| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 1587, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 3e-05, | |
| "loss": 1.7194, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 3e-05, | |
| "loss": 1.4369, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 3e-05, | |
| "loss": 1.4257, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3947, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3948, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6132307692307692, | |
| "eval_loss": 1.3087366819381714, | |
| "eval_runtime": 4.3553, | |
| "eval_samples_per_second": 114.803, | |
| "eval_steps_per_second": 14.465, | |
| "step": 529 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_exact_match": 24.4, | |
| "eval_f1": 37.614420394420414, | |
| "step": 529 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3927, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3644, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3791, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3627, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3789, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6146153846153846, | |
| "eval_loss": 1.289727807044983, | |
| "eval_runtime": 4.4281, | |
| "eval_samples_per_second": 112.915, | |
| "eval_steps_per_second": 14.227, | |
| "step": 1058 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_exact_match": 23.2, | |
| "eval_f1": 35.56775335775336, | |
| "step": 1058 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3605, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3306, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3318, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3491, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3259, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6178974358974358, | |
| "eval_loss": 1.2849316596984863, | |
| "eval_runtime": 4.875, | |
| "eval_samples_per_second": 102.565, | |
| "eval_steps_per_second": 12.923, | |
| "step": 1587 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_exact_match": 25.2, | |
| "eval_f1": 37.32698412698414, | |
| "step": 1587 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 26450, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 2.679771250074583e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |