gokuls's picture
End of training
1f2a2eb
{
"best_metric": 4.17489767074585,
"best_model_checkpoint": "HBERTv1_emb_compress_48_L10_H768_A12/checkpoint-230000",
"epoch": 1.9579953792459324,
"eval_steps": 10000,
"global_step": 238989,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 5.000000000000001e-07,
"loss": 10.2069,
"step": 500
},
{
"epoch": 0.01,
"learning_rate": 1.0000000000000002e-06,
"loss": 10.1019,
"step": 1000
},
{
"epoch": 0.01,
"learning_rate": 1.5e-06,
"loss": 10.0185,
"step": 1500
},
{
"epoch": 0.02,
"learning_rate": 2.0000000000000003e-06,
"loss": 9.9059,
"step": 2000
},
{
"epoch": 0.02,
"learning_rate": 2.5e-06,
"loss": 9.7735,
"step": 2500
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 9.6238,
"step": 3000
},
{
"epoch": 0.03,
"learning_rate": 3.5e-06,
"loss": 9.4481,
"step": 3500
},
{
"epoch": 0.03,
"learning_rate": 4.000000000000001e-06,
"loss": 9.2494,
"step": 4000
},
{
"epoch": 0.04,
"learning_rate": 4.5e-06,
"loss": 9.0314,
"step": 4500
},
{
"epoch": 0.04,
"learning_rate": 5e-06,
"loss": 8.7941,
"step": 5000
},
{
"epoch": 0.05,
"learning_rate": 5.500000000000001e-06,
"loss": 8.5483,
"step": 5500
},
{
"epoch": 0.05,
"learning_rate": 6e-06,
"loss": 8.2991,
"step": 6000
},
{
"epoch": 0.05,
"learning_rate": 6.5000000000000004e-06,
"loss": 8.0554,
"step": 6500
},
{
"epoch": 0.06,
"learning_rate": 7e-06,
"loss": 7.8282,
"step": 7000
},
{
"epoch": 0.06,
"learning_rate": 7.500000000000001e-06,
"loss": 7.6235,
"step": 7500
},
{
"epoch": 0.07,
"learning_rate": 8.000000000000001e-06,
"loss": 7.4489,
"step": 8000
},
{
"epoch": 0.07,
"learning_rate": 8.5e-06,
"loss": 7.3122,
"step": 8500
},
{
"epoch": 0.07,
"learning_rate": 9e-06,
"loss": 7.2197,
"step": 9000
},
{
"epoch": 0.08,
"learning_rate": 9.5e-06,
"loss": 7.1526,
"step": 9500
},
{
"epoch": 0.08,
"learning_rate": 1e-05,
"loss": 7.1074,
"step": 10000
},
{
"epoch": 0.08,
"eval_accuracy": 0.08278212002383048,
"eval_loss": 7.083838939666748,
"eval_runtime": 1580.6139,
"eval_samples_per_second": 195.086,
"eval_steps_per_second": 4.065,
"step": 10000
},
{
"epoch": 0.09,
"learning_rate": 9.991670692498626e-06,
"loss": 7.066,
"step": 10500
},
{
"epoch": 0.09,
"learning_rate": 9.983341384997253e-06,
"loss": 7.0215,
"step": 11000
},
{
"epoch": 0.09,
"learning_rate": 9.975012077495878e-06,
"loss": 6.9868,
"step": 11500
},
{
"epoch": 0.1,
"learning_rate": 9.966682769994504e-06,
"loss": 6.9603,
"step": 12000
},
{
"epoch": 0.1,
"learning_rate": 9.95835346249313e-06,
"loss": 6.9352,
"step": 12500
},
{
"epoch": 0.11,
"learning_rate": 9.950024154991756e-06,
"loss": 6.9027,
"step": 13000
},
{
"epoch": 0.11,
"learning_rate": 9.941694847490381e-06,
"loss": 6.8866,
"step": 13500
},
{
"epoch": 0.11,
"learning_rate": 9.933365539989006e-06,
"loss": 6.8694,
"step": 14000
},
{
"epoch": 0.12,
"learning_rate": 9.925036232487631e-06,
"loss": 6.8411,
"step": 14500
},
{
"epoch": 0.12,
"learning_rate": 9.916706924986258e-06,
"loss": 6.8226,
"step": 15000
},
{
"epoch": 0.13,
"learning_rate": 9.908377617484883e-06,
"loss": 6.8068,
"step": 15500
},
{
"epoch": 0.13,
"learning_rate": 9.90004830998351e-06,
"loss": 6.7935,
"step": 16000
},
{
"epoch": 0.14,
"learning_rate": 9.891719002482134e-06,
"loss": 6.777,
"step": 16500
},
{
"epoch": 0.14,
"learning_rate": 9.883389694980761e-06,
"loss": 6.7594,
"step": 17000
},
{
"epoch": 0.14,
"learning_rate": 9.875060387479386e-06,
"loss": 6.7509,
"step": 17500
},
{
"epoch": 0.15,
"learning_rate": 9.866731079978011e-06,
"loss": 6.7339,
"step": 18000
},
{
"epoch": 0.15,
"learning_rate": 9.858401772476636e-06,
"loss": 6.7193,
"step": 18500
},
{
"epoch": 0.16,
"learning_rate": 9.850072464975263e-06,
"loss": 6.7049,
"step": 19000
},
{
"epoch": 0.16,
"learning_rate": 9.84174315747389e-06,
"loss": 6.6998,
"step": 19500
},
{
"epoch": 0.16,
"learning_rate": 9.833413849972514e-06,
"loss": 6.6784,
"step": 20000
},
{
"epoch": 0.16,
"eval_accuracy": 0.10751173228864543,
"eval_loss": 6.679544448852539,
"eval_runtime": 1580.2444,
"eval_samples_per_second": 195.132,
"eval_steps_per_second": 4.066,
"step": 20000
},
{
"epoch": 0.17,
"learning_rate": 9.82508454247114e-06,
"loss": 6.6738,
"step": 20500
},
{
"epoch": 0.17,
"learning_rate": 9.816755234969766e-06,
"loss": 6.6716,
"step": 21000
},
{
"epoch": 0.18,
"learning_rate": 9.808425927468391e-06,
"loss": 6.6484,
"step": 21500
},
{
"epoch": 0.18,
"learning_rate": 9.800096619967016e-06,
"loss": 6.6445,
"step": 22000
},
{
"epoch": 0.18,
"learning_rate": 9.791767312465643e-06,
"loss": 6.6379,
"step": 22500
},
{
"epoch": 0.19,
"learning_rate": 9.783438004964268e-06,
"loss": 6.6333,
"step": 23000
},
{
"epoch": 0.19,
"learning_rate": 9.775108697462895e-06,
"loss": 6.6177,
"step": 23500
},
{
"epoch": 0.2,
"learning_rate": 9.76677938996152e-06,
"loss": 6.6173,
"step": 24000
},
{
"epoch": 0.2,
"learning_rate": 9.758450082460145e-06,
"loss": 6.5977,
"step": 24500
},
{
"epoch": 0.2,
"learning_rate": 9.750120774958771e-06,
"loss": 6.6106,
"step": 25000
},
{
"epoch": 0.21,
"learning_rate": 9.741791467457396e-06,
"loss": 6.5927,
"step": 25500
},
{
"epoch": 0.21,
"learning_rate": 9.733462159956021e-06,
"loss": 6.5904,
"step": 26000
},
{
"epoch": 0.22,
"learning_rate": 9.725132852454648e-06,
"loss": 6.5809,
"step": 26500
},
{
"epoch": 0.22,
"learning_rate": 9.716803544953275e-06,
"loss": 6.5684,
"step": 27000
},
{
"epoch": 0.23,
"learning_rate": 9.7084742374519e-06,
"loss": 6.5609,
"step": 27500
},
{
"epoch": 0.23,
"learning_rate": 9.700144929950525e-06,
"loss": 6.5572,
"step": 28000
},
{
"epoch": 0.23,
"learning_rate": 9.69181562244915e-06,
"loss": 6.5565,
"step": 28500
},
{
"epoch": 0.24,
"learning_rate": 9.683486314947776e-06,
"loss": 6.5462,
"step": 29000
},
{
"epoch": 0.24,
"learning_rate": 9.675157007446401e-06,
"loss": 6.5433,
"step": 29500
},
{
"epoch": 0.25,
"learning_rate": 9.666827699945028e-06,
"loss": 6.535,
"step": 30000
},
{
"epoch": 0.25,
"eval_accuracy": 0.11924252539007817,
"eval_loss": 6.532186508178711,
"eval_runtime": 1582.2038,
"eval_samples_per_second": 194.89,
"eval_steps_per_second": 4.061,
"step": 30000
},
{
"epoch": 0.25,
"learning_rate": 9.658498392443653e-06,
"loss": 6.5266,
"step": 30500
},
{
"epoch": 0.25,
"learning_rate": 9.65016908494228e-06,
"loss": 6.5226,
"step": 31000
},
{
"epoch": 0.26,
"learning_rate": 9.641839777440905e-06,
"loss": 6.5216,
"step": 31500
},
{
"epoch": 0.26,
"learning_rate": 9.63351046993953e-06,
"loss": 6.5216,
"step": 32000
},
{
"epoch": 0.27,
"learning_rate": 9.625181162438155e-06,
"loss": 6.5115,
"step": 32500
},
{
"epoch": 0.27,
"learning_rate": 9.616851854936781e-06,
"loss": 6.5013,
"step": 33000
},
{
"epoch": 0.27,
"learning_rate": 9.608522547435406e-06,
"loss": 6.4921,
"step": 33500
},
{
"epoch": 0.28,
"learning_rate": 9.600193239934033e-06,
"loss": 6.4942,
"step": 34000
},
{
"epoch": 0.28,
"learning_rate": 9.591863932432658e-06,
"loss": 6.4906,
"step": 34500
},
{
"epoch": 0.29,
"learning_rate": 9.583534624931285e-06,
"loss": 6.4855,
"step": 35000
},
{
"epoch": 0.29,
"learning_rate": 9.57520531742991e-06,
"loss": 6.4917,
"step": 35500
},
{
"epoch": 0.29,
"learning_rate": 9.566876009928535e-06,
"loss": 6.4757,
"step": 36000
},
{
"epoch": 0.3,
"learning_rate": 9.55854670242716e-06,
"loss": 6.4675,
"step": 36500
},
{
"epoch": 0.3,
"learning_rate": 9.550217394925787e-06,
"loss": 6.4655,
"step": 37000
},
{
"epoch": 0.31,
"learning_rate": 9.541888087424413e-06,
"loss": 6.4699,
"step": 37500
},
{
"epoch": 0.31,
"learning_rate": 9.533558779923038e-06,
"loss": 6.4516,
"step": 38000
},
{
"epoch": 0.32,
"learning_rate": 9.525229472421663e-06,
"loss": 6.4578,
"step": 38500
},
{
"epoch": 0.32,
"learning_rate": 9.51690016492029e-06,
"loss": 6.4497,
"step": 39000
},
{
"epoch": 0.32,
"learning_rate": 9.508570857418915e-06,
"loss": 6.4445,
"step": 39500
},
{
"epoch": 0.33,
"learning_rate": 9.50024154991754e-06,
"loss": 6.4482,
"step": 40000
},
{
"epoch": 0.33,
"eval_accuracy": 0.12672710970015436,
"eval_loss": 6.438965797424316,
"eval_runtime": 1585.6479,
"eval_samples_per_second": 194.467,
"eval_steps_per_second": 4.052,
"step": 40000
},
{
"epoch": 0.33,
"learning_rate": 9.491912242416167e-06,
"loss": 6.435,
"step": 40500
},
{
"epoch": 0.34,
"learning_rate": 9.483582934914792e-06,
"loss": 6.4366,
"step": 41000
},
{
"epoch": 0.34,
"learning_rate": 9.475253627413418e-06,
"loss": 6.4338,
"step": 41500
},
{
"epoch": 0.34,
"learning_rate": 9.466924319912043e-06,
"loss": 6.4301,
"step": 42000
},
{
"epoch": 0.35,
"learning_rate": 9.458595012410668e-06,
"loss": 6.4299,
"step": 42500
},
{
"epoch": 0.35,
"learning_rate": 9.450265704909295e-06,
"loss": 6.4188,
"step": 43000
},
{
"epoch": 0.36,
"learning_rate": 9.44193639740792e-06,
"loss": 6.4196,
"step": 43500
},
{
"epoch": 0.36,
"learning_rate": 9.433607089906545e-06,
"loss": 6.4163,
"step": 44000
},
{
"epoch": 0.36,
"learning_rate": 9.425277782405172e-06,
"loss": 6.4094,
"step": 44500
},
{
"epoch": 0.37,
"learning_rate": 9.416948474903798e-06,
"loss": 6.4041,
"step": 45000
},
{
"epoch": 0.37,
"learning_rate": 9.408619167402423e-06,
"loss": 6.3966,
"step": 45500
},
{
"epoch": 0.38,
"learning_rate": 9.400289859901048e-06,
"loss": 6.398,
"step": 46000
},
{
"epoch": 0.38,
"learning_rate": 9.391960552399673e-06,
"loss": 6.3911,
"step": 46500
},
{
"epoch": 0.39,
"learning_rate": 9.3836312448983e-06,
"loss": 6.3955,
"step": 47000
},
{
"epoch": 0.39,
"learning_rate": 9.375301937396925e-06,
"loss": 6.3969,
"step": 47500
},
{
"epoch": 0.39,
"learning_rate": 9.366972629895552e-06,
"loss": 6.3905,
"step": 48000
},
{
"epoch": 0.4,
"learning_rate": 9.358643322394177e-06,
"loss": 6.3804,
"step": 48500
},
{
"epoch": 0.4,
"learning_rate": 9.350314014892804e-06,
"loss": 6.3801,
"step": 49000
},
{
"epoch": 0.41,
"learning_rate": 9.341984707391429e-06,
"loss": 6.3822,
"step": 49500
},
{
"epoch": 0.41,
"learning_rate": 9.333655399890054e-06,
"loss": 6.3716,
"step": 50000
},
{
"epoch": 0.41,
"eval_accuracy": 0.13242329621184235,
"eval_loss": 6.371087074279785,
"eval_runtime": 1578.6347,
"eval_samples_per_second": 195.331,
"eval_steps_per_second": 4.07,
"step": 50000
},
{
"epoch": 0.41,
"learning_rate": 9.325326092388679e-06,
"loss": 6.3762,
"step": 50500
},
{
"epoch": 0.42,
"learning_rate": 9.316996784887305e-06,
"loss": 6.3691,
"step": 51000
},
{
"epoch": 0.42,
"learning_rate": 9.30866747738593e-06,
"loss": 6.3706,
"step": 51500
},
{
"epoch": 0.43,
"learning_rate": 9.300338169884557e-06,
"loss": 6.3604,
"step": 52000
},
{
"epoch": 0.43,
"learning_rate": 9.292008862383182e-06,
"loss": 6.3587,
"step": 52500
},
{
"epoch": 0.43,
"learning_rate": 9.283679554881809e-06,
"loss": 6.3511,
"step": 53000
},
{
"epoch": 0.44,
"learning_rate": 9.275350247380434e-06,
"loss": 6.3539,
"step": 53500
},
{
"epoch": 0.44,
"learning_rate": 9.267020939879059e-06,
"loss": 6.3522,
"step": 54000
},
{
"epoch": 0.45,
"learning_rate": 9.258691632377685e-06,
"loss": 6.3403,
"step": 54500
},
{
"epoch": 0.45,
"learning_rate": 9.25036232487631e-06,
"loss": 6.3472,
"step": 55000
},
{
"epoch": 0.45,
"learning_rate": 9.242033017374937e-06,
"loss": 6.3436,
"step": 55500
},
{
"epoch": 0.46,
"learning_rate": 9.233703709873562e-06,
"loss": 6.3457,
"step": 56000
},
{
"epoch": 0.46,
"learning_rate": 9.225374402372189e-06,
"loss": 6.3421,
"step": 56500
},
{
"epoch": 0.47,
"learning_rate": 9.217045094870814e-06,
"loss": 6.3383,
"step": 57000
},
{
"epoch": 0.47,
"learning_rate": 9.208715787369439e-06,
"loss": 6.3389,
"step": 57500
},
{
"epoch": 0.48,
"learning_rate": 9.200386479868064e-06,
"loss": 6.3254,
"step": 58000
},
{
"epoch": 0.48,
"learning_rate": 9.19205717236669e-06,
"loss": 6.337,
"step": 58500
},
{
"epoch": 0.48,
"learning_rate": 9.183727864865315e-06,
"loss": 6.3234,
"step": 59000
},
{
"epoch": 0.49,
"learning_rate": 9.175398557363942e-06,
"loss": 6.3284,
"step": 59500
},
{
"epoch": 0.49,
"learning_rate": 9.167069249862567e-06,
"loss": 6.3233,
"step": 60000
},
{
"epoch": 0.49,
"eval_accuracy": 0.13513090758597238,
"eval_loss": 6.321866035461426,
"eval_runtime": 1580.28,
"eval_samples_per_second": 195.127,
"eval_steps_per_second": 4.066,
"step": 60000
},
{
"epoch": 0.5,
"learning_rate": 9.158739942361194e-06,
"loss": 6.3257,
"step": 60500
},
{
"epoch": 0.5,
"learning_rate": 9.150410634859819e-06,
"loss": 6.3203,
"step": 61000
},
{
"epoch": 0.5,
"learning_rate": 9.142081327358444e-06,
"loss": 6.3176,
"step": 61500
},
{
"epoch": 0.51,
"learning_rate": 9.133752019857069e-06,
"loss": 6.3173,
"step": 62000
},
{
"epoch": 0.51,
"learning_rate": 9.125422712355696e-06,
"loss": 6.3089,
"step": 62500
},
{
"epoch": 0.52,
"learning_rate": 9.11709340485432e-06,
"loss": 6.3148,
"step": 63000
},
{
"epoch": 0.52,
"learning_rate": 9.108764097352947e-06,
"loss": 6.3113,
"step": 63500
},
{
"epoch": 0.52,
"learning_rate": 9.100434789851572e-06,
"loss": 6.2967,
"step": 64000
},
{
"epoch": 0.53,
"learning_rate": 9.092105482350199e-06,
"loss": 6.2994,
"step": 64500
},
{
"epoch": 0.53,
"learning_rate": 9.083776174848824e-06,
"loss": 6.3024,
"step": 65000
},
{
"epoch": 0.54,
"learning_rate": 9.075446867347449e-06,
"loss": 6.3,
"step": 65500
},
{
"epoch": 0.54,
"learning_rate": 9.067117559846076e-06,
"loss": 6.3047,
"step": 66000
},
{
"epoch": 0.54,
"learning_rate": 9.0587882523447e-06,
"loss": 6.2907,
"step": 66500
},
{
"epoch": 0.55,
"learning_rate": 9.050458944843327e-06,
"loss": 6.2993,
"step": 67000
},
{
"epoch": 0.55,
"learning_rate": 9.042129637341952e-06,
"loss": 6.291,
"step": 67500
},
{
"epoch": 0.56,
"learning_rate": 9.033800329840577e-06,
"loss": 6.2823,
"step": 68000
},
{
"epoch": 0.56,
"learning_rate": 9.025471022339204e-06,
"loss": 6.2873,
"step": 68500
},
{
"epoch": 0.57,
"learning_rate": 9.017141714837829e-06,
"loss": 6.2845,
"step": 69000
},
{
"epoch": 0.57,
"learning_rate": 9.008812407336454e-06,
"loss": 6.281,
"step": 69500
},
{
"epoch": 0.57,
"learning_rate": 9.00048309983508e-06,
"loss": 6.2821,
"step": 70000
},
{
"epoch": 0.57,
"eval_accuracy": 0.13829922885191706,
"eval_loss": 6.278103828430176,
"eval_runtime": 1580.0233,
"eval_samples_per_second": 195.159,
"eval_steps_per_second": 4.066,
"step": 70000
},
{
"epoch": 0.58,
"learning_rate": 8.992153792333706e-06,
"loss": 6.2812,
"step": 70500
},
{
"epoch": 0.58,
"learning_rate": 8.983824484832332e-06,
"loss": 6.2796,
"step": 71000
},
{
"epoch": 0.59,
"learning_rate": 8.975495177330957e-06,
"loss": 6.2753,
"step": 71500
},
{
"epoch": 0.59,
"learning_rate": 8.967165869829582e-06,
"loss": 6.2812,
"step": 72000
},
{
"epoch": 0.59,
"learning_rate": 8.958836562328209e-06,
"loss": 6.2701,
"step": 72500
},
{
"epoch": 0.6,
"learning_rate": 8.950507254826834e-06,
"loss": 6.2686,
"step": 73000
},
{
"epoch": 0.6,
"learning_rate": 8.942177947325459e-06,
"loss": 6.2662,
"step": 73500
},
{
"epoch": 0.61,
"learning_rate": 8.933848639824086e-06,
"loss": 6.2643,
"step": 74000
},
{
"epoch": 0.61,
"learning_rate": 8.925519332322713e-06,
"loss": 6.2631,
"step": 74500
},
{
"epoch": 0.61,
"learning_rate": 8.917190024821338e-06,
"loss": 6.2663,
"step": 75000
},
{
"epoch": 0.62,
"learning_rate": 8.908860717319963e-06,
"loss": 6.2651,
"step": 75500
},
{
"epoch": 0.62,
"learning_rate": 8.900531409818588e-06,
"loss": 6.2671,
"step": 76000
},
{
"epoch": 0.63,
"learning_rate": 8.892202102317214e-06,
"loss": 6.2603,
"step": 76500
},
{
"epoch": 0.63,
"learning_rate": 8.88387279481584e-06,
"loss": 6.259,
"step": 77000
},
{
"epoch": 0.63,
"learning_rate": 8.875543487314466e-06,
"loss": 6.2528,
"step": 77500
},
{
"epoch": 0.64,
"learning_rate": 8.867214179813091e-06,
"loss": 6.2509,
"step": 78000
},
{
"epoch": 0.64,
"learning_rate": 8.858884872311718e-06,
"loss": 6.2532,
"step": 78500
},
{
"epoch": 0.65,
"learning_rate": 8.850555564810343e-06,
"loss": 6.2467,
"step": 79000
},
{
"epoch": 0.65,
"learning_rate": 8.842226257308968e-06,
"loss": 6.2482,
"step": 79500
},
{
"epoch": 0.66,
"learning_rate": 8.833896949807593e-06,
"loss": 6.251,
"step": 80000
},
{
"epoch": 0.66,
"eval_accuracy": 0.14081867847227955,
"eval_loss": 6.243099212646484,
"eval_runtime": 1579.9481,
"eval_samples_per_second": 195.168,
"eval_steps_per_second": 4.067,
"step": 80000
},
{
"epoch": 0.66,
"learning_rate": 8.82556764230622e-06,
"loss": 6.2483,
"step": 80500
},
{
"epoch": 0.66,
"learning_rate": 8.817238334804844e-06,
"loss": 6.2448,
"step": 81000
},
{
"epoch": 0.67,
"learning_rate": 8.808909027303471e-06,
"loss": 6.2395,
"step": 81500
},
{
"epoch": 0.67,
"learning_rate": 8.800579719802096e-06,
"loss": 6.242,
"step": 82000
},
{
"epoch": 0.68,
"learning_rate": 8.792250412300723e-06,
"loss": 6.2407,
"step": 82500
},
{
"epoch": 0.68,
"learning_rate": 8.783921104799348e-06,
"loss": 6.2332,
"step": 83000
},
{
"epoch": 0.68,
"learning_rate": 8.775591797297973e-06,
"loss": 6.2353,
"step": 83500
},
{
"epoch": 0.69,
"learning_rate": 8.767262489796598e-06,
"loss": 6.2343,
"step": 84000
},
{
"epoch": 0.69,
"learning_rate": 8.758933182295224e-06,
"loss": 6.2242,
"step": 84500
},
{
"epoch": 0.7,
"learning_rate": 8.750603874793851e-06,
"loss": 6.2308,
"step": 85000
},
{
"epoch": 0.7,
"learning_rate": 8.742274567292476e-06,
"loss": 6.2251,
"step": 85500
},
{
"epoch": 0.7,
"learning_rate": 8.733945259791101e-06,
"loss": 6.2308,
"step": 86000
},
{
"epoch": 0.71,
"learning_rate": 8.725615952289728e-06,
"loss": 6.2181,
"step": 86500
},
{
"epoch": 0.71,
"learning_rate": 8.717286644788353e-06,
"loss": 6.2171,
"step": 87000
},
{
"epoch": 0.72,
"learning_rate": 8.708957337286978e-06,
"loss": 6.2271,
"step": 87500
},
{
"epoch": 0.72,
"learning_rate": 8.700628029785605e-06,
"loss": 6.2183,
"step": 88000
},
{
"epoch": 0.73,
"learning_rate": 8.69229872228423e-06,
"loss": 6.2182,
"step": 88500
},
{
"epoch": 0.73,
"learning_rate": 8.683969414782856e-06,
"loss": 6.2187,
"step": 89000
},
{
"epoch": 0.73,
"learning_rate": 8.675640107281481e-06,
"loss": 6.2098,
"step": 89500
},
{
"epoch": 0.74,
"learning_rate": 8.667310799780106e-06,
"loss": 6.2159,
"step": 90000
},
{
"epoch": 0.74,
"eval_accuracy": 0.14246835530859112,
"eval_loss": 6.211101531982422,
"eval_runtime": 1578.6619,
"eval_samples_per_second": 195.327,
"eval_steps_per_second": 4.07,
"step": 90000
},
{
"epoch": 0.74,
"learning_rate": 8.658981492278733e-06,
"loss": 6.214,
"step": 90500
},
{
"epoch": 0.75,
"learning_rate": 8.650652184777358e-06,
"loss": 6.2078,
"step": 91000
},
{
"epoch": 0.75,
"learning_rate": 8.642322877275983e-06,
"loss": 6.2058,
"step": 91500
},
{
"epoch": 0.75,
"learning_rate": 8.63399356977461e-06,
"loss": 6.2091,
"step": 92000
},
{
"epoch": 0.76,
"learning_rate": 8.625664262273236e-06,
"loss": 6.2064,
"step": 92500
},
{
"epoch": 0.76,
"learning_rate": 8.617334954771861e-06,
"loss": 6.2046,
"step": 93000
},
{
"epoch": 0.77,
"learning_rate": 8.609005647270486e-06,
"loss": 6.2047,
"step": 93500
},
{
"epoch": 0.77,
"learning_rate": 8.600676339769111e-06,
"loss": 6.2015,
"step": 94000
},
{
"epoch": 0.77,
"learning_rate": 8.592347032267738e-06,
"loss": 6.1972,
"step": 94500
},
{
"epoch": 0.78,
"learning_rate": 8.584017724766363e-06,
"loss": 6.1875,
"step": 95000
},
{
"epoch": 0.78,
"learning_rate": 8.57568841726499e-06,
"loss": 6.2048,
"step": 95500
},
{
"epoch": 0.79,
"learning_rate": 8.567359109763615e-06,
"loss": 6.1954,
"step": 96000
},
{
"epoch": 0.79,
"learning_rate": 8.559029802262241e-06,
"loss": 6.1889,
"step": 96500
},
{
"epoch": 0.79,
"learning_rate": 8.550700494760866e-06,
"loss": 6.1947,
"step": 97000
},
{
"epoch": 0.8,
"learning_rate": 8.542371187259491e-06,
"loss": 6.1846,
"step": 97500
},
{
"epoch": 0.8,
"learning_rate": 8.534041879758118e-06,
"loss": 6.1825,
"step": 98000
},
{
"epoch": 0.81,
"learning_rate": 8.525712572256743e-06,
"loss": 6.1879,
"step": 98500
},
{
"epoch": 0.81,
"learning_rate": 8.517383264755368e-06,
"loss": 6.1852,
"step": 99000
},
{
"epoch": 0.82,
"learning_rate": 8.509053957253995e-06,
"loss": 6.1843,
"step": 99500
},
{
"epoch": 0.82,
"learning_rate": 8.500724649752622e-06,
"loss": 6.1838,
"step": 100000
},
{
"epoch": 0.82,
"eval_accuracy": 0.1444293247581906,
"eval_loss": 6.177408218383789,
"eval_runtime": 1578.5991,
"eval_samples_per_second": 195.335,
"eval_steps_per_second": 4.07,
"step": 100000
},
{
"epoch": 0.82,
"learning_rate": 8.492395342251247e-06,
"loss": 6.1767,
"step": 100500
},
{
"epoch": 0.83,
"learning_rate": 8.484066034749872e-06,
"loss": 6.1704,
"step": 101000
},
{
"epoch": 0.83,
"learning_rate": 8.475736727248497e-06,
"loss": 6.1719,
"step": 101500
},
{
"epoch": 0.84,
"learning_rate": 8.467407419747123e-06,
"loss": 6.1685,
"step": 102000
},
{
"epoch": 0.84,
"learning_rate": 8.459078112245748e-06,
"loss": 6.173,
"step": 102500
},
{
"epoch": 0.84,
"learning_rate": 8.450748804744375e-06,
"loss": 6.1637,
"step": 103000
},
{
"epoch": 0.85,
"learning_rate": 8.442419497243e-06,
"loss": 6.166,
"step": 103500
},
{
"epoch": 0.85,
"learning_rate": 8.434090189741627e-06,
"loss": 6.1639,
"step": 104000
},
{
"epoch": 0.86,
"learning_rate": 8.425760882240252e-06,
"loss": 6.161,
"step": 104500
},
{
"epoch": 0.86,
"learning_rate": 8.417431574738877e-06,
"loss": 6.1524,
"step": 105000
},
{
"epoch": 0.86,
"learning_rate": 8.409102267237502e-06,
"loss": 6.1491,
"step": 105500
},
{
"epoch": 0.87,
"learning_rate": 8.400772959736128e-06,
"loss": 6.1532,
"step": 106000
},
{
"epoch": 0.87,
"learning_rate": 8.392443652234753e-06,
"loss": 6.1554,
"step": 106500
},
{
"epoch": 0.88,
"learning_rate": 8.38411434473338e-06,
"loss": 6.1568,
"step": 107000
},
{
"epoch": 0.88,
"learning_rate": 8.375785037232005e-06,
"loss": 6.1481,
"step": 107500
},
{
"epoch": 0.88,
"learning_rate": 8.367455729730632e-06,
"loss": 6.1444,
"step": 108000
},
{
"epoch": 0.89,
"learning_rate": 8.359126422229257e-06,
"loss": 6.143,
"step": 108500
},
{
"epoch": 0.89,
"learning_rate": 8.350797114727882e-06,
"loss": 6.1443,
"step": 109000
},
{
"epoch": 0.9,
"learning_rate": 8.342467807226507e-06,
"loss": 6.1405,
"step": 109500
},
{
"epoch": 0.9,
"learning_rate": 8.334138499725133e-06,
"loss": 6.1338,
"step": 110000
},
{
"epoch": 0.9,
"eval_accuracy": 0.14637375769110822,
"eval_loss": 6.134941577911377,
"eval_runtime": 1583.5442,
"eval_samples_per_second": 194.725,
"eval_steps_per_second": 4.057,
"step": 110000
},
{
"epoch": 0.91,
"learning_rate": 8.32580919222376e-06,
"loss": 6.1348,
"step": 110500
},
{
"epoch": 0.91,
"learning_rate": 8.317479884722385e-06,
"loss": 6.1383,
"step": 111000
},
{
"epoch": 0.91,
"learning_rate": 8.30915057722101e-06,
"loss": 6.1392,
"step": 111500
},
{
"epoch": 0.92,
"learning_rate": 8.300821269719637e-06,
"loss": 6.1357,
"step": 112000
},
{
"epoch": 0.92,
"learning_rate": 8.292491962218262e-06,
"loss": 6.1328,
"step": 112500
},
{
"epoch": 0.93,
"learning_rate": 8.284162654716887e-06,
"loss": 6.1303,
"step": 113000
},
{
"epoch": 0.93,
"learning_rate": 8.275833347215514e-06,
"loss": 6.1254,
"step": 113500
},
{
"epoch": 0.93,
"learning_rate": 8.267504039714139e-06,
"loss": 6.1303,
"step": 114000
},
{
"epoch": 0.94,
"learning_rate": 8.259174732212765e-06,
"loss": 6.1204,
"step": 114500
},
{
"epoch": 0.94,
"learning_rate": 8.25084542471139e-06,
"loss": 6.1251,
"step": 115000
},
{
"epoch": 0.95,
"learning_rate": 8.242516117210015e-06,
"loss": 6.1247,
"step": 115500
},
{
"epoch": 0.95,
"learning_rate": 8.234186809708642e-06,
"loss": 6.1238,
"step": 116000
},
{
"epoch": 0.95,
"learning_rate": 8.225857502207267e-06,
"loss": 6.1209,
"step": 116500
},
{
"epoch": 0.96,
"learning_rate": 8.217528194705892e-06,
"loss": 6.1138,
"step": 117000
},
{
"epoch": 0.96,
"learning_rate": 8.209198887204519e-06,
"loss": 6.111,
"step": 117500
},
{
"epoch": 0.97,
"learning_rate": 8.200869579703145e-06,
"loss": 6.1091,
"step": 118000
},
{
"epoch": 0.97,
"learning_rate": 8.19254027220177e-06,
"loss": 6.1069,
"step": 118500
},
{
"epoch": 0.97,
"learning_rate": 8.184210964700395e-06,
"loss": 6.1164,
"step": 119000
},
{
"epoch": 0.98,
"learning_rate": 8.17588165719902e-06,
"loss": 6.1065,
"step": 119500
},
{
"epoch": 0.98,
"learning_rate": 8.167552349697647e-06,
"loss": 6.1022,
"step": 120000
},
{
"epoch": 0.98,
"eval_accuracy": 0.14809341125340497,
"eval_loss": 6.093875408172607,
"eval_runtime": 1588.0161,
"eval_samples_per_second": 194.177,
"eval_steps_per_second": 4.046,
"step": 120000
},
{
"epoch": 0.99,
"learning_rate": 8.159223042196272e-06,
"loss": 6.0932,
"step": 120500
},
{
"epoch": 0.99,
"learning_rate": 8.150893734694899e-06,
"loss": 6.1008,
"step": 121000
},
{
"epoch": 1.0,
"learning_rate": 8.142564427193524e-06,
"loss": 6.0901,
"step": 121500
},
{
"epoch": 1.0,
"learning_rate": 8.13423511969215e-06,
"loss": 6.0877,
"step": 122000
},
{
"epoch": 1.0,
"learning_rate": 8.125905812190775e-06,
"loss": 6.0819,
"step": 122500
},
{
"epoch": 1.01,
"learning_rate": 8.1175765046894e-06,
"loss": 6.0769,
"step": 123000
},
{
"epoch": 1.01,
"learning_rate": 8.109247197188025e-06,
"loss": 6.0685,
"step": 123500
},
{
"epoch": 1.02,
"learning_rate": 8.100917889686652e-06,
"loss": 6.0788,
"step": 124000
},
{
"epoch": 1.02,
"learning_rate": 8.092588582185277e-06,
"loss": 6.0692,
"step": 124500
},
{
"epoch": 1.02,
"learning_rate": 8.084259274683904e-06,
"loss": 6.0667,
"step": 125000
},
{
"epoch": 1.03,
"learning_rate": 8.075929967182529e-06,
"loss": 6.0598,
"step": 125500
},
{
"epoch": 1.03,
"learning_rate": 8.067600659681156e-06,
"loss": 6.0502,
"step": 126000
},
{
"epoch": 1.04,
"learning_rate": 8.05927135217978e-06,
"loss": 6.0502,
"step": 126500
},
{
"epoch": 1.04,
"learning_rate": 8.050942044678406e-06,
"loss": 6.0323,
"step": 127000
},
{
"epoch": 1.04,
"learning_rate": 8.04261273717703e-06,
"loss": 6.0379,
"step": 127500
},
{
"epoch": 1.05,
"learning_rate": 8.034283429675657e-06,
"loss": 6.0327,
"step": 128000
},
{
"epoch": 1.05,
"learning_rate": 8.025954122174284e-06,
"loss": 6.0248,
"step": 128500
},
{
"epoch": 1.06,
"learning_rate": 8.017624814672909e-06,
"loss": 6.0334,
"step": 129000
},
{
"epoch": 1.06,
"learning_rate": 8.009295507171534e-06,
"loss": 6.02,
"step": 129500
},
{
"epoch": 1.07,
"learning_rate": 8.00096619967016e-06,
"loss": 6.0194,
"step": 130000
},
{
"epoch": 1.07,
"eval_accuracy": 0.1516894608634533,
"eval_loss": 6.007966041564941,
"eval_runtime": 1587.2858,
"eval_samples_per_second": 194.266,
"eval_steps_per_second": 4.048,
"step": 130000
},
{
"epoch": 1.07,
"learning_rate": 7.992636892168786e-06,
"loss": 6.0169,
"step": 130500
},
{
"epoch": 1.07,
"learning_rate": 7.98430758466741e-06,
"loss": 6.0068,
"step": 131000
},
{
"epoch": 1.08,
"learning_rate": 7.975978277166037e-06,
"loss": 6.0055,
"step": 131500
},
{
"epoch": 1.08,
"learning_rate": 7.967648969664662e-06,
"loss": 5.9986,
"step": 132000
},
{
"epoch": 1.09,
"learning_rate": 7.959319662163289e-06,
"loss": 6.0037,
"step": 132500
},
{
"epoch": 1.09,
"learning_rate": 7.950990354661914e-06,
"loss": 6.0003,
"step": 133000
},
{
"epoch": 1.09,
"learning_rate": 7.942661047160539e-06,
"loss": 5.9885,
"step": 133500
},
{
"epoch": 1.1,
"learning_rate": 7.934331739659166e-06,
"loss": 5.9928,
"step": 134000
},
{
"epoch": 1.1,
"learning_rate": 7.92600243215779e-06,
"loss": 5.9885,
"step": 134500
},
{
"epoch": 1.11,
"learning_rate": 7.917673124656416e-06,
"loss": 5.9769,
"step": 135000
},
{
"epoch": 1.11,
"learning_rate": 7.909343817155042e-06,
"loss": 5.9778,
"step": 135500
},
{
"epoch": 1.11,
"learning_rate": 7.901014509653669e-06,
"loss": 5.9683,
"step": 136000
},
{
"epoch": 1.12,
"learning_rate": 7.892685202152294e-06,
"loss": 5.9621,
"step": 136500
},
{
"epoch": 1.12,
"learning_rate": 7.884355894650919e-06,
"loss": 5.9623,
"step": 137000
},
{
"epoch": 1.13,
"learning_rate": 7.876026587149544e-06,
"loss": 5.9569,
"step": 137500
},
{
"epoch": 1.13,
"learning_rate": 7.86769727964817e-06,
"loss": 5.9539,
"step": 138000
},
{
"epoch": 1.13,
"learning_rate": 7.859367972146796e-06,
"loss": 5.9485,
"step": 138500
},
{
"epoch": 1.14,
"learning_rate": 7.851038664645423e-06,
"loss": 5.9401,
"step": 139000
},
{
"epoch": 1.14,
"learning_rate": 7.842709357144048e-06,
"loss": 5.945,
"step": 139500
},
{
"epoch": 1.15,
"learning_rate": 7.834380049642674e-06,
"loss": 5.9309,
"step": 140000
},
{
"epoch": 1.15,
"eval_accuracy": 0.1642394578768706,
"eval_loss": 5.91987943649292,
"eval_runtime": 1587.768,
"eval_samples_per_second": 194.207,
"eval_steps_per_second": 4.047,
"step": 140000
},
{
"epoch": 1.15,
"learning_rate": 7.8260507421413e-06,
"loss": 5.9252,
"step": 140500
},
{
"epoch": 1.16,
"learning_rate": 7.817721434639924e-06,
"loss": 5.9326,
"step": 141000
},
{
"epoch": 1.16,
"learning_rate": 7.809392127138551e-06,
"loss": 5.9209,
"step": 141500
},
{
"epoch": 1.16,
"learning_rate": 7.801062819637176e-06,
"loss": 5.917,
"step": 142000
},
{
"epoch": 1.17,
"learning_rate": 7.792733512135801e-06,
"loss": 5.92,
"step": 142500
},
{
"epoch": 1.17,
"learning_rate": 7.784404204634428e-06,
"loss": 5.9083,
"step": 143000
},
{
"epoch": 1.18,
"learning_rate": 7.776074897133054e-06,
"loss": 5.9085,
"step": 143500
},
{
"epoch": 1.18,
"learning_rate": 7.76774558963168e-06,
"loss": 5.9036,
"step": 144000
},
{
"epoch": 1.18,
"learning_rate": 7.759416282130304e-06,
"loss": 5.8982,
"step": 144500
},
{
"epoch": 1.19,
"learning_rate": 7.75108697462893e-06,
"loss": 5.899,
"step": 145000
},
{
"epoch": 1.19,
"learning_rate": 7.742757667127556e-06,
"loss": 5.8947,
"step": 145500
},
{
"epoch": 1.2,
"learning_rate": 7.734428359626181e-06,
"loss": 5.8835,
"step": 146000
},
{
"epoch": 1.2,
"learning_rate": 7.726099052124808e-06,
"loss": 5.8851,
"step": 146500
},
{
"epoch": 1.2,
"learning_rate": 7.717769744623433e-06,
"loss": 5.8866,
"step": 147000
},
{
"epoch": 1.21,
"learning_rate": 7.70944043712206e-06,
"loss": 5.8797,
"step": 147500
},
{
"epoch": 1.21,
"learning_rate": 7.701111129620684e-06,
"loss": 5.8713,
"step": 148000
},
{
"epoch": 1.22,
"learning_rate": 7.69278182211931e-06,
"loss": 5.8739,
"step": 148500
},
{
"epoch": 1.22,
"learning_rate": 7.684452514617934e-06,
"loss": 5.8633,
"step": 149000
},
{
"epoch": 1.22,
"learning_rate": 7.676123207116561e-06,
"loss": 5.8625,
"step": 149500
},
{
"epoch": 1.23,
"learning_rate": 7.667793899615186e-06,
"loss": 5.8593,
"step": 150000
},
{
"epoch": 1.23,
"eval_accuracy": 0.17686092879085152,
"eval_loss": 5.832648754119873,
"eval_runtime": 1590.4655,
"eval_samples_per_second": 193.878,
"eval_steps_per_second": 4.04,
"step": 150000
},
{
"epoch": 1.23,
"learning_rate": 7.659464592113813e-06,
"loss": 5.8526,
"step": 150500
},
{
"epoch": 1.24,
"learning_rate": 7.651135284612438e-06,
"loss": 5.8472,
"step": 151000
},
{
"epoch": 1.24,
"learning_rate": 7.642805977111065e-06,
"loss": 5.8435,
"step": 151500
},
{
"epoch": 1.25,
"learning_rate": 7.63447666960969e-06,
"loss": 5.8389,
"step": 152000
},
{
"epoch": 1.25,
"learning_rate": 7.6261473621083145e-06,
"loss": 5.8354,
"step": 152500
},
{
"epoch": 1.25,
"learning_rate": 7.61781805460694e-06,
"loss": 5.8268,
"step": 153000
},
{
"epoch": 1.26,
"learning_rate": 7.609488747105567e-06,
"loss": 5.8242,
"step": 153500
},
{
"epoch": 1.26,
"learning_rate": 7.601159439604192e-06,
"loss": 5.8141,
"step": 154000
},
{
"epoch": 1.27,
"learning_rate": 7.592830132102817e-06,
"loss": 5.8079,
"step": 154500
},
{
"epoch": 1.27,
"learning_rate": 7.584500824601443e-06,
"loss": 5.7948,
"step": 155000
},
{
"epoch": 1.27,
"learning_rate": 7.57617151710007e-06,
"loss": 5.7878,
"step": 155500
},
{
"epoch": 1.28,
"learning_rate": 7.567842209598695e-06,
"loss": 5.7851,
"step": 156000
},
{
"epoch": 1.28,
"learning_rate": 7.5595129020973205e-06,
"loss": 5.7723,
"step": 156500
},
{
"epoch": 1.29,
"learning_rate": 7.5511835945959455e-06,
"loss": 5.7576,
"step": 157000
},
{
"epoch": 1.29,
"learning_rate": 7.542854287094572e-06,
"loss": 5.7574,
"step": 157500
},
{
"epoch": 1.29,
"learning_rate": 7.534524979593197e-06,
"loss": 5.7511,
"step": 158000
},
{
"epoch": 1.3,
"learning_rate": 7.526195672091823e-06,
"loss": 5.7404,
"step": 158500
},
{
"epoch": 1.3,
"learning_rate": 7.517866364590448e-06,
"loss": 5.7245,
"step": 159000
},
{
"epoch": 1.31,
"learning_rate": 7.509537057089075e-06,
"loss": 5.7206,
"step": 159500
},
{
"epoch": 1.31,
"learning_rate": 7.5012077495877e-06,
"loss": 5.7093,
"step": 160000
},
{
"epoch": 1.31,
"eval_accuracy": 0.20403873466247577,
"eval_loss": 5.665909290313721,
"eval_runtime": 1589.5203,
"eval_samples_per_second": 193.993,
"eval_steps_per_second": 4.042,
"step": 160000
},
{
"epoch": 1.31,
"learning_rate": 7.492878442086326e-06,
"loss": 5.6974,
"step": 160500
},
{
"epoch": 1.32,
"learning_rate": 7.484549134584951e-06,
"loss": 5.6881,
"step": 161000
},
{
"epoch": 1.32,
"learning_rate": 7.476219827083577e-06,
"loss": 5.683,
"step": 161500
},
{
"epoch": 1.33,
"learning_rate": 7.467890519582202e-06,
"loss": 5.6651,
"step": 162000
},
{
"epoch": 1.33,
"learning_rate": 7.459561212080828e-06,
"loss": 5.6696,
"step": 162500
},
{
"epoch": 1.34,
"learning_rate": 7.451231904579453e-06,
"loss": 5.6537,
"step": 163000
},
{
"epoch": 1.34,
"learning_rate": 7.44290259707808e-06,
"loss": 5.6446,
"step": 163500
},
{
"epoch": 1.34,
"learning_rate": 7.434573289576706e-06,
"loss": 5.6274,
"step": 164000
},
{
"epoch": 1.35,
"learning_rate": 7.426243982075331e-06,
"loss": 5.6161,
"step": 164500
},
{
"epoch": 1.35,
"learning_rate": 7.417914674573956e-06,
"loss": 5.6102,
"step": 165000
},
{
"epoch": 1.36,
"learning_rate": 7.409585367072582e-06,
"loss": 5.5932,
"step": 165500
},
{
"epoch": 1.36,
"learning_rate": 7.401256059571208e-06,
"loss": 5.5859,
"step": 166000
},
{
"epoch": 1.36,
"learning_rate": 7.392926752069833e-06,
"loss": 5.5689,
"step": 166500
},
{
"epoch": 1.37,
"learning_rate": 7.384597444568459e-06,
"loss": 5.5703,
"step": 167000
},
{
"epoch": 1.37,
"learning_rate": 7.376268137067085e-06,
"loss": 5.5574,
"step": 167500
},
{
"epoch": 1.38,
"learning_rate": 7.367938829565711e-06,
"loss": 5.5426,
"step": 168000
},
{
"epoch": 1.38,
"learning_rate": 7.359609522064336e-06,
"loss": 5.5308,
"step": 168500
},
{
"epoch": 1.38,
"learning_rate": 7.351280214562962e-06,
"loss": 5.5224,
"step": 169000
},
{
"epoch": 1.39,
"learning_rate": 7.3429509070615875e-06,
"loss": 5.5135,
"step": 169500
},
{
"epoch": 1.39,
"learning_rate": 7.334621599560213e-06,
"loss": 5.5018,
"step": 170000
},
{
"epoch": 1.39,
"eval_accuracy": 0.23387507284031656,
"eval_loss": 5.443307399749756,
"eval_runtime": 1588.6618,
"eval_samples_per_second": 194.098,
"eval_steps_per_second": 4.044,
"step": 170000
},
{
"epoch": 1.4,
"learning_rate": 7.326292292058838e-06,
"loss": 5.4854,
"step": 170500
},
{
"epoch": 1.4,
"learning_rate": 7.317962984557464e-06,
"loss": 5.4786,
"step": 171000
},
{
"epoch": 1.41,
"learning_rate": 7.309633677056091e-06,
"loss": 5.4724,
"step": 171500
},
{
"epoch": 1.41,
"learning_rate": 7.301304369554716e-06,
"loss": 5.4575,
"step": 172000
},
{
"epoch": 1.41,
"learning_rate": 7.292975062053341e-06,
"loss": 5.4517,
"step": 172500
},
{
"epoch": 1.42,
"learning_rate": 7.284645754551967e-06,
"loss": 5.4422,
"step": 173000
},
{
"epoch": 1.42,
"learning_rate": 7.2763164470505934e-06,
"loss": 5.4359,
"step": 173500
},
{
"epoch": 1.43,
"learning_rate": 7.2679871395492184e-06,
"loss": 5.4235,
"step": 174000
},
{
"epoch": 1.43,
"learning_rate": 7.259657832047844e-06,
"loss": 5.4131,
"step": 174500
},
{
"epoch": 1.43,
"learning_rate": 7.251328524546469e-06,
"loss": 5.4043,
"step": 175000
},
{
"epoch": 1.44,
"learning_rate": 7.242999217045096e-06,
"loss": 5.3934,
"step": 175500
},
{
"epoch": 1.44,
"learning_rate": 7.234669909543721e-06,
"loss": 5.3856,
"step": 176000
},
{
"epoch": 1.45,
"learning_rate": 7.226340602042347e-06,
"loss": 5.3723,
"step": 176500
},
{
"epoch": 1.45,
"learning_rate": 7.218011294540972e-06,
"loss": 5.3631,
"step": 177000
},
{
"epoch": 1.45,
"learning_rate": 7.2096819870395985e-06,
"loss": 5.3507,
"step": 177500
},
{
"epoch": 1.46,
"learning_rate": 7.2013526795382235e-06,
"loss": 5.339,
"step": 178000
},
{
"epoch": 1.46,
"learning_rate": 7.193023372036849e-06,
"loss": 5.3276,
"step": 178500
},
{
"epoch": 1.47,
"learning_rate": 7.184694064535474e-06,
"loss": 5.3195,
"step": 179000
},
{
"epoch": 1.47,
"learning_rate": 7.176364757034101e-06,
"loss": 5.3143,
"step": 179500
},
{
"epoch": 1.47,
"learning_rate": 7.168035449532726e-06,
"loss": 5.3036,
"step": 180000
},
{
"epoch": 1.47,
"eval_accuracy": 0.25764224608894687,
"eval_loss": 5.229185581207275,
"eval_runtime": 1591.3473,
"eval_samples_per_second": 193.77,
"eval_steps_per_second": 4.037,
"step": 180000
},
{
"epoch": 1.48,
"learning_rate": 7.159706142031352e-06,
"loss": 5.2829,
"step": 180500
},
{
"epoch": 1.48,
"learning_rate": 7.151376834529977e-06,
"loss": 5.2669,
"step": 181000
},
{
"epoch": 1.49,
"learning_rate": 7.143047527028604e-06,
"loss": 5.2612,
"step": 181500
},
{
"epoch": 1.49,
"learning_rate": 7.1347182195272295e-06,
"loss": 5.2562,
"step": 182000
},
{
"epoch": 1.5,
"learning_rate": 7.1263889120258545e-06,
"loss": 5.2445,
"step": 182500
},
{
"epoch": 1.5,
"learning_rate": 7.118059604524481e-06,
"loss": 5.234,
"step": 183000
},
{
"epoch": 1.5,
"learning_rate": 7.109730297023106e-06,
"loss": 5.2214,
"step": 183500
},
{
"epoch": 1.51,
"learning_rate": 7.101400989521732e-06,
"loss": 5.2147,
"step": 184000
},
{
"epoch": 1.51,
"learning_rate": 7.093071682020357e-06,
"loss": 5.2045,
"step": 184500
},
{
"epoch": 1.52,
"learning_rate": 7.084742374518984e-06,
"loss": 5.1819,
"step": 185000
},
{
"epoch": 1.52,
"learning_rate": 7.076413067017609e-06,
"loss": 5.1804,
"step": 185500
},
{
"epoch": 1.52,
"learning_rate": 7.068083759516235e-06,
"loss": 5.1655,
"step": 186000
},
{
"epoch": 1.53,
"learning_rate": 7.05975445201486e-06,
"loss": 5.1458,
"step": 186500
},
{
"epoch": 1.53,
"learning_rate": 7.051425144513486e-06,
"loss": 5.15,
"step": 187000
},
{
"epoch": 1.54,
"learning_rate": 7.043095837012111e-06,
"loss": 5.1348,
"step": 187500
},
{
"epoch": 1.54,
"learning_rate": 7.034766529510737e-06,
"loss": 5.1179,
"step": 188000
},
{
"epoch": 1.54,
"learning_rate": 7.026437222009362e-06,
"loss": 5.1011,
"step": 188500
},
{
"epoch": 1.55,
"learning_rate": 7.018107914507989e-06,
"loss": 5.0872,
"step": 189000
},
{
"epoch": 1.55,
"learning_rate": 7.009778607006615e-06,
"loss": 5.0784,
"step": 189500
},
{
"epoch": 1.56,
"learning_rate": 7.00144929950524e-06,
"loss": 5.0629,
"step": 190000
},
{
"epoch": 1.56,
"eval_accuracy": 0.2833824729652,
"eval_loss": 4.989490509033203,
"eval_runtime": 1588.3188,
"eval_samples_per_second": 194.14,
"eval_steps_per_second": 4.045,
"step": 190000
},
{
"epoch": 1.56,
"learning_rate": 6.993119992003865e-06,
"loss": 5.0462,
"step": 190500
},
{
"epoch": 1.56,
"learning_rate": 6.984790684502491e-06,
"loss": 5.0332,
"step": 191000
},
{
"epoch": 1.57,
"learning_rate": 6.976461377001117e-06,
"loss": 5.0144,
"step": 191500
},
{
"epoch": 1.57,
"learning_rate": 6.968132069499742e-06,
"loss": 4.9984,
"step": 192000
},
{
"epoch": 1.58,
"learning_rate": 6.959802761998368e-06,
"loss": 4.9933,
"step": 192500
},
{
"epoch": 1.58,
"learning_rate": 6.951473454496994e-06,
"loss": 4.9832,
"step": 193000
},
{
"epoch": 1.59,
"learning_rate": 6.94314414699562e-06,
"loss": 4.9633,
"step": 193500
},
{
"epoch": 1.59,
"learning_rate": 6.934814839494245e-06,
"loss": 4.9562,
"step": 194000
},
{
"epoch": 1.59,
"learning_rate": 6.926485531992871e-06,
"loss": 4.9498,
"step": 194500
},
{
"epoch": 1.6,
"learning_rate": 6.9181562244914965e-06,
"loss": 4.9311,
"step": 195000
},
{
"epoch": 1.6,
"learning_rate": 6.909826916990122e-06,
"loss": 4.9127,
"step": 195500
},
{
"epoch": 1.61,
"learning_rate": 6.901497609488747e-06,
"loss": 4.9071,
"step": 196000
},
{
"epoch": 1.61,
"learning_rate": 6.893168301987373e-06,
"loss": 4.8994,
"step": 196500
},
{
"epoch": 1.61,
"learning_rate": 6.884838994485999e-06,
"loss": 4.8945,
"step": 197000
},
{
"epoch": 1.62,
"learning_rate": 6.876509686984625e-06,
"loss": 4.8747,
"step": 197500
},
{
"epoch": 1.62,
"learning_rate": 6.86818037948325e-06,
"loss": 4.8668,
"step": 198000
},
{
"epoch": 1.63,
"learning_rate": 6.859851071981876e-06,
"loss": 4.8544,
"step": 198500
},
{
"epoch": 1.63,
"learning_rate": 6.8515217644805024e-06,
"loss": 4.8514,
"step": 199000
},
{
"epoch": 1.63,
"learning_rate": 6.8431924569791274e-06,
"loss": 4.8308,
"step": 199500
},
{
"epoch": 1.64,
"learning_rate": 6.8348631494777525e-06,
"loss": 4.8311,
"step": 200000
},
{
"epoch": 1.64,
"eval_accuracy": 0.3084937610865483,
"eval_loss": 4.76377534866333,
"eval_runtime": 1587.3867,
"eval_samples_per_second": 194.254,
"eval_steps_per_second": 4.048,
"step": 200000
},
{
"epoch": 1.64,
"learning_rate": 6.826533841976378e-06,
"loss": 4.8204,
"step": 200500
},
{
"epoch": 1.65,
"learning_rate": 6.818204534475005e-06,
"loss": 4.8109,
"step": 201000
},
{
"epoch": 1.65,
"learning_rate": 6.80987522697363e-06,
"loss": 4.7963,
"step": 201500
},
{
"epoch": 1.65,
"learning_rate": 6.801545919472256e-06,
"loss": 4.7902,
"step": 202000
},
{
"epoch": 1.66,
"learning_rate": 6.793216611970881e-06,
"loss": 4.7743,
"step": 202500
},
{
"epoch": 1.66,
"learning_rate": 6.7848873044695075e-06,
"loss": 4.7668,
"step": 203000
},
{
"epoch": 1.67,
"learning_rate": 6.7765579969681325e-06,
"loss": 4.7557,
"step": 203500
},
{
"epoch": 1.67,
"learning_rate": 6.768228689466758e-06,
"loss": 4.7478,
"step": 204000
},
{
"epoch": 1.68,
"learning_rate": 6.759899381965383e-06,
"loss": 4.7326,
"step": 204500
},
{
"epoch": 1.68,
"learning_rate": 6.75157007446401e-06,
"loss": 4.7314,
"step": 205000
},
{
"epoch": 1.68,
"learning_rate": 6.743240766962635e-06,
"loss": 4.7142,
"step": 205500
},
{
"epoch": 1.69,
"learning_rate": 6.734911459461261e-06,
"loss": 4.7195,
"step": 206000
},
{
"epoch": 1.69,
"learning_rate": 6.726582151959886e-06,
"loss": 4.6984,
"step": 206500
},
{
"epoch": 1.7,
"learning_rate": 6.718252844458513e-06,
"loss": 4.6966,
"step": 207000
},
{
"epoch": 1.7,
"learning_rate": 6.709923536957138e-06,
"loss": 4.677,
"step": 207500
},
{
"epoch": 1.7,
"learning_rate": 6.7015942294557635e-06,
"loss": 4.6676,
"step": 208000
},
{
"epoch": 1.71,
"learning_rate": 6.6932649219543885e-06,
"loss": 4.6592,
"step": 208500
},
{
"epoch": 1.71,
"learning_rate": 6.684935614453015e-06,
"loss": 4.654,
"step": 209000
},
{
"epoch": 1.72,
"learning_rate": 6.676606306951641e-06,
"loss": 4.6474,
"step": 209500
},
{
"epoch": 1.72,
"learning_rate": 6.668276999450266e-06,
"loss": 4.6239,
"step": 210000
},
{
"epoch": 1.72,
"eval_accuracy": 0.32783534000173253,
"eval_loss": 4.579888820648193,
"eval_runtime": 1589.2019,
"eval_samples_per_second": 194.032,
"eval_steps_per_second": 4.043,
"step": 210000
},
{
"epoch": 1.72,
"learning_rate": 6.659947691948891e-06,
"loss": 4.6198,
"step": 210500
},
{
"epoch": 1.73,
"learning_rate": 6.651618384447518e-06,
"loss": 4.6111,
"step": 211000
},
{
"epoch": 1.73,
"learning_rate": 6.643289076946144e-06,
"loss": 4.6091,
"step": 211500
},
{
"epoch": 1.74,
"learning_rate": 6.634959769444769e-06,
"loss": 4.5936,
"step": 212000
},
{
"epoch": 1.74,
"learning_rate": 6.6266304619433945e-06,
"loss": 4.5908,
"step": 212500
},
{
"epoch": 1.75,
"learning_rate": 6.61830115444202e-06,
"loss": 4.5759,
"step": 213000
},
{
"epoch": 1.75,
"learning_rate": 6.609971846940646e-06,
"loss": 4.5622,
"step": 213500
},
{
"epoch": 1.75,
"learning_rate": 6.601642539439271e-06,
"loss": 4.5583,
"step": 214000
},
{
"epoch": 1.76,
"learning_rate": 6.593313231937897e-06,
"loss": 4.5455,
"step": 214500
},
{
"epoch": 1.76,
"learning_rate": 6.584983924436523e-06,
"loss": 4.5355,
"step": 215000
},
{
"epoch": 1.77,
"learning_rate": 6.576654616935149e-06,
"loss": 4.5207,
"step": 215500
},
{
"epoch": 1.77,
"learning_rate": 6.568325309433774e-06,
"loss": 4.524,
"step": 216000
},
{
"epoch": 1.77,
"learning_rate": 6.5599960019323996e-06,
"loss": 4.5,
"step": 216500
},
{
"epoch": 1.78,
"learning_rate": 6.551666694431026e-06,
"loss": 4.4969,
"step": 217000
},
{
"epoch": 1.78,
"learning_rate": 6.543337386929651e-06,
"loss": 4.4852,
"step": 217500
},
{
"epoch": 1.79,
"learning_rate": 6.535008079428276e-06,
"loss": 4.4751,
"step": 218000
},
{
"epoch": 1.79,
"learning_rate": 6.526678771926902e-06,
"loss": 4.4584,
"step": 218500
},
{
"epoch": 1.79,
"learning_rate": 6.518349464425529e-06,
"loss": 4.4466,
"step": 219000
},
{
"epoch": 1.8,
"learning_rate": 6.510020156924154e-06,
"loss": 4.443,
"step": 219500
},
{
"epoch": 1.8,
"learning_rate": 6.50169084942278e-06,
"loss": 4.4305,
"step": 220000
},
{
"epoch": 1.8,
"eval_accuracy": 0.3471177142483127,
"eval_loss": 4.382063865661621,
"eval_runtime": 1586.3632,
"eval_samples_per_second": 194.379,
"eval_steps_per_second": 4.05,
"step": 220000
},
{
"epoch": 1.81,
"learning_rate": 6.493361541921405e-06,
"loss": 4.4288,
"step": 220500
},
{
"epoch": 1.81,
"learning_rate": 6.485032234420031e-06,
"loss": 4.4061,
"step": 221000
},
{
"epoch": 1.81,
"learning_rate": 6.476702926918656e-06,
"loss": 4.3984,
"step": 221500
},
{
"epoch": 1.82,
"learning_rate": 6.468373619417282e-06,
"loss": 4.39,
"step": 222000
},
{
"epoch": 1.82,
"learning_rate": 6.460044311915907e-06,
"loss": 4.3744,
"step": 222500
},
{
"epoch": 1.83,
"learning_rate": 6.451715004414534e-06,
"loss": 4.3787,
"step": 223000
},
{
"epoch": 1.83,
"learning_rate": 6.443385696913159e-06,
"loss": 4.3533,
"step": 223500
},
{
"epoch": 1.84,
"learning_rate": 6.435056389411785e-06,
"loss": 4.3479,
"step": 224000
},
{
"epoch": 1.84,
"learning_rate": 6.42672708191041e-06,
"loss": 4.3442,
"step": 224500
},
{
"epoch": 1.84,
"learning_rate": 6.4183977744090365e-06,
"loss": 4.3353,
"step": 225000
},
{
"epoch": 1.85,
"learning_rate": 6.4100684669076615e-06,
"loss": 4.3166,
"step": 225500
},
{
"epoch": 1.85,
"learning_rate": 6.401739159406287e-06,
"loss": 4.3086,
"step": 226000
},
{
"epoch": 1.86,
"learning_rate": 6.393409851904914e-06,
"loss": 4.3008,
"step": 226500
},
{
"epoch": 1.86,
"learning_rate": 6.385080544403539e-06,
"loss": 4.2799,
"step": 227000
},
{
"epoch": 1.86,
"learning_rate": 6.376751236902165e-06,
"loss": 4.2744,
"step": 227500
},
{
"epoch": 1.87,
"learning_rate": 6.36842192940079e-06,
"loss": 4.2745,
"step": 228000
},
{
"epoch": 1.87,
"learning_rate": 6.3600926218994166e-06,
"loss": 4.2587,
"step": 228500
},
{
"epoch": 1.88,
"learning_rate": 6.3517633143980416e-06,
"loss": 4.2521,
"step": 229000
},
{
"epoch": 1.88,
"learning_rate": 6.343434006896667e-06,
"loss": 4.2376,
"step": 229500
},
{
"epoch": 1.88,
"learning_rate": 6.335104699395292e-06,
"loss": 4.2209,
"step": 230000
},
{
"epoch": 1.88,
"eval_accuracy": 0.3704495166952504,
"eval_loss": 4.17489767074585,
"eval_runtime": 1581.5202,
"eval_samples_per_second": 194.974,
"eval_steps_per_second": 4.063,
"step": 230000
},
{
"epoch": 1.89,
"learning_rate": 6.326775391893919e-06,
"loss": 4.2175,
"step": 230500
},
{
"epoch": 1.89,
"learning_rate": 6.318446084392544e-06,
"loss": 4.2101,
"step": 231000
},
{
"epoch": 1.9,
"learning_rate": 6.31011677689117e-06,
"loss": 4.2024,
"step": 231500
},
{
"epoch": 1.9,
"learning_rate": 6.301787469389795e-06,
"loss": 4.1912,
"step": 232000
},
{
"epoch": 1.9,
"learning_rate": 6.293458161888422e-06,
"loss": 4.1697,
"step": 232500
},
{
"epoch": 1.91,
"learning_rate": 6.285128854387047e-06,
"loss": 4.1737,
"step": 233000
},
{
"epoch": 1.91,
"learning_rate": 6.2767995468856725e-06,
"loss": 4.1519,
"step": 233500
},
{
"epoch": 1.92,
"learning_rate": 6.2684702393842975e-06,
"loss": 4.1537,
"step": 234000
},
{
"epoch": 1.92,
"learning_rate": 6.260140931882924e-06,
"loss": 4.1277,
"step": 234500
},
{
"epoch": 1.93,
"learning_rate": 6.25181162438155e-06,
"loss": 4.1319,
"step": 235000
},
{
"epoch": 1.93,
"learning_rate": 6.243482316880175e-06,
"loss": 4.1174,
"step": 235500
},
{
"epoch": 1.93,
"learning_rate": 6.2351530093788e-06,
"loss": 4.1118,
"step": 236000
},
{
"epoch": 1.94,
"learning_rate": 6.226823701877427e-06,
"loss": 4.0875,
"step": 236500
},
{
"epoch": 1.94,
"learning_rate": 6.218494394376053e-06,
"loss": 4.0778,
"step": 237000
},
{
"epoch": 1.95,
"learning_rate": 6.210165086874678e-06,
"loss": 4.0804,
"step": 237500
},
{
"epoch": 1.95,
"learning_rate": 6.2018357793733035e-06,
"loss": 4.0629,
"step": 238000
},
{
"epoch": 1.95,
"learning_rate": 6.193506471871929e-06,
"loss": 4.0587,
"step": 238500
},
{
"epoch": 1.96,
"step": 238989,
"total_flos": 2.8220932538708787e+18,
"train_loss": 5.887160467710138,
"train_runtime": 197999.2632,
"train_samples_per_second": 147.949,
"train_steps_per_second": 3.082
}
],
"logging_steps": 500,
"max_steps": 610290,
"num_train_epochs": 5,
"save_steps": 10000,
"total_flos": 2.8220932538708787e+18,
"trial_name": null,
"trial_params": null
}