| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.0, | |
| "eval_steps": 500, | |
| "global_step": 3703, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 5e-05, | |
| "loss": 1.6355, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4037, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4183, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3881, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3891, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6137948717948718, | |
| "eval_loss": 1.3015451431274414, | |
| "eval_runtime": 4.1821, | |
| "eval_samples_per_second": 119.557, | |
| "eval_steps_per_second": 15.064, | |
| "step": 529 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_exact_match": 23.2, | |
| "eval_f1": 35.17014759120022, | |
| "step": 529 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3808, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 5e-05, | |
| "loss": 1.348, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3627, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3473, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3633, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6165641025641025, | |
| "eval_loss": 1.2854759693145752, | |
| "eval_runtime": 4.1064, | |
| "eval_samples_per_second": 121.761, | |
| "eval_steps_per_second": 15.342, | |
| "step": 1058 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_exact_match": 25.8, | |
| "eval_f1": 37.509047619047635, | |
| "step": 1058 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3358, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2917, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2948, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3144, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2929, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6176923076923077, | |
| "eval_loss": 1.2953603267669678, | |
| "eval_runtime": 4.0992, | |
| "eval_samples_per_second": 121.974, | |
| "eval_steps_per_second": 15.369, | |
| "step": 1587 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_exact_match": 25.0, | |
| "eval_f1": 37.08253968253969, | |
| "step": 1587 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2906, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2216, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2459, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2474, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2518, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2361, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6045128205128205, | |
| "eval_loss": 1.3488503694534302, | |
| "eval_runtime": 4.2015, | |
| "eval_samples_per_second": 119.004, | |
| "eval_steps_per_second": 14.995, | |
| "step": 2116 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_exact_match": 25.0, | |
| "eval_f1": 36.82126984126986, | |
| "step": 2116 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1856, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1556, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1828, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1687, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1856, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6125128205128205, | |
| "eval_loss": 1.3968186378479004, | |
| "eval_runtime": 4.3081, | |
| "eval_samples_per_second": 116.06, | |
| "eval_steps_per_second": 14.624, | |
| "step": 2645 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_exact_match": 24.6, | |
| "eval_f1": 37.079523809523806, | |
| "step": 2645 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1376, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1079, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0954, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1317, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1098, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6114871794871795, | |
| "eval_loss": 1.4720605611801147, | |
| "eval_runtime": 4.5, | |
| "eval_samples_per_second": 111.11, | |
| "eval_steps_per_second": 14.0, | |
| "step": 3174 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_exact_match": 24.4, | |
| "eval_f1": 35.78015873015873, | |
| "step": 3174 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0945, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0309, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0389, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0547, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0618, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0753, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6076410256410256, | |
| "eval_loss": 1.5797967910766602, | |
| "eval_runtime": 4.8128, | |
| "eval_samples_per_second": 103.891, | |
| "eval_steps_per_second": 13.09, | |
| "step": 3703 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_exact_match": 21.8, | |
| "eval_f1": 33.15111111111111, | |
| "step": 3703 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 26450, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 6.255076307642614e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |