| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9910313901345291, | |
| "eval_steps": 500, | |
| "global_step": 444, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02242152466367713, | |
| "grad_norm": 2.390625, | |
| "learning_rate": 2.123791785182557e-05, | |
| "loss": 0.9148, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.04484304932735426, | |
| "grad_norm": 1.3828125, | |
| "learning_rate": 4.778531516660753e-05, | |
| "loss": 0.7159, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06726457399103139, | |
| "grad_norm": 1.1875, | |
| "learning_rate": 7.433271248138951e-05, | |
| "loss": 0.6679, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.08968609865470852, | |
| "grad_norm": 1.0078125, | |
| "learning_rate": 0.00010088010979617146, | |
| "loss": 0.6512, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11210762331838565, | |
| "grad_norm": 1.046875, | |
| "learning_rate": 0.00012742750711095343, | |
| "loss": 0.6596, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.13452914798206278, | |
| "grad_norm": 1.015625, | |
| "learning_rate": 0.0001539749044257354, | |
| "loss": 0.6348, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.15695067264573992, | |
| "grad_norm": 1.1015625, | |
| "learning_rate": 0.00018052230174051736, | |
| "loss": 0.6352, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.17937219730941703, | |
| "grad_norm": 1.046875, | |
| "learning_rate": 0.00018581809294315903, | |
| "loss": 0.6422, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.20179372197309417, | |
| "grad_norm": 0.8125, | |
| "learning_rate": 0.0001857624936017134, | |
| "loss": 0.6619, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2242152466367713, | |
| "grad_norm": 0.84375, | |
| "learning_rate": 0.0001856641617644384, | |
| "loss": 0.6454, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.24663677130044842, | |
| "grad_norm": 0.66015625, | |
| "learning_rate": 0.00018552315778910269, | |
| "loss": 0.6523, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.26905829596412556, | |
| "grad_norm": 0.69140625, | |
| "learning_rate": 0.0001853395682263653, | |
| "loss": 0.6485, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2914798206278027, | |
| "grad_norm": 0.9453125, | |
| "learning_rate": 0.0001851135057666494, | |
| "loss": 0.6452, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.31390134529147984, | |
| "grad_norm": 1.2109375, | |
| "learning_rate": 0.00018484510917097098, | |
| "loss": 0.6878, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.336322869955157, | |
| "grad_norm": 0.84765625, | |
| "learning_rate": 0.00018453454318576493, | |
| "loss": 0.6585, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.35874439461883406, | |
| "grad_norm": 0.62890625, | |
| "learning_rate": 0.00018418199844176086, | |
| "loss": 0.6447, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.3811659192825112, | |
| "grad_norm": 0.67578125, | |
| "learning_rate": 0.00018378769133697064, | |
| "loss": 0.618, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.40358744394618834, | |
| "grad_norm": 0.6171875, | |
| "learning_rate": 0.00018335186390385983, | |
| "loss": 0.6242, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4260089686098655, | |
| "grad_norm": 0.62109375, | |
| "learning_rate": 0.00018287478366078356, | |
| "loss": 0.62, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.4484304932735426, | |
| "grad_norm": 0.5234375, | |
| "learning_rate": 0.0001823567434477796, | |
| "loss": 0.6234, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.47085201793721976, | |
| "grad_norm": 0.546875, | |
| "learning_rate": 0.00018179806124681794, | |
| "loss": 0.6088, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.49327354260089684, | |
| "grad_norm": 0.5078125, | |
| "learning_rate": 0.00018119907998661786, | |
| "loss": 0.6069, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.515695067264574, | |
| "grad_norm": 0.51171875, | |
| "learning_rate": 0.00018056016733215274, | |
| "loss": 0.6037, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.5381165919282511, | |
| "grad_norm": 0.50390625, | |
| "learning_rate": 0.00017988171545897072, | |
| "loss": 0.5977, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.5605381165919282, | |
| "grad_norm": 0.5234375, | |
| "learning_rate": 0.00017916414081247065, | |
| "loss": 0.5903, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.5829596412556054, | |
| "grad_norm": 0.4765625, | |
| "learning_rate": 0.0001784078838522809, | |
| "loss": 0.6063, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.6053811659192825, | |
| "grad_norm": 0.58203125, | |
| "learning_rate": 0.0001776134087818975, | |
| "loss": 0.5831, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.6278026905829597, | |
| "grad_norm": 0.5546875, | |
| "learning_rate": 0.00017678120326374824, | |
| "loss": 0.6045, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.6502242152466368, | |
| "grad_norm": 0.478515625, | |
| "learning_rate": 0.00017591177811985704, | |
| "loss": 0.593, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.672645739910314, | |
| "grad_norm": 0.55859375, | |
| "learning_rate": 0.00017500566701829254, | |
| "loss": 0.5943, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.695067264573991, | |
| "grad_norm": 0.51953125, | |
| "learning_rate": 0.00017406342614559366, | |
| "loss": 0.5716, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.7174887892376681, | |
| "grad_norm": 0.466796875, | |
| "learning_rate": 0.00017308563386537258, | |
| "loss": 0.59, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.7399103139013453, | |
| "grad_norm": 0.427734375, | |
| "learning_rate": 0.00017207289036330536, | |
| "loss": 0.5801, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.7623318385650224, | |
| "grad_norm": 0.48828125, | |
| "learning_rate": 0.00017102581727872763, | |
| "loss": 0.5918, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.7847533632286996, | |
| "grad_norm": 0.5078125, | |
| "learning_rate": 0.0001699450573230617, | |
| "loss": 0.5793, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.8071748878923767, | |
| "grad_norm": 0.51953125, | |
| "learning_rate": 0.00016883127388530927, | |
| "loss": 0.5737, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.8295964125560538, | |
| "grad_norm": 0.5546875, | |
| "learning_rate": 0.00016768515062485188, | |
| "loss": 0.5792, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.852017937219731, | |
| "grad_norm": 0.48046875, | |
| "learning_rate": 0.000166507391051809, | |
| "loss": 0.5656, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.874439461883408, | |
| "grad_norm": 0.48046875, | |
| "learning_rate": 0.0001652987180952116, | |
| "loss": 0.5625, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.8968609865470852, | |
| "grad_norm": 0.54296875, | |
| "learning_rate": 0.00016405987365925574, | |
| "loss": 0.5929, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.9192825112107623, | |
| "grad_norm": 0.51953125, | |
| "learning_rate": 0.00016279161816790908, | |
| "loss": 0.5732, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.9417040358744395, | |
| "grad_norm": 0.5625, | |
| "learning_rate": 0.00016149473009814973, | |
| "loss": 0.5595, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.9641255605381166, | |
| "grad_norm": 0.490234375, | |
| "learning_rate": 0.00016017000550212353, | |
| "loss": 0.5662, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.9865470852017937, | |
| "grad_norm": 0.4765625, | |
| "learning_rate": 0.0001588182575185139, | |
| "loss": 0.5771, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.9955156950672646, | |
| "eval_loss": 0.5749066472053528, | |
| "eval_runtime": 2.1723, | |
| "eval_samples_per_second": 19.335, | |
| "eval_steps_per_second": 19.335, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 1.0089686098654709, | |
| "grad_norm": 0.640625, | |
| "learning_rate": 0.0001574403158734228, | |
| "loss": 0.5053, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.031390134529148, | |
| "grad_norm": 0.55078125, | |
| "learning_rate": 0.0001560370263710713, | |
| "loss": 0.4183, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.053811659192825, | |
| "grad_norm": 0.5703125, | |
| "learning_rate": 0.00015460925037463, | |
| "loss": 0.4084, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.0762331838565022, | |
| "grad_norm": 0.51171875, | |
| "learning_rate": 0.00015315786427749957, | |
| "loss": 0.4123, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.0986547085201794, | |
| "grad_norm": 0.498046875, | |
| "learning_rate": 0.0001516837589653656, | |
| "loss": 0.4045, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.1210762331838564, | |
| "grad_norm": 0.4921875, | |
| "learning_rate": 0.00015018783926935766, | |
| "loss": 0.4039, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.1434977578475336, | |
| "grad_norm": 0.458984375, | |
| "learning_rate": 0.00014867102341064824, | |
| "loss": 0.4171, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.1659192825112108, | |
| "grad_norm": 0.49609375, | |
| "learning_rate": 0.00014713424243683308, | |
| "loss": 0.4107, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.188340807174888, | |
| "grad_norm": 0.458984375, | |
| "learning_rate": 0.00014557843965043807, | |
| "loss": 0.4092, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.210762331838565, | |
| "grad_norm": 0.486328125, | |
| "learning_rate": 0.000144004570029904, | |
| "loss": 0.4068, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.2331838565022422, | |
| "grad_norm": 0.45703125, | |
| "learning_rate": 0.00014241359964340427, | |
| "loss": 0.4143, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.2556053811659194, | |
| "grad_norm": 0.453125, | |
| "learning_rate": 0.00014080650505585602, | |
| "loss": 0.4233, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.2780269058295963, | |
| "grad_norm": 0.4765625, | |
| "learning_rate": 0.00013918427272948742, | |
| "loss": 0.4154, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.3004484304932735, | |
| "grad_norm": 0.423828125, | |
| "learning_rate": 0.00013754789841833055, | |
| "loss": 0.415, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.3228699551569507, | |
| "grad_norm": 0.4453125, | |
| "learning_rate": 0.00013589838655701027, | |
| "loss": 0.4167, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.3452914798206277, | |
| "grad_norm": 0.44140625, | |
| "learning_rate": 0.00013423674964420517, | |
| "loss": 0.4098, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.3677130044843049, | |
| "grad_norm": 0.458984375, | |
| "learning_rate": 0.00013256400762115845, | |
| "loss": 0.4156, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 1.390134529147982, | |
| "grad_norm": 0.470703125, | |
| "learning_rate": 0.00013088118724562047, | |
| "loss": 0.4148, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.4125560538116593, | |
| "grad_norm": 0.453125, | |
| "learning_rate": 0.00012918932146160724, | |
| "loss": 0.416, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.4349775784753362, | |
| "grad_norm": 0.474609375, | |
| "learning_rate": 0.0001274894487653617, | |
| "loss": 0.4152, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.4573991031390134, | |
| "grad_norm": 0.44140625, | |
| "learning_rate": 0.00012578261256790684, | |
| "loss": 0.4252, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.4798206278026906, | |
| "grad_norm": 0.443359375, | |
| "learning_rate": 0.00012406986055458216, | |
| "loss": 0.4119, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.5022421524663678, | |
| "grad_norm": 0.427734375, | |
| "learning_rate": 0.0001223522440419565, | |
| "loss": 0.4106, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 1.5246636771300448, | |
| "grad_norm": 0.423828125, | |
| "learning_rate": 0.00012063081733251175, | |
| "loss": 0.412, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.547085201793722, | |
| "grad_norm": 0.423828125, | |
| "learning_rate": 0.00011890663706749399, | |
| "loss": 0.4008, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 1.5695067264573992, | |
| "grad_norm": 0.44140625, | |
| "learning_rate": 0.00011718076157832895, | |
| "loss": 0.4067, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.5919282511210762, | |
| "grad_norm": 0.470703125, | |
| "learning_rate": 0.00011545425023700002, | |
| "loss": 0.4087, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 1.6143497757847534, | |
| "grad_norm": 0.4296875, | |
| "learning_rate": 0.00011372816280578757, | |
| "loss": 0.4006, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.6367713004484306, | |
| "grad_norm": 0.412109375, | |
| "learning_rate": 0.00011200355878676867, | |
| "loss": 0.4059, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 1.6591928251121075, | |
| "grad_norm": 0.43359375, | |
| "learning_rate": 0.0001102814967714765, | |
| "loss": 0.4132, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.6816143497757847, | |
| "grad_norm": 0.4453125, | |
| "learning_rate": 0.0001085630337911188, | |
| "loss": 0.4101, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.704035874439462, | |
| "grad_norm": 0.423828125, | |
| "learning_rate": 0.00010684922466775389, | |
| "loss": 0.4049, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.726457399103139, | |
| "grad_norm": 0.40234375, | |
| "learning_rate": 0.00010514112136682282, | |
| "loss": 0.4077, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 1.7488789237668163, | |
| "grad_norm": 0.447265625, | |
| "learning_rate": 0.00010343977235143499, | |
| "loss": 0.4049, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.7713004484304933, | |
| "grad_norm": 0.416015625, | |
| "learning_rate": 0.00010174622193880337, | |
| "loss": 0.3886, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 1.7937219730941703, | |
| "grad_norm": 0.40625, | |
| "learning_rate": 0.0001000615096592249, | |
| "loss": 0.3985, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.8161434977578477, | |
| "grad_norm": 0.416015625, | |
| "learning_rate": 9.83866696179987e-05, | |
| "loss": 0.3987, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 1.8385650224215246, | |
| "grad_norm": 0.40625, | |
| "learning_rate": 9.672272986067473e-05, | |
| "loss": 0.4011, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.8609865470852018, | |
| "grad_norm": 0.388671875, | |
| "learning_rate": 9.507071174202163e-05, | |
| "loss": 0.3897, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 1.883408071748879, | |
| "grad_norm": 0.42578125, | |
| "learning_rate": 9.343162929910169e-05, | |
| "loss": 0.3872, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.905829596412556, | |
| "grad_norm": 0.431640625, | |
| "learning_rate": 9.180648862883761e-05, | |
| "loss": 0.3903, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.9282511210762332, | |
| "grad_norm": 0.4296875, | |
| "learning_rate": 9.019628727045275e-05, | |
| "loss": 0.4004, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.9506726457399104, | |
| "grad_norm": 0.390625, | |
| "learning_rate": 8.860201359316462e-05, | |
| "loss": 0.3857, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 1.9730941704035874, | |
| "grad_norm": 0.412109375, | |
| "learning_rate": 8.702464618950674e-05, | |
| "loss": 0.3916, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.9910313901345291, | |
| "eval_loss": 0.5398380160331726, | |
| "eval_runtime": 1.9374, | |
| "eval_samples_per_second": 21.679, | |
| "eval_steps_per_second": 21.679, | |
| "step": 444 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 669, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.145669325918044e+17, | |
| "train_batch_size": 100, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |