python-bytes-distilgpt2 / trainer_state.json
arvkevi's picture
End of training
b3b0af8
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"global_step": 7584,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.2,
"learning_rate": 4.6703586497890296e-05,
"loss": 3.3817,
"step": 500
},
{
"epoch": 0.4,
"learning_rate": 4.340717299578059e-05,
"loss": 3.2354,
"step": 1000
},
{
"epoch": 0.59,
"learning_rate": 4.011075949367088e-05,
"loss": 3.1946,
"step": 1500
},
{
"epoch": 0.79,
"learning_rate": 3.6814345991561184e-05,
"loss": 3.1541,
"step": 2000
},
{
"epoch": 0.99,
"learning_rate": 3.351793248945148e-05,
"loss": 3.1354,
"step": 2500
},
{
"epoch": 1.19,
"learning_rate": 3.0221518987341774e-05,
"loss": 3.0363,
"step": 3000
},
{
"epoch": 1.38,
"learning_rate": 2.6925105485232068e-05,
"loss": 3.0444,
"step": 3500
},
{
"epoch": 1.58,
"learning_rate": 2.3628691983122365e-05,
"loss": 3.0292,
"step": 4000
},
{
"epoch": 1.78,
"learning_rate": 2.033227848101266e-05,
"loss": 3.0086,
"step": 4500
},
{
"epoch": 1.98,
"learning_rate": 1.7035864978902955e-05,
"loss": 2.9992,
"step": 5000
},
{
"epoch": 2.18,
"learning_rate": 1.3739451476793249e-05,
"loss": 2.9701,
"step": 5500
},
{
"epoch": 2.37,
"learning_rate": 1.0443037974683544e-05,
"loss": 2.9505,
"step": 6000
},
{
"epoch": 2.57,
"learning_rate": 7.14662447257384e-06,
"loss": 2.9507,
"step": 6500
},
{
"epoch": 2.77,
"learning_rate": 3.850210970464136e-06,
"loss": 2.9395,
"step": 7000
},
{
"epoch": 2.97,
"learning_rate": 5.537974683544304e-07,
"loss": 2.9615,
"step": 7500
},
{
"epoch": 3.0,
"step": 7584,
"total_flos": 1981674556489728.0,
"train_loss": 3.0642820732502996,
"train_runtime": 1709.0811,
"train_samples_per_second": 17.75,
"train_steps_per_second": 4.437
}
],
"max_steps": 7584,
"num_train_epochs": 3,
"total_flos": 1981674556489728.0,
"trial_name": null,
"trial_params": null
}