| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "global_step": 8241, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 2.8179832544594102e-05, | |
| "loss": 1.3765, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 2.6359665089188204e-05, | |
| "loss": 0.903, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 2.4539497633782306e-05, | |
| "loss": 0.8343, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 2.271933017837641e-05, | |
| "loss": 0.7909, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 2.0899162722970516e-05, | |
| "loss": 0.7492, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 1.9078995267564617e-05, | |
| "loss": 0.6332, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 1.725882781215872e-05, | |
| "loss": 0.5434, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 1.543866035675282e-05, | |
| "loss": 0.5355, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 1.3618492901346924e-05, | |
| "loss": 0.549, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 1.1798325445941026e-05, | |
| "loss": 0.521, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9.97815799053513e-06, | |
| "loss": 0.5263, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 8.157990535129232e-06, | |
| "loss": 0.3775, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 6.337823079723334e-06, | |
| "loss": 0.3725, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 4.517655624317437e-06, | |
| "loss": 0.363, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 2.69748816891154e-06, | |
| "loss": 0.3497, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 8.773207135056426e-07, | |
| "loss": 0.3415, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 8241, | |
| "total_flos": 2.7545611910240102e+17, | |
| "train_loss": 0.6024896357623918, | |
| "train_runtime": 2829.1187, | |
| "train_samples_per_second": 139.785, | |
| "train_steps_per_second": 2.913 | |
| } | |
| ], | |
| "max_steps": 8241, | |
| "num_train_epochs": 3, | |
| "total_flos": 2.7545611910240102e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |