|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 10.0, |
|
"global_step": 17610, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 1.8e-05, |
|
"loss": 0.0167, |
|
"step": 1761 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_EG_f1": 0.9978260869565218, |
|
"eval_ET_f1": 0.9987063389391979, |
|
"eval_TE_f1": 0.9969821673525378, |
|
"eval_loss": 0.005893823690712452, |
|
"eval_overall_accuracy": 0.998933354402876, |
|
"eval_overall_f1": 0.9976327358809605, |
|
"eval_overall_precision": 0.9975202885482417, |
|
"eval_overall_recall": 0.9977452085682075, |
|
"eval_runtime": 59.0341, |
|
"eval_samples_per_second": 53.037, |
|
"eval_steps_per_second": 3.32, |
|
"step": 1761 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"learning_rate": 1.6000000000000003e-05, |
|
"loss": 0.0048, |
|
"step": 3522 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_EG_f1": 0.9950846531949754, |
|
"eval_ET_f1": 0.9974059662775615, |
|
"eval_TE_f1": 0.9942763695829926, |
|
"eval_loss": 0.021830907091498375, |
|
"eval_overall_accuracy": 0.9975506656658634, |
|
"eval_overall_f1": 0.9951538374845036, |
|
"eval_overall_precision": 0.9948174853537629, |
|
"eval_overall_recall": 0.9954904171364148, |
|
"eval_runtime": 58.5084, |
|
"eval_samples_per_second": 53.514, |
|
"eval_steps_per_second": 3.35, |
|
"step": 3522 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"learning_rate": 1.4e-05, |
|
"loss": 0.003, |
|
"step": 5283 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_EG_f1": 0.9939989088925258, |
|
"eval_ET_f1": 0.9987063389391979, |
|
"eval_TE_f1": 0.9942638623326959, |
|
"eval_loss": 0.022889936342835426, |
|
"eval_overall_accuracy": 0.9971556117410026, |
|
"eval_overall_f1": 0.9949284345768061, |
|
"eval_overall_precision": 0.9945921586300135, |
|
"eval_overall_recall": 0.9952649379932357, |
|
"eval_runtime": 57.8629, |
|
"eval_samples_per_second": 54.111, |
|
"eval_steps_per_second": 3.387, |
|
"step": 5283 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"learning_rate": 1.2e-05, |
|
"loss": 0.002, |
|
"step": 7044 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_EG_f1": 0.9961977186311787, |
|
"eval_ET_f1": 0.9987046632124352, |
|
"eval_TE_f1": 0.9958847736625515, |
|
"eval_loss": 0.017046082764863968, |
|
"eval_overall_accuracy": 0.9982222573381266, |
|
"eval_overall_f1": 0.9965054672528464, |
|
"eval_overall_precision": 0.9963931469792606, |
|
"eval_overall_recall": 0.9966178128523112, |
|
"eval_runtime": 57.0371, |
|
"eval_samples_per_second": 54.894, |
|
"eval_steps_per_second": 3.436, |
|
"step": 7044 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"learning_rate": 1e-05, |
|
"loss": 0.0013, |
|
"step": 8805 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_EG_f1": 0.9970068027210883, |
|
"eval_ET_f1": 0.9987046632124352, |
|
"eval_TE_f1": 0.9961643835616438, |
|
"eval_loss": 0.011512575671076775, |
|
"eval_overall_accuracy": 0.9984987950855292, |
|
"eval_overall_f1": 0.9969556883526891, |
|
"eval_overall_precision": 0.9970681100586378, |
|
"eval_overall_recall": 0.9968432919954904, |
|
"eval_runtime": 57.8352, |
|
"eval_samples_per_second": 54.137, |
|
"eval_steps_per_second": 3.389, |
|
"step": 8805 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"learning_rate": 8.000000000000001e-06, |
|
"loss": 0.0015, |
|
"step": 10566 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_EG_f1": 0.9948271167982575, |
|
"eval_ET_f1": 0.9980557355800389, |
|
"eval_TE_f1": 0.9948073244055754, |
|
"eval_loss": 0.01284080371260643, |
|
"eval_overall_accuracy": 0.9979457195907241, |
|
"eval_overall_f1": 0.9953802816901409, |
|
"eval_overall_precision": 0.9948198198198198, |
|
"eval_overall_recall": 0.9959413754227734, |
|
"eval_runtime": 57.2307, |
|
"eval_samples_per_second": 54.708, |
|
"eval_steps_per_second": 3.425, |
|
"step": 10566 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"learning_rate": 6e-06, |
|
"loss": 0.0003, |
|
"step": 12327 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_EG_f1": 0.9953615279672577, |
|
"eval_ET_f1": 0.9993527508090615, |
|
"eval_TE_f1": 0.9959049959049959, |
|
"eval_loss": 0.021282300353050232, |
|
"eval_overall_accuracy": 0.9981037411606685, |
|
"eval_overall_f1": 0.9962808520229911, |
|
"eval_overall_precision": 0.9959441189725101, |
|
"eval_overall_recall": 0.9966178128523112, |
|
"eval_runtime": 57.2264, |
|
"eval_samples_per_second": 54.713, |
|
"eval_steps_per_second": 3.425, |
|
"step": 12327 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"learning_rate": 4.000000000000001e-06, |
|
"loss": 0.0005, |
|
"step": 14088 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_EG_f1": 0.9961852861035423, |
|
"eval_ET_f1": 0.9993527508090615, |
|
"eval_TE_f1": 0.9961706783369804, |
|
"eval_loss": 0.01351018063724041, |
|
"eval_overall_accuracy": 0.9983802789080709, |
|
"eval_overall_f1": 0.9967309209784693, |
|
"eval_overall_precision": 0.9966185752930569, |
|
"eval_overall_recall": 0.9968432919954904, |
|
"eval_runtime": 57.0797, |
|
"eval_samples_per_second": 54.853, |
|
"eval_steps_per_second": 3.434, |
|
"step": 14088 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"learning_rate": 2.0000000000000003e-06, |
|
"loss": 0.0002, |
|
"step": 15849 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_EG_f1": 0.9972781709308656, |
|
"eval_ET_f1": 0.9993527508090615, |
|
"eval_TE_f1": 0.9969871268145714, |
|
"eval_loss": 0.014631124213337898, |
|
"eval_overall_accuracy": 0.9986568166554735, |
|
"eval_overall_f1": 0.9975197294250282, |
|
"eval_overall_precision": 0.9975197294250282, |
|
"eval_overall_recall": 0.9975197294250282, |
|
"eval_runtime": 57.0354, |
|
"eval_samples_per_second": 54.896, |
|
"eval_steps_per_second": 3.436, |
|
"step": 15849 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"learning_rate": 0.0, |
|
"loss": 0.0001, |
|
"step": 17610 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_EG_f1": 0.9970051728832018, |
|
"eval_ET_f1": 0.9993527508090615, |
|
"eval_TE_f1": 0.9967141292442497, |
|
"eval_loss": 0.014984946697950363, |
|
"eval_overall_accuracy": 0.9984987950855292, |
|
"eval_overall_f1": 0.9972942502818489, |
|
"eval_overall_precision": 0.9972942502818489, |
|
"eval_overall_recall": 0.9972942502818489, |
|
"eval_runtime": 56.8354, |
|
"eval_samples_per_second": 55.089, |
|
"eval_steps_per_second": 3.449, |
|
"step": 17610 |
|
} |
|
], |
|
"max_steps": 17610, |
|
"num_train_epochs": 10, |
|
"total_flos": 3085811947058688.0, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|