| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 368, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01358695652173913, | |
| "grad_norm": 20.125, | |
| "learning_rate": 2.702702702702703e-06, | |
| "loss": 1.708, | |
| "mean_token_accuracy": 0.6925400972366333, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.02717391304347826, | |
| "grad_norm": 12.75, | |
| "learning_rate": 5.405405405405406e-06, | |
| "loss": 1.691, | |
| "mean_token_accuracy": 0.6875800609588623, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04076086956521739, | |
| "grad_norm": 13.125, | |
| "learning_rate": 8.108108108108109e-06, | |
| "loss": 1.549, | |
| "mean_token_accuracy": 0.7068500399589539, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.05434782608695652, | |
| "grad_norm": 11.375, | |
| "learning_rate": 1.0810810810810812e-05, | |
| "loss": 1.3324, | |
| "mean_token_accuracy": 0.7429514169692993, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06793478260869565, | |
| "grad_norm": 14.875, | |
| "learning_rate": 1.3513513513513515e-05, | |
| "loss": 0.9671, | |
| "mean_token_accuracy": 0.8036805391311646, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.08152173913043478, | |
| "grad_norm": 13.5, | |
| "learning_rate": 1.6216216216216218e-05, | |
| "loss": 0.9606, | |
| "mean_token_accuracy": 0.8269321084022522, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.09510869565217392, | |
| "grad_norm": 13.625, | |
| "learning_rate": 1.891891891891892e-05, | |
| "loss": 0.6205, | |
| "mean_token_accuracy": 0.8943289399147034, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.10869565217391304, | |
| "grad_norm": 11.4375, | |
| "learning_rate": 1.9995946530314384e-05, | |
| "loss": 0.3571, | |
| "mean_token_accuracy": 0.934650433063507, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.12228260869565218, | |
| "grad_norm": 8.75, | |
| "learning_rate": 1.9971187226043746e-05, | |
| "loss": 0.4531, | |
| "mean_token_accuracy": 0.9206116795539856, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.1358695652173913, | |
| "grad_norm": 8.875, | |
| "learning_rate": 1.9923976226947417e-05, | |
| "loss": 0.4687, | |
| "mean_token_accuracy": 0.916878092288971, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.14945652173913043, | |
| "grad_norm": 5.3125, | |
| "learning_rate": 1.985441983600819e-05, | |
| "loss": 0.3238, | |
| "mean_token_accuracy": 0.9440895915031433, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.16304347826086957, | |
| "grad_norm": 16.375, | |
| "learning_rate": 1.9762674670369757e-05, | |
| "loss": 0.2045, | |
| "mean_token_accuracy": 0.9566964030265808, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1766304347826087, | |
| "grad_norm": 9.25, | |
| "learning_rate": 1.9648947308688594e-05, | |
| "loss": 0.2272, | |
| "mean_token_accuracy": 0.9550906538963317, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.19021739130434784, | |
| "grad_norm": 13.75, | |
| "learning_rate": 1.9513493825989664e-05, | |
| "loss": 0.4506, | |
| "mean_token_accuracy": 0.9193356156349182, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.20380434782608695, | |
| "grad_norm": 12.1875, | |
| "learning_rate": 1.9356619217073252e-05, | |
| "loss": 0.4804, | |
| "mean_token_accuracy": 0.9193973302841186, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.21739130434782608, | |
| "grad_norm": 6.96875, | |
| "learning_rate": 1.917867670977126e-05, | |
| "loss": 0.4535, | |
| "mean_token_accuracy": 0.9251804232597352, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.23097826086956522, | |
| "grad_norm": 12.3125, | |
| "learning_rate": 1.8980066969599216e-05, | |
| "loss": 0.5313, | |
| "mean_token_accuracy": 0.9139753341674804, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.24456521739130435, | |
| "grad_norm": 4.75, | |
| "learning_rate": 1.8761237197594945e-05, | |
| "loss": 0.2025, | |
| "mean_token_accuracy": 0.9598217248916626, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.25815217391304346, | |
| "grad_norm": 2.859375, | |
| "learning_rate": 1.852268012337514e-05, | |
| "loss": 0.4308, | |
| "mean_token_accuracy": 0.9284366369247437, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.2717391304347826, | |
| "grad_norm": 9.25, | |
| "learning_rate": 1.8264932895677195e-05, | |
| "loss": 0.4234, | |
| "mean_token_accuracy": 0.930481493473053, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.28532608695652173, | |
| "grad_norm": 7.5, | |
| "learning_rate": 1.798857587288445e-05, | |
| "loss": 0.3343, | |
| "mean_token_accuracy": 0.9406642079353332, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.29891304347826086, | |
| "grad_norm": 12.5, | |
| "learning_rate": 1.769423131625808e-05, | |
| "loss": 0.198, | |
| "mean_token_accuracy": 0.9616581201553345, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.3125, | |
| "grad_norm": 6.875, | |
| "learning_rate": 1.738256198881809e-05, | |
| "loss": 0.2704, | |
| "mean_token_accuracy": 0.9502018094062805, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.32608695652173914, | |
| "grad_norm": 2.859375, | |
| "learning_rate": 1.7054269663028232e-05, | |
| "loss": 0.208, | |
| "mean_token_accuracy": 0.9611244797706604, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.33967391304347827, | |
| "grad_norm": 3.859375, | |
| "learning_rate": 1.6710093540645056e-05, | |
| "loss": 0.364, | |
| "mean_token_accuracy": 0.9366228461265564, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.3532608695652174, | |
| "grad_norm": 5.6875, | |
| "learning_rate": 1.6350808588288964e-05, | |
| "loss": 0.1251, | |
| "mean_token_accuracy": 0.9757622718811035, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.36684782608695654, | |
| "grad_norm": 7.1875, | |
| "learning_rate": 1.597722379248512e-05, | |
| "loss": 0.2729, | |
| "mean_token_accuracy": 0.9491044163703919, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.3804347826086957, | |
| "grad_norm": 7.84375, | |
| "learning_rate": 1.559018033810316e-05, | |
| "loss": 0.3411, | |
| "mean_token_accuracy": 0.9376911044120788, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.39402173913043476, | |
| "grad_norm": 2.65625, | |
| "learning_rate": 1.5190549714297303e-05, | |
| "loss": 0.2513, | |
| "mean_token_accuracy": 0.9505762934684754, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.4076086956521739, | |
| "grad_norm": 5.875, | |
| "learning_rate": 1.4779231752211546e-05, | |
| "loss": 0.1996, | |
| "mean_token_accuracy": 0.9632942318916321, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.421195652173913, | |
| "grad_norm": 3.8125, | |
| "learning_rate": 1.4357152598868478e-05, | |
| "loss": 0.2669, | |
| "mean_token_accuracy": 0.9502842426300049, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.43478260869565216, | |
| "grad_norm": 5.34375, | |
| "learning_rate": 1.3925262631803722e-05, | |
| "loss": 0.1767, | |
| "mean_token_accuracy": 0.962754237651825, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.4483695652173913, | |
| "grad_norm": 2.328125, | |
| "learning_rate": 1.3484534319141592e-05, | |
| "loss": 0.1936, | |
| "mean_token_accuracy": 0.9623476982116699, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.46195652173913043, | |
| "grad_norm": 4.09375, | |
| "learning_rate": 1.303596002993028e-05, | |
| "loss": 0.2875, | |
| "mean_token_accuracy": 0.948505699634552, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.47554347826086957, | |
| "grad_norm": 5.0625, | |
| "learning_rate": 1.2580549799667034e-05, | |
| "loss": 0.1843, | |
| "mean_token_accuracy": 0.9624204635620117, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.4891304347826087, | |
| "grad_norm": 5.21875, | |
| "learning_rate": 1.2119329056044533e-05, | |
| "loss": 0.1725, | |
| "mean_token_accuracy": 0.9649592399597168, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.5027173913043478, | |
| "grad_norm": 8.8125, | |
| "learning_rate": 1.165333631003928e-05, | |
| "loss": 0.3102, | |
| "mean_token_accuracy": 0.9432898998260498, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.5163043478260869, | |
| "grad_norm": 11.6875, | |
| "learning_rate": 1.1183620817540985e-05, | |
| "loss": 0.2023, | |
| "mean_token_accuracy": 0.9577706575393676, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.529891304347826, | |
| "grad_norm": 3.734375, | |
| "learning_rate": 1.0711240216788036e-05, | |
| "loss": 0.3165, | |
| "mean_token_accuracy": 0.9471179962158203, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.5434782608695652, | |
| "grad_norm": 6.90625, | |
| "learning_rate": 1.0237258146928849e-05, | |
| "loss": 0.1964, | |
| "mean_token_accuracy": 0.9606888651847839, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5570652173913043, | |
| "grad_norm": 5.875, | |
| "learning_rate": 9.762741853071153e-06, | |
| "loss": 0.2211, | |
| "mean_token_accuracy": 0.9585472226142884, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.5706521739130435, | |
| "grad_norm": 5.53125, | |
| "learning_rate": 9.288759783211967e-06, | |
| "loss": 0.2484, | |
| "mean_token_accuracy": 0.9550203323364258, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.5842391304347826, | |
| "grad_norm": 2.8125, | |
| "learning_rate": 8.81637918245902e-06, | |
| "loss": 0.3283, | |
| "mean_token_accuracy": 0.9439280152320861, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.5978260869565217, | |
| "grad_norm": 3.8125, | |
| "learning_rate": 8.346663689960724e-06, | |
| "loss": 0.2486, | |
| "mean_token_accuracy": 0.953123664855957, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.6114130434782609, | |
| "grad_norm": 5.4375, | |
| "learning_rate": 7.880670943955467e-06, | |
| "loss": 0.2411, | |
| "mean_token_accuracy": 0.9570030689239502, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 2.078125, | |
| "learning_rate": 7.419450200332965e-06, | |
| "loss": 0.2028, | |
| "mean_token_accuracy": 0.9596728563308716, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.6385869565217391, | |
| "grad_norm": 2.9375, | |
| "learning_rate": 6.964039970069722e-06, | |
| "loss": 0.2145, | |
| "mean_token_accuracy": 0.9589278101921082, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.6521739130434783, | |
| "grad_norm": 6.4375, | |
| "learning_rate": 6.515465680858412e-06, | |
| "loss": 0.2646, | |
| "mean_token_accuracy": 0.9482977151870727, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.6657608695652174, | |
| "grad_norm": 5.46875, | |
| "learning_rate": 6.074737368196279e-06, | |
| "loss": 0.2332, | |
| "mean_token_accuracy": 0.9552296161651611, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.6793478260869565, | |
| "grad_norm": 1.8984375, | |
| "learning_rate": 5.642847401131526e-06, | |
| "loss": 0.2629, | |
| "mean_token_accuracy": 0.953015148639679, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.6929347826086957, | |
| "grad_norm": 6.8125, | |
| "learning_rate": 5.220768247788458e-06, | |
| "loss": 0.2226, | |
| "mean_token_accuracy": 0.9579466819763184, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.7065217391304348, | |
| "grad_norm": 2.03125, | |
| "learning_rate": 4.809450285702697e-06, | |
| "loss": 0.1219, | |
| "mean_token_accuracy": 0.9744248032569885, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.720108695652174, | |
| "grad_norm": 4.5, | |
| "learning_rate": 4.409819661896839e-06, | |
| "loss": 0.1909, | |
| "mean_token_accuracy": 0.9654230594635009, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.7336956521739131, | |
| "grad_norm": 8.375, | |
| "learning_rate": 4.022776207514885e-06, | |
| "loss": 0.183, | |
| "mean_token_accuracy": 0.9614866137504577, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.7472826086956522, | |
| "grad_norm": 2.53125, | |
| "learning_rate": 3.6491914117110405e-06, | |
| "loss": 0.2841, | |
| "mean_token_accuracy": 0.9495038151741028, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.7608695652173914, | |
| "grad_norm": 3.5, | |
| "learning_rate": 3.2899064593549477e-06, | |
| "loss": 0.147, | |
| "mean_token_accuracy": 0.9706088781356812, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7744565217391305, | |
| "grad_norm": 9.375, | |
| "learning_rate": 2.945730336971767e-06, | |
| "loss": 0.2865, | |
| "mean_token_accuracy": 0.9473696708679199, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.7880434782608695, | |
| "grad_norm": 3.140625, | |
| "learning_rate": 2.6174380111819144e-06, | |
| "loss": 0.1647, | |
| "mean_token_accuracy": 0.9646413207054139, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.8016304347826086, | |
| "grad_norm": 3.921875, | |
| "learning_rate": 2.3057686837419246e-06, | |
| "loss": 0.127, | |
| "mean_token_accuracy": 0.9736526370048523, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.8152173913043478, | |
| "grad_norm": 2.78125, | |
| "learning_rate": 2.011424127115552e-06, | |
| "loss": 0.1714, | |
| "mean_token_accuracy": 0.963629424571991, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.8288043478260869, | |
| "grad_norm": 3.515625, | |
| "learning_rate": 1.7350671043228072e-06, | |
| "loss": 0.2071, | |
| "mean_token_accuracy": 0.9631360173225403, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.842391304347826, | |
| "grad_norm": 2.375, | |
| "learning_rate": 1.4773198766248642e-06, | |
| "loss": 0.1566, | |
| "mean_token_accuracy": 0.971684205532074, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.8559782608695652, | |
| "grad_norm": 2.265625, | |
| "learning_rate": 1.2387628024050557e-06, | |
| "loss": 0.2517, | |
| "mean_token_accuracy": 0.9517509698867798, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "grad_norm": 6.5625, | |
| "learning_rate": 1.0199330304007858e-06, | |
| "loss": 0.247, | |
| "mean_token_accuracy": 0.9540893912315369, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8831521739130435, | |
| "grad_norm": 6.65625, | |
| "learning_rate": 8.213232902287438e-07, | |
| "loss": 0.1369, | |
| "mean_token_accuracy": 0.9701100349426269, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.8967391304347826, | |
| "grad_norm": 1.90625, | |
| "learning_rate": 6.433807829267491e-07, | |
| "loss": 0.1382, | |
| "mean_token_accuracy": 0.9745123624801636, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.9103260869565217, | |
| "grad_norm": 5.125, | |
| "learning_rate": 4.865061740103361e-07, | |
| "loss": 0.1746, | |
| "mean_token_accuracy": 0.9617828845977783, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.9239130434782609, | |
| "grad_norm": 2.765625, | |
| "learning_rate": 3.510526913114065e-07, | |
| "loss": 0.2696, | |
| "mean_token_accuracy": 0.9500232100486755, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.9375, | |
| "grad_norm": 3.5, | |
| "learning_rate": 2.3732532963024468e-07, | |
| "loss": 0.2069, | |
| "mean_token_accuracy": 0.9611208081245423, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 0.9510869565217391, | |
| "grad_norm": 5.6875, | |
| "learning_rate": 1.4558016399181086e-07, | |
| "loss": 0.1823, | |
| "mean_token_accuracy": 0.965398907661438, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.9646739130434783, | |
| "grad_norm": 3.359375, | |
| "learning_rate": 7.602377305258479e-08, | |
| "loss": 0.1987, | |
| "mean_token_accuracy": 0.9622387170791626, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 0.9782608695652174, | |
| "grad_norm": 2.203125, | |
| "learning_rate": 2.8812773956256034e-08, | |
| "loss": 0.1906, | |
| "mean_token_accuracy": 0.9647317886352539, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.9918478260869565, | |
| "grad_norm": 4.09375, | |
| "learning_rate": 4.053469685617595e-09, | |
| "loss": 0.2464, | |
| "mean_token_accuracy": 0.9541746735572815, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "mean_token_accuracy": 0.9636613726615906, | |
| "step": 368, | |
| "total_flos": 2.6349158744595456e+16, | |
| "train_loss": 0.35062233002289483, | |
| "train_runtime": 5984.3116, | |
| "train_samples_per_second": 0.061, | |
| "train_steps_per_second": 0.061 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 368, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.6349158744595456e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |