{ "best_metric": null, "best_model_checkpoint": null, "epoch": 2.0, "eval_steps": 500, "global_step": 3146, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.01589319771137953, "grad_norm": 0.47528788447380066, "learning_rate": 1.1904761904761903e-05, "loss": 1.4053, "step": 25 }, { "epoch": 0.03178639542275906, "grad_norm": 0.3256166875362396, "learning_rate": 2.3809523809523807e-05, "loss": 1.1769, "step": 50 }, { "epoch": 0.04767959313413859, "grad_norm": 0.40220576524734497, "learning_rate": 3.571428571428571e-05, "loss": 1.1523, "step": 75 }, { "epoch": 0.06357279084551812, "grad_norm": 0.37274038791656494, "learning_rate": 4.7619047619047614e-05, "loss": 0.909, "step": 100 }, { "epoch": 0.07946598855689765, "grad_norm": 0.3363211452960968, "learning_rate": 5.952380952380952e-05, "loss": 0.7534, "step": 125 }, { "epoch": 0.09535918626827718, "grad_norm": 0.28489017486572266, "learning_rate": 7.142857142857142e-05, "loss": 0.7111, "step": 150 }, { "epoch": 0.11125238397965671, "grad_norm": 0.25532016158103943, "learning_rate": 8.333333333333333e-05, "loss": 0.8574, "step": 175 }, { "epoch": 0.12714558169103624, "grad_norm": 0.20530685782432556, "learning_rate": 9.523809523809523e-05, "loss": 0.7174, "step": 200 }, { "epoch": 0.14303877940241577, "grad_norm": 0.294625848531723, "learning_rate": 0.00010714285714285714, "loss": 0.7769, "step": 225 }, { "epoch": 0.1589319771137953, "grad_norm": 0.3870828151702881, "learning_rate": 0.00011904761904761903, "loss": 0.7005, "step": 250 }, { "epoch": 0.17482517482517482, "grad_norm": 0.26453226804733276, "learning_rate": 0.00013095238095238093, "loss": 0.7645, "step": 275 }, { "epoch": 0.19071837253655435, "grad_norm": 0.31448787450790405, "learning_rate": 0.00014285714285714284, "loss": 0.7961, "step": 300 }, { "epoch": 0.2066115702479339, "grad_norm": 0.25471773743629456, "learning_rate": 0.00014999538207561303, "loss": 0.8038, "step": 325 }, { "epoch": 0.22250476795931343, "grad_norm": 0.2693980634212494, "learning_rate": 0.0001499434369568889, "loss": 0.7773, "step": 350 }, { "epoch": 0.23839796567069294, "grad_norm": 0.21915049850940704, "learning_rate": 0.0001498338144248584, "loss": 0.7611, "step": 375 }, { "epoch": 0.25429116338207247, "grad_norm": 0.24487756192684174, "learning_rate": 0.00014966659884640333, "loss": 0.7344, "step": 400 }, { "epoch": 0.270184361093452, "grad_norm": 0.23428066074848175, "learning_rate": 0.00014944191891273962, "loss": 0.7511, "step": 425 }, { "epoch": 0.28607755880483154, "grad_norm": 0.15835721790790558, "learning_rate": 0.00014915994754037505, "loss": 0.6929, "step": 450 }, { "epoch": 0.3019707565162111, "grad_norm": 0.2219650149345398, "learning_rate": 0.00014882090173803036, "loss": 0.7666, "step": 475 }, { "epoch": 0.3178639542275906, "grad_norm": 0.24038617312908173, "learning_rate": 0.0001484250424396267, "loss": 0.8315, "step": 500 }, { "epoch": 0.3337571519389701, "grad_norm": 0.1785655915737152, "learning_rate": 0.00014797267430346768, "loss": 0.8097, "step": 525 }, { "epoch": 0.34965034965034963, "grad_norm": 0.2057177722454071, "learning_rate": 0.0001474641454777703, "loss": 0.8053, "step": 550 }, { "epoch": 0.36554354736172917, "grad_norm": 0.15342770516872406, "learning_rate": 0.00014689984733272585, "loss": 0.7279, "step": 575 }, { "epoch": 0.3814367450731087, "grad_norm": 0.26547038555145264, "learning_rate": 0.00014628021415929646, "loss": 0.7449, "step": 600 }, { "epoch": 0.39732994278448824, "grad_norm": 0.21101844310760498, "learning_rate": 0.00014560572283497925, "loss": 0.793, "step": 625 }, { "epoch": 0.4132231404958678, "grad_norm": 0.3606005311012268, "learning_rate": 0.00014487689245679545, "loss": 0.7311, "step": 650 }, { "epoch": 0.4291163382072473, "grad_norm": 0.20686589181423187, "learning_rate": 0.00014409428394178674, "loss": 0.7273, "step": 675 }, { "epoch": 0.44500953591862685, "grad_norm": 0.25700056552886963, "learning_rate": 0.00014325849959532654, "loss": 0.7998, "step": 700 }, { "epoch": 0.46090273363000633, "grad_norm": 0.19483058154582977, "learning_rate": 0.00014237018264757795, "loss": 0.7463, "step": 725 }, { "epoch": 0.47679593134138587, "grad_norm": 0.18222449719905853, "learning_rate": 0.0001414300167584561, "loss": 0.7365, "step": 750 }, { "epoch": 0.4926891290527654, "grad_norm": 0.1947900503873825, "learning_rate": 0.00014043872549147454, "loss": 0.747, "step": 775 }, { "epoch": 0.5085823267641449, "grad_norm": 0.1856011003255844, "learning_rate": 0.00013939707175688178, "loss": 0.7638, "step": 800 }, { "epoch": 0.5244755244755245, "grad_norm": 0.20004665851593018, "learning_rate": 0.00013830585722451587, "loss": 0.7243, "step": 825 }, { "epoch": 0.540368722186904, "grad_norm": 0.1734880954027176, "learning_rate": 0.0001371659217068294, "loss": 0.7505, "step": 850 }, { "epoch": 0.5562619198982836, "grad_norm": 0.24975039064884186, "learning_rate": 0.00013597814251255891, "loss": 0.7699, "step": 875 }, { "epoch": 0.5721551176096631, "grad_norm": 0.24918633699417114, "learning_rate": 0.0001347434337715374, "loss": 0.771, "step": 900 }, { "epoch": 0.5880483153210426, "grad_norm": 0.15272338688373566, "learning_rate": 0.00013346274573116824, "loss": 0.7405, "step": 925 }, { "epoch": 0.6039415130324222, "grad_norm": 0.19538848102092743, "learning_rate": 0.0001321370640251029, "loss": 0.7981, "step": 950 }, { "epoch": 0.6198347107438017, "grad_norm": 0.21340744197368622, "learning_rate": 0.0001307674089146848, "loss": 0.7714, "step": 975 }, { "epoch": 0.6357279084551812, "grad_norm": 0.2356826215982437, "learning_rate": 0.00012935483450374346, "loss": 0.7867, "step": 1000 }, { "epoch": 0.6516211061665607, "grad_norm": 0.2050427347421646, "learning_rate": 0.0001279004279273428, "loss": 0.7375, "step": 1025 }, { "epoch": 0.6675143038779402, "grad_norm": 0.28300225734710693, "learning_rate": 0.00012640530851510827, "loss": 0.7864, "step": 1050 }, { "epoch": 0.6834075015893197, "grad_norm": 0.2704091966152191, "learning_rate": 0.00012487062692977685, "loss": 0.7685, "step": 1075 }, { "epoch": 0.6993006993006993, "grad_norm": 0.18991053104400635, "learning_rate": 0.00012329756428163224, "loss": 0.7825, "step": 1100 }, { "epoch": 0.7151938970120788, "grad_norm": 0.20676083862781525, "learning_rate": 0.00012168733121950773, "loss": 0.6943, "step": 1125 }, { "epoch": 0.7310870947234583, "grad_norm": 0.22199320793151855, "learning_rate": 0.00012004116699905555, "loss": 0.7364, "step": 1150 }, { "epoch": 0.7469802924348379, "grad_norm": 0.29109102487564087, "learning_rate": 0.00011836033852900001, "loss": 0.7073, "step": 1175 }, { "epoch": 0.7628734901462174, "grad_norm": 0.18443694710731506, "learning_rate": 0.00011664613939610873, "loss": 0.7531, "step": 1200 }, { "epoch": 0.778766687857597, "grad_norm": 0.1905236691236496, "learning_rate": 0.00011489988886963181, "loss": 0.7124, "step": 1225 }, { "epoch": 0.7946598855689765, "grad_norm": 0.17691577970981598, "learning_rate": 0.00011312293088597576, "loss": 0.7461, "step": 1250 }, { "epoch": 0.810553083280356, "grad_norm": 0.1734876185655594, "learning_rate": 0.00011131663301439294, "loss": 0.7764, "step": 1275 }, { "epoch": 0.8264462809917356, "grad_norm": 0.17549951374530792, "learning_rate": 0.00010948238540448318, "loss": 0.7874, "step": 1300 }, { "epoch": 0.8423394787031151, "grad_norm": 0.19360962510108948, "learning_rate": 0.00010762159971631704, "loss": 0.7463, "step": 1325 }, { "epoch": 0.8582326764144946, "grad_norm": 0.18942518532276154, "learning_rate": 0.00010573570803400446, "loss": 0.777, "step": 1350 }, { "epoch": 0.8741258741258742, "grad_norm": 0.1779840737581253, "learning_rate": 0.00010382616176354481, "loss": 0.6899, "step": 1375 }, { "epoch": 0.8900190718372537, "grad_norm": 0.2987983226776123, "learning_rate": 0.00010189443051580653, "loss": 0.6906, "step": 1400 }, { "epoch": 0.9059122695486331, "grad_norm": 0.2795010209083557, "learning_rate": 9.994200097549602e-05, "loss": 0.7119, "step": 1425 }, { "epoch": 0.9218054672600127, "grad_norm": 0.21874257922172546, "learning_rate": 9.797037575698653e-05, "loss": 0.7284, "step": 1450 }, { "epoch": 0.9376986649713922, "grad_norm": 0.18118910491466522, "learning_rate": 9.598107224788701e-05, "loss": 0.6999, "step": 1475 }, { "epoch": 0.9535918626827717, "grad_norm": 0.2674807608127594, "learning_rate": 9.397562144124163e-05, "loss": 0.6853, "step": 1500 }, { "epoch": 0.9694850603941513, "grad_norm": 0.24154460430145264, "learning_rate": 9.195556675725821e-05, "loss": 0.7482, "step": 1525 }, { "epoch": 0.9853782581055308, "grad_norm": 0.2160235196352005, "learning_rate": 8.992246285547248e-05, "loss": 0.7444, "step": 1550 }, { "epoch": 1.0012714558169105, "grad_norm": 0.17261384427547455, "learning_rate": 8.787787443826268e-05, "loss": 0.822, "step": 1575 }, { "epoch": 1.0171646535282899, "grad_norm": 0.17116381227970123, "learning_rate": 8.582337504663479e-05, "loss": 0.6318, "step": 1600 }, { "epoch": 1.0330578512396693, "grad_norm": 0.18086880445480347, "learning_rate": 8.376054584920558e-05, "loss": 0.639, "step": 1625 }, { "epoch": 1.048951048951049, "grad_norm": 0.23205679655075073, "learning_rate": 8.169097442531531e-05, "loss": 0.7593, "step": 1650 }, { "epoch": 1.0648442466624284, "grad_norm": 0.2291744500398636, "learning_rate": 7.961625354320664e-05, "loss": 0.7277, "step": 1675 }, { "epoch": 1.080737444373808, "grad_norm": 0.2415177822113037, "learning_rate": 7.753797993421022e-05, "loss": 0.6691, "step": 1700 }, { "epoch": 1.0966306420851875, "grad_norm": 0.22513903677463531, "learning_rate": 7.545775306388e-05, "loss": 0.7055, "step": 1725 }, { "epoch": 1.112523839796567, "grad_norm": 0.21355651319026947, "learning_rate": 7.337717390102455e-05, "loss": 0.6853, "step": 1750 }, { "epoch": 1.1284170375079465, "grad_norm": 0.25458982586860657, "learning_rate": 7.129784368558108e-05, "loss": 0.68, "step": 1775 }, { "epoch": 1.1443102352193262, "grad_norm": 0.31313472986221313, "learning_rate": 6.922136269628126e-05, "loss": 0.6753, "step": 1800 }, { "epoch": 1.1602034329307056, "grad_norm": 0.18478913605213165, "learning_rate": 6.714932901905637e-05, "loss": 0.685, "step": 1825 }, { "epoch": 1.1760966306420853, "grad_norm": 0.22722911834716797, "learning_rate": 6.508333731713026e-05, "loss": 0.74, "step": 1850 }, { "epoch": 1.1919898283534647, "grad_norm": 0.23728103935718536, "learning_rate": 6.302497760374652e-05, "loss": 0.7204, "step": 1875 }, { "epoch": 1.2078830260648443, "grad_norm": 0.30603456497192383, "learning_rate": 6.097583401847412e-05, "loss": 0.6686, "step": 1900 }, { "epoch": 1.2237762237762237, "grad_norm": 0.27340492606163025, "learning_rate": 5.893748360803385e-05, "loss": 0.6608, "step": 1925 }, { "epoch": 1.2396694214876034, "grad_norm": 0.1994735151529312, "learning_rate": 5.691149511258303e-05, "loss": 0.6903, "step": 1950 }, { "epoch": 1.2555626191989828, "grad_norm": 0.2764354646205902, "learning_rate": 5.489942775839355e-05, "loss": 0.6587, "step": 1975 }, { "epoch": 1.2714558169103625, "grad_norm": 0.2392449527978897, "learning_rate": 5.290283005785156e-05, "loss": 0.7018, "step": 2000 }, { "epoch": 1.287349014621742, "grad_norm": 0.20800232887268066, "learning_rate": 5.0923238617702786e-05, "loss": 0.6528, "step": 2025 }, { "epoch": 1.3032422123331213, "grad_norm": 0.2834407389163971, "learning_rate": 4.896217695646081e-05, "loss": 0.6919, "step": 2050 }, { "epoch": 1.319135410044501, "grad_norm": 0.22680579125881195, "learning_rate": 4.702115433188782e-05, "loss": 0.668, "step": 2075 }, { "epoch": 1.3350286077558806, "grad_norm": 0.27213868498802185, "learning_rate": 4.5101664579450994e-05, "loss": 0.7536, "step": 2100 }, { "epoch": 1.35092180546726, "grad_norm": 0.2941715121269226, "learning_rate": 4.3205184962647714e-05, "loss": 0.8361, "step": 2125 }, { "epoch": 1.3668150031786395, "grad_norm": 0.28851884603500366, "learning_rate": 4.133317503608517e-05, "loss": 0.6659, "step": 2150 }, { "epoch": 1.3827082008900191, "grad_norm": 0.2723303735256195, "learning_rate": 3.948707552218857e-05, "loss": 0.6205, "step": 2175 }, { "epoch": 1.3986013986013985, "grad_norm": 0.34859615564346313, "learning_rate": 3.766830720240305e-05, "loss": 0.7091, "step": 2200 }, { "epoch": 1.4144945963127782, "grad_norm": 0.3292659521102905, "learning_rate": 3.5878269823742584e-05, "loss": 0.6266, "step": 2225 }, { "epoch": 1.4303877940241576, "grad_norm": 0.18559393286705017, "learning_rate": 3.411834102152669e-05, "loss": 0.7348, "step": 2250 }, { "epoch": 1.4462809917355373, "grad_norm": 0.2816512882709503, "learning_rate": 3.238987525913521e-05, "loss": 0.649, "step": 2275 }, { "epoch": 1.4621741894469167, "grad_norm": 0.23303620517253876, "learning_rate": 3.0694202785596056e-05, "loss": 0.7151, "step": 2300 }, { "epoch": 1.4780673871582963, "grad_norm": 0.2300049215555191, "learning_rate": 2.9032628611808934e-05, "loss": 0.7383, "step": 2325 }, { "epoch": 1.4939605848696758, "grad_norm": 0.2333756685256958, "learning_rate": 2.7406431506192453e-05, "loss": 0.7278, "step": 2350 }, { "epoch": 1.5098537825810552, "grad_norm": 0.285395085811615, "learning_rate": 2.5816863010527768e-05, "loss": 0.694, "step": 2375 }, { "epoch": 1.5257469802924348, "grad_norm": 0.24296920001506805, "learning_rate": 2.4265146476756423e-05, "loss": 0.6776, "step": 2400 }, { "epoch": 1.5416401780038145, "grad_norm": 0.27042725682258606, "learning_rate": 2.2752476125473068e-05, "loss": 0.717, "step": 2425 }, { "epoch": 1.557533375715194, "grad_norm": 0.27852827310562134, "learning_rate": 2.1280016126838474e-05, "loss": 0.7223, "step": 2450 }, { "epoch": 1.5734265734265733, "grad_norm": 0.3818005919456482, "learning_rate": 1.984889970461933e-05, "loss": 0.5883, "step": 2475 }, { "epoch": 1.589319771137953, "grad_norm": 0.22705549001693726, "learning_rate": 1.8460228264045072e-05, "loss": 0.6397, "step": 2500 }, { "epoch": 1.6052129688493326, "grad_norm": 0.233836829662323, "learning_rate": 1.711507054415244e-05, "loss": 0.667, "step": 2525 }, { "epoch": 1.621106166560712, "grad_norm": 0.21357496082782745, "learning_rate": 1.581446179527049e-05, "loss": 0.7254, "step": 2550 }, { "epoch": 1.6369993642720915, "grad_norm": 0.2109714299440384, "learning_rate": 1.4559402982278968e-05, "loss": 0.6232, "step": 2575 }, { "epoch": 1.6528925619834711, "grad_norm": 0.23249782621860504, "learning_rate": 1.3350860014253032e-05, "loss": 0.6821, "step": 2600 }, { "epoch": 1.6687857596948508, "grad_norm": 0.22555020451545715, "learning_rate": 1.2189763001087526e-05, "loss": 0.7022, "step": 2625 }, { "epoch": 1.6846789574062302, "grad_norm": 0.2770826518535614, "learning_rate": 1.1077005537672596e-05, "loss": 0.6495, "step": 2650 }, { "epoch": 1.7005721551176096, "grad_norm": 0.32234126329421997, "learning_rate": 1.001344401617192e-05, "loss": 0.6353, "step": 2675 }, { "epoch": 1.716465352828989, "grad_norm": 0.17805339395999908, "learning_rate": 8.999896966932396e-06, "loss": 0.7182, "step": 2700 }, { "epoch": 1.7323585505403687, "grad_norm": 0.2321351170539856, "learning_rate": 8.037144428532986e-06, "loss": 0.7443, "step": 2725 }, { "epoch": 1.7482517482517483, "grad_norm": 0.24086669087409973, "learning_rate": 7.1259273474571225e-06, "loss": 0.7209, "step": 2750 }, { "epoch": 1.7641449459631278, "grad_norm": 0.2397790104150772, "learning_rate": 6.266947007850945e-06, "loss": 0.7198, "step": 2775 }, { "epoch": 1.7800381436745072, "grad_norm": 0.27960214018821716, "learning_rate": 5.460864491806263e-06, "loss": 0.6878, "step": 2800 }, { "epoch": 1.7959313413858868, "grad_norm": 0.2695086598396301, "learning_rate": 4.7083001705833735e-06, "loss": 0.7547, "step": 2825 }, { "epoch": 1.8118245390972665, "grad_norm": 0.3778378963470459, "learning_rate": 4.009833227165573e-06, "loss": 0.5804, "step": 2850 }, { "epoch": 1.827717736808646, "grad_norm": 0.28731873631477356, "learning_rate": 3.3660012105126336e-06, "loss": 0.738, "step": 2875 }, { "epoch": 1.8436109345200253, "grad_norm": 0.2649593651294708, "learning_rate": 2.777299621856283e-06, "loss": 0.6167, "step": 2900 }, { "epoch": 1.859504132231405, "grad_norm": 0.28797072172164917, "learning_rate": 2.244181533356318e-06, "loss": 0.7371, "step": 2925 }, { "epoch": 1.8753973299427846, "grad_norm": 0.3841162323951721, "learning_rate": 1.7670572394104715e-06, "loss": 0.674, "step": 2950 }, { "epoch": 1.891290527654164, "grad_norm": 0.2641618847846985, "learning_rate": 1.346293940886739e-06, "loss": 0.6912, "step": 2975 }, { "epoch": 1.9071837253655435, "grad_norm": 0.25159719586372375, "learning_rate": 9.822154625209217e-07, "loss": 0.7177, "step": 3000 }, { "epoch": 1.9230769230769231, "grad_norm": 0.2949710488319397, "learning_rate": 6.751020036970728e-07, "loss": 0.7384, "step": 3025 }, { "epoch": 1.9389701207883026, "grad_norm": 0.29616832733154297, "learning_rate": 4.2518992280245465e-07, "loss": 0.6868, "step": 3050 }, { "epoch": 1.9548633184996822, "grad_norm": 0.2574808895587921, "learning_rate": 2.3267155532321335e-07, "loss": 0.7261, "step": 3075 }, { "epoch": 1.9707565162110616, "grad_norm": 0.18125727772712708, "learning_rate": 9.769506582052722e-08, "loss": 0.7009, "step": 3100 }, { "epoch": 1.986649713922441, "grad_norm": 0.29592961072921753, "learning_rate": 2.036433390130432e-08, "loss": 0.6884, "step": 3125 } ], "logging_steps": 25, "max_steps": 3146, "num_input_tokens_seen": 0, "num_train_epochs": 2, "save_steps": 0, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 2.1800418501722112e+17, "train_batch_size": 3, "trial_name": null, "trial_params": null }