{ "best_metric": 4.951249122619629, "best_model_checkpoint": "bert_tiny_lda_100_v1_book/checkpoint-350000", "epoch": 25.0, "eval_steps": 10000, "global_step": 355850, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0351271603203597, "grad_norm": 2.3923983573913574, "learning_rate": 5e-06, "loss": 14.1521, "step": 500 }, { "epoch": 0.0702543206407194, "grad_norm": 0.9734293818473816, "learning_rate": 1e-05, "loss": 12.053, "step": 1000 }, { "epoch": 0.1053814809610791, "grad_norm": 0.640344500541687, "learning_rate": 1.5e-05, "loss": 10.8101, "step": 1500 }, { "epoch": 0.1405086412814388, "grad_norm": 0.6007124781608582, "learning_rate": 2e-05, "loss": 10.6093, "step": 2000 }, { "epoch": 0.1756358016017985, "grad_norm": 0.5658907294273376, "learning_rate": 2.5e-05, "loss": 10.5218, "step": 2500 }, { "epoch": 0.2107629619221582, "grad_norm": 0.6568773984909058, "learning_rate": 3e-05, "loss": 10.4615, "step": 3000 }, { "epoch": 0.2458901222425179, "grad_norm": 0.6278502345085144, "learning_rate": 3.5e-05, "loss": 10.4237, "step": 3500 }, { "epoch": 0.2810172825628776, "grad_norm": 0.6695737838745117, "learning_rate": 4e-05, "loss": 10.3936, "step": 4000 }, { "epoch": 0.31614444288323734, "grad_norm": 0.546693742275238, "learning_rate": 4.5e-05, "loss": 10.3677, "step": 4500 }, { "epoch": 0.351271603203597, "grad_norm": 0.5352140665054321, "learning_rate": 5e-05, "loss": 10.3469, "step": 5000 }, { "epoch": 0.38639876352395675, "grad_norm": 0.7120894193649292, "learning_rate": 5.500000000000001e-05, "loss": 10.3234, "step": 5500 }, { "epoch": 0.4215259238443164, "grad_norm": 0.5939645767211914, "learning_rate": 6e-05, "loss": 10.3114, "step": 6000 }, { "epoch": 0.45665308416467615, "grad_norm": 0.6030145883560181, "learning_rate": 6.500000000000001e-05, "loss": 10.2966, "step": 6500 }, { "epoch": 0.4917802444850358, "grad_norm": 0.6429095268249512, "learning_rate": 7e-05, "loss": 10.2865, "step": 7000 }, { "epoch": 0.5269074048053956, "grad_norm": 0.5660910606384277, "learning_rate": 7.500000000000001e-05, "loss": 10.273, "step": 7500 }, { "epoch": 0.5620345651257552, "grad_norm": 0.7221590876579285, "learning_rate": 8e-05, "loss": 10.2609, "step": 8000 }, { "epoch": 0.5971617254461149, "grad_norm": 0.6724071502685547, "learning_rate": 8.5e-05, "loss": 10.2507, "step": 8500 }, { "epoch": 0.6322888857664747, "grad_norm": 0.886584997177124, "learning_rate": 9e-05, "loss": 10.2388, "step": 9000 }, { "epoch": 0.6674160460868344, "grad_norm": 0.506639838218689, "learning_rate": 9.5e-05, "loss": 10.2341, "step": 9500 }, { "epoch": 0.702543206407194, "grad_norm": 0.677033543586731, "learning_rate": 0.0001, "loss": 10.2267, "step": 10000 }, { "epoch": 0.702543206407194, "eval_accuracy": 0.16358490563732245, "eval_loss": 10.066490173339844, "eval_runtime": 125.6442, "eval_samples_per_second": 960.697, "eval_steps_per_second": 6.009, "step": 10000 }, { "epoch": 0.7376703667275537, "grad_norm": 0.523131251335144, "learning_rate": 9.985542865404077e-05, "loss": 10.2194, "step": 10500 }, { "epoch": 0.7727975270479135, "grad_norm": 0.534009575843811, "learning_rate": 9.971085730808155e-05, "loss": 10.213, "step": 11000 }, { "epoch": 0.8079246873682732, "grad_norm": 0.541458010673523, "learning_rate": 9.956628596212231e-05, "loss": 10.2092, "step": 11500 }, { "epoch": 0.8430518476886328, "grad_norm": 0.5457577109336853, "learning_rate": 9.942171461616309e-05, "loss": 10.1907, "step": 12000 }, { "epoch": 0.8781790080089925, "grad_norm": 0.665069580078125, "learning_rate": 9.927714327020385e-05, "loss": 10.1679, "step": 12500 }, { "epoch": 0.9133061683293523, "grad_norm": 0.6227748990058899, "learning_rate": 9.913257192424462e-05, "loss": 10.139, "step": 13000 }, { "epoch": 0.948433328649712, "grad_norm": 0.6807398796081543, "learning_rate": 9.89880005782854e-05, "loss": 10.0707, "step": 13500 }, { "epoch": 0.9835604889700716, "grad_norm": 0.7968010306358337, "learning_rate": 9.884342923232616e-05, "loss": 9.929, "step": 14000 }, { "epoch": 1.0186876492904313, "grad_norm": 1.4664796590805054, "learning_rate": 9.869885788636692e-05, "loss": 9.7329, "step": 14500 }, { "epoch": 1.053814809610791, "grad_norm": 2.061105251312256, "learning_rate": 9.855428654040769e-05, "loss": 9.2312, "step": 15000 }, { "epoch": 1.0889419699311507, "grad_norm": 2.0261075496673584, "learning_rate": 9.840971519444846e-05, "loss": 8.7562, "step": 15500 }, { "epoch": 1.1240691302515105, "grad_norm": 1.7339612245559692, "learning_rate": 9.826514384848924e-05, "loss": 8.4792, "step": 16000 }, { "epoch": 1.1591962905718702, "grad_norm": 2.0800840854644775, "learning_rate": 9.812057250253e-05, "loss": 8.2434, "step": 16500 }, { "epoch": 1.1943234508922298, "grad_norm": 2.0225894451141357, "learning_rate": 9.797600115657077e-05, "loss": 7.9747, "step": 17000 }, { "epoch": 1.2294506112125896, "grad_norm": 1.8394449949264526, "learning_rate": 9.783142981061155e-05, "loss": 7.7773, "step": 17500 }, { "epoch": 1.2645777715329491, "grad_norm": 1.70760977268219, "learning_rate": 9.768685846465231e-05, "loss": 7.6317, "step": 18000 }, { "epoch": 1.299704931853309, "grad_norm": 1.7218098640441895, "learning_rate": 9.754228711869307e-05, "loss": 7.4969, "step": 18500 }, { "epoch": 1.3348320921736687, "grad_norm": 1.4962530136108398, "learning_rate": 9.739771577273384e-05, "loss": 7.3966, "step": 19000 }, { "epoch": 1.3699592524940285, "grad_norm": 1.4457088708877563, "learning_rate": 9.725314442677463e-05, "loss": 7.3034, "step": 19500 }, { "epoch": 1.405086412814388, "grad_norm": 1.4672712087631226, "learning_rate": 9.71085730808154e-05, "loss": 7.2173, "step": 20000 }, { "epoch": 1.405086412814388, "eval_accuracy": 0.48231472535951897, "eval_loss": 6.74512243270874, "eval_runtime": 125.8626, "eval_samples_per_second": 959.03, "eval_steps_per_second": 5.999, "step": 20000 }, { "epoch": 1.4402135731347478, "grad_norm": 1.3748890161514282, "learning_rate": 9.696400173485616e-05, "loss": 7.1431, "step": 20500 }, { "epoch": 1.4753407334551074, "grad_norm": 1.3886744976043701, "learning_rate": 9.681943038889692e-05, "loss": 7.0741, "step": 21000 }, { "epoch": 1.5104678937754672, "grad_norm": 1.4711204767227173, "learning_rate": 9.66748590429377e-05, "loss": 7.0049, "step": 21500 }, { "epoch": 1.545595054095827, "grad_norm": 1.3519319295883179, "learning_rate": 9.653028769697846e-05, "loss": 6.9429, "step": 22000 }, { "epoch": 1.5807222144161865, "grad_norm": 1.3266453742980957, "learning_rate": 9.638571635101923e-05, "loss": 6.8967, "step": 22500 }, { "epoch": 1.6158493747365463, "grad_norm": 1.326925277709961, "learning_rate": 9.624114500506e-05, "loss": 6.8494, "step": 23000 }, { "epoch": 1.650976535056906, "grad_norm": 1.2158633470535278, "learning_rate": 9.609657365910077e-05, "loss": 6.8091, "step": 23500 }, { "epoch": 1.6861036953772657, "grad_norm": 1.2628967761993408, "learning_rate": 9.595200231314155e-05, "loss": 6.7688, "step": 24000 }, { "epoch": 1.7212308556976255, "grad_norm": 1.189835786819458, "learning_rate": 9.580743096718231e-05, "loss": 6.7331, "step": 24500 }, { "epoch": 1.7563580160179852, "grad_norm": 1.214654803276062, "learning_rate": 9.566285962122307e-05, "loss": 6.7003, "step": 25000 }, { "epoch": 1.7914851763383448, "grad_norm": 1.1932591199874878, "learning_rate": 9.551828827526385e-05, "loss": 6.6695, "step": 25500 }, { "epoch": 1.8266123366587044, "grad_norm": 1.2289775609970093, "learning_rate": 9.537371692930462e-05, "loss": 6.6377, "step": 26000 }, { "epoch": 1.8617394969790642, "grad_norm": 1.2651808261871338, "learning_rate": 9.522914558334538e-05, "loss": 6.6072, "step": 26500 }, { "epoch": 1.896866657299424, "grad_norm": 1.2493140697479248, "learning_rate": 9.508457423738616e-05, "loss": 6.5818, "step": 27000 }, { "epoch": 1.9319938176197837, "grad_norm": 1.183844804763794, "learning_rate": 9.494000289142692e-05, "loss": 6.5558, "step": 27500 }, { "epoch": 1.9671209779401433, "grad_norm": 1.2401198148727417, "learning_rate": 9.47954315454677e-05, "loss": 6.5295, "step": 28000 }, { "epoch": 2.002248138260503, "grad_norm": 1.2366821765899658, "learning_rate": 9.465086019950846e-05, "loss": 6.514, "step": 28500 }, { "epoch": 2.0373752985808626, "grad_norm": 1.2355561256408691, "learning_rate": 9.450628885354923e-05, "loss": 6.4813, "step": 29000 }, { "epoch": 2.0725024589012224, "grad_norm": 1.257392168045044, "learning_rate": 9.436171750758999e-05, "loss": 6.4594, "step": 29500 }, { "epoch": 2.107629619221582, "grad_norm": 1.220013976097107, "learning_rate": 9.421714616163077e-05, "loss": 6.4475, "step": 30000 }, { "epoch": 2.107629619221582, "eval_accuracy": 0.5508922649988164, "eval_loss": 6.030789852142334, "eval_runtime": 124.8659, "eval_samples_per_second": 966.685, "eval_steps_per_second": 6.046, "step": 30000 }, { "epoch": 2.142756779541942, "grad_norm": 1.221318244934082, "learning_rate": 9.407257481567155e-05, "loss": 6.4263, "step": 30500 }, { "epoch": 2.1778839398623013, "grad_norm": 1.2273422479629517, "learning_rate": 9.392800346971231e-05, "loss": 6.4035, "step": 31000 }, { "epoch": 2.213011100182661, "grad_norm": 1.1708391904830933, "learning_rate": 9.378343212375307e-05, "loss": 6.39, "step": 31500 }, { "epoch": 2.248138260503021, "grad_norm": 1.1821143627166748, "learning_rate": 9.363886077779385e-05, "loss": 6.3731, "step": 32000 }, { "epoch": 2.2832654208233807, "grad_norm": 1.2215129137039185, "learning_rate": 9.349428943183462e-05, "loss": 6.3562, "step": 32500 }, { "epoch": 2.3183925811437405, "grad_norm": 1.1790250539779663, "learning_rate": 9.334971808587538e-05, "loss": 6.3393, "step": 33000 }, { "epoch": 2.3535197414641003, "grad_norm": 1.2036933898925781, "learning_rate": 9.320514673991614e-05, "loss": 6.323, "step": 33500 }, { "epoch": 2.3886469017844596, "grad_norm": 1.1474812030792236, "learning_rate": 9.306057539395692e-05, "loss": 6.3098, "step": 34000 }, { "epoch": 2.4237740621048194, "grad_norm": 1.156016230583191, "learning_rate": 9.29160040479977e-05, "loss": 6.2901, "step": 34500 }, { "epoch": 2.458901222425179, "grad_norm": 1.2413389682769775, "learning_rate": 9.277143270203846e-05, "loss": 6.2776, "step": 35000 }, { "epoch": 2.494028382745539, "grad_norm": 1.2184691429138184, "learning_rate": 9.262686135607923e-05, "loss": 6.2639, "step": 35500 }, { "epoch": 2.5291555430658983, "grad_norm": 1.1915347576141357, "learning_rate": 9.248229001012e-05, "loss": 6.2503, "step": 36000 }, { "epoch": 2.564282703386258, "grad_norm": 1.184999704360962, "learning_rate": 9.233771866416077e-05, "loss": 6.2361, "step": 36500 }, { "epoch": 2.599409863706618, "grad_norm": 1.189146876335144, "learning_rate": 9.219314731820153e-05, "loss": 6.2218, "step": 37000 }, { "epoch": 2.6345370240269776, "grad_norm": 1.1678695678710938, "learning_rate": 9.20485759722423e-05, "loss": 6.2113, "step": 37500 }, { "epoch": 2.6696641843473374, "grad_norm": 1.1987296342849731, "learning_rate": 9.190400462628307e-05, "loss": 6.1968, "step": 38000 }, { "epoch": 2.704791344667697, "grad_norm": 1.1624820232391357, "learning_rate": 9.175943328032385e-05, "loss": 6.1856, "step": 38500 }, { "epoch": 2.739918504988057, "grad_norm": 1.2285350561141968, "learning_rate": 9.161486193436461e-05, "loss": 6.1727, "step": 39000 }, { "epoch": 2.7750456653084163, "grad_norm": 1.2020213603973389, "learning_rate": 9.147029058840538e-05, "loss": 6.1638, "step": 39500 }, { "epoch": 2.810172825628776, "grad_norm": 1.1140944957733154, "learning_rate": 9.132571924244616e-05, "loss": 6.1492, "step": 40000 }, { "epoch": 2.810172825628776, "eval_accuracy": 0.5803666875902359, "eval_loss": 5.752150058746338, "eval_runtime": 126.1615, "eval_samples_per_second": 956.758, "eval_steps_per_second": 5.984, "step": 40000 }, { "epoch": 2.845299985949136, "grad_norm": 1.203220009803772, "learning_rate": 9.118114789648692e-05, "loss": 6.1363, "step": 40500 }, { "epoch": 2.8804271462694957, "grad_norm": 1.2161797285079956, "learning_rate": 9.103657655052768e-05, "loss": 6.1242, "step": 41000 }, { "epoch": 2.915554306589855, "grad_norm": 1.1613247394561768, "learning_rate": 9.089200520456845e-05, "loss": 6.116, "step": 41500 }, { "epoch": 2.950681466910215, "grad_norm": 1.2261942625045776, "learning_rate": 9.074743385860923e-05, "loss": 6.1087, "step": 42000 }, { "epoch": 2.9858086272305746, "grad_norm": 1.1461601257324219, "learning_rate": 9.060286251265e-05, "loss": 6.0986, "step": 42500 }, { "epoch": 3.0209357875509344, "grad_norm": 1.1524542570114136, "learning_rate": 9.045829116669077e-05, "loss": 6.0772, "step": 43000 }, { "epoch": 3.056062947871294, "grad_norm": 1.2379919290542603, "learning_rate": 9.031371982073153e-05, "loss": 6.0734, "step": 43500 }, { "epoch": 3.091190108191654, "grad_norm": 1.2245177030563354, "learning_rate": 9.016914847477231e-05, "loss": 6.068, "step": 44000 }, { "epoch": 3.1263172685120133, "grad_norm": 1.2027391195297241, "learning_rate": 9.002457712881307e-05, "loss": 6.0532, "step": 44500 }, { "epoch": 3.161444428832373, "grad_norm": 1.1591733694076538, "learning_rate": 8.988000578285384e-05, "loss": 6.0472, "step": 45000 }, { "epoch": 3.196571589152733, "grad_norm": 1.1632347106933594, "learning_rate": 8.973543443689461e-05, "loss": 