| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 11.0, | |
| "global_step": 10329, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 5e-05, | |
| "loss": 4.5299, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu1_score": 0.1103, | |
| "eval_bleu2_score": 0.0516, | |
| "eval_bleu3_score": 0.0302, | |
| "eval_bleu4_score": 0.019, | |
| "eval_loss": 3.4503543376922607, | |
| "eval_rougeL": 0.1058, | |
| "eval_runtime": 174.17, | |
| "eval_samples_per_second": 9.588, | |
| "eval_steps_per_second": 0.603, | |
| "step": 939 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 4.745650625699461e-05, | |
| "loss": 3.6998, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 4.4913012513989216e-05, | |
| "loss": 3.374, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu1_score": 0.1251, | |
| "eval_bleu2_score": 0.0617, | |
| "eval_bleu3_score": 0.0349, | |
| "eval_bleu4_score": 0.0207, | |
| "eval_loss": 3.306809425354004, | |
| "eval_rougeL": 0.1227, | |
| "eval_runtime": 174.9295, | |
| "eval_samples_per_second": 9.547, | |
| "eval_steps_per_second": 0.6, | |
| "step": 1878 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 4.2369518770983826e-05, | |
| "loss": 3.2617, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 3.982602502797843e-05, | |
| "loss": 2.9885, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu1_score": 0.1377, | |
| "eval_bleu2_score": 0.0693, | |
| "eval_bleu3_score": 0.0386, | |
| "eval_bleu4_score": 0.0223, | |
| "eval_loss": 3.2763712406158447, | |
| "eval_rougeL": 0.1372, | |
| "eval_runtime": 174.431, | |
| "eval_samples_per_second": 9.574, | |
| "eval_steps_per_second": 0.602, | |
| "step": 2817 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 3.7282531284973047e-05, | |
| "loss": 2.8648, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 3.473903754196765e-05, | |
| "loss": 2.6525, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bleu1_score": 0.1382, | |
| "eval_bleu2_score": 0.073, | |
| "eval_bleu3_score": 0.0432, | |
| "eval_bleu4_score": 0.027, | |
| "eval_loss": 3.3136885166168213, | |
| "eval_rougeL": 0.1429, | |
| "eval_runtime": 173.1129, | |
| "eval_samples_per_second": 9.647, | |
| "eval_steps_per_second": 0.607, | |
| "step": 3756 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 3.219554379896226e-05, | |
| "loss": 2.5232, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 2.9652050055956863e-05, | |
| "loss": 2.3911, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_bleu1_score": 0.1455, | |
| "eval_bleu2_score": 0.076, | |
| "eval_bleu3_score": 0.0451, | |
| "eval_bleu4_score": 0.0285, | |
| "eval_loss": 3.354043960571289, | |
| "eval_rougeL": 0.1431, | |
| "eval_runtime": 172.1208, | |
| "eval_samples_per_second": 9.702, | |
| "eval_steps_per_second": 0.61, | |
| "step": 4695 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "learning_rate": 2.710855631295147e-05, | |
| "loss": 2.2543, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 2.456506256994608e-05, | |
| "loss": 2.1732, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_bleu1_score": 0.1462, | |
| "eval_bleu2_score": 0.0784, | |
| "eval_bleu3_score": 0.0486, | |
| "eval_bleu4_score": 0.0326, | |
| "eval_loss": 3.422563314437866, | |
| "eval_rougeL": 0.1439, | |
| "eval_runtime": 172.488, | |
| "eval_samples_per_second": 9.682, | |
| "eval_steps_per_second": 0.609, | |
| "step": 5634 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "learning_rate": 2.2021568826940687e-05, | |
| "loss": 2.021, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 6.92, | |
| "learning_rate": 1.9478075083935294e-05, | |
| "loss": 1.966, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_bleu1_score": 0.1527, | |
| "eval_bleu2_score": 0.0836, | |
| "eval_bleu3_score": 0.0514, | |
| "eval_bleu4_score": 0.035, | |
| "eval_loss": 3.4865562915802, | |
| "eval_rougeL": 0.1446, | |
| "eval_runtime": 173.8607, | |
| "eval_samples_per_second": 9.605, | |
| "eval_steps_per_second": 0.604, | |
| "step": 6573 | |
| }, | |
| { | |
| "epoch": 7.45, | |
| "learning_rate": 1.6934581340929904e-05, | |
| "loss": 1.8372, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "learning_rate": 1.4391087597924511e-05, | |
| "loss": 1.8303, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_bleu1_score": 0.1517, | |
| "eval_bleu2_score": 0.0854, | |
| "eval_bleu3_score": 0.0551, | |
| "eval_bleu4_score": 0.0387, | |
| "eval_loss": 3.54579758644104, | |
| "eval_rougeL": 0.1532, | |
| "eval_runtime": 172.6178, | |
| "eval_samples_per_second": 9.675, | |
| "eval_steps_per_second": 0.608, | |
| "step": 7512 | |
| }, | |
| { | |
| "epoch": 8.52, | |
| "learning_rate": 1.1847593854919118e-05, | |
| "loss": 1.6837, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_bleu1_score": 0.1575, | |
| "eval_bleu2_score": 0.0881, | |
| "eval_bleu3_score": 0.0575, | |
| "eval_bleu4_score": 0.0414, | |
| "eval_loss": 3.5976877212524414, | |
| "eval_rougeL": 0.1527, | |
| "eval_runtime": 173.4085, | |
| "eval_samples_per_second": 9.63, | |
| "eval_steps_per_second": 0.606, | |
| "step": 8451 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 9.304100111913726e-06, | |
| "loss": 1.6791, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 9.58, | |
| "learning_rate": 6.760606368908333e-06, | |
| "loss": 1.591, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_bleu1_score": 0.1533, | |
| "eval_bleu2_score": 0.0887, | |
| "eval_bleu3_score": 0.0595, | |
| "eval_bleu4_score": 0.0437, | |
| "eval_loss": 3.6393702030181885, | |
| "eval_rougeL": 0.1505, | |
| "eval_runtime": 172.9918, | |
| "eval_samples_per_second": 9.654, | |
| "eval_steps_per_second": 0.607, | |
| "step": 9390 | |
| }, | |
| { | |
| "epoch": 10.12, | |
| "learning_rate": 4.217112625902941e-06, | |
| "loss": 1.5695, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 10.65, | |
| "learning_rate": 1.6736188828975481e-06, | |
| "loss": 1.5271, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_bleu1_score": 0.1599, | |
| "eval_bleu2_score": 0.0937, | |
| "eval_bleu3_score": 0.0638, | |
| "eval_bleu4_score": 0.0478, | |
| "eval_loss": 3.6564855575561523, | |
| "eval_rougeL": 0.1548, | |
| "eval_runtime": 172.8273, | |
| "eval_samples_per_second": 9.663, | |
| "eval_steps_per_second": 0.608, | |
| "step": 10329 | |
| } | |
| ], | |
| "max_steps": 10329, | |
| "num_train_epochs": 11, | |
| "total_flos": 1.0063211828871168e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |