{"step": 0, "epoch": 0, "elapsed_seconds": 74.11151157399999, "eval_loss": 0.46460649371147156, "eval_model_preparation_time": 0.0235, "eval_runtime": 33.1568, "eval_samples_per_second": 4.403, "eval_steps_per_second": 4.403}
{"step": 5, "epoch": 0.03638017280582083, "elapsed_seconds": 173.52221370200002, "loss": 0.4336805820465088, "grad_norm": 4.60339879989624, "learning_rate": 2.857142857142857e-05}
{"step": 10, "epoch": 0.07276034561164166, "elapsed_seconds": 268.747316829, "loss": 0.11467623710632324, "grad_norm": 1.2523683309555054, "learning_rate": 6.428571428571429e-05}
{"step": 15, "epoch": 0.10914051841746249, "elapsed_seconds": 363.63161735799997, "loss": 0.027027195692062377, "grad_norm": 0.11104565113782883, "learning_rate": 0.0001}
{"step": 20, "epoch": 0.14552069122328332, "elapsed_seconds": 456.31930702899996, "loss": 0.01334860473871231, "grad_norm": 1.1765533685684204, "learning_rate": 9.991016468658499e-05}
{"step": 25, "epoch": 0.18190086402910413, "elapsed_seconds": 553.1625098330001, "loss": 0.013230687379837036, "grad_norm": 0.4148058295249939, "learning_rate": 9.964098156168142e-05}
{"step": 25, "epoch": 0.18190086402910413, "elapsed_seconds": 587.516700766, "eval_loss": 0.01401277631521225, "eval_model_preparation_time": 0.0235, "eval_runtime": 33.3166, "eval_samples_per_second": 4.382, "eval_steps_per_second": 4.382}
{"step": 30, "epoch": 0.21828103683492497, "elapsed_seconds": 691.042571865, "loss": 0.012825070321559906, "grad_norm": 0.21652083098888397, "learning_rate": 9.919341791130496e-05}
{"step": 35, "epoch": 0.2546612096407458, "elapsed_seconds": 783.232846796, "loss": 0.009455868601799011, "grad_norm": 0.013632099144160748, "learning_rate": 9.85690820162878e-05}
{"step": 40, "epoch": 0.29104138244656663, "elapsed_seconds": 876.891998315, "loss": 0.004460744932293892, "grad_norm": 0.03066296875476837, "learning_rate": 9.777021737306214e-05}
{"step": 45, "epoch": 0.3274215552523875, "elapsed_seconds": 976.7768998180001, "loss": 0.004046187549829483, "grad_norm": 0.009251159615814686, "learning_rate": 9.6799694631852e-05}
{"step": 50, "epoch": 0.36380172805820826, "elapsed_seconds": 1071.410150558, "loss": 0.003928054124116898, "grad_norm": 0.06524331122636795, "learning_rate": 9.56610012812427e-05}
{"step": 50, "epoch": 0.36380172805820826, "elapsed_seconds": 1105.281500604, "eval_loss": 0.004336625803261995, "eval_model_preparation_time": 0.0235, "eval_runtime": 33.0859, "eval_samples_per_second": 4.413, "eval_steps_per_second": 4.413}
{"step": 55, "epoch": 0.4001819008640291, "elapsed_seconds": 1203.948556652, "loss": 0.004953467100858688, "grad_norm": 0.20901039242744446, "learning_rate": 9.435822911619564e-05}
{"step": 60, "epoch": 0.43656207366984995, "elapsed_seconds": 1298.434868066, "loss": 0.003239613026380539, "grad_norm": 0.06503762304782867, "learning_rate": 9.289605953454107e-05}
{"step": 65, "epoch": 0.47294224647567074, "elapsed_seconds": 1389.272554057, "loss": 0.005726096406579018, "grad_norm": 0.09419463574886322, "learning_rate": 9.127974671478432e-05}
{"step": 70, "epoch": 0.5093224192814916, "elapsed_seconds": 1486.202438671, "loss": 0.007524035125970841, "grad_norm": 0.1491200178861618, "learning_rate": 8.951509873567499e-05}
{"step": 75, "epoch": 0.5457025920873124, "elapsed_seconds": 1579.997498743, "loss": 0.008507375419139863, "grad_norm": 0.13335324823856354, "learning_rate": 8.760845670538387e-05}
{"step": 75, "epoch": 0.5457025920873124, "elapsed_seconds": 1614.079666992, "eval_loss": 0.005701708607375622, "eval_model_preparation_time": 0.0235, "eval_runtime": 33.3334, "eval_samples_per_second": 4.38, "eval_steps_per_second": 4.38}
{"step": 80, "epoch": 0.5820827648931333, "elapsed_seconds": 1725.674950112, "loss": 0.004670526832342148, "grad_norm": 0.06246928870677948, "learning_rate": 8.556667197528543e-05}
{"step": 85, "epoch": 0.6184629376989541, "elapsed_seconds": 1823.488111403, "loss": 0.00527183786034584, "grad_norm": 0.04452742263674736, "learning_rate": 8.339708152022585e-05}
{"step": 90, "epoch": 0.654843110504775, "elapsed_seconds": 1929.594004288, "loss": 0.0018029708415269851, "grad_norm": 0.03127885237336159, "learning_rate": 8.110748157374565e-05}
{"step": 95, "epoch": 0.6912232833105957, "elapsed_seconds": 2028.764289892, "loss": 0.0026896605268120766, "grad_norm": 0.03398139402270317, "learning_rate": 7.870609961299627e-05}
{"step": 100, "epoch": 0.7276034561164165, "elapsed_seconds": 2123.432278635, "loss": 0.005595289170742035, "grad_norm": 0.050191573798656464, "learning_rate": 7.620156479402066e-05}
{"step": 100, "epoch": 0.7276034561164165, "elapsed_seconds": 2157.538371656, "eval_loss": 0.004390799906104803, "eval_model_preparation_time": 0.0235, "eval_runtime": 33.2467, "eval_samples_per_second": 4.391, "eval_steps_per_second": 4.391}
{"step": 105, "epoch": 0.7639836289222374, "elapsed_seconds": 2269.563913525, "loss": 0.002706507220864296, "grad_norm": 0.044108279049396515, "learning_rate": 7.360287694363566e-05}
{"step": 110, "epoch": 0.8003638017280582, "elapsed_seconds": 2367.819704313, "loss": 0.00048029283061623573, "grad_norm": 0.001967919059097767, "learning_rate": 7.091937421934157e-05}
{"step": 115, "epoch": 0.836743974533879, "elapsed_seconds": 2463.5002859300002, "loss": 0.008446679264307023, "grad_norm": 0.09018801897764206, "learning_rate": 6.816069955346985e-05}
{"step": 120, "epoch": 0.8731241473396999, "elapsed_seconds": 2559.333547103, "loss": 0.0031481429934501646, "grad_norm": 0.08786269277334213, "learning_rate": 6.533676600214928e-05}
{"step": 125, "epoch": 0.9095043201455207, "elapsed_seconds": 2654.273697655, "loss": 0.0030761130154132845, "grad_norm": 0.05768860504031181, "learning_rate": 6.245772112360568e-05}
{"step": 125, "epoch": 0.9095043201455207, "elapsed_seconds": 2688.595476632, "eval_loss": 0.00333703332580626, "eval_model_preparation_time": 0.0235, "eval_runtime": 33.4197, "eval_samples_per_second": 4.369, "eval_steps_per_second": 4.369}
{"step": 130, "epoch": 0.9458844929513415, "elapsed_seconds": 2792.400933789, "loss": 0.005125337466597557, "grad_norm": 0.11686617881059647, "learning_rate": 5.953391051379904e-05}
{"step": 135, "epoch": 0.9822646657571623, "elapsed_seconds": 2891.630253004, "loss": 0.004830723255872726, "grad_norm": 0.0008565874304622412, "learning_rate": 5.6575840630429286e-05}
{"step": 140, "epoch": 1.0145520691223284, "elapsed_seconds": 2976.699854047, "loss": 0.003410911187529564, "grad_norm": 0.0016425653593614697, "learning_rate": 5.359414103889947e-05}
{"step": 145, "epoch": 1.050932241928149, "elapsed_seconds": 3071.215896147, "loss": 0.0023501778021454813, "grad_norm": 0.0004682157014030963, "learning_rate": 5.059952621590216e-05}
{"step": 150, "epoch": 1.08731241473397, "elapsed_seconds": 3166.382255453, "loss": 0.0037303082644939423, "grad_norm": 0.09547750651836395, "learning_rate": 4.7602757047884595e-05}
{"step": 150, "epoch": 1.08731241473397, "elapsed_seconds": 3200.668931496, "eval_loss": 0.0031906324438750744, "eval_model_preparation_time": 0.0235, "eval_runtime": 33.2054, "eval_samples_per_second": 4.397, "eval_steps_per_second": 4.397}
{"step": 155, "epoch": 1.1236925875397907, "elapsed_seconds": 3307.433728732, "loss": 0.0014592526480555535, "grad_norm": 0.0011943726567551494, "learning_rate": 4.461460216274445e-05}
{"step": 160, "epoch": 1.1600727603456116, "elapsed_seconds": 3403.6597904660002, "loss": 0.0036914896219968794, "grad_norm": 0.050135571509599686, "learning_rate": 4.1645799233707284e-05}
{"step": 165, "epoch": 1.1964529331514324, "elapsed_seconds": 3504.651429725, "loss": 0.0005274250637739897, "grad_norm": 0.0007454394362866879, "learning_rate": 3.870701639443698e-05}
{"step": 170, "epoch": 1.2328331059572533, "elapsed_seconds": 3602.498029947, "loss": 0.010010915249586106, "grad_norm": 0.06327220797538757, "learning_rate": 3.580881390403052e-05}
{"step": 175, "epoch": 1.269213278763074, "elapsed_seconds": 3707.224299824, "loss": 0.003303169831633568, "grad_norm": 0.05803873389959335, "learning_rate": 3.296160619965056e-05}
{"step": 175, "epoch": 1.269213278763074, "elapsed_seconds": 3741.623573911, "eval_loss": 0.00334456330165267, "eval_model_preparation_time": 0.0235, "eval_runtime": 33.4566, "eval_samples_per_second": 4.364, "eval_steps_per_second": 4.364}
{"step": 180, "epoch": 1.305593451568895, "elapsed_seconds": 3861.555446486, "loss": 0.004390953853726387, "grad_norm": 0.06663581728935242, "learning_rate": 3.0175624473156316e-05}
{"step": 185, "epoch": 1.3419736243747158, "elapsed_seconds": 3961.181201452, "loss": 0.004745687544345856, "grad_norm": 0.014044524170458317, "learning_rate": 2.7460879906210487e-05}
{"step": 190, "epoch": 1.3783537971805366, "elapsed_seconds": 4058.333197755, "loss": 0.004839410632848739, "grad_norm": 0.06025797501206398, "learning_rate": 2.482712769597363e-05}
{"step": 195, "epoch": 1.4147339699863575, "elapsed_seconds": 4149.416457785999, "loss": 0.003374819457530975, "grad_norm": 0.0006364663713611662, "learning_rate": 2.22838320006563e-05}
{"step": 200, "epoch": 1.4511141427921783, "elapsed_seconds": 4246.132818931, "loss": 0.0069746367633342745, "grad_norm": 0.07674790173768997, "learning_rate": 1.9840131930894333e-05}
{"step": 200, "epoch": 1.4511141427921783, "elapsed_seconds": 4280.256723756, "eval_loss": 0.0039616054855287075, "eval_model_preparation_time": 0.0235, "eval_runtime": 33.2396, "eval_samples_per_second": 4.392, "eval_steps_per_second": 4.392}
{"step": 205, "epoch": 1.4874943155979992, "elapsed_seconds": 4376.854840755999, "loss": 0.005431993305683136, "grad_norm": 0.03662000223994255, "learning_rate": 1.7504808709154104e-05}
{"step": 210, "epoch": 1.5238744884038198, "elapsed_seconds": 4469.472708171999, "loss": 0.003454185277223587, "grad_norm": 0.03310539945960045, "learning_rate": 1.5286254115177623e-05}
{"step": 215, "epoch": 1.5602546612096408, "elapsed_seconds": 4566.978258567, "loss": 0.001565326564013958, "grad_norm": 0.024926867336034775, "learning_rate": 1.3192440330856004e-05}
{"step": 220, "epoch": 1.5966348340154615, "elapsed_seconds": 4661.4993927059995, "loss": 0.0019487164914608, "grad_norm": 0.009774737991392612, "learning_rate": 1.123089129289117e-05}
{"step": 225, "epoch": 1.6330150068212825, "elapsed_seconds": 4754.554816119, "loss": 0.004187266901135444, "grad_norm": 0.030321111902594566, "learning_rate": 9.408655656187282e-06}
{"step": 225, "epoch": 1.6330150068212825, "elapsed_seconds": 4788.6018220589995, "eval_loss": 0.004435760900378227, "eval_model_preparation_time": 0.0235, "eval_runtime": 33.1944, "eval_samples_per_second": 4.398, "eval_steps_per_second": 4.398}
{"step": 230, "epoch": 1.6693951796271032, "elapsed_seconds": 4891.549733037, "loss": 0.003941274434328079, "grad_norm": 0.0819157138466835, "learning_rate": 7.732281465125907e-06}
{"step": 235, "epoch": 1.7057753524329242, "elapsed_seconds": 4987.418221182999, "loss": 0.004403543472290039, "grad_norm": 0.13472600281238556, "learning_rate": 6.207792623741249e-06}
{"step": 240, "epoch": 1.7421555252387448, "elapsed_seconds": 5084.461640371, "loss": 0.0035004250705242155, "grad_norm": 0.08246049284934998, "learning_rate": 4.840667249347824e-06}
{"step": 245, "epoch": 1.7785356980445657, "elapsed_seconds": 5180.950417845, "loss": 0.0019740713760256767, "grad_norm": 0.027418652549386024, "learning_rate": 3.63581798740511e-06}
{"step": 250, "epoch": 1.8149158708503865, "elapsed_seconds": 5274.363783143999, "loss": 0.005988799408078194, "grad_norm": 0.06135931238532066, "learning_rate": 2.59757435835567e-06}
{"step": 250, "epoch": 1.8149158708503865, "elapsed_seconds": 5308.316412030999, "eval_loss": 0.004120469558984041, "eval_model_preparation_time": 0.0235, "eval_runtime": 33.2946, "eval_samples_per_second": 4.385, "eval_steps_per_second": 4.385}
{"step": 255, "epoch": 1.8512960436562074, "elapsed_seconds": 5412.821858069, "loss": 0.003019887208938599, "grad_norm": 0.07759344577789307, "learning_rate": 1.729667199872187e-06}
{"step": 260, "epoch": 1.8876762164620282, "elapsed_seconds": 5511.523865796, "loss": 0.002903467230498791, "grad_norm": 0.10335779935121536, "learning_rate": 1.0352152604185428e-06}
{"step": 265, "epoch": 1.924056389267849, "elapsed_seconds": 5611.368567242, "loss": 0.004673852026462555, "grad_norm": 0.08141176402568817, "learning_rate": 5.167139923000553e-07}
{"step": 270, "epoch": 1.96043656207367, "elapsed_seconds": 5708.749756812999, "loss": 0.003966786339879036, "grad_norm": 0.0932755097746849, "learning_rate": 1.760265844738096e-07}
{"step": 275, "epoch": 1.9968167348794905, "elapsed_seconds": 5800.910231717, "loss": 0.004456077888607979, "grad_norm": 0.030060311779379845, "learning_rate": 1.4377267342158274e-08}
{"step": 275, "epoch": 1.9968167348794905, "elapsed_seconds": 5834.714766614, "eval_loss": 0.004110506735742092, "eval_model_preparation_time": 0.0235, "eval_runtime": 33.2296, "eval_samples_per_second": 4.394, "eval_steps_per_second": 4.394}
{"step": 276, "epoch": 2.0, "elapsed_seconds": 5881.5106768939995, "eval_loss": 0.004070464987307787, "eval_model_preparation_time": 0.0235, "eval_runtime": 33.2187, "eval_samples_per_second": 4.395, "eval_steps_per_second": 4.395}
{"step": 276, "epoch": 2.0, "elapsed_seconds": 5885.911110157999, "train_runtime": 5810.4825, "train_samples_per_second": 0.757, "train_steps_per_second": 0.048, "total_flos": 1.77759314768726e+17, "train_loss": 0.014919810814519777}
{"step": 276, "epoch": 2.0, "elapsed_seconds": 5919.986330944999, "eval_loss": 0.0031906324438750744, "eval_model_preparation_time": 0.0235, "eval_runtime": 33.1673, "eval_samples_per_second": 4.402, "eval_steps_per_second": 4.402}