josecannete's picture
adding model fine-tuned on QA (SQAC)
13ebf76
{
"best_metric": 59.382051635096076,
"best_model_checkpoint": "/data/jcanete/all_results/sqac/albeto_tiny/epochs_4_bs_16_lr_5e-5/checkpoint-2400",
"epoch": 4.0,
"global_step": 4152,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.19,
"eval_exact_match": 12.339055793991417,
"eval_f1": 28.93103280159431,
"step": 200
},
{
"epoch": 0.39,
"eval_exact_match": 23.92703862660944,
"eval_f1": 41.67915333843358,
"step": 400
},
{
"epoch": 0.48,
"learning_rate": 4.4026974951830445e-05,
"loss": 3.1377,
"step": 500
},
{
"epoch": 0.58,
"eval_exact_match": 31.86695278969957,
"eval_f1": 51.18317520944259,
"step": 600
},
{
"epoch": 0.77,
"eval_exact_match": 34.871244635193136,
"eval_f1": 53.9432519365253,
"step": 800
},
{
"epoch": 0.96,
"learning_rate": 3.800578034682081e-05,
"loss": 2.0078,
"step": 1000
},
{
"epoch": 0.96,
"eval_exact_match": 35.51502145922747,
"eval_f1": 55.221685158386705,
"step": 1000
},
{
"epoch": 1.16,
"eval_exact_match": 36.85622317596567,
"eval_f1": 55.20708770919938,
"step": 1200
},
{
"epoch": 1.35,
"eval_exact_match": 37.98283261802575,
"eval_f1": 56.614251341537866,
"step": 1400
},
{
"epoch": 1.45,
"learning_rate": 3.198458574181118e-05,
"loss": 1.4374,
"step": 1500
},
{
"epoch": 1.54,
"eval_exact_match": 39.002145922746784,
"eval_f1": 57.702126636307476,
"step": 1600
},
{
"epoch": 1.73,
"eval_exact_match": 39.64592274678112,
"eval_f1": 58.69230555304524,
"step": 1800
},
{
"epoch": 1.93,
"learning_rate": 2.596339113680154e-05,
"loss": 1.3597,
"step": 2000
},
{
"epoch": 1.93,
"eval_exact_match": 39.64592274678112,
"eval_f1": 58.05721827009323,
"step": 2000
},
{
"epoch": 2.12,
"eval_exact_match": 39.27038626609442,
"eval_f1": 57.648146496721296,
"step": 2200
},
{
"epoch": 2.31,
"eval_exact_match": 40.50429184549356,
"eval_f1": 59.382051635096076,
"step": 2400
},
{
"epoch": 2.41,
"learning_rate": 1.995423892100193e-05,
"loss": 0.9536,
"step": 2500
},
{
"epoch": 2.5,
"eval_exact_match": 40.71888412017167,
"eval_f1": 59.060573449175955,
"step": 2600
},
{
"epoch": 2.7,
"eval_exact_match": 39.8068669527897,
"eval_f1": 58.80449299317162,
"step": 2800
},
{
"epoch": 2.89,
"learning_rate": 1.3933044315992294e-05,
"loss": 0.8884,
"step": 3000
},
{
"epoch": 2.89,
"eval_exact_match": 40.772532188841204,
"eval_f1": 58.85814483853365,
"step": 3000
},
{
"epoch": 3.08,
"eval_exact_match": 40.02145922746781,
"eval_f1": 59.12527092459649,
"step": 3200
},
{
"epoch": 3.28,
"eval_exact_match": 40.933476394849784,
"eval_f1": 59.19992976262888,
"step": 3400
},
{
"epoch": 3.37,
"learning_rate": 7.91184971098266e-06,
"loss": 0.6567,
"step": 3500
},
{
"epoch": 3.47,
"eval_exact_match": 40.933476394849784,
"eval_f1": 59.280000618875235,
"step": 3600
},
{
"epoch": 3.66,
"eval_exact_match": 40.50429184549356,
"eval_f1": 58.81358724565813,
"step": 3800
},
{
"epoch": 3.85,
"learning_rate": 1.890655105973025e-06,
"loss": 0.5809,
"step": 4000
},
{
"epoch": 3.85,
"eval_exact_match": 40.50429184549356,
"eval_f1": 58.87104735095327,
"step": 4000
},
{
"epoch": 4.0,
"step": 4152,
"total_flos": 241925105088000.0,
"train_loss": 1.34779805898207,
"train_runtime": 436.6525,
"train_samples_per_second": 152.048,
"train_steps_per_second": 9.509
}
],
"max_steps": 4152,
"num_train_epochs": 4,
"total_flos": 241925105088000.0,
"trial_name": null,
"trial_params": null
}