| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 49.992, | |
| "global_step": 3100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.000967741935483871, | |
| "loss": 5.0178, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 0.0009354838709677419, | |
| "loss": 2.8191, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 0.0009032258064516129, | |
| "loss": 2.0873, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 0.0008709677419354839, | |
| "loss": 1.5077, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 0.0008387096774193549, | |
| "loss": 1.1005, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "learning_rate": 0.0008064516129032258, | |
| "loss": 0.7513, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 11.29, | |
| "learning_rate": 0.0007741935483870968, | |
| "loss": 0.5449, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 0.0007419354838709678, | |
| "loss": 0.415, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 14.51, | |
| "learning_rate": 0.0007096774193548388, | |
| "loss": 0.3257, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "learning_rate": 0.0006774193548387097, | |
| "loss": 0.2728, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 17.74, | |
| "learning_rate": 0.0006451612903225806, | |
| "loss": 0.2365, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "learning_rate": 0.0006129032258064516, | |
| "loss": 0.2159, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 20.96, | |
| "learning_rate": 0.0005806451612903226, | |
| "loss": 0.1881, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 22.58, | |
| "learning_rate": 0.0005483870967741935, | |
| "loss": 0.1687, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 24.19, | |
| "learning_rate": 0.0005161290322580645, | |
| "loss": 0.1611, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 25.8, | |
| "learning_rate": 0.0004838709677419355, | |
| "loss": 0.1469, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 27.42, | |
| "learning_rate": 0.00045161290322580643, | |
| "loss": 0.1405, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 29.03, | |
| "learning_rate": 0.00041935483870967743, | |
| "loss": 0.136, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 30.64, | |
| "learning_rate": 0.0003870967741935484, | |
| "loss": 0.1266, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 32.26, | |
| "learning_rate": 0.0003548387096774194, | |
| "loss": 0.1247, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 32.26, | |
| "eval_bleu": 4.2768, | |
| "eval_gen_len": 18.2295, | |
| "eval_loss": 0.07783249020576477, | |
| "eval_runtime": 151.6677, | |
| "eval_samples_per_second": 13.187, | |
| "eval_steps_per_second": 3.297, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 33.86, | |
| "learning_rate": 0.0003225806451612903, | |
| "loss": 0.1175, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 35.48, | |
| "learning_rate": 0.0002903225806451613, | |
| "loss": 0.1128, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 37.1, | |
| "learning_rate": 0.00025806451612903227, | |
| "loss": 0.1127, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 38.7, | |
| "learning_rate": 0.00022580645161290321, | |
| "loss": 0.106, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 40.32, | |
| "learning_rate": 0.0001935483870967742, | |
| "loss": 0.1062, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 41.93, | |
| "learning_rate": 0.00016129032258064516, | |
| "loss": 0.1026, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 43.54, | |
| "learning_rate": 0.00012903225806451613, | |
| "loss": 0.0995, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 45.16, | |
| "learning_rate": 9.67741935483871e-05, | |
| "loss": 0.0985, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 46.77, | |
| "learning_rate": 6.451612903225807e-05, | |
| "loss": 0.0958, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 48.38, | |
| "learning_rate": 3.2258064516129034e-05, | |
| "loss": 0.0942, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 49.99, | |
| "learning_rate": 0.0, | |
| "loss": 0.092, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 49.99, | |
| "step": 3100, | |
| "total_flos": 2.628606767136768e+16, | |
| "train_loss": 0.5685424493974255, | |
| "train_runtime": 3913.436, | |
| "train_samples_per_second": 25.553, | |
| "train_steps_per_second": 0.792 | |
| } | |
| ], | |
| "max_steps": 3100, | |
| "num_train_epochs": 50, | |
| "total_flos": 2.628606767136768e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |