| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 19.653179190751445, | |
| "global_step": 17000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 1.15606936416185e-05, | |
| "loss": 4.1602, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_gen_len": 15.5003, | |
| "eval_loss": 2.8024423122406006, | |
| "eval_rouge1": 0.2552, | |
| "eval_rouge2": 0.1019, | |
| "eval_rougeL": 0.2331, | |
| "eval_rougeLsum": 0.2332, | |
| "eval_runtime": 184.148, | |
| "eval_samples_per_second": 18.778, | |
| "eval_steps_per_second": 1.178, | |
| "step": 865 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 1.9835716458777003e-05, | |
| "loss": 3.1173, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 1.922725889869182e-05, | |
| "loss": 2.9624, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_gen_len": 15.8271, | |
| "eval_loss": 2.6451988220214844, | |
| "eval_rouge1": 0.2882, | |
| "eval_rouge2": 0.1183, | |
| "eval_rougeL": 0.2626, | |
| "eval_rougeLsum": 0.2626, | |
| "eval_runtime": 182.4271, | |
| "eval_samples_per_second": 18.956, | |
| "eval_steps_per_second": 1.19, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 1.8618801338606633e-05, | |
| "loss": 2.8766, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 1.801034377852145e-05, | |
| "loss": 2.8206, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_gen_len": 16.2183, | |
| "eval_loss": 2.5787582397460938, | |
| "eval_rouge1": 0.2952, | |
| "eval_rouge2": 0.1218, | |
| "eval_rougeL": 0.2684, | |
| "eval_rougeLsum": 0.2685, | |
| "eval_runtime": 182.2993, | |
| "eval_samples_per_second": 18.969, | |
| "eval_steps_per_second": 1.19, | |
| "step": 2595 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 1.7401886218436267e-05, | |
| "loss": 2.7764, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_gen_len": 16.3418, | |
| "eval_loss": 2.5332629680633545, | |
| "eval_rouge1": 0.3029, | |
| "eval_rouge2": 0.1258, | |
| "eval_rougeL": 0.2755, | |
| "eval_rougeLsum": 0.2754, | |
| "eval_runtime": 181.4496, | |
| "eval_samples_per_second": 19.058, | |
| "eval_steps_per_second": 1.196, | |
| "step": 3460 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 1.679342865835108e-05, | |
| "loss": 2.7414, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 1.6184971098265897e-05, | |
| "loss": 2.6997, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_gen_len": 16.4101, | |
| "eval_loss": 2.504000186920166, | |
| "eval_rouge1": 0.3046, | |
| "eval_rouge2": 0.1273, | |
| "eval_rougeL": 0.2773, | |
| "eval_rougeLsum": 0.2771, | |
| "eval_runtime": 181.6751, | |
| "eval_samples_per_second": 19.034, | |
| "eval_steps_per_second": 1.194, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "learning_rate": 1.5576513538180714e-05, | |
| "loss": 2.6795, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "learning_rate": 1.4968055978095528e-05, | |
| "loss": 2.6647, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_gen_len": 16.354, | |
| "eval_loss": 2.4734175205230713, | |
| "eval_rouge1": 0.3069, | |
| "eval_rouge2": 0.1298, | |
| "eval_rougeL": 0.2802, | |
| "eval_rougeLsum": 0.2801, | |
| "eval_runtime": 181.0262, | |
| "eval_samples_per_second": 19.102, | |
| "eval_steps_per_second": 1.199, | |
| "step": 5190 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "learning_rate": 1.4359598418010346e-05, | |
| "loss": 2.6287, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "learning_rate": 1.375114085792516e-05, | |
| "loss": 2.6196, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_gen_len": 16.526, | |
| "eval_loss": 2.4547228813171387, | |
| "eval_rouge1": 0.3077, | |
| "eval_rouge2": 0.131, | |
| "eval_rougeL": 0.2804, | |
| "eval_rougeLsum": 0.2804, | |
| "eval_runtime": 182.5951, | |
| "eval_samples_per_second": 18.938, | |
| "eval_steps_per_second": 1.188, | |
| "step": 6055 | |
| }, | |
| { | |
| "epoch": 7.51, | |
| "learning_rate": 1.3142683297839978e-05, | |
| "loss": 2.5916, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_gen_len": 16.4419, | |
| "eval_loss": 2.435168504714966, | |
| "eval_rouge1": 0.3107, | |
| "eval_rouge2": 0.1332, | |
| "eval_rougeL": 0.2833, | |
| "eval_rougeLsum": 0.2834, | |
| "eval_runtime": 181.1692, | |
| "eval_samples_per_second": 19.087, | |
| "eval_steps_per_second": 1.198, | |
| "step": 6920 | |
| }, | |
| { | |
| "epoch": 8.09, | |
| "learning_rate": 1.2534225737754793e-05, | |
| "loss": 2.5986, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 8.67, | |
| "learning_rate": 1.1925768177669607e-05, | |
| "loss": 2.5716, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_gen_len": 16.4023, | |
| "eval_loss": 2.4196717739105225, | |
| "eval_rouge1": 0.3092, | |
| "eval_rouge2": 0.132, | |
| "eval_rougeL": 0.2819, | |
| "eval_rougeLsum": 0.2817, | |
| "eval_runtime": 181.3331, | |
| "eval_samples_per_second": 19.07, | |
| "eval_steps_per_second": 1.197, | |
| "step": 7785 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "learning_rate": 1.1317310617584426e-05, | |
| "loss": 2.5347, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 9.83, | |
| "learning_rate": 1.070885305749924e-05, | |
| "loss": 2.548, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_gen_len": 16.3256, | |
| "eval_loss": 2.4123945236206055, | |
| "eval_rouge1": 0.3094, | |
| "eval_rouge2": 0.1329, | |
| "eval_rougeL": 0.2822, | |
| "eval_rougeLsum": 0.2823, | |
| "eval_runtime": 181.1565, | |
| "eval_samples_per_second": 19.088, | |
| "eval_steps_per_second": 1.198, | |
| "step": 8650 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "learning_rate": 1.0100395497414058e-05, | |
| "loss": 2.5237, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 10.98, | |
| "learning_rate": 9.491937937328873e-06, | |
| "loss": 2.5352, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_gen_len": 16.3774, | |
| "eval_loss": 2.402285575866699, | |
| "eval_rouge1": 0.3096, | |
| "eval_rouge2": 0.1323, | |
| "eval_rougeL": 0.2821, | |
| "eval_rougeLsum": 0.282, | |
| "eval_runtime": 181.4776, | |
| "eval_samples_per_second": 19.055, | |
| "eval_steps_per_second": 1.196, | |
| "step": 9515 | |
| }, | |
| { | |
| "epoch": 11.56, | |
| "learning_rate": 8.883480377243688e-06, | |
| "loss": 2.5087, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_gen_len": 16.4592, | |
| "eval_loss": 2.39676570892334, | |
| "eval_rouge1": 0.3107, | |
| "eval_rouge2": 0.1326, | |
| "eval_rougeL": 0.2831, | |
| "eval_rougeLsum": 0.2832, | |
| "eval_runtime": 182.2228, | |
| "eval_samples_per_second": 18.977, | |
| "eval_steps_per_second": 1.191, | |
| "step": 10380 | |
| }, | |
| { | |
| "epoch": 12.14, | |
| "learning_rate": 8.275022817158503e-06, | |
| "loss": 2.514, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 12.72, | |
| "learning_rate": 7.66656525707332e-06, | |
| "loss": 2.4865, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_gen_len": 16.4586, | |
| "eval_loss": 2.38985013961792, | |
| "eval_rouge1": 0.3117, | |
| "eval_rouge2": 0.1339, | |
| "eval_rougeL": 0.2841, | |
| "eval_rougeLsum": 0.2842, | |
| "eval_runtime": 181.3381, | |
| "eval_samples_per_second": 19.069, | |
| "eval_steps_per_second": 1.197, | |
| "step": 11245 | |
| }, | |
| { | |
| "epoch": 13.29, | |
| "learning_rate": 7.058107696988135e-06, | |
| "loss": 2.4861, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 13.87, | |
| "learning_rate": 6.449650136902952e-06, | |
| "loss": 2.5081, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_gen_len": 16.4072, | |
| "eval_loss": 2.3885083198547363, | |
| "eval_rouge1": 0.3124, | |
| "eval_rouge2": 0.1337, | |
| "eval_rougeL": 0.2845, | |
| "eval_rougeLsum": 0.2845, | |
| "eval_runtime": 181.2166, | |
| "eval_samples_per_second": 19.082, | |
| "eval_steps_per_second": 1.197, | |
| "step": 12110 | |
| }, | |
| { | |
| "epoch": 14.45, | |
| "learning_rate": 5.841192576817768e-06, | |
| "loss": 2.4748, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_gen_len": 16.4572, | |
| "eval_loss": 2.3791513442993164, | |
| "eval_rouge1": 0.3131, | |
| "eval_rouge2": 0.134, | |
| "eval_rougeL": 0.285, | |
| "eval_rougeLsum": 0.2849, | |
| "eval_runtime": 181.192, | |
| "eval_samples_per_second": 19.085, | |
| "eval_steps_per_second": 1.198, | |
| "step": 12975 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "learning_rate": 5.232735016732583e-06, | |
| "loss": 2.4641, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 15.61, | |
| "learning_rate": 4.6242774566473994e-06, | |
| "loss": 2.469, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_gen_len": 16.3907, | |
| "eval_loss": 2.3784728050231934, | |
| "eval_rouge1": 0.3119, | |
| "eval_rouge2": 0.1333, | |
| "eval_rougeL": 0.2841, | |
| "eval_rougeLsum": 0.284, | |
| "eval_runtime": 180.9918, | |
| "eval_samples_per_second": 19.106, | |
| "eval_steps_per_second": 1.199, | |
| "step": 13840 | |
| }, | |
| { | |
| "epoch": 16.18, | |
| "learning_rate": 4.0158198965622155e-06, | |
| "loss": 2.4771, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 16.76, | |
| "learning_rate": 3.4073623364770307e-06, | |
| "loss": 2.4587, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_gen_len": 16.4488, | |
| "eval_loss": 2.3765616416931152, | |
| "eval_rouge1": 0.3115, | |
| "eval_rouge2": 0.1334, | |
| "eval_rougeL": 0.2835, | |
| "eval_rougeLsum": 0.2834, | |
| "eval_runtime": 181.9602, | |
| "eval_samples_per_second": 19.004, | |
| "eval_steps_per_second": 1.193, | |
| "step": 14705 | |
| }, | |
| { | |
| "epoch": 17.34, | |
| "learning_rate": 2.798904776391847e-06, | |
| "loss": 2.4679, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 17.92, | |
| "learning_rate": 2.1904472163066628e-06, | |
| "loss": 2.4428, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_gen_len": 16.4141, | |
| "eval_loss": 2.375408411026001, | |
| "eval_rouge1": 0.3137, | |
| "eval_rouge2": 0.1346, | |
| "eval_rougeL": 0.2854, | |
| "eval_rougeLsum": 0.2854, | |
| "eval_runtime": 181.1095, | |
| "eval_samples_per_second": 19.093, | |
| "eval_steps_per_second": 1.198, | |
| "step": 15570 | |
| }, | |
| { | |
| "epoch": 18.5, | |
| "learning_rate": 1.5819896562214788e-06, | |
| "loss": 2.4548, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_gen_len": 16.4451, | |
| "eval_loss": 2.3725531101226807, | |
| "eval_rouge1": 0.3129, | |
| "eval_rouge2": 0.1341, | |
| "eval_rougeL": 0.2849, | |
| "eval_rougeLsum": 0.285, | |
| "eval_runtime": 180.98, | |
| "eval_samples_per_second": 19.107, | |
| "eval_steps_per_second": 1.199, | |
| "step": 16435 | |
| }, | |
| { | |
| "epoch": 19.08, | |
| "learning_rate": 9.735320961362946e-07, | |
| "loss": 2.4426, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 19.65, | |
| "learning_rate": 3.650745360511105e-07, | |
| "loss": 2.4389, | |
| "step": 17000 | |
| } | |
| ], | |
| "max_steps": 17300, | |
| "num_train_epochs": 20, | |
| "total_flos": 6.944051276532941e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |