mayot / trainer_state.json
AlekseyKorshuk's picture
huggingartists
c0ecdce
{
"best_metric": 1.9315155744552612,
"best_model_checkpoint": "output/mayot/checkpoint-147",
"epoch": 1.0,
"global_step": 147,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 0.00013680872155869608,
"loss": 2.7218,
"step": 5
},
{
"epoch": 0.07,
"learning_rate": 0.00013563934975719332,
"loss": 2.4506,
"step": 10
},
{
"epoch": 0.1,
"learning_rate": 0.00013370522424493186,
"loss": 2.3233,
"step": 15
},
{
"epoch": 0.14,
"learning_rate": 0.0001310284086258687,
"loss": 2.2628,
"step": 20
},
{
"epoch": 0.17,
"learning_rate": 0.0001276394387671491,
"loss": 2.169,
"step": 25
},
{
"epoch": 0.2,
"learning_rate": 0.00012357697446014183,
"loss": 2.1897,
"step": 30
},
{
"epoch": 0.24,
"learning_rate": 0.00011888735840752609,
"loss": 2.1417,
"step": 35
},
{
"epoch": 0.27,
"learning_rate": 0.00011362408756730135,
"loss": 2.091,
"step": 40
},
{
"epoch": 0.31,
"learning_rate": 0.00010784720288438084,
"loss": 2.0851,
"step": 45
},
{
"epoch": 0.34,
"learning_rate": 0.00010162260437142711,
"loss": 2.1085,
"step": 50
},
{
"epoch": 0.37,
"learning_rate": 9.502129935216876e-05,
"loss": 2.0363,
"step": 55
},
{
"epoch": 0.41,
"learning_rate": 8.811859244288883e-05,
"loss": 2.0587,
"step": 60
},
{
"epoch": 0.44,
"learning_rate": 8.099322651239897e-05,
"loss": 2.0127,
"step": 65
},
{
"epoch": 0.48,
"learning_rate": 7.372648442002871e-05,
"loss": 1.9725,
"step": 70
},
{
"epoch": 0.51,
"learning_rate": 6.640126177858445e-05,
"loss": 1.9979,
"step": 75
},
{
"epoch": 0.54,
"learning_rate": 5.910112131976733e-05,
"loss": 2.0089,
"step": 80
},
{
"epoch": 0.58,
"learning_rate": 5.190933964940903e-05,
"loss": 1.9777,
"step": 85
},
{
"epoch": 0.61,
"learning_rate": 4.4907957266698305e-05,
"loss": 2.0007,
"step": 90
},
{
"epoch": 0.65,
"learning_rate": 3.8176842684337195e-05,
"loss": 2.0053,
"step": 95
},
{
"epoch": 0.68,
"learning_rate": 3.179278132570837e-05,
"loss": 1.9833,
"step": 100
},
{
"epoch": 0.71,
"learning_rate": 2.5828599592490882e-05,
"loss": 1.983,
"step": 105
},
{
"epoch": 0.75,
"learning_rate": 2.035233409495375e-05,
"loss": 1.9408,
"step": 110
},
{
"epoch": 0.78,
"learning_rate": 1.5426455521961213e-05,
"loss": 1.9111,
"step": 115
},
{
"epoch": 0.82,
"learning_rate": 1.1107156004419736e-05,
"loss": 1.9171,
"step": 120
},
{
"epoch": 0.85,
"learning_rate": 7.4437081015931425e-06,
"loss": 1.898,
"step": 125
},
{
"epoch": 0.88,
"learning_rate": 4.477902722671392e-06,
"loss": 1.9822,
"step": 130
},
{
"epoch": 0.92,
"learning_rate": 2.24357239552258e-06,
"loss": 1.8845,
"step": 135
},
{
"epoch": 0.95,
"learning_rate": 7.662053209561833e-07,
"loss": 1.9681,
"step": 140
},
{
"epoch": 0.99,
"learning_rate": 6.26546151870577e-08,
"loss": 1.9478,
"step": 145
},
{
"epoch": 1.0,
"eval_loss": 1.9315155744552612,
"eval_runtime": 8.7552,
"eval_samples_per_second": 21.359,
"eval_steps_per_second": 2.741,
"step": 147
}
],
"max_steps": 147,
"num_train_epochs": 1,
"total_flos": 152855838720000.0,
"trial_name": null,
"trial_params": null
}