| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.0, | |
| "global_step": 9860, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.8985801217038542e-05, | |
| "loss": 2.0844, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 1.797160243407708e-05, | |
| "loss": 1.322, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.6957403651115622e-05, | |
| "loss": 1.1934, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 1.594320486815416e-05, | |
| "loss": 1.1366, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 1.4929006085192699e-05, | |
| "loss": 1.0634, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 1.3914807302231239e-05, | |
| "loss": 0.855, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 1.2900608519269777e-05, | |
| "loss": 0.8663, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 1.1886409736308317e-05, | |
| "loss": 0.8364, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 1.0872210953346858e-05, | |
| "loss": 0.835, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 9.858012170385396e-06, | |
| "loss": 0.8107, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 8.843813387423936e-06, | |
| "loss": 0.6625, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 7.829614604462476e-06, | |
| "loss": 0.6576, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 6.815415821501015e-06, | |
| "loss": 0.6617, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 5.801217038539554e-06, | |
| "loss": 0.6515, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 4.787018255578094e-06, | |
| "loss": 0.6179, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 3.7728194726166332e-06, | |
| "loss": 0.5304, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 2.7586206896551725e-06, | |
| "loss": 0.5401, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 1.7444219066937122e-06, | |
| "loss": 0.5549, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 7.302231237322515e-07, | |
| "loss": 0.5237, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 9860, | |
| "total_flos": 6.955379978528563e+16, | |
| "train_loss": 0.8512596145846539, | |
| "train_runtime": 3575.2739, | |
| "train_samples_per_second": 99.27, | |
| "train_steps_per_second": 2.758 | |
| } | |
| ], | |
| "max_steps": 9860, | |
| "num_train_epochs": 4, | |
| "total_flos": 6.955379978528563e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |