|
{ |
|
"best_metric": 24.366, |
|
"best_model_checkpoint": "/home/hpcpudu1/rds/hpc-work/data/pretrain-mds/led_pretrain/ver2/gen_model/Centrum_base_multinews_23-7-22.2/checkpoint-23000", |
|
"epoch": 8.896557858222895, |
|
"global_step": 25000, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.0, |
|
"learning_rate": 1.2000000000000002e-08, |
|
"loss": 3.7881, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"learning_rate": 1.2000000000000002e-06, |
|
"loss": 3.5568, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 2.4000000000000003e-06, |
|
"loss": 3.5213, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"learning_rate": 3.6e-06, |
|
"loss": 3.49, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 4.800000000000001e-06, |
|
"loss": 3.4612, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"learning_rate": 6e-06, |
|
"loss": 3.4476, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"learning_rate": 7.2e-06, |
|
"loss": 3.4241, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"learning_rate": 8.400000000000001e-06, |
|
"loss": 3.4164, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"learning_rate": 9.600000000000001e-06, |
|
"loss": 3.4208, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"learning_rate": 1.08e-05, |
|
"loss": 3.4008, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 0.36, |
|
"learning_rate": 1.2e-05, |
|
"loss": 3.4391, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.39, |
|
"learning_rate": 1.32e-05, |
|
"loss": 3.3908, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"learning_rate": 1.44e-05, |
|
"loss": 3.3961, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"learning_rate": 1.56e-05, |
|
"loss": 3.4042, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"learning_rate": 1.6800000000000002e-05, |
|
"loss": 3.3847, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 0.53, |
|
"learning_rate": 1.8e-05, |
|
"loss": 3.3763, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"learning_rate": 1.9200000000000003e-05, |
|
"loss": 3.3873, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"learning_rate": 2.04e-05, |
|
"loss": 3.3817, |
|
"step": 1700 |
|
}, |
|
{ |
|
"epoch": 0.64, |
|
"learning_rate": 2.16e-05, |
|
"loss": 3.363, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 0.68, |
|
"learning_rate": 2.2800000000000002e-05, |
|
"loss": 3.3631, |
|
"step": 1900 |
|
}, |
|
{ |
|
"epoch": 0.71, |
|
"learning_rate": 2.4e-05, |
|
"loss": 3.3734, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.75, |
|
"learning_rate": 2.52e-05, |
|
"loss": 3.3414, |
|
"step": 2100 |
|
}, |
|
{ |
|
"epoch": 0.78, |
|
"learning_rate": 2.64e-05, |
|
"loss": 3.384, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 0.82, |
|
"learning_rate": 2.7600000000000003e-05, |
|
"loss": 3.3718, |
|
"step": 2300 |
|
}, |
|
{ |
|
"epoch": 0.85, |
|
"learning_rate": 2.88e-05, |
|
"loss": 3.3669, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"learning_rate": 3e-05, |
|
"loss": 3.3649, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"learning_rate": 2.9866666666666666e-05, |
|
"loss": 3.3537, |
|
"step": 2600 |
|
}, |
|
{ |
|
"epoch": 0.96, |
|
"learning_rate": 2.9733333333333336e-05, |
|
"loss": 3.393, |
|
"step": 2700 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 2.96e-05, |
|
"loss": 3.3729, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 1.03, |
|
"learning_rate": 2.9466666666666667e-05, |
|
"loss": 3.3593, |
|
"step": 2900 |
|
}, |
|
{ |
|
"epoch": 1.07, |
|
"learning_rate": 2.9333333333333333e-05, |
|
"loss": 3.3256, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 1.1, |
|
"learning_rate": 2.92e-05, |
|
"loss": 3.2975, |
|
"step": 3100 |
|
}, |
|
{ |
|
"epoch": 1.14, |
|
"learning_rate": 2.9066666666666667e-05, |
|
"loss": 3.3259, |
|
"step": 3200 |
|
}, |
|
{ |
|
"epoch": 1.17, |
|
"learning_rate": 2.8933333333333333e-05, |
|
"loss": 3.3079, |
|
"step": 3300 |
|
}, |
|
{ |
|
"epoch": 1.21, |
|
"learning_rate": 2.88e-05, |
|
"loss": 3.302, |
|
"step": 3400 |
|
}, |
|
{ |
|
"epoch": 1.25, |
|
"learning_rate": 2.8666666666666668e-05, |
|
"loss": 3.2971, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 1.28, |
|
"learning_rate": 2.8533333333333333e-05, |
|
"loss": 3.3297, |
|
"step": 3600 |
|
}, |
|
{ |
|
"epoch": 1.32, |
|
"learning_rate": 2.84e-05, |
|
"loss": 3.3068, |
|
"step": 3700 |
|
}, |
|
{ |
|
"epoch": 1.35, |
|
"learning_rate": 2.8268e-05, |
|
"loss": 3.31, |
|
"step": 3800 |
|
}, |
|
{ |
|
"epoch": 1.39, |
|
"learning_rate": 2.8134666666666667e-05, |
|
"loss": 3.2908, |
|
"step": 3900 |
|
}, |
|
{ |
|
"epoch": 1.42, |
|
"learning_rate": 2.8001333333333332e-05, |
|
"loss": 3.2954, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 1.46, |
|
"learning_rate": 2.7868e-05, |
|
"loss": 3.3029, |
|
"step": 4100 |
|
}, |
|
{ |
|
"epoch": 1.49, |
|
"learning_rate": 2.7734666666666667e-05, |
|
"loss": 3.3064, |
|
"step": 4200 |
|
}, |
|
{ |
|
"epoch": 1.53, |
|
"learning_rate": 2.7601333333333333e-05, |
|
"loss": 3.2857, |
|
"step": 4300 |
|
}, |
|
{ |
|
"epoch": 1.57, |
|
"learning_rate": 2.7468e-05, |
|
"loss": 3.2781, |
|
"step": 4400 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"learning_rate": 2.7334666666666668e-05, |
|
"loss": 3.3019, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 1.64, |
|
"learning_rate": 2.7201333333333333e-05, |
|
"loss": 3.2978, |
|
"step": 4600 |
|
}, |
|
{ |
|
"epoch": 1.67, |
|
"learning_rate": 2.7068e-05, |
|
"loss": 3.3186, |
|
"step": 4700 |
|
}, |
|
{ |
|
"epoch": 1.71, |
|
"learning_rate": 2.6934666666666665e-05, |
|
"loss": 3.2843, |
|
"step": 4800 |
|
}, |
|
{ |
|
"epoch": 1.74, |
|
"learning_rate": 2.6801333333333334e-05, |
|
"loss": 3.2863, |
|
"step": 4900 |
|
}, |
|
{ |
|
"epoch": 1.78, |
|
"learning_rate": 2.6668000000000003e-05, |
|
"loss": 3.2702, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 1.78, |
|
"eval_gen_len": 277.1855, |
|
"eval_loss": 3.2853293418884277, |
|
"eval_rouge1": 44.0203, |
|
"eval_rouge2": 16.6061, |
|
"eval_rougeL": 23.3846, |
|
"eval_rougeLsum": 40.3853, |
|
"eval_runtime": 1402.4156, |
|
"eval_samples_per_second": 4.009, |
|
"eval_steps_per_second": 0.251, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 1.81, |
|
"learning_rate": 2.653466666666667e-05, |
|
"loss": 3.3022, |
|
"step": 5100 |
|
}, |
|
{ |
|
"epoch": 1.85, |
|
"learning_rate": 2.6401333333333334e-05, |
|
"loss": 3.29, |
|
"step": 5200 |
|
}, |
|
{ |
|
"epoch": 1.89, |
|
"learning_rate": 2.6268000000000003e-05, |
|
"loss": 3.2893, |
|
"step": 5300 |
|
}, |
|
{ |
|
"epoch": 1.92, |
|
"learning_rate": 2.613466666666667e-05, |
|
"loss": 3.2849, |
|
"step": 5400 |
|
}, |
|
{ |
|
"epoch": 1.96, |
|
"learning_rate": 2.6001333333333335e-05, |
|
"loss": 3.2762, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 1.96, |
|
"eval_gen_len": 288.4173, |
|
"eval_loss": 3.2852535247802734, |
|
"eval_rouge1": 44.725, |
|
"eval_rouge2": 16.9262, |
|
"eval_rougeL": 23.475, |
|
"eval_rougeLsum": 41.0003, |
|
"eval_runtime": 1344.5644, |
|
"eval_samples_per_second": 4.181, |
|
"eval_steps_per_second": 0.262, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 1.99, |
|
"learning_rate": 2.5868e-05, |
|
"loss": 3.2937, |
|
"step": 5600 |
|
}, |
|
{ |
|
"epoch": 2.03, |
|
"learning_rate": 2.573466666666667e-05, |
|
"loss": 3.2614, |
|
"step": 5700 |
|
}, |
|
{ |
|
"epoch": 2.06, |
|
"learning_rate": 2.5601333333333335e-05, |
|
"loss": 3.1995, |
|
"step": 5800 |
|
}, |
|
{ |
|
"epoch": 2.1, |
|
"learning_rate": 2.5468e-05, |
|
"loss": 3.2162, |
|
"step": 5900 |
|
}, |
|
{ |
|
"epoch": 2.14, |
|
"learning_rate": 2.5334666666666666e-05, |
|
"loss": 3.2114, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 2.14, |
|
"eval_gen_len": 257.2761, |
|
"eval_loss": 3.2857086658477783, |
|
"eval_rouge1": 44.6456, |
|
"eval_rouge2": 17.0245, |
|
"eval_rougeL": 23.7328, |
|
"eval_rougeLsum": 40.9131, |
|
"eval_runtime": 1239.4823, |
|
"eval_samples_per_second": 4.536, |
|
"eval_steps_per_second": 0.284, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 2.17, |
|
"learning_rate": 2.5201333333333336e-05, |
|
"loss": 3.202, |
|
"step": 6100 |
|
}, |
|
{ |
|
"epoch": 2.21, |
|
"learning_rate": 2.5068e-05, |
|
"loss": 3.2089, |
|
"step": 6200 |
|
}, |
|
{ |
|
"epoch": 2.24, |
|
"learning_rate": 2.4934666666666667e-05, |
|
"loss": 3.2109, |
|
"step": 6300 |
|
}, |
|
{ |
|
"epoch": 2.28, |
|
"learning_rate": 2.4801333333333333e-05, |
|
"loss": 3.2073, |
|
"step": 6400 |
|
}, |
|
{ |
|
"epoch": 2.31, |
|
"learning_rate": 2.4668e-05, |
|
"loss": 3.1981, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 2.31, |
|
"eval_gen_len": 254.8618, |
|
"eval_loss": 3.2817349433898926, |
|
"eval_rouge1": 44.7869, |
|
"eval_rouge2": 17.0849, |
|
"eval_rougeL": 23.8372, |
|
"eval_rougeLsum": 41.0669, |
|
"eval_runtime": 1197.7293, |
|
"eval_samples_per_second": 4.694, |
|
"eval_steps_per_second": 0.294, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 2.35, |
|
"learning_rate": 2.4534666666666667e-05, |
|
"loss": 3.2258, |
|
"step": 6600 |
|
}, |
|
{ |
|
"epoch": 2.38, |
|
"learning_rate": 2.4401333333333333e-05, |
|
"loss": 3.2252, |
|
"step": 6700 |
|
}, |
|
{ |
|
"epoch": 2.42, |
|
"learning_rate": 2.4268e-05, |
|
"loss": 3.2252, |
|
"step": 6800 |
|
}, |
|
{ |
|
"epoch": 2.46, |
|
"learning_rate": 2.4134666666666668e-05, |
|
"loss": 3.2218, |
|
"step": 6900 |
|
}, |
|
{ |
|
"epoch": 2.49, |
|
"learning_rate": 2.4001333333333333e-05, |
|
"loss": 3.2298, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 2.49, |
|
"eval_gen_len": 263.0854, |
|
"eval_loss": 3.2801525592803955, |
|
"eval_rouge1": 45.2657, |
|
"eval_rouge2": 17.2618, |
|
"eval_rougeL": 23.8204, |
|
"eval_rougeLsum": 41.5807, |
|
"eval_runtime": 1199.3644, |
|
"eval_samples_per_second": 4.687, |
|
"eval_steps_per_second": 0.293, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 2.53, |
|
"learning_rate": 2.3868e-05, |
|
"loss": 3.206, |
|
"step": 7100 |
|
}, |
|
{ |
|
"epoch": 2.56, |
|
"learning_rate": 2.3734666666666665e-05, |
|
"loss": 3.2062, |
|
"step": 7200 |
|
}, |
|
{ |
|
"epoch": 2.6, |
|
"learning_rate": 2.3601333333333334e-05, |
|
"loss": 3.2241, |
|
"step": 7300 |
|
}, |
|
{ |
|
"epoch": 2.63, |
|
"learning_rate": 2.3468e-05, |
|
"loss": 3.2116, |
|
"step": 7400 |
|
}, |
|
{ |
|
"epoch": 2.67, |
|
"learning_rate": 2.3334666666666665e-05, |
|
"loss": 3.2167, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 2.67, |
|
"eval_gen_len": 244.6939, |
|
"eval_loss": 3.2773149013519287, |
|
"eval_rouge1": 44.9516, |
|
"eval_rouge2": 17.0538, |
|
"eval_rougeL": 23.7894, |
|
"eval_rougeLsum": 41.1673, |
|
"eval_runtime": 1137.295, |
|
"eval_samples_per_second": 4.943, |
|
"eval_steps_per_second": 0.31, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 2.7, |
|
"learning_rate": 2.3201333333333334e-05, |
|
"loss": 3.219, |
|
"step": 7600 |
|
}, |
|
{ |
|
"epoch": 2.74, |
|
"learning_rate": 2.3068e-05, |
|
"loss": 3.2057, |
|
"step": 7700 |
|
}, |
|
{ |
|
"epoch": 2.78, |
|
"learning_rate": 2.2936e-05, |
|
"loss": 3.2029, |
|
"step": 7800 |
|
}, |
|
{ |
|
"epoch": 2.81, |
|
"learning_rate": 2.2802666666666668e-05, |
|
"loss": 3.2395, |
|
"step": 7900 |
|
}, |
|
{ |
|
"epoch": 2.85, |
|
"learning_rate": 2.2669333333333333e-05, |
|
"loss": 3.2069, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 2.85, |
|
"eval_gen_len": 245.4036, |
|
"eval_loss": 3.2712182998657227, |
|
"eval_rouge1": 45.2153, |
|
"eval_rouge2": 17.2766, |
|
"eval_rougeL": 23.9883, |
|
"eval_rougeLsum": 41.4558, |
|
"eval_runtime": 1117.2186, |
|
"eval_samples_per_second": 5.032, |
|
"eval_steps_per_second": 0.315, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 2.88, |
|
"learning_rate": 2.2536e-05, |
|
"loss": 3.2034, |
|
"step": 8100 |
|
}, |
|
{ |
|
"epoch": 2.92, |
|
"learning_rate": 2.2402666666666665e-05, |
|
"loss": 3.2163, |
|
"step": 8200 |
|
}, |
|
{ |
|
"epoch": 2.95, |
|
"learning_rate": 2.2269333333333334e-05, |
|
"loss": 3.2158, |
|
"step": 8300 |
|
}, |
|
{ |
|
"epoch": 2.99, |
|
"learning_rate": 2.2136e-05, |
|
"loss": 3.2205, |
|
"step": 8400 |
|
}, |
|
{ |
|
"epoch": 3.02, |
|
"learning_rate": 2.2002666666666665e-05, |
|
"loss": 3.1822, |
|
"step": 8500 |
|
}, |
|
{ |
|
"epoch": 3.02, |
|
"eval_gen_len": 254.6624, |
|
"eval_loss": 3.2785804271698, |
|
"eval_rouge1": 45.4747, |
|
"eval_rouge2": 17.6754, |
|
"eval_rougeL": 24.1878, |
|
"eval_rougeLsum": 41.7304, |
|
"eval_runtime": 1171.0921, |
|
"eval_samples_per_second": 4.801, |
|
"eval_steps_per_second": 0.301, |
|
"step": 8500 |
|
}, |
|
{ |
|
"epoch": 3.06, |
|
"learning_rate": 2.186933333333333e-05, |
|
"loss": 3.1397, |
|
"step": 8600 |
|
}, |
|
{ |
|
"epoch": 3.1, |
|
"learning_rate": 2.1736e-05, |
|
"loss": 3.1659, |
|
"step": 8700 |
|
}, |
|
{ |
|
"epoch": 3.13, |
|
"learning_rate": 2.1602666666666666e-05, |
|
"loss": 3.1318, |
|
"step": 8800 |
|
}, |
|
{ |
|
"epoch": 3.17, |
|
"learning_rate": 2.1469333333333335e-05, |
|
"loss": 3.162, |
|
"step": 8900 |
|
}, |
|
{ |
|
"epoch": 3.2, |
|
"learning_rate": 2.1336000000000004e-05, |
|
"loss": 3.1529, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 3.2, |
|
"eval_gen_len": 246.0157, |
|
"eval_loss": 3.2740354537963867, |
|
"eval_rouge1": 44.9033, |
|
"eval_rouge2": 17.1386, |
|
"eval_rougeL": 23.8511, |
|
"eval_rougeLsum": 41.177, |
|
"eval_runtime": 1185.3803, |
|
"eval_samples_per_second": 4.743, |
|
"eval_steps_per_second": 0.297, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 3.24, |
|
"learning_rate": 2.120266666666667e-05, |
|
"loss": 3.1241, |
|
"step": 9100 |
|
}, |
|
{ |
|
"epoch": 3.27, |
|
"learning_rate": 2.1069333333333335e-05, |
|
"loss": 3.1537, |
|
"step": 9200 |
|
}, |
|
{ |
|
"epoch": 3.31, |
|
"learning_rate": 2.0936e-05, |
|
"loss": 3.1589, |
|
"step": 9300 |
|
}, |
|
{ |
|
"epoch": 3.35, |
|
"learning_rate": 2.080266666666667e-05, |
|
"loss": 3.1415, |
|
"step": 9400 |
|
}, |
|
{ |
|
"epoch": 3.38, |
|
"learning_rate": 2.0669333333333336e-05, |
|
"loss": 3.1407, |
|
"step": 9500 |
|
}, |
|
{ |
|
"epoch": 3.38, |
|
"eval_gen_len": 243.4922, |
|
"eval_loss": 3.270354986190796, |
|
"eval_rouge1": 45.1045, |
|
"eval_rouge2": 17.2335, |
|
"eval_rougeL": 23.9124, |
|
"eval_rougeLsum": 41.3243, |
|
"eval_runtime": 1159.3428, |
|
"eval_samples_per_second": 4.849, |
|
"eval_steps_per_second": 0.304, |
|
"step": 9500 |
|
}, |
|
{ |
|
"epoch": 3.42, |
|
"learning_rate": 2.0536e-05, |
|
"loss": 3.149, |
|
"step": 9600 |
|
}, |
|
{ |
|
"epoch": 3.45, |
|
"learning_rate": 2.0402666666666667e-05, |
|
"loss": 3.1539, |
|
"step": 9700 |
|
}, |
|
{ |
|
"epoch": 3.49, |
|
"learning_rate": 2.027066666666667e-05, |
|
"loss": 3.1539, |
|
"step": 9800 |
|
}, |
|
{ |
|
"epoch": 3.52, |
|
"learning_rate": 2.0137333333333335e-05, |
|
"loss": 3.158, |
|
"step": 9900 |
|
}, |
|
{ |
|
"epoch": 3.56, |
|
"learning_rate": 2.0004e-05, |
|
"loss": 3.1376, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 3.56, |
|
"eval_gen_len": 243.8396, |
|
"eval_loss": 3.272122621536255, |
|
"eval_rouge1": 45.2694, |
|
"eval_rouge2": 17.4797, |
|
"eval_rougeL": 24.1072, |
|
"eval_rougeLsum": 41.5441, |
|
"eval_runtime": 1143.3015, |
|
"eval_samples_per_second": 4.917, |
|
"eval_steps_per_second": 0.308, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 3.59, |
|
"learning_rate": 1.987066666666667e-05, |
|
"loss": 3.1475, |
|
"step": 10100 |
|
}, |
|
{ |
|
"epoch": 3.63, |
|
"learning_rate": 1.9737333333333335e-05, |
|
"loss": 3.1553, |
|
"step": 10200 |
|
}, |
|
{ |
|
"epoch": 3.67, |
|
"learning_rate": 1.9604e-05, |
|
"loss": 3.1589, |
|
"step": 10300 |
|
}, |
|
{ |
|
"epoch": 3.7, |
|
"learning_rate": 1.9470666666666666e-05, |
|
"loss": 3.1712, |
|
"step": 10400 |
|
}, |
|
{ |
|
"epoch": 3.74, |
|
"learning_rate": 1.9337333333333335e-05, |
|
"loss": 3.1545, |
|
"step": 10500 |
|
}, |
|
{ |
|
"epoch": 3.74, |
|
"eval_gen_len": 231.1805, |
|
"eval_loss": 3.271965503692627, |
|
"eval_rouge1": 45.3105, |
|
"eval_rouge2": 17.6338, |
|
"eval_rougeL": 24.1547, |
|
"eval_rougeLsum": 41.5731, |
|
"eval_runtime": 1051.5864, |
|
"eval_samples_per_second": 5.346, |
|
"eval_steps_per_second": 0.335, |
|
"step": 10500 |
|
}, |
|
{ |
|
"epoch": 3.77, |
|
"learning_rate": 1.9204e-05, |
|
"loss": 3.1598, |
|
"step": 10600 |
|
}, |
|
{ |
|
"epoch": 3.81, |
|
"learning_rate": 1.9070666666666667e-05, |
|
"loss": 3.1684, |
|
"step": 10700 |
|
}, |
|
{ |
|
"epoch": 3.84, |
|
"learning_rate": 1.8937333333333336e-05, |
|
"loss": 3.1703, |
|
"step": 10800 |
|
}, |
|
{ |
|
"epoch": 3.88, |
|
"learning_rate": 1.8804e-05, |
|
"loss": 3.1506, |
|
"step": 10900 |
|
}, |
|
{ |
|
"epoch": 3.91, |
|
"learning_rate": 1.8670666666666667e-05, |
|
"loss": 3.1307, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 3.91, |
|
"eval_gen_len": 250.1039, |
|
"eval_loss": 3.268434524536133, |
|
"eval_rouge1": 45.4309, |
|
"eval_rouge2": 17.2665, |
|
"eval_rougeL": 23.8954, |
|
"eval_rougeLsum": 41.6518, |
|
"eval_runtime": 1131.5203, |
|
"eval_samples_per_second": 4.969, |
|
"eval_steps_per_second": 0.311, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 3.95, |
|
"learning_rate": 1.8537333333333333e-05, |
|
"loss": 3.1651, |
|
"step": 11100 |
|
}, |
|
{ |
|
"epoch": 3.99, |
|
"learning_rate": 1.8404000000000002e-05, |
|
"loss": 3.1617, |
|
"step": 11200 |
|
}, |
|
{ |
|
"epoch": 4.02, |
|
"learning_rate": 1.8270666666666668e-05, |
|
"loss": 3.1417, |
|
"step": 11300 |
|
}, |
|
{ |
|
"epoch": 4.06, |
|
"learning_rate": 1.8137333333333333e-05, |
|
"loss": 3.0753, |
|
"step": 11400 |
|
}, |
|
{ |
|
"epoch": 4.09, |
|
"learning_rate": 1.8004e-05, |
|
"loss": 3.1022, |
|
"step": 11500 |
|
}, |
|
{ |
|
"epoch": 4.09, |
|
"eval_gen_len": 242.5923, |
|
"eval_loss": 3.271904468536377, |
|
"eval_rouge1": 45.1959, |
|
"eval_rouge2": 17.4017, |
|
"eval_rougeL": 24.056, |
|
"eval_rougeLsum": 41.5363, |
|
"eval_runtime": 1125.9262, |
|
"eval_samples_per_second": 4.993, |
|
"eval_steps_per_second": 0.313, |
|
"step": 11500 |
|
}, |
|
{ |
|
"epoch": 4.13, |
|
"learning_rate": 1.7870666666666668e-05, |
|
"loss": 3.0868, |
|
"step": 11600 |
|
}, |
|
{ |
|
"epoch": 4.16, |
|
"learning_rate": 1.7737333333333334e-05, |
|
"loss": 3.1109, |
|
"step": 11700 |
|
}, |
|
{ |
|
"epoch": 4.2, |
|
"learning_rate": 1.7605333333333332e-05, |
|
"loss": 3.0823, |
|
"step": 11800 |
|
}, |
|
{ |
|
"epoch": 4.23, |
|
"learning_rate": 1.7472e-05, |
|
"loss": 3.0932, |
|
"step": 11900 |
|
}, |
|
{ |
|
"epoch": 4.27, |
|
"learning_rate": 1.7338666666666667e-05, |
|
"loss": 3.1139, |
|
"step": 12000 |
|
}, |
|
{ |
|
"epoch": 4.27, |
|
"eval_gen_len": 240.5701, |
|
"eval_loss": 3.27105712890625, |
|
"eval_rouge1": 45.3864, |
|
"eval_rouge2": 17.4653, |
|
"eval_rougeL": 24.028, |
|
"eval_rougeLsum": 41.6797, |
|
"eval_runtime": 1097.9204, |
|
"eval_samples_per_second": 5.121, |
|
"eval_steps_per_second": 0.321, |
|
"step": 12000 |
|
}, |
|
{ |
|
"epoch": 4.31, |
|
"learning_rate": 1.7205333333333333e-05, |
|
"loss": 3.1077, |
|
"step": 12100 |
|
}, |
|
{ |
|
"epoch": 4.34, |
|
"learning_rate": 1.7072000000000002e-05, |
|
"loss": 3.1018, |
|
"step": 12200 |
|
}, |
|
{ |
|
"epoch": 4.38, |
|
"learning_rate": 1.6938666666666668e-05, |
|
"loss": 3.1089, |
|
"step": 12300 |
|
}, |
|
{ |
|
"epoch": 4.41, |
|
"learning_rate": 1.6805333333333333e-05, |
|
"loss": 3.0964, |
|
"step": 12400 |
|
}, |
|
{ |
|
"epoch": 4.45, |
|
"learning_rate": 1.6672e-05, |
|
"loss": 3.0978, |
|
"step": 12500 |
|
}, |
|
{ |
|
"epoch": 4.45, |
|
"eval_gen_len": 232.1149, |
|
"eval_loss": 3.2721784114837646, |
|
"eval_rouge1": 45.5694, |
|
"eval_rouge2": 17.501, |
|
"eval_rougeL": 24.1452, |
|
"eval_rougeLsum": 41.7894, |
|
"eval_runtime": 1018.4617, |
|
"eval_samples_per_second": 5.52, |
|
"eval_steps_per_second": 0.346, |
|
"step": 12500 |
|
}, |
|
{ |
|
"epoch": 4.48, |
|
"learning_rate": 1.6538666666666668e-05, |
|
"loss": 3.0998, |
|
"step": 12600 |
|
}, |
|
{ |
|
"epoch": 4.52, |
|
"learning_rate": 1.6405333333333334e-05, |
|
"loss": 3.1115, |
|
"step": 12700 |
|
}, |
|
{ |
|
"epoch": 4.56, |
|
"learning_rate": 1.6272e-05, |
|
"loss": 3.1083, |
|
"step": 12800 |
|
}, |
|
{ |
|
"epoch": 4.59, |
|
"learning_rate": 1.6138666666666665e-05, |
|
"loss": 3.1112, |
|
"step": 12900 |
|
}, |
|
{ |
|
"epoch": 4.63, |
|
"learning_rate": 1.6005333333333334e-05, |
|
"loss": 3.1082, |
|
"step": 13000 |
|
}, |
|
{ |
|
"epoch": 4.63, |
|
"eval_gen_len": 245.1845, |
|
"eval_loss": 3.2687015533447266, |
|
"eval_rouge1": 45.504, |
|
"eval_rouge2": 17.5137, |
|
"eval_rougeL": 24.1067, |
|
"eval_rougeLsum": 41.7686, |
|
"eval_runtime": 1129.9664, |
|
"eval_samples_per_second": 4.975, |
|
"eval_steps_per_second": 0.312, |
|
"step": 13000 |
|
}, |
|
{ |
|
"epoch": 4.66, |
|
"learning_rate": 1.5872e-05, |
|
"loss": 3.0868, |
|
"step": 13100 |
|
}, |
|
{ |
|
"epoch": 4.7, |
|
"learning_rate": 1.5738666666666666e-05, |
|
"loss": 3.1087, |
|
"step": 13200 |
|
}, |
|
{ |
|
"epoch": 4.73, |
|
"learning_rate": 1.5606666666666667e-05, |
|
"loss": 3.1035, |
|
"step": 13300 |
|
}, |
|
{ |
|
"epoch": 4.77, |
|
"learning_rate": 1.5473333333333333e-05, |
|
"loss": 3.1254, |
|
"step": 13400 |
|
}, |
|
{ |
|
"epoch": 4.8, |
|
"learning_rate": 1.534e-05, |
|
"loss": 3.1059, |
|
"step": 13500 |
|
}, |
|
{ |
|
"epoch": 4.8, |
|
"eval_gen_len": 248.6327, |
|
"eval_loss": 3.268646717071533, |
|
"eval_rouge1": 45.3603, |
|
"eval_rouge2": 17.1619, |
|
"eval_rougeL": 23.8655, |
|
"eval_rougeLsum": 41.5953, |
|
"eval_runtime": 1136.3791, |
|
"eval_samples_per_second": 4.947, |
|
"eval_steps_per_second": 0.31, |
|
"step": 13500 |
|
}, |
|
{ |
|
"epoch": 4.84, |
|
"learning_rate": 1.5206666666666668e-05, |
|
"loss": 3.1107, |
|
"step": 13600 |
|
}, |
|
{ |
|
"epoch": 4.88, |
|
"learning_rate": 1.5073333333333335e-05, |
|
"loss": 3.1008, |
|
"step": 13700 |
|
}, |
|
{ |
|
"epoch": 4.91, |
|
"learning_rate": 1.4940000000000001e-05, |
|
"loss": 3.1116, |
|
"step": 13800 |
|
}, |
|
{ |
|
"epoch": 4.95, |
|
"learning_rate": 1.4806666666666668e-05, |
|
"loss": 3.1173, |
|
"step": 13900 |
|
}, |
|
{ |
|
"epoch": 4.98, |
|
"learning_rate": 1.4673333333333334e-05, |
|
"loss": 3.1141, |
|
"step": 14000 |
|
}, |
|
{ |
|
"epoch": 4.98, |
|
"eval_gen_len": 234.0194, |
|
"eval_loss": 3.265822410583496, |
|
"eval_rouge1": 45.2741, |
|
"eval_rouge2": 17.3814, |
|
"eval_rougeL": 24.0377, |
|
"eval_rougeLsum": 41.5263, |
|
"eval_runtime": 1076.885, |
|
"eval_samples_per_second": 5.221, |
|
"eval_steps_per_second": 0.327, |
|
"step": 14000 |
|
}, |
|
{ |
|
"epoch": 5.02, |
|
"learning_rate": 1.4540000000000001e-05, |
|
"loss": 3.0773, |
|
"step": 14100 |
|
}, |
|
{ |
|
"epoch": 5.05, |
|
"learning_rate": 1.4406666666666667e-05, |
|
"loss": 3.0749, |
|
"step": 14200 |
|
}, |
|
{ |
|
"epoch": 5.09, |
|
"learning_rate": 1.4273333333333334e-05, |
|
"loss": 3.0473, |
|
"step": 14300 |
|
}, |
|
{ |
|
"epoch": 5.12, |
|
"learning_rate": 1.414e-05, |
|
"loss": 3.0415, |
|
"step": 14400 |
|
}, |
|
{ |
|
"epoch": 5.16, |
|
"learning_rate": 1.4006666666666668e-05, |
|
"loss": 3.0294, |
|
"step": 14500 |
|
}, |
|
{ |
|
"epoch": 5.16, |
|
"eval_gen_len": 244.4207, |
|
"eval_loss": 3.2715883255004883, |
|
"eval_rouge1": 45.7203, |
|
"eval_rouge2": 17.5962, |
|
"eval_rougeL": 24.1367, |
|
"eval_rougeLsum": 41.9119, |
|
"eval_runtime": 1108.4185, |
|
"eval_samples_per_second": 5.072, |
|
"eval_steps_per_second": 0.318, |
|
"step": 14500 |
|
}, |
|
{ |
|
"epoch": 5.2, |
|
"learning_rate": 1.3873333333333333e-05, |
|
"loss": 3.068, |
|
"step": 14600 |
|
}, |
|
{ |
|
"epoch": 5.23, |
|
"learning_rate": 1.374e-05, |
|
"loss": 3.0561, |
|
"step": 14700 |
|
}, |
|
{ |
|
"epoch": 5.27, |
|
"learning_rate": 1.3606666666666666e-05, |
|
"loss": 3.0738, |
|
"step": 14800 |
|
}, |
|
{ |
|
"epoch": 5.3, |
|
"learning_rate": 1.3473333333333334e-05, |
|
"loss": 3.0854, |
|
"step": 14900 |
|
}, |
|
{ |
|
"epoch": 5.34, |
|
"learning_rate": 1.334e-05, |
|
"loss": 3.0613, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 5.34, |
|
"eval_gen_len": 242.0381, |
|
"eval_loss": 3.26971435546875, |
|
"eval_rouge1": 45.775, |
|
"eval_rouge2": 17.6959, |
|
"eval_rougeL": 24.1867, |
|
"eval_rougeLsum": 42.0018, |
|
"eval_runtime": 1110.4898, |
|
"eval_samples_per_second": 5.063, |
|
"eval_steps_per_second": 0.317, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 5.37, |
|
"learning_rate": 1.3206666666666667e-05, |
|
"loss": 3.0704, |
|
"step": 15100 |
|
}, |
|
{ |
|
"epoch": 5.41, |
|
"learning_rate": 1.3073333333333334e-05, |
|
"loss": 3.0419, |
|
"step": 15200 |
|
}, |
|
{ |
|
"epoch": 5.44, |
|
"learning_rate": 1.2940000000000001e-05, |
|
"loss": 3.0748, |
|
"step": 15300 |
|
}, |
|
{ |
|
"epoch": 5.48, |
|
"learning_rate": 1.2806666666666667e-05, |
|
"loss": 3.0509, |
|
"step": 15400 |
|
}, |
|
{ |
|
"epoch": 5.52, |
|
"learning_rate": 1.2673333333333335e-05, |
|
"loss": 3.0549, |
|
"step": 15500 |
|
}, |
|
{ |
|
"epoch": 5.52, |
|
"eval_gen_len": 242.5493, |
|
"eval_loss": 3.2702813148498535, |
|
"eval_rouge1": 45.8193, |
|
"eval_rouge2": 17.686, |
|
"eval_rougeL": 24.1997, |
|
"eval_rougeLsum": 42.0109, |
|
"eval_runtime": 1089.9866, |
|
"eval_samples_per_second": 5.158, |
|
"eval_steps_per_second": 0.323, |
|
"step": 15500 |
|
}, |
|
{ |
|
"epoch": 5.55, |
|
"learning_rate": 1.254e-05, |
|
"loss": 3.0902, |
|
"step": 15600 |
|
}, |
|
{ |
|
"epoch": 5.59, |
|
"learning_rate": 1.2406666666666668e-05, |
|
"loss": 3.0697, |
|
"step": 15700 |
|
}, |
|
{ |
|
"epoch": 5.62, |
|
"learning_rate": 1.2273333333333333e-05, |
|
"loss": 3.0793, |
|
"step": 15800 |
|
}, |
|
{ |
|
"epoch": 5.66, |
|
"learning_rate": 1.214e-05, |
|
"loss": 3.076, |
|
"step": 15900 |
|
}, |
|
{ |
|
"epoch": 5.69, |
|
"learning_rate": 1.2006666666666666e-05, |
|
"loss": 3.0725, |
|
"step": 16000 |
|
}, |
|
{ |
|
"epoch": 5.69, |
|
"eval_gen_len": 240.2812, |
|
"eval_loss": 3.2654964923858643, |
|
"eval_rouge1": 45.3515, |
|
"eval_rouge2": 17.3438, |
|
"eval_rougeL": 24.0586, |
|
"eval_rougeLsum": 41.6126, |
|
"eval_runtime": 1107.0741, |
|
"eval_samples_per_second": 5.078, |
|
"eval_steps_per_second": 0.318, |
|
"step": 16000 |
|
}, |
|
{ |
|
"epoch": 5.73, |
|
"learning_rate": 1.1873333333333334e-05, |
|
"loss": 3.0904, |
|
"step": 16100 |
|
}, |
|
{ |
|
"epoch": 5.76, |
|
"learning_rate": 1.174e-05, |
|
"loss": 3.082, |
|
"step": 16200 |
|
}, |
|
{ |
|
"epoch": 5.8, |
|
"learning_rate": 1.1608000000000001e-05, |
|
"loss": 3.0543, |
|
"step": 16300 |
|
}, |
|
{ |
|
"epoch": 5.84, |
|
"learning_rate": 1.1474666666666667e-05, |
|
"loss": 3.0445, |
|
"step": 16400 |
|
}, |
|
{ |
|
"epoch": 5.87, |
|
"learning_rate": 1.1341333333333334e-05, |
|
"loss": 3.0728, |
|
"step": 16500 |
|
}, |
|
{ |
|
"epoch": 5.87, |
|
"eval_gen_len": 250.455, |
|
"eval_loss": 3.2671351432800293, |
|
"eval_rouge1": 45.6791, |
|
"eval_rouge2": 17.5028, |
|
"eval_rougeL": 24.0691, |
|
"eval_rougeLsum": 41.9219, |
|
"eval_runtime": 1163.6259, |
|
"eval_samples_per_second": 4.831, |
|
"eval_steps_per_second": 0.303, |
|
"step": 16500 |
|
}, |
|
{ |
|
"epoch": 5.91, |
|
"learning_rate": 1.1208e-05, |
|
"loss": 3.0764, |
|
"step": 16600 |
|
}, |
|
{ |
|
"epoch": 5.94, |
|
"learning_rate": 1.1074666666666667e-05, |
|
"loss": 3.0822, |
|
"step": 16700 |
|
}, |
|
{ |
|
"epoch": 5.98, |
|
"learning_rate": 1.0941333333333333e-05, |
|
"loss": 3.0452, |
|
"step": 16800 |
|
}, |
|
{ |
|
"epoch": 6.01, |
|
"learning_rate": 1.0808e-05, |
|
"loss": 3.0631, |
|
"step": 16900 |
|
}, |
|
{ |
|
"epoch": 6.05, |
|
"learning_rate": 1.0674666666666666e-05, |
|
"loss": 3.0142, |
|
"step": 17000 |
|
}, |
|
{ |
|
"epoch": 6.05, |
|
"eval_gen_len": 245.6204, |
|
"eval_loss": 3.270817279815674, |
|
"eval_rouge1": 46.0287, |
|
"eval_rouge2": 17.8079, |
|
"eval_rougeL": 24.2916, |
|
"eval_rougeLsum": 42.2369, |
|
"eval_runtime": 1077.587, |
|
"eval_samples_per_second": 5.217, |
|
"eval_steps_per_second": 0.327, |
|
"step": 17000 |
|
}, |
|
{ |
|
"epoch": 6.09, |
|
"learning_rate": 1.0541333333333334e-05, |
|
"loss": 3.0106, |
|
"step": 17100 |
|
}, |
|
{ |
|
"epoch": 6.12, |
|
"learning_rate": 1.0408e-05, |
|
"loss": 3.0208, |
|
"step": 17200 |
|
}, |
|
{ |
|
"epoch": 6.16, |
|
"learning_rate": 1.0274666666666667e-05, |
|
"loss": 3.0455, |
|
"step": 17300 |
|
}, |
|
{ |
|
"epoch": 6.19, |
|
"learning_rate": 1.0141333333333332e-05, |
|
"loss": 3.0404, |
|
"step": 17400 |
|
}, |
|
{ |
|
"epoch": 6.23, |
|
"learning_rate": 1.0008e-05, |
|
"loss": 3.0312, |
|
"step": 17500 |
|
}, |
|
{ |
|
"epoch": 6.23, |
|
"eval_gen_len": 236.2234, |
|
"eval_loss": 3.270146608352661, |
|
"eval_rouge1": 45.5731, |
|
"eval_rouge2": 17.5404, |
|
"eval_rougeL": 24.0925, |
|
"eval_rougeLsum": 41.7584, |
|
"eval_runtime": 1079.8219, |
|
"eval_samples_per_second": 5.206, |
|
"eval_steps_per_second": 0.326, |
|
"step": 17500 |
|
}, |
|
{ |
|
"epoch": 6.26, |
|
"learning_rate": 9.874666666666667e-06, |
|
"loss": 3.033, |
|
"step": 17600 |
|
}, |
|
{ |
|
"epoch": 6.3, |
|
"learning_rate": 9.741333333333334e-06, |
|
"loss": 3.045, |
|
"step": 17700 |
|
}, |
|
{ |
|
"epoch": 6.33, |
|
"learning_rate": 9.608e-06, |
|
"loss": 3.0339, |
|
"step": 17800 |
|
}, |
|
{ |
|
"epoch": 6.37, |
|
"learning_rate": 9.474666666666668e-06, |
|
"loss": 3.034, |
|
"step": 17900 |
|
}, |
|
{ |
|
"epoch": 6.41, |
|
"learning_rate": 9.341333333333333e-06, |
|
"loss": 3.0231, |
|
"step": 18000 |
|
}, |
|
{ |
|
"epoch": 6.41, |
|
"eval_gen_len": 260.1686, |
|
"eval_loss": 3.271860361099243, |
|
"eval_rouge1": 46.1094, |
|
"eval_rouge2": 17.7117, |
|
"eval_rougeL": 24.1117, |
|
"eval_rougeLsum": 42.2882, |
|
"eval_runtime": 1163.8918, |
|
"eval_samples_per_second": 4.83, |
|
"eval_steps_per_second": 0.302, |
|
"step": 18000 |
|
}, |
|
{ |
|
"epoch": 6.44, |
|
"learning_rate": 9.208e-06, |
|
"loss": 3.0454, |
|
"step": 18100 |
|
}, |
|
{ |
|
"epoch": 6.48, |
|
"learning_rate": 9.074666666666666e-06, |
|
"loss": 3.0343, |
|
"step": 18200 |
|
}, |
|
{ |
|
"epoch": 6.51, |
|
"learning_rate": 8.941333333333334e-06, |
|
"loss": 3.0386, |
|
"step": 18300 |
|
}, |
|
{ |
|
"epoch": 6.55, |
|
"learning_rate": 8.808000000000001e-06, |
|
"loss": 3.0429, |
|
"step": 18400 |
|
}, |
|
{ |
|
"epoch": 6.58, |
|
"learning_rate": 8.674666666666667e-06, |
|
"loss": 3.0414, |
|
"step": 18500 |
|
}, |
|
{ |
|
"epoch": 6.58, |
|
"eval_gen_len": 245.0961, |
|
"eval_loss": 3.2702643871307373, |
|
"eval_rouge1": 45.9178, |
|
"eval_rouge2": 17.6987, |
|
"eval_rougeL": 24.1882, |
|
"eval_rougeLsum": 42.1382, |
|
"eval_runtime": 1111.098, |
|
"eval_samples_per_second": 5.06, |
|
"eval_steps_per_second": 0.317, |
|
"step": 18500 |
|
}, |
|
{ |
|
"epoch": 6.62, |
|
"learning_rate": 8.541333333333334e-06, |
|
"loss": 3.0242, |
|
"step": 18600 |
|
}, |
|
{ |
|
"epoch": 6.65, |
|
"learning_rate": 8.408e-06, |
|
"loss": 3.0449, |
|
"step": 18700 |
|
}, |
|
{ |
|
"epoch": 6.69, |
|
"learning_rate": 8.274666666666667e-06, |
|
"loss": 3.0392, |
|
"step": 18800 |
|
}, |
|
{ |
|
"epoch": 6.73, |
|
"learning_rate": 8.141333333333333e-06, |
|
"loss": 3.0336, |
|
"step": 18900 |
|
}, |
|
{ |
|
"epoch": 6.76, |
|
"learning_rate": 8.008e-06, |
|
"loss": 3.0434, |
|
"step": 19000 |
|
}, |
|
{ |
|
"epoch": 6.76, |
|
"eval_gen_len": 247.8225, |
|
"eval_loss": 3.2714767456054688, |
|
"eval_rouge1": 46.0129, |
|
"eval_rouge2": 17.7545, |
|
"eval_rougeL": 24.2235, |
|
"eval_rougeLsum": 42.245, |
|
"eval_runtime": 1091.4396, |
|
"eval_samples_per_second": 5.151, |
|
"eval_steps_per_second": 0.323, |
|
"step": 19000 |
|
}, |
|
{ |
|
"epoch": 6.8, |
|
"learning_rate": 7.874666666666666e-06, |
|
"loss": 3.0021, |
|
"step": 19100 |
|
}, |
|
{ |
|
"epoch": 6.83, |
|
"learning_rate": 7.741333333333335e-06, |
|
"loss": 3.0498, |
|
"step": 19200 |
|
}, |
|
{ |
|
"epoch": 6.87, |
|
"learning_rate": 7.608e-06, |
|
"loss": 3.0314, |
|
"step": 19300 |
|
}, |
|
{ |
|
"epoch": 6.9, |
|
"learning_rate": 7.476e-06, |
|
"loss": 3.0437, |
|
"step": 19400 |
|
}, |
|
{ |
|
"epoch": 6.94, |
|
"learning_rate": 7.342666666666667e-06, |
|
"loss": 3.0456, |
|
"step": 19500 |
|
}, |
|
{ |
|
"epoch": 6.94, |
|
"eval_gen_len": 256.9835, |
|
"eval_loss": 3.2681996822357178, |
|
"eval_rouge1": 45.8634, |
|
"eval_rouge2": 17.6462, |
|
"eval_rougeL": 24.1366, |
|
"eval_rougeLsum": 42.1194, |
|
"eval_runtime": 1174.1005, |
|
"eval_samples_per_second": 4.788, |
|
"eval_steps_per_second": 0.3, |
|
"step": 19500 |
|
}, |
|
{ |
|
"epoch": 6.97, |
|
"learning_rate": 7.209333333333334e-06, |
|
"loss": 3.0172, |
|
"step": 19600 |
|
}, |
|
{ |
|
"epoch": 7.01, |
|
"learning_rate": 7.077333333333333e-06, |
|
"loss": 3.0364, |
|
"step": 19700 |
|
}, |
|
{ |
|
"epoch": 7.05, |
|
"learning_rate": 6.944e-06, |
|
"loss": 3.0109, |
|
"step": 19800 |
|
}, |
|
{ |
|
"epoch": 7.08, |
|
"learning_rate": 6.8106666666666665e-06, |
|
"loss": 2.998, |
|
"step": 19900 |
|
}, |
|
{ |
|
"epoch": 7.12, |
|
"learning_rate": 6.677333333333334e-06, |
|
"loss": 3.0188, |
|
"step": 20000 |
|
}, |
|
{ |
|
"epoch": 7.12, |
|
"eval_gen_len": 240.1866, |
|
"eval_loss": 3.2752106189727783, |
|
"eval_rouge1": 45.8366, |
|
"eval_rouge2": 17.6771, |
|
"eval_rougeL": 24.165, |
|
"eval_rougeLsum": 42.0438, |
|
"eval_runtime": 1085.4389, |
|
"eval_samples_per_second": 5.179, |
|
"eval_steps_per_second": 0.324, |
|
"step": 20000 |
|
}, |
|
{ |
|
"epoch": 7.15, |
|
"learning_rate": 6.544e-06, |
|
"loss": 3.0119, |
|
"step": 20100 |
|
}, |
|
{ |
|
"epoch": 7.19, |
|
"learning_rate": 6.410666666666667e-06, |
|
"loss": 3.0091, |
|
"step": 20200 |
|
}, |
|
{ |
|
"epoch": 7.22, |
|
"learning_rate": 6.2773333333333334e-06, |
|
"loss": 3.0077, |
|
"step": 20300 |
|
}, |
|
{ |
|
"epoch": 7.26, |
|
"learning_rate": 6.144000000000001e-06, |
|
"loss": 2.9942, |
|
"step": 20400 |
|
}, |
|
{ |
|
"epoch": 7.3, |
|
"learning_rate": 6.010666666666667e-06, |
|
"loss": 3.0227, |
|
"step": 20500 |
|
}, |
|
{ |
|
"epoch": 7.3, |
|
"eval_gen_len": 245.8337, |
|
"eval_loss": 3.2722229957580566, |
|
"eval_rouge1": 46.0509, |
|
"eval_rouge2": 17.8248, |
|
"eval_rougeL": 24.2389, |
|
"eval_rougeLsum": 42.2681, |
|
"eval_runtime": 1093.3258, |
|
"eval_samples_per_second": 5.142, |
|
"eval_steps_per_second": 0.322, |
|
"step": 20500 |
|
}, |
|
{ |
|
"epoch": 7.33, |
|
"learning_rate": 5.877333333333334e-06, |
|
"loss": 2.9996, |
|
"step": 20600 |
|
}, |
|
{ |
|
"epoch": 7.37, |
|
"learning_rate": 5.744e-06, |
|
"loss": 3.0046, |
|
"step": 20700 |
|
}, |
|
{ |
|
"epoch": 7.4, |
|
"learning_rate": 5.610666666666667e-06, |
|
"loss": 3.0018, |
|
"step": 20800 |
|
}, |
|
{ |
|
"epoch": 7.44, |
|
"learning_rate": 5.4773333333333335e-06, |
|
"loss": 3.0096, |
|
"step": 20900 |
|
}, |
|
{ |
|
"epoch": 7.47, |
|
"learning_rate": 5.344e-06, |
|
"loss": 2.9895, |
|
"step": 21000 |
|
}, |
|
{ |
|
"epoch": 7.47, |
|
"eval_gen_len": 243.867, |
|
"eval_loss": 3.2725987434387207, |
|
"eval_rouge1": 45.7896, |
|
"eval_rouge2": 17.5833, |
|
"eval_rougeL": 24.1226, |
|
"eval_rougeLsum": 42.016, |
|
"eval_runtime": 1110.8794, |
|
"eval_samples_per_second": 5.061, |
|
"eval_steps_per_second": 0.317, |
|
"step": 21000 |
|
}, |
|
{ |
|
"epoch": 7.51, |
|
"learning_rate": 5.2106666666666665e-06, |
|
"loss": 3.0186, |
|
"step": 21100 |
|
}, |
|
{ |
|
"epoch": 7.54, |
|
"learning_rate": 5.077333333333334e-06, |
|
"loss": 3.0337, |
|
"step": 21200 |
|
}, |
|
{ |
|
"epoch": 7.58, |
|
"learning_rate": 4.9440000000000004e-06, |
|
"loss": 3.0136, |
|
"step": 21300 |
|
}, |
|
{ |
|
"epoch": 7.62, |
|
"learning_rate": 4.810666666666667e-06, |
|
"loss": 3.0109, |
|
"step": 21400 |
|
}, |
|
{ |
|
"epoch": 7.65, |
|
"learning_rate": 4.6773333333333335e-06, |
|
"loss": 3.0146, |
|
"step": 21500 |
|
}, |
|
{ |
|
"epoch": 7.65, |
|
"eval_gen_len": 244.0598, |
|
"eval_loss": 3.269317865371704, |
|
"eval_rouge1": 46.0179, |
|
"eval_rouge2": 17.6952, |
|
"eval_rougeL": 24.2204, |
|
"eval_rougeLsum": 42.2436, |
|
"eval_runtime": 1075.6479, |
|
"eval_samples_per_second": 5.227, |
|
"eval_steps_per_second": 0.327, |
|
"step": 21500 |
|
}, |
|
{ |
|
"epoch": 7.69, |
|
"learning_rate": 4.544e-06, |
|
"loss": 3.0195, |
|
"step": 21600 |
|
}, |
|
{ |
|
"epoch": 7.72, |
|
"learning_rate": 4.4106666666666666e-06, |
|
"loss": 3.0103, |
|
"step": 21700 |
|
}, |
|
{ |
|
"epoch": 7.76, |
|
"learning_rate": 4.277333333333333e-06, |
|
"loss": 3.0117, |
|
"step": 21800 |
|
}, |
|
{ |
|
"epoch": 7.79, |
|
"learning_rate": 4.144e-06, |
|
"loss": 3.0012, |
|
"step": 21900 |
|
}, |
|
{ |
|
"epoch": 7.83, |
|
"learning_rate": 4.010666666666667e-06, |
|
"loss": 3.014, |
|
"step": 22000 |
|
}, |
|
{ |
|
"epoch": 7.83, |
|
"eval_gen_len": 240.4804, |
|
"eval_loss": 3.2708346843719482, |
|
"eval_rouge1": 46.0704, |
|
"eval_rouge2": 17.75, |
|
"eval_rougeL": 24.2308, |
|
"eval_rougeLsum": 42.2591, |
|
"eval_runtime": 1055.0456, |
|
"eval_samples_per_second": 5.329, |
|
"eval_steps_per_second": 0.334, |
|
"step": 22000 |
|
}, |
|
{ |
|
"epoch": 7.86, |
|
"learning_rate": 3.8773333333333335e-06, |
|
"loss": 3.001, |
|
"step": 22100 |
|
}, |
|
{ |
|
"epoch": 7.9, |
|
"learning_rate": 3.744e-06, |
|
"loss": 3.0133, |
|
"step": 22200 |
|
}, |
|
{ |
|
"epoch": 7.94, |
|
"learning_rate": 3.6106666666666666e-06, |
|
"loss": 3.0007, |
|
"step": 22300 |
|
}, |
|
{ |
|
"epoch": 7.97, |
|
"learning_rate": 3.4773333333333336e-06, |
|
"loss": 3.0189, |
|
"step": 22400 |
|
}, |
|
{ |
|
"epoch": 8.01, |
|
"learning_rate": 3.344e-06, |
|
"loss": 3.0427, |
|
"step": 22500 |
|
}, |
|
{ |
|
"epoch": 8.01, |
|
"eval_gen_len": 242.4203, |
|
"eval_loss": 3.27339243888855, |
|
"eval_rouge1": 46.0662, |
|
"eval_rouge2": 17.7231, |
|
"eval_rougeL": 24.1915, |
|
"eval_rougeLsum": 42.2227, |
|
"eval_runtime": 1083.3051, |
|
"eval_samples_per_second": 5.19, |
|
"eval_steps_per_second": 0.325, |
|
"step": 22500 |
|
}, |
|
{ |
|
"epoch": 8.04, |
|
"learning_rate": 3.210666666666667e-06, |
|
"loss": 2.995, |
|
"step": 22600 |
|
}, |
|
{ |
|
"epoch": 8.08, |
|
"learning_rate": 3.0773333333333336e-06, |
|
"loss": 2.9946, |
|
"step": 22700 |
|
}, |
|
{ |
|
"epoch": 8.11, |
|
"learning_rate": 2.944e-06, |
|
"loss": 3.0003, |
|
"step": 22800 |
|
}, |
|
{ |
|
"epoch": 8.15, |
|
"learning_rate": 2.8106666666666666e-06, |
|
"loss": 2.9959, |
|
"step": 22900 |
|
}, |
|
{ |
|
"epoch": 8.19, |
|
"learning_rate": 2.6773333333333336e-06, |
|
"loss": 2.9835, |
|
"step": 23000 |
|
}, |
|
{ |
|
"epoch": 8.19, |
|
"eval_gen_len": 236.6266, |
|
"eval_loss": 3.273963212966919, |
|
"eval_rouge1": 46.165, |
|
"eval_rouge2": 17.8947, |
|
"eval_rougeL": 24.366, |
|
"eval_rougeLsum": 42.3521, |
|
"eval_runtime": 1047.6593, |
|
"eval_samples_per_second": 5.366, |
|
"eval_steps_per_second": 0.336, |
|
"step": 23000 |
|
}, |
|
{ |
|
"epoch": 8.22, |
|
"learning_rate": 2.544e-06, |
|
"loss": 2.9922, |
|
"step": 23100 |
|
}, |
|
{ |
|
"epoch": 8.26, |
|
"learning_rate": 2.4106666666666667e-06, |
|
"loss": 2.9937, |
|
"step": 23200 |
|
}, |
|
{ |
|
"epoch": 8.29, |
|
"learning_rate": 2.277333333333333e-06, |
|
"loss": 2.9933, |
|
"step": 23300 |
|
}, |
|
{ |
|
"epoch": 8.33, |
|
"learning_rate": 2.144e-06, |
|
"loss": 2.9921, |
|
"step": 23400 |
|
}, |
|
{ |
|
"epoch": 8.36, |
|
"learning_rate": 2.0106666666666667e-06, |
|
"loss": 2.987, |
|
"step": 23500 |
|
}, |
|
{ |
|
"epoch": 8.36, |
|
"eval_gen_len": 238.479, |
|
"eval_loss": 3.2719457149505615, |
|
"eval_rouge1": 45.9025, |
|
"eval_rouge2": 17.7625, |
|
"eval_rougeL": 24.2432, |
|
"eval_rougeLsum": 42.1257, |
|
"eval_runtime": 1104.1087, |
|
"eval_samples_per_second": 5.092, |
|
"eval_steps_per_second": 0.319, |
|
"step": 23500 |
|
}, |
|
{ |
|
"epoch": 8.4, |
|
"learning_rate": 1.8773333333333332e-06, |
|
"loss": 2.9925, |
|
"step": 23600 |
|
}, |
|
{ |
|
"epoch": 8.43, |
|
"learning_rate": 1.7440000000000002e-06, |
|
"loss": 2.9924, |
|
"step": 23700 |
|
}, |
|
{ |
|
"epoch": 8.47, |
|
"learning_rate": 1.6106666666666667e-06, |
|
"loss": 3.0224, |
|
"step": 23800 |
|
}, |
|
{ |
|
"epoch": 8.51, |
|
"learning_rate": 1.4773333333333334e-06, |
|
"loss": 3.0137, |
|
"step": 23900 |
|
}, |
|
{ |
|
"epoch": 8.54, |
|
"learning_rate": 1.344e-06, |
|
"loss": 2.9922, |
|
"step": 24000 |
|
}, |
|
{ |
|
"epoch": 8.54, |
|
"eval_gen_len": 245.2081, |
|
"eval_loss": 3.2731070518493652, |
|
"eval_rouge1": 46.1971, |
|
"eval_rouge2": 17.7962, |
|
"eval_rougeL": 24.2279, |
|
"eval_rougeLsum": 42.3853, |
|
"eval_runtime": 1087.3894, |
|
"eval_samples_per_second": 5.17, |
|
"eval_steps_per_second": 0.324, |
|
"step": 24000 |
|
}, |
|
{ |
|
"epoch": 8.58, |
|
"learning_rate": 1.2106666666666667e-06, |
|
"loss": 2.9974, |
|
"step": 24100 |
|
}, |
|
{ |
|
"epoch": 8.61, |
|
"learning_rate": 1.0773333333333332e-06, |
|
"loss": 2.9955, |
|
"step": 24200 |
|
}, |
|
{ |
|
"epoch": 8.65, |
|
"learning_rate": 9.44e-07, |
|
"loss": 2.9914, |
|
"step": 24300 |
|
}, |
|
{ |
|
"epoch": 8.68, |
|
"learning_rate": 8.106666666666667e-07, |
|
"loss": 2.9803, |
|
"step": 24400 |
|
}, |
|
{ |
|
"epoch": 8.72, |
|
"learning_rate": 6.773333333333334e-07, |
|
"loss": 2.9788, |
|
"step": 24500 |
|
}, |
|
{ |
|
"epoch": 8.72, |
|
"eval_gen_len": 240.1747, |
|
"eval_loss": 3.2718217372894287, |
|
"eval_rouge1": 46.0806, |
|
"eval_rouge2": 17.8417, |
|
"eval_rougeL": 24.3261, |
|
"eval_rougeLsum": 42.264, |
|
"eval_runtime": 1088.0734, |
|
"eval_samples_per_second": 5.167, |
|
"eval_steps_per_second": 0.324, |
|
"step": 24500 |
|
}, |
|
{ |
|
"epoch": 8.75, |
|
"learning_rate": 5.44e-07, |
|
"loss": 3.001, |
|
"step": 24600 |
|
}, |
|
{ |
|
"epoch": 8.79, |
|
"learning_rate": 4.106666666666667e-07, |
|
"loss": 2.9962, |
|
"step": 24700 |
|
}, |
|
{ |
|
"epoch": 8.83, |
|
"learning_rate": 2.7733333333333333e-07, |
|
"loss": 2.9987, |
|
"step": 24800 |
|
}, |
|
{ |
|
"epoch": 8.86, |
|
"learning_rate": 1.44e-07, |
|
"loss": 2.9703, |
|
"step": 24900 |
|
}, |
|
{ |
|
"epoch": 8.9, |
|
"learning_rate": 1.0666666666666668e-08, |
|
"loss": 2.9878, |
|
"step": 25000 |
|
}, |
|
{ |
|
"epoch": 8.9, |
|
"eval_gen_len": 242.5598, |
|
"eval_loss": 3.2715346813201904, |
|
"eval_rouge1": 46.0618, |
|
"eval_rouge2": 17.7725, |
|
"eval_rougeL": 24.2234, |
|
"eval_rougeLsum": 42.2574, |
|
"eval_runtime": 1097.8453, |
|
"eval_samples_per_second": 5.121, |
|
"eval_steps_per_second": 0.321, |
|
"step": 25000 |
|
}, |
|
{ |
|
"epoch": 8.9, |
|
"step": 25000, |
|
"total_flos": 5.0265269514797056e+17, |
|
"train_loss": 3.1441598370170594, |
|
"train_runtime": 61182.0827, |
|
"train_samples_per_second": 6.538, |
|
"train_steps_per_second": 0.409 |
|
} |
|
], |
|
"max_steps": 25000, |
|
"num_train_epochs": 9, |
|
"total_flos": 5.0265269514797056e+17, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|