mobilebert_sst2 / trainer_state.json
Alireza1044's picture
End of training
9d2338f
raw
history blame
5.71 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 10.0,
"global_step": 21050,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.24,
"learning_rate": 9.7624703087886e-06,
"loss": 221692.304,
"step": 500
},
{
"epoch": 0.48,
"learning_rate": 9.524940617577197e-06,
"loss": 0.2815,
"step": 1000
},
{
"epoch": 0.71,
"learning_rate": 9.287410926365797e-06,
"loss": 0.2438,
"step": 1500
},
{
"epoch": 0.95,
"learning_rate": 9.049881235154396e-06,
"loss": 0.2325,
"step": 2000
},
{
"epoch": 1.19,
"learning_rate": 8.812351543942995e-06,
"loss": 0.1994,
"step": 2500
},
{
"epoch": 1.43,
"learning_rate": 8.574821852731592e-06,
"loss": 0.1864,
"step": 3000
},
{
"epoch": 1.66,
"learning_rate": 8.337292161520191e-06,
"loss": 0.18,
"step": 3500
},
{
"epoch": 1.9,
"learning_rate": 8.099762470308789e-06,
"loss": 0.1788,
"step": 4000
},
{
"epoch": 2.14,
"learning_rate": 7.862232779097388e-06,
"loss": 0.1615,
"step": 4500
},
{
"epoch": 2.38,
"learning_rate": 7.624703087885986e-06,
"loss": 0.1516,
"step": 5000
},
{
"epoch": 2.61,
"learning_rate": 7.387173396674585e-06,
"loss": 0.1486,
"step": 5500
},
{
"epoch": 2.85,
"learning_rate": 7.149643705463184e-06,
"loss": 0.1462,
"step": 6000
},
{
"epoch": 3.09,
"learning_rate": 6.912114014251782e-06,
"loss": 0.134,
"step": 6500
},
{
"epoch": 3.33,
"learning_rate": 6.674584323040381e-06,
"loss": 0.1328,
"step": 7000
},
{
"epoch": 3.56,
"learning_rate": 6.437054631828979e-06,
"loss": 0.1229,
"step": 7500
},
{
"epoch": 3.8,
"learning_rate": 6.199524940617578e-06,
"loss": 0.1204,
"step": 8000
},
{
"epoch": 4.04,
"learning_rate": 5.961995249406176e-06,
"loss": 0.1323,
"step": 8500
},
{
"epoch": 4.28,
"learning_rate": 5.724465558194775e-06,
"loss": 0.1131,
"step": 9000
},
{
"epoch": 4.51,
"learning_rate": 5.486935866983374e-06,
"loss": 0.122,
"step": 9500
},
{
"epoch": 4.75,
"learning_rate": 5.2494061757719716e-06,
"loss": 0.1081,
"step": 10000
},
{
"epoch": 4.99,
"learning_rate": 5.011876484560571e-06,
"loss": 0.1061,
"step": 10500
},
{
"epoch": 5.23,
"learning_rate": 4.774346793349169e-06,
"loss": 0.0975,
"step": 11000
},
{
"epoch": 5.46,
"learning_rate": 4.536817102137768e-06,
"loss": 0.094,
"step": 11500
},
{
"epoch": 5.7,
"learning_rate": 4.299287410926366e-06,
"loss": 0.1133,
"step": 12000
},
{
"epoch": 5.94,
"learning_rate": 4.061757719714965e-06,
"loss": 0.0999,
"step": 12500
},
{
"epoch": 6.18,
"learning_rate": 3.824228028503563e-06,
"loss": 0.0899,
"step": 13000
},
{
"epoch": 6.41,
"learning_rate": 3.5866983372921617e-06,
"loss": 0.0921,
"step": 13500
},
{
"epoch": 6.65,
"learning_rate": 3.3491686460807604e-06,
"loss": 0.083,
"step": 14000
},
{
"epoch": 6.89,
"learning_rate": 3.1116389548693587e-06,
"loss": 0.0909,
"step": 14500
},
{
"epoch": 7.13,
"learning_rate": 2.874109263657958e-06,
"loss": 0.0772,
"step": 15000
},
{
"epoch": 7.36,
"learning_rate": 2.636579572446556e-06,
"loss": 0.1717,
"step": 15500
},
{
"epoch": 7.6,
"learning_rate": 2.3990498812351544e-06,
"loss": 0.0863,
"step": 16000
},
{
"epoch": 7.84,
"learning_rate": 2.161520190023753e-06,
"loss": 0.0827,
"step": 16500
},
{
"epoch": 8.08,
"learning_rate": 1.9239904988123518e-06,
"loss": 0.0726,
"step": 17000
},
{
"epoch": 8.31,
"learning_rate": 1.6864608076009503e-06,
"loss": 0.081,
"step": 17500
},
{
"epoch": 8.55,
"learning_rate": 1.448931116389549e-06,
"loss": 0.0779,
"step": 18000
},
{
"epoch": 8.79,
"learning_rate": 1.2114014251781475e-06,
"loss": 0.077,
"step": 18500
},
{
"epoch": 9.03,
"learning_rate": 9.73871733966746e-07,
"loss": 0.0747,
"step": 19000
},
{
"epoch": 9.26,
"learning_rate": 7.363420427553445e-07,
"loss": 0.0676,
"step": 19500
},
{
"epoch": 9.5,
"learning_rate": 4.988123515439431e-07,
"loss": 0.0989,
"step": 20000
},
{
"epoch": 9.74,
"learning_rate": 2.6128266033254157e-07,
"loss": 0.0746,
"step": 20500
},
{
"epoch": 9.98,
"learning_rate": 2.3752969121140145e-08,
"loss": 0.0738,
"step": 21000
},
{
"epoch": 10.0,
"step": 21050,
"total_flos": 1.05583904453376e+16,
"train_loss": 5265.971264971169,
"train_runtime": 8516.6719,
"train_samples_per_second": 79.079,
"train_steps_per_second": 2.472
}
],
"max_steps": 21050,
"num_train_epochs": 10,
"total_flos": 1.05583904453376e+16,
"trial_name": null,
"trial_params": null
}