josecannete's picture
adding model finetuned on sqac
b24fad9
{
"best_metric": 76.26399788082415,
"best_model_checkpoint": "/home/jcanete/ft-data/all_results/sqac/albeto_base_8/epochs_3_bs_16_lr_5e-5/checkpoint-2000",
"epoch": 3.0,
"global_step": 3114,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.19,
"eval_exact_match": 38.8412017167382,
"eval_f1": 59.56507187901322,
"step": 200
},
{
"epoch": 0.39,
"eval_exact_match": 45.11802575107296,
"eval_f1": 66.94860285103326,
"step": 400
},
{
"epoch": 0.48,
"learning_rate": 4.203596660244059e-05,
"loss": 2.1132,
"step": 500
},
{
"epoch": 0.58,
"eval_exact_match": 51.663090128755364,
"eval_f1": 70.0320493826151,
"step": 600
},
{
"epoch": 0.77,
"eval_exact_match": 53.59442060085837,
"eval_f1": 72.38853344284765,
"step": 800
},
{
"epoch": 0.96,
"learning_rate": 3.400770712909441e-05,
"loss": 1.2927,
"step": 1000
},
{
"epoch": 0.96,
"eval_exact_match": 55.63304721030043,
"eval_f1": 74.55898999725306,
"step": 1000
},
{
"epoch": 1.16,
"eval_exact_match": 56.223175965665234,
"eval_f1": 74.4642823357856,
"step": 1200
},
{
"epoch": 1.35,
"eval_exact_match": 54.98927038626609,
"eval_f1": 73.49170360943472,
"step": 1400
},
{
"epoch": 1.45,
"learning_rate": 2.5979447655748235e-05,
"loss": 0.8019,
"step": 1500
},
{
"epoch": 1.54,
"eval_exact_match": 55.9549356223176,
"eval_f1": 74.65040749842683,
"step": 1600
},
{
"epoch": 1.73,
"eval_exact_match": 57.457081545064376,
"eval_f1": 75.3556762232775,
"step": 1800
},
{
"epoch": 1.93,
"learning_rate": 1.7951188182402056e-05,
"loss": 0.7301,
"step": 2000
},
{
"epoch": 1.93,
"eval_exact_match": 59.17381974248927,
"eval_f1": 76.26399788082415,
"step": 2000
},
{
"epoch": 2.12,
"eval_exact_match": 57.8862660944206,
"eval_f1": 75.18740187093938,
"step": 2200
},
{
"epoch": 2.31,
"eval_exact_match": 57.77896995708154,
"eval_f1": 75.37580137031661,
"step": 2400
},
{
"epoch": 2.41,
"learning_rate": 9.922928709055877e-06,
"loss": 0.4301,
"step": 2500
},
{
"epoch": 2.5,
"eval_exact_match": 58.31545064377682,
"eval_f1": 75.85451949985745,
"step": 2600
},
{
"epoch": 2.7,
"eval_exact_match": 57.18884120171674,
"eval_f1": 75.28315608222837,
"step": 2800
},
{
"epoch": 2.89,
"learning_rate": 1.8946692357096983e-06,
"loss": 0.3568,
"step": 3000
},
{
"epoch": 2.89,
"eval_exact_match": 57.832618025751074,
"eval_f1": 75.7745668972737,
"step": 3000
},
{
"epoch": 3.0,
"step": 3114,
"total_flos": 1075319724912768.0,
"train_loss": 0.9321928299920408,
"train_runtime": 372.7049,
"train_samples_per_second": 133.602,
"train_steps_per_second": 8.355
}
],
"max_steps": 3114,
"num_train_epochs": 3,
"total_flos": 1075319724912768.0,
"trial_name": null,
"trial_params": null
}