10k_MT5_small_sum-de_GNAD / trainer_state.json
Einmalumdiewelt's picture
End of training
dca79fe
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 10.0,
"global_step": 7000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.71,
"learning_rate": 4.642857142857143e-05,
"loss": 2.8035,
"step": 500
},
{
"epoch": 1.43,
"learning_rate": 4.2857142857142856e-05,
"loss": 2.7688,
"step": 1000
},
{
"epoch": 2.14,
"learning_rate": 3.928571428571429e-05,
"loss": 2.7244,
"step": 1500
},
{
"epoch": 2.86,
"learning_rate": 3.571428571428572e-05,
"loss": 2.7084,
"step": 2000
},
{
"epoch": 3.57,
"learning_rate": 3.2142857142857144e-05,
"loss": 2.6641,
"step": 2500
},
{
"epoch": 4.29,
"learning_rate": 2.857142857142857e-05,
"loss": 2.6529,
"step": 3000
},
{
"epoch": 5.0,
"learning_rate": 2.5e-05,
"loss": 2.6412,
"step": 3500
},
{
"epoch": 5.71,
"learning_rate": 2.1428571428571428e-05,
"loss": 2.6125,
"step": 4000
},
{
"epoch": 6.43,
"learning_rate": 1.785714285714286e-05,
"loss": 2.6094,
"step": 4500
},
{
"epoch": 7.14,
"learning_rate": 1.4285714285714285e-05,
"loss": 2.5987,
"step": 5000
},
{
"epoch": 7.86,
"learning_rate": 1.0714285714285714e-05,
"loss": 2.5816,
"step": 5500
},
{
"epoch": 8.57,
"learning_rate": 7.142857142857143e-06,
"loss": 2.5776,
"step": 6000
},
{
"epoch": 9.29,
"learning_rate": 3.5714285714285714e-06,
"loss": 2.5766,
"step": 6500
},
{
"epoch": 10.0,
"learning_rate": 0.0,
"loss": 2.5662,
"step": 7000
},
{
"epoch": 10.0,
"step": 7000,
"total_flos": 3.701008317136896e+16,
"train_loss": 2.648989501953125,
"train_runtime": 4314.4712,
"train_samples_per_second": 16.224,
"train_steps_per_second": 1.622
}
],
"max_steps": 7000,
"num_train_epochs": 10,
"total_flos": 3.701008317136896e+16,
"trial_name": null,
"trial_params": null
}