 0.8585, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.00304, | |
| "grad_norm": 0.22123876214027405, | |
| "learning_rate": 5.333333333333333e-05, | |
| "loss": 0.6978, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.00312, | |
| "grad_norm": 0.24300044775009155, | |
| "learning_rate": 4.888888888888889e-05, | |
| "loss": 0.7604, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0032, | |
| "grad_norm": 0.2412249892950058, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.9841, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.00328, | |
| "grad_norm": 0.2551029324531555, | |
| "learning_rate": 4e-05, | |
| "loss": 0.7346, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.00336, | |
| "grad_norm": 0.2715751528739929, | |
| "learning_rate": 3.555555555555556e-05, | |
| "loss": 0.9846, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.00344, | |
| "grad_norm": 0.22599095106124878, | |
| "learning_rate": 3.111111111111111e-05, | |
| "loss": 0.7476, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.00352, | |
| "grad_norm": 0.26102009415626526, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 0.7947, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.0036, | |
| "grad_norm": 0.2690563499927521, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.7454, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.00368, | |
| "grad_norm": 0.2677896320819855, | |
| "learning_rate": 1.777777777777778e-05, | |
| "loss": 0.8543, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.00376, | |
| "grad_norm": 0.2913207709789276, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.771, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.00384, | |
| "grad_norm": 0.2139165699481964, | |
| "learning_rate": 8.88888888888889e-06, | |
| "loss": 0.6454, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.00392, | |
| "grad_norm": 0.3593102693557739, | |
| "learning_rate": 4.444444444444445e-06, | |
| "loss": 1.0135, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.004, | |
| "grad_norm": 0.2176782488822937, | |
| "learning_rate": 0.0, | |
| "loss": 1.0192, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.00408, | |
| "grad_norm": 0.17560550570487976, | |
| "learning_rate": 0.0, | |
| "loss": 0.4516, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.00416, | |
| "grad_norm": 0.1952800750732422, | |
| "learning_rate": 0.0, | |
| "loss": 0.9039, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.00424, | |
| "grad_norm": 0.9650332927703857, | |
| "learning_rate": 0.00019923169267707086, | |
| "loss": 1.2938, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.00432, | |
| "grad_norm": 0.2434082180261612, | |
| "learning_rate": 0.0001992156862745098, | |
| "loss": 0.6807, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.0044, | |
| "grad_norm": 0.3336096405982971, | |
| "learning_rate": 0.00019919967987194878, | |
| "loss": 1.0532, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.00448, | |
| "grad_norm": 0.21415311098098755, | |
| "learning_rate": 0.00019918367346938775, | |
| "loss": 1.1118, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.00456, | |
| "grad_norm": 0.17886948585510254, | |
| "learning_rate": 0.00019916766706682676, | |
| "loss": 0.7198, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.00464, | |
| "grad_norm": 0.1993376612663269, | |
| "learning_rate": 0.0001991516606642657, | |
| "loss": 0.8233, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.00472, | |
| "grad_norm": 0.21231591701507568, | |
| "learning_rate": 0.00019913565426170468, | |
| "loss": 0.75, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.0048, | |
| "grad_norm": 0.22594475746154785, | |
| "learning_rate": 0.00019911964785914368, | |
| "loss": 0.9078, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.00488, | |
| "grad_norm": 0.21537600457668304, | |
| "learning_rate": 0.00019910364145658266, | |
| "loss": 0.7157, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.00496, | |
| "grad_norm": 0.23297461867332458, | |
| "learning_rate": 0.0001990876350540216, | |
| "loss": 0.8651, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.00504, | |
| "grad_norm": 0.19292137026786804, | |
| "learning_rate": 0.00019907162865146058, | |
| "loss": 0.7759, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.00512, | |
| "grad_norm": 0.4145672023296356, | |
| "learning_rate": 0.00019905562224889958, | |
| "loss": 1.1774, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.0052, | |
| "grad_norm": 0.19476854801177979, | |
| "learning_rate": 0.00019903961584633856, | |
| "loss": 0.768, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.00528, | |
| "grad_norm": 0.22146064043045044, | |
| "learning_rate": 0.0001990236094437775, | |
| "loss": 0.8986, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.00536, | |
| "grad_norm": 0.2676100432872772, | |
| "learning_rate": 0.0001990076030412165, | |
| "loss": 0.8371, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.00544, | |
| "grad_norm": 0.21964198350906372, | |
| "learning_rate": 0.00019899159663865548, | |
| "loss": 0.7568, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.00552, | |
| "grad_norm": 0.18942779302597046, | |
| "learning_rate": 0.00019897559023609445, | |
| "loss": 0.8593, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.0056, | |
