| { | |
| "best_metric": 2.8276827335357666, | |
| "best_model_checkpoint": "/exports/eddie/scratch/s1970716/models/summarization/longt5_xl_sfd_bp_40/checkpoint-71", | |
| "epoch": 19.47826086956522, | |
| "eval_steps": 500, | |
| "global_step": 280, | |
| "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": 0.0873, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.001, | |
| "loss": 0.1042, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.001, | |
| "loss": 0.1006, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.001, | |
| "loss": 0.1057, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.001, | |
| "loss": 0.1157, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 0.001, | |
| "loss": 0.1099, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.001, | |
| "loss": 0.1033, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_gen_len": 279.7507418397626, | |
| "eval_loss": 3.109584331512451, | |
| "eval_rouge1": 40.3355, | |
| "eval_rouge2": 16.0557, | |
| "eval_rougeL": 27.4642, | |
| "eval_rougeLsum": 38.6436, | |
| "eval_runtime": 1755.4356, | |
| "eval_samples_per_second": 0.192, | |
| "eval_steps_per_second": 0.024, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.001, | |
| "loss": 0.0862, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.001, | |
| "loss": 0.086, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 0.001, | |
| "loss": 0.0666, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.001, | |
| "loss": 0.0732, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.001, | |
| "loss": 0.0981, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 0.001, | |
| "loss": 0.0995, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 0.001, | |
| "loss": 0.0836, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "eval_gen_len": 368.4658753709199, | |
| "eval_loss": 3.036095142364502, | |
| "eval_rouge1": 38.2411, | |
| "eval_rouge2": 16.5448, | |
| "eval_rougeL": 26.6409, | |
| "eval_rougeLsum": 36.5841, | |
| "eval_runtime": 1798.7338, | |
| "eval_samples_per_second": 0.187, | |
| "eval_steps_per_second": 0.024, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 0.001, | |
| "loss": 0.091, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 0.001, | |
| "loss": 0.072, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 0.001, | |
| "loss": 0.0779, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 0.001, | |
| "loss": 0.0598, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 0.001, | |
| "loss": 0.0624, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 0.001, | |
| "loss": 0.0823, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 0.001, | |
| "loss": 0.0717, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_gen_len": 489.2047477744807, | |
| "eval_loss": 2.9388694763183594, | |
| "eval_rouge1": 32.0114, | |
| "eval_rouge2": 13.7953, | |
| "eval_rougeL": 22.278, | |
| "eval_rougeLsum": 30.726, | |
| "eval_runtime": 1801.0686, | |
| "eval_samples_per_second": 0.187, | |
| "eval_steps_per_second": 0.024, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 0.001, | |
| "loss": 0.0677, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.001, | |
| "loss": 0.0665, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 0.001, | |
| "loss": 0.0732, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 0.001, | |
| "loss": 0.0573, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 0.001, | |
| "loss": 0.0537, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 0.001, | |
| "loss": 0.0787, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 0.001, | |
| "loss": 0.0614, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "eval_gen_len": 477.7240356083086, | |
| "eval_loss": 3.0221002101898193, | |
| "eval_rouge1": 32.969, | |
| "eval_rouge2": 13.7053, | |
| "eval_rougeL": 22.7428, | |
| "eval_rougeLsum": 31.6951, | |
| "eval_runtime": 1800.6727, | |
| "eval_samples_per_second": 0.187, | |
| "eval_steps_per_second": 0.024, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 0.001, | |
| "loss": 0.0598, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 0.001, | |
| "loss": 0.055, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 0.001, | |
| "loss": 0.0485, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "learning_rate": 0.001, | |
| "loss": 0.0482, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 0.001, | |
| "loss": 0.0586, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 0.001, | |
| "loss": 0.0612, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 0.001, | |
| "loss": 0.1275, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "eval_gen_len": 294.08902077151333, | |
| "eval_loss": 2.8276827335357666, | |
| "eval_rouge1": 40.6965, | |
| "eval_rouge2": 17.2793, | |
| "eval_rougeL": 27.8429, | |
| "eval_rougeLsum": 39.0726, | |
| "eval_runtime": 1758.653, | |
| "eval_samples_per_second": 0.192, | |
| "eval_steps_per_second": 0.024, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 0.001, | |
| "loss": 0.0615, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 5.15, | |
| "learning_rate": 0.001, | |
| "loss": 0.0422, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "learning_rate": 0.001, | |
| "loss": 0.04, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 0.001, | |
| "loss": 0.0439, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "learning_rate": 0.001, | |
| "loss": 0.0797, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 0.001, | |
| "loss": 0.1061, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "learning_rate": 0.001, | |
| "loss": 0.0778, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "learning_rate": 0.001, | |
| "loss": 0.0511, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "eval_gen_len": 476.84569732937683, | |
| "eval_loss": 3.043327808380127, | |
| "eval_rouge1": 33.6479, | |
| "eval_rouge2": 15.0729, | |
| "eval_rougeL": 23.5443, | |
| "eval_rougeLsum": 32.3304, | |
| "eval_runtime": 1800.1403, | |
| "eval_samples_per_second": 0.187, | |
| "eval_steps_per_second": 0.024, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 0.001, | |
| "loss": 0.0429, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 6.26, | |
| "learning_rate": 0.001, | |
| "loss": 0.0405, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.001, | |
| "loss": 0.0431, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "learning_rate": 0.001, | |
| "loss": 0.0481, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "learning_rate": 0.001, | |
| "loss": 0.0598, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 6.82, | |
| "learning_rate": 0.001, | |
| "loss": 0.0788, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "learning_rate": 0.001, | |
| "loss": 0.0666, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "eval_gen_len": 390.43620178041544, | |
| "eval_loss": 3.1150412559509277, | |
| "eval_rouge1": 37.743, | |
| "eval_rouge2": 16.2368, | |
| "eval_rougeL": 26.2524, | |
| "eval_rougeLsum": 36.1313, | |
| "eval_runtime": 1800.3794, | |
| "eval_samples_per_second": 0.187, | |
| "eval_steps_per_second": 0.024, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 0.001, | |
| "loss": 0.0532, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 7.23, | |
| "learning_rate": 0.001, | |
| "loss": 0.0591, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 0.001, | |
| "loss": 0.0325, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 7.51, | |
| "learning_rate": 0.001, | |
| "loss": 0.0362, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "learning_rate": 0.001, | |
| "loss": 0.0424, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "learning_rate": 0.001, | |
| "loss": 0.0405, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 7.93, | |
| "learning_rate": 0.001, | |
| "loss": 0.0398, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_gen_len": 203.40059347181008, | |
| "eval_loss": 3.2224972248077393, | |
| "eval_rouge1": 41.3177, | |
| "eval_rouge2": 16.6663, | |
| "eval_rougeL": 28.7806, | |
| "eval_rougeLsum": 39.5914, | |
| "eval_runtime": 1514.3306, | |
| "eval_samples_per_second": 0.223, | |
| "eval_steps_per_second": 0.028, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 0.001, | |
| "loss": 0.0391, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 8.21, | |
| "learning_rate": 0.001, | |
| "loss": 0.0372, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 8.35, | |
| "learning_rate": 0.001, | |
| "loss": 0.0317, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 0.001, | |
| "loss": 0.0333, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 8.63, | |
| "learning_rate": 0.001, | |
| "loss": 0.04, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "learning_rate": 0.001, | |
| "loss": 0.1656, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "learning_rate": 0.001, | |
| "loss": 0.0396, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "eval_gen_len": 123.83086053412462, | |
| "eval_loss": 3.1462180614471436, | |
| "eval_rouge1": 39.9605, | |
| "eval_rouge2": 16.6732, | |
| "eval_rougeL": 28.3459, | |
| "eval_rougeLsum": 38.226, | |
| "eval_runtime": 1008.3447, | |
| "eval_samples_per_second": 0.334, | |
| "eval_steps_per_second": 0.043, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 0.001, | |
| "loss": 0.0464, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 9.18, | |
| "learning_rate": 0.001, | |
| "loss": 0.066, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 9.32, | |
| "learning_rate": 0.001, | |
| "loss": 0.0693, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 9.46, | |
| "learning_rate": 0.001, | |
| "loss": 0.038, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 0.001, | |
| "loss": 0.0311, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "learning_rate": 0.001, | |
| "loss": 0.0331, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "learning_rate": 0.001, | |
| "loss": 0.0466, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "eval_gen_len": 298.55786350148367, | |
| "eval_loss": 3.2544760704040527, | |
| "eval_rouge1": 40.7977, | |
| "eval_rouge2": 16.9616, | |
| "eval_rougeL": 27.427, | |
| "eval_rougeLsum": 38.8973, | |
| "eval_runtime": 1774.1286, | |
| "eval_samples_per_second": 0.19, | |
| "eval_steps_per_second": 0.024, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "learning_rate": 0.001, | |
| "loss": 0.0309, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 10.16, | |
| "learning_rate": 0.001, | |
| "loss": 0.0311, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 10.3, | |
| "learning_rate": 0.001, | |
| "loss": 0.0305, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 10.43, | |
| "learning_rate": 0.001, | |
| "loss": 0.0323, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 10.57, | |
| "learning_rate": 0.001, | |
| "loss": 0.034, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 10.71, | |
| "learning_rate": 0.001, | |
| "loss": 0.0344, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 10.85, | |
| "learning_rate": 0.001, | |
| "loss": 0.0347, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 10.99, | |
| "learning_rate": 0.001, | |
| "loss": 0.043, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 10.99, | |
| "eval_gen_len": 425.13946587537095, | |
| "eval_loss": 3.3188469409942627, | |
| "eval_rouge1": 36.6349, | |
| "eval_rouge2": 16.1781, | |
| "eval_rougeL": 25.1327, | |
| "eval_rougeLsum": 35.1793, | |
| "eval_runtime": 1793.8451, | |
| "eval_samples_per_second": 0.188, | |
| "eval_steps_per_second": 0.024, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 11.13, | |
| "learning_rate": 0.001, | |
| "loss": 0.0421, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 11.27, | |
| "learning_rate": 0.001, | |
| "loss": 0.03, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 11.41, | |
| "learning_rate": 0.001, | |
| "loss": 0.0294, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 11.55, | |
| "learning_rate": 0.001, | |
| "loss": 0.0259, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 11.69, | |
| "learning_rate": 0.001, | |
| "loss": 0.0306, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 11.83, | |
| "learning_rate": 0.001, | |
| "loss": 0.0644, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 11.97, | |
| "learning_rate": 0.001, | |
| "loss": 0.0538, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 11.97, | |
| "eval_gen_len": 394.86053412462905, | |
| "eval_loss": 2.8277456760406494, | |
| "eval_rouge1": 36.7878, | |
| "eval_rouge2": 15.1186, | |
| "eval_rougeL": 24.9774, | |
| "eval_rougeLsum": 35.275, | |
| "eval_runtime": 1794.0118, | |
| "eval_samples_per_second": 0.188, | |
| "eval_steps_per_second": 0.024, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 12.1, | |
| "learning_rate": 0.001, | |
| "loss": 0.0703, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 12.24, | |
| "learning_rate": 0.001, | |
| "loss": 0.0298, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 12.38, | |
| "learning_rate": 0.001, | |
| "loss": 0.0228, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 12.52, | |
| "learning_rate": 0.001, | |
| "loss": 0.0279, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 12.66, | |
| "learning_rate": 0.001, | |
| "loss": 0.0254, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "learning_rate": 0.001, | |
| "loss": 0.0281, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 12.94, | |
| "learning_rate": 0.001, | |
| "loss": 0.028, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 12.94, | |
| "eval_gen_len": 188.1780415430267, | |
| "eval_loss": 3.4398186206817627, | |
| "eval_rouge1": 42.9644, | |
| "eval_rouge2": 18.1812, | |
| "eval_rougeL": 29.1539, | |
| "eval_rougeLsum": 41.0465, | |
| "eval_runtime": 1387.7431, | |
| "eval_samples_per_second": 0.243, | |
| "eval_steps_per_second": 0.031, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 13.08, | |
| "learning_rate": 0.001, | |
| "loss": 0.0322, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 13.22, | |
| "learning_rate": 0.001, | |
| "loss": 0.0253, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 13.36, | |
| "learning_rate": 0.001, | |
| "loss": 0.0538, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "learning_rate": 0.001, | |
| "loss": 0.0417, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 13.63, | |
| "learning_rate": 0.001, | |
| "loss": 0.0664, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 13.77, | |
| "learning_rate": 0.001, | |
| "loss": 0.1414, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 13.91, | |
| "learning_rate": 0.001, | |
| "loss": 0.1056, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 13.98, | |
| "eval_gen_len": 261.2967359050445, | |
| "eval_loss": 3.33475661277771, | |
| "eval_rouge1": 41.1626, | |
| "eval_rouge2": 17.1605, | |
| "eval_rougeL": 27.6558, | |
| "eval_rougeLsum": 39.2548, | |
| "eval_runtime": 1747.9005, | |
| "eval_samples_per_second": 0.193, | |
| "eval_steps_per_second": 0.025, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 14.05, | |
| "learning_rate": 0.001, | |
| "loss": 0.0509, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 14.19, | |
| "learning_rate": 0.001, | |
| "loss": 0.0251, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 14.33, | |
| "learning_rate": 0.001, | |
| "loss": 0.0302, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 14.47, | |
| "learning_rate": 0.001, | |
| "loss": 0.0228, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 14.61, | |
| "learning_rate": 0.001, | |
| "loss": 0.0261, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 14.75, | |
| "learning_rate": 0.001, | |
| "loss": 0.0289, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 14.89, | |
| "learning_rate": 0.001, | |
| "loss": 0.0303, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 14.96, | |
| "eval_gen_len": 231.60830860534125, | |
| "eval_loss": 3.0237529277801514, | |
| "eval_rouge1": 42.2372, | |
| "eval_rouge2": 17.7292, | |
| "eval_rougeL": 28.8099, | |
| "eval_rougeLsum": 40.3325, | |
| "eval_runtime": 1583.6175, | |
| "eval_samples_per_second": 0.213, | |
| "eval_steps_per_second": 0.027, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "learning_rate": 0.001, | |
| "loss": 0.0474, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 15.17, | |
| "learning_rate": 0.001, | |
| "loss": 0.0252, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 15.3, | |
| "learning_rate": 0.001, | |
| "loss": 0.019, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 15.44, | |
| "learning_rate": 0.001, | |
| "loss": 0.0222, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 15.58, | |
| "learning_rate": 0.001, | |
| "loss": 0.0241, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 15.72, | |
| "learning_rate": 0.001, | |
| "loss": 0.0236, | |
| "step": 226 | |
| }, | |
| { | |
| "epoch": 15.86, | |
| "learning_rate": 0.001, | |
| "loss": 0.0216, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 0.001, | |
| "loss": 0.0234, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_gen_len": 306.16023738872406, | |
| "eval_loss": 3.34845232963562, | |
| "eval_rouge1": 41.714, | |
| "eval_rouge2": 17.7161, | |
| "eval_rougeL": 27.9345, | |
| "eval_rougeLsum": 39.8519, | |
| "eval_runtime": 1759.4496, | |
| "eval_samples_per_second": 0.192, | |
| "eval_steps_per_second": 0.024, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 16.14, | |
| "learning_rate": 0.001, | |
| "loss": 0.021, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 16.28, | |
| "learning_rate": 0.001, | |
| "loss": 0.0232, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 16.42, | |
| "learning_rate": 0.001, | |
| "loss": 0.0223, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 16.56, | |
| "learning_rate": 0.001, | |
| "loss": 0.0286, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 16.7, | |
| "learning_rate": 0.001, | |
| "loss": 0.024, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 16.83, | |
| "learning_rate": 0.001, | |
| "loss": 0.0326, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 16.97, | |
| "learning_rate": 0.001, | |
| "loss": 0.0263, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 16.97, | |
| "eval_gen_len": 210.71216617210683, | |
| "eval_loss": 3.2419376373291016, | |
| "eval_rouge1": 42.0014, | |
| "eval_rouge2": 17.2719, | |
| "eval_rougeL": 28.499, | |
| "eval_rougeLsum": 40.2024, | |
| "eval_runtime": 1510.2771, | |
| "eval_samples_per_second": 0.223, | |
| "eval_steps_per_second": 0.028, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 17.11, | |
| "learning_rate": 0.001, | |
| "loss": 0.0238, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 17.25, | |
| "learning_rate": 0.001, | |
| "loss": 0.0218, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 17.39, | |
| "learning_rate": 0.001, | |
| "loss": 0.0249, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 17.53, | |
| "learning_rate": 0.001, | |
| "loss": 0.0257, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 17.67, | |
| "learning_rate": 0.001, | |
| "loss": 0.036, | |
| "step": 254 | |
| }, | |
| { | |
| "epoch": 17.81, | |
| "learning_rate": 0.001, | |
| "loss": 0.0328, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 17.95, | |
| "learning_rate": 0.001, | |
| "loss": 0.0225, | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 17.95, | |
| "eval_gen_len": 248.5786350148368, | |
| "eval_loss": 3.3452718257904053, | |
| "eval_rouge1": 41.7766, | |
| "eval_rouge2": 17.7154, | |
| "eval_rougeL": 28.4692, | |
| "eval_rougeLsum": 39.9749, | |
| "eval_runtime": 1646.4459, | |
| "eval_samples_per_second": 0.205, | |
| "eval_steps_per_second": 0.026, | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 18.09, | |
| "learning_rate": 0.001, | |
| "loss": 0.0225, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 18.23, | |
| "learning_rate": 0.001, | |
| "loss": 0.0211, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 18.37, | |
| "learning_rate": 0.001, | |
| "loss": 0.0201, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 18.5, | |
| "learning_rate": 0.001, | |
| "loss": 0.0235, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 18.64, | |
| "learning_rate": 0.001, | |
| "loss": 0.0209, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 18.78, | |
| "learning_rate": 0.001, | |
| "loss": 0.0211, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 18.92, | |
| "learning_rate": 0.001, | |
| "loss": 0.0225, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "eval_gen_len": 248.6379821958457, | |
| "eval_loss": 3.4440982341766357, | |
| "eval_rouge1": 42.1727, | |
| "eval_rouge2": 17.598, | |
| "eval_rougeL": 28.5122, | |
| "eval_rougeLsum": 40.4005, | |
| "eval_runtime": 1700.2721, | |
| "eval_samples_per_second": 0.198, | |
| "eval_steps_per_second": 0.025, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 19.06, | |
| "learning_rate": 0.001, | |
| "loss": 0.0213, | |
| "step": 274 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "learning_rate": 0.001, | |
| "loss": 0.0221, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 19.34, | |
| "learning_rate": 0.001, | |
| "loss": 0.0189, | |
| "step": 278 | |
| }, | |
| { | |
| "epoch": 19.48, | |
| "learning_rate": 0.001, | |
| "loss": 0.0211, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 19.48, | |
| "eval_gen_len": 200.37982195845697, | |
| "eval_loss": 3.321052074432373, | |
| "eval_rouge1": 42.5239, | |
| "eval_rouge2": 17.4102, | |
| "eval_rougeL": 28.6868, | |
| "eval_rougeLsum": 40.6537, | |
| "eval_runtime": 1595.3839, | |
| "eval_samples_per_second": 0.211, | |
| "eval_steps_per_second": 0.027, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 19.48, | |
| "step": 280, | |
| "total_flos": 4.895224157149471e+18, | |
| "train_loss": 0.049460863295410364, | |
| "train_runtime": 102546.6903, | |
| "train_samples_per_second": 0.716, | |
| "train_steps_per_second": 0.003 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 280, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 4.895224157149471e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |