{ "best_metric": null, "best_model_checkpoint": null, "epoch": 2.994667614646285, "eval_steps": 1000, "global_step": 1053, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.028439388553146108, "grad_norm": 1.0473670959472656, "learning_rate": 9.433962264150944e-06, "loss": 2.6607, "step": 10 }, { "epoch": 0.056878777106292217, "grad_norm": 0.4230079650878906, "learning_rate": 1.8867924528301888e-05, "loss": 2.5121, "step": 20 }, { "epoch": 0.08531816565943832, "grad_norm": 0.35547205805778503, "learning_rate": 2.830188679245283e-05, "loss": 2.3998, "step": 30 }, { "epoch": 0.11375755421258443, "grad_norm": 0.3876227140426636, "learning_rate": 3.7735849056603776e-05, "loss": 2.3461, "step": 40 }, { "epoch": 0.14219694276573053, "grad_norm": 0.38349416851997375, "learning_rate": 4.716981132075472e-05, "loss": 2.2716, "step": 50 }, { "epoch": 0.17063633131887665, "grad_norm": 0.4068184792995453, "learning_rate": 4.999395511092461e-05, "loss": 2.2223, "step": 60 }, { "epoch": 0.19907571987202274, "grad_norm": 0.4073835015296936, "learning_rate": 4.996435452798774e-05, "loss": 2.1657, "step": 70 }, { "epoch": 0.22751510842516887, "grad_norm": 0.5059598684310913, "learning_rate": 4.991011714111481e-05, "loss": 2.1248, "step": 80 }, { "epoch": 0.25595449697831496, "grad_norm": 0.47395384311676025, "learning_rate": 4.9831296476058484e-05, "loss": 2.1223, "step": 90 }, { "epoch": 0.28439388553146105, "grad_norm": 0.43479716777801514, "learning_rate": 4.9727970319299044e-05, "loss": 2.0617, "step": 100 }, { "epoch": 0.3128332740846072, "grad_norm": 0.48797106742858887, "learning_rate": 4.9600240641278496e-05, "loss": 2.0526, "step": 110 }, { "epoch": 0.3412726626377533, "grad_norm": 0.5094576478004456, "learning_rate": 4.944823349576805e-05, "loss": 2.0356, "step": 120 }, { "epoch": 0.3697120511908994, "grad_norm": 0.5222152471542358, "learning_rate": 4.9272098895468277e-05, "loss": 2.0315, "step": 130 }, { "epoch": 0.3981514397440455, "grad_norm": 0.48217442631721497, "learning_rate": 4.907201066396469e-05, "loss": 2.013, "step": 140 }, { "epoch": 0.42659082829719164, "grad_norm": 0.46154898405075073, "learning_rate": 4.8848166264184844e-05, "loss": 1.9751, "step": 150 }, { "epoch": 0.45503021685033773, "grad_norm": 0.4828225374221802, "learning_rate": 4.860078660352625e-05, "loss": 1.9866, "step": 160 }, { "epoch": 0.4834696054034838, "grad_norm": 0.4831569194793701, "learning_rate": 4.8330115815847465e-05, "loss": 1.9529, "step": 170 }, { "epoch": 0.5119089939566299, "grad_norm": 0.4747070074081421, "learning_rate": 4.803642102053746e-05, "loss": 1.9575, "step": 180 }, { "epoch": 0.540348382509776, "grad_norm": 0.4549647271633148, "learning_rate": 4.7719992058901006e-05, "loss": 1.9488, "step": 190 }, { "epoch": 0.5687877710629221, "grad_norm": 0.4767439067363739, "learning_rate": 4.7381141208120296e-05, "loss": 1.9074, "step": 200 }, { "epoch": 0.5972271596160682, "grad_norm": 0.4805672764778137, "learning_rate": 4.702020287307509e-05, "loss": 1.9286, "step": 210 }, { "epoch": 0.6256665481692144, "grad_norm": 0.47634372115135193, "learning_rate": 4.663753325632548e-05, "loss": 1.924, "step": 220 }, { "epoch": 0.6541059367223605, "grad_norm": 0.4570080637931824, "learning_rate": 4.6233510006582914e-05, "loss": 1.8953, "step": 230 }, { "epoch": 0.6825453252755066, "grad_norm": 0.47734373807907104, "learning_rate": 4.580853184601659e-05, "loss": 1.9023, "step": 240 }, { "epoch": 0.7109847138286527, "grad_norm": 0.48877596855163574, "learning_rate": 4.536301817676274e-05, "loss": 1.8832, "step": 250 }, { "epoch": 0.7394241023817988, "grad_norm": 0.46437644958496094, "learning_rate": 4.48974086670254e-05, "loss": 1.9052, "step": 260 }, { "epoch": 0.7678634909349449, "grad_norm": 0.4709625840187073, "learning_rate": 4.4412162817176965e-05, "loss": 1.867, "step": 270 }, { "epoch": 0.796302879488091, "grad_norm": 0.5039490461349487, "learning_rate": 4.39077595062868e-05, "loss": 1.8549, "step": 280 }, { "epoch": 0.8247422680412371, "grad_norm": 0.4863899350166321, "learning_rate": 4.33846965195254e-05, "loss": 1.8589, "step": 290 }, { "epoch": 0.8531816565943833, "grad_norm": 0.4797697961330414, "learning_rate": 4.2843490056910534e-05, "loss": 1.8624, "step": 300 }, { "epoch": 0.8816210451475294, "grad_norm": 0.5284737348556519, "learning_rate": 4.228467422388016e-05, "loss": 1.878, "step": 310 }, { "epoch": 0.9100604337006755, "grad_norm": 0.45375633239746094, "learning_rate": 4.1708800504194827e-05, "loss": 1.8837, "step": 320 }, { "epoch": 0.9384998222538216, "grad_norm": 0.4459105432033539, "learning_rate": 4.1116437215689784e-05, "loss": 1.8472, "step": 330 }, { "epoch": 0.9669392108069677, "grad_norm": 0.48929303884506226, "learning_rate": 4.0508168949413906e-05, "loss": 1.8261, "step": 340 }, { "epoch": 0.9953785993601137, "grad_norm": 0.4859321415424347, "learning_rate": 3.988459599270888e-05, "loss": 1.8697, "step": 350 }, { "epoch": 1.0238179879132598, "grad_norm": 0.481235533952713, "learning_rate": 3.9246333736798095e-05, "loss": 1.8783, "step": 360 }, { "epoch": 1.052257376466406, "grad_norm": 0.5238078832626343, "learning_rate": 3.859401206946982e-05, "loss": 1.7799, "step": 370 }, { "epoch": 1.080696765019552, "grad_norm": 0.5101935863494873, "learning_rate": 3.792827475345393e-05, "loss": 1.8068, "step": 380 }, { "epoch": 1.1091361535726982, "grad_norm": 0.4812924265861511, "learning_rate": 3.724977879110591e-05, "loss": 1.7974, "step": 390 }, { "epoch": 1.1375755421258442, "grad_norm": 0.5056812763214111, "learning_rate": 3.6559193776024794e-05, "loss": 1.7918, "step": 400 }, { "epoch": 1.1660149306789904, "grad_norm": 0.5049041509628296, "learning_rate": 3.585720123224512e-05, "loss": 1.7856, "step": 410 }, { "epoch": 1.1944543192321366, "grad_norm": 0.5172712802886963, "learning_rate": 3.5144493941655e-05, "loss": 1.786, "step": 420 }, { "epoch": 1.2228937077852826, "grad_norm": 0.48730871081352234, "learning_rate": 3.442177526030407e-05, "loss": 1.7553, "step": 430 }, { "epoch": 1.2513330963384286, "grad_norm": 0.5034505724906921, "learning_rate": 3.3689758424275926e-05, "loss": 1.7567, "step": 440 }, { "epoch": 1.2797724848915748, "grad_norm": 0.5002079606056213, "learning_rate": 3.294916584581027e-05, "loss": 1.7851, "step": 450 }, { "epoch": 1.308211873444721, "grad_norm": 0.5162603855133057, "learning_rate": 3.220072840036923e-05, "loss": 1.7631, "step": 460 }, { "epoch": 1.336651261997867, "grad_norm": 0.5188742876052856, "learning_rate": 3.14451847053515e-05, "loss": 1.7645, "step": 470 }, { "epoch": 1.3650906505510132, "grad_norm": 0.5118332505226135, "learning_rate": 3.068328039116616e-05, "loss": 1.7693, "step": 480 }, { "epoch": 1.3935300391041592, "grad_norm": 0.5362988710403442, "learning_rate": 2.99157673653855e-05, "loss": 1.7712, "step": 490 }, { "epoch": 1.4219694276573054, "grad_norm": 0.5271508097648621, "learning_rate": 2.9143403070702997e-05, "loss": 1.7613, "step": 500 }, { "epoch": 1.4504088162104516, "grad_norm": 0.5073094964027405, "learning_rate": 2.8366949737428817e-05, "loss": 1.7729, "step": 510 }, { "epoch": 1.4788482047635976, "grad_norm": 0.5312384366989136, "learning_rate": 2.7587173631260566e-05, "loss": 1.7649, "step": 520 }, { "epoch": 1.5072875933167436, "grad_norm": 0.5345740914344788, "learning_rate": 2.6804844297071526e-05, "loss": 1.747, "step": 530 }, { "epoch": 1.5357269818698898, "grad_norm": 0.5181931257247925, "learning_rate": 2.6020733799462754e-05, "loss": 1.7496, "step": 540 }, { "epoch": 1.564166370423036, "grad_norm": 0.5075932741165161, "learning_rate": 2.5235615960828605e-05, "loss": 1.7517, "step": 550 }, { "epoch": 1.5926057589761822, "grad_norm": 0.5280550122261047, "learning_rate": 2.4450265597687376e-05, "loss": 1.7768, "step": 560 }, { "epoch": 1.6210451475293282, "grad_norm": 0.526050865650177, "learning_rate": 2.3665457756030988e-05, "loss": 1.7369, "step": 570 }, { "epoch": 1.6494845360824741, "grad_norm": 0.5144099593162537, "learning_rate": 2.2881966946448167e-05, "loss": 1.7389, "step": 580 }, { "epoch": 1.6779239246356203, "grad_norm": 0.532425045967102, "learning_rate": 2.2100566379775967e-05, "loss": 1.7364, "step": 590 }, { "epoch": 1.7063633131887666, "grad_norm": 0.5095171332359314, "learning_rate": 2.1322027204034066e-05, "loss": 1.7447, "step": 600 }, { "epoch": 1.7348027017419125, "grad_norm": 0.5283172726631165, "learning_rate": 2.0547117743394744e-05, "loss": 1.7647, "step": 610 }, { "epoch": 1.7632420902950585, "grad_norm": 0.5820568203926086, "learning_rate": 1.9776602739939714e-05, "loss": 1.7343, "step": 620 }, { "epoch": 1.7916814788482047, "grad_norm": 0.5260512232780457, "learning_rate": 1.9011242598951962e-05, "loss": 1.7402, "step": 630 }, { "epoch": 1.820120867401351, "grad_norm": 0.5743537545204163, "learning_rate": 1.8251792638487596e-05, "loss": 1.7421, "step": 640 }, { "epoch": 1.8485602559544971, "grad_norm": 0.508335530757904, "learning_rate": 1.7499002343968098e-05, "loss": 1.7303, "step": 650 }, { "epoch": 1.8769996445076431, "grad_norm": 0.5431790947914124, "learning_rate": 1.675361462852868e-05, "loss": 1.7236, "step": 660 }, { "epoch": 1.905439033060789, "grad_norm": 0.5047234892845154, "learning_rate": 1.6016365099852735e-05, "loss": 1.7163, "step": 670 }, { "epoch": 1.9338784216139353, "grad_norm": 0.5409323573112488, "learning_rate": 1.528798133421585e-05, "loss": 1.7253, "step": 680 }, { "epoch": 1.9623178101670815, "grad_norm": 0.5161160826683044, "learning_rate": 1.4569182158455875e-05, "loss": 1.7054, "step": 690 }, { "epoch": 1.9907571987202275, "grad_norm": 0.5363497734069824, "learning_rate": 1.3860676940577594e-05, "loss": 1.7469, "step": 700 }, { "epoch": 2.0191965872733735, "grad_norm": 0.5758066177368164, "learning_rate": 1.3163164889692197e-05, "loss": 1.7598, "step": 710 }, { "epoch": 2.0476359758265197, "grad_norm": 0.5709594488143921, "learning_rate": 1.2477334365982248e-05, "loss": 1.6736, "step": 720 }, { "epoch": 2.076075364379666, "grad_norm": 0.5616790056228638, "learning_rate": 1.1803862201373342e-05, "loss": 1.6529, "step": 730 }, { "epoch": 2.104514752932812, "grad_norm": 0.5473997592926025, "learning_rate": 1.1143413031582645e-05, "loss": 1.6734, "step": 740 }, { "epoch": 2.132954141485958, "grad_norm": 0.559503436088562, "learning_rate": 1.0496638640203774e-05, "loss": 1.6702, "step": 750 }, { "epoch": 2.161393530039104, "grad_norm": 0.5698120594024658, "learning_rate": 9.864177315474968e-06, "loss": 1.6613, "step": 760 }, { "epoch": 2.1898329185922503, "grad_norm": 0.5590789318084717, "learning_rate": 9.246653220365778e-06, "loss": 1.6898, "step": 770 }, { "epoch": 2.2182723071453965, "grad_norm": 0.5687025785446167, "learning_rate": 8.644675776603476e-06, "loss": 1.6688, "step": 780 }, { "epoch": 2.2467116956985427, "grad_norm": 0.5703499913215637, "learning_rate": 8.058839063247447e-06, "loss": 1.6761, "step": 790 }, { "epoch": 2.2751510842516884, "grad_norm": 0.586949348449707, "learning_rate": 7.489721230404842e-06, "loss": 1.7022, "step": 800 }, { "epoch": 2.3035904728048346, "grad_norm": 0.5651352405548096, "learning_rate": 6.937883928666255e-06, "loss": 1.6778, "step": 810 }, { "epoch": 2.332029861357981, "grad_norm": 0.5906292200088501, "learning_rate": 6.403871754824373e-06, "loss": 1.6517, "step": 820 }, { "epoch": 2.360469249911127, "grad_norm": 0.5772544145584106, "learning_rate": 5.8882117144227115e-06, "loss": 1.6696, "step": 830 }, { "epoch": 2.3889086384642733, "grad_norm": 0.5833613872528076, "learning_rate": 5.391412701664744e-06, "loss": 1.6452, "step": 840 }, { "epoch": 2.417348027017419, "grad_norm": 0.5983288884162903, "learning_rate": 4.91396499719681e-06, "loss": 1.6739, "step": 850 }, { "epoch": 2.4457874155705652, "grad_norm": 0.5902206897735596, "learning_rate": 4.456339784260247e-06, "loss": 1.6582, "step": 860 }, { "epoch": 2.4742268041237114, "grad_norm": 0.5683711767196655, "learning_rate": 4.018988683690461e-06, "loss": 1.6483, "step": 870 }, { "epoch": 2.502666192676857, "grad_norm": 0.5805079936981201, "learning_rate": 3.6023433082216755e-06, "loss": 1.6731, "step": 880 }, { "epoch": 2.5311055812300034, "grad_norm": 0.5799385905265808, "learning_rate": 3.2068148365372806e-06, "loss": 1.6705, "step": 890 }, { "epoch": 2.5595449697831496, "grad_norm": 0.6097893118858337, "learning_rate": 2.832793607486087e-06, "loss": 1.6851, "step": 900 }, { "epoch": 2.587984358336296, "grad_norm": 0.5920665860176086, "learning_rate": 2.4806487348650485e-06, "loss": 1.6464, "step": 910 }, { "epoch": 2.616423746889442, "grad_norm": 0.5650290846824646, "learning_rate": 2.150727743148473e-06, "loss": 1.6576, "step": 920 }, { "epoch": 2.6448631354425878, "grad_norm": 0.5883274674415588, "learning_rate": 1.8433562245233349e-06, "loss": 1.6304, "step": 930 }, { "epoch": 2.673302523995734, "grad_norm": 0.5887704491615295, "learning_rate": 1.5588375175691117e-06, "loss": 1.6674, "step": 940 }, { "epoch": 2.70174191254888, "grad_norm": 0.576253354549408, "learning_rate": 1.2974524078991995e-06, "loss": 1.6493, "step": 950 }, { "epoch": 2.7301813011020264, "grad_norm": 0.5727583169937134, "learning_rate": 1.0594588510594445e-06, "loss": 1.6529, "step": 960 }, { "epoch": 2.7586206896551726, "grad_norm": 0.5819412469863892, "learning_rate": 8.450917179571305e-07, "loss": 1.6612, "step": 970 }, { "epoch": 2.7870600782083184, "grad_norm": 0.570379376411438, "learning_rate": 6.545625630717783e-07, "loss": 1.6637, "step": 980 }, { "epoch": 2.8154994667614646, "grad_norm": 0.5768330693244934, "learning_rate": 4.880594156763896e-07, "loss": 1.6784, "step": 990 }, { "epoch": 2.8439388553146108, "grad_norm": 0.5636200904846191, "learning_rate": 3.4574659427528133e-07, "loss": 1.6679, "step": 1000 }, { "epoch": 2.8439388553146108, "eval_loss": 1.758003830909729, "eval_runtime": 181.1788, "eval_samples_per_second": 55.194, "eval_steps_per_second": 1.728, "step": 1000 }, { "epoch": 2.872378243867757, "grad_norm": 0.5933385491371155, "learning_rate": 2.2776454444153328e-07, "loss": 1.6689, "step": 1010 }, { "epoch": 2.900817632420903, "grad_norm": 0.5807960033416748, "learning_rate": 1.342297002141918e-07, "loss": 1.6333, "step": 1020 }, { "epoch": 2.929257020974049, "grad_norm": 0.5805134773254395, "learning_rate": 6.523436919190773e-08, "loss": 1.6701, "step": 1030 }, { "epoch": 2.957696409527195, "grad_norm": 0.5947986841201782, "learning_rate": 2.0846641436497726e-08, "loss": 1.652, "step": 1040 }, { "epoch": 2.9861357980803414, "grad_norm": 0.5900915861129761, "learning_rate": 1.1103222762542941e-09, "loss": 1.6704, "step": 1050 }, { "epoch": 2.994667614646285, "step": 1053, "total_flos": 4.229776342129836e+18, "train_loss": 1.8177586702200084, "train_runtime": 12857.7751, "train_samples_per_second": 20.999, "train_steps_per_second": 0.082 } ], "logging_steps": 10, "max_steps": 1053, "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": 4.229776342129836e+18, "train_batch_size": 4, "trial_name": null, "trial_params": null }