|
{ |
|
"best_metric": 0.8122448979591836, |
|
"best_model_checkpoint": "/hdd1/mujeen/retrieval_prf/output/labeler_multi.train.v4.1_nq.dev.v4.1_rlmulti_title_wq/checkpoint-300", |
|
"epoch": 1.0, |
|
"global_step": 3080, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 1.903896103896104e-05, |
|
"loss": 0.2844, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"eval_accuracy": 0.7714285714285715, |
|
"eval_loss": 0.5584465861320496, |
|
"eval_runtime": 3.5617, |
|
"eval_samples_per_second": 137.576, |
|
"eval_steps_per_second": 17.408, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 1.8064935064935067e-05, |
|
"loss": 0.2673, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"eval_accuracy": 0.8122448979591836, |
|
"eval_loss": 0.5982859134674072, |
|
"eval_runtime": 3.1189, |
|
"eval_samples_per_second": 157.106, |
|
"eval_steps_per_second": 19.879, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"learning_rate": 1.7090909090909092e-05, |
|
"loss": 0.2653, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"eval_accuracy": 0.8040816326530612, |
|
"eval_loss": 0.5716066956520081, |
|
"eval_runtime": 3.0903, |
|
"eval_samples_per_second": 158.56, |
|
"eval_steps_per_second": 20.063, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"learning_rate": 1.6116883116883118e-05, |
|
"loss": 0.2453, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"eval_accuracy": 0.7836734693877551, |
|
"eval_loss": 0.6890708208084106, |
|
"eval_runtime": 2.9422, |
|
"eval_samples_per_second": 166.541, |
|
"eval_steps_per_second": 21.073, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.24, |
|
"learning_rate": 1.514935064935065e-05, |
|
"loss": 0.2234, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 0.24, |
|
"eval_accuracy": 0.8, |
|
"eval_loss": 0.6545684337615967, |
|
"eval_runtime": 3.0903, |
|
"eval_samples_per_second": 158.559, |
|
"eval_steps_per_second": 20.063, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"learning_rate": 1.4175324675324675e-05, |
|
"loss": 0.2117, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"eval_accuracy": 0.7938775510204081, |
|
"eval_loss": 0.6074094772338867, |
|
"eval_runtime": 3.4718, |
|
"eval_samples_per_second": 141.137, |
|
"eval_steps_per_second": 17.858, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 0.34, |
|
"learning_rate": 1.3201298701298702e-05, |
|
"loss": 0.1867, |
|
"step": 1050 |
|
}, |
|
{ |
|
"epoch": 0.34, |
|
"eval_accuracy": 0.7938775510204081, |
|
"eval_loss": 0.7075100541114807, |
|
"eval_runtime": 3.2481, |
|
"eval_samples_per_second": 150.857, |
|
"eval_steps_per_second": 19.088, |
|
"step": 1050 |
|
}, |
|
{ |
|
"epoch": 0.39, |
|
"learning_rate": 1.2227272727272728e-05, |
|
"loss": 0.179, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 0.39, |
|
"eval_accuracy": 0.8040816326530612, |
|
"eval_loss": 0.6317723989486694, |
|
"eval_runtime": 3.0265, |
|
"eval_samples_per_second": 161.903, |
|
"eval_steps_per_second": 20.486, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 0.44, |
|
"learning_rate": 1.1253246753246754e-05, |
|
"loss": 0.1718, |
|
"step": 1350 |
|
}, |
|
{ |
|
"epoch": 0.44, |
|
"eval_accuracy": 0.7836734693877551, |
|
"eval_loss": 0.7868985533714294, |
|
"eval_runtime": 3.0224, |
|
"eval_samples_per_second": 162.125, |
|
"eval_steps_per_second": 20.514, |
|
"step": 1350 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"learning_rate": 1.027922077922078e-05, |
|
"loss": 0.1713, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"eval_accuracy": 0.7775510204081633, |
|
"eval_loss": 0.6507557034492493, |
|
"eval_runtime": 3.6832, |
|
"eval_samples_per_second": 133.036, |
|
"eval_steps_per_second": 16.833, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"learning_rate": 9.318181818181819e-06, |
|
"loss": 0.1502, |
|
"step": 1650 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"eval_accuracy": 0.7836734693877551, |
|
"eval_loss": 0.738645613193512, |
|
"eval_runtime": 3.2984, |
|
"eval_samples_per_second": 148.555, |
|
"eval_steps_per_second": 18.797, |
|
"step": 1650 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"learning_rate": 8.344155844155845e-06, |
|
"loss": 0.1541, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"eval_accuracy": 0.7877551020408163, |
|
"eval_loss": 0.8032358884811401, |
|
"eval_runtime": 3.6265, |
|
"eval_samples_per_second": 135.118, |
|
"eval_steps_per_second": 17.097, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 0.63, |
|
"learning_rate": 7.370129870129871e-06, |
|
"loss": 0.1358, |
|
"step": 1950 |
|
}, |
|
{ |
|
"epoch": 0.63, |
|
"eval_accuracy": 0.7959183673469388, |
|
"eval_loss": 0.7314993739128113, |
|
"eval_runtime": 3.7106, |
|
"eval_samples_per_second": 132.054, |
|
"eval_steps_per_second": 16.709, |
|
"step": 1950 |
|
}, |
|
{ |
|
"epoch": 0.68, |
|
"learning_rate": 6.3961038961038964e-06, |
|
"loss": 0.1251, |
|
"step": 2100 |
|
}, |
|
{ |
|
"epoch": 0.68, |
|
"eval_accuracy": 0.7816326530612245, |
|
"eval_loss": 0.9869228601455688, |
|
"eval_runtime": 3.3927, |
|
"eval_samples_per_second": 144.427, |
|
"eval_steps_per_second": 18.274, |
|
"step": 2100 |
|
}, |
|
{ |
|
"epoch": 0.73, |
|
"learning_rate": 5.422077922077923e-06, |
|
"loss": 0.1244, |
|
"step": 2250 |
|
}, |
|
{ |
|
"epoch": 0.73, |
|
"eval_accuracy": 0.7836734693877551, |
|
"eval_loss": 0.8370808362960815, |
|
"eval_runtime": 3.5148, |
|
"eval_samples_per_second": 139.41, |
|
"eval_steps_per_second": 17.64, |
|
"step": 2250 |
|
}, |
|
{ |
|
"epoch": 0.78, |
|
"learning_rate": 4.448051948051948e-06, |
|
"loss": 0.1223, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 0.78, |
|
"eval_accuracy": 0.7816326530612245, |
|
"eval_loss": 0.833113968372345, |
|
"eval_runtime": 3.1172, |
|
"eval_samples_per_second": 157.192, |
|
"eval_steps_per_second": 19.89, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 0.83, |
|
"learning_rate": 3.474025974025974e-06, |
|
"loss": 0.096, |
|
"step": 2550 |
|
}, |
|
{ |
|
"epoch": 0.83, |
|
"eval_accuracy": 0.7979591836734694, |
|
"eval_loss": 0.8976467847824097, |
|
"eval_runtime": 3.0168, |
|
"eval_samples_per_second": 162.426, |
|
"eval_steps_per_second": 20.552, |
|
"step": 2550 |
|
}, |
|
{ |
|
"epoch": 0.88, |
|
"learning_rate": 2.5e-06, |
|
"loss": 0.1113, |
|
"step": 2700 |
|
}, |
|
{ |
|
"epoch": 0.88, |
|
"eval_accuracy": 0.7755102040816326, |
|
"eval_loss": 0.96781325340271, |
|
"eval_runtime": 3.3837, |
|
"eval_samples_per_second": 144.814, |
|
"eval_steps_per_second": 18.323, |
|
"step": 2700 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"learning_rate": 1.525974025974026e-06, |
|
"loss": 0.094, |
|
"step": 2850 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"eval_accuracy": 0.7816326530612245, |
|
"eval_loss": 1.0002238750457764, |
|
"eval_runtime": 2.9984, |
|
"eval_samples_per_second": 163.421, |
|
"eval_steps_per_second": 20.678, |
|
"step": 2850 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"learning_rate": 5.51948051948052e-07, |
|
"loss": 0.1084, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"eval_accuracy": 0.7836734693877551, |
|
"eval_loss": 1.0199074745178223, |
|
"eval_runtime": 3.3069, |
|
"eval_samples_per_second": 148.174, |
|
"eval_steps_per_second": 18.748, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"step": 3080, |
|
"total_flos": 2.867459611823923e+16, |
|
"train_loss": 0.16962762838834292, |
|
"train_runtime": 2524.7018, |
|
"train_samples_per_second": 39.038, |
|
"train_steps_per_second": 1.22 |
|
} |
|
], |
|
"max_steps": 3080, |
|
"num_train_epochs": 1, |
|
"total_flos": 2.867459611823923e+16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|