| "grad_norm": 0.21514566242694855, | |
| "learning_rate": 0.0001989595838335334, | |
| "loss": 0.6191, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.00568, | |
| "grad_norm": 0.21467895805835724, | |
| "learning_rate": 0.0001989435774309724, | |
| "loss": 0.4907, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.00576, | |
| "grad_norm": 0.20334534347057343, | |
| "learning_rate": 0.00019892757102841138, | |
| "loss": 0.6906, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.00584, | |
| "grad_norm": 0.21853458881378174, | |
| "learning_rate": 0.00019891156462585035, | |
| "loss": 0.8808, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.00592, | |
| "grad_norm": 0.2489083856344223, | |
| "learning_rate": 0.00019889555822328933, | |
| "loss": 0.9344, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.006, | |
| "grad_norm": 0.28158295154571533, | |
| "learning_rate": 0.0001988795518207283, | |
| "loss": 0.6637, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.00608, | |
| "grad_norm": 0.267078161239624, | |
| "learning_rate": 0.00019886354541816728, | |
| "loss": 0.7328, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.00616, | |
| "grad_norm": 0.2382790893316269, | |
| "learning_rate": 0.00019884753901560625, | |
| "loss": 0.9646, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.00624, | |
| "grad_norm": 0.26280778646469116, | |
| "learning_rate": 0.00019883153261304523, | |
| "loss": 0.7376, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.00632, | |
| "grad_norm": 0.2713547945022583, | |
| "learning_rate": 0.0001988155262104842, | |
| "loss": 1.0505, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.0064, | |
| "grad_norm": 0.23571620881557465, | |
| "learning_rate": 0.00019879951980792318, | |
| "loss": 0.8017, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.00648, | |
| "grad_norm": 0.2187693864107132, | |
| "learning_rate": 0.00019878351340536215, | |
| "loss": 0.6597, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.00656, | |
| "grad_norm": 0.22543781995773315, | |
| "learning_rate": 0.00019876750700280113, | |
| "loss": 0.8349, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.00664, | |
| "grad_norm": 0.23452317714691162, | |
| "learning_rate": 0.0001987515006002401, | |
| "loss": 0.8549, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.00672, | |
| "grad_norm": 0.20848535001277924, | |
| "learning_rate": 0.00019873549419767908, | |
| "loss": 1.0107, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.0068, | |
| "grad_norm": 0.19626936316490173, | |
| "learning_rate": 0.00019871948779511805, | |
| "loss": 0.8725, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.00688, | |
| "grad_norm": 0.17012615501880646, | |
| "learning_rate": 0.00019870348139255703, | |
| "loss": 0.6128, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.00696, | |
| "grad_norm": 0.21431602537631989, | |
| "learning_rate": 0.000198687474989996, | |
| "loss": 0.8825, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.00704, | |
| "grad_norm": 0.18353329598903656, | |
| "learning_rate": 0.000198671468587435, | |
| "loss": 0.621, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.00712, | |
| "grad_norm": 0.20687150955200195, | |
| "learning_rate": 0.00019865546218487395, | |
| "loss": 0.8183, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.0072, | |
| "grad_norm": 0.1761425882577896, | |
| "learning_rate": 0.00019863945578231293, | |
| "loss": 1.1441, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.00728, | |
| "grad_norm": 0.2332560122013092, | |
| "learning_rate": 0.0001986234493797519, | |
| "loss": 0.8049, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.00736, | |
| "grad_norm": 0.2265244722366333, | |
| "learning_rate": 0.0001986074429771909, | |
| "loss": 0.5617, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.00744, | |
| "grad_norm": 0.22966669499874115, | |
| "learning_rate": 0.00019859143657462985, | |
| "loss": 0.8118, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.00752, | |
| "grad_norm": 0.1718137413263321, | |
| "learning_rate": 0.00019857543017206883, | |
| "loss": 0.6905, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.0076, | |
| "grad_norm": 0.1688937246799469, | |
| "learning_rate": 0.0001985594237695078, | |
| "loss": 0.532, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.00768, | |
| "grad_norm": 0.263818621635437, | |
| "learning_rate": 0.0001985434173669468, | |
| "loss": 0.5655, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.00776, | |
| "grad_norm": 0.20886196196079254, | |
| "learning_rate": 0.00019852741096438575, | |
| "loss": 0.655, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.00784, | |
| "grad_norm": 0.23567333817481995, | |
| "learning_rate": 0.00019851140456182473, | |
| "loss": 0.7477, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.00792, | |
| "grad_norm": 0.22903132438659668, | |
| "learning_rate": 0.00019849539815926373, | |
| "loss": 0.764, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.008, | |
| "grad_norm": 0.1896105259656906, | |
| "learning_rate": 0.0001984793917567027, | |
| "loss": 0.7269, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 12500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.039229864103936e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |