| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 29.986564299424185, | |
| "eval_steps": 500, | |
| "global_step": 11700, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_f1": 0.9034, | |
| "eval_gen_len": 26.296727272727274, | |
| "eval_loss": 1.8257849216461182, | |
| "eval_precision": 0.9049, | |
| "eval_recall": 0.9023, | |
| "eval_rouge1": 0.4338, | |
| "eval_rouge2": 0.1906, | |
| "eval_rougeL": 0.3496, | |
| "eval_rougeLsum": 0.3498, | |
| "eval_runtime": 513.5292, | |
| "eval_samples_per_second": 5.355, | |
| "eval_steps_per_second": 0.335, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1.3589743589743592e-05, | |
| "loss": 2.1621, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_f1": 0.9054, | |
| "eval_gen_len": 26.272727272727273, | |
| "eval_loss": 1.7537195682525635, | |
| "eval_precision": 0.9068, | |
| "eval_recall": 0.9044, | |
| "eval_rouge1": 0.4449, | |
| "eval_rouge2": 0.2005, | |
| "eval_rougeL": 0.3633, | |
| "eval_rougeLsum": 0.3633, | |
| "eval_runtime": 505.2109, | |
| "eval_samples_per_second": 5.443, | |
| "eval_steps_per_second": 0.34, | |
| "step": 781 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 7.17948717948718e-06, | |
| "loss": 1.8794, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_f1": 0.9066, | |
| "eval_gen_len": 26.434545454545454, | |
| "eval_loss": 1.726783275604248, | |
| "eval_precision": 0.9078, | |
| "eval_recall": 0.9058, | |
| "eval_rouge1": 0.4518, | |
| "eval_rouge2": 0.2061, | |
| "eval_rougeL": 0.3696, | |
| "eval_rougeLsum": 0.3695, | |
| "eval_runtime": 507.6007, | |
| "eval_samples_per_second": 5.418, | |
| "eval_steps_per_second": 0.339, | |
| "step": 1172 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 7.692307692307694e-07, | |
| "loss": 1.8271, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_f1": 0.9069, | |
| "eval_gen_len": 26.39709090909091, | |
| "eval_loss": 1.7157036066055298, | |
| "eval_precision": 0.9082, | |
| "eval_recall": 0.906, | |
| "eval_rouge1": 0.4539, | |
| "eval_rouge2": 0.2075, | |
| "eval_rougeL": 0.3716, | |
| "eval_rougeLsum": 0.3714, | |
| "eval_runtime": 505.7397, | |
| "eval_samples_per_second": 5.438, | |
| "eval_steps_per_second": 0.34, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_f1": 0.9074, | |
| "eval_gen_len": 26.301454545454547, | |
| "eval_loss": 1.703265905380249, | |
| "eval_precision": 0.9087, | |
| "eval_recall": 0.9065, | |
| "eval_rouge1": 0.4561, | |
| "eval_rouge2": 0.2098, | |
| "eval_rougeL": 0.3735, | |
| "eval_rougeLsum": 0.3734, | |
| "eval_runtime": 505.7439, | |
| "eval_samples_per_second": 5.438, | |
| "eval_steps_per_second": 0.34, | |
| "step": 1951 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 2.9059829059829063e-06, | |
| "loss": 1.8067, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_f1": 0.9077, | |
| "eval_gen_len": 26.438909090909092, | |
| "eval_loss": 1.6896997690200806, | |
| "eval_precision": 0.9089, | |
| "eval_recall": 0.9069, | |
| "eval_rouge1": 0.4592, | |
| "eval_rouge2": 0.2114, | |
| "eval_rougeL": 0.3762, | |
| "eval_rougeLsum": 0.3759, | |
| "eval_runtime": 506.0324, | |
| "eval_samples_per_second": 5.434, | |
| "eval_steps_per_second": 0.34, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 3.974358974358974e-06, | |
| "loss": 1.7833, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_f1": 0.9079, | |
| "eval_gen_len": 26.374545454545455, | |
| "eval_loss": 1.6819266080856323, | |
| "eval_precision": 0.9092, | |
| "eval_recall": 0.9071, | |
| "eval_rouge1": 0.4598, | |
| "eval_rouge2": 0.2115, | |
| "eval_rougeL": 0.3764, | |
| "eval_rougeLsum": 0.376, | |
| "eval_runtime": 506.9661, | |
| "eval_samples_per_second": 5.424, | |
| "eval_steps_per_second": 0.339, | |
| "step": 2731 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 7.692307692307694e-07, | |
| "loss": 1.7683, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_f1": 0.9083, | |
| "eval_gen_len": 26.620363636363635, | |
| "eval_loss": 1.676251769065857, | |
| "eval_precision": 0.9094, | |
| "eval_recall": 0.9076, | |
| "eval_rouge1": 0.4621, | |
| "eval_rouge2": 0.2133, | |
| "eval_rougeL": 0.3791, | |
| "eval_rougeLsum": 0.3789, | |
| "eval_runtime": 506.5847, | |
| "eval_samples_per_second": 5.429, | |
| "eval_steps_per_second": 0.34, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "learning_rate": 5.042735042735043e-06, | |
| "loss": 1.7559, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_f1": 0.9086, | |
| "eval_gen_len": 26.424, | |
| "eval_loss": 1.666217565536499, | |
| "eval_precision": 0.9098, | |
| "eval_recall": 0.9078, | |
| "eval_rouge1": 0.4632, | |
| "eval_rouge2": 0.215, | |
| "eval_rougeL": 0.38, | |
| "eval_rougeLsum": 0.3799, | |
| "eval_runtime": 507.5176, | |
| "eval_samples_per_second": 5.419, | |
| "eval_steps_per_second": 0.339, | |
| "step": 3511 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_f1": 0.9089, | |
| "eval_gen_len": 26.542545454545454, | |
| "eval_loss": 1.6593670845031738, | |
| "eval_precision": 0.9099, | |
| "eval_recall": 0.9082, | |
| "eval_rouge1": 0.4651, | |
| "eval_rouge2": 0.2168, | |
| "eval_rougeL": 0.3812, | |
| "eval_rougeLsum": 0.3812, | |
| "eval_runtime": 501.1134, | |
| "eval_samples_per_second": 5.488, | |
| "eval_steps_per_second": 0.343, | |
| "step": 3902 | |
| }, | |
| { | |
| "epoch": 10.25, | |
| "learning_rate": 2.9059829059829063e-06, | |
| "loss": 1.7357, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_f1": 0.9091, | |
| "eval_gen_len": 26.605090909090908, | |
| "eval_loss": 1.6554700136184692, | |
| "eval_precision": 0.91, | |
| "eval_recall": 0.9086, | |
| "eval_rouge1": 0.4663, | |
| "eval_rouge2": 0.2178, | |
| "eval_rougeL": 0.3824, | |
| "eval_rougeLsum": 0.3823, | |
| "eval_runtime": 504.1076, | |
| "eval_samples_per_second": 5.455, | |
| "eval_steps_per_second": 0.341, | |
| "step": 4293 | |
| }, | |
| { | |
| "epoch": 11.53, | |
| "learning_rate": 7.692307692307694e-07, | |
| "loss": 1.7297, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_f1": 0.9092, | |
| "eval_gen_len": 26.439272727272726, | |
| "eval_loss": 1.6508440971374512, | |
| "eval_precision": 0.9103, | |
| "eval_recall": 0.9084, | |
| "eval_rouge1": 0.4668, | |
| "eval_rouge2": 0.2175, | |
| "eval_rougeL": 0.3823, | |
| "eval_rougeLsum": 0.3822, | |
| "eval_runtime": 506.6299, | |
| "eval_samples_per_second": 5.428, | |
| "eval_steps_per_second": 0.339, | |
| "step": 4680 | |
| }, | |
| { | |
| "epoch": 12.82, | |
| "learning_rate": 3.974358974358974e-06, | |
| "loss": 1.7165, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_f1": 0.9094, | |
| "eval_gen_len": 26.638545454545454, | |
| "eval_loss": 1.6451035737991333, | |
| "eval_precision": 0.9103, | |
| "eval_recall": 0.9089, | |
| "eval_rouge1": 0.4687, | |
| "eval_rouge2": 0.2191, | |
| "eval_rougeL": 0.3834, | |
| "eval_rougeLsum": 0.3834, | |
| "eval_runtime": 509.0115, | |
| "eval_samples_per_second": 5.403, | |
| "eval_steps_per_second": 0.338, | |
| "step": 5071 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_f1": 0.9095, | |
| "eval_gen_len": 26.415636363636363, | |
| "eval_loss": 1.6405242681503296, | |
| "eval_precision": 0.9106, | |
| "eval_recall": 0.9087, | |
| "eval_rouge1": 0.4691, | |
| "eval_rouge2": 0.2193, | |
| "eval_rougeL": 0.3845, | |
| "eval_rougeLsum": 0.3844, | |
| "eval_runtime": 502.7756, | |
| "eval_samples_per_second": 5.47, | |
| "eval_steps_per_second": 0.342, | |
| "step": 5462 | |
| }, | |
| { | |
| "epoch": 14.1, | |
| "learning_rate": 2.371794871794872e-06, | |
| "loss": 1.7068, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_f1": 0.9097, | |
| "eval_gen_len": 26.45709090909091, | |
| "eval_loss": 1.6383482217788696, | |
| "eval_precision": 0.9108, | |
| "eval_recall": 0.9089, | |
| "eval_rouge1": 0.4699, | |
| "eval_rouge2": 0.2204, | |
| "eval_rougeL": 0.3853, | |
| "eval_rougeLsum": 0.3853, | |
| "eval_runtime": 506.608, | |
| "eval_samples_per_second": 5.428, | |
| "eval_steps_per_second": 0.34, | |
| "step": 5853 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 7.692307692307694e-07, | |
| "loss": 1.7004, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_f1": 0.9097, | |
| "eval_gen_len": 26.424727272727274, | |
| "eval_loss": 1.6346381902694702, | |
| "eval_precision": 0.9108, | |
| "eval_recall": 0.9089, | |
| "eval_rouge1": 0.4703, | |
| "eval_rouge2": 0.2204, | |
| "eval_rougeL": 0.385, | |
| "eval_rougeLsum": 0.385, | |
| "eval_runtime": 506.5023, | |
| "eval_samples_per_second": 5.429, | |
| "eval_steps_per_second": 0.34, | |
| "step": 6240 | |
| }, | |
| { | |
| "epoch": 16.66, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 1.6923, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_f1": 0.9099, | |
| "eval_gen_len": 26.443636363636365, | |
| "eval_loss": 1.6305155754089355, | |
| "eval_precision": 0.911, | |
| "eval_recall": 0.9091, | |
| "eval_rouge1": 0.4706, | |
| "eval_rouge2": 0.221, | |
| "eval_rougeL": 0.3855, | |
| "eval_rougeLsum": 0.3856, | |
| "eval_runtime": 506.0984, | |
| "eval_samples_per_second": 5.434, | |
| "eval_steps_per_second": 0.34, | |
| "step": 6631 | |
| }, | |
| { | |
| "epoch": 17.94, | |
| "learning_rate": 2.0512820512820513e-06, | |
| "loss": 1.6839, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_f1": 0.9098, | |
| "eval_gen_len": 26.612, | |
| "eval_loss": 1.6285477876663208, | |
| "eval_precision": 0.9106, | |
| "eval_recall": 0.9094, | |
| "eval_rouge1": 0.4712, | |
| "eval_rouge2": 0.2215, | |
| "eval_rougeL": 0.3862, | |
| "eval_rougeLsum": 0.3864, | |
| "eval_runtime": 509.5093, | |
| "eval_samples_per_second": 5.397, | |
| "eval_steps_per_second": 0.338, | |
| "step": 7022 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_f1": 0.9099, | |
| "eval_gen_len": 26.529090909090908, | |
| "eval_loss": 1.6263303756713867, | |
| "eval_precision": 0.9108, | |
| "eval_recall": 0.9093, | |
| "eval_rouge1": 0.4709, | |
| "eval_rouge2": 0.2217, | |
| "eval_rougeL": 0.3862, | |
| "eval_rougeLsum": 0.3864, | |
| "eval_runtime": 510.6771, | |
| "eval_samples_per_second": 5.385, | |
| "eval_steps_per_second": 0.337, | |
| "step": 7413 | |
| }, | |
| { | |
| "epoch": 19.22, | |
| "learning_rate": 7.692307692307694e-07, | |
| "loss": 1.6743, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_f1": 0.91, | |
| "eval_gen_len": 26.42509090909091, | |
| "eval_loss": 1.620501160621643, | |
| "eval_precision": 0.9111, | |
| "eval_recall": 0.9092, | |
| "eval_rouge1": 0.4727, | |
| "eval_rouge2": 0.2223, | |
| "eval_rougeL": 0.3876, | |
| "eval_rougeLsum": 0.3876, | |
| "eval_runtime": 508.7027, | |
| "eval_samples_per_second": 5.406, | |
| "eval_steps_per_second": 0.338, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 20.51, | |
| "learning_rate": 6.324786324786325e-06, | |
| "loss": 1.6692, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_f1": 0.9102, | |
| "eval_gen_len": 26.748363636363635, | |
| "eval_loss": 1.6152820587158203, | |
| "eval_precision": 0.911, | |
| "eval_recall": 0.9098, | |
| "eval_rouge1": 0.4737, | |
| "eval_rouge2": 0.2229, | |
| "eval_rougeL": 0.388, | |
| "eval_rougeLsum": 0.388, | |
| "eval_runtime": 510.0415, | |
| "eval_samples_per_second": 5.392, | |
| "eval_steps_per_second": 0.337, | |
| "step": 8191 | |
| }, | |
| { | |
| "epoch": 21.79, | |
| "learning_rate": 5.470085470085471e-06, | |
| "loss": 1.6568, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_f1": 0.9103, | |
| "eval_gen_len": 26.532, | |
| "eval_loss": 1.6104176044464111, | |
| "eval_precision": 0.9113, | |
| "eval_recall": 0.9096, | |
| "eval_rouge1": 0.4733, | |
| "eval_rouge2": 0.2221, | |
| "eval_rougeL": 0.3885, | |
| "eval_rougeLsum": 0.3886, | |
| "eval_runtime": 508.5365, | |
| "eval_samples_per_second": 5.408, | |
| "eval_steps_per_second": 0.338, | |
| "step": 8582 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_f1": 0.9104, | |
| "eval_gen_len": 26.82, | |
| "eval_loss": 1.6056216955184937, | |
| "eval_precision": 0.911, | |
| "eval_recall": 0.9101, | |
| "eval_rouge1": 0.4756, | |
| "eval_rouge2": 0.2236, | |
| "eval_rougeL": 0.3891, | |
| "eval_rougeLsum": 0.3891, | |
| "eval_runtime": 510.3093, | |
| "eval_samples_per_second": 5.389, | |
| "eval_steps_per_second": 0.337, | |
| "step": 8973 | |
| }, | |
| { | |
| "epoch": 23.07, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 1.6418, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_f1": 0.9106, | |
| "eval_gen_len": 26.55127272727273, | |
| "eval_loss": 1.6021137237548828, | |
| "eval_precision": 0.9115, | |
| "eval_recall": 0.91, | |
| "eval_rouge1": 0.476, | |
| "eval_rouge2": 0.2246, | |
| "eval_rougeL": 0.3903, | |
| "eval_rougeLsum": 0.3903, | |
| "eval_runtime": 506.3054, | |
| "eval_samples_per_second": 5.432, | |
| "eval_steps_per_second": 0.34, | |
| "step": 9360 | |
| }, | |
| { | |
| "epoch": 24.36, | |
| "learning_rate": 3.760683760683761e-06, | |
| "loss": 1.6319, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_f1": 0.9105, | |
| "eval_gen_len": 26.437454545454546, | |
| "eval_loss": 1.5995395183563232, | |
| "eval_precision": 0.9116, | |
| "eval_recall": 0.9098, | |
| "eval_rouge1": 0.4751, | |
| "eval_rouge2": 0.2245, | |
| "eval_rougeL": 0.3905, | |
| "eval_rougeLsum": 0.3905, | |
| "eval_runtime": 501.0862, | |
| "eval_samples_per_second": 5.488, | |
| "eval_steps_per_second": 0.343, | |
| "step": 9751 | |
| }, | |
| { | |
| "epoch": 25.64, | |
| "learning_rate": 2.9059829059829063e-06, | |
| "loss": 1.624, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_f1": 0.9107, | |
| "eval_gen_len": 26.60181818181818, | |
| "eval_loss": 1.597448468208313, | |
| "eval_precision": 0.9116, | |
| "eval_recall": 0.9101, | |
| "eval_rouge1": 0.4756, | |
| "eval_rouge2": 0.2247, | |
| "eval_rougeL": 0.3903, | |
| "eval_rougeLsum": 0.3904, | |
| "eval_runtime": 500.0077, | |
| "eval_samples_per_second": 5.5, | |
| "eval_steps_per_second": 0.344, | |
| "step": 10142 | |
| }, | |
| { | |
| "epoch": 26.92, | |
| "learning_rate": 2.0512820512820513e-06, | |
| "loss": 1.6184, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_f1": 0.9106, | |
| "eval_gen_len": 26.483272727272727, | |
| "eval_loss": 1.5952799320220947, | |
| "eval_precision": 0.9116, | |
| "eval_recall": 0.9099, | |
| "eval_rouge1": 0.4747, | |
| "eval_rouge2": 0.2231, | |
| "eval_rougeL": 0.3899, | |
| "eval_rougeLsum": 0.3899, | |
| "eval_runtime": 505.7256, | |
| "eval_samples_per_second": 5.438, | |
| "eval_steps_per_second": 0.34, | |
| "step": 10533 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_f1": 0.9107, | |
| "eval_gen_len": 26.560363636363636, | |
| "eval_loss": 1.594251036643982, | |
| "eval_precision": 0.9116, | |
| "eval_recall": 0.9102, | |
| "eval_rouge1": 0.4758, | |
| "eval_rouge2": 0.2243, | |
| "eval_rougeL": 0.3907, | |
| "eval_rougeLsum": 0.3908, | |
| "eval_runtime": 503.0877, | |
| "eval_samples_per_second": 5.466, | |
| "eval_steps_per_second": 0.342, | |
| "step": 10923 | |
| }, | |
| { | |
| "epoch": 28.2, | |
| "learning_rate": 1.1965811965811968e-06, | |
| "loss": 1.6126, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_f1": 0.9108, | |
| "eval_gen_len": 26.577454545454547, | |
| "eval_loss": 1.5936089754104614, | |
| "eval_precision": 0.9117, | |
| "eval_recall": 0.9103, | |
| "eval_rouge1": 0.4776, | |
| "eval_rouge2": 0.226, | |
| "eval_rougeL": 0.3926, | |
| "eval_rougeLsum": 0.3926, | |
| "eval_runtime": 506.8028, | |
| "eval_samples_per_second": 5.426, | |
| "eval_steps_per_second": 0.339, | |
| "step": 11314 | |
| }, | |
| { | |
| "epoch": 29.47, | |
| "learning_rate": 3.4188034188034194e-07, | |
| "loss": 1.6148, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 29.99, | |
| "eval_f1": 0.9108, | |
| "eval_gen_len": 26.555636363636363, | |
| "eval_loss": 1.5934168100357056, | |
| "eval_precision": 0.9117, | |
| "eval_recall": 0.9103, | |
| "eval_rouge1": 0.4774, | |
| "eval_rouge2": 0.2259, | |
| "eval_rougeL": 0.3926, | |
| "eval_rougeLsum": 0.3926, | |
| "eval_runtime": 505.9746, | |
| "eval_samples_per_second": 5.435, | |
| "eval_steps_per_second": 0.34, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 29.99, | |
| "step": 11700, | |
| "total_flos": 2.1629605976148541e+18, | |
| "train_loss": 0.37363835000584267, | |
| "train_runtime": 14987.4083, | |
| "train_samples_per_second": 100.084, | |
| "train_steps_per_second": 0.781 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 11700, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "total_flos": 2.1629605976148541e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |