| { | |
| "best_metric": 1.3323031663894653, | |
| "best_model_checkpoint": "longt5_xl_summ_screen_bp_10/checkpoint-57", | |
| "epoch": 9.73913043478261, | |
| "eval_steps": 500, | |
| "global_step": 140, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.001, | |
| "loss": 2.8751, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.001, | |
| "loss": 3.5717, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.001, | |
| "loss": 2.8585, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.001, | |
| "loss": 2.5104, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.001, | |
| "loss": 3.2659, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 0.001, | |
| "loss": 3.4634, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.001, | |
| "loss": 2.4559, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_gen_len": 511.0, | |
| "eval_loss": 2.0707387924194336, | |
| "eval_rouge1": 11.7833, | |
| "eval_rouge2": 1.6011, | |
| "eval_rougeL": 11.1858, | |
| "eval_rougeLsum": 10.3025, | |
| "eval_runtime": 1812.8885, | |
| "eval_samples_per_second": 0.186, | |
| "eval_steps_per_second": 0.024, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.001, | |
| "loss": 2.1517, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.001, | |
| "loss": 2.2029, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 0.001, | |
| "loss": 2.1161, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.001, | |
| "loss": 1.9513, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.001, | |
| "loss": 1.7095, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 0.001, | |
| "loss": 1.6535, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 0.001, | |
| "loss": 1.6238, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "eval_gen_len": 511.0, | |
| "eval_loss": 1.5286704301834106, | |
| "eval_rouge1": 19.0489, | |
| "eval_rouge2": 4.687, | |
| "eval_rougeL": 16.6504, | |
| "eval_rougeLsum": 17.1808, | |
| "eval_runtime": 1807.4269, | |
| "eval_samples_per_second": 0.187, | |
| "eval_steps_per_second": 0.024, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 0.001, | |
| "loss": 1.5804, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 0.001, | |
| "loss": 1.511, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 0.001, | |
| "loss": 1.4961, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 0.001, | |
| "loss": 1.4334, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 0.001, | |
| "loss": 1.3994, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 0.001, | |
| "loss": 1.4018, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 0.001, | |
| "loss": 1.3964, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_gen_len": 511.0, | |
| "eval_loss": 1.3520147800445557, | |
| "eval_rouge1": 21.9994, | |
| "eval_rouge2": 5.8519, | |
| "eval_rougeL": 18.9231, | |
| "eval_rougeLsum": 19.958, | |
| "eval_runtime": 1809.4299, | |
| "eval_samples_per_second": 0.187, | |
| "eval_steps_per_second": 0.024, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 0.001, | |
| "loss": 1.3428, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.001, | |
| "loss": 1.3034, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 0.001, | |
| "loss": 1.4137, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 0.001, | |
| "loss": 1.4083, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 0.001, | |
| "loss": 1.3075, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 0.001, | |
| "loss": 1.2527, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 0.001, | |
| "loss": 1.2538, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "eval_gen_len": 497.2455621301775, | |
| "eval_loss": 1.3323031663894653, | |
| "eval_rouge1": 22.9554, | |
| "eval_rouge2": 6.4509, | |
| "eval_rougeL": 19.7437, | |
| "eval_rougeLsum": 20.923, | |
| "eval_runtime": 1810.7532, | |
| "eval_samples_per_second": 0.187, | |
| "eval_steps_per_second": 0.024, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 0.001, | |
| "loss": 1.2028, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 0.001, | |
| "loss": 1.0981, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 0.001, | |
| "loss": 1.1033, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "learning_rate": 0.001, | |
| "loss": 1.1303, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 0.001, | |
| "loss": 1.1675, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 0.001, | |
| "loss": 1.3701, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 0.001, | |
| "loss": 1.277, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "eval_gen_len": 507.2278106508876, | |
| "eval_loss": 1.5462373495101929, | |
| "eval_rouge1": 14.6326, | |
| "eval_rouge2": 3.6509, | |
| "eval_rougeL": 12.4805, | |
| "eval_rougeLsum": 13.5001, | |
| "eval_runtime": 1806.4311, | |
| "eval_samples_per_second": 0.187, | |
| "eval_steps_per_second": 0.024, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 0.001, | |
| "loss": 1.3884, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 5.15, | |
| "learning_rate": 0.001, | |
| "loss": 1.0428, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "learning_rate": 0.001, | |
| "loss": 1.0266, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 0.001, | |
| "loss": 1.0247, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "learning_rate": 0.001, | |
| "loss": 0.9732, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 0.001, | |
| "loss": 1.0042, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "learning_rate": 0.001, | |
| "loss": 1.0099, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "learning_rate": 0.001, | |
| "loss": 1.0071, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "eval_gen_len": 429.7721893491124, | |
| "eval_loss": 1.3604055643081665, | |
| "eval_rouge1": 29.5352, | |
| "eval_rouge2": 9.9544, | |
| "eval_rougeL": 22.1073, | |
| "eval_rougeLsum": 28.1204, | |
| "eval_runtime": 1808.4033, | |
| "eval_samples_per_second": 0.187, | |
| "eval_steps_per_second": 0.024, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 0.001, | |
| "loss": 0.8375, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 6.26, | |
| "learning_rate": 0.001, | |
| "loss": 0.8301, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.001, | |
| "loss": 0.8551, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "learning_rate": 0.001, | |
| "loss": 0.823, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "learning_rate": 0.001, | |
| "loss": 0.8783, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 6.82, | |
| "learning_rate": 0.001, | |
| "loss": 0.885, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "learning_rate": 0.001, | |
| "loss": 0.8685, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "eval_gen_len": 451.78402366863907, | |
| "eval_loss": 1.4360722303390503, | |
| "eval_rouge1": 31.0337, | |
| "eval_rouge2": 10.6724, | |
| "eval_rougeL": 22.3815, | |
| "eval_rougeLsum": 29.6325, | |
| "eval_runtime": 1808.8854, | |
| "eval_samples_per_second": 0.187, | |
| "eval_steps_per_second": 0.024, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 0.001, | |
| "loss": 0.7653, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 7.23, | |
| "learning_rate": 0.001, | |
| "loss": 0.7402, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 0.001, | |
| "loss": 0.7582, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 7.51, | |
| "learning_rate": 0.001, | |
| "loss": 0.7518, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "learning_rate": 0.001, | |
| "loss": 0.7486, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "learning_rate": 0.001, | |
| "loss": 0.7645, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 7.93, | |
| "learning_rate": 0.001, | |
| "loss": 0.7498, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_gen_len": 473.896449704142, | |
| "eval_loss": 1.530242681503296, | |
| "eval_rouge1": 28.433, | |
| "eval_rouge2": 8.4887, | |
| "eval_rougeL": 21.3588, | |
| "eval_rougeLsum": 26.6817, | |
| "eval_runtime": 1807.993, | |
| "eval_samples_per_second": 0.187, | |
| "eval_steps_per_second": 0.024, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 0.001, | |
| "loss": 0.6877, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 8.21, | |
| "learning_rate": 0.001, | |
| "loss": 0.6278, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 8.35, | |
| "learning_rate": 0.001, | |
| "loss": 0.6602, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 0.001, | |
| "loss": 0.6408, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 8.63, | |
| "learning_rate": 0.001, | |
| "loss": 0.6514, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "learning_rate": 0.001, | |
| "loss": 0.6434, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "learning_rate": 0.001, | |
| "loss": 0.6226, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "eval_gen_len": 358.76627218934914, | |
| "eval_loss": 1.628932237625122, | |
| "eval_rouge1": 37.251, | |
| "eval_rouge2": 12.8214, | |
| "eval_rougeL": 24.8704, | |
| "eval_rougeLsum": 36.0027, | |
| "eval_runtime": 1807.901, | |
| "eval_samples_per_second": 0.187, | |
| "eval_steps_per_second": 0.024, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 0.001, | |
| "loss": 0.5826, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 9.18, | |
| "learning_rate": 0.001, | |
| "loss": 0.5105, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 9.32, | |
| "learning_rate": 0.001, | |
| "loss": 0.5395, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 9.46, | |
| "learning_rate": 0.001, | |
| "loss": 0.5103, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 0.001, | |
| "loss": 0.5377, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "learning_rate": 0.001, | |
| "loss": 0.5558, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "eval_gen_len": 284.0266272189349, | |
| "eval_loss": 1.5811121463775635, | |
| "eval_rouge1": 35.4657, | |
| "eval_rouge2": 12.0036, | |
| "eval_rougeL": 24.7787, | |
| "eval_rougeLsum": 34.3775, | |
| "eval_runtime": 1740.8347, | |
| "eval_samples_per_second": 0.194, | |
| "eval_steps_per_second": 0.025, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "step": 140, | |
| "total_flos": 2.447850236380324e+18, | |
| "train_loss": 1.2823251613548823, | |
| "train_runtime": 53785.754, | |
| "train_samples_per_second": 0.683, | |
| "train_steps_per_second": 0.003 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 140, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 2.447850236380324e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |