| { | |
| "best_metric": 3.929802894592285, | |
| "best_model_checkpoint": "output/Chinese-poetry/SikuGPT2-POEM/checkpoint-5000", | |
| "epoch": 3.0, | |
| "global_step": 9810, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.745158002038736e-05, | |
| "loss": 4.361, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 4.490316004077472e-05, | |
| "loss": 4.2153, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.235474006116208e-05, | |
| "loss": 4.1768, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 3.980632008154944e-05, | |
| "loss": 4.1459, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 3.72579001019368e-05, | |
| "loss": 4.1214, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 3.4709480122324164e-05, | |
| "loss": 4.0975, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 3.2161060142711516e-05, | |
| "loss": 4.0607, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 2.9612640163098882e-05, | |
| "loss": 4.0215, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 2.7064220183486238e-05, | |
| "loss": 4.007, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 2.45158002038736e-05, | |
| "loss": 4.0004, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_loss": 3.929802894592285, | |
| "eval_runtime": 283.0836, | |
| "eval_samples_per_second": 7.69, | |
| "eval_steps_per_second": 0.964, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 2.196738022426096e-05, | |
| "loss": 3.9871, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 1.9418960244648318e-05, | |
| "loss": 3.9853, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 1.6870540265035677e-05, | |
| "loss": 3.9716, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 1.4322120285423038e-05, | |
| "loss": 3.9384, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 1.1773700305810397e-05, | |
| "loss": 3.928, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 9.225280326197758e-06, | |
| "loss": 3.9257, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 6.676860346585118e-06, | |
| "loss": 3.92, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 4.128440366972477e-06, | |
| "loss": 3.9186, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 1.580020387359837e-06, | |
| "loss": 3.9191, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 9810, | |
| "total_flos": 3.0753026998272e+16, | |
| "train_loss": 4.032964795469387, | |
| "train_runtime": 12111.7268, | |
| "train_samples_per_second": 4.859, | |
| "train_steps_per_second": 0.81 | |
| } | |
| ], | |
| "max_steps": 9810, | |
| "num_train_epochs": 3, | |
| "total_flos": 3.0753026998272e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |