Ubuntu
added finetuned categorical classification model + more evaluated dataset modifications
2e0c701
raw
history blame
4.15 kB
{
"best_metric": 0.21237443387508392,
"best_model_checkpoint": "finetuned_entity_categorical_classification/checkpoint-4728",
"epoch": 6.0,
"eval_steps": 500,
"global_step": 9456,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.32,
"learning_rate": 1.957698815566836e-05,
"loss": 1.5567,
"step": 500
},
{
"epoch": 0.63,
"learning_rate": 1.915397631133672e-05,
"loss": 0.3944,
"step": 1000
},
{
"epoch": 0.95,
"learning_rate": 1.873096446700508e-05,
"loss": 0.2773,
"step": 1500
},
{
"epoch": 1.0,
"eval_accuracy": 0.9374900840869427,
"eval_loss": 0.2187376469373703,
"eval_runtime": 2.2114,
"eval_samples_per_second": 2850.256,
"eval_steps_per_second": 178.169,
"step": 1576
},
{
"epoch": 1.27,
"learning_rate": 1.830795262267344e-05,
"loss": 0.1997,
"step": 2000
},
{
"epoch": 1.59,
"learning_rate": 1.7884940778341796e-05,
"loss": 0.202,
"step": 2500
},
{
"epoch": 1.9,
"learning_rate": 1.7461928934010152e-05,
"loss": 0.1797,
"step": 3000
},
{
"epoch": 2.0,
"eval_accuracy": 0.9452641599238458,
"eval_loss": 0.22844311594963074,
"eval_runtime": 2.2403,
"eval_samples_per_second": 2813.437,
"eval_steps_per_second": 175.868,
"step": 3152
},
{
"epoch": 2.22,
"learning_rate": 1.7038917089678512e-05,
"loss": 0.148,
"step": 3500
},
{
"epoch": 2.54,
"learning_rate": 1.661590524534687e-05,
"loss": 0.1357,
"step": 4000
},
{
"epoch": 2.86,
"learning_rate": 1.619289340101523e-05,
"loss": 0.1525,
"step": 4500
},
{
"epoch": 3.0,
"eval_accuracy": 0.9481199428843408,
"eval_loss": 0.21237443387508392,
"eval_runtime": 2.2281,
"eval_samples_per_second": 2828.834,
"eval_steps_per_second": 176.83,
"step": 4728
},
{
"epoch": 3.17,
"learning_rate": 1.576988155668359e-05,
"loss": 0.1218,
"step": 5000
},
{
"epoch": 3.49,
"learning_rate": 1.5346869712351946e-05,
"loss": 0.1147,
"step": 5500
},
{
"epoch": 3.81,
"learning_rate": 1.4923857868020306e-05,
"loss": 0.1195,
"step": 6000
},
{
"epoch": 4.0,
"eval_accuracy": 0.94859590671109,
"eval_loss": 0.2216227501630783,
"eval_runtime": 2.2255,
"eval_samples_per_second": 2832.161,
"eval_steps_per_second": 177.038,
"step": 6304
},
{
"epoch": 4.12,
"learning_rate": 1.4500846023688663e-05,
"loss": 0.114,
"step": 6500
},
{
"epoch": 4.44,
"learning_rate": 1.4077834179357023e-05,
"loss": 0.104,
"step": 7000
},
{
"epoch": 4.76,
"learning_rate": 1.3654822335025382e-05,
"loss": 0.0936,
"step": 7500
},
{
"epoch": 5.0,
"eval_accuracy": 0.9462160875773441,
"eval_loss": 0.24313929677009583,
"eval_runtime": 2.1752,
"eval_samples_per_second": 2897.6,
"eval_steps_per_second": 181.129,
"step": 7880
},
{
"epoch": 5.08,
"learning_rate": 1.323181049069374e-05,
"loss": 0.099,
"step": 8000
},
{
"epoch": 5.39,
"learning_rate": 1.28087986463621e-05,
"loss": 0.0929,
"step": 8500
},
{
"epoch": 5.71,
"learning_rate": 1.2385786802030457e-05,
"loss": 0.0872,
"step": 9000
},
{
"epoch": 6.0,
"eval_accuracy": 0.9465333967951769,
"eval_loss": 0.25265371799468994,
"eval_runtime": 2.2222,
"eval_samples_per_second": 2836.351,
"eval_steps_per_second": 177.3,
"step": 9456
}
],
"logging_steps": 500,
"max_steps": 23640,
"num_train_epochs": 15,
"save_steps": 500,
"total_flos": 603345335922576.0,
"trial_name": null,
"trial_params": null
}