| { | |
| "best_metric": 2.3254711627960205, | |
| "best_model_checkpoint": "/home/co-ou1/rds/hpc-work/models/longt5_xl_sfd_4096/longt5_xl_sfd_4096_e10/checkpoint-28", | |
| "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": 3.2585, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.001, | |
| "loss": 3.3242, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.001, | |
| "loss": 3.6288, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.001, | |
| "loss": 3.1948, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.001, | |
| "loss": 2.9524, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 0.001, | |
| "loss": 2.8889, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.001, | |
| "loss": 3.0332, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_loss": 2.5423872470855713, | |
| "eval_runtime": 96.2671, | |
| "eval_samples_per_second": 3.511, | |
| "eval_steps_per_second": 0.447, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.001, | |
| "loss": 2.6074, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.001, | |
| "loss": 2.6808, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 0.001, | |
| "loss": 2.6085, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.001, | |
| "loss": 2.5063, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.001, | |
| "loss": 2.4614, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 0.001, | |
| "loss": 2.4519, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 0.001, | |
| "loss": 2.4105, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "eval_loss": 2.3254711627960205, | |
| "eval_runtime": 99.9342, | |
| "eval_samples_per_second": 3.382, | |
| "eval_steps_per_second": 0.43, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 0.001, | |
| "loss": 2.1628, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 0.001, | |
| "loss": 2.0701, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 0.001, | |
| "loss": 2.0992, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 0.001, | |
| "loss": 2.0401, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 0.001, | |
| "loss": 2.0299, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 0.001, | |
| "loss": 2.0812, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 0.001, | |
| "loss": 2.0496, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_loss": 2.3419768810272217, | |
| "eval_runtime": 99.9699, | |
| "eval_samples_per_second": 3.381, | |
| "eval_steps_per_second": 0.43, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 0.001, | |
| "loss": 1.9994, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.001, | |
| "loss": 1.7276, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 0.001, | |
| "loss": 1.7639, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 0.001, | |
| "loss": 1.7624, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 0.001, | |
| "loss": 1.7726, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 0.001, | |
| "loss": 1.7218, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 0.001, | |
| "loss": 1.7473, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "eval_loss": 2.352036476135254, | |
| "eval_runtime": 109.1375, | |
| "eval_samples_per_second": 3.097, | |
| "eval_steps_per_second": 0.394, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 0.001, | |
| "loss": 1.6586, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 0.001, | |
| "loss": 1.3888, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 0.001, | |
| "loss": 1.4192, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "learning_rate": 0.001, | |
| "loss": 1.4003, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 0.001, | |
| "loss": 1.4405, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 0.001, | |
| "loss": 1.3766, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 0.001, | |
| "loss": 1.4007, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "eval_loss": 2.4979982376098633, | |
| "eval_runtime": 101.842, | |
| "eval_samples_per_second": 3.319, | |
| "eval_steps_per_second": 0.422, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 0.001, | |
| "loss": 1.3547, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 5.15, | |
| "learning_rate": 0.001, | |
| "loss": 1.3243, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "learning_rate": 0.001, | |
| "loss": 1.3494, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 0.001, | |
| "loss": 1.3982, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "learning_rate": 0.001, | |
| "loss": 1.3294, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 0.001, | |
| "loss": 1.404, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "learning_rate": 0.001, | |
| "loss": 1.371, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "learning_rate": 0.001, | |
| "loss": 1.3809, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "eval_loss": 2.4785053730010986, | |
| "eval_runtime": 78.2069, | |
| "eval_samples_per_second": 4.322, | |
| "eval_steps_per_second": 0.55, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 0.001, | |
| "loss": 1.0798, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 6.26, | |
| "learning_rate": 0.001, | |
| "loss": 1.0476, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.001, | |
| "loss": 1.111, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "learning_rate": 0.001, | |
| "loss": 1.0734, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "learning_rate": 0.001, | |
| "loss": 1.0563, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 6.82, | |
| "learning_rate": 0.001, | |
| "loss": 1.1215, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "learning_rate": 0.001, | |
| "loss": 1.1153, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "eval_loss": 2.732642650604248, | |
| "eval_runtime": 78.0611, | |
| "eval_samples_per_second": 4.33, | |
| "eval_steps_per_second": 0.551, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 0.001, | |
| "loss": 0.9032, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 7.23, | |
| "learning_rate": 0.001, | |
| "loss": 0.8517, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 0.001, | |
| "loss": 0.8711, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 7.51, | |
| "learning_rate": 0.001, | |
| "loss": 0.8849, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "learning_rate": 0.001, | |
| "loss": 0.898, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "learning_rate": 0.001, | |
| "loss": 0.9153, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 7.93, | |
| "learning_rate": 0.001, | |
| "loss": 0.9129, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 2.9232447147369385, | |
| "eval_runtime": 78.4249, | |
| "eval_samples_per_second": 4.31, | |
| "eval_steps_per_second": 0.548, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 0.001, | |
| "loss": 0.7377, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 8.21, | |
| "learning_rate": 0.001, | |
| "loss": 0.6558, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 8.35, | |
| "learning_rate": 0.001, | |
| "loss": 0.7047, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 0.001, | |
| "loss": 0.7382, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 8.63, | |
| "learning_rate": 0.001, | |
| "loss": 0.6919, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "learning_rate": 0.001, | |
| "loss": 0.7257, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "learning_rate": 0.001, | |
| "loss": 0.7118, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "eval_loss": 3.047579526901245, | |
| "eval_runtime": 78.168, | |
| "eval_samples_per_second": 4.324, | |
| "eval_steps_per_second": 0.55, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 0.001, | |
| "loss": 0.6401, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 9.18, | |
| "learning_rate": 0.001, | |
| "loss": 0.5032, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 9.32, | |
| "learning_rate": 0.001, | |
| "loss": 0.548, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 9.46, | |
| "learning_rate": 0.001, | |
| "loss": 0.5218, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 0.001, | |
| "loss": 0.5744, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "learning_rate": 0.001, | |
| "loss": 0.5883, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "eval_loss": 3.3644142150878906, | |
| "eval_runtime": 78.3345, | |
| "eval_samples_per_second": 4.315, | |
| "eval_steps_per_second": 0.549, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "step": 140, | |
| "total_flos": 2.447850236380324e+18, | |
| "train_loss": 1.5745136559009552, | |
| "train_runtime": 36489.2213, | |
| "train_samples_per_second": 1.007, | |
| "train_steps_per_second": 0.004 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 140, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 2.447850236380324e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |