{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 3.0, "eval_steps": 200, "global_step": 1947, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "entropy": 1.1166099309921265, "epoch": 0.0015408320493066256, "grad_norm": 2.048114538192749, "learning_rate": 0.0, "loss": 1.8954, "mean_token_accuracy": 0.6198432445526123, "num_tokens": 65392.0, "step": 1 }, { "entropy": 1.067945957183838, "epoch": 0.0030816640986132513, "grad_norm": 2.459822416305542, "learning_rate": 2.0408163265306121e-07, "loss": 1.9595, "mean_token_accuracy": 0.6000674962997437, "num_tokens": 127860.0, "step": 2 }, { "entropy": 1.076798915863037, "epoch": 0.004622496147919877, "grad_norm": 3.8358404636383057, "learning_rate": 4.0816326530612243e-07, "loss": 2.0135, "mean_token_accuracy": 0.5936726033687592, "num_tokens": 188378.0, "step": 3 }, { "entropy": 1.0898915529251099, "epoch": 0.0061633281972265025, "grad_norm": 2.5504703521728516, "learning_rate": 6.122448979591837e-07, "loss": 2.0316, "mean_token_accuracy": 0.5925821363925934, "num_tokens": 247429.0, "step": 4 }, { "entropy": 1.0819076895713806, "epoch": 0.007704160246533128, "grad_norm": 2.4035096168518066, "learning_rate": 8.163265306122449e-07, "loss": 2.0117, "mean_token_accuracy": 0.5945810675621033, "num_tokens": 305427.0, "step": 5 }, { "entropy": 1.114140510559082, "epoch": 0.009244992295839754, "grad_norm": 2.3264708518981934, "learning_rate": 1.0204081632653063e-06, "loss": 2.0568, "mean_token_accuracy": 0.5870096981525421, "num_tokens": 362248.0, "step": 6 }, { "entropy": 1.1025208234786987, "epoch": 0.01078582434514638, "grad_norm": 2.3177919387817383, "learning_rate": 1.2244897959183673e-06, "loss": 2.0433, "mean_token_accuracy": 0.5885131657123566, "num_tokens": 418035.0, "step": 7 }, { "entropy": 1.1331512928009033, "epoch": 0.012326656394453005, "grad_norm": 2.244300603866577, "learning_rate": 1.4285714285714286e-06, "loss": 2.1019, "mean_token_accuracy": 0.5755126178264618, "num_tokens": 472726.0, "step": 8 }, { "entropy": 1.1354270577430725, "epoch": 0.01386748844375963, "grad_norm": 1.9873753786087036, "learning_rate": 1.6326530612244897e-06, "loss": 2.051, "mean_token_accuracy": 0.5843476057052612, "num_tokens": 526017.0, "step": 9 }, { "entropy": 1.1884055733680725, "epoch": 0.015408320493066256, "grad_norm": 2.2195703983306885, "learning_rate": 1.8367346938775512e-06, "loss": 2.1282, "mean_token_accuracy": 0.5666418969631195, "num_tokens": 577023.0, "step": 10 }, { "entropy": 1.1007508039474487, "epoch": 0.01694915254237288, "grad_norm": 3.1744253635406494, "learning_rate": 2.0408163265306125e-06, "loss": 2.4744, "mean_token_accuracy": 0.5443819761276245, "num_tokens": 595368.0, "step": 11 }, { "entropy": 1.0786568522453308, "epoch": 0.01848998459167951, "grad_norm": 6.220513343811035, "learning_rate": 2.244897959183674e-06, "loss": 3.1811, "mean_token_accuracy": 0.5258253216743469, "num_tokens": 604807.0, "step": 12 }, { "entropy": 1.1419293284416199, "epoch": 0.020030816640986132, "grad_norm": 1.8345328569412231, "learning_rate": 2.4489795918367347e-06, "loss": 2.0926, "mean_token_accuracy": 0.5589998364448547, "num_tokens": 641042.0, "step": 13 }, { "entropy": 1.1164453625679016, "epoch": 0.02157164869029276, "grad_norm": 1.4737261533737183, "learning_rate": 2.6530612244897964e-06, "loss": 1.8581, "mean_token_accuracy": 0.6111711859703064, "num_tokens": 705227.0, "step": 14 }, { "entropy": 1.1013811826705933, "epoch": 0.023112480739599383, "grad_norm": 1.4611196517944336, "learning_rate": 2.8571428571428573e-06, "loss": 1.8555, "mean_token_accuracy": 0.6112561523914337, "num_tokens": 766428.0, "step": 15 }, { "entropy": 1.1715907454490662, "epoch": 0.02465331278890601, "grad_norm": 1.3365057706832886, "learning_rate": 3.0612244897959185e-06, "loss": 1.8977, "mean_token_accuracy": 0.5997171998023987, "num_tokens": 825996.0, "step": 16 }, { "entropy": 1.1750763058662415, "epoch": 0.026194144838212634, "grad_norm": 1.2849830389022827, "learning_rate": 3.2653061224489794e-06, "loss": 1.8625, "mean_token_accuracy": 0.5998840034008026, "num_tokens": 884283.0, "step": 17 }, { "entropy": 1.2399238348007202, "epoch": 0.02773497688751926, "grad_norm": 1.133266806602478, "learning_rate": 3.469387755102041e-06, "loss": 1.8934, "mean_token_accuracy": 0.5958154201507568, "num_tokens": 941447.0, "step": 18 }, { "entropy": 1.268789529800415, "epoch": 0.029275808936825885, "grad_norm": 1.02197265625, "learning_rate": 3.6734693877551024e-06, "loss": 1.8573, "mean_token_accuracy": 0.5979815423488617, "num_tokens": 997567.0, "step": 19 }, { "entropy": 1.3028552532196045, "epoch": 0.030816640986132512, "grad_norm": 0.9115889668464661, "learning_rate": 3.877551020408164e-06, "loss": 1.8567, "mean_token_accuracy": 0.5965172350406647, "num_tokens": 1052606.0, "step": 20 }, { "entropy": 1.3288394808769226, "epoch": 0.032357473035439135, "grad_norm": 0.8141210079193115, "learning_rate": 4.081632653061225e-06, "loss": 1.823, "mean_token_accuracy": 0.5941728949546814, "num_tokens": 1106350.0, "step": 21 }, { "entropy": 1.3815255761146545, "epoch": 0.03389830508474576, "grad_norm": 0.7993112206459045, "learning_rate": 4.2857142857142855e-06, "loss": 1.8388, "mean_token_accuracy": 0.5887587070465088, "num_tokens": 1158133.0, "step": 22 }, { "entropy": 1.285710632801056, "epoch": 0.03543913713405239, "grad_norm": 1.0425184965133667, "learning_rate": 4.489795918367348e-06, "loss": 1.931, "mean_token_accuracy": 0.5787847340106964, "num_tokens": 1188104.0, "step": 23 }, { "entropy": 1.2443310022354126, "epoch": 0.03697996918335902, "grad_norm": 3.2980475425720215, "learning_rate": 4.693877551020409e-06, "loss": 2.5356, "mean_token_accuracy": 0.5419443547725677, "num_tokens": 1198641.0, "step": 24 }, { "entropy": 1.2641022205352783, "epoch": 0.03852080123266564, "grad_norm": 4.0346245765686035, "learning_rate": 4.897959183673469e-06, "loss": 2.7635, "mean_token_accuracy": 0.5503197908401489, "num_tokens": 1206619.0, "step": 25 }, { "entropy": 1.2999488711357117, "epoch": 0.040061633281972264, "grad_norm": 0.5223360061645508, "learning_rate": 5.1020408163265315e-06, "loss": 1.565, "mean_token_accuracy": 0.6481170952320099, "num_tokens": 1271890.0, "step": 26 }, { "entropy": 1.3394767045974731, "epoch": 0.04160246533127889, "grad_norm": 0.5582960844039917, "learning_rate": 5.306122448979593e-06, "loss": 1.6087, "mean_token_accuracy": 0.6302163600921631, "num_tokens": 1334763.0, "step": 27 }, { "entropy": 1.3502419590950012, "epoch": 0.04314329738058552, "grad_norm": 0.5407151579856873, "learning_rate": 5.510204081632653e-06, "loss": 1.6039, "mean_token_accuracy": 0.6266109645366669, "num_tokens": 1395637.0, "step": 28 }, { "entropy": 1.355388581752777, "epoch": 0.04468412942989214, "grad_norm": 0.5162605047225952, "learning_rate": 5.7142857142857145e-06, "loss": 1.5536, "mean_token_accuracy": 0.6347056925296783, "num_tokens": 1454741.0, "step": 29 }, { "entropy": 1.3951674699783325, "epoch": 0.046224961479198766, "grad_norm": 0.5207234621047974, "learning_rate": 5.918367346938776e-06, "loss": 1.5851, "mean_token_accuracy": 0.6318900883197784, "num_tokens": 1512507.0, "step": 30 }, { "entropy": 1.410614252090454, "epoch": 0.04776579352850539, "grad_norm": 0.5331891775131226, "learning_rate": 6.122448979591837e-06, "loss": 1.5731, "mean_token_accuracy": 0.6373177468776703, "num_tokens": 1568901.0, "step": 31 }, { "entropy": 1.415243685245514, "epoch": 0.04930662557781202, "grad_norm": 0.5451644659042358, "learning_rate": 6.326530612244899e-06, "loss": 1.5559, "mean_token_accuracy": 0.6380660533905029, "num_tokens": 1624048.0, "step": 32 }, { "entropy": 1.3981574177742004, "epoch": 0.05084745762711865, "grad_norm": 0.6062421798706055, "learning_rate": 6.530612244897959e-06, "loss": 1.5207, "mean_token_accuracy": 0.6433392465114594, "num_tokens": 1678060.0, "step": 33 }, { "entropy": 1.4160131216049194, "epoch": 0.05238828967642527, "grad_norm": 0.6023133993148804, "learning_rate": 6.734693877551021e-06, "loss": 1.5165, "mean_token_accuracy": 0.6423613429069519, "num_tokens": 1730755.0, "step": 34 }, { "entropy": 1.4293076992034912, "epoch": 0.053929121725731895, "grad_norm": 0.5575998425483704, "learning_rate": 6.938775510204082e-06, "loss": 1.5304, "mean_token_accuracy": 0.6452022194862366, "num_tokens": 1774984.0, "step": 35 }, { "entropy": 1.4634100794792175, "epoch": 0.05546995377503852, "grad_norm": 1.9630212783813477, "learning_rate": 7.1428571428571436e-06, "loss": 1.7235, "mean_token_accuracy": 0.6272614598274231, "num_tokens": 1786336.0, "step": 36 }, { "entropy": 1.3920773267745972, "epoch": 0.05701078582434515, "grad_norm": 2.1560463905334473, "learning_rate": 7.346938775510205e-06, "loss": 1.6868, "mean_token_accuracy": 0.6434301733970642, "num_tokens": 1795508.0, "step": 37 }, { "entropy": 1.3546090126037598, "epoch": 0.05855161787365177, "grad_norm": 0.5001200437545776, "learning_rate": 7.551020408163265e-06, "loss": 1.4208, "mean_token_accuracy": 0.680578351020813, "num_tokens": 1831917.0, "step": 38 }, { "entropy": 1.3750313520431519, "epoch": 0.060092449922958396, "grad_norm": 0.3193170428276062, "learning_rate": 7.755102040816327e-06, "loss": 1.3912, "mean_token_accuracy": 0.6743785738945007, "num_tokens": 1896054.0, "step": 39 }, { "entropy": 1.3467950224876404, "epoch": 0.061633281972265024, "grad_norm": 0.3405231833457947, "learning_rate": 7.959183673469388e-06, "loss": 1.3571, "mean_token_accuracy": 0.675582766532898, "num_tokens": 1957607.0, "step": 40 }, { "entropy": 1.4068493247032166, "epoch": 0.06317411402157165, "grad_norm": 0.27743232250213623, "learning_rate": 8.16326530612245e-06, "loss": 1.3901, "mean_token_accuracy": 0.6676525771617889, "num_tokens": 2017678.0, "step": 41 }, { "entropy": 1.38527250289917, "epoch": 0.06471494607087827, "grad_norm": 0.2759312093257904, "learning_rate": 8.36734693877551e-06, "loss": 1.3653, "mean_token_accuracy": 0.6725476086139679, "num_tokens": 2076346.0, "step": 42 }, { "entropy": 1.4152355194091797, "epoch": 0.0662557781201849, "grad_norm": 0.2714540362358093, "learning_rate": 8.571428571428571e-06, "loss": 1.3756, "mean_token_accuracy": 0.6703079342842102, "num_tokens": 2133806.0, "step": 43 }, { "entropy": 1.4345101714134216, "epoch": 0.06779661016949153, "grad_norm": 0.25623807311058044, "learning_rate": 8.775510204081633e-06, "loss": 1.3943, "mean_token_accuracy": 0.6705135107040405, "num_tokens": 2190022.0, "step": 44 }, { "entropy": 1.4259186387062073, "epoch": 0.06933744221879815, "grad_norm": 0.2355126142501831, "learning_rate": 8.979591836734695e-06, "loss": 1.3828, "mean_token_accuracy": 0.6708194315433502, "num_tokens": 2245027.0, "step": 45 }, { "entropy": 1.423492670059204, "epoch": 0.07087827426810478, "grad_norm": 0.24491247534751892, "learning_rate": 9.183673469387756e-06, "loss": 1.3612, "mean_token_accuracy": 0.669874757528305, "num_tokens": 2298522.0, "step": 46 }, { "entropy": 1.4575979709625244, "epoch": 0.0724191063174114, "grad_norm": 0.2854325473308563, "learning_rate": 9.387755102040818e-06, "loss": 1.3973, "mean_token_accuracy": 0.6609207093715668, "num_tokens": 2350213.0, "step": 47 }, { "entropy": 1.4179415702819824, "epoch": 0.07395993836671803, "grad_norm": 0.456508994102478, "learning_rate": 9.591836734693878e-06, "loss": 1.3787, "mean_token_accuracy": 0.6705058515071869, "num_tokens": 2386852.0, "step": 48 }, { "entropy": 1.4314310550689697, "epoch": 0.07550077041602465, "grad_norm": 1.5667709112167358, "learning_rate": 9.795918367346939e-06, "loss": 1.3767, "mean_token_accuracy": 0.6764441132545471, "num_tokens": 2398051.0, "step": 49 }, { "entropy": 1.2926554679870605, "epoch": 0.07704160246533127, "grad_norm": 1.480271339416504, "learning_rate": 1e-05, "loss": 1.2987, "mean_token_accuracy": 0.7148274779319763, "num_tokens": 2406476.0, "step": 50 }, { "entropy": 1.3785969018936157, "epoch": 0.07858243451463791, "grad_norm": 0.2692592442035675, "learning_rate": 1.0204081632653063e-05, "loss": 1.3064, "mean_token_accuracy": 0.6913812458515167, "num_tokens": 2471840.0, "step": 51 }, { "entropy": 1.37378591299057, "epoch": 0.08012326656394453, "grad_norm": 0.22790347039699554, "learning_rate": 1.0408163265306123e-05, "loss": 1.3175, "mean_token_accuracy": 0.6800710260868073, "num_tokens": 2534347.0, "step": 52 }, { "entropy": 1.358722746372223, "epoch": 0.08166409861325115, "grad_norm": 0.2471814900636673, "learning_rate": 1.0612244897959186e-05, "loss": 1.2947, "mean_token_accuracy": 0.6801523864269257, "num_tokens": 2594781.0, "step": 53 }, { "entropy": 1.3550601601600647, "epoch": 0.08320493066255778, "grad_norm": 0.21714967489242554, "learning_rate": 1.0816326530612246e-05, "loss": 1.2922, "mean_token_accuracy": 0.6850081384181976, "num_tokens": 2653914.0, "step": 54 }, { "entropy": 1.3757457733154297, "epoch": 0.0847457627118644, "grad_norm": 0.19867496192455292, "learning_rate": 1.1020408163265306e-05, "loss": 1.3202, "mean_token_accuracy": 0.6776906251907349, "num_tokens": 2711834.0, "step": 55 }, { "entropy": 1.3785210847854614, "epoch": 0.08628659476117104, "grad_norm": 0.20601148903369904, "learning_rate": 1.1224489795918367e-05, "loss": 1.3264, "mean_token_accuracy": 0.6730702221393585, "num_tokens": 2768665.0, "step": 56 }, { "entropy": 1.390173852443695, "epoch": 0.08782742681047766, "grad_norm": 0.20482276380062103, "learning_rate": 1.1428571428571429e-05, "loss": 1.3356, "mean_token_accuracy": 0.6767943799495697, "num_tokens": 2824387.0, "step": 57 }, { "entropy": 1.339047372341156, "epoch": 0.08936825885978428, "grad_norm": 0.1985456794500351, "learning_rate": 1.1632653061224491e-05, "loss": 1.2938, "mean_token_accuracy": 0.6821994483470917, "num_tokens": 2878621.0, "step": 58 }, { "entropy": 1.3643373847007751, "epoch": 0.09090909090909091, "grad_norm": 0.19926731288433075, "learning_rate": 1.1836734693877552e-05, "loss": 1.3254, "mean_token_accuracy": 0.6756160259246826, "num_tokens": 2931140.0, "step": 59 }, { "entropy": 1.3715867400169373, "epoch": 0.09244992295839753, "grad_norm": 0.2262343019247055, "learning_rate": 1.2040816326530614e-05, "loss": 1.35, "mean_token_accuracy": 0.6744479238986969, "num_tokens": 2973875.0, "step": 60 }, { "entropy": 1.256079614162445, "epoch": 0.09399075500770417, "grad_norm": 0.9030208587646484, "learning_rate": 1.2244897959183674e-05, "loss": 1.2319, "mean_token_accuracy": 0.7004675567150116, "num_tokens": 2985544.0, "step": 61 }, { "entropy": 1.1448688507080078, "epoch": 0.09553158705701079, "grad_norm": 0.6902526617050171, "learning_rate": 1.2448979591836736e-05, "loss": 1.143, "mean_token_accuracy": 0.7256202101707458, "num_tokens": 2994772.0, "step": 62 }, { "entropy": 1.1552671194076538, "epoch": 0.0970724191063174, "grad_norm": 0.30490317940711975, "learning_rate": 1.2653061224489798e-05, "loss": 1.2507, "mean_token_accuracy": 0.7333579063415527, "num_tokens": 3031135.0, "step": 63 }, { "entropy": 1.2912518382072449, "epoch": 0.09861325115562404, "grad_norm": 0.24345964193344116, "learning_rate": 1.2857142857142859e-05, "loss": 1.257, "mean_token_accuracy": 0.6965952813625336, "num_tokens": 3095552.0, "step": 64 }, { "entropy": 1.2751701474189758, "epoch": 0.10015408320493066, "grad_norm": 0.1774371713399887, "learning_rate": 1.3061224489795918e-05, "loss": 1.2426, "mean_token_accuracy": 0.6905304491519928, "num_tokens": 3156859.0, "step": 65 }, { "entropy": 1.2700653672218323, "epoch": 0.1016949152542373, "grad_norm": 0.18590854108333588, "learning_rate": 1.326530612244898e-05, "loss": 1.2515, "mean_token_accuracy": 0.6929566860198975, "num_tokens": 3216491.0, "step": 66 }, { "entropy": 1.2973484992980957, "epoch": 0.10323574730354391, "grad_norm": 0.1647886484861374, "learning_rate": 1.3469387755102042e-05, "loss": 1.2796, "mean_token_accuracy": 0.6860228478908539, "num_tokens": 3274786.0, "step": 67 }, { "entropy": 1.3115774989128113, "epoch": 0.10477657935285054, "grad_norm": 0.18149203062057495, "learning_rate": 1.3673469387755102e-05, "loss": 1.2923, "mean_token_accuracy": 0.6832408607006073, "num_tokens": 3331847.0, "step": 68 }, { "entropy": 1.283263087272644, "epoch": 0.10631741140215717, "grad_norm": 0.17151886224746704, "learning_rate": 1.3877551020408165e-05, "loss": 1.2619, "mean_token_accuracy": 0.6859965324401855, "num_tokens": 3387610.0, "step": 69 }, { "entropy": 1.3135151267051697, "epoch": 0.10785824345146379, "grad_norm": 0.18241697549819946, "learning_rate": 1.4081632653061225e-05, "loss": 1.2911, "mean_token_accuracy": 0.6818918287754059, "num_tokens": 3442333.0, "step": 70 }, { "entropy": 1.298197329044342, "epoch": 0.10939907550077041, "grad_norm": 0.18023963272571564, "learning_rate": 1.4285714285714287e-05, "loss": 1.2874, "mean_token_accuracy": 0.6836556196212769, "num_tokens": 3495838.0, "step": 71 }, { "entropy": 1.3163077235221863, "epoch": 0.11093990755007704, "grad_norm": 0.18750514090061188, "learning_rate": 1.448979591836735e-05, "loss": 1.3002, "mean_token_accuracy": 0.6805817186832428, "num_tokens": 3547396.0, "step": 72 }, { "entropy": 1.218782901763916, "epoch": 0.11248073959938366, "grad_norm": 0.38678720593452454, "learning_rate": 1.469387755102041e-05, "loss": 1.2962, "mean_token_accuracy": 0.690745621919632, "num_tokens": 3574144.0, "step": 73 }, { "entropy": 1.0390467643737793, "epoch": 0.1140215716486903, "grad_norm": 0.7937684059143066, "learning_rate": 1.4897959183673472e-05, "loss": 1.1271, "mean_token_accuracy": 0.7160314321517944, "num_tokens": 3584999.0, "step": 74 }, { "entropy": 0.9610435366630554, "epoch": 0.11556240369799692, "grad_norm": 0.8527204990386963, "learning_rate": 1.510204081632653e-05, "loss": 1.0902, "mean_token_accuracy": 0.7375245690345764, "num_tokens": 3593190.0, "step": 75 }, { "entropy": 1.180212378501892, "epoch": 0.11710323574730354, "grad_norm": 0.3046119809150696, "learning_rate": 1.530612244897959e-05, "loss": 1.1615, "mean_token_accuracy": 0.7213225066661835, "num_tokens": 3658425.0, "step": 76 }, { "entropy": 1.2005605697631836, "epoch": 0.11864406779661017, "grad_norm": 0.2118452787399292, "learning_rate": 1.5510204081632655e-05, "loss": 1.1994, "mean_token_accuracy": 0.7007608711719513, "num_tokens": 3721095.0, "step": 77 }, { "entropy": 1.2498918175697327, "epoch": 0.12018489984591679, "grad_norm": 0.18598024547100067, "learning_rate": 1.5714285714285715e-05, "loss": 1.2323, "mean_token_accuracy": 0.6948350071907043, "num_tokens": 3781860.0, "step": 78 }, { "entropy": 1.2422747611999512, "epoch": 0.12172573189522343, "grad_norm": 0.16501612961292267, "learning_rate": 1.5918367346938776e-05, "loss": 1.2134, "mean_token_accuracy": 0.6985945403575897, "num_tokens": 3841240.0, "step": 79 }, { "entropy": 1.2970284223556519, "epoch": 0.12326656394453005, "grad_norm": 0.17338834702968597, "learning_rate": 1.612244897959184e-05, "loss": 1.2742, "mean_token_accuracy": 0.6833244264125824, "num_tokens": 3899408.0, "step": 80 }, { "entropy": 1.259686291217804, "epoch": 0.12480739599383667, "grad_norm": 0.17090970277786255, "learning_rate": 1.63265306122449e-05, "loss": 1.2426, "mean_token_accuracy": 0.6896885335445404, "num_tokens": 3956509.0, "step": 81 }, { "entropy": 1.3052712678909302, "epoch": 0.1263482280431433, "grad_norm": 0.17299887537956238, "learning_rate": 1.653061224489796e-05, "loss": 1.2913, "mean_token_accuracy": 0.6865420937538147, "num_tokens": 4012444.0, "step": 82 }, { "entropy": 1.2694194912910461, "epoch": 0.12788906009244994, "grad_norm": 0.1667395681142807, "learning_rate": 1.673469387755102e-05, "loss": 1.2409, "mean_token_accuracy": 0.6935325860977173, "num_tokens": 4067267.0, "step": 83 }, { "entropy": 1.300751805305481, "epoch": 0.12942989214175654, "grad_norm": 0.16546165943145752, "learning_rate": 1.6938775510204085e-05, "loss": 1.274, "mean_token_accuracy": 0.6836924254894257, "num_tokens": 4120520.0, "step": 84 }, { "entropy": 1.301823616027832, "epoch": 0.13097072419106318, "grad_norm": 0.16247878968715668, "learning_rate": 1.7142857142857142e-05, "loss": 1.2906, "mean_token_accuracy": 0.6820458769798279, "num_tokens": 4171418.0, "step": 85 }, { "entropy": 1.216541349887848, "epoch": 0.1325115562403698, "grad_norm": 0.46898865699768066, "learning_rate": 1.7346938775510206e-05, "loss": 1.2005, "mean_token_accuracy": 0.6984381675720215, "num_tokens": 4188871.0, "step": 86 }, { "entropy": 1.1531486511230469, "epoch": 0.13405238828967642, "grad_norm": 0.4649360179901123, "learning_rate": 1.7551020408163266e-05, "loss": 1.126, "mean_token_accuracy": 0.7286891639232635, "num_tokens": 4198587.0, "step": 87 }, { "entropy": 1.029190182685852, "epoch": 0.13559322033898305, "grad_norm": 0.3149832785129547, "learning_rate": 1.7755102040816327e-05, "loss": 1.072, "mean_token_accuracy": 0.7597673535346985, "num_tokens": 4234927.0, "step": 88 }, { "entropy": 1.1297488808631897, "epoch": 0.13713405238828968, "grad_norm": 0.19498305022716522, "learning_rate": 1.795918367346939e-05, "loss": 1.129, "mean_token_accuracy": 0.720356673002243, "num_tokens": 4299170.0, "step": 89 }, { "entropy": 1.1790211200714111, "epoch": 0.1386748844375963, "grad_norm": 0.1875903606414795, "learning_rate": 1.816326530612245e-05, "loss": 1.1893, "mean_token_accuracy": 0.7015982568264008, "num_tokens": 4360899.0, "step": 90 }, { "entropy": 1.1867957711219788, "epoch": 0.14021571648690292, "grad_norm": 0.16463136672973633, "learning_rate": 1.836734693877551e-05, "loss": 1.1932, "mean_token_accuracy": 0.7002697587013245, "num_tokens": 4421216.0, "step": 91 }, { "entropy": 1.2240930795669556, "epoch": 0.14175654853620956, "grad_norm": 0.15712255239486694, "learning_rate": 1.8571428571428575e-05, "loss": 1.2418, "mean_token_accuracy": 0.6914651393890381, "num_tokens": 4480182.0, "step": 92 }, { "entropy": 1.2126803398132324, "epoch": 0.14329738058551617, "grad_norm": 0.16738064587116241, "learning_rate": 1.8775510204081636e-05, "loss": 1.2153, "mean_token_accuracy": 0.6958963871002197, "num_tokens": 4538023.0, "step": 93 }, { "entropy": 1.2484999895095825, "epoch": 0.1448382126348228, "grad_norm": 0.16801941394805908, "learning_rate": 1.8979591836734696e-05, "loss": 1.2541, "mean_token_accuracy": 0.6865383684635162, "num_tokens": 4594685.0, "step": 94 }, { "entropy": 1.2355953454971313, "epoch": 0.14637904468412943, "grad_norm": 0.17386969923973083, "learning_rate": 1.9183673469387756e-05, "loss": 1.2535, "mean_token_accuracy": 0.6880841255187988, "num_tokens": 4650161.0, "step": 95 }, { "entropy": 1.217761754989624, "epoch": 0.14791987673343607, "grad_norm": 0.15721255540847778, "learning_rate": 1.9387755102040817e-05, "loss": 1.222, "mean_token_accuracy": 0.6962701380252838, "num_tokens": 4704489.0, "step": 96 }, { "entropy": 1.246911883354187, "epoch": 0.14946070878274267, "grad_norm": 0.16272477805614471, "learning_rate": 1.9591836734693877e-05, "loss": 1.2519, "mean_token_accuracy": 0.6880429983139038, "num_tokens": 4757276.0, "step": 97 }, { "entropy": 1.1872326731681824, "epoch": 0.1510015408320493, "grad_norm": 0.2594091594219208, "learning_rate": 1.979591836734694e-05, "loss": 1.2833, "mean_token_accuracy": 0.6884508728981018, "num_tokens": 4793718.0, "step": 98 }, { "entropy": 0.9742604792118073, "epoch": 0.15254237288135594, "grad_norm": 0.5665745735168457, "learning_rate": 2e-05, "loss": 1.0833, "mean_token_accuracy": 0.7222688496112823, "num_tokens": 4804890.0, "step": 99 }, { "entropy": 0.9292000234127045, "epoch": 0.15408320493066255, "grad_norm": 0.6358155608177185, "learning_rate": 1.9999985565696826e-05, "loss": 1.0191, "mean_token_accuracy": 0.7465425431728363, "num_tokens": 4813209.0, "step": 100 }, { "entropy": 1.1387584209442139, "epoch": 0.15562403697996918, "grad_norm": 0.31268125772476196, "learning_rate": 1.9999942262828967e-05, "loss": 1.093, "mean_token_accuracy": 0.7350356578826904, "num_tokens": 4878712.0, "step": 101 }, { "entropy": 1.138710618019104, "epoch": 0.15716486902927582, "grad_norm": 0.22164300084114075, "learning_rate": 1.999987009152143e-05, "loss": 1.1012, "mean_token_accuracy": 0.7202340066432953, "num_tokens": 4941928.0, "step": 102 }, { "entropy": 1.1990888118743896, "epoch": 0.15870570107858242, "grad_norm": 0.20463788509368896, "learning_rate": 1.9999769051982578e-05, "loss": 1.1626, "mean_token_accuracy": 0.7095012664794922, "num_tokens": 5003137.0, "step": 103 }, { "entropy": 1.2298290729522705, "epoch": 0.16024653312788906, "grad_norm": 0.17075873911380768, "learning_rate": 1.999963914450408e-05, "loss": 1.1957, "mean_token_accuracy": 0.697584867477417, "num_tokens": 5062986.0, "step": 104 }, { "entropy": 1.2346097826957703, "epoch": 0.1617873651771957, "grad_norm": 0.18050366640090942, "learning_rate": 1.9999480369460968e-05, "loss": 1.212, "mean_token_accuracy": 0.6979473829269409, "num_tokens": 5121593.0, "step": 105 }, { "entropy": 1.2363342642784119, "epoch": 0.1633281972265023, "grad_norm": 0.17311757802963257, "learning_rate": 1.9999292727311608e-05, "loss": 1.2046, "mean_token_accuracy": 0.6956012547016144, "num_tokens": 5179009.0, "step": 106 }, { "entropy": 1.2573286890983582, "epoch": 0.16486902927580893, "grad_norm": 0.18123400211334229, "learning_rate": 1.9999076218597693e-05, "loss": 1.2428, "mean_token_accuracy": 0.6897366940975189, "num_tokens": 5235259.0, "step": 107 }, { "entropy": 1.256956696510315, "epoch": 0.16640986132511557, "grad_norm": 0.195603147149086, "learning_rate": 1.9998830843944253e-05, "loss": 1.2504, "mean_token_accuracy": 0.6899679899215698, "num_tokens": 5290188.0, "step": 108 }, { "entropy": 1.2548956871032715, "epoch": 0.1679506933744222, "grad_norm": 0.172136127948761, "learning_rate": 1.999855660405965e-05, "loss": 1.2481, "mean_token_accuracy": 0.6929168105125427, "num_tokens": 5343806.0, "step": 109 }, { "entropy": 1.2929682731628418, "epoch": 0.1694915254237288, "grad_norm": 0.17192070186138153, "learning_rate": 1.9998253499735577e-05, "loss": 1.2855, "mean_token_accuracy": 0.6829306781291962, "num_tokens": 5394426.0, "step": 110 }, { "entropy": 1.1173404455184937, "epoch": 0.17103235747303544, "grad_norm": 0.5861712694168091, "learning_rate": 1.9997921531847052e-05, "loss": 1.079, "mean_token_accuracy": 0.7174108326435089, "num_tokens": 5407305.0, "step": 111 }, { "entropy": 1.0678060054779053, "epoch": 0.17257318952234207, "grad_norm": 0.4236045181751251, "learning_rate": 1.999756070135243e-05, "loss": 1.0474, "mean_token_accuracy": 0.7443316876888275, "num_tokens": 5417042.0, "step": 112 }, { "entropy": 0.9383814334869385, "epoch": 0.17411402157164868, "grad_norm": 0.2648269236087799, "learning_rate": 1.999717100929336e-05, "loss": 0.9483, "mean_token_accuracy": 0.7777544260025024, "num_tokens": 5453557.0, "step": 113 }, { "entropy": 1.1191928386688232, "epoch": 0.17565485362095531, "grad_norm": 0.16897857189178467, "learning_rate": 1.9996752456794846e-05, "loss": 1.1005, "mean_token_accuracy": 0.7230362594127655, "num_tokens": 5517500.0, "step": 114 }, { "entropy": 1.160162329673767, "epoch": 0.17719568567026195, "grad_norm": 0.1650782823562622, "learning_rate": 1.9996305045065182e-05, "loss": 1.1709, "mean_token_accuracy": 0.7032356262207031, "num_tokens": 5578703.0, "step": 115 }, { "entropy": 1.1630686521530151, "epoch": 0.17873651771956856, "grad_norm": 0.17935232818126678, "learning_rate": 1.9995828775395985e-05, "loss": 1.1666, "mean_token_accuracy": 0.7042007744312286, "num_tokens": 5638141.0, "step": 116 }, { "entropy": 1.2075429558753967, "epoch": 0.1802773497688752, "grad_norm": 0.1684490442276001, "learning_rate": 1.999532364916218e-05, "loss": 1.2184, "mean_token_accuracy": 0.6971616744995117, "num_tokens": 5696285.0, "step": 117 }, { "entropy": 1.188731074333191, "epoch": 0.18181818181818182, "grad_norm": 0.17314885556697845, "learning_rate": 1.9994789667821996e-05, "loss": 1.2057, "mean_token_accuracy": 0.7001489996910095, "num_tokens": 5753430.0, "step": 118 }, { "entropy": 1.1645244359970093, "epoch": 0.18335901386748846, "grad_norm": 0.17938730120658875, "learning_rate": 1.9994226832916957e-05, "loss": 1.1743, "mean_token_accuracy": 0.7026138603687286, "num_tokens": 5809561.0, "step": 119 }, { "entropy": 1.1541482210159302, "epoch": 0.18489984591679506, "grad_norm": 0.1760181337594986, "learning_rate": 1.9993635146071903e-05, "loss": 1.166, "mean_token_accuracy": 0.7081045210361481, "num_tokens": 5864541.0, "step": 120 }, { "entropy": 1.2357128262519836, "epoch": 0.1864406779661017, "grad_norm": 0.16163229942321777, "learning_rate": 1.9993014608994935e-05, "loss": 1.2507, "mean_token_accuracy": 0.6905570328235626, "num_tokens": 5918211.0, "step": 121 }, { "entropy": 1.2535192370414734, "epoch": 0.18798151001540833, "grad_norm": 0.16839131712913513, "learning_rate": 1.999236522347747e-05, "loss": 1.2696, "mean_token_accuracy": 0.6859842836856842, "num_tokens": 5970070.0, "step": 122 }, { "entropy": 1.0935485363006592, "epoch": 0.18952234206471494, "grad_norm": 0.22030942142009735, "learning_rate": 1.999168699139418e-05, "loss": 1.1045, "mean_token_accuracy": 0.7160505056381226, "num_tokens": 6002386.0, "step": 123 }, { "entropy": 1.050820231437683, "epoch": 0.19106317411402157, "grad_norm": 0.3832346796989441, "learning_rate": 1.999097991470304e-05, "loss": 1.0827, "mean_token_accuracy": 0.7192369997501373, "num_tokens": 6013405.0, "step": 124 }, { "entropy": 0.9642382264137268, "epoch": 0.1926040061633282, "grad_norm": 0.46868908405303955, "learning_rate": 1.9990243995445278e-05, "loss": 0.9826, "mean_token_accuracy": 0.7579763233661652, "num_tokens": 6021543.0, "step": 125 }, { "entropy": 1.015868365764618, "epoch": 0.1941448382126348, "grad_norm": 0.22065670788288116, "learning_rate": 1.9989479235745385e-05, "loss": 0.9808, "mean_token_accuracy": 0.7591089010238647, "num_tokens": 6087019.0, "step": 126 }, { "entropy": 1.132751226425171, "epoch": 0.19568567026194145, "grad_norm": 0.18450111150741577, "learning_rate": 1.9988685637811126e-05, "loss": 1.1186, "mean_token_accuracy": 0.7130588889122009, "num_tokens": 6149986.0, "step": 127 }, { "entropy": 1.1610320806503296, "epoch": 0.19722650231124808, "grad_norm": 0.18639154732227325, "learning_rate": 1.9987863203933493e-05, "loss": 1.1517, "mean_token_accuracy": 0.7091838419437408, "num_tokens": 6210887.0, "step": 128 }, { "entropy": 1.1671220064163208, "epoch": 0.1987673343605547, "grad_norm": 0.17959916591644287, "learning_rate": 1.9987011936486752e-05, "loss": 1.1543, "mean_token_accuracy": 0.7124265432357788, "num_tokens": 6270248.0, "step": 129 }, { "entropy": 1.176917016506195, "epoch": 0.20030816640986132, "grad_norm": 0.17876574397087097, "learning_rate": 1.9986131837928383e-05, "loss": 1.1556, "mean_token_accuracy": 0.7080089747905731, "num_tokens": 6328355.0, "step": 130 }, { "entropy": 1.1821573972702026, "epoch": 0.20184899845916796, "grad_norm": 0.17650888860225677, "learning_rate": 1.9985222910799112e-05, "loss": 1.1645, "mean_token_accuracy": 0.7038737237453461, "num_tokens": 6385384.0, "step": 131 }, { "entropy": 1.1822656393051147, "epoch": 0.2033898305084746, "grad_norm": 0.1748703569173813, "learning_rate": 1.998428515772289e-05, "loss": 1.1729, "mean_token_accuracy": 0.704448401927948, "num_tokens": 6441278.0, "step": 132 }, { "entropy": 1.175943911075592, "epoch": 0.2049306625577812, "grad_norm": 0.183792844414711, "learning_rate": 1.9983318581406876e-05, "loss": 1.1659, "mean_token_accuracy": 0.7057514190673828, "num_tokens": 6495931.0, "step": 133 }, { "entropy": 1.1840004324913025, "epoch": 0.20647149460708783, "grad_norm": 0.17157047986984253, "learning_rate": 1.9982323184641436e-05, "loss": 1.1735, "mean_token_accuracy": 0.7048531770706177, "num_tokens": 6548988.0, "step": 134 }, { "entropy": 1.2075117230415344, "epoch": 0.20801232665639446, "grad_norm": 0.1793193519115448, "learning_rate": 1.9981298970300143e-05, "loss": 1.2008, "mean_token_accuracy": 0.6988723576068878, "num_tokens": 6599572.0, "step": 135 }, { "entropy": 1.1009886264801025, "epoch": 0.20955315870570107, "grad_norm": 0.3026435077190399, "learning_rate": 1.9980245941339768e-05, "loss": 1.1082, "mean_token_accuracy": 0.7155754566192627, "num_tokens": 6618661.0, "step": 136 }, { "entropy": 0.9974360466003418, "epoch": 0.2110939907550077, "grad_norm": 0.38361963629722595, "learning_rate": 1.9979164100800253e-05, "loss": 1.0176, "mean_token_accuracy": 0.750533252954483, "num_tokens": 6628355.0, "step": 137 }, { "entropy": 0.9594066739082336, "epoch": 0.21263482280431434, "grad_norm": 0.27171510457992554, "learning_rate": 1.9978053451804724e-05, "loss": 0.9693, "mean_token_accuracy": 0.7687053084373474, "num_tokens": 6664725.0, "step": 138 }, { "entropy": 1.0177362561225891, "epoch": 0.21417565485362094, "grad_norm": 0.22558711469173431, "learning_rate": 1.9976913997559468e-05, "loss": 1.023, "mean_token_accuracy": 0.7411064803600311, "num_tokens": 6729215.0, "step": 139 }, { "entropy": 1.1301628351211548, "epoch": 0.21571648690292758, "grad_norm": 0.2184857875108719, "learning_rate": 1.9975745741353927e-05, "loss": 1.136, "mean_token_accuracy": 0.711577296257019, "num_tokens": 6790587.0, "step": 140 }, { "entropy": 1.098094642162323, "epoch": 0.2172573189522342, "grad_norm": 0.18565914034843445, "learning_rate": 1.9974548686560703e-05, "loss": 1.1092, "mean_token_accuracy": 0.71519935131073, "num_tokens": 6850257.0, "step": 141 }, { "entropy": 1.1329447627067566, "epoch": 0.21879815100154082, "grad_norm": 0.18037377297878265, "learning_rate": 1.9973322836635517e-05, "loss": 1.1351, "mean_token_accuracy": 0.7124396860599518, "num_tokens": 6908677.0, "step": 142 }, { "entropy": 1.158177673816681, "epoch": 0.22033898305084745, "grad_norm": 0.19469338655471802, "learning_rate": 1.9972068195117235e-05, "loss": 1.1567, "mean_token_accuracy": 0.7074559926986694, "num_tokens": 6965978.0, "step": 143 }, { "entropy": 1.1673192381858826, "epoch": 0.2218798151001541, "grad_norm": 0.1876979023218155, "learning_rate": 1.9970784765627824e-05, "loss": 1.1646, "mean_token_accuracy": 0.7072589695453644, "num_tokens": 7022154.0, "step": 144 }, { "entropy": 1.1412739157676697, "epoch": 0.22342064714946072, "grad_norm": 0.1967936009168625, "learning_rate": 1.9969472551872373e-05, "loss": 1.129, "mean_token_accuracy": 0.7160259783267975, "num_tokens": 7077119.0, "step": 145 }, { "entropy": 1.1912404298782349, "epoch": 0.22496147919876733, "grad_norm": 0.18448729813098907, "learning_rate": 1.996813155763906e-05, "loss": 1.1924, "mean_token_accuracy": 0.7000617384910583, "num_tokens": 7130691.0, "step": 146 }, { "entropy": 1.2259633541107178, "epoch": 0.22650231124807396, "grad_norm": 0.18025259673595428, "learning_rate": 1.996676178679914e-05, "loss": 1.2474, "mean_token_accuracy": 0.6905113458633423, "num_tokens": 7182603.0, "step": 147 }, { "entropy": 1.0584537982940674, "epoch": 0.2280431432973806, "grad_norm": 0.27708500623703003, "learning_rate": 1.996536324330696e-05, "loss": 1.1396, "mean_token_accuracy": 0.7160138189792633, "num_tokens": 7209362.0, "step": 148 }, { "entropy": 1.054779291152954, "epoch": 0.2295839753466872, "grad_norm": 0.4102008044719696, "learning_rate": 1.9963935931199917e-05, "loss": 1.0795, "mean_token_accuracy": 0.7257434725761414, "num_tokens": 7219546.0, "step": 149 }, { "entropy": 0.9692428112030029, "epoch": 0.23112480739599384, "grad_norm": 0.550530195236206, "learning_rate": 1.9962479854598463e-05, "loss": 0.9631, "mean_token_accuracy": 0.7574954628944397, "num_tokens": 7227157.0, "step": 150 }, { "entropy": 1.0272763967514038, "epoch": 0.23266563944530047, "grad_norm": 0.24715633690357208, "learning_rate": 1.9960995017706082e-05, "loss": 1.0136, "mean_token_accuracy": 0.7478936314582825, "num_tokens": 7292560.0, "step": 151 }, { "entropy": 1.1122183799743652, "epoch": 0.23420647149460708, "grad_norm": 0.2311318814754486, "learning_rate": 1.9959481424809302e-05, "loss": 1.1101, "mean_token_accuracy": 0.718651682138443, "num_tokens": 7355587.0, "step": 152 }, { "entropy": 1.1275506019592285, "epoch": 0.2357473035439137, "grad_norm": 0.22032763063907623, "learning_rate": 1.9957939080277647e-05, "loss": 1.1256, "mean_token_accuracy": 0.7155501246452332, "num_tokens": 7416571.0, "step": 153 }, { "entropy": 1.1241522431373596, "epoch": 0.23728813559322035, "grad_norm": 0.3403574824333191, "learning_rate": 1.995636798856365e-05, "loss": 1.1217, "mean_token_accuracy": 0.7166844606399536, "num_tokens": 7476001.0, "step": 154 }, { "entropy": 1.1330779194831848, "epoch": 0.23882896764252695, "grad_norm": 0.19036191701889038, "learning_rate": 1.995476815420284e-05, "loss": 1.1294, "mean_token_accuracy": 0.7147082686424255, "num_tokens": 7533973.0, "step": 155 }, { "entropy": 1.1975545287132263, "epoch": 0.24036979969183359, "grad_norm": 0.18437211215496063, "learning_rate": 1.9953139581813708e-05, "loss": 1.193, "mean_token_accuracy": 0.7005789577960968, "num_tokens": 7590754.0, "step": 156 }, { "entropy": 1.1681327819824219, "epoch": 0.24191063174114022, "grad_norm": 0.18428149819374084, "learning_rate": 1.9951482276097726e-05, "loss": 1.1429, "mean_token_accuracy": 0.7097501456737518, "num_tokens": 7646518.0, "step": 157 }, { "entropy": 1.1553452014923096, "epoch": 0.24345146379044685, "grad_norm": 0.2273171842098236, "learning_rate": 1.9949796241839293e-05, "loss": 1.1369, "mean_token_accuracy": 0.7106545865535736, "num_tokens": 7701180.0, "step": 158 }, { "entropy": 1.214854657649994, "epoch": 0.24499229583975346, "grad_norm": 0.1945824772119522, "learning_rate": 1.9948081483905763e-05, "loss": 1.193, "mean_token_accuracy": 0.699378490447998, "num_tokens": 7754330.0, "step": 159 }, { "entropy": 1.2159176468849182, "epoch": 0.2465331278890601, "grad_norm": 0.20264406502246857, "learning_rate": 1.99463380072474e-05, "loss": 1.2107, "mean_token_accuracy": 0.6962764263153076, "num_tokens": 7803703.0, "step": 160 }, { "entropy": 0.9955940544605255, "epoch": 0.24807395993836673, "grad_norm": 0.5065114498138428, "learning_rate": 1.9944565816897378e-05, "loss": 1.0485, "mean_token_accuracy": 0.7292615175247192, "num_tokens": 7816185.0, "step": 161 }, { "entropy": 0.9848976731300354, "epoch": 0.24961479198767333, "grad_norm": 0.4581451117992401, "learning_rate": 1.9942764917971767e-05, "loss": 1.035, "mean_token_accuracy": 0.7453111410140991, "num_tokens": 7825730.0, "step": 162 }, { "entropy": 0.921267569065094, "epoch": 0.25115562403698, "grad_norm": 0.2724779546260834, "learning_rate": 1.9940935315669504e-05, "loss": 0.9492, "mean_token_accuracy": 0.7745375335216522, "num_tokens": 7862142.0, "step": 163 }, { "entropy": 1.0672119855880737, "epoch": 0.2526964560862866, "grad_norm": 0.19953328371047974, "learning_rate": 1.9939077015272404e-05, "loss": 1.0531, "mean_token_accuracy": 0.7322631180286407, "num_tokens": 7926249.0, "step": 164 }, { "entropy": 1.1129704713821411, "epoch": 0.2542372881355932, "grad_norm": 0.2005014568567276, "learning_rate": 1.993719002214512e-05, "loss": 1.1125, "mean_token_accuracy": 0.7144196927547455, "num_tokens": 7987628.0, "step": 165 }, { "entropy": 1.1186891794204712, "epoch": 0.25577812018489987, "grad_norm": 0.20594757795333862, "learning_rate": 1.9935274341735132e-05, "loss": 1.1179, "mean_token_accuracy": 0.7143732309341431, "num_tokens": 8047381.0, "step": 166 }, { "entropy": 1.1395350694656372, "epoch": 0.2573189522342065, "grad_norm": 0.2135891616344452, "learning_rate": 1.9933329979572743e-05, "loss": 1.1322, "mean_token_accuracy": 0.7125391364097595, "num_tokens": 8105841.0, "step": 167 }, { "entropy": 1.11935293674469, "epoch": 0.2588597842835131, "grad_norm": 0.19619043171405792, "learning_rate": 1.993135694127106e-05, "loss": 1.0976, "mean_token_accuracy": 0.7168008983135223, "num_tokens": 8163187.0, "step": 168 }, { "entropy": 1.1408463716506958, "epoch": 0.26040061633281975, "grad_norm": 0.20186994969844818, "learning_rate": 1.9929355232525976e-05, "loss": 1.1299, "mean_token_accuracy": 0.7123763561248779, "num_tokens": 8219549.0, "step": 169 }, { "entropy": 1.1392396092414856, "epoch": 0.26194144838212635, "grad_norm": 0.20524179935455322, "learning_rate": 1.992732485911613e-05, "loss": 1.1264, "mean_token_accuracy": 0.7149041295051575, "num_tokens": 8274908.0, "step": 170 }, { "entropy": 1.2090429067611694, "epoch": 0.26348228043143296, "grad_norm": 0.21786709129810333, "learning_rate": 1.9925265826902936e-05, "loss": 1.2242, "mean_token_accuracy": 0.6943202316761017, "num_tokens": 8329030.0, "step": 171 }, { "entropy": 1.2005067467689514, "epoch": 0.2650231124807396, "grad_norm": 0.19501744210720062, "learning_rate": 1.9923178141830536e-05, "loss": 1.2015, "mean_token_accuracy": 0.6987924575805664, "num_tokens": 8381157.0, "step": 172 }, { "entropy": 1.1063753962516785, "epoch": 0.2665639445300462, "grad_norm": 0.24943861365318298, "learning_rate": 1.9921061809925777e-05, "loss": 1.1267, "mean_token_accuracy": 0.7235679626464844, "num_tokens": 8413467.0, "step": 173 }, { "entropy": 1.0343735218048096, "epoch": 0.26810477657935283, "grad_norm": 0.48326385021209717, "learning_rate": 1.991891683729822e-05, "loss": 1.012, "mean_token_accuracy": 0.7309278249740601, "num_tokens": 8424245.0, "step": 174 }, { "entropy": 0.9369077980518341, "epoch": 0.2696456086286595, "grad_norm": 0.5269757509231567, "learning_rate": 1.99167432301401e-05, "loss": 0.9464, "mean_token_accuracy": 0.7636696994304657, "num_tokens": 8432197.0, "step": 175 }, { "entropy": 0.9736452400684357, "epoch": 0.2711864406779661, "grad_norm": 0.2456178516149521, "learning_rate": 1.991454099472632e-05, "loss": 0.9606, "mean_token_accuracy": 0.7564698159694672, "num_tokens": 8497363.0, "step": 176 }, { "entropy": 1.0599616765975952, "epoch": 0.2727272727272727, "grad_norm": 0.231217160820961, "learning_rate": 1.9912310137414426e-05, "loss": 1.0666, "mean_token_accuracy": 0.7249644696712494, "num_tokens": 8559582.0, "step": 177 }, { "entropy": 1.082266926765442, "epoch": 0.27426810477657937, "grad_norm": 0.21956586837768555, "learning_rate": 1.991005066464459e-05, "loss": 1.0951, "mean_token_accuracy": 0.7203294932842255, "num_tokens": 8619821.0, "step": 178 }, { "entropy": 1.0771585702896118, "epoch": 0.275808936825886, "grad_norm": 0.2282981276512146, "learning_rate": 1.9907762582939596e-05, "loss": 1.0909, "mean_token_accuracy": 0.7187336981296539, "num_tokens": 8678732.0, "step": 179 }, { "entropy": 1.1041181087493896, "epoch": 0.2773497688751926, "grad_norm": 0.22502928972244263, "learning_rate": 1.9905445898904822e-05, "loss": 1.1195, "mean_token_accuracy": 0.7161448001861572, "num_tokens": 8736503.0, "step": 180 }, { "entropy": 1.1314849257469177, "epoch": 0.27889060092449924, "grad_norm": 0.21399584412574768, "learning_rate": 1.9903100619228202e-05, "loss": 1.1355, "mean_token_accuracy": 0.7128588259220123, "num_tokens": 8793095.0, "step": 181 }, { "entropy": 1.1143299341201782, "epoch": 0.28043143297380585, "grad_norm": 0.2203298956155777, "learning_rate": 1.990072675068024e-05, "loss": 1.1113, "mean_token_accuracy": 0.7164505124092102, "num_tokens": 8848250.0, "step": 182 }, { "entropy": 1.170916497707367, "epoch": 0.28197226502311246, "grad_norm": 0.21634458005428314, "learning_rate": 1.9898324300113963e-05, "loss": 1.1761, "mean_token_accuracy": 0.7033270001411438, "num_tokens": 8902092.0, "step": 183 }, { "entropy": 1.1767056584358215, "epoch": 0.2835130970724191, "grad_norm": 0.22482049465179443, "learning_rate": 1.9895893274464907e-05, "loss": 1.1676, "mean_token_accuracy": 0.7075644433498383, "num_tokens": 8954569.0, "step": 184 }, { "entropy": 1.1388468146324158, "epoch": 0.2850539291217257, "grad_norm": 0.23414495587348938, "learning_rate": 1.9893433680751105e-05, "loss": 1.1763, "mean_token_accuracy": 0.7097490131855011, "num_tokens": 8995903.0, "step": 185 }, { "entropy": 1.0113065838813782, "epoch": 0.28659476117103233, "grad_norm": 0.46144774556159973, "learning_rate": 1.989094552607306e-05, "loss": 1.0453, "mean_token_accuracy": 0.7320194244384766, "num_tokens": 9007488.0, "step": 186 }, { "entropy": 0.9664472341537476, "epoch": 0.288135593220339, "grad_norm": 0.44434186816215515, "learning_rate": 1.9888428817613735e-05, "loss": 0.978, "mean_token_accuracy": 0.7519363760948181, "num_tokens": 9016505.0, "step": 187 }, { "entropy": 0.8530955016613007, "epoch": 0.2896764252696456, "grad_norm": 0.341478556394577, "learning_rate": 1.988588356263851e-05, "loss": 0.8642, "mean_token_accuracy": 0.7919975817203522, "num_tokens": 9052653.0, "step": 188 }, { "entropy": 1.0581720471382141, "epoch": 0.29121725731895226, "grad_norm": 0.2557825744152069, "learning_rate": 1.9883309768495183e-05, "loss": 1.0399, "mean_token_accuracy": 0.735775351524353, "num_tokens": 9117044.0, "step": 189 }, { "entropy": 1.0796369910240173, "epoch": 0.29275808936825887, "grad_norm": 0.2361701875925064, "learning_rate": 1.988070744261394e-05, "loss": 1.05, "mean_token_accuracy": 0.7296320796012878, "num_tokens": 9178662.0, "step": 190 }, { "entropy": 1.1076009273529053, "epoch": 0.2942989214175655, "grad_norm": 0.2139163762331009, "learning_rate": 1.987807659250733e-05, "loss": 1.0932, "mean_token_accuracy": 0.7201550602912903, "num_tokens": 9238404.0, "step": 191 }, { "entropy": 1.1215635538101196, "epoch": 0.29583975346687214, "grad_norm": 0.2232576608657837, "learning_rate": 1.987541722577025e-05, "loss": 1.1023, "mean_token_accuracy": 0.7168191075325012, "num_tokens": 9296877.0, "step": 192 }, { "entropy": 1.1226049065589905, "epoch": 0.29738058551617874, "grad_norm": 0.21469469368457794, "learning_rate": 1.987272935007993e-05, "loss": 1.0976, "mean_token_accuracy": 0.7163324356079102, "num_tokens": 9354088.0, "step": 193 }, { "entropy": 1.1206071376800537, "epoch": 0.29892141756548535, "grad_norm": 0.22862546145915985, "learning_rate": 1.9870012973195883e-05, "loss": 1.1249, "mean_token_accuracy": 0.7151551246643066, "num_tokens": 9410120.0, "step": 194 }, { "entropy": 1.1243339776992798, "epoch": 0.300462249614792, "grad_norm": 0.22732380032539368, "learning_rate": 1.986726810295992e-05, "loss": 1.1362, "mean_token_accuracy": 0.7093969881534576, "num_tokens": 9465129.0, "step": 195 }, { "entropy": 1.1757338643074036, "epoch": 0.3020030816640986, "grad_norm": 0.23412038385868073, "learning_rate": 1.9864494747296084e-05, "loss": 1.1891, "mean_token_accuracy": 0.6986605226993561, "num_tokens": 9518840.0, "step": 196 }, { "entropy": 1.2109919786453247, "epoch": 0.3035439137134052, "grad_norm": 0.23390990495681763, "learning_rate": 1.986169291421068e-05, "loss": 1.2179, "mean_token_accuracy": 0.6933363676071167, "num_tokens": 9570265.0, "step": 197 }, { "entropy": 1.0063912868499756, "epoch": 0.3050847457627119, "grad_norm": 0.3012617230415344, "learning_rate": 1.9858862611792206e-05, "loss": 1.0516, "mean_token_accuracy": 0.7311294376850128, "num_tokens": 9597264.0, "step": 198 }, { "entropy": 1.0119141638278961, "epoch": 0.3066255778120185, "grad_norm": 0.4625769853591919, "learning_rate": 1.9856003848211345e-05, "loss": 1.0178, "mean_token_accuracy": 0.7336307764053345, "num_tokens": 9607969.0, "step": 199 }, { "entropy": 0.9427486658096313, "epoch": 0.3081664098613251, "grad_norm": 0.5139606595039368, "learning_rate": 1.985311663172096e-05, "loss": 0.9464, "mean_token_accuracy": 0.7629606425762177, "num_tokens": 9616053.0, "step": 200 }, { "epoch": 0.3081664098613251, "eval_entropy": 1.0635582298602697, "eval_loss": 1.067421555519104, "eval_mean_token_accuracy": 0.7274574404055217, "eval_num_tokens": 9616053.0, "eval_runtime": 194.8656, "eval_samples_per_second": 4.352, "eval_steps_per_second": 1.088, "step": 200 }, { "entropy": 0.9491062164306641, "epoch": 0.30970724191063176, "grad_norm": 0.3510213792324066, "learning_rate": 1.9850200970656032e-05, "loss": 0.9369, "mean_token_accuracy": 0.7646206319332123, "num_tokens": 9681001.0, "step": 201 }, { "entropy": 1.0170182287693024, "epoch": 0.31124807395993837, "grad_norm": 0.25812211632728577, "learning_rate": 1.9847256873433672e-05, "loss": 1.0133, "mean_token_accuracy": 0.7366096675395966, "num_tokens": 9743070.0, "step": 202 }, { "entropy": 1.0601570010185242, "epoch": 0.31278890600924497, "grad_norm": 0.25494879484176636, "learning_rate": 1.9844284348553077e-05, "loss": 1.0724, "mean_token_accuracy": 0.7239386439323425, "num_tokens": 9803498.0, "step": 203 }, { "entropy": 1.0632287859916687, "epoch": 0.31432973805855163, "grad_norm": 0.22975307703018188, "learning_rate": 1.984128340459552e-05, "loss": 1.063, "mean_token_accuracy": 0.7281982004642487, "num_tokens": 9862516.0, "step": 204 }, { "entropy": 1.088874876499176, "epoch": 0.31587057010785824, "grad_norm": 0.21664854884147644, "learning_rate": 1.9838254050224294e-05, "loss": 1.082, "mean_token_accuracy": 0.722571462392807, "num_tokens": 9920264.0, "step": 205 }, { "entropy": 1.0779345631599426, "epoch": 0.31741140215716485, "grad_norm": 0.2171836495399475, "learning_rate": 1.9835196294184736e-05, "loss": 1.0685, "mean_token_accuracy": 0.7242911159992218, "num_tokens": 9976837.0, "step": 206 }, { "entropy": 1.0976250171661377, "epoch": 0.3189522342064715, "grad_norm": 0.22422601282596588, "learning_rate": 1.9832110145304152e-05, "loss": 1.0963, "mean_token_accuracy": 0.7186403572559357, "num_tokens": 10032471.0, "step": 207 }, { "entropy": 1.1256202459335327, "epoch": 0.3204930662557781, "grad_norm": 0.23694950342178345, "learning_rate": 1.982899561249183e-05, "loss": 1.1184, "mean_token_accuracy": 0.7144766449928284, "num_tokens": 10087031.0, "step": 208 }, { "entropy": 1.1143254041671753, "epoch": 0.3220338983050847, "grad_norm": 0.23961246013641357, "learning_rate": 1.9825852704738993e-05, "loss": 1.1128, "mean_token_accuracy": 0.716560035943985, "num_tokens": 10139889.0, "step": 209 }, { "entropy": 1.085246056318283, "epoch": 0.3235747303543914, "grad_norm": 0.2500257194042206, "learning_rate": 1.982268143111877e-05, "loss": 1.1202, "mean_token_accuracy": 0.7176263928413391, "num_tokens": 10182211.0, "step": 210 }, { "entropy": 0.9768007695674896, "epoch": 0.325115562403698, "grad_norm": 0.5144143104553223, "learning_rate": 1.9819481800786192e-05, "loss": 0.9822, "mean_token_accuracy": 0.7403228580951691, "num_tokens": 10194079.0, "step": 211 }, { "entropy": 0.9559390544891357, "epoch": 0.3266563944530046, "grad_norm": 0.5147581100463867, "learning_rate": 1.9816253822978147e-05, "loss": 0.9518, "mean_token_accuracy": 0.7585658431053162, "num_tokens": 10203531.0, "step": 212 }, { "entropy": 0.8551934659481049, "epoch": 0.32819722650231126, "grad_norm": 0.34743639826774597, "learning_rate": 1.981299750701335e-05, "loss": 0.8728, "mean_token_accuracy": 0.7880162298679352, "num_tokens": 10239851.0, "step": 213 }, { "entropy": 0.9948636293411255, "epoch": 0.32973805855161786, "grad_norm": 0.26092249155044556, "learning_rate": 1.980971286229234e-05, "loss": 0.9913, "mean_token_accuracy": 0.7437352240085602, "num_tokens": 10303667.0, "step": 214 }, { "entropy": 1.0513219237327576, "epoch": 0.3312788906009245, "grad_norm": 0.23448386788368225, "learning_rate": 1.9806399898297422e-05, "loss": 1.038, "mean_token_accuracy": 0.7291106581687927, "num_tokens": 10364991.0, "step": 215 }, { "entropy": 1.0699206590652466, "epoch": 0.33281972265023113, "grad_norm": 0.2362591177225113, "learning_rate": 1.9803058624592665e-05, "loss": 1.0617, "mean_token_accuracy": 0.7278167009353638, "num_tokens": 10424762.0, "step": 216 }, { "entropy": 1.0874415636062622, "epoch": 0.33436055469953774, "grad_norm": 0.2652667462825775, "learning_rate": 1.979968905082386e-05, "loss": 1.0759, "mean_token_accuracy": 0.723151445388794, "num_tokens": 10483332.0, "step": 217 }, { "entropy": 1.104274034500122, "epoch": 0.3359013867488444, "grad_norm": 0.24471069872379303, "learning_rate": 1.9796291186718494e-05, "loss": 1.092, "mean_token_accuracy": 0.7172232568264008, "num_tokens": 10540566.0, "step": 218 }, { "entropy": 1.0903122425079346, "epoch": 0.337442218798151, "grad_norm": 0.2417384684085846, "learning_rate": 1.9792865042085732e-05, "loss": 1.0716, "mean_token_accuracy": 0.7256182134151459, "num_tokens": 10596727.0, "step": 219 }, { "entropy": 1.1015235781669617, "epoch": 0.3389830508474576, "grad_norm": 0.24188995361328125, "learning_rate": 1.9789410626816373e-05, "loss": 1.092, "mean_token_accuracy": 0.7184254229068756, "num_tokens": 10651668.0, "step": 220 }, { "entropy": 1.081334114074707, "epoch": 0.3405238828967643, "grad_norm": 0.23943038284778595, "learning_rate": 1.978592795088283e-05, "loss": 1.0726, "mean_token_accuracy": 0.7234618067741394, "num_tokens": 10705319.0, "step": 221 }, { "entropy": 1.0891441106796265, "epoch": 0.3420647149460709, "grad_norm": 0.24722956120967865, "learning_rate": 1.9782417024339108e-05, "loss": 1.0901, "mean_token_accuracy": 0.7194399833679199, "num_tokens": 10756901.0, "step": 222 }, { "entropy": 1.0877742767333984, "epoch": 0.3436055469953775, "grad_norm": 0.31949999928474426, "learning_rate": 1.977887785732076e-05, "loss": 1.1102, "mean_token_accuracy": 0.7140873968601227, "num_tokens": 10785631.0, "step": 223 }, { "entropy": 0.9610119462013245, "epoch": 0.34514637904468415, "grad_norm": 0.4947839677333832, "learning_rate": 1.977531046004487e-05, "loss": 0.9514, "mean_token_accuracy": 0.7553706765174866, "num_tokens": 10796491.0, "step": 224 }, { "entropy": 0.910370945930481, "epoch": 0.34668721109399075, "grad_norm": 0.5601955652236938, "learning_rate": 1.9771714842810014e-05, "loss": 0.9072, "mean_token_accuracy": 0.770256370306015, "num_tokens": 10804616.0, "step": 225 }, { "entropy": 0.8703013956546783, "epoch": 0.34822804314329736, "grad_norm": 0.29865214228630066, "learning_rate": 1.976809101599624e-05, "loss": 0.8716, "mean_token_accuracy": 0.7767088413238525, "num_tokens": 10869996.0, "step": 226 }, { "entropy": 0.982592910528183, "epoch": 0.349768875192604, "grad_norm": 0.30609431862831116, "learning_rate": 1.976443899006503e-05, "loss": 1.0072, "mean_token_accuracy": 0.7355456054210663, "num_tokens": 10932960.0, "step": 227 }, { "entropy": 1.0524355173110962, "epoch": 0.35130970724191063, "grad_norm": 0.27863365411758423, "learning_rate": 1.9760758775559275e-05, "loss": 1.0781, "mean_token_accuracy": 0.7221183180809021, "num_tokens": 10993924.0, "step": 228 }, { "entropy": 1.0301344394683838, "epoch": 0.35285053929121724, "grad_norm": 0.2593282163143158, "learning_rate": 1.9757050383103235e-05, "loss": 1.047, "mean_token_accuracy": 0.72962486743927, "num_tokens": 11053583.0, "step": 229 }, { "entropy": 1.065339982509613, "epoch": 0.3543913713405239, "grad_norm": 0.26160895824432373, "learning_rate": 1.9753313823402536e-05, "loss": 1.0712, "mean_token_accuracy": 0.7224406003952026, "num_tokens": 11112023.0, "step": 230 }, { "entropy": 1.1065597534179688, "epoch": 0.3559322033898305, "grad_norm": 0.26307591795921326, "learning_rate": 1.974954910724409e-05, "loss": 1.0977, "mean_token_accuracy": 0.7173341512680054, "num_tokens": 11169221.0, "step": 231 }, { "entropy": 1.1034944653511047, "epoch": 0.3574730354391371, "grad_norm": 0.26805076003074646, "learning_rate": 1.9745756245496118e-05, "loss": 1.0901, "mean_token_accuracy": 0.7187582552433014, "num_tokens": 11225222.0, "step": 232 }, { "entropy": 1.1268479228019714, "epoch": 0.35901386748844377, "grad_norm": 0.2681247889995575, "learning_rate": 1.974193524910808e-05, "loss": 1.1049, "mean_token_accuracy": 0.7182426452636719, "num_tokens": 11279956.0, "step": 233 }, { "entropy": 1.0887632369995117, "epoch": 0.3605546995377504, "grad_norm": 0.29170849919319153, "learning_rate": 1.9738086129110655e-05, "loss": 1.0786, "mean_token_accuracy": 0.7238105833530426, "num_tokens": 11333064.0, "step": 234 }, { "entropy": 1.1011694073677063, "epoch": 0.362095531587057, "grad_norm": 0.2837817072868347, "learning_rate": 1.973420889661573e-05, "loss": 1.0956, "mean_token_accuracy": 0.7217880189418793, "num_tokens": 11381996.0, "step": 235 }, { "entropy": 0.9933000206947327, "epoch": 0.36363636363636365, "grad_norm": 0.5370975732803345, "learning_rate": 1.973030356281632e-05, "loss": 1.0112, "mean_token_accuracy": 0.7330097854137421, "num_tokens": 11394016.0, "step": 236 }, { "entropy": 0.9695306122303009, "epoch": 0.36517719568567025, "grad_norm": 0.5804514288902283, "learning_rate": 1.9726370138986583e-05, "loss": 0.974, "mean_token_accuracy": 0.7604009211063385, "num_tokens": 11403142.0, "step": 237 }, { "entropy": 0.8962224125862122, "epoch": 0.3667180277349769, "grad_norm": 0.38159844279289246, "learning_rate": 1.972240863648177e-05, "loss": 0.8974, "mean_token_accuracy": 0.7816522419452667, "num_tokens": 11439413.0, "step": 238 }, { "entropy": 0.8871813714504242, "epoch": 0.3682588597842835, "grad_norm": 0.27964475750923157, "learning_rate": 1.9718419066738186e-05, "loss": 0.8722, "mean_token_accuracy": 0.7721313834190369, "num_tokens": 11503810.0, "step": 239 }, { "entropy": 1.013273149728775, "epoch": 0.3697996918335901, "grad_norm": 0.2918948531150818, "learning_rate": 1.971440144127316e-05, "loss": 1.0156, "mean_token_accuracy": 0.7358708679676056, "num_tokens": 11565064.0, "step": 240 }, { "entropy": 1.0177216529846191, "epoch": 0.3713405238828968, "grad_norm": 0.2734447121620178, "learning_rate": 1.9710355771685017e-05, "loss": 1.0097, "mean_token_accuracy": 0.7384256720542908, "num_tokens": 11624621.0, "step": 241 }, { "entropy": 1.0399941504001617, "epoch": 0.3728813559322034, "grad_norm": 0.28031691908836365, "learning_rate": 1.9706282069653043e-05, "loss": 1.0194, "mean_token_accuracy": 0.7333246171474457, "num_tokens": 11683132.0, "step": 242 }, { "entropy": 1.0615828037261963, "epoch": 0.37442218798151, "grad_norm": 0.2719002962112427, "learning_rate": 1.970218034693745e-05, "loss": 1.0509, "mean_token_accuracy": 0.7279288172721863, "num_tokens": 11740628.0, "step": 243 }, { "entropy": 1.080402910709381, "epoch": 0.37596302003081666, "grad_norm": 0.27639034390449524, "learning_rate": 1.9698050615379333e-05, "loss": 1.073, "mean_token_accuracy": 0.7242756187915802, "num_tokens": 11797105.0, "step": 244 }, { "entropy": 1.0980858206748962, "epoch": 0.37750385208012327, "grad_norm": 0.27693477272987366, "learning_rate": 1.9693892886900656e-05, "loss": 1.0986, "mean_token_accuracy": 0.7176824510097504, "num_tokens": 11852704.0, "step": 245 }, { "entropy": 1.0788218975067139, "epoch": 0.3790446841294299, "grad_norm": 0.2742116153240204, "learning_rate": 1.96897071735042e-05, "loss": 1.0664, "mean_token_accuracy": 0.7258268594741821, "num_tokens": 11907152.0, "step": 246 }, { "entropy": 1.0470447540283203, "epoch": 0.38058551617873654, "grad_norm": 0.2885188162326813, "learning_rate": 1.9685493487273542e-05, "loss": 1.0355, "mean_token_accuracy": 0.7303601801395416, "num_tokens": 11959949.0, "step": 247 }, { "entropy": 1.0550857186317444, "epoch": 0.38212634822804314, "grad_norm": 0.32689306139945984, "learning_rate": 1.9681251840372998e-05, "loss": 1.0542, "mean_token_accuracy": 0.7259317636489868, "num_tokens": 12001181.0, "step": 248 }, { "entropy": 1.0112228989601135, "epoch": 0.38366718027734975, "grad_norm": 0.5260847210884094, "learning_rate": 1.967698224504762e-05, "loss": 0.9824, "mean_token_accuracy": 0.741569459438324, "num_tokens": 12012595.0, "step": 249 }, { "entropy": 0.9345259070396423, "epoch": 0.3852080123266564, "grad_norm": 0.5733681917190552, "learning_rate": 1.9672684713623125e-05, "loss": 0.9189, "mean_token_accuracy": 0.7717365026473999, "num_tokens": 12020669.0, "step": 250 }, { "entropy": 0.8381962180137634, "epoch": 0.386748844375963, "grad_norm": 0.3640625774860382, "learning_rate": 1.9668359258505898e-05, "loss": 0.8431, "mean_token_accuracy": 0.7844701111316681, "num_tokens": 12086114.0, "step": 251 }, { "entropy": 0.9385179281234741, "epoch": 0.3882896764252696, "grad_norm": 0.3491758704185486, "learning_rate": 1.9664005892182917e-05, "loss": 0.9738, "mean_token_accuracy": 0.7425862550735474, "num_tokens": 12149107.0, "step": 252 }, { "entropy": 0.9768623113632202, "epoch": 0.3898305084745763, "grad_norm": 0.3209327161312103, "learning_rate": 1.965962462722175e-05, "loss": 1.0033, "mean_token_accuracy": 0.736761599779129, "num_tokens": 12209988.0, "step": 253 }, { "entropy": 1.0103687047958374, "epoch": 0.3913713405238829, "grad_norm": 0.29868221282958984, "learning_rate": 1.9655215476270494e-05, "loss": 1.0116, "mean_token_accuracy": 0.7367973625659943, "num_tokens": 12269218.0, "step": 254 }, { "entropy": 1.0466511249542236, "epoch": 0.3929121725731895, "grad_norm": 0.2733313739299774, "learning_rate": 1.9650778452057752e-05, "loss": 1.0447, "mean_token_accuracy": 0.7304053902626038, "num_tokens": 12327285.0, "step": 255 }, { "entropy": 1.0733622312545776, "epoch": 0.39445300462249616, "grad_norm": 0.29512375593185425, "learning_rate": 1.96463135673926e-05, "loss": 1.0654, "mean_token_accuracy": 0.7233293652534485, "num_tokens": 12384237.0, "step": 256 }, { "entropy": 1.0944142937660217, "epoch": 0.39599383667180277, "grad_norm": 0.29497018456459045, "learning_rate": 1.9641820835164532e-05, "loss": 1.081, "mean_token_accuracy": 0.7230397164821625, "num_tokens": 12440054.0, "step": 257 }, { "entropy": 1.0875163078308105, "epoch": 0.3975346687211094, "grad_norm": 0.310628205537796, "learning_rate": 1.9637300268343442e-05, "loss": 1.0724, "mean_token_accuracy": 0.7245276570320129, "num_tokens": 12494692.0, "step": 258 }, { "entropy": 1.080467164516449, "epoch": 0.39907550077041604, "grad_norm": 0.31932008266448975, "learning_rate": 1.963275187997958e-05, "loss": 1.0719, "mean_token_accuracy": 0.7233053743839264, "num_tokens": 12547748.0, "step": 259 }, { "entropy": 1.0637614727020264, "epoch": 0.40061633281972264, "grad_norm": 0.32736918330192566, "learning_rate": 1.9628175683203505e-05, "loss": 1.044, "mean_token_accuracy": 0.731007993221283, "num_tokens": 12598492.0, "step": 260 }, { "entropy": 0.9434312880039215, "epoch": 0.40215716486902925, "grad_norm": 0.5189025402069092, "learning_rate": 1.962357169122606e-05, "loss": 0.9993, "mean_token_accuracy": 0.7429157197475433, "num_tokens": 12614518.0, "step": 261 }, { "entropy": 0.9680747389793396, "epoch": 0.4036979969183359, "grad_norm": 0.5959228277206421, "learning_rate": 1.9618939917338328e-05, "loss": 1.0098, "mean_token_accuracy": 0.7487044930458069, "num_tokens": 12624048.0, "step": 262 }, { "entropy": 0.8484733402729034, "epoch": 0.4052388289676425, "grad_norm": 0.41596829891204834, "learning_rate": 1.9614280374911597e-05, "loss": 0.8408, "mean_token_accuracy": 0.7905826270580292, "num_tokens": 12660341.0, "step": 263 }, { "entropy": 0.9251594841480255, "epoch": 0.4067796610169492, "grad_norm": 0.3314366340637207, "learning_rate": 1.9609593077397317e-05, "loss": 0.9092, "mean_token_accuracy": 0.7638004422187805, "num_tokens": 12724504.0, "step": 264 }, { "entropy": 0.971444845199585, "epoch": 0.4083204930662558, "grad_norm": 0.3264803886413574, "learning_rate": 1.960487803832706e-05, "loss": 0.955, "mean_token_accuracy": 0.7484413683414459, "num_tokens": 12785717.0, "step": 265 }, { "entropy": 1.0450551509857178, "epoch": 0.4098613251155624, "grad_norm": 0.3105798065662384, "learning_rate": 1.9600135271312484e-05, "loss": 1.0335, "mean_token_accuracy": 0.7315234839916229, "num_tokens": 12845189.0, "step": 266 }, { "entropy": 1.039156973361969, "epoch": 0.41140215716486905, "grad_norm": 0.3253927528858185, "learning_rate": 1.9595364790045303e-05, "loss": 1.0175, "mean_token_accuracy": 0.7354311645030975, "num_tokens": 12903339.0, "step": 267 }, { "entropy": 1.0598442554473877, "epoch": 0.41294298921417566, "grad_norm": 0.30677759647369385, "learning_rate": 1.959056660829723e-05, "loss": 1.0294, "mean_token_accuracy": 0.732042521238327, "num_tokens": 12960499.0, "step": 268 }, { "entropy": 1.0747231841087341, "epoch": 0.41448382126348227, "grad_norm": 0.31351029872894287, "learning_rate": 1.9585740739919946e-05, "loss": 1.052, "mean_token_accuracy": 0.7296299636363983, "num_tokens": 13016529.0, "step": 269 }, { "entropy": 1.0649322867393494, "epoch": 0.41602465331278893, "grad_norm": 0.3158623278141022, "learning_rate": 1.9580887198845055e-05, "loss": 1.0453, "mean_token_accuracy": 0.7278242707252502, "num_tokens": 13071322.0, "step": 270 }, { "entropy": 1.0961998105049133, "epoch": 0.41756548536209553, "grad_norm": 0.32320836186408997, "learning_rate": 1.957600599908406e-05, "loss": 1.1137, "mean_token_accuracy": 0.7124568521976471, "num_tokens": 13124708.0, "step": 271 }, { "entropy": 1.0882747769355774, "epoch": 0.41910631741140214, "grad_norm": 0.3281896710395813, "learning_rate": 1.9571097154728306e-05, "loss": 1.1027, "mean_token_accuracy": 0.7179121971130371, "num_tokens": 13175306.0, "step": 272 }, { "entropy": 1.0482059121131897, "epoch": 0.4206471494607088, "grad_norm": 0.5430480241775513, "learning_rate": 1.956616067994894e-05, "loss": 1.0447, "mean_token_accuracy": 0.7217666804790497, "num_tokens": 13189065.0, "step": 273 }, { "entropy": 1.0423285365104675, "epoch": 0.4221879815100154, "grad_norm": 0.5716179013252258, "learning_rate": 1.9561196588996873e-05, "loss": 1.0252, "mean_token_accuracy": 0.7324531376361847, "num_tokens": 13199203.0, "step": 274 }, { "entropy": 0.9190036356449127, "epoch": 0.423728813559322, "grad_norm": 0.6330918073654175, "learning_rate": 1.9556204896202748e-05, "loss": 0.889, "mean_token_accuracy": 0.7794653475284576, "num_tokens": 13207154.0, "step": 275 }, { "entropy": 0.8193510174751282, "epoch": 0.4252696456086287, "grad_norm": 0.3744107186794281, "learning_rate": 1.9551185615976883e-05, "loss": 0.8134, "mean_token_accuracy": 0.7874249219894409, "num_tokens": 13272365.0, "step": 276 }, { "entropy": 0.9770923852920532, "epoch": 0.4268104776579353, "grad_norm": 0.3651951551437378, "learning_rate": 1.9546138762809246e-05, "loss": 0.998, "mean_token_accuracy": 0.7396715581417084, "num_tokens": 13334351.0, "step": 277 }, { "entropy": 0.9495940506458282, "epoch": 0.4283513097072419, "grad_norm": 0.35534170269966125, "learning_rate": 1.9541064351269398e-05, "loss": 0.9685, "mean_token_accuracy": 0.746786892414093, "num_tokens": 13394652.0, "step": 278 }, { "entropy": 0.977106124162674, "epoch": 0.42989214175654855, "grad_norm": 0.335166335105896, "learning_rate": 1.9535962396006453e-05, "loss": 0.9847, "mean_token_accuracy": 0.7417066395282745, "num_tokens": 13453802.0, "step": 279 }, { "entropy": 1.011274516582489, "epoch": 0.43143297380585516, "grad_norm": 0.31974202394485474, "learning_rate": 1.9530832911749046e-05, "loss": 1.0196, "mean_token_accuracy": 0.7337155044078827, "num_tokens": 13511711.0, "step": 280 }, { "entropy": 1.0511541366577148, "epoch": 0.43297380585516176, "grad_norm": 0.32504236698150635, "learning_rate": 1.952567591330529e-05, "loss": 1.0479, "mean_token_accuracy": 0.7302321195602417, "num_tokens": 13568288.0, "step": 281 }, { "entropy": 1.02443066239357, "epoch": 0.4345146379044684, "grad_norm": 0.3239118754863739, "learning_rate": 1.952049141556271e-05, "loss": 1.0097, "mean_token_accuracy": 0.7399837672710419, "num_tokens": 13623656.0, "step": 282 }, { "entropy": 1.1016828417778015, "epoch": 0.43605546995377503, "grad_norm": 0.3372076153755188, "learning_rate": 1.9515279433488235e-05, "loss": 1.0722, "mean_token_accuracy": 0.7235656380653381, "num_tokens": 13677943.0, "step": 283 }, { "entropy": 1.1099361181259155, "epoch": 0.43759630200308164, "grad_norm": 0.341728150844574, "learning_rate": 1.951003998212813e-05, "loss": 1.1065, "mean_token_accuracy": 0.716703861951828, "num_tokens": 13730693.0, "step": 284 }, { "entropy": 1.1150873899459839, "epoch": 0.4391371340523883, "grad_norm": 0.3469105064868927, "learning_rate": 1.9504773076607963e-05, "loss": 1.1105, "mean_token_accuracy": 0.7193589806556702, "num_tokens": 13781172.0, "step": 285 }, { "entropy": 0.9573420584201813, "epoch": 0.4406779661016949, "grad_norm": 0.5325751900672913, "learning_rate": 1.949947873213255e-05, "loss": 0.9723, "mean_token_accuracy": 0.7387649118900299, "num_tokens": 13797790.0, "step": 286 }, { "entropy": 0.9526257812976837, "epoch": 0.4422187981510015, "grad_norm": 0.5918636322021484, "learning_rate": 1.9494156963985934e-05, "loss": 0.9529, "mean_token_accuracy": 0.7535216510295868, "num_tokens": 13807679.0, "step": 287 }, { "entropy": 0.8371843099594116, "epoch": 0.4437596302003082, "grad_norm": 0.6081588268280029, "learning_rate": 1.948880778753131e-05, "loss": 0.8404, "mean_token_accuracy": 0.7867316603660583, "num_tokens": 13844095.0, "step": 288 }, { "entropy": 0.8966417014598846, "epoch": 0.4453004622496148, "grad_norm": 0.3648871183395386, "learning_rate": 1.948343121821101e-05, "loss": 0.8799, "mean_token_accuracy": 0.7706982493400574, "num_tokens": 13908687.0, "step": 289 }, { "entropy": 0.9836355745792389, "epoch": 0.44684129429892144, "grad_norm": 0.348431795835495, "learning_rate": 1.947802727154644e-05, "loss": 0.9657, "mean_token_accuracy": 0.7476580142974854, "num_tokens": 13970523.0, "step": 290 }, { "entropy": 1.0435945987701416, "epoch": 0.44838212634822805, "grad_norm": 0.3355500102043152, "learning_rate": 1.9472595963138037e-05, "loss": 1.0094, "mean_token_accuracy": 0.7352299392223358, "num_tokens": 14030623.0, "step": 291 }, { "entropy": 1.0529168248176575, "epoch": 0.44992295839753466, "grad_norm": 0.35061150789260864, "learning_rate": 1.9467137308665237e-05, "loss": 1.0163, "mean_token_accuracy": 0.7340034544467926, "num_tokens": 14089351.0, "step": 292 }, { "entropy": 1.0424723029136658, "epoch": 0.4514637904468413, "grad_norm": 0.35146647691726685, "learning_rate": 1.9461651323886408e-05, "loss": 1.0132, "mean_token_accuracy": 0.7337992489337921, "num_tokens": 14146819.0, "step": 293 }, { "entropy": 1.0192838311195374, "epoch": 0.4530046224961479, "grad_norm": 0.3441980183124542, "learning_rate": 1.9456138024638832e-05, "loss": 1.0344, "mean_token_accuracy": 0.7295454144477844, "num_tokens": 14203200.0, "step": 294 }, { "entropy": 1.0251617431640625, "epoch": 0.45454545454545453, "grad_norm": 0.35392656922340393, "learning_rate": 1.9450597426838627e-05, "loss": 1.033, "mean_token_accuracy": 0.7327487468719482, "num_tokens": 14258355.0, "step": 295 }, { "entropy": 1.0295028686523438, "epoch": 0.4560862865947612, "grad_norm": 0.3656595051288605, "learning_rate": 1.9445029546480735e-05, "loss": 1.0476, "mean_token_accuracy": 0.7284654974937439, "num_tokens": 14312136.0, "step": 296 }, { "entropy": 1.0395074486732483, "epoch": 0.4576271186440678, "grad_norm": 0.38253554701805115, "learning_rate": 1.943943439963884e-05, "loss": 1.0511, "mean_token_accuracy": 0.7257314622402191, "num_tokens": 14364043.0, "step": 297 }, { "entropy": 1.0373750925064087, "epoch": 0.4591679506933744, "grad_norm": 0.4564608633518219, "learning_rate": 1.9433812002465364e-05, "loss": 1.0573, "mean_token_accuracy": 0.7324310839176178, "num_tokens": 14392824.0, "step": 298 }, { "entropy": 0.9486433863639832, "epoch": 0.46070878274268107, "grad_norm": 0.5727681517601013, "learning_rate": 1.9428162371191373e-05, "loss": 0.9199, "mean_token_accuracy": 0.7553496062755585, "num_tokens": 14403670.0, "step": 299 }, { "entropy": 0.9305461645126343, "epoch": 0.4622496147919877, "grad_norm": 0.6422452926635742, "learning_rate": 1.9422485522126574e-05, "loss": 0.9038, "mean_token_accuracy": 0.7690128684043884, "num_tokens": 14411917.0, "step": 300 }, { "entropy": 0.788120448589325, "epoch": 0.4637904468412943, "grad_norm": 0.5457755327224731, "learning_rate": 1.941678147165923e-05, "loss": 0.7176, "mean_token_accuracy": 0.8139662742614746, "num_tokens": 14477453.0, "step": 301 }, { "entropy": 0.9291627109050751, "epoch": 0.46533127889060094, "grad_norm": 0.3988177180290222, "learning_rate": 1.9411050236256145e-05, "loss": 0.9519, "mean_token_accuracy": 0.7496409714221954, "num_tokens": 14540771.0, "step": 302 }, { "entropy": 0.9507352709770203, "epoch": 0.46687211093990755, "grad_norm": 0.4390845000743866, "learning_rate": 1.94052918324626e-05, "loss": 0.9801, "mean_token_accuracy": 0.7419291734695435, "num_tokens": 14601666.0, "step": 303 }, { "entropy": 0.9616167545318604, "epoch": 0.46841294298921415, "grad_norm": 0.425306111574173, "learning_rate": 1.9399506276902297e-05, "loss": 0.9882, "mean_token_accuracy": 0.7444433867931366, "num_tokens": 14660953.0, "step": 304 }, { "entropy": 0.9728153645992279, "epoch": 0.4699537750385208, "grad_norm": 0.3965277671813965, "learning_rate": 1.9393693586277332e-05, "loss": 0.976, "mean_token_accuracy": 0.7464949190616608, "num_tokens": 14718995.0, "step": 305 }, { "entropy": 0.9875591993331909, "epoch": 0.4714946070878274, "grad_norm": 0.37671324610710144, "learning_rate": 1.9387853777368136e-05, "loss": 0.9826, "mean_token_accuracy": 0.7432833909988403, "num_tokens": 14776032.0, "step": 306 }, { "entropy": 1.0266414880752563, "epoch": 0.47303543913713403, "grad_norm": 0.38023555278778076, "learning_rate": 1.9381986867033418e-05, "loss": 1.0162, "mean_token_accuracy": 0.7359950244426727, "num_tokens": 14831861.0, "step": 307 }, { "entropy": 1.054717779159546, "epoch": 0.4745762711864407, "grad_norm": 0.377249538898468, "learning_rate": 1.9376092872210135e-05, "loss": 1.0326, "mean_token_accuracy": 0.730976015329361, "num_tokens": 14886415.0, "step": 308 }, { "entropy": 1.0463762879371643, "epoch": 0.4761171032357473, "grad_norm": 0.36740073561668396, "learning_rate": 1.9370171809913426e-05, "loss": 1.028, "mean_token_accuracy": 0.7364760339260101, "num_tokens": 14939669.0, "step": 309 }, { "entropy": 0.9874489009380341, "epoch": 0.4776579352850539, "grad_norm": 0.37306901812553406, "learning_rate": 1.936422369723657e-05, "loss": 0.9682, "mean_token_accuracy": 0.7482308745384216, "num_tokens": 14990671.0, "step": 310 }, { "entropy": 0.9723087549209595, "epoch": 0.47919876733436056, "grad_norm": 0.5721508860588074, "learning_rate": 1.9358248551350945e-05, "loss": 1.0665, "mean_token_accuracy": 0.7290179431438446, "num_tokens": 15008661.0, "step": 311 }, { "entropy": 0.8790964484214783, "epoch": 0.48073959938366717, "grad_norm": 0.6861698627471924, "learning_rate": 1.9352246389505963e-05, "loss": 0.9063, "mean_token_accuracy": 0.7682745158672333, "num_tokens": 15018039.0, "step": 312 }, { "entropy": 0.7997311949729919, "epoch": 0.48228043143297383, "grad_norm": 0.5569450259208679, "learning_rate": 1.9346217229029026e-05, "loss": 0.7924, "mean_token_accuracy": 0.798639714717865, "num_tokens": 15054391.0, "step": 313 }, { "entropy": 0.8583049178123474, "epoch": 0.48382126348228044, "grad_norm": 0.46146467328071594, "learning_rate": 1.9340161087325483e-05, "loss": 0.8608, "mean_token_accuracy": 0.773988276720047, "num_tokens": 15119095.0, "step": 314 }, { "entropy": 0.9356653094291687, "epoch": 0.48536209553158705, "grad_norm": 0.4175281822681427, "learning_rate": 1.933407798187857e-05, "loss": 0.9103, "mean_token_accuracy": 0.7580974102020264, "num_tokens": 15181120.0, "step": 315 }, { "entropy": 0.9827759563922882, "epoch": 0.4869029275808937, "grad_norm": 0.3664589524269104, "learning_rate": 1.9327967930249362e-05, "loss": 0.9577, "mean_token_accuracy": 0.7468955516815186, "num_tokens": 15241233.0, "step": 316 }, { "entropy": 1.017783522605896, "epoch": 0.4884437596302003, "grad_norm": 0.37999436259269714, "learning_rate": 1.9321830950076732e-05, "loss": 0.996, "mean_token_accuracy": 0.7350513637065887, "num_tokens": 15299960.0, "step": 317 }, { "entropy": 0.978701263666153, "epoch": 0.4899845916795069, "grad_norm": 0.3845328688621521, "learning_rate": 1.931566705907728e-05, "loss": 0.9458, "mean_token_accuracy": 0.7519245147705078, "num_tokens": 15357359.0, "step": 318 }, { "entropy": 0.9867020845413208, "epoch": 0.4915254237288136, "grad_norm": 0.40629714727401733, "learning_rate": 1.9309476275045306e-05, "loss": 0.9748, "mean_token_accuracy": 0.7458271980285645, "num_tokens": 15413561.0, "step": 319 }, { "entropy": 1.0241209268569946, "epoch": 0.4930662557781202, "grad_norm": 0.39157143235206604, "learning_rate": 1.9303258615852737e-05, "loss": 1.0182, "mean_token_accuracy": 0.7341704368591309, "num_tokens": 15468633.0, "step": 320 }, { "entropy": 1.0174873769283295, "epoch": 0.4946070878274268, "grad_norm": 0.4042147994041443, "learning_rate": 1.9297014099449095e-05, "loss": 1.0062, "mean_token_accuracy": 0.7380396127700806, "num_tokens": 15522187.0, "step": 321 }, { "entropy": 1.0111220479011536, "epoch": 0.49614791987673346, "grad_norm": 0.44198617339134216, "learning_rate": 1.929074274386142e-05, "loss": 1.0163, "mean_token_accuracy": 0.7355753183364868, "num_tokens": 15573193.0, "step": 322 }, { "entropy": 0.9588909447193146, "epoch": 0.49768875192604006, "grad_norm": 0.619880199432373, "learning_rate": 1.9284444567194246e-05, "loss": 0.9881, "mean_token_accuracy": 0.7364840805530548, "num_tokens": 15592293.0, "step": 323 }, { "entropy": 1.0024993121623993, "epoch": 0.49922958397534667, "grad_norm": 0.6752212643623352, "learning_rate": 1.927811958762953e-05, "loss": 1.0069, "mean_token_accuracy": 0.7398483753204346, "num_tokens": 15602871.0, "step": 324 }, { "entropy": 0.9022562503814697, "epoch": 0.5007704160246533, "grad_norm": 0.7096251845359802, "learning_rate": 1.9271767823426607e-05, "loss": 0.8933, "mean_token_accuracy": 0.7769883275032043, "num_tokens": 15610857.0, "step": 325 }, { "entropy": 0.7723791003227234, "epoch": 0.50231124807396, "grad_norm": 0.5227532386779785, "learning_rate": 1.9265389292922133e-05, "loss": 0.7118, "mean_token_accuracy": 0.8148288726806641, "num_tokens": 15676226.0, "step": 326 }, { "entropy": 0.8748729228973389, "epoch": 0.5038520801232665, "grad_norm": 0.47217506170272827, "learning_rate": 1.9258984014530043e-05, "loss": 0.867, "mean_token_accuracy": 0.7683269679546356, "num_tokens": 15739065.0, "step": 327 }, { "entropy": 0.9348930716514587, "epoch": 0.5053929121725732, "grad_norm": 0.4359320402145386, "learning_rate": 1.925255200674148e-05, "loss": 0.9455, "mean_token_accuracy": 0.750718742609024, "num_tokens": 15799434.0, "step": 328 }, { "entropy": 0.9439020156860352, "epoch": 0.5069337442218799, "grad_norm": 0.4252333343029022, "learning_rate": 1.9246093288124748e-05, "loss": 0.9505, "mean_token_accuracy": 0.7492730915546417, "num_tokens": 15858171.0, "step": 329 }, { "entropy": 0.9464855492115021, "epoch": 0.5084745762711864, "grad_norm": 0.43666476011276245, "learning_rate": 1.9239607877325275e-05, "loss": 0.957, "mean_token_accuracy": 0.7486556768417358, "num_tokens": 15915778.0, "step": 330 }, { "entropy": 0.9750763177871704, "epoch": 0.5100154083204931, "grad_norm": 0.4290591776371002, "learning_rate": 1.9233095793065534e-05, "loss": 0.9857, "mean_token_accuracy": 0.7403514981269836, "num_tokens": 15972280.0, "step": 331 }, { "entropy": 0.9656431376934052, "epoch": 0.5115562403697997, "grad_norm": 0.4063112437725067, "learning_rate": 1.9226557054145008e-05, "loss": 0.9412, "mean_token_accuracy": 0.7532393634319305, "num_tokens": 16027711.0, "step": 332 }, { "entropy": 1.0258893966674805, "epoch": 0.5130970724191063, "grad_norm": 0.4059493839740753, "learning_rate": 1.921999167944012e-05, "loss": 1.0032, "mean_token_accuracy": 0.7398551404476166, "num_tokens": 16081906.0, "step": 333 }, { "entropy": 1.0106145739555359, "epoch": 0.514637904468413, "grad_norm": 0.4083745777606964, "learning_rate": 1.92133996879042e-05, "loss": 0.9892, "mean_token_accuracy": 0.7434115707874298, "num_tokens": 16134225.0, "step": 334 }, { "entropy": 0.9619610011577606, "epoch": 0.5161787365177196, "grad_norm": 0.4626840353012085, "learning_rate": 1.9206781098567403e-05, "loss": 0.96, "mean_token_accuracy": 0.7546554505825043, "num_tokens": 16175563.0, "step": 335 }, { "entropy": 0.99931401014328, "epoch": 0.5177195685670262, "grad_norm": 0.691291868686676, "learning_rate": 1.9200135930536672e-05, "loss": 1.0129, "mean_token_accuracy": 0.7296529710292816, "num_tokens": 16187683.0, "step": 336 }, { "entropy": 0.9438363015651703, "epoch": 0.5192604006163328, "grad_norm": 0.6817367672920227, "learning_rate": 1.9193464202995688e-05, "loss": 0.9587, "mean_token_accuracy": 0.7613221406936646, "num_tokens": 16196830.0, "step": 337 }, { "entropy": 0.75175541639328, "epoch": 0.5208012326656395, "grad_norm": 0.6863135099411011, "learning_rate": 1.918676593520479e-05, "loss": 0.6223, "mean_token_accuracy": 0.8124595582485199, "num_tokens": 16233037.0, "step": 338 }, { "entropy": 0.8207379579544067, "epoch": 0.522342064714946, "grad_norm": 0.4774440824985504, "learning_rate": 1.918004114650095e-05, "loss": 0.8234, "mean_token_accuracy": 0.7817659676074982, "num_tokens": 16296917.0, "step": 339 }, { "entropy": 0.9065532386302948, "epoch": 0.5238828967642527, "grad_norm": 0.4492279291152954, "learning_rate": 1.9173289856297693e-05, "loss": 0.8951, "mean_token_accuracy": 0.758342981338501, "num_tokens": 16358669.0, "step": 340 }, { "entropy": 0.9596481621265411, "epoch": 0.5254237288135594, "grad_norm": 0.42398199439048767, "learning_rate": 1.9166512084085055e-05, "loss": 0.9462, "mean_token_accuracy": 0.7498698234558105, "num_tokens": 16418551.0, "step": 341 }, { "entropy": 0.9638877213001251, "epoch": 0.5269645608628659, "grad_norm": 0.42384251952171326, "learning_rate": 1.915970784942952e-05, "loss": 0.932, "mean_token_accuracy": 0.7555508613586426, "num_tokens": 16477024.0, "step": 342 }, { "entropy": 0.9978684782981873, "epoch": 0.5285053929121726, "grad_norm": 0.44009339809417725, "learning_rate": 1.915287717197396e-05, "loss": 0.9734, "mean_token_accuracy": 0.7456685304641724, "num_tokens": 16534254.0, "step": 343 }, { "entropy": 0.9786334335803986, "epoch": 0.5300462249614792, "grad_norm": 0.4255545139312744, "learning_rate": 1.9146020071437597e-05, "loss": 0.9591, "mean_token_accuracy": 0.7458925843238831, "num_tokens": 16590374.0, "step": 344 }, { "entropy": 0.9350001811981201, "epoch": 0.5315870570107858, "grad_norm": 0.4049918055534363, "learning_rate": 1.913913656761592e-05, "loss": 0.9193, "mean_token_accuracy": 0.7574960589408875, "num_tokens": 16645390.0, "step": 345 }, { "entropy": 0.995197594165802, "epoch": 0.5331278890600925, "grad_norm": 0.44920212030410767, "learning_rate": 1.9132226680380642e-05, "loss": 0.9979, "mean_token_accuracy": 0.7424894869327545, "num_tokens": 16699007.0, "step": 346 }, { "entropy": 0.9318511784076691, "epoch": 0.5346687211093991, "grad_norm": 0.48988234996795654, "learning_rate": 1.9125290429679652e-05, "loss": 0.9333, "mean_token_accuracy": 0.7542620003223419, "num_tokens": 16750907.0, "step": 347 }, { "entropy": 0.96406489610672, "epoch": 0.5362095531587057, "grad_norm": 0.596936047077179, "learning_rate": 1.911832783553693e-05, "loss": 0.8968, "mean_token_accuracy": 0.7544503211975098, "num_tokens": 16774354.0, "step": 348 }, { "entropy": 1.0202385187149048, "epoch": 0.5377503852080123, "grad_norm": 0.6295598149299622, "learning_rate": 1.9111338918052525e-05, "loss": 0.9881, "mean_token_accuracy": 0.7420793175697327, "num_tokens": 16785100.0, "step": 349 }, { "entropy": 0.9698611497879028, "epoch": 0.539291217257319, "grad_norm": 0.712397575378418, "learning_rate": 1.9104323697402462e-05, "loss": 0.9438, "mean_token_accuracy": 0.7687529921531677, "num_tokens": 16793234.0, "step": 350 }, { "entropy": 0.7430060803890228, "epoch": 0.5408320493066255, "grad_norm": 0.47920989990234375, "learning_rate": 1.9097282193838704e-05, "loss": 0.7129, "mean_token_accuracy": 0.8102925419807434, "num_tokens": 16858300.0, "step": 351 }, { "entropy": 0.906875342130661, "epoch": 0.5423728813559322, "grad_norm": 0.46368905901908875, "learning_rate": 1.9090214427689094e-05, "loss": 0.919, "mean_token_accuracy": 0.7573151588439941, "num_tokens": 16920599.0, "step": 352 }, { "entropy": 0.8942819833755493, "epoch": 0.5439137134052389, "grad_norm": 0.4281911253929138, "learning_rate": 1.9083120419357286e-05, "loss": 0.9053, "mean_token_accuracy": 0.7580410242080688, "num_tokens": 16981201.0, "step": 353 }, { "entropy": 0.8971975743770599, "epoch": 0.5454545454545454, "grad_norm": 0.4326510727405548, "learning_rate": 1.9076000189322694e-05, "loss": 0.9029, "mean_token_accuracy": 0.762725442647934, "num_tokens": 17040238.0, "step": 354 }, { "entropy": 0.9092396199703217, "epoch": 0.5469953775038521, "grad_norm": 0.43044859170913696, "learning_rate": 1.9068853758140432e-05, "loss": 0.9227, "mean_token_accuracy": 0.759487122297287, "num_tokens": 17098172.0, "step": 355 }, { "entropy": 0.9509382843971252, "epoch": 0.5485362095531587, "grad_norm": 0.4300386607646942, "learning_rate": 1.9061681146441245e-05, "loss": 0.9465, "mean_token_accuracy": 0.7506694495677948, "num_tokens": 17155115.0, "step": 356 }, { "entropy": 0.9467697143554688, "epoch": 0.5500770416024653, "grad_norm": 0.46839937567710876, "learning_rate": 1.905448237493147e-05, "loss": 0.9468, "mean_token_accuracy": 0.7517959475517273, "num_tokens": 17210904.0, "step": 357 }, { "entropy": 0.9165197610855103, "epoch": 0.551617873651772, "grad_norm": 0.46427878737449646, "learning_rate": 1.904725746439295e-05, "loss": 0.902, "mean_token_accuracy": 0.7628574967384338, "num_tokens": 17265367.0, "step": 358 }, { "entropy": 0.9306047558784485, "epoch": 0.5531587057010786, "grad_norm": 0.47638171911239624, "learning_rate": 1.9040006435683e-05, "loss": 0.8957, "mean_token_accuracy": 0.766440212726593, "num_tokens": 17318391.0, "step": 359 }, { "entropy": 0.9474429786205292, "epoch": 0.5546995377503852, "grad_norm": 0.515444278717041, "learning_rate": 1.903272930973433e-05, "loss": 0.9292, "mean_token_accuracy": 0.7564335465431213, "num_tokens": 17367502.0, "step": 360 }, { "entropy": 0.9583354294300079, "epoch": 0.5562403697996918, "grad_norm": 0.6589438915252686, "learning_rate": 1.9025426107554987e-05, "loss": 1.0037, "mean_token_accuracy": 0.730921059846878, "num_tokens": 17379941.0, "step": 361 }, { "entropy": 0.9522935152053833, "epoch": 0.5577812018489985, "grad_norm": 0.7380706071853638, "learning_rate": 1.9018096850228297e-05, "loss": 0.9704, "mean_token_accuracy": 0.7632938921451569, "num_tokens": 17388887.0, "step": 362 }, { "entropy": 0.7399669289588928, "epoch": 0.559322033898305, "grad_norm": 0.6387145519256592, "learning_rate": 1.9010741558912803e-05, "loss": 0.6723, "mean_token_accuracy": 0.8114824295043945, "num_tokens": 17425084.0, "step": 363 }, { "entropy": 0.8060667216777802, "epoch": 0.5608628659476117, "grad_norm": 0.4969497621059418, "learning_rate": 1.9003360254842207e-05, "loss": 0.7808, "mean_token_accuracy": 0.7916131019592285, "num_tokens": 17489858.0, "step": 364 }, { "entropy": 0.8900258541107178, "epoch": 0.5624036979969184, "grad_norm": 0.4824165105819702, "learning_rate": 1.8995952959325312e-05, "loss": 0.8742, "mean_token_accuracy": 0.7675205171108246, "num_tokens": 17552019.0, "step": 365 }, { "entropy": 0.8560473322868347, "epoch": 0.5639445300462249, "grad_norm": 0.47723862528800964, "learning_rate": 1.898851969374594e-05, "loss": 0.8351, "mean_token_accuracy": 0.7786802053451538, "num_tokens": 17612349.0, "step": 366 }, { "entropy": 0.9277917146682739, "epoch": 0.5654853620955316, "grad_norm": 0.4568464159965515, "learning_rate": 1.8981060479562893e-05, "loss": 0.8982, "mean_token_accuracy": 0.7602544128894806, "num_tokens": 17671415.0, "step": 367 }, { "entropy": 0.942799985408783, "epoch": 0.5670261941448382, "grad_norm": 0.45853057503700256, "learning_rate": 1.8973575338309884e-05, "loss": 0.9259, "mean_token_accuracy": 0.757209300994873, "num_tokens": 17729215.0, "step": 368 }, { "entropy": 0.9606009423732758, "epoch": 0.5685670261941448, "grad_norm": 0.48883330821990967, "learning_rate": 1.8966064291595472e-05, "loss": 0.9509, "mean_token_accuracy": 0.7495742440223694, "num_tokens": 17786036.0, "step": 369 }, { "entropy": 0.9659188985824585, "epoch": 0.5701078582434514, "grad_norm": 0.4622895419597626, "learning_rate": 1.8958527361103005e-05, "loss": 0.9597, "mean_token_accuracy": 0.7489015460014343, "num_tokens": 17841832.0, "step": 370 }, { "entropy": 0.9491275548934937, "epoch": 0.5716486902927581, "grad_norm": 0.49132686853408813, "learning_rate": 1.8950964568590548e-05, "loss": 0.9425, "mean_token_accuracy": 0.7568167746067047, "num_tokens": 17896435.0, "step": 371 }, { "entropy": 0.9638166129589081, "epoch": 0.5731895223420647, "grad_norm": 0.4718216359615326, "learning_rate": 1.8943375935890834e-05, "loss": 0.9435, "mean_token_accuracy": 0.754529744386673, "num_tokens": 17949375.0, "step": 372 }, { "entropy": 0.9775730967521667, "epoch": 0.5747303543913713, "grad_norm": 0.5234924554824829, "learning_rate": 1.8935761484911183e-05, "loss": 0.951, "mean_token_accuracy": 0.7504199743270874, "num_tokens": 17997335.0, "step": 373 }, { "entropy": 1.0391918420791626, "epoch": 0.576271186440678, "grad_norm": 0.6570227742195129, "learning_rate": 1.8928121237633454e-05, "loss": 0.9957, "mean_token_accuracy": 0.7411914765834808, "num_tokens": 18008773.0, "step": 374 }, { "entropy": 0.9248173236846924, "epoch": 0.5778120184899846, "grad_norm": 0.7648917436599731, "learning_rate": 1.8920455216113975e-05, "loss": 0.9391, "mean_token_accuracy": 0.769658774137497, "num_tokens": 18016845.0, "step": 375 }, { "entropy": 0.7248341739177704, "epoch": 0.5793528505392912, "grad_norm": 0.5560222268104553, "learning_rate": 1.8912763442483484e-05, "loss": 0.6856, "mean_token_accuracy": 0.8159162402153015, "num_tokens": 18081948.0, "step": 376 }, { "entropy": 0.8317951560020447, "epoch": 0.5808936825885979, "grad_norm": 0.4836423099040985, "learning_rate": 1.8905045938947063e-05, "loss": 0.8387, "mean_token_accuracy": 0.7765500545501709, "num_tokens": 18144120.0, "step": 377 }, { "entropy": 0.8737204968929291, "epoch": 0.5824345146379045, "grad_norm": 0.5062094330787659, "learning_rate": 1.8897302727784063e-05, "loss": 0.892, "mean_token_accuracy": 0.7635113596916199, "num_tokens": 18204222.0, "step": 378 }, { "entropy": 0.8914177417755127, "epoch": 0.5839753466872111, "grad_norm": 0.5173004269599915, "learning_rate": 1.8889533831348058e-05, "loss": 0.9088, "mean_token_accuracy": 0.7585304379463196, "num_tokens": 18263002.0, "step": 379 }, { "entropy": 0.8768998682498932, "epoch": 0.5855161787365177, "grad_norm": 0.507412850856781, "learning_rate": 1.8881739272066772e-05, "loss": 0.8936, "mean_token_accuracy": 0.7639376819133759, "num_tokens": 18320555.0, "step": 380 }, { "entropy": 0.9562215507030487, "epoch": 0.5870570107858244, "grad_norm": 0.4737968146800995, "learning_rate": 1.8873919072442003e-05, "loss": 0.9437, "mean_token_accuracy": 0.7496383488178253, "num_tokens": 18377094.0, "step": 381 }, { "entropy": 0.8826883137226105, "epoch": 0.588597842835131, "grad_norm": 0.4925789535045624, "learning_rate": 1.8866073255049588e-05, "loss": 0.8444, "mean_token_accuracy": 0.7783036530017853, "num_tokens": 18432700.0, "step": 382 }, { "entropy": 0.9323877692222595, "epoch": 0.5901386748844376, "grad_norm": 0.5395112037658691, "learning_rate": 1.8858201842539303e-05, "loss": 0.8828, "mean_token_accuracy": 0.7662901282310486, "num_tokens": 18487160.0, "step": 383 }, { "entropy": 0.9540666341781616, "epoch": 0.5916795069337443, "grad_norm": 0.5016188621520996, "learning_rate": 1.8850304857634817e-05, "loss": 0.9513, "mean_token_accuracy": 0.7508966028690338, "num_tokens": 18540360.0, "step": 384 }, { "entropy": 0.929486483335495, "epoch": 0.5932203389830508, "grad_norm": 0.5618557929992676, "learning_rate": 1.884238232313363e-05, "loss": 0.9283, "mean_token_accuracy": 0.7576351761817932, "num_tokens": 18589702.0, "step": 385 }, { "entropy": 0.9237514138221741, "epoch": 0.5947611710323575, "grad_norm": 0.6265010237693787, "learning_rate": 1.8834434261906992e-05, "loss": 0.9266, "mean_token_accuracy": 0.7474926114082336, "num_tokens": 18602539.0, "step": 386 }, { "entropy": 0.9524433016777039, "epoch": 0.5963020030816641, "grad_norm": 0.6790736317634583, "learning_rate": 1.8826460696899848e-05, "loss": 0.9509, "mean_token_accuracy": 0.7592921555042267, "num_tokens": 18611995.0, "step": 387 }, { "entropy": 0.6810170710086823, "epoch": 0.5978428351309707, "grad_norm": 0.7948004603385925, "learning_rate": 1.8818461651130767e-05, "loss": 0.6613, "mean_token_accuracy": 0.8255063593387604, "num_tokens": 18648394.0, "step": 388 }, { "entropy": 0.7073840498924255, "epoch": 0.5993836671802774, "grad_norm": 0.51377934217453, "learning_rate": 1.8810437147691883e-05, "loss": 0.6954, "mean_token_accuracy": 0.8139230906963348, "num_tokens": 18713161.0, "step": 389 }, { "entropy": 0.8280047178268433, "epoch": 0.600924499229584, "grad_norm": 0.5041950345039368, "learning_rate": 1.880238720974882e-05, "loss": 0.8094, "mean_token_accuracy": 0.7832685708999634, "num_tokens": 18775076.0, "step": 390 }, { "entropy": 0.8993235230445862, "epoch": 0.6024653312788906, "grad_norm": 0.5131763815879822, "learning_rate": 1.8794311860540623e-05, "loss": 0.8628, "mean_token_accuracy": 0.7681954205036163, "num_tokens": 18835235.0, "step": 391 }, { "entropy": 0.9165942072868347, "epoch": 0.6040061633281972, "grad_norm": 0.5255646109580994, "learning_rate": 1.87862111233797e-05, "loss": 0.8734, "mean_token_accuracy": 0.7684797048568726, "num_tokens": 18893835.0, "step": 392 }, { "entropy": 0.909884512424469, "epoch": 0.6055469953775039, "grad_norm": 0.4816731810569763, "learning_rate": 1.8778085021651756e-05, "loss": 0.894, "mean_token_accuracy": 0.7627480030059814, "num_tokens": 18951088.0, "step": 393 }, { "entropy": 0.8697177171707153, "epoch": 0.6070878274268104, "grad_norm": 0.5432159900665283, "learning_rate": 1.8769933578815707e-05, "loss": 0.8784, "mean_token_accuracy": 0.7715619504451752, "num_tokens": 19007181.0, "step": 394 }, { "entropy": 0.8743345737457275, "epoch": 0.6086286594761171, "grad_norm": 0.6134296655654907, "learning_rate": 1.8761756818403634e-05, "loss": 0.8791, "mean_token_accuracy": 0.7680772542953491, "num_tokens": 19062043.0, "step": 395 }, { "entropy": 0.9726457297801971, "epoch": 0.6101694915254238, "grad_norm": 0.5376118421554565, "learning_rate": 1.8753554764020707e-05, "loss": 0.9787, "mean_token_accuracy": 0.7459292411804199, "num_tokens": 19115464.0, "step": 396 }, { "entropy": 1.0112932920455933, "epoch": 0.6117103235747303, "grad_norm": 0.5408052206039429, "learning_rate": 1.8745327439345113e-05, "loss": 0.993, "mean_token_accuracy": 0.7409583926200867, "num_tokens": 19167109.0, "step": 397 }, { "entropy": 0.9847508072853088, "epoch": 0.613251155624037, "grad_norm": 0.673870325088501, "learning_rate": 1.8737074868127997e-05, "loss": 0.8994, "mean_token_accuracy": 0.7529029250144958, "num_tokens": 19190902.0, "step": 398 }, { "entropy": 0.9997061491012573, "epoch": 0.6147919876733436, "grad_norm": 0.6840930581092834, "learning_rate": 1.872879707419337e-05, "loss": 0.9803, "mean_token_accuracy": 0.7452785074710846, "num_tokens": 19201376.0, "step": 399 }, { "entropy": 0.8578898906707764, "epoch": 0.6163328197226502, "grad_norm": 0.7989295125007629, "learning_rate": 1.872049408143808e-05, "loss": 0.8303, "mean_token_accuracy": 0.792082816362381, "num_tokens": 19209480.0, "step": 400 }, { "epoch": 0.6163328197226502, "eval_entropy": 0.8946358238188725, "eval_loss": 0.8694416880607605, "eval_mean_token_accuracy": 0.7718655528887263, "eval_num_tokens": 19209480.0, "eval_runtime": 192.8164, "eval_samples_per_second": 4.398, "eval_steps_per_second": 1.099, "step": 400 }, { "entropy": 0.8017340004444122, "epoch": 0.6178736517719569, "grad_norm": 0.5804023742675781, "learning_rate": 1.87121659138317e-05, "loss": 0.742, "mean_token_accuracy": 0.8034992218017578, "num_tokens": 19274440.0, "step": 401 }, { "entropy": 0.8292969465255737, "epoch": 0.6194144838212635, "grad_norm": 0.5548847317695618, "learning_rate": 1.8703812595416503e-05, "loss": 0.8284, "mean_token_accuracy": 0.7784838080406189, "num_tokens": 19336682.0, "step": 402 }, { "entropy": 0.7891210615634918, "epoch": 0.6209553158705701, "grad_norm": 0.608067512512207, "learning_rate": 1.8695434150307342e-05, "loss": 0.8026, "mean_token_accuracy": 0.7863937616348267, "num_tokens": 19397257.0, "step": 403 }, { "entropy": 0.8787820339202881, "epoch": 0.6224961479198767, "grad_norm": 0.5980172157287598, "learning_rate": 1.8687030602691628e-05, "loss": 0.9058, "mean_token_accuracy": 0.7615223526954651, "num_tokens": 19456225.0, "step": 404 }, { "entropy": 0.872244656085968, "epoch": 0.6240369799691834, "grad_norm": 0.5046532154083252, "learning_rate": 1.867860197682923e-05, "loss": 0.8821, "mean_token_accuracy": 0.7672625184059143, "num_tokens": 19513965.0, "step": 405 }, { "entropy": 0.9049012362957001, "epoch": 0.6255778120184899, "grad_norm": 0.5250270962715149, "learning_rate": 1.8670148297052416e-05, "loss": 0.8875, "mean_token_accuracy": 0.7671936750411987, "num_tokens": 19570648.0, "step": 406 }, { "entropy": 0.8916016221046448, "epoch": 0.6271186440677966, "grad_norm": 0.5394876599311829, "learning_rate": 1.866166958776578e-05, "loss": 0.8622, "mean_token_accuracy": 0.7718310952186584, "num_tokens": 19626371.0, "step": 407 }, { "entropy": 0.9408760964870453, "epoch": 0.6286594761171033, "grad_norm": 0.5322009325027466, "learning_rate": 1.865316587344618e-05, "loss": 0.9308, "mean_token_accuracy": 0.7591669857501984, "num_tokens": 19681062.0, "step": 408 }, { "entropy": 0.8904439210891724, "epoch": 0.6302003081664098, "grad_norm": 0.5848644375801086, "learning_rate": 1.8644637178642642e-05, "loss": 0.8831, "mean_token_accuracy": 0.7689194083213806, "num_tokens": 19734376.0, "step": 409 }, { "entropy": 0.8647785186767578, "epoch": 0.6317411402157165, "grad_norm": 0.6137943863868713, "learning_rate": 1.863608352797633e-05, "loss": 0.8648, "mean_token_accuracy": 0.7740585207939148, "num_tokens": 19784741.0, "step": 410 }, { "entropy": 0.8906062841415405, "epoch": 0.6332819722650231, "grad_norm": 0.7085247039794922, "learning_rate": 1.862750494614044e-05, "loss": 0.9341, "mean_token_accuracy": 0.7520233690738678, "num_tokens": 19799172.0, "step": 411 }, { "entropy": 0.9226191639900208, "epoch": 0.6348228043143297, "grad_norm": 0.741629958152771, "learning_rate": 1.8618901457900138e-05, "loss": 0.943, "mean_token_accuracy": 0.7592947781085968, "num_tokens": 19808459.0, "step": 412 }, { "entropy": 0.7389196455478668, "epoch": 0.6363636363636364, "grad_norm": 0.6515417098999023, "learning_rate": 1.8610273088092497e-05, "loss": 0.604, "mean_token_accuracy": 0.8199951648712158, "num_tokens": 19844715.0, "step": 413 }, { "entropy": 0.7575202584266663, "epoch": 0.637904468412943, "grad_norm": 0.5690957307815552, "learning_rate": 1.8601619861626424e-05, "loss": 0.708, "mean_token_accuracy": 0.8081709444522858, "num_tokens": 19909360.0, "step": 414 }, { "entropy": 0.8621223270893097, "epoch": 0.6394453004622496, "grad_norm": 0.5643359422683716, "learning_rate": 1.859294180348257e-05, "loss": 0.8325, "mean_token_accuracy": 0.77853062748909, "num_tokens": 19971047.0, "step": 415 }, { "entropy": 0.8522298038005829, "epoch": 0.6409861325115562, "grad_norm": 0.5687861442565918, "learning_rate": 1.858423893871328e-05, "loss": 0.8388, "mean_token_accuracy": 0.7769237756729126, "num_tokens": 20030931.0, "step": 416 }, { "entropy": 0.8671985268592834, "epoch": 0.6425269645608629, "grad_norm": 0.5722052454948425, "learning_rate": 1.8575511292442518e-05, "loss": 0.8615, "mean_token_accuracy": 0.7710536420345306, "num_tokens": 20089470.0, "step": 417 }, { "entropy": 0.866868793964386, "epoch": 0.6440677966101694, "grad_norm": 0.5465446710586548, "learning_rate": 1.8566758889865775e-05, "loss": 0.8648, "mean_token_accuracy": 0.7716164588928223, "num_tokens": 20146909.0, "step": 418 }, { "entropy": 0.8640634715557098, "epoch": 0.6456086286594761, "grad_norm": 0.5354492664337158, "learning_rate": 1.855798175625002e-05, "loss": 0.8519, "mean_token_accuracy": 0.7745667994022369, "num_tokens": 20203267.0, "step": 419 }, { "entropy": 0.8851080536842346, "epoch": 0.6471494607087828, "grad_norm": 0.571625292301178, "learning_rate": 1.854917991693362e-05, "loss": 0.8565, "mean_token_accuracy": 0.7719822525978088, "num_tokens": 20258570.0, "step": 420 }, { "entropy": 0.9429062008857727, "epoch": 0.6486902927580893, "grad_norm": 0.5721187591552734, "learning_rate": 1.8540353397326253e-05, "loss": 0.9221, "mean_token_accuracy": 0.7575022578239441, "num_tokens": 20312376.0, "step": 421 }, { "entropy": 0.8565718233585358, "epoch": 0.650231124807396, "grad_norm": 0.5945937633514404, "learning_rate": 1.853150222290885e-05, "loss": 0.8439, "mean_token_accuracy": 0.7773304283618927, "num_tokens": 20364497.0, "step": 422 }, { "entropy": 0.935333788394928, "epoch": 0.6517719568567026, "grad_norm": 0.7657658457756042, "learning_rate": 1.852262641923352e-05, "loss": 0.8638, "mean_token_accuracy": 0.7589642703533173, "num_tokens": 20395906.0, "step": 423 }, { "entropy": 1.0185968279838562, "epoch": 0.6533127889060092, "grad_norm": 0.712435245513916, "learning_rate": 1.8513726011923472e-05, "loss": 1.0081, "mean_token_accuracy": 0.7379648685455322, "num_tokens": 20406640.0, "step": 424 }, { "entropy": 0.9407070875167847, "epoch": 0.6548536209553159, "grad_norm": 0.7736991047859192, "learning_rate": 1.850480102667294e-05, "loss": 0.9308, "mean_token_accuracy": 0.7727629840373993, "num_tokens": 20414621.0, "step": 425 }, { "entropy": 0.6616452038288116, "epoch": 0.6563944530046225, "grad_norm": 0.6516114473342896, "learning_rate": 1.8495851489247114e-05, "loss": 0.6407, "mean_token_accuracy": 0.8275282084941864, "num_tokens": 20479850.0, "step": 426 }, { "entropy": 0.826431155204773, "epoch": 0.6579352850539292, "grad_norm": 0.5665984749794006, "learning_rate": 1.8486877425482062e-05, "loss": 0.8146, "mean_token_accuracy": 0.7849454283714294, "num_tokens": 20542183.0, "step": 427 }, { "entropy": 0.8647668063640594, "epoch": 0.6594761171032357, "grad_norm": 0.5338972806930542, "learning_rate": 1.847787886128465e-05, "loss": 0.8403, "mean_token_accuracy": 0.7751559615135193, "num_tokens": 20602625.0, "step": 428 }, { "entropy": 0.8133454322814941, "epoch": 0.6610169491525424, "grad_norm": 0.5385761260986328, "learning_rate": 1.8468855822632485e-05, "loss": 0.8025, "mean_token_accuracy": 0.7849075198173523, "num_tokens": 20661560.0, "step": 429 }, { "entropy": 0.8264959156513214, "epoch": 0.662557781201849, "grad_norm": 0.5496692657470703, "learning_rate": 1.8459808335573818e-05, "loss": 0.8193, "mean_token_accuracy": 0.7849914729595184, "num_tokens": 20719312.0, "step": 430 }, { "entropy": 0.8110030591487885, "epoch": 0.6640986132511556, "grad_norm": 0.5584554672241211, "learning_rate": 1.845073642622749e-05, "loss": 0.8021, "mean_token_accuracy": 0.787286251783371, "num_tokens": 20776014.0, "step": 431 }, { "entropy": 0.8392120003700256, "epoch": 0.6656394453004623, "grad_norm": 0.5671530961990356, "learning_rate": 1.844164012078283e-05, "loss": 0.839, "mean_token_accuracy": 0.782294511795044, "num_tokens": 20831698.0, "step": 432 }, { "entropy": 0.8605882525444031, "epoch": 0.6671802773497689, "grad_norm": 0.5795756578445435, "learning_rate": 1.8432519445499607e-05, "loss": 0.8439, "mean_token_accuracy": 0.7784620225429535, "num_tokens": 20886410.0, "step": 433 }, { "entropy": 0.8011252284049988, "epoch": 0.6687211093990755, "grad_norm": 0.5871370434761047, "learning_rate": 1.8423374426707944e-05, "loss": 0.7818, "mean_token_accuracy": 0.7930239140987396, "num_tokens": 20939730.0, "step": 434 }, { "entropy": 0.8682892918586731, "epoch": 0.6702619414483821, "grad_norm": 0.6662325859069824, "learning_rate": 1.841420509080823e-05, "loss": 0.8437, "mean_token_accuracy": 0.7802730202674866, "num_tokens": 20990194.0, "step": 435 }, { "entropy": 0.9680300951004028, "epoch": 0.6718027734976888, "grad_norm": 0.7482401728630066, "learning_rate": 1.840501146427106e-05, "loss": 0.9523, "mean_token_accuracy": 0.7476081550121307, "num_tokens": 21004702.0, "step": 436 }, { "entropy": 0.9813300669193268, "epoch": 0.6733436055469953, "grad_norm": 0.7837386131286621, "learning_rate": 1.839579357363716e-05, "loss": 0.9636, "mean_token_accuracy": 0.7548449635505676, "num_tokens": 21014245.0, "step": 437 }, { "entropy": 0.7060564160346985, "epoch": 0.674884437596302, "grad_norm": 0.7585479617118835, "learning_rate": 1.838655144551729e-05, "loss": 0.5799, "mean_token_accuracy": 0.8305248618125916, "num_tokens": 21050635.0, "step": 438 }, { "entropy": 0.7064327299594879, "epoch": 0.6764252696456087, "grad_norm": 0.5556596517562866, "learning_rate": 1.8377285106592185e-05, "loss": 0.6714, "mean_token_accuracy": 0.8200050294399261, "num_tokens": 21114890.0, "step": 439 }, { "entropy": 0.8079176545143127, "epoch": 0.6779661016949152, "grad_norm": 0.7182942628860474, "learning_rate": 1.8367994583612475e-05, "loss": 0.7853, "mean_token_accuracy": 0.7897604703903198, "num_tokens": 21176276.0, "step": 440 }, { "entropy": 0.8258574903011322, "epoch": 0.6795069337442219, "grad_norm": 0.5770600438117981, "learning_rate": 1.8358679903398602e-05, "loss": 0.8035, "mean_token_accuracy": 0.7848893404006958, "num_tokens": 21235910.0, "step": 441 }, { "entropy": 0.7989886701107025, "epoch": 0.6810477657935285, "grad_norm": 0.5379884243011475, "learning_rate": 1.8349341092840757e-05, "loss": 0.7927, "mean_token_accuracy": 0.7887792587280273, "num_tokens": 21294239.0, "step": 442 }, { "entropy": 0.8333309888839722, "epoch": 0.6825885978428351, "grad_norm": 0.5697650909423828, "learning_rate": 1.833997817889878e-05, "loss": 0.8263, "mean_token_accuracy": 0.781194806098938, "num_tokens": 21351439.0, "step": 443 }, { "entropy": 0.8270282447338104, "epoch": 0.6841294298921418, "grad_norm": 0.5892913341522217, "learning_rate": 1.83305911886021e-05, "loss": 0.8364, "mean_token_accuracy": 0.7788231670856476, "num_tokens": 21407607.0, "step": 444 }, { "entropy": 0.8070065379142761, "epoch": 0.6856702619414484, "grad_norm": 0.6114411950111389, "learning_rate": 1.832118014904965e-05, "loss": 0.8141, "mean_token_accuracy": 0.7864700853824615, "num_tokens": 21462587.0, "step": 445 }, { "entropy": 0.8420366048812866, "epoch": 0.687211093990755, "grad_norm": 0.6106347441673279, "learning_rate": 1.8311745087409785e-05, "loss": 0.8241, "mean_token_accuracy": 0.78264319896698, "num_tokens": 21515966.0, "step": 446 }, { "entropy": 0.8704203963279724, "epoch": 0.6887519260400616, "grad_norm": 0.6348656415939331, "learning_rate": 1.8302286030920217e-05, "loss": 0.8465, "mean_token_accuracy": 0.7795785367488861, "num_tokens": 21567511.0, "step": 447 }, { "entropy": 0.9107683897018433, "epoch": 0.6902927580893683, "grad_norm": 0.7761650681495667, "learning_rate": 1.8292803006887923e-05, "loss": 0.9195, "mean_token_accuracy": 0.7585366070270538, "num_tokens": 21586118.0, "step": 448 }, { "entropy": 0.9850309491157532, "epoch": 0.6918335901386748, "grad_norm": 0.7343606352806091, "learning_rate": 1.8283296042689075e-05, "loss": 0.9963, "mean_token_accuracy": 0.7452509999275208, "num_tokens": 21596203.0, "step": 449 }, { "entropy": 0.845465362071991, "epoch": 0.6933744221879815, "grad_norm": 0.7953429222106934, "learning_rate": 1.8273765165768954e-05, "loss": 0.8246, "mean_token_accuracy": 0.7946197092533112, "num_tokens": 21604026.0, "step": 450 }, { "entropy": 0.6899242997169495, "epoch": 0.6949152542372882, "grad_norm": 0.7047879099845886, "learning_rate": 1.8264210403641866e-05, "loss": 0.6512, "mean_token_accuracy": 0.8257539868354797, "num_tokens": 21669113.0, "step": 451 }, { "entropy": 0.7450542449951172, "epoch": 0.6964560862865947, "grad_norm": 0.7083922624588013, "learning_rate": 1.825463178389108e-05, "loss": 0.7155, "mean_token_accuracy": 0.8063499331474304, "num_tokens": 21731438.0, "step": 452 }, { "entropy": 0.8329744338989258, "epoch": 0.6979969183359014, "grad_norm": 0.5745351910591125, "learning_rate": 1.8245029334168743e-05, "loss": 0.794, "mean_token_accuracy": 0.7891736626625061, "num_tokens": 21791893.0, "step": 453 }, { "entropy": 0.8555083870887756, "epoch": 0.699537750385208, "grad_norm": 0.5600628852844238, "learning_rate": 1.8235403082195782e-05, "loss": 0.8251, "mean_token_accuracy": 0.7778974771499634, "num_tokens": 21850878.0, "step": 454 }, { "entropy": 0.8051244616508484, "epoch": 0.7010785824345146, "grad_norm": 0.5757558345794678, "learning_rate": 1.822575305576185e-05, "loss": 0.7757, "mean_token_accuracy": 0.7921045124530792, "num_tokens": 21908750.0, "step": 455 }, { "entropy": 0.7910523414611816, "epoch": 0.7026194144838213, "grad_norm": 0.5763965249061584, "learning_rate": 1.8216079282725225e-05, "loss": 0.7756, "mean_token_accuracy": 0.7945888340473175, "num_tokens": 21965506.0, "step": 456 }, { "entropy": 0.8260907530784607, "epoch": 0.7041602465331279, "grad_norm": 0.6227362751960754, "learning_rate": 1.820638179101274e-05, "loss": 0.8452, "mean_token_accuracy": 0.7780827879905701, "num_tokens": 22021252.0, "step": 457 }, { "entropy": 0.7528596818447113, "epoch": 0.7057010785824345, "grad_norm": 0.6396445631980896, "learning_rate": 1.8196660608619704e-05, "loss": 0.7676, "mean_token_accuracy": 0.7970865070819855, "num_tokens": 22075735.0, "step": 458 }, { "entropy": 0.8271595537662506, "epoch": 0.7072419106317411, "grad_norm": 0.7084936499595642, "learning_rate": 1.818691576360982e-05, "loss": 0.8186, "mean_token_accuracy": 0.7815895974636078, "num_tokens": 22128709.0, "step": 459 }, { "entropy": 0.8314042687416077, "epoch": 0.7087827426810478, "grad_norm": 0.6556583642959595, "learning_rate": 1.8177147284115092e-05, "loss": 0.8102, "mean_token_accuracy": 0.7880943715572357, "num_tokens": 22176843.0, "step": 460 }, { "entropy": 0.982695996761322, "epoch": 0.7103235747303543, "grad_norm": 0.6853083968162537, "learning_rate": 1.8167355198335758e-05, "loss": 0.9564, "mean_token_accuracy": 0.7456813454627991, "num_tokens": 22188925.0, "step": 461 }, { "entropy": 0.9514792263507843, "epoch": 0.711864406779661, "grad_norm": 0.8346017599105835, "learning_rate": 1.8157539534540213e-05, "loss": 0.9405, "mean_token_accuracy": 0.7600831389427185, "num_tokens": 22198418.0, "step": 462 }, { "entropy": 0.7071046829223633, "epoch": 0.7134052388289677, "grad_norm": 0.8096746802330017, "learning_rate": 1.8147700321064897e-05, "loss": 0.5169, "mean_token_accuracy": 0.8306362330913544, "num_tokens": 22234828.0, "step": 463 }, { "entropy": 0.7370755672454834, "epoch": 0.7149460708782742, "grad_norm": 0.6118667125701904, "learning_rate": 1.8137837586314264e-05, "loss": 0.6871, "mean_token_accuracy": 0.8157793581485748, "num_tokens": 22299033.0, "step": 464 }, { "entropy": 0.8091652095317841, "epoch": 0.7164869029275809, "grad_norm": 0.6329115033149719, "learning_rate": 1.812795135876064e-05, "loss": 0.7714, "mean_token_accuracy": 0.7933105826377869, "num_tokens": 22360395.0, "step": 465 }, { "entropy": 0.7994483709335327, "epoch": 0.7180277349768875, "grad_norm": 0.6834350228309631, "learning_rate": 1.8118041666944202e-05, "loss": 0.7902, "mean_token_accuracy": 0.789707362651825, "num_tokens": 22420219.0, "step": 466 }, { "entropy": 0.7602736055850983, "epoch": 0.7195685670261941, "grad_norm": 0.6498889327049255, "learning_rate": 1.8108108539472833e-05, "loss": 0.7601, "mean_token_accuracy": 0.797406405210495, "num_tokens": 22478851.0, "step": 467 }, { "entropy": 0.7720863819122314, "epoch": 0.7211093990755008, "grad_norm": 0.6744052767753601, "learning_rate": 1.8098152005022097e-05, "loss": 0.7823, "mean_token_accuracy": 0.789202094078064, "num_tokens": 22536293.0, "step": 468 }, { "entropy": 0.7686461210250854, "epoch": 0.7226502311248074, "grad_norm": 0.5915248394012451, "learning_rate": 1.808817209233512e-05, "loss": 0.7656, "mean_token_accuracy": 0.7952660024166107, "num_tokens": 22592791.0, "step": 469 }, { "entropy": 0.8062160015106201, "epoch": 0.724191063174114, "grad_norm": 0.5982757806777954, "learning_rate": 1.8078168830222516e-05, "loss": 0.8111, "mean_token_accuracy": 0.7873937785625458, "num_tokens": 22648101.0, "step": 470 }, { "entropy": 0.8693538904190063, "epoch": 0.7257318952234206, "grad_norm": 0.6183062791824341, "learning_rate": 1.806814224756231e-05, "loss": 0.8523, "mean_token_accuracy": 0.7774045765399933, "num_tokens": 22702267.0, "step": 471 }, { "entropy": 0.861188530921936, "epoch": 0.7272727272727273, "grad_norm": 0.6635674238204956, "learning_rate": 1.8058092373299855e-05, "loss": 0.8454, "mean_token_accuracy": 0.7741262316703796, "num_tokens": 22754592.0, "step": 472 }, { "entropy": 0.8723970055580139, "epoch": 0.7288135593220338, "grad_norm": 0.7629942893981934, "learning_rate": 1.804801923644773e-05, "loss": 0.8246, "mean_token_accuracy": 0.7707898914813995, "num_tokens": 22787041.0, "step": 473 }, { "entropy": 0.9248522818088531, "epoch": 0.7303543913713405, "grad_norm": 0.7278069257736206, "learning_rate": 1.803792286608568e-05, "loss": 0.9439, "mean_token_accuracy": 0.7531997263431549, "num_tokens": 22798024.0, "step": 474 }, { "entropy": 0.8349332809448242, "epoch": 0.7318952234206472, "grad_norm": 0.8063945770263672, "learning_rate": 1.8027803291360515e-05, "loss": 0.8294, "mean_token_accuracy": 0.7951104938983917, "num_tokens": 22805685.0, "step": 475 }, { "entropy": 0.5646562576293945, "epoch": 0.7334360554699538, "grad_norm": 0.7194772958755493, "learning_rate": 1.801766054148604e-05, "loss": 0.5426, "mean_token_accuracy": 0.8527752757072449, "num_tokens": 22871092.0, "step": 476 }, { "entropy": 0.7421990931034088, "epoch": 0.7349768875192604, "grad_norm": 0.7087520956993103, "learning_rate": 1.8007494645742967e-05, "loss": 0.7129, "mean_token_accuracy": 0.8082769215106964, "num_tokens": 22934058.0, "step": 477 }, { "entropy": 0.7825068831443787, "epoch": 0.736517719568567, "grad_norm": 0.6618267297744751, "learning_rate": 1.799730563347881e-05, "loss": 0.7487, "mean_token_accuracy": 0.7990700900554657, "num_tokens": 22994785.0, "step": 478 }, { "entropy": 0.8546741306781769, "epoch": 0.7380585516178737, "grad_norm": 0.6195006370544434, "learning_rate": 1.798709353410783e-05, "loss": 0.8075, "mean_token_accuracy": 0.7834112048149109, "num_tokens": 23053878.0, "step": 479 }, { "entropy": 0.8485706746578217, "epoch": 0.7395993836671803, "grad_norm": 0.6405356526374817, "learning_rate": 1.797685837711094e-05, "loss": 0.8025, "mean_token_accuracy": 0.7870714366436005, "num_tokens": 23111715.0, "step": 480 }, { "entropy": 0.7568847239017487, "epoch": 0.7411402157164869, "grad_norm": 0.6282846331596375, "learning_rate": 1.7966600192035605e-05, "loss": 0.7351, "mean_token_accuracy": 0.8037236332893372, "num_tokens": 23168319.0, "step": 481 }, { "entropy": 0.7854646444320679, "epoch": 0.7426810477657936, "grad_norm": 0.6779035925865173, "learning_rate": 1.795631900849578e-05, "loss": 0.7943, "mean_token_accuracy": 0.791608065366745, "num_tokens": 23223907.0, "step": 482 }, { "entropy": 0.7525529861450195, "epoch": 0.7442218798151001, "grad_norm": 0.702568769454956, "learning_rate": 1.7946014856171805e-05, "loss": 0.761, "mean_token_accuracy": 0.8006133437156677, "num_tokens": 23278247.0, "step": 483 }, { "entropy": 0.7724007964134216, "epoch": 0.7457627118644068, "grad_norm": 0.7008646130561829, "learning_rate": 1.7935687764810337e-05, "loss": 0.7658, "mean_token_accuracy": 0.79779052734375, "num_tokens": 23331080.0, "step": 484 }, { "entropy": 0.7161285579204559, "epoch": 0.7473035439137135, "grad_norm": 0.6937347650527954, "learning_rate": 1.7925337764224246e-05, "loss": 0.6884, "mean_token_accuracy": 0.820366233587265, "num_tokens": 23379886.0, "step": 485 }, { "entropy": 0.9700843095779419, "epoch": 0.74884437596302, "grad_norm": 0.6991108655929565, "learning_rate": 1.7914964884292543e-05, "loss": 0.9768, "mean_token_accuracy": 0.739799290895462, "num_tokens": 23392288.0, "step": 486 }, { "entropy": 0.9146089255809784, "epoch": 0.7503852080123267, "grad_norm": 0.7797329425811768, "learning_rate": 1.7904569154960287e-05, "loss": 0.9, "mean_token_accuracy": 0.7767994999885559, "num_tokens": 23401498.0, "step": 487 }, { "entropy": 0.6269094944000244, "epoch": 0.7519260400616333, "grad_norm": 0.8323846459388733, "learning_rate": 1.7894150606238498e-05, "loss": 0.4585, "mean_token_accuracy": 0.8512799441814423, "num_tokens": 23437708.0, "step": 488 }, { "entropy": 0.5978611707687378, "epoch": 0.7534668721109399, "grad_norm": 0.6352093815803528, "learning_rate": 1.7883709268204073e-05, "loss": 0.5475, "mean_token_accuracy": 0.8530621826648712, "num_tokens": 23502494.0, "step": 489 }, { "entropy": 0.7137824594974518, "epoch": 0.7550077041602465, "grad_norm": 0.7269333600997925, "learning_rate": 1.78732451709997e-05, "loss": 0.6963, "mean_token_accuracy": 0.8102832436561584, "num_tokens": 23564574.0, "step": 490 }, { "entropy": 0.6773908138275146, "epoch": 0.7565485362095532, "grad_norm": 0.7038426399230957, "learning_rate": 1.7862758344833773e-05, "loss": 0.6724, "mean_token_accuracy": 0.8185820877552032, "num_tokens": 23624772.0, "step": 491 }, { "entropy": 0.7352079749107361, "epoch": 0.7580893682588598, "grad_norm": 0.746961236000061, "learning_rate": 1.7852248819980297e-05, "loss": 0.7416, "mean_token_accuracy": 0.7989521026611328, "num_tokens": 23683404.0, "step": 492 }, { "entropy": 0.808477133512497, "epoch": 0.7596302003081664, "grad_norm": 0.6438852548599243, "learning_rate": 1.7841716626778803e-05, "loss": 0.7943, "mean_token_accuracy": 0.7888232171535492, "num_tokens": 23740830.0, "step": 493 }, { "entropy": 0.787971556186676, "epoch": 0.7611710323574731, "grad_norm": 0.663276731967926, "learning_rate": 1.783116179563427e-05, "loss": 0.7586, "mean_token_accuracy": 0.79994997382164, "num_tokens": 23797200.0, "step": 494 }, { "entropy": 0.7890351414680481, "epoch": 0.7627118644067796, "grad_norm": 0.7261046767234802, "learning_rate": 1.7820584357017018e-05, "loss": 0.7573, "mean_token_accuracy": 0.7964493036270142, "num_tokens": 23852447.0, "step": 495 }, { "entropy": 0.8015623688697815, "epoch": 0.7642526964560863, "grad_norm": 0.653415322303772, "learning_rate": 1.7809984341462637e-05, "loss": 0.7883, "mean_token_accuracy": 0.7900203168392181, "num_tokens": 23906059.0, "step": 496 }, { "entropy": 0.7217371761798859, "epoch": 0.765793528505393, "grad_norm": 0.7963257431983948, "learning_rate": 1.7799361779571904e-05, "loss": 0.7148, "mean_token_accuracy": 0.8100500404834747, "num_tokens": 23957985.0, "step": 497 }, { "entropy": 0.8242279291152954, "epoch": 0.7673343605546995, "grad_norm": 0.8738524317741394, "learning_rate": 1.7788716702010666e-05, "loss": 0.7565, "mean_token_accuracy": 0.7780761122703552, "num_tokens": 23987583.0, "step": 498 }, { "entropy": 0.9009746313095093, "epoch": 0.7688751926040062, "grad_norm": 0.7654175758361816, "learning_rate": 1.7778049139509785e-05, "loss": 0.9183, "mean_token_accuracy": 0.760333776473999, "num_tokens": 23997886.0, "step": 499 }, { "entropy": 0.9009602069854736, "epoch": 0.7704160246533128, "grad_norm": 0.8123462200164795, "learning_rate": 1.776735912286502e-05, "loss": 0.89, "mean_token_accuracy": 0.773427426815033, "num_tokens": 24005773.0, "step": 500 }, { "entropy": 0.5591139495372772, "epoch": 0.7719568567026194, "grad_norm": 0.6246494054794312, "learning_rate": 1.7756646682936964e-05, "loss": 0.5206, "mean_token_accuracy": 0.8593753576278687, "num_tokens": 24071037.0, "step": 501 }, { "entropy": 0.7369069159030914, "epoch": 0.773497688751926, "grad_norm": 0.6613090634346008, "learning_rate": 1.7745911850650942e-05, "loss": 0.699, "mean_token_accuracy": 0.8115769624710083, "num_tokens": 24133520.0, "step": 502 }, { "entropy": 0.7538451552391052, "epoch": 0.7750385208012327, "grad_norm": 0.6689869165420532, "learning_rate": 1.7735154656996908e-05, "loss": 0.707, "mean_token_accuracy": 0.8136107921600342, "num_tokens": 24194169.0, "step": 503 }, { "entropy": 0.7538253664970398, "epoch": 0.7765793528505393, "grad_norm": 0.6423341035842896, "learning_rate": 1.7724375133029388e-05, "loss": 0.7271, "mean_token_accuracy": 0.8060763478279114, "num_tokens": 24253227.0, "step": 504 }, { "entropy": 0.7402171790599823, "epoch": 0.7781201848998459, "grad_norm": 0.6659181118011475, "learning_rate": 1.7713573309867365e-05, "loss": 0.7244, "mean_token_accuracy": 0.8078895509243011, "num_tokens": 24310885.0, "step": 505 }, { "entropy": 0.7854810357093811, "epoch": 0.7796610169491526, "grad_norm": 0.6640326380729675, "learning_rate": 1.7702749218694196e-05, "loss": 0.7693, "mean_token_accuracy": 0.7952100038528442, "num_tokens": 24367524.0, "step": 506 }, { "entropy": 0.7721377611160278, "epoch": 0.7812018489984591, "grad_norm": 0.7011154890060425, "learning_rate": 1.7691902890757528e-05, "loss": 0.7713, "mean_token_accuracy": 0.7958495318889618, "num_tokens": 24423239.0, "step": 507 }, { "entropy": 0.7653034329414368, "epoch": 0.7827426810477658, "grad_norm": 0.6725184917449951, "learning_rate": 1.7681034357369193e-05, "loss": 0.7452, "mean_token_accuracy": 0.8026959300041199, "num_tokens": 24477890.0, "step": 508 }, { "entropy": 0.6879469156265259, "epoch": 0.7842835130970724, "grad_norm": 0.750636875629425, "learning_rate": 1.767014364990513e-05, "loss": 0.658, "mean_token_accuracy": 0.8219696581363678, "num_tokens": 24531055.0, "step": 509 }, { "entropy": 0.7235906720161438, "epoch": 0.785824345146379, "grad_norm": 0.7249462008476257, "learning_rate": 1.7659230799805298e-05, "loss": 0.7059, "mean_token_accuracy": 0.8166292905807495, "num_tokens": 24579344.0, "step": 510 }, { "entropy": 0.9760016798973083, "epoch": 0.7873651771956857, "grad_norm": 0.6822357773780823, "learning_rate": 1.7648295838573576e-05, "loss": 0.9809, "mean_token_accuracy": 0.7427650392055511, "num_tokens": 24591477.0, "step": 511 }, { "entropy": 0.9477180540561676, "epoch": 0.7889060092449923, "grad_norm": 0.7591356635093689, "learning_rate": 1.763733879777767e-05, "loss": 0.9278, "mean_token_accuracy": 0.7647373378276825, "num_tokens": 24600957.0, "step": 512 }, { "entropy": 0.6506048738956451, "epoch": 0.7904468412942989, "grad_norm": 0.8880473375320435, "learning_rate": 1.762635970904903e-05, "loss": 0.4754, "mean_token_accuracy": 0.8406797051429749, "num_tokens": 24637279.0, "step": 513 }, { "entropy": 0.7013395130634308, "epoch": 0.7919876733436055, "grad_norm": 0.7321651577949524, "learning_rate": 1.7615358604082752e-05, "loss": 0.676, "mean_token_accuracy": 0.8179242610931396, "num_tokens": 24700837.0, "step": 514 }, { "entropy": 0.6935074031352997, "epoch": 0.7935285053929122, "grad_norm": 0.6866543889045715, "learning_rate": 1.7604335514637502e-05, "loss": 0.6637, "mean_token_accuracy": 0.8201083838939667, "num_tokens": 24761743.0, "step": 515 }, { "entropy": 0.7435300052165985, "epoch": 0.7950693374422187, "grad_norm": 0.6453403234481812, "learning_rate": 1.7593290472535394e-05, "loss": 0.7184, "mean_token_accuracy": 0.8051873743534088, "num_tokens": 24821118.0, "step": 516 }, { "entropy": 0.7380551099777222, "epoch": 0.7966101694915254, "grad_norm": 0.6698354482650757, "learning_rate": 1.7582223509661932e-05, "loss": 0.7134, "mean_token_accuracy": 0.8071314990520477, "num_tokens": 24879292.0, "step": 517 }, { "entropy": 0.7413027882575989, "epoch": 0.7981510015408321, "grad_norm": 0.7364901900291443, "learning_rate": 1.7571134657965888e-05, "loss": 0.7299, "mean_token_accuracy": 0.8062470257282257, "num_tokens": 24936254.0, "step": 518 }, { "entropy": 0.7514287531375885, "epoch": 0.7996918335901386, "grad_norm": 0.6998208165168762, "learning_rate": 1.756002394945924e-05, "loss": 0.7412, "mean_token_accuracy": 0.8062129616737366, "num_tokens": 24992309.0, "step": 519 }, { "entropy": 0.690136194229126, "epoch": 0.8012326656394453, "grad_norm": 0.7281439304351807, "learning_rate": 1.7548891416217048e-05, "loss": 0.6737, "mean_token_accuracy": 0.8218989372253418, "num_tokens": 25047464.0, "step": 520 }, { "entropy": 0.722637414932251, "epoch": 0.802773497688752, "grad_norm": 0.7087018489837646, "learning_rate": 1.7537737090377383e-05, "loss": 0.7035, "mean_token_accuracy": 0.8149236142635345, "num_tokens": 25101641.0, "step": 521 }, { "entropy": 0.7352562546730042, "epoch": 0.8043143297380585, "grad_norm": 0.7064898610115051, "learning_rate": 1.7526561004141235e-05, "loss": 0.7166, "mean_token_accuracy": 0.8091699481010437, "num_tokens": 25154075.0, "step": 522 }, { "entropy": 0.7511390447616577, "epoch": 0.8058551617873652, "grad_norm": 0.7767520546913147, "learning_rate": 1.7515363189772406e-05, "loss": 0.6836, "mean_token_accuracy": 0.8123564124107361, "num_tokens": 25189053.0, "step": 523 }, { "entropy": 0.9335349500179291, "epoch": 0.8073959938366718, "grad_norm": 0.6873849034309387, "learning_rate": 1.7504143679597425e-05, "loss": 0.9461, "mean_token_accuracy": 0.749202162027359, "num_tokens": 25200256.0, "step": 524 }, { "entropy": 0.8755460977554321, "epoch": 0.8089368258859785, "grad_norm": 0.7882755398750305, "learning_rate": 1.7492902506005457e-05, "loss": 0.8625, "mean_token_accuracy": 0.7861902415752411, "num_tokens": 25208586.0, "step": 525 }, { "entropy": 0.5491880625486374, "epoch": 0.810477657935285, "grad_norm": 0.6882697343826294, "learning_rate": 1.7481639701448196e-05, "loss": 0.49, "mean_token_accuracy": 0.8685620725154877, "num_tokens": 25273808.0, "step": 526 }, { "entropy": 0.6520794630050659, "epoch": 0.8120184899845917, "grad_norm": 0.7281544208526611, "learning_rate": 1.7470355298439795e-05, "loss": 0.6195, "mean_token_accuracy": 0.8309482634067535, "num_tokens": 25335991.0, "step": 527 }, { "entropy": 0.7125169038772583, "epoch": 0.8135593220338984, "grad_norm": 0.693480372428894, "learning_rate": 1.745904932955675e-05, "loss": 0.6989, "mean_token_accuracy": 0.811470240354538, "num_tokens": 25396320.0, "step": 528 }, { "entropy": 0.6760894060134888, "epoch": 0.8151001540832049, "grad_norm": 0.776375949382782, "learning_rate": 1.744772182743782e-05, "loss": 0.6601, "mean_token_accuracy": 0.819790780544281, "num_tokens": 25455335.0, "step": 529 }, { "entropy": 0.6996972262859344, "epoch": 0.8166409861325116, "grad_norm": 0.6720384955406189, "learning_rate": 1.7436372824783924e-05, "loss": 0.6745, "mean_token_accuracy": 0.8199342787265778, "num_tokens": 25513025.0, "step": 530 }, { "entropy": 0.696536511182785, "epoch": 0.8181818181818182, "grad_norm": 0.6708771586418152, "learning_rate": 1.742500235435805e-05, "loss": 0.664, "mean_token_accuracy": 0.8238098919391632, "num_tokens": 25569501.0, "step": 531 }, { "entropy": 0.6889917850494385, "epoch": 0.8197226502311248, "grad_norm": 0.7183478474617004, "learning_rate": 1.7413610448985165e-05, "loss": 0.6677, "mean_token_accuracy": 0.8225856721401215, "num_tokens": 25625033.0, "step": 532 }, { "entropy": 0.6948089301586151, "epoch": 0.8212634822804314, "grad_norm": 0.7366759777069092, "learning_rate": 1.740219714155211e-05, "loss": 0.6934, "mean_token_accuracy": 0.8153162598609924, "num_tokens": 25679482.0, "step": 533 }, { "entropy": 0.6685558259487152, "epoch": 0.8228043143297381, "grad_norm": 0.7622361183166504, "learning_rate": 1.739076246500751e-05, "loss": 0.6616, "mean_token_accuracy": 0.8251914381980896, "num_tokens": 25732237.0, "step": 534 }, { "entropy": 0.6849299371242523, "epoch": 0.8243451463790447, "grad_norm": 0.7437466979026794, "learning_rate": 1.7379306452361684e-05, "loss": 0.643, "mean_token_accuracy": 0.8277880847454071, "num_tokens": 25776466.0, "step": 535 }, { "entropy": 0.9510533511638641, "epoch": 0.8258859784283513, "grad_norm": 0.6636925339698792, "learning_rate": 1.7367829136686546e-05, "loss": 0.9665, "mean_token_accuracy": 0.7426698505878448, "num_tokens": 25788566.0, "step": 536 }, { "entropy": 0.8790594935417175, "epoch": 0.827426810477658, "grad_norm": 0.7535960674285889, "learning_rate": 1.7356330551115506e-05, "loss": 0.872, "mean_token_accuracy": 0.778728723526001, "num_tokens": 25797844.0, "step": 537 }, { "entropy": 0.6035102307796478, "epoch": 0.8289676425269645, "grad_norm": 0.8696148991584778, "learning_rate": 1.734481072884338e-05, "loss": 0.414, "mean_token_accuracy": 0.8510608375072479, "num_tokens": 25834014.0, "step": 538 }, { "entropy": 0.6078549027442932, "epoch": 0.8305084745762712, "grad_norm": 0.6585686206817627, "learning_rate": 1.7333269703126285e-05, "loss": 0.5663, "mean_token_accuracy": 0.8472914099693298, "num_tokens": 25898420.0, "step": 539 }, { "entropy": 0.7674084007740021, "epoch": 0.8320493066255779, "grad_norm": 0.7441838383674622, "learning_rate": 1.7321707507281557e-05, "loss": 0.7329, "mean_token_accuracy": 0.8020242154598236, "num_tokens": 25960276.0, "step": 540 }, { "entropy": 0.7200008034706116, "epoch": 0.8335901386748844, "grad_norm": 0.7320507168769836, "learning_rate": 1.731012417468764e-05, "loss": 0.6797, "mean_token_accuracy": 0.8153003752231598, "num_tokens": 26020307.0, "step": 541 }, { "entropy": 0.6903385818004608, "epoch": 0.8351309707241911, "grad_norm": 0.6639702916145325, "learning_rate": 1.7298519738784006e-05, "loss": 0.6635, "mean_token_accuracy": 0.8248816728591919, "num_tokens": 26078999.0, "step": 542 }, { "entropy": 0.7102160453796387, "epoch": 0.8366718027734977, "grad_norm": 0.7269941568374634, "learning_rate": 1.728689423307105e-05, "loss": 0.7084, "mean_token_accuracy": 0.8125104308128357, "num_tokens": 26136446.0, "step": 543 }, { "entropy": 0.7490622401237488, "epoch": 0.8382126348228043, "grad_norm": 0.7038019895553589, "learning_rate": 1.727524769110997e-05, "loss": 0.7412, "mean_token_accuracy": 0.8020261824131012, "num_tokens": 26192721.0, "step": 544 }, { "entropy": 0.7403735518455505, "epoch": 0.8397534668721109, "grad_norm": 0.6925029754638672, "learning_rate": 1.7263580146522718e-05, "loss": 0.7238, "mean_token_accuracy": 0.8082438409328461, "num_tokens": 26248033.0, "step": 545 }, { "entropy": 0.7227424383163452, "epoch": 0.8412942989214176, "grad_norm": 0.7426387071609497, "learning_rate": 1.7251891632991874e-05, "loss": 0.6923, "mean_token_accuracy": 0.8167887032032013, "num_tokens": 26302010.0, "step": 546 }, { "entropy": 0.7162514925003052, "epoch": 0.8428351309707242, "grad_norm": 0.777336061000824, "learning_rate": 1.7240182184260544e-05, "loss": 0.6853, "mean_token_accuracy": 0.8196493685245514, "num_tokens": 26353840.0, "step": 547 }, { "entropy": 0.790680319070816, "epoch": 0.8443759630200308, "grad_norm": 0.8462156653404236, "learning_rate": 1.7228451834132275e-05, "loss": 0.7323, "mean_token_accuracy": 0.7941605448722839, "num_tokens": 26387500.0, "step": 548 }, { "entropy": 0.9754643440246582, "epoch": 0.8459167950693375, "grad_norm": 0.7932902574539185, "learning_rate": 1.7216700616470946e-05, "loss": 0.9827, "mean_token_accuracy": 0.7436946928501129, "num_tokens": 26398450.0, "step": 549 }, { "entropy": 0.8407149612903595, "epoch": 0.847457627118644, "grad_norm": 0.8748205304145813, "learning_rate": 1.7204928565200693e-05, "loss": 0.86, "mean_token_accuracy": 0.7867599427700043, "num_tokens": 26406512.0, "step": 550 }, { "entropy": 0.49032340943813324, "epoch": 0.8489984591679507, "grad_norm": 0.644839346408844, "learning_rate": 1.719313571430579e-05, "loss": 0.4348, "mean_token_accuracy": 0.8851594924926758, "num_tokens": 26471838.0, "step": 551 }, { "entropy": 0.6103380024433136, "epoch": 0.8505392912172574, "grad_norm": 0.6948839426040649, "learning_rate": 1.7181322097830545e-05, "loss": 0.5774, "mean_token_accuracy": 0.8420035243034363, "num_tokens": 26534573.0, "step": 552 }, { "entropy": 0.6644564270973206, "epoch": 0.8520801232665639, "grad_norm": 0.7183674573898315, "learning_rate": 1.7169487749879227e-05, "loss": 0.6464, "mean_token_accuracy": 0.8260222375392914, "num_tokens": 26595109.0, "step": 553 }, { "entropy": 0.6987314224243164, "epoch": 0.8536209553158706, "grad_norm": 0.7199399471282959, "learning_rate": 1.715763270461595e-05, "loss": 0.6674, "mean_token_accuracy": 0.8216021656990051, "num_tokens": 26654343.0, "step": 554 }, { "entropy": 0.7333724498748779, "epoch": 0.8551617873651772, "grad_norm": 0.728376030921936, "learning_rate": 1.7145756996264575e-05, "loss": 0.7083, "mean_token_accuracy": 0.8135379552841187, "num_tokens": 26712477.0, "step": 555 }, { "entropy": 0.7530547082424164, "epoch": 0.8567026194144838, "grad_norm": 0.7060726284980774, "learning_rate": 1.7133860659108618e-05, "loss": 0.7185, "mean_token_accuracy": 0.8116485178470612, "num_tokens": 26769132.0, "step": 556 }, { "entropy": 0.7373248338699341, "epoch": 0.8582434514637904, "grad_norm": 0.7002667784690857, "learning_rate": 1.7121943727491147e-05, "loss": 0.7104, "mean_token_accuracy": 0.8102265000343323, "num_tokens": 26824449.0, "step": 557 }, { "entropy": 0.7134297788143158, "epoch": 0.8597842835130971, "grad_norm": 0.7338776588439941, "learning_rate": 1.711000623581468e-05, "loss": 0.6832, "mean_token_accuracy": 0.8183204829692841, "num_tokens": 26878571.0, "step": 558 }, { "entropy": 0.699675440788269, "epoch": 0.8613251155624037, "grad_norm": 0.7845198512077332, "learning_rate": 1.70980482185411e-05, "loss": 0.6791, "mean_token_accuracy": 0.8194474875926971, "num_tokens": 26931239.0, "step": 559 }, { "entropy": 0.726429671049118, "epoch": 0.8628659476117103, "grad_norm": 0.8232124447822571, "learning_rate": 1.7086069710191526e-05, "loss": 0.6805, "mean_token_accuracy": 0.8197901546955109, "num_tokens": 26974958.0, "step": 560 }, { "entropy": 0.9344125390052795, "epoch": 0.864406779661017, "grad_norm": 0.6651447415351868, "learning_rate": 1.7074070745346247e-05, "loss": 0.9419, "mean_token_accuracy": 0.7485838234424591, "num_tokens": 26987008.0, "step": 561 }, { "entropy": 0.9255008101463318, "epoch": 0.8659476117103235, "grad_norm": 0.7322109341621399, "learning_rate": 1.7062051358644605e-05, "loss": 0.9493, "mean_token_accuracy": 0.7632151544094086, "num_tokens": 26996504.0, "step": 562 }, { "entropy": 0.6246985048055649, "epoch": 0.8674884437596302, "grad_norm": 0.7603177428245544, "learning_rate": 1.705001158478489e-05, "loss": 0.3903, "mean_token_accuracy": 0.8497866094112396, "num_tokens": 27032979.0, "step": 563 }, { "entropy": 0.5345063209533691, "epoch": 0.8690292758089369, "grad_norm": 0.64854496717453, "learning_rate": 1.7037951458524255e-05, "loss": 0.4949, "mean_token_accuracy": 0.8682925403118134, "num_tokens": 27097417.0, "step": 564 }, { "entropy": 0.6441138982772827, "epoch": 0.8705701078582434, "grad_norm": 0.7372936606407166, "learning_rate": 1.7025871014678596e-05, "loss": 0.6181, "mean_token_accuracy": 0.8319627940654755, "num_tokens": 27158719.0, "step": 565 }, { "entropy": 0.6668389737606049, "epoch": 0.8721109399075501, "grad_norm": 0.7285493016242981, "learning_rate": 1.701377028812248e-05, "loss": 0.6332, "mean_token_accuracy": 0.8282293081283569, "num_tokens": 27218522.0, "step": 566 }, { "entropy": 0.7052689492702484, "epoch": 0.8736517719568567, "grad_norm": 0.7078289985656738, "learning_rate": 1.7001649313789014e-05, "loss": 0.693, "mean_token_accuracy": 0.8162044286727905, "num_tokens": 27276878.0, "step": 567 }, { "entropy": 0.6681920289993286, "epoch": 0.8751926040061633, "grad_norm": 0.6736162900924683, "learning_rate": 1.698950812666976e-05, "loss": 0.6353, "mean_token_accuracy": 0.8305443525314331, "num_tokens": 27334185.0, "step": 568 }, { "entropy": 0.6766898036003113, "epoch": 0.8767334360554699, "grad_norm": 0.7200175523757935, "learning_rate": 1.6977346761814636e-05, "loss": 0.6599, "mean_token_accuracy": 0.8269561231136322, "num_tokens": 27390418.0, "step": 569 }, { "entropy": 0.6802597641944885, "epoch": 0.8782742681047766, "grad_norm": 0.7749627828598022, "learning_rate": 1.696516525433181e-05, "loss": 0.6581, "mean_token_accuracy": 0.8244665861129761, "num_tokens": 27445573.0, "step": 570 }, { "entropy": 0.7498743236064911, "epoch": 0.8798151001540832, "grad_norm": 0.780486524105072, "learning_rate": 1.6952963639387588e-05, "loss": 0.7233, "mean_token_accuracy": 0.8108309805393219, "num_tokens": 27499432.0, "step": 571 }, { "entropy": 0.6869001090526581, "epoch": 0.8813559322033898, "grad_norm": 0.762568473815918, "learning_rate": 1.6940741952206342e-05, "loss": 0.6615, "mean_token_accuracy": 0.8296073079109192, "num_tokens": 27551319.0, "step": 572 }, { "entropy": 0.8140328526496887, "epoch": 0.8828967642526965, "grad_norm": 0.8863934874534607, "learning_rate": 1.692850022807037e-05, "loss": 0.6692, "mean_token_accuracy": 0.7926450073719025, "num_tokens": 27581556.0, "step": 573 }, { "entropy": 1.0258130431175232, "epoch": 0.884437596302003, "grad_norm": 0.677437424659729, "learning_rate": 1.691623850231983e-05, "loss": 1.0122, "mean_token_accuracy": 0.7408553957939148, "num_tokens": 27592188.0, "step": 574 }, { "entropy": 0.8583033084869385, "epoch": 0.8859784283513097, "grad_norm": 0.7575722932815552, "learning_rate": 1.690395681035261e-05, "loss": 0.8547, "mean_token_accuracy": 0.791776716709137, "num_tokens": 27600092.0, "step": 575 }, { "entropy": 0.5025186091661453, "epoch": 0.8875192604006163, "grad_norm": 0.6801319718360901, "learning_rate": 1.6891655187624255e-05, "loss": 0.4789, "mean_token_accuracy": 0.8720715939998627, "num_tokens": 27665439.0, "step": 576 }, { "entropy": 0.6525769829750061, "epoch": 0.889060092449923, "grad_norm": 0.6978858709335327, "learning_rate": 1.687933366964782e-05, "loss": 0.6299, "mean_token_accuracy": 0.8293664157390594, "num_tokens": 27728022.0, "step": 577 }, { "entropy": 0.6260074973106384, "epoch": 0.8906009244992296, "grad_norm": 0.7134965658187866, "learning_rate": 1.686699229199382e-05, "loss": 0.5968, "mean_token_accuracy": 0.8384524583816528, "num_tokens": 27788488.0, "step": 578 }, { "entropy": 0.6372815668582916, "epoch": 0.8921417565485362, "grad_norm": 0.7224052548408508, "learning_rate": 1.6854631090290088e-05, "loss": 0.6071, "mean_token_accuracy": 0.8380266427993774, "num_tokens": 27847683.0, "step": 579 }, { "entropy": 0.7136280834674835, "epoch": 0.8936825885978429, "grad_norm": 0.7144384980201721, "learning_rate": 1.6842250100221693e-05, "loss": 0.6859, "mean_token_accuracy": 0.8178804516792297, "num_tokens": 27905725.0, "step": 580 }, { "entropy": 0.672081470489502, "epoch": 0.8952234206471494, "grad_norm": 0.7466240525245667, "learning_rate": 1.682984935753083e-05, "loss": 0.6524, "mean_token_accuracy": 0.8262945413589478, "num_tokens": 27962688.0, "step": 581 }, { "entropy": 0.6782833337783813, "epoch": 0.8967642526964561, "grad_norm": 0.694524347782135, "learning_rate": 1.6817428898016707e-05, "loss": 0.6603, "mean_token_accuracy": 0.8267903029918671, "num_tokens": 28018607.0, "step": 582 }, { "entropy": 0.6789102256298065, "epoch": 0.8983050847457628, "grad_norm": 0.7944988012313843, "learning_rate": 1.6804988757535463e-05, "loss": 0.647, "mean_token_accuracy": 0.82933709025383, "num_tokens": 28073308.0, "step": 583 }, { "entropy": 0.6067968010902405, "epoch": 0.8998459167950693, "grad_norm": 0.7370564341545105, "learning_rate": 1.6792528972000054e-05, "loss": 0.5797, "mean_token_accuracy": 0.8465811908245087, "num_tokens": 28126359.0, "step": 584 }, { "entropy": 0.6229593455791473, "epoch": 0.901386748844376, "grad_norm": 0.763624906539917, "learning_rate": 1.6780049577380143e-05, "loss": 0.6079, "mean_token_accuracy": 0.8382472097873688, "num_tokens": 28177523.0, "step": 585 }, { "entropy": 0.7953560650348663, "epoch": 0.9029275808936826, "grad_norm": 0.9308640956878662, "learning_rate": 1.6767550609702e-05, "loss": 0.6917, "mean_token_accuracy": 0.7996573448181152, "num_tokens": 28197007.0, "step": 586 }, { "entropy": 0.9025672078132629, "epoch": 0.9044684129429892, "grad_norm": 0.7176588773727417, "learning_rate": 1.675503210504841e-05, "loss": 0.9065, "mean_token_accuracy": 0.7694394588470459, "num_tokens": 28207097.0, "step": 587 }, { "entropy": 0.6275964975357056, "epoch": 0.9060092449922958, "grad_norm": 0.7716622948646545, "learning_rate": 1.674249409955854e-05, "loss": 0.4108, "mean_token_accuracy": 0.8464445471763611, "num_tokens": 28243776.0, "step": 588 }, { "entropy": 0.5849349349737167, "epoch": 0.9075500770416025, "grad_norm": 0.7051457166671753, "learning_rate": 1.672993662942787e-05, "loss": 0.5513, "mean_token_accuracy": 0.854127049446106, "num_tokens": 28308424.0, "step": 589 }, { "entropy": 0.5962132811546326, "epoch": 0.9090909090909091, "grad_norm": 0.7009429931640625, "learning_rate": 1.6717359730908072e-05, "loss": 0.5631, "mean_token_accuracy": 0.8466776609420776, "num_tokens": 28369952.0, "step": 590 }, { "entropy": 0.67216357588768, "epoch": 0.9106317411402157, "grad_norm": 0.7449541687965393, "learning_rate": 1.670476344030689e-05, "loss": 0.6531, "mean_token_accuracy": 0.8238305449485779, "num_tokens": 28429531.0, "step": 591 }, { "entropy": 0.6723234057426453, "epoch": 0.9121725731895224, "grad_norm": 0.7844904065132141, "learning_rate": 1.6692147793988062e-05, "loss": 0.6545, "mean_token_accuracy": 0.8221221566200256, "num_tokens": 28487988.0, "step": 592 }, { "entropy": 0.6659137904644012, "epoch": 0.9137134052388289, "grad_norm": 0.7571461200714111, "learning_rate": 1.6679512828371203e-05, "loss": 0.6337, "mean_token_accuracy": 0.8331311941146851, "num_tokens": 28545375.0, "step": 593 }, { "entropy": 0.7139880657196045, "epoch": 0.9152542372881356, "grad_norm": 0.7411863803863525, "learning_rate": 1.6666858579931702e-05, "loss": 0.6814, "mean_token_accuracy": 0.8206351101398468, "num_tokens": 28601706.0, "step": 594 }, { "entropy": 0.6355736255645752, "epoch": 0.9167950693374423, "grad_norm": 0.8011767864227295, "learning_rate": 1.6654185085200597e-05, "loss": 0.6078, "mean_token_accuracy": 0.839038759469986, "num_tokens": 28657023.0, "step": 595 }, { "entropy": 0.6448324918746948, "epoch": 0.9183359013867488, "grad_norm": 0.7486951351165771, "learning_rate": 1.6641492380764516e-05, "loss": 0.6333, "mean_token_accuracy": 0.8344825506210327, "num_tokens": 28711036.0, "step": 596 }, { "entropy": 0.6731320023536682, "epoch": 0.9198767334360555, "grad_norm": 0.8220444917678833, "learning_rate": 1.662878050326552e-05, "loss": 0.6481, "mean_token_accuracy": 0.8276066482067108, "num_tokens": 28762913.0, "step": 597 }, { "entropy": 0.8168574869632721, "epoch": 0.9214175654853621, "grad_norm": 0.8065090179443359, "learning_rate": 1.6616049489401026e-05, "loss": 0.6977, "mean_token_accuracy": 0.7923493683338165, "num_tokens": 28787100.0, "step": 598 }, { "entropy": 0.9794042408466339, "epoch": 0.9229583975346687, "grad_norm": 0.7213971018791199, "learning_rate": 1.6603299375923706e-05, "loss": 0.9912, "mean_token_accuracy": 0.738404393196106, "num_tokens": 28797543.0, "step": 599 }, { "entropy": 0.8631232976913452, "epoch": 0.9244992295839753, "grad_norm": 0.8058370351791382, "learning_rate": 1.659053019964135e-05, "loss": 0.8389, "mean_token_accuracy": 0.787352055311203, "num_tokens": 28805338.0, "step": 600 }, { "epoch": 0.9244992295839753, "eval_entropy": 0.698361540880968, "eval_loss": 0.6631878614425659, "eval_mean_token_accuracy": 0.8249830879130453, "eval_num_tokens": 28805338.0, "eval_runtime": 193.9971, "eval_samples_per_second": 4.371, "eval_steps_per_second": 1.093, "step": 600 }, { "entropy": 0.44869251549243927, "epoch": 0.926040061633282, "grad_norm": 0.698883593082428, "learning_rate": 1.6577741997416797e-05, "loss": 0.4226, "mean_token_accuracy": 0.8867907226085663, "num_tokens": 28870525.0, "step": 601 }, { "entropy": 0.5745980441570282, "epoch": 0.9275808936825886, "grad_norm": 0.71623694896698, "learning_rate": 1.6564934806167806e-05, "loss": 0.5481, "mean_token_accuracy": 0.8494601249694824, "num_tokens": 28932719.0, "step": 602 }, { "entropy": 0.6373375952243805, "epoch": 0.9291217257318952, "grad_norm": 0.7258870005607605, "learning_rate": 1.6552108662866943e-05, "loss": 0.6091, "mean_token_accuracy": 0.8373463451862335, "num_tokens": 28993314.0, "step": 603 }, { "entropy": 0.654396116733551, "epoch": 0.9306625577812019, "grad_norm": 0.7040342092514038, "learning_rate": 1.653926360454151e-05, "loss": 0.6339, "mean_token_accuracy": 0.8326483070850372, "num_tokens": 29052545.0, "step": 604 }, { "entropy": 0.6312633156776428, "epoch": 0.9322033898305084, "grad_norm": 0.7308303713798523, "learning_rate": 1.6526399668273388e-05, "loss": 0.6018, "mean_token_accuracy": 0.8398293554782867, "num_tokens": 29110537.0, "step": 605 }, { "entropy": 0.6911543011665344, "epoch": 0.9337442218798151, "grad_norm": 0.76470547914505, "learning_rate": 1.6513516891198974e-05, "loss": 0.6672, "mean_token_accuracy": 0.8256165385246277, "num_tokens": 29167407.0, "step": 606 }, { "entropy": 0.6972342133522034, "epoch": 0.9352850539291218, "grad_norm": 0.7460659742355347, "learning_rate": 1.650061531050905e-05, "loss": 0.668, "mean_token_accuracy": 0.8260082006454468, "num_tokens": 29223069.0, "step": 607 }, { "entropy": 0.6684848368167877, "epoch": 0.9368258859784283, "grad_norm": 0.8038926720619202, "learning_rate": 1.6487694963448683e-05, "loss": 0.6345, "mean_token_accuracy": 0.833026796579361, "num_tokens": 29277544.0, "step": 608 }, { "entropy": 0.6131243705749512, "epoch": 0.938366718027735, "grad_norm": 0.7687508463859558, "learning_rate": 1.647475588731711e-05, "loss": 0.5828, "mean_token_accuracy": 0.8472865223884583, "num_tokens": 29330579.0, "step": 609 }, { "entropy": 0.6474324762821198, "epoch": 0.9399075500770416, "grad_norm": 0.8162680864334106, "learning_rate": 1.6461798119467648e-05, "loss": 0.6197, "mean_token_accuracy": 0.8395942151546478, "num_tokens": 29374536.0, "step": 610 }, { "entropy": 0.9134557545185089, "epoch": 0.9414483821263482, "grad_norm": 0.6496542692184448, "learning_rate": 1.6448821697307556e-05, "loss": 0.9446, "mean_token_accuracy": 0.7478476464748383, "num_tokens": 29386520.0, "step": 611 }, { "entropy": 0.8894510269165039, "epoch": 0.9429892141756548, "grad_norm": 0.8307142853736877, "learning_rate": 1.6435826658297964e-05, "loss": 0.9102, "mean_token_accuracy": 0.7711392343044281, "num_tokens": 29395606.0, "step": 612 }, { "entropy": 0.5839551389217377, "epoch": 0.9445300462249615, "grad_norm": 0.7365360856056213, "learning_rate": 1.6422813039953735e-05, "loss": 0.3818, "mean_token_accuracy": 0.8538520038127899, "num_tokens": 29431774.0, "step": 613 }, { "entropy": 0.5446498394012451, "epoch": 0.9460708782742681, "grad_norm": 0.683037519454956, "learning_rate": 1.6409780879843377e-05, "loss": 0.5289, "mean_token_accuracy": 0.8602617383003235, "num_tokens": 29495807.0, "step": 614 }, { "entropy": 0.5717845559120178, "epoch": 0.9476117103235747, "grad_norm": 0.7217373251914978, "learning_rate": 1.6396730215588913e-05, "loss": 0.5607, "mean_token_accuracy": 0.8498973250389099, "num_tokens": 29557165.0, "step": 615 }, { "entropy": 0.6545683145523071, "epoch": 0.9491525423728814, "grad_norm": 0.7049498558044434, "learning_rate": 1.63836610848658e-05, "loss": 0.6265, "mean_token_accuracy": 0.8321782350540161, "num_tokens": 29616960.0, "step": 616 }, { "entropy": 0.688260555267334, "epoch": 0.9506933744221879, "grad_norm": 0.7355931997299194, "learning_rate": 1.6370573525402785e-05, "loss": 0.6514, "mean_token_accuracy": 0.8257284760475159, "num_tokens": 29675374.0, "step": 617 }, { "entropy": 0.6752356588840485, "epoch": 0.9522342064714946, "grad_norm": 0.7711451053619385, "learning_rate": 1.635746757498184e-05, "loss": 0.6238, "mean_token_accuracy": 0.8350069224834442, "num_tokens": 29732618.0, "step": 618 }, { "entropy": 0.6314889192581177, "epoch": 0.9537750385208013, "grad_norm": 0.7626258730888367, "learning_rate": 1.6344343271438013e-05, "loss": 0.5897, "mean_token_accuracy": 0.8429074585437775, "num_tokens": 29788853.0, "step": 619 }, { "entropy": 0.6534005403518677, "epoch": 0.9553158705701078, "grad_norm": 0.7811862826347351, "learning_rate": 1.6331200652659337e-05, "loss": 0.6272, "mean_token_accuracy": 0.83578160405159, "num_tokens": 29843941.0, "step": 620 }, { "entropy": 0.6513084769248962, "epoch": 0.9568567026194145, "grad_norm": 0.7964636087417603, "learning_rate": 1.6318039756586724e-05, "loss": 0.6302, "mean_token_accuracy": 0.8340610563755035, "num_tokens": 29897749.0, "step": 621 }, { "entropy": 0.59810009598732, "epoch": 0.9583975346687211, "grad_norm": 0.8360713720321655, "learning_rate": 1.6304860621213846e-05, "loss": 0.5775, "mean_token_accuracy": 0.8483818173408508, "num_tokens": 29949669.0, "step": 622 }, { "entropy": 0.7558655440807343, "epoch": 0.9599383667180277, "grad_norm": 0.8585265278816223, "learning_rate": 1.6291663284587028e-05, "loss": 0.6282, "mean_token_accuracy": 0.8062027394771576, "num_tokens": 29980757.0, "step": 623 }, { "entropy": 0.9364363551139832, "epoch": 0.9614791987673343, "grad_norm": 0.7263184785842896, "learning_rate": 1.627844778480514e-05, "loss": 0.9292, "mean_token_accuracy": 0.7614871561527252, "num_tokens": 29991291.0, "step": 624 }, { "entropy": 0.8774539828300476, "epoch": 0.963020030816641, "grad_norm": 0.8621702790260315, "learning_rate": 1.62652141600195e-05, "loss": 0.8941, "mean_token_accuracy": 0.7821542620658875, "num_tokens": 29999090.0, "step": 625 }, { "entropy": 0.41135573387145996, "epoch": 0.9645608628659477, "grad_norm": 0.6784238815307617, "learning_rate": 1.6251962448433722e-05, "loss": 0.3615, "mean_token_accuracy": 0.9026855826377869, "num_tokens": 30064541.0, "step": 626 }, { "entropy": 0.5890232920646667, "epoch": 0.9661016949152542, "grad_norm": 0.7306105494499207, "learning_rate": 1.6238692688303664e-05, "loss": 0.5652, "mean_token_accuracy": 0.8460389077663422, "num_tokens": 30127478.0, "step": 627 }, { "entropy": 0.5777334868907928, "epoch": 0.9676425269645609, "grad_norm": 0.7378655672073364, "learning_rate": 1.6225404917937265e-05, "loss": 0.5609, "mean_token_accuracy": 0.8478977680206299, "num_tokens": 30188192.0, "step": 628 }, { "entropy": 0.6169094145298004, "epoch": 0.9691833590138675, "grad_norm": 0.7236188054084778, "learning_rate": 1.621209917569447e-05, "loss": 0.6083, "mean_token_accuracy": 0.837787002325058, "num_tokens": 30247415.0, "step": 629 }, { "entropy": 0.6121735572814941, "epoch": 0.9707241910631741, "grad_norm": 0.7361787557601929, "learning_rate": 1.6198775499987106e-05, "loss": 0.5862, "mean_token_accuracy": 0.8431510031223297, "num_tokens": 30305286.0, "step": 630 }, { "entropy": 0.618719607591629, "epoch": 0.9722650231124808, "grad_norm": 0.7578607797622681, "learning_rate": 1.618543392927876e-05, "loss": 0.5899, "mean_token_accuracy": 0.843658059835434, "num_tokens": 30361953.0, "step": 631 }, { "entropy": 0.60792475938797, "epoch": 0.9738058551617874, "grad_norm": 0.8163373470306396, "learning_rate": 1.6172074502084695e-05, "loss": 0.5915, "mean_token_accuracy": 0.8434663116931915, "num_tokens": 30417462.0, "step": 632 }, { "entropy": 0.6062297523021698, "epoch": 0.975346687211094, "grad_norm": 0.7789714932441711, "learning_rate": 1.615869725697171e-05, "loss": 0.5831, "mean_token_accuracy": 0.8471483290195465, "num_tokens": 30471810.0, "step": 633 }, { "entropy": 0.5688799917697906, "epoch": 0.9768875192604006, "grad_norm": 0.8991982340812683, "learning_rate": 1.6145302232558047e-05, "loss": 0.5423, "mean_token_accuracy": 0.8569185435771942, "num_tokens": 30524622.0, "step": 634 }, { "entropy": 0.595861554145813, "epoch": 0.9784283513097073, "grad_norm": 0.831292986869812, "learning_rate": 1.6131889467513283e-05, "loss": 0.5319, "mean_token_accuracy": 0.8578867018222809, "num_tokens": 30568005.0, "step": 635 }, { "entropy": 0.9480359852313995, "epoch": 0.9799691833590138, "grad_norm": 0.6692772507667542, "learning_rate": 1.611845900055819e-05, "loss": 0.9501, "mean_token_accuracy": 0.7472974061965942, "num_tokens": 30579950.0, "step": 636 }, { "entropy": 0.9357363879680634, "epoch": 0.9815100154083205, "grad_norm": 0.8290296196937561, "learning_rate": 1.610501087046466e-05, "loss": 0.9322, "mean_token_accuracy": 0.7675319910049438, "num_tokens": 30589389.0, "step": 637 }, { "entropy": 0.6061297953128815, "epoch": 0.9830508474576272, "grad_norm": 0.7980325818061829, "learning_rate": 1.6091545116055572e-05, "loss": 0.3882, "mean_token_accuracy": 0.8537670373916626, "num_tokens": 30625727.0, "step": 638 }, { "entropy": 0.5440765023231506, "epoch": 0.9845916795069337, "grad_norm": 0.6868157386779785, "learning_rate": 1.607806177620468e-05, "loss": 0.5066, "mean_token_accuracy": 0.8655721247196198, "num_tokens": 30689315.0, "step": 639 }, { "entropy": 0.5708615779876709, "epoch": 0.9861325115562404, "grad_norm": 0.7537672519683838, "learning_rate": 1.606456088983651e-05, "loss": 0.5445, "mean_token_accuracy": 0.8516312837600708, "num_tokens": 30749926.0, "step": 640 }, { "entropy": 0.6079775094985962, "epoch": 0.987673343605547, "grad_norm": 0.7927120327949524, "learning_rate": 1.6051042495926235e-05, "loss": 0.5812, "mean_token_accuracy": 0.8408438265323639, "num_tokens": 30808926.0, "step": 641 }, { "entropy": 0.6239073872566223, "epoch": 0.9892141756548536, "grad_norm": 0.7729496955871582, "learning_rate": 1.6037506633499573e-05, "loss": 0.5981, "mean_token_accuracy": 0.8412908017635345, "num_tokens": 30866565.0, "step": 642 }, { "entropy": 0.6447944939136505, "epoch": 0.9907550077041603, "grad_norm": 0.7715206146240234, "learning_rate": 1.602395334163268e-05, "loss": 0.6219, "mean_token_accuracy": 0.8359026610851288, "num_tokens": 30923134.0, "step": 643 }, { "entropy": 0.6211671233177185, "epoch": 0.9922958397534669, "grad_norm": 0.7731960415840149, "learning_rate": 1.601038265945201e-05, "loss": 0.5962, "mean_token_accuracy": 0.8442643880844116, "num_tokens": 30978416.0, "step": 644 }, { "entropy": 0.6186486780643463, "epoch": 0.9938366718027735, "grad_norm": 0.7861031889915466, "learning_rate": 1.5996794626134245e-05, "loss": 0.5772, "mean_token_accuracy": 0.8488798439502716, "num_tokens": 31032274.0, "step": 645 }, { "entropy": 0.651924192905426, "epoch": 0.9953775038520801, "grad_norm": 0.8575247526168823, "learning_rate": 1.5983189280906134e-05, "loss": 0.6268, "mean_token_accuracy": 0.8363270461559296, "num_tokens": 31083523.0, "step": 646 }, { "entropy": 0.8319452106952667, "epoch": 0.9969183359013868, "grad_norm": 0.9899511933326721, "learning_rate": 1.5969566663044412e-05, "loss": 0.6979, "mean_token_accuracy": 0.7964890599250793, "num_tokens": 31102834.0, "step": 647 }, { "entropy": 0.9037694334983826, "epoch": 0.9984591679506933, "grad_norm": 0.7732399106025696, "learning_rate": 1.595592681187568e-05, "loss": 0.8672, "mean_token_accuracy": 0.7712464332580566, "num_tokens": 31112729.0, "step": 648 }, { "entropy": 0.5949493050575256, "epoch": 1.0, "grad_norm": 0.9259129762649536, "learning_rate": 1.594226976677629e-05, "loss": 0.3855, "mean_token_accuracy": 0.8546124696731567, "num_tokens": 31129167.0, "step": 649 }, { "entropy": 0.38511183857917786, "epoch": 1.0015408320493067, "grad_norm": 0.7485491037368774, "learning_rate": 1.592859556717223e-05, "loss": 0.365, "mean_token_accuracy": 0.9024272561073303, "num_tokens": 31194542.0, "step": 650 }, { "entropy": 0.5483366847038269, "epoch": 1.0030816640986133, "grad_norm": 0.7529429197311401, "learning_rate": 1.5914904252539e-05, "loss": 0.5298, "mean_token_accuracy": 0.8576961159706116, "num_tokens": 31257393.0, "step": 651 }, { "entropy": 0.5589179992675781, "epoch": 1.00462249614792, "grad_norm": 0.8143649101257324, "learning_rate": 1.5901195862401522e-05, "loss": 0.5282, "mean_token_accuracy": 0.8583163022994995, "num_tokens": 31318323.0, "step": 652 }, { "entropy": 0.5900787711143494, "epoch": 1.0061633281972264, "grad_norm": 0.7329776287078857, "learning_rate": 1.5887470436334008e-05, "loss": 0.5459, "mean_token_accuracy": 0.8500621318817139, "num_tokens": 31377462.0, "step": 653 }, { "entropy": 0.586925745010376, "epoch": 1.007704160246533, "grad_norm": 0.745000958442688, "learning_rate": 1.5873728013959852e-05, "loss": 0.5465, "mean_token_accuracy": 0.8534631133079529, "num_tokens": 31435381.0, "step": 654 }, { "entropy": 0.5842992067337036, "epoch": 1.0092449922958397, "grad_norm": 0.7797932028770447, "learning_rate": 1.585996863495151e-05, "loss": 0.5583, "mean_token_accuracy": 0.8501728177070618, "num_tokens": 31492172.0, "step": 655 }, { "entropy": 0.5318429470062256, "epoch": 1.0107858243451464, "grad_norm": 0.7459152936935425, "learning_rate": 1.5846192339030388e-05, "loss": 0.4935, "mean_token_accuracy": 0.867554783821106, "num_tokens": 31547941.0, "step": 656 }, { "entropy": 0.5382826328277588, "epoch": 1.012326656394453, "grad_norm": 0.8312380909919739, "learning_rate": 1.583239916596674e-05, "loss": 0.5165, "mean_token_accuracy": 0.8626798093318939, "num_tokens": 31602476.0, "step": 657 }, { "entropy": 0.5388518571853638, "epoch": 1.0138674884437597, "grad_norm": 0.8041343688964844, "learning_rate": 1.581858915557953e-05, "loss": 0.5099, "mean_token_accuracy": 0.865503579378128, "num_tokens": 31655415.0, "step": 658 }, { "entropy": 0.47057364881038666, "epoch": 1.0154083204930662, "grad_norm": 0.7942938208580017, "learning_rate": 1.5804762347736333e-05, "loss": 0.4475, "mean_token_accuracy": 0.8850530982017517, "num_tokens": 31705541.0, "step": 659 }, { "entropy": 0.8692333698272705, "epoch": 1.0169491525423728, "grad_norm": 0.6918618083000183, "learning_rate": 1.5790918782353217e-05, "loss": 0.8365, "mean_token_accuracy": 0.7646225392818451, "num_tokens": 31717810.0, "step": 660 }, { "entropy": 0.8948358297348022, "epoch": 1.0184899845916795, "grad_norm": 0.7985566258430481, "learning_rate": 1.5777058499394627e-05, "loss": 0.8584, "mean_token_accuracy": 0.7732578217983246, "num_tokens": 31727221.0, "step": 661 }, { "entropy": 0.548029214143753, "epoch": 1.0200308166409862, "grad_norm": 0.8122413158416748, "learning_rate": 1.5763181538873267e-05, "loss": 0.3372, "mean_token_accuracy": 0.8704034388065338, "num_tokens": 31763555.0, "step": 662 }, { "entropy": 0.4705689549446106, "epoch": 1.0215716486902928, "grad_norm": 0.8264371752738953, "learning_rate": 1.574928794084999e-05, "loss": 0.4725, "mean_token_accuracy": 0.87197145819664, "num_tokens": 31827100.0, "step": 663 }, { "entropy": 0.5360741317272186, "epoch": 1.0231124807395995, "grad_norm": 0.9143874645233154, "learning_rate": 1.5735377745433672e-05, "loss": 0.5505, "mean_token_accuracy": 0.8534023463726044, "num_tokens": 31887725.0, "step": 664 }, { "entropy": 0.537407785654068, "epoch": 1.024653312788906, "grad_norm": 0.7521181106567383, "learning_rate": 1.5721450992781115e-05, "loss": 0.514, "mean_token_accuracy": 0.8586986362934113, "num_tokens": 31946880.0, "step": 665 }, { "entropy": 0.5650250315666199, "epoch": 1.0261941448382126, "grad_norm": 0.7906454205513, "learning_rate": 1.5707507723096913e-05, "loss": 0.5192, "mean_token_accuracy": 0.8605929315090179, "num_tokens": 32004890.0, "step": 666 }, { "entropy": 0.6054283082485199, "epoch": 1.0277349768875192, "grad_norm": 0.7783249020576477, "learning_rate": 1.5693547976633335e-05, "loss": 0.5674, "mean_token_accuracy": 0.8510688543319702, "num_tokens": 32061819.0, "step": 667 }, { "entropy": 0.5594584941864014, "epoch": 1.029275808936826, "grad_norm": 0.7411470413208008, "learning_rate": 1.5679571793690233e-05, "loss": 0.5238, "mean_token_accuracy": 0.8604991734027863, "num_tokens": 32117705.0, "step": 668 }, { "entropy": 0.5125472545623779, "epoch": 1.0308166409861326, "grad_norm": 0.7793480157852173, "learning_rate": 1.566557921461489e-05, "loss": 0.4804, "mean_token_accuracy": 0.872340202331543, "num_tokens": 32172229.0, "step": 669 }, { "entropy": 0.5083071291446686, "epoch": 1.0323574730354392, "grad_norm": 0.8317343592643738, "learning_rate": 1.5651570279801934e-05, "loss": 0.4858, "mean_token_accuracy": 0.8715094029903412, "num_tokens": 32225099.0, "step": 670 }, { "entropy": 0.5021253377199173, "epoch": 1.0338983050847457, "grad_norm": 0.878305196762085, "learning_rate": 1.5637545029693218e-05, "loss": 0.4882, "mean_token_accuracy": 0.870223730802536, "num_tokens": 32275153.0, "step": 671 }, { "entropy": 0.8189936876296997, "epoch": 1.0354391371340523, "grad_norm": 0.8239673376083374, "learning_rate": 1.562350350477767e-05, "loss": 0.8369, "mean_token_accuracy": 0.7698549926280975, "num_tokens": 32289757.0, "step": 672 }, { "entropy": 0.8288593292236328, "epoch": 1.036979969183359, "grad_norm": 0.846693754196167, "learning_rate": 1.5609445745591224e-05, "loss": 0.842, "mean_token_accuracy": 0.7777789235115051, "num_tokens": 32299887.0, "step": 673 }, { "entropy": 0.7751916646957397, "epoch": 1.0385208012326657, "grad_norm": 0.9892095923423767, "learning_rate": 1.559537179271667e-05, "loss": 0.7732, "mean_token_accuracy": 0.8064650595188141, "num_tokens": 32307583.0, "step": 674 }, { "entropy": 0.4583739787340164, "epoch": 1.0400616332819723, "grad_norm": 0.738777220249176, "learning_rate": 1.5581281686783544e-05, "loss": 0.4097, "mean_token_accuracy": 0.8911742866039276, "num_tokens": 32372963.0, "step": 675 }, { "entropy": 0.5399772524833679, "epoch": 1.041602465331279, "grad_norm": 0.8354852199554443, "learning_rate": 1.5567175468468025e-05, "loss": 0.4899, "mean_token_accuracy": 0.8662234842777252, "num_tokens": 32435291.0, "step": 676 }, { "entropy": 0.5799676477909088, "epoch": 1.0431432973805854, "grad_norm": 0.7529307007789612, "learning_rate": 1.5553053178492797e-05, "loss": 0.5465, "mean_token_accuracy": 0.8534370064735413, "num_tokens": 32495583.0, "step": 677 }, { "entropy": 0.5323314666748047, "epoch": 1.044684129429892, "grad_norm": 0.9122039079666138, "learning_rate": 1.553891485762694e-05, "loss": 0.5303, "mean_token_accuracy": 0.858908474445343, "num_tokens": 32554602.0, "step": 678 }, { "entropy": 0.5444338321685791, "epoch": 1.0462249614791987, "grad_norm": 0.9685908555984497, "learning_rate": 1.5524760546685816e-05, "loss": 0.5386, "mean_token_accuracy": 0.857380598783493, "num_tokens": 32612513.0, "step": 679 }, { "entropy": 0.5407070219516754, "epoch": 1.0477657935285054, "grad_norm": 0.8127376437187195, "learning_rate": 1.5510590286530954e-05, "loss": 0.5157, "mean_token_accuracy": 0.8641003966331482, "num_tokens": 32669200.0, "step": 680 }, { "entropy": 0.6231144070625305, "epoch": 1.049306625577812, "grad_norm": 0.7623326778411865, "learning_rate": 1.5496404118069922e-05, "loss": 0.5715, "mean_token_accuracy": 0.8507666289806366, "num_tokens": 32724958.0, "step": 681 }, { "entropy": 0.58161860704422, "epoch": 1.0508474576271187, "grad_norm": 0.8177782297134399, "learning_rate": 1.54822020822562e-05, "loss": 0.5281, "mean_token_accuracy": 0.8615186810493469, "num_tokens": 32779592.0, "step": 682 }, { "entropy": 0.5430444478988647, "epoch": 1.0523882896764252, "grad_norm": 0.818202018737793, "learning_rate": 1.546798422008909e-05, "loss": 0.4949, "mean_token_accuracy": 0.8686383962631226, "num_tokens": 32832856.0, "step": 683 }, { "entropy": 0.4598069339990616, "epoch": 1.0539291217257318, "grad_norm": 0.7800855040550232, "learning_rate": 1.5453750572613586e-05, "loss": 0.4371, "mean_token_accuracy": 0.8847869038581848, "num_tokens": 32883638.0, "step": 684 }, { "entropy": 0.797127366065979, "epoch": 1.0554699537750385, "grad_norm": 0.9509226679801941, "learning_rate": 1.5439501180920237e-05, "loss": 0.6938, "mean_token_accuracy": 0.8037258386611938, "num_tokens": 32901259.0, "step": 685 }, { "entropy": 0.865061491727829, "epoch": 1.0570107858243452, "grad_norm": 0.859535276889801, "learning_rate": 1.542523608614505e-05, "loss": 0.8615, "mean_token_accuracy": 0.7789372503757477, "num_tokens": 32910856.0, "step": 686 }, { "entropy": 0.5668277591466904, "epoch": 1.0585516178736518, "grad_norm": 0.9801860451698303, "learning_rate": 1.5410955329469374e-05, "loss": 0.3997, "mean_token_accuracy": 0.8548542261123657, "num_tokens": 32947206.0, "step": 687 }, { "entropy": 0.4717353731393814, "epoch": 1.0600924499229585, "grad_norm": 0.77371746301651, "learning_rate": 1.5396658952119757e-05, "loss": 0.4638, "mean_token_accuracy": 0.8724038600921631, "num_tokens": 33011228.0, "step": 688 }, { "entropy": 0.5378332436084747, "epoch": 1.061633281972265, "grad_norm": 0.7687921524047852, "learning_rate": 1.538234699536785e-05, "loss": 0.5027, "mean_token_accuracy": 0.8628409802913666, "num_tokens": 33072405.0, "step": 689 }, { "entropy": 0.5185485929250717, "epoch": 1.0631741140215716, "grad_norm": 0.6975071430206299, "learning_rate": 1.5368019500530272e-05, "loss": 0.4847, "mean_token_accuracy": 0.8707943260669708, "num_tokens": 33131863.0, "step": 690 }, { "entropy": 0.5804604887962341, "epoch": 1.0647149460708782, "grad_norm": 0.7812047004699707, "learning_rate": 1.5353676508968514e-05, "loss": 0.5333, "mean_token_accuracy": 0.8570082783699036, "num_tokens": 33189893.0, "step": 691 }, { "entropy": 0.5502882301807404, "epoch": 1.066255778120185, "grad_norm": 0.8083896636962891, "learning_rate": 1.5339318062088785e-05, "loss": 0.5132, "mean_token_accuracy": 0.8637231290340424, "num_tokens": 33246819.0, "step": 692 }, { "entropy": 0.5330832600593567, "epoch": 1.0677966101694916, "grad_norm": 0.8046830296516418, "learning_rate": 1.5324944201341924e-05, "loss": 0.5132, "mean_token_accuracy": 0.8649774193763733, "num_tokens": 33302501.0, "step": 693 }, { "entropy": 0.49690790474414825, "epoch": 1.0693374422187982, "grad_norm": 0.8592949509620667, "learning_rate": 1.5310554968223263e-05, "loss": 0.4744, "mean_token_accuracy": 0.8758134543895721, "num_tokens": 33357005.0, "step": 694 }, { "entropy": 0.509696364402771, "epoch": 1.0708782742681047, "grad_norm": 0.8768734335899353, "learning_rate": 1.5296150404272513e-05, "loss": 0.4892, "mean_token_accuracy": 0.8723560273647308, "num_tokens": 33410189.0, "step": 695 }, { "entropy": 0.5134027302265167, "epoch": 1.0724191063174113, "grad_norm": 0.8068960309028625, "learning_rate": 1.528173055107364e-05, "loss": 0.4836, "mean_token_accuracy": 0.8753958344459534, "num_tokens": 33461145.0, "step": 696 }, { "entropy": 0.7712198793888092, "epoch": 1.073959938366718, "grad_norm": 0.9698675870895386, "learning_rate": 1.526729545025476e-05, "loss": 0.6651, "mean_token_accuracy": 0.8036609292030334, "num_tokens": 33480199.0, "step": 697 }, { "entropy": 0.8632596731185913, "epoch": 1.0755007704160247, "grad_norm": 0.7761237621307373, "learning_rate": 1.5252845143487981e-05, "loss": 0.8343, "mean_token_accuracy": 0.7790936231613159, "num_tokens": 33490677.0, "step": 698 }, { "entropy": 0.7482123076915741, "epoch": 1.0770416024653313, "grad_norm": 0.8898210525512695, "learning_rate": 1.5238379672489342e-05, "loss": 0.7268, "mean_token_accuracy": 0.8161284327507019, "num_tokens": 33498407.0, "step": 699 }, { "entropy": 0.39576229453086853, "epoch": 1.078582434514638, "grad_norm": 0.6851622462272644, "learning_rate": 1.5223899079018629e-05, "loss": 0.3441, "mean_token_accuracy": 0.9112300872802734, "num_tokens": 33563758.0, "step": 700 }, { "entropy": 0.4991697669029236, "epoch": 1.0801232665639446, "grad_norm": 0.7685014605522156, "learning_rate": 1.5209403404879305e-05, "loss": 0.4552, "mean_token_accuracy": 0.8750397861003876, "num_tokens": 33626550.0, "step": 701 }, { "entropy": 0.5168411135673523, "epoch": 1.081664098613251, "grad_norm": 0.7935958504676819, "learning_rate": 1.5194892691918357e-05, "loss": 0.4914, "mean_token_accuracy": 0.8659385144710541, "num_tokens": 33687309.0, "step": 702 }, { "entropy": 0.49022042751312256, "epoch": 1.0832049306625577, "grad_norm": 0.9493620991706848, "learning_rate": 1.518036698202619e-05, "loss": 0.4926, "mean_token_accuracy": 0.8671912550926208, "num_tokens": 33746503.0, "step": 703 }, { "entropy": 0.5128056406974792, "epoch": 1.0847457627118644, "grad_norm": 0.9016753435134888, "learning_rate": 1.5165826317136509e-05, "loss": 0.5177, "mean_token_accuracy": 0.8621093332767487, "num_tokens": 33804195.0, "step": 704 }, { "entropy": 0.51957868039608, "epoch": 1.086286594761171, "grad_norm": 0.8502915501594543, "learning_rate": 1.5151270739226185e-05, "loss": 0.506, "mean_token_accuracy": 0.8663561046123505, "num_tokens": 33860734.0, "step": 705 }, { "entropy": 0.557144433259964, "epoch": 1.0878274268104777, "grad_norm": 0.7723761796951294, "learning_rate": 1.513670029031514e-05, "loss": 0.5307, "mean_token_accuracy": 0.8613586127758026, "num_tokens": 33916408.0, "step": 706 }, { "entropy": 0.5457858443260193, "epoch": 1.0893682588597842, "grad_norm": 0.7769936323165894, "learning_rate": 1.5122115012466234e-05, "loss": 0.4988, "mean_token_accuracy": 0.8691689074039459, "num_tokens": 33970998.0, "step": 707 }, { "entropy": 0.5049193501472473, "epoch": 1.0909090909090908, "grad_norm": 0.7501549124717712, "learning_rate": 1.5107514947785127e-05, "loss": 0.4634, "mean_token_accuracy": 0.8792309463024139, "num_tokens": 34024251.0, "step": 708 }, { "entropy": 0.5133844763040543, "epoch": 1.0924499229583975, "grad_norm": 0.7769955992698669, "learning_rate": 1.5092900138420175e-05, "loss": 0.4793, "mean_token_accuracy": 0.8768480122089386, "num_tokens": 34075516.0, "step": 709 }, { "entropy": 0.7724987864494324, "epoch": 1.0939907550077042, "grad_norm": 1.022909164428711, "learning_rate": 1.5078270626562293e-05, "loss": 0.7119, "mean_token_accuracy": 0.8000074923038483, "num_tokens": 34093840.0, "step": 710 }, { "entropy": 0.8061621189117432, "epoch": 1.0955315870570108, "grad_norm": 0.9969980716705322, "learning_rate": 1.5063626454444844e-05, "loss": 0.8214, "mean_token_accuracy": 0.7831631600856781, "num_tokens": 34103359.0, "step": 711 }, { "entropy": 0.4548729956150055, "epoch": 1.0970724191063175, "grad_norm": 0.7245083451271057, "learning_rate": 1.5048967664343512e-05, "loss": 0.2567, "mean_token_accuracy": 0.8876852691173553, "num_tokens": 34139621.0, "step": 712 }, { "entropy": 0.37555451691150665, "epoch": 1.0986132511556241, "grad_norm": 0.6814975738525391, "learning_rate": 1.5034294298576183e-05, "loss": 0.3627, "mean_token_accuracy": 0.9032894968986511, "num_tokens": 34204567.0, "step": 713 }, { "entropy": 0.534715086221695, "epoch": 1.1001540832049306, "grad_norm": 0.774013102054596, "learning_rate": 1.501960639950281e-05, "loss": 0.5206, "mean_token_accuracy": 0.8597505986690521, "num_tokens": 34266870.0, "step": 714 }, { "entropy": 0.49478381872177124, "epoch": 1.1016949152542372, "grad_norm": 0.7558149099349976, "learning_rate": 1.5004904009525328e-05, "loss": 0.4651, "mean_token_accuracy": 0.8748617768287659, "num_tokens": 34327299.0, "step": 715 }, { "entropy": 0.500090479850769, "epoch": 1.103235747303544, "grad_norm": 0.7926544547080994, "learning_rate": 1.4990187171087469e-05, "loss": 0.4561, "mean_token_accuracy": 0.8768334686756134, "num_tokens": 34386233.0, "step": 716 }, { "entropy": 0.5259290337562561, "epoch": 1.1047765793528506, "grad_norm": 0.7527215480804443, "learning_rate": 1.497545592667471e-05, "loss": 0.4982, "mean_token_accuracy": 0.8702469766139984, "num_tokens": 34443937.0, "step": 717 }, { "entropy": 0.49939432740211487, "epoch": 1.1063174114021572, "grad_norm": 0.7811996340751648, "learning_rate": 1.4960710318814087e-05, "loss": 0.4626, "mean_token_accuracy": 0.8765758574008942, "num_tokens": 34500535.0, "step": 718 }, { "entropy": 0.45766058564186096, "epoch": 1.1078582434514639, "grad_norm": 0.7926631569862366, "learning_rate": 1.4945950390074124e-05, "loss": 0.4186, "mean_token_accuracy": 0.8882555961608887, "num_tokens": 34555961.0, "step": 719 }, { "entropy": 0.519398421049118, "epoch": 1.1093990755007703, "grad_norm": 0.8115248084068298, "learning_rate": 1.4931176183064677e-05, "loss": 0.4804, "mean_token_accuracy": 0.8735583424568176, "num_tokens": 34609996.0, "step": 720 }, { "entropy": 0.4945348650217056, "epoch": 1.110939907550077, "grad_norm": 0.8003508448600769, "learning_rate": 1.4916387740436822e-05, "loss": 0.4606, "mean_token_accuracy": 0.878229558467865, "num_tokens": 34661989.0, "step": 721 }, { "entropy": 0.6867009401321411, "epoch": 1.1124807395993837, "grad_norm": 0.957168459892273, "learning_rate": 1.490158510488273e-05, "loss": 0.5668, "mean_token_accuracy": 0.8165560662746429, "num_tokens": 34689741.0, "step": 722 }, { "entropy": 0.8584219515323639, "epoch": 1.1140215716486903, "grad_norm": 0.8001411557197571, "learning_rate": 1.488676831913555e-05, "loss": 0.8415, "mean_token_accuracy": 0.7759655714035034, "num_tokens": 34700882.0, "step": 723 }, { "entropy": 0.8001158833503723, "epoch": 1.115562403697997, "grad_norm": 0.8829438090324402, "learning_rate": 1.4871937425969271e-05, "loss": 0.7656, "mean_token_accuracy": 0.8017398118972778, "num_tokens": 34709118.0, "step": 724 }, { "entropy": 0.3839167058467865, "epoch": 1.1171032357473036, "grad_norm": 0.6857571601867676, "learning_rate": 1.4857092468198626e-05, "loss": 0.3513, "mean_token_accuracy": 0.9060083627700806, "num_tokens": 34774320.0, "step": 725 }, { "entropy": 0.4647670388221741, "epoch": 1.11864406779661, "grad_norm": 0.7705230712890625, "learning_rate": 1.4842233488678927e-05, "loss": 0.4403, "mean_token_accuracy": 0.8788951933383942, "num_tokens": 34836399.0, "step": 726 }, { "entropy": 0.4834244102239609, "epoch": 1.1201848998459167, "grad_norm": 0.8724203705787659, "learning_rate": 1.4827360530305987e-05, "loss": 0.4844, "mean_token_accuracy": 0.8694550395011902, "num_tokens": 34896564.0, "step": 727 }, { "entropy": 0.538922131061554, "epoch": 1.1217257318952234, "grad_norm": 0.8357527852058411, "learning_rate": 1.4812473636015958e-05, "loss": 0.5285, "mean_token_accuracy": 0.8579253852367401, "num_tokens": 34955575.0, "step": 728 }, { "entropy": 0.4942777156829834, "epoch": 1.12326656394453, "grad_norm": 0.8175280690193176, "learning_rate": 1.479757284878523e-05, "loss": 0.4724, "mean_token_accuracy": 0.8735999464988708, "num_tokens": 35013559.0, "step": 729 }, { "entropy": 0.502753347158432, "epoch": 1.1248073959938367, "grad_norm": 0.7818880081176758, "learning_rate": 1.4782658211630301e-05, "loss": 0.4777, "mean_token_accuracy": 0.8728020191192627, "num_tokens": 35070454.0, "step": 730 }, { "entropy": 0.4793979078531265, "epoch": 1.1263482280431434, "grad_norm": 0.8295426964759827, "learning_rate": 1.476772976760765e-05, "loss": 0.4482, "mean_token_accuracy": 0.8818789124488831, "num_tokens": 35126138.0, "step": 731 }, { "entropy": 0.5165483057498932, "epoch": 1.1278890600924498, "grad_norm": 0.8202833533287048, "learning_rate": 1.4752787559813612e-05, "loss": 0.4743, "mean_token_accuracy": 0.8751756548881531, "num_tokens": 35180510.0, "step": 732 }, { "entropy": 0.46216820180416107, "epoch": 1.1294298921417565, "grad_norm": 0.7621103525161743, "learning_rate": 1.473783163138426e-05, "loss": 0.4254, "mean_token_accuracy": 0.888803631067276, "num_tokens": 35233572.0, "step": 733 }, { "entropy": 0.4571734368801117, "epoch": 1.1309707241910631, "grad_norm": 0.834879457950592, "learning_rate": 1.4722862025495275e-05, "loss": 0.4304, "mean_token_accuracy": 0.8871630728244781, "num_tokens": 35282280.0, "step": 734 }, { "entropy": 0.8614612519741058, "epoch": 1.1325115562403698, "grad_norm": 0.7209380269050598, "learning_rate": 1.4707878785361822e-05, "loss": 0.874, "mean_token_accuracy": 0.7654021084308624, "num_tokens": 35294478.0, "step": 735 }, { "entropy": 0.8393189013004303, "epoch": 1.1340523882896765, "grad_norm": 0.8367904424667358, "learning_rate": 1.4692881954238429e-05, "loss": 0.8283, "mean_token_accuracy": 0.7811445593833923, "num_tokens": 35304067.0, "step": 736 }, { "entropy": 0.4956230968236923, "epoch": 1.1355932203389831, "grad_norm": 0.7117910385131836, "learning_rate": 1.4677871575418854e-05, "loss": 0.308, "mean_token_accuracy": 0.876295268535614, "num_tokens": 35340453.0, "step": 737 }, { "entropy": 0.41725610196590424, "epoch": 1.1371340523882898, "grad_norm": 0.7252728343009949, "learning_rate": 1.4662847692235973e-05, "loss": 0.4061, "mean_token_accuracy": 0.8912385404109955, "num_tokens": 35404365.0, "step": 738 }, { "entropy": 0.48795732855796814, "epoch": 1.1386748844375962, "grad_norm": 0.7916492223739624, "learning_rate": 1.4647810348061642e-05, "loss": 0.4777, "mean_token_accuracy": 0.8708635866641998, "num_tokens": 35465502.0, "step": 739 }, { "entropy": 0.5548458695411682, "epoch": 1.140215716486903, "grad_norm": 0.8022623062133789, "learning_rate": 1.4632759586306575e-05, "loss": 0.5387, "mean_token_accuracy": 0.8543573021888733, "num_tokens": 35524869.0, "step": 740 }, { "entropy": 0.504782184958458, "epoch": 1.1417565485362096, "grad_norm": 0.7173783183097839, "learning_rate": 1.4617695450420226e-05, "loss": 0.4726, "mean_token_accuracy": 0.8743169009685516, "num_tokens": 35583080.0, "step": 741 }, { "entropy": 0.47813500463962555, "epoch": 1.1432973805855162, "grad_norm": 0.7812519669532776, "learning_rate": 1.4602617983890657e-05, "loss": 0.429, "mean_token_accuracy": 0.8844589293003082, "num_tokens": 35640165.0, "step": 742 }, { "entropy": 0.47655612230300903, "epoch": 1.1448382126348229, "grad_norm": 0.7747393846511841, "learning_rate": 1.458752723024441e-05, "loss": 0.4292, "mean_token_accuracy": 0.8855638206005096, "num_tokens": 35696260.0, "step": 743 }, { "entropy": 0.4685034900903702, "epoch": 1.1463790446841293, "grad_norm": 0.7956759333610535, "learning_rate": 1.4572423233046386e-05, "loss": 0.4253, "mean_token_accuracy": 0.8888247907161713, "num_tokens": 35751458.0, "step": 744 }, { "entropy": 0.5143736451864243, "epoch": 1.147919876733436, "grad_norm": 0.8767870664596558, "learning_rate": 1.4557306035899728e-05, "loss": 0.4871, "mean_token_accuracy": 0.8718785047531128, "num_tokens": 35805271.0, "step": 745 }, { "entropy": 0.47573940455913544, "epoch": 1.1494607087827426, "grad_norm": 0.9272811412811279, "learning_rate": 1.454217568244567e-05, "loss": 0.4642, "mean_token_accuracy": 0.8770708739757538, "num_tokens": 35857068.0, "step": 746 }, { "entropy": 0.6228059530258179, "epoch": 1.1510015408320493, "grad_norm": 0.9019546508789062, "learning_rate": 1.4527032216363432e-05, "loss": 0.511, "mean_token_accuracy": 0.8412439227104187, "num_tokens": 35889995.0, "step": 747 }, { "entropy": 0.8587234020233154, "epoch": 1.152542372881356, "grad_norm": 0.8710352182388306, "learning_rate": 1.45118756813701e-05, "loss": 0.8321, "mean_token_accuracy": 0.7817944586277008, "num_tokens": 35900568.0, "step": 748 }, { "entropy": 0.7896102964878082, "epoch": 1.1540832049306626, "grad_norm": 0.9539501070976257, "learning_rate": 1.4496706121220467e-05, "loss": 0.7667, "mean_token_accuracy": 0.8035923838615417, "num_tokens": 35908403.0, "step": 749 }, { "entropy": 0.3005313277244568, "epoch": 1.1556240369799693, "grad_norm": 0.6367896795272827, "learning_rate": 1.4481523579706947e-05, "loss": 0.255, "mean_token_accuracy": 0.9328800737857819, "num_tokens": 35973933.0, "step": 750 }, { "entropy": 0.4734940826892853, "epoch": 1.1571648690292757, "grad_norm": 0.7782305479049683, "learning_rate": 1.4466328100659417e-05, "loss": 0.4419, "mean_token_accuracy": 0.8790837824344635, "num_tokens": 36037105.0, "step": 751 }, { "entropy": 0.4792975187301636, "epoch": 1.1587057010785824, "grad_norm": 0.773251473903656, "learning_rate": 1.445111972794511e-05, "loss": 0.4583, "mean_token_accuracy": 0.8759827315807343, "num_tokens": 36097710.0, "step": 752 }, { "entropy": 0.4866265654563904, "epoch": 1.160246533127889, "grad_norm": 0.8030644655227661, "learning_rate": 1.443589850546848e-05, "loss": 0.475, "mean_token_accuracy": 0.8721277117729187, "num_tokens": 36156674.0, "step": 753 }, { "entropy": 0.5128634572029114, "epoch": 1.1617873651771957, "grad_norm": 0.8594357967376709, "learning_rate": 1.442066447717107e-05, "loss": 0.514, "mean_token_accuracy": 0.8652056157588959, "num_tokens": 36214383.0, "step": 754 }, { "entropy": 0.4585960805416107, "epoch": 1.1633281972265024, "grad_norm": 0.8608696460723877, "learning_rate": 1.4405417687031396e-05, "loss": 0.449, "mean_token_accuracy": 0.8811867833137512, "num_tokens": 36270944.0, "step": 755 }, { "entropy": 0.49556758999824524, "epoch": 1.1648690292758088, "grad_norm": 0.8167732357978821, "learning_rate": 1.4390158179064827e-05, "loss": 0.4664, "mean_token_accuracy": 0.8779769241809845, "num_tokens": 36326371.0, "step": 756 }, { "entropy": 0.46801814436912537, "epoch": 1.1664098613251155, "grad_norm": 0.8128525018692017, "learning_rate": 1.437488599732342e-05, "loss": 0.4351, "mean_token_accuracy": 0.8851093947887421, "num_tokens": 36380661.0, "step": 757 }, { "entropy": 0.5233100205659866, "epoch": 1.1679506933744221, "grad_norm": 0.816537082195282, "learning_rate": 1.4359601185895851e-05, "loss": 0.4773, "mean_token_accuracy": 0.874967098236084, "num_tokens": 36433354.0, "step": 758 }, { "entropy": 0.5481535941362381, "epoch": 1.1694915254237288, "grad_norm": 0.8139190077781677, "learning_rate": 1.4344303788907228e-05, "loss": 0.4972, "mean_token_accuracy": 0.8667415380477905, "num_tokens": 36476212.0, "step": 759 }, { "entropy": 0.8281311094760895, "epoch": 1.1710323574730355, "grad_norm": 0.8269695043563843, "learning_rate": 1.4328993850519008e-05, "loss": 0.8723, "mean_token_accuracy": 0.7645474970340729, "num_tokens": 36488275.0, "step": 760 }, { "entropy": 0.8274376392364502, "epoch": 1.1725731895223421, "grad_norm": 0.9323627352714539, "learning_rate": 1.4313671414928854e-05, "loss": 0.8556, "mean_token_accuracy": 0.784521758556366, "num_tokens": 36497683.0, "step": 761 }, { "entropy": 0.47460444271564484, "epoch": 1.1741140215716488, "grad_norm": 0.8500837683677673, "learning_rate": 1.4298336526370497e-05, "loss": 0.2813, "mean_token_accuracy": 0.8786875903606415, "num_tokens": 36533950.0, "step": 762 }, { "entropy": 0.49033865332603455, "epoch": 1.1756548536209552, "grad_norm": 0.7310845255851746, "learning_rate": 1.4282989229113627e-05, "loss": 0.4691, "mean_token_accuracy": 0.8725214004516602, "num_tokens": 36597536.0, "step": 763 }, { "entropy": 0.4831033945083618, "epoch": 1.177195685670262, "grad_norm": 0.7862340807914734, "learning_rate": 1.4267629567463748e-05, "loss": 0.4512, "mean_token_accuracy": 0.8784531354904175, "num_tokens": 36658829.0, "step": 764 }, { "entropy": 0.5110671520233154, "epoch": 1.1787365177195686, "grad_norm": 0.7511234879493713, "learning_rate": 1.4252257585762067e-05, "loss": 0.4768, "mean_token_accuracy": 0.8747625946998596, "num_tokens": 36718447.0, "step": 765 }, { "entropy": 0.4997715801000595, "epoch": 1.1802773497688752, "grad_norm": 0.8037508130073547, "learning_rate": 1.4236873328385348e-05, "loss": 0.4607, "mean_token_accuracy": 0.876869797706604, "num_tokens": 36776807.0, "step": 766 }, { "entropy": 0.4882461577653885, "epoch": 1.1818181818181819, "grad_norm": 0.7692822217941284, "learning_rate": 1.42214768397458e-05, "loss": 0.4581, "mean_token_accuracy": 0.8785678744316101, "num_tokens": 36834154.0, "step": 767 }, { "entropy": 0.5376203060150146, "epoch": 1.1833590138674885, "grad_norm": 0.8024082779884338, "learning_rate": 1.4206068164290944e-05, "loss": 0.5105, "mean_token_accuracy": 0.8673704862594604, "num_tokens": 36890375.0, "step": 768 }, { "entropy": 0.49318012595176697, "epoch": 1.184899845916795, "grad_norm": 0.9149112701416016, "learning_rate": 1.4190647346503475e-05, "loss": 0.4699, "mean_token_accuracy": 0.8768426477909088, "num_tokens": 36945455.0, "step": 769 }, { "entropy": 0.4536915123462677, "epoch": 1.1864406779661016, "grad_norm": 0.827889084815979, "learning_rate": 1.4175214430901143e-05, "loss": 0.4109, "mean_token_accuracy": 0.8915937840938568, "num_tokens": 36999173.0, "step": 770 }, { "entropy": 0.5189929008483887, "epoch": 1.1879815100154083, "grad_norm": 0.825063169002533, "learning_rate": 1.4159769462036628e-05, "loss": 0.4707, "mean_token_accuracy": 0.8771334290504456, "num_tokens": 37050722.0, "step": 771 }, { "entropy": 0.8017478287220001, "epoch": 1.189522342064715, "grad_norm": 0.8828855752944946, "learning_rate": 1.41443124844974e-05, "loss": 0.643, "mean_token_accuracy": 0.8088609874248505, "num_tokens": 37073384.0, "step": 772 }, { "entropy": 0.866552084684372, "epoch": 1.1910631741140216, "grad_norm": 0.8147509098052979, "learning_rate": 1.4128843542905597e-05, "loss": 0.8217, "mean_token_accuracy": 0.7802944481372833, "num_tokens": 37083826.0, "step": 773 }, { "entropy": 0.7168662548065186, "epoch": 1.1926040061633283, "grad_norm": 0.9703771471977234, "learning_rate": 1.4113362681917908e-05, "loss": 0.724, "mean_token_accuracy": 0.8253074586391449, "num_tokens": 37091164.0, "step": 774 }, { "entropy": 0.3031690716743469, "epoch": 1.1941448382126347, "grad_norm": 0.6680529713630676, "learning_rate": 1.409786994622541e-05, "loss": 0.274, "mean_token_accuracy": 0.9257787764072418, "num_tokens": 37156522.0, "step": 775 }, { "entropy": 0.48454025387763977, "epoch": 1.1956856702619414, "grad_norm": 0.8303526043891907, "learning_rate": 1.4082365380553475e-05, "loss": 0.4817, "mean_token_accuracy": 0.8703080713748932, "num_tokens": 37219074.0, "step": 776 }, { "entropy": 0.45749431848526, "epoch": 1.197226502311248, "grad_norm": 0.7669852375984192, "learning_rate": 1.4066849029661625e-05, "loss": 0.452, "mean_token_accuracy": 0.8781610429286957, "num_tokens": 37279716.0, "step": 777 }, { "entropy": 0.4727403521537781, "epoch": 1.1987673343605547, "grad_norm": 0.8001810908317566, "learning_rate": 1.4051320938343402e-05, "loss": 0.4647, "mean_token_accuracy": 0.87689608335495, "num_tokens": 37339052.0, "step": 778 }, { "entropy": 0.48017077147960663, "epoch": 1.2003081664098614, "grad_norm": 0.7450380921363831, "learning_rate": 1.4035781151426242e-05, "loss": 0.4585, "mean_token_accuracy": 0.8771277368068695, "num_tokens": 37397154.0, "step": 779 }, { "entropy": 0.4952477961778641, "epoch": 1.201848998459168, "grad_norm": 0.7423424124717712, "learning_rate": 1.4020229713771343e-05, "loss": 0.46, "mean_token_accuracy": 0.8789591491222382, "num_tokens": 37454120.0, "step": 780 }, { "entropy": 0.49579760432243347, "epoch": 1.2033898305084745, "grad_norm": 0.7668017148971558, "learning_rate": 1.4004666670273536e-05, "loss": 0.4608, "mean_token_accuracy": 0.8781505525112152, "num_tokens": 37509840.0, "step": 781 }, { "entropy": 0.4964725375175476, "epoch": 1.2049306625577811, "grad_norm": 0.79903644323349, "learning_rate": 1.3989092065861165e-05, "loss": 0.4595, "mean_token_accuracy": 0.8807966411113739, "num_tokens": 37564334.0, "step": 782 }, { "entropy": 0.421910360455513, "epoch": 1.2064714946070878, "grad_norm": 0.7612454891204834, "learning_rate": 1.3973505945495936e-05, "loss": 0.3862, "mean_token_accuracy": 0.8989439308643341, "num_tokens": 37617452.0, "step": 783 }, { "entropy": 0.42465008795261383, "epoch": 1.2080123266563945, "grad_norm": 0.8894613981246948, "learning_rate": 1.3957908354172811e-05, "loss": 0.4037, "mean_token_accuracy": 0.8938982784748077, "num_tokens": 37667864.0, "step": 784 }, { "entropy": 0.8407284319400787, "epoch": 1.2095531587057011, "grad_norm": 0.833299994468689, "learning_rate": 1.3942299336919858e-05, "loss": 0.7949, "mean_token_accuracy": 0.7771202623844147, "num_tokens": 37681611.0, "step": 785 }, { "entropy": 0.8040720820426941, "epoch": 1.2110939907550078, "grad_norm": 0.9501211643218994, "learning_rate": 1.392667893879814e-05, "loss": 0.797, "mean_token_accuracy": 0.7994080483913422, "num_tokens": 37690900.0, "step": 786 }, { "entropy": 0.5073088109493256, "epoch": 1.2126348228043144, "grad_norm": 0.7460045218467712, "learning_rate": 1.391104720490156e-05, "loss": 0.3184, "mean_token_accuracy": 0.880297064781189, "num_tokens": 37727162.0, "step": 787 }, { "entropy": 0.39003658294677734, "epoch": 1.214175654853621, "grad_norm": 0.6633943319320679, "learning_rate": 1.3895404180356766e-05, "loss": 0.3688, "mean_token_accuracy": 0.9011155366897583, "num_tokens": 37791814.0, "step": 788 }, { "entropy": 0.4445952922105789, "epoch": 1.2157164869029276, "grad_norm": 0.7473528981208801, "learning_rate": 1.3879749910322986e-05, "loss": 0.427, "mean_token_accuracy": 0.8845164477825165, "num_tokens": 37853580.0, "step": 789 }, { "entropy": 0.46676959097385406, "epoch": 1.2172573189522342, "grad_norm": 0.8127845525741577, "learning_rate": 1.3864084439991914e-05, "loss": 0.4515, "mean_token_accuracy": 0.8777066469192505, "num_tokens": 37913958.0, "step": 790 }, { "entropy": 0.4629710018634796, "epoch": 1.2187981510015409, "grad_norm": 0.8322455883026123, "learning_rate": 1.384840781458758e-05, "loss": 0.4407, "mean_token_accuracy": 0.8820155262947083, "num_tokens": 37972945.0, "step": 791 }, { "entropy": 0.45592716336250305, "epoch": 1.2203389830508475, "grad_norm": 0.7671761512756348, "learning_rate": 1.383272007936622e-05, "loss": 0.4327, "mean_token_accuracy": 0.884373277425766, "num_tokens": 38030611.0, "step": 792 }, { "entropy": 0.4890872687101364, "epoch": 1.221879815100154, "grad_norm": 0.8471571803092957, "learning_rate": 1.3817021279616132e-05, "loss": 0.4672, "mean_token_accuracy": 0.8757957518100739, "num_tokens": 38087075.0, "step": 793 }, { "entropy": 0.45407524704933167, "epoch": 1.2234206471494606, "grad_norm": 0.7958045601844788, "learning_rate": 1.3801311460657574e-05, "loss": 0.4237, "mean_token_accuracy": 0.8882387280464172, "num_tokens": 38142324.0, "step": 794 }, { "entropy": 0.5034683346748352, "epoch": 1.2249614791987673, "grad_norm": 0.7985876798629761, "learning_rate": 1.3785590667842592e-05, "loss": 0.483, "mean_token_accuracy": 0.8762783110141754, "num_tokens": 38196160.0, "step": 795 }, { "entropy": 0.4172234684228897, "epoch": 1.226502311248074, "grad_norm": 0.7916297912597656, "learning_rate": 1.3769858946554933e-05, "loss": 0.3771, "mean_token_accuracy": 0.9000651836395264, "num_tokens": 38247894.0, "step": 796 }, { "entropy": 0.7011696696281433, "epoch": 1.2280431432973806, "grad_norm": 0.9339724779129028, "learning_rate": 1.375411634220988e-05, "loss": 0.5611, "mean_token_accuracy": 0.823423832654953, "num_tokens": 38270455.0, "step": 797 }, { "entropy": 0.8860751688480377, "epoch": 1.2295839753466873, "grad_norm": 0.8616496324539185, "learning_rate": 1.3738362900254138e-05, "loss": 0.8452, "mean_token_accuracy": 0.7746357917785645, "num_tokens": 38280586.0, "step": 798 }, { "entropy": 0.7556107640266418, "epoch": 1.231124807395994, "grad_norm": 0.9149807691574097, "learning_rate": 1.3722598666165701e-05, "loss": 0.7485, "mean_token_accuracy": 0.8119445741176605, "num_tokens": 38288207.0, "step": 799 }, { "entropy": 0.31863054633140564, "epoch": 1.2326656394453004, "grad_norm": 0.6011478900909424, "learning_rate": 1.3706823685453712e-05, "loss": 0.2876, "mean_token_accuracy": 0.9248139560222626, "num_tokens": 38353568.0, "step": 800 }, { "epoch": 1.2326656394453004, "eval_entropy": 0.5544024425435741, "eval_loss": 0.5433660745620728, "eval_mean_token_accuracy": 0.8578476703391885, "eval_num_tokens": 38353568.0, "eval_runtime": 195.1582, "eval_samples_per_second": 4.345, "eval_steps_per_second": 1.086, "step": 800 }, { "entropy": 0.44650672376155853, "epoch": 1.234206471494607, "grad_norm": 0.7474241256713867, "learning_rate": 1.3691038003658346e-05, "loss": 0.4209, "mean_token_accuracy": 0.8845660984516144, "num_tokens": 38416064.0, "step": 801 }, { "entropy": 0.4330813139677048, "epoch": 1.2357473035439137, "grad_norm": 0.7525302767753601, "learning_rate": 1.367524166635066e-05, "loss": 0.4122, "mean_token_accuracy": 0.8893899619579315, "num_tokens": 38476828.0, "step": 802 }, { "entropy": 0.45970623195171356, "epoch": 1.2372881355932204, "grad_norm": 0.7583761811256409, "learning_rate": 1.3659434719132485e-05, "loss": 0.4388, "mean_token_accuracy": 0.8824505507946014, "num_tokens": 38536204.0, "step": 803 }, { "entropy": 0.43146783113479614, "epoch": 1.238828967642527, "grad_norm": 0.8368088006973267, "learning_rate": 1.364361720763627e-05, "loss": 0.4222, "mean_token_accuracy": 0.8876582384109497, "num_tokens": 38594301.0, "step": 804 }, { "entropy": 0.40138348937034607, "epoch": 1.2403697996918335, "grad_norm": 0.7409807443618774, "learning_rate": 1.362778917752497e-05, "loss": 0.3789, "mean_token_accuracy": 0.8983252346515656, "num_tokens": 38651530.0, "step": 805 }, { "entropy": 0.4525478482246399, "epoch": 1.2419106317411401, "grad_norm": 0.7792746424674988, "learning_rate": 1.3611950674491898e-05, "loss": 0.4228, "mean_token_accuracy": 0.8899587690830231, "num_tokens": 38707627.0, "step": 806 }, { "entropy": 0.44481711089611053, "epoch": 1.2434514637904468, "grad_norm": 0.7641570568084717, "learning_rate": 1.3596101744260608e-05, "loss": 0.4127, "mean_token_accuracy": 0.889973908662796, "num_tokens": 38762549.0, "step": 807 }, { "entropy": 0.4337643086910248, "epoch": 1.2449922958397535, "grad_norm": 0.7690803408622742, "learning_rate": 1.3580242432584753e-05, "loss": 0.4019, "mean_token_accuracy": 0.8952448666095734, "num_tokens": 38816076.0, "step": 808 }, { "entropy": 0.4499708116054535, "epoch": 1.2465331278890601, "grad_norm": 0.7820766568183899, "learning_rate": 1.3564372785247955e-05, "loss": 0.4227, "mean_token_accuracy": 0.8917208611965179, "num_tokens": 38866889.0, "step": 809 }, { "entropy": 0.8099448978900909, "epoch": 1.2480739599383668, "grad_norm": 0.8400903344154358, "learning_rate": 1.3548492848063675e-05, "loss": 0.7941, "mean_token_accuracy": 0.7805911600589752, "num_tokens": 38880511.0, "step": 810 }, { "entropy": 0.7892324030399323, "epoch": 1.2496147919876734, "grad_norm": 0.984506368637085, "learning_rate": 1.3532602666875074e-05, "loss": 0.8046, "mean_token_accuracy": 0.7955529987812042, "num_tokens": 38889669.0, "step": 811 }, { "entropy": 0.47676628828048706, "epoch": 1.25115562403698, "grad_norm": 0.735542356967926, "learning_rate": 1.3516702287554896e-05, "loss": 0.2825, "mean_token_accuracy": 0.8778612017631531, "num_tokens": 38925834.0, "step": 812 }, { "entropy": 0.4004654735326767, "epoch": 1.2526964560862865, "grad_norm": 0.6938637495040894, "learning_rate": 1.3500791756005318e-05, "loss": 0.3735, "mean_token_accuracy": 0.899282306432724, "num_tokens": 38990131.0, "step": 813 }, { "entropy": 0.40139979124069214, "epoch": 1.2542372881355932, "grad_norm": 0.7138944268226624, "learning_rate": 1.3484871118157826e-05, "loss": 0.382, "mean_token_accuracy": 0.8947394788265228, "num_tokens": 39051890.0, "step": 814 }, { "entropy": 0.4400578439235687, "epoch": 1.2557781201848999, "grad_norm": 0.7868729829788208, "learning_rate": 1.3468940419973086e-05, "loss": 0.4257, "mean_token_accuracy": 0.8858998119831085, "num_tokens": 39111521.0, "step": 815 }, { "entropy": 0.4499833732843399, "epoch": 1.2573189522342065, "grad_norm": 0.7770370841026306, "learning_rate": 1.34529997074408e-05, "loss": 0.4245, "mean_token_accuracy": 0.8864297568798065, "num_tokens": 39169610.0, "step": 816 }, { "entropy": 0.47642141580581665, "epoch": 1.258859784283513, "grad_norm": 0.7956764101982117, "learning_rate": 1.3437049026579587e-05, "loss": 0.4495, "mean_token_accuracy": 0.8820413053035736, "num_tokens": 39226553.0, "step": 817 }, { "entropy": 0.46077868342399597, "epoch": 1.2604006163328196, "grad_norm": 0.7801942825317383, "learning_rate": 1.3421088423436834e-05, "loss": 0.4287, "mean_token_accuracy": 0.8885772228240967, "num_tokens": 39282471.0, "step": 818 }, { "entropy": 0.43582627177238464, "epoch": 1.2619414483821263, "grad_norm": 0.759852945804596, "learning_rate": 1.3405117944088584e-05, "loss": 0.4068, "mean_token_accuracy": 0.8926430940628052, "num_tokens": 39337417.0, "step": 819 }, { "entropy": 0.4582117646932602, "epoch": 1.263482280431433, "grad_norm": 0.790057361125946, "learning_rate": 1.3389137634639381e-05, "loss": 0.4306, "mean_token_accuracy": 0.8894909918308258, "num_tokens": 39391082.0, "step": 820 }, { "entropy": 0.4662276804447174, "epoch": 1.2650231124807396, "grad_norm": 0.7576630115509033, "learning_rate": 1.3373147541222155e-05, "loss": 0.4383, "mean_token_accuracy": 0.8869408965110779, "num_tokens": 39442909.0, "step": 821 }, { "entropy": 0.6310765743255615, "epoch": 1.2665639445300463, "grad_norm": 0.97206050157547, "learning_rate": 1.3357147709998075e-05, "loss": 0.5062, "mean_token_accuracy": 0.8329560160636902, "num_tokens": 39471213.0, "step": 822 }, { "entropy": 0.8367979824542999, "epoch": 1.268104776579353, "grad_norm": 0.8188115358352661, "learning_rate": 1.3341138187156423e-05, "loss": 0.8389, "mean_token_accuracy": 0.7726362645626068, "num_tokens": 39482086.0, "step": 823 }, { "entropy": 0.7895478308200836, "epoch": 1.2696456086286596, "grad_norm": 0.8811002373695374, "learning_rate": 1.3325119018914463e-05, "loss": 0.7599, "mean_token_accuracy": 0.8091971576213837, "num_tokens": 39490414.0, "step": 824 }, { "entropy": 0.2997692674398422, "epoch": 1.271186440677966, "grad_norm": 0.5948421955108643, "learning_rate": 1.33090902515173e-05, "loss": 0.2667, "mean_token_accuracy": 0.9265863597393036, "num_tokens": 39555873.0, "step": 825 }, { "entropy": 0.3979836106300354, "epoch": 1.2727272727272727, "grad_norm": 0.7248284816741943, "learning_rate": 1.329305193123775e-05, "loss": 0.3724, "mean_token_accuracy": 0.8993945419788361, "num_tokens": 39618580.0, "step": 826 }, { "entropy": 0.4760018289089203, "epoch": 1.2742681047765794, "grad_norm": 0.7751103639602661, "learning_rate": 1.327700410437621e-05, "loss": 0.4558, "mean_token_accuracy": 0.8784486055374146, "num_tokens": 39679344.0, "step": 827 }, { "entropy": 0.4436483681201935, "epoch": 1.275808936825886, "grad_norm": 0.7836160063743591, "learning_rate": 1.326094681726052e-05, "loss": 0.4256, "mean_token_accuracy": 0.8851683735847473, "num_tokens": 39738833.0, "step": 828 }, { "entropy": 0.4085596650838852, "epoch": 1.2773497688751925, "grad_norm": 0.7240772247314453, "learning_rate": 1.3244880116245829e-05, "loss": 0.3876, "mean_token_accuracy": 0.898107647895813, "num_tokens": 39797116.0, "step": 829 }, { "entropy": 0.4794706255197525, "epoch": 1.2788906009244991, "grad_norm": 0.84328693151474, "learning_rate": 1.3228804047714462e-05, "loss": 0.4645, "mean_token_accuracy": 0.8769466876983643, "num_tokens": 39854159.0, "step": 830 }, { "entropy": 0.4302190691232681, "epoch": 1.2804314329738058, "grad_norm": 0.7760748267173767, "learning_rate": 1.3212718658075794e-05, "loss": 0.4107, "mean_token_accuracy": 0.8913137316703796, "num_tokens": 39910052.0, "step": 831 }, { "entropy": 0.4301987886428833, "epoch": 1.2819722650231125, "grad_norm": 0.7803808450698853, "learning_rate": 1.3196623993766097e-05, "loss": 0.4009, "mean_token_accuracy": 0.8943925201892853, "num_tokens": 39964710.0, "step": 832 }, { "entropy": 0.38972750306129456, "epoch": 1.2835130970724191, "grad_norm": 0.7722770571708679, "learning_rate": 1.318052010124843e-05, "loss": 0.3561, "mean_token_accuracy": 0.9080211222171783, "num_tokens": 40018039.0, "step": 833 }, { "entropy": 0.4272724986076355, "epoch": 1.2850539291217258, "grad_norm": 0.7889288067817688, "learning_rate": 1.316440702701248e-05, "loss": 0.3986, "mean_token_accuracy": 0.8972617089748383, "num_tokens": 40069112.0, "step": 834 }, { "entropy": 0.7185744941234589, "epoch": 1.2865947611710324, "grad_norm": 1.0198885202407837, "learning_rate": 1.3148284817574455e-05, "loss": 0.5886, "mean_token_accuracy": 0.8196066319942474, "num_tokens": 40089741.0, "step": 835 }, { "entropy": 0.7998925149440765, "epoch": 1.288135593220339, "grad_norm": 0.9434200525283813, "learning_rate": 1.3132153519476915e-05, "loss": 0.7912, "mean_token_accuracy": 0.793085902929306, "num_tokens": 40099207.0, "step": 836 }, { "entropy": 0.4744259715080261, "epoch": 1.2896764252696455, "grad_norm": 0.6773841977119446, "learning_rate": 1.3116013179288677e-05, "loss": 0.2963, "mean_token_accuracy": 0.8818413019180298, "num_tokens": 40135601.0, "step": 837 }, { "entropy": 0.3928452879190445, "epoch": 1.2912172573189522, "grad_norm": 0.654312252998352, "learning_rate": 1.3099863843604656e-05, "loss": 0.3572, "mean_token_accuracy": 0.9036343395709991, "num_tokens": 40199832.0, "step": 838 }, { "entropy": 0.4163826107978821, "epoch": 1.2927580893682589, "grad_norm": 0.730822741985321, "learning_rate": 1.308370555904573e-05, "loss": 0.3973, "mean_token_accuracy": 0.8953453898429871, "num_tokens": 40261023.0, "step": 839 }, { "entropy": 0.4021114557981491, "epoch": 1.2942989214175655, "grad_norm": 0.7476876974105835, "learning_rate": 1.3067538372258612e-05, "loss": 0.379, "mean_token_accuracy": 0.897333413362503, "num_tokens": 40320621.0, "step": 840 }, { "entropy": 0.4273197650909424, "epoch": 1.2958397534668722, "grad_norm": 0.7884918451309204, "learning_rate": 1.3051362329915722e-05, "loss": 0.41, "mean_token_accuracy": 0.8914981186389923, "num_tokens": 40378943.0, "step": 841 }, { "entropy": 0.4499085247516632, "epoch": 1.2973805855161786, "grad_norm": 0.811513364315033, "learning_rate": 1.3035177478715035e-05, "loss": 0.4342, "mean_token_accuracy": 0.8849261403083801, "num_tokens": 40436147.0, "step": 842 }, { "entropy": 0.4429740458726883, "epoch": 1.2989214175654853, "grad_norm": 0.7707722187042236, "learning_rate": 1.3018983865379968e-05, "loss": 0.425, "mean_token_accuracy": 0.8887284696102142, "num_tokens": 40492286.0, "step": 843 }, { "entropy": 0.42749491333961487, "epoch": 1.300462249614792, "grad_norm": 0.7581144571304321, "learning_rate": 1.3002781536659216e-05, "loss": 0.4049, "mean_token_accuracy": 0.8956300020217896, "num_tokens": 40547414.0, "step": 844 }, { "entropy": 0.4313061237335205, "epoch": 1.3020030816640986, "grad_norm": 0.7869265675544739, "learning_rate": 1.2986570539326655e-05, "loss": 0.4048, "mean_token_accuracy": 0.8957211375236511, "num_tokens": 40601040.0, "step": 845 }, { "entropy": 0.41661545634269714, "epoch": 1.3035439137134053, "grad_norm": 0.7707061171531677, "learning_rate": 1.2970350920181164e-05, "loss": 0.3853, "mean_token_accuracy": 0.89975705742836, "num_tokens": 40652460.0, "step": 846 }, { "entropy": 0.6730538308620453, "epoch": 1.305084745762712, "grad_norm": 0.9090076088905334, "learning_rate": 1.2954122726046534e-05, "loss": 0.5005, "mean_token_accuracy": 0.8284515738487244, "num_tokens": 40679036.0, "step": 847 }, { "entropy": 0.8497120440006256, "epoch": 1.3066255778120186, "grad_norm": 0.8843027949333191, "learning_rate": 1.2937886003771293e-05, "loss": 0.8298, "mean_token_accuracy": 0.7785083055496216, "num_tokens": 40689746.0, "step": 848 }, { "entropy": 0.7782982587814331, "epoch": 1.308166409861325, "grad_norm": 0.9840623140335083, "learning_rate": 1.2921640800228596e-05, "loss": 0.7625, "mean_token_accuracy": 0.7978587448596954, "num_tokens": 40698048.0, "step": 849 }, { "entropy": 0.2738802433013916, "epoch": 1.3097072419106317, "grad_norm": 0.5794438719749451, "learning_rate": 1.2905387162316081e-05, "loss": 0.2454, "mean_token_accuracy": 0.9354879260063171, "num_tokens": 40763476.0, "step": 850 }, { "entropy": 0.4480188488960266, "epoch": 1.3112480739599384, "grad_norm": 0.7069799304008484, "learning_rate": 1.288912513695574e-05, "loss": 0.414, "mean_token_accuracy": 0.8867010176181793, "num_tokens": 40826259.0, "step": 851 }, { "entropy": 0.481309249997139, "epoch": 1.312788906009245, "grad_norm": 0.789355993270874, "learning_rate": 1.2872854771093766e-05, "loss": 0.4631, "mean_token_accuracy": 0.8770294487476349, "num_tokens": 40886820.0, "step": 852 }, { "entropy": 0.46446023881435394, "epoch": 1.3143297380585517, "grad_norm": 0.8139018416404724, "learning_rate": 1.285657611170045e-05, "loss": 0.4529, "mean_token_accuracy": 0.8790455162525177, "num_tokens": 40945829.0, "step": 853 }, { "entropy": 0.4370674341917038, "epoch": 1.3158705701078581, "grad_norm": 0.7433444857597351, "learning_rate": 1.2840289205769998e-05, "loss": 0.4194, "mean_token_accuracy": 0.8887923657894135, "num_tokens": 41003629.0, "step": 854 }, { "entropy": 0.41878949105739594, "epoch": 1.3174114021571648, "grad_norm": 0.7435742020606995, "learning_rate": 1.2823994100320448e-05, "loss": 0.3999, "mean_token_accuracy": 0.8961624801158905, "num_tokens": 41060262.0, "step": 855 }, { "entropy": 0.44066862761974335, "epoch": 1.3189522342064715, "grad_norm": 0.7865291237831116, "learning_rate": 1.2807690842393494e-05, "loss": 0.4175, "mean_token_accuracy": 0.8918332457542419, "num_tokens": 41115799.0, "step": 856 }, { "entropy": 0.39282794296741486, "epoch": 1.3204930662557781, "grad_norm": 0.803318202495575, "learning_rate": 1.2791379479054373e-05, "loss": 0.3603, "mean_token_accuracy": 0.9052494168281555, "num_tokens": 41170357.0, "step": 857 }, { "entropy": 0.41222554445266724, "epoch": 1.3220338983050848, "grad_norm": 0.7733611464500427, "learning_rate": 1.2775060057391716e-05, "loss": 0.377, "mean_token_accuracy": 0.9018443822860718, "num_tokens": 41223252.0, "step": 858 }, { "entropy": 0.4220001995563507, "epoch": 1.3235747303543914, "grad_norm": 0.8222111463546753, "learning_rate": 1.2758732624517419e-05, "loss": 0.3954, "mean_token_accuracy": 0.8987040817737579, "num_tokens": 41273020.0, "step": 859 }, { "entropy": 0.8162117600440979, "epoch": 1.325115562403698, "grad_norm": 0.7654094696044922, "learning_rate": 1.2742397227566503e-05, "loss": 0.833, "mean_token_accuracy": 0.7689952254295349, "num_tokens": 41286417.0, "step": 860 }, { "entropy": 0.7719217538833618, "epoch": 1.3266563944530045, "grad_norm": 0.9776811003684998, "learning_rate": 1.2726053913696988e-05, "loss": 0.7811, "mean_token_accuracy": 0.7979236841201782, "num_tokens": 41296008.0, "step": 861 }, { "entropy": 0.4699357897043228, "epoch": 1.3281972265023112, "grad_norm": 0.6748993992805481, "learning_rate": 1.2709702730089737e-05, "loss": 0.289, "mean_token_accuracy": 0.8834750652313232, "num_tokens": 41332374.0, "step": 862 }, { "entropy": 0.40801048278808594, "epoch": 1.3297380585516179, "grad_norm": 0.691907525062561, "learning_rate": 1.2693343723948342e-05, "loss": 0.3747, "mean_token_accuracy": 0.8976681530475616, "num_tokens": 41396349.0, "step": 863 }, { "entropy": 0.4460611492395401, "epoch": 1.3312788906009245, "grad_norm": 0.7525284290313721, "learning_rate": 1.2676976942498971e-05, "loss": 0.4233, "mean_token_accuracy": 0.8859855532646179, "num_tokens": 41457734.0, "step": 864 }, { "entropy": 0.3976954519748688, "epoch": 1.3328197226502312, "grad_norm": 0.7365103363990784, "learning_rate": 1.2660602432990245e-05, "loss": 0.3793, "mean_token_accuracy": 0.8987604379653931, "num_tokens": 41517469.0, "step": 865 }, { "entropy": 0.3683518022298813, "epoch": 1.3343605546995376, "grad_norm": 0.7921053767204285, "learning_rate": 1.2644220242693088e-05, "loss": 0.3482, "mean_token_accuracy": 0.9066147804260254, "num_tokens": 41575631.0, "step": 866 }, { "entropy": 0.4082944989204407, "epoch": 1.3359013867488443, "grad_norm": 0.7084032297134399, "learning_rate": 1.2627830418900601e-05, "loss": 0.3843, "mean_token_accuracy": 0.8989914059638977, "num_tokens": 41632796.0, "step": 867 }, { "entropy": 0.4350254088640213, "epoch": 1.337442218798151, "grad_norm": 0.7979709506034851, "learning_rate": 1.2611433008927925e-05, "loss": 0.4167, "mean_token_accuracy": 0.8907610774040222, "num_tokens": 41688795.0, "step": 868 }, { "entropy": 0.4128717631101608, "epoch": 1.3389830508474576, "grad_norm": 0.7808943390846252, "learning_rate": 1.2595028060112094e-05, "loss": 0.3926, "mean_token_accuracy": 0.8977866768836975, "num_tokens": 41743641.0, "step": 869 }, { "entropy": 0.471362829208374, "epoch": 1.3405238828967643, "grad_norm": 0.8013387322425842, "learning_rate": 1.2578615619811905e-05, "loss": 0.4473, "mean_token_accuracy": 0.8849197924137115, "num_tokens": 41797149.0, "step": 870 }, { "entropy": 0.38501258194446564, "epoch": 1.342064714946071, "grad_norm": 0.7567100524902344, "learning_rate": 1.2562195735407792e-05, "loss": 0.344, "mean_token_accuracy": 0.9102269113063812, "num_tokens": 41848638.0, "step": 871 }, { "entropy": 0.7385390102863312, "epoch": 1.3436055469953776, "grad_norm": 1.0213241577148438, "learning_rate": 1.254576845430167e-05, "loss": 0.5706, "mean_token_accuracy": 0.819618433713913, "num_tokens": 41870647.0, "step": 872 }, { "entropy": 0.8167250752449036, "epoch": 1.3451463790446843, "grad_norm": 0.7880586981773376, "learning_rate": 1.2529333823916807e-05, "loss": 0.8027, "mean_token_accuracy": 0.7863408327102661, "num_tokens": 41881182.0, "step": 873 }, { "entropy": 0.7786854207515717, "epoch": 1.3466872110939907, "grad_norm": 0.9015839099884033, "learning_rate": 1.25128918916977e-05, "loss": 0.7487, "mean_token_accuracy": 0.8104383945465088, "num_tokens": 41889180.0, "step": 874 }, { "entropy": 0.27611875534057617, "epoch": 1.3482280431432974, "grad_norm": 0.5754743814468384, "learning_rate": 1.2496442705109908e-05, "loss": 0.2387, "mean_token_accuracy": 0.9362891614437103, "num_tokens": 41954547.0, "step": 875 }, { "entropy": 0.4768066257238388, "epoch": 1.349768875192604, "grad_norm": 0.7236468195915222, "learning_rate": 1.2479986311639943e-05, "loss": 0.4499, "mean_token_accuracy": 0.8818181753158569, "num_tokens": 42017129.0, "step": 876 }, { "entropy": 0.38842248916625977, "epoch": 1.3513097072419107, "grad_norm": 0.7140432596206665, "learning_rate": 1.2463522758795117e-05, "loss": 0.3631, "mean_token_accuracy": 0.9041881263256073, "num_tokens": 42077340.0, "step": 877 }, { "entropy": 0.4372783750295639, "epoch": 1.3528505392912171, "grad_norm": 0.7644354701042175, "learning_rate": 1.2447052094103416e-05, "loss": 0.4179, "mean_token_accuracy": 0.8868769109249115, "num_tokens": 42136103.0, "step": 878 }, { "entropy": 0.4129698872566223, "epoch": 1.3543913713405238, "grad_norm": 0.8230661153793335, "learning_rate": 1.2430574365113352e-05, "loss": 0.4018, "mean_token_accuracy": 0.8951138556003571, "num_tokens": 42193683.0, "step": 879 }, { "entropy": 0.4583929628133774, "epoch": 1.3559322033898304, "grad_norm": 0.8477357625961304, "learning_rate": 1.2414089619393832e-05, "loss": 0.4446, "mean_token_accuracy": 0.8850149512290955, "num_tokens": 42250166.0, "step": 880 }, { "entropy": 0.42514078319072723, "epoch": 1.357473035439137, "grad_norm": 0.7742024660110474, "learning_rate": 1.239759790453402e-05, "loss": 0.4034, "mean_token_accuracy": 0.8939701318740845, "num_tokens": 42305422.0, "step": 881 }, { "entropy": 0.43692414462566376, "epoch": 1.3590138674884438, "grad_norm": 0.7721869349479675, "learning_rate": 1.2381099268143199e-05, "loss": 0.4041, "mean_token_accuracy": 0.8935768902301788, "num_tokens": 42359509.0, "step": 882 }, { "entropy": 0.4236985445022583, "epoch": 1.3605546995377504, "grad_norm": 0.7781355977058411, "learning_rate": 1.236459375785063e-05, "loss": 0.3964, "mean_token_accuracy": 0.8964767456054688, "num_tokens": 42411962.0, "step": 883 }, { "entropy": 0.4711737185716629, "epoch": 1.362095531587057, "grad_norm": 0.7977170944213867, "learning_rate": 1.2348081421305427e-05, "loss": 0.3976, "mean_token_accuracy": 0.8948805630207062, "num_tokens": 42454491.0, "step": 884 }, { "entropy": 0.8195341527462006, "epoch": 1.3636363636363638, "grad_norm": 0.7822806239128113, "learning_rate": 1.2331562306176395e-05, "loss": 0.8376, "mean_token_accuracy": 0.7726299166679382, "num_tokens": 42466388.0, "step": 885 }, { "entropy": 0.7698968052864075, "epoch": 1.3651771956856702, "grad_norm": 0.9670037031173706, "learning_rate": 1.2315036460151926e-05, "loss": 0.7678, "mean_token_accuracy": 0.803615003824234, "num_tokens": 42475584.0, "step": 886 }, { "entropy": 0.4222251623868942, "epoch": 1.3667180277349769, "grad_norm": 0.7450445294380188, "learning_rate": 1.2298503930939829e-05, "loss": 0.2356, "mean_token_accuracy": 0.8953434526920319, "num_tokens": 42511948.0, "step": 887 }, { "entropy": 0.3312455415725708, "epoch": 1.3682588597842835, "grad_norm": 0.6879040002822876, "learning_rate": 1.2281964766267216e-05, "loss": 0.3053, "mean_token_accuracy": 0.9169019460678101, "num_tokens": 42576388.0, "step": 888 }, { "entropy": 0.3973444849252701, "epoch": 1.3697996918335902, "grad_norm": 0.751072883605957, "learning_rate": 1.2265419013880346e-05, "loss": 0.3795, "mean_token_accuracy": 0.8975928425788879, "num_tokens": 42637890.0, "step": 889 }, { "entropy": 0.3813963383436203, "epoch": 1.3713405238828968, "grad_norm": 0.7097960710525513, "learning_rate": 1.2248866721544505e-05, "loss": 0.3581, "mean_token_accuracy": 0.9052862524986267, "num_tokens": 42697376.0, "step": 890 }, { "entropy": 0.41053755581378937, "epoch": 1.3728813559322033, "grad_norm": 0.7727629542350769, "learning_rate": 1.2232307937043848e-05, "loss": 0.3863, "mean_token_accuracy": 0.8964643180370331, "num_tokens": 42755446.0, "step": 891 }, { "entropy": 0.469132199883461, "epoch": 1.37442218798151, "grad_norm": 0.748324453830719, "learning_rate": 1.2215742708181286e-05, "loss": 0.439, "mean_token_accuracy": 0.8842993676662445, "num_tokens": 42812541.0, "step": 892 }, { "entropy": 0.44538769125938416, "epoch": 1.3759630200308166, "grad_norm": 0.7505574822425842, "learning_rate": 1.2199171082778318e-05, "loss": 0.4241, "mean_token_accuracy": 0.89022496342659, "num_tokens": 42868610.0, "step": 893 }, { "entropy": 0.40506914258003235, "epoch": 1.3775038520801233, "grad_norm": 0.7094132304191589, "learning_rate": 1.2182593108674925e-05, "loss": 0.387, "mean_token_accuracy": 0.9000474810600281, "num_tokens": 42923683.0, "step": 894 }, { "entropy": 0.40512815117836, "epoch": 1.37904468412943, "grad_norm": 0.8306067585945129, "learning_rate": 1.2166008833729399e-05, "loss": 0.3843, "mean_token_accuracy": 0.899463564157486, "num_tokens": 42977714.0, "step": 895 }, { "entropy": 0.3866468518972397, "epoch": 1.3805855161787366, "grad_norm": 0.7784432768821716, "learning_rate": 1.2149418305818237e-05, "loss": 0.3609, "mean_token_accuracy": 0.9052821397781372, "num_tokens": 43030217.0, "step": 896 }, { "entropy": 0.4827163517475128, "epoch": 1.3821263482280433, "grad_norm": 0.8654159307479858, "learning_rate": 1.213282157283598e-05, "loss": 0.3963, "mean_token_accuracy": 0.8923806250095367, "num_tokens": 43070331.0, "step": 897 }, { "entropy": 0.8445702791213989, "epoch": 1.3836671802773497, "grad_norm": 0.8935002684593201, "learning_rate": 1.2116218682695083e-05, "loss": 0.8357, "mean_token_accuracy": 0.7772089838981628, "num_tokens": 43081653.0, "step": 898 }, { "entropy": 0.7546714246273041, "epoch": 1.3852080123266564, "grad_norm": 0.9885482788085938, "learning_rate": 1.2099609683325775e-05, "loss": 0.739, "mean_token_accuracy": 0.815660685300827, "num_tokens": 43090047.0, "step": 899 }, { "entropy": 0.32818150520324707, "epoch": 1.386748844375963, "grad_norm": 0.6306811571121216, "learning_rate": 1.2082994622675922e-05, "loss": 0.2899, "mean_token_accuracy": 0.922254353761673, "num_tokens": 43155202.0, "step": 900 }, { "entropy": 0.46521347761154175, "epoch": 1.3882896764252697, "grad_norm": 0.7398255467414856, "learning_rate": 1.2066373548710887e-05, "loss": 0.4349, "mean_token_accuracy": 0.8845932483673096, "num_tokens": 43217789.0, "step": 901 }, { "entropy": 0.39698898792266846, "epoch": 1.3898305084745763, "grad_norm": 0.724245548248291, "learning_rate": 1.2049746509413402e-05, "loss": 0.3746, "mean_token_accuracy": 0.9008102416992188, "num_tokens": 43278384.0, "step": 902 }, { "entropy": 0.3833140581846237, "epoch": 1.3913713405238828, "grad_norm": 0.7090548872947693, "learning_rate": 1.2033113552783402e-05, "loss": 0.3615, "mean_token_accuracy": 0.9017447531223297, "num_tokens": 43337777.0, "step": 903 }, { "entropy": 0.36817893385887146, "epoch": 1.3929121725731894, "grad_norm": 0.7253274321556091, "learning_rate": 1.2016474726837921e-05, "loss": 0.3419, "mean_token_accuracy": 0.9086888432502747, "num_tokens": 43396118.0, "step": 904 }, { "entropy": 0.39049533009529114, "epoch": 1.394453004622496, "grad_norm": 0.7645336985588074, "learning_rate": 1.1999830079610928e-05, "loss": 0.3717, "mean_token_accuracy": 0.9027009606361389, "num_tokens": 43453064.0, "step": 905 }, { "entropy": 0.4061228930950165, "epoch": 1.3959938366718028, "grad_norm": 0.8080176115036011, "learning_rate": 1.1983179659153203e-05, "loss": 0.3902, "mean_token_accuracy": 0.8980417549610138, "num_tokens": 43508861.0, "step": 906 }, { "entropy": 0.39780253171920776, "epoch": 1.3975346687211094, "grad_norm": 0.774815022945404, "learning_rate": 1.1966523513532187e-05, "loss": 0.3817, "mean_token_accuracy": 0.8998930156230927, "num_tokens": 43563514.0, "step": 907 }, { "entropy": 0.39100252091884613, "epoch": 1.399075500770416, "grad_norm": 0.7468082308769226, "learning_rate": 1.194986169083185e-05, "loss": 0.3597, "mean_token_accuracy": 0.9072011709213257, "num_tokens": 43616887.0, "step": 908 }, { "entropy": 0.36847782135009766, "epoch": 1.4006163328197228, "grad_norm": 0.7422658205032349, "learning_rate": 1.1933194239152553e-05, "loss": 0.3329, "mean_token_accuracy": 0.9105934500694275, "num_tokens": 43668338.0, "step": 909 }, { "entropy": 0.7146350741386414, "epoch": 1.4021571648690292, "grad_norm": 0.9117757678031921, "learning_rate": 1.191652120661091e-05, "loss": 0.5754, "mean_token_accuracy": 0.821476936340332, "num_tokens": 43690132.0, "step": 910 }, { "entropy": 0.8205354809761047, "epoch": 1.4036979969183359, "grad_norm": 0.9925346970558167, "learning_rate": 1.1899842641339635e-05, "loss": 0.82, "mean_token_accuracy": 0.787765234708786, "num_tokens": 43699600.0, "step": 911 }, { "entropy": 0.4861857295036316, "epoch": 1.4052388289676425, "grad_norm": 0.6760313510894775, "learning_rate": 1.1883158591487429e-05, "loss": 0.2877, "mean_token_accuracy": 0.8770120739936829, "num_tokens": 43735789.0, "step": 912 }, { "entropy": 0.3503331243991852, "epoch": 1.4067796610169492, "grad_norm": 0.6076130867004395, "learning_rate": 1.1866469105218811e-05, "loss": 0.3255, "mean_token_accuracy": 0.9108181297779083, "num_tokens": 43800085.0, "step": 913 }, { "entropy": 0.40374404191970825, "epoch": 1.4083204930662558, "grad_norm": 0.7126924991607666, "learning_rate": 1.1849774230714011e-05, "loss": 0.3823, "mean_token_accuracy": 0.8982253670692444, "num_tokens": 43861431.0, "step": 914 }, { "entropy": 0.454937145113945, "epoch": 1.4098613251155623, "grad_norm": 0.75928795337677, "learning_rate": 1.1833074016168798e-05, "loss": 0.4369, "mean_token_accuracy": 0.8855842649936676, "num_tokens": 43921132.0, "step": 915 }, { "entropy": 0.40782007575035095, "epoch": 1.411402157164869, "grad_norm": 0.7798514366149902, "learning_rate": 1.1816368509794365e-05, "loss": 0.3928, "mean_token_accuracy": 0.8965611159801483, "num_tokens": 43979758.0, "step": 916 }, { "entropy": 0.4174739420413971, "epoch": 1.4129429892141756, "grad_norm": 0.8058932423591614, "learning_rate": 1.1799657759817184e-05, "loss": 0.4067, "mean_token_accuracy": 0.8925335109233856, "num_tokens": 44036939.0, "step": 917 }, { "entropy": 0.40514054894447327, "epoch": 1.4144838212634823, "grad_norm": 0.7427037954330444, "learning_rate": 1.1782941814478858e-05, "loss": 0.382, "mean_token_accuracy": 0.9008095264434814, "num_tokens": 44092877.0, "step": 918 }, { "entropy": 0.4333883374929428, "epoch": 1.416024653312789, "grad_norm": 0.7668015360832214, "learning_rate": 1.1766220722035993e-05, "loss": 0.4101, "mean_token_accuracy": 0.8929409682750702, "num_tokens": 44147830.0, "step": 919 }, { "entropy": 0.42132972180843353, "epoch": 1.4175654853620956, "grad_norm": 0.7287042140960693, "learning_rate": 1.1749494530760052e-05, "loss": 0.399, "mean_token_accuracy": 0.8993971347808838, "num_tokens": 44201608.0, "step": 920 }, { "entropy": 0.3894237130880356, "epoch": 1.4191063174114023, "grad_norm": 0.7756037712097168, "learning_rate": 1.1732763288937219e-05, "loss": 0.3543, "mean_token_accuracy": 0.9097053408622742, "num_tokens": 44253458.0, "step": 921 }, { "entropy": 0.6754321753978729, "epoch": 1.420647149460709, "grad_norm": 0.9316450953483582, "learning_rate": 1.1716027044868257e-05, "loss": 0.5151, "mean_token_accuracy": 0.8225883543491364, "num_tokens": 44283339.0, "step": 922 }, { "entropy": 0.8336122632026672, "epoch": 1.4221879815100154, "grad_norm": 0.8210532069206238, "learning_rate": 1.169928584686837e-05, "loss": 0.8229, "mean_token_accuracy": 0.7762010395526886, "num_tokens": 44294165.0, "step": 923 }, { "entropy": 0.7512508928775787, "epoch": 1.423728813559322, "grad_norm": 1.0454026460647583, "learning_rate": 1.1682539743267063e-05, "loss": 0.7144, "mean_token_accuracy": 0.8158195912837982, "num_tokens": 44301988.0, "step": 924 }, { "entropy": 0.27729617059230804, "epoch": 1.4252696456086287, "grad_norm": 0.5692864656448364, "learning_rate": 1.1665788782408006e-05, "loss": 0.2563, "mean_token_accuracy": 0.9310206174850464, "num_tokens": 44367306.0, "step": 925 }, { "entropy": 0.4247916489839554, "epoch": 1.4268104776579353, "grad_norm": 0.756807804107666, "learning_rate": 1.1649033012648885e-05, "loss": 0.4039, "mean_token_accuracy": 0.8925371468067169, "num_tokens": 44429615.0, "step": 926 }, { "entropy": 0.3788754791021347, "epoch": 1.4283513097072418, "grad_norm": 0.693475604057312, "learning_rate": 1.1632272482361273e-05, "loss": 0.3503, "mean_token_accuracy": 0.9060869812965393, "num_tokens": 44490032.0, "step": 927 }, { "entropy": 0.4086735248565674, "epoch": 1.4298921417565484, "grad_norm": 0.7713301181793213, "learning_rate": 1.1615507239930487e-05, "loss": 0.397, "mean_token_accuracy": 0.896946907043457, "num_tokens": 44549165.0, "step": 928 }, { "entropy": 0.37942081689834595, "epoch": 1.431432973805855, "grad_norm": 0.8043217062950134, "learning_rate": 1.1598737333755443e-05, "loss": 0.3598, "mean_token_accuracy": 0.9031563103199005, "num_tokens": 44607112.0, "step": 929 }, { "entropy": 0.3588341772556305, "epoch": 1.4329738058551618, "grad_norm": 0.7778260707855225, "learning_rate": 1.1581962812248524e-05, "loss": 0.3456, "mean_token_accuracy": 0.9091442823410034, "num_tokens": 44663848.0, "step": 930 }, { "entropy": 0.4175126850605011, "epoch": 1.4345146379044684, "grad_norm": 0.7695613503456116, "learning_rate": 1.1565183723835436e-05, "loss": 0.3936, "mean_token_accuracy": 0.8993442952632904, "num_tokens": 44719472.0, "step": 931 }, { "entropy": 0.42201538383960724, "epoch": 1.436055469953775, "grad_norm": 0.7489761710166931, "learning_rate": 1.1548400116955065e-05, "loss": 0.3882, "mean_token_accuracy": 0.8985090255737305, "num_tokens": 44773741.0, "step": 932 }, { "entropy": 0.42232660949230194, "epoch": 1.4375963020030817, "grad_norm": 0.8299871683120728, "learning_rate": 1.1531612040059355e-05, "loss": 0.4031, "mean_token_accuracy": 0.8977280259132385, "num_tokens": 44826507.0, "step": 933 }, { "entropy": 0.4164632707834244, "epoch": 1.4391371340523884, "grad_norm": 0.7929196357727051, "learning_rate": 1.1514819541613133e-05, "loss": 0.3426, "mean_token_accuracy": 0.9099384844303131, "num_tokens": 44871579.0, "step": 934 }, { "entropy": 0.8137744665145874, "epoch": 1.4406779661016949, "grad_norm": 0.8349602222442627, "learning_rate": 1.1498022670094011e-05, "loss": 0.836, "mean_token_accuracy": 0.776858389377594, "num_tokens": 44883402.0, "step": 935 }, { "entropy": 0.7474218606948853, "epoch": 1.4422187981510015, "grad_norm": 1.1329175233840942, "learning_rate": 1.148122147399221e-05, "loss": 0.7506, "mean_token_accuracy": 0.8033492565155029, "num_tokens": 44892335.0, "step": 936 }, { "entropy": 0.4825110137462616, "epoch": 1.4437596302003082, "grad_norm": 0.6411798000335693, "learning_rate": 1.1464416001810443e-05, "loss": 0.2616, "mean_token_accuracy": 0.880749374628067, "num_tokens": 44928661.0, "step": 937 }, { "entropy": 0.3676704466342926, "epoch": 1.4453004622496148, "grad_norm": 0.6815663576126099, "learning_rate": 1.144760630206377e-05, "loss": 0.3391, "mean_token_accuracy": 0.9089425504207611, "num_tokens": 44992656.0, "step": 938 }, { "entropy": 0.4128652364015579, "epoch": 1.4468412942989215, "grad_norm": 0.745421290397644, "learning_rate": 1.1430792423279444e-05, "loss": 0.3815, "mean_token_accuracy": 0.8975734114646912, "num_tokens": 45054155.0, "step": 939 }, { "entropy": 0.3845049738883972, "epoch": 1.448382126348228, "grad_norm": 0.7619598507881165, "learning_rate": 1.1413974413996794e-05, "loss": 0.3542, "mean_token_accuracy": 0.9027822017669678, "num_tokens": 45114116.0, "step": 940 }, { "entropy": 0.39585810899734497, "epoch": 1.4499229583975346, "grad_norm": 0.7438348531723022, "learning_rate": 1.1397152322767073e-05, "loss": 0.3752, "mean_token_accuracy": 0.9013730883598328, "num_tokens": 45172668.0, "step": 941 }, { "entropy": 0.36634060740470886, "epoch": 1.4514637904468413, "grad_norm": 0.7499401569366455, "learning_rate": 1.1380326198153307e-05, "loss": 0.3467, "mean_token_accuracy": 0.909209817647934, "num_tokens": 45230063.0, "step": 942 }, { "entropy": 0.37479494512081146, "epoch": 1.453004622496148, "grad_norm": 0.7598559856414795, "learning_rate": 1.1363496088730177e-05, "loss": 0.354, "mean_token_accuracy": 0.9076713025569916, "num_tokens": 45286414.0, "step": 943 }, { "entropy": 0.3652294874191284, "epoch": 1.4545454545454546, "grad_norm": 0.7774335145950317, "learning_rate": 1.134666204308386e-05, "loss": 0.3389, "mean_token_accuracy": 0.9108863174915314, "num_tokens": 45341389.0, "step": 944 }, { "entropy": 0.36526212096214294, "epoch": 1.4560862865947612, "grad_norm": 0.747968852519989, "learning_rate": 1.1329824109811903e-05, "loss": 0.3408, "mean_token_accuracy": 0.9113625586032867, "num_tokens": 45395057.0, "step": 945 }, { "entropy": 0.3238263577222824, "epoch": 1.457627118644068, "grad_norm": 0.7077100872993469, "learning_rate": 1.131298233752307e-05, "loss": 0.2909, "mean_token_accuracy": 0.9227835237979889, "num_tokens": 45446828.0, "step": 946 }, { "entropy": 0.6864329278469086, "epoch": 1.4591679506933743, "grad_norm": 0.8856323957443237, "learning_rate": 1.1296136774837214e-05, "loss": 0.5353, "mean_token_accuracy": 0.824316531419754, "num_tokens": 45473587.0, "step": 947 }, { "entropy": 0.8172440528869629, "epoch": 1.460708782742681, "grad_norm": 0.917468786239624, "learning_rate": 1.1279287470385126e-05, "loss": 0.8051, "mean_token_accuracy": 0.785345733165741, "num_tokens": 45484287.0, "step": 948 }, { "entropy": 0.7301540672779083, "epoch": 1.4622496147919877, "grad_norm": 1.0162670612335205, "learning_rate": 1.1262434472808396e-05, "loss": 0.7379, "mean_token_accuracy": 0.8171815872192383, "num_tokens": 45492196.0, "step": 949 }, { "entropy": 0.2563215270638466, "epoch": 1.4637904468412943, "grad_norm": 0.6327465176582336, "learning_rate": 1.1245577830759283e-05, "loss": 0.2365, "mean_token_accuracy": 0.9373209178447723, "num_tokens": 45557581.0, "step": 950 }, { "entropy": 0.35213179886341095, "epoch": 1.465331278890601, "grad_norm": 0.6872291564941406, "learning_rate": 1.1228717592900562e-05, "loss": 0.3362, "mean_token_accuracy": 0.9097125232219696, "num_tokens": 45620047.0, "step": 951 }, { "entropy": 0.417670413851738, "epoch": 1.4668721109399074, "grad_norm": 0.7068012952804565, "learning_rate": 1.1211853807905388e-05, "loss": 0.3993, "mean_token_accuracy": 0.8955115675926208, "num_tokens": 45680823.0, "step": 952 }, { "entropy": 0.37370190024375916, "epoch": 1.468412942989214, "grad_norm": 0.7390000224113464, "learning_rate": 1.1194986524457165e-05, "loss": 0.354, "mean_token_accuracy": 0.9048081338405609, "num_tokens": 45740336.0, "step": 953 }, { "entropy": 0.4056796580553055, "epoch": 1.4699537750385208, "grad_norm": 0.7538642883300781, "learning_rate": 1.117811579124938e-05, "loss": 0.3859, "mean_token_accuracy": 0.8975527882575989, "num_tokens": 45798857.0, "step": 954 }, { "entropy": 0.40699411928653717, "epoch": 1.4714946070878274, "grad_norm": 0.7222437262535095, "learning_rate": 1.1161241656985498e-05, "loss": 0.3864, "mean_token_accuracy": 0.8986087739467621, "num_tokens": 45856109.0, "step": 955 }, { "entropy": 0.36612842977046967, "epoch": 1.473035439137134, "grad_norm": 0.7051148414611816, "learning_rate": 1.1144364170378784e-05, "loss": 0.3385, "mean_token_accuracy": 0.912442684173584, "num_tokens": 45912149.0, "step": 956 }, { "entropy": 0.4002978354692459, "epoch": 1.4745762711864407, "grad_norm": 0.7502301931381226, "learning_rate": 1.1127483380152198e-05, "loss": 0.3681, "mean_token_accuracy": 0.9032047986984253, "num_tokens": 45966801.0, "step": 957 }, { "entropy": 0.3787462115287781, "epoch": 1.4761171032357474, "grad_norm": 0.774571418762207, "learning_rate": 1.1110599335038221e-05, "loss": 0.3545, "mean_token_accuracy": 0.908636599779129, "num_tokens": 46019809.0, "step": 958 }, { "entropy": 0.3629244267940521, "epoch": 1.4776579352850538, "grad_norm": 0.8472306728363037, "learning_rate": 1.1093712083778748e-05, "loss": 0.3421, "mean_token_accuracy": 0.9125185012817383, "num_tokens": 46069973.0, "step": 959 }, { "entropy": 0.8554102182388306, "epoch": 1.4791987673343605, "grad_norm": 0.8839317560195923, "learning_rate": 1.1076821675124906e-05, "loss": 0.8414, "mean_token_accuracy": 0.7659036815166473, "num_tokens": 46083761.0, "step": 960 }, { "entropy": 0.8131800293922424, "epoch": 1.4807395993836672, "grad_norm": 0.9876549243927002, "learning_rate": 1.1059928157836965e-05, "loss": 0.7975, "mean_token_accuracy": 0.788355678319931, "num_tokens": 46093612.0, "step": 961 }, { "entropy": 0.48841363936662674, "epoch": 1.4822804314329738, "grad_norm": 0.6355686187744141, "learning_rate": 1.1043031580684147e-05, "loss": 0.2554, "mean_token_accuracy": 0.8803221881389618, "num_tokens": 46130163.0, "step": 962 }, { "entropy": 0.288593053817749, "epoch": 1.4838212634822805, "grad_norm": 0.5839232206344604, "learning_rate": 1.1026131992444517e-05, "loss": 0.2693, "mean_token_accuracy": 0.928693026304245, "num_tokens": 46194923.0, "step": 963 }, { "entropy": 0.39798495173454285, "epoch": 1.485362095531587, "grad_norm": 0.7667365074157715, "learning_rate": 1.1009229441904831e-05, "loss": 0.3762, "mean_token_accuracy": 0.8990342915058136, "num_tokens": 46257227.0, "step": 964 }, { "entropy": 0.38045597076416016, "epoch": 1.4869029275808936, "grad_norm": 0.7131215333938599, "learning_rate": 1.0992323977860399e-05, "loss": 0.3511, "mean_token_accuracy": 0.9059851169586182, "num_tokens": 46317856.0, "step": 965 }, { "entropy": 0.3962727040052414, "epoch": 1.4884437596302003, "grad_norm": 0.74609375, "learning_rate": 1.0975415649114938e-05, "loss": 0.3802, "mean_token_accuracy": 0.8992033004760742, "num_tokens": 46377036.0, "step": 966 }, { "entropy": 0.34950587153434753, "epoch": 1.489984591679507, "grad_norm": 0.8463436365127563, "learning_rate": 1.0958504504480434e-05, "loss": 0.3379, "mean_token_accuracy": 0.9116087853908539, "num_tokens": 46434781.0, "step": 967 }, { "entropy": 0.36471377313137054, "epoch": 1.4915254237288136, "grad_norm": 0.8563871383666992, "learning_rate": 1.0941590592777007e-05, "loss": 0.3557, "mean_token_accuracy": 0.9060058891773224, "num_tokens": 46491469.0, "step": 968 }, { "entropy": 0.40816226601600647, "epoch": 1.4930662557781202, "grad_norm": 0.8167893290519714, "learning_rate": 1.0924673962832768e-05, "loss": 0.3846, "mean_token_accuracy": 0.8983428478240967, "num_tokens": 46547011.0, "step": 969 }, { "entropy": 0.41549915075302124, "epoch": 1.494607087827427, "grad_norm": 0.8257809281349182, "learning_rate": 1.090775466348366e-05, "loss": 0.3965, "mean_token_accuracy": 0.8992454707622528, "num_tokens": 46601179.0, "step": 970 }, { "entropy": 0.4026498645544052, "epoch": 1.4961479198767336, "grad_norm": 0.759606659412384, "learning_rate": 1.0890832743573351e-05, "loss": 0.3697, "mean_token_accuracy": 0.903455525636673, "num_tokens": 46653543.0, "step": 971 }, { "entropy": 0.49864207208156586, "epoch": 1.49768875192604, "grad_norm": 0.7768622040748596, "learning_rate": 1.087390825195306e-05, "loss": 0.376, "mean_token_accuracy": 0.8915912806987762, "num_tokens": 46691837.0, "step": 972 }, { "entropy": 0.8279995918273926, "epoch": 1.4992295839753467, "grad_norm": 0.8682583570480347, "learning_rate": 1.0856981237481439e-05, "loss": 0.8251, "mean_token_accuracy": 0.7789612114429474, "num_tokens": 46702935.0, "step": 973 }, { "entropy": 0.7947720289230347, "epoch": 1.5007704160246533, "grad_norm": 1.0684847831726074, "learning_rate": 1.0840051749024416e-05, "loss": 0.7971, "mean_token_accuracy": 0.7961967587471008, "num_tokens": 46711278.0, "step": 974 }, { "entropy": 0.25718145072460175, "epoch": 1.50231124807396, "grad_norm": 0.5429683923721313, "learning_rate": 1.0823119835455069e-05, "loss": 0.2298, "mean_token_accuracy": 0.9405795037746429, "num_tokens": 46776276.0, "step": 975 }, { "entropy": 0.3737444281578064, "epoch": 1.5038520801232664, "grad_norm": 0.7193872928619385, "learning_rate": 1.0806185545653472e-05, "loss": 0.3396, "mean_token_accuracy": 0.9088635146617889, "num_tokens": 46838635.0, "step": 976 }, { "entropy": 0.34834375977516174, "epoch": 1.505392912172573, "grad_norm": 0.6925857067108154, "learning_rate": 1.0789248928506557e-05, "loss": 0.3171, "mean_token_accuracy": 0.914468914270401, "num_tokens": 46898949.0, "step": 977 }, { "entropy": 0.3738424777984619, "epoch": 1.5069337442218798, "grad_norm": 0.7419418096542358, "learning_rate": 1.0772310032907976e-05, "loss": 0.3552, "mean_token_accuracy": 0.9043752253055573, "num_tokens": 46957818.0, "step": 978 }, { "entropy": 0.40171681344509125, "epoch": 1.5084745762711864, "grad_norm": 0.8369203209877014, "learning_rate": 1.0755368907757969e-05, "loss": 0.3855, "mean_token_accuracy": 0.8979126811027527, "num_tokens": 47015564.0, "step": 979 }, { "entropy": 0.37567710876464844, "epoch": 1.510015408320493, "grad_norm": 0.8059343695640564, "learning_rate": 1.0738425601963192e-05, "loss": 0.3631, "mean_token_accuracy": 0.9055216014385223, "num_tokens": 47072229.0, "step": 980 }, { "entropy": 0.35540175437927246, "epoch": 1.5115562403697997, "grad_norm": 0.7904871702194214, "learning_rate": 1.0721480164436614e-05, "loss": 0.3377, "mean_token_accuracy": 0.9116830825805664, "num_tokens": 47127679.0, "step": 981 }, { "entropy": 0.362382709980011, "epoch": 1.5130970724191064, "grad_norm": 0.8135955929756165, "learning_rate": 1.0704532644097351e-05, "loss": 0.3314, "mean_token_accuracy": 0.9122683107852936, "num_tokens": 47181838.0, "step": 982 }, { "entropy": 0.3377160280942917, "epoch": 1.514637904468413, "grad_norm": 0.7314626574516296, "learning_rate": 1.068758308987053e-05, "loss": 0.3127, "mean_token_accuracy": 0.9181984066963196, "num_tokens": 47234568.0, "step": 983 }, { "entropy": 0.42299166321754456, "epoch": 1.5161787365177197, "grad_norm": 0.7578260898590088, "learning_rate": 1.0670631550687157e-05, "loss": 0.3383, "mean_token_accuracy": 0.9069967567920685, "num_tokens": 47276133.0, "step": 984 }, { "entropy": 0.8614127039909363, "epoch": 1.5177195685670262, "grad_norm": 0.8317596316337585, "learning_rate": 1.0653678075483955e-05, "loss": 0.8745, "mean_token_accuracy": 0.7693269848823547, "num_tokens": 47287907.0, "step": 985 }, { "entropy": 0.8086737096309662, "epoch": 1.5192604006163328, "grad_norm": 0.9432604908943176, "learning_rate": 1.063672271320325e-05, "loss": 0.7729, "mean_token_accuracy": 0.7933334708213806, "num_tokens": 47297273.0, "step": 986 }, { "entropy": 0.44593362510204315, "epoch": 1.5208012326656395, "grad_norm": 0.5813491344451904, "learning_rate": 1.0619765512792812e-05, "loss": 0.2112, "mean_token_accuracy": 0.8902988135814667, "num_tokens": 47333578.0, "step": 987 }, { "entropy": 0.3512614071369171, "epoch": 1.522342064714946, "grad_norm": 0.6390202641487122, "learning_rate": 1.0602806523205708e-05, "loss": 0.3199, "mean_token_accuracy": 0.9138855636119843, "num_tokens": 47397908.0, "step": 988 }, { "entropy": 0.34990255534648895, "epoch": 1.5238828967642526, "grad_norm": 0.6937991380691528, "learning_rate": 1.0585845793400184e-05, "loss": 0.3253, "mean_token_accuracy": 0.9132742285728455, "num_tokens": 47459658.0, "step": 989 }, { "entropy": 0.3791058659553528, "epoch": 1.5254237288135593, "grad_norm": 0.7132191061973572, "learning_rate": 1.0568883372339498e-05, "loss": 0.3575, "mean_token_accuracy": 0.9050051271915436, "num_tokens": 47519961.0, "step": 990 }, { "entropy": 0.37965862452983856, "epoch": 1.526964560862866, "grad_norm": 0.7970967292785645, "learning_rate": 1.05519193089918e-05, "loss": 0.3651, "mean_token_accuracy": 0.9041184484958649, "num_tokens": 47578894.0, "step": 991 }, { "entropy": 0.3858550190925598, "epoch": 1.5285053929121726, "grad_norm": 0.8011000752449036, "learning_rate": 1.0534953652329976e-05, "loss": 0.3794, "mean_token_accuracy": 0.9002048671245575, "num_tokens": 47636737.0, "step": 992 }, { "entropy": 0.3851163983345032, "epoch": 1.5300462249614792, "grad_norm": 0.8061558604240417, "learning_rate": 1.051798645133151e-05, "loss": 0.3623, "mean_token_accuracy": 0.9048239588737488, "num_tokens": 47693496.0, "step": 993 }, { "entropy": 0.38555918633937836, "epoch": 1.531587057010786, "grad_norm": 0.8438881039619446, "learning_rate": 1.0501017754978348e-05, "loss": 0.3683, "mean_token_accuracy": 0.904256135225296, "num_tokens": 47749013.0, "step": 994 }, { "entropy": 0.3430805206298828, "epoch": 1.5331278890600926, "grad_norm": 0.7053744792938232, "learning_rate": 1.0484047612256751e-05, "loss": 0.3175, "mean_token_accuracy": 0.9177311956882477, "num_tokens": 47803188.0, "step": 995 }, { "entropy": 0.3603765368461609, "epoch": 1.5346687211093992, "grad_norm": 0.8055171966552734, "learning_rate": 1.0467076072157159e-05, "loss": 0.3317, "mean_token_accuracy": 0.9144122004508972, "num_tokens": 47855582.0, "step": 996 }, { "entropy": 0.5837174504995346, "epoch": 1.5362095531587057, "grad_norm": 0.8311460018157959, "learning_rate": 1.0450103183674038e-05, "loss": 0.4246, "mean_token_accuracy": 0.8504509627819061, "num_tokens": 47888259.0, "step": 997 }, { "entropy": 0.844998687505722, "epoch": 1.5377503852080123, "grad_norm": 0.8031079173088074, "learning_rate": 1.0433128995805757e-05, "loss": 0.8209, "mean_token_accuracy": 0.7818261384963989, "num_tokens": 47899011.0, "step": 998 }, { "entropy": 0.7535492777824402, "epoch": 1.539291217257319, "grad_norm": 0.9656753540039062, "learning_rate": 1.0416153557554425e-05, "loss": 0.7243, "mean_token_accuracy": 0.8147108852863312, "num_tokens": 47907077.0, "step": 999 }, { "entropy": 0.25707101821899414, "epoch": 1.5408320493066254, "grad_norm": 0.542732298374176, "learning_rate": 1.0399176917925773e-05, "loss": 0.2401, "mean_token_accuracy": 0.9368680715560913, "num_tokens": 47972531.0, "step": 1000 }, { "epoch": 1.5408320493066254, "eval_entropy": 0.4848950245189217, "eval_loss": 0.46216994524002075, "eval_mean_token_accuracy": 0.8793219122684227, "eval_num_tokens": 47972531.0, "eval_runtime": 193.1191, "eval_samples_per_second": 4.391, "eval_steps_per_second": 1.098, "step": 1000 }, { "entropy": 0.3445664942264557, "epoch": 1.542372881355932, "grad_norm": 0.6921026110649109, "learning_rate": 1.0382199125928984e-05, "loss": 0.318, "mean_token_accuracy": 0.9153912663459778, "num_tokens": 48035311.0, "step": 1001 }, { "entropy": 0.4004209637641907, "epoch": 1.5439137134052388, "grad_norm": 0.7541795969009399, "learning_rate": 1.0365220230576592e-05, "loss": 0.3781, "mean_token_accuracy": 0.8989389538764954, "num_tokens": 48096082.0, "step": 1002 }, { "entropy": 0.39692945778369904, "epoch": 1.5454545454545454, "grad_norm": 0.707336962223053, "learning_rate": 1.0348240280884285e-05, "loss": 0.3712, "mean_token_accuracy": 0.9014990329742432, "num_tokens": 48155297.0, "step": 1003 }, { "entropy": 0.4170571565628052, "epoch": 1.546995377503852, "grad_norm": 0.8258360624313354, "learning_rate": 1.0331259325870824e-05, "loss": 0.4065, "mean_token_accuracy": 0.8948065042495728, "num_tokens": 48213433.0, "step": 1004 }, { "entropy": 0.3253302574157715, "epoch": 1.5485362095531587, "grad_norm": 0.7430322766304016, "learning_rate": 1.0314277414557854e-05, "loss": 0.308, "mean_token_accuracy": 0.9161421656608582, "num_tokens": 48270526.0, "step": 1005 }, { "entropy": 0.35776950418949127, "epoch": 1.5500770416024654, "grad_norm": 0.9052730798721313, "learning_rate": 1.0297294595969789e-05, "loss": 0.3489, "mean_token_accuracy": 0.9069467782974243, "num_tokens": 48326509.0, "step": 1006 }, { "entropy": 0.3406979739665985, "epoch": 1.551617873651772, "grad_norm": 0.7956295609474182, "learning_rate": 1.0280310919133657e-05, "loss": 0.3261, "mean_token_accuracy": 0.9148518741130829, "num_tokens": 48381381.0, "step": 1007 }, { "entropy": 0.3595014810562134, "epoch": 1.5531587057010787, "grad_norm": 0.7672415971755981, "learning_rate": 1.0263326433078966e-05, "loss": 0.3319, "mean_token_accuracy": 0.9143193960189819, "num_tokens": 48434774.0, "step": 1008 }, { "entropy": 0.33557215332984924, "epoch": 1.5546995377503852, "grad_norm": 0.7263293862342834, "learning_rate": 1.0246341186837558e-05, "loss": 0.3049, "mean_token_accuracy": 0.9215916097164154, "num_tokens": 48485890.0, "step": 1009 }, { "entropy": 0.7645518183708191, "epoch": 1.5562403697996918, "grad_norm": 0.8935337662696838, "learning_rate": 1.0229355229443483e-05, "loss": 0.6878, "mean_token_accuracy": 0.8076842725276947, "num_tokens": 48503105.0, "step": 1010 }, { "entropy": 0.7724018096923828, "epoch": 1.5577812018489985, "grad_norm": 1.1145007610321045, "learning_rate": 1.0212368609932816e-05, "loss": 0.7781, "mean_token_accuracy": 0.79161736369133, "num_tokens": 48512960.0, "step": 1011 }, { "entropy": 0.4961482882499695, "epoch": 1.559322033898305, "grad_norm": 0.6207665205001831, "learning_rate": 1.0195381377343576e-05, "loss": 0.2793, "mean_token_accuracy": 0.876209944486618, "num_tokens": 48549513.0, "step": 1012 }, { "entropy": 0.3310953974723816, "epoch": 1.5608628659476116, "grad_norm": 0.6216629147529602, "learning_rate": 1.0178393580715524e-05, "loss": 0.2924, "mean_token_accuracy": 0.9209280610084534, "num_tokens": 48613973.0, "step": 1013 }, { "entropy": 0.31613166630268097, "epoch": 1.5624036979969183, "grad_norm": 0.6636056900024414, "learning_rate": 1.0161405269090068e-05, "loss": 0.2876, "mean_token_accuracy": 0.9232576191425323, "num_tokens": 48675626.0, "step": 1014 }, { "entropy": 0.37533411383628845, "epoch": 1.563944530046225, "grad_norm": 0.7052972316741943, "learning_rate": 1.0144416491510095e-05, "loss": 0.3436, "mean_token_accuracy": 0.9088415503501892, "num_tokens": 48735609.0, "step": 1015 }, { "entropy": 0.37137773633003235, "epoch": 1.5654853620955316, "grad_norm": 0.7529043555259705, "learning_rate": 1.0127427297019836e-05, "loss": 0.3505, "mean_token_accuracy": 0.9066370129585266, "num_tokens": 48794082.0, "step": 1016 }, { "entropy": 0.3492370694875717, "epoch": 1.5670261941448382, "grad_norm": 0.7730548977851868, "learning_rate": 1.0110437734664728e-05, "loss": 0.3357, "mean_token_accuracy": 0.9121277332305908, "num_tokens": 48851517.0, "step": 1017 }, { "entropy": 0.33483393490314484, "epoch": 1.568567026194145, "grad_norm": 0.7642796039581299, "learning_rate": 1.0093447853491276e-05, "loss": 0.3164, "mean_token_accuracy": 0.9159277677536011, "num_tokens": 48907896.0, "step": 1018 }, { "entropy": 0.34964515268802643, "epoch": 1.5701078582434516, "grad_norm": 0.8219295740127563, "learning_rate": 1.0076457702546889e-05, "loss": 0.3295, "mean_token_accuracy": 0.9133083522319794, "num_tokens": 48963051.0, "step": 1019 }, { "entropy": 0.33040787279605865, "epoch": 1.5716486902927582, "grad_norm": 0.7802428603172302, "learning_rate": 1.0059467330879776e-05, "loss": 0.3104, "mean_token_accuracy": 0.9193694889545441, "num_tokens": 49016756.0, "step": 1020 }, { "entropy": 0.3538467437028885, "epoch": 1.5731895223420647, "grad_norm": 0.8245171308517456, "learning_rate": 1.0042476787538763e-05, "loss": 0.3402, "mean_token_accuracy": 0.9127759337425232, "num_tokens": 49068790.0, "step": 1021 }, { "entropy": 0.5715163946151733, "epoch": 1.5747303543913713, "grad_norm": 0.736468493938446, "learning_rate": 1.0025486121573183e-05, "loss": 0.3825, "mean_token_accuracy": 0.8445075452327728, "num_tokens": 49099688.0, "step": 1022 }, { "entropy": 0.8114289045333862, "epoch": 1.576271186440678, "grad_norm": 0.8105292916297913, "learning_rate": 1.0008495382032724e-05, "loss": 0.7993, "mean_token_accuracy": 0.7841039597988129, "num_tokens": 49110581.0, "step": 1023 }, { "entropy": 0.7813067734241486, "epoch": 1.5778120184899846, "grad_norm": 0.9903311729431152, "learning_rate": 9.99150461796728e-06, "loss": 0.7567, "mean_token_accuracy": 0.8036144077777863, "num_tokens": 49118771.0, "step": 1024 }, { "entropy": 0.21948610246181488, "epoch": 1.579352850539291, "grad_norm": 0.4682918190956116, "learning_rate": 9.974513878426819e-06, "loss": 0.1938, "mean_token_accuracy": 0.9476452469825745, "num_tokens": 49184019.0, "step": 1025 }, { "entropy": 0.4056839495897293, "epoch": 1.5808936825885977, "grad_norm": 0.7340463995933533, "learning_rate": 9.957523212461239e-06, "loss": 0.3634, "mean_token_accuracy": 0.9028380215167999, "num_tokens": 49246440.0, "step": 1026 }, { "entropy": 0.3764192461967468, "epoch": 1.5824345146379044, "grad_norm": 0.746769368648529, "learning_rate": 9.94053266912023e-06, "loss": 0.3399, "mean_token_accuracy": 0.9114930927753448, "num_tokens": 49306802.0, "step": 1027 }, { "entropy": 0.3610509932041168, "epoch": 1.583975346687211, "grad_norm": 0.6806610226631165, "learning_rate": 9.923542297453113e-06, "loss": 0.3339, "mean_token_accuracy": 0.9127370417118073, "num_tokens": 49365799.0, "step": 1028 }, { "entropy": 0.39256730675697327, "epoch": 1.5855161787365177, "grad_norm": 0.8117960095405579, "learning_rate": 9.906552146508727e-06, "loss": 0.3817, "mean_token_accuracy": 0.9008910655975342, "num_tokens": 49423651.0, "step": 1029 }, { "entropy": 0.3685729652643204, "epoch": 1.5870570107858244, "grad_norm": 0.8762190937995911, "learning_rate": 9.889562265335272e-06, "loss": 0.3686, "mean_token_accuracy": 0.904299795627594, "num_tokens": 49480401.0, "step": 1030 }, { "entropy": 0.34513573348522186, "epoch": 1.588597842835131, "grad_norm": 0.8298799991607666, "learning_rate": 9.872572702980166e-06, "loss": 0.3346, "mean_token_accuracy": 0.9125572443008423, "num_tokens": 49536169.0, "step": 1031 }, { "entropy": 0.33883893489837646, "epoch": 1.5901386748844377, "grad_norm": 0.8555536866188049, "learning_rate": 9.85558350848991e-06, "loss": 0.3345, "mean_token_accuracy": 0.9122505187988281, "num_tokens": 49590923.0, "step": 1032 }, { "entropy": 0.35447658598423004, "epoch": 1.5916795069337444, "grad_norm": 0.7886487245559692, "learning_rate": 9.838594730909936e-06, "loss": 0.3312, "mean_token_accuracy": 0.9144788980484009, "num_tokens": 49643964.0, "step": 1033 }, { "entropy": 0.3405940979719162, "epoch": 1.5932203389830508, "grad_norm": 0.7601723670959473, "learning_rate": 9.82160641928448e-06, "loss": 0.3081, "mean_token_accuracy": 0.920476108789444, "num_tokens": 49692040.0, "step": 1034 }, { "entropy": 0.8418119847774506, "epoch": 1.5947611710323575, "grad_norm": 0.7786232233047485, "learning_rate": 9.804618622656427e-06, "loss": 0.8424, "mean_token_accuracy": 0.7697497606277466, "num_tokens": 49704578.0, "step": 1035 }, { "entropy": 0.7578488886356354, "epoch": 1.5963020030816641, "grad_norm": 0.9874365329742432, "learning_rate": 9.787631390067184e-06, "loss": 0.7475, "mean_token_accuracy": 0.8077180981636047, "num_tokens": 49714022.0, "step": 1036 }, { "entropy": 0.45306579768657684, "epoch": 1.5978428351309706, "grad_norm": 0.6082413196563721, "learning_rate": 9.770644770556524e-06, "loss": 0.218, "mean_token_accuracy": 0.8870800733566284, "num_tokens": 49750269.0, "step": 1037 }, { "entropy": 0.2920878082513809, "epoch": 1.5993836671802772, "grad_norm": 0.5904483199119568, "learning_rate": 9.753658813162444e-06, "loss": 0.2578, "mean_token_accuracy": 0.9311122596263885, "num_tokens": 49814683.0, "step": 1038 }, { "entropy": 0.39498430490493774, "epoch": 1.600924499229584, "grad_norm": 0.7297495603561401, "learning_rate": 9.736673566921036e-06, "loss": 0.3511, "mean_token_accuracy": 0.9074370265007019, "num_tokens": 49876208.0, "step": 1039 }, { "entropy": 0.40525902807712555, "epoch": 1.6024653312788906, "grad_norm": 0.7360152006149292, "learning_rate": 9.719689080866347e-06, "loss": 0.3632, "mean_token_accuracy": 0.9040480256080627, "num_tokens": 49935937.0, "step": 1040 }, { "entropy": 0.3885108381509781, "epoch": 1.6040061633281972, "grad_norm": 0.7293180227279663, "learning_rate": 9.702705404030215e-06, "loss": 0.3674, "mean_token_accuracy": 0.9036713540554047, "num_tokens": 49994449.0, "step": 1041 }, { "entropy": 0.349113792181015, "epoch": 1.605546995377504, "grad_norm": 0.7722710967063904, "learning_rate": 9.68572258544215e-06, "loss": 0.3387, "mean_token_accuracy": 0.9110961854457855, "num_tokens": 50051772.0, "step": 1042 }, { "entropy": 0.3593316972255707, "epoch": 1.6070878274268106, "grad_norm": 0.8864720463752747, "learning_rate": 9.668740674129178e-06, "loss": 0.3553, "mean_token_accuracy": 0.9074713587760925, "num_tokens": 50108103.0, "step": 1043 }, { "entropy": 0.3632139265537262, "epoch": 1.6086286594761172, "grad_norm": 0.8885848522186279, "learning_rate": 9.651759719115716e-06, "loss": 0.3615, "mean_token_accuracy": 0.9066572189331055, "num_tokens": 50163273.0, "step": 1044 }, { "entropy": 0.35053060948848724, "epoch": 1.6101694915254239, "grad_norm": 0.8592848181724548, "learning_rate": 9.634779769423412e-06, "loss": 0.3411, "mean_token_accuracy": 0.9128399193286896, "num_tokens": 50217084.0, "step": 1045 }, { "entropy": 0.33947597444057465, "epoch": 1.6117103235747303, "grad_norm": 0.7777562737464905, "learning_rate": 9.617800874071014e-06, "loss": 0.3208, "mean_token_accuracy": 0.9181949198246002, "num_tokens": 50268777.0, "step": 1046 }, { "entropy": 0.645783856511116, "epoch": 1.613251155624037, "grad_norm": 0.8590431809425354, "learning_rate": 9.600823082074232e-06, "loss": 0.5074, "mean_token_accuracy": 0.8292523324489594, "num_tokens": 50297246.0, "step": 1047 }, { "entropy": 0.8267770707607269, "epoch": 1.6147919876733436, "grad_norm": 0.9315404295921326, "learning_rate": 9.583846442445578e-06, "loss": 0.8495, "mean_token_accuracy": 0.772677093744278, "num_tokens": 50307863.0, "step": 1048 }, { "entropy": 0.7290515005588531, "epoch": 1.61633281972265, "grad_norm": 0.9870967864990234, "learning_rate": 9.566871004194245e-06, "loss": 0.7181, "mean_token_accuracy": 0.8157494366168976, "num_tokens": 50315849.0, "step": 1049 }, { "entropy": 0.2804524898529053, "epoch": 1.6178736517719567, "grad_norm": 0.5471501350402832, "learning_rate": 9.549896816325964e-06, "loss": 0.2418, "mean_token_accuracy": 0.937265157699585, "num_tokens": 50381280.0, "step": 1050 }, { "entropy": 0.3936946392059326, "epoch": 1.6194144838212634, "grad_norm": 0.7559870481491089, "learning_rate": 9.532923927842844e-06, "loss": 0.351, "mean_token_accuracy": 0.9062430560588837, "num_tokens": 50444421.0, "step": 1051 }, { "entropy": 0.38441602885723114, "epoch": 1.62095531587057, "grad_norm": 0.7274194359779358, "learning_rate": 9.515952387743252e-06, "loss": 0.341, "mean_token_accuracy": 0.9085601270198822, "num_tokens": 50505516.0, "step": 1052 }, { "entropy": 0.3519275337457657, "epoch": 1.6224961479198767, "grad_norm": 0.7034042477607727, "learning_rate": 9.498982245021655e-06, "loss": 0.3187, "mean_token_accuracy": 0.9157246947288513, "num_tokens": 50564826.0, "step": 1053 }, { "entropy": 0.3666542172431946, "epoch": 1.6240369799691834, "grad_norm": 0.7544987201690674, "learning_rate": 9.482013548668492e-06, "loss": 0.3416, "mean_token_accuracy": 0.9082333445549011, "num_tokens": 50622862.0, "step": 1054 }, { "entropy": 0.3718187212944031, "epoch": 1.62557781201849, "grad_norm": 0.8253666162490845, "learning_rate": 9.465046347670029e-06, "loss": 0.3564, "mean_token_accuracy": 0.9065527319908142, "num_tokens": 50679704.0, "step": 1055 }, { "entropy": 0.3798713982105255, "epoch": 1.6271186440677967, "grad_norm": 0.8765273094177246, "learning_rate": 9.448080691008202e-06, "loss": 0.3732, "mean_token_accuracy": 0.9023167192935944, "num_tokens": 50735439.0, "step": 1056 }, { "entropy": 0.337435781955719, "epoch": 1.6286594761171034, "grad_norm": 0.926148533821106, "learning_rate": 9.431116627660505e-06, "loss": 0.3392, "mean_token_accuracy": 0.9114595651626587, "num_tokens": 50789983.0, "step": 1057 }, { "entropy": 0.3304522633552551, "epoch": 1.6302003081664098, "grad_norm": 0.8181936144828796, "learning_rate": 9.414154206599821e-06, "loss": 0.314, "mean_token_accuracy": 0.9184253215789795, "num_tokens": 50843084.0, "step": 1058 }, { "entropy": 0.33874472975730896, "epoch": 1.6317411402157165, "grad_norm": 0.8375577926635742, "learning_rate": 9.397193476794293e-06, "loss": 0.3231, "mean_token_accuracy": 0.9178401529788971, "num_tokens": 50892516.0, "step": 1059 }, { "entropy": 0.8699841499328613, "epoch": 1.6332819722650231, "grad_norm": 0.8347638249397278, "learning_rate": 9.380234487207192e-06, "loss": 0.8932, "mean_token_accuracy": 0.7636078000068665, "num_tokens": 50904590.0, "step": 1060 }, { "entropy": 0.7998379766941071, "epoch": 1.6348228043143296, "grad_norm": 0.9536826014518738, "learning_rate": 9.363277286796751e-06, "loss": 0.777, "mean_token_accuracy": 0.7976020276546478, "num_tokens": 50913828.0, "step": 1061 }, { "entropy": 0.44171367585659027, "epoch": 1.6363636363636362, "grad_norm": 0.5470724105834961, "learning_rate": 9.346321924516045e-06, "loss": 0.2214, "mean_token_accuracy": 0.8897871673107147, "num_tokens": 50950130.0, "step": 1062 }, { "entropy": 0.3512416183948517, "epoch": 1.637904468412943, "grad_norm": 0.6541381478309631, "learning_rate": 9.329368449312847e-06, "loss": 0.3114, "mean_token_accuracy": 0.9165794253349304, "num_tokens": 51014136.0, "step": 1063 }, { "entropy": 0.391769215464592, "epoch": 1.6394453004622496, "grad_norm": 0.7257230877876282, "learning_rate": 9.312416910129471e-06, "loss": 0.3441, "mean_token_accuracy": 0.9092686772346497, "num_tokens": 51075167.0, "step": 1064 }, { "entropy": 0.3659151941537857, "epoch": 1.6409861325115562, "grad_norm": 0.6776227951049805, "learning_rate": 9.295467355902652e-06, "loss": 0.33, "mean_token_accuracy": 0.9114861786365509, "num_tokens": 51134530.0, "step": 1065 }, { "entropy": 0.35517890751361847, "epoch": 1.642526964560863, "grad_norm": 0.6937512159347534, "learning_rate": 9.278519835563387e-06, "loss": 0.3244, "mean_token_accuracy": 0.9151228368282318, "num_tokens": 51192721.0, "step": 1066 }, { "entropy": 0.41514265537261963, "epoch": 1.6440677966101696, "grad_norm": 0.770442008972168, "learning_rate": 9.26157439803681e-06, "loss": 0.39, "mean_token_accuracy": 0.8971673548221588, "num_tokens": 51249969.0, "step": 1067 }, { "entropy": 0.35523325204849243, "epoch": 1.6456086286594762, "grad_norm": 0.7676479816436768, "learning_rate": 9.244631092242036e-06, "loss": 0.3375, "mean_token_accuracy": 0.9110822379589081, "num_tokens": 51306218.0, "step": 1068 }, { "entropy": 0.3445281982421875, "epoch": 1.6471494607087829, "grad_norm": 0.9324392676353455, "learning_rate": 9.227689967092025e-06, "loss": 0.3448, "mean_token_accuracy": 0.9092360436916351, "num_tokens": 51361255.0, "step": 1069 }, { "entropy": 0.36460405588150024, "epoch": 1.6486902927580893, "grad_norm": 0.9453696012496948, "learning_rate": 9.210751071493448e-06, "loss": 0.3637, "mean_token_accuracy": 0.90715092420578, "num_tokens": 51414902.0, "step": 1070 }, { "entropy": 0.2937265485525131, "epoch": 1.650231124807396, "grad_norm": 0.7435469627380371, "learning_rate": 9.193814454346531e-06, "loss": 0.2754, "mean_token_accuracy": 0.9294047951698303, "num_tokens": 51466422.0, "step": 1071 }, { "entropy": 0.7024309039115906, "epoch": 1.6517719568567026, "grad_norm": 0.9840456247329712, "learning_rate": 9.17688016454493e-06, "loss": 0.5215, "mean_token_accuracy": 0.833113819360733, "num_tokens": 51488287.0, "step": 1072 }, { "entropy": 0.8041163980960846, "epoch": 1.653312788906009, "grad_norm": 0.9132399559020996, "learning_rate": 9.159948250975586e-06, "loss": 0.7903, "mean_token_accuracy": 0.7864421308040619, "num_tokens": 51498659.0, "step": 1073 }, { "entropy": 0.787934422492981, "epoch": 1.6548536209553157, "grad_norm": 0.988511323928833, "learning_rate": 9.143018762518566e-06, "loss": 0.7844, "mean_token_accuracy": 0.8032382428646088, "num_tokens": 51506464.0, "step": 1074 }, { "entropy": 0.24526293575763702, "epoch": 1.6563944530046224, "grad_norm": 0.5172885656356812, "learning_rate": 9.126091748046943e-06, "loss": 0.2325, "mean_token_accuracy": 0.9380636513233185, "num_tokens": 51571954.0, "step": 1075 }, { "entropy": 0.37184441089630127, "epoch": 1.657935285053929, "grad_norm": 0.6445520520210266, "learning_rate": 9.10916725642665e-06, "loss": 0.3412, "mean_token_accuracy": 0.906610518693924, "num_tokens": 51635210.0, "step": 1076 }, { "entropy": 0.3850744813680649, "epoch": 1.6594761171032357, "grad_norm": 0.7502169013023376, "learning_rate": 9.092245336516341e-06, "loss": 0.3493, "mean_token_accuracy": 0.9067317247390747, "num_tokens": 51696139.0, "step": 1077 }, { "entropy": 0.394161656498909, "epoch": 1.6610169491525424, "grad_norm": 0.7372358441352844, "learning_rate": 9.075326037167239e-06, "loss": 0.3531, "mean_token_accuracy": 0.9078056216239929, "num_tokens": 51755526.0, "step": 1078 }, { "entropy": 0.34868955612182617, "epoch": 1.662557781201849, "grad_norm": 0.6892228722572327, "learning_rate": 9.058409407222996e-06, "loss": 0.3142, "mean_token_accuracy": 0.9194933474063873, "num_tokens": 51813640.0, "step": 1079 }, { "entropy": 0.3345145881175995, "epoch": 1.6640986132511557, "grad_norm": 0.7090299725532532, "learning_rate": 9.041495495519567e-06, "loss": 0.3065, "mean_token_accuracy": 0.9204688668251038, "num_tokens": 51870529.0, "step": 1080 }, { "entropy": 0.3954104334115982, "epoch": 1.6656394453004624, "grad_norm": 0.7869145274162292, "learning_rate": 9.024584350885067e-06, "loss": 0.3864, "mean_token_accuracy": 0.9008515477180481, "num_tokens": 51926397.0, "step": 1081 }, { "entropy": 0.30294016003608704, "epoch": 1.667180277349769, "grad_norm": 0.8656772375106812, "learning_rate": 9.007676022139603e-06, "loss": 0.3013, "mean_token_accuracy": 0.9214164018630981, "num_tokens": 51981277.0, "step": 1082 }, { "entropy": 0.30725713074207306, "epoch": 1.6687211093990755, "grad_norm": 0.799132227897644, "learning_rate": 8.990770558095174e-06, "loss": 0.2923, "mean_token_accuracy": 0.9233315587043762, "num_tokens": 52034812.0, "step": 1083 }, { "entropy": 0.31454668939113617, "epoch": 1.6702619414483821, "grad_norm": 0.8759317994117737, "learning_rate": 8.973868007555486e-06, "loss": 0.3067, "mean_token_accuracy": 0.9218968749046326, "num_tokens": 52085706.0, "step": 1084 }, { "entropy": 0.7917650938034058, "epoch": 1.6718027734976888, "grad_norm": 1.0133821964263916, "learning_rate": 8.956968419315858e-06, "loss": 0.6898, "mean_token_accuracy": 0.7974615395069122, "num_tokens": 52103050.0, "step": 1085 }, { "entropy": 0.7632021605968475, "epoch": 1.6733436055469952, "grad_norm": 1.0454987287521362, "learning_rate": 8.940071842163038e-06, "loss": 0.7623, "mean_token_accuracy": 0.7988005876541138, "num_tokens": 52112803.0, "step": 1086 }, { "entropy": 0.40629252046346664, "epoch": 1.674884437596302, "grad_norm": 0.6303904056549072, "learning_rate": 8.923178324875095e-06, "loss": 0.2067, "mean_token_accuracy": 0.8983080685138702, "num_tokens": 52149110.0, "step": 1087 }, { "entropy": 0.31238096952438354, "epoch": 1.6764252696456086, "grad_norm": 0.593475341796875, "learning_rate": 8.906287916221259e-06, "loss": 0.2909, "mean_token_accuracy": 0.9225495159626007, "num_tokens": 52213352.0, "step": 1088 }, { "entropy": 0.39309366047382355, "epoch": 1.6779661016949152, "grad_norm": 0.677842915058136, "learning_rate": 8.889400664961782e-06, "loss": 0.3533, "mean_token_accuracy": 0.9059905111789703, "num_tokens": 52274948.0, "step": 1089 }, { "entropy": 0.3507690131664276, "epoch": 1.6795069337442219, "grad_norm": 0.6674107909202576, "learning_rate": 8.872516619847805e-06, "loss": 0.3164, "mean_token_accuracy": 0.9168327748775482, "num_tokens": 52334852.0, "step": 1090 }, { "entropy": 0.3438497483730316, "epoch": 1.6810477657935285, "grad_norm": 0.7160604596138, "learning_rate": 8.855635829621217e-06, "loss": 0.3019, "mean_token_accuracy": 0.921341598033905, "num_tokens": 52393282.0, "step": 1091 }, { "entropy": 0.34969571232795715, "epoch": 1.6825885978428352, "grad_norm": 0.6950910091400146, "learning_rate": 8.838758343014504e-06, "loss": 0.3242, "mean_token_accuracy": 0.9147965610027313, "num_tokens": 52450591.0, "step": 1092 }, { "entropy": 0.32281017303466797, "epoch": 1.6841294298921419, "grad_norm": 0.6904616355895996, "learning_rate": 8.821884208750623e-06, "loss": 0.3007, "mean_token_accuracy": 0.921626627445221, "num_tokens": 52506813.0, "step": 1093 }, { "entropy": 0.334498792886734, "epoch": 1.6856702619414485, "grad_norm": 0.8081304430961609, "learning_rate": 8.80501347554284e-06, "loss": 0.3123, "mean_token_accuracy": 0.918409675359726, "num_tokens": 52561846.0, "step": 1094 }, { "entropy": 0.34316758811473846, "epoch": 1.687211093990755, "grad_norm": 0.8207107186317444, "learning_rate": 8.788146192094615e-06, "loss": 0.3287, "mean_token_accuracy": 0.912713348865509, "num_tokens": 52615655.0, "step": 1095 }, { "entropy": 0.31875038146972656, "epoch": 1.6887519260400616, "grad_norm": 0.7835222482681274, "learning_rate": 8.771282407099441e-06, "loss": 0.3008, "mean_token_accuracy": 0.9232972264289856, "num_tokens": 52667288.0, "step": 1096 }, { "entropy": 0.6310667395591736, "epoch": 1.6902927580893683, "grad_norm": 0.7859800457954407, "learning_rate": 8.754422169240719e-06, "loss": 0.4456, "mean_token_accuracy": 0.8376898169517517, "num_tokens": 52694726.0, "step": 1097 }, { "entropy": 0.8151070773601532, "epoch": 1.6918335901386747, "grad_norm": 0.8819522261619568, "learning_rate": 8.737565527191609e-06, "loss": 0.785, "mean_token_accuracy": 0.7864404618740082, "num_tokens": 52705238.0, "step": 1098 }, { "entropy": 0.7531603872776031, "epoch": 1.6933744221879814, "grad_norm": 0.9799526929855347, "learning_rate": 8.72071252961488e-06, "loss": 0.731, "mean_token_accuracy": 0.8146850764751434, "num_tokens": 52713052.0, "step": 1099 }, { "entropy": 0.22662878036499023, "epoch": 1.694915254237288, "grad_norm": 0.6186622381210327, "learning_rate": 8.703863225162788e-06, "loss": 0.2209, "mean_token_accuracy": 0.9405584037303925, "num_tokens": 52778523.0, "step": 1100 }, { "entropy": 0.3498546928167343, "epoch": 1.6964560862865947, "grad_norm": 0.8092815279960632, "learning_rate": 8.687017662476933e-06, "loss": 0.3466, "mean_token_accuracy": 0.9061404764652252, "num_tokens": 52841451.0, "step": 1101 }, { "entropy": 0.3361303210258484, "epoch": 1.6979969183359014, "grad_norm": 0.7111431956291199, "learning_rate": 8.670175890188099e-06, "loss": 0.3251, "mean_token_accuracy": 0.9138661623001099, "num_tokens": 52902208.0, "step": 1102 }, { "entropy": 0.3433539718389511, "epoch": 1.699537750385208, "grad_norm": 0.7117166519165039, "learning_rate": 8.653337956916141e-06, "loss": 0.321, "mean_token_accuracy": 0.916059672832489, "num_tokens": 52961490.0, "step": 1103 }, { "entropy": 0.34065185487270355, "epoch": 1.7010785824345147, "grad_norm": 0.6860400438308716, "learning_rate": 8.636503911269828e-06, "loss": 0.3106, "mean_token_accuracy": 0.9183832406997681, "num_tokens": 53019547.0, "step": 1104 }, { "entropy": 0.4002942442893982, "epoch": 1.7026194144838214, "grad_norm": 0.7329869866371155, "learning_rate": 8.619673801846697e-06, "loss": 0.3603, "mean_token_accuracy": 0.9070156514644623, "num_tokens": 53076323.0, "step": 1105 }, { "entropy": 0.3659636676311493, "epoch": 1.704160246533128, "grad_norm": 0.7212053537368774, "learning_rate": 8.60284767723293e-06, "loss": 0.3259, "mean_token_accuracy": 0.9164974093437195, "num_tokens": 53132007.0, "step": 1106 }, { "entropy": 0.34479376673698425, "epoch": 1.7057010785824345, "grad_norm": 0.6869596838951111, "learning_rate": 8.586025586003206e-06, "loss": 0.3036, "mean_token_accuracy": 0.9214841723442078, "num_tokens": 53186582.0, "step": 1107 }, { "entropy": 0.361681267619133, "epoch": 1.7072419106317411, "grad_norm": 0.7313923239707947, "learning_rate": 8.569207576720558e-06, "loss": 0.3376, "mean_token_accuracy": 0.9141926765441895, "num_tokens": 53239446.0, "step": 1108 }, { "entropy": 0.4005378633737564, "epoch": 1.7087827426810478, "grad_norm": 0.7842838764190674, "learning_rate": 8.552393697936236e-06, "loss": 0.3479, "mean_token_accuracy": 0.9100260436534882, "num_tokens": 53284297.0, "step": 1109 }, { "entropy": 0.8214946687221527, "epoch": 1.7103235747303542, "grad_norm": 0.8296878933906555, "learning_rate": 8.535583998189558e-06, "loss": 0.8493, "mean_token_accuracy": 0.7627301514148712, "num_tokens": 53296628.0, "step": 1110 }, { "entropy": 0.7719273865222931, "epoch": 1.711864406779661, "grad_norm": 1.0810925960540771, "learning_rate": 8.518778526007793e-06, "loss": 0.7949, "mean_token_accuracy": 0.791944146156311, "num_tokens": 53306053.0, "step": 1111 }, { "entropy": 0.45345036685466766, "epoch": 1.7134052388289676, "grad_norm": 0.720853865146637, "learning_rate": 8.501977329905992e-06, "loss": 0.2351, "mean_token_accuracy": 0.8878662288188934, "num_tokens": 53342564.0, "step": 1112 }, { "entropy": 0.2870091497898102, "epoch": 1.7149460708782742, "grad_norm": 0.7339043617248535, "learning_rate": 8.485180458386865e-06, "loss": 0.2786, "mean_token_accuracy": 0.9256813526153564, "num_tokens": 53406454.0, "step": 1113 }, { "entropy": 0.33046725392341614, "epoch": 1.7164869029275809, "grad_norm": 0.7356221079826355, "learning_rate": 8.46838795994065e-06, "loss": 0.3188, "mean_token_accuracy": 0.9139839112758636, "num_tokens": 53467929.0, "step": 1114 }, { "entropy": 0.3778156489133835, "epoch": 1.7180277349768875, "grad_norm": 0.7638092041015625, "learning_rate": 8.451599883044937e-06, "loss": 0.3646, "mean_token_accuracy": 0.9047285914421082, "num_tokens": 53527725.0, "step": 1115 }, { "entropy": 0.35918310284614563, "epoch": 1.7195685670261942, "grad_norm": 0.7192323207855225, "learning_rate": 8.434816276164567e-06, "loss": 0.3386, "mean_token_accuracy": 0.9100305438041687, "num_tokens": 53586067.0, "step": 1116 }, { "entropy": 0.36497363448143005, "epoch": 1.7211093990755009, "grad_norm": 0.7084750533103943, "learning_rate": 8.41803718775148e-06, "loss": 0.3313, "mean_token_accuracy": 0.9137909114360809, "num_tokens": 53643298.0, "step": 1117 }, { "entropy": 0.31923724710941315, "epoch": 1.7226502311248075, "grad_norm": 0.6817907094955444, "learning_rate": 8.40126266624456e-06, "loss": 0.2851, "mean_token_accuracy": 0.926692008972168, "num_tokens": 53699611.0, "step": 1118 }, { "entropy": 0.3879624158143997, "epoch": 1.724191063174114, "grad_norm": 0.733768880367279, "learning_rate": 8.384492760069518e-06, "loss": 0.3459, "mean_token_accuracy": 0.911109209060669, "num_tokens": 53755016.0, "step": 1119 }, { "entropy": 0.3642096370458603, "epoch": 1.7257318952234206, "grad_norm": 0.7516413331031799, "learning_rate": 8.367727517638729e-06, "loss": 0.3285, "mean_token_accuracy": 0.9140211343765259, "num_tokens": 53809146.0, "step": 1120 }, { "entropy": 0.31566159427165985, "epoch": 1.7272727272727273, "grad_norm": 0.7236393094062805, "learning_rate": 8.35096698735112e-06, "loss": 0.2946, "mean_token_accuracy": 0.9242013096809387, "num_tokens": 53861288.0, "step": 1121 }, { "entropy": 0.5456562638282776, "epoch": 1.7288135593220337, "grad_norm": 0.8535805940628052, "learning_rate": 8.334211217591997e-06, "loss": 0.3949, "mean_token_accuracy": 0.8496732711791992, "num_tokens": 53893067.0, "step": 1122 }, { "entropy": 0.8311359286308289, "epoch": 1.7303543913713404, "grad_norm": 0.81272292137146, "learning_rate": 8.317460256732937e-06, "loss": 0.819, "mean_token_accuracy": 0.7775570750236511, "num_tokens": 53904149.0, "step": 1123 }, { "entropy": 0.7292285561561584, "epoch": 1.731895223420647, "grad_norm": 1.0489131212234497, "learning_rate": 8.300714153131632e-06, "loss": 0.7107, "mean_token_accuracy": 0.8185764849185944, "num_tokens": 53912437.0, "step": 1124 }, { "entropy": 0.25574297457933426, "epoch": 1.7334360554699537, "grad_norm": 0.6379603743553162, "learning_rate": 8.283972955131747e-06, "loss": 0.2477, "mean_token_accuracy": 0.9351848661899567, "num_tokens": 53977729.0, "step": 1125 }, { "entropy": 0.35072144865989685, "epoch": 1.7349768875192604, "grad_norm": 0.8015511631965637, "learning_rate": 8.267236711062785e-06, "loss": 0.3505, "mean_token_accuracy": 0.906651109457016, "num_tokens": 54040257.0, "step": 1126 }, { "entropy": 0.3238384276628494, "epoch": 1.736517719568567, "grad_norm": 0.79253089427948, "learning_rate": 8.25050546923995e-06, "loss": 0.3209, "mean_token_accuracy": 0.9140223264694214, "num_tokens": 54100869.0, "step": 1127 }, { "entropy": 0.32733292877674103, "epoch": 1.7380585516178737, "grad_norm": 0.740621030330658, "learning_rate": 8.233779277964009e-06, "loss": 0.3181, "mean_token_accuracy": 0.9163314998149872, "num_tokens": 54160014.0, "step": 1128 }, { "entropy": 0.3758051246404648, "epoch": 1.7395993836671804, "grad_norm": 0.7292847633361816, "learning_rate": 8.217058185521145e-06, "loss": 0.352, "mean_token_accuracy": 0.9074923098087311, "num_tokens": 54217939.0, "step": 1129 }, { "entropy": 0.3331088721752167, "epoch": 1.741140215716487, "grad_norm": 0.6709481477737427, "learning_rate": 8.20034224018282e-06, "loss": 0.3032, "mean_token_accuracy": 0.9211255013942719, "num_tokens": 54274743.0, "step": 1130 }, { "entropy": 0.3980555832386017, "epoch": 1.7426810477657937, "grad_norm": 0.7230737209320068, "learning_rate": 8.183631490205636e-06, "loss": 0.3628, "mean_token_accuracy": 0.9095185101032257, "num_tokens": 54330393.0, "step": 1131 }, { "entropy": 0.33550257980823517, "epoch": 1.7442218798151001, "grad_norm": 0.75874924659729, "learning_rate": 8.166925983831206e-06, "loss": 0.2937, "mean_token_accuracy": 0.9234819710254669, "num_tokens": 54384794.0, "step": 1132 }, { "entropy": 0.36489437520504, "epoch": 1.7457627118644068, "grad_norm": 0.7615087032318115, "learning_rate": 8.15022576928599e-06, "loss": 0.3286, "mean_token_accuracy": 0.9143741130828857, "num_tokens": 54437775.0, "step": 1133 }, { "entropy": 0.3092028647661209, "epoch": 1.7473035439137135, "grad_norm": 0.6920976638793945, "learning_rate": 8.13353089478119e-06, "loss": 0.2797, "mean_token_accuracy": 0.9282291531562805, "num_tokens": 54488682.0, "step": 1134 }, { "entropy": 0.7341200709342957, "epoch": 1.74884437596302, "grad_norm": 0.8389025926589966, "learning_rate": 8.116841408512576e-06, "loss": 0.5791, "mean_token_accuracy": 0.821784108877182, "num_tokens": 54507496.0, "step": 1135 }, { "entropy": 0.8187333047389984, "epoch": 1.7503852080123266, "grad_norm": 1.053586721420288, "learning_rate": 8.100157358660367e-06, "loss": 0.8343, "mean_token_accuracy": 0.780840277671814, "num_tokens": 54517408.0, "step": 1136 }, { "entropy": 0.44415565580129623, "epoch": 1.7519260400616332, "grad_norm": 0.7669696807861328, "learning_rate": 8.083478793389092e-06, "loss": 0.2369, "mean_token_accuracy": 0.8863668143749237, "num_tokens": 54553651.0, "step": 1137 }, { "entropy": 0.31461529433727264, "epoch": 1.7534668721109399, "grad_norm": 0.7532101273536682, "learning_rate": 8.066805760847449e-06, "loss": 0.3006, "mean_token_accuracy": 0.9190893471240997, "num_tokens": 54617873.0, "step": 1138 }, { "entropy": 0.3316240608692169, "epoch": 1.7550077041602465, "grad_norm": 0.739212155342102, "learning_rate": 8.05013830916815e-06, "loss": 0.3232, "mean_token_accuracy": 0.9124544560909271, "num_tokens": 54679287.0, "step": 1139 }, { "entropy": 0.33145758509635925, "epoch": 1.7565485362095532, "grad_norm": 0.7246508598327637, "learning_rate": 8.033476486467818e-06, "loss": 0.3204, "mean_token_accuracy": 0.9137423932552338, "num_tokens": 54738995.0, "step": 1140 }, { "entropy": 0.34054356813430786, "epoch": 1.7580893682588599, "grad_norm": 0.7414966821670532, "learning_rate": 8.0168203408468e-06, "loss": 0.3229, "mean_token_accuracy": 0.91623854637146, "num_tokens": 54797420.0, "step": 1141 }, { "entropy": 0.3051619678735733, "epoch": 1.7596302003081665, "grad_norm": 0.714688241481781, "learning_rate": 8.000169920389074e-06, "loss": 0.2875, "mean_token_accuracy": 0.9250382483005524, "num_tokens": 54854647.0, "step": 1142 }, { "entropy": 0.3334145247936249, "epoch": 1.7611710323574732, "grad_norm": 0.6729596257209778, "learning_rate": 7.98352527316208e-06, "loss": 0.3136, "mean_token_accuracy": 0.9188911020755768, "num_tokens": 54910721.0, "step": 1143 }, { "entropy": 0.3574065864086151, "epoch": 1.7627118644067796, "grad_norm": 0.7091937065124512, "learning_rate": 7.9668864472166e-06, "loss": 0.3299, "mean_token_accuracy": 0.9163220524787903, "num_tokens": 54965588.0, "step": 1144 }, { "entropy": 0.34926196932792664, "epoch": 1.7642526964560863, "grad_norm": 0.7368658185005188, "learning_rate": 7.950253490586603e-06, "loss": 0.3231, "mean_token_accuracy": 0.9183654487133026, "num_tokens": 55019053.0, "step": 1145 }, { "entropy": 0.34225286543369293, "epoch": 1.765793528505393, "grad_norm": 0.7421087622642517, "learning_rate": 7.933626451289116e-06, "loss": 0.3098, "mean_token_accuracy": 0.9193829298019409, "num_tokens": 55070584.0, "step": 1146 }, { "entropy": 0.6113008260726929, "epoch": 1.7673343605546994, "grad_norm": 0.8109264969825745, "learning_rate": 7.91700537732408e-06, "loss": 0.435, "mean_token_accuracy": 0.8457348644733429, "num_tokens": 55097623.0, "step": 1147 }, { "entropy": 0.8490845561027527, "epoch": 1.768875192604006, "grad_norm": 0.7930845618247986, "learning_rate": 7.900390316674229e-06, "loss": 0.8261, "mean_token_accuracy": 0.7768031358718872, "num_tokens": 55108232.0, "step": 1148 }, { "entropy": 0.7317865192890167, "epoch": 1.7704160246533127, "grad_norm": 0.9962891340255737, "learning_rate": 7.883781317304917e-06, "loss": 0.7295, "mean_token_accuracy": 0.8197780251502991, "num_tokens": 55116042.0, "step": 1149 }, { "entropy": 0.20063777267932892, "epoch": 1.7719568567026194, "grad_norm": 0.5051599740982056, "learning_rate": 7.867178427164024e-06, "loss": 0.1744, "mean_token_accuracy": 0.9541084468364716, "num_tokens": 55181565.0, "step": 1150 }, { "entropy": 0.3469012826681137, "epoch": 1.773497688751926, "grad_norm": 0.6832678914070129, "learning_rate": 7.850581694181765e-06, "loss": 0.3245, "mean_token_accuracy": 0.9123489260673523, "num_tokens": 55244687.0, "step": 1151 }, { "entropy": 0.3052908182144165, "epoch": 1.7750385208012327, "grad_norm": 0.6433122754096985, "learning_rate": 7.833991166270604e-06, "loss": 0.2859, "mean_token_accuracy": 0.9236160814762115, "num_tokens": 55305483.0, "step": 1152 }, { "entropy": 0.3522382080554962, "epoch": 1.7765793528505394, "grad_norm": 0.7220339775085449, "learning_rate": 7.817406891325078e-06, "loss": 0.3269, "mean_token_accuracy": 0.9121283292770386, "num_tokens": 55364874.0, "step": 1153 }, { "entropy": 0.3382432460784912, "epoch": 1.778120184899846, "grad_norm": 0.7256735563278198, "learning_rate": 7.800828917221683e-06, "loss": 0.3261, "mean_token_accuracy": 0.9150277376174927, "num_tokens": 55422949.0, "step": 1154 }, { "entropy": 0.32726964354515076, "epoch": 1.7796610169491527, "grad_norm": 0.7550110816955566, "learning_rate": 7.78425729181872e-06, "loss": 0.3106, "mean_token_accuracy": 0.9206837117671967, "num_tokens": 55479847.0, "step": 1155 }, { "entropy": 0.3342910408973694, "epoch": 1.7812018489984591, "grad_norm": 0.7865113019943237, "learning_rate": 7.767692062956156e-06, "loss": 0.3173, "mean_token_accuracy": 0.9160670638084412, "num_tokens": 55535589.0, "step": 1156 }, { "entropy": 0.29955407977104187, "epoch": 1.7827426810477658, "grad_norm": 0.7191455960273743, "learning_rate": 7.751133278455498e-06, "loss": 0.2834, "mean_token_accuracy": 0.927176296710968, "num_tokens": 55590166.0, "step": 1157 }, { "entropy": 0.3289608210325241, "epoch": 1.7842835130970724, "grad_norm": 0.7315102815628052, "learning_rate": 7.734580986119656e-06, "loss": 0.3129, "mean_token_accuracy": 0.9189658463001251, "num_tokens": 55643080.0, "step": 1158 }, { "entropy": 0.32803474366664886, "epoch": 1.7858243451463789, "grad_norm": 0.7050526142120361, "learning_rate": 7.718035233732786e-06, "loss": 0.3059, "mean_token_accuracy": 0.9219446182250977, "num_tokens": 55690810.0, "step": 1159 }, { "entropy": 0.8268380463123322, "epoch": 1.7873651771956856, "grad_norm": 0.8682421445846558, "learning_rate": 7.701496069060177e-06, "loss": 0.8528, "mean_token_accuracy": 0.7650494575500488, "num_tokens": 55702960.0, "step": 1160 }, { "entropy": 0.7485423684120178, "epoch": 1.7889060092449922, "grad_norm": 0.9973227977752686, "learning_rate": 7.684963539848077e-06, "loss": 0.7348, "mean_token_accuracy": 0.8046754896640778, "num_tokens": 55712498.0, "step": 1161 }, { "entropy": 0.45589151978492737, "epoch": 1.7904468412942989, "grad_norm": 0.5466704368591309, "learning_rate": 7.668437693823609e-06, "loss": 0.2277, "mean_token_accuracy": 0.8884575068950653, "num_tokens": 55748905.0, "step": 1162 }, { "entropy": 0.3084145039319992, "epoch": 1.7919876733436055, "grad_norm": 0.5886192321777344, "learning_rate": 7.651918578694578e-06, "loss": 0.2769, "mean_token_accuracy": 0.9248026311397552, "num_tokens": 55812602.0, "step": 1163 }, { "entropy": 0.34693779051303864, "epoch": 1.7935285053929122, "grad_norm": 0.6566057801246643, "learning_rate": 7.635406242149372e-06, "loss": 0.3126, "mean_token_accuracy": 0.9156429767608643, "num_tokens": 55873659.0, "step": 1164 }, { "entropy": 0.33607685565948486, "epoch": 1.7950693374422189, "grad_norm": 0.6553991436958313, "learning_rate": 7.618900731856806e-06, "loss": 0.313, "mean_token_accuracy": 0.918339729309082, "num_tokens": 55933068.0, "step": 1165 }, { "entropy": 0.3453989028930664, "epoch": 1.7966101694915255, "grad_norm": 0.6951221823692322, "learning_rate": 7.602402095465984e-06, "loss": 0.3205, "mean_token_accuracy": 0.9167520999908447, "num_tokens": 55991270.0, "step": 1166 }, { "entropy": 0.3035885989665985, "epoch": 1.7981510015408322, "grad_norm": 0.7221125960350037, "learning_rate": 7.5859103806061694e-06, "loss": 0.2884, "mean_token_accuracy": 0.9244616031646729, "num_tokens": 56048398.0, "step": 1167 }, { "entropy": 0.30423004925251007, "epoch": 1.7996918335901386, "grad_norm": 0.7508959770202637, "learning_rate": 7.56942563488665e-06, "loss": 0.2872, "mean_token_accuracy": 0.9247471988201141, "num_tokens": 56104379.0, "step": 1168 }, { "entropy": 0.3253948390483856, "epoch": 1.8012326656394453, "grad_norm": 0.797300398349762, "learning_rate": 7.552947905896586e-06, "loss": 0.311, "mean_token_accuracy": 0.9199561774730682, "num_tokens": 56159431.0, "step": 1169 }, { "entropy": 0.32777033746242523, "epoch": 1.802773497688752, "grad_norm": 0.7618340253829956, "learning_rate": 7.5364772412048885e-06, "loss": 0.3077, "mean_token_accuracy": 0.9213409721851349, "num_tokens": 56213207.0, "step": 1170 }, { "entropy": 0.31758512556552887, "epoch": 1.8043143297380584, "grad_norm": 0.792239248752594, "learning_rate": 7.5200136883600615e-06, "loss": 0.3101, "mean_token_accuracy": 0.9205600023269653, "num_tokens": 56264923.0, "step": 1171 }, { "entropy": 0.5625988245010376, "epoch": 1.805855161787365, "grad_norm": 0.7457577586174011, "learning_rate": 7.503557294890096e-06, "loss": 0.4213, "mean_token_accuracy": 0.8459064960479736, "num_tokens": 56294814.0, "step": 1172 }, { "entropy": 0.8463826477527618, "epoch": 1.8073959938366717, "grad_norm": 0.9401266574859619, "learning_rate": 7.487108108302301e-06, "loss": 0.8462, "mean_token_accuracy": 0.7743797898292542, "num_tokens": 56305936.0, "step": 1173 }, { "entropy": 0.7501126825809479, "epoch": 1.8089368258859784, "grad_norm": 1.050044059753418, "learning_rate": 7.470666176083193e-06, "loss": 0.7201, "mean_token_accuracy": 0.8119775652885437, "num_tokens": 56314017.0, "step": 1174 }, { "entropy": 0.24493146687746048, "epoch": 1.810477657935285, "grad_norm": 0.5191584229469299, "learning_rate": 7.454231545698334e-06, "loss": 0.225, "mean_token_accuracy": 0.9387161433696747, "num_tokens": 56379338.0, "step": 1175 }, { "entropy": 0.3639131486415863, "epoch": 1.8120184899845917, "grad_norm": 0.6978549957275391, "learning_rate": 7.4378042645922145e-06, "loss": 0.3347, "mean_token_accuracy": 0.9115226864814758, "num_tokens": 56441930.0, "step": 1176 }, { "entropy": 0.34345391392707825, "epoch": 1.8135593220338984, "grad_norm": 0.6720646023750305, "learning_rate": 7.421384380188098e-06, "loss": 0.3189, "mean_token_accuracy": 0.9158399999141693, "num_tokens": 56502411.0, "step": 1177 }, { "entropy": 0.3525722771883011, "epoch": 1.815100154083205, "grad_norm": 0.7247658371925354, "learning_rate": 7.40497193988791e-06, "loss": 0.3256, "mean_token_accuracy": 0.913998931646347, "num_tokens": 56561361.0, "step": 1178 }, { "entropy": 0.3189808875322342, "epoch": 1.8166409861325117, "grad_norm": 0.6512379050254822, "learning_rate": 7.388566991072077e-06, "loss": 0.3002, "mean_token_accuracy": 0.9223353862762451, "num_tokens": 56619029.0, "step": 1179 }, { "entropy": 0.3647850751876831, "epoch": 1.8181818181818183, "grad_norm": 0.7366579174995422, "learning_rate": 7.372169581099398e-06, "loss": 0.3405, "mean_token_accuracy": 0.9109580218791962, "num_tokens": 56675583.0, "step": 1180 }, { "entropy": 0.3276207894086838, "epoch": 1.8197226502311248, "grad_norm": 0.736655592918396, "learning_rate": 7.355779757306915e-06, "loss": 0.3083, "mean_token_accuracy": 0.9203782081604004, "num_tokens": 56731178.0, "step": 1181 }, { "entropy": 0.3154364824295044, "epoch": 1.8212634822804314, "grad_norm": 0.7973605394363403, "learning_rate": 7.339397567009758e-06, "loss": 0.2958, "mean_token_accuracy": 0.9231442809104919, "num_tokens": 56785684.0, "step": 1182 }, { "entropy": 0.29507899284362793, "epoch": 1.822804314329738, "grad_norm": 0.7493900060653687, "learning_rate": 7.323023057501033e-06, "loss": 0.2797, "mean_token_accuracy": 0.9262272715568542, "num_tokens": 56838843.0, "step": 1183 }, { "entropy": 0.28842417895793915, "epoch": 1.8243451463790445, "grad_norm": 0.7861377596855164, "learning_rate": 7.3066562760516615e-06, "loss": 0.2829, "mean_token_accuracy": 0.9275560081005096, "num_tokens": 56889063.0, "step": 1184 }, { "entropy": 0.83814537525177, "epoch": 1.8258859784283512, "grad_norm": 0.867538332939148, "learning_rate": 7.290297269910267e-06, "loss": 0.8642, "mean_token_accuracy": 0.7656330168247223, "num_tokens": 56901948.0, "step": 1185 }, { "entropy": 0.7643576264381409, "epoch": 1.8274268104776579, "grad_norm": 1.034626841545105, "learning_rate": 7.273946086303019e-06, "loss": 0.7699, "mean_token_accuracy": 0.8047168254852295, "num_tokens": 56911016.0, "step": 1186 }, { "entropy": 0.44265709817409515, "epoch": 1.8289676425269645, "grad_norm": 0.6064550876617432, "learning_rate": 7.257602772433499e-06, "loss": 0.2157, "mean_token_accuracy": 0.8880203366279602, "num_tokens": 56947340.0, "step": 1187 }, { "entropy": 0.2743529826402664, "epoch": 1.8305084745762712, "grad_norm": 0.6046243906021118, "learning_rate": 7.241267375482584e-06, "loss": 0.2532, "mean_token_accuracy": 0.9319040775299072, "num_tokens": 57011472.0, "step": 1188 }, { "entropy": 0.35499759018421173, "epoch": 1.8320493066255779, "grad_norm": 0.6965934634208679, "learning_rate": 7.224939942608287e-06, "loss": 0.3217, "mean_token_accuracy": 0.9132830202579498, "num_tokens": 57072774.0, "step": 1189 }, { "entropy": 0.34940867125988007, "epoch": 1.8335901386748845, "grad_norm": 0.6641712784767151, "learning_rate": 7.208620520945627e-06, "loss": 0.3194, "mean_token_accuracy": 0.9154815673828125, "num_tokens": 57132639.0, "step": 1190 }, { "entropy": 0.33172978460788727, "epoch": 1.8351309707241912, "grad_norm": 0.6846923232078552, "learning_rate": 7.192309157606508e-06, "loss": 0.2981, "mean_token_accuracy": 0.9215838611125946, "num_tokens": 57191138.0, "step": 1191 }, { "entropy": 0.33018895983695984, "epoch": 1.8366718027734978, "grad_norm": 0.6413344740867615, "learning_rate": 7.176005899679556e-06, "loss": 0.3023, "mean_token_accuracy": 0.9203876852989197, "num_tokens": 57248436.0, "step": 1192 }, { "entropy": 0.31592249870300293, "epoch": 1.8382126348228043, "grad_norm": 0.7025464773178101, "learning_rate": 7.159710794230006e-06, "loss": 0.2902, "mean_token_accuracy": 0.9242518544197083, "num_tokens": 57304486.0, "step": 1193 }, { "entropy": 0.2940081059932709, "epoch": 1.839753466872111, "grad_norm": 0.7215101718902588, "learning_rate": 7.143423888299554e-06, "loss": 0.2706, "mean_token_accuracy": 0.9283051192760468, "num_tokens": 57359514.0, "step": 1194 }, { "entropy": 0.3201545923948288, "epoch": 1.8412942989214176, "grad_norm": 0.7634199857711792, "learning_rate": 7.127145228906234e-06, "loss": 0.3061, "mean_token_accuracy": 0.9216424524784088, "num_tokens": 57413455.0, "step": 1195 }, { "entropy": 0.3037295937538147, "epoch": 1.842835130970724, "grad_norm": 0.7958951592445374, "learning_rate": 7.110874863044266e-06, "loss": 0.2956, "mean_token_accuracy": 0.9250476062297821, "num_tokens": 57465680.0, "step": 1196 }, { "entropy": 0.559918999671936, "epoch": 1.8443759630200307, "grad_norm": 0.8269791007041931, "learning_rate": 7.094612837683923e-06, "loss": 0.4033, "mean_token_accuracy": 0.851285308599472, "num_tokens": 57494512.0, "step": 1197 }, { "entropy": 0.7931991219520569, "epoch": 1.8459167950693374, "grad_norm": 0.9588444828987122, "learning_rate": 7.078359199771407e-06, "loss": 0.7969, "mean_token_accuracy": 0.7860711812973022, "num_tokens": 57504802.0, "step": 1198 }, { "entropy": 0.736746609210968, "epoch": 1.847457627118644, "grad_norm": 1.2912700176239014, "learning_rate": 7.06211399622871e-06, "loss": 0.7347, "mean_token_accuracy": 0.8093047142028809, "num_tokens": 57512581.0, "step": 1199 }, { "entropy": 0.20605303347110748, "epoch": 1.8489984591679507, "grad_norm": 0.5237095952033997, "learning_rate": 7.045877273953467e-06, "loss": 0.1918, "mean_token_accuracy": 0.9488480687141418, "num_tokens": 57577952.0, "step": 1200 }, { "epoch": 1.8489984591679507, "eval_entropy": 0.4412136339354065, "eval_loss": 0.4193393886089325, "eval_mean_token_accuracy": 0.8906174892524503, "eval_num_tokens": 57577952.0, "eval_runtime": 192.3164, "eval_samples_per_second": 4.409, "eval_steps_per_second": 1.102, "step": 1200 }, { "entropy": 0.31882497668266296, "epoch": 1.8505392912172574, "grad_norm": 0.6260185241699219, "learning_rate": 7.029649079818836e-06, "loss": 0.2991, "mean_token_accuracy": 0.9194160997867584, "num_tokens": 57640556.0, "step": 1201 }, { "entropy": 0.33837608993053436, "epoch": 1.852080123266564, "grad_norm": 0.6468535661697388, "learning_rate": 7.013429460673349e-06, "loss": 0.3167, "mean_token_accuracy": 0.9150879681110382, "num_tokens": 57701258.0, "step": 1202 }, { "entropy": 0.3163376599550247, "epoch": 1.8536209553158707, "grad_norm": 0.6739808917045593, "learning_rate": 6.997218463340786e-06, "loss": 0.2873, "mean_token_accuracy": 0.9233696460723877, "num_tokens": 57760584.0, "step": 1203 }, { "entropy": 0.3459166884422302, "epoch": 1.8551617873651773, "grad_norm": 0.6643741130828857, "learning_rate": 6.981016134620036e-06, "loss": 0.3093, "mean_token_accuracy": 0.9205880165100098, "num_tokens": 57818632.0, "step": 1204 }, { "entropy": 0.3165028691291809, "epoch": 1.8567026194144838, "grad_norm": 0.6545714735984802, "learning_rate": 6.964822521284966e-06, "loss": 0.2897, "mean_token_accuracy": 0.924932062625885, "num_tokens": 57875616.0, "step": 1205 }, { "entropy": 0.31493738293647766, "epoch": 1.8582434514637904, "grad_norm": 0.6993868350982666, "learning_rate": 6.9486376700842835e-06, "loss": 0.286, "mean_token_accuracy": 0.9251686632633209, "num_tokens": 57931721.0, "step": 1206 }, { "entropy": 0.30782550573349, "epoch": 1.859784283513097, "grad_norm": 0.7384077906608582, "learning_rate": 6.932461627741393e-06, "loss": 0.2937, "mean_token_accuracy": 0.923955500125885, "num_tokens": 57986519.0, "step": 1207 }, { "entropy": 0.3146795183420181, "epoch": 1.8613251155624035, "grad_norm": 0.7505443096160889, "learning_rate": 6.9162944409542744e-06, "loss": 0.2877, "mean_token_accuracy": 0.9256703853607178, "num_tokens": 58039752.0, "step": 1208 }, { "entropy": 0.3815041482448578, "epoch": 1.8628659476117102, "grad_norm": 0.873999297618866, "learning_rate": 6.900136156395347e-06, "loss": 0.3242, "mean_token_accuracy": 0.9146091938018799, "num_tokens": 58083961.0, "step": 1209 }, { "entropy": 0.8513643741607666, "epoch": 1.8644067796610169, "grad_norm": 0.8273122310638428, "learning_rate": 6.883986820711321e-06, "loss": 0.8447, "mean_token_accuracy": 0.7715265452861786, "num_tokens": 58095766.0, "step": 1210 }, { "entropy": 0.8173656761646271, "epoch": 1.8659476117103235, "grad_norm": 1.004235029220581, "learning_rate": 6.86784648052309e-06, "loss": 0.7968, "mean_token_accuracy": 0.7937957048416138, "num_tokens": 58105016.0, "step": 1211 }, { "entropy": 0.42676909267902374, "epoch": 1.8674884437596302, "grad_norm": 0.6359408497810364, "learning_rate": 6.851715182425551e-06, "loss": 0.1922, "mean_token_accuracy": 0.8958799839019775, "num_tokens": 58141322.0, "step": 1212 }, { "entropy": 0.2746562063694, "epoch": 1.8690292758089369, "grad_norm": 0.617998480796814, "learning_rate": 6.8355929729875216e-06, "loss": 0.2598, "mean_token_accuracy": 0.9307920634746552, "num_tokens": 58205688.0, "step": 1213 }, { "entropy": 0.32358184456825256, "epoch": 1.8705701078582435, "grad_norm": 0.680891215801239, "learning_rate": 6.819479898751571e-06, "loss": 0.3073, "mean_token_accuracy": 0.9173025488853455, "num_tokens": 58267455.0, "step": 1214 }, { "entropy": 0.31531183421611786, "epoch": 1.8721109399075502, "grad_norm": 0.673389732837677, "learning_rate": 6.803376006233903e-06, "loss": 0.3018, "mean_token_accuracy": 0.9213504791259766, "num_tokens": 58327669.0, "step": 1215 }, { "entropy": 0.3178231567144394, "epoch": 1.8736517719568568, "grad_norm": 0.7139577865600586, "learning_rate": 6.7872813419242055e-06, "loss": 0.3008, "mean_token_accuracy": 0.9209749400615692, "num_tokens": 58386394.0, "step": 1216 }, { "entropy": 0.2867015600204468, "epoch": 1.8751926040061633, "grad_norm": 0.6729170680046082, "learning_rate": 6.771195952285541e-06, "loss": 0.2644, "mean_token_accuracy": 0.9294416904449463, "num_tokens": 58443932.0, "step": 1217 }, { "entropy": 0.3235973119735718, "epoch": 1.87673343605547, "grad_norm": 0.6962930560112, "learning_rate": 6.7551198837541746e-06, "loss": 0.2992, "mean_token_accuracy": 0.9223398864269257, "num_tokens": 58500425.0, "step": 1218 }, { "entropy": 0.3091260492801666, "epoch": 1.8782742681047766, "grad_norm": 0.7034379243850708, "learning_rate": 6.739053182739484e-06, "loss": 0.2892, "mean_token_accuracy": 0.9256251752376556, "num_tokens": 58555766.0, "step": 1219 }, { "entropy": 0.3304354101419449, "epoch": 1.879815100154083, "grad_norm": 0.6990631818771362, "learning_rate": 6.722995895623791e-06, "loss": 0.2995, "mean_token_accuracy": 0.9235360324382782, "num_tokens": 58609637.0, "step": 1220 }, { "entropy": 0.30506083369255066, "epoch": 1.8813559322033897, "grad_norm": 0.7290918231010437, "learning_rate": 6.706948068762251e-06, "loss": 0.2839, "mean_token_accuracy": 0.9259768128395081, "num_tokens": 58661007.0, "step": 1221 }, { "entropy": 0.593474343419075, "epoch": 1.8828967642526964, "grad_norm": 0.7843896746635437, "learning_rate": 6.690909748482704e-06, "loss": 0.4119, "mean_token_accuracy": 0.8549410998821259, "num_tokens": 58686061.0, "step": 1222 }, { "entropy": 0.8467982709407806, "epoch": 1.884437596302003, "grad_norm": 0.8586987853050232, "learning_rate": 6.674880981085541e-06, "loss": 0.8182, "mean_token_accuracy": 0.7808684408664703, "num_tokens": 58696508.0, "step": 1223 }, { "entropy": 0.7150152325630188, "epoch": 1.8859784283513097, "grad_norm": 1.0447903871536255, "learning_rate": 6.658861812843578e-06, "loss": 0.7222, "mean_token_accuracy": 0.8166452348232269, "num_tokens": 58704544.0, "step": 1224 }, { "entropy": 0.2190827950835228, "epoch": 1.8875192604006163, "grad_norm": 0.5412318110466003, "learning_rate": 6.642852290001927e-06, "loss": 0.2064, "mean_token_accuracy": 0.9468713104724884, "num_tokens": 58770038.0, "step": 1225 }, { "entropy": 0.33239249885082245, "epoch": 1.889060092449923, "grad_norm": 0.7460736632347107, "learning_rate": 6.626852458777846e-06, "loss": 0.3172, "mean_token_accuracy": 0.9151608943939209, "num_tokens": 58833089.0, "step": 1226 }, { "entropy": 0.297749400138855, "epoch": 1.8906009244992297, "grad_norm": 0.7197936773300171, "learning_rate": 6.610862365360622e-06, "loss": 0.2883, "mean_token_accuracy": 0.9236486554145813, "num_tokens": 58893848.0, "step": 1227 }, { "entropy": 0.3031459450721741, "epoch": 1.8921417565485363, "grad_norm": 0.6871368885040283, "learning_rate": 6.594882055911419e-06, "loss": 0.2888, "mean_token_accuracy": 0.9235569834709167, "num_tokens": 58953222.0, "step": 1228 }, { "entropy": 0.2984921336174011, "epoch": 1.893682588597843, "grad_norm": 0.6936859488487244, "learning_rate": 6.57891157656317e-06, "loss": 0.2774, "mean_token_accuracy": 0.9263814091682434, "num_tokens": 59011181.0, "step": 1229 }, { "entropy": 0.34733277559280396, "epoch": 1.8952234206471494, "grad_norm": 0.6781644821166992, "learning_rate": 6.562950973420416e-06, "loss": 0.3267, "mean_token_accuracy": 0.9140821099281311, "num_tokens": 59067848.0, "step": 1230 }, { "entropy": 0.3121741861104965, "epoch": 1.896764252696456, "grad_norm": 0.695829451084137, "learning_rate": 6.547000292559201e-06, "loss": 0.2922, "mean_token_accuracy": 0.923791378736496, "num_tokens": 59123481.0, "step": 1231 }, { "entropy": 0.2985233664512634, "epoch": 1.8983050847457628, "grad_norm": 0.7175043225288391, "learning_rate": 6.531059580026917e-06, "loss": 0.2727, "mean_token_accuracy": 0.9289233386516571, "num_tokens": 59177870.0, "step": 1232 }, { "entropy": 0.30560559034347534, "epoch": 1.8998459167950692, "grad_norm": 0.6815491318702698, "learning_rate": 6.515128881842177e-06, "loss": 0.286, "mean_token_accuracy": 0.9270285367965698, "num_tokens": 59230810.0, "step": 1233 }, { "entropy": 0.26893389225006104, "epoch": 1.9013867488443759, "grad_norm": 0.697843611240387, "learning_rate": 6.499208243994683e-06, "loss": 0.244, "mean_token_accuracy": 0.9367959201335907, "num_tokens": 59281185.0, "step": 1234 }, { "entropy": 0.7556061744689941, "epoch": 1.9029275808936825, "grad_norm": 0.8869484066963196, "learning_rate": 6.483297712445107e-06, "loss": 0.6328, "mean_token_accuracy": 0.8129903972148895, "num_tokens": 59297626.0, "step": 1235 }, { "entropy": 0.7472674250602722, "epoch": 1.9044684129429892, "grad_norm": 0.9711276292800903, "learning_rate": 6.467397333124928e-06, "loss": 0.736, "mean_token_accuracy": 0.8033357858657837, "num_tokens": 59307309.0, "step": 1236 }, { "entropy": 0.44511231034994125, "epoch": 1.9060092449922958, "grad_norm": 0.6329291462898254, "learning_rate": 6.451507151936332e-06, "loss": 0.2234, "mean_token_accuracy": 0.8891220986843109, "num_tokens": 59343651.0, "step": 1237 }, { "entropy": 0.3080284595489502, "epoch": 1.9075500770416025, "grad_norm": 0.6630970239639282, "learning_rate": 6.435627214752047e-06, "loss": 0.2809, "mean_token_accuracy": 0.9240678250789642, "num_tokens": 59407991.0, "step": 1238 }, { "entropy": 0.3280094861984253, "epoch": 1.9090909090909092, "grad_norm": 0.7059434652328491, "learning_rate": 6.419757567415249e-06, "loss": 0.3163, "mean_token_accuracy": 0.917273223400116, "num_tokens": 59469665.0, "step": 1239 }, { "entropy": 0.32820703089237213, "epoch": 1.9106317411402158, "grad_norm": 0.7249607443809509, "learning_rate": 6.403898255739394e-06, "loss": 0.3143, "mean_token_accuracy": 0.918062299489975, "num_tokens": 59529956.0, "step": 1240 }, { "entropy": 0.31428663432598114, "epoch": 1.9121725731895225, "grad_norm": 0.6790779232978821, "learning_rate": 6.3880493255081035e-06, "loss": 0.2907, "mean_token_accuracy": 0.9227578938007355, "num_tokens": 59588970.0, "step": 1241 }, { "entropy": 0.32538874447345734, "epoch": 1.913713405238829, "grad_norm": 0.6491665840148926, "learning_rate": 6.372210822475034e-06, "loss": 0.3062, "mean_token_accuracy": 0.9210338592529297, "num_tokens": 59646954.0, "step": 1242 }, { "entropy": 0.3042677789926529, "epoch": 1.9152542372881356, "grad_norm": 0.7241361737251282, "learning_rate": 6.356382792363733e-06, "loss": 0.2823, "mean_token_accuracy": 0.9257973730564117, "num_tokens": 59703847.0, "step": 1243 }, { "entropy": 0.30474627017974854, "epoch": 1.9167950693374423, "grad_norm": 0.7272719144821167, "learning_rate": 6.340565280867519e-06, "loss": 0.2894, "mean_token_accuracy": 0.9242174029350281, "num_tokens": 59759650.0, "step": 1244 }, { "entropy": 0.30433104932308197, "epoch": 1.9183359013867487, "grad_norm": 0.7028172612190247, "learning_rate": 6.324758333649343e-06, "loss": 0.2918, "mean_token_accuracy": 0.9245812296867371, "num_tokens": 59814237.0, "step": 1245 }, { "entropy": 0.2813384532928467, "epoch": 1.9198767334360554, "grad_norm": 0.7520090937614441, "learning_rate": 6.308961996341658e-06, "loss": 0.2597, "mean_token_accuracy": 0.9334626197814941, "num_tokens": 59867256.0, "step": 1246 }, { "entropy": 0.29306626319885254, "epoch": 1.921417565485362, "grad_norm": 0.6643139719963074, "learning_rate": 6.293176314546292e-06, "loss": 0.2708, "mean_token_accuracy": 0.9301292300224304, "num_tokens": 59918333.0, "step": 1247 }, { "entropy": 0.7159150838851929, "epoch": 1.9229583975346687, "grad_norm": 0.9409279823303223, "learning_rate": 6.277401333834301e-06, "loss": 0.5317, "mean_token_accuracy": 0.8293076157569885, "num_tokens": 59938818.0, "step": 1248 }, { "entropy": 0.74931800365448, "epoch": 1.9244992295839753, "grad_norm": 0.954762876033783, "learning_rate": 6.261637099745863e-06, "loss": 0.7636, "mean_token_accuracy": 0.8029882609844208, "num_tokens": 59947677.0, "step": 1249 }, { "entropy": 0.19187350571155548, "epoch": 1.926040061633282, "grad_norm": 0.4301447868347168, "learning_rate": 6.245883657790122e-06, "loss": 0.1677, "mean_token_accuracy": 0.9547729194164276, "num_tokens": 60013059.0, "step": 1250 }, { "entropy": 0.3203873783349991, "epoch": 1.9275808936825887, "grad_norm": 0.6475263833999634, "learning_rate": 6.230141053445067e-06, "loss": 0.2971, "mean_token_accuracy": 0.9205348491668701, "num_tokens": 60075632.0, "step": 1251 }, { "entropy": 0.30758143961429596, "epoch": 1.9291217257318953, "grad_norm": 0.6333567500114441, "learning_rate": 6.214409332157408e-06, "loss": 0.2899, "mean_token_accuracy": 0.9225760996341705, "num_tokens": 60136140.0, "step": 1252 }, { "entropy": 0.3118809908628464, "epoch": 1.930662557781202, "grad_norm": 0.6756778955459595, "learning_rate": 6.198688539342431e-06, "loss": 0.2889, "mean_token_accuracy": 0.9239360392093658, "num_tokens": 60195205.0, "step": 1253 }, { "entropy": 0.30591122806072235, "epoch": 1.9322033898305084, "grad_norm": 0.9068213105201721, "learning_rate": 6.18297872038387e-06, "loss": 0.2826, "mean_token_accuracy": 0.9258987009525299, "num_tokens": 60253115.0, "step": 1254 }, { "entropy": 0.31331922113895416, "epoch": 1.933744221879815, "grad_norm": 0.7114205956459045, "learning_rate": 6.167279920633784e-06, "loss": 0.2943, "mean_token_accuracy": 0.9221931397914886, "num_tokens": 60309874.0, "step": 1255 }, { "entropy": 0.3171158879995346, "epoch": 1.9352850539291218, "grad_norm": 0.6912866830825806, "learning_rate": 6.1515921854124225e-06, "loss": 0.3008, "mean_token_accuracy": 0.9227756261825562, "num_tokens": 60365509.0, "step": 1256 }, { "entropy": 0.3163105547428131, "epoch": 1.9368258859784282, "grad_norm": 0.7657924890518188, "learning_rate": 6.135915560008088e-06, "loss": 0.3025, "mean_token_accuracy": 0.9220530092716217, "num_tokens": 60419873.0, "step": 1257 }, { "entropy": 0.3090207129716873, "epoch": 1.9383667180277349, "grad_norm": 0.7843337655067444, "learning_rate": 6.1202500896770175e-06, "loss": 0.2894, "mean_token_accuracy": 0.9252520501613617, "num_tokens": 60472672.0, "step": 1258 }, { "entropy": 0.3415939509868622, "epoch": 1.9399075500770415, "grad_norm": 0.8058279156684875, "learning_rate": 6.104595819643235e-06, "loss": 0.3013, "mean_token_accuracy": 0.9214912950992584, "num_tokens": 60518984.0, "step": 1259 }, { "entropy": 0.8558387756347656, "epoch": 1.9414483821263482, "grad_norm": 0.7863738536834717, "learning_rate": 6.088952795098442e-06, "loss": 0.8701, "mean_token_accuracy": 0.7623797655105591, "num_tokens": 60531358.0, "step": 1260 }, { "entropy": 0.768841952085495, "epoch": 1.9429892141756548, "grad_norm": 0.9549652934074402, "learning_rate": 6.073321061201865e-06, "loss": 0.7367, "mean_token_accuracy": 0.8041465580463409, "num_tokens": 60540510.0, "step": 1261 }, { "entropy": 0.4262720122933388, "epoch": 1.9445300462249615, "grad_norm": 0.6279612183570862, "learning_rate": 6.057700663080144e-06, "loss": 0.2177, "mean_token_accuracy": 0.8912063241004944, "num_tokens": 60576850.0, "step": 1262 }, { "entropy": 0.2629564329981804, "epoch": 1.9460708782742682, "grad_norm": 0.5478349924087524, "learning_rate": 6.042091645827194e-06, "loss": 0.2431, "mean_token_accuracy": 0.9337383508682251, "num_tokens": 60641188.0, "step": 1263 }, { "entropy": 0.2915605157613754, "epoch": 1.9476117103235748, "grad_norm": 0.5911458730697632, "learning_rate": 6.026494054504068e-06, "loss": 0.2694, "mean_token_accuracy": 0.928007960319519, "num_tokens": 60702921.0, "step": 1264 }, { "entropy": 0.319914847612381, "epoch": 1.9491525423728815, "grad_norm": 0.613426923751831, "learning_rate": 6.010907934138838e-06, "loss": 0.2951, "mean_token_accuracy": 0.9231474697589874, "num_tokens": 60762972.0, "step": 1265 }, { "entropy": 0.33114229142665863, "epoch": 1.950693374422188, "grad_norm": 0.6802210807800293, "learning_rate": 5.995333329726465e-06, "loss": 0.3091, "mean_token_accuracy": 0.91966712474823, "num_tokens": 60821659.0, "step": 1266 }, { "entropy": 0.30552470684051514, "epoch": 1.9522342064714946, "grad_norm": 0.6489818692207336, "learning_rate": 5.979770286228659e-06, "loss": 0.2839, "mean_token_accuracy": 0.9277896285057068, "num_tokens": 60879095.0, "step": 1267 }, { "entropy": 0.30958497524261475, "epoch": 1.9537750385208013, "grad_norm": 0.6942368745803833, "learning_rate": 5.964218848573762e-06, "loss": 0.2862, "mean_token_accuracy": 0.9276029169559479, "num_tokens": 60935249.0, "step": 1268 }, { "entropy": 0.299513041973114, "epoch": 1.9553158705701077, "grad_norm": 0.7170417904853821, "learning_rate": 5.9486790616566e-06, "loss": 0.2784, "mean_token_accuracy": 0.9274436235427856, "num_tokens": 60990273.0, "step": 1269 }, { "entropy": 0.3134317696094513, "epoch": 1.9568567026194144, "grad_norm": 0.7497215270996094, "learning_rate": 5.933150970338377e-06, "loss": 0.2931, "mean_token_accuracy": 0.9236960113048553, "num_tokens": 61044058.0, "step": 1270 }, { "entropy": 0.308674618601799, "epoch": 1.958397534668721, "grad_norm": 0.7462538480758667, "learning_rate": 5.917634619446526e-06, "loss": 0.29, "mean_token_accuracy": 0.9238549172878265, "num_tokens": 61096163.0, "step": 1271 }, { "entropy": 0.5035629943013191, "epoch": 1.9599383667180277, "grad_norm": 0.7945431470870972, "learning_rate": 5.902130053774593e-06, "loss": 0.3512, "mean_token_accuracy": 0.8704022765159607, "num_tokens": 61128953.0, "step": 1272 }, { "entropy": 0.7921432256698608, "epoch": 1.9614791987673343, "grad_norm": 0.9472218155860901, "learning_rate": 5.8866373180820976e-06, "loss": 0.7742, "mean_token_accuracy": 0.7898263037204742, "num_tokens": 61139839.0, "step": 1273 }, { "entropy": 0.7231127917766571, "epoch": 1.963020030816641, "grad_norm": 1.1658846139907837, "learning_rate": 5.871156457094405e-06, "loss": 0.7354, "mean_token_accuracy": 0.8083641529083252, "num_tokens": 61147815.0, "step": 1274 }, { "entropy": 0.24946749210357666, "epoch": 1.9645608628659477, "grad_norm": 0.588226318359375, "learning_rate": 5.855687515502604e-06, "loss": 0.2364, "mean_token_accuracy": 0.936528205871582, "num_tokens": 61212698.0, "step": 1275 }, { "entropy": 0.2893551290035248, "epoch": 1.9661016949152543, "grad_norm": 0.6046372652053833, "learning_rate": 5.840230537963376e-06, "loss": 0.2674, "mean_token_accuracy": 0.9298784136772156, "num_tokens": 61274585.0, "step": 1276 }, { "entropy": 0.31454887986183167, "epoch": 1.967642526964561, "grad_norm": 0.6693950891494751, "learning_rate": 5.824785569098858e-06, "loss": 0.2933, "mean_token_accuracy": 0.9208047091960907, "num_tokens": 61334530.0, "step": 1277 }, { "entropy": 0.30817975103855133, "epoch": 1.9691833590138677, "grad_norm": 0.6441570520401001, "learning_rate": 5.8093526534965315e-06, "loss": 0.2895, "mean_token_accuracy": 0.9234423339366913, "num_tokens": 61393173.0, "step": 1278 }, { "entropy": 0.29360994696617126, "epoch": 1.970724191063174, "grad_norm": 0.6320379972457886, "learning_rate": 5.7939318357090615e-06, "loss": 0.2721, "mean_token_accuracy": 0.9294228255748749, "num_tokens": 61450594.0, "step": 1279 }, { "entropy": 0.3366210460662842, "epoch": 1.9722650231124808, "grad_norm": 0.7085033059120178, "learning_rate": 5.778523160254199e-06, "loss": 0.3186, "mean_token_accuracy": 0.9187261164188385, "num_tokens": 61506946.0, "step": 1280 }, { "entropy": 0.2882189452648163, "epoch": 1.9738058551617874, "grad_norm": 0.6458709239959717, "learning_rate": 5.7631266716146535e-06, "loss": 0.2647, "mean_token_accuracy": 0.9312797486782074, "num_tokens": 61562351.0, "step": 1281 }, { "entropy": 0.29485076665878296, "epoch": 1.9753466872110939, "grad_norm": 0.6839344501495361, "learning_rate": 5.747742414237936e-06, "loss": 0.276, "mean_token_accuracy": 0.9292584955692291, "num_tokens": 61616547.0, "step": 1282 }, { "entropy": 0.30333346128463745, "epoch": 1.9768875192604005, "grad_norm": 0.709102988243103, "learning_rate": 5.732370432536258e-06, "loss": 0.2797, "mean_token_accuracy": 0.9283732175827026, "num_tokens": 61669097.0, "step": 1283 }, { "entropy": 0.5675656199455261, "epoch": 1.9784283513097072, "grad_norm": 0.850753664970398, "learning_rate": 5.7170107708863755e-06, "loss": 0.3832, "mean_token_accuracy": 0.8471953272819519, "num_tokens": 61700794.0, "step": 1284 }, { "entropy": 0.8300711512565613, "epoch": 1.9799691833590138, "grad_norm": 0.9054953455924988, "learning_rate": 5.701663473629504e-06, "loss": 0.8313, "mean_token_accuracy": 0.7761167585849762, "num_tokens": 61712033.0, "step": 1285 }, { "entropy": 0.74921715259552, "epoch": 1.9815100154083205, "grad_norm": 1.0180690288543701, "learning_rate": 5.686328585071148e-06, "loss": 0.7205, "mean_token_accuracy": 0.8090031147003174, "num_tokens": 61721400.0, "step": 1286 }, { "entropy": 0.4384000673890114, "epoch": 1.9830508474576272, "grad_norm": 0.5729281306266785, "learning_rate": 5.671006149480993e-06, "loss": 0.1836, "mean_token_accuracy": 0.8942499756813049, "num_tokens": 61757763.0, "step": 1287 }, { "entropy": 0.2904321402311325, "epoch": 1.9845916795069338, "grad_norm": 0.6420186758041382, "learning_rate": 5.655696211092776e-06, "loss": 0.2724, "mean_token_accuracy": 0.9257166683673859, "num_tokens": 61821687.0, "step": 1288 }, { "entropy": 0.3157305121421814, "epoch": 1.9861325115562405, "grad_norm": 0.698717474937439, "learning_rate": 5.640398814104156e-06, "loss": 0.3018, "mean_token_accuracy": 0.9204666018486023, "num_tokens": 61882622.0, "step": 1289 }, { "entropy": 0.30187617242336273, "epoch": 1.9876733436055471, "grad_norm": 0.6387500166893005, "learning_rate": 5.6251140026765815e-06, "loss": 0.2832, "mean_token_accuracy": 0.9260503351688385, "num_tokens": 61941738.0, "step": 1290 }, { "entropy": 0.31735581159591675, "epoch": 1.9892141756548536, "grad_norm": 0.6921149492263794, "learning_rate": 5.609841820935179e-06, "loss": 0.2984, "mean_token_accuracy": 0.9216937124729156, "num_tokens": 61999343.0, "step": 1291 }, { "entropy": 0.2762303650379181, "epoch": 1.9907550077041603, "grad_norm": 0.6270156502723694, "learning_rate": 5.594582312968602e-06, "loss": 0.2495, "mean_token_accuracy": 0.933588981628418, "num_tokens": 62055714.0, "step": 1292 }, { "entropy": 0.2908019870519638, "epoch": 1.992295839753467, "grad_norm": 0.6835677027702332, "learning_rate": 5.579335522828932e-06, "loss": 0.274, "mean_token_accuracy": 0.9293809533119202, "num_tokens": 62110874.0, "step": 1293 }, { "entropy": 0.2953278124332428, "epoch": 1.9938366718027734, "grad_norm": 0.7074451446533203, "learning_rate": 5.564101494531526e-06, "loss": 0.2754, "mean_token_accuracy": 0.9287161231040955, "num_tokens": 62164428.0, "step": 1294 }, { "entropy": 0.26990465819835663, "epoch": 1.99537750385208, "grad_norm": 0.6848849654197693, "learning_rate": 5.548880272054894e-06, "loss": 0.2484, "mean_token_accuracy": 0.935890793800354, "num_tokens": 62215307.0, "step": 1295 }, { "entropy": 0.7249132394790649, "epoch": 1.9969183359013867, "grad_norm": 0.876919686794281, "learning_rate": 5.5336718993405825e-06, "loss": 0.5875, "mean_token_accuracy": 0.8232371509075165, "num_tokens": 62232092.0, "step": 1296 }, { "entropy": 0.8052437007427216, "epoch": 1.9984591679506933, "grad_norm": 1.015784502029419, "learning_rate": 5.518476420293055e-06, "loss": 0.7846, "mean_token_accuracy": 0.7933857142925262, "num_tokens": 62241778.0, "step": 1297 }, { "entropy": 0.4652930349111557, "epoch": 2.0, "grad_norm": 0.6145098209381104, "learning_rate": 5.503293878779534e-06, "loss": 0.2241, "mean_token_accuracy": 0.8783581256866455, "num_tokens": 62258265.0, "step": 1298 }, { "entropy": 0.22899405658245087, "epoch": 2.0015408320493067, "grad_norm": 0.47440457344055176, "learning_rate": 5.488124318629908e-06, "loss": 0.2082, "mean_token_accuracy": 0.9442063271999359, "num_tokens": 62323341.0, "step": 1299 }, { "entropy": 0.3021625876426697, "epoch": 2.0030816640986133, "grad_norm": 0.5773673057556152, "learning_rate": 5.4729677836365695e-06, "loss": 0.2692, "mean_token_accuracy": 0.9278303980827332, "num_tokens": 62385636.0, "step": 1300 }, { "entropy": 0.2921716868877411, "epoch": 2.00462249614792, "grad_norm": 0.5950532555580139, "learning_rate": 5.4578243175543344e-06, "loss": 0.2717, "mean_token_accuracy": 0.9279645681381226, "num_tokens": 62446179.0, "step": 1301 }, { "entropy": 0.3086947053670883, "epoch": 2.0061633281972266, "grad_norm": 0.6225762367248535, "learning_rate": 5.442693964100276e-06, "loss": 0.2876, "mean_token_accuracy": 0.9246253073215485, "num_tokens": 62505432.0, "step": 1302 }, { "entropy": 0.2774394303560257, "epoch": 2.0077041602465333, "grad_norm": 0.6049097180366516, "learning_rate": 5.427576766953615e-06, "loss": 0.2516, "mean_token_accuracy": 0.9323847889900208, "num_tokens": 62563478.0, "step": 1303 }, { "entropy": 0.28888455033302307, "epoch": 2.00924499229584, "grad_norm": 0.6376782059669495, "learning_rate": 5.412472769755595e-06, "loss": 0.2694, "mean_token_accuracy": 0.9298742711544037, "num_tokens": 62620239.0, "step": 1304 }, { "entropy": 0.26781243085861206, "epoch": 2.010785824345146, "grad_norm": 0.639708399772644, "learning_rate": 5.397382016109347e-06, "loss": 0.2459, "mean_token_accuracy": 0.936321496963501, "num_tokens": 62675677.0, "step": 1305 }, { "entropy": 0.2921280264854431, "epoch": 2.012326656394453, "grad_norm": 0.6806097030639648, "learning_rate": 5.3823045495797775e-06, "loss": 0.2689, "mean_token_accuracy": 0.9315831661224365, "num_tokens": 62730017.0, "step": 1306 }, { "entropy": 0.2616446763277054, "epoch": 2.0138674884437595, "grad_norm": 0.6515586376190186, "learning_rate": 5.367240413693429e-06, "loss": 0.2407, "mean_token_accuracy": 0.9376666843891144, "num_tokens": 62782488.0, "step": 1307 }, { "entropy": 0.32007932662963867, "epoch": 2.015408320493066, "grad_norm": 0.6376478672027588, "learning_rate": 5.352189651938362e-06, "loss": 0.2521, "mean_token_accuracy": 0.9316624999046326, "num_tokens": 62826630.0, "step": 1308 }, { "entropy": 0.8238591849803925, "epoch": 2.016949152542373, "grad_norm": 0.7824305891990662, "learning_rate": 5.337152307764031e-06, "loss": 0.7807, "mean_token_accuracy": 0.785490095615387, "num_tokens": 62838432.0, "step": 1309 }, { "entropy": 0.7276124954223633, "epoch": 2.0184899845916795, "grad_norm": 0.995245099067688, "learning_rate": 5.3221284245811485e-06, "loss": 0.6315, "mean_token_accuracy": 0.8279706239700317, "num_tokens": 62847734.0, "step": 1310 }, { "entropy": 0.4208463802933693, "epoch": 2.020030816640986, "grad_norm": 0.6513915657997131, "learning_rate": 5.3071180457615755e-06, "loss": 0.198, "mean_token_accuracy": 0.8966114521026611, "num_tokens": 62883989.0, "step": 1311 }, { "entropy": 0.21689476817846298, "epoch": 2.021571648690293, "grad_norm": 0.5401560068130493, "learning_rate": 5.2921212146381815e-06, "loss": 0.2009, "mean_token_accuracy": 0.9446318447589874, "num_tokens": 62948206.0, "step": 1312 }, { "entropy": 0.2780810594558716, "epoch": 2.0231124807395995, "grad_norm": 0.6525228023529053, "learning_rate": 5.2771379745047255e-06, "loss": 0.2646, "mean_token_accuracy": 0.9299938380718231, "num_tokens": 63009596.0, "step": 1313 }, { "entropy": 0.2826758474111557, "epoch": 2.024653312788906, "grad_norm": 0.6853611469268799, "learning_rate": 5.2621683686157445e-06, "loss": 0.2757, "mean_token_accuracy": 0.9257222712039948, "num_tokens": 63069387.0, "step": 1314 }, { "entropy": 0.29403775185346603, "epoch": 2.026194144838213, "grad_norm": 0.6772225499153137, "learning_rate": 5.2472124401863936e-06, "loss": 0.2822, "mean_token_accuracy": 0.9272791147232056, "num_tokens": 63127718.0, "step": 1315 }, { "entropy": 0.28130224347114563, "epoch": 2.0277349768875195, "grad_norm": 0.6929225921630859, "learning_rate": 5.2322702323923544e-06, "loss": 0.2678, "mean_token_accuracy": 0.9293368458747864, "num_tokens": 63184782.0, "step": 1316 }, { "entropy": 0.274248406291008, "epoch": 2.0292758089368257, "grad_norm": 0.6942499876022339, "learning_rate": 5.217341788369699e-06, "loss": 0.251, "mean_token_accuracy": 0.9343591034412384, "num_tokens": 63240830.0, "step": 1317 }, { "entropy": 0.2827557474374771, "epoch": 2.0308166409861323, "grad_norm": 0.6554699540138245, "learning_rate": 5.2024271512147706e-06, "loss": 0.263, "mean_token_accuracy": 0.9322314858436584, "num_tokens": 63295812.0, "step": 1318 }, { "entropy": 0.26224808394908905, "epoch": 2.032357473035439, "grad_norm": 0.6780679225921631, "learning_rate": 5.187526363984048e-06, "loss": 0.2426, "mean_token_accuracy": 0.9364317655563354, "num_tokens": 63349707.0, "step": 1319 }, { "entropy": 0.2535221353173256, "epoch": 2.0338983050847457, "grad_norm": 0.6134437918663025, "learning_rate": 5.172639469694014e-06, "loss": 0.2332, "mean_token_accuracy": 0.9407214522361755, "num_tokens": 63402159.0, "step": 1320 }, { "entropy": 0.4696149379014969, "epoch": 2.0354391371340523, "grad_norm": 0.6936350464820862, "learning_rate": 5.157766511321074e-06, "loss": 0.3345, "mean_token_accuracy": 0.8919719755649567, "num_tokens": 63437872.0, "step": 1321 }, { "entropy": 0.6997888684272766, "epoch": 2.036979969183359, "grad_norm": 0.9067084193229675, "learning_rate": 5.142907531801378e-06, "loss": 0.6969, "mean_token_accuracy": 0.8123534917831421, "num_tokens": 63449250.0, "step": 1322 }, { "entropy": 0.6268580555915833, "epoch": 2.0385208012326657, "grad_norm": 1.1152706146240234, "learning_rate": 5.1280625740307296e-06, "loss": 0.6227, "mean_token_accuracy": 0.8330096900463104, "num_tokens": 63457387.0, "step": 1323 }, { "entropy": 0.18317259848117828, "epoch": 2.0400616332819723, "grad_norm": 0.4544285535812378, "learning_rate": 5.113231680864454e-06, "loss": 0.1651, "mean_token_accuracy": 0.9565164744853973, "num_tokens": 63522767.0, "step": 1324 }, { "entropy": 0.26821014285087585, "epoch": 2.041602465331279, "grad_norm": 0.5494820475578308, "learning_rate": 5.098414895117273e-06, "loss": 0.2483, "mean_token_accuracy": 0.9329530596733093, "num_tokens": 63585550.0, "step": 1325 }, { "entropy": 0.26001501083374023, "epoch": 2.0431432973805856, "grad_norm": 0.6235042214393616, "learning_rate": 5.083612259563182e-06, "loss": 0.2415, "mean_token_accuracy": 0.9339691698551178, "num_tokens": 63646305.0, "step": 1326 }, { "entropy": 0.27473975718021393, "epoch": 2.0446841294298923, "grad_norm": 0.5840004682540894, "learning_rate": 5.0688238169353265e-06, "loss": 0.2559, "mean_token_accuracy": 0.932563304901123, "num_tokens": 63705545.0, "step": 1327 }, { "entropy": 0.25361980497837067, "epoch": 2.046224961479199, "grad_norm": 0.5901036858558655, "learning_rate": 5.054049609925878e-06, "loss": 0.2324, "mean_token_accuracy": 0.9379728734493256, "num_tokens": 63763526.0, "step": 1328 }, { "entropy": 0.29010894894599915, "epoch": 2.047765793528505, "grad_norm": 0.6225770711898804, "learning_rate": 5.039289681185913e-06, "loss": 0.2736, "mean_token_accuracy": 0.9289602637290955, "num_tokens": 63820366.0, "step": 1329 }, { "entropy": 0.29254359006881714, "epoch": 2.049306625577812, "grad_norm": 0.6333932280540466, "learning_rate": 5.024544073325297e-06, "loss": 0.2735, "mean_token_accuracy": 0.9283972084522247, "num_tokens": 63876102.0, "step": 1330 }, { "entropy": 0.26660867780447006, "epoch": 2.0508474576271185, "grad_norm": 0.6312490701675415, "learning_rate": 5.009812828912535e-06, "loss": 0.2524, "mean_token_accuracy": 0.9342674314975739, "num_tokens": 63930754.0, "step": 1331 }, { "entropy": 0.2602839469909668, "epoch": 2.052388289676425, "grad_norm": 0.6170662641525269, "learning_rate": 4.995095990474679e-06, "loss": 0.2449, "mean_token_accuracy": 0.9368104040622711, "num_tokens": 63984031.0, "step": 1332 }, { "entropy": 0.23418794572353363, "epoch": 2.053929121725732, "grad_norm": 0.6175838708877563, "learning_rate": 4.9803936004971874e-06, "loss": 0.2132, "mean_token_accuracy": 0.9437951743602753, "num_tokens": 64034844.0, "step": 1333 }, { "entropy": 0.661401093006134, "epoch": 2.0554699537750385, "grad_norm": 0.8260689377784729, "learning_rate": 4.96570570142382e-06, "loss": 0.5589, "mean_token_accuracy": 0.837171196937561, "num_tokens": 64050868.0, "step": 1334 }, { "entropy": 0.6861454546451569, "epoch": 2.057010785824345, "grad_norm": 0.9896096587181091, "learning_rate": 4.951032335656493e-06, "loss": 0.673, "mean_token_accuracy": 0.820949912071228, "num_tokens": 64060259.0, "step": 1335 }, { "entropy": 0.40395066142082214, "epoch": 2.058551617873652, "grad_norm": 0.5506385564804077, "learning_rate": 4.9363735455551606e-06, "loss": 0.1858, "mean_token_accuracy": 0.8980529606342316, "num_tokens": 64096601.0, "step": 1336 }, { "entropy": 0.2659422606229782, "epoch": 2.0600924499229585, "grad_norm": 0.5541557669639587, "learning_rate": 4.921729373437708e-06, "loss": 0.2543, "mean_token_accuracy": 0.9318082630634308, "num_tokens": 64160619.0, "step": 1337 }, { "entropy": 0.2715007960796356, "epoch": 2.061633281972265, "grad_norm": 0.6016919612884521, "learning_rate": 4.907099861579827e-06, "loss": 0.2541, "mean_token_accuracy": 0.9318198263645172, "num_tokens": 64222407.0, "step": 1338 }, { "entropy": 0.315929114818573, "epoch": 2.063174114021572, "grad_norm": 0.7051572799682617, "learning_rate": 4.892485052214873e-06, "loss": 0.3002, "mean_token_accuracy": 0.9212778806686401, "num_tokens": 64282396.0, "step": 1339 }, { "entropy": 0.2900830805301666, "epoch": 2.0647149460708785, "grad_norm": 0.6468941569328308, "learning_rate": 4.8778849875337716e-06, "loss": 0.2755, "mean_token_accuracy": 0.9277671873569489, "num_tokens": 64341060.0, "step": 1340 }, { "entropy": 0.24873962998390198, "epoch": 2.066255778120185, "grad_norm": 0.5713239908218384, "learning_rate": 4.863299709684861e-06, "loss": 0.2253, "mean_token_accuracy": 0.9401804804801941, "num_tokens": 64398579.0, "step": 1341 }, { "entropy": 0.2576713413000107, "epoch": 2.0677966101694913, "grad_norm": 0.6067392230033875, "learning_rate": 4.848729260773818e-06, "loss": 0.2311, "mean_token_accuracy": 0.9404649436473846, "num_tokens": 64455092.0, "step": 1342 }, { "entropy": 0.2854810208082199, "epoch": 2.069337442218798, "grad_norm": 0.6339489817619324, "learning_rate": 4.834173682863492e-06, "loss": 0.2626, "mean_token_accuracy": 0.9316912591457367, "num_tokens": 64510475.0, "step": 1343 }, { "entropy": 0.2700754255056381, "epoch": 2.0708782742681047, "grad_norm": 0.6219398975372314, "learning_rate": 4.819633017973812e-06, "loss": 0.2513, "mean_token_accuracy": 0.9353560209274292, "num_tokens": 64564744.0, "step": 1344 }, { "entropy": 0.25295204669237137, "epoch": 2.0724191063174113, "grad_norm": 0.5881286263465881, "learning_rate": 4.805107308081647e-06, "loss": 0.2309, "mean_token_accuracy": 0.9401419758796692, "num_tokens": 64617539.0, "step": 1345 }, { "entropy": 0.3167862296104431, "epoch": 2.073959938366718, "grad_norm": 0.661246120929718, "learning_rate": 4.790596595120699e-06, "loss": 0.2545, "mean_token_accuracy": 0.9298122823238373, "num_tokens": 64661865.0, "step": 1346 }, { "entropy": 0.7306373119354248, "epoch": 2.0755007704160247, "grad_norm": 0.8886509537696838, "learning_rate": 4.776100920981374e-06, "loss": 0.7104, "mean_token_accuracy": 0.8056884407997131, "num_tokens": 64673482.0, "step": 1347 }, { "entropy": 0.6297113597393036, "epoch": 2.0770416024653313, "grad_norm": 1.1017975807189941, "learning_rate": 4.761620327510662e-06, "loss": 0.5848, "mean_token_accuracy": 0.8449733853340149, "num_tokens": 64681716.0, "step": 1348 }, { "entropy": 0.20797918736934662, "epoch": 2.078582434514638, "grad_norm": 0.4912075400352478, "learning_rate": 4.747154856512016e-06, "loss": 0.1892, "mean_token_accuracy": 0.9481683373451233, "num_tokens": 64746996.0, "step": 1349 }, { "entropy": 0.2747947722673416, "epoch": 2.0801232665639446, "grad_norm": 0.5923528075218201, "learning_rate": 4.732704549745247e-06, "loss": 0.2534, "mean_token_accuracy": 0.9319902658462524, "num_tokens": 64809468.0, "step": 1350 }, { "entropy": 0.2684113085269928, "epoch": 2.0816640986132513, "grad_norm": 0.5506814122200012, "learning_rate": 4.718269448926361e-06, "loss": 0.2512, "mean_token_accuracy": 0.9313322603702545, "num_tokens": 64870033.0, "step": 1351 }, { "entropy": 0.2880490571260452, "epoch": 2.083204930662558, "grad_norm": 0.6906023621559143, "learning_rate": 4.703849595727491e-06, "loss": 0.269, "mean_token_accuracy": 0.9290967881679535, "num_tokens": 64929104.0, "step": 1352 }, { "entropy": 0.2729254215955734, "epoch": 2.084745762711864, "grad_norm": 0.5979353189468384, "learning_rate": 4.689445031776737e-06, "loss": 0.2519, "mean_token_accuracy": 0.9344661235809326, "num_tokens": 64986963.0, "step": 1353 }, { "entropy": 0.2874646335840225, "epoch": 2.086286594761171, "grad_norm": 0.6630030274391174, "learning_rate": 4.675055798658077e-06, "loss": 0.2655, "mean_token_accuracy": 0.9298277497291565, "num_tokens": 65043712.0, "step": 1354 }, { "entropy": 0.2813657224178314, "epoch": 2.0878274268104775, "grad_norm": 0.6261852979660034, "learning_rate": 4.66068193791122e-06, "loss": 0.2604, "mean_token_accuracy": 0.9313385784626007, "num_tokens": 65099292.0, "step": 1355 }, { "entropy": 0.2706831693649292, "epoch": 2.089368258859784, "grad_norm": 0.6456521153450012, "learning_rate": 4.646323491031492e-06, "loss": 0.2514, "mean_token_accuracy": 0.9346813261508942, "num_tokens": 65153737.0, "step": 1356 }, { "entropy": 0.24387428164482117, "epoch": 2.090909090909091, "grad_norm": 0.6099643111228943, "learning_rate": 4.6319804994697284e-06, "loss": 0.218, "mean_token_accuracy": 0.9441450834274292, "num_tokens": 65206518.0, "step": 1357 }, { "entropy": 0.33616239577531815, "epoch": 2.0924499229583975, "grad_norm": 0.7115060091018677, "learning_rate": 4.6176530046321534e-06, "loss": 0.2651, "mean_token_accuracy": 0.9259044528007507, "num_tokens": 65248629.0, "step": 1358 }, { "entropy": 0.7511560022830963, "epoch": 2.093990755007704, "grad_norm": 0.9472434520721436, "learning_rate": 4.603341047880244e-06, "loss": 0.7425, "mean_token_accuracy": 0.7929057776927948, "num_tokens": 65260520.0, "step": 1359 }, { "entropy": 0.6816588342189789, "epoch": 2.095531587057011, "grad_norm": 1.1621543169021606, "learning_rate": 4.5890446705306315e-06, "loss": 0.6573, "mean_token_accuracy": 0.8230663537979126, "num_tokens": 65269916.0, "step": 1360 }, { "entropy": 0.39982566237449646, "epoch": 2.0970724191063175, "grad_norm": 0.5265263915061951, "learning_rate": 4.574763913854949e-06, "loss": 0.1973, "mean_token_accuracy": 0.8938726782798767, "num_tokens": 65306313.0, "step": 1361 }, { "entropy": 0.2670069709420204, "epoch": 2.098613251155624, "grad_norm": 0.5190237164497375, "learning_rate": 4.560498819079766e-06, "loss": 0.2444, "mean_token_accuracy": 0.9335883557796478, "num_tokens": 65370662.0, "step": 1362 }, { "entropy": 0.2681414484977722, "epoch": 2.100154083204931, "grad_norm": 0.5945353507995605, "learning_rate": 4.546249427386417e-06, "loss": 0.2503, "mean_token_accuracy": 0.9334887266159058, "num_tokens": 65431861.0, "step": 1363 }, { "entropy": 0.26357516646385193, "epoch": 2.1016949152542375, "grad_norm": 0.6125047206878662, "learning_rate": 4.532015779910911e-06, "loss": 0.2451, "mean_token_accuracy": 0.9346965253353119, "num_tokens": 65491504.0, "step": 1364 }, { "entropy": 0.29018092155456543, "epoch": 2.103235747303544, "grad_norm": 0.6380603909492493, "learning_rate": 4.5177979177438035e-06, "loss": 0.2757, "mean_token_accuracy": 0.9266772270202637, "num_tokens": 65549963.0, "step": 1365 }, { "entropy": 0.278698205947876, "epoch": 2.1047765793528503, "grad_norm": 0.6689497828483582, "learning_rate": 4.503595881930083e-06, "loss": 0.2644, "mean_token_accuracy": 0.930324524641037, "num_tokens": 65607331.0, "step": 1366 }, { "entropy": 0.274218425154686, "epoch": 2.106317411402157, "grad_norm": 0.6123804450035095, "learning_rate": 4.489409713469046e-06, "loss": 0.2594, "mean_token_accuracy": 0.932293564081192, "num_tokens": 65663587.0, "step": 1367 }, { "entropy": 0.2900755703449249, "epoch": 2.1078582434514637, "grad_norm": 0.6385893225669861, "learning_rate": 4.475239453314184e-06, "loss": 0.2727, "mean_token_accuracy": 0.9291305840015411, "num_tokens": 65718693.0, "step": 1368 }, { "entropy": 0.28037160634994507, "epoch": 2.1093990755007703, "grad_norm": 0.6351566910743713, "learning_rate": 4.461085142373064e-06, "loss": 0.2588, "mean_token_accuracy": 0.9338101446628571, "num_tokens": 65772606.0, "step": 1369 }, { "entropy": 0.26645004749298096, "epoch": 2.110939907550077, "grad_norm": 0.682149350643158, "learning_rate": 4.446946821507205e-06, "loss": 0.2427, "mean_token_accuracy": 0.9355919361114502, "num_tokens": 65824726.0, "step": 1370 }, { "entropy": 0.43963515758514404, "epoch": 2.1124807395993837, "grad_norm": 0.656681478023529, "learning_rate": 4.432824531531979e-06, "loss": 0.313, "mean_token_accuracy": 0.8970727026462555, "num_tokens": 65859816.0, "step": 1371 }, { "entropy": 0.7406118512153625, "epoch": 2.1140215716486903, "grad_norm": 1.0602004528045654, "learning_rate": 4.418718313216461e-06, "loss": 0.731, "mean_token_accuracy": 0.8037857115268707, "num_tokens": 65870814.0, "step": 1372 }, { "entropy": 0.6153799891471863, "epoch": 2.115562403697997, "grad_norm": 1.2402033805847168, "learning_rate": 4.404628207283338e-06, "loss": 0.584, "mean_token_accuracy": 0.8442805409431458, "num_tokens": 65878978.0, "step": 1373 }, { "entropy": 0.20543889701366425, "epoch": 2.1171032357473036, "grad_norm": 0.5004975199699402, "learning_rate": 4.390554254408778e-06, "loss": 0.1829, "mean_token_accuracy": 0.9498016536235809, "num_tokens": 65944122.0, "step": 1374 }, { "entropy": 0.29180993139743805, "epoch": 2.1186440677966103, "grad_norm": 0.5827184915542603, "learning_rate": 4.376496495222332e-06, "loss": 0.2666, "mean_token_accuracy": 0.9274333715438843, "num_tokens": 66006380.0, "step": 1375 }, { "entropy": 0.28782977163791656, "epoch": 2.120184899845917, "grad_norm": 0.6263056993484497, "learning_rate": 4.3624549703067895e-06, "loss": 0.2695, "mean_token_accuracy": 0.9270564913749695, "num_tokens": 66066987.0, "step": 1376 }, { "entropy": 0.2860146760940552, "epoch": 2.1217257318952236, "grad_norm": 0.6401958465576172, "learning_rate": 4.348429720198065e-06, "loss": 0.2682, "mean_token_accuracy": 0.9290086328983307, "num_tokens": 66126187.0, "step": 1377 }, { "entropy": 0.2842288911342621, "epoch": 2.12326656394453, "grad_norm": 0.6363611817359924, "learning_rate": 4.334420785385113e-06, "loss": 0.2615, "mean_token_accuracy": 0.930781900882721, "num_tokens": 66184105.0, "step": 1378 }, { "entropy": 0.2788894772529602, "epoch": 2.1248073959938365, "grad_norm": 0.6523936986923218, "learning_rate": 4.320428206309771e-06, "loss": 0.2618, "mean_token_accuracy": 0.9308032989501953, "num_tokens": 66240960.0, "step": 1379 }, { "entropy": 0.2570837438106537, "epoch": 2.126348228043143, "grad_norm": 0.6444106101989746, "learning_rate": 4.3064520233666665e-06, "loss": 0.2347, "mean_token_accuracy": 0.9380612075328827, "num_tokens": 66296644.0, "step": 1380 }, { "entropy": 0.2527001053094864, "epoch": 2.12788906009245, "grad_norm": 0.6190477013587952, "learning_rate": 4.292492276903091e-06, "loss": 0.235, "mean_token_accuracy": 0.9390477538108826, "num_tokens": 66350987.0, "step": 1381 }, { "entropy": 0.26765842735767365, "epoch": 2.1294298921417565, "grad_norm": 0.66351717710495, "learning_rate": 4.278549007218885e-06, "loss": 0.2501, "mean_token_accuracy": 0.9358106553554535, "num_tokens": 66403654.0, "step": 1382 }, { "entropy": 0.26185592263936996, "epoch": 2.130970724191063, "grad_norm": 0.6378859877586365, "learning_rate": 4.264622254566328e-06, "loss": 0.2443, "mean_token_accuracy": 0.9379086494445801, "num_tokens": 66451571.0, "step": 1383 }, { "entropy": 0.7261480093002319, "epoch": 2.13251155624037, "grad_norm": 0.9007375240325928, "learning_rate": 4.2507120591500125e-06, "loss": 0.7232, "mean_token_accuracy": 0.7948670387268066, "num_tokens": 66463700.0, "step": 1384 }, { "entropy": 0.6538786888122559, "epoch": 2.1340523882896765, "grad_norm": 1.111098289489746, "learning_rate": 4.2368184611267345e-06, "loss": 0.6307, "mean_token_accuracy": 0.8333071768283844, "num_tokens": 66473166.0, "step": 1385 }, { "entropy": 0.3823787346482277, "epoch": 2.135593220338983, "grad_norm": 0.5833919644355774, "learning_rate": 4.222941500605375e-06, "loss": 0.1824, "mean_token_accuracy": 0.9040353000164032, "num_tokens": 66509454.0, "step": 1386 }, { "entropy": 0.25638750940561295, "epoch": 2.13713405238829, "grad_norm": 0.5601032376289368, "learning_rate": 4.209081217646785e-06, "loss": 0.2395, "mean_token_accuracy": 0.9330670535564423, "num_tokens": 66573584.0, "step": 1387 }, { "entropy": 0.2562527656555176, "epoch": 2.1386748844375965, "grad_norm": 0.608620822429657, "learning_rate": 4.195237652263669e-06, "loss": 0.2352, "mean_token_accuracy": 0.9361146092414856, "num_tokens": 66634889.0, "step": 1388 }, { "entropy": 0.2574703246355057, "epoch": 2.140215716486903, "grad_norm": 0.6372120380401611, "learning_rate": 4.181410844420473e-06, "loss": 0.24, "mean_token_accuracy": 0.9370808303356171, "num_tokens": 66694486.0, "step": 1389 }, { "entropy": 0.2659468948841095, "epoch": 2.1417565485362093, "grad_norm": 0.6072478890419006, "learning_rate": 4.1676008340332596e-06, "loss": 0.2439, "mean_token_accuracy": 0.9343952238559723, "num_tokens": 66752869.0, "step": 1390 }, { "entropy": 0.2744990140199661, "epoch": 2.143297380585516, "grad_norm": 0.5877384543418884, "learning_rate": 4.153807660969614e-06, "loss": 0.256, "mean_token_accuracy": 0.9329849183559418, "num_tokens": 66810250.0, "step": 1391 }, { "entropy": 0.25395648181438446, "epoch": 2.1448382126348227, "grad_norm": 0.6395219564437866, "learning_rate": 4.1400313650484955e-06, "loss": 0.2312, "mean_token_accuracy": 0.9387475252151489, "num_tokens": 66866642.0, "step": 1392 }, { "entropy": 0.23837856203317642, "epoch": 2.1463790446841293, "grad_norm": 0.5517833232879639, "learning_rate": 4.126271986040152e-06, "loss": 0.2199, "mean_token_accuracy": 0.9417974948883057, "num_tokens": 66921899.0, "step": 1393 }, { "entropy": 0.27944718301296234, "epoch": 2.147919876733436, "grad_norm": 0.630229115486145, "learning_rate": 4.112529563665992e-06, "loss": 0.2538, "mean_token_accuracy": 0.9341011643409729, "num_tokens": 66975840.0, "step": 1394 }, { "entropy": 0.22655341029167175, "epoch": 2.1494607087827426, "grad_norm": 0.6160709261894226, "learning_rate": 4.0988041375984784e-06, "loss": 0.2088, "mean_token_accuracy": 0.9459251165390015, "num_tokens": 67028010.0, "step": 1395 }, { "entropy": 0.5129545107483864, "epoch": 2.1510015408320493, "grad_norm": 0.7403218746185303, "learning_rate": 4.085095747461005e-06, "loss": 0.3618, "mean_token_accuracy": 0.8590262234210968, "num_tokens": 67059893.0, "step": 1396 }, { "entropy": 0.7404741644859314, "epoch": 2.152542372881356, "grad_norm": 0.9581102728843689, "learning_rate": 4.071404432827777e-06, "loss": 0.7524, "mean_token_accuracy": 0.7928842604160309, "num_tokens": 67071264.0, "step": 1397 }, { "entropy": 0.6353053152561188, "epoch": 2.1540832049306626, "grad_norm": 1.1496427059173584, "learning_rate": 4.05773023322371e-06, "loss": 0.6401, "mean_token_accuracy": 0.8395758867263794, "num_tokens": 67079658.0, "step": 1398 }, { "entropy": 0.18487559258937836, "epoch": 2.1556240369799693, "grad_norm": 0.48300209641456604, "learning_rate": 4.04407318812432e-06, "loss": 0.1689, "mean_token_accuracy": 0.954853892326355, "num_tokens": 67145104.0, "step": 1399 }, { "entropy": 0.2655838802456856, "epoch": 2.157164869029276, "grad_norm": 0.5780742168426514, "learning_rate": 4.030433336955591e-06, "loss": 0.2473, "mean_token_accuracy": 0.9330219626426697, "num_tokens": 67208199.0, "step": 1400 }, { "epoch": 2.157164869029276, "eval_entropy": 0.39462503752196737, "eval_loss": 0.40249186754226685, "eval_mean_token_accuracy": 0.8954614218113557, "eval_num_tokens": 67208199.0, "eval_runtime": 192.5985, "eval_samples_per_second": 4.403, "eval_steps_per_second": 1.101, "step": 1400 }, { "entropy": 0.2815244793891907, "epoch": 2.1587057010785826, "grad_norm": 0.6471758484840393, "learning_rate": 4.01681071909387e-06, "loss": 0.2632, "mean_token_accuracy": 0.9300617575645447, "num_tokens": 67269419.0, "step": 1401 }, { "entropy": 0.2736307382583618, "epoch": 2.1602465331278893, "grad_norm": 0.620723307132721, "learning_rate": 4.003205373865758e-06, "loss": 0.2567, "mean_token_accuracy": 0.9314944446086884, "num_tokens": 67329112.0, "step": 1402 }, { "entropy": 0.29094424843788147, "epoch": 2.1617873651771955, "grad_norm": 0.6229572892189026, "learning_rate": 3.98961734054799e-06, "loss": 0.2626, "mean_token_accuracy": 0.930304765701294, "num_tokens": 67387631.0, "step": 1403 }, { "entropy": 0.25587955117225647, "epoch": 2.163328197226502, "grad_norm": 0.6329513192176819, "learning_rate": 3.976046658367325e-06, "loss": 0.24, "mean_token_accuracy": 0.9369424283504486, "num_tokens": 67444928.0, "step": 1404 }, { "entropy": 0.256832480430603, "epoch": 2.164869029275809, "grad_norm": 0.6385379433631897, "learning_rate": 3.962493366500429e-06, "loss": 0.2377, "mean_token_accuracy": 0.9366275370121002, "num_tokens": 67501005.0, "step": 1405 }, { "entropy": 0.26805379986763, "epoch": 2.1664098613251155, "grad_norm": 0.6095177531242371, "learning_rate": 3.948957504073767e-06, "loss": 0.2397, "mean_token_accuracy": 0.9386517405509949, "num_tokens": 67555876.0, "step": 1406 }, { "entropy": 0.25142184644937515, "epoch": 2.167950693374422, "grad_norm": 0.5781794786453247, "learning_rate": 3.9354391101634946e-06, "loss": 0.226, "mean_token_accuracy": 0.9416385293006897, "num_tokens": 67609120.0, "step": 1407 }, { "entropy": 0.2777843400835991, "epoch": 2.169491525423729, "grad_norm": 0.6565898060798645, "learning_rate": 3.921938223795323e-06, "loss": 0.2562, "mean_token_accuracy": 0.9339181780815125, "num_tokens": 67659751.0, "step": 1408 }, { "entropy": 0.747272402048111, "epoch": 2.1710323574730355, "grad_norm": 0.8041735887527466, "learning_rate": 3.908454883944431e-06, "loss": 0.7028, "mean_token_accuracy": 0.8018701076507568, "num_tokens": 67674041.0, "step": 1409 }, { "entropy": 0.6972972452640533, "epoch": 2.172573189522342, "grad_norm": 1.1609035730361938, "learning_rate": 3.894989129535339e-06, "loss": 0.6715, "mean_token_accuracy": 0.822966068983078, "num_tokens": 67683555.0, "step": 1410 }, { "entropy": 0.382010854780674, "epoch": 2.174114021571649, "grad_norm": 0.5979192852973938, "learning_rate": 3.881540999441811e-06, "loss": 0.1936, "mean_token_accuracy": 0.9008991122245789, "num_tokens": 67719703.0, "step": 1411 }, { "entropy": 0.20545747876167297, "epoch": 2.1756548536209555, "grad_norm": 0.48670148849487305, "learning_rate": 3.868110532486723e-06, "loss": 0.1831, "mean_token_accuracy": 0.9502158761024475, "num_tokens": 67783950.0, "step": 1412 }, { "entropy": 0.27266305685043335, "epoch": 2.177195685670262, "grad_norm": 0.5930047631263733, "learning_rate": 3.854697767441955e-06, "loss": 0.2602, "mean_token_accuracy": 0.9289792776107788, "num_tokens": 67845455.0, "step": 1413 }, { "entropy": 0.2815416008234024, "epoch": 2.1787365177195683, "grad_norm": 0.6394755244255066, "learning_rate": 3.841302743028291e-06, "loss": 0.2625, "mean_token_accuracy": 0.9302110970020294, "num_tokens": 67905218.0, "step": 1414 }, { "entropy": 0.27659402787685394, "epoch": 2.180277349768875, "grad_norm": 0.6452198624610901, "learning_rate": 3.827925497915305e-06, "loss": 0.2594, "mean_token_accuracy": 0.930750846862793, "num_tokens": 67963641.0, "step": 1415 }, { "entropy": 0.27051785588264465, "epoch": 2.1818181818181817, "grad_norm": 0.6735694408416748, "learning_rate": 3.8145660707212395e-06, "loss": 0.2542, "mean_token_accuracy": 0.932025820016861, "num_tokens": 68020850.0, "step": 1416 }, { "entropy": 0.28830039501190186, "epoch": 2.1833590138674883, "grad_norm": 0.617725133895874, "learning_rate": 3.8012245000128987e-06, "loss": 0.2695, "mean_token_accuracy": 0.9311622083187103, "num_tokens": 68076948.0, "step": 1417 }, { "entropy": 0.23558755218982697, "epoch": 2.184899845916795, "grad_norm": 0.6158937215805054, "learning_rate": 3.7879008243055292e-06, "loss": 0.2188, "mean_token_accuracy": 0.9412027895450592, "num_tokens": 68131849.0, "step": 1418 }, { "entropy": 0.2628171965479851, "epoch": 2.1864406779661016, "grad_norm": 0.6315849423408508, "learning_rate": 3.7745950820627355e-06, "loss": 0.243, "mean_token_accuracy": 0.9356712698936462, "num_tokens": 68185109.0, "step": 1419 }, { "entropy": 0.2477075606584549, "epoch": 2.1879815100154083, "grad_norm": 0.6181500554084778, "learning_rate": 3.761307311696337e-06, "loss": 0.2259, "mean_token_accuracy": 0.9419459700584412, "num_tokens": 68236008.0, "step": 1420 }, { "entropy": 0.6779249012470245, "epoch": 2.189522342064715, "grad_norm": 0.8723528981208801, "learning_rate": 3.7480375515662783e-06, "loss": 0.6076, "mean_token_accuracy": 0.827238142490387, "num_tokens": 68252779.0, "step": 1421 }, { "entropy": 0.7026145160198212, "epoch": 2.1910631741140216, "grad_norm": 0.9763786196708679, "learning_rate": 3.7347858399805035e-06, "loss": 0.6576, "mean_token_accuracy": 0.8191916942596436, "num_tokens": 68262929.0, "step": 1422 }, { "entropy": 0.5927427113056183, "epoch": 2.1926040061633283, "grad_norm": 1.2117995023727417, "learning_rate": 3.721552215194859e-06, "loss": 0.5701, "mean_token_accuracy": 0.85160231590271, "num_tokens": 68270646.0, "step": 1423 }, { "entropy": 0.1597815603017807, "epoch": 2.194144838212635, "grad_norm": 0.44914698600769043, "learning_rate": 3.7083367154129756e-06, "loss": 0.137, "mean_token_accuracy": 0.963602751493454, "num_tokens": 68336182.0, "step": 1424 }, { "entropy": 0.2425483614206314, "epoch": 2.1956856702619416, "grad_norm": 0.5718766450881958, "learning_rate": 3.6951393787861587e-06, "loss": 0.2206, "mean_token_accuracy": 0.9405915439128876, "num_tokens": 68399758.0, "step": 1425 }, { "entropy": 0.29848912358283997, "epoch": 2.1972265023112483, "grad_norm": 0.6604275107383728, "learning_rate": 3.6819602434132796e-06, "loss": 0.278, "mean_token_accuracy": 0.926274299621582, "num_tokens": 68460860.0, "step": 1426 }, { "entropy": 0.275689959526062, "epoch": 2.1987673343605545, "grad_norm": 0.6368737816810608, "learning_rate": 3.6687993473406667e-06, "loss": 0.2563, "mean_token_accuracy": 0.9310835897922516, "num_tokens": 68520287.0, "step": 1427 }, { "entropy": 0.28144803643226624, "epoch": 2.200308166409861, "grad_norm": 0.6633134484291077, "learning_rate": 3.6556567285619917e-06, "loss": 0.2631, "mean_token_accuracy": 0.9295243322849274, "num_tokens": 68578340.0, "step": 1428 }, { "entropy": 0.27228376269340515, "epoch": 2.201848998459168, "grad_norm": 0.6567227840423584, "learning_rate": 3.642532425018164e-06, "loss": 0.2566, "mean_token_accuracy": 0.9334557354450226, "num_tokens": 68635213.0, "step": 1429 }, { "entropy": 0.2537154257297516, "epoch": 2.2033898305084745, "grad_norm": 0.617355227470398, "learning_rate": 3.6294264745972176e-06, "loss": 0.2296, "mean_token_accuracy": 0.9395292401313782, "num_tokens": 68690913.0, "step": 1430 }, { "entropy": 0.26835574209690094, "epoch": 2.204930662557781, "grad_norm": 0.6711882948875427, "learning_rate": 3.6163389151342033e-06, "loss": 0.2472, "mean_token_accuracy": 0.9352254867553711, "num_tokens": 68745333.0, "step": 1431 }, { "entropy": 0.23033834248781204, "epoch": 2.206471494607088, "grad_norm": 0.6628974080085754, "learning_rate": 3.6032697844110896e-06, "loss": 0.2056, "mean_token_accuracy": 0.946390688419342, "num_tokens": 68798101.0, "step": 1432 }, { "entropy": 0.2205931842327118, "epoch": 2.2080123266563945, "grad_norm": 0.6665132641792297, "learning_rate": 3.5902191201566273e-06, "loss": 0.2015, "mean_token_accuracy": 0.9476175010204315, "num_tokens": 68847411.0, "step": 1433 }, { "entropy": 0.7446451783180237, "epoch": 2.209553158705701, "grad_norm": 0.8768050670623779, "learning_rate": 3.577186960046264e-06, "loss": 0.7525, "mean_token_accuracy": 0.78983074426651, "num_tokens": 68860113.0, "step": 1434 }, { "entropy": 0.7080598473548889, "epoch": 2.211093990755008, "grad_norm": 1.0656903982162476, "learning_rate": 3.5641733417020373e-06, "loss": 0.6516, "mean_token_accuracy": 0.8265104293823242, "num_tokens": 68869555.0, "step": 1435 }, { "entropy": 0.3680233955383301, "epoch": 2.2126348228043144, "grad_norm": 0.5372891426086426, "learning_rate": 3.551178302692445e-06, "loss": 0.1764, "mean_token_accuracy": 0.9134046733379364, "num_tokens": 68905843.0, "step": 1436 }, { "entropy": 0.23981043696403503, "epoch": 2.214175654853621, "grad_norm": 0.5442007184028625, "learning_rate": 3.5382018805323593e-06, "loss": 0.2255, "mean_token_accuracy": 0.9388779699802399, "num_tokens": 68969759.0, "step": 1437 }, { "entropy": 0.27815620601177216, "epoch": 2.2157164869029278, "grad_norm": 0.7055981159210205, "learning_rate": 3.5252441126828895e-06, "loss": 0.2704, "mean_token_accuracy": 0.9280113279819489, "num_tokens": 69031007.0, "step": 1438 }, { "entropy": 0.2680676430463791, "epoch": 2.2172573189522344, "grad_norm": 0.6343545317649841, "learning_rate": 3.512305036551319e-06, "loss": 0.2535, "mean_token_accuracy": 0.9319189786911011, "num_tokens": 69090685.0, "step": 1439 }, { "entropy": 0.27185797691345215, "epoch": 2.2187981510015407, "grad_norm": 0.6375028491020203, "learning_rate": 3.499384689490951e-06, "loss": 0.2512, "mean_token_accuracy": 0.9329533874988556, "num_tokens": 69149265.0, "step": 1440 }, { "entropy": 0.2786538451910019, "epoch": 2.2203389830508473, "grad_norm": 0.6248424053192139, "learning_rate": 3.4864831088010277e-06, "loss": 0.2606, "mean_token_accuracy": 0.9309795200824738, "num_tokens": 69206620.0, "step": 1441 }, { "entropy": 0.2525939717888832, "epoch": 2.221879815100154, "grad_norm": 0.5856837034225464, "learning_rate": 3.4736003317266155e-06, "loss": 0.2344, "mean_token_accuracy": 0.9375196695327759, "num_tokens": 69262695.0, "step": 1442 }, { "entropy": 0.23156806826591492, "epoch": 2.2234206471494606, "grad_norm": 0.6234540939331055, "learning_rate": 3.460736395458495e-06, "loss": 0.217, "mean_token_accuracy": 0.9429470598697662, "num_tokens": 69317534.0, "step": 1443 }, { "entropy": 0.2595859318971634, "epoch": 2.2249614791987673, "grad_norm": 0.6207613945007324, "learning_rate": 3.447891337133059e-06, "loss": 0.2382, "mean_token_accuracy": 0.9371338188648224, "num_tokens": 69371125.0, "step": 1444 }, { "entropy": 0.24260136485099792, "epoch": 2.226502311248074, "grad_norm": 0.667349636554718, "learning_rate": 3.4350651938322e-06, "loss": 0.2251, "mean_token_accuracy": 0.9406746029853821, "num_tokens": 69422762.0, "step": 1445 }, { "entropy": 0.6157099604606628, "epoch": 2.2280431432973806, "grad_norm": 0.8225474953651428, "learning_rate": 3.422258002583205e-06, "loss": 0.4613, "mean_token_accuracy": 0.84516242146492, "num_tokens": 69445220.0, "step": 1446 }, { "entropy": 0.7145560085773468, "epoch": 2.2295839753466873, "grad_norm": 1.0082060098648071, "learning_rate": 3.409469800358649e-06, "loss": 0.6903, "mean_token_accuracy": 0.8081543445587158, "num_tokens": 69455839.0, "step": 1447 }, { "entropy": 0.5811224281787872, "epoch": 2.231124807395994, "grad_norm": 1.1465386152267456, "learning_rate": 3.396700624076298e-06, "loss": 0.5453, "mean_token_accuracy": 0.8567440509796143, "num_tokens": 69463749.0, "step": 1448 }, { "entropy": 0.17817386239767075, "epoch": 2.2326656394453006, "grad_norm": 0.4506602883338928, "learning_rate": 3.3839505105989747e-06, "loss": 0.1617, "mean_token_accuracy": 0.955828070640564, "num_tokens": 69529158.0, "step": 1449 }, { "entropy": 0.22832895070314407, "epoch": 2.2342064714946073, "grad_norm": 0.5680158734321594, "learning_rate": 3.3712194967344835e-06, "loss": 0.2128, "mean_token_accuracy": 0.9404067695140839, "num_tokens": 69591648.0, "step": 1450 }, { "entropy": 0.2722720801830292, "epoch": 2.2357473035439135, "grad_norm": 0.5898522734642029, "learning_rate": 3.358507619235485e-06, "loss": 0.2524, "mean_token_accuracy": 0.9322571456432343, "num_tokens": 69652135.0, "step": 1451 }, { "entropy": 0.28071150183677673, "epoch": 2.23728813559322, "grad_norm": 0.6295030117034912, "learning_rate": 3.3458149147994025e-06, "loss": 0.267, "mean_token_accuracy": 0.9302556812763214, "num_tokens": 69711148.0, "step": 1452 }, { "entropy": 0.2679489105939865, "epoch": 2.238828967642527, "grad_norm": 0.618219792842865, "learning_rate": 3.333141420068304e-06, "loss": 0.2444, "mean_token_accuracy": 0.9339344501495361, "num_tokens": 69768669.0, "step": 1453 }, { "entropy": 0.25124359875917435, "epoch": 2.2403697996918335, "grad_norm": 0.5773463249206543, "learning_rate": 3.3204871716288e-06, "loss": 0.2351, "mean_token_accuracy": 0.9382041990756989, "num_tokens": 69824936.0, "step": 1454 }, { "entropy": 0.2642394155263901, "epoch": 2.24191063174114, "grad_norm": 0.7066903114318848, "learning_rate": 3.307852206011939e-06, "loss": 0.2444, "mean_token_accuracy": 0.9355450868606567, "num_tokens": 69880015.0, "step": 1455 }, { "entropy": 0.24926749616861343, "epoch": 2.243451463790447, "grad_norm": 0.6083498001098633, "learning_rate": 3.295236559693112e-06, "loss": 0.2307, "mean_token_accuracy": 0.9405523240566254, "num_tokens": 69933773.0, "step": 1456 }, { "entropy": 0.2238684520125389, "epoch": 2.2449922958397535, "grad_norm": 0.630627453327179, "learning_rate": 3.2826402690919314e-06, "loss": 0.2089, "mean_token_accuracy": 0.9445151388645172, "num_tokens": 69985469.0, "step": 1457 }, { "entropy": 0.4656059592962265, "epoch": 2.24653312788906, "grad_norm": 0.7466793060302734, "learning_rate": 3.27006337057213e-06, "loss": 0.3476, "mean_token_accuracy": 0.8754791617393494, "num_tokens": 70018677.0, "step": 1458 }, { "entropy": 0.7115910649299622, "epoch": 2.248073959938367, "grad_norm": 0.984247624874115, "learning_rate": 3.257505900441462e-06, "loss": 0.7218, "mean_token_accuracy": 0.7991906404495239, "num_tokens": 70030111.0, "step": 1459 }, { "entropy": 0.6457723379135132, "epoch": 2.2496147919876734, "grad_norm": 1.2326693534851074, "learning_rate": 3.2449678949515937e-06, "loss": 0.6256, "mean_token_accuracy": 0.8373067677021027, "num_tokens": 70039229.0, "step": 1460 }, { "entropy": 0.3808843791484833, "epoch": 2.25115562403698, "grad_norm": 0.6148693561553955, "learning_rate": 3.232449390298e-06, "loss": 0.2159, "mean_token_accuracy": 0.8961623013019562, "num_tokens": 70075461.0, "step": 1461 }, { "entropy": 0.24383320659399033, "epoch": 2.2526964560862868, "grad_norm": 0.5533859133720398, "learning_rate": 3.2199504226198587e-06, "loss": 0.2238, "mean_token_accuracy": 0.9391647577285767, "num_tokens": 70139674.0, "step": 1462 }, { "entropy": 0.262772873044014, "epoch": 2.2542372881355934, "grad_norm": 0.6193530559539795, "learning_rate": 3.2074710279999468e-06, "loss": 0.249, "mean_token_accuracy": 0.932528167963028, "num_tokens": 70201366.0, "step": 1463 }, { "entropy": 0.27143052220344543, "epoch": 2.2557781201848996, "grad_norm": 0.6293825507164001, "learning_rate": 3.1950112424645385e-06, "loss": 0.2506, "mean_token_accuracy": 0.9330002665519714, "num_tokens": 70261409.0, "step": 1464 }, { "entropy": 0.2818099111318588, "epoch": 2.2573189522342063, "grad_norm": 0.6376467347145081, "learning_rate": 3.182571101983297e-06, "loss": 0.265, "mean_token_accuracy": 0.9297927618026733, "num_tokens": 70320032.0, "step": 1465 }, { "entropy": 0.26861701905727386, "epoch": 2.258859784283513, "grad_norm": 0.5939739346504211, "learning_rate": 3.1701506424691763e-06, "loss": 0.247, "mean_token_accuracy": 0.9354894161224365, "num_tokens": 70377425.0, "step": 1466 }, { "entropy": 0.24589116126298904, "epoch": 2.2604006163328196, "grad_norm": 0.6053591966629028, "learning_rate": 3.1577498997783074e-06, "loss": 0.2301, "mean_token_accuracy": 0.9404363632202148, "num_tokens": 70433723.0, "step": 1467 }, { "entropy": 0.28789544105529785, "epoch": 2.2619414483821263, "grad_norm": 0.6438174843788147, "learning_rate": 3.145368909709915e-06, "loss": 0.2635, "mean_token_accuracy": 0.931659072637558, "num_tokens": 70488776.0, "step": 1468 }, { "entropy": 0.27902862429618835, "epoch": 2.263482280431433, "grad_norm": 0.6373434066772461, "learning_rate": 3.1330077080061837e-06, "loss": 0.2579, "mean_token_accuracy": 0.9329128265380859, "num_tokens": 70542528.0, "step": 1469 }, { "entropy": 0.2589965686202049, "epoch": 2.2650231124807396, "grad_norm": 0.6570436358451843, "learning_rate": 3.1206663303521833e-06, "loss": 0.2428, "mean_token_accuracy": 0.9367062449455261, "num_tokens": 70594780.0, "step": 1470 }, { "entropy": 0.3786054775118828, "epoch": 2.2665639445300463, "grad_norm": 0.7282187938690186, "learning_rate": 3.1083448123757475e-06, "loss": 0.2774, "mean_token_accuracy": 0.9178656339645386, "num_tokens": 70633887.0, "step": 1471 }, { "entropy": 0.7325589060783386, "epoch": 2.268104776579353, "grad_norm": 0.9133981466293335, "learning_rate": 3.0960431896473877e-06, "loss": 0.7196, "mean_token_accuracy": 0.796262800693512, "num_tokens": 70645484.0, "step": 1472 }, { "entropy": 0.6551516652107239, "epoch": 2.2696456086286596, "grad_norm": 1.1065183877944946, "learning_rate": 3.0837614976801742e-06, "loss": 0.6247, "mean_token_accuracy": 0.8348782956600189, "num_tokens": 70653835.0, "step": 1473 }, { "entropy": 0.19297419488430023, "epoch": 2.2711864406779663, "grad_norm": 0.4424774944782257, "learning_rate": 3.0714997719296346e-06, "loss": 0.1719, "mean_token_accuracy": 0.9538041353225708, "num_tokens": 70719284.0, "step": 1474 }, { "entropy": 0.2774417996406555, "epoch": 2.2727272727272725, "grad_norm": 0.5882661938667297, "learning_rate": 3.0592580477936606e-06, "loss": 0.2598, "mean_token_accuracy": 0.9315089881420135, "num_tokens": 70782078.0, "step": 1475 }, { "entropy": 0.2586042284965515, "epoch": 2.2742681047765796, "grad_norm": 0.6098074913024902, "learning_rate": 3.0470363606124133e-06, "loss": 0.2372, "mean_token_accuracy": 0.9366777241230011, "num_tokens": 70842435.0, "step": 1476 }, { "entropy": 0.24963774532079697, "epoch": 2.275808936825886, "grad_norm": 0.6047325730323792, "learning_rate": 3.0348347456681937e-06, "loss": 0.2318, "mean_token_accuracy": 0.9381052255630493, "num_tokens": 70901218.0, "step": 1477 }, { "entropy": 0.25247659534215927, "epoch": 2.2773497688751925, "grad_norm": 0.717073917388916, "learning_rate": 3.0226532381853646e-06, "loss": 0.2311, "mean_token_accuracy": 0.9390464723110199, "num_tokens": 70958971.0, "step": 1478 }, { "entropy": 0.27279072999954224, "epoch": 2.278890600924499, "grad_norm": 0.6475360989570618, "learning_rate": 3.0104918733302415e-06, "loss": 0.2557, "mean_token_accuracy": 0.9326714277267456, "num_tokens": 71015654.0, "step": 1479 }, { "entropy": 0.27284960448741913, "epoch": 2.280431432973806, "grad_norm": 0.6354268789291382, "learning_rate": 2.998350686210989e-06, "loss": 0.2534, "mean_token_accuracy": 0.9349430203437805, "num_tokens": 71071198.0, "step": 1480 }, { "entropy": 0.23902900516986847, "epoch": 2.2819722650231125, "grad_norm": 0.6508108377456665, "learning_rate": 2.986229711877523e-06, "loss": 0.2219, "mean_token_accuracy": 0.9403153359889984, "num_tokens": 71125497.0, "step": 1481 }, { "entropy": 0.23876427114009857, "epoch": 2.283513097072419, "grad_norm": 0.6615651249885559, "learning_rate": 2.9741289853214063e-06, "loss": 0.2194, "mean_token_accuracy": 0.9419302046298981, "num_tokens": 71178171.0, "step": 1482 }, { "entropy": 0.3287089392542839, "epoch": 2.285053929121726, "grad_norm": 0.6898131370544434, "learning_rate": 2.9620485414757504e-06, "loss": 0.2667, "mean_token_accuracy": 0.9268409907817841, "num_tokens": 71220702.0, "step": 1483 }, { "entropy": 0.7483394145965576, "epoch": 2.2865947611710324, "grad_norm": 0.918687641620636, "learning_rate": 2.9499884152151126e-06, "loss": 0.7464, "mean_token_accuracy": 0.7941856384277344, "num_tokens": 71232461.0, "step": 1484 }, { "entropy": 0.652903288602829, "epoch": 2.288135593220339, "grad_norm": 1.1444138288497925, "learning_rate": 2.937948641355398e-06, "loss": 0.6375, "mean_token_accuracy": 0.8314222693443298, "num_tokens": 71241525.0, "step": 1485 }, { "entropy": 0.37026333808898926, "epoch": 2.2896764252696458, "grad_norm": 0.5919217467308044, "learning_rate": 2.9259292546537543e-06, "loss": 0.1829, "mean_token_accuracy": 0.8996927440166473, "num_tokens": 71277738.0, "step": 1486 }, { "entropy": 0.27045267820358276, "epoch": 2.2912172573189524, "grad_norm": 0.6202074885368347, "learning_rate": 2.913930289808474e-06, "loss": 0.2525, "mean_token_accuracy": 0.9313071966171265, "num_tokens": 71341606.0, "step": 1487 }, { "entropy": 0.2806621640920639, "epoch": 2.2927580893682586, "grad_norm": 0.5852921605110168, "learning_rate": 2.9019517814589015e-06, "loss": 0.2672, "mean_token_accuracy": 0.9282805919647217, "num_tokens": 71403072.0, "step": 1488 }, { "entropy": 0.2509538531303406, "epoch": 2.2942989214175653, "grad_norm": 0.6007704734802246, "learning_rate": 2.8899937641853216e-06, "loss": 0.233, "mean_token_accuracy": 0.9378871619701385, "num_tokens": 71462744.0, "step": 1489 }, { "entropy": 0.2554765045642853, "epoch": 2.295839753466872, "grad_norm": 0.629080593585968, "learning_rate": 2.878056272508858e-06, "loss": 0.2399, "mean_token_accuracy": 0.9377570152282715, "num_tokens": 71520964.0, "step": 1490 }, { "entropy": 0.2701399698853493, "epoch": 2.2973805855161786, "grad_norm": 0.6563435196876526, "learning_rate": 2.866139340891384e-06, "loss": 0.2517, "mean_token_accuracy": 0.9340987205505371, "num_tokens": 71577941.0, "step": 1491 }, { "entropy": 0.2514049932360649, "epoch": 2.2989214175654853, "grad_norm": 0.6138540506362915, "learning_rate": 2.8542430037354274e-06, "loss": 0.2342, "mean_token_accuracy": 0.937827080488205, "num_tokens": 71633958.0, "step": 1492 }, { "entropy": 0.2354898899793625, "epoch": 2.300462249614792, "grad_norm": 0.6244626641273499, "learning_rate": 2.842367295384053e-06, "loss": 0.2139, "mean_token_accuracy": 0.9432496726512909, "num_tokens": 71688821.0, "step": 1493 }, { "entropy": 0.24723467975854874, "epoch": 2.3020030816640986, "grad_norm": 0.6349254250526428, "learning_rate": 2.8305122501207784e-06, "loss": 0.224, "mean_token_accuracy": 0.941044956445694, "num_tokens": 71742422.0, "step": 1494 }, { "entropy": 0.22920794039964676, "epoch": 2.3035439137134053, "grad_norm": 0.6570906639099121, "learning_rate": 2.8186779021694577e-06, "loss": 0.2117, "mean_token_accuracy": 0.945864737033844, "num_tokens": 71794010.0, "step": 1495 }, { "entropy": 0.5653776824474335, "epoch": 2.305084745762712, "grad_norm": 0.6935083270072937, "learning_rate": 2.806864285694213e-06, "loss": 0.3839, "mean_token_accuracy": 0.8572343289852142, "num_tokens": 71820643.0, "step": 1496 }, { "entropy": 0.7196711897850037, "epoch": 2.3066255778120186, "grad_norm": 1.0404874086380005, "learning_rate": 2.795071434799307e-06, "loss": 0.7054, "mean_token_accuracy": 0.8087341785430908, "num_tokens": 71831479.0, "step": 1497 }, { "entropy": 0.6494724452495575, "epoch": 2.3081664098613253, "grad_norm": 1.162358045578003, "learning_rate": 2.7832993835290556e-06, "loss": 0.6211, "mean_token_accuracy": 0.8353652358055115, "num_tokens": 71839994.0, "step": 1498 }, { "entropy": 0.23404507339000702, "epoch": 2.309707241910632, "grad_norm": 0.5278438329696655, "learning_rate": 2.77154816586773e-06, "loss": 0.2145, "mean_token_accuracy": 0.9423441588878632, "num_tokens": 71905129.0, "step": 1499 }, { "entropy": 0.295854389667511, "epoch": 2.3112480739599386, "grad_norm": 0.6462519764900208, "learning_rate": 2.7598178157394582e-06, "loss": 0.2781, "mean_token_accuracy": 0.9250317215919495, "num_tokens": 71967316.0, "step": 1500 }, { "entropy": 0.2788437008857727, "epoch": 2.312788906009245, "grad_norm": 0.6046594977378845, "learning_rate": 2.7481083670081277e-06, "loss": 0.2585, "mean_token_accuracy": 0.9316633343696594, "num_tokens": 72027387.0, "step": 1501 }, { "entropy": 0.25335264950990677, "epoch": 2.3143297380585515, "grad_norm": 0.6255109906196594, "learning_rate": 2.7364198534772824e-06, "loss": 0.2388, "mean_token_accuracy": 0.9364058673381805, "num_tokens": 72086125.0, "step": 1502 }, { "entropy": 0.2589180916547775, "epoch": 2.315870570107858, "grad_norm": 0.6145830154418945, "learning_rate": 2.7247523088900342e-06, "loss": 0.2358, "mean_token_accuracy": 0.9382800757884979, "num_tokens": 72143787.0, "step": 1503 }, { "entropy": 0.25899001210927963, "epoch": 2.317411402157165, "grad_norm": 0.5723229050636292, "learning_rate": 2.7131057669289574e-06, "loss": 0.2352, "mean_token_accuracy": 0.9398485720157623, "num_tokens": 72200352.0, "step": 1504 }, { "entropy": 0.262570858001709, "epoch": 2.3189522342064715, "grad_norm": 0.6248596906661987, "learning_rate": 2.701480261215994e-06, "loss": 0.2417, "mean_token_accuracy": 0.9355814158916473, "num_tokens": 72255735.0, "step": 1505 }, { "entropy": 0.2573477327823639, "epoch": 2.320493066255778, "grad_norm": 0.606480062007904, "learning_rate": 2.6898758253123613e-06, "loss": 0.2417, "mean_token_accuracy": 0.9373492300510406, "num_tokens": 72309701.0, "step": 1506 }, { "entropy": 0.23493606597185135, "epoch": 2.3220338983050848, "grad_norm": 0.603860080242157, "learning_rate": 2.6782924927184484e-06, "loss": 0.2067, "mean_token_accuracy": 0.9453189671039581, "num_tokens": 72361851.0, "step": 1507 }, { "entropy": 0.35989321768283844, "epoch": 2.3235747303543914, "grad_norm": 0.7018561959266663, "learning_rate": 2.666730296873716e-06, "loss": 0.2565, "mean_token_accuracy": 0.9262537062168121, "num_tokens": 72401412.0, "step": 1508 }, { "entropy": 0.7204102277755737, "epoch": 2.325115562403698, "grad_norm": 0.9544089436531067, "learning_rate": 2.655189271156625e-06, "loss": 0.7279, "mean_token_accuracy": 0.7986856400966644, "num_tokens": 72413625.0, "step": 1509 }, { "entropy": 0.6605339050292969, "epoch": 2.3266563944530048, "grad_norm": 1.1096111536026, "learning_rate": 2.6436694488844984e-06, "loss": 0.6254, "mean_token_accuracy": 0.8324555456638336, "num_tokens": 72423056.0, "step": 1510 }, { "entropy": 0.342689573764801, "epoch": 2.3281972265023114, "grad_norm": 0.7207785248756409, "learning_rate": 2.632170863313459e-06, "loss": 0.1746, "mean_token_accuracy": 0.9167031347751617, "num_tokens": 72459369.0, "step": 1511 }, { "entropy": 0.24534016102552414, "epoch": 2.3297380585516176, "grad_norm": 0.5555769801139832, "learning_rate": 2.620693547638318e-06, "loss": 0.2344, "mean_token_accuracy": 0.9356741905212402, "num_tokens": 72523242.0, "step": 1512 }, { "entropy": 0.2615051865577698, "epoch": 2.3312788906009247, "grad_norm": 0.5824722051620483, "learning_rate": 2.609237534992494e-06, "loss": 0.2452, "mean_token_accuracy": 0.9346528351306915, "num_tokens": 72584389.0, "step": 1513 }, { "entropy": 0.2765926122665405, "epoch": 2.332819722650231, "grad_norm": 0.6457837820053101, "learning_rate": 2.5978028584478974e-06, "loss": 0.2594, "mean_token_accuracy": 0.9288662672042847, "num_tokens": 72643699.0, "step": 1514 }, { "entropy": 0.2497108280658722, "epoch": 2.3343605546995376, "grad_norm": 0.8988678455352783, "learning_rate": 2.5863895510148373e-06, "loss": 0.2311, "mean_token_accuracy": 0.9391140639781952, "num_tokens": 72701754.0, "step": 1515 }, { "entropy": 0.25990601629018784, "epoch": 2.3359013867488443, "grad_norm": 0.6833863854408264, "learning_rate": 2.574997645641951e-06, "loss": 0.2384, "mean_token_accuracy": 0.9364548921585083, "num_tokens": 72758633.0, "step": 1516 }, { "entropy": 0.27156101167201996, "epoch": 2.337442218798151, "grad_norm": 0.613694429397583, "learning_rate": 2.563627175216078e-06, "loss": 0.2478, "mean_token_accuracy": 0.9350405037403107, "num_tokens": 72814480.0, "step": 1517 }, { "entropy": 0.2519235610961914, "epoch": 2.3389830508474576, "grad_norm": 0.6676573753356934, "learning_rate": 2.5522781725621814e-06, "loss": 0.2379, "mean_token_accuracy": 0.9364201128482819, "num_tokens": 72869413.0, "step": 1518 }, { "entropy": 0.2854577898979187, "epoch": 2.3405238828967643, "grad_norm": 0.6554715037345886, "learning_rate": 2.540950670443252e-06, "loss": 0.2694, "mean_token_accuracy": 0.9301974177360535, "num_tokens": 72923255.0, "step": 1519 }, { "entropy": 0.24593069404363632, "epoch": 2.342064714946071, "grad_norm": 0.6606329083442688, "learning_rate": 2.5296447015602087e-06, "loss": 0.231, "mean_token_accuracy": 0.9395478069782257, "num_tokens": 72975047.0, "step": 1520 }, { "entropy": 0.5490841865539551, "epoch": 2.3436055469953776, "grad_norm": 0.7288292050361633, "learning_rate": 2.5183602985518073e-06, "loss": 0.3831, "mean_token_accuracy": 0.860439121723175, "num_tokens": 72999076.0, "step": 1521 }, { "entropy": 0.7099651992321014, "epoch": 2.3451463790446843, "grad_norm": 1.117872953414917, "learning_rate": 2.507097493994548e-06, "loss": 0.7042, "mean_token_accuracy": 0.8150413632392883, "num_tokens": 73009518.0, "step": 1522 }, { "entropy": 0.6294797956943512, "epoch": 2.346687211093991, "grad_norm": 1.2296448945999146, "learning_rate": 2.4958563204025766e-06, "loss": 0.6327, "mean_token_accuracy": 0.8341982960700989, "num_tokens": 73017225.0, "step": 1523 }, { "entropy": 0.175405815243721, "epoch": 2.3482280431432976, "grad_norm": 0.42226842045783997, "learning_rate": 2.484636810227593e-06, "loss": 0.1573, "mean_token_accuracy": 0.957149088382721, "num_tokens": 73082531.0, "step": 1524 }, { "entropy": 0.2637258321046829, "epoch": 2.349768875192604, "grad_norm": 0.5758313536643982, "learning_rate": 2.473438995858767e-06, "loss": 0.2398, "mean_token_accuracy": 0.9358624815940857, "num_tokens": 73145066.0, "step": 1525 }, { "entropy": 0.24301956593990326, "epoch": 2.3513097072419105, "grad_norm": 0.5524592995643616, "learning_rate": 2.46226290962262e-06, "loss": 0.2217, "mean_token_accuracy": 0.9402364492416382, "num_tokens": 73205635.0, "step": 1526 }, { "entropy": 0.2705647796392441, "epoch": 2.352850539291217, "grad_norm": 0.6035783886909485, "learning_rate": 2.4511085837829596e-06, "loss": 0.2471, "mean_token_accuracy": 0.9343385696411133, "num_tokens": 73264700.0, "step": 1527 }, { "entropy": 0.28328903019428253, "epoch": 2.354391371340524, "grad_norm": 0.6158072352409363, "learning_rate": 2.4399760505407646e-06, "loss": 0.2615, "mean_token_accuracy": 0.931071549654007, "num_tokens": 73322536.0, "step": 1528 }, { "entropy": 0.24955600500106812, "epoch": 2.3559322033898304, "grad_norm": 0.6854059100151062, "learning_rate": 2.4288653420341134e-06, "loss": 0.2334, "mean_token_accuracy": 0.9384049773216248, "num_tokens": 73379345.0, "step": 1529 }, { "entropy": 0.2646782249212265, "epoch": 2.357473035439137, "grad_norm": 0.638177752494812, "learning_rate": 2.417776490338074e-06, "loss": 0.2413, "mean_token_accuracy": 0.9371777176856995, "num_tokens": 73435079.0, "step": 1530 }, { "entropy": 0.2561024874448776, "epoch": 2.3590138674884438, "grad_norm": 0.6500886082649231, "learning_rate": 2.40670952746461e-06, "loss": 0.2376, "mean_token_accuracy": 0.9381046891212463, "num_tokens": 73489705.0, "step": 1531 }, { "entropy": 0.25076724588871, "epoch": 2.3605546995377504, "grad_norm": 0.6180400252342224, "learning_rate": 2.3956644853625e-06, "loss": 0.2323, "mean_token_accuracy": 0.939531683921814, "num_tokens": 73543047.0, "step": 1532 }, { "entropy": 0.24244731664657593, "epoch": 2.362095531587057, "grad_norm": 0.6364676356315613, "learning_rate": 2.3846413959172477e-06, "loss": 0.2197, "mean_token_accuracy": 0.9428375959396362, "num_tokens": 73594139.0, "step": 1533 }, { "entropy": 0.6497230231761932, "epoch": 2.3636363636363638, "grad_norm": 0.8085006475448608, "learning_rate": 2.373640290950974e-06, "loss": 0.5235, "mean_token_accuracy": 0.8416623771190643, "num_tokens": 73611988.0, "step": 1534 }, { "entropy": 0.6385592222213745, "epoch": 2.3651771956856704, "grad_norm": 1.0738314390182495, "learning_rate": 2.362661202222336e-06, "loss": 0.6144, "mean_token_accuracy": 0.8364743292331696, "num_tokens": 73621711.0, "step": 1535 }, { "entropy": 0.3804338425397873, "epoch": 2.366718027734977, "grad_norm": 0.5635444521903992, "learning_rate": 2.351704161426427e-06, "loss": 0.2011, "mean_token_accuracy": 0.9063582420349121, "num_tokens": 73658092.0, "step": 1536 }, { "entropy": 0.2421955093741417, "epoch": 2.3682588597842837, "grad_norm": 0.5953739285469055, "learning_rate": 2.3407692001947034e-06, "loss": 0.2271, "mean_token_accuracy": 0.9383822977542877, "num_tokens": 73722117.0, "step": 1537 }, { "entropy": 0.2541443258523941, "epoch": 2.36979969183359, "grad_norm": 0.5702321529388428, "learning_rate": 2.329856350094872e-06, "loss": 0.2325, "mean_token_accuracy": 0.9368578195571899, "num_tokens": 73783499.0, "step": 1538 }, { "entropy": 0.23893849551677704, "epoch": 2.3713405238828966, "grad_norm": 0.5579627752304077, "learning_rate": 2.3189656426308115e-06, "loss": 0.223, "mean_token_accuracy": 0.9400564432144165, "num_tokens": 73843278.0, "step": 1539 }, { "entropy": 0.2538415193557739, "epoch": 2.3728813559322033, "grad_norm": 0.6138657331466675, "learning_rate": 2.3080971092424754e-06, "loss": 0.2372, "mean_token_accuracy": 0.9373043775558472, "num_tokens": 73901767.0, "step": 1540 }, { "entropy": 0.24629691243171692, "epoch": 2.37442218798151, "grad_norm": 0.6165222525596619, "learning_rate": 2.2972507813058044e-06, "loss": 0.2303, "mean_token_accuracy": 0.9391351044178009, "num_tokens": 73959155.0, "step": 1541 }, { "entropy": 0.25437335669994354, "epoch": 2.3759630200308166, "grad_norm": 0.6082466244697571, "learning_rate": 2.286426690132636e-06, "loss": 0.2367, "mean_token_accuracy": 0.9389703273773193, "num_tokens": 74015525.0, "step": 1542 }, { "entropy": 0.2611647844314575, "epoch": 2.3775038520801233, "grad_norm": 0.6499186158180237, "learning_rate": 2.275624866970614e-06, "loss": 0.2486, "mean_token_accuracy": 0.935027152299881, "num_tokens": 74070747.0, "step": 1543 }, { "entropy": 0.2427680641412735, "epoch": 2.37904468412943, "grad_norm": 0.6346345543861389, "learning_rate": 2.2648453430030927e-06, "loss": 0.2299, "mean_token_accuracy": 0.9394168853759766, "num_tokens": 74124813.0, "step": 1544 }, { "entropy": 0.23947027325630188, "epoch": 2.3805855161787366, "grad_norm": 0.6029489636421204, "learning_rate": 2.2540881493490642e-06, "loss": 0.2173, "mean_token_accuracy": 0.9435958564281464, "num_tokens": 74177025.0, "step": 1545 }, { "entropy": 0.4965398907661438, "epoch": 2.3821263482280433, "grad_norm": 0.7748445272445679, "learning_rate": 2.2433533170630383e-06, "loss": 0.3248, "mean_token_accuracy": 0.8632934987545013, "num_tokens": 74208089.0, "step": 1546 }, { "entropy": 0.6901237964630127, "epoch": 2.38366718027735, "grad_norm": 1.1187803745269775, "learning_rate": 2.2326408771349827e-06, "loss": 0.6656, "mean_token_accuracy": 0.8183472454547882, "num_tokens": 74218626.0, "step": 1547 }, { "entropy": 0.6560049951076508, "epoch": 2.3852080123266566, "grad_norm": 1.23043692111969, "learning_rate": 2.2219508604902175e-06, "loss": 0.6434, "mean_token_accuracy": 0.8360656797885895, "num_tokens": 74226609.0, "step": 1548 }, { "entropy": 0.1751701682806015, "epoch": 2.386748844375963, "grad_norm": 0.4322517216205597, "learning_rate": 2.211283297989334e-06, "loss": 0.1581, "mean_token_accuracy": 0.9560669660568237, "num_tokens": 74292042.0, "step": 1549 }, { "entropy": 0.2662368640303612, "epoch": 2.3882896764252695, "grad_norm": 0.5881085395812988, "learning_rate": 2.2006382204280997e-06, "loss": 0.2434, "mean_token_accuracy": 0.9347570240497589, "num_tokens": 74354840.0, "step": 1550 }, { "entropy": 0.263676181435585, "epoch": 2.389830508474576, "grad_norm": 0.5921332240104675, "learning_rate": 2.1900156585373654e-06, "loss": 0.243, "mean_token_accuracy": 0.9348765909671783, "num_tokens": 74415262.0, "step": 1551 }, { "entropy": 0.2547618895769119, "epoch": 2.391371340523883, "grad_norm": 0.6185798048973083, "learning_rate": 2.1794156429829847e-06, "loss": 0.2351, "mean_token_accuracy": 0.9368638098239899, "num_tokens": 74474289.0, "step": 1552 }, { "entropy": 0.26434384286403656, "epoch": 2.3929121725731894, "grad_norm": 0.6006340980529785, "learning_rate": 2.168838204365733e-06, "loss": 0.2378, "mean_token_accuracy": 0.9358274042606354, "num_tokens": 74532218.0, "step": 1553 }, { "entropy": 0.24956142157316208, "epoch": 2.394453004622496, "grad_norm": 0.600708544254303, "learning_rate": 2.158283373221196e-06, "loss": 0.2281, "mean_token_accuracy": 0.938880443572998, "num_tokens": 74588812.0, "step": 1554 }, { "entropy": 0.25000983476638794, "epoch": 2.3959938366718028, "grad_norm": 0.6079732775688171, "learning_rate": 2.1477511800197036e-06, "loss": 0.2315, "mean_token_accuracy": 0.9386750161647797, "num_tokens": 74644292.0, "step": 1555 }, { "entropy": 0.23283596336841583, "epoch": 2.3975346687211094, "grad_norm": 0.5873470902442932, "learning_rate": 2.137241655166227e-06, "loss": 0.2114, "mean_token_accuracy": 0.9454014301300049, "num_tokens": 74698562.0, "step": 1556 }, { "entropy": 0.23425346612930298, "epoch": 2.399075500770416, "grad_norm": 0.6262191534042358, "learning_rate": 2.1267548290003003e-06, "loss": 0.2133, "mean_token_accuracy": 0.9436473250389099, "num_tokens": 74751485.0, "step": 1557 }, { "entropy": 0.3123002052307129, "epoch": 2.4006163328197228, "grad_norm": 0.6270663142204285, "learning_rate": 2.11629073179593e-06, "loss": 0.2459, "mean_token_accuracy": 0.9346550703048706, "num_tokens": 74796174.0, "step": 1558 }, { "entropy": 0.7426881492137909, "epoch": 2.4021571648690294, "grad_norm": 0.9359362125396729, "learning_rate": 2.1058493937615065e-06, "loss": 0.7416, "mean_token_accuracy": 0.7952556014060974, "num_tokens": 74807971.0, "step": 1559 }, { "entropy": 0.6721793115139008, "epoch": 2.403697996918336, "grad_norm": 1.0983130931854248, "learning_rate": 2.0954308450397177e-06, "loss": 0.647, "mean_token_accuracy": 0.8295380473136902, "num_tokens": 74817250.0, "step": 1560 }, { "entropy": 0.40329837799072266, "epoch": 2.4052388289676427, "grad_norm": 0.6112974882125854, "learning_rate": 2.08503511570746e-06, "loss": 0.1745, "mean_token_accuracy": 0.8969709873199463, "num_tokens": 74853523.0, "step": 1561 }, { "entropy": 0.26027220487594604, "epoch": 2.406779661016949, "grad_norm": 0.5507169365882874, "learning_rate": 2.0746622357757574e-06, "loss": 0.2455, "mean_token_accuracy": 0.9339165687561035, "num_tokens": 74917550.0, "step": 1562 }, { "entropy": 0.24951893091201782, "epoch": 2.4083204930662556, "grad_norm": 0.5790901184082031, "learning_rate": 2.0643122351896662e-06, "loss": 0.2338, "mean_token_accuracy": 0.9361053109169006, "num_tokens": 74978878.0, "step": 1563 }, { "entropy": 0.24017132818698883, "epoch": 2.4098613251155623, "grad_norm": 0.554193377494812, "learning_rate": 2.0539851438281976e-06, "loss": 0.2221, "mean_token_accuracy": 0.9397900104522705, "num_tokens": 75038424.0, "step": 1564 }, { "entropy": 0.27013589441776276, "epoch": 2.411402157164869, "grad_norm": 0.6306830048561096, "learning_rate": 2.043680991504222e-06, "loss": 0.2598, "mean_token_accuracy": 0.9326454102993011, "num_tokens": 75096900.0, "step": 1565 }, { "entropy": 0.23652830719947815, "epoch": 2.4129429892141756, "grad_norm": 0.5943522453308105, "learning_rate": 2.0333998079643992e-06, "loss": 0.2182, "mean_token_accuracy": 0.9413471817970276, "num_tokens": 75154315.0, "step": 1566 }, { "entropy": 0.2699623703956604, "epoch": 2.4144838212634823, "grad_norm": 0.6176144480705261, "learning_rate": 2.0231416228890654e-06, "loss": 0.254, "mean_token_accuracy": 0.9338555335998535, "num_tokens": 75210577.0, "step": 1567 }, { "entropy": 0.26578523218631744, "epoch": 2.416024653312789, "grad_norm": 0.5946397185325623, "learning_rate": 2.0129064658921737e-06, "loss": 0.2456, "mean_token_accuracy": 0.9357764720916748, "num_tokens": 75265801.0, "step": 1568 }, { "entropy": 0.27489112317562103, "epoch": 2.4175654853620956, "grad_norm": 0.6303104758262634, "learning_rate": 2.002694366521193e-06, "loss": 0.2645, "mean_token_accuracy": 0.9323308765888214, "num_tokens": 75319884.0, "step": 1569 }, { "entropy": 0.2588546723127365, "epoch": 2.4191063174114023, "grad_norm": 0.6532129049301147, "learning_rate": 1.992505354257036e-06, "loss": 0.2406, "mean_token_accuracy": 0.9377377331256866, "num_tokens": 75371889.0, "step": 1570 }, { "entropy": 0.5060567781329155, "epoch": 2.420647149460709, "grad_norm": 0.6853481531143188, "learning_rate": 1.982339458513962e-06, "loss": 0.351, "mean_token_accuracy": 0.8632612526416779, "num_tokens": 75400642.0, "step": 1571 }, { "entropy": 0.7123809158802032, "epoch": 2.4221879815100156, "grad_norm": 1.0239479541778564, "learning_rate": 1.9721967086394864e-06, "loss": 0.6949, "mean_token_accuracy": 0.8083038032054901, "num_tokens": 75411397.0, "step": 1572 }, { "entropy": 0.6637684404850006, "epoch": 2.423728813559322, "grad_norm": 1.2155050039291382, "learning_rate": 1.962077133914324e-06, "loss": 0.679, "mean_token_accuracy": 0.8252193331718445, "num_tokens": 75419208.0, "step": 1573 }, { "entropy": 0.16242142021656036, "epoch": 2.425269645608629, "grad_norm": 0.42186927795410156, "learning_rate": 1.9519807635522726e-06, "loss": 0.1466, "mean_token_accuracy": 0.9614668488502502, "num_tokens": 75484578.0, "step": 1574 }, { "entropy": 0.2761717587709427, "epoch": 2.426810477657935, "grad_norm": 0.6021881103515625, "learning_rate": 1.9419076267001468e-06, "loss": 0.2649, "mean_token_accuracy": 0.928144097328186, "num_tokens": 75547200.0, "step": 1575 }, { "entropy": 0.2508213371038437, "epoch": 2.428351309707242, "grad_norm": 0.5573890209197998, "learning_rate": 1.931857752437689e-06, "loss": 0.2326, "mean_token_accuracy": 0.9374187588691711, "num_tokens": 75607960.0, "step": 1576 }, { "entropy": 0.27591779828071594, "epoch": 2.4298921417565484, "grad_norm": 0.6986901760101318, "learning_rate": 1.921831169777485e-06, "loss": 0.2623, "mean_token_accuracy": 0.9301189184188843, "num_tokens": 75667212.0, "step": 1577 }, { "entropy": 0.24910682439804077, "epoch": 2.431432973805855, "grad_norm": 0.5909239649772644, "learning_rate": 1.9118279076648825e-06, "loss": 0.2327, "mean_token_accuracy": 0.9393655359745026, "num_tokens": 75725213.0, "step": 1578 }, { "entropy": 0.2501323074102402, "epoch": 2.4329738058551618, "grad_norm": 0.5909350514411926, "learning_rate": 1.9018479949779057e-06, "loss": 0.2311, "mean_token_accuracy": 0.9381533265113831, "num_tokens": 75782069.0, "step": 1579 }, { "entropy": 0.26621200144290924, "epoch": 2.4345146379044684, "grad_norm": 0.6371167302131653, "learning_rate": 1.8918914605271687e-06, "loss": 0.2506, "mean_token_accuracy": 0.9349822998046875, "num_tokens": 75837755.0, "step": 1580 }, { "entropy": 0.251947320997715, "epoch": 2.436055469953775, "grad_norm": 0.621184766292572, "learning_rate": 1.8819583330558022e-06, "loss": 0.2303, "mean_token_accuracy": 0.9397964477539062, "num_tokens": 75892124.0, "step": 1581 }, { "entropy": 0.2719549089670181, "epoch": 2.4375963020030817, "grad_norm": 0.6790603399276733, "learning_rate": 1.872048641239359e-06, "loss": 0.2526, "mean_token_accuracy": 0.9353083074092865, "num_tokens": 75944929.0, "step": 1582 }, { "entropy": 0.2641405165195465, "epoch": 2.4391371340523884, "grad_norm": 0.648227334022522, "learning_rate": 1.8621624136857397e-06, "loss": 0.2309, "mean_token_accuracy": 0.937990128993988, "num_tokens": 75992418.0, "step": 1583 }, { "entropy": 0.7206456959247589, "epoch": 2.440677966101695, "grad_norm": 0.9154735803604126, "learning_rate": 1.8522996789351033e-06, "loss": 0.7164, "mean_token_accuracy": 0.7995591759681702, "num_tokens": 76004553.0, "step": 1584 }, { "entropy": 0.6811261177062988, "epoch": 2.4422187981510017, "grad_norm": 1.131035566329956, "learning_rate": 1.84246046545979e-06, "loss": 0.6482, "mean_token_accuracy": 0.8291007578372955, "num_tokens": 76013832.0, "step": 1585 }, { "entropy": 0.3556508719921112, "epoch": 2.443759630200308, "grad_norm": 0.5545229315757751, "learning_rate": 1.8326448016642452e-06, "loss": 0.1772, "mean_token_accuracy": 0.9110731184482574, "num_tokens": 76050137.0, "step": 1586 }, { "entropy": 0.254609078168869, "epoch": 2.4453004622496146, "grad_norm": 0.5179657936096191, "learning_rate": 1.822852715884913e-06, "loss": 0.2332, "mean_token_accuracy": 0.9370172917842865, "num_tokens": 76113874.0, "step": 1587 }, { "entropy": 0.26701243221759796, "epoch": 2.4468412942989213, "grad_norm": 0.5767444968223572, "learning_rate": 1.8130842363901845e-06, "loss": 0.2487, "mean_token_accuracy": 0.9339492917060852, "num_tokens": 76174837.0, "step": 1588 }, { "entropy": 0.2435910925269127, "epoch": 2.448382126348228, "grad_norm": 0.5995512008666992, "learning_rate": 1.8033393913802966e-06, "loss": 0.233, "mean_token_accuracy": 0.9374135434627533, "num_tokens": 76234365.0, "step": 1589 }, { "entropy": 0.2381136417388916, "epoch": 2.4499229583975346, "grad_norm": 0.5683373212814331, "learning_rate": 1.7936182089872623e-06, "loss": 0.2172, "mean_token_accuracy": 0.9419918060302734, "num_tokens": 76292893.0, "step": 1590 }, { "entropy": 0.24961048364639282, "epoch": 2.4514637904468413, "grad_norm": 0.5724102258682251, "learning_rate": 1.7839207172747786e-06, "loss": 0.2274, "mean_token_accuracy": 0.9395546913146973, "num_tokens": 76350192.0, "step": 1591 }, { "entropy": 0.28244762122631073, "epoch": 2.453004622496148, "grad_norm": 0.631631076335907, "learning_rate": 1.7742469442381549e-06, "loss": 0.2648, "mean_token_accuracy": 0.9312800168991089, "num_tokens": 76406533.0, "step": 1592 }, { "entropy": 0.2498076856136322, "epoch": 2.4545454545454546, "grad_norm": 0.5993589162826538, "learning_rate": 1.7645969178042188e-06, "loss": 0.2222, "mean_token_accuracy": 0.9414167404174805, "num_tokens": 76461640.0, "step": 1593 }, { "entropy": 0.2545156478881836, "epoch": 2.4560862865947612, "grad_norm": 0.6535375714302063, "learning_rate": 1.7549706658312592e-06, "loss": 0.2355, "mean_token_accuracy": 0.9390829503536224, "num_tokens": 76515381.0, "step": 1594 }, { "entropy": 0.2493821531534195, "epoch": 2.457627118644068, "grad_norm": 0.6400819420814514, "learning_rate": 1.7453682161089203e-06, "loss": 0.2268, "mean_token_accuracy": 0.9418010413646698, "num_tokens": 76567161.0, "step": 1595 }, { "entropy": 0.5598728358745575, "epoch": 2.4591679506933746, "grad_norm": 0.7247159481048584, "learning_rate": 1.7357895963581373e-06, "loss": 0.3626, "mean_token_accuracy": 0.8574254810810089, "num_tokens": 76593525.0, "step": 1596 }, { "entropy": 0.6760543882846832, "epoch": 2.4607087827426812, "grad_norm": 1.0015288591384888, "learning_rate": 1.7262348342310497e-06, "loss": 0.6497, "mean_token_accuracy": 0.824762761592865, "num_tokens": 76603906.0, "step": 1597 }, { "entropy": 0.6248387098312378, "epoch": 2.462249614791988, "grad_norm": 1.1855807304382324, "learning_rate": 1.7167039573109258e-06, "loss": 0.6073, "mean_token_accuracy": 0.8443897366523743, "num_tokens": 76611812.0, "step": 1598 }, { "entropy": 0.17723935842514038, "epoch": 2.463790446841294, "grad_norm": 0.45227673649787903, "learning_rate": 1.7071969931120768e-06, "loss": 0.1592, "mean_token_accuracy": 0.9579584896564484, "num_tokens": 76677259.0, "step": 1599 }, { "entropy": 0.26041559875011444, "epoch": 2.4653312788906008, "grad_norm": 0.5885992646217346, "learning_rate": 1.697713969079785e-06, "loss": 0.251, "mean_token_accuracy": 0.9320547878742218, "num_tokens": 76740240.0, "step": 1600 }, { "epoch": 2.4653312788906008, "eval_entropy": 0.3808202345697385, "eval_loss": 0.39292529225349426, "eval_mean_token_accuracy": 0.8983020318566628, "eval_num_tokens": 76740240.0, "eval_runtime": 192.0958, "eval_samples_per_second": 4.414, "eval_steps_per_second": 1.104, "step": 1600 }, { "entropy": 0.24790535867214203, "epoch": 2.4668721109399074, "grad_norm": 0.5550276041030884, "learning_rate": 1.6882549125902159e-06, "loss": 0.2288, "mean_token_accuracy": 0.9384972453117371, "num_tokens": 76801278.0, "step": 1601 }, { "entropy": 0.27136991918087006, "epoch": 2.468412942989214, "grad_norm": 0.6173387169837952, "learning_rate": 1.678819850950355e-06, "loss": 0.2541, "mean_token_accuracy": 0.932819664478302, "num_tokens": 76860877.0, "step": 1602 }, { "entropy": 0.2569790482521057, "epoch": 2.4699537750385208, "grad_norm": 0.7231326699256897, "learning_rate": 1.6694088113979034e-06, "loss": 0.234, "mean_token_accuracy": 0.938458651304245, "num_tokens": 76919302.0, "step": 1603 }, { "entropy": 0.2534871771931648, "epoch": 2.4714946070878274, "grad_norm": 0.6045148968696594, "learning_rate": 1.660021821101222e-06, "loss": 0.235, "mean_token_accuracy": 0.9384705126285553, "num_tokens": 76976487.0, "step": 1604 }, { "entropy": 0.25306981801986694, "epoch": 2.473035439137134, "grad_norm": 0.6134238839149475, "learning_rate": 1.6506589071592427e-06, "loss": 0.2353, "mean_token_accuracy": 0.937341570854187, "num_tokens": 77032407.0, "step": 1605 }, { "entropy": 0.24122758209705353, "epoch": 2.4745762711864407, "grad_norm": 0.5992720127105713, "learning_rate": 1.6413200966013965e-06, "loss": 0.2153, "mean_token_accuracy": 0.9439417719841003, "num_tokens": 77087327.0, "step": 1606 }, { "entropy": 0.22580602765083313, "epoch": 2.4761171032357474, "grad_norm": 0.6415355205535889, "learning_rate": 1.6320054163875288e-06, "loss": 0.2017, "mean_token_accuracy": 0.9473961889743805, "num_tokens": 77140733.0, "step": 1607 }, { "entropy": 0.2227180451154709, "epoch": 2.477657935285054, "grad_norm": 0.612894594669342, "learning_rate": 1.6227148934078195e-06, "loss": 0.2086, "mean_token_accuracy": 0.9445943236351013, "num_tokens": 77190713.0, "step": 1608 }, { "entropy": 0.741747260093689, "epoch": 2.4791987673343607, "grad_norm": 0.9104878306388855, "learning_rate": 1.6134485544827116e-06, "loss": 0.7692, "mean_token_accuracy": 0.7881543636322021, "num_tokens": 77203628.0, "step": 1609 }, { "entropy": 0.6632975041866302, "epoch": 2.480739599383667, "grad_norm": 1.093065619468689, "learning_rate": 1.60420642636284e-06, "loss": 0.6401, "mean_token_accuracy": 0.8290397524833679, "num_tokens": 77213273.0, "step": 1610 }, { "entropy": 0.3935920149087906, "epoch": 2.482280431432974, "grad_norm": 0.6205646991729736, "learning_rate": 1.5949885357289375e-06, "loss": 0.188, "mean_token_accuracy": 0.9022161066532135, "num_tokens": 77249706.0, "step": 1611 }, { "entropy": 0.22564923763275146, "epoch": 2.4838212634822803, "grad_norm": 0.536590039730072, "learning_rate": 1.5857949091917735e-06, "loss": 0.2094, "mean_token_accuracy": 0.9423957765102386, "num_tokens": 77314167.0, "step": 1612 }, { "entropy": 0.2778533548116684, "epoch": 2.485362095531587, "grad_norm": 0.5649770498275757, "learning_rate": 1.5766255732920576e-06, "loss": 0.2617, "mean_token_accuracy": 0.9312827885150909, "num_tokens": 77375593.0, "step": 1613 }, { "entropy": 0.2655201256275177, "epoch": 2.4869029275808936, "grad_norm": 0.6241227984428406, "learning_rate": 1.5674805545003934e-06, "loss": 0.2499, "mean_token_accuracy": 0.9324565529823303, "num_tokens": 77435427.0, "step": 1614 }, { "entropy": 0.2570401430130005, "epoch": 2.4884437596302003, "grad_norm": 0.5743791460990906, "learning_rate": 1.5583598792171727e-06, "loss": 0.2417, "mean_token_accuracy": 0.9356797933578491, "num_tokens": 77494053.0, "step": 1615 }, { "entropy": 0.2758473753929138, "epoch": 2.489984591679507, "grad_norm": 0.6510738134384155, "learning_rate": 1.5492635737725138e-06, "loss": 0.2584, "mean_token_accuracy": 0.9311342835426331, "num_tokens": 77551376.0, "step": 1616 }, { "entropy": 0.2631407752633095, "epoch": 2.4915254237288136, "grad_norm": 0.6303712129592896, "learning_rate": 1.5401916644261816e-06, "loss": 0.2426, "mean_token_accuracy": 0.9367357790470123, "num_tokens": 77607467.0, "step": 1617 }, { "entropy": 0.24975191056728363, "epoch": 2.4930662557781202, "grad_norm": 0.6305494904518127, "learning_rate": 1.5311441773675173e-06, "loss": 0.2312, "mean_token_accuracy": 0.9393047094345093, "num_tokens": 77662359.0, "step": 1618 }, { "entropy": 0.2325933426618576, "epoch": 2.494607087827427, "grad_norm": 0.6517180800437927, "learning_rate": 1.5221211387153522e-06, "loss": 0.2171, "mean_token_accuracy": 0.9429868161678314, "num_tokens": 77715919.0, "step": 1619 }, { "entropy": 0.24089017510414124, "epoch": 2.4961479198767336, "grad_norm": 0.6395618319511414, "learning_rate": 1.5131225745179423e-06, "loss": 0.2188, "mean_token_accuracy": 0.9427703619003296, "num_tokens": 77767623.0, "step": 1620 }, { "entropy": 0.4312155097723007, "epoch": 2.4976887519260402, "grad_norm": 0.7245088815689087, "learning_rate": 1.5041485107528885e-06, "loss": 0.3022, "mean_token_accuracy": 0.8879015445709229, "num_tokens": 77800847.0, "step": 1621 }, { "entropy": 0.7008659839630127, "epoch": 2.499229583975347, "grad_norm": 1.0753662586212158, "learning_rate": 1.4951989733270622e-06, "loss": 0.6688, "mean_token_accuracy": 0.812565952539444, "num_tokens": 77811541.0, "step": 1622 }, { "entropy": 0.6402573585510254, "epoch": 2.500770416024653, "grad_norm": 1.2609517574310303, "learning_rate": 1.4862739880765309e-06, "loss": 0.6142, "mean_token_accuracy": 0.8419844806194305, "num_tokens": 77819615.0, "step": 1623 }, { "entropy": 0.17620781809091568, "epoch": 2.50231124807396, "grad_norm": 0.4153822362422943, "learning_rate": 1.4773735807664824e-06, "loss": 0.1544, "mean_token_accuracy": 0.9582631587982178, "num_tokens": 77885043.0, "step": 1624 }, { "entropy": 0.23164799064397812, "epoch": 2.5038520801232664, "grad_norm": 0.5407425761222839, "learning_rate": 1.4684977770911535e-06, "loss": 0.2119, "mean_token_accuracy": 0.9420932531356812, "num_tokens": 77947901.0, "step": 1625 }, { "entropy": 0.2571442648768425, "epoch": 2.505392912172573, "grad_norm": 0.5600372552871704, "learning_rate": 1.4596466026737489e-06, "loss": 0.2364, "mean_token_accuracy": 0.9366209805011749, "num_tokens": 78008722.0, "step": 1626 }, { "entropy": 0.25656041502952576, "epoch": 2.5069337442218798, "grad_norm": 0.5601813197135925, "learning_rate": 1.4508200830663833e-06, "loss": 0.2421, "mean_token_accuracy": 0.9354088604450226, "num_tokens": 78068284.0, "step": 1627 }, { "entropy": 0.2537119537591934, "epoch": 2.5084745762711864, "grad_norm": 0.5964526534080505, "learning_rate": 1.4420182437499809e-06, "loss": 0.2348, "mean_token_accuracy": 0.936476081609726, "num_tokens": 78126557.0, "step": 1628 }, { "entropy": 0.25967881828546524, "epoch": 2.510015408320493, "grad_norm": 0.6520641446113586, "learning_rate": 1.4332411101342259e-06, "loss": 0.24, "mean_token_accuracy": 0.936371922492981, "num_tokens": 78183790.0, "step": 1629 }, { "entropy": 0.26733292639255524, "epoch": 2.5115562403697997, "grad_norm": 0.5961201786994934, "learning_rate": 1.4244887075574844e-06, "loss": 0.2498, "mean_token_accuracy": 0.9324370622634888, "num_tokens": 78240084.0, "step": 1630 }, { "entropy": 0.2514963746070862, "epoch": 2.5130970724191064, "grad_norm": 0.6373652815818787, "learning_rate": 1.4157610612867202e-06, "loss": 0.2286, "mean_token_accuracy": 0.9403288662433624, "num_tokens": 78295280.0, "step": 1631 }, { "entropy": 0.2533837705850601, "epoch": 2.514637904468413, "grad_norm": 0.5919271111488342, "learning_rate": 1.4070581965174323e-06, "loss": 0.2293, "mean_token_accuracy": 0.9403780400753021, "num_tokens": 78349070.0, "step": 1632 }, { "entropy": 0.23580175638198853, "epoch": 2.5161787365177197, "grad_norm": 0.6381546854972839, "learning_rate": 1.3983801383735784e-06, "loss": 0.2139, "mean_token_accuracy": 0.9445627331733704, "num_tokens": 78400753.0, "step": 1633 }, { "entropy": 0.5720073878765106, "epoch": 2.517719568567026, "grad_norm": 0.89513099193573, "learning_rate": 1.389726911907502e-06, "loss": 0.4326, "mean_token_accuracy": 0.8543025553226471, "num_tokens": 78422248.0, "step": 1634 }, { "entropy": 0.6621885597705841, "epoch": 2.519260400616333, "grad_norm": 1.01436185836792, "learning_rate": 1.3810985420998624e-06, "loss": 0.6308, "mean_token_accuracy": 0.8321410715579987, "num_tokens": 78432030.0, "step": 1635 }, { "entropy": 0.3978841304779053, "epoch": 2.5208012326656393, "grad_norm": 0.6065288782119751, "learning_rate": 1.3724950538595616e-06, "loss": 0.1948, "mean_token_accuracy": 0.9033833146095276, "num_tokens": 78468392.0, "step": 1636 }, { "entropy": 0.20841382443904877, "epoch": 2.522342064714946, "grad_norm": 0.5003097653388977, "learning_rate": 1.3639164720236708e-06, "loss": 0.1905, "mean_token_accuracy": 0.9482710361480713, "num_tokens": 78532506.0, "step": 1637 }, { "entropy": 0.24048367142677307, "epoch": 2.5238828967642526, "grad_norm": 0.5504543781280518, "learning_rate": 1.3553628213573589e-06, "loss": 0.225, "mean_token_accuracy": 0.937564492225647, "num_tokens": 78593894.0, "step": 1638 }, { "entropy": 0.2742376923561096, "epoch": 2.5254237288135593, "grad_norm": 0.6072757244110107, "learning_rate": 1.3468341265538243e-06, "loss": 0.2642, "mean_token_accuracy": 0.930324137210846, "num_tokens": 78653435.0, "step": 1639 }, { "entropy": 0.24892327934503555, "epoch": 2.526964560862866, "grad_norm": 0.6219666004180908, "learning_rate": 1.338330412234221e-06, "loss": 0.2319, "mean_token_accuracy": 0.9380239844322205, "num_tokens": 78711625.0, "step": 1640 }, { "entropy": 0.21802204102277756, "epoch": 2.5285053929121726, "grad_norm": 0.5473381876945496, "learning_rate": 1.3298517029475854e-06, "loss": 0.1979, "mean_token_accuracy": 0.9461008906364441, "num_tokens": 78768753.0, "step": 1641 }, { "entropy": 0.2487521842122078, "epoch": 2.5300462249614792, "grad_norm": 0.691769003868103, "learning_rate": 1.3213980231707691e-06, "loss": 0.2342, "mean_token_accuracy": 0.9384419918060303, "num_tokens": 78824805.0, "step": 1642 }, { "entropy": 0.26256680488586426, "epoch": 2.531587057010786, "grad_norm": 0.5857059359550476, "learning_rate": 1.3129693973083734e-06, "loss": 0.2489, "mean_token_accuracy": 0.9359204471111298, "num_tokens": 78879624.0, "step": 1643 }, { "entropy": 0.2355850338935852, "epoch": 2.5331278890600926, "grad_norm": 0.6057854890823364, "learning_rate": 1.3045658496926594e-06, "loss": 0.2151, "mean_token_accuracy": 0.9433797895908356, "num_tokens": 78932976.0, "step": 1644 }, { "entropy": 0.22897650301456451, "epoch": 2.5346687211093992, "grad_norm": 0.6639989614486694, "learning_rate": 1.2961874045835011e-06, "loss": 0.212, "mean_token_accuracy": 0.9453070163726807, "num_tokens": 78983480.0, "step": 1645 }, { "entropy": 0.7509763240814209, "epoch": 2.536209553158706, "grad_norm": 0.8632983565330505, "learning_rate": 1.2878340861682993e-06, "loss": 0.7471, "mean_token_accuracy": 0.7912810742855072, "num_tokens": 78997784.0, "step": 1646 }, { "entropy": 0.6809007227420807, "epoch": 2.537750385208012, "grad_norm": 1.1274269819259644, "learning_rate": 1.279505918561923e-06, "loss": 0.6446, "mean_token_accuracy": 0.8273549675941467, "num_tokens": 79008038.0, "step": 1647 }, { "entropy": 0.6138690412044525, "epoch": 2.539291217257319, "grad_norm": 1.2140837907791138, "learning_rate": 1.2712029258066338e-06, "loss": 0.6069, "mean_token_accuracy": 0.8417371213436127, "num_tokens": 79016055.0, "step": 1648 }, { "entropy": 0.19548485428094864, "epoch": 2.5408320493066254, "grad_norm": 0.5157225131988525, "learning_rate": 1.2629251318720092e-06, "loss": 0.1802, "mean_token_accuracy": 0.9513345062732697, "num_tokens": 79081395.0, "step": 1649 }, { "entropy": 0.27277927100658417, "epoch": 2.542372881355932, "grad_norm": 0.5577870011329651, "learning_rate": 1.2546725606548871e-06, "loss": 0.2527, "mean_token_accuracy": 0.9308972954750061, "num_tokens": 79144116.0, "step": 1650 }, { "entropy": 0.24077941477298737, "epoch": 2.5439137134052388, "grad_norm": 0.5634075999259949, "learning_rate": 1.2464452359792932e-06, "loss": 0.223, "mean_token_accuracy": 0.940977156162262, "num_tokens": 79204920.0, "step": 1651 }, { "entropy": 0.26552577316761017, "epoch": 2.5454545454545454, "grad_norm": 0.5863950252532959, "learning_rate": 1.238243181596367e-06, "loss": 0.2507, "mean_token_accuracy": 0.933573842048645, "num_tokens": 79264166.0, "step": 1652 }, { "entropy": 0.2663063704967499, "epoch": 2.546995377503852, "grad_norm": 0.6204710006713867, "learning_rate": 1.230066421184296e-06, "loss": 0.2516, "mean_token_accuracy": 0.934893399477005, "num_tokens": 79322253.0, "step": 1653 }, { "entropy": 0.26322437822818756, "epoch": 2.5485362095531587, "grad_norm": 0.605076789855957, "learning_rate": 1.2219149783482464e-06, "loss": 0.2476, "mean_token_accuracy": 0.9338852763175964, "num_tokens": 79379314.0, "step": 1654 }, { "entropy": 0.27077604830265045, "epoch": 2.5500770416024654, "grad_norm": 0.6867201328277588, "learning_rate": 1.2137888766202998e-06, "loss": 0.2567, "mean_token_accuracy": 0.9329261779785156, "num_tokens": 79435240.0, "step": 1655 }, { "entropy": 0.2478635460138321, "epoch": 2.551617873651772, "grad_norm": 0.594097375869751, "learning_rate": 1.2056881394593788e-06, "loss": 0.2305, "mean_token_accuracy": 0.9395008385181427, "num_tokens": 79490016.0, "step": 1656 }, { "entropy": 0.26441776752471924, "epoch": 2.5531587057010787, "grad_norm": 0.6184442639350891, "learning_rate": 1.1976127902511825e-06, "loss": 0.2479, "mean_token_accuracy": 0.9364229738712311, "num_tokens": 79543326.0, "step": 1657 }, { "entropy": 0.21989805996418, "epoch": 2.554699537750385, "grad_norm": 0.5758084058761597, "learning_rate": 1.189562852308117e-06, "loss": 0.2034, "mean_token_accuracy": 0.9473601877689362, "num_tokens": 79594239.0, "step": 1658 }, { "entropy": 0.6643407940864563, "epoch": 2.556240369799692, "grad_norm": 0.8047322630882263, "learning_rate": 1.1815383488692345e-06, "loss": 0.5431, "mean_token_accuracy": 0.8358457684516907, "num_tokens": 79611690.0, "step": 1659 }, { "entropy": 0.7003147602081299, "epoch": 2.5577812018489983, "grad_norm": 1.0712970495224, "learning_rate": 1.1735393031001551e-06, "loss": 0.6671, "mean_token_accuracy": 0.8230919241905212, "num_tokens": 79621248.0, "step": 1660 }, { "entropy": 0.39504458755254745, "epoch": 2.559322033898305, "grad_norm": 0.5714684128761292, "learning_rate": 1.1655657380930107e-06, "loss": 0.2128, "mean_token_accuracy": 0.8952510356903076, "num_tokens": 79657455.0, "step": 1661 }, { "entropy": 0.2527701109647751, "epoch": 2.5608628659476116, "grad_norm": 0.5644436478614807, "learning_rate": 1.1576176768663706e-06, "loss": 0.2321, "mean_token_accuracy": 0.9367364346981049, "num_tokens": 79721484.0, "step": 1662 }, { "entropy": 0.24136274307966232, "epoch": 2.5624036979969183, "grad_norm": 0.5415719151496887, "learning_rate": 1.1496951423651848e-06, "loss": 0.2257, "mean_token_accuracy": 0.9384903013706207, "num_tokens": 79782634.0, "step": 1663 }, { "entropy": 0.2612917572259903, "epoch": 2.563944530046225, "grad_norm": 0.5944631099700928, "learning_rate": 1.1417981574607006e-06, "loss": 0.2443, "mean_token_accuracy": 0.9344781339168549, "num_tokens": 79842259.0, "step": 1664 }, { "entropy": 0.23901791870594025, "epoch": 2.5654853620955316, "grad_norm": 0.5921348929405212, "learning_rate": 1.1339267449504143e-06, "loss": 0.2183, "mean_token_accuracy": 0.9413757622241974, "num_tokens": 79900491.0, "step": 1665 }, { "entropy": 0.2507576271891594, "epoch": 2.5670261941448382, "grad_norm": 0.5888736844062805, "learning_rate": 1.1260809275579966e-06, "loss": 0.2308, "mean_token_accuracy": 0.9379897713661194, "num_tokens": 79957457.0, "step": 1666 }, { "entropy": 0.2534849941730499, "epoch": 2.568567026194145, "grad_norm": 0.5988259315490723, "learning_rate": 1.1182607279332314e-06, "loss": 0.2327, "mean_token_accuracy": 0.9390934407711029, "num_tokens": 80013281.0, "step": 1667 }, { "entropy": 0.2675626575946808, "epoch": 2.5701078582434516, "grad_norm": 0.6686258316040039, "learning_rate": 1.1104661686519436e-06, "loss": 0.2485, "mean_token_accuracy": 0.9349306523799896, "num_tokens": 80068070.0, "step": 1668 }, { "entropy": 0.24536453187465668, "epoch": 2.571648690292758, "grad_norm": 0.5922843217849731, "learning_rate": 1.1026972722159412e-06, "loss": 0.227, "mean_token_accuracy": 0.940820038318634, "num_tokens": 80121462.0, "step": 1669 }, { "entropy": 0.2267133742570877, "epoch": 2.573189522342065, "grad_norm": 0.5579976439476013, "learning_rate": 1.0949540610529396e-06, "loss": 0.2115, "mean_token_accuracy": 0.9449692666530609, "num_tokens": 80172779.0, "step": 1670 }, { "entropy": 0.515229694545269, "epoch": 2.574730354391371, "grad_norm": 0.7066289782524109, "learning_rate": 1.0872365575165168e-06, "loss": 0.3353, "mean_token_accuracy": 0.8645601868629456, "num_tokens": 80200290.0, "step": 1671 }, { "entropy": 0.6711301207542419, "epoch": 2.576271186440678, "grad_norm": 1.0317108631134033, "learning_rate": 1.0795447838860273e-06, "loss": 0.6317, "mean_token_accuracy": 0.8312576711177826, "num_tokens": 80210699.0, "step": 1672 }, { "entropy": 0.6207587122917175, "epoch": 2.5778120184899844, "grad_norm": 1.2237693071365356, "learning_rate": 1.0718787623665504e-06, "loss": 0.6253, "mean_token_accuracy": 0.8388943672180176, "num_tokens": 80218497.0, "step": 1673 }, { "entropy": 0.18502859771251678, "epoch": 2.579352850539291, "grad_norm": 0.4228017330169678, "learning_rate": 1.0642385150888213e-06, "loss": 0.1652, "mean_token_accuracy": 0.9547512531280518, "num_tokens": 80283868.0, "step": 1674 }, { "entropy": 0.2795054018497467, "epoch": 2.5808936825885977, "grad_norm": 0.6087064743041992, "learning_rate": 1.0566240641091686e-06, "loss": 0.2627, "mean_token_accuracy": 0.9295134246349335, "num_tokens": 80346484.0, "step": 1675 }, { "entropy": 0.2647228538990021, "epoch": 2.5824345146379044, "grad_norm": 0.6077116131782532, "learning_rate": 1.049035431409452e-06, "loss": 0.2492, "mean_token_accuracy": 0.9334661662578583, "num_tokens": 80407142.0, "step": 1676 }, { "entropy": 0.25552527606487274, "epoch": 2.583975346687211, "grad_norm": 0.569162130355835, "learning_rate": 1.0414726388969964e-06, "loss": 0.2363, "mean_token_accuracy": 0.9373045563697815, "num_tokens": 80466031.0, "step": 1677 }, { "entropy": 0.2259029746055603, "epoch": 2.5855161787365177, "grad_norm": 0.555417537689209, "learning_rate": 1.03393570840453e-06, "loss": 0.2095, "mean_token_accuracy": 0.9431947767734528, "num_tokens": 80523729.0, "step": 1678 }, { "entropy": 0.27299466729164124, "epoch": 2.5870570107858244, "grad_norm": 0.6381082534790039, "learning_rate": 1.0264246616901197e-06, "loss": 0.2516, "mean_token_accuracy": 0.9336928427219391, "num_tokens": 80580379.0, "step": 1679 }, { "entropy": 0.24026623368263245, "epoch": 2.588597842835131, "grad_norm": 0.5867028832435608, "learning_rate": 1.018939520437111e-06, "loss": 0.2259, "mean_token_accuracy": 0.9407189190387726, "num_tokens": 80635930.0, "step": 1680 }, { "entropy": 0.25235386192798615, "epoch": 2.5901386748844377, "grad_norm": 0.65633624792099, "learning_rate": 1.0114803062540647e-06, "loss": 0.2363, "mean_token_accuracy": 0.9378034174442291, "num_tokens": 80690173.0, "step": 1681 }, { "entropy": 0.24471809715032578, "epoch": 2.5916795069337444, "grad_norm": 0.606451153755188, "learning_rate": 1.0040470406746916e-06, "loss": 0.2306, "mean_token_accuracy": 0.9401601254940033, "num_tokens": 80743098.0, "step": 1682 }, { "entropy": 0.2924126982688904, "epoch": 2.593220338983051, "grad_norm": 0.6483690142631531, "learning_rate": 9.966397451577914e-07, "loss": 0.2269, "mean_token_accuracy": 0.9367680847644806, "num_tokens": 80787680.0, "step": 1683 }, { "entropy": 0.7727250158786774, "epoch": 2.5947611710323573, "grad_norm": 1.0054664611816406, "learning_rate": 9.892584410871996e-07, "loss": 0.7899, "mean_token_accuracy": 0.7854651808738708, "num_tokens": 80799680.0, "step": 1684 }, { "entropy": 0.6798030734062195, "epoch": 2.5963020030816644, "grad_norm": 1.1979365348815918, "learning_rate": 9.81903149771707e-07, "loss": 0.6402, "mean_token_accuracy": 0.8262642920017242, "num_tokens": 80809031.0, "step": 1685 }, { "entropy": 0.39006907492876053, "epoch": 2.5978428351309706, "grad_norm": 0.5220504999160767, "learning_rate": 9.745738924450143e-07, "loss": 0.2081, "mean_token_accuracy": 0.8990786969661713, "num_tokens": 80845361.0, "step": 1686 }, { "entropy": 0.23079591989517212, "epoch": 2.5993836671802772, "grad_norm": 0.6694757342338562, "learning_rate": 9.672706902656704e-07, "loss": 0.2097, "mean_token_accuracy": 0.9432952404022217, "num_tokens": 80910059.0, "step": 1687 }, { "entropy": 0.2380344197154045, "epoch": 2.600924499229584, "grad_norm": 0.5708295702934265, "learning_rate": 9.59993564317e-07, "loss": 0.2252, "mean_token_accuracy": 0.9379231631755829, "num_tokens": 80971903.0, "step": 1688 }, { "entropy": 0.25591251254081726, "epoch": 2.6024653312788906, "grad_norm": 0.6017871499061584, "learning_rate": 9.52742535607053e-07, "loss": 0.2356, "mean_token_accuracy": 0.936410516500473, "num_tokens": 81031905.0, "step": 1689 }, { "entropy": 0.272662490606308, "epoch": 2.6040061633281972, "grad_norm": 0.592627763748169, "learning_rate": 9.455176250685338e-07, "loss": 0.2568, "mean_token_accuracy": 0.931819885969162, "num_tokens": 81090667.0, "step": 1690 }, { "entropy": 0.2180783674120903, "epoch": 2.605546995377504, "grad_norm": 0.5514487624168396, "learning_rate": 9.383188535587573e-07, "loss": 0.2017, "mean_token_accuracy": 0.9472089111804962, "num_tokens": 81148128.0, "step": 1691 }, { "entropy": 0.23975679278373718, "epoch": 2.6070878274268106, "grad_norm": 0.5956898331642151, "learning_rate": 9.311462418595718e-07, "loss": 0.2185, "mean_token_accuracy": 0.9423589706420898, "num_tokens": 81204547.0, "step": 1692 }, { "entropy": 0.2641661465167999, "epoch": 2.608628659476117, "grad_norm": 0.6434975862503052, "learning_rate": 9.239998106773074e-07, "loss": 0.2433, "mean_token_accuracy": 0.9368337392807007, "num_tokens": 81259974.0, "step": 1693 }, { "entropy": 0.2416614294052124, "epoch": 2.610169491525424, "grad_norm": 0.5753040313720703, "learning_rate": 9.168795806427156e-07, "loss": 0.2184, "mean_token_accuracy": 0.9430364668369293, "num_tokens": 81314183.0, "step": 1694 }, { "entropy": 0.2364598661661148, "epoch": 2.61171032357473, "grad_norm": 0.6098959445953369, "learning_rate": 9.097855723109083e-07, "loss": 0.2144, "mean_token_accuracy": 0.9441978335380554, "num_tokens": 81366829.0, "step": 1695 }, { "entropy": 0.39678433537483215, "epoch": 2.613251155624037, "grad_norm": 0.6541060209274292, "learning_rate": 9.027178061612984e-07, "loss": 0.2804, "mean_token_accuracy": 0.9137652218341827, "num_tokens": 81404898.0, "step": 1696 }, { "entropy": 0.7222594320774078, "epoch": 2.6147919876733434, "grad_norm": 1.0389997959136963, "learning_rate": 8.956763025975413e-07, "loss": 0.7065, "mean_token_accuracy": 0.8054831027984619, "num_tokens": 81416132.0, "step": 1697 }, { "entropy": 0.6383982002735138, "epoch": 2.61633281972265, "grad_norm": 1.2334141731262207, "learning_rate": 8.886610819474772e-07, "loss": 0.6258, "mean_token_accuracy": 0.8379530906677246, "num_tokens": 81424249.0, "step": 1698 }, { "entropy": 0.16487031430006027, "epoch": 2.6178736517719567, "grad_norm": 0.42047828435897827, "learning_rate": 8.81672164463071e-07, "loss": 0.146, "mean_token_accuracy": 0.9604553282260895, "num_tokens": 81489566.0, "step": 1699 }, { "entropy": 0.23015689849853516, "epoch": 2.6194144838212634, "grad_norm": 0.5252680778503418, "learning_rate": 8.747095703203523e-07, "loss": 0.2193, "mean_token_accuracy": 0.9395607709884644, "num_tokens": 81552083.0, "step": 1700 }, { "entropy": 0.26038722693920135, "epoch": 2.62095531587057, "grad_norm": 0.5519822239875793, "learning_rate": 8.677733196193605e-07, "loss": 0.2438, "mean_token_accuracy": 0.934130072593689, "num_tokens": 81612753.0, "step": 1701 }, { "entropy": 0.24006182700395584, "epoch": 2.6224961479198767, "grad_norm": 0.5355826020240784, "learning_rate": 8.608634323840848e-07, "loss": 0.2257, "mean_token_accuracy": 0.9392687678337097, "num_tokens": 81672005.0, "step": 1702 }, { "entropy": 0.25130411982536316, "epoch": 2.6240369799691834, "grad_norm": 0.5445948839187622, "learning_rate": 8.53979928562404e-07, "loss": 0.2305, "mean_token_accuracy": 0.939100831747055, "num_tokens": 81729991.0, "step": 1703 }, { "entropy": 0.25270039588212967, "epoch": 2.62557781201849, "grad_norm": 0.5955731868743896, "learning_rate": 8.471228280260401e-07, "loss": 0.2368, "mean_token_accuracy": 0.9383566975593567, "num_tokens": 81786860.0, "step": 1704 }, { "entropy": 0.2432229369878769, "epoch": 2.6271186440677967, "grad_norm": 0.5946387052536011, "learning_rate": 8.402921505704842e-07, "loss": 0.224, "mean_token_accuracy": 0.9409800469875336, "num_tokens": 81842667.0, "step": 1705 }, { "entropy": 0.2509583681821823, "epoch": 2.6286594761171034, "grad_norm": 0.6055125594139099, "learning_rate": 8.334879159149478e-07, "loss": 0.2349, "mean_token_accuracy": 0.940176248550415, "num_tokens": 81897407.0, "step": 1706 }, { "entropy": 0.276890367269516, "epoch": 2.63020030816641, "grad_norm": 0.6234329342842102, "learning_rate": 8.267101437023084e-07, "loss": 0.2598, "mean_token_accuracy": 0.9331550598144531, "num_tokens": 81950795.0, "step": 1707 }, { "entropy": 0.23030610382556915, "epoch": 2.6317411402157163, "grad_norm": 0.6149360537528992, "learning_rate": 8.199588534990521e-07, "loss": 0.2098, "mean_token_accuracy": 0.9441168308258057, "num_tokens": 82001996.0, "step": 1708 }, { "entropy": 0.5932334959506989, "epoch": 2.6332819722650234, "grad_norm": 0.7792565822601318, "learning_rate": 8.132340647952119e-07, "loss": 0.4259, "mean_token_accuracy": 0.8467635214328766, "num_tokens": 82025630.0, "step": 1709 }, { "entropy": 0.6788563132286072, "epoch": 2.6348228043143296, "grad_norm": 1.0804792642593384, "learning_rate": 8.065357970043163e-07, "loss": 0.6358, "mean_token_accuracy": 0.828337550163269, "num_tokens": 82035508.0, "step": 1710 }, { "entropy": 0.3760470300912857, "epoch": 2.6363636363636362, "grad_norm": 0.45538073778152466, "learning_rate": 7.998640694633297e-07, "loss": 0.1517, "mean_token_accuracy": 0.9101597964763641, "num_tokens": 82072010.0, "step": 1711 }, { "entropy": 0.2237078994512558, "epoch": 2.637904468412943, "grad_norm": 0.525282621383667, "learning_rate": 7.932189014326008e-07, "loss": 0.2069, "mean_token_accuracy": 0.944471001625061, "num_tokens": 82136240.0, "step": 1712 }, { "entropy": 0.23701447248458862, "epoch": 2.6394453004622496, "grad_norm": 0.5296241044998169, "learning_rate": 7.866003120958022e-07, "loss": 0.2164, "mean_token_accuracy": 0.9415386319160461, "num_tokens": 82197511.0, "step": 1713 }, { "entropy": 0.2406381219625473, "epoch": 2.6409861325115562, "grad_norm": 0.5624717473983765, "learning_rate": 7.800083205598796e-07, "loss": 0.2285, "mean_token_accuracy": 0.9395492374897003, "num_tokens": 82257175.0, "step": 1714 }, { "entropy": 0.24990498274564743, "epoch": 2.642526964560863, "grad_norm": 0.5809083580970764, "learning_rate": 7.734429458549941e-07, "loss": 0.2331, "mean_token_accuracy": 0.9379479885101318, "num_tokens": 82315561.0, "step": 1715 }, { "entropy": 0.25641772150993347, "epoch": 2.6440677966101696, "grad_norm": 0.6739305853843689, "learning_rate": 7.669042069344678e-07, "loss": 0.2446, "mean_token_accuracy": 0.935545027256012, "num_tokens": 82372925.0, "step": 1716 }, { "entropy": 0.2460051327943802, "epoch": 2.645608628659476, "grad_norm": 0.5788447260856628, "learning_rate": 7.60392122674728e-07, "loss": 0.2222, "mean_token_accuracy": 0.9397186934947968, "num_tokens": 82429412.0, "step": 1717 }, { "entropy": 0.2642780989408493, "epoch": 2.647149460708783, "grad_norm": 0.6441606283187866, "learning_rate": 7.539067118752541e-07, "loss": 0.2513, "mean_token_accuracy": 0.9348781406879425, "num_tokens": 82484906.0, "step": 1718 }, { "entropy": 0.23527581989765167, "epoch": 2.648690292758089, "grad_norm": 0.61605304479599, "learning_rate": 7.474479932585221e-07, "loss": 0.2176, "mean_token_accuracy": 0.9419949650764465, "num_tokens": 82538947.0, "step": 1719 }, { "entropy": 0.242695651948452, "epoch": 2.650231124807396, "grad_norm": 0.6713213324546814, "learning_rate": 7.410159854699584e-07, "loss": 0.2203, "mean_token_accuracy": 0.9422966837882996, "num_tokens": 82590918.0, "step": 1720 }, { "entropy": 0.548020750284195, "epoch": 2.6517719568567024, "grad_norm": 0.7737093567848206, "learning_rate": 7.346107070778674e-07, "loss": 0.3777, "mean_token_accuracy": 0.8572807013988495, "num_tokens": 82615194.0, "step": 1721 }, { "entropy": 0.6853159964084625, "epoch": 2.653312788906009, "grad_norm": 1.0621554851531982, "learning_rate": 7.282321765733957e-07, "loss": 0.6744, "mean_token_accuracy": 0.8118784129619598, "num_tokens": 82625794.0, "step": 1722 }, { "entropy": 0.6466008424758911, "epoch": 2.6548536209553157, "grad_norm": 1.2829952239990234, "learning_rate": 7.218804123704715e-07, "loss": 0.6427, "mean_token_accuracy": 0.8291504383087158, "num_tokens": 82633838.0, "step": 1723 }, { "entropy": 0.19802788645029068, "epoch": 2.6563944530046224, "grad_norm": 0.47456225752830505, "learning_rate": 7.155554328057556e-07, "loss": 0.1778, "mean_token_accuracy": 0.9505210816860199, "num_tokens": 82698953.0, "step": 1724 }, { "entropy": 0.2834305018186569, "epoch": 2.657935285053929, "grad_norm": 0.6096200346946716, "learning_rate": 7.092572561385836e-07, "loss": 0.2717, "mean_token_accuracy": 0.928535521030426, "num_tokens": 82761261.0, "step": 1725 }, { "entropy": 0.2659959942102432, "epoch": 2.6594761171032357, "grad_norm": 0.5810233950614929, "learning_rate": 7.029859005509087e-07, "loss": 0.2482, "mean_token_accuracy": 0.9359013140201569, "num_tokens": 82821794.0, "step": 1726 }, { "entropy": 0.27526599168777466, "epoch": 2.6610169491525424, "grad_norm": 0.5995367169380188, "learning_rate": 6.967413841472625e-07, "loss": 0.2597, "mean_token_accuracy": 0.9324967861175537, "num_tokens": 82881050.0, "step": 1727 }, { "entropy": 0.25813791155815125, "epoch": 2.662557781201849, "grad_norm": 0.6224498152732849, "learning_rate": 6.905237249546947e-07, "loss": 0.2409, "mean_token_accuracy": 0.9369837045669556, "num_tokens": 82939158.0, "step": 1728 }, { "entropy": 0.26024875044822693, "epoch": 2.6640986132511557, "grad_norm": 0.6341143250465393, "learning_rate": 6.843329409227206e-07, "loss": 0.2398, "mean_token_accuracy": 0.9376099109649658, "num_tokens": 82996054.0, "step": 1729 }, { "entropy": 0.2460588961839676, "epoch": 2.6656394453004624, "grad_norm": 0.6055461168289185, "learning_rate": 6.781690499232729e-07, "loss": 0.2285, "mean_token_accuracy": 0.9401125907897949, "num_tokens": 83051845.0, "step": 1730 }, { "entropy": 0.24418766051530838, "epoch": 2.667180277349769, "grad_norm": 0.6128745675086975, "learning_rate": 6.720320697506399e-07, "loss": 0.2243, "mean_token_accuracy": 0.9417503178119659, "num_tokens": 83106412.0, "step": 1731 }, { "entropy": 0.2403196170926094, "epoch": 2.6687211093990753, "grad_norm": 0.5712807178497314, "learning_rate": 6.659220181214343e-07, "loss": 0.2183, "mean_token_accuracy": 0.9417198002338409, "num_tokens": 83159720.0, "step": 1732 }, { "entropy": 0.2347535714507103, "epoch": 2.6702619414483824, "grad_norm": 0.6209002137184143, "learning_rate": 6.598389126745209e-07, "loss": 0.2134, "mean_token_accuracy": 0.9437917172908783, "num_tokens": 83210435.0, "step": 1733 }, { "entropy": 0.6867729127407074, "epoch": 2.6718027734976886, "grad_norm": 0.8483384847640991, "learning_rate": 6.537827709709765e-07, "loss": 0.5674, "mean_token_accuracy": 0.8303823471069336, "num_tokens": 83227261.0, "step": 1734 }, { "entropy": 0.6366385221481323, "epoch": 2.6733436055469952, "grad_norm": 1.1306341886520386, "learning_rate": 6.477536104940407e-07, "loss": 0.6284, "mean_token_accuracy": 0.8363052308559418, "num_tokens": 83236883.0, "step": 1735 }, { "entropy": 0.38439562916755676, "epoch": 2.674884437596302, "grad_norm": 0.6276645064353943, "learning_rate": 6.417514486490572e-07, "loss": 0.201, "mean_token_accuracy": 0.9052903056144714, "num_tokens": 83273369.0, "step": 1736 }, { "entropy": 0.23202861845493317, "epoch": 2.6764252696456086, "grad_norm": 0.49214962124824524, "learning_rate": 6.357763027634323e-07, "loss": 0.2158, "mean_token_accuracy": 0.941438227891922, "num_tokens": 83338033.0, "step": 1737 }, { "entropy": 0.2383202686905861, "epoch": 2.6779661016949152, "grad_norm": 0.5271440744400024, "learning_rate": 6.298281900865788e-07, "loss": 0.2177, "mean_token_accuracy": 0.9411033391952515, "num_tokens": 83399776.0, "step": 1738 }, { "entropy": 0.2624776363372803, "epoch": 2.679506933744222, "grad_norm": 0.5833832025527954, "learning_rate": 6.239071277898678e-07, "loss": 0.2424, "mean_token_accuracy": 0.9341283440589905, "num_tokens": 83459884.0, "step": 1739 }, { "entropy": 0.26702384650707245, "epoch": 2.6810477657935285, "grad_norm": 0.5795257687568665, "learning_rate": 6.18013132966585e-07, "loss": 0.2466, "mean_token_accuracy": 0.9346422851085663, "num_tokens": 83518612.0, "step": 1740 }, { "entropy": 0.2458302527666092, "epoch": 2.682588597842835, "grad_norm": 0.5442808866500854, "learning_rate": 6.121462226318675e-07, "loss": 0.2325, "mean_token_accuracy": 0.9388407468795776, "num_tokens": 83576041.0, "step": 1741 }, { "entropy": 0.24486291408538818, "epoch": 2.684129429892142, "grad_norm": 0.5804866552352905, "learning_rate": 6.063064137226693e-07, "loss": 0.2252, "mean_token_accuracy": 0.9408155679702759, "num_tokens": 83632271.0, "step": 1742 }, { "entropy": 0.2432895302772522, "epoch": 2.6856702619414485, "grad_norm": 0.5875058770179749, "learning_rate": 6.004937230977048e-07, "loss": 0.2246, "mean_token_accuracy": 0.9418061077594757, "num_tokens": 83687275.0, "step": 1743 }, { "entropy": 0.25835683941841125, "epoch": 2.687211093990755, "grad_norm": 0.6558237075805664, "learning_rate": 5.947081675374033e-07, "loss": 0.2426, "mean_token_accuracy": 0.9377466440200806, "num_tokens": 83740972.0, "step": 1744 }, { "entropy": 0.21970681846141815, "epoch": 2.6887519260400614, "grad_norm": 0.5938444137573242, "learning_rate": 5.889497637438557e-07, "loss": 0.1959, "mean_token_accuracy": 0.9483272433280945, "num_tokens": 83792954.0, "step": 1745 }, { "entropy": 0.394988089799881, "epoch": 2.6902927580893685, "grad_norm": 0.689375638961792, "learning_rate": 5.832185283407732e-07, "loss": 0.293, "mean_token_accuracy": 0.9130075573921204, "num_tokens": 83831080.0, "step": 1746 }, { "entropy": 0.6911328732967377, "epoch": 2.6918335901386747, "grad_norm": 1.0117076635360718, "learning_rate": 5.775144778734299e-07, "loss": 0.6793, "mean_token_accuracy": 0.816473662853241, "num_tokens": 83841994.0, "step": 1747 }, { "entropy": 0.6352590024471283, "epoch": 2.6933744221879814, "grad_norm": 1.2464711666107178, "learning_rate": 5.71837628808628e-07, "loss": 0.6335, "mean_token_accuracy": 0.8378670513629913, "num_tokens": 83850101.0, "step": 1748 }, { "entropy": 0.18002597987651825, "epoch": 2.694915254237288, "grad_norm": 0.48154088854789734, "learning_rate": 5.661879975346385e-07, "loss": 0.1634, "mean_token_accuracy": 0.955785483121872, "num_tokens": 83915326.0, "step": 1749 }, { "entropy": 0.23977802693843842, "epoch": 2.6964560862865947, "grad_norm": 0.5332176685333252, "learning_rate": 5.605656003611593e-07, "loss": 0.2186, "mean_token_accuracy": 0.94035804271698, "num_tokens": 83977596.0, "step": 1750 }, { "entropy": 0.25191135704517365, "epoch": 2.6979969183359014, "grad_norm": 0.5899388194084167, "learning_rate": 5.549704535192679e-07, "loss": 0.2414, "mean_token_accuracy": 0.9351645708084106, "num_tokens": 84038092.0, "step": 1751 }, { "entropy": 0.2588416188955307, "epoch": 2.699537750385208, "grad_norm": 0.6052764058113098, "learning_rate": 5.494025731613739e-07, "loss": 0.2414, "mean_token_accuracy": 0.9357598721981049, "num_tokens": 84097190.0, "step": 1752 }, { "entropy": 0.2345101609826088, "epoch": 2.7010785824345147, "grad_norm": 0.5940964221954346, "learning_rate": 5.438619753611707e-07, "loss": 0.2228, "mean_token_accuracy": 0.9421818554401398, "num_tokens": 84154691.0, "step": 1753 }, { "entropy": 0.24392719566822052, "epoch": 2.7026194144838214, "grad_norm": 0.5733761787414551, "learning_rate": 5.383486761135926e-07, "loss": 0.2271, "mean_token_accuracy": 0.9413121342658997, "num_tokens": 84211267.0, "step": 1754 }, { "entropy": 0.2313271090388298, "epoch": 2.704160246533128, "grad_norm": 0.5617684721946716, "learning_rate": 5.328626913347656e-07, "loss": 0.2085, "mean_token_accuracy": 0.9439406096935272, "num_tokens": 84266938.0, "step": 1755 }, { "entropy": 0.23264553397893906, "epoch": 2.7057010785824342, "grad_norm": 0.6086382269859314, "learning_rate": 5.274040368619648e-07, "loss": 0.2101, "mean_token_accuracy": 0.9439109563827515, "num_tokens": 84321479.0, "step": 1756 }, { "entropy": 0.24136173725128174, "epoch": 2.7072419106317414, "grad_norm": 0.5760769248008728, "learning_rate": 5.219727284535625e-07, "loss": 0.2193, "mean_token_accuracy": 0.9409843981266022, "num_tokens": 84374377.0, "step": 1757 }, { "entropy": 0.243970587849617, "epoch": 2.7087827426810476, "grad_norm": 0.6367390155792236, "learning_rate": 5.165687817889919e-07, "loss": 0.2242, "mean_token_accuracy": 0.9394833147525787, "num_tokens": 84422713.0, "step": 1758 }, { "entropy": 0.7259567975997925, "epoch": 2.7103235747303542, "grad_norm": 0.9312436580657959, "learning_rate": 5.111922124686919e-07, "loss": 0.7355, "mean_token_accuracy": 0.7991748154163361, "num_tokens": 84434677.0, "step": 1759 }, { "entropy": 0.6972218751907349, "epoch": 2.711864406779661, "grad_norm": 1.1447217464447021, "learning_rate": 5.058430360140676e-07, "loss": 0.6695, "mean_token_accuracy": 0.8162078559398651, "num_tokens": 84443994.0, "step": 1760 }, { "entropy": 0.3889486566185951, "epoch": 2.7134052388289676, "grad_norm": 0.5358933210372925, "learning_rate": 5.00521267867451e-07, "loss": 0.1988, "mean_token_accuracy": 0.8977777063846588, "num_tokens": 84480388.0, "step": 1761 }, { "entropy": 0.2232784479856491, "epoch": 2.714946070878274, "grad_norm": 0.46755993366241455, "learning_rate": 4.952269233920404e-07, "loss": 0.2085, "mean_token_accuracy": 0.9442306756973267, "num_tokens": 84545141.0, "step": 1762 }, { "entropy": 0.26034878194332123, "epoch": 2.716486902927581, "grad_norm": 0.568641722202301, "learning_rate": 4.899600178718722e-07, "loss": 0.2456, "mean_token_accuracy": 0.9323916435241699, "num_tokens": 84607134.0, "step": 1763 }, { "entropy": 0.27605627477169037, "epoch": 2.7180277349768875, "grad_norm": 0.5931583642959595, "learning_rate": 4.84720566511766e-07, "loss": 0.258, "mean_token_accuracy": 0.9305887818336487, "num_tokens": 84667453.0, "step": 1764 }, { "entropy": 0.22657682746648788, "epoch": 2.719568567026194, "grad_norm": 0.5558061599731445, "learning_rate": 4.795085844372927e-07, "loss": 0.2044, "mean_token_accuracy": 0.9447420835494995, "num_tokens": 84726359.0, "step": 1765 }, { "entropy": 0.24315686523914337, "epoch": 2.721109399075501, "grad_norm": 0.5873855352401733, "learning_rate": 4.7432408669471517e-07, "loss": 0.2272, "mean_token_accuracy": 0.9379754066467285, "num_tokens": 84784228.0, "step": 1766 }, { "entropy": 0.24541175365447998, "epoch": 2.7226502311248075, "grad_norm": 0.5791314244270325, "learning_rate": 4.6916708825095467e-07, "loss": 0.2259, "mean_token_accuracy": 0.9409752190113068, "num_tokens": 84840933.0, "step": 1767 }, { "entropy": 0.25566938519477844, "epoch": 2.724191063174114, "grad_norm": 0.6120162606239319, "learning_rate": 4.64037603993549e-07, "loss": 0.2384, "mean_token_accuracy": 0.9378076791763306, "num_tokens": 84896607.0, "step": 1768 }, { "entropy": 0.2608935236930847, "epoch": 2.7257318952234204, "grad_norm": 0.6048621535301208, "learning_rate": 4.5893564873060514e-07, "loss": 0.2412, "mean_token_accuracy": 0.936864048242569, "num_tokens": 84951155.0, "step": 1769 }, { "entropy": 0.2432524710893631, "epoch": 2.7272727272727275, "grad_norm": 0.6126618981361389, "learning_rate": 4.538612371907547e-07, "loss": 0.224, "mean_token_accuracy": 0.9417798817157745, "num_tokens": 85004088.0, "step": 1770 }, { "entropy": 0.30137620866298676, "epoch": 2.7288135593220337, "grad_norm": 0.6704943180084229, "learning_rate": 4.488143840231174e-07, "loss": 0.2265, "mean_token_accuracy": 0.9362540543079376, "num_tokens": 85048129.0, "step": 1771 }, { "entropy": 0.774356335401535, "epoch": 2.7303543913713404, "grad_norm": 1.015096664428711, "learning_rate": 4.4379510379725543e-07, "loss": 0.7836, "mean_token_accuracy": 0.7838968932628632, "num_tokens": 85059958.0, "step": 1772 }, { "entropy": 0.6917970776557922, "epoch": 2.731895223420647, "grad_norm": 1.2822597026824951, "learning_rate": 4.388034110031303e-07, "loss": 0.6778, "mean_token_accuracy": 0.8205705285072327, "num_tokens": 85068354.0, "step": 1773 }, { "entropy": 0.17790217697620392, "epoch": 2.7334360554699537, "grad_norm": 0.4204992949962616, "learning_rate": 4.3383932005106486e-07, "loss": 0.1581, "mean_token_accuracy": 0.9564289450645447, "num_tokens": 85133816.0, "step": 1774 }, { "entropy": 0.24884001165628433, "epoch": 2.7349768875192604, "grad_norm": 0.5733755826950073, "learning_rate": 4.28902845271697e-07, "loss": 0.2335, "mean_token_accuracy": 0.936320573091507, "num_tokens": 85196761.0, "step": 1775 }, { "entropy": 0.25355830788612366, "epoch": 2.736517719568567, "grad_norm": 0.5826796293258667, "learning_rate": 4.2399400091594154e-07, "loss": 0.2415, "mean_token_accuracy": 0.9359521865844727, "num_tokens": 85257638.0, "step": 1776 }, { "entropy": 0.2601975053548813, "epoch": 2.7380585516178737, "grad_norm": 0.5810043811798096, "learning_rate": 4.1911280115494726e-07, "loss": 0.2421, "mean_token_accuracy": 0.9346910417079926, "num_tokens": 85317089.0, "step": 1777 }, { "entropy": 0.2513424754142761, "epoch": 2.7395993836671804, "grad_norm": 0.5819321870803833, "learning_rate": 4.1425926008005877e-07, "loss": 0.2274, "mean_token_accuracy": 0.9397516250610352, "num_tokens": 85375040.0, "step": 1778 }, { "entropy": 0.2290833443403244, "epoch": 2.741140215716487, "grad_norm": 0.538281261920929, "learning_rate": 4.0943339170277265e-07, "loss": 0.206, "mean_token_accuracy": 0.9445286691188812, "num_tokens": 85431835.0, "step": 1779 }, { "entropy": 0.22881657630205154, "epoch": 2.7426810477657937, "grad_norm": 0.6062558889389038, "learning_rate": 4.0463520995469575e-07, "loss": 0.2093, "mean_token_accuracy": 0.9457751214504242, "num_tokens": 85487519.0, "step": 1780 }, { "entropy": 0.24028239399194717, "epoch": 2.7442218798151004, "grad_norm": 0.6191934943199158, "learning_rate": 3.998647286875146e-07, "loss": 0.2214, "mean_token_accuracy": 0.9413374960422516, "num_tokens": 85542094.0, "step": 1781 }, { "entropy": 0.22394400089979172, "epoch": 2.7457627118644066, "grad_norm": 0.5988538861274719, "learning_rate": 3.9512196167294293e-07, "loss": 0.2078, "mean_token_accuracy": 0.9459616541862488, "num_tokens": 85594892.0, "step": 1782 }, { "entropy": 0.23484297096729279, "epoch": 2.7473035439137137, "grad_norm": 0.6519277691841125, "learning_rate": 3.904069226026852e-07, "loss": 0.2165, "mean_token_accuracy": 0.9433141052722931, "num_tokens": 85645017.0, "step": 1783 }, { "entropy": 0.7283403873443604, "epoch": 2.74884437596302, "grad_norm": 0.8996860384941101, "learning_rate": 3.857196250884021e-07, "loss": 0.7164, "mean_token_accuracy": 0.7997124493122101, "num_tokens": 85658957.0, "step": 1784 }, { "entropy": 0.691051185131073, "epoch": 2.7503852080123266, "grad_norm": 1.0835669040679932, "learning_rate": 3.810600826616706e-07, "loss": 0.6693, "mean_token_accuracy": 0.8262257874011993, "num_tokens": 85668477.0, "step": 1785 }, { "entropy": 0.35412753373384476, "epoch": 2.751926040061633, "grad_norm": 0.5555875897407532, "learning_rate": 3.7642830877394067e-07, "loss": 0.1766, "mean_token_accuracy": 0.9121626615524292, "num_tokens": 85704957.0, "step": 1786 }, { "entropy": 0.21338580548763275, "epoch": 2.75346687211094, "grad_norm": 0.5053078532218933, "learning_rate": 3.718243167964974e-07, "loss": 0.1944, "mean_token_accuracy": 0.9471817910671234, "num_tokens": 85769188.0, "step": 1787 }, { "entropy": 0.24695396423339844, "epoch": 2.7550077041602465, "grad_norm": 0.5581739544868469, "learning_rate": 3.6724812002042145e-07, "loss": 0.2397, "mean_token_accuracy": 0.9360271990299225, "num_tokens": 85830698.0, "step": 1788 }, { "entropy": 0.22890283167362213, "epoch": 2.756548536209553, "grad_norm": 0.534565806388855, "learning_rate": 3.626997316565584e-07, "loss": 0.2152, "mean_token_accuracy": 0.9431581795215607, "num_tokens": 85890562.0, "step": 1789 }, { "entropy": 0.266547828912735, "epoch": 2.75808936825886, "grad_norm": 0.6342723369598389, "learning_rate": 3.5817916483547045e-07, "loss": 0.2523, "mean_token_accuracy": 0.9327424168586731, "num_tokens": 85949120.0, "step": 1790 }, { "entropy": 0.25985246896743774, "epoch": 2.7596302003081665, "grad_norm": 0.5999288558959961, "learning_rate": 3.53686432607403e-07, "loss": 0.2416, "mean_token_accuracy": 0.9376472532749176, "num_tokens": 86006565.0, "step": 1791 }, { "entropy": 0.26401883363723755, "epoch": 2.761171032357473, "grad_norm": 0.6171002388000488, "learning_rate": 3.4922154794225004e-07, "loss": 0.246, "mean_token_accuracy": 0.9358735680580139, "num_tokens": 86063021.0, "step": 1792 }, { "entropy": 0.2620973289012909, "epoch": 2.7627118644067794, "grad_norm": 0.6453427076339722, "learning_rate": 3.4478452372950886e-07, "loss": 0.2352, "mean_token_accuracy": 0.9385650753974915, "num_tokens": 86118383.0, "step": 1793 }, { "entropy": 0.28122495114803314, "epoch": 2.7642526964560865, "grad_norm": 0.6764447689056396, "learning_rate": 3.4037537277825215e-07, "loss": 0.2588, "mean_token_accuracy": 0.9327581226825714, "num_tokens": 86172532.0, "step": 1794 }, { "entropy": 0.22514335811138153, "epoch": 2.7657935285053927, "grad_norm": 0.5706218481063843, "learning_rate": 3.3599410781708255e-07, "loss": 0.2028, "mean_token_accuracy": 0.9468427896499634, "num_tokens": 86224755.0, "step": 1795 }, { "entropy": 0.4913194179534912, "epoch": 2.7673343605546994, "grad_norm": 0.7127416729927063, "learning_rate": 3.3164074149410143e-07, "loss": 0.3189, "mean_token_accuracy": 0.8653839528560638, "num_tokens": 86255960.0, "step": 1796 }, { "entropy": 0.7059801518917084, "epoch": 2.768875192604006, "grad_norm": 1.1203405857086182, "learning_rate": 3.273152863768747e-07, "loss": 0.6903, "mean_token_accuracy": 0.8142432272434235, "num_tokens": 86266887.0, "step": 1797 }, { "entropy": 0.5995681583881378, "epoch": 2.7704160246533127, "grad_norm": 1.3747857809066772, "learning_rate": 3.2301775495238277e-07, "loss": 0.5841, "mean_token_accuracy": 0.850159615278244, "num_tokens": 86274944.0, "step": 1798 }, { "entropy": 0.17438115924596786, "epoch": 2.7719568567026194, "grad_norm": 0.44927504658699036, "learning_rate": 3.1874815962700254e-07, "loss": 0.1628, "mean_token_accuracy": 0.9563201069831848, "num_tokens": 86340361.0, "step": 1799 }, { "entropy": 0.26155252754688263, "epoch": 2.773497688751926, "grad_norm": 0.5286547541618347, "learning_rate": 3.1450651272645905e-07, "loss": 0.2425, "mean_token_accuracy": 0.93459752202034, "num_tokens": 86403314.0, "step": 1800 }, { "epoch": 2.773497688751926, "eval_entropy": 0.376958379100235, "eval_loss": 0.3802676200866699, "eval_mean_token_accuracy": 0.9015121547118673, "eval_num_tokens": 86403314.0, "eval_runtime": 192.1819, "eval_samples_per_second": 4.412, "eval_steps_per_second": 1.103, "step": 1800 }, { "entropy": 0.26145754754543304, "epoch": 2.7750385208012327, "grad_norm": 0.5491335988044739, "learning_rate": 3.102928264957994e-07, "loss": 0.2462, "mean_token_accuracy": 0.9326633214950562, "num_tokens": 86464230.0, "step": 1801 }, { "entropy": 0.24571394175291061, "epoch": 2.7765793528505394, "grad_norm": 0.5322064161300659, "learning_rate": 3.0610711309934647e-07, "loss": 0.2235, "mean_token_accuracy": 0.9398948848247528, "num_tokens": 86523377.0, "step": 1802 }, { "entropy": 0.2483087107539177, "epoch": 2.778120184899846, "grad_norm": 0.6105037927627563, "learning_rate": 3.0194938462067e-07, "loss": 0.2316, "mean_token_accuracy": 0.9376359283924103, "num_tokens": 86581139.0, "step": 1803 }, { "entropy": 0.24025192856788635, "epoch": 2.7796610169491527, "grad_norm": 0.5946200489997864, "learning_rate": 2.9781965306255324e-07, "loss": 0.2207, "mean_token_accuracy": 0.9410625100135803, "num_tokens": 86637771.0, "step": 1804 }, { "entropy": 0.23250900954008102, "epoch": 2.7812018489984593, "grad_norm": 0.5691198706626892, "learning_rate": 2.9371793034695726e-07, "loss": 0.2163, "mean_token_accuracy": 0.9431920349597931, "num_tokens": 86693416.0, "step": 1805 }, { "entropy": 0.24413946270942688, "epoch": 2.7827426810477656, "grad_norm": 0.6086665987968445, "learning_rate": 2.8964422831498363e-07, "loss": 0.2229, "mean_token_accuracy": 0.941428154706955, "num_tokens": 86747838.0, "step": 1806 }, { "entropy": 0.24999262392520905, "epoch": 2.7842835130970727, "grad_norm": 0.6491571664810181, "learning_rate": 2.855985587268428e-07, "loss": 0.2281, "mean_token_accuracy": 0.9404013752937317, "num_tokens": 86800630.0, "step": 1807 }, { "entropy": 0.2150523066520691, "epoch": 2.785824345146379, "grad_norm": 0.5805530548095703, "learning_rate": 2.8158093326181577e-07, "loss": 0.201, "mean_token_accuracy": 0.9478375613689423, "num_tokens": 86850566.0, "step": 1808 }, { "entropy": 0.7322630882263184, "epoch": 2.7873651771956856, "grad_norm": 0.9141660332679749, "learning_rate": 2.7759136351823033e-07, "loss": 0.7486, "mean_token_accuracy": 0.7953679263591766, "num_tokens": 86862841.0, "step": 1809 }, { "entropy": 0.6772502958774567, "epoch": 2.788906009244992, "grad_norm": 1.2626620531082153, "learning_rate": 2.7362986101341803e-07, "loss": 0.6458, "mean_token_accuracy": 0.8290519118309021, "num_tokens": 86872067.0, "step": 1810 }, { "entropy": 0.36897940933704376, "epoch": 2.790446841294299, "grad_norm": 0.5323114395141602, "learning_rate": 2.696964371836841e-07, "loss": 0.1795, "mean_token_accuracy": 0.9048473238945007, "num_tokens": 86908485.0, "step": 1811 }, { "entropy": 0.21528515219688416, "epoch": 2.7919876733436055, "grad_norm": 0.5447700023651123, "learning_rate": 2.6579110338427416e-07, "loss": 0.1969, "mean_token_accuracy": 0.9468072354793549, "num_tokens": 86973135.0, "step": 1812 }, { "entropy": 0.26428667455911636, "epoch": 2.793528505392912, "grad_norm": 0.6150845289230347, "learning_rate": 2.6191387088934426e-07, "loss": 0.2479, "mean_token_accuracy": 0.933162122964859, "num_tokens": 87035075.0, "step": 1813 }, { "entropy": 0.26965518295764923, "epoch": 2.795069337442219, "grad_norm": 0.5836346745491028, "learning_rate": 2.580647508919232e-07, "loss": 0.2539, "mean_token_accuracy": 0.9340353906154633, "num_tokens": 87095199.0, "step": 1814 }, { "entropy": 0.2657649740576744, "epoch": 2.7966101694915255, "grad_norm": 0.5993187427520752, "learning_rate": 2.542437545038845e-07, "loss": 0.2474, "mean_token_accuracy": 0.9347180724143982, "num_tokens": 87153818.0, "step": 1815 }, { "entropy": 0.25718512386083603, "epoch": 2.798151001540832, "grad_norm": 0.6187759637832642, "learning_rate": 2.504508927559113e-07, "loss": 0.2382, "mean_token_accuracy": 0.9381674826145172, "num_tokens": 87211333.0, "step": 1816 }, { "entropy": 0.2776934802532196, "epoch": 2.7996918335901384, "grad_norm": 0.6109705567359924, "learning_rate": 2.4668617659746707e-07, "loss": 0.258, "mean_token_accuracy": 0.9325273334980011, "num_tokens": 87267814.0, "step": 1817 }, { "entropy": 0.2373834177851677, "epoch": 2.8012326656394455, "grad_norm": 0.5786963105201721, "learning_rate": 2.4294961689676487e-07, "loss": 0.2215, "mean_token_accuracy": 0.9423523247241974, "num_tokens": 87323169.0, "step": 1818 }, { "entropy": 0.23898613452911377, "epoch": 2.8027734976887517, "grad_norm": 0.6233100891113281, "learning_rate": 2.392412244407294e-07, "loss": 0.219, "mean_token_accuracy": 0.942132830619812, "num_tokens": 87376901.0, "step": 1819 }, { "entropy": 0.23915192484855652, "epoch": 2.8043143297380584, "grad_norm": 0.5856971740722656, "learning_rate": 2.3556100993497255e-07, "loss": 0.2181, "mean_token_accuracy": 0.9436657428741455, "num_tokens": 87428884.0, "step": 1820 }, { "entropy": 0.43303021788597107, "epoch": 2.805855161787365, "grad_norm": 0.7108818292617798, "learning_rate": 2.319089840037614e-07, "loss": 0.2918, "mean_token_accuracy": 0.896206796169281, "num_tokens": 87463492.0, "step": 1821 }, { "entropy": 0.6812084019184113, "epoch": 2.8073959938366717, "grad_norm": 1.0441434383392334, "learning_rate": 2.282851571899869e-07, "loss": 0.674, "mean_token_accuracy": 0.8143326640129089, "num_tokens": 87474483.0, "step": 1822 }, { "entropy": 0.6394663155078888, "epoch": 2.8089368258859784, "grad_norm": 1.1900475025177002, "learning_rate": 2.24689539955133e-07, "loss": 0.6078, "mean_token_accuracy": 0.8410816192626953, "num_tokens": 87482692.0, "step": 1823 }, { "entropy": 0.1645928919315338, "epoch": 2.810477657935285, "grad_norm": 0.42593640089035034, "learning_rate": 2.2112214267924093e-07, "loss": 0.1468, "mean_token_accuracy": 0.9601805210113525, "num_tokens": 87548143.0, "step": 1824 }, { "entropy": 0.258357509970665, "epoch": 2.8120184899845917, "grad_norm": 0.5449671149253845, "learning_rate": 2.1758297566089382e-07, "loss": 0.2432, "mean_token_accuracy": 0.9336381554603577, "num_tokens": 87611104.0, "step": 1825 }, { "entropy": 0.24557043612003326, "epoch": 2.8135593220338984, "grad_norm": 0.5442591309547424, "learning_rate": 2.1407204911717105e-07, "loss": 0.2305, "mean_token_accuracy": 0.9374613761901855, "num_tokens": 87671872.0, "step": 1826 }, { "entropy": 0.2492477372288704, "epoch": 2.815100154083205, "grad_norm": 0.5654980540275574, "learning_rate": 2.1058937318363059e-07, "loss": 0.2347, "mean_token_accuracy": 0.9362650513648987, "num_tokens": 87731251.0, "step": 1827 }, { "entropy": 0.2608130872249603, "epoch": 2.8166409861325117, "grad_norm": 0.5570374131202698, "learning_rate": 2.0713495791427006e-07, "loss": 0.2422, "mean_token_accuracy": 0.9371358156204224, "num_tokens": 87789294.0, "step": 1828 }, { "entropy": 0.26433993875980377, "epoch": 2.8181818181818183, "grad_norm": 0.605560839176178, "learning_rate": 2.0370881328150682e-07, "loss": 0.2486, "mean_token_accuracy": 0.9345421195030212, "num_tokens": 87846349.0, "step": 1829 }, { "entropy": 0.2453429400920868, "epoch": 2.8197226502311246, "grad_norm": 0.5925305485725403, "learning_rate": 2.0031094917614235e-07, "loss": 0.23, "mean_token_accuracy": 0.9401512444019318, "num_tokens": 87902329.0, "step": 1830 }, { "entropy": 0.2733774185180664, "epoch": 2.8212634822804317, "grad_norm": 0.6067247986793518, "learning_rate": 1.9694137540733682e-07, "loss": 0.2561, "mean_token_accuracy": 0.9338645339012146, "num_tokens": 87957322.0, "step": 1831 }, { "entropy": 0.21875058114528656, "epoch": 2.822804314329738, "grad_norm": 0.6225065588951111, "learning_rate": 1.9360010170257903e-07, "loss": 0.1971, "mean_token_accuracy": 0.9484537839889526, "num_tokens": 88011073.0, "step": 1832 }, { "entropy": 0.23803409188985825, "epoch": 2.8243451463790445, "grad_norm": 0.6153862476348877, "learning_rate": 1.9028713770766205e-07, "loss": 0.2197, "mean_token_accuracy": 0.9425828158855438, "num_tokens": 88062311.0, "step": 1833 }, { "entropy": 0.6820586919784546, "epoch": 2.825885978428351, "grad_norm": 1.0285003185272217, "learning_rate": 1.8700249298665096e-07, "loss": 0.5734, "mean_token_accuracy": 0.8344188630580902, "num_tokens": 88078115.0, "step": 1834 }, { "entropy": 0.6986259818077087, "epoch": 2.827426810477658, "grad_norm": 1.0879470109939575, "learning_rate": 1.8374617702185627e-07, "loss": 0.6711, "mean_token_accuracy": 0.8229444921016693, "num_tokens": 88087522.0, "step": 1835 }, { "entropy": 0.38869892060756683, "epoch": 2.8289676425269645, "grad_norm": 0.5157177448272705, "learning_rate": 1.8051819921380942e-07, "loss": 0.1706, "mean_token_accuracy": 0.8999703824520111, "num_tokens": 88123883.0, "step": 1836 }, { "entropy": 0.22053170204162598, "epoch": 2.830508474576271, "grad_norm": 0.46418797969818115, "learning_rate": 1.7731856888123068e-07, "loss": 0.2033, "mean_token_accuracy": 0.9448665380477905, "num_tokens": 88188072.0, "step": 1837 }, { "entropy": 0.2943509370088577, "epoch": 2.832049306625578, "grad_norm": 0.6046449542045593, "learning_rate": 1.7414729526101015e-07, "loss": 0.279, "mean_token_accuracy": 0.9245781600475311, "num_tokens": 88249469.0, "step": 1838 }, { "entropy": 0.24692001938819885, "epoch": 2.8335901386748845, "grad_norm": 0.570534884929657, "learning_rate": 1.7100438750817018e-07, "loss": 0.2339, "mean_token_accuracy": 0.9380026757717133, "num_tokens": 88309027.0, "step": 1839 }, { "entropy": 0.2606980800628662, "epoch": 2.835130970724191, "grad_norm": 0.6054999828338623, "learning_rate": 1.678898546958485e-07, "loss": 0.2417, "mean_token_accuracy": 0.9352309107780457, "num_tokens": 88367275.0, "step": 1840 }, { "entropy": 0.2505849003791809, "epoch": 2.836671802773498, "grad_norm": 0.5535808205604553, "learning_rate": 1.6480370581526629e-07, "loss": 0.2301, "mean_token_accuracy": 0.9382922053337097, "num_tokens": 88424541.0, "step": 1841 }, { "entropy": 0.24715616554021835, "epoch": 2.8382126348228045, "grad_norm": 0.6135382652282715, "learning_rate": 1.617459497757068e-07, "loss": 0.229, "mean_token_accuracy": 0.9403978288173676, "num_tokens": 88480774.0, "step": 1842 }, { "entropy": 0.2590566575527191, "epoch": 2.8397534668721107, "grad_norm": 0.604911208152771, "learning_rate": 1.5871659540448448e-07, "loss": 0.2421, "mean_token_accuracy": 0.9363018870353699, "num_tokens": 88535918.0, "step": 1843 }, { "entropy": 0.2313525602221489, "epoch": 2.841294298921418, "grad_norm": 0.5814886093139648, "learning_rate": 1.557156514469238e-07, "loss": 0.2123, "mean_token_accuracy": 0.9438813924789429, "num_tokens": 88589665.0, "step": 1844 }, { "entropy": 0.24521317332983017, "epoch": 2.842835130970724, "grad_norm": 0.6069482564926147, "learning_rate": 1.527431265663304e-07, "loss": 0.2233, "mean_token_accuracy": 0.9416478276252747, "num_tokens": 88641436.0, "step": 1845 }, { "entropy": 0.5723627209663391, "epoch": 2.8443759630200307, "grad_norm": 0.7586327195167542, "learning_rate": 1.4979902934397106e-07, "loss": 0.4077, "mean_token_accuracy": 0.849659651517868, "num_tokens": 88668031.0, "step": 1846 }, { "entropy": 0.7314127385616302, "epoch": 2.8459167950693374, "grad_norm": 1.116696834564209, "learning_rate": 1.468833682790427e-07, "loss": 0.7087, "mean_token_accuracy": 0.809020608663559, "num_tokens": 88678564.0, "step": 1847 }, { "entropy": 0.6085841655731201, "epoch": 2.847457627118644, "grad_norm": 1.3344122171401978, "learning_rate": 1.4399615178865457e-07, "loss": 0.5934, "mean_token_accuracy": 0.8431450724601746, "num_tokens": 88686462.0, "step": 1848 }, { "entropy": 0.1716984361410141, "epoch": 2.8489984591679507, "grad_norm": 0.4527837336063385, "learning_rate": 1.4113738820779598e-07, "loss": 0.156, "mean_token_accuracy": 0.957435131072998, "num_tokens": 88751684.0, "step": 1849 }, { "entropy": 0.25488555431365967, "epoch": 2.8505392912172574, "grad_norm": 0.5517792105674744, "learning_rate": 1.3830708578931984e-07, "loss": 0.2437, "mean_token_accuracy": 0.9336680173873901, "num_tokens": 88814241.0, "step": 1850 }, { "entropy": 0.2673380821943283, "epoch": 2.852080123266564, "grad_norm": 0.5604245662689209, "learning_rate": 1.3550525270391578e-07, "loss": 0.2538, "mean_token_accuracy": 0.9339485168457031, "num_tokens": 88874825.0, "step": 1851 }, { "entropy": 0.25461234897375107, "epoch": 2.8536209553158707, "grad_norm": 0.5861278176307678, "learning_rate": 1.3273189704008371e-07, "loss": 0.2326, "mean_token_accuracy": 0.9375563263893127, "num_tokens": 88934087.0, "step": 1852 }, { "entropy": 0.24759557098150253, "epoch": 2.8551617873651773, "grad_norm": 0.56315016746521, "learning_rate": 1.2998702680411702e-07, "loss": 0.2281, "mean_token_accuracy": 0.9393981695175171, "num_tokens": 88992141.0, "step": 1853 }, { "entropy": 0.24846789985895157, "epoch": 2.8567026194144836, "grad_norm": 0.5648965239524841, "learning_rate": 1.2727064992007154e-07, "loss": 0.2242, "mean_token_accuracy": 0.9400011897087097, "num_tokens": 89049078.0, "step": 1854 }, { "entropy": 0.2557980716228485, "epoch": 2.8582434514637907, "grad_norm": 0.5934321880340576, "learning_rate": 1.2458277422974895e-07, "loss": 0.233, "mean_token_accuracy": 0.9389745593070984, "num_tokens": 89104750.0, "step": 1855 }, { "entropy": 0.22746802121400833, "epoch": 2.859784283513097, "grad_norm": 0.6075649261474609, "learning_rate": 1.219234074926734e-07, "loss": 0.2137, "mean_token_accuracy": 0.9444398880004883, "num_tokens": 89159215.0, "step": 1856 }, { "entropy": 0.25382618606090546, "epoch": 2.8613251155624035, "grad_norm": 0.6322683691978455, "learning_rate": 1.192925573860637e-07, "loss": 0.2338, "mean_token_accuracy": 0.9395424425601959, "num_tokens": 89212226.0, "step": 1857 }, { "entropy": 0.2413768172264099, "epoch": 2.86286594761171, "grad_norm": 0.6692529916763306, "learning_rate": 1.1669023150481906e-07, "loss": 0.2205, "mean_token_accuracy": 0.9427884817123413, "num_tokens": 89262695.0, "step": 1858 }, { "entropy": 0.6754532754421234, "epoch": 2.864406779661017, "grad_norm": 0.9175548553466797, "learning_rate": 1.1411643736149336e-07, "loss": 0.6655, "mean_token_accuracy": 0.8178496956825256, "num_tokens": 89276369.0, "step": 1859 }, { "entropy": 0.6442925035953522, "epoch": 2.8659476117103235, "grad_norm": 1.1688687801361084, "learning_rate": 1.1157118238626752e-07, "loss": 0.6128, "mean_token_accuracy": 0.8373392820358276, "num_tokens": 89285292.0, "step": 1860 }, { "entropy": 0.3336997888982296, "epoch": 2.86748844375963, "grad_norm": 0.448986679315567, "learning_rate": 1.0905447392694058e-07, "loss": 0.1371, "mean_token_accuracy": 0.9106951951980591, "num_tokens": 89321391.0, "step": 1861 }, { "entropy": 0.25814466178417206, "epoch": 2.869029275808937, "grad_norm": 0.5221923589706421, "learning_rate": 1.0656631924889749e-07, "loss": 0.2423, "mean_token_accuracy": 0.9343613684177399, "num_tokens": 89385510.0, "step": 1862 }, { "entropy": 0.2599526643753052, "epoch": 2.8705701078582435, "grad_norm": 0.5753408074378967, "learning_rate": 1.0410672553509693e-07, "loss": 0.2482, "mean_token_accuracy": 0.9337214231491089, "num_tokens": 89446943.0, "step": 1863 }, { "entropy": 0.24414828419685364, "epoch": 2.87211093990755, "grad_norm": 0.5700128078460693, "learning_rate": 1.016756998860402e-07, "loss": 0.2218, "mean_token_accuracy": 0.9394818842411041, "num_tokens": 89506741.0, "step": 1864 }, { "entropy": 0.24816292524337769, "epoch": 2.873651771956857, "grad_norm": 0.5681974291801453, "learning_rate": 9.927324931976012e-08, "loss": 0.2279, "mean_token_accuracy": 0.9388836920261383, "num_tokens": 89565223.0, "step": 1865 }, { "entropy": 0.23571822047233582, "epoch": 2.8751926040061635, "grad_norm": 0.5685381293296814, "learning_rate": 9.689938077179883e-08, "loss": 0.2184, "mean_token_accuracy": 0.9409510493278503, "num_tokens": 89622396.0, "step": 1866 }, { "entropy": 0.24782782047986984, "epoch": 2.8767334360554697, "grad_norm": 0.6174454689025879, "learning_rate": 9.455410109518116e-08, "loss": 0.2275, "mean_token_accuracy": 0.9403086006641388, "num_tokens": 89678189.0, "step": 1867 }, { "entropy": 0.2325315624475479, "epoch": 2.878274268104777, "grad_norm": 0.6042900085449219, "learning_rate": 9.223741706040456e-08, "loss": 0.2126, "mean_token_accuracy": 0.9441110491752625, "num_tokens": 89732832.0, "step": 1868 }, { "entropy": 0.25510983169078827, "epoch": 2.879815100154083, "grad_norm": 0.6471202969551086, "learning_rate": 8.994933535541151e-08, "loss": 0.235, "mean_token_accuracy": 0.9372577965259552, "num_tokens": 89785899.0, "step": 1869 }, { "entropy": 0.2209894061088562, "epoch": 2.8813559322033897, "grad_norm": 0.5568343997001648, "learning_rate": 8.7689862585576e-08, "loss": 0.1973, "mean_token_accuracy": 0.9486824572086334, "num_tokens": 89836406.0, "step": 1870 }, { "entropy": 0.6428464651107788, "epoch": 2.8828967642526964, "grad_norm": 0.7991012930870056, "learning_rate": 8.545900527368146e-08, "loss": 0.527, "mean_token_accuracy": 0.8371274173259735, "num_tokens": 89854677.0, "step": 1871 }, { "entropy": 0.6634705364704132, "epoch": 2.884437596302003, "grad_norm": 1.0526570081710815, "learning_rate": 8.325676985990073e-08, "loss": 0.6239, "mean_token_accuracy": 0.834095686674118, "num_tokens": 89864710.0, "step": 1872 }, { "entropy": 0.6403382420539856, "epoch": 2.8859784283513097, "grad_norm": 1.3258261680603027, "learning_rate": 8.108316270178162e-08, "loss": 0.633, "mean_token_accuracy": 0.8379532992839813, "num_tokens": 89872397.0, "step": 1873 }, { "entropy": 0.17890149354934692, "epoch": 2.8875192604006163, "grad_norm": 0.4529952108860016, "learning_rate": 7.893819007422476e-08, "loss": 0.1627, "mean_token_accuracy": 0.9562974274158478, "num_tokens": 89937927.0, "step": 1874 }, { "entropy": 0.21776309609413147, "epoch": 2.889060092449923, "grad_norm": 0.5150484442710876, "learning_rate": 7.682185816946686e-08, "loss": 0.198, "mean_token_accuracy": 0.944935530424118, "num_tokens": 90001339.0, "step": 1875 }, { "entropy": 0.2691236734390259, "epoch": 2.8906009244992297, "grad_norm": 0.5945441722869873, "learning_rate": 7.47341730970641e-08, "loss": 0.2555, "mean_token_accuracy": 0.9324967265129089, "num_tokens": 90062153.0, "step": 1876 }, { "entropy": 0.2551061511039734, "epoch": 2.8921417565485363, "grad_norm": 0.5539755821228027, "learning_rate": 7.267514088387107e-08, "loss": 0.2342, "mean_token_accuracy": 0.9358517229557037, "num_tokens": 90121514.0, "step": 1877 }, { "entropy": 0.25509992986917496, "epoch": 2.893682588597843, "grad_norm": 0.6054633259773254, "learning_rate": 7.064476747402626e-08, "loss": 0.2368, "mean_token_accuracy": 0.9374858438968658, "num_tokens": 90179531.0, "step": 1878 }, { "entropy": 0.24346482008695602, "epoch": 2.8952234206471497, "grad_norm": 0.5595784783363342, "learning_rate": 6.864305872893773e-08, "loss": 0.2204, "mean_token_accuracy": 0.941181093454361, "num_tokens": 90236350.0, "step": 1879 }, { "entropy": 0.26041021198034286, "epoch": 2.896764252696456, "grad_norm": 0.6528160572052002, "learning_rate": 6.667002042725745e-08, "loss": 0.2443, "mean_token_accuracy": 0.9356866478919983, "num_tokens": 90291859.0, "step": 1880 }, { "entropy": 0.24197784811258316, "epoch": 2.898305084745763, "grad_norm": 0.6037113070487976, "learning_rate": 6.472565826487142e-08, "loss": 0.2232, "mean_token_accuracy": 0.9414427280426025, "num_tokens": 90345971.0, "step": 1881 }, { "entropy": 0.21962495148181915, "epoch": 2.899845916795069, "grad_norm": 0.5880447030067444, "learning_rate": 6.280997785488407e-08, "loss": 0.2036, "mean_token_accuracy": 0.9465470910072327, "num_tokens": 90398525.0, "step": 1882 }, { "entropy": 0.2979128360748291, "epoch": 2.901386748844376, "grad_norm": 0.6129584908485413, "learning_rate": 6.092298472759716e-08, "loss": 0.2252, "mean_token_accuracy": 0.9365226030349731, "num_tokens": 90440896.0, "step": 1883 }, { "entropy": 0.7383750677108765, "epoch": 2.9029275808936825, "grad_norm": 0.9985675811767578, "learning_rate": 5.9064684330496505e-08, "loss": 0.7341, "mean_token_accuracy": 0.795375406742096, "num_tokens": 90452811.0, "step": 1884 }, { "entropy": 0.6677963733673096, "epoch": 2.904468412942989, "grad_norm": 1.0567837953567505, "learning_rate": 5.7235082028235243e-08, "loss": 0.6365, "mean_token_accuracy": 0.8310219347476959, "num_tokens": 90462021.0, "step": 1885 }, { "entropy": 0.3611072488129139, "epoch": 2.906009244992296, "grad_norm": 0.48863357305526733, "learning_rate": 5.5434183102622827e-08, "loss": 0.1488, "mean_token_accuracy": 0.9107751548290253, "num_tokens": 90498346.0, "step": 1886 }, { "entropy": 0.22675172984600067, "epoch": 2.9075500770416025, "grad_norm": 0.5303062796592712, "learning_rate": 5.366199275260164e-08, "loss": 0.2126, "mean_token_accuracy": 0.9419028162956238, "num_tokens": 90562832.0, "step": 1887 }, { "entropy": 0.26351960003376007, "epoch": 2.909090909090909, "grad_norm": 0.5480736494064331, "learning_rate": 5.191851609423926e-08, "loss": 0.2464, "mean_token_accuracy": 0.933190256357193, "num_tokens": 90624426.0, "step": 1888 }, { "entropy": 0.2577715367078781, "epoch": 2.910631741140216, "grad_norm": 0.5512451529502869, "learning_rate": 5.020375816070844e-08, "loss": 0.2449, "mean_token_accuracy": 0.935043066740036, "num_tokens": 90684417.0, "step": 1889 }, { "entropy": 0.2424110770225525, "epoch": 2.9121725731895225, "grad_norm": 0.5736238956451416, "learning_rate": 4.851772390227827e-08, "loss": 0.2251, "mean_token_accuracy": 0.9398631751537323, "num_tokens": 90742956.0, "step": 1890 }, { "entropy": 0.26667867600917816, "epoch": 2.9137134052388287, "grad_norm": 0.5649935007095337, "learning_rate": 4.6860418186291945e-08, "loss": 0.247, "mean_token_accuracy": 0.9349241554737091, "num_tokens": 90800448.0, "step": 1891 }, { "entropy": 0.23199106007814407, "epoch": 2.915254237288136, "grad_norm": 0.5333300828933716, "learning_rate": 4.5231845797163444e-08, "loss": 0.2115, "mean_token_accuracy": 0.9442923367023468, "num_tokens": 90856862.0, "step": 1892 }, { "entropy": 0.24683550000190735, "epoch": 2.916795069337442, "grad_norm": 0.5959057807922363, "learning_rate": 4.3632011436351984e-08, "loss": 0.2348, "mean_token_accuracy": 0.9395556449890137, "num_tokens": 90912302.0, "step": 1893 }, { "entropy": 0.22807347774505615, "epoch": 2.9183359013867487, "grad_norm": 0.5403938293457031, "learning_rate": 4.2060919722356486e-08, "loss": 0.2052, "mean_token_accuracy": 0.9457634091377258, "num_tokens": 90966462.0, "step": 1894 }, { "entropy": 0.21041585505008698, "epoch": 2.9198767334360554, "grad_norm": 0.5747525691986084, "learning_rate": 4.0518575190701124e-08, "loss": 0.1907, "mean_token_accuracy": 0.9502281248569489, "num_tokens": 91018621.0, "step": 1895 }, { "entropy": 0.5144686698913574, "epoch": 2.921417565485362, "grad_norm": 0.7806263566017151, "learning_rate": 3.900498229391869e-08, "loss": 0.3574, "mean_token_accuracy": 0.8598963022232056, "num_tokens": 91047448.0, "step": 1896 }, { "entropy": 0.7269408702850342, "epoch": 2.9229583975346687, "grad_norm": 1.0266807079315186, "learning_rate": 3.7520145401540585e-08, "loss": 0.6972, "mean_token_accuracy": 0.8104406297206879, "num_tokens": 91057932.0, "step": 1897 }, { "entropy": 0.6216744780540466, "epoch": 2.9244992295839753, "grad_norm": 1.2407283782958984, "learning_rate": 3.606406880008462e-08, "loss": 0.6062, "mean_token_accuracy": 0.8461348414421082, "num_tokens": 91065686.0, "step": 1898 }, { "entropy": 0.17726363986730576, "epoch": 2.926040061633282, "grad_norm": 0.43995338678359985, "learning_rate": 3.463675669304056e-08, "loss": 0.161, "mean_token_accuracy": 0.9564174711704254, "num_tokens": 91131041.0, "step": 1899 }, { "entropy": 0.2492859959602356, "epoch": 2.9275808936825887, "grad_norm": 0.550543487071991, "learning_rate": 3.323821320086018e-08, "loss": 0.2368, "mean_token_accuracy": 0.9356253147125244, "num_tokens": 91193572.0, "step": 1900 }, { "entropy": 0.2755107283592224, "epoch": 2.9291217257318953, "grad_norm": 0.6278766393661499, "learning_rate": 3.1868442360943886e-08, "loss": 0.2608, "mean_token_accuracy": 0.9310534596443176, "num_tokens": 91253914.0, "step": 1901 }, { "entropy": 0.23254124075174332, "epoch": 2.930662557781202, "grad_norm": 0.5361241698265076, "learning_rate": 3.052744812762742e-08, "loss": 0.2167, "mean_token_accuracy": 0.9417022168636322, "num_tokens": 91312811.0, "step": 1902 }, { "entropy": 0.27059856057167053, "epoch": 2.9322033898305087, "grad_norm": 0.588931679725647, "learning_rate": 2.9215234372176327e-08, "loss": 0.2527, "mean_token_accuracy": 0.9331872165203094, "num_tokens": 91370451.0, "step": 1903 }, { "entropy": 0.2548794150352478, "epoch": 2.933744221879815, "grad_norm": 0.6018422842025757, "learning_rate": 2.7931804882767033e-08, "loss": 0.2358, "mean_token_accuracy": 0.9376505017280579, "num_tokens": 91426961.0, "step": 1904 }, { "entropy": 0.25298070907592773, "epoch": 2.935285053929122, "grad_norm": 0.6014219522476196, "learning_rate": 2.667716336448356e-08, "loss": 0.2347, "mean_token_accuracy": 0.939649760723114, "num_tokens": 91482370.0, "step": 1905 }, { "entropy": 0.2329748049378395, "epoch": 2.936825885978428, "grad_norm": 0.5712085962295532, "learning_rate": 2.545131343929974e-08, "loss": 0.2101, "mean_token_accuracy": 0.9439662396907806, "num_tokens": 91536527.0, "step": 1906 }, { "entropy": 0.2213410809636116, "epoch": 2.938366718027735, "grad_norm": 0.6426869630813599, "learning_rate": 2.425425864607367e-08, "loss": 0.2018, "mean_token_accuracy": 0.9454615116119385, "num_tokens": 91589199.0, "step": 1907 }, { "entropy": 0.37572143971920013, "epoch": 2.9399075500770415, "grad_norm": 0.637571394443512, "learning_rate": 2.3086002440535493e-08, "loss": 0.2695, "mean_token_accuracy": 0.9244523346424103, "num_tokens": 91630002.0, "step": 1908 }, { "entropy": 0.7305442094802856, "epoch": 2.941448382126348, "grad_norm": 0.9639086723327637, "learning_rate": 2.194654819527853e-08, "loss": 0.7208, "mean_token_accuracy": 0.8050745129585266, "num_tokens": 91641491.0, "step": 1909 }, { "entropy": 0.6477666795253754, "epoch": 2.942989214175655, "grad_norm": 1.203036904335022, "learning_rate": 2.0835899199748156e-08, "loss": 0.6221, "mean_token_accuracy": 0.8373731672763824, "num_tokens": 91650625.0, "step": 1910 }, { "entropy": 0.36851122975349426, "epoch": 2.9445300462249615, "grad_norm": 0.5714613199234009, "learning_rate": 1.9754058660234053e-08, "loss": 0.1868, "mean_token_accuracy": 0.9093700647354126, "num_tokens": 91686838.0, "step": 1911 }, { "entropy": 0.2334386110305786, "epoch": 2.946070878274268, "grad_norm": 0.5066110491752625, "learning_rate": 1.8701029699857987e-08, "loss": 0.2175, "mean_token_accuracy": 0.9405451416969299, "num_tokens": 91750698.0, "step": 1912 }, { "entropy": 0.26111191511154175, "epoch": 2.947611710323575, "grad_norm": 0.5532861948013306, "learning_rate": 1.7676815358568244e-08, "loss": 0.2448, "mean_token_accuracy": 0.93499955534935, "num_tokens": 91812076.0, "step": 1913 }, { "entropy": 0.25407469272613525, "epoch": 2.9491525423728815, "grad_norm": 0.5516828298568726, "learning_rate": 1.668141859312855e-08, "loss": 0.2395, "mean_token_accuracy": 0.9356538355350494, "num_tokens": 91871732.0, "step": 1914 }, { "entropy": 0.2703789919614792, "epoch": 2.9506933744221877, "grad_norm": 0.6570232510566711, "learning_rate": 1.5714842277111396e-08, "loss": 0.2544, "mean_token_accuracy": 0.9323164224624634, "num_tokens": 91930014.0, "step": 1915 }, { "entropy": 0.26819583773612976, "epoch": 2.952234206471495, "grad_norm": 0.6301263570785522, "learning_rate": 1.477708920088694e-08, "loss": 0.2499, "mean_token_accuracy": 0.9336446523666382, "num_tokens": 91987095.0, "step": 1916 }, { "entropy": 0.28731438517570496, "epoch": 2.953775038520801, "grad_norm": 0.6531516909599304, "learning_rate": 1.3868162071618562e-08, "loss": 0.273, "mean_token_accuracy": 0.927110880613327, "num_tokens": 92043285.0, "step": 1917 }, { "entropy": 0.21292252838611603, "epoch": 2.9553158705701077, "grad_norm": 0.5727885961532593, "learning_rate": 1.298806351325066e-08, "loss": 0.1969, "mean_token_accuracy": 0.9486594796180725, "num_tokens": 92098430.0, "step": 1918 }, { "entropy": 0.25203070044517517, "epoch": 2.9568567026194144, "grad_norm": 0.5724814534187317, "learning_rate": 1.2136796066507527e-08, "loss": 0.235, "mean_token_accuracy": 0.9383587539196014, "num_tokens": 92152251.0, "step": 1919 }, { "entropy": 0.23450759053230286, "epoch": 2.958397534668721, "grad_norm": 0.6066413521766663, "learning_rate": 1.1314362188877825e-08, "loss": 0.2138, "mean_token_accuracy": 0.9445883929729462, "num_tokens": 92204253.0, "step": 1920 }, { "entropy": 0.4577551484107971, "epoch": 2.9599383667180277, "grad_norm": 0.7476376295089722, "learning_rate": 1.0520764254614568e-08, "loss": 0.3323, "mean_token_accuracy": 0.879408061504364, "num_tokens": 92236608.0, "step": 1921 }, { "entropy": 0.7138011455535889, "epoch": 2.9614791987673343, "grad_norm": 1.098020315170288, "learning_rate": 9.75600455472292e-09, "loss": 0.7053, "mean_token_accuracy": 0.8112631142139435, "num_tokens": 92247384.0, "step": 1922 }, { "entropy": 0.6499614417552948, "epoch": 2.963020030816641, "grad_norm": 1.2709518671035767, "learning_rate": 9.02008529695908e-09, "loss": 0.632, "mean_token_accuracy": 0.8359144926071167, "num_tokens": 92255573.0, "step": 1923 }, { "entropy": 0.1541329249739647, "epoch": 2.9645608628659477, "grad_norm": 0.35676565766334534, "learning_rate": 8.313008605819183e-09, "loss": 0.1317, "mean_token_accuracy": 0.9634535908699036, "num_tokens": 92321019.0, "step": 1924 }, { "entropy": 0.2638404816389084, "epoch": 2.9661016949152543, "grad_norm": 0.554985523223877, "learning_rate": 7.634776522534858e-09, "loss": 0.246, "mean_token_accuracy": 0.9346879422664642, "num_tokens": 92383649.0, "step": 1925 }, { "entropy": 0.2514384835958481, "epoch": 2.967642526964561, "grad_norm": 0.5062116980552673, "learning_rate": 6.985391005066566e-09, "loss": 0.2329, "mean_token_accuracy": 0.9361793398857117, "num_tokens": 92444282.0, "step": 1926 }, { "entropy": 0.230585977435112, "epoch": 2.9691833590138677, "grad_norm": 0.5410606265068054, "learning_rate": 6.364853928101378e-09, "loss": 0.211, "mean_token_accuracy": 0.9431191980838776, "num_tokens": 92503414.0, "step": 1927 }, { "entropy": 0.2593122124671936, "epoch": 2.970724191063174, "grad_norm": 0.6049841046333313, "learning_rate": 5.773167083041875e-09, "loss": 0.2378, "mean_token_accuracy": 0.9374564290046692, "num_tokens": 92561335.0, "step": 1928 }, { "entropy": 0.2290169894695282, "epoch": 2.972265023112481, "grad_norm": 0.5895904898643494, "learning_rate": 5.21033217800726e-09, "loss": 0.2096, "mean_token_accuracy": 0.9442637860774994, "num_tokens": 92618190.0, "step": 1929 }, { "entropy": 0.23503664135932922, "epoch": 2.973805855161787, "grad_norm": 0.5582882165908813, "learning_rate": 4.676350837822252e-09, "loss": 0.2137, "mean_token_accuracy": 0.9429614841938019, "num_tokens": 92673993.0, "step": 1930 }, { "entropy": 0.22940292954444885, "epoch": 2.975346687211094, "grad_norm": 0.5877745747566223, "learning_rate": 4.171224604017088e-09, "loss": 0.2112, "mean_token_accuracy": 0.9460145235061646, "num_tokens": 92728536.0, "step": 1931 }, { "entropy": 0.24386203289031982, "epoch": 2.9768875192604005, "grad_norm": 0.5994710922241211, "learning_rate": 3.69495493481975e-09, "loss": 0.2235, "mean_token_accuracy": 0.9419347643852234, "num_tokens": 92781620.0, "step": 1932 }, { "entropy": 0.31206244230270386, "epoch": 2.978428351309707, "grad_norm": 0.6426786184310913, "learning_rate": 3.2475432051559676e-09, "loss": 0.2415, "mean_token_accuracy": 0.9348886907100677, "num_tokens": 92827282.0, "step": 1933 }, { "entropy": 0.7239157855510712, "epoch": 2.979969183359014, "grad_norm": 1.090960144996643, "learning_rate": 2.8289907066403334e-09, "loss": 0.7261, "mean_token_accuracy": 0.7995174825191498, "num_tokens": 92838695.0, "step": 1934 }, { "entropy": 0.6590081453323364, "epoch": 2.9815100154083205, "grad_norm": 1.2538681030273438, "learning_rate": 2.4392986475751945e-09, "loss": 0.641, "mean_token_accuracy": 0.8298991620540619, "num_tokens": 92847808.0, "step": 1935 }, { "entropy": 0.3319293484091759, "epoch": 2.983050847457627, "grad_norm": 0.5089672803878784, "learning_rate": 2.078468152948432e-09, "loss": 0.1612, "mean_token_accuracy": 0.9167825877666473, "num_tokens": 92884185.0, "step": 1936 }, { "entropy": 0.26884694397449493, "epoch": 2.984591679506934, "grad_norm": 0.5721727609634399, "learning_rate": 1.7465002644256878e-09, "loss": 0.2563, "mean_token_accuracy": 0.9313720464706421, "num_tokens": 92947668.0, "step": 1937 }, { "entropy": 0.2452154979109764, "epoch": 2.9861325115562405, "grad_norm": 0.5684868097305298, "learning_rate": 1.4433959403525877e-09, "loss": 0.2299, "mean_token_accuracy": 0.9387290775775909, "num_tokens": 93008158.0, "step": 1938 }, { "entropy": 0.273072749376297, "epoch": 2.987673343605547, "grad_norm": 0.5967512130737305, "learning_rate": 1.1691560557491876e-09, "loss": 0.2557, "mean_token_accuracy": 0.9319887161254883, "num_tokens": 93067002.0, "step": 1939 }, { "entropy": 0.2558070570230484, "epoch": 2.989214175654854, "grad_norm": 0.6098358631134033, "learning_rate": 9.237814023088653e-10, "loss": 0.2334, "mean_token_accuracy": 0.9392768442630768, "num_tokens": 93124638.0, "step": 1940 }, { "entropy": 0.24199791997671127, "epoch": 2.99075500770416, "grad_norm": 0.6085664629936218, "learning_rate": 7.072726883927683e-10, "loss": 0.2281, "mean_token_accuracy": 0.939205527305603, "num_tokens": 93181185.0, "step": 1941 }, { "entropy": 0.2380695343017578, "epoch": 2.992295839753467, "grad_norm": 0.5850419402122498, "learning_rate": 5.196305390320344e-10, "loss": 0.2217, "mean_token_accuracy": 0.942044734954834, "num_tokens": 93236450.0, "step": 1942 }, { "entropy": 0.2501203119754791, "epoch": 2.9938366718027734, "grad_norm": 0.5775289535522461, "learning_rate": 3.60855495922241e-10, "loss": 0.2331, "mean_token_accuracy": 0.9386317431926727, "num_tokens": 93290313.0, "step": 1943 }, { "entropy": 0.22279025614261627, "epoch": 2.99537750385208, "grad_norm": 0.5858542919158936, "learning_rate": 2.3094801742673534e-10, "loss": 0.2036, "mean_token_accuracy": 0.9464101493358612, "num_tokens": 93341848.0, "step": 1944 }, { "entropy": 0.604118287563324, "epoch": 2.9969183359013867, "grad_norm": 0.7746883034706116, "learning_rate": 1.2990847856886313e-10, "loss": 0.4856, "mean_token_accuracy": 0.8481014370918274, "num_tokens": 93360743.0, "step": 1945 }, { "entropy": 0.6788574457168579, "epoch": 2.9984591679506933, "grad_norm": 1.0303720235824585, "learning_rate": 5.773717103529919e-11, "loss": 0.6587, "mean_token_accuracy": 0.8206154108047485, "num_tokens": 93370904.0, "step": 1946 }, { "entropy": 0.36017024517059326, "epoch": 3.0, "grad_norm": 0.5015812516212463, "learning_rate": 1.4434303176047437e-11, "loss": 0.1777, "mean_token_accuracy": 0.9141302406787872, "num_tokens": 93387442.0, "step": 1947 } ], "logging_steps": 1, "max_steps": 1947, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 200, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 1.030716897669782e+19, "train_batch_size": 8, "trial_name": null, "trial_params": null }