|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 29.0, |
|
"eval_steps": 500, |
|
"global_step": 15341, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.19, |
|
"learning_rate": 3e-05, |
|
"loss": 1.7194, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"learning_rate": 3e-05, |
|
"loss": 1.4369, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"learning_rate": 3e-05, |
|
"loss": 1.4257, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"learning_rate": 3e-05, |
|
"loss": 1.3947, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"learning_rate": 3e-05, |
|
"loss": 1.3948, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.6132307692307692, |
|
"eval_loss": 1.3087366819381714, |
|
"eval_runtime": 4.3553, |
|
"eval_samples_per_second": 114.803, |
|
"eval_steps_per_second": 14.465, |
|
"step": 529 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_exact_match": 24.4, |
|
"eval_f1": 37.614420394420414, |
|
"step": 529 |
|
}, |
|
{ |
|
"epoch": 1.13, |
|
"learning_rate": 3e-05, |
|
"loss": 1.3927, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 1.32, |
|
"learning_rate": 3e-05, |
|
"loss": 1.3644, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 1.51, |
|
"learning_rate": 3e-05, |
|
"loss": 1.3791, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 1.7, |
|
"learning_rate": 3e-05, |
|
"loss": 1.3627, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 1.89, |
|
"learning_rate": 3e-05, |
|
"loss": 1.3789, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.6146153846153846, |
|
"eval_loss": 1.289727807044983, |
|
"eval_runtime": 4.4281, |
|
"eval_samples_per_second": 112.915, |
|
"eval_steps_per_second": 14.227, |
|
"step": 1058 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_exact_match": 23.2, |
|
"eval_f1": 35.56775335775336, |
|
"step": 1058 |
|
}, |
|
{ |
|
"epoch": 2.08, |
|
"learning_rate": 3e-05, |
|
"loss": 1.3605, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 2.27, |
|
"learning_rate": 3e-05, |
|
"loss": 1.3306, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 2.46, |
|
"learning_rate": 3e-05, |
|
"loss": 1.3318, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 2.65, |
|
"learning_rate": 3e-05, |
|
"loss": 1.3491, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 2.84, |
|
"learning_rate": 3e-05, |
|
"loss": 1.3259, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.6178974358974358, |
|
"eval_loss": 1.2849316596984863, |
|
"eval_runtime": 4.875, |
|
"eval_samples_per_second": 102.565, |
|
"eval_steps_per_second": 12.923, |
|
"step": 1587 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_exact_match": 25.2, |
|
"eval_f1": 37.32698412698414, |
|
"step": 1587 |
|
}, |
|
{ |
|
"epoch": 3.02, |
|
"learning_rate": 3e-05, |
|
"loss": 1.3255, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 3.21, |
|
"learning_rate": 3e-05, |
|
"loss": 1.28, |
|
"step": 1700 |
|
}, |
|
{ |
|
"epoch": 3.4, |
|
"learning_rate": 3e-05, |
|
"loss": 1.2995, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 3.59, |
|
"learning_rate": 3e-05, |
|
"loss": 1.2991, |
|
"step": 1900 |
|
}, |
|
{ |
|
"epoch": 3.78, |
|
"learning_rate": 3e-05, |
|
"loss": 1.3024, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 3.97, |
|
"learning_rate": 3e-05, |
|
"loss": 1.2853, |
|
"step": 2100 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.615948717948718, |
|
"eval_loss": 1.316924810409546, |
|
"eval_runtime": 5.3385, |
|
"eval_samples_per_second": 93.659, |
|
"eval_steps_per_second": 11.801, |
|
"step": 2116 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_exact_match": 23.8, |
|
"eval_f1": 36.284920634920645, |
|
"step": 2116 |
|
}, |
|
{ |
|
"epoch": 4.16, |
|
"learning_rate": 3e-05, |
|
"loss": 1.2575, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 4.35, |
|
"learning_rate": 3e-05, |
|
"loss": 1.2261, |
|
"step": 2300 |
|
}, |
|
{ |
|
"epoch": 4.54, |
|
"learning_rate": 3e-05, |
|
"loss": 1.2536, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 4.73, |
|
"learning_rate": 3e-05, |
|
"loss": 1.2376, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 4.91, |
|
"learning_rate": 3e-05, |
|
"loss": 1.2556, |
|
"step": 2600 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.6131794871794872, |
|
"eval_loss": 1.3531945943832397, |
|
"eval_runtime": 4.4382, |
|
"eval_samples_per_second": 112.659, |
|
"eval_steps_per_second": 14.195, |
|
"step": 2645 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_exact_match": 25.8, |
|
"eval_f1": 37.67571428571429, |
|
"step": 2645 |
|
}, |
|
{ |
|
"epoch": 5.1, |
|
"learning_rate": 3e-05, |
|
"loss": 1.2195, |
|
"step": 2700 |
|
}, |
|
{ |
|
"epoch": 5.29, |
|
"learning_rate": 3e-05, |
|
"loss": 1.2003, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 5.48, |
|
"learning_rate": 3e-05, |
|
"loss": 1.1864, |
|
"step": 2900 |
|
}, |
|
{ |
|
"epoch": 5.67, |
|
"learning_rate": 3e-05, |
|
"loss": 1.2231, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 5.86, |
|
"learning_rate": 3e-05, |
|
"loss": 1.1972, |
|
"step": 3100 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.6125641025641025, |
|
"eval_loss": 1.4135066270828247, |
|
"eval_runtime": 5.0312, |
|
"eval_samples_per_second": 99.38, |
|
"eval_steps_per_second": 12.522, |
|
"step": 3174 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_exact_match": 24.0, |
|
"eval_f1": 36.140952380952385, |
|
"step": 3174 |
|
}, |
|
{ |
|
"epoch": 6.05, |
|
"learning_rate": 3e-05, |
|
"loss": 1.1889, |
|
"step": 3200 |
|
}, |
|
{ |
|
"epoch": 6.24, |
|
"learning_rate": 3e-05, |
|
"loss": 1.1471, |
|
"step": 3300 |
|
}, |
|
{ |
|
"epoch": 6.43, |
|
"learning_rate": 3e-05, |
|
"loss": 1.1509, |
|
"step": 3400 |
|
}, |
|
{ |
|
"epoch": 6.62, |
|
"learning_rate": 3e-05, |
|
"loss": 1.1654, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 6.81, |
|
"learning_rate": 3e-05, |
|
"loss": 1.1716, |
|
"step": 3600 |
|
}, |
|
{ |
|
"epoch": 6.99, |
|
"learning_rate": 3e-05, |
|
"loss": 1.1839, |
|
"step": 3700 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_accuracy": 0.6081025641025641, |
|
"eval_loss": 1.500714898109436, |
|
"eval_runtime": 4.4677, |
|
"eval_samples_per_second": 111.916, |
|
"eval_steps_per_second": 14.101, |
|
"step": 3703 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_exact_match": 22.8, |
|
"eval_f1": 33.44777777777779, |
|
"step": 3703 |
|
}, |
|
{ |
|
"epoch": 7.18, |
|
"learning_rate": 3e-05, |
|
"loss": 1.1059, |
|
"step": 3800 |
|
}, |
|
{ |
|
"epoch": 7.37, |
|
"learning_rate": 3e-05, |
|
"loss": 1.1146, |
|
"step": 3900 |
|
}, |
|
{ |
|
"epoch": 7.56, |
|
"learning_rate": 3e-05, |
|
"loss": 1.1121, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 7.75, |
|
"learning_rate": 3e-05, |
|
"loss": 1.1286, |
|
"step": 4100 |
|
}, |
|
{ |
|
"epoch": 7.94, |
|
"learning_rate": 3e-05, |
|
"loss": 1.1334, |
|
"step": 4200 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.6074358974358974, |
|
"eval_loss": 1.524243950843811, |
|
"eval_runtime": 4.2161, |
|
"eval_samples_per_second": 118.592, |
|
"eval_steps_per_second": 14.943, |
|
"step": 4232 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_exact_match": 23.2, |
|
"eval_f1": 34.507792207792214, |
|
"step": 4232 |
|
}, |
|
{ |
|
"epoch": 8.13, |
|
"learning_rate": 3e-05, |
|
"loss": 1.1072, |
|
"step": 4300 |
|
}, |
|
{ |
|
"epoch": 8.32, |
|
"learning_rate": 3e-05, |
|
"loss": 1.0626, |
|
"step": 4400 |
|
}, |
|
{ |
|
"epoch": 8.51, |
|
"learning_rate": 3e-05, |
|
"loss": 1.0842, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 8.7, |
|
"learning_rate": 3e-05, |
|
"loss": 1.0812, |
|
"step": 4600 |
|
}, |
|
{ |
|
"epoch": 8.88, |
|
"learning_rate": 3e-05, |
|
"loss": 1.0966, |
|
"step": 4700 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_accuracy": 0.5802564102564103, |
|
"eval_loss": 1.6106736660003662, |
|
"eval_runtime": 4.2194, |
|
"eval_samples_per_second": 118.5, |
|
"eval_steps_per_second": 14.931, |
|
"step": 4761 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_exact_match": 22.8, |
|
"eval_f1": 34.45666666666666, |
|
"step": 4761 |
|
}, |
|
{ |
|
"epoch": 9.07, |
|
"learning_rate": 3e-05, |
|
"loss": 1.0755, |
|
"step": 4800 |
|
}, |
|
{ |
|
"epoch": 9.26, |
|
"learning_rate": 3e-05, |
|
"loss": 1.0189, |
|
"step": 4900 |
|
}, |
|
{ |
|
"epoch": 9.45, |
|
"learning_rate": 3e-05, |
|
"loss": 1.0565, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 9.64, |
|
"learning_rate": 3e-05, |
|
"loss": 1.0635, |
|
"step": 5100 |
|
}, |
|
{ |
|
"epoch": 9.83, |
|
"learning_rate": 3e-05, |
|
"loss": 1.0485, |
|
"step": 5200 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_accuracy": 0.6049230769230769, |
|
"eval_loss": 1.6749216318130493, |
|
"eval_runtime": 4.8521, |
|
"eval_samples_per_second": 103.048, |
|
"eval_steps_per_second": 12.984, |
|
"step": 5290 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_exact_match": 22.8, |
|
"eval_f1": 34.3535064935065, |
|
"step": 5290 |
|
}, |
|
{ |
|
"epoch": 10.02, |
|
"learning_rate": 3e-05, |
|
"loss": 1.0585, |
|
"step": 5300 |
|
}, |
|
{ |
|
"epoch": 10.21, |
|
"learning_rate": 3e-05, |
|
"loss": 1.0098, |
|
"step": 5400 |
|
}, |
|
{ |
|
"epoch": 10.4, |
|
"learning_rate": 3e-05, |
|
"loss": 0.9935, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 10.59, |
|
"learning_rate": 3e-05, |
|
"loss": 1.0178, |
|
"step": 5600 |
|
}, |
|
{ |
|
"epoch": 10.78, |
|
"learning_rate": 3e-05, |
|
"loss": 1.0217, |
|
"step": 5700 |
|
}, |
|
{ |
|
"epoch": 10.96, |
|
"learning_rate": 3e-05, |
|
"loss": 1.021, |
|
"step": 5800 |
|
}, |
|
{ |
|
"epoch": 11.0, |
|
"eval_accuracy": 0.6014871794871794, |
|
"eval_loss": 1.7323671579360962, |
|
"eval_runtime": 4.4139, |
|
"eval_samples_per_second": 113.28, |
|
"eval_steps_per_second": 14.273, |
|
"step": 5819 |
|
}, |
|
{ |
|
"epoch": 11.0, |
|
"eval_exact_match": 22.4, |
|
"eval_f1": 34.060490620490626, |
|
"step": 5819 |
|
}, |
|
{ |
|
"epoch": 11.15, |
|
"learning_rate": 3e-05, |
|
"loss": 0.9788, |
|
"step": 5900 |
|
}, |
|
{ |
|
"epoch": 11.34, |
|
"learning_rate": 3e-05, |
|
"loss": 0.9799, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 11.53, |
|
"learning_rate": 3e-05, |
|
"loss": 0.9749, |
|
"step": 6100 |
|
}, |
|
{ |
|
"epoch": 11.72, |
|
"learning_rate": 3e-05, |
|
"loss": 0.981, |
|
"step": 6200 |
|
}, |
|
{ |
|
"epoch": 11.91, |
|
"learning_rate": 3e-05, |
|
"loss": 0.9918, |
|
"step": 6300 |
|
}, |
|
{ |
|
"epoch": 12.0, |
|
"eval_accuracy": 0.6007179487179487, |
|
"eval_loss": 1.7632156610488892, |
|
"eval_runtime": 4.4251, |
|
"eval_samples_per_second": 112.992, |
|
"eval_steps_per_second": 14.237, |
|
"step": 6348 |
|
}, |
|
{ |
|
"epoch": 12.0, |
|
"eval_exact_match": 22.4, |
|
"eval_f1": 33.017142857142865, |
|
"step": 6348 |
|
}, |
|
{ |
|
"epoch": 12.1, |
|
"learning_rate": 3e-05, |
|
"loss": 0.967, |
|
"step": 6400 |
|
}, |
|
{ |
|
"epoch": 12.29, |
|
"learning_rate": 3e-05, |
|
"loss": 0.9341, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 12.48, |
|
"learning_rate": 3e-05, |
|
"loss": 0.9656, |
|
"step": 6600 |
|
}, |
|
{ |
|
"epoch": 12.67, |
|
"learning_rate": 3e-05, |
|
"loss": 0.9519, |
|
"step": 6700 |
|
}, |
|
{ |
|
"epoch": 12.85, |
|
"learning_rate": 3e-05, |
|
"loss": 0.947, |
|
"step": 6800 |
|
}, |
|
{ |
|
"epoch": 13.0, |
|
"eval_accuracy": 0.6010769230769231, |
|
"eval_loss": 1.8302537202835083, |
|
"eval_runtime": 4.849, |
|
"eval_samples_per_second": 103.115, |
|
"eval_steps_per_second": 12.992, |
|
"step": 6877 |
|
}, |
|
{ |
|
"epoch": 13.0, |
|
"eval_exact_match": 21.6, |
|
"eval_f1": 32.89746031746034, |
|
"step": 6877 |
|
}, |
|
{ |
|
"epoch": 13.04, |
|
"learning_rate": 3e-05, |
|
"loss": 0.9461, |
|
"step": 6900 |
|
}, |
|
{ |
|
"epoch": 13.23, |
|
"learning_rate": 3e-05, |
|
"loss": 0.9153, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 13.42, |
|
"learning_rate": 3e-05, |
|
"loss": 0.9166, |
|
"step": 7100 |
|
}, |
|
{ |
|
"epoch": 13.61, |
|
"learning_rate": 3e-05, |
|
"loss": 0.9266, |
|
"step": 7200 |
|
}, |
|
{ |
|
"epoch": 13.8, |
|
"learning_rate": 3e-05, |
|
"loss": 0.9089, |
|
"step": 7300 |
|
}, |
|
{ |
|
"epoch": 13.99, |
|
"learning_rate": 3e-05, |
|
"loss": 0.9376, |
|
"step": 7400 |
|
}, |
|
{ |
|
"epoch": 14.0, |
|
"eval_accuracy": 0.5990769230769231, |
|
"eval_loss": 1.8872731924057007, |
|
"eval_runtime": 4.1087, |
|
"eval_samples_per_second": 121.694, |
|
"eval_steps_per_second": 15.333, |
|
"step": 7406 |
|
}, |
|
{ |
|
"epoch": 14.0, |
|
"eval_exact_match": 20.8, |
|
"eval_f1": 32.45952380952382, |
|
"step": 7406 |
|
}, |
|
{ |
|
"epoch": 14.18, |
|
"learning_rate": 3e-05, |
|
"loss": 0.8792, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 14.37, |
|
"learning_rate": 3e-05, |
|
"loss": 0.8787, |
|
"step": 7600 |
|
}, |
|
{ |
|
"epoch": 14.56, |
|
"learning_rate": 3e-05, |
|
"loss": 0.874, |
|
"step": 7700 |
|
}, |
|
{ |
|
"epoch": 14.74, |
|
"learning_rate": 3e-05, |
|
"loss": 0.9123, |
|
"step": 7800 |
|
}, |
|
{ |
|
"epoch": 14.93, |
|
"learning_rate": 3e-05, |
|
"loss": 0.898, |
|
"step": 7900 |
|
}, |
|
{ |
|
"epoch": 15.0, |
|
"eval_accuracy": 0.5975897435897436, |
|
"eval_loss": 1.96884024143219, |
|
"eval_runtime": 4.5009, |
|
"eval_samples_per_second": 111.09, |
|
"eval_steps_per_second": 13.997, |
|
"step": 7935 |
|
}, |
|
{ |
|
"epoch": 15.0, |
|
"eval_exact_match": 20.6, |
|
"eval_f1": 31.090952380952388, |
|
"step": 7935 |
|
}, |
|
{ |
|
"epoch": 15.12, |
|
"learning_rate": 3e-05, |
|
"loss": 0.8631, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 15.31, |
|
"learning_rate": 3e-05, |
|
"loss": 0.8353, |
|
"step": 8100 |
|
}, |
|
{ |
|
"epoch": 15.5, |
|
"learning_rate": 3e-05, |
|
"loss": 0.8673, |
|
"step": 8200 |
|
}, |
|
{ |
|
"epoch": 15.69, |
|
"learning_rate": 3e-05, |
|
"loss": 0.8515, |
|
"step": 8300 |
|
}, |
|
{ |
|
"epoch": 15.88, |
|
"learning_rate": 3e-05, |
|
"loss": 0.8559, |
|
"step": 8400 |
|
}, |
|
{ |
|
"epoch": 16.0, |
|
"eval_accuracy": 0.5987692307692307, |
|
"eval_loss": 1.9724328517913818, |
|
"eval_runtime": 4.5341, |
|
"eval_samples_per_second": 110.275, |
|
"eval_steps_per_second": 13.895, |
|
"step": 8464 |
|
}, |
|
{ |
|
"epoch": 16.0, |
|
"eval_exact_match": 21.4, |
|
"eval_f1": 32.21255411255413, |
|
"step": 8464 |
|
}, |
|
{ |
|
"epoch": 16.07, |
|
"learning_rate": 3e-05, |
|
"loss": 0.8508, |
|
"step": 8500 |
|
}, |
|
{ |
|
"epoch": 16.26, |
|
"learning_rate": 3e-05, |
|
"loss": 0.8122, |
|
"step": 8600 |
|
}, |
|
{ |
|
"epoch": 16.45, |
|
"learning_rate": 3e-05, |
|
"loss": 0.8223, |
|
"step": 8700 |
|
}, |
|
{ |
|
"epoch": 16.64, |
|
"learning_rate": 3e-05, |
|
"loss": 0.8275, |
|
"step": 8800 |
|
}, |
|
{ |
|
"epoch": 16.82, |
|
"learning_rate": 3e-05, |
|
"loss": 0.8348, |
|
"step": 8900 |
|
}, |
|
{ |
|
"epoch": 17.0, |
|
"eval_accuracy": 0.5713846153846154, |
|
"eval_loss": 1.9815243482589722, |
|
"eval_runtime": 5.0655, |
|
"eval_samples_per_second": 98.706, |
|
"eval_steps_per_second": 12.437, |
|
"step": 8993 |
|
}, |
|
{ |
|
"epoch": 17.0, |
|
"eval_exact_match": 20.0, |
|
"eval_f1": 30.670173160173174, |
|
"step": 8993 |
|
}, |
|
{ |
|
"epoch": 17.01, |
|
"learning_rate": 3e-05, |
|
"loss": 0.837, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 17.2, |
|
"learning_rate": 3e-05, |
|
"loss": 0.7764, |
|
"step": 9100 |
|
}, |
|
{ |
|
"epoch": 17.39, |
|
"learning_rate": 3e-05, |
|
"loss": 0.7817, |
|
"step": 9200 |
|
}, |
|
{ |
|
"epoch": 17.58, |
|
"learning_rate": 3e-05, |
|
"loss": 0.8068, |
|
"step": 9300 |
|
}, |
|
{ |
|
"epoch": 17.77, |
|
"learning_rate": 3e-05, |
|
"loss": 0.7968, |
|
"step": 9400 |
|
}, |
|
{ |
|
"epoch": 17.96, |
|
"learning_rate": 3e-05, |
|
"loss": 0.8106, |
|
"step": 9500 |
|
}, |
|
{ |
|
"epoch": 18.0, |
|
"eval_accuracy": 0.598, |
|
"eval_loss": 2.038572072982788, |
|
"eval_runtime": 4.2168, |
|
"eval_samples_per_second": 118.574, |
|
"eval_steps_per_second": 14.94, |
|
"step": 9522 |
|
}, |
|
{ |
|
"epoch": 18.0, |
|
"eval_exact_match": 19.2, |
|
"eval_f1": 30.44984126984128, |
|
"step": 9522 |
|
}, |
|
{ |
|
"epoch": 18.15, |
|
"learning_rate": 3e-05, |
|
"loss": 0.7697, |
|
"step": 9600 |
|
}, |
|
{ |
|
"epoch": 18.34, |
|
"learning_rate": 3e-05, |
|
"loss": 0.7429, |
|
"step": 9700 |
|
}, |
|
{ |
|
"epoch": 18.53, |
|
"learning_rate": 3e-05, |
|
"loss": 0.7468, |
|
"step": 9800 |
|
}, |
|
{ |
|
"epoch": 18.71, |
|
"learning_rate": 3e-05, |
|
"loss": 0.7667, |
|
"step": 9900 |
|
}, |
|
{ |
|
"epoch": 18.9, |
|
"learning_rate": 3e-05, |
|
"loss": 0.7848, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 19.0, |
|
"eval_accuracy": 0.5963589743589743, |
|
"eval_loss": 2.0627450942993164, |
|
"eval_runtime": 4.2486, |
|
"eval_samples_per_second": 117.685, |
|
"eval_steps_per_second": 14.828, |
|
"step": 10051 |
|
}, |
|
{ |
|
"epoch": 19.0, |
|
"eval_exact_match": 19.0, |
|
"eval_f1": 29.635238095238098, |
|
"step": 10051 |
|
}, |
|
{ |
|
"epoch": 19.09, |
|
"learning_rate": 3e-05, |
|
"loss": 0.7537, |
|
"step": 10100 |
|
}, |
|
{ |
|
"epoch": 19.28, |
|
"learning_rate": 3e-05, |
|
"loss": 0.718, |
|
"step": 10200 |
|
}, |
|
{ |
|
"epoch": 19.47, |
|
"learning_rate": 3e-05, |
|
"loss": 0.7241, |
|
"step": 10300 |
|
}, |
|
{ |
|
"epoch": 19.66, |
|
"learning_rate": 3e-05, |
|
"loss": 0.7421, |
|
"step": 10400 |
|
}, |
|
{ |
|
"epoch": 19.85, |
|
"learning_rate": 3e-05, |
|
"loss": 0.745, |
|
"step": 10500 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"eval_accuracy": 0.5965641025641025, |
|
"eval_loss": 2.0825321674346924, |
|
"eval_runtime": 4.9023, |
|
"eval_samples_per_second": 101.993, |
|
"eval_steps_per_second": 12.851, |
|
"step": 10580 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"eval_exact_match": 16.0, |
|
"eval_f1": 25.784563976747563, |
|
"step": 10580 |
|
}, |
|
{ |
|
"epoch": 20.04, |
|
"learning_rate": 3e-05, |
|
"loss": 0.7384, |
|
"step": 10600 |
|
}, |
|
{ |
|
"epoch": 20.23, |
|
"learning_rate": 3e-05, |
|
"loss": 0.6786, |
|
"step": 10700 |
|
}, |
|
{ |
|
"epoch": 20.42, |
|
"learning_rate": 3e-05, |
|
"loss": 0.6921, |
|
"step": 10800 |
|
}, |
|
{ |
|
"epoch": 20.6, |
|
"learning_rate": 3e-05, |
|
"loss": 0.7092, |
|
"step": 10900 |
|
}, |
|
{ |
|
"epoch": 20.79, |
|
"learning_rate": 3e-05, |
|
"loss": 0.7248, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 20.98, |
|
"learning_rate": 3e-05, |
|
"loss": 0.7208, |
|
"step": 11100 |
|
}, |
|
{ |
|
"epoch": 21.0, |
|
"eval_accuracy": 0.595948717948718, |
|
"eval_loss": 2.0958657264709473, |
|
"eval_runtime": 4.2167, |
|
"eval_samples_per_second": 118.575, |
|
"eval_steps_per_second": 14.941, |
|
"step": 11109 |
|
}, |
|
{ |
|
"epoch": 21.0, |
|
"eval_exact_match": 9.4, |
|
"eval_f1": 15.48481664400564, |
|
"step": 11109 |
|
}, |
|
{ |
|
"epoch": 21.17, |
|
"learning_rate": 3e-05, |
|
"loss": 0.6604, |
|
"step": 11200 |
|
}, |
|
{ |
|
"epoch": 21.36, |
|
"learning_rate": 3e-05, |
|
"loss": 0.6599, |
|
"step": 11300 |
|
}, |
|
{ |
|
"epoch": 21.55, |
|
"learning_rate": 3e-05, |
|
"loss": 0.6775, |
|
"step": 11400 |
|
}, |
|
{ |
|
"epoch": 21.74, |
|
"learning_rate": 3e-05, |
|
"loss": 0.7001, |
|
"step": 11500 |
|
}, |
|
{ |
|
"epoch": 21.93, |
|
"learning_rate": 3e-05, |
|
"loss": 0.6842, |
|
"step": 11600 |
|
}, |
|
{ |
|
"epoch": 22.0, |
|
"eval_accuracy": 0.5967692307692307, |
|
"eval_loss": 2.153414487838745, |
|
"eval_runtime": 4.8531, |
|
"eval_samples_per_second": 103.026, |
|
"eval_steps_per_second": 12.981, |
|
"step": 11638 |
|
}, |
|
{ |
|
"epoch": 22.0, |
|
"eval_exact_match": 9.2, |
|
"eval_f1": 17.03231000578938, |
|
"step": 11638 |
|
}, |
|
{ |
|
"epoch": 22.12, |
|
"learning_rate": 3e-05, |
|
"loss": 0.6544, |
|
"step": 11700 |
|
}, |
|
{ |
|
"epoch": 22.31, |
|
"learning_rate": 3e-05, |
|
"loss": 0.641, |
|
"step": 11800 |
|
}, |
|
{ |
|
"epoch": 22.5, |
|
"learning_rate": 3e-05, |
|
"loss": 0.6369, |
|
"step": 11900 |
|
}, |
|
{ |
|
"epoch": 22.68, |
|
"learning_rate": 3e-05, |
|
"loss": 0.6468, |
|
"step": 12000 |
|
}, |
|
{ |
|
"epoch": 22.87, |
|
"learning_rate": 3e-05, |
|
"loss": 0.6661, |
|
"step": 12100 |
|
}, |
|
{ |
|
"epoch": 23.0, |
|
"eval_accuracy": 0.5974871794871794, |
|
"eval_loss": 2.179178237915039, |
|
"eval_runtime": 4.8444, |
|
"eval_samples_per_second": 103.211, |
|
"eval_steps_per_second": 13.005, |
|
"step": 12167 |
|
}, |
|
{ |
|
"epoch": 23.0, |
|
"eval_exact_match": 3.6, |
|
"eval_f1": 7.6552922676511725, |
|
"step": 12167 |
|
}, |
|
{ |
|
"epoch": 23.06, |
|
"learning_rate": 3e-05, |
|
"loss": 0.6418, |
|
"step": 12200 |
|
}, |
|
{ |
|
"epoch": 23.25, |
|
"learning_rate": 3e-05, |
|
"loss": 0.6012, |
|
"step": 12300 |
|
}, |
|
{ |
|
"epoch": 23.44, |
|
"learning_rate": 3e-05, |
|
"loss": 0.6211, |
|
"step": 12400 |
|
}, |
|
{ |
|
"epoch": 23.63, |
|
"learning_rate": 3e-05, |
|
"loss": 0.6174, |
|
"step": 12500 |
|
}, |
|
{ |
|
"epoch": 23.82, |
|
"learning_rate": 3e-05, |
|
"loss": 0.6193, |
|
"step": 12600 |
|
}, |
|
{ |
|
"epoch": 24.0, |
|
"eval_accuracy": 0.5967179487179487, |
|
"eval_loss": 2.152991771697998, |
|
"eval_runtime": 4.429, |
|
"eval_samples_per_second": 112.891, |
|
"eval_steps_per_second": 14.224, |
|
"step": 12696 |
|
}, |
|
{ |
|
"epoch": 24.0, |
|
"eval_exact_match": 3.2, |
|
"eval_f1": 7.263414468771945, |
|
"step": 12696 |
|
}, |
|
{ |
|
"epoch": 24.01, |
|
"learning_rate": 3e-05, |
|
"loss": 0.6329, |
|
"step": 12700 |
|
}, |
|
{ |
|
"epoch": 24.2, |
|
"learning_rate": 3e-05, |
|
"loss": 0.572, |
|
"step": 12800 |
|
}, |
|
{ |
|
"epoch": 24.39, |
|
"learning_rate": 3e-05, |
|
"loss": 0.585, |
|
"step": 12900 |
|
}, |
|
{ |
|
"epoch": 24.57, |
|
"learning_rate": 3e-05, |
|
"loss": 0.5894, |
|
"step": 13000 |
|
}, |
|
{ |
|
"epoch": 24.76, |
|
"learning_rate": 3e-05, |
|
"loss": 0.6002, |
|
"step": 13100 |
|
}, |
|
{ |
|
"epoch": 24.95, |
|
"learning_rate": 3e-05, |
|
"loss": 0.6064, |
|
"step": 13200 |
|
}, |
|
{ |
|
"epoch": 25.0, |
|
"eval_accuracy": 0.5957948717948718, |
|
"eval_loss": 2.1720352172851562, |
|
"eval_runtime": 4.1244, |
|
"eval_samples_per_second": 121.23, |
|
"eval_steps_per_second": 15.275, |
|
"step": 13225 |
|
}, |
|
{ |
|
"epoch": 25.0, |
|
"eval_exact_match": 1.0, |
|
"eval_f1": 2.817526608207056, |
|
"step": 13225 |
|
}, |
|
{ |
|
"epoch": 25.14, |
|
"learning_rate": 3e-05, |
|
"loss": 0.5603, |
|
"step": 13300 |
|
}, |
|
{ |
|
"epoch": 25.33, |
|
"learning_rate": 3e-05, |
|
"loss": 0.5438, |
|
"step": 13400 |
|
}, |
|
{ |
|
"epoch": 25.52, |
|
"learning_rate": 3e-05, |
|
"loss": 0.5718, |
|
"step": 13500 |
|
}, |
|
{ |
|
"epoch": 25.71, |
|
"learning_rate": 3e-05, |
|
"loss": 0.5804, |
|
"step": 13600 |
|
}, |
|
{ |
|
"epoch": 25.9, |
|
"learning_rate": 3e-05, |
|
"loss": 0.5776, |
|
"step": 13700 |
|
}, |
|
{ |
|
"epoch": 26.0, |
|
"eval_accuracy": 0.5965641025641025, |
|
"eval_loss": 2.216242790222168, |
|
"eval_runtime": 4.8581, |
|
"eval_samples_per_second": 102.921, |
|
"eval_steps_per_second": 12.968, |
|
"step": 13754 |
|
}, |
|
{ |
|
"epoch": 26.0, |
|
"eval_exact_match": 0.6, |
|
"eval_f1": 1.5737055770144102, |
|
"step": 13754 |
|
}, |
|
{ |
|
"epoch": 26.09, |
|
"learning_rate": 3e-05, |
|
"loss": 0.5613, |
|
"step": 13800 |
|
}, |
|
{ |
|
"epoch": 26.28, |
|
"learning_rate": 3e-05, |
|
"loss": 0.5267, |
|
"step": 13900 |
|
}, |
|
{ |
|
"epoch": 26.47, |
|
"learning_rate": 3e-05, |
|
"loss": 0.542, |
|
"step": 14000 |
|
}, |
|
{ |
|
"epoch": 26.65, |
|
"learning_rate": 3e-05, |
|
"loss": 0.5457, |
|
"step": 14100 |
|
}, |
|
{ |
|
"epoch": 26.84, |
|
"learning_rate": 3e-05, |
|
"loss": 0.5492, |
|
"step": 14200 |
|
}, |
|
{ |
|
"epoch": 27.0, |
|
"eval_accuracy": 0.5861538461538461, |
|
"eval_loss": 2.238172769546509, |
|
"eval_runtime": 4.6504, |
|
"eval_samples_per_second": 107.516, |
|
"eval_steps_per_second": 13.547, |
|
"step": 14283 |
|
}, |
|
{ |
|
"epoch": 27.0, |
|
"eval_exact_match": 0.4, |
|
"eval_f1": 1.9589310673272307, |
|
"step": 14283 |
|
}, |
|
{ |
|
"epoch": 27.03, |
|
"learning_rate": 3e-05, |
|
"loss": 0.5488, |
|
"step": 14300 |
|
}, |
|
{ |
|
"epoch": 27.22, |
|
"learning_rate": 3e-05, |
|
"loss": 0.5016, |
|
"step": 14400 |
|
}, |
|
{ |
|
"epoch": 27.41, |
|
"learning_rate": 3e-05, |
|
"loss": 0.5115, |
|
"step": 14500 |
|
}, |
|
{ |
|
"epoch": 27.6, |
|
"learning_rate": 3e-05, |
|
"loss": 0.5219, |
|
"step": 14600 |
|
}, |
|
{ |
|
"epoch": 27.79, |
|
"learning_rate": 3e-05, |
|
"loss": 0.5388, |
|
"step": 14700 |
|
}, |
|
{ |
|
"epoch": 27.98, |
|
"learning_rate": 3e-05, |
|
"loss": 0.5256, |
|
"step": 14800 |
|
}, |
|
{ |
|
"epoch": 28.0, |
|
"eval_accuracy": 0.5963076923076923, |
|
"eval_loss": 2.2273213863372803, |
|
"eval_runtime": 4.4195, |
|
"eval_samples_per_second": 113.136, |
|
"eval_steps_per_second": 14.255, |
|
"step": 14812 |
|
}, |
|
{ |
|
"epoch": 28.0, |
|
"eval_exact_match": 0.6, |
|
"eval_f1": 1.254701854861382, |
|
"step": 14812 |
|
}, |
|
{ |
|
"epoch": 28.17, |
|
"learning_rate": 3e-05, |
|
"loss": 0.4814, |
|
"step": 14900 |
|
}, |
|
{ |
|
"epoch": 28.36, |
|
"learning_rate": 3e-05, |
|
"loss": 0.4943, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 28.54, |
|
"learning_rate": 3e-05, |
|
"loss": 0.4899, |
|
"step": 15100 |
|
}, |
|
{ |
|
"epoch": 28.73, |
|
"learning_rate": 3e-05, |
|
"loss": 0.5047, |
|
"step": 15200 |
|
}, |
|
{ |
|
"epoch": 28.92, |
|
"learning_rate": 3e-05, |
|
"loss": 0.5128, |
|
"step": 15300 |
|
}, |
|
{ |
|
"epoch": 29.0, |
|
"eval_accuracy": 0.5947692307692307, |
|
"eval_loss": 2.244802713394165, |
|
"eval_runtime": 4.2174, |
|
"eval_samples_per_second": 118.556, |
|
"eval_steps_per_second": 14.938, |
|
"step": 15341 |
|
}, |
|
{ |
|
"epoch": 29.0, |
|
"eval_exact_match": 0.0, |
|
"eval_f1": 0.3342622661004701, |
|
"step": 15341 |
|
} |
|
], |
|
"logging_steps": 100, |
|
"max_steps": 26450, |
|
"num_train_epochs": 50, |
|
"save_steps": 500, |
|
"total_flos": 2.5909359576581407e+18, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|