| { | |
| "best_metric": 96.7742, | |
| "best_model_checkpoint": "/data/tir/projects/tir7/user_data/priyansk/qa_tapex_e2_codet5p-220m_latex/checkpoint-2400", | |
| "epoch": 1.9998255577306998, | |
| "global_step": 3224, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 9.070409429280397e-05, | |
| "loss": 1.8932, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_exact_match": 69.7581, | |
| "eval_loss": 0.006803931202739477, | |
| "eval_runtime": 182.7837, | |
| "eval_samples_per_second": 2.735, | |
| "eval_steps_per_second": 0.175, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 8.140818858560795e-05, | |
| "loss": 0.0034, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_exact_match": 86.4919, | |
| "eval_loss": 0.0015783495036885142, | |
| "eval_runtime": 139.3342, | |
| "eval_samples_per_second": 3.588, | |
| "eval_steps_per_second": 0.23, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 7.211228287841191e-05, | |
| "loss": 0.0012, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_exact_match": 92.5403, | |
| "eval_loss": 0.00091337546473369, | |
| "eval_runtime": 129.2451, | |
| "eval_samples_per_second": 3.869, | |
| "eval_steps_per_second": 0.248, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 6.28163771712159e-05, | |
| "loss": 0.0008, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_exact_match": 94.3548, | |
| "eval_loss": 0.0006592237623408437, | |
| "eval_runtime": 130.2486, | |
| "eval_samples_per_second": 3.839, | |
| "eval_steps_per_second": 0.246, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 5.352047146401985e-05, | |
| "loss": 0.0006, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_exact_match": 94.5565, | |
| "eval_loss": 0.0005109157646074891, | |
| "eval_runtime": 127.9056, | |
| "eval_samples_per_second": 3.909, | |
| "eval_steps_per_second": 0.25, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 4.422456575682382e-05, | |
| "loss": 0.0004, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_exact_match": 95.3629, | |
| "eval_loss": 0.0004209030594211072, | |
| "eval_runtime": 129.0896, | |
| "eval_samples_per_second": 3.873, | |
| "eval_steps_per_second": 0.248, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 3.4928660049627796e-05, | |
| "loss": 0.0004, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_exact_match": 95.1613, | |
| "eval_loss": 0.00042048218892887235, | |
| "eval_runtime": 129.0282, | |
| "eval_samples_per_second": 3.875, | |
| "eval_steps_per_second": 0.248, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 2.563275434243176e-05, | |
| "loss": 0.0003, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "eval_exact_match": 96.7742, | |
| "eval_loss": 0.0003429962380323559, | |
| "eval_runtime": 127.2593, | |
| "eval_samples_per_second": 3.929, | |
| "eval_steps_per_second": 0.251, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 1.6336848635235734e-05, | |
| "loss": 0.0003, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "eval_exact_match": 96.7742, | |
| "eval_loss": 0.00032032810850068927, | |
| "eval_runtime": 126.2849, | |
| "eval_samples_per_second": 3.959, | |
| "eval_steps_per_second": 0.253, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 7.040942928039701e-06, | |
| "loss": 0.0003, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "eval_exact_match": 96.371, | |
| "eval_loss": 0.0003172786091454327, | |
| "eval_runtime": 128.5411, | |
| "eval_samples_per_second": 3.89, | |
| "eval_steps_per_second": 0.249, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 3224, | |
| "total_flos": 4.0211486895557837e+18, | |
| "train_loss": 0.17688602654646998, | |
| "train_runtime": 168482.5606, | |
| "train_samples_per_second": 19.598, | |
| "train_steps_per_second": 0.019 | |
| } | |
| ], | |
| "max_steps": 3224, | |
| "num_train_epochs": 2, | |
| "total_flos": 4.0211486895557837e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |