gokuls's picture
End of training
12e2f23
{
"best_metric": 2.2642505168914795,
"best_model_checkpoint": "bert_12_layer_model_v1_complete_training_new_120/checkpoint-100000",
"epoch": 0.8401251863867997,
"global_step": 102544,
"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": 2.4457,
"step": 500
},
{
"epoch": 0.01,
"learning_rate": 1.0000000000000002e-06,
"loss": 2.4345,
"step": 1000
},
{
"epoch": 0.01,
"learning_rate": 1.5e-06,
"loss": 2.4321,
"step": 1500
},
{
"epoch": 0.02,
"learning_rate": 2.0000000000000003e-06,
"loss": 2.4263,
"step": 2000
},
{
"epoch": 0.02,
"learning_rate": 2.5e-06,
"loss": 2.425,
"step": 2500
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 2.4301,
"step": 3000
},
{
"epoch": 0.03,
"learning_rate": 3.5e-06,
"loss": 2.4237,
"step": 3500
},
{
"epoch": 0.03,
"learning_rate": 4.000000000000001e-06,
"loss": 2.4301,
"step": 4000
},
{
"epoch": 0.04,
"learning_rate": 4.5e-06,
"loss": 2.4274,
"step": 4500
},
{
"epoch": 0.04,
"learning_rate": 5e-06,
"loss": 2.4339,
"step": 5000
},
{
"epoch": 0.05,
"learning_rate": 5.500000000000001e-06,
"loss": 2.4271,
"step": 5500
},
{
"epoch": 0.05,
"learning_rate": 6e-06,
"loss": 2.4326,
"step": 6000
},
{
"epoch": 0.05,
"learning_rate": 6.5000000000000004e-06,
"loss": 2.4352,
"step": 6500
},
{
"epoch": 0.06,
"learning_rate": 7e-06,
"loss": 2.4296,
"step": 7000
},
{
"epoch": 0.06,
"learning_rate": 7.500000000000001e-06,
"loss": 2.4327,
"step": 7500
},
{
"epoch": 0.07,
"learning_rate": 8.000000000000001e-06,
"loss": 2.4353,
"step": 8000
},
{
"epoch": 0.07,
"learning_rate": 8.5e-06,
"loss": 2.4334,
"step": 8500
},
{
"epoch": 0.07,
"learning_rate": 9e-06,
"loss": 2.4362,
"step": 9000
},
{
"epoch": 0.08,
"learning_rate": 9.5e-06,
"loss": 2.4367,
"step": 9500
},
{
"epoch": 0.08,
"learning_rate": 1e-05,
"loss": 2.4425,
"step": 10000
},
{
"epoch": 0.08,
"eval_accuracy": 0.5641066783899575,
"eval_loss": 2.383796453475952,
"eval_runtime": 1886.5077,
"eval_samples_per_second": 163.453,
"eval_steps_per_second": 3.406,
"step": 10000
},
{
"epoch": 0.09,
"learning_rate": 9.991670692498626e-06,
"loss": 2.4407,
"step": 10500
},
{
"epoch": 0.09,
"learning_rate": 9.983341384997253e-06,
"loss": 2.453,
"step": 11000
},
{
"epoch": 0.09,
"learning_rate": 9.975012077495878e-06,
"loss": 2.4437,
"step": 11500
},
{
"epoch": 0.1,
"learning_rate": 9.966682769994504e-06,
"loss": 2.4417,
"step": 12000
},
{
"epoch": 0.1,
"learning_rate": 9.95835346249313e-06,
"loss": 2.438,
"step": 12500
},
{
"epoch": 0.11,
"learning_rate": 9.950024154991756e-06,
"loss": 2.4354,
"step": 13000
},
{
"epoch": 0.11,
"learning_rate": 9.941694847490381e-06,
"loss": 2.4396,
"step": 13500
},
{
"epoch": 0.11,
"learning_rate": 9.933365539989006e-06,
"loss": 2.4402,
"step": 14000
},
{
"epoch": 0.12,
"learning_rate": 9.925036232487631e-06,
"loss": 2.4302,
"step": 14500
},
{
"epoch": 0.12,
"learning_rate": 9.916706924986258e-06,
"loss": 2.4353,
"step": 15000
},
{
"epoch": 0.13,
"learning_rate": 9.908377617484883e-06,
"loss": 2.4302,
"step": 15500
},
{
"epoch": 0.13,
"learning_rate": 9.90004830998351e-06,
"loss": 2.4342,
"step": 16000
},
{
"epoch": 0.14,
"learning_rate": 9.891719002482134e-06,
"loss": 2.4367,
"step": 16500
},
{
"epoch": 0.14,
"learning_rate": 9.883389694980761e-06,
"loss": 2.4308,
"step": 17000
},
{
"epoch": 0.14,
"learning_rate": 9.875060387479386e-06,
"loss": 2.4341,
"step": 17500
},
{
"epoch": 0.15,
"learning_rate": 9.866731079978011e-06,
"loss": 2.434,
"step": 18000
},
{
"epoch": 0.15,
"learning_rate": 9.858401772476636e-06,
"loss": 2.4361,
"step": 18500
},
{
"epoch": 0.16,
"learning_rate": 9.850072464975263e-06,
"loss": 2.4313,
"step": 19000
},
{
"epoch": 0.16,
"learning_rate": 9.84174315747389e-06,
"loss": 2.4304,
"step": 19500
},
{
"epoch": 0.16,
"learning_rate": 9.833413849972514e-06,
"loss": 2.4415,
"step": 20000
},
{
"epoch": 0.16,
"eval_accuracy": 0.5658001427669171,
"eval_loss": 2.370534896850586,
"eval_runtime": 1888.3249,
"eval_samples_per_second": 163.296,
"eval_steps_per_second": 3.402,
"step": 20000
},
{
"epoch": 0.17,
"learning_rate": 9.82508454247114e-06,
"loss": 2.4238,
"step": 20500
},
{
"epoch": 0.17,
"learning_rate": 9.816755234969766e-06,
"loss": 2.4273,
"step": 21000
},
{
"epoch": 0.18,
"learning_rate": 9.808425927468391e-06,
"loss": 2.4231,
"step": 21500
},
{
"epoch": 0.18,
"learning_rate": 9.800096619967016e-06,
"loss": 2.4225,
"step": 22000
},
{
"epoch": 0.18,
"learning_rate": 9.791767312465643e-06,
"loss": 2.4157,
"step": 22500
},
{
"epoch": 0.19,
"learning_rate": 9.783438004964268e-06,
"loss": 2.415,
"step": 23000
},
{
"epoch": 0.19,
"learning_rate": 9.775108697462895e-06,
"loss": 2.4223,
"step": 23500
},
{
"epoch": 0.2,
"learning_rate": 9.76677938996152e-06,
"loss": 2.4173,
"step": 24000
},
{
"epoch": 0.2,
"learning_rate": 9.758450082460145e-06,
"loss": 2.4165,
"step": 24500
},
{
"epoch": 0.2,
"learning_rate": 9.750120774958771e-06,
"loss": 2.4105,
"step": 25000
},
{
"epoch": 0.21,
"learning_rate": 9.741791467457396e-06,
"loss": 2.4157,
"step": 25500
},
{
"epoch": 0.21,
"learning_rate": 9.733462159956021e-06,
"loss": 2.4165,
"step": 26000
},
{
"epoch": 0.22,
"learning_rate": 9.725132852454648e-06,
"loss": 2.4188,
"step": 26500
},
{
"epoch": 0.22,
"learning_rate": 9.716803544953275e-06,
"loss": 2.4175,
"step": 27000
},
{
"epoch": 0.23,
"learning_rate": 9.7084742374519e-06,
"loss": 2.4159,
"step": 27500
},
{
"epoch": 0.23,
"learning_rate": 9.700144929950525e-06,
"loss": 2.41,
"step": 28000
},
{
"epoch": 0.23,
"learning_rate": 9.69181562244915e-06,
"loss": 2.4157,
"step": 28500
},
{
"epoch": 0.24,
"learning_rate": 9.683486314947776e-06,
"loss": 2.4125,
"step": 29000
},
{
"epoch": 0.24,
"learning_rate": 9.675157007446401e-06,
"loss": 2.4142,
"step": 29500
},
{
"epoch": 0.25,
"learning_rate": 9.666827699945028e-06,
"loss": 2.4103,
"step": 30000
},
{
"epoch": 0.25,
"eval_accuracy": 0.5680406910642719,
"eval_loss": 2.353705883026123,
"eval_runtime": 1889.4384,
"eval_samples_per_second": 163.2,
"eval_steps_per_second": 3.4,
"step": 30000
},
{
"epoch": 0.25,
"learning_rate": 9.658498392443653e-06,
"loss": 2.404,
"step": 30500
},
{
"epoch": 0.25,
"learning_rate": 9.65016908494228e-06,
"loss": 2.4095,
"step": 31000
},
{
"epoch": 0.26,
"learning_rate": 9.641839777440905e-06,
"loss": 2.4095,
"step": 31500
},
{
"epoch": 0.26,
"learning_rate": 9.63351046993953e-06,
"loss": 2.4109,
"step": 32000
},
{
"epoch": 0.27,
"learning_rate": 9.625181162438155e-06,
"loss": 2.4055,
"step": 32500
},
{
"epoch": 0.27,
"learning_rate": 9.616851854936781e-06,
"loss": 2.4101,
"step": 33000
},
{
"epoch": 0.27,
"learning_rate": 9.608522547435406e-06,
"loss": 2.4073,
"step": 33500
},
{
"epoch": 0.28,
"learning_rate": 9.600193239934033e-06,
"loss": 2.4037,
"step": 34000
},
{
"epoch": 0.28,
"learning_rate": 9.591863932432658e-06,
"loss": 2.4054,
"step": 34500
},
{
"epoch": 0.29,
"learning_rate": 9.583534624931285e-06,
"loss": 2.4044,
"step": 35000
},
{
"epoch": 0.29,
"learning_rate": 9.57520531742991e-06,
"loss": 2.4103,
"step": 35500
},
{
"epoch": 0.29,
"learning_rate": 9.566876009928535e-06,
"loss": 2.4036,
"step": 36000
},
{
"epoch": 0.3,
"learning_rate": 9.55854670242716e-06,
"loss": 2.4054,
"step": 36500
},
{
"epoch": 0.3,
"learning_rate": 9.550217394925787e-06,
"loss": 2.4005,
"step": 37000
},
{
"epoch": 0.31,
"learning_rate": 9.541888087424413e-06,
"loss": 2.4017,
"step": 37500
},
{
"epoch": 0.31,
"learning_rate": 9.533558779923038e-06,
"loss": 2.3948,
"step": 38000
},
{
"epoch": 0.32,
"learning_rate": 9.525229472421663e-06,
"loss": 2.3888,
"step": 38500
},
{
"epoch": 0.32,
"learning_rate": 9.51690016492029e-06,
"loss": 2.3953,
"step": 39000
},
{
"epoch": 0.32,
"learning_rate": 9.508570857418915e-06,
"loss": 2.3939,
"step": 39500
},
{
"epoch": 0.33,
"learning_rate": 9.50024154991754e-06,
"loss": 2.4068,
"step": 40000
},
{
"epoch": 0.33,
"eval_accuracy": 0.5696406007169104,
"eval_loss": 2.3429617881774902,
"eval_runtime": 1888.2965,
"eval_samples_per_second": 163.299,
"eval_steps_per_second": 3.403,
"step": 40000
},
{
"epoch": 0.33,
"learning_rate": 9.491912242416167e-06,
"loss": 2.3953,
"step": 40500
},
{
"epoch": 0.34,
"learning_rate": 9.483582934914792e-06,
"loss": 2.3931,
"step": 41000
},
{
"epoch": 0.34,
"learning_rate": 9.475253627413418e-06,
"loss": 2.3936,
"step": 41500
},
{
"epoch": 0.34,
"learning_rate": 9.466924319912043e-06,
"loss": 2.3972,
"step": 42000
},
{
"epoch": 0.35,
"learning_rate": 9.458595012410668e-06,
"loss": 2.3896,
"step": 42500
},
{
"epoch": 0.35,
"learning_rate": 9.450265704909295e-06,
"loss": 2.3909,
"step": 43000
},
{
"epoch": 0.36,
"learning_rate": 9.44193639740792e-06,
"loss": 2.3827,
"step": 43500
},
{
"epoch": 0.36,
"learning_rate": 9.433607089906545e-06,
"loss": 2.39,
"step": 44000
},
{
"epoch": 0.36,
"learning_rate": 9.425277782405172e-06,
"loss": 2.3878,
"step": 44500
},
{
"epoch": 0.37,
"learning_rate": 9.416948474903798e-06,
"loss": 2.3894,
"step": 45000
},
{
"epoch": 0.37,
"learning_rate": 9.408619167402423e-06,
"loss": 2.3928,
"step": 45500
},
{
"epoch": 0.38,
"learning_rate": 9.400289859901048e-06,
"loss": 2.3804,
"step": 46000
},
{
"epoch": 0.38,
"learning_rate": 9.391960552399673e-06,
"loss": 2.3819,
"step": 46500
},
{
"epoch": 0.39,
"learning_rate": 9.3836312448983e-06,
"loss": 2.3814,
"step": 47000
},
{
"epoch": 0.39,
"learning_rate": 9.375301937396925e-06,
"loss": 2.3826,
"step": 47500
},
{
"epoch": 0.39,
"learning_rate": 9.366972629895552e-06,
"loss": 2.381,
"step": 48000
},
{
"epoch": 0.4,
"learning_rate": 9.358643322394177e-06,
"loss": 2.3846,
"step": 48500
},
{
"epoch": 0.4,
"learning_rate": 9.350314014892804e-06,
"loss": 2.3789,
"step": 49000
},
{
"epoch": 0.41,
"learning_rate": 9.341984707391429e-06,
"loss": 2.3813,
"step": 49500
},
{
"epoch": 0.41,
"learning_rate": 9.333655399890054e-06,
"loss": 2.3823,
"step": 50000
},
{
"epoch": 0.41,
"eval_accuracy": 0.5718559029214005,
"eval_loss": 2.3248536586761475,
"eval_runtime": 1901.4727,
"eval_samples_per_second": 162.167,
"eval_steps_per_second": 3.379,
"step": 50000
},
{
"epoch": 0.41,
"learning_rate": 9.325326092388679e-06,
"loss": 2.3785,
"step": 50500
},
{
"epoch": 0.42,
"learning_rate": 9.316996784887305e-06,
"loss": 2.3815,
"step": 51000
},
{
"epoch": 0.42,
"learning_rate": 9.30866747738593e-06,
"loss": 2.3797,
"step": 51500
},
{
"epoch": 0.43,
"learning_rate": 9.300338169884557e-06,
"loss": 2.3823,
"step": 52000
},
{
"epoch": 0.43,
"learning_rate": 9.292008862383182e-06,
"loss": 2.3773,
"step": 52500
},
{
"epoch": 0.43,
"learning_rate": 9.283679554881809e-06,
"loss": 2.3758,
"step": 53000
},
{
"epoch": 0.44,
"learning_rate": 9.275350247380434e-06,
"loss": 2.3766,
"step": 53500
},
{
"epoch": 0.44,
"learning_rate": 9.267020939879059e-06,
"loss": 2.3712,
"step": 54000
},
{
"epoch": 0.45,
"learning_rate": 9.258691632377685e-06,
"loss": 2.3737,
"step": 54500
},
{
"epoch": 0.45,
"learning_rate": 9.25036232487631e-06,
"loss": 2.3714,
"step": 55000
},
{
"epoch": 0.45,
"learning_rate": 9.242033017374937e-06,
"loss": 2.3737,
"step": 55500
},
{
"epoch": 0.46,
"learning_rate": 9.233703709873562e-06,
"loss": 2.3686,
"step": 56000
},
{
"epoch": 0.46,
"learning_rate": 9.225374402372189e-06,
"loss": 2.3706,
"step": 56500
},
{
"epoch": 0.47,
"learning_rate": 9.217045094870814e-06,
"loss": 2.364,
"step": 57000
},
{
"epoch": 0.47,
"learning_rate": 9.208715787369439e-06,
"loss": 2.3707,
"step": 57500
},
{
"epoch": 0.48,
"learning_rate": 9.200386479868064e-06,
"loss": 2.3669,
"step": 58000
},
{
"epoch": 0.48,
"learning_rate": 9.19205717236669e-06,
"loss": 2.3636,
"step": 58500
},
{
"epoch": 0.48,
"learning_rate": 9.183727864865315e-06,
"loss": 2.3722,
"step": 59000
},
{
"epoch": 0.49,
"learning_rate": 9.175398557363942e-06,
"loss": 2.3718,
"step": 59500
},
{
"epoch": 0.49,
"learning_rate": 9.167069249862567e-06,
"loss": 2.3729,
"step": 60000
},
{
"epoch": 0.49,
"eval_accuracy": 0.5732608497518056,
"eval_loss": 2.314063549041748,
"eval_runtime": 1886.4966,
"eval_samples_per_second": 163.454,
"eval_steps_per_second": 3.406,
"step": 60000
},
{
"epoch": 0.5,
"learning_rate": 9.158739942361194e-06,
"loss": 2.3651,
"step": 60500
},
{
"epoch": 0.5,
"learning_rate": 9.150410634859819e-06,
"loss": 2.3607,
"step": 61000
},
{
"epoch": 0.5,
"learning_rate": 9.142081327358444e-06,
"loss": 2.3685,
"step": 61500
},
{
"epoch": 0.51,
"learning_rate": 9.133752019857069e-06,
"loss": 2.3583,
"step": 62000
},
{
"epoch": 0.51,
"learning_rate": 9.125422712355696e-06,
"loss": 2.3659,
"step": 62500
},
{
"epoch": 0.52,
"learning_rate": 9.11709340485432e-06,
"loss": 2.366,
"step": 63000
},
{
"epoch": 0.52,
"learning_rate": 9.108764097352947e-06,
"loss": 2.3559,
"step": 63500
},
{
"epoch": 0.52,
"learning_rate": 9.100434789851572e-06,
"loss": 2.3576,
"step": 64000
},
{
"epoch": 0.53,
"learning_rate": 9.092105482350199e-06,
"loss": 2.3592,
"step": 64500
},
{
"epoch": 0.53,
"learning_rate": 9.083776174848824e-06,
"loss": 2.3564,
"step": 65000
},
{
"epoch": 0.54,
"learning_rate": 9.075446867347449e-06,
"loss": 2.3668,
"step": 65500
},
{
"epoch": 0.54,
"learning_rate": 9.067117559846076e-06,
"loss": 2.3609,
"step": 66000
},
{
"epoch": 0.54,
"learning_rate": 9.0587882523447e-06,
"loss": 2.3578,
"step": 66500
},
{
"epoch": 0.55,
"learning_rate": 9.050458944843327e-06,
"loss": 2.3497,
"step": 67000
},
{
"epoch": 0.55,
"learning_rate": 9.042129637341952e-06,
"loss": 2.3556,
"step": 67500
},
{
"epoch": 0.56,
"learning_rate": 9.033800329840577e-06,
"loss": 2.3559,
"step": 68000
},
{
"epoch": 0.56,
"learning_rate": 9.025471022339204e-06,
"loss": 2.3615,
"step": 68500
},
{
"epoch": 0.57,
"learning_rate": 9.017141714837829e-06,
"loss": 2.3548,
"step": 69000
},
{
"epoch": 0.57,
"learning_rate": 9.008812407336454e-06,
"loss": 2.358,
"step": 69500
},
{
"epoch": 0.57,
"learning_rate": 9.00048309983508e-06,
"loss": 2.3516,
"step": 70000
},
{
"epoch": 0.57,
"eval_accuracy": 0.5751267112637455,
"eval_loss": 2.2986373901367188,
"eval_runtime": 1879.9393,
"eval_samples_per_second": 164.024,
"eval_steps_per_second": 3.418,
"step": 70000
},
{
"epoch": 0.58,
"learning_rate": 8.992153792333706e-06,
"loss": 2.3482,
"step": 70500
},
{
"epoch": 0.58,
"learning_rate": 8.983824484832332e-06,
"loss": 2.3494,
"step": 71000
},
{
"epoch": 0.59,
"learning_rate": 8.975495177330957e-06,
"loss": 2.3546,
"step": 71500
},
{
"epoch": 0.59,
"learning_rate": 8.967165869829582e-06,
"loss": 2.3475,
"step": 72000
},
{
"epoch": 0.59,
"learning_rate": 8.958836562328209e-06,
"loss": 2.3514,
"step": 72500
},
{
"epoch": 0.6,
"learning_rate": 8.950507254826834e-06,
"loss": 2.3452,
"step": 73000
},
{
"epoch": 0.6,
"learning_rate": 8.942177947325459e-06,
"loss": 2.3511,
"step": 73500
},
{
"epoch": 0.61,
"learning_rate": 8.933848639824086e-06,
"loss": 2.349,
"step": 74000
},
{
"epoch": 0.61,
"learning_rate": 8.925519332322713e-06,
"loss": 2.3471,
"step": 74500
},
{
"epoch": 0.61,
"learning_rate": 8.917190024821338e-06,
"loss": 2.3513,
"step": 75000
},
{
"epoch": 0.62,
"learning_rate": 8.908860717319963e-06,
"loss": 2.3449,
"step": 75500
},
{
"epoch": 0.62,
"learning_rate": 8.900531409818588e-06,
"loss": 2.3421,
"step": 76000
},
{
"epoch": 0.63,
"learning_rate": 8.892202102317214e-06,
"loss": 2.3417,
"step": 76500
},
{
"epoch": 0.63,
"learning_rate": 8.88387279481584e-06,
"loss": 2.3402,
"step": 77000
},
{
"epoch": 0.63,
"learning_rate": 8.875543487314466e-06,
"loss": 2.3407,
"step": 77500
},
{
"epoch": 0.64,
"learning_rate": 8.867214179813091e-06,
"loss": 2.3362,
"step": 78000
},
{
"epoch": 0.64,
"learning_rate": 8.858884872311718e-06,
"loss": 2.3428,
"step": 78500
},
{
"epoch": 0.65,
"learning_rate": 8.850555564810343e-06,
"loss": 2.3359,
"step": 79000
},
{
"epoch": 0.65,
"learning_rate": 8.842226257308968e-06,
"loss": 2.3455,
"step": 79500
},
{
"epoch": 0.66,
"learning_rate": 8.833896949807593e-06,
"loss": 2.342,
"step": 80000
},
{
"epoch": 0.66,
"eval_accuracy": 0.5763929795108652,
"eval_loss": 2.287837266921997,
"eval_runtime": 1881.1252,
"eval_samples_per_second": 163.921,
"eval_steps_per_second": 3.416,
"step": 80000
},
{
"epoch": 0.66,
"learning_rate": 8.82556764230622e-06,
"loss": 2.3395,
"step": 80500
},
{
"epoch": 0.66,
"learning_rate": 8.817238334804844e-06,
"loss": 2.3416,
"step": 81000
},
{
"epoch": 0.67,
"learning_rate": 8.808909027303471e-06,
"loss": 2.3374,
"step": 81500
},
{
"epoch": 0.67,
"learning_rate": 8.800579719802096e-06,
"loss": 2.3336,
"step": 82000
},
{
"epoch": 0.68,
"learning_rate": 8.792250412300723e-06,
"loss": 2.3265,
"step": 82500
},
{
"epoch": 0.68,
"learning_rate": 8.783921104799348e-06,
"loss": 2.3378,
"step": 83000
},
{
"epoch": 0.68,
"learning_rate": 8.775591797297973e-06,
"loss": 2.337,
"step": 83500
},
{
"epoch": 0.69,
"learning_rate": 8.767262489796598e-06,
"loss": 2.3341,
"step": 84000
},
{
"epoch": 0.69,
"learning_rate": 8.758933182295224e-06,
"loss": 2.3307,
"step": 84500
},
{
"epoch": 0.7,
"learning_rate": 8.750603874793851e-06,
"loss": 2.331,
"step": 85000
},
{
"epoch": 0.7,
"learning_rate": 8.742274567292476e-06,
"loss": 2.3355,
"step": 85500
},
{
"epoch": 0.7,
"learning_rate": 8.733945259791101e-06,
"loss": 2.3284,
"step": 86000
},
{
"epoch": 0.71,
"learning_rate": 8.725615952289728e-06,
"loss": 2.3304,
"step": 86500
},
{
"epoch": 0.71,
"learning_rate": 8.717286644788353e-06,
"loss": 2.3335,
"step": 87000
},
{
"epoch": 0.72,
"learning_rate": 8.708957337286978e-06,
"loss": 2.3298,
"step": 87500
},
{
"epoch": 0.72,
"learning_rate": 8.700628029785605e-06,
"loss": 2.3258,
"step": 88000
},
{
"epoch": 0.73,
"learning_rate": 8.69229872228423e-06,
"loss": 2.3312,
"step": 88500
},
{
"epoch": 0.73,
"learning_rate": 8.683969414782856e-06,
"loss": 2.3273,
"step": 89000
},
{
"epoch": 0.73,
"learning_rate": 8.675640107281481e-06,
"loss": 2.3228,
"step": 89500
},
{
"epoch": 0.74,
"learning_rate": 8.667310799780106e-06,
"loss": 2.3265,
"step": 90000
},
{
"epoch": 0.74,
"eval_accuracy": 0.5782202967194991,
"eval_loss": 2.2734427452087402,
"eval_runtime": 1878.8142,
"eval_samples_per_second": 164.123,
"eval_steps_per_second": 3.42,
"step": 90000
},
{
"epoch": 0.74,
"learning_rate": 8.658981492278733e-06,
"loss": 2.3211,
"step": 90500
},
{
"epoch": 0.75,
"learning_rate": 8.650652184777358e-06,
"loss": 2.3245,
"step": 91000
},
{
"epoch": 0.75,
"learning_rate": 8.642322877275983e-06,
"loss": 2.3294,
"step": 91500
},
{
"epoch": 0.75,
"learning_rate": 8.63399356977461e-06,
"loss": 2.3317,
"step": 92000
},
{
"epoch": 0.76,
"learning_rate": 8.625664262273236e-06,
"loss": 2.3273,
"step": 92500
},
{
"epoch": 0.76,
"learning_rate": 8.617334954771861e-06,
"loss": 2.3451,
"step": 93000
},
{
"epoch": 0.77,
"learning_rate": 8.609005647270486e-06,
"loss": 2.3355,
"step": 93500
},
{
"epoch": 0.77,
"learning_rate": 8.600676339769111e-06,
"loss": 2.3318,
"step": 94000
},
{
"epoch": 0.77,
"learning_rate": 8.592347032267738e-06,
"loss": 2.3263,
"step": 94500
},
{
"epoch": 0.78,
"learning_rate": 8.584017724766363e-06,
"loss": 2.319,
"step": 95000
},
{
"epoch": 0.78,
"learning_rate": 8.57568841726499e-06,
"loss": 2.3219,
"step": 95500
},
{
"epoch": 0.79,
"learning_rate": 8.567359109763615e-06,
"loss": 2.3192,
"step": 96000
},
{
"epoch": 0.79,
"learning_rate": 8.559029802262241e-06,
"loss": 2.3167,
"step": 96500
},
{
"epoch": 0.79,
"learning_rate": 8.550700494760866e-06,
"loss": 2.3177,
"step": 97000
},
{
"epoch": 0.8,
"learning_rate": 8.542371187259491e-06,
"loss": 2.3187,
"step": 97500
},
{
"epoch": 0.8,
"learning_rate": 8.534041879758118e-06,
"loss": 2.319,
"step": 98000
},
{
"epoch": 0.81,
"learning_rate": 8.525712572256743e-06,
"loss": 2.3227,
"step": 98500
},
{
"epoch": 0.81,
"learning_rate": 8.517383264755368e-06,
"loss": 2.3193,
"step": 99000
},
{
"epoch": 0.82,
"learning_rate": 8.509053957253995e-06,
"loss": 2.3139,
"step": 99500
},
{
"epoch": 0.82,
"learning_rate": 8.500724649752622e-06,
"loss": 2.3158,
"step": 100000
},
{
"epoch": 0.82,
"eval_accuracy": 0.5795734098420514,
"eval_loss": 2.2642505168914795,
"eval_runtime": 1878.2718,
"eval_samples_per_second": 164.17,
"eval_steps_per_second": 3.421,
"step": 100000
},
{
"epoch": 0.82,
"learning_rate": 8.492395342251247e-06,
"loss": 2.3384,
"step": 100500
},
{
"epoch": 0.83,
"learning_rate": 8.484066034749872e-06,
"loss": 2.3293,
"step": 101000
},
{
"epoch": 0.83,
"learning_rate": 8.475736727248497e-06,
"loss": 2.3386,
"step": 101500
},
{
"epoch": 0.84,
"learning_rate": 8.467407419747123e-06,
"loss": 2.3319,
"step": 102000
},
{
"epoch": 0.84,
"learning_rate": 8.459078112245748e-06,
"loss": 2.3322,
"step": 102500
},
{
"epoch": 0.84,
"step": 102544,
"total_flos": 1.4583593352424325e+18,
"train_loss": 2.3798838982364097,
"train_runtime": 107983.5978,
"train_samples_per_second": 271.28,
"train_steps_per_second": 5.652
}
],
"max_steps": 610290,
"num_train_epochs": 5,
"total_flos": 1.4583593352424325e+18,
"trial_name": null,
"trial_params": null
}