| { | |
| "best_metric": 0.5314118934504415, | |
| "best_model_checkpoint": "./results/checkpoint-15120", | |
| "epoch": 18.0, | |
| "eval_steps": 500, | |
| "global_step": 15120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.9761904761904762e-05, | |
| "loss": 7.0613, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.2605280876159668, | |
| "eval_rouge1": 0.5053086747478719, | |
| "eval_rouge2": 0.24615333793097546, | |
| "eval_rougeL": 0.4512970323707136, | |
| "eval_rougeLsum": 0.4656617846478366, | |
| "eval_runtime": 772.9001, | |
| "eval_samples_per_second": 1.938, | |
| "eval_steps_per_second": 0.122, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 4.9498746867167923e-05, | |
| "loss": 1.4331, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 4.793233082706767e-05, | |
| "loss": 1.3665, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.2270965576171875, | |
| "eval_rouge1": 0.5124476660885933, | |
| "eval_rouge2": 0.256176861062982, | |
| "eval_rougeL": 0.4600482066009496, | |
| "eval_rougeLsum": 0.4746519151030747, | |
| "eval_runtime": 1039.1433, | |
| "eval_samples_per_second": 1.442, | |
| "eval_steps_per_second": 0.09, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 4.6365914786967416e-05, | |
| "loss": 1.3381, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 4.4799498746867166e-05, | |
| "loss": 1.3009, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.2114347219467163, | |
| "eval_rouge1": 0.5179075668199837, | |
| "eval_rouge2": 0.2612977945518593, | |
| "eval_rougeL": 0.46524322885156827, | |
| "eval_rougeLsum": 0.4802301912620997, | |
| "eval_runtime": 1018.7441, | |
| "eval_samples_per_second": 1.47, | |
| "eval_steps_per_second": 0.092, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 4.323308270676692e-05, | |
| "loss": 1.2594, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.203908085823059, | |
| "eval_rouge1": 0.5229139335151443, | |
| "eval_rouge2": 0.26558551302892763, | |
| "eval_rougeL": 0.46774305498128443, | |
| "eval_rougeLsum": 0.483920816134904, | |
| "eval_runtime": 926.8973, | |
| "eval_samples_per_second": 1.616, | |
| "eval_steps_per_second": 0.101, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.247, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 4.0100250626566415e-05, | |
| "loss": 1.227, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.1987696886062622, | |
| "eval_rouge1": 0.5247159372903829, | |
| "eval_rouge2": 0.2668774790757894, | |
| "eval_rougeL": 0.4698131046976711, | |
| "eval_rougeLsum": 0.4859852789981728, | |
| "eval_runtime": 1138.215, | |
| "eval_samples_per_second": 1.316, | |
| "eval_steps_per_second": 0.083, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "learning_rate": 3.8533834586466165e-05, | |
| "loss": 1.1951, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.95, | |
| "learning_rate": 3.6967418546365914e-05, | |
| "loss": 1.1918, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.1965585947036743, | |
| "eval_rouge1": 0.5238465835630357, | |
| "eval_rouge2": 0.2697694029030245, | |
| "eval_rougeL": 0.4700972375246304, | |
| "eval_rougeLsum": 0.48604440949884575, | |
| "eval_runtime": 1144.1344, | |
| "eval_samples_per_second": 1.309, | |
| "eval_steps_per_second": 0.082, | |
| "step": 5040 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "learning_rate": 3.540100250626567e-05, | |
| "loss": 1.1696, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.193439245223999, | |
| "eval_rouge1": 0.5261420787321425, | |
| "eval_rouge2": 0.2701869159088741, | |
| "eval_rougeL": 0.472288301825879, | |
| "eval_rougeLsum": 0.48796847398006715, | |
| "eval_runtime": 913.4514, | |
| "eval_samples_per_second": 1.64, | |
| "eval_steps_per_second": 0.103, | |
| "step": 5880 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 3.3834586466165414e-05, | |
| "loss": 1.1472, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 3.2268170426065164e-05, | |
| "loss": 1.1285, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.1932892799377441, | |
| "eval_rouge1": 0.5236558096283529, | |
| "eval_rouge2": 0.26931250040124155, | |
| "eval_rougeL": 0.4700827158260693, | |
| "eval_rougeLsum": 0.48489767885463625, | |
| "eval_runtime": 1155.7348, | |
| "eval_samples_per_second": 1.296, | |
| "eval_steps_per_second": 0.081, | |
| "step": 6720 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 3.0701754385964913e-05, | |
| "loss": 1.1234, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "learning_rate": 2.9135338345864667e-05, | |
| "loss": 1.1153, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 1.1960569620132446, | |
| "eval_rouge1": 0.5263030048004651, | |
| "eval_rouge2": 0.27081863504950743, | |
| "eval_rougeL": 0.4724221317085258, | |
| "eval_rougeLsum": 0.48803341513882764, | |
| "eval_runtime": 983.851, | |
| "eval_samples_per_second": 1.523, | |
| "eval_steps_per_second": 0.096, | |
| "step": 7560 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 2.756892230576441e-05, | |
| "loss": 1.0927, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.1961216926574707, | |
| "eval_rouge1": 0.5253667627324876, | |
| "eval_rouge2": 0.2690944896569006, | |
| "eval_rougeL": 0.4720451976404024, | |
| "eval_rougeLsum": 0.48738350139979847, | |
| "eval_runtime": 932.6258, | |
| "eval_samples_per_second": 1.606, | |
| "eval_steps_per_second": 0.101, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 10.12, | |
| "learning_rate": 2.6002506265664163e-05, | |
| "loss": 1.0933, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 10.71, | |
| "learning_rate": 2.443609022556391e-05, | |
| "loss": 1.0661, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 1.2010161876678467, | |
| "eval_rouge1": 0.5234197228179973, | |
| "eval_rouge2": 0.2683998445240433, | |
| "eval_rougeL": 0.4697712822647654, | |
| "eval_rougeLsum": 0.48542105989790263, | |
| "eval_runtime": 1250.3048, | |
| "eval_samples_per_second": 1.198, | |
| "eval_steps_per_second": 0.075, | |
| "step": 9240 | |
| }, | |
| { | |
| "epoch": 11.31, | |
| "learning_rate": 2.2869674185463662e-05, | |
| "loss": 1.0663, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 11.9, | |
| "learning_rate": 2.130325814536341e-05, | |
| "loss": 1.0634, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.2002513408660889, | |
| "eval_rouge1": 0.5259472773406924, | |
| "eval_rouge2": 0.2723400096809272, | |
| "eval_rougeL": 0.47293541812561896, | |
| "eval_rougeLsum": 0.4885362784921822, | |
| "eval_runtime": 1027.3406, | |
| "eval_samples_per_second": 1.458, | |
| "eval_steps_per_second": 0.091, | |
| "step": 10080 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "learning_rate": 1.9736842105263158e-05, | |
| "loss": 1.046, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 1.2019047737121582, | |
| "eval_rouge1": 0.527682777778989, | |
| "eval_rouge2": 0.27261485269714236, | |
| "eval_rougeL": 0.4747461246549739, | |
| "eval_rougeLsum": 0.49067235332780945, | |
| "eval_runtime": 1023.6212, | |
| "eval_samples_per_second": 1.463, | |
| "eval_steps_per_second": 0.092, | |
| "step": 10920 | |
| }, | |
| { | |
| "epoch": 13.1, | |
| "learning_rate": 1.8170426065162908e-05, | |
| "loss": 1.0454, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 13.69, | |
| "learning_rate": 1.6604010025062658e-05, | |
| "loss": 1.0273, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 1.2045047283172607, | |
| "eval_rouge1": 0.5309321460587708, | |
| "eval_rouge2": 0.27493073812627356, | |
| "eval_rougeL": 0.4776371159725116, | |
| "eval_rougeLsum": 0.49401653255083877, | |
| "eval_runtime": 933.4164, | |
| "eval_samples_per_second": 1.605, | |
| "eval_steps_per_second": 0.101, | |
| "step": 11760 | |
| }, | |
| { | |
| "epoch": 14.29, | |
| "learning_rate": 1.5037593984962406e-05, | |
| "loss": 1.0257, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 14.88, | |
| "learning_rate": 1.3471177944862157e-05, | |
| "loss": 1.0218, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 1.207719326019287, | |
| "eval_rouge1": 0.5295059932432493, | |
| "eval_rouge2": 0.2727809059549531, | |
| "eval_rougeL": 0.4770545143695025, | |
| "eval_rougeLsum": 0.49245995561354516, | |
| "eval_runtime": 874.8747, | |
| "eval_samples_per_second": 1.712, | |
| "eval_steps_per_second": 0.107, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 15.48, | |
| "learning_rate": 1.1904761904761905e-05, | |
| "loss": 1.0208, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 1.2094707489013672, | |
| "eval_rouge1": 0.5303496124431821, | |
| "eval_rouge2": 0.27278313644220603, | |
| "eval_rougeL": 0.47754216166274177, | |
| "eval_rougeLsum": 0.49284317669377947, | |
| "eval_runtime": 939.0978, | |
| "eval_samples_per_second": 1.595, | |
| "eval_steps_per_second": 0.1, | |
| "step": 13440 | |
| }, | |
| { | |
| "epoch": 16.07, | |
| "learning_rate": 1.0338345864661655e-05, | |
| "loss": 1.0093, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 1.003, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 1.2109801769256592, | |
| "eval_rouge1": 0.5301463905288382, | |
| "eval_rouge2": 0.2726100969561255, | |
| "eval_rougeL": 0.47718245520328006, | |
| "eval_rougeLsum": 0.49294754454933043, | |
| "eval_runtime": 829.8724, | |
| "eval_samples_per_second": 1.805, | |
| "eval_steps_per_second": 0.113, | |
| "step": 14280 | |
| }, | |
| { | |
| "epoch": 17.26, | |
| "learning_rate": 7.205513784461153e-06, | |
| "loss": 1.0002, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "learning_rate": 5.639097744360902e-06, | |
| "loss": 1.003, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 1.209855556488037, | |
| "eval_rouge1": 0.5314118934504415, | |
| "eval_rouge2": 0.27293095312101445, | |
| "eval_rougeL": 0.47805026323896327, | |
| "eval_rougeLsum": 0.49408516728917673, | |
| "eval_runtime": 882.3325, | |
| "eval_samples_per_second": 1.698, | |
| "eval_steps_per_second": 0.107, | |
| "step": 15120 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 16800, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 3.682281994080768e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |