gro-ner / trainer_state.json
mirikwa's picture
Upload trainer_state.json
2d5888d
raw
history blame
3.51 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 5.0,
"global_step": 36725,
"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": 0.0085,
"step": 7345
},
{
"epoch": 1.0,
"eval_EG_f1": 0.9962079135863944,
"eval_ET_f1": 0.9989276960784315,
"eval_TE_f1": 0.9956351941189985,
"eval_loss": 0.002211587270721793,
"eval_overall_accuracy": 0.9995236481225324,
"eval_overall_f1": 0.9969235482594433,
"eval_overall_precision": 0.9969871827605576,
"eval_overall_recall": 0.9968599218809834,
"eval_runtime": 329.5116,
"eval_samples_per_second": 39.625,
"eval_steps_per_second": 2.479,
"step": 7345
},
{
"epoch": 2.0,
"learning_rate": 1.6000000000000003e-05,
"loss": 0.0014,
"step": 14690
},
{
"epoch": 2.0,
"eval_EG_f1": 0.9987363584147042,
"eval_ET_f1": 0.9995787861382348,
"eval_TE_f1": 0.9985064910197986,
"eval_loss": 0.0008891239413060248,
"eval_overall_accuracy": 0.9998259483524637,
"eval_overall_f1": 0.9989405562717792,
"eval_overall_precision": 0.998927805575411,
"eval_overall_recall": 0.9989533072936612,
"eval_runtime": 314.9732,
"eval_samples_per_second": 41.454,
"eval_steps_per_second": 2.594,
"step": 14690
},
{
"epoch": 3.0,
"learning_rate": 1.4e-05,
"loss": 0.0008,
"step": 22035
},
{
"epoch": 3.0,
"eval_EG_f1": 0.9983534367221903,
"eval_ET_f1": 0.9996170636440224,
"eval_TE_f1": 0.9983917904732731,
"eval_loss": 0.0008875366183929145,
"eval_overall_accuracy": 0.999835108965492,
"eval_overall_f1": 0.9987874146403727,
"eval_overall_precision": 0.9987491703681013,
"eval_overall_recall": 0.9988256618416685,
"eval_runtime": 316.3291,
"eval_samples_per_second": 41.277,
"eval_steps_per_second": 2.583,
"step": 22035
},
{
"epoch": 4.0,
"learning_rate": 1.2e-05,
"loss": 0.0005,
"step": 29380
},
{
"epoch": 4.0,
"eval_EG_f1": 0.9991574754901962,
"eval_ET_f1": 0.9996553968679404,
"eval_TE_f1": 0.9989277782032626,
"eval_loss": 0.000532182864844799,
"eval_overall_accuracy": 0.9998717514176049,
"eval_overall_f1": 0.9992469014462045,
"eval_overall_precision": 0.9992341468395793,
"eval_overall_recall": 0.9992596563784433,
"eval_runtime": 328.7608,
"eval_samples_per_second": 39.716,
"eval_steps_per_second": 2.485,
"step": 29380
},
{
"epoch": 5.0,
"learning_rate": 1e-05,
"loss": 0.0003,
"step": 36725
},
{
"epoch": 5.0,
"eval_EG_f1": 0.9992341272880447,
"eval_ET_f1": 0.999846825457609,
"eval_TE_f1": 0.9992724487842237,
"eval_loss": 0.00026392185827717185,
"eval_overall_accuracy": 0.9999404560153166,
"eval_overall_f1": 0.999451131562488,
"eval_overall_precision": 0.9994383743490248,
"eval_overall_recall": 0.9994638891016313,
"eval_runtime": 321.322,
"eval_samples_per_second": 40.635,
"eval_steps_per_second": 2.543,
"step": 36725
}
],
"max_steps": 73450,
"num_train_epochs": 10,
"total_flos": 8945844807242016.0,
"trial_name": null,
"trial_params": null
}