{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.9992659652556888, "eval_steps": 500, "global_step": 1021, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.009787129924149743, "grad_norm": 90.15490074334751, "learning_rate": 4.854368932038835e-07, "loss": 1.5489, "step": 10 }, { "epoch": 0.019574259848299486, "grad_norm": 58.23453270743763, "learning_rate": 9.70873786407767e-07, "loss": 1.4928, "step": 20 }, { "epoch": 0.029361389772449228, "grad_norm": 16.767612840452053, "learning_rate": 1.4563106796116506e-06, "loss": 1.3729, "step": 30 }, { "epoch": 0.03914851969659897, "grad_norm": 11.079321586087117, "learning_rate": 1.941747572815534e-06, "loss": 1.2475, "step": 40 }, { "epoch": 0.048935649620748714, "grad_norm": 6.953520974579646, "learning_rate": 2.427184466019418e-06, "loss": 1.1889, "step": 50 }, { "epoch": 0.058722779544898455, "grad_norm": 3.390499745911967, "learning_rate": 2.912621359223301e-06, "loss": 1.1427, "step": 60 }, { "epoch": 0.0685099094690482, "grad_norm": 2.234261881443784, "learning_rate": 3.398058252427185e-06, "loss": 1.0896, "step": 70 }, { "epoch": 0.07829703939319795, "grad_norm": 2.5807948782523944, "learning_rate": 3.883495145631068e-06, "loss": 1.104, "step": 80 }, { "epoch": 0.0880841693173477, "grad_norm": 2.356465079616129, "learning_rate": 4.368932038834952e-06, "loss": 1.0746, "step": 90 }, { "epoch": 0.09787129924149743, "grad_norm": 2.4234601720621023, "learning_rate": 4.854368932038836e-06, "loss": 1.0793, "step": 100 }, { "epoch": 0.10765842916564718, "grad_norm": 2.117132974644545, "learning_rate": 4.999282701856139e-06, "loss": 1.0692, "step": 110 }, { "epoch": 0.11744555908979691, "grad_norm": 1.8840717844600037, "learning_rate": 4.995770395678171e-06, "loss": 1.0731, "step": 120 }, { "epoch": 0.12723268901394666, "grad_norm": 2.5216433561534317, "learning_rate": 4.989335440737587e-06, "loss": 1.0464, "step": 130 }, { "epoch": 0.1370198189380964, "grad_norm": 2.4579666246587424, "learning_rate": 4.979985372628657e-06, "loss": 1.0434, "step": 140 }, { "epoch": 0.14680694886224616, "grad_norm": 2.3287908973842755, "learning_rate": 4.967731140661878e-06, "loss": 1.0441, "step": 150 }, { "epoch": 0.1565940787863959, "grad_norm": 2.0436369039725224, "learning_rate": 4.952587095041882e-06, "loss": 1.0386, "step": 160 }, { "epoch": 0.16638120871054562, "grad_norm": 2.5733520388523945, "learning_rate": 4.934570970062765e-06, "loss": 1.0565, "step": 170 }, { "epoch": 0.1761683386346954, "grad_norm": 1.8256543734485644, "learning_rate": 4.913703863340504e-06, "loss": 1.0441, "step": 180 }, { "epoch": 0.18595546855884512, "grad_norm": 1.7844329038469422, "learning_rate": 4.890010211106795e-06, "loss": 1.0396, "step": 190 }, { "epoch": 0.19574259848299486, "grad_norm": 3.2695230626975613, "learning_rate": 4.86351775959322e-06, "loss": 1.0295, "step": 200 }, { "epoch": 0.20552972840714462, "grad_norm": 2.6115311678055653, "learning_rate": 4.834257532539292e-06, "loss": 1.0465, "step": 210 }, { "epoch": 0.21531685833129435, "grad_norm": 2.422852080700085, "learning_rate": 4.802263794862385e-06, "loss": 1.017, "step": 220 }, { "epoch": 0.2251039882554441, "grad_norm": 2.2578865515295283, "learning_rate": 4.767574012532121e-06, "loss": 1.0176, "step": 230 }, { "epoch": 0.23489111817959382, "grad_norm": 2.087619836151659, "learning_rate": 4.730228808696201e-06, "loss": 1.0174, "step": 240 }, { "epoch": 0.24467824810374358, "grad_norm": 1.999324934826496, "learning_rate": 4.690271916109034e-06, "loss": 1.0174, "step": 250 }, { "epoch": 0.2544653780278933, "grad_norm": 3.372686449261057, "learning_rate": 4.647750125918909e-06, "loss": 1.0168, "step": 260 }, { "epoch": 0.2642525079520431, "grad_norm": 2.5211594090463088, "learning_rate": 4.6027132328736515e-06, "loss": 1.0351, "step": 270 }, { "epoch": 0.2740396378761928, "grad_norm": 3.6300783567036525, "learning_rate": 4.555213977008946e-06, "loss": 1.0105, "step": 280 }, { "epoch": 0.28382676780034255, "grad_norm": 1.6572768717333426, "learning_rate": 4.50530798188761e-06, "loss": 1.0227, "step": 290 }, { "epoch": 0.2936138977244923, "grad_norm": 2.021540908150653, "learning_rate": 4.453053689462131e-06, "loss": 1.0178, "step": 300 }, { "epoch": 0.303401027648642, "grad_norm": 1.8651477439042456, "learning_rate": 4.398512291636768e-06, "loss": 1.0091, "step": 310 }, { "epoch": 0.3131881575727918, "grad_norm": 2.43880182053301, "learning_rate": 4.341747658609332e-06, "loss": 1.0196, "step": 320 }, { "epoch": 0.32297528749694154, "grad_norm": 2.8064493559277963, "learning_rate": 4.282826264076587e-06, "loss": 1.0142, "step": 330 }, { "epoch": 0.33276241742109125, "grad_norm": 1.9574404119807658, "learning_rate": 4.221817107390847e-06, "loss": 1.0014, "step": 340 }, { "epoch": 0.342549547345241, "grad_norm": 2.163660779817407, "learning_rate": 4.15879163275892e-06, "loss": 1.0297, "step": 350 }, { "epoch": 0.3523366772693908, "grad_norm": 1.9653596257387018, "learning_rate": 4.093823645578037e-06, "loss": 1.0155, "step": 360 }, { "epoch": 0.3621238071935405, "grad_norm": 2.2959542048619253, "learning_rate": 4.02698922600672e-06, "loss": 1.0136, "step": 370 }, { "epoch": 0.37191093711769024, "grad_norm": 1.7061830149355224, "learning_rate": 3.958366639871826e-06, "loss": 0.9889, "step": 380 }, { "epoch": 0.38169806704184, "grad_norm": 2.299910460880925, "learning_rate": 3.888036247016073e-06, "loss": 0.9959, "step": 390 }, { "epoch": 0.3914851969659897, "grad_norm": 1.9452961119067358, "learning_rate": 3.81608040719339e-06, "loss": 1.0049, "step": 400 }, { "epoch": 0.4012723268901395, "grad_norm": 2.744356349117842, "learning_rate": 3.7425833836222947e-06, "loss": 1.0018, "step": 410 }, { "epoch": 0.41105945681428924, "grad_norm": 2.477879387395352, "learning_rate": 3.6676312443102323e-06, "loss": 1.0152, "step": 420 }, { "epoch": 0.42084658673843894, "grad_norm": 2.1636493383526676, "learning_rate": 3.591311761264433e-06, "loss": 1.0056, "step": 430 }, { "epoch": 0.4306337166625887, "grad_norm": 2.743033130906874, "learning_rate": 3.513714307707321e-06, "loss": 0.9988, "step": 440 }, { "epoch": 0.4404208465867384, "grad_norm": 2.3096838919036093, "learning_rate": 3.4349297534168242e-06, "loss": 0.9973, "step": 450 }, { "epoch": 0.4502079765108882, "grad_norm": 2.7080342141867586, "learning_rate": 3.3550503583141726e-06, "loss": 0.9841, "step": 460 }, { "epoch": 0.45999510643503794, "grad_norm": 1.8384006858146424, "learning_rate": 3.274169664423768e-06, "loss": 0.9991, "step": 470 }, { "epoch": 0.46978223635918764, "grad_norm": 1.7788391832635966, "learning_rate": 3.192382386331667e-06, "loss": 0.9992, "step": 480 }, { "epoch": 0.4795693662833374, "grad_norm": 1.9349340294062531, "learning_rate": 3.109784300270943e-06, "loss": 1.004, "step": 490 }, { "epoch": 0.48935649620748717, "grad_norm": 2.089365671821631, "learning_rate": 3.0264721319638176e-06, "loss": 0.9892, "step": 500 }, { "epoch": 0.4991436261316369, "grad_norm": 1.6583587633830374, "learning_rate": 2.9425434433518985e-06, "loss": 0.9991, "step": 510 }, { "epoch": 0.5089307560557866, "grad_norm": 2.164226641765232, "learning_rate": 2.8580965183471794e-06, "loss": 0.9946, "step": 520 }, { "epoch": 0.5187178859799364, "grad_norm": 2.595883151442786, "learning_rate": 2.773230247737569e-06, "loss": 1.0007, "step": 530 }, { "epoch": 0.5285050159040862, "grad_norm": 1.9135033210807688, "learning_rate": 2.6880440133817563e-06, "loss": 0.997, "step": 540 }, { "epoch": 0.5382921458282359, "grad_norm": 2.1997366611393776, "learning_rate": 2.602637571829009e-06, "loss": 0.9916, "step": 550 }, { "epoch": 0.5480792757523856, "grad_norm": 4.036295629554255, "learning_rate": 2.517110937500185e-06, "loss": 0.9921, "step": 560 }, { "epoch": 0.5578664056765353, "grad_norm": 1.9466096395859314, "learning_rate": 2.431564265566781e-06, "loss": 0.9998, "step": 570 }, { "epoch": 0.5676535356006851, "grad_norm": 2.487288664151329, "learning_rate": 2.346097734665143e-06, "loss": 0.9871, "step": 580 }, { "epoch": 0.5774406655248349, "grad_norm": 1.816928462189231, "learning_rate": 2.2608114295832053e-06, "loss": 0.9979, "step": 590 }, { "epoch": 0.5872277954489846, "grad_norm": 1.75653406126234, "learning_rate": 2.175805224057129e-06, "loss": 0.9724, "step": 600 }, { "epoch": 0.5970149253731343, "grad_norm": 1.7976791598972546, "learning_rate": 2.0911786638150873e-06, "loss": 0.9937, "step": 610 }, { "epoch": 0.606802055297284, "grad_norm": 1.9243691484831966, "learning_rate": 2.0070308500051715e-06, "loss": 0.9859, "step": 620 }, { "epoch": 0.6165891852214338, "grad_norm": 2.7481202060087835, "learning_rate": 1.9234603231439e-06, "loss": 0.9841, "step": 630 }, { "epoch": 0.6263763151455836, "grad_norm": 2.003686173024267, "learning_rate": 1.84056494772127e-06, "loss": 0.9749, "step": 640 }, { "epoch": 0.6361634450697333, "grad_norm": 1.6161005997468445, "learning_rate": 1.7584417975974535e-06, "loss": 0.985, "step": 650 }, { "epoch": 0.6459505749938831, "grad_norm": 1.5994212448277128, "learning_rate": 1.6771870423253473e-06, "loss": 0.9823, "step": 660 }, { "epoch": 0.6557377049180327, "grad_norm": 2.947926766692084, "learning_rate": 1.5968958345321178e-06, "loss": 0.9905, "step": 670 }, { "epoch": 0.6655248348421825, "grad_norm": 2.496926900199072, "learning_rate": 1.517662198491599e-06, "loss": 0.9802, "step": 680 }, { "epoch": 0.6753119647663323, "grad_norm": 2.50313272426719, "learning_rate": 1.4395789200180343e-06, "loss": 0.9839, "step": 690 }, { "epoch": 0.685099094690482, "grad_norm": 3.162725883476327, "learning_rate": 1.362737437810114e-06, "loss": 0.9746, "step": 700 }, { "epoch": 0.6948862246146318, "grad_norm": 2.5094677251656696, "learning_rate": 1.287227736372538e-06, "loss": 0.9662, "step": 710 }, { "epoch": 0.7046733545387815, "grad_norm": 1.7194609557385765, "learning_rate": 1.2131382406404866e-06, "loss": 0.9654, "step": 720 }, { "epoch": 0.7144604844629312, "grad_norm": 2.088741834771206, "learning_rate": 1.1405557124304338e-06, "loss": 0.9687, "step": 730 }, { "epoch": 0.724247614387081, "grad_norm": 2.3589729813581637, "learning_rate": 1.0695651488385168e-06, "loss": 0.9764, "step": 740 }, { "epoch": 0.7340347443112307, "grad_norm": 1.4080735212450242, "learning_rate": 1.0002496827054806e-06, "loss": 0.9734, "step": 750 }, { "epoch": 0.7438218742353805, "grad_norm": 2.028266231817523, "learning_rate": 9.326904852647345e-07, "loss": 0.9731, "step": 760 }, { "epoch": 0.7536090041595302, "grad_norm": 1.9325878087268091, "learning_rate": 8.669666710875319e-07, "loss": 0.9717, "step": 770 }, { "epoch": 0.76339613408368, "grad_norm": 2.6471850221409583, "learning_rate": 8.031552054365905e-07, "loss": 0.9721, "step": 780 }, { "epoch": 0.7731832640078297, "grad_norm": 1.8267904470848466, "learning_rate": 7.413308141366254e-07, "loss": 0.9925, "step": 790 }, { "epoch": 0.7829703939319794, "grad_norm": 1.7349835377240377, "learning_rate": 6.815658960673782e-07, "loss": 0.9581, "step": 800 }, { "epoch": 0.7927575238561292, "grad_norm": 2.037859066803747, "learning_rate": 6.239304383815706e-07, "loss": 0.9699, "step": 810 }, { "epoch": 0.802544653780279, "grad_norm": 2.1130579933367257, "learning_rate": 5.684919345471029e-07, "loss": 0.9905, "step": 820 }, { "epoch": 0.8123317837044287, "grad_norm": 1.946465156584691, "learning_rate": 5.15315305309455e-07, "loss": 0.973, "step": 830 }, { "epoch": 0.8221189136285785, "grad_norm": 2.5768060337551972, "learning_rate": 4.644628226668485e-07, "loss": 0.9582, "step": 840 }, { "epoch": 0.8319060435527281, "grad_norm": 2.173860689008899, "learning_rate": 4.159940369472015e-07, "loss": 0.9729, "step": 850 }, { "epoch": 0.8416931734768779, "grad_norm": 2.291543898620316, "learning_rate": 3.699657070722698e-07, "loss": 0.9761, "step": 860 }, { "epoch": 0.8514803034010276, "grad_norm": 2.088024718750774, "learning_rate": 3.2643173409063976e-07, "loss": 0.9818, "step": 870 }, { "epoch": 0.8612674333251774, "grad_norm": 1.8060057642166407, "learning_rate": 2.854430980574002e-07, "loss": 0.9636, "step": 880 }, { "epoch": 0.8710545632493272, "grad_norm": 1.749866063660457, "learning_rate": 2.4704779833442993e-07, "loss": 0.9636, "step": 890 }, { "epoch": 0.8808416931734768, "grad_norm": 2.0741463604546913, "learning_rate": 2.1129079738118424e-07, "loss": 0.963, "step": 900 }, { "epoch": 0.8906288230976266, "grad_norm": 1.4388364613303446, "learning_rate": 1.782139681018244e-07, "loss": 0.9757, "step": 910 }, { "epoch": 0.9004159530217763, "grad_norm": 2.3114929534129343, "learning_rate": 1.4785604481034639e-07, "loss": 0.9678, "step": 920 }, { "epoch": 0.9102030829459261, "grad_norm": 2.0415525131838432, "learning_rate": 1.202525778711172e-07, "loss": 0.9595, "step": 930 }, { "epoch": 0.9199902128700759, "grad_norm": 2.3293806080394104, "learning_rate": 9.54358920679524e-08, "loss": 0.98, "step": 940 }, { "epoch": 0.9297773427942256, "grad_norm": 2.0830083897095406, "learning_rate": 7.343504875047814e-08, "loss": 0.964, "step": 950 }, { "epoch": 0.9395644727183753, "grad_norm": 2.218326205399271, "learning_rate": 5.427581180210639e-08, "loss": 0.979, "step": 960 }, { "epoch": 0.949351602642525, "grad_norm": 2.153253403337433, "learning_rate": 3.798061746947995e-08, "loss": 0.9578, "step": 970 }, { "epoch": 0.9591387325666748, "grad_norm": 1.611398342662154, "learning_rate": 2.456854808871201e-08, "loss": 0.9759, "step": 980 }, { "epoch": 0.9689258624908246, "grad_norm": 2.2214603407537883, "learning_rate": 1.4055309739195166e-08, "loss": 0.9691, "step": 990 }, { "epoch": 0.9787129924149743, "grad_norm": 1.8872382897952824, "learning_rate": 6.453213851142226e-09, "loss": 0.9866, "step": 1000 }, { "epoch": 0.9885001223391241, "grad_norm": 3.466854317296078, "learning_rate": 1.7711627883998383e-09, "loss": 0.9735, "step": 1010 }, { "epoch": 0.9982872522632737, "grad_norm": 1.8136326675400594, "learning_rate": 1.463942341850544e-11, "loss": 0.9689, "step": 1020 }, { "epoch": 0.9992659652556888, "step": 1021, "total_flos": 1.136081069573931e+18, "train_loss": 1.019411426333092, "train_runtime": 37962.3304, "train_samples_per_second": 3.444, "train_steps_per_second": 0.027 } ], "logging_steps": 10, "max_steps": 1021, "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": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 1.136081069573931e+18, "train_batch_size": 4, "trial_name": null, "trial_params": null }