es-bert-xnli / trainer_state.json
WillHeld's picture
End of training
4d721fc
raw
history blame
6.57 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.0,
"global_step": 24544,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04,
"learning_rate": 4.898142112125163e-05,
"loss": 0.7857,
"step": 500
},
{
"epoch": 0.08,
"learning_rate": 4.7962842242503265e-05,
"loss": 0.6857,
"step": 1000
},
{
"epoch": 0.12,
"learning_rate": 4.694426336375489e-05,
"loss": 0.6523,
"step": 1500
},
{
"epoch": 0.16,
"learning_rate": 4.592568448500652e-05,
"loss": 0.6259,
"step": 2000
},
{
"epoch": 0.2,
"learning_rate": 4.490710560625815e-05,
"loss": 0.6288,
"step": 2500
},
{
"epoch": 0.24,
"learning_rate": 4.3888526727509784e-05,
"loss": 0.6145,
"step": 3000
},
{
"epoch": 0.29,
"learning_rate": 4.286994784876141e-05,
"loss": 0.6104,
"step": 3500
},
{
"epoch": 0.33,
"learning_rate": 4.185136897001304e-05,
"loss": 0.5997,
"step": 4000
},
{
"epoch": 0.37,
"learning_rate": 4.0832790091264675e-05,
"loss": 0.5836,
"step": 4500
},
{
"epoch": 0.41,
"learning_rate": 3.98142112125163e-05,
"loss": 0.5834,
"step": 5000
},
{
"epoch": 0.45,
"learning_rate": 3.879563233376793e-05,
"loss": 0.5685,
"step": 5500
},
{
"epoch": 0.49,
"learning_rate": 3.777705345501956e-05,
"loss": 0.567,
"step": 6000
},
{
"epoch": 0.53,
"learning_rate": 3.675847457627119e-05,
"loss": 0.5703,
"step": 6500
},
{
"epoch": 0.57,
"learning_rate": 3.5739895697522816e-05,
"loss": 0.5648,
"step": 7000
},
{
"epoch": 0.61,
"learning_rate": 3.4721316818774444e-05,
"loss": 0.5652,
"step": 7500
},
{
"epoch": 0.65,
"learning_rate": 3.370273794002607e-05,
"loss": 0.5497,
"step": 8000
},
{
"epoch": 0.69,
"learning_rate": 3.2684159061277707e-05,
"loss": 0.5488,
"step": 8500
},
{
"epoch": 0.73,
"learning_rate": 3.1665580182529335e-05,
"loss": 0.5484,
"step": 9000
},
{
"epoch": 0.77,
"learning_rate": 3.064700130378096e-05,
"loss": 0.54,
"step": 9500
},
{
"epoch": 0.81,
"learning_rate": 2.9628422425032598e-05,
"loss": 0.5385,
"step": 10000
},
{
"epoch": 0.86,
"learning_rate": 2.8609843546284226e-05,
"loss": 0.5391,
"step": 10500
},
{
"epoch": 0.9,
"learning_rate": 2.7591264667535854e-05,
"loss": 0.5348,
"step": 11000
},
{
"epoch": 0.94,
"learning_rate": 2.6572685788787482e-05,
"loss": 0.534,
"step": 11500
},
{
"epoch": 0.98,
"learning_rate": 2.5554106910039117e-05,
"loss": 0.5347,
"step": 12000
},
{
"epoch": 1.02,
"learning_rate": 2.4535528031290745e-05,
"loss": 0.4633,
"step": 12500
},
{
"epoch": 1.06,
"learning_rate": 2.3516949152542376e-05,
"loss": 0.3967,
"step": 13000
},
{
"epoch": 1.1,
"learning_rate": 2.2498370273794004e-05,
"loss": 0.3887,
"step": 13500
},
{
"epoch": 1.14,
"learning_rate": 2.1479791395045636e-05,
"loss": 0.3927,
"step": 14000
},
{
"epoch": 1.18,
"learning_rate": 2.0461212516297264e-05,
"loss": 0.376,
"step": 14500
},
{
"epoch": 1.22,
"learning_rate": 1.944263363754889e-05,
"loss": 0.3855,
"step": 15000
},
{
"epoch": 1.26,
"learning_rate": 1.842405475880052e-05,
"loss": 0.3765,
"step": 15500
},
{
"epoch": 1.3,
"learning_rate": 1.740547588005215e-05,
"loss": 0.3804,
"step": 16000
},
{
"epoch": 1.34,
"learning_rate": 1.6386897001303783e-05,
"loss": 0.3763,
"step": 16500
},
{
"epoch": 1.39,
"learning_rate": 1.536831812255541e-05,
"loss": 0.3846,
"step": 17000
},
{
"epoch": 1.43,
"learning_rate": 1.4349739243807042e-05,
"loss": 0.372,
"step": 17500
},
{
"epoch": 1.47,
"learning_rate": 1.333116036505867e-05,
"loss": 0.3704,
"step": 18000
},
{
"epoch": 1.51,
"learning_rate": 1.23125814863103e-05,
"loss": 0.3683,
"step": 18500
},
{
"epoch": 1.55,
"learning_rate": 1.1294002607561931e-05,
"loss": 0.3812,
"step": 19000
},
{
"epoch": 1.59,
"learning_rate": 1.027542372881356e-05,
"loss": 0.3731,
"step": 19500
},
{
"epoch": 1.63,
"learning_rate": 9.25684485006519e-06,
"loss": 0.3745,
"step": 20000
},
{
"epoch": 1.67,
"learning_rate": 8.238265971316819e-06,
"loss": 0.3662,
"step": 20500
},
{
"epoch": 1.71,
"learning_rate": 7.219687092568449e-06,
"loss": 0.3588,
"step": 21000
},
{
"epoch": 1.75,
"learning_rate": 6.2011082138200785e-06,
"loss": 0.375,
"step": 21500
},
{
"epoch": 1.79,
"learning_rate": 5.182529335071708e-06,
"loss": 0.3599,
"step": 22000
},
{
"epoch": 1.83,
"learning_rate": 4.163950456323338e-06,
"loss": 0.3732,
"step": 22500
},
{
"epoch": 1.87,
"learning_rate": 3.1453715775749674e-06,
"loss": 0.3631,
"step": 23000
},
{
"epoch": 1.91,
"learning_rate": 2.126792698826597e-06,
"loss": 0.3546,
"step": 23500
},
{
"epoch": 1.96,
"learning_rate": 1.108213820078227e-06,
"loss": 0.3589,
"step": 24000
},
{
"epoch": 2.0,
"learning_rate": 8.963494132985658e-08,
"loss": 0.3586,
"step": 24500
},
{
"epoch": 2.0,
"step": 24544,
"total_flos": 5.166258268431053e+16,
"train_loss": 0.4794288249261376,
"train_runtime": 6378.689,
"train_samples_per_second": 123.129,
"train_steps_per_second": 3.848
}
],
"max_steps": 24544,
"num_train_epochs": 2,
"total_flos": 5.166258268431053e+16,
"trial_name": null,
"trial_params": null
}