byt5-sentfix / trainer_state.json
Wikidepia's picture
Initial commit
de5f1a8
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.362265145153057,
"global_step": 22000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 4.958833506232608e-05,
"loss": 0.3018,
"step": 500
},
{
"epoch": 0.02,
"learning_rate": 4.9176670124652144e-05,
"loss": 0.1198,
"step": 1000
},
{
"epoch": 0.02,
"learning_rate": 4.876500518697822e-05,
"loss": 0.1041,
"step": 1500
},
{
"epoch": 0.03,
"learning_rate": 4.835334024930429e-05,
"loss": 0.0934,
"step": 2000
},
{
"epoch": 0.04,
"learning_rate": 4.794167531163036e-05,
"loss": 0.0893,
"step": 2500
},
{
"epoch": 0.05,
"learning_rate": 4.753001037395643e-05,
"loss": 0.0827,
"step": 3000
},
{
"epoch": 0.06,
"learning_rate": 4.71183454362825e-05,
"loss": 0.0794,
"step": 3500
},
{
"epoch": 0.07,
"learning_rate": 4.670668049860858e-05,
"loss": 0.0773,
"step": 4000
},
{
"epoch": 0.07,
"learning_rate": 4.6295015560934644e-05,
"loss": 0.0761,
"step": 4500
},
{
"epoch": 0.08,
"learning_rate": 4.588335062326072e-05,
"loss": 0.0719,
"step": 5000
},
{
"epoch": 0.09,
"learning_rate": 4.547168568558679e-05,
"loss": 0.0702,
"step": 5500
},
{
"epoch": 0.1,
"learning_rate": 4.506002074791286e-05,
"loss": 0.0681,
"step": 6000
},
{
"epoch": 0.11,
"learning_rate": 4.4648355810238935e-05,
"loss": 0.0684,
"step": 6500
},
{
"epoch": 0.12,
"learning_rate": 4.423669087256501e-05,
"loss": 0.0651,
"step": 7000
},
{
"epoch": 0.12,
"learning_rate": 4.382502593489107e-05,
"loss": 0.0642,
"step": 7500
},
{
"epoch": 0.13,
"learning_rate": 4.3413360997217144e-05,
"loss": 0.0651,
"step": 8000
},
{
"epoch": 0.14,
"learning_rate": 4.300169605954322e-05,
"loss": 0.064,
"step": 8500
},
{
"epoch": 0.15,
"learning_rate": 4.2590031121869286e-05,
"loss": 0.061,
"step": 9000
},
{
"epoch": 0.16,
"learning_rate": 4.217836618419536e-05,
"loss": 0.0623,
"step": 9500
},
{
"epoch": 0.16,
"learning_rate": 4.1766701246521435e-05,
"loss": 0.0589,
"step": 10000
},
{
"epoch": 0.17,
"learning_rate": 4.13550363088475e-05,
"loss": 0.058,
"step": 10500
},
{
"epoch": 0.18,
"learning_rate": 4.094337137117358e-05,
"loss": 0.0581,
"step": 11000
},
{
"epoch": 0.19,
"learning_rate": 4.053170643349965e-05,
"loss": 0.0574,
"step": 11500
},
{
"epoch": 0.2,
"learning_rate": 4.012004149582572e-05,
"loss": 0.0582,
"step": 12000
},
{
"epoch": 0.21,
"learning_rate": 3.9708376558151786e-05,
"loss": 0.0561,
"step": 12500
},
{
"epoch": 0.21,
"learning_rate": 3.929671162047786e-05,
"loss": 0.0557,
"step": 13000
},
{
"epoch": 0.22,
"learning_rate": 3.8885046682803935e-05,
"loss": 0.0549,
"step": 13500
},
{
"epoch": 0.23,
"learning_rate": 3.847338174513e-05,
"loss": 0.0556,
"step": 14000
},
{
"epoch": 0.24,
"learning_rate": 3.806171680745608e-05,
"loss": 0.0535,
"step": 14500
},
{
"epoch": 0.25,
"learning_rate": 3.765005186978215e-05,
"loss": 0.0539,
"step": 15000
},
{
"epoch": 0.26,
"learning_rate": 3.723838693210822e-05,
"loss": 0.0544,
"step": 15500
},
{
"epoch": 0.26,
"learning_rate": 3.682672199443429e-05,
"loss": 0.0532,
"step": 16000
},
{
"epoch": 0.27,
"learning_rate": 3.641505705676037e-05,
"loss": 0.0541,
"step": 16500
},
{
"epoch": 0.28,
"learning_rate": 3.6003392119086435e-05,
"loss": 0.0523,
"step": 17000
},
{
"epoch": 0.29,
"learning_rate": 3.55917271814125e-05,
"loss": 0.0518,
"step": 17500
},
{
"epoch": 0.3,
"learning_rate": 3.518006224373858e-05,
"loss": 0.0518,
"step": 18000
},
{
"epoch": 0.3,
"learning_rate": 3.476839730606465e-05,
"loss": 0.0508,
"step": 18500
},
{
"epoch": 0.31,
"learning_rate": 3.435673236839072e-05,
"loss": 0.0522,
"step": 19000
},
{
"epoch": 0.32,
"learning_rate": 3.3945067430716793e-05,
"loss": 0.0518,
"step": 19500
},
{
"epoch": 0.33,
"learning_rate": 3.353340249304287e-05,
"loss": 0.0506,
"step": 20000
},
{
"epoch": 0.34,
"learning_rate": 3.3121737555368935e-05,
"loss": 0.0503,
"step": 20500
},
{
"epoch": 0.35,
"learning_rate": 3.271007261769501e-05,
"loss": 0.0502,
"step": 21000
},
{
"epoch": 0.35,
"learning_rate": 3.2298407680021084e-05,
"loss": 0.0497,
"step": 21500
},
{
"epoch": 0.36,
"learning_rate": 3.1886742742347145e-05,
"loss": 0.051,
"step": 22000
}
],
"max_steps": 60729,
"num_train_epochs": 1,
"total_flos": 8.854507190040883e+16,
"trial_name": null,
"trial_params": null
}