{ "best_metric": 3.870356321334839, "best_model_checkpoint": "bert_tiny_lda_20_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": 2.401348352432251, "learning_rate": 5e-06, "loss": 12.5679, "step": 500 }, { "epoch": 0.0702543206407194, "grad_norm": 0.991630494594574, "learning_rate": 1e-05, "loss": 10.4787, "step": 1000 }, { "epoch": 0.1053814809610791, "grad_norm": 0.6601514220237732, "learning_rate": 1.5e-05, "loss": 9.2323, "step": 1500 }, { "epoch": 0.1405086412814388, "grad_norm": 0.6543694734573364, "learning_rate": 2e-05, "loss": 9.0277, "step": 2000 }, { "epoch": 0.1756358016017985, "grad_norm": 0.6228306293487549, "learning_rate": 2.5e-05, "loss": 8.937, "step": 2500 }, { "epoch": 0.2107629619221582, "grad_norm": 0.7250866889953613, "learning_rate": 3e-05, "loss": 8.8752, "step": 3000 }, { "epoch": 0.2458901222425179, "grad_norm": 0.6583071947097778, "learning_rate": 3.5e-05, "loss": 8.8359, "step": 3500 }, { "epoch": 0.2810172825628776, "grad_norm": 0.6191977858543396, "learning_rate": 4e-05, "loss": 8.8043, "step": 4000 }, { "epoch": 0.31614444288323734, "grad_norm": 0.7777481079101562, "learning_rate": 4.5e-05, "loss": 8.7784, "step": 4500 }, { "epoch": 0.351271603203597, "grad_norm": 0.6216750144958496, "learning_rate": 5e-05, "loss": 8.7572, "step": 5000 }, { "epoch": 0.38639876352395675, "grad_norm": 0.709453284740448, "learning_rate": 5.500000000000001e-05, "loss": 8.7338, "step": 5500 }, { "epoch": 0.4215259238443164, "grad_norm": 0.6589279174804688, "learning_rate": 6e-05, "loss": 8.7227, "step": 6000 }, { "epoch": 0.45665308416467615, "grad_norm": 0.6572352647781372, "learning_rate": 6.500000000000001e-05, "loss": 8.7075, "step": 6500 }, { "epoch": 0.4917802444850358, "grad_norm": 0.6236314177513123, "learning_rate": 7e-05, "loss": 8.6975, "step": 7000 }, { "epoch": 0.5269074048053956, "grad_norm": 0.6040647625923157, "learning_rate": 7.500000000000001e-05, "loss": 8.6844, "step": 7500 }, { "epoch": 0.5620345651257552, "grad_norm": 0.6199480295181274, "learning_rate": 8e-05, "loss": 8.6726, "step": 8000 }, { "epoch": 0.5971617254461149, "grad_norm": 0.8181083798408508, "learning_rate": 8.5e-05, "loss": 8.6617, "step": 8500 }, { "epoch": 0.6322888857664747, "grad_norm": 0.848405659198761, "learning_rate": 9e-05, "loss": 8.6408, "step": 9000 }, { "epoch": 0.6674160460868344, "grad_norm": 0.7515135407447815, "learning_rate": 9.5e-05, "loss": 8.6193, "step": 9500 }, { "epoch": 0.702543206407194, "grad_norm": 0.646868109703064, "learning_rate": 0.0001, "loss": 8.6049, "step": 10000 }, { "epoch": 0.702543206407194, "eval_accuracy": 0.16525696846602408, "eval_loss": 8.437688827514648, "eval_runtime": 188.4328, "eval_samples_per_second": 640.579, "eval_steps_per_second": 4.007, "step": 10000 }, { "epoch": 0.7376703667275537, "grad_norm": 0.6523381471633911, "learning_rate": 9.985542865404077e-05, "loss": 8.5787, "step": 10500 }, { "epoch": 0.7727975270479135, "grad_norm": 0.8462009429931641, "learning_rate": 9.971085730808155e-05, "loss": 8.5023, "step": 11000 }, { "epoch": 0.8079246873682732, "grad_norm": 1.039638876914978, "learning_rate": 9.956628596212231e-05, "loss": 8.4001, "step": 11500 }, { "epoch": 0.8430518476886328, "grad_norm": 1.497228741645813, "learning_rate": 9.942171461616309e-05, "loss": 8.2407, "step": 12000 }, { "epoch": 0.8781790080089925, "grad_norm": 2.7516579627990723, "learning_rate": 9.927714327020385e-05, "loss": 7.7421, "step": 12500 }, { "epoch": 0.9133061683293523, "grad_norm": 2.322057008743286, "learning_rate": 9.913257192424462e-05, "loss": 7.3284, "step": 13000 }, { "epoch": 0.948433328649712, "grad_norm": 2.196650743484497, "learning_rate": 9.89880005782854e-05, "loss": 7.0796, "step": 13500 }, { "epoch": 0.9835604889700716, "grad_norm": 2.592818021774292, "learning_rate": 9.884342923232616e-05, "loss": 6.8093, "step": 14000 }, { "epoch": 1.0186876492904313, "grad_norm": 1.907196044921875, "learning_rate": 9.869885788636692e-05, "loss": 6.5757, "step": 14500 }, { "epoch": 1.053814809610791, "grad_norm": 1.8518394231796265, "learning_rate": 9.855428654040769e-05, "loss": 6.4011, "step": 15000 }, { "epoch": 1.0889419699311507, "grad_norm": 1.7491087913513184, "learning_rate": 9.840971519444846e-05, "loss": 6.2571, "step": 15500 }, { "epoch": 1.1240691302515105, "grad_norm": 1.8714861869812012, "learning_rate": 9.826514384848924e-05, "loss": 6.1394, "step": 16000 }, { "epoch": 1.1591962905718702, "grad_norm": 1.67587411403656, "learning_rate": 9.812057250253e-05, "loss": 6.0508, "step": 16500 }, { "epoch": 1.1943234508922298, "grad_norm": 1.788169264793396, "learning_rate": 9.797600115657077e-05, "loss": 5.9706, "step": 17000 }, { "epoch": 1.2294506112125896, "grad_norm": 1.6467270851135254, "learning_rate": 9.783142981061155e-05, "loss": 5.8968, "step": 17500 }, { "epoch": 1.2645777715329491, "grad_norm": 1.5956337451934814, "learning_rate": 9.768685846465231e-05, "loss": 5.8423, "step": 18000 }, { "epoch": 1.299704931853309, "grad_norm": 1.600663423538208, "learning_rate": 9.754228711869307e-05, "loss": 5.7809, "step": 18500 }, { "epoch": 1.3348320921736687, "grad_norm": 1.5106139183044434, "learning_rate": 9.739771577273384e-05, "loss": 5.7406, "step": 19000 }, { "epoch": 1.3699592524940285, "grad_norm": 1.454498529434204, "learning_rate": 9.725314442677463e-05, "loss": 5.7009, "step": 19500 }, { "epoch": 1.405086412814388, "grad_norm": 1.3555099964141846, "learning_rate": 9.71085730808154e-05, "loss": 5.6593, "step": 20000 }, { "epoch": 1.405086412814388, "eval_accuracy": 0.5031426620018882, "eval_loss": 5.224310398101807, "eval_runtime": 207.539, "eval_samples_per_second": 581.606, "eval_steps_per_second": 3.638, "step": 20000 }, { "epoch": 1.4402135731347478, "grad_norm": 1.3797138929367065, "learning_rate": 9.696400173485616e-05, "loss": 5.6226, "step": 20500 }, { "epoch": 1.4753407334551074, "grad_norm": 1.51322340965271, "learning_rate": 9.681943038889692e-05, "loss": 5.5895, "step": 21000 }, { "epoch": 1.5104678937754672, "grad_norm": 1.534635066986084, "learning_rate": 9.66748590429377e-05, "loss": 5.5538, "step": 21500 }, { "epoch": 1.545595054095827, "grad_norm": 1.3956483602523804, "learning_rate": 9.653028769697846e-05, "loss": 5.5185, "step": 22000 }, { "epoch": 1.5807222144161865, "grad_norm": 1.2456926107406616, "learning_rate": 9.638571635101923e-05, "loss": 5.4931, "step": 22500 }, { "epoch": 1.6158493747365463, "grad_norm": 1.247243046760559, "learning_rate": 9.624114500506e-05, "loss": 5.4628, "step": 23000 }, { "epoch": 1.650976535056906, "grad_norm": 1.3580514192581177, "learning_rate": 9.609657365910077e-05, "loss": 5.435, "step": 23500 }, { "epoch": 1.6861036953772657, "grad_norm": 1.2038990259170532, "learning_rate": 9.595200231314155e-05, "loss": 5.4073, "step": 24000 }, { "epoch": 1.7212308556976255, "grad_norm": 1.1903976202011108, "learning_rate": 9.580743096718231e-05, "loss": 5.3826, "step": 24500 }, { "epoch": 1.7563580160179852, "grad_norm": 1.1388156414031982, "learning_rate": 9.566285962122307e-05, "loss": 5.3595, "step": 25000 }, { "epoch": 1.7914851763383448, "grad_norm": 1.1945675611495972, "learning_rate": 9.551828827526385e-05, "loss": 5.3363, "step": 25500 }, { "epoch": 1.8266123366587044, "grad_norm": 1.135754942893982, "learning_rate": 9.537371692930462e-05, "loss": 5.311, "step": 26000 }, { "epoch": 1.8617394969790642, "grad_norm": 1.1858739852905273, "learning_rate": 9.522914558334538e-05, "loss": 5.287, "step": 26500 }, { "epoch": 1.896866657299424, "grad_norm": 1.1977298259735107, "learning_rate": 9.508457423738616e-05, "loss": 5.2691, "step": 27000 }, { "epoch": 1.9319938176197837, "grad_norm": 1.166898250579834, "learning_rate": 9.494000289142692e-05, "loss": 5.2483, "step": 27500 }, { "epoch": 1.9671209779401433, "grad_norm": 1.1156128644943237, "learning_rate": 9.47954315454677e-05, "loss": 5.2268, "step": 28000 }, { "epoch": 2.002248138260503, "grad_norm": 1.1086357831954956, "learning_rate": 9.465086019950846e-05, "loss": 5.2143, "step": 28500 }, { "epoch": 2.0373752985808626, "grad_norm": 1.1779756546020508, "learning_rate": 9.450628885354923e-05, "loss": 5.1871, "step": 29000 }, { "epoch": 2.0725024589012224, "grad_norm": 1.152138113975525, "learning_rate": 9.436171750758999e-05, "loss": 5.1703, "step": 29500 }, { "epoch": 2.107629619221582, "grad_norm": 1.1608511209487915, "learning_rate": 9.421714616163077e-05, "loss": 5.1606, "step": 30000 }, { "epoch": 2.107629619221582, "eval_accuracy": 0.5589214761805081, "eval_loss": 4.777843475341797, "eval_runtime": 188.3176, "eval_samples_per_second": 640.97, "eval_steps_per_second": 4.009, "step": 30000 }, { "epoch": 2.142756779541942, "grad_norm": 1.1260640621185303, "learning_rate": 9.407257481567155e-05, "loss": 5.1429, "step": 30500 }, { "epoch": 2.1778839398623013, "grad_norm": 1.090111255645752, "learning_rate": 9.392800346971231e-05, "loss": 5.1246, "step": 31000 }, { "epoch": 2.213011100182661, "grad_norm": 1.1177226305007935, "learning_rate": 9.378343212375307e-05, "loss": 5.1149, "step": 31500 }, { "epoch": 2.248138260503021, "grad_norm": 1.1309847831726074, "learning_rate": 9.363886077779385e-05, "loss": 5.1005, "step": 32000 }, { "epoch": 2.2832654208233807, "grad_norm": 1.1482443809509277, "learning_rate": 9.349428943183462e-05, "loss": 5.0876, "step": 32500 }, { "epoch": 2.3183925811437405, "grad_norm": 1.05221426486969, "learning_rate": 9.334971808587538e-05, "loss": 5.0705, "step": 33000 }, { "epoch": 2.3535197414641003, "grad_norm": 1.1037867069244385, "learning_rate": 9.320514673991614e-05, "loss": 5.0596, "step": 33500 }, { "epoch": 2.3886469017844596, "grad_norm": 1.011998176574707, "learning_rate": 9.306057539395692e-05, "loss": 5.0488, "step": 34000 }, { "epoch": 2.4237740621048194, "grad_norm": 1.0158193111419678, "learning_rate": 9.29160040479977e-05, "loss": 5.0341, "step": 34500 }, { "epoch": 2.458901222425179, "grad_norm": 1.1270122528076172, "learning_rate": 9.277143270203846e-05, "loss": 5.0239, "step": 35000 }, { "epoch": 2.494028382745539, "grad_norm": 1.0856341123580933, "learning_rate": 9.262686135607923e-05, "loss": 5.013, "step": 35500 }, { "epoch": 2.5291555430658983, "grad_norm": 1.0190181732177734, "learning_rate": 9.248229001012e-05, "loss": 5.0015, "step": 36000 }, { "epoch": 2.564282703386258, "grad_norm": 1.0561621189117432, "learning_rate": 9.233771866416077e-05, "loss": 4.9899, "step": 36500 }, { "epoch": 2.599409863706618, "grad_norm": 1.1023633480072021, "learning_rate": 9.219314731820153e-05, "loss": 4.9781, "step": 37000 }, { "epoch": 2.6345370240269776, "grad_norm": 1.042366862297058, "learning_rate": 9.20485759722423e-05, "loss": 4.9684, "step": 37500 }, { "epoch": 2.6696641843473374, "grad_norm": 1.111264705657959, "learning_rate": 9.190400462628307e-05, "loss": 4.9564, "step": 38000 }, { "epoch": 2.704791344667697, "grad_norm": 1.0811132192611694, "learning_rate": 9.175943328032385e-05, "loss": 4.9462, "step": 38500 }, { "epoch": 2.739918504988057, "grad_norm": 1.1237092018127441, "learning_rate": 9.161486193436461e-05, "loss": 4.9347, "step": 39000 }, { "epoch": 2.7750456653084163, "grad_norm": 1.0572726726531982, "learning_rate": 9.147029058840538e-05, "loss": 4.9287, "step": 39500 }, { "epoch": 2.810172825628776, "grad_norm": 1.0520048141479492, "learning_rate": 9.132571924244616e-05, "loss": 4.9169, "step": 40000 }, { "epoch": 2.810172825628776, "eval_accuracy": 0.5885351968519732, "eval_loss": 4.553859710693359, "eval_runtime": 193.9608, "eval_samples_per_second": 622.322, "eval_steps_per_second": 3.893, "step": 40000 }, { "epoch": 2.845299985949136, "grad_norm": 1.1296918392181396, "learning_rate": 9.118114789648692e-05, "loss": 4.9052, "step": 40500 }, { "epoch": 2.8804271462694957, "grad_norm": 1.0899595022201538, "learning_rate": 9.103657655052768e-05, "loss": 4.8964, "step": 41000 }, { "epoch": 2.915554306589855, "grad_norm": 1.0832406282424927, "learning_rate": 9.089200520456845e-05, "loss": 4.8891, "step": 41500 }, { "epoch": 2.950681466910215, "grad_norm": 1.029417872428894, "learning_rate": 9.074743385860923e-05, "loss": 4.8836, "step": 42000 }, { "epoch": 2.9858086272305746, "grad_norm": 1.104624629020691, "learning_rate": 9.060286251265e-05, "loss": 4.8744, "step": 42500 }, { "epoch": 3.0209357875509344, "grad_norm": 1.0334268808364868, "learning_rate": 9.045829116669077e-05, "loss": 4.8585, "step": 43000 }, { "epoch": 3.056062947871294, "grad_norm": 1.063277244567871, "learning_rate": 9.031371982073153e-05, "loss": 4.8544, "step": 43500 }, { "epoch": 3.091190108191654, "grad_norm": 1.0381839275360107, "learning_rate": 9.016914847477231e-05, "loss": 4.8499, "step": 44000 }, { "epoch": 3.1263172685120133, "grad_norm": 1.0323851108551025, "learning_rate": 9.002457712881307e-05, "loss": 4.8409, "step": 44500 }, { "epoch": 3.161444428832373, "grad_norm": 1.1177279949188232, "learning_rate": 8.988000578285384e-05, "loss": 4.8332, "step": 45000 }, { "epoch": 3.196571589152733, "grad_norm": 1.0623455047607422, "learning_rate": 8.973543443689461e-05, "loss": 4.8241, "step": 45500 }, { "epoch": 3.2316987494730927, "grad_norm": 1.064278244972229, "learning_rate": 8.959086309093538e-05, "loss": 4.8188, "step": 46000 }, { "epoch": 3.2668259097934524, "grad_norm": 1.0583056211471558, "learning_rate": 8.944629174497616e-05, "loss": 4.8121, "step": 46500 }, { "epoch": 3.301953070113812, "grad_norm": 1.0297889709472656, "learning_rate": 8.930172039901692e-05, "loss": 4.8025, "step": 47000 }, { "epoch": 3.3370802304341716, "grad_norm": 1.0724543333053589, "learning_rate": 8.915714905305768e-05, "loss": 4.7964, "step": 47500 }, { "epoch": 3.3722073907545314, "grad_norm": 1.085640788078308, "learning_rate": 8.901257770709845e-05, "loss": 4.7917, "step": 48000 }, { "epoch": 3.407334551074891, "grad_norm": 1.0612136125564575, "learning_rate": 8.886800636113923e-05, "loss": 4.7803, "step": 48500 }, { "epoch": 3.442461711395251, "grad_norm": 1.0585534572601318, "learning_rate": 8.872343501518e-05, "loss": 4.7736, "step": 49000 }, { "epoch": 3.4775888717156107, "grad_norm": 1.1188209056854248, "learning_rate": 8.857886366922077e-05, "loss": 4.7655, "step": 49500 }, { "epoch": 3.51271603203597, "grad_norm": 1.0340315103530884, "learning_rate": 8.843429232326153e-05, "loss": 4.7607, "step": 50000 }, { "epoch": 3.51271603203597, "eval_accuracy": 0.6087781463533943, "eval_loss": 4.408504962921143, "eval_runtime": 242.209, "eval_samples_per_second": 498.355, "eval_steps_per_second": 3.117, "step": 50000 }, { "epoch": 3.54784319235633, "grad_norm": 1.106341004371643, "learning_rate": 8.828972097730231e-05, "loss": 4.7532, "step": 50500 }, { "epoch": 3.5829703526766896, "grad_norm": 1.0758745670318604, "learning_rate": 8.814514963134307e-05, "loss": 4.7378, "step": 51000 }, { "epoch": 3.6180975129970494, "grad_norm": 1.0182976722717285, "learning_rate": 8.800057828538384e-05, "loss": 4.7372, "step": 51500 }, { "epoch": 3.6532246733174087, "grad_norm": 1.0748611688613892, "learning_rate": 8.78560069394246e-05, "loss": 4.729, "step": 52000 }, { "epoch": 3.6883518336377685, "grad_norm": 1.011732578277588, "learning_rate": 8.771143559346538e-05, "loss": 4.723, "step": 52500 }, { "epoch": 3.7234789939581283, "grad_norm": 1.0253151655197144, "learning_rate": 8.756686424750616e-05, "loss": 4.7201, "step": 53000 }, { "epoch": 3.758606154278488, "grad_norm": 1.0731598138809204, "learning_rate": 8.742229290154692e-05, "loss": 4.714, "step": 53500 }, { "epoch": 3.793733314598848, "grad_norm": 1.0363712310791016, "learning_rate": 8.727772155558768e-05, "loss": 4.704, "step": 54000 }, { "epoch": 3.8288604749192077, "grad_norm": 1.0688296556472778, "learning_rate": 8.713315020962846e-05, "loss": 4.703, "step": 54500 }, { "epoch": 3.8639876352395675, "grad_norm": 1.059349536895752, "learning_rate": 8.698857886366923e-05, "loss": 4.6937, "step": 55000 }, { "epoch": 3.899114795559927, "grad_norm": 1.0599170923233032, "learning_rate": 8.684400751770999e-05, "loss": 4.6886, "step": 55500 }, { "epoch": 3.9342419558802866, "grad_norm": 1.0922009944915771, "learning_rate": 8.669943617175075e-05, "loss": 4.6845, "step": 56000 }, { "epoch": 3.9693691162006464, "grad_norm": 1.0112320184707642, "learning_rate": 8.655486482579153e-05, "loss": 4.681, "step": 56500 }, { "epoch": 4.004496276521006, "grad_norm": 1.111264705657959, "learning_rate": 8.641029347983231e-05, "loss": 4.6714, "step": 57000 }, { "epoch": 4.0396234368413655, "grad_norm": 1.1196843385696411, "learning_rate": 8.626572213387307e-05, "loss": 4.6633, "step": 57500 }, { "epoch": 4.074750597161725, "grad_norm": 1.0570870637893677, "learning_rate": 8.612115078791384e-05, "loss": 4.6594, "step": 58000 }, { "epoch": 4.109877757482085, "grad_norm": 1.1056139469146729, "learning_rate": 8.597657944195461e-05, "loss": 4.6538, "step": 58500 }, { "epoch": 4.145004917802445, "grad_norm": 1.0164746046066284, "learning_rate": 8.583200809599538e-05, "loss": 4.6481, "step": 59000 }, { "epoch": 4.180132078122805, "grad_norm": 1.035692811012268, "learning_rate": 8.568743675003614e-05, "loss": 4.6458, "step": 59500 }, { "epoch": 4.215259238443164, "grad_norm": 1.0868757963180542, "learning_rate": 8.55428654040769e-05, "loss": 4.6405, "step": 60000 }, { "epoch": 4.215259238443164, "eval_accuracy": 0.6215926740910587, "eval_loss": 4.302714824676514, "eval_runtime": 170.0085, "eval_samples_per_second": 710.0, "eval_steps_per_second": 4.441, "step": 60000 }, { "epoch": 4.250386398763524, "grad_norm": 1.1007674932479858, "learning_rate": 8.539829405811768e-05, "loss": 4.637, "step": 60500 }, { "epoch": 4.285513559083884, "grad_norm": 1.0472089052200317, "learning_rate": 8.525372271215846e-05, "loss": 4.6283, "step": 61000 }, { "epoch": 4.320640719404244, "grad_norm": 1.102898359298706, "learning_rate": 8.510915136619923e-05, "loss": 4.6212, "step": 61500 }, { "epoch": 4.355767879724603, "grad_norm": 1.1299731731414795, "learning_rate": 8.496458002023999e-05, "loss": 4.6199, "step": 62000 }, { "epoch": 4.3908950400449624, "grad_norm": 1.1199188232421875, "learning_rate": 8.482000867428075e-05, "loss": 4.6187, "step": 62500 }, { "epoch": 4.426022200365322, "grad_norm": 1.071740984916687, "learning_rate": 8.467543732832153e-05, "loss": 4.6139, "step": 63000 }, { "epoch": 4.461149360685682, "grad_norm": 1.076743721961975, "learning_rate": 8.45308659823623e-05, "loss": 4.6059, "step": 63500 }, { "epoch": 4.496276521006042, "grad_norm": 1.1580578088760376, "learning_rate": 8.438629463640307e-05, "loss": 4.603, "step": 64000 }, { "epoch": 4.531403681326402, "grad_norm": 1.0820659399032593, "learning_rate": 8.424172329044384e-05, "loss": 4.5982, "step": 64500 }, { "epoch": 4.566530841646761, "grad_norm": 1.100649356842041, "learning_rate": 8.409715194448461e-05, "loss": 4.5922, "step": 65000 }, { "epoch": 4.601658001967121, "grad_norm": 1.1021618843078613, "learning_rate": 8.395258059852538e-05, "loss": 4.59, "step": 65500 }, { "epoch": 4.636785162287481, "grad_norm": 1.172073245048523, "learning_rate": 8.380800925256614e-05, "loss": 4.5828, "step": 66000 }, { "epoch": 4.671912322607841, "grad_norm": 1.086607575416565, "learning_rate": 8.36634379066069e-05, "loss": 4.5825, "step": 66500 }, { "epoch": 4.7070394829282005, "grad_norm": 1.1319900751113892, "learning_rate": 8.351886656064768e-05, "loss": 4.5746, "step": 67000 }, { "epoch": 4.742166643248559, "grad_norm": 1.1135094165802002, "learning_rate": 8.337429521468846e-05, "loss": 4.5729, "step": 67500 }, { "epoch": 4.777293803568919, "grad_norm": 1.0919950008392334, "learning_rate": 8.322972386872923e-05, "loss": 4.5706, "step": 68000 }, { "epoch": 4.812420963889279, "grad_norm": 1.1084672212600708, "learning_rate": 8.308515252276999e-05, "loss": 4.5695, "step": 68500 }, { "epoch": 4.847548124209639, "grad_norm": 1.0930005311965942, "learning_rate": 8.294058117681077e-05, "loss": 4.5611, "step": 69000 }, { "epoch": 4.8826752845299985, "grad_norm": 1.1142082214355469, "learning_rate": 8.279600983085153e-05, "loss": 4.5606, "step": 69500 }, { "epoch": 4.917802444850358, "grad_norm": 1.1282103061676025, "learning_rate": 8.26514384848923e-05, "loss": 4.5578, "step": 70000 }, { "epoch": 4.917802444850358, "eval_accuracy": 0.6306798098194751, "eval_loss": 4.215607643127441, "eval_runtime": 170.9535, "eval_samples_per_second": 706.075, "eval_steps_per_second": 4.416, "step": 70000 }, { "epoch": 4.952929605170718, "grad_norm": 1.1085766553878784, "learning_rate": 8.250686713893306e-05, "loss": 4.5556, "step": 70500 }, { "epoch": 4.988056765491078, "grad_norm": 1.0641589164733887, "learning_rate": 8.236229579297384e-05, "loss": 4.5485, "step": 71000 }, { "epoch": 5.023183925811438, "grad_norm": 1.1384668350219727, "learning_rate": 8.221772444701461e-05, "loss": 4.547, "step": 71500 }, { "epoch": 5.0583110861317975, "grad_norm": 1.1709734201431274, "learning_rate": 8.207315310105538e-05, "loss": 4.5395, "step": 72000 }, { "epoch": 5.093438246452157, "grad_norm": 1.095879077911377, "learning_rate": 8.192858175509614e-05, "loss": 4.5347, "step": 72500 }, { "epoch": 5.128565406772516, "grad_norm": 1.069482684135437, "learning_rate": 8.178401040913692e-05, "loss": 4.5313, "step": 73000 }, { "epoch": 5.163692567092876, "grad_norm": 1.0765794515609741, "learning_rate": 8.163943906317768e-05, "loss": 4.5303, "step": 73500 }, { "epoch": 5.198819727413236, "grad_norm": 1.0767077207565308, "learning_rate": 8.149486771721845e-05, "loss": 4.526, "step": 74000 }, { "epoch": 5.2339468877335955, "grad_norm": 1.1215113401412964, "learning_rate": 8.135029637125921e-05, "loss": 4.5273, "step": 74500 }, { "epoch": 5.269074048053955, "grad_norm": 1.1744107007980347, "learning_rate": 8.120572502529999e-05, "loss": 4.5235, "step": 75000 }, { "epoch": 5.304201208374315, "grad_norm": 1.0483390092849731, "learning_rate": 8.106115367934077e-05, "loss": 4.5214, "step": 75500 }, { "epoch": 5.339328368694675, "grad_norm": 1.0660243034362793, "learning_rate": 8.091658233338153e-05, "loss": 4.5179, "step": 76000 }, { "epoch": 5.374455529015035, "grad_norm": 1.0934308767318726, "learning_rate": 8.07720109874223e-05, "loss": 4.5145, "step": 76500 }, { "epoch": 5.409582689335394, "grad_norm": 1.0717884302139282, "learning_rate": 8.062743964146306e-05, "loss": 4.5094, "step": 77000 }, { "epoch": 5.444709849655754, "grad_norm": 1.1847779750823975, "learning_rate": 8.048286829550384e-05, "loss": 4.5064, "step": 77500 }, { "epoch": 5.479837009976113, "grad_norm": 1.1186975240707397, "learning_rate": 8.03382969495446e-05, "loss": 4.5085, "step": 78000 }, { "epoch": 5.514964170296473, "grad_norm": 1.0433039665222168, "learning_rate": 8.019372560358536e-05, "loss": 4.5051, "step": 78500 }, { "epoch": 5.550091330616833, "grad_norm": 1.1157066822052002, "learning_rate": 8.004915425762614e-05, "loss": 4.5005, "step": 79000 }, { "epoch": 5.5852184909371925, "grad_norm": 1.0628628730773926, "learning_rate": 7.990458291166692e-05, "loss": 4.4998, "step": 79500 }, { "epoch": 5.620345651257552, "grad_norm": 1.0906633138656616, "learning_rate": 7.976001156570768e-05, "loss": 4.496, "step": 80000 }, { "epoch": 5.620345651257552, "eval_accuracy": 0.6374694361204982, "eval_loss": 4.161874294281006, "eval_runtime": 170.5073, "eval_samples_per_second": 707.923, "eval_steps_per_second": 4.428, "step": 80000 }, { "epoch": 5.655472811577912, "grad_norm": 1.056024193763733, "learning_rate": 7.961544021974845e-05, "loss": 4.4924, "step": 80500 }, { "epoch": 5.690599971898272, "grad_norm": 1.1981945037841797, "learning_rate": 7.947086887378921e-05, "loss": 4.4989, "step": 81000 }, { "epoch": 5.725727132218632, "grad_norm": 1.0816618204116821, "learning_rate": 7.932629752782999e-05, "loss": 4.4925, "step": 81500 }, { "epoch": 5.760854292538991, "grad_norm": 1.0585711002349854, "learning_rate": 7.918172618187075e-05, "loss": 4.4865, "step": 82000 }, { "epoch": 5.795981452859351, "grad_norm": 1.109006643295288, "learning_rate": 7.903715483591153e-05, "loss": 4.488, "step": 82500 }, { "epoch": 5.83110861317971, "grad_norm": 1.0382167100906372, "learning_rate": 7.88925834899523e-05, "loss": 4.4858, "step": 83000 }, { "epoch": 5.866235773500071, "grad_norm": 1.1258573532104492, "learning_rate": 7.874801214399307e-05, "loss": 4.4826, "step": 83500 }, { "epoch": 5.90136293382043, "grad_norm": 1.082244634628296, "learning_rate": 7.860344079803384e-05, "loss": 4.4799, "step": 84000 }, { "epoch": 5.936490094140789, "grad_norm": 1.0842386484146118, "learning_rate": 7.84588694520746e-05, "loss": 4.4774, "step": 84500 }, { "epoch": 5.971617254461149, "grad_norm": 1.0564007759094238, "learning_rate": 7.831429810611536e-05, "loss": 4.4757, "step": 85000 }, { "epoch": 6.006744414781509, "grad_norm": 1.0922231674194336, "learning_rate": 7.816972676015614e-05, "loss": 4.4736, "step": 85500 }, { "epoch": 6.041871575101869, "grad_norm": 1.053505301475525, "learning_rate": 7.802515541419692e-05, "loss": 4.4682, "step": 86000 }, { "epoch": 6.076998735422229, "grad_norm": 1.0822075605392456, "learning_rate": 7.788058406823768e-05, "loss": 4.4697, "step": 86500 }, { "epoch": 6.112125895742588, "grad_norm": 1.097996473312378, "learning_rate": 7.773601272227845e-05, "loss": 4.464, "step": 87000 }, { "epoch": 6.147253056062948, "grad_norm": 1.080566167831421, "learning_rate": 7.759144137631922e-05, "loss": 4.4631, "step": 87500 }, { "epoch": 6.182380216383308, "grad_norm": 1.0531387329101562, "learning_rate": 7.744687003035999e-05, "loss": 4.4623, "step": 88000 }, { "epoch": 6.217507376703667, "grad_norm": 1.0348528623580933, "learning_rate": 7.730229868440075e-05, "loss": 4.4593, "step": 88500 }, { "epoch": 6.252634537024027, "grad_norm": 1.0266953706741333, "learning_rate": 7.715772733844152e-05, "loss": 4.4566, "step": 89000 }, { "epoch": 6.287761697344386, "grad_norm": 1.1049429178237915, "learning_rate": 7.70131559924823e-05, "loss": 4.4565, "step": 89500 }, { "epoch": 6.322888857664746, "grad_norm": 1.0704232454299927, "learning_rate": 7.686858464652307e-05, "loss": 4.457, "step": 90000 }, { "epoch": 6.322888857664746, "eval_accuracy": 0.6424581114867136, "eval_loss": 4.125554084777832, "eval_runtime": 177.4427, "eval_samples_per_second": 680.254, "eval_steps_per_second": 4.255, "step": 90000 }, { "epoch": 6.358016017985106, "grad_norm": 1.0814696550369263, "learning_rate": 7.672401330056384e-05, "loss": 4.4542, "step": 90500 }, { "epoch": 6.393143178305466, "grad_norm": 1.0255705118179321, "learning_rate": 7.65794419546046e-05, "loss": 4.4513, "step": 91000 }, { "epoch": 6.4282703386258255, "grad_norm": 1.0077965259552002, "learning_rate": 7.643487060864536e-05, "loss": 4.4486, "step": 91500 }, { "epoch": 6.463397498946185, "grad_norm": 1.0889065265655518, "learning_rate": 7.629029926268614e-05, "loss": 4.4488, "step": 92000 }, { "epoch": 6.498524659266545, "grad_norm": 1.1086533069610596, "learning_rate": 7.61457279167269e-05, "loss": 4.4487, "step": 92500 }, { "epoch": 6.533651819586905, "grad_norm": 1.0889889001846313, "learning_rate": 7.600115657076767e-05, "loss": 4.4469, "step": 93000 }, { "epoch": 6.568778979907265, "grad_norm": 1.1585131883621216, "learning_rate": 7.585658522480845e-05, "loss": 4.4424, "step": 93500 }, { "epoch": 6.603906140227624, "grad_norm": 1.032514214515686, "learning_rate": 7.571201387884922e-05, "loss": 4.4392, "step": 94000 }, { "epoch": 6.639033300547983, "grad_norm": 1.0205252170562744, "learning_rate": 7.556744253288999e-05, "loss": 4.438, "step": 94500 }, { "epoch": 6.674160460868343, "grad_norm": 1.0339618921279907, "learning_rate": 7.542287118693075e-05, "loss": 4.4356, "step": 95000 }, { "epoch": 6.709287621188703, "grad_norm": 1.0968862771987915, "learning_rate": 7.527829984097152e-05, "loss": 4.4361, "step": 95500 }, { "epoch": 6.744414781509063, "grad_norm": 1.0627433061599731, "learning_rate": 7.51337284950123e-05, "loss": 4.4399, "step": 96000 }, { "epoch": 6.7795419418294225, "grad_norm": 1.084746241569519, "learning_rate": 7.498915714905306e-05, "loss": 4.43, "step": 96500 }, { "epoch": 6.814669102149782, "grad_norm": 1.095624566078186, "learning_rate": 7.484458580309382e-05, "loss": 4.4325, "step": 97000 }, { "epoch": 6.849796262470142, "grad_norm": 1.1057868003845215, "learning_rate": 7.47000144571346e-05, "loss": 4.4249, "step": 97500 }, { "epoch": 6.884923422790502, "grad_norm": 1.0699888467788696, "learning_rate": 7.455544311117538e-05, "loss": 4.4278, "step": 98000 }, { "epoch": 6.920050583110862, "grad_norm": 1.1115961074829102, "learning_rate": 7.441087176521614e-05, "loss": 4.4231, "step": 98500 }, { "epoch": 6.955177743431221, "grad_norm": 1.0538578033447266, "learning_rate": 7.42663004192569e-05, "loss": 4.424, "step": 99000 }, { "epoch": 6.99030490375158, "grad_norm": 1.101834774017334, "learning_rate": 7.412172907329767e-05, "loss": 4.4193, "step": 99500 }, { "epoch": 7.02543206407194, "grad_norm": 1.0599411725997925, "learning_rate": 7.397715772733845e-05, "loss": 4.4199, "step": 100000 }, { "epoch": 7.02543206407194, "eval_accuracy": 0.6468197671632273, "eval_loss": 4.091779708862305, "eval_runtime": 174.7024, "eval_samples_per_second": 690.923, "eval_steps_per_second": 4.322, "step": 100000 }, { "epoch": 7.0605592243923, "grad_norm": 1.0605195760726929, "learning_rate": 7.383258638137921e-05, "loss": 4.4184, "step": 100500 }, { "epoch": 7.09568638471266, "grad_norm": 1.0498496294021606, "learning_rate": 7.368801503541999e-05, "loss": 4.4179, "step": 101000 }, { "epoch": 7.1308135450330195, "grad_norm": 1.1070677042007446, "learning_rate": 7.354344368946075e-05, "loss": 4.4162, "step": 101500 }, { "epoch": 7.165940705353379, "grad_norm": 1.0741323232650757, "learning_rate": 7.339887234350153e-05, "loss": 4.4132, "step": 102000 }, { "epoch": 7.201067865673739, "grad_norm": 1.1206176280975342, "learning_rate": 7.32543009975423e-05, "loss": 4.4119, "step": 102500 }, { "epoch": 7.236195025994099, "grad_norm": 1.0241475105285645, "learning_rate": 7.310972965158306e-05, "loss": 4.4124, "step": 103000 }, { "epoch": 7.271322186314459, "grad_norm": 1.0716612339019775, "learning_rate": 7.296515830562382e-05, "loss": 4.4126, "step": 103500 }, { "epoch": 7.306449346634818, "grad_norm": 1.1098811626434326, "learning_rate": 7.28205869596646e-05, "loss": 4.4081, "step": 104000 }, { "epoch": 7.341576506955178, "grad_norm": 1.0627920627593994, "learning_rate": 7.267601561370536e-05, "loss": 4.4017, "step": 104500 }, { "epoch": 7.376703667275537, "grad_norm": 1.067907452583313, "learning_rate": 7.253144426774614e-05, "loss": 4.4066, "step": 105000 }, { "epoch": 7.411830827595897, "grad_norm": 1.0581320524215698, "learning_rate": 7.23868729217869e-05, "loss": 4.408, "step": 105500 }, { "epoch": 7.446957987916257, "grad_norm": 1.0649940967559814, "learning_rate": 7.224230157582768e-05, "loss": 4.4061, "step": 106000 }, { "epoch": 7.482085148236616, "grad_norm": 1.0771982669830322, "learning_rate": 7.209773022986845e-05, "loss": 4.4027, "step": 106500 }, { "epoch": 7.517212308556976, "grad_norm": 1.0529967546463013, "learning_rate": 7.195315888390921e-05, "loss": 4.3994, "step": 107000 }, { "epoch": 7.552339468877336, "grad_norm": 1.0561673641204834, "learning_rate": 7.180858753794997e-05, "loss": 4.3991, "step": 107500 }, { "epoch": 7.587466629197696, "grad_norm": 1.036828875541687, "learning_rate": 7.166401619199075e-05, "loss": 4.3979, "step": 108000 }, { "epoch": 7.6225937895180556, "grad_norm": 1.1498559713363647, "learning_rate": 7.151944484603153e-05, "loss": 4.3992, "step": 108500 }, { "epoch": 7.657720949838415, "grad_norm": 1.0244455337524414, "learning_rate": 7.13748735000723e-05, "loss": 4.3946, "step": 109000 }, { "epoch": 7.692848110158775, "grad_norm": 1.100959300994873, "learning_rate": 7.123030215411306e-05, "loss": 4.396, "step": 109500 }, { "epoch": 7.727975270479135, "grad_norm": 1.1434977054595947, "learning_rate": 7.108573080815382e-05, "loss": 4.3953, "step": 110000 }, { "epoch": 7.727975270479135, "eval_accuracy": 0.6503580778208442, "eval_loss": 4.067699432373047, "eval_runtime": 179.0849, "eval_samples_per_second": 674.016, "eval_steps_per_second": 4.216, "step": 110000 }, { "epoch": 7.763102430799494, "grad_norm": 1.0459327697753906, "learning_rate": 7.09411594621946e-05, "loss": 4.3932, "step": 110500 }, { "epoch": 7.798229591119854, "grad_norm": 1.0255916118621826, "learning_rate": 7.079658811623536e-05, "loss": 4.3898, "step": 111000 }, { "epoch": 7.833356751440213, "grad_norm": 1.0919266939163208, "learning_rate": 7.065201677027613e-05, "loss": 4.3885, "step": 111500 }, { "epoch": 7.868483911760573, "grad_norm": 1.0899759531021118, "learning_rate": 7.05074454243169e-05, "loss": 4.3919, "step": 112000 }, { "epoch": 7.903611072080933, "grad_norm": 1.015812873840332, "learning_rate": 7.036287407835768e-05, "loss": 4.3895, "step": 112500 }, { "epoch": 7.938738232401293, "grad_norm": 1.0552341938018799, "learning_rate": 7.021830273239845e-05, "loss": 4.3834, "step": 113000 }, { "epoch": 7.9738653927216525, "grad_norm": 1.042982816696167, "learning_rate": 7.007373138643921e-05, "loss": 4.3838, "step": 113500 }, { "epoch": 8.008992553042011, "grad_norm": 1.0557994842529297, "learning_rate": 6.992916004047997e-05, "loss": 4.3844, "step": 114000 }, { "epoch": 8.044119713362372, "grad_norm": 1.089377999305725, "learning_rate": 6.978458869452075e-05, "loss": 4.3795, "step": 114500 }, { "epoch": 8.079246873682731, "grad_norm": 1.0726460218429565, "learning_rate": 6.964001734856152e-05, "loss": 4.3785, "step": 115000 }, { "epoch": 8.114374034003092, "grad_norm": 1.0248174667358398, "learning_rate": 6.949544600260228e-05, "loss": 4.3778, "step": 115500 }, { "epoch": 8.14950119432345, "grad_norm": 1.0784227848052979, "learning_rate": 6.935087465664306e-05, "loss": 4.3791, "step": 116000 }, { "epoch": 8.184628354643811, "grad_norm": 1.0527198314666748, "learning_rate": 6.920630331068383e-05, "loss": 4.3782, "step": 116500 }, { "epoch": 8.21975551496417, "grad_norm": 1.0497902631759644, "learning_rate": 6.90617319647246e-05, "loss": 4.3753, "step": 117000 }, { "epoch": 8.25488267528453, "grad_norm": 1.0928125381469727, "learning_rate": 6.891716061876536e-05, "loss": 4.3753, "step": 117500 }, { "epoch": 8.29000983560489, "grad_norm": 1.079317331314087, "learning_rate": 6.877258927280613e-05, "loss": 4.3696, "step": 118000 }, { "epoch": 8.325136995925249, "grad_norm": 1.0786110162734985, "learning_rate": 6.86280179268469e-05, "loss": 4.3698, "step": 118500 }, { "epoch": 8.36026415624561, "grad_norm": 1.0580191612243652, "learning_rate": 6.848344658088767e-05, "loss": 4.3706, "step": 119000 }, { "epoch": 8.395391316565968, "grad_norm": 1.0516324043273926, "learning_rate": 6.833887523492845e-05, "loss": 4.37, "step": 119500 }, { "epoch": 8.430518476886329, "grad_norm": 1.0617542266845703, "learning_rate": 6.819430388896921e-05, "loss": 4.3703, "step": 120000 }, { "epoch": 8.430518476886329, "eval_accuracy": 0.6537571905928443, "eval_loss": 4.044128894805908, "eval_runtime": 184.2009, "eval_samples_per_second": 655.295, "eval_steps_per_second": 4.099, "step": 120000 }, { "epoch": 8.465645637206688, "grad_norm": 1.052673101425171, "learning_rate": 6.804973254300999e-05, "loss": 4.3688, "step": 120500 }, { "epoch": 8.500772797527048, "grad_norm": 1.0831342935562134, "learning_rate": 6.790516119705075e-05, "loss": 4.3699, "step": 121000 }, { "epoch": 8.535899957847407, "grad_norm": 1.0830225944519043, "learning_rate": 6.776058985109152e-05, "loss": 4.369, "step": 121500 }, { "epoch": 8.571027118167768, "grad_norm": 1.0491890907287598, "learning_rate": 6.761601850513228e-05, "loss": 4.3659, "step": 122000 }, { "epoch": 8.606154278488127, "grad_norm": 1.042027473449707, "learning_rate": 6.747144715917306e-05, "loss": 4.3636, "step": 122500 }, { "epoch": 8.641281438808488, "grad_norm": 0.9986875653266907, "learning_rate": 6.732687581321382e-05, "loss": 4.363, "step": 123000 }, { "epoch": 8.676408599128846, "grad_norm": 1.0622915029525757, "learning_rate": 6.71823044672546e-05, "loss": 4.3587, "step": 123500 }, { "epoch": 8.711535759449205, "grad_norm": 1.0308351516723633, "learning_rate": 6.703773312129536e-05, "loss": 4.3566, "step": 124000 }, { "epoch": 8.746662919769566, "grad_norm": 1.0938032865524292, "learning_rate": 6.689316177533613e-05, "loss": 4.3528, "step": 124500 }, { "epoch": 8.781790080089925, "grad_norm": 1.0253453254699707, "learning_rate": 6.67485904293769e-05, "loss": 4.3592, "step": 125000 }, { "epoch": 8.816917240410286, "grad_norm": 1.0983203649520874, "learning_rate": 6.660401908341767e-05, "loss": 4.3561, "step": 125500 }, { "epoch": 8.852044400730644, "grad_norm": 1.0602632761001587, "learning_rate": 6.645944773745843e-05, "loss": 4.3554, "step": 126000 }, { "epoch": 8.887171561051005, "grad_norm": 1.144900918006897, "learning_rate": 6.631487639149921e-05, "loss": 4.3554, "step": 126500 }, { "epoch": 8.922298721371364, "grad_norm": 1.0928415060043335, "learning_rate": 6.617030504553999e-05, "loss": 4.3529, "step": 127000 }, { "epoch": 8.957425881691725, "grad_norm": 1.0777556896209717, "learning_rate": 6.602573369958075e-05, "loss": 4.3585, "step": 127500 }, { "epoch": 8.992553042012084, "grad_norm": 1.1269131898880005, "learning_rate": 6.588116235362152e-05, "loss": 4.3569, "step": 128000 }, { "epoch": 9.027680202332444, "grad_norm": 1.0965328216552734, "learning_rate": 6.573659100766228e-05, "loss": 4.3489, "step": 128500 }, { "epoch": 9.062807362652803, "grad_norm": 1.0457299947738647, "learning_rate": 6.559201966170306e-05, "loss": 4.3476, "step": 129000 }, { "epoch": 9.097934522973162, "grad_norm": 1.1396300792694092, "learning_rate": 6.544744831574382e-05, "loss": 4.3503, "step": 129500 }, { "epoch": 9.133061683293523, "grad_norm": 1.1194312572479248, "learning_rate": 6.530287696978458e-05, "loss": 4.3437, "step": 130000 }, { "epoch": 9.133061683293523, "eval_accuracy": 0.6559576129359359, "eval_loss": 4.029501914978027, "eval_runtime": 184.5396, "eval_samples_per_second": 654.093, "eval_steps_per_second": 4.091, "step": 130000 }, { "epoch": 9.168188843613882, "grad_norm": 1.0390701293945312, "learning_rate": 6.515830562382536e-05, "loss": 4.3458, "step": 130500 }, { "epoch": 9.203316003934242, "grad_norm": 1.0907750129699707, "learning_rate": 6.501373427786614e-05, "loss": 4.3485, "step": 131000 }, { "epoch": 9.238443164254601, "grad_norm": 1.0733038187026978, "learning_rate": 6.48691629319069e-05, "loss": 4.3458, "step": 131500 }, { "epoch": 9.273570324574962, "grad_norm": 1.0722750425338745, "learning_rate": 6.472459158594767e-05, "loss": 4.3484, "step": 132000 }, { "epoch": 9.30869748489532, "grad_norm": 1.0519601106643677, "learning_rate": 6.458002023998843e-05, "loss": 4.3443, "step": 132500 }, { "epoch": 9.343824645215681, "grad_norm": 1.0446197986602783, "learning_rate": 6.443544889402921e-05, "loss": 4.3417, "step": 133000 }, { "epoch": 9.37895180553604, "grad_norm": 1.063036561012268, "learning_rate": 6.429087754806997e-05, "loss": 4.3402, "step": 133500 }, { "epoch": 9.414078965856401, "grad_norm": 1.056117296218872, "learning_rate": 6.414630620211074e-05, "loss": 4.3412, "step": 134000 }, { "epoch": 9.44920612617676, "grad_norm": 1.1823673248291016, "learning_rate": 6.400173485615151e-05, "loss": 4.3368, "step": 134500 }, { "epoch": 9.484333286497119, "grad_norm": 1.0423256158828735, "learning_rate": 6.385716351019229e-05, "loss": 4.3362, "step": 135000 }, { "epoch": 9.51946044681748, "grad_norm": 1.0792287588119507, "learning_rate": 6.371259216423306e-05, "loss": 4.3383, "step": 135500 }, { "epoch": 9.554587607137838, "grad_norm": 1.029358983039856, "learning_rate": 6.356802081827382e-05, "loss": 4.3367, "step": 136000 }, { "epoch": 9.589714767458199, "grad_norm": 1.0576837062835693, "learning_rate": 6.342344947231458e-05, "loss": 4.3376, "step": 136500 }, { "epoch": 9.624841927778558, "grad_norm": 1.0619704723358154, "learning_rate": 6.327887812635536e-05, "loss": 4.3347, "step": 137000 }, { "epoch": 9.659969088098919, "grad_norm": 1.0426596403121948, "learning_rate": 6.313430678039613e-05, "loss": 4.3352, "step": 137500 }, { "epoch": 9.695096248419278, "grad_norm": 1.0416324138641357, "learning_rate": 6.29897354344369e-05, "loss": 4.3294, "step": 138000 }, { "epoch": 9.730223408739638, "grad_norm": 1.0752136707305908, "learning_rate": 6.284516408847767e-05, "loss": 4.3363, "step": 138500 }, { "epoch": 9.765350569059997, "grad_norm": 1.0406975746154785, "learning_rate": 6.270059274251843e-05, "loss": 4.3305, "step": 139000 }, { "epoch": 9.800477729380358, "grad_norm": 1.0854167938232422, "learning_rate": 6.255602139655921e-05, "loss": 4.3319, "step": 139500 }, { "epoch": 9.835604889700717, "grad_norm": 1.0566082000732422, "learning_rate": 6.241145005059997e-05, "loss": 4.3295, "step": 140000 }, { "epoch": 9.835604889700717, "eval_accuracy": 0.6593799691761195, "eval_loss": 4.008376121520996, "eval_runtime": 187.0577, "eval_samples_per_second": 645.288, "eval_steps_per_second": 4.036, "step": 140000 }, { "epoch": 9.870732050021076, "grad_norm": 1.103357195854187, "learning_rate": 6.226687870464074e-05, "loss": 4.3305, "step": 140500 }, { "epoch": 9.905859210341436, "grad_norm": 1.1054251194000244, "learning_rate": 6.212230735868151e-05, "loss": 4.3275, "step": 141000 }, { "epoch": 9.940986370661795, "grad_norm": 1.1275943517684937, "learning_rate": 6.197773601272228e-05, "loss": 4.3281, "step": 141500 }, { "epoch": 9.976113530982156, "grad_norm": 1.091591238975525, "learning_rate": 6.183316466676306e-05, "loss": 4.3247, "step": 142000 }, { "epoch": 10.011240691302515, "grad_norm": 1.0717936754226685, "learning_rate": 6.168859332080382e-05, "loss": 4.3276, "step": 142500 }, { "epoch": 10.046367851622875, "grad_norm": 1.0560671091079712, "learning_rate": 6.154402197484458e-05, "loss": 4.3265, "step": 143000 }, { "epoch": 10.081495011943234, "grad_norm": 1.0523194074630737, "learning_rate": 6.139945062888536e-05, "loss": 4.322, "step": 143500 }, { "epoch": 10.116622172263595, "grad_norm": 1.1264257431030273, "learning_rate": 6.125487928292613e-05, "loss": 4.3201, "step": 144000 }, { "epoch": 10.151749332583954, "grad_norm": 1.1297935247421265, "learning_rate": 6.111030793696689e-05, "loss": 4.3212, "step": 144500 }, { "epoch": 10.186876492904315, "grad_norm": 1.0614631175994873, "learning_rate": 6.096573659100766e-05, "loss": 4.3213, "step": 145000 }, { "epoch": 10.222003653224673, "grad_norm": 1.099090576171875, "learning_rate": 6.082116524504844e-05, "loss": 4.3229, "step": 145500 }, { "epoch": 10.257130813545032, "grad_norm": 1.0678114891052246, "learning_rate": 6.067659389908921e-05, "loss": 4.3223, "step": 146000 }, { "epoch": 10.292257973865393, "grad_norm": 1.0615801811218262, "learning_rate": 6.053202255312997e-05, "loss": 4.3162, "step": 146500 }, { "epoch": 10.327385134185752, "grad_norm": 1.112459421157837, "learning_rate": 6.0387451207170744e-05, "loss": 4.3172, "step": 147000 }, { "epoch": 10.362512294506113, "grad_norm": 1.1005117893218994, "learning_rate": 6.024287986121151e-05, "loss": 4.3179, "step": 147500 }, { "epoch": 10.397639454826471, "grad_norm": 1.0875022411346436, "learning_rate": 6.009830851525228e-05, "loss": 4.3147, "step": 148000 }, { "epoch": 10.432766615146832, "grad_norm": 1.1111528873443604, "learning_rate": 5.995373716929304e-05, "loss": 4.319, "step": 148500 }, { "epoch": 10.467893775467191, "grad_norm": 1.0767004489898682, "learning_rate": 5.980916582333381e-05, "loss": 4.313, "step": 149000 }, { "epoch": 10.503020935787552, "grad_norm": 1.0752588510513306, "learning_rate": 5.966459447737459e-05, "loss": 4.3125, "step": 149500 }, { "epoch": 10.53814809610791, "grad_norm": 1.0421361923217773, "learning_rate": 5.952002313141536e-05, "loss": 4.3125, "step": 150000 }, { "epoch": 10.53814809610791, "eval_accuracy": 0.6611775395959194, "eval_loss": 3.995457649230957, "eval_runtime": 180.9695, "eval_samples_per_second": 666.996, "eval_steps_per_second": 4.172, "step": 150000 }, { "epoch": 10.573275256428271, "grad_norm": 1.0622786283493042, "learning_rate": 5.9375451785456126e-05, "loss": 4.3179, "step": 150500 }, { "epoch": 10.60840241674863, "grad_norm": 1.0611732006072998, "learning_rate": 5.9230880439496896e-05, "loss": 4.3112, "step": 151000 }, { "epoch": 10.643529577068989, "grad_norm": 1.0922372341156006, "learning_rate": 5.908630909353766e-05, "loss": 4.3122, "step": 151500 }, { "epoch": 10.67865673738935, "grad_norm": 1.028787612915039, "learning_rate": 5.894173774757843e-05, "loss": 4.3117, "step": 152000 }, { "epoch": 10.713783897709709, "grad_norm": 1.1244584321975708, "learning_rate": 5.8797166401619195e-05, "loss": 4.3087, "step": 152500 }, { "epoch": 10.74891105803007, "grad_norm": 1.1056017875671387, "learning_rate": 5.865259505565997e-05, "loss": 4.3103, "step": 153000 }, { "epoch": 10.784038218350428, "grad_norm": 1.0762313604354858, "learning_rate": 5.8508023709700743e-05, "loss": 4.3124, "step": 153500 }, { "epoch": 10.819165378670789, "grad_norm": 1.0612378120422363, "learning_rate": 5.8363452363741514e-05, "loss": 4.31, "step": 154000 }, { "epoch": 10.854292538991148, "grad_norm": 1.060267686843872, "learning_rate": 5.821888101778228e-05, "loss": 4.31, "step": 154500 }, { "epoch": 10.889419699311508, "grad_norm": 1.0424984693527222, "learning_rate": 5.807430967182305e-05, "loss": 4.308, "step": 155000 }, { "epoch": 10.924546859631867, "grad_norm": 1.0900484323501587, "learning_rate": 5.792973832586381e-05, "loss": 4.3084, "step": 155500 }, { "epoch": 10.959674019952226, "grad_norm": 1.0835260152816772, "learning_rate": 5.7785166979904584e-05, "loss": 4.3074, "step": 156000 }, { "epoch": 10.994801180272587, "grad_norm": 1.079627275466919, "learning_rate": 5.764059563394536e-05, "loss": 4.3057, "step": 156500 }, { "epoch": 11.029928340592946, "grad_norm": 1.1124992370605469, "learning_rate": 5.7496024287986125e-05, "loss": 4.2991, "step": 157000 }, { "epoch": 11.065055500913306, "grad_norm": 1.0468932390213013, "learning_rate": 5.7351452942026896e-05, "loss": 4.3031, "step": 157500 }, { "epoch": 11.100182661233665, "grad_norm": 1.1354649066925049, "learning_rate": 5.720688159606766e-05, "loss": 4.3033, "step": 158000 }, { "epoch": 11.135309821554026, "grad_norm": 1.083852767944336, "learning_rate": 5.706231025010843e-05, "loss": 4.2994, "step": 158500 }, { "epoch": 11.170436981874385, "grad_norm": 1.0800615549087524, "learning_rate": 5.6917738904149195e-05, "loss": 4.2994, "step": 159000 }, { "epoch": 11.205564142194746, "grad_norm": 1.0894683599472046, "learning_rate": 5.6773167558189966e-05, "loss": 4.3027, "step": 159500 }, { "epoch": 11.240691302515105, "grad_norm": 1.1090705394744873, "learning_rate": 5.6628596212230736e-05, "loss": 4.3048, "step": 160000 }, { "epoch": 11.240691302515105, "eval_accuracy": 0.662732521317256, "eval_loss": 3.984241008758545, "eval_runtime": 181.0627, "eval_samples_per_second": 666.653, "eval_steps_per_second": 4.17, "step": 160000 }, { "epoch": 11.275818462835465, "grad_norm": 1.0897754430770874, "learning_rate": 5.6484024866271514e-05, "loss": 4.2973, "step": 160500 }, { "epoch": 11.310945623155824, "grad_norm": 1.0397636890411377, "learning_rate": 5.633945352031228e-05, "loss": 4.3019, "step": 161000 }, { "epoch": 11.346072783476183, "grad_norm": 1.0997382402420044, "learning_rate": 5.619488217435305e-05, "loss": 4.2982, "step": 161500 }, { "epoch": 11.381199943796544, "grad_norm": 1.106771469116211, "learning_rate": 5.605031082839381e-05, "loss": 4.2984, "step": 162000 }, { "epoch": 11.416327104116903, "grad_norm": 1.1233487129211426, "learning_rate": 5.5905739482434584e-05, "loss": 4.2955, "step": 162500 }, { "epoch": 11.451454264437263, "grad_norm": 1.042432427406311, "learning_rate": 5.576116813647535e-05, "loss": 4.2917, "step": 163000 }, { "epoch": 11.486581424757622, "grad_norm": 1.0913660526275635, "learning_rate": 5.561659679051612e-05, "loss": 4.295, "step": 163500 }, { "epoch": 11.521708585077983, "grad_norm": 1.1289492845535278, "learning_rate": 5.5472025444556896e-05, "loss": 4.295, "step": 164000 }, { "epoch": 11.556835745398342, "grad_norm": 1.1050851345062256, "learning_rate": 5.532745409859767e-05, "loss": 4.2935, "step": 164500 }, { "epoch": 11.591962905718702, "grad_norm": 1.0680861473083496, "learning_rate": 5.518288275263843e-05, "loss": 4.2934, "step": 165000 }, { "epoch": 11.627090066039061, "grad_norm": 1.0729119777679443, "learning_rate": 5.50383114066792e-05, "loss": 4.2942, "step": 165500 }, { "epoch": 11.662217226359422, "grad_norm": 1.0583761930465698, "learning_rate": 5.4893740060719965e-05, "loss": 4.2921, "step": 166000 }, { "epoch": 11.69734438667978, "grad_norm": 1.0220874547958374, "learning_rate": 5.4749168714760736e-05, "loss": 4.2928, "step": 166500 }, { "epoch": 11.73247154700014, "grad_norm": 1.050709843635559, "learning_rate": 5.46045973688015e-05, "loss": 4.2886, "step": 167000 }, { "epoch": 11.7675987073205, "grad_norm": 1.0787092447280884, "learning_rate": 5.446002602284227e-05, "loss": 4.2891, "step": 167500 }, { "epoch": 11.80272586764086, "grad_norm": 1.0705547332763672, "learning_rate": 5.431545467688305e-05, "loss": 4.2902, "step": 168000 }, { "epoch": 11.83785302796122, "grad_norm": 1.0446017980575562, "learning_rate": 5.417088333092382e-05, "loss": 4.286, "step": 168500 }, { "epoch": 11.872980188281579, "grad_norm": 1.1501891613006592, "learning_rate": 5.402631198496458e-05, "loss": 4.2871, "step": 169000 }, { "epoch": 11.90810734860194, "grad_norm": 1.14247465133667, "learning_rate": 5.3881740639005354e-05, "loss": 4.2869, "step": 169500 }, { "epoch": 11.943234508922298, "grad_norm": 1.073425054550171, "learning_rate": 5.373716929304612e-05, "loss": 4.2863, "step": 170000 }, { "epoch": 11.943234508922298, "eval_accuracy": 0.6645155616613816, "eval_loss": 3.9726850986480713, "eval_runtime": 181.6423, "eval_samples_per_second": 664.526, "eval_steps_per_second": 4.157, "step": 170000 }, { "epoch": 11.978361669242659, "grad_norm": 1.1048774719238281, "learning_rate": 5.359259794708689e-05, "loss": 4.2873, "step": 170500 }, { "epoch": 12.013488829563018, "grad_norm": 1.075563669204712, "learning_rate": 5.344802660112765e-05, "loss": 4.2864, "step": 171000 }, { "epoch": 12.048615989883379, "grad_norm": 1.0765880346298218, "learning_rate": 5.330345525516843e-05, "loss": 4.285, "step": 171500 }, { "epoch": 12.083743150203738, "grad_norm": 1.0824280977249146, "learning_rate": 5.31588839092092e-05, "loss": 4.2813, "step": 172000 }, { "epoch": 12.118870310524096, "grad_norm": 1.0891708135604858, "learning_rate": 5.3014312563249965e-05, "loss": 4.2838, "step": 172500 }, { "epoch": 12.153997470844457, "grad_norm": 1.0725291967391968, "learning_rate": 5.2869741217290736e-05, "loss": 4.2873, "step": 173000 }, { "epoch": 12.189124631164816, "grad_norm": 1.0460180044174194, "learning_rate": 5.272516987133151e-05, "loss": 4.2834, "step": 173500 }, { "epoch": 12.224251791485177, "grad_norm": 1.0577491521835327, "learning_rate": 5.258059852537227e-05, "loss": 4.2777, "step": 174000 }, { "epoch": 12.259378951805536, "grad_norm": 1.06594717502594, "learning_rate": 5.243602717941304e-05, "loss": 4.2835, "step": 174500 }, { "epoch": 12.294506112125896, "grad_norm": 1.0660723447799683, "learning_rate": 5.229145583345382e-05, "loss": 4.2784, "step": 175000 }, { "epoch": 12.329633272446255, "grad_norm": 1.1456482410430908, "learning_rate": 5.214688448749458e-05, "loss": 4.2834, "step": 175500 }, { "epoch": 12.364760432766616, "grad_norm": 1.0873714685440063, "learning_rate": 5.2002313141535354e-05, "loss": 4.2788, "step": 176000 }, { "epoch": 12.399887593086975, "grad_norm": 1.0916754007339478, "learning_rate": 5.185774179557612e-05, "loss": 4.2776, "step": 176500 }, { "epoch": 12.435014753407334, "grad_norm": 1.058746099472046, "learning_rate": 5.171317044961689e-05, "loss": 4.2797, "step": 177000 }, { "epoch": 12.470141913727694, "grad_norm": 1.0694043636322021, "learning_rate": 5.156859910365765e-05, "loss": 4.2762, "step": 177500 }, { "epoch": 12.505269074048053, "grad_norm": 1.1144299507141113, "learning_rate": 5.1424027757698423e-05, "loss": 4.2798, "step": 178000 }, { "epoch": 12.540396234368414, "grad_norm": 1.096498727798462, "learning_rate": 5.127945641173919e-05, "loss": 4.2798, "step": 178500 }, { "epoch": 12.575523394688773, "grad_norm": 1.0573207139968872, "learning_rate": 5.113488506577997e-05, "loss": 4.2763, "step": 179000 }, { "epoch": 12.610650555009133, "grad_norm": 1.0426855087280273, "learning_rate": 5.0990313719820736e-05, "loss": 4.2757, "step": 179500 }, { "epoch": 12.645777715329492, "grad_norm": 1.0871822834014893, "learning_rate": 5.0845742373861507e-05, "loss": 4.276, "step": 180000 }, { "epoch": 12.645777715329492, "eval_accuracy": 0.6663402908282602, "eval_loss": 3.9591915607452393, "eval_runtime": 177.7384, "eval_samples_per_second": 679.122, "eval_steps_per_second": 4.248, "step": 180000 }, { "epoch": 12.680904875649853, "grad_norm": 1.1127347946166992, "learning_rate": 5.070117102790227e-05, "loss": 4.2741, "step": 180500 }, { "epoch": 12.716032035970212, "grad_norm": 1.1301934719085693, "learning_rate": 5.055659968194304e-05, "loss": 4.2746, "step": 181000 }, { "epoch": 12.751159196290573, "grad_norm": 1.1162948608398438, "learning_rate": 5.0412028335983805e-05, "loss": 4.2751, "step": 181500 }, { "epoch": 12.786286356610931, "grad_norm": 1.0459678173065186, "learning_rate": 5.0267456990024576e-05, "loss": 4.2747, "step": 182000 }, { "epoch": 12.82141351693129, "grad_norm": 1.0574650764465332, "learning_rate": 5.0122885644065354e-05, "loss": 4.2764, "step": 182500 }, { "epoch": 12.856540677251651, "grad_norm": 1.1084128618240356, "learning_rate": 4.997831429810612e-05, "loss": 4.2721, "step": 183000 }, { "epoch": 12.89166783757201, "grad_norm": 1.1307094097137451, "learning_rate": 4.983374295214689e-05, "loss": 4.2741, "step": 183500 }, { "epoch": 12.92679499789237, "grad_norm": 1.0601236820220947, "learning_rate": 4.968917160618766e-05, "loss": 4.2704, "step": 184000 }, { "epoch": 12.96192215821273, "grad_norm": 1.0482027530670166, "learning_rate": 4.954460026022842e-05, "loss": 4.2748, "step": 184500 }, { "epoch": 12.99704931853309, "grad_norm": 1.1317024230957031, "learning_rate": 4.9400028914269194e-05, "loss": 4.2716, "step": 185000 }, { "epoch": 13.032176478853449, "grad_norm": 1.0516389608383179, "learning_rate": 4.9255457568309965e-05, "loss": 4.2673, "step": 185500 }, { "epoch": 13.06730363917381, "grad_norm": 1.083932638168335, "learning_rate": 4.9110886222350736e-05, "loss": 4.2667, "step": 186000 }, { "epoch": 13.102430799494169, "grad_norm": 1.0865064859390259, "learning_rate": 4.89663148763915e-05, "loss": 4.2687, "step": 186500 }, { "epoch": 13.13755795981453, "grad_norm": 1.0824753046035767, "learning_rate": 4.882174353043227e-05, "loss": 4.2644, "step": 187000 }, { "epoch": 13.172685120134888, "grad_norm": 1.0726726055145264, "learning_rate": 4.867717218447304e-05, "loss": 4.2661, "step": 187500 }, { "epoch": 13.207812280455247, "grad_norm": 1.0734553337097168, "learning_rate": 4.853260083851381e-05, "loss": 4.2665, "step": 188000 }, { "epoch": 13.242939440775608, "grad_norm": 1.1151655912399292, "learning_rate": 4.8388029492554576e-05, "loss": 4.2662, "step": 188500 }, { "epoch": 13.278066601095967, "grad_norm": 1.075623631477356, "learning_rate": 4.824345814659535e-05, "loss": 4.2652, "step": 189000 }, { "epoch": 13.313193761416327, "grad_norm": 1.0625180006027222, "learning_rate": 4.809888680063612e-05, "loss": 4.2677, "step": 189500 }, { "epoch": 13.348320921736686, "grad_norm": 1.089568853378296, "learning_rate": 4.795431545467688e-05, "loss": 4.2651, "step": 190000 }, { "epoch": 13.348320921736686, "eval_accuracy": 0.6669317028733285, "eval_loss": 3.954268217086792, "eval_runtime": 175.2236, "eval_samples_per_second": 688.869, "eval_steps_per_second": 4.309, "step": 190000 }, { "epoch": 13.383448082057047, "grad_norm": 1.1519585847854614, "learning_rate": 4.780974410871765e-05, "loss": 4.2602, "step": 190500 }, { "epoch": 13.418575242377406, "grad_norm": 1.1269229650497437, "learning_rate": 4.766517276275842e-05, "loss": 4.2659, "step": 191000 }, { "epoch": 13.453702402697767, "grad_norm": 1.0760694742202759, "learning_rate": 4.7520601416799194e-05, "loss": 4.2612, "step": 191500 }, { "epoch": 13.488829563018125, "grad_norm": 1.0666991472244263, "learning_rate": 4.737603007083996e-05, "loss": 4.2648, "step": 192000 }, { "epoch": 13.523956723338486, "grad_norm": 1.0701134204864502, "learning_rate": 4.723145872488073e-05, "loss": 4.2652, "step": 192500 }, { "epoch": 13.559083883658845, "grad_norm": 1.1397043466567993, "learning_rate": 4.70868873789215e-05, "loss": 4.2629, "step": 193000 }, { "epoch": 13.594211043979204, "grad_norm": 1.1100988388061523, "learning_rate": 4.694231603296227e-05, "loss": 4.2636, "step": 193500 }, { "epoch": 13.629338204299565, "grad_norm": 1.108329176902771, "learning_rate": 4.6797744687003034e-05, "loss": 4.2596, "step": 194000 }, { "epoch": 13.664465364619923, "grad_norm": 1.1289511919021606, "learning_rate": 4.665317334104381e-05, "loss": 4.2594, "step": 194500 }, { "epoch": 13.699592524940284, "grad_norm": 1.1572721004486084, "learning_rate": 4.6508601995084576e-05, "loss": 4.2611, "step": 195000 }, { "epoch": 13.734719685260643, "grad_norm": 1.0808353424072266, "learning_rate": 4.6364030649125346e-05, "loss": 4.2613, "step": 195500 }, { "epoch": 13.769846845581004, "grad_norm": 1.1108957529067993, "learning_rate": 4.621945930316611e-05, "loss": 4.261, "step": 196000 }, { "epoch": 13.804974005901363, "grad_norm": 1.1699885129928589, "learning_rate": 4.607488795720689e-05, "loss": 4.2585, "step": 196500 }, { "epoch": 13.840101166221723, "grad_norm": 1.0974141359329224, "learning_rate": 4.593031661124765e-05, "loss": 4.2586, "step": 197000 }, { "epoch": 13.875228326542082, "grad_norm": 1.0936520099639893, "learning_rate": 4.578574526528842e-05, "loss": 4.2573, "step": 197500 }, { "epoch": 13.910355486862443, "grad_norm": 1.0823609828948975, "learning_rate": 4.564117391932919e-05, "loss": 4.2583, "step": 198000 }, { "epoch": 13.945482647182802, "grad_norm": 1.1502684354782104, "learning_rate": 4.5496602573369964e-05, "loss": 4.2577, "step": 198500 }, { "epoch": 13.98060980750316, "grad_norm": 1.0798330307006836, "learning_rate": 4.535203122741073e-05, "loss": 4.2613, "step": 199000 }, { "epoch": 14.015736967823521, "grad_norm": 1.0376824140548706, "learning_rate": 4.52074598814515e-05, "loss": 4.255, "step": 199500 }, { "epoch": 14.05086412814388, "grad_norm": 1.1139241456985474, "learning_rate": 4.506288853549227e-05, "loss": 4.2573, "step": 200000 }, { "epoch": 14.05086412814388, "eval_accuracy": 0.6683491505244313, "eval_loss": 3.9437615871429443, "eval_runtime": 175.8986, "eval_samples_per_second": 686.225, "eval_steps_per_second": 4.292, "step": 200000 }, { "epoch": 14.08599128846424, "grad_norm": 1.0487502813339233, "learning_rate": 4.491831718953304e-05, "loss": 4.2527, "step": 200500 }, { "epoch": 14.1211184487846, "grad_norm": 1.0906695127487183, "learning_rate": 4.4773745843573805e-05, "loss": 4.2552, "step": 201000 }, { "epoch": 14.15624560910496, "grad_norm": 1.0857034921646118, "learning_rate": 4.4629174497614575e-05, "loss": 4.2534, "step": 201500 }, { "epoch": 14.19137276942532, "grad_norm": 1.1025879383087158, "learning_rate": 4.4484603151655346e-05, "loss": 4.2504, "step": 202000 }, { "epoch": 14.22649992974568, "grad_norm": 1.088425636291504, "learning_rate": 4.434003180569612e-05, "loss": 4.2517, "step": 202500 }, { "epoch": 14.261627090066039, "grad_norm": 1.1351035833358765, "learning_rate": 4.419546045973688e-05, "loss": 4.2536, "step": 203000 }, { "epoch": 14.2967542503864, "grad_norm": 1.1109907627105713, "learning_rate": 4.405088911377765e-05, "loss": 4.251, "step": 203500 }, { "epoch": 14.331881410706758, "grad_norm": 1.0982377529144287, "learning_rate": 4.390631776781842e-05, "loss": 4.2554, "step": 204000 }, { "epoch": 14.367008571027117, "grad_norm": 1.0929986238479614, "learning_rate": 4.3761746421859187e-05, "loss": 4.2515, "step": 204500 }, { "epoch": 14.402135731347478, "grad_norm": 1.1218113899230957, "learning_rate": 4.361717507589996e-05, "loss": 4.2497, "step": 205000 }, { "epoch": 14.437262891667837, "grad_norm": 1.0992532968521118, "learning_rate": 4.347260372994073e-05, "loss": 4.2532, "step": 205500 }, { "epoch": 14.472390051988198, "grad_norm": 1.100758671760559, "learning_rate": 4.33280323839815e-05, "loss": 4.2502, "step": 206000 }, { "epoch": 14.507517212308557, "grad_norm": 1.1153535842895508, "learning_rate": 4.318346103802226e-05, "loss": 4.2492, "step": 206500 }, { "epoch": 14.542644372628917, "grad_norm": 1.1434364318847656, "learning_rate": 4.3038889692063034e-05, "loss": 4.2469, "step": 207000 }, { "epoch": 14.577771532949276, "grad_norm": 1.091856837272644, "learning_rate": 4.2894318346103804e-05, "loss": 4.2482, "step": 207500 }, { "epoch": 14.612898693269637, "grad_norm": 1.1288974285125732, "learning_rate": 4.2749747000144575e-05, "loss": 4.2503, "step": 208000 }, { "epoch": 14.648025853589996, "grad_norm": 1.0787211656570435, "learning_rate": 4.260517565418534e-05, "loss": 4.249, "step": 208500 }, { "epoch": 14.683153013910356, "grad_norm": 1.0996347665786743, "learning_rate": 4.246060430822611e-05, "loss": 4.248, "step": 209000 }, { "epoch": 14.718280174230715, "grad_norm": 1.080690860748291, "learning_rate": 4.231603296226688e-05, "loss": 4.2491, "step": 209500 }, { "epoch": 14.753407334551074, "grad_norm": 1.1203533411026, "learning_rate": 4.217146161630765e-05, "loss": 4.247, "step": 210000 }, { "epoch": 14.753407334551074, "eval_accuracy": 0.6699330503644155, "eval_loss": 3.934297800064087, "eval_runtime": 171.9163, "eval_samples_per_second": 702.121, "eval_steps_per_second": 4.392, "step": 210000 }, { "epoch": 14.788534494871435, "grad_norm": 1.16519033908844, "learning_rate": 4.2026890270348416e-05, "loss": 4.2459, "step": 210500 }, { "epoch": 14.823661655191794, "grad_norm": 1.153714895248413, "learning_rate": 4.1882318924389186e-05, "loss": 4.2457, "step": 211000 }, { "epoch": 14.858788815512154, "grad_norm": 1.1054308414459229, "learning_rate": 4.173774757842996e-05, "loss": 4.2473, "step": 211500 }, { "epoch": 14.893915975832513, "grad_norm": 1.119315266609192, "learning_rate": 4.159317623247073e-05, "loss": 4.2445, "step": 212000 }, { "epoch": 14.929043136152874, "grad_norm": 1.1489713191986084, "learning_rate": 4.144860488651149e-05, "loss": 4.2482, "step": 212500 }, { "epoch": 14.964170296473233, "grad_norm": 1.1230578422546387, "learning_rate": 4.130403354055227e-05, "loss": 4.2467, "step": 213000 }, { "epoch": 14.999297456793593, "grad_norm": 1.1186426877975464, "learning_rate": 4.1159462194593033e-05, "loss": 4.2466, "step": 213500 }, { "epoch": 15.034424617113952, "grad_norm": 1.1236460208892822, "learning_rate": 4.1014890848633804e-05, "loss": 4.24, "step": 214000 }, { "epoch": 15.069551777434313, "grad_norm": 1.1264896392822266, "learning_rate": 4.087031950267457e-05, "loss": 4.2432, "step": 214500 }, { "epoch": 15.104678937754672, "grad_norm": 1.0915526151657104, "learning_rate": 4.0725748156715346e-05, "loss": 4.2421, "step": 215000 }, { "epoch": 15.13980609807503, "grad_norm": 1.109480857849121, "learning_rate": 4.058117681075611e-05, "loss": 4.2378, "step": 215500 }, { "epoch": 15.174933258395392, "grad_norm": 1.1790484189987183, "learning_rate": 4.043660546479688e-05, "loss": 4.2431, "step": 216000 }, { "epoch": 15.21006041871575, "grad_norm": 1.1003268957138062, "learning_rate": 4.0292034118837645e-05, "loss": 4.2433, "step": 216500 }, { "epoch": 15.245187579036111, "grad_norm": 1.1183347702026367, "learning_rate": 4.014746277287842e-05, "loss": 4.2411, "step": 217000 }, { "epoch": 15.28031473935647, "grad_norm": 1.0823845863342285, "learning_rate": 4.0002891426919186e-05, "loss": 4.24, "step": 217500 }, { "epoch": 15.31544189967683, "grad_norm": 1.1573807001113892, "learning_rate": 3.985832008095996e-05, "loss": 4.2414, "step": 218000 }, { "epoch": 15.35056905999719, "grad_norm": 1.1436432600021362, "learning_rate": 3.971374873500072e-05, "loss": 4.2407, "step": 218500 }, { "epoch": 15.38569622031755, "grad_norm": 1.1000200510025024, "learning_rate": 3.95691773890415e-05, "loss": 4.2407, "step": 219000 }, { "epoch": 15.42082338063791, "grad_norm": 1.1054102182388306, "learning_rate": 3.942460604308226e-05, "loss": 4.2409, "step": 219500 }, { "epoch": 15.45595054095827, "grad_norm": 1.093224048614502, "learning_rate": 3.928003469712303e-05, "loss": 4.2387, "step": 220000 }, { "epoch": 15.45595054095827, "eval_accuracy": 0.6712017302931698, "eval_loss": 3.9273860454559326, "eval_runtime": 170.7541, "eval_samples_per_second": 706.9, "eval_steps_per_second": 4.422, "step": 220000 }, { "epoch": 15.491077701278629, "grad_norm": 1.1378525495529175, "learning_rate": 3.9135463351163804e-05, "loss": 4.24, "step": 220500 }, { "epoch": 15.526204861598988, "grad_norm": 1.1722012758255005, "learning_rate": 3.899089200520457e-05, "loss": 4.2409, "step": 221000 }, { "epoch": 15.561332021919348, "grad_norm": 1.1050492525100708, "learning_rate": 3.884632065924534e-05, "loss": 4.2383, "step": 221500 }, { "epoch": 15.596459182239707, "grad_norm": 1.156582236289978, "learning_rate": 3.870174931328611e-05, "loss": 4.2387, "step": 222000 }, { "epoch": 15.631586342560068, "grad_norm": 1.0647773742675781, "learning_rate": 3.855717796732688e-05, "loss": 4.2349, "step": 222500 }, { "epoch": 15.666713502880427, "grad_norm": 1.1113219261169434, "learning_rate": 3.8412606621367644e-05, "loss": 4.2374, "step": 223000 }, { "epoch": 15.701840663200787, "grad_norm": 1.137458086013794, "learning_rate": 3.8268035275408415e-05, "loss": 4.2379, "step": 223500 }, { "epoch": 15.736967823521146, "grad_norm": 1.1807042360305786, "learning_rate": 3.812346392944918e-05, "loss": 4.2356, "step": 224000 }, { "epoch": 15.772094983841507, "grad_norm": 1.1037286520004272, "learning_rate": 3.797889258348996e-05, "loss": 4.2323, "step": 224500 }, { "epoch": 15.807222144161866, "grad_norm": 1.1280908584594727, "learning_rate": 3.783432123753072e-05, "loss": 4.2321, "step": 225000 }, { "epoch": 15.842349304482227, "grad_norm": 1.146420955657959, "learning_rate": 3.768974989157149e-05, "loss": 4.2319, "step": 225500 }, { "epoch": 15.877476464802585, "grad_norm": 1.13649582862854, "learning_rate": 3.754517854561226e-05, "loss": 4.2371, "step": 226000 }, { "epoch": 15.912603625122944, "grad_norm": 1.1646398305892944, "learning_rate": 3.740060719965303e-05, "loss": 4.2377, "step": 226500 }, { "epoch": 15.947730785443305, "grad_norm": 1.1516343355178833, "learning_rate": 3.72560358536938e-05, "loss": 4.2348, "step": 227000 }, { "epoch": 15.982857945763664, "grad_norm": 1.1276516914367676, "learning_rate": 3.711146450773457e-05, "loss": 4.2353, "step": 227500 }, { "epoch": 16.017985106084023, "grad_norm": 1.1549897193908691, "learning_rate": 3.696689316177534e-05, "loss": 4.2308, "step": 228000 }, { "epoch": 16.053112266404383, "grad_norm": 1.1366312503814697, "learning_rate": 3.682232181581611e-05, "loss": 4.2286, "step": 228500 }, { "epoch": 16.088239426724744, "grad_norm": 1.1602461338043213, "learning_rate": 3.667775046985687e-05, "loss": 4.2309, "step": 229000 }, { "epoch": 16.123366587045105, "grad_norm": 1.1252471208572388, "learning_rate": 3.6533179123897644e-05, "loss": 4.2313, "step": 229500 }, { "epoch": 16.158493747365462, "grad_norm": 1.1889277696609497, "learning_rate": 3.6388607777938415e-05, "loss": 4.2331, "step": 230000 }, { "epoch": 16.158493747365462, "eval_accuracy": 0.6717989554009071, "eval_loss": 3.9226460456848145, "eval_runtime": 169.808, "eval_samples_per_second": 710.838, "eval_steps_per_second": 4.446, "step": 230000 }, { "epoch": 16.193620907685823, "grad_norm": 1.148395299911499, "learning_rate": 3.6244036431979186e-05, "loss": 4.2269, "step": 230500 }, { "epoch": 16.228748068006183, "grad_norm": 1.094735860824585, "learning_rate": 3.609946508601995e-05, "loss": 4.2316, "step": 231000 }, { "epoch": 16.26387522832654, "grad_norm": 1.13967764377594, "learning_rate": 3.595489374006073e-05, "loss": 4.226, "step": 231500 }, { "epoch": 16.2990023886469, "grad_norm": 1.0968010425567627, "learning_rate": 3.581032239410149e-05, "loss": 4.2305, "step": 232000 }, { "epoch": 16.33412954896726, "grad_norm": 1.1561118364334106, "learning_rate": 3.566575104814226e-05, "loss": 4.2275, "step": 232500 }, { "epoch": 16.369256709287622, "grad_norm": 1.1207199096679688, "learning_rate": 3.5521179702183026e-05, "loss": 4.229, "step": 233000 }, { "epoch": 16.40438386960798, "grad_norm": 1.1718597412109375, "learning_rate": 3.5376608356223804e-05, "loss": 4.2281, "step": 233500 }, { "epoch": 16.43951102992834, "grad_norm": 1.1647950410842896, "learning_rate": 3.523203701026457e-05, "loss": 4.2274, "step": 234000 }, { "epoch": 16.4746381902487, "grad_norm": 1.1305009126663208, "learning_rate": 3.508746566430534e-05, "loss": 4.2297, "step": 234500 }, { "epoch": 16.50976535056906, "grad_norm": 1.2222501039505005, "learning_rate": 3.49428943183461e-05, "loss": 4.2276, "step": 235000 }, { "epoch": 16.54489251088942, "grad_norm": 1.1193674802780151, "learning_rate": 3.479832297238687e-05, "loss": 4.2242, "step": 235500 }, { "epoch": 16.58001967120978, "grad_norm": 1.0854778289794922, "learning_rate": 3.4653751626427644e-05, "loss": 4.2289, "step": 236000 }, { "epoch": 16.61514683153014, "grad_norm": 1.1885193586349487, "learning_rate": 3.4509180280468415e-05, "loss": 4.2267, "step": 236500 }, { "epoch": 16.650273991850497, "grad_norm": 1.1464821100234985, "learning_rate": 3.436460893450918e-05, "loss": 4.2327, "step": 237000 }, { "epoch": 16.685401152170858, "grad_norm": 1.142699122428894, "learning_rate": 3.422003758854995e-05, "loss": 4.2274, "step": 237500 }, { "epoch": 16.72052831249122, "grad_norm": 1.159665822982788, "learning_rate": 3.407546624259072e-05, "loss": 4.2261, "step": 238000 }, { "epoch": 16.75565547281158, "grad_norm": 1.1871224641799927, "learning_rate": 3.393089489663149e-05, "loss": 4.2275, "step": 238500 }, { "epoch": 16.790782633131936, "grad_norm": 1.121037244796753, "learning_rate": 3.378632355067226e-05, "loss": 4.2243, "step": 239000 }, { "epoch": 16.825909793452297, "grad_norm": 1.0763566493988037, "learning_rate": 3.3641752204713026e-05, "loss": 4.225, "step": 239500 }, { "epoch": 16.861036953772658, "grad_norm": 1.108519434928894, "learning_rate": 3.3497180858753797e-05, "loss": 4.2238, "step": 240000 }, { "epoch": 16.861036953772658, "eval_accuracy": 0.6726590955875544, "eval_loss": 3.916060447692871, "eval_runtime": 170.4139, "eval_samples_per_second": 708.311, "eval_steps_per_second": 4.43, "step": 240000 }, { "epoch": 16.89616411409302, "grad_norm": 1.1777600049972534, "learning_rate": 3.335260951279456e-05, "loss": 4.2243, "step": 240500 }, { "epoch": 16.931291274413375, "grad_norm": 1.1039535999298096, "learning_rate": 3.320803816683534e-05, "loss": 4.2264, "step": 241000 }, { "epoch": 16.966418434733736, "grad_norm": 1.175890564918518, "learning_rate": 3.30634668208761e-05, "loss": 4.223, "step": 241500 }, { "epoch": 17.001545595054097, "grad_norm": 1.146806240081787, "learning_rate": 3.291889547491687e-05, "loss": 4.2262, "step": 242000 }, { "epoch": 17.036672755374454, "grad_norm": 1.1392848491668701, "learning_rate": 3.277432412895764e-05, "loss": 4.2245, "step": 242500 }, { "epoch": 17.071799915694815, "grad_norm": 1.0995248556137085, "learning_rate": 3.2629752782998414e-05, "loss": 4.2215, "step": 243000 }, { "epoch": 17.106927076015175, "grad_norm": 1.1238758563995361, "learning_rate": 3.248518143703918e-05, "loss": 4.2218, "step": 243500 }, { "epoch": 17.142054236335536, "grad_norm": 1.144672155380249, "learning_rate": 3.234061009107995e-05, "loss": 4.2203, "step": 244000 }, { "epoch": 17.177181396655893, "grad_norm": 1.173998475074768, "learning_rate": 3.219603874512072e-05, "loss": 4.2178, "step": 244500 }, { "epoch": 17.212308556976254, "grad_norm": 1.145478367805481, "learning_rate": 3.205146739916149e-05, "loss": 4.2219, "step": 245000 }, { "epoch": 17.247435717296614, "grad_norm": 1.1230450868606567, "learning_rate": 3.1906896053202255e-05, "loss": 4.2191, "step": 245500 }, { "epoch": 17.282562877616975, "grad_norm": 1.127146601676941, "learning_rate": 3.1762324707243026e-05, "loss": 4.2239, "step": 246000 }, { "epoch": 17.317690037937332, "grad_norm": 1.0922625064849854, "learning_rate": 3.1617753361283796e-05, "loss": 4.2181, "step": 246500 }, { "epoch": 17.352817198257693, "grad_norm": 1.1170495748519897, "learning_rate": 3.147318201532457e-05, "loss": 4.2171, "step": 247000 }, { "epoch": 17.387944358578054, "grad_norm": 1.1932189464569092, "learning_rate": 3.132861066936533e-05, "loss": 4.2222, "step": 247500 }, { "epoch": 17.42307151889841, "grad_norm": 1.1148924827575684, "learning_rate": 3.11840393234061e-05, "loss": 4.2137, "step": 248000 }, { "epoch": 17.45819867921877, "grad_norm": 1.1644034385681152, "learning_rate": 3.103946797744687e-05, "loss": 4.2172, "step": 248500 }, { "epoch": 17.493325839539132, "grad_norm": 1.193620204925537, "learning_rate": 3.0894896631487643e-05, "loss": 4.2145, "step": 249000 }, { "epoch": 17.528452999859493, "grad_norm": 1.236822247505188, "learning_rate": 3.075032528552841e-05, "loss": 4.2147, "step": 249500 }, { "epoch": 17.56358016017985, "grad_norm": 1.1341967582702637, "learning_rate": 3.060575393956918e-05, "loss": 4.2171, "step": 250000 }, { "epoch": 17.56358016017985, "eval_accuracy": 0.6734997894059586, "eval_loss": 3.9105637073516846, "eval_runtime": 174.9584, "eval_samples_per_second": 689.913, "eval_steps_per_second": 4.315, "step": 250000 }, { "epoch": 17.59870732050021, "grad_norm": 1.1993333101272583, "learning_rate": 3.046118259360995e-05, "loss": 4.2219, "step": 250500 }, { "epoch": 17.63383448082057, "grad_norm": 1.1576712131500244, "learning_rate": 3.0316611247650716e-05, "loss": 4.2142, "step": 251000 }, { "epoch": 17.668961641140932, "grad_norm": 1.1532392501831055, "learning_rate": 3.0172039901691484e-05, "loss": 4.2166, "step": 251500 }, { "epoch": 17.70408880146129, "grad_norm": 1.1168920993804932, "learning_rate": 3.0027468555732258e-05, "loss": 4.2176, "step": 252000 }, { "epoch": 17.73921596178165, "grad_norm": 1.170613408088684, "learning_rate": 2.9882897209773025e-05, "loss": 4.217, "step": 252500 }, { "epoch": 17.77434312210201, "grad_norm": 1.1426963806152344, "learning_rate": 2.9738325863813793e-05, "loss": 4.2146, "step": 253000 }, { "epoch": 17.809470282422367, "grad_norm": 1.1551812887191772, "learning_rate": 2.959375451785456e-05, "loss": 4.2184, "step": 253500 }, { "epoch": 17.844597442742728, "grad_norm": 1.0944799184799194, "learning_rate": 2.9449183171895334e-05, "loss": 4.2135, "step": 254000 }, { "epoch": 17.87972460306309, "grad_norm": 1.1788231134414673, "learning_rate": 2.93046118259361e-05, "loss": 4.2194, "step": 254500 }, { "epoch": 17.91485176338345, "grad_norm": 1.1297837495803833, "learning_rate": 2.916004047997687e-05, "loss": 4.216, "step": 255000 }, { "epoch": 17.949978923703807, "grad_norm": 1.1463844776153564, "learning_rate": 2.9015469134017636e-05, "loss": 4.2166, "step": 255500 }, { "epoch": 17.985106084024167, "grad_norm": 1.165405035018921, "learning_rate": 2.887089778805841e-05, "loss": 4.2143, "step": 256000 }, { "epoch": 18.020233244344528, "grad_norm": 1.1273869276046753, "learning_rate": 2.8726326442099178e-05, "loss": 4.2151, "step": 256500 }, { "epoch": 18.05536040466489, "grad_norm": 1.1423300504684448, "learning_rate": 2.8581755096139945e-05, "loss": 4.2151, "step": 257000 }, { "epoch": 18.090487564985246, "grad_norm": 1.1430673599243164, "learning_rate": 2.843718375018072e-05, "loss": 4.2096, "step": 257500 }, { "epoch": 18.125614725305606, "grad_norm": 1.1666117906570435, "learning_rate": 2.8292612404221487e-05, "loss": 4.2137, "step": 258000 }, { "epoch": 18.160741885625967, "grad_norm": 1.1632168292999268, "learning_rate": 2.8148041058262254e-05, "loss": 4.2133, "step": 258500 }, { "epoch": 18.195869045946324, "grad_norm": 1.17476487159729, "learning_rate": 2.800346971230302e-05, "loss": 4.2126, "step": 259000 }, { "epoch": 18.230996206266685, "grad_norm": 1.14350163936615, "learning_rate": 2.7858898366343792e-05, "loss": 4.2132, "step": 259500 }, { "epoch": 18.266123366587045, "grad_norm": 1.1448757648468018, "learning_rate": 2.7714327020384563e-05, "loss": 4.2098, "step": 260000 }, { "epoch": 18.266123366587045, "eval_accuracy": 0.6740192392662758, "eval_loss": 3.9046037197113037, "eval_runtime": 174.8606, "eval_samples_per_second": 690.298, "eval_steps_per_second": 4.318, "step": 260000 }, { "epoch": 18.301250526907406, "grad_norm": 1.1282014846801758, "learning_rate": 2.756975567442533e-05, "loss": 4.2105, "step": 260500 }, { "epoch": 18.336377687227763, "grad_norm": 1.1444168090820312, "learning_rate": 2.7425184328466098e-05, "loss": 4.2075, "step": 261000 }, { "epoch": 18.371504847548124, "grad_norm": 1.139543890953064, "learning_rate": 2.728061298250687e-05, "loss": 4.2128, "step": 261500 }, { "epoch": 18.406632007868485, "grad_norm": 1.0938020944595337, "learning_rate": 2.7136041636547636e-05, "loss": 4.2105, "step": 262000 }, { "epoch": 18.441759168188845, "grad_norm": 1.12821364402771, "learning_rate": 2.6991470290588407e-05, "loss": 4.2082, "step": 262500 }, { "epoch": 18.476886328509202, "grad_norm": 1.2308045625686646, "learning_rate": 2.6846898944629178e-05, "loss": 4.2088, "step": 263000 }, { "epoch": 18.512013488829563, "grad_norm": 1.1535277366638184, "learning_rate": 2.6702327598669945e-05, "loss": 4.2061, "step": 263500 }, { "epoch": 18.547140649149924, "grad_norm": 1.2038856744766235, "learning_rate": 2.6557756252710713e-05, "loss": 4.2056, "step": 264000 }, { "epoch": 18.58226780947028, "grad_norm": 1.2072213888168335, "learning_rate": 2.641318490675148e-05, "loss": 4.2071, "step": 264500 }, { "epoch": 18.61739496979064, "grad_norm": 1.1688348054885864, "learning_rate": 2.6268613560792254e-05, "loss": 4.2116, "step": 265000 }, { "epoch": 18.652522130111002, "grad_norm": 1.150309443473816, "learning_rate": 2.612404221483302e-05, "loss": 4.2087, "step": 265500 }, { "epoch": 18.687649290431363, "grad_norm": 1.167677402496338, "learning_rate": 2.597947086887379e-05, "loss": 4.2063, "step": 266000 }, { "epoch": 18.72277645075172, "grad_norm": 1.1538523435592651, "learning_rate": 2.5834899522914556e-05, "loss": 4.2083, "step": 266500 }, { "epoch": 18.75790361107208, "grad_norm": 1.1187360286712646, "learning_rate": 2.569032817695533e-05, "loss": 4.2078, "step": 267000 }, { "epoch": 18.79303077139244, "grad_norm": 1.1682405471801758, "learning_rate": 2.5545756830996098e-05, "loss": 4.21, "step": 267500 }, { "epoch": 18.828157931712802, "grad_norm": 1.1744188070297241, "learning_rate": 2.5401185485036865e-05, "loss": 4.2087, "step": 268000 }, { "epoch": 18.86328509203316, "grad_norm": 1.2228838205337524, "learning_rate": 2.5256614139077633e-05, "loss": 4.2074, "step": 268500 }, { "epoch": 18.89841225235352, "grad_norm": 1.1564033031463623, "learning_rate": 2.5112042793118407e-05, "loss": 4.2082, "step": 269000 }, { "epoch": 18.93353941267388, "grad_norm": 1.17214834690094, "learning_rate": 2.4967471447159174e-05, "loss": 4.2073, "step": 269500 }, { "epoch": 18.968666572994238, "grad_norm": 1.1211931705474854, "learning_rate": 2.4822900101199945e-05, "loss": 4.2083, "step": 270000 }, { "epoch": 18.968666572994238, "eval_accuracy": 0.6749243403605761, "eval_loss": 3.9001495838165283, "eval_runtime": 172.4666, "eval_samples_per_second": 699.88, "eval_steps_per_second": 4.378, "step": 270000 }, { "epoch": 19.0037937333146, "grad_norm": 1.1477597951889038, "learning_rate": 2.4678328755240712e-05, "loss": 4.2062, "step": 270500 }, { "epoch": 19.03892089363496, "grad_norm": 1.1509101390838623, "learning_rate": 2.4533757409281483e-05, "loss": 4.2039, "step": 271000 }, { "epoch": 19.07404805395532, "grad_norm": 1.2401434183120728, "learning_rate": 2.438918606332225e-05, "loss": 4.2057, "step": 271500 }, { "epoch": 19.109175214275677, "grad_norm": 1.15226411819458, "learning_rate": 2.424461471736302e-05, "loss": 4.2043, "step": 272000 }, { "epoch": 19.144302374596037, "grad_norm": 1.1487555503845215, "learning_rate": 2.410004337140379e-05, "loss": 4.2043, "step": 272500 }, { "epoch": 19.179429534916398, "grad_norm": 1.1425518989562988, "learning_rate": 2.395547202544456e-05, "loss": 4.2042, "step": 273000 }, { "epoch": 19.21455669523676, "grad_norm": 1.1883976459503174, "learning_rate": 2.3810900679485327e-05, "loss": 4.2007, "step": 273500 }, { "epoch": 19.249683855557116, "grad_norm": 1.154150366783142, "learning_rate": 2.3666329333526098e-05, "loss": 4.2018, "step": 274000 }, { "epoch": 19.284811015877477, "grad_norm": 1.1688642501831055, "learning_rate": 2.3521757987566865e-05, "loss": 4.2022, "step": 274500 }, { "epoch": 19.319938176197837, "grad_norm": 1.1799802780151367, "learning_rate": 2.3377186641607636e-05, "loss": 4.2049, "step": 275000 }, { "epoch": 19.355065336518194, "grad_norm": 1.185250163078308, "learning_rate": 2.3232615295648403e-05, "loss": 4.2054, "step": 275500 }, { "epoch": 19.390192496838555, "grad_norm": 1.1517564058303833, "learning_rate": 2.3088043949689174e-05, "loss": 4.1997, "step": 276000 }, { "epoch": 19.425319657158916, "grad_norm": 1.1448290348052979, "learning_rate": 2.2943472603729945e-05, "loss": 4.2046, "step": 276500 }, { "epoch": 19.460446817479276, "grad_norm": 1.154484510421753, "learning_rate": 2.2798901257770712e-05, "loss": 4.2005, "step": 277000 }, { "epoch": 19.495573977799634, "grad_norm": 1.1633652448654175, "learning_rate": 2.265432991181148e-05, "loss": 4.2043, "step": 277500 }, { "epoch": 19.530701138119994, "grad_norm": 1.1757423877716064, "learning_rate": 2.250975856585225e-05, "loss": 4.1993, "step": 278000 }, { "epoch": 19.565828298440355, "grad_norm": 1.1893761157989502, "learning_rate": 2.2365187219893018e-05, "loss": 4.2035, "step": 278500 }, { "epoch": 19.600955458760716, "grad_norm": 1.1733102798461914, "learning_rate": 2.2220615873933785e-05, "loss": 4.1985, "step": 279000 }, { "epoch": 19.636082619081073, "grad_norm": 1.1549720764160156, "learning_rate": 2.2076044527974556e-05, "loss": 4.2019, "step": 279500 }, { "epoch": 19.671209779401433, "grad_norm": 1.1371748447418213, "learning_rate": 2.1931473182015323e-05, "loss": 4.1991, "step": 280000 }, { "epoch": 19.671209779401433, "eval_accuracy": 0.6758713727890701, "eval_loss": 3.8948512077331543, "eval_runtime": 175.0894, "eval_samples_per_second": 689.396, "eval_steps_per_second": 4.312, "step": 280000 }, { "epoch": 19.706336939721794, "grad_norm": 1.1210827827453613, "learning_rate": 2.1786901836056094e-05, "loss": 4.2014, "step": 280500 }, { "epoch": 19.74146410004215, "grad_norm": 1.1898761987686157, "learning_rate": 2.164233049009686e-05, "loss": 4.202, "step": 281000 }, { "epoch": 19.776591260362512, "grad_norm": 1.1539230346679688, "learning_rate": 2.1497759144137632e-05, "loss": 4.2023, "step": 281500 }, { "epoch": 19.811718420682872, "grad_norm": 1.1996328830718994, "learning_rate": 2.1353187798178403e-05, "loss": 4.1997, "step": 282000 }, { "epoch": 19.846845581003233, "grad_norm": 1.1640974283218384, "learning_rate": 2.120861645221917e-05, "loss": 4.2007, "step": 282500 }, { "epoch": 19.88197274132359, "grad_norm": 1.2094998359680176, "learning_rate": 2.106404510625994e-05, "loss": 4.1994, "step": 283000 }, { "epoch": 19.91709990164395, "grad_norm": 1.1988486051559448, "learning_rate": 2.091947376030071e-05, "loss": 4.2026, "step": 283500 }, { "epoch": 19.95222706196431, "grad_norm": 1.1877211332321167, "learning_rate": 2.077490241434148e-05, "loss": 4.1979, "step": 284000 }, { "epoch": 19.987354222284672, "grad_norm": 1.2073891162872314, "learning_rate": 2.0630331068382247e-05, "loss": 4.1992, "step": 284500 }, { "epoch": 20.02248138260503, "grad_norm": 1.194408655166626, "learning_rate": 2.0485759722423017e-05, "loss": 4.2016, "step": 285000 }, { "epoch": 20.05760854292539, "grad_norm": 1.164766550064087, "learning_rate": 2.0341188376463785e-05, "loss": 4.1971, "step": 285500 }, { "epoch": 20.09273570324575, "grad_norm": 1.1763235330581665, "learning_rate": 2.0196617030504556e-05, "loss": 4.1975, "step": 286000 }, { "epoch": 20.127862863566108, "grad_norm": 1.1697255373001099, "learning_rate": 2.0052045684545323e-05, "loss": 4.1976, "step": 286500 }, { "epoch": 20.16299002388647, "grad_norm": 1.1412349939346313, "learning_rate": 1.9907474338586094e-05, "loss": 4.1962, "step": 287000 }, { "epoch": 20.19811718420683, "grad_norm": 1.1671453714370728, "learning_rate": 1.976290299262686e-05, "loss": 4.1944, "step": 287500 }, { "epoch": 20.23324434452719, "grad_norm": 1.152823567390442, "learning_rate": 1.9618331646667632e-05, "loss": 4.1954, "step": 288000 }, { "epoch": 20.268371504847547, "grad_norm": 1.156847596168518, "learning_rate": 1.9473760300708403e-05, "loss": 4.1934, "step": 288500 }, { "epoch": 20.303498665167908, "grad_norm": 1.177436351776123, "learning_rate": 1.932918895474917e-05, "loss": 4.198, "step": 289000 }, { "epoch": 20.33862582548827, "grad_norm": 1.2419123649597168, "learning_rate": 1.918461760878994e-05, "loss": 4.1968, "step": 289500 }, { "epoch": 20.37375298580863, "grad_norm": 1.2177331447601318, "learning_rate": 1.9040046262830708e-05, "loss": 4.1961, "step": 290000 }, { "epoch": 20.37375298580863, "eval_accuracy": 0.6766310761415485, "eval_loss": 3.8902509212493896, "eval_runtime": 172.8568, "eval_samples_per_second": 698.301, "eval_steps_per_second": 4.368, "step": 290000 }, { "epoch": 20.408880146128986, "grad_norm": 1.1365113258361816, "learning_rate": 1.889547491687148e-05, "loss": 4.1927, "step": 290500 }, { "epoch": 20.444007306449347, "grad_norm": 1.233569860458374, "learning_rate": 1.8750903570912246e-05, "loss": 4.1982, "step": 291000 }, { "epoch": 20.479134466769708, "grad_norm": 1.171462059020996, "learning_rate": 1.8606332224953017e-05, "loss": 4.1931, "step": 291500 }, { "epoch": 20.514261627090065, "grad_norm": 1.1427992582321167, "learning_rate": 1.8461760878993785e-05, "loss": 4.1951, "step": 292000 }, { "epoch": 20.549388787410425, "grad_norm": 1.143660545349121, "learning_rate": 1.8317189533034555e-05, "loss": 4.1915, "step": 292500 }, { "epoch": 20.584515947730786, "grad_norm": 1.175752878189087, "learning_rate": 1.8172618187075323e-05, "loss": 4.196, "step": 293000 }, { "epoch": 20.619643108051147, "grad_norm": 1.1572779417037964, "learning_rate": 1.8028046841116093e-05, "loss": 4.1966, "step": 293500 }, { "epoch": 20.654770268371504, "grad_norm": 1.1428428888320923, "learning_rate": 1.788347549515686e-05, "loss": 4.1918, "step": 294000 }, { "epoch": 20.689897428691864, "grad_norm": 1.1450142860412598, "learning_rate": 1.7738904149197628e-05, "loss": 4.1943, "step": 294500 }, { "epoch": 20.725024589012225, "grad_norm": 1.2521476745605469, "learning_rate": 1.75943328032384e-05, "loss": 4.1939, "step": 295000 }, { "epoch": 20.760151749332586, "grad_norm": 1.1601628065109253, "learning_rate": 1.7449761457279166e-05, "loss": 4.1959, "step": 295500 }, { "epoch": 20.795278909652943, "grad_norm": 1.154974341392517, "learning_rate": 1.7305190111319937e-05, "loss": 4.1912, "step": 296000 }, { "epoch": 20.830406069973304, "grad_norm": 1.1588667631149292, "learning_rate": 1.7160618765360705e-05, "loss": 4.1935, "step": 296500 }, { "epoch": 20.865533230293664, "grad_norm": 1.1577287912368774, "learning_rate": 1.7016047419401475e-05, "loss": 4.1893, "step": 297000 }, { "epoch": 20.90066039061402, "grad_norm": 1.1805323362350464, "learning_rate": 1.6871476073442243e-05, "loss": 4.1938, "step": 297500 }, { "epoch": 20.935787550934382, "grad_norm": 1.1622014045715332, "learning_rate": 1.6726904727483014e-05, "loss": 4.1936, "step": 298000 }, { "epoch": 20.970914711254743, "grad_norm": 1.1905090808868408, "learning_rate": 1.658233338152378e-05, "loss": 4.1928, "step": 298500 }, { "epoch": 21.006041871575103, "grad_norm": 1.232733130455017, "learning_rate": 1.6437762035564552e-05, "loss": 4.1937, "step": 299000 }, { "epoch": 21.04116903189546, "grad_norm": 1.20850670337677, "learning_rate": 1.629319068960532e-05, "loss": 4.1926, "step": 299500 }, { "epoch": 21.07629619221582, "grad_norm": 1.175947666168213, "learning_rate": 1.614861934364609e-05, "loss": 4.1893, "step": 300000 }, { "epoch": 21.07629619221582, "eval_accuracy": 0.6772383458706313, "eval_loss": 3.886443853378296, "eval_runtime": 172.3941, "eval_samples_per_second": 700.175, "eval_steps_per_second": 4.38, "step": 300000 }, { "epoch": 21.111423352536182, "grad_norm": 1.220378041267395, "learning_rate": 1.6004047997686857e-05, "loss": 4.1903, "step": 300500 }, { "epoch": 21.14655051285654, "grad_norm": 1.2184807062149048, "learning_rate": 1.5859476651727628e-05, "loss": 4.1925, "step": 301000 }, { "epoch": 21.1816776731769, "grad_norm": 1.228394627571106, "learning_rate": 1.57149053057684e-05, "loss": 4.1905, "step": 301500 }, { "epoch": 21.21680483349726, "grad_norm": 1.151199460029602, "learning_rate": 1.5570333959809166e-05, "loss": 4.1945, "step": 302000 }, { "epoch": 21.25193199381762, "grad_norm": 1.1613242626190186, "learning_rate": 1.5425762613849937e-05, "loss": 4.1859, "step": 302500 }, { "epoch": 21.287059154137978, "grad_norm": 1.1827828884124756, "learning_rate": 1.5281191267890704e-05, "loss": 4.189, "step": 303000 }, { "epoch": 21.32218631445834, "grad_norm": 1.1835259199142456, "learning_rate": 1.5136619921931475e-05, "loss": 4.1895, "step": 303500 }, { "epoch": 21.3573134747787, "grad_norm": 1.1795170307159424, "learning_rate": 1.4992048575972243e-05, "loss": 4.1865, "step": 304000 }, { "epoch": 21.39244063509906, "grad_norm": 1.1692734956741333, "learning_rate": 1.4847477230013013e-05, "loss": 4.19, "step": 304500 }, { "epoch": 21.427567795419417, "grad_norm": 1.1868877410888672, "learning_rate": 1.470290588405378e-05, "loss": 4.1885, "step": 305000 }, { "epoch": 21.462694955739778, "grad_norm": 1.2531903982162476, "learning_rate": 1.4558334538094551e-05, "loss": 4.1903, "step": 305500 }, { "epoch": 21.49782211606014, "grad_norm": 1.1918978691101074, "learning_rate": 1.4413763192135319e-05, "loss": 4.1888, "step": 306000 }, { "epoch": 21.532949276380496, "grad_norm": 1.2377437353134155, "learning_rate": 1.4269191846176088e-05, "loss": 4.1872, "step": 306500 }, { "epoch": 21.568076436700856, "grad_norm": 1.2057268619537354, "learning_rate": 1.4124620500216857e-05, "loss": 4.1863, "step": 307000 }, { "epoch": 21.603203597021217, "grad_norm": 1.1783188581466675, "learning_rate": 1.3980049154257626e-05, "loss": 4.1883, "step": 307500 }, { "epoch": 21.638330757341578, "grad_norm": 1.205777645111084, "learning_rate": 1.3835477808298397e-05, "loss": 4.194, "step": 308000 }, { "epoch": 21.673457917661935, "grad_norm": 1.2125111818313599, "learning_rate": 1.3690906462339164e-05, "loss": 4.1904, "step": 308500 }, { "epoch": 21.708585077982296, "grad_norm": 1.1315722465515137, "learning_rate": 1.3546335116379935e-05, "loss": 4.1891, "step": 309000 }, { "epoch": 21.743712238302656, "grad_norm": 1.2444932460784912, "learning_rate": 1.3401763770420702e-05, "loss": 4.1868, "step": 309500 }, { "epoch": 21.778839398623017, "grad_norm": 1.1555064916610718, "learning_rate": 1.3257192424461473e-05, "loss": 4.1866, "step": 310000 }, { "epoch": 21.778839398623017, "eval_accuracy": 0.6779464498329975, "eval_loss": 3.8807756900787354, "eval_runtime": 173.1336, "eval_samples_per_second": 697.184, "eval_steps_per_second": 4.361, "step": 310000 }, { "epoch": 21.813966558943374, "grad_norm": 1.2265560626983643, "learning_rate": 1.311262107850224e-05, "loss": 4.1856, "step": 310500 }, { "epoch": 21.849093719263735, "grad_norm": 1.2370774745941162, "learning_rate": 1.2968049732543011e-05, "loss": 4.1836, "step": 311000 }, { "epoch": 21.884220879584095, "grad_norm": 1.1513094902038574, "learning_rate": 1.2823478386583779e-05, "loss": 4.1901, "step": 311500 }, { "epoch": 21.919348039904452, "grad_norm": 1.204253077507019, "learning_rate": 1.267890704062455e-05, "loss": 4.185, "step": 312000 }, { "epoch": 21.954475200224813, "grad_norm": 1.1831001043319702, "learning_rate": 1.2534335694665317e-05, "loss": 4.1858, "step": 312500 }, { "epoch": 21.989602360545174, "grad_norm": 1.2059133052825928, "learning_rate": 1.2389764348706088e-05, "loss": 4.1883, "step": 313000 }, { "epoch": 22.024729520865534, "grad_norm": 1.1907854080200195, "learning_rate": 1.2245193002746857e-05, "loss": 4.1852, "step": 313500 }, { "epoch": 22.05985668118589, "grad_norm": 1.2312885522842407, "learning_rate": 1.2100621656787626e-05, "loss": 4.1829, "step": 314000 }, { "epoch": 22.094983841506252, "grad_norm": 1.1898822784423828, "learning_rate": 1.1956050310828395e-05, "loss": 4.1874, "step": 314500 }, { "epoch": 22.130111001826613, "grad_norm": 1.2186199426651, "learning_rate": 1.1811478964869162e-05, "loss": 4.183, "step": 315000 }, { "epoch": 22.165238162146974, "grad_norm": 1.2078900337219238, "learning_rate": 1.1666907618909931e-05, "loss": 4.1819, "step": 315500 }, { "epoch": 22.20036532246733, "grad_norm": 1.2073997259140015, "learning_rate": 1.15223362729507e-05, "loss": 4.1822, "step": 316000 }, { "epoch": 22.23549248278769, "grad_norm": 1.219303846359253, "learning_rate": 1.137776492699147e-05, "loss": 4.1844, "step": 316500 }, { "epoch": 22.270619643108052, "grad_norm": 1.2415733337402344, "learning_rate": 1.123319358103224e-05, "loss": 4.1847, "step": 317000 }, { "epoch": 22.30574680342841, "grad_norm": 1.1647149324417114, "learning_rate": 1.108862223507301e-05, "loss": 4.1825, "step": 317500 }, { "epoch": 22.34087396374877, "grad_norm": 1.1433378458023071, "learning_rate": 1.0944050889113779e-05, "loss": 4.1837, "step": 318000 }, { "epoch": 22.37600112406913, "grad_norm": 1.1504170894622803, "learning_rate": 1.0799479543154548e-05, "loss": 4.1855, "step": 318500 }, { "epoch": 22.41112828438949, "grad_norm": 1.1801626682281494, "learning_rate": 1.0654908197195317e-05, "loss": 4.1811, "step": 319000 }, { "epoch": 22.44625544470985, "grad_norm": 1.1882909536361694, "learning_rate": 1.0510336851236086e-05, "loss": 4.1822, "step": 319500 }, { "epoch": 22.48138260503021, "grad_norm": 1.2363790273666382, "learning_rate": 1.0365765505276855e-05, "loss": 4.181, "step": 320000 }, { "epoch": 22.48138260503021, "eval_accuracy": 0.6783526657157467, "eval_loss": 3.8781731128692627, "eval_runtime": 176.1055, "eval_samples_per_second": 685.419, "eval_steps_per_second": 4.287, "step": 320000 }, { "epoch": 22.51650976535057, "grad_norm": 1.2691586017608643, "learning_rate": 1.0221194159317624e-05, "loss": 4.1823, "step": 320500 }, { "epoch": 22.55163692567093, "grad_norm": 1.2211499214172363, "learning_rate": 1.0076622813358393e-05, "loss": 4.187, "step": 321000 }, { "epoch": 22.586764085991287, "grad_norm": 1.190129280090332, "learning_rate": 9.932051467399162e-06, "loss": 4.1841, "step": 321500 }, { "epoch": 22.621891246311648, "grad_norm": 1.1977702379226685, "learning_rate": 9.787480121439931e-06, "loss": 4.1801, "step": 322000 }, { "epoch": 22.65701840663201, "grad_norm": 1.1786447763442993, "learning_rate": 9.6429087754807e-06, "loss": 4.184, "step": 322500 }, { "epoch": 22.692145566952366, "grad_norm": 1.199123501777649, "learning_rate": 9.49833742952147e-06, "loss": 4.1827, "step": 323000 }, { "epoch": 22.727272727272727, "grad_norm": 1.1751512289047241, "learning_rate": 9.353766083562238e-06, "loss": 4.1809, "step": 323500 }, { "epoch": 22.762399887593087, "grad_norm": 1.17415189743042, "learning_rate": 9.209194737603008e-06, "loss": 4.1782, "step": 324000 }, { "epoch": 22.797527047913448, "grad_norm": 1.1726998090744019, "learning_rate": 9.064623391643777e-06, "loss": 4.1839, "step": 324500 }, { "epoch": 22.832654208233805, "grad_norm": 1.2341679334640503, "learning_rate": 8.920052045684546e-06, "loss": 4.1831, "step": 325000 }, { "epoch": 22.867781368554166, "grad_norm": 1.1989631652832031, "learning_rate": 8.775480699725315e-06, "loss": 4.1818, "step": 325500 }, { "epoch": 22.902908528874526, "grad_norm": 1.1842224597930908, "learning_rate": 8.630909353766084e-06, "loss": 4.184, "step": 326000 }, { "epoch": 22.938035689194887, "grad_norm": 1.1631402969360352, "learning_rate": 8.486338007806853e-06, "loss": 4.181, "step": 326500 }, { "epoch": 22.973162849515244, "grad_norm": 1.1952308416366577, "learning_rate": 8.341766661847622e-06, "loss": 4.177, "step": 327000 }, { "epoch": 23.008290009835605, "grad_norm": 1.172516107559204, "learning_rate": 8.197195315888391e-06, "loss": 4.1812, "step": 327500 }, { "epoch": 23.043417170155966, "grad_norm": 1.1752017736434937, "learning_rate": 8.05262396992916e-06, "loss": 4.1792, "step": 328000 }, { "epoch": 23.078544330476323, "grad_norm": 1.2063802480697632, "learning_rate": 7.90805262396993e-06, "loss": 4.176, "step": 328500 }, { "epoch": 23.113671490796683, "grad_norm": 1.2511104345321655, "learning_rate": 7.763481278010698e-06, "loss": 4.1802, "step": 329000 }, { "epoch": 23.148798651117044, "grad_norm": 1.218986988067627, "learning_rate": 7.618909932051468e-06, "loss": 4.1833, "step": 329500 }, { "epoch": 23.183925811437405, "grad_norm": 1.1786144971847534, "learning_rate": 7.474338586092237e-06, "loss": 4.18, "step": 330000 }, { "epoch": 23.183925811437405, "eval_accuracy": 0.6784669032674464, "eval_loss": 3.876251697540283, "eval_runtime": 176.2511, "eval_samples_per_second": 684.853, "eval_steps_per_second": 4.284, "step": 330000 }, { "epoch": 23.219052971757762, "grad_norm": 1.1755656003952026, "learning_rate": 7.3297672401330065e-06, "loss": 4.1755, "step": 330500 }, { "epoch": 23.254180132078123, "grad_norm": 1.2084829807281494, "learning_rate": 7.1851958941737756e-06, "loss": 4.1796, "step": 331000 }, { "epoch": 23.289307292398483, "grad_norm": 1.1508269309997559, "learning_rate": 7.040624548214545e-06, "loss": 4.1783, "step": 331500 }, { "epoch": 23.324434452718844, "grad_norm": 1.2164191007614136, "learning_rate": 6.896053202255314e-06, "loss": 4.1836, "step": 332000 }, { "epoch": 23.3595616130392, "grad_norm": 1.1581343412399292, "learning_rate": 6.751481856296083e-06, "loss": 4.1798, "step": 332500 }, { "epoch": 23.39468877335956, "grad_norm": 1.2312132120132446, "learning_rate": 6.606910510336851e-06, "loss": 4.175, "step": 333000 }, { "epoch": 23.429815933679922, "grad_norm": 1.177598237991333, "learning_rate": 6.46233916437762e-06, "loss": 4.1789, "step": 333500 }, { "epoch": 23.46494309400028, "grad_norm": 1.1932919025421143, "learning_rate": 6.317767818418389e-06, "loss": 4.1778, "step": 334000 }, { "epoch": 23.50007025432064, "grad_norm": 1.1783565282821655, "learning_rate": 6.173196472459159e-06, "loss": 4.1799, "step": 334500 }, { "epoch": 23.535197414641, "grad_norm": 1.188306450843811, "learning_rate": 6.028625126499928e-06, "loss": 4.1806, "step": 335000 }, { "epoch": 23.57032457496136, "grad_norm": 1.2344175577163696, "learning_rate": 5.884053780540697e-06, "loss": 4.1768, "step": 335500 }, { "epoch": 23.60545173528172, "grad_norm": 1.2231345176696777, "learning_rate": 5.739482434581466e-06, "loss": 4.1762, "step": 336000 }, { "epoch": 23.64057889560208, "grad_norm": 1.2006125450134277, "learning_rate": 5.5949110886222355e-06, "loss": 4.1732, "step": 336500 }, { "epoch": 23.67570605592244, "grad_norm": 1.2157808542251587, "learning_rate": 5.4503397426630046e-06, "loss": 4.1786, "step": 337000 }, { "epoch": 23.7108332162428, "grad_norm": 1.1912636756896973, "learning_rate": 5.305768396703774e-06, "loss": 4.1725, "step": 337500 }, { "epoch": 23.745960376563158, "grad_norm": 1.1852861642837524, "learning_rate": 5.161197050744543e-06, "loss": 4.1754, "step": 338000 }, { "epoch": 23.78108753688352, "grad_norm": 1.1866953372955322, "learning_rate": 5.016625704785312e-06, "loss": 4.1732, "step": 338500 }, { "epoch": 23.81621469720388, "grad_norm": 1.1770223379135132, "learning_rate": 4.872054358826081e-06, "loss": 4.1774, "step": 339000 }, { "epoch": 23.851341857524236, "grad_norm": 1.1820812225341797, "learning_rate": 4.72748301286685e-06, "loss": 4.1773, "step": 339500 }, { "epoch": 23.886469017844597, "grad_norm": 1.173677921295166, "learning_rate": 4.582911666907619e-06, "loss": 4.1771, "step": 340000 }, { "epoch": 23.886469017844597, "eval_accuracy": 0.6789761727430269, "eval_loss": 3.873131275177002, "eval_runtime": 177.5077, "eval_samples_per_second": 680.004, "eval_steps_per_second": 4.253, "step": 340000 }, { "epoch": 23.921596178164958, "grad_norm": 1.1811652183532715, "learning_rate": 4.438340320948388e-06, "loss": 4.1748, "step": 340500 }, { "epoch": 23.956723338485318, "grad_norm": 1.157711386680603, "learning_rate": 4.293768974989157e-06, "loss": 4.1772, "step": 341000 }, { "epoch": 23.991850498805675, "grad_norm": 1.2078789472579956, "learning_rate": 4.149197629029926e-06, "loss": 4.1761, "step": 341500 }, { "epoch": 24.026977659126036, "grad_norm": 1.1596558094024658, "learning_rate": 4.004626283070695e-06, "loss": 4.1777, "step": 342000 }, { "epoch": 24.062104819446397, "grad_norm": 1.1860870122909546, "learning_rate": 3.8600549371114645e-06, "loss": 4.177, "step": 342500 }, { "epoch": 24.097231979766757, "grad_norm": 1.1694058179855347, "learning_rate": 3.7154835911522336e-06, "loss": 4.1733, "step": 343000 }, { "epoch": 24.132359140087114, "grad_norm": 1.214502215385437, "learning_rate": 3.5709122451930026e-06, "loss": 4.1766, "step": 343500 }, { "epoch": 24.167486300407475, "grad_norm": 1.1541764736175537, "learning_rate": 3.426340899233772e-06, "loss": 4.1786, "step": 344000 }, { "epoch": 24.202613460727836, "grad_norm": 1.1734694242477417, "learning_rate": 3.2817695532745412e-06, "loss": 4.1762, "step": 344500 }, { "epoch": 24.237740621048193, "grad_norm": 1.16045081615448, "learning_rate": 3.1371982073153103e-06, "loss": 4.1794, "step": 345000 }, { "epoch": 24.272867781368554, "grad_norm": 1.1962388753890991, "learning_rate": 2.9926268613560794e-06, "loss": 4.176, "step": 345500 }, { "epoch": 24.307994941688914, "grad_norm": 1.1952061653137207, "learning_rate": 2.8480555153968485e-06, "loss": 4.1739, "step": 346000 }, { "epoch": 24.343122102009275, "grad_norm": 1.176952600479126, "learning_rate": 2.7034841694376176e-06, "loss": 4.1712, "step": 346500 }, { "epoch": 24.378249262329632, "grad_norm": 1.1439900398254395, "learning_rate": 2.5589128234783866e-06, "loss": 4.1751, "step": 347000 }, { "epoch": 24.413376422649993, "grad_norm": 1.2053487300872803, "learning_rate": 2.414341477519156e-06, "loss": 4.1733, "step": 347500 }, { "epoch": 24.448503582970353, "grad_norm": 1.1910380125045776, "learning_rate": 2.269770131559925e-06, "loss": 4.1759, "step": 348000 }, { "epoch": 24.483630743290714, "grad_norm": 1.1746999025344849, "learning_rate": 2.125198785600694e-06, "loss": 4.1723, "step": 348500 }, { "epoch": 24.51875790361107, "grad_norm": 1.1580783128738403, "learning_rate": 1.980627439641463e-06, "loss": 4.1712, "step": 349000 }, { "epoch": 24.553885063931432, "grad_norm": 1.1594668626785278, "learning_rate": 1.8360560936822323e-06, "loss": 4.172, "step": 349500 }, { "epoch": 24.589012224251793, "grad_norm": 1.1586041450500488, "learning_rate": 1.6914847477230013e-06, "loss": 4.1765, "step": 350000 }, { "epoch": 24.589012224251793, "eval_accuracy": 0.6795139107528111, "eval_loss": 3.870356321334839, "eval_runtime": 173.1855, "eval_samples_per_second": 696.975, "eval_steps_per_second": 4.359, "step": 350000 }, { "epoch": 24.62413938457215, "grad_norm": 1.2085744142532349, "learning_rate": 1.5469134017637704e-06, "loss": 4.1728, "step": 350500 }, { "epoch": 24.65926654489251, "grad_norm": 1.1842459440231323, "learning_rate": 1.4023420558045395e-06, "loss": 4.1723, "step": 351000 }, { "epoch": 24.69439370521287, "grad_norm": 1.1857490539550781, "learning_rate": 1.2577707098453088e-06, "loss": 4.1726, "step": 351500 }, { "epoch": 24.72952086553323, "grad_norm": 1.2067129611968994, "learning_rate": 1.1131993638860779e-06, "loss": 4.1739, "step": 352000 }, { "epoch": 24.76464802585359, "grad_norm": 1.1831903457641602, "learning_rate": 9.68628017926847e-07, "loss": 4.1733, "step": 352500 }, { "epoch": 24.79977518617395, "grad_norm": 1.1799302101135254, "learning_rate": 8.240566719676161e-07, "loss": 4.1731, "step": 353000 }, { "epoch": 24.83490234649431, "grad_norm": 1.1894813776016235, "learning_rate": 6.794853260083851e-07, "loss": 4.1729, "step": 353500 }, { "epoch": 24.870029506814667, "grad_norm": 1.1784272193908691, "learning_rate": 5.349139800491543e-07, "loss": 4.1742, "step": 354000 }, { "epoch": 24.905156667135028, "grad_norm": 1.1680693626403809, "learning_rate": 3.903426340899234e-07, "loss": 4.1732, "step": 354500 }, { "epoch": 24.94028382745539, "grad_norm": 1.1994397640228271, "learning_rate": 2.457712881306925e-07, "loss": 4.1689, "step": 355000 }, { "epoch": 24.97541098777575, "grad_norm": 1.157138705253601, "learning_rate": 1.0119994217146162e-07, "loss": 4.1728, "step": 355500 }, { "epoch": 25.0, "step": 355850, "total_flos": 3.039073706855117e+18, "train_loss": 4.565901666523048, "train_runtime": 125705.0232, "train_samples_per_second": 452.914, "train_steps_per_second": 2.831 } ], "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": 3.039073706855117e+18, "train_batch_size": 160, "trial_name": null, "trial_params": null }