{ "best_metric": 4.475473880767822, "best_model_checkpoint": "bert_tiny_lda_50_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": 2.40334153175354, "learning_rate": 5e-06, "loss": 13.4934, "step": 500 }, { "epoch": 0.0702543206407194, "grad_norm": 0.9581671953201294, "learning_rate": 1e-05, "loss": 11.3759, "step": 1000 }, { "epoch": 0.1053814809610791, "grad_norm": 0.6576395034790039, "learning_rate": 1.5e-05, "loss": 10.1407, "step": 1500 }, { "epoch": 0.1405086412814388, "grad_norm": 0.5959604382514954, "learning_rate": 2e-05, "loss": 9.9392, "step": 2000 }, { "epoch": 0.1756358016017985, "grad_norm": 0.6270022392272949, "learning_rate": 2.5e-05, "loss": 9.848, "step": 2500 }, { "epoch": 0.2107629619221582, "grad_norm": 0.7302377223968506, "learning_rate": 3e-05, "loss": 9.7871, "step": 3000 }, { "epoch": 0.2458901222425179, "grad_norm": 0.6443547010421753, "learning_rate": 3.5e-05, "loss": 9.7484, "step": 3500 }, { "epoch": 0.2810172825628776, "grad_norm": 0.6018112301826477, "learning_rate": 4e-05, "loss": 9.7174, "step": 4000 }, { "epoch": 0.31614444288323734, "grad_norm": 0.6648190021514893, "learning_rate": 4.5e-05, "loss": 9.691, "step": 4500 }, { "epoch": 0.351271603203597, "grad_norm": 0.5837637186050415, "learning_rate": 5e-05, "loss": 9.669, "step": 5000 }, { "epoch": 0.38639876352395675, "grad_norm": 0.6552785038948059, "learning_rate": 5.500000000000001e-05, "loss": 9.6451, "step": 5500 }, { "epoch": 0.4215259238443164, "grad_norm": 0.5583786368370056, "learning_rate": 6e-05, "loss": 9.6335, "step": 6000 }, { "epoch": 0.45665308416467615, "grad_norm": 0.5866196751594543, "learning_rate": 6.500000000000001e-05, "loss": 9.6176, "step": 6500 }, { "epoch": 0.4917802444850358, "grad_norm": 0.6038596630096436, "learning_rate": 7e-05, "loss": 9.607, "step": 7000 }, { "epoch": 0.5269074048053956, "grad_norm": 0.6006375551223755, "learning_rate": 7.500000000000001e-05, "loss": 9.5937, "step": 7500 }, { "epoch": 0.5620345651257552, "grad_norm": 0.7146994471549988, "learning_rate": 8e-05, "loss": 9.5818, "step": 8000 }, { "epoch": 0.5971617254461149, "grad_norm": 0.5217270851135254, "learning_rate": 8.5e-05, "loss": 9.5715, "step": 8500 }, { "epoch": 0.6322888857664747, "grad_norm": 0.9519511461257935, "learning_rate": 9e-05, "loss": 9.5605, "step": 9000 }, { "epoch": 0.6674160460868344, "grad_norm": 0.5103440284729004, "learning_rate": 9.5e-05, "loss": 9.5559, "step": 9500 }, { "epoch": 0.702543206407194, "grad_norm": 0.5671107172966003, "learning_rate": 0.0001, "loss": 9.5487, "step": 10000 }, { "epoch": 0.702543206407194, "eval_accuracy": 0.16431585271721819, "eval_loss": 9.386157989501953, "eval_runtime": 124.5623, "eval_samples_per_second": 969.041, "eval_steps_per_second": 6.061, "step": 10000 }, { "epoch": 0.7376703667275537, "grad_norm": 0.5918739438056946, "learning_rate": 9.985542865404077e-05, "loss": 9.5338, "step": 10500 }, { "epoch": 0.7727975270479135, "grad_norm": 0.6293185949325562, "learning_rate": 9.971085730808155e-05, "loss": 9.5028, "step": 11000 }, { "epoch": 0.8079246873682732, "grad_norm": 0.5953327417373657, "learning_rate": 9.956628596212231e-05, "loss": 9.4615, "step": 11500 }, { "epoch": 0.8430518476886328, "grad_norm": 0.7317783832550049, "learning_rate": 9.942171461616309e-05, "loss": 9.3984, "step": 12000 }, { "epoch": 0.8781790080089925, "grad_norm": 0.9626119136810303, "learning_rate": 9.927714327020385e-05, "loss": 9.2707, "step": 12500 }, { "epoch": 0.9133061683293523, "grad_norm": 1.1582441329956055, "learning_rate": 9.913257192424462e-05, "loss": 9.1004, "step": 13000 }, { "epoch": 0.948433328649712, "grad_norm": 1.7291252613067627, "learning_rate": 9.89880005782854e-05, "loss": 8.8314, "step": 13500 }, { "epoch": 0.9835604889700716, "grad_norm": 2.173393964767456, "learning_rate": 9.884342923232616e-05, "loss": 8.3401, "step": 14000 }, { "epoch": 1.0186876492904313, "grad_norm": 3.0057594776153564, "learning_rate": 9.869885788636692e-05, "loss": 8.0003, "step": 14500 }, { "epoch": 1.053814809610791, "grad_norm": 2.066540479660034, "learning_rate": 9.855428654040769e-05, "loss": 7.7386, "step": 15000 }, { "epoch": 1.0889419699311507, "grad_norm": 2.1406211853027344, "learning_rate": 9.840971519444846e-05, "loss": 7.4399, "step": 15500 }, { "epoch": 1.1240691302515105, "grad_norm": 1.7433831691741943, "learning_rate": 9.826514384848924e-05, "loss": 7.2149, "step": 16000 }, { "epoch": 1.1591962905718702, "grad_norm": 1.8667043447494507, "learning_rate": 9.812057250253e-05, "loss": 7.0522, "step": 16500 }, { "epoch": 1.1943234508922298, "grad_norm": 1.668195366859436, "learning_rate": 9.797600115657077e-05, "loss": 6.9198, "step": 17000 }, { "epoch": 1.2294506112125896, "grad_norm": 1.7096377611160278, "learning_rate": 9.783142981061155e-05, "loss": 6.807, "step": 17500 }, { "epoch": 1.2645777715329491, "grad_norm": 1.5724433660507202, "learning_rate": 9.768685846465231e-05, "loss": 6.7254, "step": 18000 }, { "epoch": 1.299704931853309, "grad_norm": 1.721472978591919, "learning_rate": 9.754228711869307e-05, "loss": 6.6404, "step": 18500 }, { "epoch": 1.3348320921736687, "grad_norm": 1.5330699682235718, "learning_rate": 9.739771577273384e-05, "loss": 6.5817, "step": 19000 }, { "epoch": 1.3699592524940285, "grad_norm": 1.552282452583313, "learning_rate": 9.725314442677463e-05, "loss": 6.5258, "step": 19500 }, { "epoch": 1.405086412814388, "grad_norm": 1.527437448501587, "learning_rate": 9.71085730808154e-05, "loss": 6.4629, "step": 20000 }, { "epoch": 1.405086412814388, "eval_accuracy": 0.4932245270787259, "eval_loss": 5.999148368835449, "eval_runtime": 125.0428, "eval_samples_per_second": 965.318, "eval_steps_per_second": 6.038, "step": 20000 }, { "epoch": 1.4402135731347478, "grad_norm": 1.5886305570602417, "learning_rate": 9.696400173485616e-05, "loss": 6.4097, "step": 20500 }, { "epoch": 1.4753407334551074, "grad_norm": 1.415772557258606, "learning_rate": 9.681943038889692e-05, "loss": 6.3632, "step": 21000 }, { "epoch": 1.5104678937754672, "grad_norm": 1.4513144493103027, "learning_rate": 9.66748590429377e-05, "loss": 6.3156, "step": 21500 }, { "epoch": 1.545595054095827, "grad_norm": 1.352752923965454, "learning_rate": 9.653028769697846e-05, "loss": 6.2674, "step": 22000 }, { "epoch": 1.5807222144161865, "grad_norm": 1.3439993858337402, "learning_rate": 9.638571635101923e-05, "loss": 6.2332, "step": 22500 }, { "epoch": 1.6158493747365463, "grad_norm": 1.3796255588531494, "learning_rate": 9.624114500506e-05, "loss": 6.1972, "step": 23000 }, { "epoch": 1.650976535056906, "grad_norm": 1.277636170387268, "learning_rate": 9.609657365910077e-05, "loss": 6.1648, "step": 23500 }, { "epoch": 1.6861036953772657, "grad_norm": 1.309707760810852, "learning_rate": 9.595200231314155e-05, "loss": 6.1311, "step": 24000 }, { "epoch": 1.7212308556976255, "grad_norm": 1.3859153985977173, "learning_rate": 9.580743096718231e-05, "loss": 6.1035, "step": 24500 }, { "epoch": 1.7563580160179852, "grad_norm": 1.31857430934906, "learning_rate": 9.566285962122307e-05, "loss": 6.0761, "step": 25000 }, { "epoch": 1.7914851763383448, "grad_norm": 1.3341494798660278, "learning_rate": 9.551828827526385e-05, "loss": 6.0495, "step": 25500 }, { "epoch": 1.8266123366587044, "grad_norm": 1.2407506704330444, "learning_rate": 9.537371692930462e-05, "loss": 6.0213, "step": 26000 }, { "epoch": 1.8617394969790642, "grad_norm": 1.2589643001556396, "learning_rate": 9.522914558334538e-05, "loss": 5.9932, "step": 26500 }, { "epoch": 1.896866657299424, "grad_norm": 1.2918974161148071, "learning_rate": 9.508457423738616e-05, "loss": 5.9701, "step": 27000 }, { "epoch": 1.9319938176197837, "grad_norm": 1.3162486553192139, "learning_rate": 9.494000289142692e-05, "loss": 5.9439, "step": 27500 }, { "epoch": 1.9671209779401433, "grad_norm": 1.336181640625, "learning_rate": 9.47954315454677e-05, "loss": 5.9207, "step": 28000 }, { "epoch": 2.002248138260503, "grad_norm": 1.2838622331619263, "learning_rate": 9.465086019950846e-05, "loss": 5.9056, "step": 28500 }, { "epoch": 2.0373752985808626, "grad_norm": 1.3492305278778076, "learning_rate": 9.450628885354923e-05, "loss": 5.8753, "step": 29000 }, { "epoch": 2.0725024589012224, "grad_norm": 1.2521207332611084, "learning_rate": 9.436171750758999e-05, "loss": 5.8571, "step": 29500 }, { "epoch": 2.107629619221582, "grad_norm": 1.2759764194488525, "learning_rate": 9.421714616163077e-05, "loss": 5.845, "step": 30000 }, { "epoch": 2.107629619221582, "eval_accuracy": 0.5586971077379357, "eval_loss": 5.438508987426758, "eval_runtime": 124.1856, "eval_samples_per_second": 971.981, "eval_steps_per_second": 6.08, "step": 30000 }, { "epoch": 2.142756779541942, "grad_norm": 1.2237348556518555, "learning_rate": 9.407257481567155e-05, "loss": 5.8244, "step": 30500 }, { "epoch": 2.1778839398623013, "grad_norm": 1.2119154930114746, "learning_rate": 9.392800346971231e-05, "loss": 5.8055, "step": 31000 }, { "epoch": 2.213011100182661, "grad_norm": 1.2271583080291748, "learning_rate": 9.378343212375307e-05, "loss": 5.7932, "step": 31500 }, { "epoch": 2.248138260503021, "grad_norm": 1.226678490638733, "learning_rate": 9.363886077779385e-05, "loss": 5.7755, "step": 32000 }, { "epoch": 2.2832654208233807, "grad_norm": 1.1767781972885132, "learning_rate": 9.349428943183462e-05, "loss": 5.7616, "step": 32500 }, { "epoch": 2.3183925811437405, "grad_norm": 1.2638683319091797, "learning_rate": 9.334971808587538e-05, "loss": 5.7425, "step": 33000 }, { "epoch": 2.3535197414641003, "grad_norm": 1.2280012369155884, "learning_rate": 9.320514673991614e-05, "loss": 5.7277, "step": 33500 }, { "epoch": 2.3886469017844596, "grad_norm": 1.2539173364639282, "learning_rate": 9.306057539395692e-05, "loss": 5.7162, "step": 34000 }, { "epoch": 2.4237740621048194, "grad_norm": 1.2175486087799072, "learning_rate": 9.29160040479977e-05, "loss": 5.6991, "step": 34500 }, { "epoch": 2.458901222425179, "grad_norm": 1.3080875873565674, "learning_rate": 9.277143270203846e-05, "loss": 5.6867, "step": 35000 }, { "epoch": 2.494028382745539, "grad_norm": 1.1766318082809448, "learning_rate": 9.262686135607923e-05, "loss": 5.673, "step": 35500 }, { "epoch": 2.5291555430658983, "grad_norm": 1.175809383392334, "learning_rate": 9.248229001012e-05, "loss": 5.6597, "step": 36000 }, { "epoch": 2.564282703386258, "grad_norm": 1.2161694765090942, "learning_rate": 9.233771866416077e-05, "loss": 5.646, "step": 36500 }, { "epoch": 2.599409863706618, "grad_norm": 1.2190661430358887, "learning_rate": 9.219314731820153e-05, "loss": 5.6334, "step": 37000 }, { "epoch": 2.6345370240269776, "grad_norm": 1.2153582572937012, "learning_rate": 9.20485759722423e-05, "loss": 5.6221, "step": 37500 }, { "epoch": 2.6696641843473374, "grad_norm": 1.2843390703201294, "learning_rate": 9.190400462628307e-05, "loss": 5.6106, "step": 38000 }, { "epoch": 2.704791344667697, "grad_norm": 1.147096872329712, "learning_rate": 9.175943328032385e-05, "loss": 5.5993, "step": 38500 }, { "epoch": 2.739918504988057, "grad_norm": 1.1578583717346191, "learning_rate": 9.161486193436461e-05, "loss": 5.5856, "step": 39000 }, { "epoch": 2.7750456653084163, "grad_norm": 1.215349793434143, "learning_rate": 9.147029058840538e-05, "loss": 5.5779, "step": 39500 }, { "epoch": 2.810172825628776, "grad_norm": 1.2331188917160034, "learning_rate": 9.132571924244616e-05, "loss": 5.564, "step": 40000 }, { "epoch": 2.810172825628776, "eval_accuracy": 0.5911347016336529, "eval_loss": 5.186133861541748, "eval_runtime": 124.8191, "eval_samples_per_second": 967.047, "eval_steps_per_second": 6.049, "step": 40000 }, { "epoch": 2.845299985949136, "grad_norm": 1.2439841032028198, "learning_rate": 9.118114789648692e-05, "loss": 5.55, "step": 40500 }, { "epoch": 2.8804271462694957, "grad_norm": 1.2326741218566895, "learning_rate": 9.103657655052768e-05, "loss": 5.5416, "step": 41000 }, { "epoch": 2.915554306589855, "grad_norm": 1.1872031688690186, "learning_rate": 9.089200520456845e-05, "loss": 5.5325, "step": 41500 }, { "epoch": 2.950681466910215, "grad_norm": 1.2260159254074097, "learning_rate": 9.074743385860923e-05, "loss": 5.5263, "step": 42000 }, { "epoch": 2.9858086272305746, "grad_norm": 1.231950283050537, "learning_rate": 9.060286251265e-05, "loss": 5.5162, "step": 42500 }, { "epoch": 3.0209357875509344, "grad_norm": 1.2354462146759033, "learning_rate": 9.045829116669077e-05, "loss": 5.4962, "step": 43000 }, { "epoch": 3.056062947871294, "grad_norm": 1.1429723501205444, "learning_rate": 9.031371982073153e-05, "loss": 5.4938, "step": 43500 }, { "epoch": 3.091190108191654, "grad_norm": 1.156050443649292, "learning_rate": 9.016914847477231e-05, "loss": 5.4874, "step": 44000 }, { "epoch": 3.1263172685120133, "grad_norm": 1.219225287437439, "learning_rate": 9.002457712881307e-05, "loss": 5.4773, "step": 44500 }, { "epoch": 3.161444428832373, "grad_norm": 1.2014859914779663, "learning_rate": 8.988000578285384e-05, "loss": 5.4706, "step": 45000 }, { "epoch": 3.196571589152733, "grad_norm": 1.1764975786209106, "learning_rate": 8.973543443689461e-05, "loss": 5.4602, "step": 45500 }, { "epoch": 3.2316987494730927, "grad_norm": 1.1808688640594482, "learning_rate": 8.959086309093538e-05, "loss": 5.4557, "step": 46000 }, { "epoch": 3.2668259097934524, "grad_norm": 1.1370998620986938, "learning_rate": 8.944629174497616e-05, "loss": 5.4478, "step": 46500 }, { "epoch": 3.301953070113812, "grad_norm": 1.1912189722061157, "learning_rate": 8.930172039901692e-05, "loss": 5.4389, "step": 47000 }, { "epoch": 3.3370802304341716, "grad_norm": 1.1792778968811035, "learning_rate": 8.915714905305768e-05, "loss": 5.4314, "step": 47500 }, { "epoch": 3.3722073907545314, "grad_norm": 1.1954056024551392, "learning_rate": 8.901257770709845e-05, "loss": 5.4261, "step": 48000 }, { "epoch": 3.407334551074891, "grad_norm": 1.1304168701171875, "learning_rate": 8.886800636113923e-05, "loss": 5.4172, "step": 48500 }, { "epoch": 3.442461711395251, "grad_norm": 1.199284315109253, "learning_rate": 8.872343501518e-05, "loss": 5.4098, "step": 49000 }, { "epoch": 3.4775888717156107, "grad_norm": 1.169266939163208, "learning_rate": 8.857886366922077e-05, "loss": 5.4032, "step": 49500 }, { "epoch": 3.51271603203597, "grad_norm": 1.151167631149292, "learning_rate": 8.843429232326153e-05, "loss": 5.4009, "step": 50000 }, { "epoch": 3.51271603203597, "eval_accuracy": 0.6106977696928932, "eval_loss": 5.036365509033203, "eval_runtime": 125.4036, "eval_samples_per_second": 962.54, "eval_steps_per_second": 6.021, "step": 50000 }, { "epoch": 3.54784319235633, "grad_norm": 1.2317692041397095, "learning_rate": 8.828972097730231e-05, "loss": 5.3944, "step": 50500 }, { "epoch": 3.5829703526766896, "grad_norm": 1.2115331888198853, "learning_rate": 8.814514963134307e-05, "loss": 5.3848, "step": 51000 }, { "epoch": 3.6180975129970494, "grad_norm": 1.1387019157409668, "learning_rate": 8.800057828538384e-05, "loss": 5.3839, "step": 51500 }, { "epoch": 3.6532246733174087, "grad_norm": 1.2434635162353516, "learning_rate": 8.78560069394246e-05, "loss": 5.3758, "step": 52000 }, { "epoch": 3.6883518336377685, "grad_norm": 1.2112048864364624, "learning_rate": 8.771143559346538e-05, "loss": 5.3677, "step": 52500 }, { "epoch": 3.7234789939581283, "grad_norm": 1.1720407009124756, "learning_rate": 8.756686424750616e-05, "loss": 5.3654, "step": 53000 }, { "epoch": 3.758606154278488, "grad_norm": 1.2057770490646362, "learning_rate": 8.742229290154692e-05, "loss": 5.3604, "step": 53500 }, { "epoch": 3.793733314598848, "grad_norm": 1.1396489143371582, "learning_rate": 8.727772155558768e-05, "loss": 5.3494, "step": 54000 }, { "epoch": 3.8288604749192077, "grad_norm": 1.1905720233917236, "learning_rate": 8.713315020962846e-05, "loss": 5.35, "step": 54500 }, { "epoch": 3.8639876352395675, "grad_norm": 1.1290558576583862, "learning_rate": 8.698857886366923e-05, "loss": 5.3396, "step": 55000 }, { "epoch": 3.899114795559927, "grad_norm": 1.1912096738815308, "learning_rate": 8.684400751770999e-05, "loss": 5.3341, "step": 55500 }, { "epoch": 3.9342419558802866, "grad_norm": 1.08969247341156, "learning_rate": 8.669943617175075e-05, "loss": 5.3302, "step": 56000 }, { "epoch": 3.9693691162006464, "grad_norm": 1.213199257850647, "learning_rate": 8.655486482579153e-05, "loss": 5.3272, "step": 56500 }, { "epoch": 4.004496276521006, "grad_norm": 1.1836055517196655, "learning_rate": 8.641029347983231e-05, "loss": 5.3178, "step": 57000 }, { "epoch": 4.0396234368413655, "grad_norm": 1.1996716260910034, "learning_rate": 8.626572213387307e-05, "loss": 5.3077, "step": 57500 }, { "epoch": 4.074750597161725, "grad_norm": 1.167160153388977, "learning_rate": 8.612115078791384e-05, "loss": 5.3063, "step": 58000 }, { "epoch": 4.109877757482085, "grad_norm": 1.1981871128082275, "learning_rate": 8.597657944195461e-05, "loss": 5.3014, "step": 58500 }, { "epoch": 4.145004917802445, "grad_norm": 1.1067780256271362, "learning_rate": 8.583200809599538e-05, "loss": 5.2962, "step": 59000 }, { "epoch": 4.180132078122805, "grad_norm": 1.1463693380355835, "learning_rate": 8.568743675003614e-05, "loss": 5.2952, "step": 59500 }, { "epoch": 4.215259238443164, "grad_norm": 1.1945289373397827, "learning_rate": 8.55428654040769e-05, "loss": 5.2879, "step": 60000 }, { "epoch": 4.215259238443164, "eval_accuracy": 0.6220534115425067, "eval_loss": 4.9377336502075195, "eval_runtime": 125.3777, "eval_samples_per_second": 962.739, "eval_steps_per_second": 6.022, "step": 60000 }, { "epoch": 4.250386398763524, "grad_norm": 1.2011545896530151, "learning_rate": 8.539829405811768e-05, "loss": 5.2879, "step": 60500 }, { "epoch": 4.285513559083884, "grad_norm": 1.1138263940811157, "learning_rate": 8.525372271215846e-05, "loss": 5.2791, "step": 61000 }, { "epoch": 4.320640719404244, "grad_norm": 1.1305272579193115, "learning_rate": 8.510915136619923e-05, "loss": 5.2726, "step": 61500 }, { "epoch": 4.355767879724603, "grad_norm": 1.1887916326522827, "learning_rate": 8.496458002023999e-05, "loss": 5.2701, "step": 62000 }, { "epoch": 4.3908950400449624, "grad_norm": 1.1760022640228271, "learning_rate": 8.482000867428075e-05, "loss": 5.2705, "step": 62500 }, { "epoch": 4.426022200365322, "grad_norm": 1.2318981885910034, "learning_rate": 8.467543732832153e-05, "loss": 5.2659, "step": 63000 }, { "epoch": 4.461149360685682, "grad_norm": 1.2214834690093994, "learning_rate": 8.45308659823623e-05, "loss": 5.2574, "step": 63500 }, { "epoch": 4.496276521006042, "grad_norm": 1.16779625415802, "learning_rate": 8.438629463640307e-05, "loss": 5.2546, "step": 64000 }, { "epoch": 4.531403681326402, "grad_norm": 1.2083040475845337, "learning_rate": 8.424172329044384e-05, "loss": 5.2484, "step": 64500 }, { "epoch": 4.566530841646761, "grad_norm": 1.1498740911483765, "learning_rate": 8.409715194448461e-05, "loss": 5.2444, "step": 65000 }, { "epoch": 4.601658001967121, "grad_norm": 1.1557725667953491, "learning_rate": 8.395258059852538e-05, "loss": 5.2407, "step": 65500 }, { "epoch": 4.636785162287481, "grad_norm": 1.2024227380752563, "learning_rate": 8.380800925256614e-05, "loss": 5.2348, "step": 66000 }, { "epoch": 4.671912322607841, "grad_norm": 1.1286647319793701, "learning_rate": 8.36634379066069e-05, "loss": 5.234, "step": 66500 }, { "epoch": 4.7070394829282005, "grad_norm": 1.170233130455017, "learning_rate": 8.351886656064768e-05, "loss": 5.226, "step": 67000 }, { "epoch": 4.742166643248559, "grad_norm": 1.29181969165802, "learning_rate": 8.337429521468846e-05, "loss": 5.2242, "step": 67500 }, { "epoch": 4.777293803568919, "grad_norm": 1.2092469930648804, "learning_rate": 8.322972386872923e-05, "loss": 5.2207, "step": 68000 }, { "epoch": 4.812420963889279, "grad_norm": 1.2143335342407227, "learning_rate": 8.308515252276999e-05, "loss": 5.2183, "step": 68500 }, { "epoch": 4.847548124209639, "grad_norm": 1.159454345703125, "learning_rate": 8.294058117681077e-05, "loss": 5.2092, "step": 69000 }, { "epoch": 4.8826752845299985, "grad_norm": 1.3042088747024536, "learning_rate": 8.279600983085153e-05, "loss": 5.2089, "step": 69500 }, { "epoch": 4.917802444850358, "grad_norm": 1.2692782878875732, "learning_rate": 8.26514384848923e-05, "loss": 5.2055, "step": 70000 }, { "epoch": 4.917802444850358, "eval_accuracy": 0.6305966975078529, "eval_loss": 4.848171234130859, "eval_runtime": 125.6729, "eval_samples_per_second": 960.478, "eval_steps_per_second": 6.008, "step": 70000 }, { "epoch": 4.952929605170718, "grad_norm": 1.1479058265686035, "learning_rate": 8.250686713893306e-05, "loss": 5.2027, "step": 70500 }, { "epoch": 4.988056765491078, "grad_norm": 1.2258899211883545, "learning_rate": 8.236229579297384e-05, "loss": 5.196, "step": 71000 }, { "epoch": 5.023183925811438, "grad_norm": 1.182870626449585, "learning_rate": 8.221772444701461e-05, "loss": 5.1946, "step": 71500 }, { "epoch": 5.0583110861317975, "grad_norm": 1.2557169198989868, "learning_rate": 8.207315310105538e-05, "loss": 5.1844, "step": 72000 }, { "epoch": 5.093438246452157, "grad_norm": 1.2197999954223633, "learning_rate": 8.192858175509614e-05, "loss": 5.1802, "step": 72500 }, { "epoch": 5.128565406772516, "grad_norm": 1.1851375102996826, "learning_rate": 8.178401040913692e-05, "loss": 5.1774, "step": 73000 }, { "epoch": 5.163692567092876, "grad_norm": 1.2506704330444336, "learning_rate": 8.163943906317768e-05, "loss": 5.1752, "step": 73500 }, { "epoch": 5.198819727413236, "grad_norm": 1.2497591972351074, "learning_rate": 8.149486771721845e-05, "loss": 5.1693, "step": 74000 }, { "epoch": 5.2339468877335955, "grad_norm": 1.2156429290771484, "learning_rate": 8.135029637125921e-05, "loss": 5.173, "step": 74500 }, { "epoch": 5.269074048053955, "grad_norm": 1.1860578060150146, "learning_rate": 8.120572502529999e-05, "loss": 5.1676, "step": 75000 }, { "epoch": 5.304201208374315, "grad_norm": 1.195095419883728, "learning_rate": 8.106115367934077e-05, "loss": 5.1644, "step": 75500 }, { "epoch": 5.339328368694675, "grad_norm": 1.236399531364441, "learning_rate": 8.091658233338153e-05, "loss": 5.1609, "step": 76000 }, { "epoch": 5.374455529015035, "grad_norm": 1.1603856086730957, "learning_rate": 8.07720109874223e-05, "loss": 5.1573, "step": 76500 }, { "epoch": 5.409582689335394, "grad_norm": 1.2362587451934814, "learning_rate": 8.062743964146306e-05, "loss": 5.153, "step": 77000 }, { "epoch": 5.444709849655754, "grad_norm": 1.2191047668457031, "learning_rate": 8.048286829550384e-05, "loss": 5.1474, "step": 77500 }, { "epoch": 5.479837009976113, "grad_norm": 1.2990015745162964, "learning_rate": 8.03382969495446e-05, "loss": 5.148, "step": 78000 }, { "epoch": 5.514964170296473, "grad_norm": 1.244593858718872, "learning_rate": 8.019372560358536e-05, "loss": 5.1456, "step": 78500 }, { "epoch": 5.550091330616833, "grad_norm": 1.2601698637008667, "learning_rate": 8.004915425762614e-05, "loss": 5.1423, "step": 79000 }, { "epoch": 5.5852184909371925, "grad_norm": 1.245548963546753, "learning_rate": 7.990458291166692e-05, "loss": 5.141, "step": 79500 }, { "epoch": 5.620345651257552, "grad_norm": 1.22415292263031, "learning_rate": 7.976001156570768e-05, "loss": 5.1368, "step": 80000 }, { "epoch": 5.620345651257552, "eval_accuracy": 0.63645054842254, "eval_loss": 4.786564350128174, "eval_runtime": 125.4195, "eval_samples_per_second": 962.418, "eval_steps_per_second": 6.02, "step": 80000 }, { "epoch": 5.655472811577912, "grad_norm": 1.2277495861053467, "learning_rate": 7.961544021974845e-05, "loss": 5.1324, "step": 80500 }, { "epoch": 5.690599971898272, "grad_norm": 1.1786996126174927, "learning_rate": 7.947086887378921e-05, "loss": 5.1379, "step": 81000 }, { "epoch": 5.725727132218632, "grad_norm": 1.2532355785369873, "learning_rate": 7.932629752782999e-05, "loss": 5.1321, "step": 81500 }, { "epoch": 5.760854292538991, "grad_norm": 1.3051321506500244, "learning_rate": 7.918172618187075e-05, "loss": 5.1261, "step": 82000 }, { "epoch": 5.795981452859351, "grad_norm": 1.1728847026824951, "learning_rate": 7.903715483591153e-05, "loss": 5.1279, "step": 82500 }, { "epoch": 5.83110861317971, "grad_norm": 1.1523922681808472, "learning_rate": 7.88925834899523e-05, "loss": 5.1246, "step": 83000 }, { "epoch": 5.866235773500071, "grad_norm": 1.3072059154510498, "learning_rate": 7.874801214399307e-05, "loss": 5.1203, "step": 83500 }, { "epoch": 5.90136293382043, "grad_norm": 1.197030782699585, "learning_rate": 7.860344079803384e-05, "loss": 5.119, "step": 84000 }, { "epoch": 5.936490094140789, "grad_norm": 1.2470386028289795, "learning_rate": 7.84588694520746e-05, "loss": 5.1162, "step": 84500 }, { "epoch": 5.971617254461149, "grad_norm": 1.2944002151489258, "learning_rate": 7.831429810611536e-05, "loss": 5.114, "step": 85000 }, { "epoch": 6.006744414781509, "grad_norm": 1.2087780237197876, "learning_rate": 7.816972676015614e-05, "loss": 5.1122, "step": 85500 }, { "epoch": 6.041871575101869, "grad_norm": 1.2677130699157715, "learning_rate": 7.802515541419692e-05, "loss": 5.1059, "step": 86000 }, { "epoch": 6.076998735422229, "grad_norm": 1.1628944873809814, "learning_rate": 7.788058406823768e-05, "loss": 5.1059, "step": 86500 }, { "epoch": 6.112125895742588, "grad_norm": 1.2843151092529297, "learning_rate": 7.773601272227845e-05, "loss": 5.0998, "step": 87000 }, { "epoch": 6.147253056062948, "grad_norm": 1.1976428031921387, "learning_rate": 7.759144137631922e-05, "loss": 5.0986, "step": 87500 }, { "epoch": 6.182380216383308, "grad_norm": 1.2090280055999756, "learning_rate": 7.744687003035999e-05, "loss": 5.098, "step": 88000 }, { "epoch": 6.217507376703667, "grad_norm": 1.2936923503875732, "learning_rate": 7.730229868440075e-05, "loss": 5.0963, "step": 88500 }, { "epoch": 6.252634537024027, "grad_norm": 1.2826613187789917, "learning_rate": 7.715772733844152e-05, "loss": 5.0923, "step": 89000 }, { "epoch": 6.287761697344386, "grad_norm": 1.2612204551696777, "learning_rate": 7.70131559924823e-05, "loss": 5.093, "step": 89500 }, { "epoch": 6.322888857664746, "grad_norm": 1.181904911994934, "learning_rate": 7.686858464652307e-05, "loss": 5.0913, "step": 90000 }, { "epoch": 6.322888857664746, "eval_accuracy": 0.6413607271882584, "eval_loss": 4.74521017074585, "eval_runtime": 125.572, "eval_samples_per_second": 961.25, "eval_steps_per_second": 6.012, "step": 90000 }, { "epoch": 6.358016017985106, "grad_norm": 1.196683645248413, "learning_rate": 7.672401330056384e-05, "loss": 5.0892, "step": 90500 }, { "epoch": 6.393143178305466, "grad_norm": 1.2156459093093872, "learning_rate": 7.65794419546046e-05, "loss": 5.0856, "step": 91000 }, { "epoch": 6.4282703386258255, "grad_norm": 1.2786281108856201, "learning_rate": 7.643487060864536e-05, "loss": 5.0837, "step": 91500 }, { "epoch": 6.463397498946185, "grad_norm": 1.2408231496810913, "learning_rate": 7.629029926268614e-05, "loss": 5.0823, "step": 92000 }, { "epoch": 6.498524659266545, "grad_norm": 1.1885371208190918, "learning_rate": 7.61457279167269e-05, "loss": 5.0816, "step": 92500 }, { "epoch": 6.533651819586905, "grad_norm": 1.3527296781539917, "learning_rate": 7.600115657076767e-05, "loss": 5.0825, "step": 93000 }, { "epoch": 6.568778979907265, "grad_norm": 1.3230575323104858, "learning_rate": 7.585658522480845e-05, "loss": 5.0783, "step": 93500 }, { "epoch": 6.603906140227624, "grad_norm": 1.25337553024292, "learning_rate": 7.571201387884922e-05, "loss": 5.0744, "step": 94000 }, { "epoch": 6.639033300547983, "grad_norm": 1.2015641927719116, "learning_rate": 7.556744253288999e-05, "loss": 5.0726, "step": 94500 }, { "epoch": 6.674160460868343, "grad_norm": 1.231350302696228, "learning_rate": 7.542287118693075e-05, "loss": 5.0696, "step": 95000 }, { "epoch": 6.709287621188703, "grad_norm": 1.2066929340362549, "learning_rate": 7.527829984097152e-05, "loss": 5.0698, "step": 95500 }, { "epoch": 6.744414781509063, "grad_norm": 1.2691816091537476, "learning_rate": 7.51337284950123e-05, "loss": 5.0748, "step": 96000 }, { "epoch": 6.7795419418294225, "grad_norm": 1.1830929517745972, "learning_rate": 7.498915714905306e-05, "loss": 5.0652, "step": 96500 }, { "epoch": 6.814669102149782, "grad_norm": 1.2081068754196167, "learning_rate": 7.484458580309382e-05, "loss": 5.0657, "step": 97000 }, { "epoch": 6.849796262470142, "grad_norm": 1.164678692817688, "learning_rate": 7.47000144571346e-05, "loss": 5.059, "step": 97500 }, { "epoch": 6.884923422790502, "grad_norm": 1.2889105081558228, "learning_rate": 7.455544311117538e-05, "loss": 5.0608, "step": 98000 }, { "epoch": 6.920050583110862, "grad_norm": 1.1974706649780273, "learning_rate": 7.441087176521614e-05, "loss": 5.0567, "step": 98500 }, { "epoch": 6.955177743431221, "grad_norm": 1.2864340543746948, "learning_rate": 7.42663004192569e-05, "loss": 5.0566, "step": 99000 }, { "epoch": 6.99030490375158, "grad_norm": 1.2518681287765503, "learning_rate": 7.412172907329767e-05, "loss": 5.0538, "step": 99500 }, { "epoch": 7.02543206407194, "grad_norm": 1.170648217201233, "learning_rate": 7.397715772733845e-05, "loss": 5.0527, "step": 100000 }, { "epoch": 7.02543206407194, "eval_accuracy": 0.6454498729990131, "eval_loss": 4.707531929016113, "eval_runtime": 124.5703, "eval_samples_per_second": 968.979, "eval_steps_per_second": 6.061, "step": 100000 }, { "epoch": 7.0605592243923, "grad_norm": 1.2881115674972534, "learning_rate": 7.383258638137921e-05, "loss": 5.0492, "step": 100500 }, { "epoch": 7.09568638471266, "grad_norm": 1.2099517583847046, "learning_rate": 7.368801503541999e-05, "loss": 5.0504, "step": 101000 }, { "epoch": 7.1308135450330195, "grad_norm": 1.174860954284668, "learning_rate": 7.354344368946075e-05, "loss": 5.0485, "step": 101500 }, { "epoch": 7.165940705353379, "grad_norm": 1.228942632675171, "learning_rate": 7.339887234350153e-05, "loss": 5.0463, "step": 102000 }, { "epoch": 7.201067865673739, "grad_norm": 1.1392688751220703, "learning_rate": 7.32543009975423e-05, "loss": 5.0442, "step": 102500 }, { "epoch": 7.236195025994099, "grad_norm": 1.3000644445419312, "learning_rate": 7.310972965158306e-05, "loss": 5.0422, "step": 103000 }, { "epoch": 7.271322186314459, "grad_norm": 1.1535313129425049, "learning_rate": 7.296515830562382e-05, "loss": 5.0446, "step": 103500 }, { "epoch": 7.306449346634818, "grad_norm": 1.1499958038330078, "learning_rate": 7.28205869596646e-05, "loss": 5.0371, "step": 104000 }, { "epoch": 7.341576506955178, "grad_norm": 1.1743727922439575, "learning_rate": 7.267601561370536e-05, "loss": 5.0342, "step": 104500 }, { "epoch": 7.376703667275537, "grad_norm": 1.1658422946929932, "learning_rate": 7.253144426774614e-05, "loss": 5.0373, "step": 105000 }, { "epoch": 7.411830827595897, "grad_norm": 1.1941930055618286, "learning_rate": 7.23868729217869e-05, "loss": 5.0397, "step": 105500 }, { "epoch": 7.446957987916257, "grad_norm": 1.26359224319458, "learning_rate": 7.224230157582768e-05, "loss": 5.0384, "step": 106000 }, { "epoch": 7.482085148236616, "grad_norm": 1.2290655374526978, "learning_rate": 7.209773022986845e-05, "loss": 5.0342, "step": 106500 }, { "epoch": 7.517212308556976, "grad_norm": 1.2265841960906982, "learning_rate": 7.195315888390921e-05, "loss": 5.0289, "step": 107000 }, { "epoch": 7.552339468877336, "grad_norm": 1.1834629774093628, "learning_rate": 7.180858753794997e-05, "loss": 5.0281, "step": 107500 }, { "epoch": 7.587466629197696, "grad_norm": 1.2507638931274414, "learning_rate": 7.166401619199075e-05, "loss": 5.0296, "step": 108000 }, { "epoch": 7.6225937895180556, "grad_norm": 1.186900019645691, "learning_rate": 7.151944484603153e-05, "loss": 5.0313, "step": 108500 }, { "epoch": 7.657720949838415, "grad_norm": 1.2016674280166626, "learning_rate": 7.13748735000723e-05, "loss": 5.0265, "step": 109000 }, { "epoch": 7.692848110158775, "grad_norm": 1.1423753499984741, "learning_rate": 7.123030215411306e-05, "loss": 5.0263, "step": 109500 }, { "epoch": 7.727975270479135, "grad_norm": 1.1622411012649536, "learning_rate": 7.108573080815382e-05, "loss": 5.0258, "step": 110000 }, { "epoch": 7.727975270479135, "eval_accuracy": 0.6488509030575409, "eval_loss": 4.6829681396484375, "eval_runtime": 125.5323, "eval_samples_per_second": 961.553, "eval_steps_per_second": 6.014, "step": 110000 }, { "epoch": 7.763102430799494, "grad_norm": 1.1935741901397705, "learning_rate": 7.09411594621946e-05, "loss": 5.0236, "step": 110500 }, { "epoch": 7.798229591119854, "grad_norm": 1.149841070175171, "learning_rate": 7.079658811623536e-05, "loss": 5.02, "step": 111000 }, { "epoch": 7.833356751440213, "grad_norm": 1.2027647495269775, "learning_rate": 7.065201677027613e-05, "loss": 5.0204, "step": 111500 }, { "epoch": 7.868483911760573, "grad_norm": 1.1929402351379395, "learning_rate": 7.05074454243169e-05, "loss": 5.0214, "step": 112000 }, { "epoch": 7.903611072080933, "grad_norm": 1.1726230382919312, "learning_rate": 7.036287407835768e-05, "loss": 5.0194, "step": 112500 }, { "epoch": 7.938738232401293, "grad_norm": 1.195603370666504, "learning_rate": 7.021830273239845e-05, "loss": 5.0131, "step": 113000 }, { "epoch": 7.9738653927216525, "grad_norm": 1.2208681106567383, "learning_rate": 7.007373138643921e-05, "loss": 5.0138, "step": 113500 }, { "epoch": 8.008992553042011, "grad_norm": 1.1930869817733765, "learning_rate": 6.992916004047997e-05, "loss": 5.0143, "step": 114000 }, { "epoch": 8.044119713362372, "grad_norm": 1.2132607698440552, "learning_rate": 6.978458869452075e-05, "loss": 5.0082, "step": 114500 }, { "epoch": 8.079246873682731, "grad_norm": 1.1897276639938354, "learning_rate": 6.964001734856152e-05, "loss": 5.0071, "step": 115000 }, { "epoch": 8.114374034003092, "grad_norm": 1.2654327154159546, "learning_rate": 6.949544600260228e-05, "loss": 5.0071, "step": 115500 }, { "epoch": 8.14950119432345, "grad_norm": 1.1854734420776367, "learning_rate": 6.935087465664306e-05, "loss": 5.0074, "step": 116000 }, { "epoch": 8.184628354643811, "grad_norm": 1.2320798635482788, "learning_rate": 6.920630331068383e-05, "loss": 5.0078, "step": 116500 }, { "epoch": 8.21975551496417, "grad_norm": 1.1847695112228394, "learning_rate": 6.90617319647246e-05, "loss": 5.0034, "step": 117000 }, { "epoch": 8.25488267528453, "grad_norm": 1.2522870302200317, "learning_rate": 6.891716061876536e-05, "loss": 5.0033, "step": 117500 }, { "epoch": 8.29000983560489, "grad_norm": 1.2357484102249146, "learning_rate": 6.877258927280613e-05, "loss": 4.9994, "step": 118000 }, { "epoch": 8.325136995925249, "grad_norm": 1.2570139169692993, "learning_rate": 6.86280179268469e-05, "loss": 4.9982, "step": 118500 }, { "epoch": 8.36026415624561, "grad_norm": 1.1642590761184692, "learning_rate": 6.848344658088767e-05, "loss": 4.9996, "step": 119000 }, { "epoch": 8.395391316565968, "grad_norm": 1.2592113018035889, "learning_rate": 6.833887523492845e-05, "loss": 4.9985, "step": 119500 }, { "epoch": 8.430518476886329, "grad_norm": 1.2361762523651123, "learning_rate": 6.819430388896921e-05, "loss": 4.9991, "step": 120000 }, { "epoch": 8.430518476886329, "eval_accuracy": 0.6518944310752692, "eval_loss": 4.657070159912109, "eval_runtime": 125.4785, "eval_samples_per_second": 961.966, "eval_steps_per_second": 6.017, "step": 120000 }, { "epoch": 8.465645637206688, "grad_norm": 1.1777372360229492, "learning_rate": 6.804973254300999e-05, "loss": 4.9993, "step": 120500 }, { "epoch": 8.500772797527048, "grad_norm": 1.2769887447357178, "learning_rate": 6.790516119705075e-05, "loss": 4.9978, "step": 121000 }, { "epoch": 8.535899957847407, "grad_norm": 1.1881922483444214, "learning_rate": 6.776058985109152e-05, "loss": 4.9987, "step": 121500 }, { "epoch": 8.571027118167768, "grad_norm": 1.2238999605178833, "learning_rate": 6.761601850513228e-05, "loss": 4.994, "step": 122000 }, { "epoch": 8.606154278488127, "grad_norm": 1.2310227155685425, "learning_rate": 6.747144715917306e-05, "loss": 4.9934, "step": 122500 }, { "epoch": 8.641281438808488, "grad_norm": 1.2682673931121826, "learning_rate": 6.732687581321382e-05, "loss": 4.9916, "step": 123000 }, { "epoch": 8.676408599128846, "grad_norm": 1.2880076169967651, "learning_rate": 6.71823044672546e-05, "loss": 4.9862, "step": 123500 }, { "epoch": 8.711535759449205, "grad_norm": 1.2074475288391113, "learning_rate": 6.703773312129536e-05, "loss": 4.9855, "step": 124000 }, { "epoch": 8.746662919769566, "grad_norm": 1.185457706451416, "learning_rate": 6.689316177533613e-05, "loss": 4.9822, "step": 124500 }, { "epoch": 8.781790080089925, "grad_norm": 1.2463042736053467, "learning_rate": 6.67485904293769e-05, "loss": 4.9853, "step": 125000 }, { "epoch": 8.816917240410286, "grad_norm": 1.2737094163894653, "learning_rate": 6.660401908341767e-05, "loss": 4.9849, "step": 125500 }, { "epoch": 8.852044400730644, "grad_norm": 1.1822280883789062, "learning_rate": 6.645944773745843e-05, "loss": 4.9839, "step": 126000 }, { "epoch": 8.887171561051005, "grad_norm": 1.1642378568649292, "learning_rate": 6.631487639149921e-05, "loss": 4.9836, "step": 126500 }, { "epoch": 8.922298721371364, "grad_norm": 1.2464725971221924, "learning_rate": 6.617030504553999e-05, "loss": 4.9817, "step": 127000 }, { "epoch": 8.957425881691725, "grad_norm": 1.2214902639389038, "learning_rate": 6.602573369958075e-05, "loss": 4.9871, "step": 127500 }, { "epoch": 8.992553042012084, "grad_norm": 1.2652417421340942, "learning_rate": 6.588116235362152e-05, "loss": 4.9838, "step": 128000 }, { "epoch": 9.027680202332444, "grad_norm": 1.2146638631820679, "learning_rate": 6.573659100766228e-05, "loss": 4.9763, "step": 128500 }, { "epoch": 9.062807362652803, "grad_norm": 1.125086784362793, "learning_rate": 6.559201966170306e-05, "loss": 4.9764, "step": 129000 }, { "epoch": 9.097934522973162, "grad_norm": 1.2171489000320435, "learning_rate": 6.544744831574382e-05, "loss": 4.9772, "step": 129500 }, { "epoch": 9.133061683293523, "grad_norm": 1.1570535898208618, "learning_rate": 6.530287696978458e-05, "loss": 4.9708, "step": 130000 }, { "epoch": 9.133061683293523, "eval_accuracy": 0.6542186608492178, "eval_loss": 4.642580509185791, "eval_runtime": 125.2179, "eval_samples_per_second": 963.967, "eval_steps_per_second": 6.029, "step": 130000 }, { "epoch": 9.168188843613882, "grad_norm": 1.1505987644195557, "learning_rate": 6.515830562382536e-05, "loss": 4.9737, "step": 130500 }, { "epoch": 9.203316003934242, "grad_norm": 1.1545060873031616, "learning_rate": 6.501373427786614e-05, "loss": 4.9758, "step": 131000 }, { "epoch": 9.238443164254601, "grad_norm": 1.2373590469360352, "learning_rate": 6.48691629319069e-05, "loss": 4.9741, "step": 131500 }, { "epoch": 9.273570324574962, "grad_norm": 1.1195789575576782, "learning_rate": 6.472459158594767e-05, "loss": 4.9737, "step": 132000 }, { "epoch": 9.30869748489532, "grad_norm": 1.1722861528396606, "learning_rate": 6.458002023998843e-05, "loss": 4.9722, "step": 132500 }, { "epoch": 9.343824645215681, "grad_norm": 1.198255181312561, "learning_rate": 6.443544889402921e-05, "loss": 4.9703, "step": 133000 }, { "epoch": 9.37895180553604, "grad_norm": 1.2369712591171265, "learning_rate": 6.429087754806997e-05, "loss": 4.9685, "step": 133500 }, { "epoch": 9.414078965856401, "grad_norm": 1.1678396463394165, "learning_rate": 6.414630620211074e-05, "loss": 4.9701, "step": 134000 }, { "epoch": 9.44920612617676, "grad_norm": 1.2125905752182007, "learning_rate": 6.400173485615151e-05, "loss": 4.9649, "step": 134500 }, { "epoch": 9.484333286497119, "grad_norm": 1.256878137588501, "learning_rate": 6.385716351019229e-05, "loss": 4.9632, "step": 135000 }, { "epoch": 9.51946044681748, "grad_norm": 1.147707462310791, "learning_rate": 6.371259216423306e-05, "loss": 4.9647, "step": 135500 }, { "epoch": 9.554587607137838, "grad_norm": 1.1856908798217773, "learning_rate": 6.356802081827382e-05, "loss": 4.9652, "step": 136000 }, { "epoch": 9.589714767458199, "grad_norm": 1.1790053844451904, "learning_rate": 6.342344947231458e-05, "loss": 4.9661, "step": 136500 }, { "epoch": 9.624841927778558, "grad_norm": 1.1994777917861938, "learning_rate": 6.327887812635536e-05, "loss": 4.9623, "step": 137000 }, { "epoch": 9.659969088098919, "grad_norm": 1.1758321523666382, "learning_rate": 6.313430678039613e-05, "loss": 4.9636, "step": 137500 }, { "epoch": 9.695096248419278, "grad_norm": 1.2180451154708862, "learning_rate": 6.29897354344369e-05, "loss": 4.9555, "step": 138000 }, { "epoch": 9.730223408739638, "grad_norm": 1.2169358730316162, "learning_rate": 6.284516408847767e-05, "loss": 4.9646, "step": 138500 }, { "epoch": 9.765350569059997, "grad_norm": 1.1795868873596191, "learning_rate": 6.270059274251843e-05, "loss": 4.9582, "step": 139000 }, { "epoch": 9.800477729380358, "grad_norm": 1.1503639221191406, "learning_rate": 6.255602139655921e-05, "loss": 4.9592, "step": 139500 }, { "epoch": 9.835604889700717, "grad_norm": 1.2159857749938965, "learning_rate": 6.241145005059997e-05, "loss": 4.9566, "step": 140000 }, { "epoch": 9.835604889700717, "eval_accuracy": 0.657344947022373, "eval_loss": 4.620968341827393, "eval_runtime": 125.5523, "eval_samples_per_second": 961.4, "eval_steps_per_second": 6.013, "step": 140000 }, { "epoch": 9.870732050021076, "grad_norm": 1.173851728439331, "learning_rate": 6.226687870464074e-05, "loss": 4.9572, "step": 140500 }, { "epoch": 9.905859210341436, "grad_norm": 1.201773762702942, "learning_rate": 6.212230735868151e-05, "loss": 4.9554, "step": 141000 }, { "epoch": 9.940986370661795, "grad_norm": 1.1786022186279297, "learning_rate": 6.197773601272228e-05, "loss": 4.9541, "step": 141500 }, { "epoch": 9.976113530982156, "grad_norm": 1.3232065439224243, "learning_rate": 6.183316466676306e-05, "loss": 4.951, "step": 142000 }, { "epoch": 10.011240691302515, "grad_norm": 1.241267204284668, "learning_rate": 6.168859332080382e-05, "loss": 4.9545, "step": 142500 }, { "epoch": 10.046367851622875, "grad_norm": 1.1741814613342285, "learning_rate": 6.154402197484458e-05, "loss": 4.9516, "step": 143000 }, { "epoch": 10.081495011943234, "grad_norm": 1.1933821439743042, "learning_rate": 6.139945062888536e-05, "loss": 4.9474, "step": 143500 }, { "epoch": 10.116622172263595, "grad_norm": 1.2142330408096313, "learning_rate": 6.125487928292613e-05, "loss": 4.9453, "step": 144000 }, { "epoch": 10.151749332583954, "grad_norm": 1.2016196250915527, "learning_rate": 6.111030793696689e-05, "loss": 4.9495, "step": 144500 }, { "epoch": 10.186876492904315, "grad_norm": 1.259665846824646, "learning_rate": 6.096573659100766e-05, "loss": 4.9482, "step": 145000 }, { "epoch": 10.222003653224673, "grad_norm": 1.1649497747421265, "learning_rate": 6.082116524504844e-05, "loss": 4.9495, "step": 145500 }, { "epoch": 10.257130813545032, "grad_norm": 1.202059030532837, "learning_rate": 6.067659389908921e-05, "loss": 4.9483, "step": 146000 }, { "epoch": 10.292257973865393, "grad_norm": 1.177734375, "learning_rate": 6.053202255312997e-05, "loss": 4.943, "step": 146500 }, { "epoch": 10.327385134185752, "grad_norm": 1.2007008790969849, "learning_rate": 6.0387451207170744e-05, "loss": 4.9435, "step": 147000 }, { "epoch": 10.362512294506113, "grad_norm": 1.1573467254638672, "learning_rate": 6.024287986121151e-05, "loss": 4.9433, "step": 147500 }, { "epoch": 10.397639454826471, "grad_norm": 1.1788324117660522, "learning_rate": 6.009830851525228e-05, "loss": 4.9399, "step": 148000 }, { "epoch": 10.432766615146832, "grad_norm": 1.2692636251449585, "learning_rate": 5.995373716929304e-05, "loss": 4.9454, "step": 148500 }, { "epoch": 10.467893775467191, "grad_norm": 1.132515549659729, "learning_rate": 5.980916582333381e-05, "loss": 4.9393, "step": 149000 }, { "epoch": 10.503020935787552, "grad_norm": 1.18707275390625, "learning_rate": 5.966459447737459e-05, "loss": 4.9381, "step": 149500 }, { "epoch": 10.53814809610791, "grad_norm": 1.2118017673492432, "learning_rate": 5.952002313141536e-05, "loss": 4.9396, "step": 150000 }, { "epoch": 10.53814809610791, "eval_accuracy": 0.6589358687639654, "eval_loss": 4.607008934020996, "eval_runtime": 125.4136, "eval_samples_per_second": 962.464, "eval_steps_per_second": 6.02, "step": 150000 }, { "epoch": 10.573275256428271, "grad_norm": 1.2168219089508057, "learning_rate": 5.9375451785456126e-05, "loss": 4.9441, "step": 150500 }, { "epoch": 10.60840241674863, "grad_norm": 1.183809757232666, "learning_rate": 5.9230880439496896e-05, "loss": 4.9385, "step": 151000 }, { "epoch": 10.643529577068989, "grad_norm": 1.2794101238250732, "learning_rate": 5.908630909353766e-05, "loss": 4.9378, "step": 151500 }, { "epoch": 10.67865673738935, "grad_norm": 1.1689482927322388, "learning_rate": 5.894173774757843e-05, "loss": 4.9391, "step": 152000 }, { "epoch": 10.713783897709709, "grad_norm": 1.1655418872833252, "learning_rate": 5.8797166401619195e-05, "loss": 4.9348, "step": 152500 }, { "epoch": 10.74891105803007, "grad_norm": 1.1817426681518555, "learning_rate": 5.865259505565997e-05, "loss": 4.9358, "step": 153000 }, { "epoch": 10.784038218350428, "grad_norm": 1.1901863813400269, "learning_rate": 5.8508023709700743e-05, "loss": 4.9375, "step": 153500 }, { "epoch": 10.819165378670789, "grad_norm": 1.1427429914474487, "learning_rate": 5.8363452363741514e-05, "loss": 4.9348, "step": 154000 }, { "epoch": 10.854292538991148, "grad_norm": 1.1928921937942505, "learning_rate": 5.821888101778228e-05, "loss": 4.9356, "step": 154500 }, { "epoch": 10.889419699311508, "grad_norm": 1.1977267265319824, "learning_rate": 5.807430967182305e-05, "loss": 4.935, "step": 155000 }, { "epoch": 10.924546859631867, "grad_norm": 1.2637089490890503, "learning_rate": 5.792973832586381e-05, "loss": 4.934, "step": 155500 }, { "epoch": 10.959674019952226, "grad_norm": 1.1878407001495361, "learning_rate": 5.7785166979904584e-05, "loss": 4.9338, "step": 156000 }, { "epoch": 10.994801180272587, "grad_norm": 1.1838468313217163, "learning_rate": 5.764059563394536e-05, "loss": 4.9306, "step": 156500 }, { "epoch": 11.029928340592946, "grad_norm": 1.1852989196777344, "learning_rate": 5.7496024287986125e-05, "loss": 4.9238, "step": 157000 }, { "epoch": 11.065055500913306, "grad_norm": 1.2436959743499756, "learning_rate": 5.7351452942026896e-05, "loss": 4.9279, "step": 157500 }, { "epoch": 11.100182661233665, "grad_norm": 1.1707674264907837, "learning_rate": 5.720688159606766e-05, "loss": 4.9269, "step": 158000 }, { "epoch": 11.135309821554026, "grad_norm": 1.2317562103271484, "learning_rate": 5.706231025010843e-05, "loss": 4.9253, "step": 158500 }, { "epoch": 11.170436981874385, "grad_norm": 1.2182191610336304, "learning_rate": 5.6917738904149195e-05, "loss": 4.9264, "step": 159000 }, { "epoch": 11.205564142194746, "grad_norm": 1.186889410018921, "learning_rate": 5.6773167558189966e-05, "loss": 4.929, "step": 159500 }, { "epoch": 11.240691302515105, "grad_norm": 1.269489049911499, "learning_rate": 5.6628596212230736e-05, "loss": 4.929, "step": 160000 }, { "epoch": 11.240691302515105, "eval_accuracy": 0.6604041903355252, "eval_loss": 4.59634256362915, "eval_runtime": 125.6574, "eval_samples_per_second": 960.596, "eval_steps_per_second": 6.008, "step": 160000 }, { "epoch": 11.275818462835465, "grad_norm": 1.1887845993041992, "learning_rate": 5.6484024866271514e-05, "loss": 4.9234, "step": 160500 }, { "epoch": 11.310945623155824, "grad_norm": 1.206114411354065, "learning_rate": 5.633945352031228e-05, "loss": 4.9278, "step": 161000 }, { "epoch": 11.346072783476183, "grad_norm": 1.1506130695343018, "learning_rate": 5.619488217435305e-05, "loss": 4.9239, "step": 161500 }, { "epoch": 11.381199943796544, "grad_norm": 1.25631582736969, "learning_rate": 5.605031082839381e-05, "loss": 4.9232, "step": 162000 }, { "epoch": 11.416327104116903, "grad_norm": 1.2387101650238037, "learning_rate": 5.5905739482434584e-05, "loss": 4.9209, "step": 162500 }, { "epoch": 11.451454264437263, "grad_norm": 1.1939486265182495, "learning_rate": 5.576116813647535e-05, "loss": 4.9167, "step": 163000 }, { "epoch": 11.486581424757622, "grad_norm": 1.2466892004013062, "learning_rate": 5.561659679051612e-05, "loss": 4.9204, "step": 163500 }, { "epoch": 11.521708585077983, "grad_norm": 1.1698747873306274, "learning_rate": 5.5472025444556896e-05, "loss": 4.9204, "step": 164000 }, { "epoch": 11.556835745398342, "grad_norm": 1.1395771503448486, "learning_rate": 5.532745409859767e-05, "loss": 4.9181, "step": 164500 }, { "epoch": 11.591962905718702, "grad_norm": 1.203838586807251, "learning_rate": 5.518288275263843e-05, "loss": 4.9186, "step": 165000 }, { "epoch": 11.627090066039061, "grad_norm": 1.149657964706421, "learning_rate": 5.50383114066792e-05, "loss": 4.9185, "step": 165500 }, { "epoch": 11.662217226359422, "grad_norm": 1.1450413465499878, "learning_rate": 5.4893740060719965e-05, "loss": 4.9161, "step": 166000 }, { "epoch": 11.69734438667978, "grad_norm": 1.209080696105957, "learning_rate": 5.4749168714760736e-05, "loss": 4.9185, "step": 166500 }, { "epoch": 11.73247154700014, "grad_norm": 1.1621487140655518, "learning_rate": 5.46045973688015e-05, "loss": 4.9124, "step": 167000 }, { "epoch": 11.7675987073205, "grad_norm": 1.189600944519043, "learning_rate": 5.446002602284227e-05, "loss": 4.9149, "step": 167500 }, { "epoch": 11.80272586764086, "grad_norm": 1.2446949481964111, "learning_rate": 5.431545467688305e-05, "loss": 4.9156, "step": 168000 }, { "epoch": 11.83785302796122, "grad_norm": 1.1775277853012085, "learning_rate": 5.417088333092382e-05, "loss": 4.9111, "step": 168500 }, { "epoch": 11.872980188281579, "grad_norm": 1.2621877193450928, "learning_rate": 5.402631198496458e-05, "loss": 4.9111, "step": 169000 }, { "epoch": 11.90810734860194, "grad_norm": 1.1973707675933838, "learning_rate": 5.3881740639005354e-05, "loss": 4.9099, "step": 169500 }, { "epoch": 11.943234508922298, "grad_norm": 1.222820520401001, "learning_rate": 5.373716929304612e-05, "loss": 4.9134, "step": 170000 }, { "epoch": 11.943234508922298, "eval_accuracy": 0.6622467923008861, "eval_loss": 4.58348274230957, "eval_runtime": 127.4415, "eval_samples_per_second": 947.149, "eval_steps_per_second": 5.924, "step": 170000 }, { "epoch": 11.978361669242659, "grad_norm": 1.319101095199585, "learning_rate": 5.359259794708689e-05, "loss": 4.9114, "step": 170500 }, { "epoch": 12.013488829563018, "grad_norm": 1.2069631814956665, "learning_rate": 5.344802660112765e-05, "loss": 4.9127, "step": 171000 }, { "epoch": 12.048615989883379, "grad_norm": 1.2296098470687866, "learning_rate": 5.330345525516843e-05, "loss": 4.9083, "step": 171500 }, { "epoch": 12.083743150203738, "grad_norm": 1.2166742086410522, "learning_rate": 5.31588839092092e-05, "loss": 4.9066, "step": 172000 }, { "epoch": 12.118870310524096, "grad_norm": 1.1931017637252808, "learning_rate": 5.3014312563249965e-05, "loss": 4.908, "step": 172500 }, { "epoch": 12.153997470844457, "grad_norm": 1.1866223812103271, "learning_rate": 5.2869741217290736e-05, "loss": 4.9124, "step": 173000 }, { "epoch": 12.189124631164816, "grad_norm": 1.2110309600830078, "learning_rate": 5.272516987133151e-05, "loss": 4.9067, "step": 173500 }, { "epoch": 12.224251791485177, "grad_norm": 1.2086114883422852, "learning_rate": 5.258059852537227e-05, "loss": 4.9019, "step": 174000 }, { "epoch": 12.259378951805536, "grad_norm": 1.2596267461776733, "learning_rate": 5.243602717941304e-05, "loss": 4.9069, "step": 174500 }, { "epoch": 12.294506112125896, "grad_norm": 1.171631932258606, "learning_rate": 5.229145583345382e-05, "loss": 4.9008, "step": 175000 }, { "epoch": 12.329633272446255, "grad_norm": 1.24288809299469, "learning_rate": 5.214688448749458e-05, "loss": 4.908, "step": 175500 }, { "epoch": 12.364760432766616, "grad_norm": 1.178281545639038, "learning_rate": 5.2002313141535354e-05, "loss": 4.9039, "step": 176000 }, { "epoch": 12.399887593086975, "grad_norm": 1.1844769716262817, "learning_rate": 5.185774179557612e-05, "loss": 4.9021, "step": 176500 }, { "epoch": 12.435014753407334, "grad_norm": 1.2377811670303345, "learning_rate": 5.171317044961689e-05, "loss": 4.9053, "step": 177000 }, { "epoch": 12.470141913727694, "grad_norm": 1.175628900527954, "learning_rate": 5.156859910365765e-05, "loss": 4.9019, "step": 177500 }, { "epoch": 12.505269074048053, "grad_norm": 1.1951406002044678, "learning_rate": 5.1424027757698423e-05, "loss": 4.9038, "step": 178000 }, { "epoch": 12.540396234368414, "grad_norm": 1.271713376045227, "learning_rate": 5.127945641173919e-05, "loss": 4.9058, "step": 178500 }, { "epoch": 12.575523394688773, "grad_norm": 1.193676471710205, "learning_rate": 5.113488506577997e-05, "loss": 4.9016, "step": 179000 }, { "epoch": 12.610650555009133, "grad_norm": 1.164211630821228, "learning_rate": 5.0990313719820736e-05, "loss": 4.9004, "step": 179500 }, { "epoch": 12.645777715329492, "grad_norm": 1.228807806968689, "learning_rate": 5.0845742373861507e-05, "loss": 4.901, "step": 180000 }, { "epoch": 12.645777715329492, "eval_accuracy": 0.6639364887616751, "eval_loss": 4.568302631378174, "eval_runtime": 126.2785, "eval_samples_per_second": 955.872, "eval_steps_per_second": 5.979, "step": 180000 }, { "epoch": 12.680904875649853, "grad_norm": 1.250861644744873, "learning_rate": 5.070117102790227e-05, "loss": 4.8987, "step": 180500 }, { "epoch": 12.716032035970212, "grad_norm": 1.2661194801330566, "learning_rate": 5.055659968194304e-05, "loss": 4.8986, "step": 181000 }, { "epoch": 12.751159196290573, "grad_norm": 1.210195541381836, "learning_rate": 5.0412028335983805e-05, "loss": 4.9008, "step": 181500 }, { "epoch": 12.786286356610931, "grad_norm": 1.276237964630127, "learning_rate": 5.0267456990024576e-05, "loss": 4.8976, "step": 182000 }, { "epoch": 12.82141351693129, "grad_norm": 1.1311328411102295, "learning_rate": 5.0122885644065354e-05, "loss": 4.9002, "step": 182500 }, { "epoch": 12.856540677251651, "grad_norm": 1.2184550762176514, "learning_rate": 4.997831429810612e-05, "loss": 4.8966, "step": 183000 }, { "epoch": 12.89166783757201, "grad_norm": 1.1910830736160278, "learning_rate": 4.983374295214689e-05, "loss": 4.896, "step": 183500 }, { "epoch": 12.92679499789237, "grad_norm": 1.3047518730163574, "learning_rate": 4.968917160618766e-05, "loss": 4.894, "step": 184000 }, { "epoch": 12.96192215821273, "grad_norm": 1.2237051725387573, "learning_rate": 4.954460026022842e-05, "loss": 4.8988, "step": 184500 }, { "epoch": 12.99704931853309, "grad_norm": 1.2142150402069092, "learning_rate": 4.9400028914269194e-05, "loss": 4.8967, "step": 185000 }, { "epoch": 13.032176478853449, "grad_norm": 1.2646372318267822, "learning_rate": 4.9255457568309965e-05, "loss": 4.8924, "step": 185500 }, { "epoch": 13.06730363917381, "grad_norm": 1.1980942487716675, "learning_rate": 4.9110886222350736e-05, "loss": 4.8912, "step": 186000 }, { "epoch": 13.102430799494169, "grad_norm": 1.2629406452178955, "learning_rate": 4.89663148763915e-05, "loss": 4.8918, "step": 186500 }, { "epoch": 13.13755795981453, "grad_norm": 1.2157944440841675, "learning_rate": 4.882174353043227e-05, "loss": 4.8881, "step": 187000 }, { "epoch": 13.172685120134888, "grad_norm": 1.2210171222686768, "learning_rate": 4.867717218447304e-05, "loss": 4.8893, "step": 187500 }, { "epoch": 13.207812280455247, "grad_norm": 1.165700912475586, "learning_rate": 4.853260083851381e-05, "loss": 4.8909, "step": 188000 }, { "epoch": 13.242939440775608, "grad_norm": 1.1676913499832153, "learning_rate": 4.8388029492554576e-05, "loss": 4.8908, "step": 188500 }, { "epoch": 13.278066601095967, "grad_norm": 1.1220760345458984, "learning_rate": 4.824345814659535e-05, "loss": 4.8893, "step": 189000 }, { "epoch": 13.313193761416327, "grad_norm": 1.22340989112854, "learning_rate": 4.809888680063612e-05, "loss": 4.892, "step": 189500 }, { "epoch": 13.348320921736686, "grad_norm": 1.2241616249084473, "learning_rate": 4.795431545467688e-05, "loss": 4.8884, "step": 190000 }, { "epoch": 13.348320921736686, "eval_accuracy": 0.6645560421281556, "eval_loss": 4.5643630027771, "eval_runtime": 126.1255, "eval_samples_per_second": 957.031, "eval_steps_per_second": 5.986, "step": 190000 }, { "epoch": 13.383448082057047, "grad_norm": 1.2446415424346924, "learning_rate": 4.780974410871765e-05, "loss": 4.8847, "step": 190500 }, { "epoch": 13.418575242377406, "grad_norm": 1.2117501497268677, "learning_rate": 4.766517276275842e-05, "loss": 4.8893, "step": 191000 }, { "epoch": 13.453702402697767, "grad_norm": 1.1914085149765015, "learning_rate": 4.7520601416799194e-05, "loss": 4.8863, "step": 191500 }, { "epoch": 13.488829563018125, "grad_norm": 1.1365530490875244, "learning_rate": 4.737603007083996e-05, "loss": 4.8872, "step": 192000 }, { "epoch": 13.523956723338486, "grad_norm": 1.1694480180740356, "learning_rate": 4.723145872488073e-05, "loss": 4.8879, "step": 192500 }, { "epoch": 13.559083883658845, "grad_norm": 1.2067158222198486, "learning_rate": 4.70868873789215e-05, "loss": 4.8866, "step": 193000 }, { "epoch": 13.594211043979204, "grad_norm": 1.2343794107437134, "learning_rate": 4.694231603296227e-05, "loss": 4.8864, "step": 193500 }, { "epoch": 13.629338204299565, "grad_norm": 1.1796584129333496, "learning_rate": 4.6797744687003034e-05, "loss": 4.8841, "step": 194000 }, { "epoch": 13.664465364619923, "grad_norm": 1.2234842777252197, "learning_rate": 4.665317334104381e-05, "loss": 4.883, "step": 194500 }, { "epoch": 13.699592524940284, "grad_norm": 1.2463550567626953, "learning_rate": 4.6508601995084576e-05, "loss": 4.8843, "step": 195000 }, { "epoch": 13.734719685260643, "grad_norm": 1.3087186813354492, "learning_rate": 4.6364030649125346e-05, "loss": 4.8846, "step": 195500 }, { "epoch": 13.769846845581004, "grad_norm": 1.2503262758255005, "learning_rate": 4.621945930316611e-05, "loss": 4.8841, "step": 196000 }, { "epoch": 13.804974005901363, "grad_norm": 1.1892198324203491, "learning_rate": 4.607488795720689e-05, "loss": 4.8813, "step": 196500 }, { "epoch": 13.840101166221723, "grad_norm": 1.2064417600631714, "learning_rate": 4.593031661124765e-05, "loss": 4.884, "step": 197000 }, { "epoch": 13.875228326542082, "grad_norm": 1.2562015056610107, "learning_rate": 4.578574526528842e-05, "loss": 4.8795, "step": 197500 }, { "epoch": 13.910355486862443, "grad_norm": 1.1253942251205444, "learning_rate": 4.564117391932919e-05, "loss": 4.8805, "step": 198000 }, { "epoch": 13.945482647182802, "grad_norm": 1.2661116123199463, "learning_rate": 4.5496602573369964e-05, "loss": 4.8795, "step": 198500 }, { "epoch": 13.98060980750316, "grad_norm": 1.2310277223587036, "learning_rate": 4.535203122741073e-05, "loss": 4.8857, "step": 199000 }, { "epoch": 14.015736967823521, "grad_norm": 1.2121599912643433, "learning_rate": 4.52074598814515e-05, "loss": 4.8799, "step": 199500 }, { "epoch": 14.05086412814388, "grad_norm": 1.243844985961914, "learning_rate": 4.506288853549227e-05, "loss": 4.8823, "step": 200000 }, { "epoch": 14.05086412814388, "eval_accuracy": 0.6661261289249786, "eval_loss": 4.552921772003174, "eval_runtime": 126.4818, "eval_samples_per_second": 954.335, "eval_steps_per_second": 5.969, "step": 200000 }, { "epoch": 14.08599128846424, "grad_norm": 1.2012193202972412, "learning_rate": 4.491831718953304e-05, "loss": 4.876, "step": 200500 }, { "epoch": 14.1211184487846, "grad_norm": 1.1868584156036377, "learning_rate": 4.4773745843573805e-05, "loss": 4.8795, "step": 201000 }, { "epoch": 14.15624560910496, "grad_norm": 1.2169872522354126, "learning_rate": 4.4629174497614575e-05, "loss": 4.8748, "step": 201500 }, { "epoch": 14.19137276942532, "grad_norm": 1.1852657794952393, "learning_rate": 4.4484603151655346e-05, "loss": 4.8739, "step": 202000 }, { "epoch": 14.22649992974568, "grad_norm": 1.1942535638809204, "learning_rate": 4.434003180569612e-05, "loss": 4.8751, "step": 202500 }, { "epoch": 14.261627090066039, "grad_norm": 1.215712308883667, "learning_rate": 4.419546045973688e-05, "loss": 4.8773, "step": 203000 }, { "epoch": 14.2967542503864, "grad_norm": 1.241031527519226, "learning_rate": 4.405088911377765e-05, "loss": 4.8742, "step": 203500 }, { "epoch": 14.331881410706758, "grad_norm": 1.2578802108764648, "learning_rate": 4.390631776781842e-05, "loss": 4.8776, "step": 204000 }, { "epoch": 14.367008571027117, "grad_norm": 1.2334762811660767, "learning_rate": 4.3761746421859187e-05, "loss": 4.8745, "step": 204500 }, { "epoch": 14.402135731347478, "grad_norm": 1.2310295104980469, "learning_rate": 4.361717507589996e-05, "loss": 4.8738, "step": 205000 }, { "epoch": 14.437262891667837, "grad_norm": 1.2107402086257935, "learning_rate": 4.347260372994073e-05, "loss": 4.8741, "step": 205500 }, { "epoch": 14.472390051988198, "grad_norm": 1.1998326778411865, "learning_rate": 4.33280323839815e-05, "loss": 4.8733, "step": 206000 }, { "epoch": 14.507517212308557, "grad_norm": 1.2449408769607544, "learning_rate": 4.318346103802226e-05, "loss": 4.8718, "step": 206500 }, { "epoch": 14.542644372628917, "grad_norm": 1.1739016771316528, "learning_rate": 4.3038889692063034e-05, "loss": 4.8709, "step": 207000 }, { "epoch": 14.577771532949276, "grad_norm": 1.3018107414245605, "learning_rate": 4.2894318346103804e-05, "loss": 4.8723, "step": 207500 }, { "epoch": 14.612898693269637, "grad_norm": 1.2145353555679321, "learning_rate": 4.2749747000144575e-05, "loss": 4.8731, "step": 208000 }, { "epoch": 14.648025853589996, "grad_norm": 1.2300605773925781, "learning_rate": 4.260517565418534e-05, "loss": 4.8708, "step": 208500 }, { "epoch": 14.683153013910356, "grad_norm": 1.2295414209365845, "learning_rate": 4.246060430822611e-05, "loss": 4.8705, "step": 209000 }, { "epoch": 14.718280174230715, "grad_norm": 1.2211674451828003, "learning_rate": 4.231603296226688e-05, "loss": 4.8702, "step": 209500 }, { "epoch": 14.753407334551074, "grad_norm": 1.1789523363113403, "learning_rate": 4.217146161630765e-05, "loss": 4.8706, "step": 210000 }, { "epoch": 14.753407334551074, "eval_accuracy": 0.6672022115746427, "eval_loss": 4.544014930725098, "eval_runtime": 126.1167, "eval_samples_per_second": 957.098, "eval_steps_per_second": 5.987, "step": 210000 }, { "epoch": 14.788534494871435, "grad_norm": 1.236729383468628, "learning_rate": 4.2026890270348416e-05, "loss": 4.8686, "step": 210500 }, { "epoch": 14.823661655191794, "grad_norm": 1.2526315450668335, "learning_rate": 4.1882318924389186e-05, "loss": 4.8675, "step": 211000 }, { "epoch": 14.858788815512154, "grad_norm": 1.2497429847717285, "learning_rate": 4.173774757842996e-05, "loss": 4.8696, "step": 211500 }, { "epoch": 14.893915975832513, "grad_norm": 1.226439118385315, "learning_rate": 4.159317623247073e-05, "loss": 4.8665, "step": 212000 }, { "epoch": 14.929043136152874, "grad_norm": 1.28299880027771, "learning_rate": 4.144860488651149e-05, "loss": 4.8706, "step": 212500 }, { "epoch": 14.964170296473233, "grad_norm": 1.228131651878357, "learning_rate": 4.130403354055227e-05, "loss": 4.8699, "step": 213000 }, { "epoch": 14.999297456793593, "grad_norm": 1.2444589138031006, "learning_rate": 4.1159462194593033e-05, "loss": 4.8685, "step": 213500 }, { "epoch": 15.034424617113952, "grad_norm": 1.2919272184371948, "learning_rate": 4.1014890848633804e-05, "loss": 4.8624, "step": 214000 }, { "epoch": 15.069551777434313, "grad_norm": 1.2361222505569458, "learning_rate": 4.087031950267457e-05, "loss": 4.8659, "step": 214500 }, { "epoch": 15.104678937754672, "grad_norm": 1.2248547077178955, "learning_rate": 4.0725748156715346e-05, "loss": 4.8655, "step": 215000 }, { "epoch": 15.13980609807503, "grad_norm": 1.2208781242370605, "learning_rate": 4.058117681075611e-05, "loss": 4.8612, "step": 215500 }, { "epoch": 15.174933258395392, "grad_norm": 1.1718521118164062, "learning_rate": 4.043660546479688e-05, "loss": 4.8652, "step": 216000 }, { "epoch": 15.21006041871575, "grad_norm": 1.2290360927581787, "learning_rate": 4.0292034118837645e-05, "loss": 4.8649, "step": 216500 }, { "epoch": 15.245187579036111, "grad_norm": 1.1924952268600464, "learning_rate": 4.014746277287842e-05, "loss": 4.8645, "step": 217000 }, { "epoch": 15.28031473935647, "grad_norm": 1.1877204179763794, "learning_rate": 4.0002891426919186e-05, "loss": 4.8615, "step": 217500 }, { "epoch": 15.31544189967683, "grad_norm": 1.2351772785186768, "learning_rate": 3.985832008095996e-05, "loss": 4.8627, "step": 218000 }, { "epoch": 15.35056905999719, "grad_norm": 1.2018719911575317, "learning_rate": 3.971374873500072e-05, "loss": 4.8637, "step": 218500 }, { "epoch": 15.38569622031755, "grad_norm": 1.24372398853302, "learning_rate": 3.95691773890415e-05, "loss": 4.8628, "step": 219000 }, { "epoch": 15.42082338063791, "grad_norm": 1.1655218601226807, "learning_rate": 3.942460604308226e-05, "loss": 4.8624, "step": 219500 }, { "epoch": 15.45595054095827, "grad_norm": 1.2862427234649658, "learning_rate": 3.928003469712303e-05, "loss": 4.8607, "step": 220000 }, { "epoch": 15.45595054095827, "eval_accuracy": 0.6688299408630574, "eval_loss": 4.5351457595825195, "eval_runtime": 125.5363, "eval_samples_per_second": 961.523, "eval_steps_per_second": 6.014, "step": 220000 }, { "epoch": 15.491077701278629, "grad_norm": 1.2285141944885254, "learning_rate": 3.9135463351163804e-05, "loss": 4.8619, "step": 220500 }, { "epoch": 15.526204861598988, "grad_norm": 1.221633791923523, "learning_rate": 3.899089200520457e-05, "loss": 4.8629, "step": 221000 }, { "epoch": 15.561332021919348, "grad_norm": 1.2441167831420898, "learning_rate": 3.884632065924534e-05, "loss": 4.861, "step": 221500 }, { "epoch": 15.596459182239707, "grad_norm": 1.2674016952514648, "learning_rate": 3.870174931328611e-05, "loss": 4.8629, "step": 222000 }, { "epoch": 15.631586342560068, "grad_norm": 1.1846632957458496, "learning_rate": 3.855717796732688e-05, "loss": 4.8589, "step": 222500 }, { "epoch": 15.666713502880427, "grad_norm": 1.1672122478485107, "learning_rate": 3.8412606621367644e-05, "loss": 4.8602, "step": 223000 }, { "epoch": 15.701840663200787, "grad_norm": 1.270010232925415, "learning_rate": 3.8268035275408415e-05, "loss": 4.8606, "step": 223500 }, { "epoch": 15.736967823521146, "grad_norm": 1.2058274745941162, "learning_rate": 3.812346392944918e-05, "loss": 4.8572, "step": 224000 }, { "epoch": 15.772094983841507, "grad_norm": 1.2046891450881958, "learning_rate": 3.797889258348996e-05, "loss": 4.8569, "step": 224500 }, { "epoch": 15.807222144161866, "grad_norm": 1.2354998588562012, "learning_rate": 3.783432123753072e-05, "loss": 4.8539, "step": 225000 }, { "epoch": 15.842349304482227, "grad_norm": 1.2390248775482178, "learning_rate": 3.768974989157149e-05, "loss": 4.8559, "step": 225500 }, { "epoch": 15.877476464802585, "grad_norm": 1.2330007553100586, "learning_rate": 3.754517854561226e-05, "loss": 4.8597, "step": 226000 }, { "epoch": 15.912603625122944, "grad_norm": 1.2747974395751953, "learning_rate": 3.740060719965303e-05, "loss": 4.8588, "step": 226500 }, { "epoch": 15.947730785443305, "grad_norm": 1.2696510553359985, "learning_rate": 3.72560358536938e-05, "loss": 4.8557, "step": 227000 }, { "epoch": 15.982857945763664, "grad_norm": 1.298751711845398, "learning_rate": 3.711146450773457e-05, "loss": 4.8577, "step": 227500 }, { "epoch": 16.017985106084023, "grad_norm": 1.2625064849853516, "learning_rate": 3.696689316177534e-05, "loss": 4.8527, "step": 228000 }, { "epoch": 16.053112266404383, "grad_norm": 1.2783324718475342, "learning_rate": 3.682232181581611e-05, "loss": 4.8509, "step": 228500 }, { "epoch": 16.088239426724744, "grad_norm": 1.2679533958435059, "learning_rate": 3.667775046985687e-05, "loss": 4.8525, "step": 229000 }, { "epoch": 16.123366587045105, "grad_norm": 1.2338887453079224, "learning_rate": 3.6533179123897644e-05, "loss": 4.8531, "step": 229500 }, { "epoch": 16.158493747365462, "grad_norm": 1.2011520862579346, "learning_rate": 3.6388607777938415e-05, "loss": 4.8551, "step": 230000 }, { "epoch": 16.158493747365462, "eval_accuracy": 0.6691269185696531, "eval_loss": 4.528936386108398, "eval_runtime": 140.3839, "eval_samples_per_second": 859.828, "eval_steps_per_second": 5.378, "step": 230000 }, { "epoch": 16.193620907685823, "grad_norm": 1.224282145500183, "learning_rate": 3.6244036431979186e-05, "loss": 4.85, "step": 230500 }, { "epoch": 16.228748068006183, "grad_norm": 1.3068790435791016, "learning_rate": 3.609946508601995e-05, "loss": 4.8521, "step": 231000 }, { "epoch": 16.26387522832654, "grad_norm": 1.2348567247390747, "learning_rate": 3.595489374006073e-05, "loss": 4.8483, "step": 231500 }, { "epoch": 16.2990023886469, "grad_norm": 1.1817059516906738, "learning_rate": 3.581032239410149e-05, "loss": 4.8532, "step": 232000 }, { "epoch": 16.33412954896726, "grad_norm": 1.21579110622406, "learning_rate": 3.566575104814226e-05, "loss": 4.8491, "step": 232500 }, { "epoch": 16.369256709287622, "grad_norm": 1.2975022792816162, "learning_rate": 3.5521179702183026e-05, "loss": 4.8494, "step": 233000 }, { "epoch": 16.40438386960798, "grad_norm": 1.1813398599624634, "learning_rate": 3.5376608356223804e-05, "loss": 4.8502, "step": 233500 }, { "epoch": 16.43951102992834, "grad_norm": 1.2424049377441406, "learning_rate": 3.523203701026457e-05, "loss": 4.8485, "step": 234000 }, { "epoch": 16.4746381902487, "grad_norm": 1.2299352884292603, "learning_rate": 3.508746566430534e-05, "loss": 4.8518, "step": 234500 }, { "epoch": 16.50976535056906, "grad_norm": 1.1961103677749634, "learning_rate": 3.49428943183461e-05, "loss": 4.8479, "step": 235000 }, { "epoch": 16.54489251088942, "grad_norm": 1.2365537881851196, "learning_rate": 3.479832297238687e-05, "loss": 4.8465, "step": 235500 }, { "epoch": 16.58001967120978, "grad_norm": 1.2166712284088135, "learning_rate": 3.4653751626427644e-05, "loss": 4.8506, "step": 236000 }, { "epoch": 16.61514683153014, "grad_norm": 1.269130825996399, "learning_rate": 3.4509180280468415e-05, "loss": 4.8463, "step": 236500 }, { "epoch": 16.650273991850497, "grad_norm": 1.2526973485946655, "learning_rate": 3.436460893450918e-05, "loss": 4.854, "step": 237000 }, { "epoch": 16.685401152170858, "grad_norm": 1.266256332397461, "learning_rate": 3.422003758854995e-05, "loss": 4.8488, "step": 237500 }, { "epoch": 16.72052831249122, "grad_norm": 1.2287768125534058, "learning_rate": 3.407546624259072e-05, "loss": 4.8481, "step": 238000 }, { "epoch": 16.75565547281158, "grad_norm": 1.2123924493789673, "learning_rate": 3.393089489663149e-05, "loss": 4.8491, "step": 238500 }, { "epoch": 16.790782633131936, "grad_norm": 1.2437433004379272, "learning_rate": 3.378632355067226e-05, "loss": 4.8455, "step": 239000 }, { "epoch": 16.825909793452297, "grad_norm": 1.2102261781692505, "learning_rate": 3.3641752204713026e-05, "loss": 4.8461, "step": 239500 }, { "epoch": 16.861036953772658, "grad_norm": 1.2507513761520386, "learning_rate": 3.3497180858753797e-05, "loss": 4.8461, "step": 240000 }, { "epoch": 16.861036953772658, "eval_accuracy": 0.670154007681976, "eval_loss": 4.523779392242432, "eval_runtime": 139.3663, "eval_samples_per_second": 866.106, "eval_steps_per_second": 5.417, "step": 240000 }, { "epoch": 16.89616411409302, "grad_norm": 1.2576217651367188, "learning_rate": 3.335260951279456e-05, "loss": 4.8448, "step": 240500 }, { "epoch": 16.931291274413375, "grad_norm": 1.1840087175369263, "learning_rate": 3.320803816683534e-05, "loss": 4.8504, "step": 241000 }, { "epoch": 16.966418434733736, "grad_norm": 1.219841480255127, "learning_rate": 3.30634668208761e-05, "loss": 4.8442, "step": 241500 }, { "epoch": 17.001545595054097, "grad_norm": 1.324711561203003, "learning_rate": 3.291889547491687e-05, "loss": 4.8492, "step": 242000 }, { "epoch": 17.036672755374454, "grad_norm": 1.2644296884536743, "learning_rate": 3.277432412895764e-05, "loss": 4.8466, "step": 242500 }, { "epoch": 17.071799915694815, "grad_norm": 1.2209409475326538, "learning_rate": 3.2629752782998414e-05, "loss": 4.8436, "step": 243000 }, { "epoch": 17.106927076015175, "grad_norm": 1.240304946899414, "learning_rate": 3.248518143703918e-05, "loss": 4.8422, "step": 243500 }, { "epoch": 17.142054236335536, "grad_norm": 1.2072173357009888, "learning_rate": 3.234061009107995e-05, "loss": 4.8406, "step": 244000 }, { "epoch": 17.177181396655893, "grad_norm": 1.2032321691513062, "learning_rate": 3.219603874512072e-05, "loss": 4.8399, "step": 244500 }, { "epoch": 17.212308556976254, "grad_norm": 1.2202192544937134, "learning_rate": 3.205146739916149e-05, "loss": 4.8439, "step": 245000 }, { "epoch": 17.247435717296614, "grad_norm": 1.2092161178588867, "learning_rate": 3.1906896053202255e-05, "loss": 4.8409, "step": 245500 }, { "epoch": 17.282562877616975, "grad_norm": 1.19815993309021, "learning_rate": 3.1762324707243026e-05, "loss": 4.8447, "step": 246000 }, { "epoch": 17.317690037937332, "grad_norm": 1.241196632385254, "learning_rate": 3.1617753361283796e-05, "loss": 4.8415, "step": 246500 }, { "epoch": 17.352817198257693, "grad_norm": 1.2433580160140991, "learning_rate": 3.147318201532457e-05, "loss": 4.8392, "step": 247000 }, { "epoch": 17.387944358578054, "grad_norm": 1.2644190788269043, "learning_rate": 3.132861066936533e-05, "loss": 4.8436, "step": 247500 }, { "epoch": 17.42307151889841, "grad_norm": 1.201249599456787, "learning_rate": 3.11840393234061e-05, "loss": 4.8339, "step": 248000 }, { "epoch": 17.45819867921877, "grad_norm": 1.3017292022705078, "learning_rate": 3.103946797744687e-05, "loss": 4.839, "step": 248500 }, { "epoch": 17.493325839539132, "grad_norm": 1.2682198286056519, "learning_rate": 3.0894896631487643e-05, "loss": 4.8368, "step": 249000 }, { "epoch": 17.528452999859493, "grad_norm": 1.2241114377975464, "learning_rate": 3.075032528552841e-05, "loss": 4.8371, "step": 249500 }, { "epoch": 17.56358016017985, "grad_norm": 1.1952778100967407, "learning_rate": 3.060575393956918e-05, "loss": 4.8365, "step": 250000 }, { "epoch": 17.56358016017985, "eval_accuracy": 0.6707725782357475, "eval_loss": 4.5180792808532715, "eval_runtime": 142.6244, "eval_samples_per_second": 846.321, "eval_steps_per_second": 5.294, "step": 250000 }, { "epoch": 17.59870732050021, "grad_norm": 1.3638995885849, "learning_rate": 3.046118259360995e-05, "loss": 4.8434, "step": 250500 }, { "epoch": 17.63383448082057, "grad_norm": 1.2575321197509766, "learning_rate": 3.0316611247650716e-05, "loss": 4.8335, "step": 251000 }, { "epoch": 17.668961641140932, "grad_norm": 1.225792407989502, "learning_rate": 3.0172039901691484e-05, "loss": 4.8384, "step": 251500 }, { "epoch": 17.70408880146129, "grad_norm": 1.3035650253295898, "learning_rate": 3.0027468555732258e-05, "loss": 4.8388, "step": 252000 }, { "epoch": 17.73921596178165, "grad_norm": 1.2521038055419922, "learning_rate": 2.9882897209773025e-05, "loss": 4.8373, "step": 252500 }, { "epoch": 17.77434312210201, "grad_norm": 1.2092581987380981, "learning_rate": 2.9738325863813793e-05, "loss": 4.835, "step": 253000 }, { "epoch": 17.809470282422367, "grad_norm": 1.2128022909164429, "learning_rate": 2.959375451785456e-05, "loss": 4.8384, "step": 253500 }, { "epoch": 17.844597442742728, "grad_norm": 1.3103076219558716, "learning_rate": 2.9449183171895334e-05, "loss": 4.8346, "step": 254000 }, { "epoch": 17.87972460306309, "grad_norm": 1.2214813232421875, "learning_rate": 2.93046118259361e-05, "loss": 4.8399, "step": 254500 }, { "epoch": 17.91485176338345, "grad_norm": 1.2692084312438965, "learning_rate": 2.916004047997687e-05, "loss": 4.8368, "step": 255000 }, { "epoch": 17.949978923703807, "grad_norm": 1.257507562637329, "learning_rate": 2.9015469134017636e-05, "loss": 4.8382, "step": 255500 }, { "epoch": 17.985106084024167, "grad_norm": 1.2650429010391235, "learning_rate": 2.887089778805841e-05, "loss": 4.8344, "step": 256000 }, { "epoch": 18.020233244344528, "grad_norm": 1.1996506452560425, "learning_rate": 2.8726326442099178e-05, "loss": 4.8362, "step": 256500 }, { "epoch": 18.05536040466489, "grad_norm": 1.244161605834961, "learning_rate": 2.8581755096139945e-05, "loss": 4.8355, "step": 257000 }, { "epoch": 18.090487564985246, "grad_norm": 1.2256453037261963, "learning_rate": 2.843718375018072e-05, "loss": 4.8318, "step": 257500 }, { "epoch": 18.125614725305606, "grad_norm": 1.2700932025909424, "learning_rate": 2.8292612404221487e-05, "loss": 4.8341, "step": 258000 }, { "epoch": 18.160741885625967, "grad_norm": 1.227278709411621, "learning_rate": 2.8148041058262254e-05, "loss": 4.8326, "step": 258500 }, { "epoch": 18.195869045946324, "grad_norm": 1.237528681755066, "learning_rate": 2.800346971230302e-05, "loss": 4.8338, "step": 259000 }, { "epoch": 18.230996206266685, "grad_norm": 1.278845191001892, "learning_rate": 2.7858898366343792e-05, "loss": 4.8332, "step": 259500 }, { "epoch": 18.266123366587045, "grad_norm": 1.2519580125808716, "learning_rate": 2.7714327020384563e-05, "loss": 4.831, "step": 260000 }, { "epoch": 18.266123366587045, "eval_accuracy": 0.6715583971956572, "eval_loss": 4.512533664703369, "eval_runtime": 138.5668, "eval_samples_per_second": 871.103, "eval_steps_per_second": 5.449, "step": 260000 }, { "epoch": 18.301250526907406, "grad_norm": 1.2077840566635132, "learning_rate": 2.756975567442533e-05, "loss": 4.8311, "step": 260500 }, { "epoch": 18.336377687227763, "grad_norm": 1.2855967283248901, "learning_rate": 2.7425184328466098e-05, "loss": 4.8283, "step": 261000 }, { "epoch": 18.371504847548124, "grad_norm": 1.2155075073242188, "learning_rate": 2.728061298250687e-05, "loss": 4.8347, "step": 261500 }, { "epoch": 18.406632007868485, "grad_norm": 1.2062431573867798, "learning_rate": 2.7136041636547636e-05, "loss": 4.8312, "step": 262000 }, { "epoch": 18.441759168188845, "grad_norm": 1.2710195779800415, "learning_rate": 2.6991470290588407e-05, "loss": 4.8291, "step": 262500 }, { "epoch": 18.476886328509202, "grad_norm": 1.3219693899154663, "learning_rate": 2.6846898944629178e-05, "loss": 4.8304, "step": 263000 }, { "epoch": 18.512013488829563, "grad_norm": 1.2231227159500122, "learning_rate": 2.6702327598669945e-05, "loss": 4.8274, "step": 263500 }, { "epoch": 18.547140649149924, "grad_norm": 1.2857335805892944, "learning_rate": 2.6557756252710713e-05, "loss": 4.8272, "step": 264000 }, { "epoch": 18.58226780947028, "grad_norm": 1.2383677959442139, "learning_rate": 2.641318490675148e-05, "loss": 4.8276, "step": 264500 }, { "epoch": 18.61739496979064, "grad_norm": 1.263206124305725, "learning_rate": 2.6268613560792254e-05, "loss": 4.831, "step": 265000 }, { "epoch": 18.652522130111002, "grad_norm": 1.2725106477737427, "learning_rate": 2.612404221483302e-05, "loss": 4.8289, "step": 265500 }, { "epoch": 18.687649290431363, "grad_norm": 1.2537930011749268, "learning_rate": 2.597947086887379e-05, "loss": 4.8268, "step": 266000 }, { "epoch": 18.72277645075172, "grad_norm": 1.2264420986175537, "learning_rate": 2.5834899522914556e-05, "loss": 4.8299, "step": 266500 }, { "epoch": 18.75790361107208, "grad_norm": 1.1995207071304321, "learning_rate": 2.569032817695533e-05, "loss": 4.8297, "step": 267000 }, { "epoch": 18.79303077139244, "grad_norm": 1.234877586364746, "learning_rate": 2.5545756830996098e-05, "loss": 4.8308, "step": 267500 }, { "epoch": 18.828157931712802, "grad_norm": 1.2074074745178223, "learning_rate": 2.5401185485036865e-05, "loss": 4.8289, "step": 268000 }, { "epoch": 18.86328509203316, "grad_norm": 1.345072865486145, "learning_rate": 2.5256614139077633e-05, "loss": 4.8268, "step": 268500 }, { "epoch": 18.89841225235352, "grad_norm": 1.2629283666610718, "learning_rate": 2.5112042793118407e-05, "loss": 4.8289, "step": 269000 }, { "epoch": 18.93353941267388, "grad_norm": 1.228371024131775, "learning_rate": 2.4967471447159174e-05, "loss": 4.8281, "step": 269500 }, { "epoch": 18.968666572994238, "grad_norm": 1.233062982559204, "learning_rate": 2.4822900101199945e-05, "loss": 4.8298, "step": 270000 }, { "epoch": 18.968666572994238, "eval_accuracy": 0.6723839472789862, "eval_loss": 4.508018493652344, "eval_runtime": 136.0801, "eval_samples_per_second": 887.022, "eval_steps_per_second": 5.548, "step": 270000 }, { "epoch": 19.0037937333146, "grad_norm": 1.3693454265594482, "learning_rate": 2.4678328755240712e-05, "loss": 4.8269, "step": 270500 }, { "epoch": 19.03892089363496, "grad_norm": 1.310105323791504, "learning_rate": 2.4533757409281483e-05, "loss": 4.8242, "step": 271000 }, { "epoch": 19.07404805395532, "grad_norm": 1.2209819555282593, "learning_rate": 2.438918606332225e-05, "loss": 4.8255, "step": 271500 }, { "epoch": 19.109175214275677, "grad_norm": 1.2459584474563599, "learning_rate": 2.424461471736302e-05, "loss": 4.8238, "step": 272000 }, { "epoch": 19.144302374596037, "grad_norm": 1.2385457754135132, "learning_rate": 2.410004337140379e-05, "loss": 4.8248, "step": 272500 }, { "epoch": 19.179429534916398, "grad_norm": 1.3180341720581055, "learning_rate": 2.395547202544456e-05, "loss": 4.8253, "step": 273000 }, { "epoch": 19.21455669523676, "grad_norm": 1.225286602973938, "learning_rate": 2.3810900679485327e-05, "loss": 4.8206, "step": 273500 }, { "epoch": 19.249683855557116, "grad_norm": 1.2601996660232544, "learning_rate": 2.3666329333526098e-05, "loss": 4.8227, "step": 274000 }, { "epoch": 19.284811015877477, "grad_norm": 1.2234896421432495, "learning_rate": 2.3521757987566865e-05, "loss": 4.8237, "step": 274500 }, { "epoch": 19.319938176197837, "grad_norm": 1.2657651901245117, "learning_rate": 2.3377186641607636e-05, "loss": 4.8252, "step": 275000 }, { "epoch": 19.355065336518194, "grad_norm": 1.251855731010437, "learning_rate": 2.3232615295648403e-05, "loss": 4.8267, "step": 275500 }, { "epoch": 19.390192496838555, "grad_norm": 1.2351305484771729, "learning_rate": 2.3088043949689174e-05, "loss": 4.8192, "step": 276000 }, { "epoch": 19.425319657158916, "grad_norm": 1.2252118587493896, "learning_rate": 2.2943472603729945e-05, "loss": 4.8251, "step": 276500 }, { "epoch": 19.460446817479276, "grad_norm": 1.2224647998809814, "learning_rate": 2.2798901257770712e-05, "loss": 4.8202, "step": 277000 }, { "epoch": 19.495573977799634, "grad_norm": 1.242702841758728, "learning_rate": 2.265432991181148e-05, "loss": 4.8236, "step": 277500 }, { "epoch": 19.530701138119994, "grad_norm": 1.2677148580551147, "learning_rate": 2.250975856585225e-05, "loss": 4.8203, "step": 278000 }, { "epoch": 19.565828298440355, "grad_norm": 1.3082236051559448, "learning_rate": 2.2365187219893018e-05, "loss": 4.8243, "step": 278500 }, { "epoch": 19.600955458760716, "grad_norm": 1.269713044166565, "learning_rate": 2.2220615873933785e-05, "loss": 4.8195, "step": 279000 }, { "epoch": 19.636082619081073, "grad_norm": 1.246695876121521, "learning_rate": 2.2076044527974556e-05, "loss": 4.8237, "step": 279500 }, { "epoch": 19.671209779401433, "grad_norm": 1.250370979309082, "learning_rate": 2.1931473182015323e-05, "loss": 4.82, "step": 280000 }, { "epoch": 19.671209779401433, "eval_accuracy": 0.6732291231476323, "eval_loss": 4.502586364746094, "eval_runtime": 136.5332, "eval_samples_per_second": 884.078, "eval_steps_per_second": 5.53, "step": 280000 }, { "epoch": 19.706336939721794, "grad_norm": 1.2715970277786255, "learning_rate": 2.1786901836056094e-05, "loss": 4.823, "step": 280500 }, { "epoch": 19.74146410004215, "grad_norm": 1.2604025602340698, "learning_rate": 2.164233049009686e-05, "loss": 4.8229, "step": 281000 }, { "epoch": 19.776591260362512, "grad_norm": 1.2717580795288086, "learning_rate": 2.1497759144137632e-05, "loss": 4.821, "step": 281500 }, { "epoch": 19.811718420682872, "grad_norm": 1.2805583477020264, "learning_rate": 2.1353187798178403e-05, "loss": 4.8195, "step": 282000 }, { "epoch": 19.846845581003233, "grad_norm": 1.2786941528320312, "learning_rate": 2.120861645221917e-05, "loss": 4.8188, "step": 282500 }, { "epoch": 19.88197274132359, "grad_norm": 1.2619529962539673, "learning_rate": 2.106404510625994e-05, "loss": 4.8183, "step": 283000 }, { "epoch": 19.91709990164395, "grad_norm": 1.3264570236206055, "learning_rate": 2.091947376030071e-05, "loss": 4.8225, "step": 283500 }, { "epoch": 19.95222706196431, "grad_norm": 1.3388673067092896, "learning_rate": 2.077490241434148e-05, "loss": 4.8182, "step": 284000 }, { "epoch": 19.987354222284672, "grad_norm": 1.2571300268173218, "learning_rate": 2.0630331068382247e-05, "loss": 4.8186, "step": 284500 }, { "epoch": 20.02248138260503, "grad_norm": 1.260873794555664, "learning_rate": 2.0485759722423017e-05, "loss": 4.8216, "step": 285000 }, { "epoch": 20.05760854292539, "grad_norm": 1.2571736574172974, "learning_rate": 2.0341188376463785e-05, "loss": 4.8159, "step": 285500 }, { "epoch": 20.09273570324575, "grad_norm": 1.285014033317566, "learning_rate": 2.0196617030504556e-05, "loss": 4.8155, "step": 286000 }, { "epoch": 20.127862863566108, "grad_norm": 1.2193158864974976, "learning_rate": 2.0052045684545323e-05, "loss": 4.8167, "step": 286500 }, { "epoch": 20.16299002388647, "grad_norm": 1.2150843143463135, "learning_rate": 1.9907474338586094e-05, "loss": 4.8149, "step": 287000 }, { "epoch": 20.19811718420683, "grad_norm": 1.2412134408950806, "learning_rate": 1.976290299262686e-05, "loss": 4.8146, "step": 287500 }, { "epoch": 20.23324434452719, "grad_norm": 1.2893328666687012, "learning_rate": 1.9618331646667632e-05, "loss": 4.8165, "step": 288000 }, { "epoch": 20.268371504847547, "grad_norm": 1.259504795074463, "learning_rate": 1.9473760300708403e-05, "loss": 4.8139, "step": 288500 }, { "epoch": 20.303498665167908, "grad_norm": 1.2881910800933838, "learning_rate": 1.932918895474917e-05, "loss": 4.8192, "step": 289000 }, { "epoch": 20.33862582548827, "grad_norm": 1.3316067457199097, "learning_rate": 1.918461760878994e-05, "loss": 4.8155, "step": 289500 }, { "epoch": 20.37375298580863, "grad_norm": 1.2800790071487427, "learning_rate": 1.9040046262830708e-05, "loss": 4.8168, "step": 290000 }, { "epoch": 20.37375298580863, "eval_accuracy": 0.6737959000986882, "eval_loss": 4.496816158294678, "eval_runtime": 134.4386, "eval_samples_per_second": 897.852, "eval_steps_per_second": 5.616, "step": 290000 }, { "epoch": 20.408880146128986, "grad_norm": 1.2663326263427734, "learning_rate": 1.889547491687148e-05, "loss": 4.8151, "step": 290500 }, { "epoch": 20.444007306449347, "grad_norm": 1.2812772989273071, "learning_rate": 1.8750903570912246e-05, "loss": 4.8177, "step": 291000 }, { "epoch": 20.479134466769708, "grad_norm": 1.2344677448272705, "learning_rate": 1.8606332224953017e-05, "loss": 4.8119, "step": 291500 }, { "epoch": 20.514261627090065, "grad_norm": 1.2526355981826782, "learning_rate": 1.8461760878993785e-05, "loss": 4.8153, "step": 292000 }, { "epoch": 20.549388787410425, "grad_norm": 1.2275665998458862, "learning_rate": 1.8317189533034555e-05, "loss": 4.812, "step": 292500 }, { "epoch": 20.584515947730786, "grad_norm": 1.287656545639038, "learning_rate": 1.8172618187075323e-05, "loss": 4.8157, "step": 293000 }, { "epoch": 20.619643108051147, "grad_norm": 1.3148868083953857, "learning_rate": 1.8028046841116093e-05, "loss": 4.8162, "step": 293500 }, { "epoch": 20.654770268371504, "grad_norm": 1.3041707277297974, "learning_rate": 1.788347549515686e-05, "loss": 4.8126, "step": 294000 }, { "epoch": 20.689897428691864, "grad_norm": 1.2801120281219482, "learning_rate": 1.7738904149197628e-05, "loss": 4.8153, "step": 294500 }, { "epoch": 20.725024589012225, "grad_norm": 1.2438610792160034, "learning_rate": 1.75943328032384e-05, "loss": 4.8138, "step": 295000 }, { "epoch": 20.760151749332586, "grad_norm": 1.2917931079864502, "learning_rate": 1.7449761457279166e-05, "loss": 4.8161, "step": 295500 }, { "epoch": 20.795278909652943, "grad_norm": 1.3261077404022217, "learning_rate": 1.7305190111319937e-05, "loss": 4.8102, "step": 296000 }, { "epoch": 20.830406069973304, "grad_norm": 1.2542301416397095, "learning_rate": 1.7160618765360705e-05, "loss": 4.8137, "step": 296500 }, { "epoch": 20.865533230293664, "grad_norm": 1.212733268737793, "learning_rate": 1.7016047419401475e-05, "loss": 4.8082, "step": 297000 }, { "epoch": 20.90066039061402, "grad_norm": 1.261491298675537, "learning_rate": 1.6871476073442243e-05, "loss": 4.8146, "step": 297500 }, { "epoch": 20.935787550934382, "grad_norm": 1.2800852060317993, "learning_rate": 1.6726904727483014e-05, "loss": 4.813, "step": 298000 }, { "epoch": 20.970914711254743, "grad_norm": 1.2858755588531494, "learning_rate": 1.658233338152378e-05, "loss": 4.815, "step": 298500 }, { "epoch": 21.006041871575103, "grad_norm": 1.3007336854934692, "learning_rate": 1.6437762035564552e-05, "loss": 4.8123, "step": 299000 }, { "epoch": 21.04116903189546, "grad_norm": 1.2589894533157349, "learning_rate": 1.629319068960532e-05, "loss": 4.8127, "step": 299500 }, { "epoch": 21.07629619221582, "grad_norm": 1.2507349252700806, "learning_rate": 1.614861934364609e-05, "loss": 4.8087, "step": 300000 }, { "epoch": 21.07629619221582, "eval_accuracy": 0.6745277977453018, "eval_loss": 4.493993759155273, "eval_runtime": 135.2696, "eval_samples_per_second": 892.336, "eval_steps_per_second": 5.581, "step": 300000 }, { "epoch": 21.111423352536182, "grad_norm": 1.3088561296463013, "learning_rate": 1.6004047997686857e-05, "loss": 4.8094, "step": 300500 }, { "epoch": 21.14655051285654, "grad_norm": 1.2399375438690186, "learning_rate": 1.5859476651727628e-05, "loss": 4.8104, "step": 301000 }, { "epoch": 21.1816776731769, "grad_norm": 1.2422513961791992, "learning_rate": 1.57149053057684e-05, "loss": 4.8105, "step": 301500 }, { "epoch": 21.21680483349726, "grad_norm": 1.1982223987579346, "learning_rate": 1.5570333959809166e-05, "loss": 4.8143, "step": 302000 }, { "epoch": 21.25193199381762, "grad_norm": 1.2185834646224976, "learning_rate": 1.5425762613849937e-05, "loss": 4.8068, "step": 302500 }, { "epoch": 21.287059154137978, "grad_norm": 1.3271536827087402, "learning_rate": 1.5281191267890704e-05, "loss": 4.8093, "step": 303000 }, { "epoch": 21.32218631445834, "grad_norm": 1.2299219369888306, "learning_rate": 1.5136619921931475e-05, "loss": 4.8083, "step": 303500 }, { "epoch": 21.3573134747787, "grad_norm": 1.263633370399475, "learning_rate": 1.4992048575972243e-05, "loss": 4.8062, "step": 304000 }, { "epoch": 21.39244063509906, "grad_norm": 1.2906148433685303, "learning_rate": 1.4847477230013013e-05, "loss": 4.8111, "step": 304500 }, { "epoch": 21.427567795419417, "grad_norm": 1.2754861116409302, "learning_rate": 1.470290588405378e-05, "loss": 4.8077, "step": 305000 }, { "epoch": 21.462694955739778, "grad_norm": 1.2857871055603027, "learning_rate": 1.4558334538094551e-05, "loss": 4.8082, "step": 305500 }, { "epoch": 21.49782211606014, "grad_norm": 1.3503425121307373, "learning_rate": 1.4413763192135319e-05, "loss": 4.8071, "step": 306000 }, { "epoch": 21.532949276380496, "grad_norm": 1.2912927865982056, "learning_rate": 1.4269191846176088e-05, "loss": 4.8067, "step": 306500 }, { "epoch": 21.568076436700856, "grad_norm": 1.2711697816848755, "learning_rate": 1.4124620500216857e-05, "loss": 4.8069, "step": 307000 }, { "epoch": 21.603203597021217, "grad_norm": 1.3177986145019531, "learning_rate": 1.3980049154257626e-05, "loss": 4.8072, "step": 307500 }, { "epoch": 21.638330757341578, "grad_norm": 1.2948944568634033, "learning_rate": 1.3835477808298397e-05, "loss": 4.815, "step": 308000 }, { "epoch": 21.673457917661935, "grad_norm": 1.2391667366027832, "learning_rate": 1.3690906462339164e-05, "loss": 4.8091, "step": 308500 }, { "epoch": 21.708585077982296, "grad_norm": 1.2883468866348267, "learning_rate": 1.3546335116379935e-05, "loss": 4.8072, "step": 309000 }, { "epoch": 21.743712238302656, "grad_norm": 1.3043705224990845, "learning_rate": 1.3401763770420702e-05, "loss": 4.8036, "step": 309500 }, { "epoch": 21.778839398623017, "grad_norm": 1.2346197366714478, "learning_rate": 1.3257192424461473e-05, "loss": 4.8065, "step": 310000 }, { "epoch": 21.778839398623017, "eval_accuracy": 0.6752954696307744, "eval_loss": 4.48726224899292, "eval_runtime": 139.5292, "eval_samples_per_second": 865.095, "eval_steps_per_second": 5.411, "step": 310000 }, { "epoch": 21.813966558943374, "grad_norm": 1.3174384832382202, "learning_rate": 1.311262107850224e-05, "loss": 4.8037, "step": 310500 }, { "epoch": 21.849093719263735, "grad_norm": 1.2581169605255127, "learning_rate": 1.2968049732543011e-05, "loss": 4.8036, "step": 311000 }, { "epoch": 21.884220879584095, "grad_norm": 1.2498201131820679, "learning_rate": 1.2823478386583779e-05, "loss": 4.8085, "step": 311500 }, { "epoch": 21.919348039904452, "grad_norm": 1.2525370121002197, "learning_rate": 1.267890704062455e-05, "loss": 4.804, "step": 312000 }, { "epoch": 21.954475200224813, "grad_norm": 1.2862482070922852, "learning_rate": 1.2534335694665317e-05, "loss": 4.8054, "step": 312500 }, { "epoch": 21.989602360545174, "grad_norm": 1.230281949043274, "learning_rate": 1.2389764348706088e-05, "loss": 4.8076, "step": 313000 }, { "epoch": 22.024729520865534, "grad_norm": 1.294276475906372, "learning_rate": 1.2245193002746857e-05, "loss": 4.8032, "step": 313500 }, { "epoch": 22.05985668118589, "grad_norm": 1.3363004922866821, "learning_rate": 1.2100621656787626e-05, "loss": 4.8027, "step": 314000 }, { "epoch": 22.094983841506252, "grad_norm": 1.3100172281265259, "learning_rate": 1.1956050310828395e-05, "loss": 4.8059, "step": 314500 }, { "epoch": 22.130111001826613, "grad_norm": 1.2850050926208496, "learning_rate": 1.1811478964869162e-05, "loss": 4.8029, "step": 315000 }, { "epoch": 22.165238162146974, "grad_norm": 1.2826613187789917, "learning_rate": 1.1666907618909931e-05, "loss": 4.8002, "step": 315500 }, { "epoch": 22.20036532246733, "grad_norm": 1.2550464868545532, "learning_rate": 1.15223362729507e-05, "loss": 4.8019, "step": 316000 }, { "epoch": 22.23549248278769, "grad_norm": 1.22563898563385, "learning_rate": 1.137776492699147e-05, "loss": 4.8032, "step": 316500 }, { "epoch": 22.270619643108052, "grad_norm": 1.2782310247421265, "learning_rate": 1.123319358103224e-05, "loss": 4.8045, "step": 317000 }, { "epoch": 22.30574680342841, "grad_norm": 1.2926520109176636, "learning_rate": 1.108862223507301e-05, "loss": 4.8031, "step": 317500 }, { "epoch": 22.34087396374877, "grad_norm": 1.2573494911193848, "learning_rate": 1.0944050889113779e-05, "loss": 4.8027, "step": 318000 }, { "epoch": 22.37600112406913, "grad_norm": 1.275132179260254, "learning_rate": 1.0799479543154548e-05, "loss": 4.8044, "step": 318500 }, { "epoch": 22.41112828438949, "grad_norm": 1.2956249713897705, "learning_rate": 1.0654908197195317e-05, "loss": 4.8005, "step": 319000 }, { "epoch": 22.44625544470985, "grad_norm": 1.2665064334869385, "learning_rate": 1.0510336851236086e-05, "loss": 4.8007, "step": 319500 }, { "epoch": 22.48138260503021, "grad_norm": 1.3498852252960205, "learning_rate": 1.0365765505276855e-05, "loss": 4.7999, "step": 320000 }, { "epoch": 22.48138260503021, "eval_accuracy": 0.6754618608508858, "eval_loss": 4.484737873077393, "eval_runtime": 136.5284, "eval_samples_per_second": 884.109, "eval_steps_per_second": 5.53, "step": 320000 }, { "epoch": 22.51650976535057, "grad_norm": 1.308068871498108, "learning_rate": 1.0221194159317624e-05, "loss": 4.8019, "step": 320500 }, { "epoch": 22.55163692567093, "grad_norm": 1.3431742191314697, "learning_rate": 1.0076622813358393e-05, "loss": 4.806, "step": 321000 }, { "epoch": 22.586764085991287, "grad_norm": 1.2667869329452515, "learning_rate": 9.932051467399162e-06, "loss": 4.8039, "step": 321500 }, { "epoch": 22.621891246311648, "grad_norm": 1.209094762802124, "learning_rate": 9.787480121439931e-06, "loss": 4.8, "step": 322000 }, { "epoch": 22.65701840663201, "grad_norm": 1.2730494737625122, "learning_rate": 9.6429087754807e-06, "loss": 4.8026, "step": 322500 }, { "epoch": 22.692145566952366, "grad_norm": 1.3531582355499268, "learning_rate": 9.49833742952147e-06, "loss": 4.8005, "step": 323000 }, { "epoch": 22.727272727272727, "grad_norm": 1.2946540117263794, "learning_rate": 9.353766083562238e-06, "loss": 4.7994, "step": 323500 }, { "epoch": 22.762399887593087, "grad_norm": 1.2677605152130127, "learning_rate": 9.209194737603008e-06, "loss": 4.7982, "step": 324000 }, { "epoch": 22.797527047913448, "grad_norm": 1.3184878826141357, "learning_rate": 9.064623391643777e-06, "loss": 4.8019, "step": 324500 }, { "epoch": 22.832654208233805, "grad_norm": 1.250824213027954, "learning_rate": 8.920052045684546e-06, "loss": 4.8016, "step": 325000 }, { "epoch": 22.867781368554166, "grad_norm": 1.3487441539764404, "learning_rate": 8.775480699725315e-06, "loss": 4.8011, "step": 325500 }, { "epoch": 22.902908528874526, "grad_norm": 1.2791815996170044, "learning_rate": 8.630909353766084e-06, "loss": 4.8026, "step": 326000 }, { "epoch": 22.938035689194887, "grad_norm": 1.257918357849121, "learning_rate": 8.486338007806853e-06, "loss": 4.8003, "step": 326500 }, { "epoch": 22.973162849515244, "grad_norm": 1.2757270336151123, "learning_rate": 8.341766661847622e-06, "loss": 4.7951, "step": 327000 }, { "epoch": 23.008290009835605, "grad_norm": 1.2635070085525513, "learning_rate": 8.197195315888391e-06, "loss": 4.8015, "step": 327500 }, { "epoch": 23.043417170155966, "grad_norm": 1.2947243452072144, "learning_rate": 8.05262396992916e-06, "loss": 4.8004, "step": 328000 }, { "epoch": 23.078544330476323, "grad_norm": 1.2774442434310913, "learning_rate": 7.90805262396993e-06, "loss": 4.7951, "step": 328500 }, { "epoch": 23.113671490796683, "grad_norm": 1.2870813608169556, "learning_rate": 7.763481278010698e-06, "loss": 4.799, "step": 329000 }, { "epoch": 23.148798651117044, "grad_norm": 1.3238667249679565, "learning_rate": 7.618909932051468e-06, "loss": 4.8019, "step": 329500 }, { "epoch": 23.183925811437405, "grad_norm": 1.2685827016830444, "learning_rate": 7.474338586092237e-06, "loss": 4.798, "step": 330000 }, { "epoch": 23.183925811437405, "eval_accuracy": 0.6758814341201937, "eval_loss": 4.481825351715088, "eval_runtime": 137.5505, "eval_samples_per_second": 877.539, "eval_steps_per_second": 5.489, "step": 330000 }, { "epoch": 23.219052971757762, "grad_norm": 1.3054661750793457, "learning_rate": 7.3297672401330065e-06, "loss": 4.7937, "step": 330500 }, { "epoch": 23.254180132078123, "grad_norm": 1.2993879318237305, "learning_rate": 7.1851958941737756e-06, "loss": 4.7985, "step": 331000 }, { "epoch": 23.289307292398483, "grad_norm": 1.2552520036697388, "learning_rate": 7.040624548214545e-06, "loss": 4.798, "step": 331500 }, { "epoch": 23.324434452718844, "grad_norm": 1.2615892887115479, "learning_rate": 6.896053202255314e-06, "loss": 4.8014, "step": 332000 }, { "epoch": 23.3595616130392, "grad_norm": 1.2777199745178223, "learning_rate": 6.751481856296083e-06, "loss": 4.7981, "step": 332500 }, { "epoch": 23.39468877335956, "grad_norm": 1.2996366024017334, "learning_rate": 6.606910510336851e-06, "loss": 4.794, "step": 333000 }, { "epoch": 23.429815933679922, "grad_norm": 1.2594910860061646, "learning_rate": 6.46233916437762e-06, "loss": 4.7968, "step": 333500 }, { "epoch": 23.46494309400028, "grad_norm": 1.358947992324829, "learning_rate": 6.317767818418389e-06, "loss": 4.7968, "step": 334000 }, { "epoch": 23.50007025432064, "grad_norm": 1.2598515748977661, "learning_rate": 6.173196472459159e-06, "loss": 4.7983, "step": 334500 }, { "epoch": 23.535197414641, "grad_norm": 1.243940830230713, "learning_rate": 6.028625126499928e-06, "loss": 4.8009, "step": 335000 }, { "epoch": 23.57032457496136, "grad_norm": 1.2729179859161377, "learning_rate": 5.884053780540697e-06, "loss": 4.7973, "step": 335500 }, { "epoch": 23.60545173528172, "grad_norm": 1.276167631149292, "learning_rate": 5.739482434581466e-06, "loss": 4.7946, "step": 336000 }, { "epoch": 23.64057889560208, "grad_norm": 1.301024317741394, "learning_rate": 5.5949110886222355e-06, "loss": 4.7929, "step": 336500 }, { "epoch": 23.67570605592244, "grad_norm": 1.3131747245788574, "learning_rate": 5.4503397426630046e-06, "loss": 4.7966, "step": 337000 }, { "epoch": 23.7108332162428, "grad_norm": 1.2923544645309448, "learning_rate": 5.305768396703774e-06, "loss": 4.7921, "step": 337500 }, { "epoch": 23.745960376563158, "grad_norm": 1.3842499256134033, "learning_rate": 5.161197050744543e-06, "loss": 4.7927, "step": 338000 }, { "epoch": 23.78108753688352, "grad_norm": 1.2835748195648193, "learning_rate": 5.016625704785312e-06, "loss": 4.7935, "step": 338500 }, { "epoch": 23.81621469720388, "grad_norm": 1.265910267829895, "learning_rate": 4.872054358826081e-06, "loss": 4.7959, "step": 339000 }, { "epoch": 23.851341857524236, "grad_norm": 1.2558317184448242, "learning_rate": 4.72748301286685e-06, "loss": 4.7955, "step": 339500 }, { "epoch": 23.886469017844597, "grad_norm": 1.2944668531417847, "learning_rate": 4.582911666907619e-06, "loss": 4.7951, "step": 340000 }, { "epoch": 23.886469017844597, "eval_accuracy": 0.6763557003560086, "eval_loss": 4.478089809417725, "eval_runtime": 136.4338, "eval_samples_per_second": 884.722, "eval_steps_per_second": 5.534, "step": 340000 }, { "epoch": 23.921596178164958, "grad_norm": 1.274920105934143, "learning_rate": 4.438340320948388e-06, "loss": 4.794, "step": 340500 }, { "epoch": 23.956723338485318, "grad_norm": 1.2540873289108276, "learning_rate": 4.293768974989157e-06, "loss": 4.796, "step": 341000 }, { "epoch": 23.991850498805675, "grad_norm": 1.3056056499481201, "learning_rate": 4.149197629029926e-06, "loss": 4.7953, "step": 341500 }, { "epoch": 24.026977659126036, "grad_norm": 1.2587436437606812, "learning_rate": 4.004626283070695e-06, "loss": 4.7968, "step": 342000 }, { "epoch": 24.062104819446397, "grad_norm": 1.273866891860962, "learning_rate": 3.8600549371114645e-06, "loss": 4.7943, "step": 342500 }, { "epoch": 24.097231979766757, "grad_norm": 1.2160638570785522, "learning_rate": 3.7154835911522336e-06, "loss": 4.7928, "step": 343000 }, { "epoch": 24.132359140087114, "grad_norm": 1.2497260570526123, "learning_rate": 3.5709122451930026e-06, "loss": 4.7953, "step": 343500 }, { "epoch": 24.167486300407475, "grad_norm": 1.2620630264282227, "learning_rate": 3.426340899233772e-06, "loss": 4.7975, "step": 344000 }, { "epoch": 24.202613460727836, "grad_norm": 1.2412761449813843, "learning_rate": 3.2817695532745412e-06, "loss": 4.7947, "step": 344500 }, { "epoch": 24.237740621048193, "grad_norm": 1.2190656661987305, "learning_rate": 3.1371982073153103e-06, "loss": 4.7977, "step": 345000 }, { "epoch": 24.272867781368554, "grad_norm": 1.2885154485702515, "learning_rate": 2.9926268613560794e-06, "loss": 4.7933, "step": 345500 }, { "epoch": 24.307994941688914, "grad_norm": 1.2771856784820557, "learning_rate": 2.8480555153968485e-06, "loss": 4.7928, "step": 346000 }, { "epoch": 24.343122102009275, "grad_norm": 1.298667311668396, "learning_rate": 2.7034841694376176e-06, "loss": 4.7909, "step": 346500 }, { "epoch": 24.378249262329632, "grad_norm": 1.2902612686157227, "learning_rate": 2.5589128234783866e-06, "loss": 4.7941, "step": 347000 }, { "epoch": 24.413376422649993, "grad_norm": 1.3076661825180054, "learning_rate": 2.414341477519156e-06, "loss": 4.7917, "step": 347500 }, { "epoch": 24.448503582970353, "grad_norm": 1.2441986799240112, "learning_rate": 2.269770131559925e-06, "loss": 4.7944, "step": 348000 }, { "epoch": 24.483630743290714, "grad_norm": 1.2361170053482056, "learning_rate": 2.125198785600694e-06, "loss": 4.7908, "step": 348500 }, { "epoch": 24.51875790361107, "grad_norm": 1.2300372123718262, "learning_rate": 1.980627439641463e-06, "loss": 4.7903, "step": 349000 }, { "epoch": 24.553885063931432, "grad_norm": 1.2479238510131836, "learning_rate": 1.8360560936822323e-06, "loss": 4.791, "step": 349500 }, { "epoch": 24.589012224251793, "grad_norm": 1.2789229154586792, "learning_rate": 1.6914847477230013e-06, "loss": 4.7946, "step": 350000 }, { "epoch": 24.589012224251793, "eval_accuracy": 0.6768395395626101, "eval_loss": 4.475473880767822, "eval_runtime": 136.401, "eval_samples_per_second": 884.935, "eval_steps_per_second": 5.535, "step": 350000 }, { "epoch": 24.62413938457215, "grad_norm": 1.290574312210083, "learning_rate": 1.5469134017637704e-06, "loss": 4.7913, "step": 350500 }, { "epoch": 24.65926654489251, "grad_norm": 1.2345504760742188, "learning_rate": 1.4023420558045395e-06, "loss": 4.7912, "step": 351000 }, { "epoch": 24.69439370521287, "grad_norm": 1.2862014770507812, "learning_rate": 1.2577707098453088e-06, "loss": 4.7922, "step": 351500 }, { "epoch": 24.72952086553323, "grad_norm": 1.3092856407165527, "learning_rate": 1.1131993638860779e-06, "loss": 4.7912, "step": 352000 }, { "epoch": 24.76464802585359, "grad_norm": 1.29878568649292, "learning_rate": 9.68628017926847e-07, "loss": 4.7924, "step": 352500 }, { "epoch": 24.79977518617395, "grad_norm": 1.2417224645614624, "learning_rate": 8.240566719676161e-07, "loss": 4.7923, "step": 353000 }, { "epoch": 24.83490234649431, "grad_norm": 1.25477135181427, "learning_rate": 6.794853260083851e-07, "loss": 4.7905, "step": 353500 }, { "epoch": 24.870029506814667, "grad_norm": 1.244131088256836, "learning_rate": 5.349139800491543e-07, "loss": 4.7931, "step": 354000 }, { "epoch": 24.905156667135028, "grad_norm": 1.2467916011810303, "learning_rate": 3.903426340899234e-07, "loss": 4.7926, "step": 354500 }, { "epoch": 24.94028382745539, "grad_norm": 1.3050836324691772, "learning_rate": 2.457712881306925e-07, "loss": 4.7885, "step": 355000 }, { "epoch": 24.97541098777575, "grad_norm": 1.197683572769165, "learning_rate": 1.0119994217146162e-07, "loss": 4.7916, "step": 355500 }, { "epoch": 25.0, "step": 355850, "total_flos": 3.041765415786701e+18, "train_loss": 5.218619978394052, "train_runtime": 91292.1269, "train_samples_per_second": 623.641, "train_steps_per_second": 3.898 } ], "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": 3.041765415786701e+18, "train_batch_size": 160, "trial_name": null, "trial_params": null }