6.0353, "step": 45500 }, { "epoch": 3.2316987494730927, "grad_norm": 1.1952688694000244, "learning_rate": 8.959086309093538e-05, "loss": 6.0297, "step": 46000 }, { "epoch": 3.2668259097934524, "grad_norm": 1.2016774415969849, "learning_rate": 8.944629174497616e-05, "loss": 6.0198, "step": 46500 }, { "epoch": 3.301953070113812, "grad_norm": 1.1584421396255493, "learning_rate": 8.930172039901692e-05, "loss": 6.0094, "step": 47000 }, { "epoch": 3.3370802304341716, "grad_norm": 1.182223916053772, "learning_rate": 8.915714905305768e-05, "loss": 6.0024, "step": 47500 }, { "epoch": 3.3722073907545314, "grad_norm": 1.1740553379058838, "learning_rate": 8.901257770709845e-05, "loss": 5.9956, "step": 48000 }, { "epoch": 3.407334551074891, "grad_norm": 1.1617754697799683, "learning_rate": 8.886800636113923e-05, "loss": 5.9826, "step": 48500 }, { "epoch": 3.442461711395251, "grad_norm": 1.1171213388442993, "learning_rate": 8.872343501518e-05, "loss": 5.9699, "step": 49000 }, { "epoch": 3.4775888717156107, "grad_norm": 1.188246488571167, "learning_rate": 8.857886366922077e-05, "loss": 5.9637, "step": 49500 }, { "epoch": 3.51271603203597, "grad_norm": 1.1380078792572021, "learning_rate": 8.843429232326153e-05, "loss": 5.9566, "step": 50000 }, { "epoch": 3.51271603203597, "eval_accuracy": 0.600941284375621, "eval_loss": 5.570221424102783, "eval_runtime": 126.6911, "eval_samples_per_second": 952.758, "eval_steps_per_second": 5.959, "step": 50000 }, { "epoch": 3.54784319235633, "grad_norm": 1.2439249753952026, "learning_rate": 8.828972097730231e-05, "loss": 5.9511, "step": 50500 }, { "epoch": 3.5829703526766896, "grad_norm": 1.1788381338119507, "learning_rate": 8.814514963134307e-05, "loss": 5.9378, "step": 51000 }, { "epoch": 3.6180975129970494, "grad_norm": 1.1215739250183105, "learning_rate": 8.800057828538384e-05, "loss": 5.9362, "step": 51500 }, { "epoch": 3.6532246733174087, "grad_norm": 1.1777937412261963, "learning_rate": 8.78560069394246e-05, "loss": 5.9268, "step": 52000 }, { "epoch": 3.6883518336377685, "grad_norm": 1.1265531778335571, "learning_rate": 8.771143559346538e-05, "loss": 5.9175, "step": 52500 }, { "epoch": 3.7234789939581283, "grad_norm": 1.1818273067474365, "learning_rate": 8.756686424750616e-05, "loss": 5.9152, "step": 53000 }, { "epoch": 3.758606154278488, "grad_norm": 1.156366229057312, "learning_rate": 8.742229290154692e-05, "loss": 5.9103, "step": 53500 }, { "epoch": 3.793733314598848, "grad_norm": 1.1763765811920166, "learning_rate": 8.727772155558768e-05, "loss": 5.8977, "step": 54000 }, { "epoch": 3.8288604749192077, "grad_norm": 1.1385083198547363, "learning_rate": 8.713315020962846e-05, "loss": 5.8947, "step": 54500 }, { "epoch": 3.8639876352395675, "grad_norm": 1.2357500791549683, "learning_rate": 8.698857886366923e-05, "loss": 5.8851, "step": 55000 }, { "epoch": 3.899114795559927, "grad_norm": 1.1894572973251343, "learning_rate": 8.684400751770999e-05, "loss": 5.8796, "step": 55500 }, { "epoch": 3.9342419558802866, "grad_norm": 1.1474850177764893, "learning_rate": 8.669943617175075e-05, "loss": 5.8733, "step": 56000 }, { "epoch": 3.9693691162006464, "grad_norm": 1.1875358819961548, "learning_rate": 8.655486482579153e-05, "loss": 5.8684, "step": 56500 }, { "epoch": 4.004496276521006, "grad_norm": 1.1427265405654907, "learning_rate": 8.641029347983231e-05, "loss": 5.8589, "step": 57000 }, { "epoch": 4.0396234368413655, "grad_norm": 1.1655430793762207, "learning_rate": 8.626572213387307e-05, "loss": 5.8494, "step": 57500 }, { "epoch": 4.074750597161725, "grad_norm": 1.1547647714614868, "learning_rate": 8.612115078791384e-05, "loss": 5.8459, "step": 58000 }, { "epoch": 4.109877757482085, "grad_norm": 1.1440244913101196, "learning_rate": 8.597657944195461e-05, "loss": 5.841, "step": 58500 }, { "epoch": 4.145004917802445, "grad_norm": 1.1749067306518555, "learning_rate": 8.583200809599538e-05, "loss": 5.8332, "step": 59000 }, { "epoch": 4.180132078122805, "grad_norm": 1.1220006942749023, "learning_rate": 8.568743675003614e-05, "loss": 5.8332, "step": 59500 }, { "epoch": 4.215259238443164, "grad_norm": 1.1797089576721191, "learning_rate": 8.55428654040769e-05, "loss": 5.8259, "step": 60000 }, { "epoch": 4.215259238443164, "eval_accuracy": 0.6137438703389166, "eval_loss": 5.454646110534668, "eval_runtime": 125.9485, "eval_samples_per_second": 958.376, "eval_steps_per_second": 5.995, "step": 60000 }, { "epoch": 4.250386398763524, "grad_norm": 1.1942250728607178, "learning_rate": 8.539829405811768e-05, "loss": 5.8235, "step": 60500 }, { "epoch": 4.285513559083884, "grad_norm": 1.1691994667053223, "learning_rate": 8.525372271215846e-05, "loss": 5.8149, "step": 61000 }, { "epoch": 4.320640719404244, "grad_norm": 1.1399987936019897, "learning_rate": 8.510915136619923e-05, "loss": 5.8084, "step": 61500 }, { "epoch": 4.355767879724603, "grad_norm": 1.1549854278564453, "learning_rate": 8.496458002023999e-05, "loss": 5.8046, "step": 62000 }, { "epoch": 4.3908950400449624, "grad_norm": 1.1478040218353271, "learning_rate": 8.482000867428075e-05, "loss": 5.8053, "step": 62500 }, { "epoch": 4.426022200365322, "grad_norm": 1.2138196229934692, "learning_rate": 8.467543732832153e-05, "loss": 5.8002, "step": 63000 }, { "epoch": 4.461149360685682, "grad_norm": 1.203627586364746, "learning_rate": 8.45308659823623e-05, "loss": 5.7914, "step": 63500 }, { "epoch": 4.496276521006042, "grad_norm": 1.1575429439544678, "learning_rate": 8.438629463640307e-05, "loss": 5.7888, "step": 64000 }, { "epoch": 4.531403681326402, "grad_norm": 1.2016817331314087, "learning_rate": 8.424172329044384e-05, "loss": 5.7841, "step": 64500 }, { "epoch": 4.566530841646761, "grad_norm": 1.126859188079834, "learning_rate": 8.409715194448461e-05, "loss": 5.7793, "step": 65000 }, { "epoch": 4.601658001967121, "grad_norm": 1.1709495782852173, "learning_rate": 8.395258059852538e-05, "loss": 5.7754, "step": 65500 }, { "epoch": 4.636785162287481, "grad_norm": 1.2062314748764038, "learning_rate": 8.380800925256614e-05, "loss": 5.7684, "step": 66000 }, { "epoch": 4.671912322607841, "grad_norm": 1.195332646369934, "learning_rate": 8.36634379066069e-05, "loss": 5.7681, "step": 66500 }, { "epoch": 4.7070394829282005, "grad_norm": 1.1682814359664917, "learning_rate": 8.351886656064768e-05, "loss": 5.7583, "step": 67000 }, { "epoch": 4.742166643248559, "grad_norm": 1.1663920879364014, "learning_rate": 8.337429521468846e-05, "loss": 5.7591, "step": 67500 }, { "epoch": 4.777293803568919, "grad_norm": 1.1364179849624634, "learning_rate": 8.322972386872923e-05, "loss": 5.7557, "step": 68000 }, { "epoch": 4.812420963889279, "grad_norm": 1.1614656448364258, "learning_rate": 8.308515252276999e-05, "loss": 5.7535, "step": 68500 }, { "epoch": 4.847548124209639, "grad_norm": 1.2386486530303955, "learning_rate": 8.294058117681077e-05, "loss": 5.7443, "step": 69000 }, { "epoch": 4.8826752845299985, "grad_norm": 1.135480284690857, "learning_rate": 8.279600983085153e-05, "loss": 5.7433, "step": 69500 }, { "epoch": 4.917802444850358, "grad_norm": 1.1809818744659424, "learning_rate": 8.26514384848923e-05, "loss": 5.7396, "step": 70000 }, { "epoch": 4.917802444850358, "eval_accuracy": 0.6238602002028918, "eval_loss": 5.367178916931152, "eval_runtime": 126.4827, "eval_samples_per_second": 954.328, "eval_steps_per_second": 5.969, "step": 70000 }, { "epoch": 4.952929605170718, "grad_norm": 1.1215052604675293, "learning_rate": 8.250686713893306e-05, "loss": 5.7359, "step": 70500 }, { "epoch": 4.988056765491078, "grad_norm": 1.2044142484664917, "learning_rate": 8.236229579297384e-05, "loss": 5.7301, "step": 71000 }, { "epoch": 5.023183925811438, "grad_norm": 1.1988801956176758, "learning_rate": 8.221772444701461e-05, "loss": 5.7281, "step": 71500 }, { "epoch": 5.0583110861317975, "grad_norm": 1.1432907581329346, "learning_rate": 8.207315310105538e-05, "loss": 5.7195, "step": 72000 }, { "epoch": 5.093438246452157, "grad_norm": 1.1418726444244385, "learning_rate": 8.192858175509614e-05, "loss": 5.7137, "step": 72500 }, { "epoch": 5.128565406772516, "grad_norm": 1.1230759620666504, "learning_rate": 8.178401040913692e-05, "loss": 5.7094, "step": 73000 }, { "epoch": 5.163692567092876, "grad_norm": 1.1333376169204712, "learning_rate": 8.163943906317768e-05, "loss": 5.7077, "step": 73500 }, { "epoch": 5.198819727413236, "grad_norm": 1.1893682479858398, "learning_rate": 8.149486771721845e-05, "loss": 5.7031, "step": 74000 }, { "epoch": 5.2339468877335955, "grad_norm": 1.1560386419296265, "learning_rate": 8.135029637125921e-05, "loss": 5.705, "step": 74500 }, { "epoch": 5.269074048053955, "grad_norm": 1.2150088548660278, "learning_rate": 8.120572502529999e-05, "loss": 5.7003, "step": 75000 }, { "epoch": 5.304201208374315, "grad_norm": 1.1376268863677979, "learning_rate": 8.106115367934077e-05, "loss": 5.6964, "step": 75500 }, { "epoch": 5.339328368694675, "grad_norm": 1.1558517217636108, "learning_rate": 8.091658233338153e-05, "loss": 5.6926, "step": 76000 }, { "epoch": 5.374455529015035, "grad_norm": 1.2448821067810059, "learning_rate": 8.07720109874223e-05, "loss": 5.6902, "step": 76500 }, { "epoch": 5.409582689335394, "grad_norm": 1.2126572132110596, "learning_rate": 8.062743964146306e-05, "loss": 5.6846, "step": 77000 }, { "epoch": 5.444709849655754, "grad_norm": 1.2237986326217651, "learning_rate": 8.048286829550384e-05, "loss": 5.6792, "step": 77500 }, { "epoch": 5.479837009976113, "grad_norm": 1.1750723123550415, "learning_rate": 8.03382969495446e-05, "loss": 5.6804, "step": 78000 }, { "epoch": 5.514964170296473, "grad_norm": 1.176378846168518, "learning_rate": 8.019372560358536e-05, "loss": 5.6773, "step": 78500 }, { "epoch": 5.550091330616833, "grad_norm": 1.1815321445465088, "learning_rate": 8.004915425762614e-05, "loss": 5.6733, "step": 79000 }, { "epoch": 5.5852184909371925, "grad_norm": 1.2011853456497192, "learning_rate": 7.990458291166692e-05, "loss": 5.67, "step": 79500 }, { "epoch": 5.620345651257552, "grad_norm": 1.1479657888412476, "learning_rate": 7.976001156570768e-05, "loss": 5.6664, "step": 80000 }, { "epoch": 5.620345651257552, "eval_accuracy": 0.6312497210348582, "eval_loss": 5.307363033294678, "eval_runtime": 126.1055, "eval_samples_per_second": 957.183, "eval_steps_per_second": 5.987, "step": 80000 }, { "epoch": 5.655472811577912, "grad_norm": 1.166767954826355, "learning_rate": 7.961544021974845e-05, "loss": 5.663, "step": 80500 }, { "epoch": 5.690599971898272, "grad_norm": 1.1171019077301025, "learning_rate": 7.947086887378921e-05, "loss": 5.6688, "step": 81000 }, { "epoch": 5.725727132218632, "grad_norm": 1.1799217462539673, "learning_rate": 7.932629752782999e-05, "loss": 5.6605, "step": 81500 }, { "epoch": 5.760854292538991, "grad_norm": 1.1959232091903687, "learning_rate": 7.918172618187075e-05, "loss": 5.6538, "step": 82000 }, { "epoch": 5.795981452859351, "grad_norm": 1.1740916967391968, "learning_rate": 7.903715483591153e-05, "loss": 5.6553, "step": 82500 }, { "epoch": 5.83110861317971, "grad_norm": 1.1855255365371704, "learning_rate": 7.88925834899523e-05, "loss": 5.6524, "step": 83000 }, { "epoch": 5.866235773500071, "grad_norm": 1.200687289237976, "learning_rate": 7.874801214399307e-05, "loss": 5.6493, "step": 83500 }, { "epoch": 5.90136293382043, "grad_norm": 1.1658035516738892, "learning_rate": 7.860344079803384e-05, "loss": 5.6458, "step": 84000 }, { "epoch": 5.936490094140789, "grad_norm": 1.134080410003662, "learning_rate": 7.84588694520746e-05, "loss": 5.6432, "step": 84500 }, { "epoch": 5.971617254461149, "grad_norm": 1.2157056331634521, "learning_rate": 7.831429810611536e-05, "loss": 5.6425, "step": 85000 }, { "epoch": 6.006744414781509, "grad_norm": 1.2122453451156616, "learning_rate": 7.816972676015614e-05, "loss": 5.6396, "step": 85500 }, { "epoch": 6.041871575101869, "grad_norm": 1.1487025022506714, "learning_rate": 7.802515541419692e-05, "loss": 5.6319, "step": 86000 }, { "epoch": 6.076998735422229, "grad_norm": 1.0889257192611694, "learning_rate": 7.788058406823768e-05, "loss": 5.6311, "step": 86500 }, { "epoch": 6.112125895742588, "grad_norm": 1.1648160219192505, "learning_rate": 7.773601272227845e-05, "loss": 5.626, "step": 87000 }, { "epoch": 6.147253056062948, "grad_norm": 1.1622391939163208, "learning_rate": 7.759144137631922e-05, "loss": 5.6254, "step": 87500 }, { "epoch": 6.182380216383308, "grad_norm": 1.139646291732788, "learning_rate": 7.744687003035999e-05, "loss": 5.624, "step": 88000 }, { "epoch": 6.217507376703667, "grad_norm": 1.1604758501052856, "learning_rate": 7.730229868440075e-05, "loss": 5.6197, "step": 88500 }, { "epoch": 6.252634537024027, "grad_norm": 1.1392582654953003, "learning_rate": 7.715772733844152e-05, "loss": 5.6168, "step": 89000 }, { "epoch": 6.287761697344386, "grad_norm": 1.2056643962860107, "learning_rate": 7.70131559924823e-05, "loss": 5.618, "step": 89500 }, { "epoch": 6.322888857664746, "grad_norm": 1.1435514688491821, "learning_rate": 7.686858464652307e-05, "loss": 5.6155, "step": 90000 }, { "epoch": 6.322888857664746, "eval_accuracy": 0.6366031945913234, "eval_loss": 5.262214183807373, "eval_runtime": 126.3601, "eval_samples_per_second": 955.254, "eval_steps_per_second": 5.975, "step": 90000 }, { "epoch": 6.358016017985106, "grad_norm": 1.1635953187942505, "learning_rate": 7.672401330056384e-05, "loss": 5.6131, "step": 90500 }, { "epoch": 6.393143178305466, "grad_norm": 1.1402812004089355, "learning_rate": 7.65794419546046e-05, "loss": 5.6101, "step": 91000 }, { "epoch": 6.4282703386258255, "grad_norm": 1.1227062940597534, "learning_rate": 7.643487060864536e-05, "loss": 5.6073, "step": 91500 }, { "epoch": 6.463397498946185, "grad_norm": 1.2081722021102905, "learning_rate": 7.629029926268614e-05, "loss": 5.6058, "step": 92000 }, { "epoch": 6.498524659266545, "grad_norm": 1.1431390047073364, "learning_rate": 7.61457279167269e-05, "loss": 5.6047, "step": 92500 }, { "epoch": 6.533651819586905, "grad_norm": 1.1397309303283691, "learning_rate": 7.600115657076767e-05, "loss": 5.6031, "step": 93000 }, { "epoch": 6.568778979907265, "grad_norm": 1.2204087972640991, "learning_rate": 7.585658522480845e-05, "loss": 5.5988, "step": 93500 }, { "epoch": 6.603906140227624, "grad_norm": 1.1261354684829712, "learning_rate": 7.571201387884922e-05, "loss": 5.5956, "step": 94000 }, { "epoch": 6.639033300547983, "grad_norm": 1.1716442108154297, "learning_rate": 7.556744253288999e-05, "loss": 5.5936, "step": 94500 }, { "epoch": 6.674160460868343, "grad_norm": 1.1498794555664062, "learning_rate": 7.542287118693075e-05, "loss": 5.5911, "step": 95000 }, { "epoch": 6.709287621188703, "grad_norm": 1.1422091722488403, "learning_rate": 7.527829984097152e-05, "loss": 5.5895, "step": 95500 }, { "epoch": 6.744414781509063, "grad_norm": 1.1866816282272339, "learning_rate": 7.51337284950123e-05, "loss": 5.5945, "step": 96000 }, { "epoch": 6.7795419418294225, "grad_norm": 1.1697109937667847, "learning_rate": 7.498915714905306e-05, "loss": 5.5832, "step": 96500 }, { "epoch": 6.814669102149782, "grad_norm": 1.1553871631622314, "learning_rate": 7.484458580309382e-05, "loss": 5.585, "step": 97000 }, { "epoch": 6.849796262470142, "grad_norm": 1.1596871614456177, "learning_rate": 7.47000144571346e-05, "loss": 5.5776, "step": 97500 }, { "epoch": 6.884923422790502, "grad_norm": 1.1737241744995117, "learning_rate": 7.455544311117538e-05, "loss": 5.5789, "step": 98000 }, { "epoch": 6.920050583110862, "grad_norm": 1.1882907152175903, "learning_rate": 7.441087176521614e-05, "loss": 5.5753, "step": 98500 }, { "epoch": 6.955177743431221, "grad_norm": 1.1603020429611206, "learning_rate": 7.42663004192569e-05, "loss": 5.5745, "step": 99000 }, { "epoch": 6.99030490375158, "grad_norm": 1.142120599746704, "learning_rate": 7.412172907329767e-05, "loss": 5.5709, "step": 99500 }, { "epoch": 7.02543206407194, "grad_norm": 1.1305012702941895, "learning_rate": 7.397715772733845e-05, "loss": 5.5704, "step": 100000 }, { "epoch": 7.02543206407194, "eval_accuracy": 0.6416274953957863, "eval_loss": 5.21771764755249, "eval_runtime": 126.1825, "eval_samples_per_second": 956.599, "eval_steps_per_second": 5.983, "step": 100000 }, { "epoch": 7.0605592243923, "grad_norm": 1.1883624792099, "learning_rate": 7.383258638137921e-05, "loss": 5.5674, "step": 100500 }, { "epoch": 7.09568638471266, "grad_norm": 1.1774970293045044, "learning_rate": 7.368801503541999e-05, "loss": 5.5672, "step": 101000 }, { "epoch": 7.1308135450330195, "grad_norm": 1.2163723707199097, "learning_rate": 7.354344368946075e-05, "loss": 5.5656, "step": 101500 }, { "epoch": 7.165940705353379, "grad_norm": 1.162907361984253, "learning_rate": 7.339887234350153e-05, "loss": 5.561, "step": 102000 }, { "epoch": 7.201067865673739, "grad_norm": 1.1895023584365845, "learning_rate": 7.32543009975423e-05, "loss": 5.5598, "step": 102500 }, { "epoch": 7.236195025994099, "grad_norm": 1.147109866142273, "learning_rate": 7.310972965158306e-05, "loss": 5.5594, "step": 103000 }, { "epoch": 7.271322186314459, "grad_norm": 1.148010015487671, "learning_rate": 7.296515830562382e-05, "loss": 5.5601, "step": 103500 }, { "epoch": 7.306449346634818, "grad_norm": 1.158715844154358, "learning_rate": 7.28205869596646e-05, "loss": 5.5536, "step": 104000 }, { "epoch": 7.341576506955178, "grad_norm": 1.1228585243225098, "learning_rate": 7.267601561370536e-05, "loss": 5.5471, "step": 104500 }, { "epoch": 7.376703667275537, "grad_norm": 1.188887357711792, "learning_rate": 7.253144426774614e-05, "loss": 5.5538, "step": 105000 }, { "epoch": 7.411830827595897, "grad_norm": 1.1512972116470337, "learning_rate": 7.23868729217869e-05, "loss": 5.5546, "step": 105500 }, { "epoch": 7.446957987916257, "grad_norm": 1.1770893335342407, "learning_rate": 7.224230157582768e-05, "loss": 5.5529, "step": 106000 }, { "epoch": 7.482085148236616, "grad_norm": 1.2063955068588257, "learning_rate": 7.209773022986845e-05, "loss": 5.5475, "step": 106500 }, { "epoch": 7.517212308556976, "grad_norm": 1.1827296018600464, "learning_rate": 7.195315888390921e-05, "loss": 5.5428, "step": 107000 }, { "epoch": 7.552339468877336, "grad_norm": 1.1626031398773193, "learning_rate": 7.180858753794997e-05, "loss": 5.5431, "step": 107500 }, { "epoch": 7.587466629197696, "grad_norm": 1.1745115518569946, "learning_rate": 7.166401619199075e-05, "loss": 5.5416, "step": 108000 }, { "epoch": 7.6225937895180556, "grad_norm": 1.155131459236145, "learning_rate": 7.151944484603153e-05, "loss": 5.5438, "step": 108500 }, { "epoch": 7.657720949838415, "grad_norm": 1.1387735605239868, "learning_rate": 7.13748735000723e-05, "loss": 5.5384, "step": 109000 }, { "epoch": 7.692848110158775, "grad_norm": 1.2025445699691772, "learning_rate": 7.123030215411306e-05, "loss": 5.5382, "step": 109500 }, { "epoch": 7.727975270479135, "grad_norm": 1.1272386312484741, "learning_rate": 7.108573080815382e-05, "loss": 5.5381, "step": 110000 }, { "epoch": 7.727975270479135, "eval_accuracy": 0.6459911230157876, "eval_loss": 5.186947822570801, "eval_runtime": 127.0503, "eval_samples_per_second": 950.065, "eval_steps_per_second": 5.943, "step": 110000 }, { "epoch": 7.763102430799494, "grad_norm": 1.151397466659546, "learning_rate": 7.09411594621946e-05, "loss": 5.5362, "step": 110500 }, { "epoch": 7.798229591119854, "grad_norm": 1.2249643802642822, "learning_rate": 7.079658811623536e-05, "loss": 5.532, "step": 111000 }, { "epoch": 7.833356751440213, "grad_norm": 1.1769822835922241, "learning_rate": 7.065201677027613e-05, "loss": 5.5312, "step": 111500 }, { "epoch": 7.868483911760573, "grad_norm": 1.1882879734039307, "learning_rate": 7.05074454243169e-05, "loss": 5.5319, "step": 112000 }, { "epoch": 7.903611072080933, "grad_norm": 1.0995951890945435, "learning_rate": 7.036287407835768e-05, "loss": 5.5326, "step": 112500 }, { "epoch": 7.938738232401293, "grad_norm": 1.1851023435592651, "learning_rate": 7.021830273239845e-05, "loss": 5.5233, "step": 113000 }, { "epoch": 7.9738653927216525, "grad_norm": 1.1201878786087036, "learning_rate": 7.007373138643921e-05, "loss": 5.5234, "step": 113500 }, { "epoch": 8.008992553042011, "grad_norm": 1.132328987121582, "learning_rate": 6.992916004047997e-05, "loss": 5.5244, "step": 114000 }, { "epoch": 8.044119713362372, "grad_norm": 1.1754218339920044, "learning_rate": 6.978458869452075e-05, "loss": 5.5191, "step": 114500 }, { "epoch": 8.079246873682731, "grad_norm": 1.2175325155258179, "learning_rate": 6.964001734856152e-05, "loss": 5.5171, "step": 115000 }, { "epoch": 8.114374034003092, "grad_norm": 1.1968905925750732, "learning_rate": 6.949544600260228e-05, "loss": 5.5166, "step": 115500 }, { "epoch": 8.14950119432345, "grad_norm": 1.1892062425613403, "learning_rate": 6.935087465664306e-05, "loss": 5.518, "step": 116000 }, { "epoch": 8.184628354643811, "grad_norm": 1.2034764289855957, "learning_rate": 6.920630331068383e-05, "loss": 5.5178, "step": 116500 }, { "epoch": 8.21975551496417, "grad_norm": 1.1592258214950562, "learning_rate": 6.90617319647246e-05, "loss": 5.5125, "step": 117000 }, { "epoch": 8.25488267528453, "grad_norm": 1.1501221656799316, "learning_rate": 6.891716061876536e-05, "loss": 5.513, "step": 117500 }, { "epoch": 8.29000983560489, "grad_norm": 1.19719398021698, "learning_rate": 6.877258927280613e-05, "loss": 5.5069, "step": 118000 }, { "epoch": 8.325136995925249, "grad_norm": 1.1488349437713623, "learning_rate": 6.86280179268469e-05, "loss": 5.5066, "step": 118500 }, { "epoch": 8.36026415624561, "grad_norm": 1.2181166410446167, "learning_rate": 6.848344658088767e-05, "loss": 5.5085, "step": 119000 }, { "epoch": 8.395391316565968, "grad_norm": 1.2024781703948975, "learning_rate": 6.833887523492845e-05, "loss": 5.5072, "step": 119500 }, { "epoch": 8.430518476886329, "grad_norm": 1.1667537689208984, "learning_rate": 6.819430388896921e-05, "loss": 5.5072, "step": 120000 }, { "epoch": 8.430518476886329, "eval_accuracy": 0.6494610822515428, "eval_loss": 5.157218933105469, "eval_runtime": 126.0791, "eval_samples_per_second": 957.383, "eval_steps_per_second": 5.988, "step": 120000 }, { "epoch": 8.465645637206688, "grad_norm": 1.1520241498947144, "learning_rate": 6.804973254300999e-05, "loss": 5.5051, "step": 120500 }, { "epoch": 8.500772797527048, "grad_norm": 1.2032662630081177, "learning_rate": 6.790516119705075e-05, "loss": 5.5049, "step": 121000 }, { "epoch": 8.535899957847407, "grad_norm": 1.175516128540039, "learning_rate": 6.776058985109152e-05, "loss": 5.506, "step": 121500 }, { "epoch": 8.571027118167768, "grad_norm": 1.1792590618133545, "learning_rate": 6.761601850513228e-05, "loss": 5.5012, "step": 122000 }, { "epoch": 8.606154278488127, "grad_norm": 1.1310416460037231, "learning_rate": 6.747144715917306e-05, "loss": 5.5011, "step": 122500 }, { "epoch": 8.641281438808488, "grad_norm": 1.1733070611953735, "learning_rate": 6.732687581321382e-05, "loss": 5.4989, "step": 123000 }, { "epoch": 8.676408599128846, "grad_norm": 1.19496750831604, "learning_rate": 6.71823044672546e-05, "loss": 5.4927, "step": 123500 }, { "epoch": 8.711535759449205, "grad_norm": 1.1653318405151367, "learning_rate": 6.703773312129536e-05, "loss": 5.4922, "step": 124000 }, { "epoch": 8.746662919769566, "grad_norm": 1.1700496673583984, "learning_rate": 6.689316177533613e-05, "loss": 5.4876, "step": 124500 }, { "epoch": 8.781790080089925, "grad_norm": 1.1617351770401, "learning_rate": 6.67485904293769e-05, "loss": 5.4924, "step": 125000 }, { "epoch": 8.816917240410286, "grad_norm": 1.1872036457061768, "learning_rate": 6.660401908341767e-05, "loss": 5.4911, "step": 125500 }, { "epoch": 8.852044400730644, "grad_norm": 1.1824781894683838, "learning_rate": 6.645944773745843e-05, "loss": 5.4903, "step": 126000 }, { "epoch": 8.887171561051005, "grad_norm": 1.1772197484970093, "learning_rate": 6.631487639149921e-05, "loss": 5.4895, "step": 126500 }, { "epoch": 8.922298721371364, "grad_norm": 1.2576428651809692, "learning_rate": 6.617030504553999e-05, "loss": 5.4869, "step": 127000 }, { "epoch": 8.957425881691725, "grad_norm": 1.1482747793197632, "learning_rate": 6.602573369958075e-05, "loss": 5.4923, "step": 127500 }, { "epoch": 8.992553042012084, "grad_norm": 1.1709399223327637, "learning_rate": 6.588116235362152e-05, "loss": 5.4898, "step": 128000 }, { "epoch": 9.027680202332444, "grad_norm": 1.2026115655899048, "learning_rate": 6.573659100766228e-05, "loss": 5.4814, "step": 128500 }, { "epoch": 9.062807362652803, "grad_norm": 1.1449785232543945, "learning_rate": 6.559201966170306e-05, "loss": 5.4804, "step": 129000 }, { "epoch": 9.097934522973162, "grad_norm": 1.2136037349700928, "learning_rate": 6.544744831574382e-05, "loss": 5.4817, "step": 129500 }, { "epoch": 9.133061683293523, "grad_norm": 1.1829079389572144, "learning_rate": 6.530287696978458e-05, "loss": 5.476, "step": 130000 }, { "epoch": 9.133061683293523, "eval_accuracy": 0.651950457160933, "eval_loss": 5.139862537384033, "eval_runtime": 126.0544, "eval_samples_per_second": 957.571, "eval_steps_per_second": 5.989, "step": 130000 }, { "epoch": 9.168188843613882, "grad_norm": 1.1634808778762817, "learning_rate": 6.515830562382536e-05, "loss": 5.4777, "step": 130500 }, { "epoch": 9.203316003934242, "grad_norm": 1.1809887886047363, "learning_rate": 6.501373427786614e-05, "loss": 5.4791, "step": 131000 }, { "epoch": 9.238443164254601, "grad_norm": 1.1817798614501953, "learning_rate": 6.48691629319069e-05, "loss": 5.4783, "step": 131500 }, { "epoch": 9.273570324574962, "grad_norm": 1.114335298538208, "learning_rate": 6.472459158594767e-05, "loss": 5.4785, "step": 132000 }, { "epoch": 9.30869748489532, "grad_norm": 1.2152668237686157, "learning_rate": 6.458002023998843e-05, "loss": 5.4777, "step": 132500 }, { "epoch": 9.343824645215681, "grad_norm": 1.186461091041565, "learning_rate": 6.443544889402921e-05, "loss": 5.4736, "step": 133000 }, { "epoch": 9.37895180553604, "grad_norm": 1.2176927328109741, "learning_rate": 6.429087754806997e-05, "loss": 5.4737, "step": 133500 }, { "epoch": 9.414078965856401, "grad_norm": 1.1699422597885132, "learning_rate": 6.414630620211074e-05, "loss": 5.4728, "step": 134000 }, { "epoch": 9.44920612617676, "grad_norm": 1.173329472541809, "learning_rate": 6.400173485615151e-05, "loss": 5.4672, "step": 134500 }, { "epoch": 9.484333286497119, "grad_norm": 1.191834568977356, "learning_rate": 6.385716351019229e-05, "loss": 5.4668, "step": 135000 }, { "epoch": 9.51946044681748, "grad_norm": 1.1519852876663208, "learning_rate": 6.371259216423306e-05, "loss": 5.4688, "step": 135500 }, { "epoch": 9.554587607137838, "grad_norm": 1.1618045568466187, "learning_rate": 6.356802081827382e-05, "loss": 5.4679, "step": 136000 }, { "epoch": 9.589714767458199, "grad_norm": 1.1754469871520996, "learning_rate": 6.342344947231458e-05, "loss": 5.4675, "step": 136500 }, { "epoch": 9.624841927778558, "grad_norm": 1.1873865127563477, "learning_rate": 6.327887812635536e-05, "loss": 5.4647, "step": 137000 }, { "epoch": 9.659969088098919, "grad_norm": 1.1484997272491455, "learning_rate": 6.313430678039613e-05, "loss": 5.4644, "step": 137500 }, { "epoch": 9.695096248419278, "grad_norm": 1.21599280834198, "learning_rate": 6.29897354344369e-05, "loss": 5.4576, "step": 138000 }, { "epoch": 9.730223408739638, "grad_norm": 1.205138921737671, "learning_rate": 6.284516408847767e-05, "loss": 5.4654, "step": 138500 }, { "epoch": 9.765350569059997, "grad_norm": 1.1492888927459717, "learning_rate": 6.270059274251843e-05, "loss": 5.4601, "step": 139000 }, { "epoch": 9.800477729380358, "grad_norm": 1.1409257650375366, "learning_rate": 6.255602139655921e-05, "loss": 5.4613, "step": 139500 }, { "epoch": 9.835604889700717, "grad_norm": 1.179166555404663, "learning_rate": 6.241145005059997e-05, "loss": 5.4586, "step": 140000 }, { "epoch": 9.835604889700717, "eval_accuracy": 0.6553756299489875, "eval_loss": 5.114386558532715, "eval_runtime": 127.5499, "eval_samples_per_second": 946.343, "eval_steps_per_second": 5.919, "step": 140000 }, { "epoch": 9.870732050021076, "grad_norm": 1.2062571048736572, "learning_rate": 6.226687870464074e-05, "loss": 5.4591, "step": 140500 }, { "epoch": 9.905859210341436, "grad_norm": 1.133453369140625, "learning_rate": 6.212230735868151e-05, "loss": 5.4572, "step": 141000 }, { "epoch": 9.940986370661795, "grad_norm": 1.146584153175354, "learning_rate": 6.197773601272228e-05, "loss": 5.4572, "step": 141500 }, { "epoch": 9.976113530982156, "grad_norm": 1.1408936977386475, "learning_rate": 6.183316466676306e-05, "loss": 5.4527, "step": 142000 }, { "epoch": 10.011240691302515, "grad_norm": 1.190815806388855, "learning_rate": 6.168859332080382e-05, "loss": 5.4565, "step": 142500 }, { "epoch": 10.046367851622875, "grad_norm": 1.1872814893722534, "learning_rate": 6.154402197484458e-05, "loss": 5.4533, "step": 143000 }, { "epoch": 10.081495011943234, "grad_norm": 1.1626367568969727, "learning_rate": 6.139945062888536e-05, "loss": 5.4486, "step": 143500 }, { "epoch": 10.116622172263595, "grad_norm": 1.1665513515472412, "learning_rate": 6.125487928292613e-05, "loss": 5.4483, "step": 144000 }, { "epoch": 10.151749332583954, "grad_norm": 1.1779884099960327, "learning_rate": 6.111030793696689e-05, "loss": 5.4488, "step": 144500 }, { "epoch": 10.186876492904315, "grad_norm": 1.2090669870376587, "learning_rate": 6.096573659100766e-05, "loss": 5.4484, "step": 145000 }, { "epoch": 10.222003653224673, "grad_norm": 1.177783489227295, "learning_rate": 6.082116524504844e-05, "loss": 5.4495, "step": 145500 }, { "epoch": 10.257130813545032, "grad_norm": 1.1222484111785889, "learning_rate": 6.067659389908921e-05, "loss": 5.4499, "step": 146000 }, { "epoch": 10.292257973865393, "grad_norm": 1.1884996891021729, "learning_rate": 6.053202255312997e-05, "loss": 5.4434, "step": 146500 }, { "epoch": 10.327385134185752, "grad_norm": 1.181362271308899, "learning_rate": 6.0387451207170744e-05, "loss": 5.4437, "step": 147000 }, { "epoch": 10.362512294506113, "grad_norm": 1.1353344917297363, "learning_rate": 6.024287986121151e-05, "loss": 5.4445, "step": 147500 }, { "epoch": 10.397639454826471, "grad_norm": 1.149796485900879, "learning_rate": 6.009830851525228e-05, "loss": 5.4407, "step": 148000 }, { "epoch": 10.432766615146832, "grad_norm": 1.1714903116226196, "learning_rate": 5.995373716929304e-05, "loss": 5.445, "step": 148500 }, { "epoch": 10.467893775467191, "grad_norm": 1.1469001770019531, "learning_rate": 5.980916582333381e-05, "loss": 5.4408, "step": 149000 }, { "epoch": 10.503020935787552, "grad_norm": 1.1510165929794312, "learning_rate": 5.966459447737459e-05, "loss": 5.4378, "step": 149500 }, { "epoch": 10.53814809610791, "grad_norm": 1.1331630945205688, "learning_rate": 5.952002313141536e-05, "loss": 5.4395, "step": 150000 }, { "epoch": 10.53814809610791, "eval_accuracy": 0.6572613131285701, "eval_loss": 5.097984790802002, "eval_runtime": 126.1907, "eval_samples_per_second": 956.537, "eval_steps_per_second": 5.983, "step": 150000 }, { "epoch": 10.573275256428271, "grad_norm": 1.1784754991531372, "learning_rate": 5.9375451785456126e-05, "loss": 5.4446, "step": 150500 }, { "epoch": 10.60840241674863, "grad_norm": 1.2072874307632446, "learning_rate": 5.9230880439496896e-05, "loss": 5.4382, "step": 151000 }, { "epoch": 10.643529577068989, "grad_norm": 1.2143604755401611, "learning_rate": 5.908630909353766e-05, "loss": 5.4386, "step": 151500 }, { "epoch": 10.67865673738935, "grad_norm": 1.1461095809936523, "learning_rate": 5.894173774757843e-05, "loss": 5.4369, "step": 152000 }, { "epoch": 10.713783897709709, "grad_norm": 1.17250657081604, "learning_rate": 5.8797166401619195e-05, "loss": 5.434, "step": 152500 }, { "epoch": 10.74891105803007, "grad_norm": 1.1867562532424927, "learning_rate": 5.865259505565997e-05, "loss": 5.4354, "step": 153000 }, { "epoch": 10.784038218350428, "grad_norm": 1.164304256439209, "learning_rate": 5.8508023709700743e-05, "loss": 5.4367, "step": 153500 }, { "epoch": 10.819165378670789, "grad_norm": 1.1781518459320068, "learning_rate": 5.8363452363741514e-05, "loss": 5.433, "step": 154000 }, { "epoch": 10.854292538991148, "grad_norm": 1.1799439191818237, "learning_rate": 5.821888101778228e-05, "loss": 5.4332, "step": 154500 }, { "epoch": 10.889419699311508, "grad_norm": 1.193841576576233, "learning_rate": 5.807430967182305e-05, "loss": 5.4329, "step": 155000 }, { "epoch": 10.924546859631867, "grad_norm": 1.179405689239502, "learning_rate": 5.792973832586381e-05, "loss": 5.4336, "step": 155500 }, { "epoch": 10.959674019952226, "grad_norm": 1.1666207313537598, "learning_rate": 5.7785166979904584e-05, "loss": 5.4322, "step": 156000 }, { "epoch": 10.994801180272587, "grad_norm": 1.1545931100845337, "learning_rate": 5.764059563394536e-05, "loss": 5.4297, "step": 156500 }, { "epoch": 11.029928340592946, "grad_norm": 1.214898943901062, "learning_rate": 5.7496024287986125e-05, "loss": 5.4227, "step": 157000 }, { "epoch": 11.065055500913306, "grad_norm": 1.1414653062820435, "learning_rate": 5.7351452942026896e-05, "loss": 5.4272, "step": 157500 }, { "epoch": 11.100182661233665, "grad_norm": 1.167088270187378, "learning_rate": 5.720688159606766e-05, "loss": 5.4254, "step": 158000 }, { "epoch": 11.135309821554026, "grad_norm": 1.1644870042800903, "learning_rate": 5.706231025010843e-05, "loss": 5.4236, "step": 158500 }, { "epoch": 11.170436981874385, "grad_norm": 1.177067518234253, "learning_rate": 5.6917738904149195e-05, "loss": 5.424, "step": 159000 }, { "epoch": 11.205564142194746, "grad_norm": 1.1633716821670532, "learning_rate": 5.6773167558189966e-05, "loss": 5.4264, "step": 159500 }, { "epoch": 11.240691302515105, "grad_norm": 1.1675889492034912, "learning_rate": 5.6628596212230736e-05, "loss": 5.4279, "step": 160000 }, { "epoch": 11.240691302515105, "eval_accuracy": 0.6587713141993589, "eval_loss": 5.085355758666992, "eval_runtime": 127.6699, "eval_samples_per_second": 945.454, "eval_steps_per_second": 5.914, "step": 160000 }, { "epoch": 11.275818462835465, "grad_norm": 1.1570359468460083, "learning_rate": 5.6484024866271514e-05, "loss": 5.4212, "step": 160500 }, { "epoch": 11.310945623155824, "grad_norm": 1.1573740243911743, "learning_rate": 5.633945352031228e-05, "loss": 5.4259, "step": 161000 }, { "epoch": 11.346072783476183, "grad_norm": 1.145591378211975, "learning_rate": 5.619488217435305e-05, "loss": 5.421, "step": 161500 }, { "epoch": 11.381199943796544, "grad_norm": 1.172127604484558, "learning_rate": 5.605031082839381e-05, "loss": 5.4217, "step": 162000 }, { "epoch": 11.416327104116903, "grad_norm": 1.1558443307876587, "learning_rate": 5.5905739482434584e-05, "loss": 5.4173, "step": 162500 }, { "epoch": 11.451454264437263, "grad_norm": 1.1067453622817993, "learning_rate": 5.576116813647535e-05, "loss": 5.4134, "step": 163000 }, { "epoch": 11.486581424757622, "grad_norm": 1.2098469734191895, "learning_rate": 5.561659679051612e-05, "loss": 5.417, "step": 163500 }, { "epoch": 11.521708585077983, "grad_norm": 1.1795800924301147, "learning_rate": 5.5472025444556896e-05, "loss": 5.4172, "step": 164000 }, { "epoch": 11.556835745398342, "grad_norm": 1.1982141733169556, "learning_rate": 5.532745409859767e-05, "loss": 5.4164, "step": 164500 }, { "epoch": 11.591962905718702, "grad_norm": 1.1516317129135132, "learning_rate": 5.518288275263843e-05, "loss": 5.416, "step": 165000 }, { "epoch": 11.627090066039061, "grad_norm": 1.1769285202026367, "learning_rate": 5.50383114066792e-05, "loss": 5.4167, "step": 165500 }, { "epoch": 11.662217226359422, "grad_norm": 1.1476564407348633, "learning_rate": 5.4893740060719965e-05, "loss": 5.4124, "step": 166000 }, { "epoch": 11.69734438667978, "grad_norm": 1.1805812120437622, "learning_rate": 5.4749168714760736e-05, "loss": 5.4143, "step": 166500 }, { "epoch": 11.73247154700014, "grad_norm": 1.1290351152420044, "learning_rate": 5.46045973688015e-05, "loss": 5.4084, "step": 167000 }, { "epoch": 11.7675987073205, "grad_norm": 1.169985294342041, "learning_rate": 5.446002602284227e-05, "loss": 5.4093, "step": 167500 }, { "epoch": 11.80272586764086, "grad_norm": 1.1833126544952393, "learning_rate": 5.431545467688305e-05, "loss": 5.4114, "step": 168000 }, { "epoch": 11.83785302796122, "grad_norm": 1.2261061668395996, "learning_rate": 5.417088333092382e-05, "loss": 5.4071, "step": 168500 }, { "epoch": 11.872980188281579, "grad_norm": 1.1827682256698608, "learning_rate": 5.402631198496458e-05, "loss": 5.4074, "step": 169000 }, { "epoch": 11.90810734860194, "grad_norm": 1.1952297687530518, "learning_rate": 5.3881740639005354e-05, "loss": 5.4073, "step": 169500 }, { "epoch": 11.943234508922298, "grad_norm": 1.1410478353500366, "learning_rate": 5.373716929304612e-05, "loss": 5.4084, "step": 170000 }, { "epoch": 11.943234508922298, "eval_accuracy": 0.6609756014030679, "eval_loss": 5.069380760192871, "eval_runtime": 128.1719, "eval_samples_per_second": 941.751, "eval_steps_per_second": 5.891, "step": 170000 }, { "epoch": 11.978361669242659, "grad_norm": 1.2068756818771362, "learning_rate": 5.359259794708689e-05, "loss": 5.4064, "step": 170500 }, { "epoch": 12.013488829563018, "grad_norm": 1.202238917350769, "learning_rate": 5.344802660112765e-05, "loss": 5.4087, "step": 171000 }, { "epoch": 12.048615989883379, "grad_norm": 1.1819404363632202, "learning_rate": 5.330345525516843e-05, "loss": 5.4046, "step": 171500 }, { "epoch": 12.083743150203738, "grad_norm": 1.1757937669754028, "learning_rate": 5.31588839092092e-05, "loss": 5.4025, "step": 172000 }, { "epoch": 12.118870310524096, "grad_norm": 1.201640009880066, "learning_rate": 5.3014312563249965e-05, "loss": 5.4035, "step": 172500 }, { "epoch": 12.153997470844457, "grad_norm": 1.151092290878296, "learning_rate": 5.2869741217290736e-05, "loss": 5.4076, "step": 173000 }, { "epoch": 12.189124631164816, "grad_norm": 1.1719902753829956, "learning_rate": 5.272516987133151e-05, "loss": 5.4034, "step": 173500 }, { "epoch": 12.224251791485177, "grad_norm": 1.179118275642395, "learning_rate": 5.258059852537227e-05, "loss": 5.3971, "step": 174000 }, { "epoch": 12.259378951805536, "grad_norm": 1.1940786838531494, "learning_rate": 5.243602717941304e-05, "loss": 5.4024, "step": 174500 }, { "epoch": 12.294506112125896, "grad_norm": 1.1985485553741455, "learning_rate": 5.229145583345382e-05, "loss": 5.3962, "step": 175000 }, { "epoch": 12.329633272446255, "grad_norm": 1.1246628761291504, "learning_rate": 5.214688448749458e-05, "loss": 5.4027, "step": 175500 }, { "epoch": 12.364760432766616, "grad_norm": 1.1976138353347778, "learning_rate": 5.2002313141535354e-05, "loss": 5.3989, "step": 176000 }, { "epoch": 12.399887593086975, "grad_norm": 1.1607234477996826, "learning_rate": 5.185774179557612e-05, "loss": 5.3976, "step": 176500 }, { "epoch": 12.435014753407334, "grad_norm": 1.173844814300537, "learning_rate": 5.171317044961689e-05, "loss": 5.3997, "step": 177000 }, { "epoch": 12.470141913727694, "grad_norm": 1.1754404306411743, "learning_rate": 5.156859910365765e-05, "loss": 5.3969, "step": 177500 }, { "epoch": 12.505269074048053, "grad_norm": 1.1915861368179321, "learning_rate": 5.1424027757698423e-05, "loss": 5.3993, "step": 178000 }, { "epoch": 12.540396234368414, "grad_norm": 1.216575264930725, "learning_rate": 5.127945641173919e-05, "loss": 5.3997, "step": 178500 }, { "epoch": 12.575523394688773, "grad_norm": 1.1933749914169312, "learning_rate": 5.113488506577997e-05, "loss": 5.3953, "step": 179000 }, { "epoch": 12.610650555009133, "grad_norm": 1.1539580821990967, "learning_rate": 5.0990313719820736e-05, "loss": 5.3951, "step": 179500 }, { "epoch": 12.645777715329492, "grad_norm": 1.1963214874267578, "learning_rate": 5.0845742373861507e-05, "loss": 5.3943, "step": 180000 }, { "epoch": 12.645777715329492, "eval_accuracy": 0.6627021865306387, "eval_loss": 5.054413318634033, "eval_runtime": 127.5991, "eval_samples_per_second": 945.979, "eval_steps_per_second": 5.917, "step": 180000 }, { "epoch": 12.680904875649853, "grad_norm": 1.217602252960205, "learning_rate": 5.070117102790227e-05, "loss": 5.3934, "step": 180500 }, { "epoch": 12.716032035970212, "grad_norm": 1.2188109159469604, "learning_rate": 5.055659968194304e-05, "loss": 5.3941, "step": 181000 }, { "epoch": 12.751159196290573, "grad_norm": 1.1604007482528687, "learning_rate": 5.0412028335983805e-05, "loss": 5.3931, "step": 181500 }, { "epoch": 12.786286356610931, "grad_norm": 1.1772806644439697, "learning_rate": 5.0267456990024576e-05, "loss": 5.3924, "step": 182000 }, { "epoch": 12.82141351693129, "grad_norm": 1.1824551820755005, "learning_rate": 5.0122885644065354e-05, "loss": 5.3952, "step": 182500 }, { "epoch": 12.856540677251651, "grad_norm": 1.203250765800476, "learning_rate": 4.997831429810612e-05, "loss": 5.3908, "step": 183000 }, { "epoch": 12.89166783757201, "grad_norm": 1.1858001947402954, "learning_rate": 4.983374295214689e-05, "loss": 5.3911, "step": 183500 }, { "epoch": 12.92679499789237, "grad_norm": 1.1396986246109009, "learning_rate": 4.968917160618766e-05, "loss": 5.3878, "step": 184000 }, { "epoch": 12.96192215821273, "grad_norm": 1.155434250831604, "learning_rate": 4.954460026022842e-05, "loss": 5.3912, "step": 184500 }, { "epoch": 12.99704931853309, "grad_norm": 1.1653069257736206, "learning_rate": 4.9400028914269194e-05, "loss": 5.3908, "step": 185000 }, { "epoch": 13.032176478853449, "grad_norm": 1.2010760307312012, "learning_rate": 4.9255457568309965e-05, "loss": 5.385, "step": 185500 }, { "epoch": 13.06730363917381, "grad_norm": 1.2106468677520752, "learning_rate": 4.9110886222350736e-05, "loss": 5.3855, "step": 186000 }, { "epoch": 13.102430799494169, "grad_norm": 1.1811589002609253, "learning_rate": 4.89663148763915e-05, "loss": 5.386, "step": 186500 }, { "epoch": 13.13755795981453, "grad_norm": 1.1451094150543213, "learning_rate": 4.882174353043227e-05, "loss": 5.3811, "step": 187000 }, { "epoch": 13.172685120134888, "grad_norm": 1.2099153995513916, "learning_rate": 4.867717218447304e-05, "loss": 5.384, "step": 187500 }, { "epoch": 13.207812280455247, "grad_norm": 1.1451085805892944, "learning_rate": 4.853260083851381e-05, "loss": 5.3843, "step": 188000 }, { "epoch": 13.242939440775608, "grad_norm": 1.1829298734664917, "learning_rate": 4.8388029492554576e-05, "loss": 5.3837, "step": 188500 }, { "epoch": 13.278066601095967, "grad_norm": 1.133914589881897, "learning_rate": 4.824345814659535e-05, "loss": 5.381, "step": 189000 }, { "epoch": 13.313193761416327, "grad_norm": 1.1763981580734253, "learning_rate": 4.809888680063612e-05, "loss": 5.3856, "step": 189500 }, { "epoch": 13.348320921736686, "grad_norm": 1.1866949796676636, "learning_rate": 4.795431545467688e-05, "loss": 5.3829, "step": 190000 }, { "epoch": 13.348320921736686, "eval_accuracy": 0.6635887965904411, "eval_loss": 5.047701358795166, "eval_runtime": 126.6818, "eval_samples_per_second": 952.828, "eval_steps_per_second": 5.96, "step": 190000 }, { "epoch": 13.383448082057047, "grad_norm": 1.1739898920059204, "learning_rate": 4.780974410871765e-05, "loss": 5.3778, "step": 190500 }, { "epoch": 13.418575242377406, "grad_norm": 1.1885180473327637, "learning_rate": 4.766517276275842e-05, "loss": 5.3809, "step": 191000 }, { "epoch": 13.453702402697767, "grad_norm": 1.1664230823516846, "learning_rate": 4.7520601416799194e-05, "loss": 5.3792, "step": 191500 }, { "epoch": 13.488829563018125, "grad_norm": 1.1393156051635742, "learning_rate": 4.737603007083996e-05, "loss": 5.3803, "step": 192000 }, { "epoch": 13.523956723338486, "grad_norm": 1.1409962177276611, "learning_rate": 4.723145872488073e-05, "loss": 5.3821, "step": 192500 }, { "epoch": 13.559083883658845, "grad_norm": 1.267981767654419, "learning_rate": 4.70868873789215e-05, "loss": 5.3795, "step": 193000 }, { "epoch": 13.594211043979204, "grad_norm": 1.1796077489852905, "learning_rate": 4.694231603296227e-05, "loss": 5.3789, "step": 193500 }, { "epoch": 13.629338204299565, "grad_norm": 1.2097166776657104, "learning_rate": 4.6797744687003034e-05, "loss": 5.3763, "step": 194000 }, { "epoch": 13.664465364619923, "grad_norm": 1.1931965351104736, "learning_rate": 4.665317334104381e-05, "loss": 5.375, "step": 194500 }, { "epoch": 13.699592524940284, "grad_norm": 1.1692981719970703, "learning_rate": 4.6508601995084576e-05, "loss": 5.3771, "step": 195000 }, { "epoch": 13.734719685260643, "grad_norm": 1.1764898300170898, "learning_rate": 4.6364030649125346e-05, "loss": 5.3764, "step": 195500 }, { "epoch": 13.769846845581004, "grad_norm": 1.1638944149017334, "learning_rate": 4.621945930316611e-05, "loss": 5.3777, "step": 196000 }, { "epoch": 13.804974005901363, "grad_norm": 1.2403713464736938, "learning_rate": 4.607488795720689e-05, "loss": 5.3738, "step": 196500 }, { "epoch": 13.840101166221723, "grad_norm": 1.2131552696228027, "learning_rate": 4.593031661124765e-05, "loss": 5.3749, "step": 197000 }, { "epoch": 13.875228326542082, "grad_norm": 1.1957006454467773, "learning_rate": 4.578574526528842e-05, "loss": 5.3721, "step": 197500 }, { "epoch": 13.910355486862443, "grad_norm": 1.1719543933868408, "learning_rate": 4.564117391932919e-05, "loss": 5.3727, "step": 198000 }, { "epoch": 13.945482647182802, "grad_norm": 1.2030900716781616, "learning_rate": 4.5496602573369964e-05, "loss": 5.3725, "step": 198500 }, { "epoch": 13.98060980750316, "grad_norm": 1.1576215028762817, "learning_rate": 4.535203122741073e-05, "loss": 5.378, "step": 199000 }, { "epoch": 14.015736967823521, "grad_norm": 1.1843230724334717, "learning_rate": 4.52074598814515e-05, "loss": 5.371, "step": 199500 }, { "epoch": 14.05086412814388, "grad_norm": 1.2949912548065186, "learning_rate": 4.506288853549227e-05, "loss": 5.374, "step": 200000 }, { "epoch": 14.05086412814388, "eval_accuracy": 0.6652399780865722, "eval_loss": 5.036068916320801, "eval_runtime": 126.4369, "eval_samples_per_second": 954.674, "eval_steps_per_second": 5.971, "step": 200000 }, { "epoch": 14.08599128846424, "grad_norm": 1.1451940536499023, "learning_rate": 4.491831718953304e-05, "loss": 5.3675, "step": 200500 }, { "epoch": 14.1211184487846, "grad_norm": 1.2271783351898193, "learning_rate": 4.4773745843573805e-05, "loss": 5.3711, "step": 201000 }, { "epoch": 14.15624560910496, "grad_norm": 1.2067863941192627, "learning_rate": 4.4629174497614575e-05, "loss": 5.3676, "step": 201500 }, { "epoch": 14.19137276942532, "grad_norm": 1.195374846458435, "learning_rate": 4.4484603151655346e-05, "loss": 5.3657, "step": 202000 }, { "epoch": 14.22649992974568, "grad_norm": 1.1865936517715454, "learning_rate": 4.434003180569612e-05, "loss": 5.3663, "step": 202500 }, { "epoch": 14.261627090066039, "grad_norm": 1.1512318849563599, "learning_rate": 4.419546045973688e-05, "loss": 5.3691, "step": 203000 }, { "epoch": 14.2967542503864, "grad_norm": 1.2092339992523193, "learning_rate": 4.405088911377765e-05, "loss": 5.3662, "step": 203500 }, { "epoch": 14.331881410706758, "grad_norm": 1.2129069566726685, "learning_rate": 4.390631776781842e-05, "loss": 5.3698, "step": 204000 }, { "epoch": 14.367008571027117, "grad_norm": 1.1823105812072754, "learning_rate": 4.3761746421859187e-05, "loss": 5.367, "step": 204500 }, { "epoch": 14.402135731347478, "grad_norm": 1.1684224605560303, "learning_rate": 4.361717507589996e-05, "loss": 5.3645, "step": 205000 }, { "epoch": 14.437262891667837, "grad_norm": 1.1812078952789307, "learning_rate": 4.347260372994073e-05, "loss": 5.3675, "step": 205500 }, { "epoch": 14.472390051988198, "grad_norm": 1.1713589429855347, "learning_rate": 4.33280323839815e-05, "loss": 5.3661, "step": 206000 }, { "epoch": 14.507517212308557, "grad_norm": 1.2227543592453003, "learning_rate": 4.318346103802226e-05, "loss": 5.3626, "step": 206500 }, { "epoch": 14.542644372628917, "grad_norm": 1.1285825967788696, "learning_rate": 4.3038889692063034e-05, "loss": 5.3615, "step": 207000 }, { "epoch": 14.577771532949276, "grad_norm": 1.1806303262710571, "learning_rate": 4.2894318346103804e-05, "loss": 5.3626, "step": 207500 }, { "epoch": 14.612898693269637, "grad_norm": 1.2527612447738647, "learning_rate": 4.2749747000144575e-05, "loss": 5.3632, "step": 208000 }, { "epoch": 14.648025853589996, "grad_norm": 1.140224814414978, "learning_rate": 4.260517565418534e-05, "loss": 5.3613, "step": 208500 }, { "epoch": 14.683153013910356, "grad_norm": 1.2229194641113281, "learning_rate": 4.246060430822611e-05, "loss": 5.3628, "step": 209000 }, { "epoch": 14.718280174230715, "grad_norm": 1.212386965751648, "learning_rate": 4.231603296226688e-05, "loss": 5.3616, "step": 209500 }, { "epoch": 14.753407334551074, "grad_norm": 1.183121681213379, "learning_rate": 4.217146161630765e-05, "loss": 5.3602, "step": 210000 }, { "epoch": 14.753407334551074, "eval_accuracy": 0.6666274526157097, "eval_loss": 5.025681018829346, "eval_runtime": 127.2546, "eval_samples_per_second": 948.539, "eval_steps_per_second": 5.933, "step": 210000 }, { "epoch": 14.788534494871435, "grad_norm": 1.1668003797531128, "learning_rate": 4.2026890270348416e-05, "loss": 5.3591, "step": 210500 }, { "epoch": 14.823661655191794, "grad_norm": 1.2255350351333618, "learning_rate": 4.1882318924389186e-05, "loss": 5.3586, "step": 211000 }, { "epoch": 14.858788815512154, "grad_norm": 1.13953697681427, "learning_rate": 4.173774757842996e-05, "loss": 5.3612, "step": 211500 }, { "epoch": 14.893915975832513, "grad_norm": 1.1586263179779053, "learning_rate": 4.159317623247073e-05, "loss": 5.3586, "step": 212000 }, { "epoch": 14.929043136152874, "grad_norm": 1.226428508758545, "learning_rate": 4.144860488651149e-05, "loss": 5.3609, "step": 212500 }, { "epoch": 14.964170296473233, "grad_norm": 1.157700538635254, "learning_rate": 4.130403354055227e-05, "loss": 5.3607, "step": 213000 }, { "epoch": 14.999297456793593, "grad_norm": 1.1561223268508911, "learning_rate": 4.1159462194593033e-05, "loss": 5.3604, "step": 213500 }, { "epoch": 15.034424617113952, "grad_norm": 1.180484414100647, "learning_rate": 4.1014890848633804e-05, "loss": 5.3533, "step": 214000 }, { "epoch": 15.069551777434313, "grad_norm": 1.195053219795227, "learning_rate": 4.087031950267457e-05, "loss": 5.3553, "step": 214500 }, { "epoch": 15.104678937754672, "grad_norm": 1.1267015933990479, "learning_rate": 4.0725748156715346e-05, "loss": 5.3568, "step": 215000 }, { "epoch": 15.13980609807503, "grad_norm": 1.1654679775238037, "learning_rate": 4.058117681075611e-05, "loss": 5.3516, "step": 215500 }, { "epoch": 15.174933258395392, "grad_norm": 1.1776633262634277, "learning_rate": 4.043660546479688e-05, "loss": 5.3564, "step": 216000 }, { "epoch": 15.21006041871575, "grad_norm": 1.201765775680542, "learning_rate": 4.0292034118837645e-05, "loss": 5.3562, "step": 216500 }, { "epoch": 15.245187579036111, "grad_norm": 1.2297497987747192, "learning_rate": 4.014746277287842e-05, "loss": 5.3537, "step": 217000 }, { "epoch": 15.28031473935647, "grad_norm": 1.2239570617675781, "learning_rate": 4.0002891426919186e-05, "loss": 5.3525, "step": 217500 }, { "epoch": 15.31544189967683, "grad_norm": 1.2314610481262207, "learning_rate": 3.985832008095996e-05, "loss": 5.3533, "step": 218000 }, { "epoch": 15.35056905999719, "grad_norm": 1.1848505735397339, "learning_rate": 3.971374873500072e-05, "loss": 5.3534, "step": 218500 }, { "epoch": 15.38569622031755, "grad_norm": 1.2602753639221191, "learning_rate": 3.95691773890415e-05, "loss": 5.3526, "step": 219000 }, { "epoch": 15.42082338063791, "grad_norm": 1.1745126247406006, "learning_rate": 3.942460604308226e-05, "loss": 5.3526, "step": 219500 }, { "epoch": 15.45595054095827, "grad_norm": 1.188979148864746, "learning_rate": 3.928003469712303e-05, "loss": 5.3506, "step": 220000 }, { "epoch": 15.45595054095827, "eval_accuracy": 0.6680702098599409, "eval_loss": 5.015492916107178, "eval_runtime": 127.3339, "eval_samples_per_second": 947.948, "eval_steps_per_second": 5.929, "step": 220000 }, { "epoch": 15.491077701278629, "grad_norm": 1.2011260986328125, "learning_rate": 3.9135463351163804e-05, "loss": 5.3519, "step": 220500 }, { "epoch": 15.526204861598988, "grad_norm": 1.175593614578247, "learning_rate": 3.899089200520457e-05, "loss": 5.3532, "step": 221000 }, { "epoch": 15.561332021919348, "grad_norm": 1.2360161542892456, "learning_rate": 3.884632065924534e-05, "loss": 5.3508, "step": 221500 }, { "epoch": 15.596459182239707, "grad_norm": 1.2163788080215454, "learning_rate": 3.870174931328611e-05, "loss": 5.3516, "step": 222000 }, { "epoch": 15.631586342560068, "grad_norm": 1.181384801864624, "learning_rate": 3.855717796732688e-05, "loss": 5.3485, "step": 222500 }, { "epoch": 15.666713502880427, "grad_norm": 1.2437771558761597, "learning_rate": 3.8412606621367644e-05, "loss": 5.3497, "step": 223000 }, { "epoch": 15.701840663200787, "grad_norm": 1.2318360805511475, "learning_rate": 3.8268035275408415e-05, "loss": 5.3503, "step": 223500 }, { "epoch": 15.736967823521146, "grad_norm": 1.2126661539077759, "learning_rate": 3.812346392944918e-05, "loss": 5.3474, "step": 224000 }, { "epoch": 15.772094983841507, "grad_norm": 1.1528905630111694, "learning_rate": 3.797889258348996e-05, "loss": 5.3465, "step": 224500 }, { "epoch": 15.807222144161866, "grad_norm": 1.217221975326538, "learning_rate": 3.783432123753072e-05, "loss": 5.3439, "step": 225000 }, { "epoch": 15.842349304482227, "grad_norm": 1.2265459299087524, "learning_rate": 3.768974989157149e-05, "loss": 5.3444, "step": 225500 }, { "epoch": 15.877476464802585, "grad_norm": 1.1632280349731445, "learning_rate": 3.754517854561226e-05, "loss": 5.3486, "step": 226000 }, { "epoch": 15.912603625122944, "grad_norm": 1.2100650072097778, "learning_rate": 3.740060719965303e-05, "loss": 5.3494, "step": 226500 }, { "epoch": 15.947730785443305, "grad_norm": 1.181597113609314, "learning_rate": 3.72560358536938e-05, "loss": 5.3461, "step": 227000 }, { "epoch": 15.982857945763664, "grad_norm": 1.2156709432601929, "learning_rate": 3.711146450773457e-05, "loss": 5.3468, "step": 227500 }, { "epoch": 16.017985106084023, "grad_norm": 1.2033458948135376, "learning_rate": 3.696689316177534e-05, "loss": 5.343, "step": 228000 }, { "epoch": 16.053112266404383, "grad_norm": 1.2176637649536133, "learning_rate": 3.682232181581611e-05, "loss": 5.3408, "step": 228500 }, { "epoch": 16.088239426724744, "grad_norm": 1.246570348739624, "learning_rate": 3.667775046985687e-05, "loss": 5.341, "step": 229000 }, { "epoch": 16.123366587045105, "grad_norm": 1.1855249404907227, "learning_rate": 3.6533179123897644e-05, "loss": 5.3418, "step": 229500 }, { "epoch": 16.158493747365462, "grad_norm": 1.1909562349319458, "learning_rate": 3.6388607777938415e-05, "loss": 5.3443, "step": 230000 }, { "epoch": 16.158493747365462, "eval_accuracy": 0.6687147634926975, "eval_loss": 5.0102643966674805, "eval_runtime": 151.6637, "eval_samples_per_second": 795.879, "eval_steps_per_second": 4.978, "step": 230000 }, { "epoch": 16.193620907685823, "grad_norm": 1.1971921920776367, "learning_rate": 3.6244036431979186e-05, "loss": 5.3402, "step": 230500 }, { "epoch": 16.228748068006183, "grad_norm": 1.2171639204025269, "learning_rate": 3.609946508601995e-05, "loss": 5.3427, "step": 231000 }, { "epoch": 16.26387522832654, "grad_norm": 1.193764328956604, "learning_rate": 3.595489374006073e-05, "loss": 5.3376, "step": 231500 }, { "epoch": 16.2990023886469, "grad_norm": 1.1574360132217407, "learning_rate": 3.581032239410149e-05, "loss": 5.3433, "step": 232000 }, { "epoch": 16.33412954896726, "grad_norm": 1.2496017217636108, "learning_rate": 3.566575104814226e-05, "loss": 5.3392, "step": 232500 }, { "epoch": 16.369256709287622, "grad_norm": 1.1936050653457642, "learning_rate": 3.5521179702183026e-05, "loss": 5.3379, "step": 233000 }, { "epoch": 16.40438386960798, "grad_norm": 1.1835389137268066, "learning_rate": 3.5376608356223804e-05, "loss": 5.3394, "step": 233500 }, { "epoch": 16.43951102992834, "grad_norm": 1.1349616050720215, "learning_rate": 3.523203701026457e-05, "loss": 5.3375, "step": 234000 }, { "epoch": 16.4746381902487, "grad_norm": 1.2075680494308472, "learning_rate": 3.508746566430534e-05, "loss": 5.3402, "step": 234500 }, { "epoch": 16.50976535056906, "grad_norm": 1.1728594303131104, "learning_rate": 3.49428943183461e-05, "loss": 5.3366, "step": 235000 }, { "epoch": 16.54489251088942, "grad_norm": 1.204835295677185, "learning_rate": 3.479832297238687e-05, "loss": 5.3349, "step": 235500 }, { "epoch": 16.58001967120978, "grad_norm": 1.172888159751892, "learning_rate": 3.4653751626427644e-05, "loss": 5.3396, "step": 236000 }, { "epoch": 16.61514683153014, "grad_norm": 1.2482709884643555, "learning_rate": 3.4509180280468415e-05, "loss": 5.3357, "step": 236500 }, { "epoch": 16.650273991850497, "grad_norm": 1.2217960357666016, "learning_rate": 3.436460893450918e-05, "loss": 5.343, "step": 237000 }, { "epoch": 16.685401152170858, "grad_norm": 1.237485408782959, "learning_rate": 3.422003758854995e-05, "loss": 5.3368, "step": 237500 }, { "epoch": 16.72052831249122, "grad_norm": 1.2395271062850952, "learning_rate": 3.407546624259072e-05, "loss": 5.337, "step": 238000 }, { "epoch": 16.75565547281158, "grad_norm": 1.228334665298462, "learning_rate": 3.393089489663149e-05, "loss": 5.3385, "step": 238500 }, { "epoch": 16.790782633131936, "grad_norm": 1.183610200881958, "learning_rate": 3.378632355067226e-05, "loss": 5.3355, "step": 239000 }, { "epoch": 16.825909793452297, "grad_norm": 1.1948573589324951, "learning_rate": 3.3641752204713026e-05, "loss": 5.3355, "step": 239500 }, { "epoch": 16.861036953772658, "grad_norm": 1.1900110244750977, "learning_rate": 3.3497180858753797e-05, "loss": 5.3334, "step": 240000 }, { "epoch": 16.861036953772658, "eval_accuracy": 0.6697086668751692, "eval_loss": 5.0030341148376465, "eval_runtime": 143.0107, "eval_samples_per_second": 844.035, "eval_steps_per_second": 5.279, "step": 240000 }, { "epoch": 16.89616411409302, "grad_norm": 1.1953723430633545, "learning_rate": 3.335260951279456e-05, "loss": 5.3337, "step": 240500 }, { "epoch": 16.931291274413375, "grad_norm": 1.2045830488204956, "learning_rate": 3.320803816683534e-05, "loss": 5.3394, "step": 241000 }, { "epoch": 16.966418434733736, "grad_norm": 1.238985300064087, "learning_rate": 3.30634668208761e-05, "loss": 5.3327, "step": 241500 }, { "epoch": 17.001545595054097, "grad_norm": 1.215276837348938, "learning_rate": 3.291889547491687e-05, "loss": 5.3371, "step": 242000 }, { "epoch": 17.036672755374454, "grad_norm": 1.2557088136672974, "learning_rate": 3.277432412895764e-05, "loss": 5.3354, "step": 242500 }, { "epoch": 17.071799915694815, "grad_norm": 1.2225830554962158, "learning_rate": 3.2629752782998414e-05, "loss": 5.3329, "step": 243000 }, { "epoch": 17.106927076015175, "grad_norm": 1.2303576469421387, "learning_rate": 3.248518143703918e-05, "loss": 5.3323, "step": 243500 }, { "epoch": 17.142054236335536, "grad_norm": 1.2233577966690063, "learning_rate": 3.234061009107995e-05, "loss": 5.3292, "step": 244000 }, { "epoch": 17.177181396655893, "grad_norm": 1.226540207862854, "learning_rate": 3.219603874512072e-05, "loss": 5.3287, "step": 244500 }, { "epoch": 17.212308556976254, "grad_norm": 1.2142741680145264, "learning_rate": 3.205146739916149e-05, "loss": 5.3323, "step": 245000 }, { "epoch": 17.247435717296614, "grad_norm": 1.2251989841461182, "learning_rate": 3.1906896053202255e-05, "loss": 5.3295, "step": 245500 }, { "epoch": 17.282562877616975, "grad_norm": 1.2491041421890259, "learning_rate": 3.1762324707243026e-05, "loss": 5.3327, "step": 246000 }, { "epoch": 17.317690037937332, "grad_norm": 1.1981154680252075, "learning_rate": 3.1617753361283796e-05, "loss": 5.3294, "step": 246500 }, { "epoch": 17.352817198257693, "grad_norm": 1.1962498426437378, "learning_rate": 3.147318201532457e-05, "loss": 5.3272, "step": 247000 }, { "epoch": 17.387944358578054, "grad_norm": 1.2217235565185547, "learning_rate": 3.132861066936533e-05, "loss": 5.3306, "step": 247500 }, { "epoch": 17.42307151889841, "grad_norm": 1.2277535200119019, "learning_rate": 3.11840393234061e-05, "loss": 5.3226, "step": 248000 }, { "epoch": 17.45819867921877, "grad_norm": 1.1723650693893433, "learning_rate": 3.103946797744687e-05, "loss": 5.3262, "step": 248500 }, { "epoch": 17.493325839539132, "grad_norm": 1.2194057703018188, "learning_rate": 3.0894896631487643e-05, "loss": 5.3244, "step": 249000 }, { "epoch": 17.528452999859493, "grad_norm": 1.2663850784301758, "learning_rate": 3.075032528552841e-05, "loss": 5.326, "step": 249500 }, { "epoch": 17.56358016017985, "grad_norm": 1.1696590185165405, "learning_rate": 3.060575393956918e-05, "loss": 5.3252, "step": 250000 }, { "epoch": 17.56358016017985, "eval_accuracy": 0.6704656707267438, "eval_loss": 4.996424674987793, "eval_runtime": 143.7761, "eval_samples_per_second": 839.541, "eval_steps_per_second": 5.251, "step": 250000 }, { "epoch": 17.59870732050021, "grad_norm": 1.252528190612793, "learning_rate": 3.046118259360995e-05, "loss": 5.3312, "step": 250500 }, { "epoch": 17.63383448082057, "grad_norm": 1.219990849494934, "learning_rate": 3.0316611247650716e-05, "loss": 5.3237, "step": 251000 }, { "epoch": 17.668961641140932, "grad_norm": 1.182396650314331, "learning_rate": 3.0172039901691484e-05, "loss": 5.3259, "step": 251500 }, { "epoch": 17.70408880146129, "grad_norm": 1.208756923675537, "learning_rate": 3.0027468555732258e-05, "loss": 5.3271, "step": 252000 }, { "epoch": 17.73921596178165, "grad_norm": 1.214400053024292, "learning_rate": 2.9882897209773025e-05, "loss": 5.3255, "step": 252500 }, { "epoch": 17.77434312210201, "grad_norm": 1.1972471475601196, "learning_rate": 2.9738325863813793e-05, "loss": 5.3242, "step": 253000 }, { "epoch": 17.809470282422367, "grad_norm": 1.2323074340820312, "learning_rate": 2.959375451785456e-05, "loss": 5.3267, "step": 253500 }, { "epoch": 17.844597442742728, "grad_norm": 1.2003233432769775, "learning_rate": 2.9449183171895334e-05, "loss": 5.3235, "step": 254000 }, { "epoch": 17.87972460306309, "grad_norm": 1.2235249280929565, "learning_rate": 2.93046118259361e-05, "loss": 5.3278, "step": 254500 }, { "epoch": 17.91485176338345, "grad_norm": 1.2437875270843506, "learning_rate": 2.916004047997687e-05, "loss": 5.3248, "step": 255000 }, { "epoch": 17.949978923703807, "grad_norm": 1.2585877180099487, "learning_rate": 2.9015469134017636e-05, "loss": 5.3251, "step": 255500 }, { "epoch": 17.985106084024167, "grad_norm": 1.205811858177185, "learning_rate": 2.887089778805841e-05, "loss": 5.3232, "step": 256000 }, { "epoch": 18.020233244344528, "grad_norm": 1.1605521440505981, "learning_rate": 2.8726326442099178e-05, "loss": 5.3241, "step": 256500 }, { "epoch": 18.05536040466489, "grad_norm": 1.2066062688827515, "learning_rate": 2.8581755096139945e-05, "loss": 5.3235, "step": 257000 }, { "epoch": 18.090487564985246, "grad_norm": 1.213157057762146, "learning_rate": 2.843718375018072e-05, "loss": 5.3186, "step": 257500 }, { "epoch": 18.125614725305606, "grad_norm": 1.2181349992752075, "learning_rate": 2.8292612404221487e-05, "loss": 5.3224, "step": 258000 }, { "epoch": 18.160741885625967, "grad_norm": 1.2189512252807617, "learning_rate": 2.8148041058262254e-05, "loss": 5.3212, "step": 258500 }, { "epoch": 18.195869045946324, "grad_norm": 1.218788981437683, "learning_rate": 2.800346971230302e-05, "loss": 5.3208, "step": 259000 }, { "epoch": 18.230996206266685, "grad_norm": 1.2304424047470093, "learning_rate": 2.7858898366343792e-05, "loss": 5.3215, "step": 259500 }, { "epoch": 18.266123366587045, "grad_norm": 1.231799602508545, "learning_rate": 2.7714327020384563e-05, "loss": 5.3187, "step": 260000 }, { "epoch": 18.266123366587045, "eval_accuracy": 0.6711306107089177, "eval_loss": 4.99044942855835, "eval_runtime": 140.4251, "eval_samples_per_second": 859.576, "eval_steps_per_second": 5.377, "step": 260000 }, { "epoch": 18.301250526907406, "grad_norm": 1.2224138975143433, "learning_rate": 2.756975567442533e-05, "loss": 5.3189, "step": 260500 }, { "epoch": 18.336377687227763, "grad_norm": 1.2453559637069702, "learning_rate": 2.7425184328466098e-05, "loss": 5.3161, "step": 261000 }, { "epoch": 18.371504847548124, "grad_norm": 1.2224081754684448, "learning_rate": 2.728061298250687e-05, "loss": 5.3222, "step": 261500 }, { "epoch": 18.406632007868485, "grad_norm": 1.207418441772461, "learning_rate": 2.7136041636547636e-05, "loss": 5.3192, "step": 262000 }, { "epoch": 18.441759168188845, "grad_norm": 1.233441948890686, "learning_rate": 2.6991470290588407e-05, "loss": 5.315, "step": 262500 }, { "epoch": 18.476886328509202, "grad_norm": 1.2389020919799805, "learning_rate": 2.6846898944629178e-05, "loss": 5.3162, "step": 263000 }, { "epoch": 18.512013488829563, "grad_norm": 1.208122968673706, "learning_rate": 2.6702327598669945e-05, "loss": 5.3158, "step": 263500 }, { "epoch": 18.547140649149924, "grad_norm": 1.2151379585266113, "learning_rate": 2.6557756252710713e-05, "loss": 5.315, "step": 264000 }, { "epoch": 18.58226780947028, "grad_norm": 1.2691776752471924, "learning_rate": 2.641318490675148e-05, "loss": 5.3157, "step": 264500 }, { "epoch": 18.61739496979064, "grad_norm": 1.2351784706115723, "learning_rate": 2.6268613560792254e-05, "loss": 5.3195, "step": 265000 }, { "epoch": 18.652522130111002, "grad_norm": 1.2182989120483398, "learning_rate": 2.612404221483302e-05, "loss": 5.316, "step": 265500 }, { "epoch": 18.687649290431363, "grad_norm": 1.2158135175704956, "learning_rate": 2.597947086887379e-05, "loss": 5.3147, "step": 266000 }, { "epoch": 18.72277645075172, "grad_norm": 1.2246290445327759, "learning_rate": 2.5834899522914556e-05, "loss": 5.317, "step": 266500 }, { "epoch": 18.75790361107208, "grad_norm": 1.1721491813659668, "learning_rate": 2.569032817695533e-05, "loss": 5.3172, "step": 267000 }, { "epoch": 18.79303077139244, "grad_norm": 1.2415876388549805, "learning_rate": 2.5545756830996098e-05, "loss": 5.3177, "step": 267500 }, { "epoch": 18.828157931712802, "grad_norm": 1.2287744283676147, "learning_rate": 2.5401185485036865e-05, "loss": 5.3174, "step": 268000 }, { "epoch": 18.86328509203316, "grad_norm": 1.2606596946716309, "learning_rate": 2.5256614139077633e-05, "loss": 5.3151, "step": 268500 }, { "epoch": 18.89841225235352, "grad_norm": 1.1996958255767822, "learning_rate": 2.5112042793118407e-05, "loss": 5.3157, "step": 269000 }, { "epoch": 18.93353941267388, "grad_norm": 1.259444236755371, "learning_rate": 2.4967471447159174e-05, "loss": 5.3152, "step": 269500 }, { "epoch": 18.968666572994238, "grad_norm": 1.1930986642837524, "learning_rate": 2.4822900101199945e-05, "loss": 5.3167, "step": 270000 }, { "epoch": 18.968666572994238, "eval_accuracy": 0.6723161969411823, "eval_loss": 4.984919548034668, "eval_runtime": 135.8309, "eval_samples_per_second": 888.649, "eval_steps_per_second": 5.558, "step": 270000 }, { "epoch": 19.0037937333146, "grad_norm": 1.2136319875717163, "learning_rate": 2.4678328755240712e-05, "loss": 5.3147, "step": 270500 }, { "epoch": 19.03892089363496, "grad_norm": 1.2361167669296265, "learning_rate": 2.4533757409281483e-05, "loss": 5.3116, "step": 271000 }, { "epoch": 19.07404805395532, "grad_norm": 1.246456503868103, "learning_rate": 2.438918606332225e-05, "loss": 5.314, "step": 271500 }, { "epoch": 19.109175214275677, "grad_norm": 1.2121821641921997, "learning_rate": 2.424461471736302e-05, "loss": 5.3114, "step": 272000 }, { "epoch": 19.144302374596037, "grad_norm": 1.2199236154556274, "learning_rate": 2.410004337140379e-05, "loss": 5.3115, "step": 272500 }, { "epoch": 19.179429534916398, "grad_norm": 1.2184224128723145, "learning_rate": 2.395547202544456e-05, "loss": 5.3126, "step": 273000 }, { "epoch": 19.21455669523676, "grad_norm": 1.208341360092163, "learning_rate": 2.3810900679485327e-05, "loss": 5.3078, "step": 273500 }, { "epoch": 19.249683855557116, "grad_norm": 1.194269061088562, "learning_rate": 2.3666329333526098e-05, "loss": 5.3096, "step": 274000 }, { "epoch": 19.284811015877477, "grad_norm": 1.2189325094223022, "learning_rate": 2.3521757987566865e-05, "loss": 5.3102, "step": 274500 }, { "epoch": 19.319938176197837, "grad_norm": 1.228743076324463, "learning_rate": 2.3377186641607636e-05, "loss": 5.3135, "step": 275000 }, { "epoch": 19.355065336518194, "grad_norm": 1.2371536493301392, "learning_rate": 2.3232615295648403e-05, "loss": 5.3119, "step": 275500 }, { "epoch": 19.390192496838555, "grad_norm": 1.1781997680664062, "learning_rate": 2.3088043949689174e-05, "loss": 5.3066, "step": 276000 }, { "epoch": 19.425319657158916, "grad_norm": 1.2210091352462769, "learning_rate": 2.2943472603729945e-05, "loss": 5.312, "step": 276500 }, { "epoch": 19.460446817479276, "grad_norm": 1.2152116298675537, "learning_rate": 2.2798901257770712e-05, "loss": 5.3064, "step": 277000 }, { "epoch": 19.495573977799634, "grad_norm": 1.2245073318481445, "learning_rate": 2.265432991181148e-05, "loss": 5.3114, "step": 277500 }, { "epoch": 19.530701138119994, "grad_norm": 1.224804162979126, "learning_rate": 2.250975856585225e-05, "loss": 5.3069, "step": 278000 }, { "epoch": 19.565828298440355, "grad_norm": 1.244660496711731, "learning_rate": 2.2365187219893018e-05, "loss": 5.3118, "step": 278500 }, { "epoch": 19.600955458760716, "grad_norm": 1.1914337873458862, "learning_rate": 2.2220615873933785e-05, "loss": 5.3074, "step": 279000 }, { "epoch": 19.636082619081073, "grad_norm": 1.2195658683776855, "learning_rate": 2.2076044527974556e-05, "loss": 5.3106, "step": 279500 }, { "epoch": 19.671209779401433, "grad_norm": 1.2408769130706787, "learning_rate": 2.1931473182015323e-05, "loss": 5.3068, "step": 280000 }, { "epoch": 19.671209779401433, "eval_accuracy": 0.6730934476560286, "eval_loss": 4.979102611541748, "eval_runtime": 137.147, "eval_samples_per_second": 880.121, "eval_steps_per_second": 5.505, "step": 280000 }, { "epoch": 19.706336939721794, "grad_norm": 1.188286304473877, "learning_rate": 2.1786901836056094e-05, "loss": 5.3089, "step": 280500 }, { "epoch": 19.74146410004215, "grad_norm": 1.2085764408111572, "learning_rate": 2.164233049009686e-05, "loss": 5.3089, "step": 281000 }, { "epoch": 19.776591260362512, "grad_norm": 1.2351000308990479, "learning_rate": 2.1497759144137632e-05, "loss": 5.3088, "step": 281500 }, { "epoch": 19.811718420682872, "grad_norm": 1.2089718580245972, "learning_rate": 2.1353187798178403e-05, "loss": 5.3068, "step": 282000 }, { "epoch": 19.846845581003233, "grad_norm": 1.2146202325820923, "learning_rate": 2.120861645221917e-05, "loss": 5.3065, "step": 282500 }, { "epoch": 19.88197274132359, "grad_norm": 1.2239059209823608, "learning_rate": 2.106404510625994e-05, "loss": 5.3051, "step": 283000 }, { "epoch": 19.91709990164395, "grad_norm": 1.2181216478347778, "learning_rate": 2.091947376030071e-05, "loss": 5.31, "step": 283500 }, { "epoch": 19.95222706196431, "grad_norm": 1.2403028011322021, "learning_rate": 2.077490241434148e-05, "loss": 5.3054, "step": 284000 }, { "epoch": 19.987354222284672, "grad_norm": 1.212841510772705, "learning_rate": 2.0630331068382247e-05, "loss": 5.3057, "step": 284500 }, { "epoch": 20.02248138260503, "grad_norm": 1.2400531768798828, "learning_rate": 2.0485759722423017e-05, "loss": 5.3087, "step": 285000 }, { "epoch": 20.05760854292539, "grad_norm": 1.2261024713516235, "learning_rate": 2.0341188376463785e-05, "loss": 5.3034, "step": 285500 }, { "epoch": 20.09273570324575, "grad_norm": 1.224755048751831, "learning_rate": 2.0196617030504556e-05, "loss": 5.3035, "step": 286000 }, { "epoch": 20.127862863566108, "grad_norm": 1.2355103492736816, "learning_rate": 2.0052045684545323e-05, "loss": 5.3027, "step": 286500 }, { "epoch": 20.16299002388647, "grad_norm": 1.2008854150772095, "learning_rate": 1.9907474338586094e-05, "loss": 5.3036, "step": 287000 }, { "epoch": 20.19811718420683, "grad_norm": 1.2236711978912354, "learning_rate": 1.976290299262686e-05, "loss": 5.3004, "step": 287500 }, { "epoch": 20.23324434452719, "grad_norm": 1.1880850791931152, "learning_rate": 1.9618331646667632e-05, "loss": 5.3029, "step": 288000 }, { "epoch": 20.268371504847547, "grad_norm": 1.2427552938461304, "learning_rate": 1.9473760300708403e-05, "loss": 5.3001, "step": 288500 }, { "epoch": 20.303498665167908, "grad_norm": 1.244379997253418, "learning_rate": 1.932918895474917e-05, "loss": 5.3047, "step": 289000 }, { "epoch": 20.33862582548827, "grad_norm": 1.2730302810668945, "learning_rate": 1.918461760878994e-05, "loss": 5.3019, "step": 289500 }, { "epoch": 20.37375298580863, "grad_norm": 1.256911039352417, "learning_rate": 1.9040046262830708e-05, "loss": 5.3031, "step": 290000 }, { "epoch": 20.37375298580863, "eval_accuracy": 0.673595711038036, "eval_loss": 4.974025249481201, "eval_runtime": 136.5088, "eval_samples_per_second": 884.236, "eval_steps_per_second": 5.531, "step": 290000 }, { "epoch": 20.408880146128986, "grad_norm": 1.2491388320922852, "learning_rate": 1.889547491687148e-05, "loss": 5.3009, "step": 290500 }, { "epoch": 20.444007306449347, "grad_norm": 1.1806505918502808, "learning_rate": 1.8750903570912246e-05, "loss": 5.304, "step": 291000 }, { "epoch": 20.479134466769708, "grad_norm": 1.2364420890808105, "learning_rate": 1.8606332224953017e-05, "loss": 5.2997, "step": 291500 }, { "epoch": 20.514261627090065, "grad_norm": 1.2333416938781738, "learning_rate": 1.8461760878993785e-05, "loss": 5.3018, "step": 292000 }, { "epoch": 20.549388787410425, "grad_norm": 1.2524642944335938, "learning_rate": 1.8317189533034555e-05, "loss": 5.2984, "step": 292500 }, { "epoch": 20.584515947730786, "grad_norm": 1.260603904724121, "learning_rate": 1.8172618187075323e-05, "loss": 5.302, "step": 293000 }, { "epoch": 20.619643108051147, "grad_norm": 1.19582998752594, "learning_rate": 1.8028046841116093e-05, "loss": 5.302, "step": 293500 }, { "epoch": 20.654770268371504, "grad_norm": 1.2455157041549683, "learning_rate": 1.788347549515686e-05, "loss": 5.2985, "step": 294000 }, { "epoch": 20.689897428691864, "grad_norm": 1.2068105936050415, "learning_rate": 1.7738904149197628e-05, "loss": 5.3018, "step": 294500 }, { "epoch": 20.725024589012225, "grad_norm": 1.233468770980835, "learning_rate": 1.75943328032384e-05, "loss": 5.2999, "step": 295000 }, { "epoch": 20.760151749332586, "grad_norm": 1.2510418891906738, "learning_rate": 1.7449761457279166e-05, "loss": 5.3028, "step": 295500 }, { "epoch": 20.795278909652943, "grad_norm": 1.2423063516616821, "learning_rate": 1.7305190111319937e-05, "loss": 5.2983, "step": 296000 }, { "epoch": 20.830406069973304, "grad_norm": 1.2420071363449097, "learning_rate": 1.7160618765360705e-05, "loss": 5.3004, "step": 296500 }, { "epoch": 20.865533230293664, "grad_norm": 1.2328987121582031, "learning_rate": 1.7016047419401475e-05, "loss": 5.296, "step": 297000 }, { "epoch": 20.90066039061402, "grad_norm": 1.2324236631393433, "learning_rate": 1.6871476073442243e-05, "loss": 5.3016, "step": 297500 }, { "epoch": 20.935787550934382, "grad_norm": 1.2644219398498535, "learning_rate": 1.6726904727483014e-05, "loss": 5.3012, "step": 298000 }, { "epoch": 20.970914711254743, "grad_norm": 1.2247650623321533, "learning_rate": 1.658233338152378e-05, "loss": 5.3003, "step": 298500 }, { "epoch": 21.006041871575103, "grad_norm": 1.2335166931152344, "learning_rate": 1.6437762035564552e-05, "loss": 5.2991, "step": 299000 }, { "epoch": 21.04116903189546, "grad_norm": 1.2490075826644897, "learning_rate": 1.629319068960532e-05, "loss": 5.3007, "step": 299500 }, { "epoch": 21.07629619221582, "grad_norm": 1.223919153213501, "learning_rate": 1.614861934364609e-05, "loss": 5.2947, "step": 300000 }, { "epoch": 21.07629619221582, "eval_accuracy": 0.6741870096208519, "eval_loss": 4.970097064971924, "eval_runtime": 138.1759, "eval_samples_per_second": 873.568, "eval_steps_per_second": 5.464, "step": 300000 }, { "epoch": 21.111423352536182, "grad_norm": 1.228028655052185, "learning_rate": 1.6004047997686857e-05, "loss": 5.2953, "step": 300500 }, { "epoch": 21.14655051285654, "grad_norm": 1.247745394706726, "learning_rate": 1.5859476651727628e-05, "loss": 5.2983, "step": 301000 }, { "epoch": 21.1816776731769, "grad_norm": 1.2382663488388062, "learning_rate": 1.57149053057684e-05, "loss": 5.2973, "step": 301500 }, { "epoch": 21.21680483349726, "grad_norm": 1.1911544799804688, "learning_rate": 1.5570333959809166e-05, "loss": 5.3, "step": 302000 }, { "epoch": 21.25193199381762, "grad_norm": 1.1953802108764648, "learning_rate": 1.5425762613849937e-05, "loss": 5.2933, "step": 302500 }, { "epoch": 21.287059154137978, "grad_norm": 1.3020567893981934, "learning_rate": 1.5281191267890704e-05, "loss": 5.2966, "step": 303000 }, { "epoch": 21.32218631445834, "grad_norm": 1.21441650390625, "learning_rate": 1.5136619921931475e-05, "loss": 5.2945, "step": 303500 }, { "epoch": 21.3573134747787, "grad_norm": 1.2315086126327515, "learning_rate": 1.4992048575972243e-05, "loss": 5.2922, "step": 304000 }, { "epoch": 21.39244063509906, "grad_norm": 1.2811838388442993, "learning_rate": 1.4847477230013013e-05, "loss": 5.2968, "step": 304500 }, { "epoch": 21.427567795419417, "grad_norm": 1.1895486116409302, "learning_rate": 1.470290588405378e-05, "loss": 5.2947, "step": 305000 }, { "epoch": 21.462694955739778, "grad_norm": 1.2534794807434082, "learning_rate": 1.4558334538094551e-05, "loss": 5.2944, "step": 305500 }, { "epoch": 21.49782211606014, "grad_norm": 1.2384119033813477, "learning_rate": 1.4413763192135319e-05, "loss": 5.2942, "step": 306000 }, { "epoch": 21.532949276380496, "grad_norm": 1.248732328414917, "learning_rate": 1.4269191846176088e-05, "loss": 5.2929, "step": 306500 }, { "epoch": 21.568076436700856, "grad_norm": 1.2607676982879639, "learning_rate": 1.4124620500216857e-05, "loss": 5.2923, "step": 307000 }, { "epoch": 21.603203597021217, "grad_norm": 1.2328699827194214, "learning_rate": 1.3980049154257626e-05, "loss": 5.2944, "step": 307500 }, { "epoch": 21.638330757341578, "grad_norm": 1.3178867101669312, "learning_rate": 1.3835477808298397e-05, "loss": 5.2995, "step": 308000 }, { "epoch": 21.673457917661935, "grad_norm": 1.2518072128295898, "learning_rate": 1.3690906462339164e-05, "loss": 5.2962, "step": 308500 }, { "epoch": 21.708585077982296, "grad_norm": 1.2616263628005981, "learning_rate": 1.3546335116379935e-05, "loss": 5.2931, "step": 309000 }, { "epoch": 21.743712238302656, "grad_norm": 1.2531579732894897, "learning_rate": 1.3401763770420702e-05, "loss": 5.2923, "step": 309500 }, { "epoch": 21.778839398623017, "grad_norm": 1.2060049772262573, "learning_rate": 1.3257192424461473e-05, "loss": 5.2931, "step": 310000 }, { "epoch": 21.778839398623017, "eval_accuracy": 0.6751666815268325, "eval_loss": 4.963305473327637, "eval_runtime": 139.1336, "eval_samples_per_second": 867.555, "eval_steps_per_second": 5.426, "step": 310000 }, { "epoch": 21.813966558943374, "grad_norm": 1.2173222303390503, "learning_rate": 1.311262107850224e-05, "loss": 5.2891, "step": 310500 }, { "epoch": 21.849093719263735, "grad_norm": 1.2453060150146484, "learning_rate": 1.2968049732543011e-05, "loss": 5.2891, "step": 311000 }, { "epoch": 21.884220879584095, "grad_norm": 1.2876172065734863, "learning_rate": 1.2823478386583779e-05, "loss": 5.2943, "step": 311500 }, { "epoch": 21.919348039904452, "grad_norm": 1.2135957479476929, "learning_rate": 1.267890704062455e-05, "loss": 5.2907, "step": 312000 }, { "epoch": 21.954475200224813, "grad_norm": 1.2596029043197632, "learning_rate": 1.2534335694665317e-05, "loss": 5.2918, "step": 312500 }, { "epoch": 21.989602360545174, "grad_norm": 1.2608625888824463, "learning_rate": 1.2389764348706088e-05, "loss": 5.2936, "step": 313000 }, { "epoch": 22.024729520865534, "grad_norm": 1.2269704341888428, "learning_rate": 1.2245193002746857e-05, "loss": 5.2908, "step": 313500 }, { "epoch": 22.05985668118589, "grad_norm": 1.2162054777145386, "learning_rate": 1.2100621656787626e-05, "loss": 5.2887, "step": 314000 }, { "epoch": 22.094983841506252, "grad_norm": 1.2698053121566772, "learning_rate": 1.1956050310828395e-05, "loss": 5.2924, "step": 314500 }, { "epoch": 22.130111001826613, "grad_norm": 1.2472206354141235, "learning_rate": 1.1811478964869162e-05, "loss": 5.2906, "step": 315000 }, { "epoch": 22.165238162146974, "grad_norm": 1.2648873329162598, "learning_rate": 1.1666907618909931e-05, "loss": 5.2864, "step": 315500 }, { "epoch": 22.20036532246733, "grad_norm": 1.2268308401107788, "learning_rate": 1.15223362729507e-05, "loss": 5.2865, "step": 316000 }, { "epoch": 22.23549248278769, "grad_norm": 1.209009051322937, "learning_rate": 1.137776492699147e-05, "loss": 5.2903, "step": 316500 }, { "epoch": 22.270619643108052, "grad_norm": 1.2333182096481323, "learning_rate": 1.123319358103224e-05, "loss": 5.2905, "step": 317000 }, { "epoch": 22.30574680342841, "grad_norm": 1.2603861093521118, "learning_rate": 1.108862223507301e-05, "loss": 5.2891, "step": 317500 }, { "epoch": 22.34087396374877, "grad_norm": 1.2085554599761963, "learning_rate": 1.0944050889113779e-05, "loss": 5.288, "step": 318000 }, { "epoch": 22.37600112406913, "grad_norm": 1.25042724609375, "learning_rate": 1.0799479543154548e-05, "loss": 5.2906, "step": 318500 }, { "epoch": 22.41112828438949, "grad_norm": 1.2044715881347656, "learning_rate": 1.0654908197195317e-05, "loss": 5.2864, "step": 319000 }, { "epoch": 22.44625544470985, "grad_norm": 1.2200154066085815, "learning_rate": 1.0510336851236086e-05, "loss": 5.2873, "step": 319500 }, { "epoch": 22.48138260503021, "grad_norm": 1.2526922225952148, "learning_rate": 1.0365765505276855e-05, "loss": 5.2875, "step": 320000 }, { "epoch": 22.48138260503021, "eval_accuracy": 0.6755681665268015, "eval_loss": 4.960219860076904, "eval_runtime": 138.2027, "eval_samples_per_second": 873.398, "eval_steps_per_second": 5.463, "step": 320000 }, { "epoch": 22.51650976535057, "grad_norm": 1.2500373125076294, "learning_rate": 1.0221194159317624e-05, "loss": 5.2864, "step": 320500 }, { "epoch": 22.55163692567093, "grad_norm": 1.2759400606155396, "learning_rate": 1.0076622813358393e-05, "loss": 5.2924, "step": 321000 }, { "epoch": 22.586764085991287, "grad_norm": 1.2182376384735107, "learning_rate": 9.932051467399162e-06, "loss": 5.2895, "step": 321500 }, { "epoch": 22.621891246311648, "grad_norm": 1.2039501667022705, "learning_rate": 9.787480121439931e-06, "loss": 5.2868, "step": 322000 }, { "epoch": 22.65701840663201, "grad_norm": 1.2825946807861328, "learning_rate": 9.6429087754807e-06, "loss": 5.2896, "step": 322500 }, { "epoch": 22.692145566952366, "grad_norm": 1.2612876892089844, "learning_rate": 9.49833742952147e-06, "loss": 5.2871, "step": 323000 }, { "epoch": 22.727272727272727, "grad_norm": 1.2150561809539795, "learning_rate": 9.353766083562238e-06, "loss": 5.2862, "step": 323500 }, { "epoch": 22.762399887593087, "grad_norm": 1.2840204238891602, "learning_rate": 9.209194737603008e-06, "loss": 5.2846, "step": 324000 }, { "epoch": 22.797527047913448, "grad_norm": 1.2217748165130615, "learning_rate": 9.064623391643777e-06, "loss": 5.289, "step": 324500 }, { "epoch": 22.832654208233805, "grad_norm": 1.2241199016571045, "learning_rate": 8.920052045684546e-06, "loss": 5.2877, "step": 325000 }, { "epoch": 22.867781368554166, "grad_norm": 1.315966010093689, "learning_rate": 8.775480699725315e-06, "loss": 5.2863, "step": 325500 }, { "epoch": 22.902908528874526, "grad_norm": 1.2527564764022827, "learning_rate": 8.630909353766084e-06, "loss": 5.2894, "step": 326000 }, { "epoch": 22.938035689194887, "grad_norm": 1.2263576984405518, "learning_rate": 8.486338007806853e-06, "loss": 5.286, "step": 326500 }, { "epoch": 22.973162849515244, "grad_norm": 1.2176321744918823, "learning_rate": 8.341766661847622e-06, "loss": 5.281, "step": 327000 }, { "epoch": 23.008290009835605, "grad_norm": 1.2201576232910156, "learning_rate": 8.197195315888391e-06, "loss": 5.2868, "step": 327500 }, { "epoch": 23.043417170155966, "grad_norm": 1.2183256149291992, "learning_rate": 8.05262396992916e-06, "loss": 5.2854, "step": 328000 }, { "epoch": 23.078544330476323, "grad_norm": 1.2602317333221436, "learning_rate": 7.90805262396993e-06, "loss": 5.2809, "step": 328500 }, { "epoch": 23.113671490796683, "grad_norm": 1.2807954549789429, "learning_rate": 7.763481278010698e-06, "loss": 5.2854, "step": 329000 }, { "epoch": 23.148798651117044, "grad_norm": 1.2522610425949097, "learning_rate": 7.618909932051468e-06, "loss": 5.2887, "step": 329500 }, { "epoch": 23.183925811437405, "grad_norm": 1.2843894958496094, "learning_rate": 7.474338586092237e-06, "loss": 5.2841, "step": 330000 }, { "epoch": 23.183925811437405, "eval_accuracy": 0.6757566799440172, "eval_loss": 4.958164691925049, "eval_runtime": 135.9164, "eval_samples_per_second": 888.09, "eval_steps_per_second": 5.555, "step": 330000 }, { "epoch": 23.219052971757762, "grad_norm": 1.2468485832214355, "learning_rate": 7.3297672401330065e-06, "loss": 5.2804, "step": 330500 }, { "epoch": 23.254180132078123, "grad_norm": 1.2579010725021362, "learning_rate": 7.1851958941737756e-06, "loss": 5.2836, "step": 331000 }, { "epoch": 23.289307292398483, "grad_norm": 1.2422467470169067, "learning_rate": 7.040624548214545e-06, "loss": 5.2841, "step": 331500 }, { "epoch": 23.324434452718844, "grad_norm": 1.2477812767028809, "learning_rate": 6.896053202255314e-06, "loss": 5.289, "step": 332000 }, { "epoch": 23.3595616130392, "grad_norm": 1.2169203758239746, "learning_rate": 6.751481856296083e-06, "loss": 5.2844, "step": 332500 }, { "epoch": 23.39468877335956, "grad_norm": 1.2749801874160767, "learning_rate": 6.606910510336851e-06, "loss": 5.2797, "step": 333000 }, { "epoch": 23.429815933679922, "grad_norm": 1.276930809020996, "learning_rate": 6.46233916437762e-06, "loss": 5.2826, "step": 333500 }, { "epoch": 23.46494309400028, "grad_norm": 1.2267518043518066, "learning_rate": 6.317767818418389e-06, "loss": 5.283, "step": 334000 }, { "epoch": 23.50007025432064, "grad_norm": 1.226639986038208, "learning_rate": 6.173196472459159e-06, "loss": 5.2849, "step": 334500 }, { "epoch": 23.535197414641, "grad_norm": 1.2478587627410889, "learning_rate": 6.028625126499928e-06, "loss": 5.2861, "step": 335000 }, { "epoch": 23.57032457496136, "grad_norm": 1.2578195333480835, "learning_rate": 5.884053780540697e-06, "loss": 5.2833, "step": 335500 }, { "epoch": 23.60545173528172, "grad_norm": 1.2274130582809448, "learning_rate": 5.739482434581466e-06, "loss": 5.28, "step": 336000 }, { "epoch": 23.64057889560208, "grad_norm": 1.2471580505371094, "learning_rate": 5.5949110886222355e-06, "loss": 5.2777, "step": 336500 }, { "epoch": 23.67570605592244, "grad_norm": 1.2308429479599, "learning_rate": 5.4503397426630046e-06, "loss": 5.2824, "step": 337000 }, { "epoch": 23.7108332162428, "grad_norm": 1.2559664249420166, "learning_rate": 5.305768396703774e-06, "loss": 5.2769, "step": 337500 }, { "epoch": 23.745960376563158, "grad_norm": 1.240325689315796, "learning_rate": 5.161197050744543e-06, "loss": 5.2797, "step": 338000 }, { "epoch": 23.78108753688352, "grad_norm": 1.214863896369934, "learning_rate": 5.016625704785312e-06, "loss": 5.2787, "step": 338500 }, { "epoch": 23.81621469720388, "grad_norm": 1.2319936752319336, "learning_rate": 4.872054358826081e-06, "loss": 5.2812, "step": 339000 }, { "epoch": 23.851341857524236, "grad_norm": 1.2701442241668701, "learning_rate": 4.72748301286685e-06, "loss": 5.2818, "step": 339500 }, { "epoch": 23.886469017844597, "grad_norm": 1.2425583600997925, "learning_rate": 4.582911666907619e-06, "loss": 5.2815, "step": 340000 }, { "epoch": 23.886469017844597, "eval_accuracy": 0.6762319351039592, "eval_loss": 4.954134941101074, "eval_runtime": 136.6188, "eval_samples_per_second": 883.524, "eval_steps_per_second": 5.526, "step": 340000 }, { "epoch": 23.921596178164958, "grad_norm": 1.2548600435256958, "learning_rate": 4.438340320948388e-06, "loss": 5.2785, "step": 340500 }, { "epoch": 23.956723338485318, "grad_norm": 1.2147016525268555, "learning_rate": 4.293768974989157e-06, "loss": 5.2813, "step": 341000 }, { "epoch": 23.991850498805675, "grad_norm": 1.2386934757232666, "learning_rate": 4.149197629029926e-06, "loss": 5.2818, "step": 341500 }, { "epoch": 24.026977659126036, "grad_norm": 1.2231061458587646, "learning_rate": 4.004626283070695e-06, "loss": 5.2822, "step": 342000 }, { "epoch": 24.062104819446397, "grad_norm": 1.2466256618499756, "learning_rate": 3.8600549371114645e-06, "loss": 5.2797, "step": 342500 }, { "epoch": 24.097231979766757, "grad_norm": 1.2280278205871582, "learning_rate": 3.7154835911522336e-06, "loss": 5.2789, "step": 343000 }, { "epoch": 24.132359140087114, "grad_norm": 1.2528438568115234, "learning_rate": 3.5709122451930026e-06, "loss": 5.281, "step": 343500 }, { "epoch": 24.167486300407475, "grad_norm": 1.2286638021469116, "learning_rate": 3.426340899233772e-06, "loss": 5.2833, "step": 344000 }, { "epoch": 24.202613460727836, "grad_norm": 1.2314958572387695, "learning_rate": 3.2817695532745412e-06, "loss": 5.2812, "step": 344500 }, { "epoch": 24.237740621048193, "grad_norm": 1.2333118915557861, "learning_rate": 3.1371982073153103e-06, "loss": 5.2836, "step": 345000 }, { "epoch": 24.272867781368554, "grad_norm": 1.2485569715499878, "learning_rate": 2.9926268613560794e-06, "loss": 5.2794, "step": 345500 }, { "epoch": 24.307994941688914, "grad_norm": 1.2749316692352295, "learning_rate": 2.8480555153968485e-06, "loss": 5.2781, "step": 346000 }, { "epoch": 24.343122102009275, "grad_norm": 1.2548277378082275, "learning_rate": 2.7034841694376176e-06, "loss": 5.2759, "step": 346500 }, { "epoch": 24.378249262329632, "grad_norm": 1.210546612739563, "learning_rate": 2.5589128234783866e-06, "loss": 5.2801, "step": 347000 }, { "epoch": 24.413376422649993, "grad_norm": 1.2708075046539307, "learning_rate": 2.414341477519156e-06, "loss": 5.2769, "step": 347500 }, { "epoch": 24.448503582970353, "grad_norm": 1.2050740718841553, "learning_rate": 2.269770131559925e-06, "loss": 5.2798, "step": 348000 }, { "epoch": 24.483630743290714, "grad_norm": 1.2496392726898193, "learning_rate": 2.125198785600694e-06, "loss": 5.2776, "step": 348500 }, { "epoch": 24.51875790361107, "grad_norm": 1.1686387062072754, "learning_rate": 1.980627439641463e-06, "loss": 5.2757, "step": 349000 }, { "epoch": 24.553885063931432, "grad_norm": 1.217787265777588, "learning_rate": 1.8360560936822323e-06, "loss": 5.2762, "step": 349500 }, { "epoch": 24.589012224251793, "grad_norm": 1.2311830520629883, "learning_rate": 1.6914847477230013e-06, "loss": 5.2811, "step": 350000 }, { "epoch": 24.589012224251793, "eval_accuracy": 0.6765809424306541, "eval_loss": 4.951249122619629, "eval_runtime": 135.3493, "eval_samples_per_second": 891.811, "eval_steps_per_second": 5.578, "step": 350000 }, { "epoch": 24.62413938457215, "grad_norm": 1.2537038326263428, "learning_rate": 1.5469134017637704e-06, "loss": 5.2773, "step": 350500 }, { "epoch": 24.65926654489251, "grad_norm": 1.2281949520111084, "learning_rate": 1.4023420558045395e-06, "loss": 5.2753, "step": 351000 }, { "epoch": 24.69439370521287, "grad_norm": 1.255089521408081, "learning_rate": 1.2577707098453088e-06, "loss": 5.2779, "step": 351500 }, { "epoch": 24.72952086553323, "grad_norm": 1.29042387008667, "learning_rate": 1.1131993638860779e-06, "loss": 5.2788, "step": 352000 }, { "epoch": 24.76464802585359, "grad_norm": 1.2561049461364746, "learning_rate": 9.68628017926847e-07, "loss": 5.2783, "step": 352500 }, { "epoch": 24.79977518617395, "grad_norm": 1.2399394512176514, "learning_rate": 8.240566719676161e-07, "loss": 5.2784, "step": 353000 }, { "epoch": 24.83490234649431, "grad_norm": 1.1991074085235596, "learning_rate": 6.794853260083851e-07, "loss": 5.2762, "step": 353500 }, { "epoch": 24.870029506814667, "grad_norm": 1.2151962518692017, "learning_rate": 5.349139800491543e-07, "loss": 5.2782, "step": 354000 }, { "epoch": 24.905156667135028, "grad_norm": 1.2174469232559204, "learning_rate": 3.903426340899234e-07, "loss": 5.278, "step": 354500 }, { "epoch": 24.94028382745539, "grad_norm": 1.2640910148620605, "learning_rate": 2.457712881306925e-07, "loss": 5.2723, "step": 355000 }, { "epoch": 24.97541098777575, "grad_norm": 1.1888768672943115, "learning_rate": 1.0119994217146162e-07, "loss": 5.278, "step": 355500 }, { "epoch": 25.0, "step": 355850, "total_flos": 3.046251597339341e+18, "train_loss": 5.746741898249789, "train_runtime": 91431.3002, "train_samples_per_second": 622.692, "train_steps_per_second": 3.892 } ], "logging_steps": 500, "max_steps": 355850, "num_input_tokens_seen": 0, "num_train_epochs": 25, "save_steps": 10000, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 3.046251597339341e+18, "train_batch_size": 160, "trial_name": null, "trial_params": null }