{ "best_metric": null, "best_model_checkpoint": null, "epoch": 1.0, "eval_steps": 500, "global_step": 598, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0016722408026755853, "grad_norm": 5899.944994453189, "learning_rate": 4.9999999999999996e-06, "loss": 22.9452, "step": 1 }, { "epoch": 0.008361204013377926, "grad_norm": 3832.174007639646, "learning_rate": 2.4999999999999998e-05, "loss": 17.2666, "step": 5 }, { "epoch": 0.016722408026755852, "grad_norm": 1842.7301746786366, "learning_rate": 4.9999999999999996e-05, "loss": 16.1655, "step": 10 }, { "epoch": 0.02508361204013378, "grad_norm": 342.882874624202, "learning_rate": 7.5e-05, "loss": 6.9604, "step": 15 }, { "epoch": 0.033444816053511704, "grad_norm": 648.6675239239157, "learning_rate": 9.999999999999999e-05, "loss": 5.362, "step": 20 }, { "epoch": 0.04180602006688963, "grad_norm": 642.7533463105359, "learning_rate": 0.000125, "loss": 7.6165, "step": 25 }, { "epoch": 0.05016722408026756, "grad_norm": 124.26540956463161, "learning_rate": 0.00015, "loss": 5.5364, "step": 30 }, { "epoch": 0.05852842809364549, "grad_norm": 527.1226263500417, "learning_rate": 0.000175, "loss": 6.2687, "step": 35 }, { "epoch": 0.06688963210702341, "grad_norm": 235.15362781581305, "learning_rate": 0.00019999999999999998, "loss": 6.8242, "step": 40 }, { "epoch": 0.07525083612040134, "grad_norm": 232.04999588659172, "learning_rate": 0.000225, "loss": 5.0975, "step": 45 }, { "epoch": 0.08361204013377926, "grad_norm": 193.92802731203975, "learning_rate": 0.00025, "loss": 4.363, "step": 50 }, { "epoch": 0.09197324414715718, "grad_norm": 87.32996989626787, "learning_rate": 0.00027499999999999996, "loss": 4.492, "step": 55 }, { "epoch": 0.10033444816053512, "grad_norm": 157.51219806377262, "learning_rate": 0.0003, "loss": 5.2962, "step": 60 }, { "epoch": 0.10869565217391304, "grad_norm": 178.03521535228302, "learning_rate": 0.0002999360698660697, "loss": 5.2349, "step": 65 }, { "epoch": 0.11705685618729098, "grad_norm": 82.65266770623266, "learning_rate": 0.0002997443339584391, "loss": 4.8057, "step": 70 }, { "epoch": 0.1254180602006689, "grad_norm": 48.21827434938864, "learning_rate": 0.0002994249557131383, "loss": 4.6267, "step": 75 }, { "epoch": 0.13377926421404682, "grad_norm": 101.47726512121915, "learning_rate": 0.00029897820736875393, "loss": 3.7307, "step": 80 }, { "epoch": 0.14214046822742474, "grad_norm": 84.76151041346198, "learning_rate": 0.0002984044697343724, "loss": 3.3221, "step": 85 }, { "epoch": 0.1505016722408027, "grad_norm": 35.2161154461866, "learning_rate": 0.00029770423186497794, "loss": 2.7633, "step": 90 }, { "epoch": 0.1588628762541806, "grad_norm": 39.923892471272296, "learning_rate": 0.0002968780906445808, "loss": 2.1568, "step": 95 }, { "epoch": 0.16722408026755853, "grad_norm": 72.82838678501577, "learning_rate": 0.00029592675027743253, "loss": 2.3427, "step": 100 }, { "epoch": 0.17558528428093645, "grad_norm": 94.37110002387033, "learning_rate": 0.00029485102168776096, "loss": 6.1606, "step": 105 }, { "epoch": 0.18394648829431437, "grad_norm": 373.81535831887, "learning_rate": 0.0002936518218285368, "loss": 10.4362, "step": 110 }, { "epoch": 0.19230769230769232, "grad_norm": 221.23048878415432, "learning_rate": 0.0002923301728998616, "loss": 16.4553, "step": 115 }, { "epoch": 0.20066889632107024, "grad_norm": 395.57153442034314, "learning_rate": 0.0002908872014776423, "loss": 43.516, "step": 120 }, { "epoch": 0.20903010033444816, "grad_norm": 286.615099772232, "learning_rate": 0.0002893241375532959, "loss": 28.7129, "step": 125 }, { "epoch": 0.21739130434782608, "grad_norm": 82.2777934481911, "learning_rate": 0.00028764231348530267, "loss": 27.8137, "step": 130 }, { "epoch": 0.225752508361204, "grad_norm": 170.02097895100943, "learning_rate": 0.00028584316286350164, "loss": 21.8417, "step": 135 }, { "epoch": 0.23411371237458195, "grad_norm": 202.04237746771835, "learning_rate": 0.0002839282192870955, "loss": 19.1217, "step": 140 }, { "epoch": 0.24247491638795987, "grad_norm": 34.74067140580764, "learning_rate": 0.00028189911505740843, "loss": 15.3816, "step": 145 }, { "epoch": 0.2508361204013378, "grad_norm": 59.32768633148153, "learning_rate": 0.0002797575797865093, "loss": 13.5017, "step": 150 }, { "epoch": 0.2591973244147157, "grad_norm": 45.03643469009911, "learning_rate": 0.0002775054389228871, "loss": 11.8208, "step": 155 }, { "epoch": 0.26755852842809363, "grad_norm": 29.78986842292205, "learning_rate": 0.0002751446121954359, "loss": 9.8787, "step": 160 }, { "epoch": 0.27591973244147155, "grad_norm": 19.68751651825531, "learning_rate": 0.00027267711197707386, "loss": 8.1994, "step": 165 }, { "epoch": 0.2842809364548495, "grad_norm": 16.564772134956407, "learning_rate": 0.0002701050415693934, "loss": 7.3116, "step": 170 }, { "epoch": 0.29264214046822745, "grad_norm": 18.01034438896602, "learning_rate": 0.0002674305934098032, "loss": 6.7163, "step": 175 }, { "epoch": 0.3010033444816054, "grad_norm": 12.65556929182177, "learning_rate": 0.00026465604720269026, "loss": 6.2962, "step": 180 }, { "epoch": 0.3093645484949833, "grad_norm": 9.795061211752262, "learning_rate": 0.0002617837679761962, "loss": 6.0851, "step": 185 }, { "epoch": 0.3177257525083612, "grad_norm": 9.27750296099571, "learning_rate": 0.0002588162040662628, "loss": 5.8121, "step": 190 }, { "epoch": 0.32608695652173914, "grad_norm": 7.581802949350765, "learning_rate": 0.0002557558850296661, "loss": 5.5399, "step": 195 }, { "epoch": 0.33444816053511706, "grad_norm": 10.289352311810807, "learning_rate": 0.00025260541948781784, "loss": 5.3443, "step": 200 }, { "epoch": 0.342809364548495, "grad_norm": 6.606792882694537, "learning_rate": 0.00024936749290317196, "loss": 5.0023, "step": 205 }, { "epoch": 0.3511705685618729, "grad_norm": 5.306167947282732, "learning_rate": 0.00024604486529013106, "loss": 4.8264, "step": 210 }, { "epoch": 0.3595317725752508, "grad_norm": 8.160432944309026, "learning_rate": 0.00024264036886240603, "loss": 4.8126, "step": 215 }, { "epoch": 0.36789297658862874, "grad_norm": 9.350662398086465, "learning_rate": 0.00023915690561883125, "loss": 4.5927, "step": 220 }, { "epoch": 0.3762541806020067, "grad_norm": 6.717782881890457, "learning_rate": 0.00023559744486969622, "loss": 4.3447, "step": 225 }, { "epoch": 0.38461538461538464, "grad_norm": 10.834440165150719, "learning_rate": 0.00023196502070569963, "loss": 4.7644, "step": 230 }, { "epoch": 0.39297658862876256, "grad_norm": 9.530980370244805, "learning_rate": 0.00022826272941168556, "loss": 4.5336, "step": 235 }, { "epoch": 0.4013377926421405, "grad_norm": 9.96662219085004, "learning_rate": 0.00022449372682736428, "loss": 4.1642, "step": 240 }, { "epoch": 0.4096989966555184, "grad_norm": 7.17474089656695, "learning_rate": 0.00022066122565726914, "loss": 4.1165, "step": 245 }, { "epoch": 0.4180602006688963, "grad_norm": 5.991132986370517, "learning_rate": 0.00021676849273224132, "loss": 3.9771, "step": 250 }, { "epoch": 0.42642140468227424, "grad_norm": 6.305690862174887, "learning_rate": 0.00021281884622477757, "loss": 3.8831, "step": 255 }, { "epoch": 0.43478260869565216, "grad_norm": 5.754061162264476, "learning_rate": 0.00020881565282061386, "loss": 3.7495, "step": 260 }, { "epoch": 0.4431438127090301, "grad_norm": 5.539202432443519, "learning_rate": 0.00020476232484895653, "loss": 3.5914, "step": 265 }, { "epoch": 0.451505016722408, "grad_norm": 6.726459928222644, "learning_rate": 0.00020066231737380684, "loss": 3.5781, "step": 270 }, { "epoch": 0.459866220735786, "grad_norm": 5.078063325340097, "learning_rate": 0.00019651912524885814, "loss": 3.5243, "step": 275 }, { "epoch": 0.4682274247491639, "grad_norm": 6.020230646512102, "learning_rate": 0.00019233628013847636, "loss": 3.386, "step": 280 }, { "epoch": 0.4765886287625418, "grad_norm": 4.216873427535516, "learning_rate": 0.00018811734750730306, "loss": 3.4131, "step": 285 }, { "epoch": 0.48494983277591974, "grad_norm": 4.224871355162302, "learning_rate": 0.000183865923581047, "loss": 3.2989, "step": 290 }, { "epoch": 0.49331103678929766, "grad_norm": 4.645722524256328, "learning_rate": 0.00017958563228105473, "loss": 3.227, "step": 295 }, { "epoch": 0.5016722408026756, "grad_norm": 4.526907081748694, "learning_rate": 0.000175280122135274, "loss": 3.1885, "step": 300 }, { "epoch": 0.5100334448160535, "grad_norm": 3.8079799771786558, "learning_rate": 0.00017095306316824145, "loss": 3.1247, "step": 305 }, { "epoch": 0.5183946488294314, "grad_norm": 3.7709095979542058, "learning_rate": 0.00016660814377274764, "loss": 3.1044, "step": 310 }, { "epoch": 0.5267558528428093, "grad_norm": 3.7946474549046267, "learning_rate": 0.00016224906756584414, "loss": 3.0063, "step": 315 }, { "epoch": 0.5351170568561873, "grad_norm": 4.245301792150929, "learning_rate": 0.00015787955023187394, "loss": 2.9188, "step": 320 }, { "epoch": 0.5434782608695652, "grad_norm": 4.017268668767216, "learning_rate": 0.00015350331635521528, "loss": 2.9589, "step": 325 }, { "epoch": 0.5518394648829431, "grad_norm": 3.333420461437936, "learning_rate": 0.00014912409624543942, "loss": 2.8643, "step": 330 }, { "epoch": 0.560200668896321, "grad_norm": 3.2798450150057508, "learning_rate": 0.000144745622757588, "loss": 2.7767, "step": 335 }, { "epoch": 0.568561872909699, "grad_norm": 3.664786491151442, "learning_rate": 0.00014037162811028105, "loss": 2.6968, "step": 340 }, { "epoch": 0.5769230769230769, "grad_norm": 3.0151240913774635, "learning_rate": 0.00013600584070436663, "loss": 2.7462, "step": 345 }, { "epoch": 0.5852842809364549, "grad_norm": 4.300668720491546, "learning_rate": 0.0001316519819448257, "loss": 2.6857, "step": 350 }, { "epoch": 0.5936454849498328, "grad_norm": 6.856565698153125, "learning_rate": 0.0001273137630686397, "loss": 2.6735, "step": 355 }, { "epoch": 0.6020066889632107, "grad_norm": 3.6559482252841904, "learning_rate": 0.0001229948819813257, "loss": 2.6167, "step": 360 }, { "epoch": 0.6103678929765887, "grad_norm": 3.549605686334841, "learning_rate": 0.00011869902010483487, "loss": 2.5546, "step": 365 }, { "epoch": 0.6187290969899666, "grad_norm": 3.008473775307977, "learning_rate": 0.00011442983923950197, "loss": 2.5618, "step": 370 }, { "epoch": 0.6270903010033445, "grad_norm": 3.493843480596192, "learning_rate": 0.00011019097844272025, "loss": 2.4395, "step": 375 }, { "epoch": 0.6354515050167224, "grad_norm": 3.022476308004379, "learning_rate": 0.00010598605092700247, "loss": 2.4327, "step": 380 }, { "epoch": 0.6438127090301003, "grad_norm": 3.033725108847574, "learning_rate": 0.00010181864098007187, "loss": 2.4019, "step": 385 }, { "epoch": 0.6521739130434783, "grad_norm": 2.730168253261192, "learning_rate": 9.76923009096091e-05, "loss": 2.3644, "step": 390 }, { "epoch": 0.6605351170568562, "grad_norm": 2.5487378528989986, "learning_rate": 9.361054801525882e-05, "loss": 2.3956, "step": 395 }, { "epoch": 0.6688963210702341, "grad_norm": 3.0219826290571214, "learning_rate": 8.9576861590477e-05, "loss": 2.3118, "step": 400 }, { "epoch": 0.677257525083612, "grad_norm": 2.3724800707991442, "learning_rate": 8.559467995677535e-05, "loss": 2.3294, "step": 405 }, { "epoch": 0.68561872909699, "grad_norm": 3.218242001430894, "learning_rate": 8.166739753288954e-05, "loss": 2.2735, "step": 410 }, { "epoch": 0.6939799331103679, "grad_norm": 2.618273238863685, "learning_rate": 7.77983619413708e-05, "loss": 2.2728, "step": 415 }, { "epoch": 0.7023411371237458, "grad_norm": 2.1838152570753016, "learning_rate": 7.399087115506633e-05, "loss": 2.2301, "step": 420 }, { "epoch": 0.7107023411371237, "grad_norm": 2.3293111793801473, "learning_rate": 7.024817068592166e-05, "loss": 2.2563, "step": 425 }, { "epoch": 0.7190635451505016, "grad_norm": 3.3354279232331505, "learning_rate": 6.657345081850015e-05, "loss": 2.2324, "step": 430 }, { "epoch": 0.7274247491638796, "grad_norm": 2.4637443882850474, "learning_rate": 6.296984389057886e-05, "loss": 2.1572, "step": 435 }, { "epoch": 0.7357859531772575, "grad_norm": 2.66480391294473, "learning_rate": 5.944042162313818e-05, "loss": 2.1485, "step": 440 }, { "epoch": 0.7441471571906354, "grad_norm": 2.369277046664397, "learning_rate": 5.59881925020216e-05, "loss": 2.1156, "step": 445 }, { "epoch": 0.7525083612040134, "grad_norm": 2.404749403147846, "learning_rate": 5.2616099213496704e-05, "loss": 2.0883, "step": 450 }, { "epoch": 0.7608695652173914, "grad_norm": 2.151600123558966, "learning_rate": 4.93270161359043e-05, "loss": 2.0871, "step": 455 }, { "epoch": 0.7692307692307693, "grad_norm": 1.8485834869843754, "learning_rate": 4.612374688953315e-05, "loss": 2.0717, "step": 460 }, { "epoch": 0.7775919732441472, "grad_norm": 2.015234182046728, "learning_rate": 4.300902194680902e-05, "loss": 2.0691, "step": 465 }, { "epoch": 0.7859531772575251, "grad_norm": 1.8364808711885607, "learning_rate": 3.998549630483525e-05, "loss": 2.0201, "step": 470 }, { "epoch": 0.794314381270903, "grad_norm": 1.9757575481092078, "learning_rate": 3.7055747222268256e-05, "loss": 2.0128, "step": 475 }, { "epoch": 0.802675585284281, "grad_norm": 1.957601461432362, "learning_rate": 3.422227202245778e-05, "loss": 1.975, "step": 480 }, { "epoch": 0.8110367892976589, "grad_norm": 1.8124541787764386, "learning_rate": 3.1487485964723966e-05, "loss": 1.9249, "step": 485 }, { "epoch": 0.8193979933110368, "grad_norm": 1.8489721306641316, "learning_rate": 2.885372018558604e-05, "loss": 1.9786, "step": 490 }, { "epoch": 0.8277591973244147, "grad_norm": 