{ "best_metric": 3.566912889480591, "best_model_checkpoint": "bert_12_layer_model_v2_complete_training_new_48/checkpoint-200000", "epoch": 1.6385734650739812, "global_step": 200001, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0, "learning_rate": 5.000000000000001e-07, "loss": 10.0453, "step": 500 }, { "epoch": 0.01, "learning_rate": 1.0000000000000002e-06, "loss": 9.255, "step": 1000 }, { "epoch": 0.01, "learning_rate": 1.5e-06, "loss": 8.8535, "step": 1500 }, { "epoch": 0.02, "learning_rate": 2.0000000000000003e-06, "loss": 8.387, "step": 2000 }, { "epoch": 0.02, "learning_rate": 2.5e-06, "loss": 7.8881, "step": 2500 }, { "epoch": 0.02, "learning_rate": 3e-06, "loss": 7.4779, "step": 3000 }, { "epoch": 0.03, "learning_rate": 3.5e-06, "loss": 7.2099, "step": 3500 }, { "epoch": 0.03, "learning_rate": 4.000000000000001e-06, "loss": 7.0705, "step": 4000 }, { "epoch": 0.04, "learning_rate": 4.5e-06, "loss": 6.9682, "step": 4500 }, { "epoch": 0.04, "learning_rate": 5e-06, "loss": 6.9046, "step": 5000 }, { "epoch": 0.05, "learning_rate": 5.500000000000001e-06, "loss": 6.8446, "step": 5500 }, { "epoch": 0.05, "learning_rate": 6e-06, "loss": 6.7981, "step": 6000 }, { "epoch": 0.05, "learning_rate": 6.5000000000000004e-06, "loss": 6.7663, "step": 6500 }, { "epoch": 0.06, "learning_rate": 7e-06, "loss": 6.7301, "step": 7000 }, { "epoch": 0.06, "learning_rate": 7.500000000000001e-06, "loss": 6.7009, "step": 7500 }, { "epoch": 0.07, "learning_rate": 8.000000000000001e-06, "loss": 6.6649, "step": 8000 }, { "epoch": 0.07, "learning_rate": 8.5e-06, "loss": 6.6373, "step": 8500 }, { "epoch": 0.07, "learning_rate": 9e-06, "loss": 6.6173, "step": 9000 }, { "epoch": 0.08, "learning_rate": 9.5e-06, "loss": 6.5929, "step": 9500 }, { "epoch": 0.08, "learning_rate": 1e-05, "loss": 6.5761, "step": 10000 }, { "epoch": 0.08, "eval_accuracy": 0.1270219986877099, "eval_loss": 6.539961814880371, "eval_runtime": 1773.0544, "eval_samples_per_second": 173.912, "eval_steps_per_second": 3.624, "step": 10000 }, { "epoch": 0.09, "learning_rate": 9.991670692498626e-06, "loss": 6.5543, "step": 10500 }, { "epoch": 0.09, "learning_rate": 9.983341384997253e-06, "loss": 6.5419, "step": 11000 }, { "epoch": 0.09, "learning_rate": 9.975012077495878e-06, "loss": 6.5189, "step": 11500 }, { "epoch": 0.1, "learning_rate": 9.966682769994504e-06, "loss": 6.5035, "step": 12000 }, { "epoch": 0.1, "learning_rate": 9.95835346249313e-06, "loss": 6.4785, "step": 12500 }, { "epoch": 0.11, "learning_rate": 9.950024154991756e-06, "loss": 6.4681, "step": 13000 }, { "epoch": 0.11, "learning_rate": 9.941694847490381e-06, "loss": 6.4574, "step": 13500 }, { "epoch": 0.11, "learning_rate": 9.933365539989006e-06, "loss": 6.4455, "step": 14000 }, { "epoch": 0.12, "learning_rate": 9.925036232487631e-06, "loss": 6.4259, "step": 14500 }, { "epoch": 0.12, "learning_rate": 9.916706924986258e-06, "loss": 6.415, "step": 15000 }, { "epoch": 0.13, "learning_rate": 9.908377617484883e-06, "loss": 6.3972, "step": 15500 }, { "epoch": 0.13, "learning_rate": 9.90004830998351e-06, "loss": 6.3957, "step": 16000 }, { "epoch": 0.14, "learning_rate": 9.891719002482134e-06, "loss": 6.3916, "step": 16500 }, { "epoch": 0.14, "learning_rate": 9.883389694980761e-06, "loss": 6.3714, "step": 17000 }, { "epoch": 0.14, "learning_rate": 9.875060387479386e-06, "loss": 6.371, "step": 17500 }, { "epoch": 0.15, "learning_rate": 9.866731079978011e-06, "loss": 6.3623, "step": 18000 }, { "epoch": 0.15, "learning_rate": 9.858401772476636e-06, "loss": 6.3593, "step": 18500 }, { "epoch": 0.16, "learning_rate": 9.850072464975263e-06, "loss": 6.3421, "step": 19000 }, { "epoch": 0.16, "learning_rate": 9.84174315747389e-06, "loss": 6.3338, "step": 19500 }, { "epoch": 0.16, "learning_rate": 9.833413849972514e-06, "loss": 6.3285, "step": 20000 }, { "epoch": 0.16, "eval_accuracy": 0.14105833813706292, "eval_loss": 6.304974555969238, "eval_runtime": 1769.5872, "eval_samples_per_second": 174.253, "eval_steps_per_second": 3.631, "step": 20000 }, { "epoch": 0.17, "learning_rate": 9.82508454247114e-06, "loss": 6.3208, "step": 20500 }, { "epoch": 0.17, "learning_rate": 9.816755234969766e-06, "loss": 6.3102, "step": 21000 }, { "epoch": 0.18, "learning_rate": 9.808425927468391e-06, "loss": 6.3147, "step": 21500 }, { "epoch": 0.18, "learning_rate": 9.800096619967016e-06, "loss": 6.2978, "step": 22000 }, { "epoch": 0.18, "learning_rate": 9.791767312465643e-06, "loss": 6.3007, "step": 22500 }, { "epoch": 0.19, "learning_rate": 9.783438004964268e-06, "loss": 6.2824, "step": 23000 }, { "epoch": 0.19, "learning_rate": 9.775108697462895e-06, "loss": 6.2786, "step": 23500 }, { "epoch": 0.2, "learning_rate": 9.76677938996152e-06, "loss": 6.2721, "step": 24000 }, { "epoch": 0.2, "learning_rate": 9.758450082460145e-06, "loss": 6.2734, "step": 24500 }, { "epoch": 0.2, "learning_rate": 9.750120774958771e-06, "loss": 6.2678, "step": 25000 }, { "epoch": 0.21, "learning_rate": 9.741791467457396e-06, "loss": 6.2592, "step": 25500 }, { "epoch": 0.21, "learning_rate": 9.733462159956021e-06, "loss": 6.2565, "step": 26000 }, { "epoch": 0.22, "learning_rate": 9.725132852454648e-06, "loss": 6.2606, "step": 26500 }, { "epoch": 0.22, "learning_rate": 9.716803544953275e-06, "loss": 6.2541, "step": 27000 }, { "epoch": 0.23, "learning_rate": 9.7084742374519e-06, "loss": 6.2446, "step": 27500 }, { "epoch": 0.23, "learning_rate": 9.700144929950525e-06, "loss": 6.2423, "step": 28000 }, { "epoch": 0.23, "learning_rate": 9.69181562244915e-06, "loss": 6.2407, "step": 28500 }, { "epoch": 0.24, "learning_rate": 9.683486314947776e-06, "loss": 6.2356, "step": 29000 }, { "epoch": 0.24, "learning_rate": 9.675157007446401e-06, "loss": 6.2292, "step": 29500 }, { "epoch": 0.25, "learning_rate": 9.666827699945028e-06, "loss": 6.2279, "step": 30000 }, { "epoch": 0.25, "eval_accuracy": 0.14494122493690972, "eval_loss": 6.2128376960754395, "eval_runtime": 1769.2009, "eval_samples_per_second": 174.291, "eval_steps_per_second": 3.632, "step": 30000 }, { "epoch": 0.25, "learning_rate": 9.658498392443653e-06, "loss": 6.2207, "step": 30500 }, { "epoch": 0.25, "learning_rate": 9.65016908494228e-06, "loss": 6.2211, "step": 31000 }, { "epoch": 0.26, "learning_rate": 9.641839777440905e-06, "loss": 6.2216, "step": 31500 }, { "epoch": 0.26, "learning_rate": 9.63351046993953e-06, "loss": 6.2127, "step": 32000 }, { "epoch": 0.27, "learning_rate": 9.625181162438155e-06, "loss": 6.2086, "step": 32500 }, { "epoch": 0.27, "learning_rate": 9.616851854936781e-06, "loss": 6.2051, "step": 33000 }, { "epoch": 0.27, "learning_rate": 9.608522547435406e-06, "loss": 6.2005, "step": 33500 }, { "epoch": 0.28, "learning_rate": 9.600193239934033e-06, "loss": 6.1998, "step": 34000 }, { "epoch": 0.28, "learning_rate": 9.591863932432658e-06, "loss": 6.1927, "step": 34500 }, { "epoch": 0.29, "learning_rate": 9.583534624931285e-06, "loss": 6.1945, "step": 35000 }, { "epoch": 0.29, "learning_rate": 9.57520531742991e-06, "loss": 6.1996, "step": 35500 }, { "epoch": 0.29, "learning_rate": 9.566876009928535e-06, "loss": 6.192, "step": 36000 }, { "epoch": 0.3, "learning_rate": 9.55854670242716e-06, "loss": 6.189, "step": 36500 }, { "epoch": 0.3, "learning_rate": 9.550217394925787e-06, "loss": 6.1852, "step": 37000 }, { "epoch": 0.31, "learning_rate": 9.541888087424413e-06, "loss": 6.1824, "step": 37500 }, { "epoch": 0.31, "learning_rate": 9.533558779923038e-06, "loss": 6.1714, "step": 38000 }, { "epoch": 0.32, "learning_rate": 9.525229472421663e-06, "loss": 6.1748, "step": 38500 }, { "epoch": 0.32, "learning_rate": 9.51690016492029e-06, "loss": 6.1673, "step": 39000 }, { "epoch": 0.32, "learning_rate": 9.508570857418915e-06, "loss": 6.1711, "step": 39500 }, { "epoch": 0.33, "learning_rate": 9.50024154991754e-06, "loss": 6.1754, "step": 40000 }, { "epoch": 0.33, "eval_accuracy": 0.14783782769864162, "eval_loss": 6.153486251831055, "eval_runtime": 1794.205, "eval_samples_per_second": 171.862, "eval_steps_per_second": 3.581, "step": 40000 }, { "epoch": 0.33, "learning_rate": 9.491912242416167e-06, "loss": 6.1656, "step": 40500 }, { "epoch": 0.34, "learning_rate": 9.483582934914792e-06, "loss": 6.1643, "step": 41000 }, { "epoch": 0.34, "learning_rate": 9.475253627413418e-06, "loss": 6.1612, "step": 41500 }, { "epoch": 0.34, "learning_rate": 9.466924319912043e-06, "loss": 6.1645, "step": 42000 }, { "epoch": 0.35, "learning_rate": 9.458595012410668e-06, "loss": 6.1522, "step": 42500 }, { "epoch": 0.35, "learning_rate": 9.450265704909295e-06, "loss": 6.1511, "step": 43000 }, { "epoch": 0.36, "learning_rate": 9.44193639740792e-06, "loss": 6.1476, "step": 43500 }, { "epoch": 0.36, "learning_rate": 9.433607089906545e-06, "loss": 6.1509, "step": 44000 }, { "epoch": 0.36, "learning_rate": 9.425277782405172e-06, "loss": 6.147, "step": 44500 }, { "epoch": 0.37, "learning_rate": 9.416948474903798e-06, "loss": 6.145, "step": 45000 }, { "epoch": 0.37, "learning_rate": 9.408619167402423e-06, "loss": 6.1471, "step": 45500 }, { "epoch": 0.38, "learning_rate": 9.400289859901048e-06, "loss": 6.1362, "step": 46000 }, { "epoch": 0.38, "learning_rate": 9.391960552399673e-06, "loss": 6.1362, "step": 46500 }, { "epoch": 0.39, "learning_rate": 9.3836312448983e-06, "loss": 6.1306, "step": 47000 }, { "epoch": 0.39, "learning_rate": 9.375301937396925e-06, "loss": 6.1322, "step": 47500 }, { "epoch": 0.39, "learning_rate": 9.366972629895552e-06, "loss": 6.1332, "step": 48000 }, { "epoch": 0.4, "learning_rate": 9.358643322394177e-06, "loss": 6.1274, "step": 48500 }, { "epoch": 0.4, "learning_rate": 9.350314014892804e-06, "loss": 6.1276, "step": 49000 }, { "epoch": 0.41, "learning_rate": 9.341984707391429e-06, "loss": 6.1283, "step": 49500 }, { "epoch": 0.41, "learning_rate": 9.333655399890054e-06, "loss": 6.1291, "step": 50000 }, { "epoch": 0.41, "eval_accuracy": 0.14881116882917134, "eval_loss": 6.118132591247559, "eval_runtime": 1770.8476, "eval_samples_per_second": 174.129, "eval_steps_per_second": 3.628, "step": 50000 }, { "epoch": 0.41, "learning_rate": 9.325326092388679e-06, "loss": 6.1206, "step": 50500 }, { "epoch": 0.42, "learning_rate": 9.316996784887305e-06, "loss": 6.1214, "step": 51000 }, { "epoch": 0.42, "learning_rate": 9.30866747738593e-06, "loss": 6.1251, "step": 51500 }, { "epoch": 0.43, "learning_rate": 9.300338169884557e-06, "loss": 6.1199, "step": 52000 }, { "epoch": 0.43, "learning_rate": 9.292008862383182e-06, "loss": 6.116, "step": 52500 }, { "epoch": 0.43, "learning_rate": 9.283679554881809e-06, "loss": 6.1199, "step": 53000 }, { "epoch": 0.44, "learning_rate": 9.275350247380434e-06, "loss": 6.1131, "step": 53500 }, { "epoch": 0.44, "learning_rate": 9.267020939879059e-06, "loss": 6.1152, "step": 54000 }, { "epoch": 0.45, "learning_rate": 9.258691632377685e-06, "loss": 6.1146, "step": 54500 }, { "epoch": 0.45, "learning_rate": 9.25036232487631e-06, "loss": 6.1063, "step": 55000 }, { "epoch": 0.45, "learning_rate": 9.242033017374937e-06, "loss": 6.1138, "step": 55500 }, { "epoch": 0.46, "learning_rate": 9.233703709873562e-06, "loss": 6.1071, "step": 56000 }, { "epoch": 0.46, "learning_rate": 9.225374402372189e-06, "loss": 6.103, "step": 56500 }, { "epoch": 0.47, "learning_rate": 9.217045094870814e-06, "loss": 6.0942, "step": 57000 }, { "epoch": 0.47, "learning_rate": 9.208715787369439e-06, "loss": 6.1056, "step": 57500 }, { "epoch": 0.48, "learning_rate": 9.200386479868064e-06, "loss": 6.1059, "step": 58000 }, { "epoch": 0.48, "learning_rate": 9.19205717236669e-06, "loss": 6.1024, "step": 58500 }, { "epoch": 0.48, "learning_rate": 9.183727864865315e-06, "loss": 6.0976, "step": 59000 }, { "epoch": 0.49, "learning_rate": 9.175398557363942e-06, "loss": 6.1034, "step": 59500 }, { "epoch": 0.49, "learning_rate": 9.167069249862567e-06, "loss": 6.1008, "step": 60000 }, { "epoch": 0.49, "eval_accuracy": 0.14947371229012402, "eval_loss": 6.084630012512207, "eval_runtime": 1770.4458, "eval_samples_per_second": 174.169, "eval_steps_per_second": 3.629, "step": 60000 }, { "epoch": 0.5, "learning_rate": 9.158739942361194e-06, "loss": 6.0913, "step": 60500 }, { "epoch": 0.5, "learning_rate": 9.150410634859819e-06, "loss": 6.0894, "step": 61000 }, { "epoch": 0.5, "learning_rate": 9.142081327358444e-06, "loss": 6.0932, "step": 61500 }, { "epoch": 0.51, "learning_rate": 9.133752019857069e-06, "loss": 6.0844, "step": 62000 }, { "epoch": 0.51, "learning_rate": 9.125422712355696e-06, "loss": 6.0899, "step": 62500 }, { "epoch": 0.52, "learning_rate": 9.11709340485432e-06, "loss": 6.0903, "step": 63000 }, { "epoch": 0.52, "learning_rate": 9.108764097352947e-06, "loss": 6.0793, "step": 63500 }, { "epoch": 0.52, "learning_rate": 9.100434789851572e-06, "loss": 6.0804, "step": 64000 }, { "epoch": 0.53, "learning_rate": 9.092105482350199e-06, "loss": 6.0832, "step": 64500 }, { "epoch": 0.53, "learning_rate": 9.083776174848824e-06, "loss": 6.0816, "step": 65000 }, { "epoch": 0.54, "learning_rate": 9.075446867347449e-06, "loss": 6.0844, "step": 65500 }, { "epoch": 0.54, "learning_rate": 9.067117559846076e-06, "loss": 6.0856, "step": 66000 }, { "epoch": 0.54, "learning_rate": 9.0587882523447e-06, "loss": 6.0749, "step": 66500 }, { "epoch": 0.55, "learning_rate": 9.050458944843327e-06, "loss": 6.0755, "step": 67000 }, { "epoch": 0.55, "learning_rate": 9.042129637341952e-06, "loss": 6.0746, "step": 67500 }, { "epoch": 0.56, "learning_rate": 9.033800329840577e-06, "loss": 6.0801, "step": 68000 }, { "epoch": 0.56, "learning_rate": 9.025471022339204e-06, "loss": 6.0765, "step": 68500 }, { "epoch": 0.57, "learning_rate": 9.017141714837829e-06, "loss": 6.0728, "step": 69000 }, { "epoch": 0.57, "learning_rate": 9.008812407336454e-06, "loss": 6.0765, "step": 69500 }, { "epoch": 0.57, "learning_rate": 9.00048309983508e-06, "loss": 6.0716, "step": 70000 }, { "epoch": 0.57, "eval_accuracy": 0.15036080325943188, "eval_loss": 6.060856819152832, "eval_runtime": 1772.2261, "eval_samples_per_second": 173.994, "eval_steps_per_second": 3.625, "step": 70000 }, { "epoch": 0.58, "learning_rate": 8.992153792333706e-06, "loss": 6.0667, "step": 70500 }, { "epoch": 0.58, "learning_rate": 8.983824484832332e-06, "loss": 6.0603, "step": 71000 }, { "epoch": 0.59, "learning_rate": 8.975495177330957e-06, "loss": 6.0689, "step": 71500 }, { "epoch": 0.59, "learning_rate": 8.967165869829582e-06, "loss": 6.0649, "step": 72000 }, { "epoch": 0.59, "learning_rate": 8.958836562328209e-06, "loss": 6.0598, "step": 72500 }, { "epoch": 0.6, "learning_rate": 8.950507254826834e-06, "loss": 6.0489, "step": 73000 }, { "epoch": 0.6, "learning_rate": 8.942177947325459e-06, "loss": 6.0486, "step": 73500 }, { "epoch": 0.61, "learning_rate": 8.933848639824086e-06, "loss": 6.0301, "step": 74000 }, { "epoch": 0.61, "learning_rate": 8.925519332322713e-06, "loss": 6.0153, "step": 74500 }, { "epoch": 0.61, "learning_rate": 8.917190024821338e-06, "loss": 6.0084, "step": 75000 }, { "epoch": 0.62, "learning_rate": 8.908860717319963e-06, "loss": 5.9887, "step": 75500 }, { "epoch": 0.62, "learning_rate": 8.900531409818588e-06, "loss": 5.9798, "step": 76000 }, { "epoch": 0.63, "learning_rate": 8.892202102317214e-06, "loss": 5.962, "step": 76500 }, { "epoch": 0.63, "learning_rate": 8.88387279481584e-06, "loss": 5.9522, "step": 77000 }, { "epoch": 0.63, "learning_rate": 8.875543487314466e-06, "loss": 5.9392, "step": 77500 }, { "epoch": 0.64, "learning_rate": 8.867214179813091e-06, "loss": 5.9297, "step": 78000 }, { "epoch": 0.64, "learning_rate": 8.858884872311718e-06, "loss": 5.9253, "step": 78500 }, { "epoch": 0.65, "learning_rate": 8.850555564810343e-06, "loss": 5.9142, "step": 79000 }, { "epoch": 0.65, "learning_rate": 8.842226257308968e-06, "loss": 5.9113, "step": 79500 }, { "epoch": 0.66, "learning_rate": 8.833896949807593e-06, "loss": 5.9041, "step": 80000 }, { "epoch": 0.66, "eval_accuracy": 0.15765009726720394, "eval_loss": 5.868770122528076, "eval_runtime": 1769.7047, "eval_samples_per_second": 174.242, "eval_steps_per_second": 3.631, "step": 80000 }, { "epoch": 0.66, "learning_rate": 8.82556764230622e-06, "loss": 5.8956, "step": 80500 }, { "epoch": 0.66, "learning_rate": 8.817238334804844e-06, "loss": 5.8901, "step": 81000 }, { "epoch": 0.67, "learning_rate": 8.808909027303471e-06, "loss": 5.88, "step": 81500 }, { "epoch": 0.67, "learning_rate": 8.800579719802096e-06, "loss": 5.8778, "step": 82000 }, { "epoch": 0.68, "learning_rate": 8.792250412300723e-06, "loss": 5.8655, "step": 82500 }, { "epoch": 0.68, "learning_rate": 8.783921104799348e-06, "loss": 5.8693, "step": 83000 }, { "epoch": 0.68, "learning_rate": 8.775591797297973e-06, "loss": 5.8597, "step": 83500 }, { "epoch": 0.69, "learning_rate": 8.767262489796598e-06, "loss": 5.8539, "step": 84000 }, { "epoch": 0.69, "learning_rate": 8.758933182295224e-06, "loss": 5.8448, "step": 84500 }, { "epoch": 0.7, "learning_rate": 8.750603874793851e-06, "loss": 5.8444, "step": 85000 }, { "epoch": 0.7, "learning_rate": 8.742274567292476e-06, "loss": 5.8402, "step": 85500 }, { "epoch": 0.7, "learning_rate": 8.733945259791101e-06, "loss": 5.8319, "step": 86000 }, { "epoch": 0.71, "learning_rate": 8.725615952289728e-06, "loss": 5.8329, "step": 86500 }, { "epoch": 0.71, "learning_rate": 8.717286644788353e-06, "loss": 5.826, "step": 87000 }, { "epoch": 0.72, "learning_rate": 8.708957337286978e-06, "loss": 5.8248, "step": 87500 }, { "epoch": 0.72, "learning_rate": 8.700628029785605e-06, "loss": 5.8162, "step": 88000 }, { "epoch": 0.73, "learning_rate": 8.69229872228423e-06, "loss": 5.8119, "step": 88500 }, { "epoch": 0.73, "learning_rate": 8.683969414782856e-06, "loss": 5.8093, "step": 89000 }, { "epoch": 0.73, "learning_rate": 8.675640107281481e-06, "loss": 5.8021, "step": 89500 }, { "epoch": 0.74, "learning_rate": 8.667310799780106e-06, "loss": 5.7999, "step": 90000 }, { "epoch": 0.74, "eval_accuracy": 0.16906430476040712, "eval_loss": 5.7594709396362305, "eval_runtime": 1781.037, "eval_samples_per_second": 173.133, "eval_steps_per_second": 3.607, "step": 90000 }, { "epoch": 0.74, "learning_rate": 8.658981492278733e-06, "loss": 5.7898, "step": 90500 }, { "epoch": 0.75, "learning_rate": 8.650652184777358e-06, "loss": 5.7883, "step": 91000 }, { "epoch": 0.75, "learning_rate": 8.642322877275983e-06, "loss": 5.7818, "step": 91500 }, { "epoch": 0.75, "learning_rate": 8.63399356977461e-06, "loss": 5.7883, "step": 92000 }, { "epoch": 0.76, "learning_rate": 8.625664262273236e-06, "loss": 5.773, "step": 92500 }, { "epoch": 0.76, "learning_rate": 8.617334954771861e-06, "loss": 5.7716, "step": 93000 }, { "epoch": 0.77, "learning_rate": 8.609005647270486e-06, "loss": 5.7626, "step": 93500 }, { "epoch": 0.77, "learning_rate": 8.600676339769111e-06, "loss": 5.7566, "step": 94000 }, { "epoch": 0.77, "learning_rate": 8.592347032267738e-06, "loss": 5.7611, "step": 94500 }, { "epoch": 0.78, "learning_rate": 8.584017724766363e-06, "loss": 5.7489, "step": 95000 }, { "epoch": 0.78, "learning_rate": 8.57568841726499e-06, "loss": 5.7494, "step": 95500 }, { "epoch": 0.79, "learning_rate": 8.567359109763615e-06, "loss": 5.7464, "step": 96000 }, { "epoch": 0.79, "learning_rate": 8.559029802262241e-06, "loss": 5.7394, "step": 96500 }, { "epoch": 0.79, "learning_rate": 8.550700494760866e-06, "loss": 5.7333, "step": 97000 }, { "epoch": 0.8, "learning_rate": 8.542371187259491e-06, "loss": 5.7274, "step": 97500 }, { "epoch": 0.8, "learning_rate": 8.534041879758118e-06, "loss": 5.7245, "step": 98000 }, { "epoch": 0.81, "learning_rate": 8.525712572256743e-06, "loss": 5.7222, "step": 98500 }, { "epoch": 0.81, "learning_rate": 8.517383264755368e-06, "loss": 5.7152, "step": 99000 }, { "epoch": 0.82, "learning_rate": 8.509053957253995e-06, "loss": 5.7061, "step": 99500 }, { "epoch": 0.82, "learning_rate": 8.500724649752622e-06, "loss": 5.6997, "step": 100000 }, { "epoch": 0.82, "eval_accuracy": 0.1828346855228615, "eval_loss": 5.646856307983398, "eval_runtime": 1786.155, "eval_samples_per_second": 172.637, "eval_steps_per_second": 3.597, "step": 100000 }, { "epoch": 0.82, "learning_rate": 8.492395342251247e-06, "loss": 5.6965, "step": 100500 }, { "epoch": 0.83, "learning_rate": 8.484066034749872e-06, "loss": 5.6886, "step": 101000 }, { "epoch": 0.83, "learning_rate": 8.475736727248497e-06, "loss": 5.6897, "step": 101500 }, { "epoch": 0.84, "learning_rate": 8.467407419747123e-06, "loss": 5.6804, "step": 102000 }, { "epoch": 0.84, "learning_rate": 8.459078112245748e-06, "loss": 5.6781, "step": 102500 }, { "epoch": 0.84, "learning_rate": 8.450748804744375e-06, "loss": 5.6756, "step": 103000 }, { "epoch": 0.85, "learning_rate": 8.442419497243e-06, "loss": 5.6682, "step": 103500 }, { "epoch": 0.85, "learning_rate": 8.434090189741627e-06, "loss": 5.6595, "step": 104000 }, { "epoch": 0.86, "learning_rate": 8.425760882240252e-06, "loss": 5.6659, "step": 104500 }, { "epoch": 0.86, "learning_rate": 8.417431574738877e-06, "loss": 5.6533, "step": 105000 }, { "epoch": 0.86, "learning_rate": 8.409102267237502e-06, "loss": 5.655, "step": 105500 }, { "epoch": 0.87, "learning_rate": 8.400772959736128e-06, "loss": 5.643, "step": 106000 }, { "epoch": 0.87, "learning_rate": 8.392443652234753e-06, "loss": 5.6374, "step": 106500 }, { "epoch": 0.88, "learning_rate": 8.38411434473338e-06, "loss": 5.6284, "step": 107000 }, { "epoch": 0.88, "learning_rate": 8.375785037232005e-06, "loss": 5.6242, "step": 107500 }, { "epoch": 0.88, "learning_rate": 8.367455729730632e-06, "loss": 5.6235, "step": 108000 }, { "epoch": 0.89, "learning_rate": 8.359126422229257e-06, "loss": 5.6139, "step": 108500 }, { "epoch": 0.89, "learning_rate": 8.350797114727882e-06, "loss": 5.6102, "step": 109000 }, { "epoch": 0.9, "learning_rate": 8.342467807226507e-06, "loss": 5.6069, "step": 109500 }, { "epoch": 0.9, "learning_rate": 8.334138499725133e-06, "loss": 5.6002, "step": 110000 }, { "epoch": 0.9, "eval_accuracy": 0.19633756823302348, "eval_loss": 5.535816192626953, "eval_runtime": 1769.3441, "eval_samples_per_second": 174.277, "eval_steps_per_second": 3.631, "step": 110000 }, { "epoch": 0.91, "learning_rate": 8.32580919222376e-06, "loss": 5.6003, "step": 110500 }, { "epoch": 0.91, "learning_rate": 8.317479884722385e-06, "loss": 5.5931, "step": 111000 }, { "epoch": 0.91, "learning_rate": 8.30915057722101e-06, "loss": 5.5869, "step": 111500 }, { "epoch": 0.92, "learning_rate": 8.300821269719637e-06, "loss": 5.5814, "step": 112000 }, { "epoch": 0.92, "learning_rate": 8.292491962218262e-06, "loss": 5.5825, "step": 112500 }, { "epoch": 0.93, "learning_rate": 8.284162654716887e-06, "loss": 5.5723, "step": 113000 }, { "epoch": 0.93, "learning_rate": 8.275833347215514e-06, "loss": 5.566, "step": 113500 }, { "epoch": 0.93, "learning_rate": 8.267504039714139e-06, "loss": 5.5676, "step": 114000 }, { "epoch": 0.94, "learning_rate": 8.259174732212765e-06, "loss": 5.5587, "step": 114500 }, { "epoch": 0.94, "learning_rate": 8.25084542471139e-06, "loss": 5.5548, "step": 115000 }, { "epoch": 0.95, "learning_rate": 8.242516117210015e-06, "loss": 5.5457, "step": 115500 }, { "epoch": 0.95, "learning_rate": 8.234186809708642e-06, "loss": 5.5372, "step": 116000 }, { "epoch": 0.95, "learning_rate": 8.225857502207267e-06, "loss": 5.5288, "step": 116500 }, { "epoch": 0.96, "learning_rate": 8.217528194705892e-06, "loss": 5.5216, "step": 117000 }, { "epoch": 0.96, "learning_rate": 8.209198887204519e-06, "loss": 5.5147, "step": 117500 }, { "epoch": 0.97, "learning_rate": 8.200869579703145e-06, "loss": 5.5009, "step": 118000 }, { "epoch": 0.97, "learning_rate": 8.19254027220177e-06, "loss": 5.4972, "step": 118500 }, { "epoch": 0.97, "learning_rate": 8.184210964700395e-06, "loss": 5.4823, "step": 119000 }, { "epoch": 0.98, "learning_rate": 8.17588165719902e-06, "loss": 5.455, "step": 119500 }, { "epoch": 0.98, "learning_rate": 8.167552349697647e-06, "loss": 5.4372, "step": 120000 }, { "epoch": 0.98, "eval_accuracy": 0.22528347504121415, "eval_loss": 5.311297416687012, "eval_runtime": 1769.0434, "eval_samples_per_second": 174.307, "eval_steps_per_second": 3.632, "step": 120000 }, { "epoch": 0.99, "learning_rate": 8.159223042196272e-06, "loss": 5.4131, "step": 120500 }, { "epoch": 0.99, "learning_rate": 8.150893734694899e-06, "loss": 5.3892, "step": 121000 }, { "epoch": 1.0, "learning_rate": 8.142564427193524e-06, "loss": 5.3796, "step": 121500 }, { "epoch": 1.0, "learning_rate": 8.13423511969215e-06, "loss": 5.3521, "step": 122000 }, { "epoch": 1.0, "learning_rate": 8.125905812190775e-06, "loss": 5.335, "step": 122500 }, { "epoch": 1.01, "learning_rate": 8.1175765046894e-06, "loss": 5.3171, "step": 123000 }, { "epoch": 1.01, "learning_rate": 8.109247197188025e-06, "loss": 5.3011, "step": 123500 }, { "epoch": 1.02, "learning_rate": 8.100917889686652e-06, "loss": 5.2787, "step": 124000 }, { "epoch": 1.02, "learning_rate": 8.092588582185277e-06, "loss": 5.2604, "step": 124500 }, { "epoch": 1.02, "learning_rate": 8.084259274683904e-06, "loss": 5.2429, "step": 125000 }, { "epoch": 1.03, "learning_rate": 8.075929967182529e-06, "loss": 5.2277, "step": 125500 }, { "epoch": 1.03, "learning_rate": 8.067600659681156e-06, "loss": 5.1998, "step": 126000 }, { "epoch": 1.04, "learning_rate": 8.05927135217978e-06, "loss": 5.1848, "step": 126500 }, { "epoch": 1.04, "learning_rate": 8.050942044678406e-06, "loss": 5.1681, "step": 127000 }, { "epoch": 1.04, "learning_rate": 8.04261273717703e-06, "loss": 5.1464, "step": 127500 }, { "epoch": 1.05, "learning_rate": 8.034283429675657e-06, "loss": 5.126, "step": 128000 }, { "epoch": 1.05, "learning_rate": 8.025954122174284e-06, "loss": 5.1044, "step": 128500 }, { "epoch": 1.06, "learning_rate": 8.017624814672909e-06, "loss": 5.0931, "step": 129000 }, { "epoch": 1.06, "learning_rate": 8.009295507171534e-06, "loss": 5.0696, "step": 129500 }, { "epoch": 1.07, "learning_rate": 8.00096619967016e-06, "loss": 5.0465, "step": 130000 }, { "epoch": 1.07, "eval_accuracy": 0.2742936938062577, "eval_loss": 4.876473426818848, "eval_runtime": 1769.978, "eval_samples_per_second": 174.215, "eval_steps_per_second": 3.63, "step": 130000 }, { "epoch": 1.07, "learning_rate": 7.992636892168786e-06, "loss": 5.0325, "step": 130500 }, { "epoch": 1.07, "learning_rate": 7.98430758466741e-06, "loss": 5.0227, "step": 131000 }, { "epoch": 1.08, "learning_rate": 7.975978277166037e-06, "loss": 5.005, "step": 131500 }, { "epoch": 1.08, "learning_rate": 7.967648969664662e-06, "loss": 4.9835, "step": 132000 }, { "epoch": 1.09, "learning_rate": 7.959319662163289e-06, "loss": 4.9714, "step": 132500 }, { "epoch": 1.09, "learning_rate": 7.950990354661914e-06, "loss": 4.9533, "step": 133000 }, { "epoch": 1.09, "learning_rate": 7.942661047160539e-06, "loss": 4.9397, "step": 133500 }, { "epoch": 1.1, "learning_rate": 7.934331739659166e-06, "loss": 4.9215, "step": 134000 }, { "epoch": 1.1, "learning_rate": 7.92600243215779e-06, "loss": 4.9123, "step": 134500 }, { "epoch": 1.11, "learning_rate": 7.917673124656416e-06, "loss": 4.8854, "step": 135000 }, { "epoch": 1.11, "learning_rate": 7.909343817155042e-06, "loss": 4.8883, "step": 135500 }, { "epoch": 1.11, "learning_rate": 7.901014509653669e-06, "loss": 4.8687, "step": 136000 }, { "epoch": 1.12, "learning_rate": 7.892685202152294e-06, "loss": 4.8497, "step": 136500 }, { "epoch": 1.12, "learning_rate": 7.884355894650919e-06, "loss": 4.8467, "step": 137000 }, { "epoch": 1.13, "learning_rate": 7.876026587149544e-06, "loss": 4.8269, "step": 137500 }, { "epoch": 1.13, "learning_rate": 7.86769727964817e-06, "loss": 4.8099, "step": 138000 }, { "epoch": 1.13, "learning_rate": 7.859367972146796e-06, "loss": 4.8007, "step": 138500 }, { "epoch": 1.14, "learning_rate": 7.851038664645423e-06, "loss": 4.7846, "step": 139000 }, { "epoch": 1.14, "learning_rate": 7.842709357144048e-06, "loss": 4.759, "step": 139500 }, { "epoch": 1.15, "learning_rate": 7.834380049642674e-06, "loss": 4.7373, "step": 140000 }, { "epoch": 1.15, "eval_accuracy": 0.30952878562145963, "eval_loss": 4.553632736206055, "eval_runtime": 1771.8203, "eval_samples_per_second": 174.033, "eval_steps_per_second": 3.626, "step": 140000 }, { "epoch": 1.15, "learning_rate": 7.8260507421413e-06, "loss": 4.7121, "step": 140500 }, { "epoch": 1.16, "learning_rate": 7.817721434639924e-06, "loss": 4.6851, "step": 141000 }, { "epoch": 1.16, "learning_rate": 7.809392127138551e-06, "loss": 4.653, "step": 141500 }, { "epoch": 1.16, "learning_rate": 7.801062819637176e-06, "loss": 4.6298, "step": 142000 }, { "epoch": 1.17, "learning_rate": 7.792733512135801e-06, "loss": 4.6164, "step": 142500 }, { "epoch": 1.17, "learning_rate": 7.784404204634428e-06, "loss": 4.5905, "step": 143000 }, { "epoch": 1.18, "learning_rate": 7.776074897133054e-06, "loss": 4.5768, "step": 143500 }, { "epoch": 1.18, "learning_rate": 7.76774558963168e-06, "loss": 4.5647, "step": 144000 }, { "epoch": 1.18, "learning_rate": 7.759416282130304e-06, "loss": 4.5377, "step": 144500 }, { "epoch": 1.19, "learning_rate": 7.75108697462893e-06, "loss": 4.517, "step": 145000 }, { "epoch": 1.19, "learning_rate": 7.742757667127556e-06, "loss": 4.4965, "step": 145500 }, { "epoch": 1.2, "learning_rate": 7.734428359626181e-06, "loss": 4.4825, "step": 146000 }, { "epoch": 1.2, "learning_rate": 7.726099052124808e-06, "loss": 4.477, "step": 146500 }, { "epoch": 1.2, "learning_rate": 7.717769744623433e-06, "loss": 4.4546, "step": 147000 }, { "epoch": 1.21, "learning_rate": 7.70944043712206e-06, "loss": 4.4352, "step": 147500 }, { "epoch": 1.21, "learning_rate": 7.701111129620684e-06, "loss": 4.4242, "step": 148000 }, { "epoch": 1.22, "learning_rate": 7.69278182211931e-06, "loss": 4.4192, "step": 148500 }, { "epoch": 1.22, "learning_rate": 7.684452514617934e-06, "loss": 4.3991, "step": 149000 }, { "epoch": 1.22, "learning_rate": 7.676123207116561e-06, "loss": 4.397, "step": 149500 }, { "epoch": 1.23, "learning_rate": 7.667793899615186e-06, "loss": 4.3779, "step": 150000 }, { "epoch": 1.23, "eval_accuracy": 0.34167024571868826, "eval_loss": 4.2078070640563965, "eval_runtime": 1771.0899, "eval_samples_per_second": 174.105, "eval_steps_per_second": 3.628, "step": 150000 }, { "epoch": 1.23, "learning_rate": 7.659464592113813e-06, "loss": 4.3608, "step": 150500 }, { "epoch": 1.24, "learning_rate": 7.651135284612438e-06, "loss": 4.3514, "step": 151000 }, { "epoch": 1.24, "learning_rate": 7.642805977111065e-06, "loss": 4.3322, "step": 151500 }, { "epoch": 1.25, "learning_rate": 7.63447666960969e-06, "loss": 4.323, "step": 152000 }, { "epoch": 1.25, "learning_rate": 7.6261473621083145e-06, "loss": 4.3135, "step": 152500 }, { "epoch": 1.25, "learning_rate": 7.61781805460694e-06, "loss": 4.2928, "step": 153000 }, { "epoch": 1.26, "learning_rate": 7.609488747105567e-06, "loss": 4.2904, "step": 153500 }, { "epoch": 1.26, "learning_rate": 7.601159439604192e-06, "loss": 4.2735, "step": 154000 }, { "epoch": 1.27, "learning_rate": 7.592830132102817e-06, "loss": 4.262, "step": 154500 }, { "epoch": 1.27, "learning_rate": 7.584500824601443e-06, "loss": 4.2449, "step": 155000 }, { "epoch": 1.27, "learning_rate": 7.57617151710007e-06, "loss": 4.2219, "step": 155500 }, { "epoch": 1.28, "learning_rate": 7.567842209598695e-06, "loss": 4.2211, "step": 156000 }, { "epoch": 1.28, "learning_rate": 7.5595129020973205e-06, "loss": 4.2092, "step": 156500 }, { "epoch": 1.29, "learning_rate": 7.5511835945959455e-06, "loss": 4.2009, "step": 157000 }, { "epoch": 1.29, "learning_rate": 7.542854287094572e-06, "loss": 4.1755, "step": 157500 }, { "epoch": 1.29, "learning_rate": 7.534524979593197e-06, "loss": 4.171, "step": 158000 }, { "epoch": 1.3, "learning_rate": 7.526195672091823e-06, "loss": 4.1454, "step": 158500 }, { "epoch": 1.3, "learning_rate": 7.517866364590448e-06, "loss": 4.1508, "step": 159000 }, { "epoch": 1.31, "learning_rate": 7.509537057089075e-06, "loss": 4.1358, "step": 159500 }, { "epoch": 1.31, "learning_rate": 7.5012077495877e-06, "loss": 4.1299, "step": 160000 }, { "epoch": 1.31, "eval_accuracy": 0.3630001313259987, "eval_loss": 3.990952491760254, "eval_runtime": 1772.6319, "eval_samples_per_second": 173.954, "eval_steps_per_second": 3.625, "step": 160000 }, { "epoch": 1.31, "learning_rate": 7.492878442086326e-06, "loss": 4.1212, "step": 160500 }, { "epoch": 1.32, "learning_rate": 7.484549134584951e-06, "loss": 4.1031, "step": 161000 }, { "epoch": 1.32, "learning_rate": 7.476219827083577e-06, "loss": 4.091, "step": 161500 }, { "epoch": 1.33, "learning_rate": 7.467890519582202e-06, "loss": 4.0882, "step": 162000 }, { "epoch": 1.33, "learning_rate": 7.459561212080828e-06, "loss": 4.0743, "step": 162500 }, { "epoch": 1.34, "learning_rate": 7.451231904579453e-06, "loss": 4.0662, "step": 163000 }, { "epoch": 1.34, "learning_rate": 7.44290259707808e-06, "loss": 4.0632, "step": 163500 }, { "epoch": 1.34, "learning_rate": 7.434573289576706e-06, "loss": 4.047, "step": 164000 }, { "epoch": 1.35, "learning_rate": 7.426243982075331e-06, "loss": 4.0359, "step": 164500 }, { "epoch": 1.35, "learning_rate": 7.417914674573956e-06, "loss": 4.0316, "step": 165000 }, { "epoch": 1.36, "learning_rate": 7.409585367072582e-06, "loss": 4.0246, "step": 165500 }, { "epoch": 1.36, "learning_rate": 7.401256059571208e-06, "loss": 4.0144, "step": 166000 }, { "epoch": 1.36, "learning_rate": 7.392926752069833e-06, "loss": 4.004, "step": 166500 }, { "epoch": 1.37, "learning_rate": 7.384597444568459e-06, "loss": 3.9944, "step": 167000 }, { "epoch": 1.37, "learning_rate": 7.376268137067085e-06, "loss": 3.9907, "step": 167500 }, { "epoch": 1.38, "learning_rate": 7.367938829565711e-06, "loss": 3.983, "step": 168000 }, { "epoch": 1.38, "learning_rate": 7.359609522064336e-06, "loss": 3.9743, "step": 168500 }, { "epoch": 1.38, "learning_rate": 7.351280214562962e-06, "loss": 3.9681, "step": 169000 }, { "epoch": 1.39, "learning_rate": 7.3429509070615875e-06, "loss": 3.9624, "step": 169500 }, { "epoch": 1.39, "learning_rate": 7.334621599560213e-06, "loss": 3.9585, "step": 170000 }, { "epoch": 1.39, "eval_accuracy": 0.37979738910319555, "eval_loss": 3.834655284881592, "eval_runtime": 1771.4872, "eval_samples_per_second": 174.066, "eval_steps_per_second": 3.627, "step": 170000 }, { "epoch": 1.4, "learning_rate": 7.326292292058838e-06, "loss": 3.9474, "step": 170500 }, { "epoch": 1.4, "learning_rate": 7.317962984557464e-06, "loss": 3.9524, "step": 171000 }, { "epoch": 1.41, "learning_rate": 7.309633677056091e-06, "loss": 3.9369, "step": 171500 }, { "epoch": 1.41, "learning_rate": 7.301304369554716e-06, "loss": 3.9296, "step": 172000 }, { "epoch": 1.41, "learning_rate": 7.292975062053341e-06, "loss": 3.919, "step": 172500 }, { "epoch": 1.42, "learning_rate": 7.284645754551967e-06, "loss": 3.9186, "step": 173000 }, { "epoch": 1.42, "learning_rate": 7.2763164470505934e-06, "loss": 3.9164, "step": 173500 }, { "epoch": 1.43, "learning_rate": 7.2679871395492184e-06, "loss": 3.9066, "step": 174000 }, { "epoch": 1.43, "learning_rate": 7.259657832047844e-06, "loss": 3.8948, "step": 174500 }, { "epoch": 1.43, "learning_rate": 7.251328524546469e-06, "loss": 3.8933, "step": 175000 }, { "epoch": 1.44, "learning_rate": 7.242999217045096e-06, "loss": 3.8933, "step": 175500 }, { "epoch": 1.44, "learning_rate": 7.234669909543721e-06, "loss": 3.8846, "step": 176000 }, { "epoch": 1.45, "learning_rate": 7.226340602042347e-06, "loss": 3.8778, "step": 176500 }, { "epoch": 1.45, "learning_rate": 7.218011294540972e-06, "loss": 3.864, "step": 177000 }, { "epoch": 1.45, "learning_rate": 7.2096819870395985e-06, "loss": 3.8679, "step": 177500 }, { "epoch": 1.46, "learning_rate": 7.2013526795382235e-06, "loss": 3.8657, "step": 178000 }, { "epoch": 1.46, "learning_rate": 7.193023372036849e-06, "loss": 3.8664, "step": 178500 }, { "epoch": 1.47, "learning_rate": 7.184694064535474e-06, "loss": 3.8488, "step": 179000 }, { "epoch": 1.47, "learning_rate": 7.176364757034101e-06, "loss": 3.8404, "step": 179500 }, { "epoch": 1.47, "learning_rate": 7.168035449532726e-06, "loss": 3.8423, "step": 180000 }, { "epoch": 1.47, "eval_accuracy": 0.3910571234040146, "eval_loss": 3.7274010181427, "eval_runtime": 1771.2182, "eval_samples_per_second": 174.093, "eval_steps_per_second": 3.627, "step": 180000 }, { "epoch": 1.48, "learning_rate": 7.159706142031352e-06, "loss": 3.8346, "step": 180500 }, { "epoch": 1.48, "learning_rate": 7.151376834529977e-06, "loss": 3.8321, "step": 181000 }, { "epoch": 1.49, "learning_rate": 7.143047527028604e-06, "loss": 3.8192, "step": 181500 }, { "epoch": 1.49, "learning_rate": 7.1347182195272295e-06, "loss": 3.8109, "step": 182000 }, { "epoch": 1.5, "learning_rate": 7.1263889120258545e-06, "loss": 3.8079, "step": 182500 }, { "epoch": 1.5, "learning_rate": 7.118059604524481e-06, "loss": 3.8101, "step": 183000 }, { "epoch": 1.5, "learning_rate": 7.109730297023106e-06, "loss": 3.8074, "step": 183500 }, { "epoch": 1.51, "learning_rate": 7.101400989521732e-06, "loss": 3.7989, "step": 184000 }, { "epoch": 1.51, "learning_rate": 7.093071682020357e-06, "loss": 3.8001, "step": 184500 }, { "epoch": 1.52, "learning_rate": 7.084742374518984e-06, "loss": 3.7933, "step": 185000 }, { "epoch": 1.52, "learning_rate": 7.076413067017609e-06, "loss": 3.7792, "step": 185500 }, { "epoch": 1.52, "learning_rate": 7.068083759516235e-06, "loss": 3.7799, "step": 186000 }, { "epoch": 1.53, "learning_rate": 7.05975445201486e-06, "loss": 3.7805, "step": 186500 }, { "epoch": 1.53, "learning_rate": 7.051425144513486e-06, "loss": 3.7711, "step": 187000 }, { "epoch": 1.54, "learning_rate": 7.043095837012111e-06, "loss": 3.7634, "step": 187500 }, { "epoch": 1.54, "learning_rate": 7.034766529510737e-06, "loss": 3.7684, "step": 188000 }, { "epoch": 1.54, "learning_rate": 7.026437222009362e-06, "loss": 3.7505, "step": 188500 }, { "epoch": 1.55, "learning_rate": 7.018107914507989e-06, "loss": 3.7544, "step": 189000 }, { "epoch": 1.55, "learning_rate": 7.009778607006615e-06, "loss": 3.7554, "step": 189500 }, { "epoch": 1.56, "learning_rate": 7.00144929950524e-06, "loss": 3.7403, "step": 190000 }, { "epoch": 1.56, "eval_accuracy": 0.3996292753509408, "eval_loss": 3.64223575592041, "eval_runtime": 1768.8198, "eval_samples_per_second": 174.329, "eval_steps_per_second": 3.632, "step": 190000 }, { "epoch": 1.56, "learning_rate": 6.993119992003865e-06, "loss": 3.7381, "step": 190500 }, { "epoch": 1.56, "learning_rate": 6.984790684502491e-06, "loss": 3.7377, "step": 191000 }, { "epoch": 1.57, "learning_rate": 6.976461377001117e-06, "loss": 3.741, "step": 191500 }, { "epoch": 1.57, "learning_rate": 6.968132069499742e-06, "loss": 3.7326, "step": 192000 }, { "epoch": 1.58, "learning_rate": 6.959802761998368e-06, "loss": 3.725, "step": 192500 }, { "epoch": 1.58, "learning_rate": 6.951473454496994e-06, "loss": 3.729, "step": 193000 }, { "epoch": 1.59, "learning_rate": 6.94314414699562e-06, "loss": 3.7222, "step": 193500 }, { "epoch": 1.59, "learning_rate": 6.934814839494245e-06, "loss": 3.7134, "step": 194000 }, { "epoch": 1.59, "learning_rate": 6.926485531992871e-06, "loss": 3.7125, "step": 194500 }, { "epoch": 1.6, "learning_rate": 6.9181562244914965e-06, "loss": 3.7017, "step": 195000 }, { "epoch": 1.6, "learning_rate": 6.909826916990122e-06, "loss": 3.7072, "step": 195500 }, { "epoch": 1.61, "learning_rate": 6.901497609488747e-06, "loss": 3.7006, "step": 196000 }, { "epoch": 1.61, "learning_rate": 6.893168301987373e-06, "loss": 3.7001, "step": 196500 }, { "epoch": 1.61, "learning_rate": 6.884838994485999e-06, "loss": 3.6864, "step": 197000 }, { "epoch": 1.62, "learning_rate": 6.876509686984625e-06, "loss": 3.6866, "step": 197500 }, { "epoch": 1.62, "learning_rate": 6.86818037948325e-06, "loss": 3.694, "step": 198000 }, { "epoch": 1.63, "learning_rate": 6.859851071981876e-06, "loss": 3.6737, "step": 198500 }, { "epoch": 1.63, "learning_rate": 6.8515217644805024e-06, "loss": 3.6765, "step": 199000 }, { "epoch": 1.63, "learning_rate": 6.8431924569791274e-06, "loss": 3.6702, "step": 199500 }, { "epoch": 1.64, "learning_rate": 6.8348631494777525e-06, "loss": 3.6767, "step": 200000 }, { "epoch": 1.64, "eval_accuracy": 0.40780666387337433, "eval_loss": 3.566912889480591, "eval_runtime": 1768.7487, "eval_samples_per_second": 174.336, "eval_steps_per_second": 3.633, "step": 200000 }, { "epoch": 1.64, "step": 200001, "total_flos": 2.810618651129938e+18, "train_loss": 5.382437963459396, "train_runtime": 198262.7738, "train_samples_per_second": 147.752, "train_steps_per_second": 3.078 } ], "max_steps": 610290, "num_train_epochs": 5, "total_flos": 2.810618651129938e+18, "trial_name": null, "trial_params": null }