1.9783817862893105, "learning_rate": 2.6323219711697368e-05, "loss": 1.9287, "step": 495 }, { "epoch": 0.8361204013377926, "grad_norm": 1.9022882119693414, "learning_rate": 2.3898141546180678e-05, "loss": 1.9572, "step": 500 }, { "epoch": 0.8444816053511706, "grad_norm": 2.1784717203571673, "learning_rate": 2.1580552829994785e-05, "loss": 1.9073, "step": 505 }, { "epoch": 0.8528428093645485, "grad_norm": 1.7051803845257447, "learning_rate": 1.9372429079900052e-05, "loss": 1.9092, "step": 510 }, { "epoch": 0.8612040133779264, "grad_norm": 1.8732499048540139, "learning_rate": 1.727565250452414e-05, "loss": 1.8558, "step": 515 }, { "epoch": 0.8695652173913043, "grad_norm": 1.4473569152727979, "learning_rate": 1.52920103999642e-05, "loss": 1.8463, "step": 520 }, { "epoch": 0.8779264214046822, "grad_norm": 1.4559349207330088, "learning_rate": 1.3423193626292405e-05, "loss": 1.8972, "step": 525 }, { "epoch": 0.8862876254180602, "grad_norm": 1.456537147004704, "learning_rate": 1.1670795166263846e-05, "loss": 1.8839, "step": 530 }, { "epoch": 0.8946488294314381, "grad_norm": 1.4711682274882243, "learning_rate": 1.003630876745522e-05, "loss": 1.8899, "step": 535 }, { "epoch": 0.903010033444816, "grad_norm": 1.3419796942254214, "learning_rate": 8.521127668991612e-06, "loss": 1.822, "step": 540 }, { "epoch": 0.9113712374581939, "grad_norm": 1.4154994584871663, "learning_rate": 7.126543413947045e-06, "loss": 1.8409, "step": 545 }, { "epoch": 0.919732441471572, "grad_norm": 1.5056657282270582, "learning_rate": 5.853744748430883e-06, "loss": 1.8138, "step": 550 }, { "epoch": 0.9280936454849499, "grad_norm": 1.3367881313162435, "learning_rate": 4.703816608298533e-06, "loss": 1.8748, "step": 555 }, { "epoch": 0.9364548494983278, "grad_norm": 1.4438743199653445, "learning_rate": 3.6777391943500877e-06, "loss": 1.8279, "step": 560 }, { "epoch": 0.9448160535117057, "grad_norm": 1.3948588971264737, "learning_rate": 2.7763871368055137e-06, "loss": 1.8959, "step": 565 }, { "epoch": 0.9531772575250836, "grad_norm": 1.2272041835054792, "learning_rate": 2.000528749768199e-06, "loss": 1.8279, "step": 570 }, { "epoch": 0.9615384615384616, "grad_norm": 1.2479607608150853, "learning_rate": 1.3508253763127974e-06, "loss": 1.8121, "step": 575 }, { "epoch": 0.9698996655518395, "grad_norm": 1.387198687782991, "learning_rate": 8.278308247549992e-07, "loss": 1.8638, "step": 580 }, { "epoch": 0.9782608695652174, "grad_norm": 1.2432235153355073, "learning_rate": 4.3199089658449425e-07, "loss": 1.8267, "step": 585 }, { "epoch": 0.9866220735785953, "grad_norm": 1.3076524438940778, "learning_rate": 1.6364300646291018e-07, "loss": 1.7537, "step": 590 }, { "epoch": 0.9949832775919732, "grad_norm": 1.278244790033875, "learning_rate": 2.30158946110115e-08, "loss": 1.8, "step": 595 }, { "epoch": 1.0, "eval_loss": 4.838874340057373, "eval_runtime": 1.4359, "eval_samples_per_second": 3.482, "eval_steps_per_second": 0.696, "step": 598 }, { "epoch": 1.0, "step": 598, "total_flos": 20570041221120.0, "train_loss": 5.0949278754932825, "train_runtime": 8333.1384, "train_samples_per_second": 2.295, "train_steps_per_second": 0.072 } ], "logging_steps": 5, "max_steps": 598, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 100, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 20570041221120.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }