| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "global_step": 14754, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 2.8983326555510372e-05, | |
| "loss": 2.4671, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 2.796665311102074e-05, | |
| "loss": 1.5004, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 2.694997966653111e-05, | |
| "loss": 1.3819, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 2.593330622204148e-05, | |
| "loss": 1.3173, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 2.491663277755185e-05, | |
| "loss": 1.2579, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 2.389995933306222e-05, | |
| "loss": 1.2089, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 2.288328588857259e-05, | |
| "loss": 1.1875, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 2.1866612444082963e-05, | |
| "loss": 1.1345, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 2.084993899959333e-05, | |
| "loss": 1.0905, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.9833265555103702e-05, | |
| "loss": 1.0827, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 1.8816592110614073e-05, | |
| "loss": 1.1219, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 1.779991866612444e-05, | |
| "loss": 1.0689, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 1.6783245221634812e-05, | |
| "loss": 1.0331, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 1.576657177714518e-05, | |
| "loss": 1.0555, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1.4749898332655551e-05, | |
| "loss": 0.9443, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 1.373322488816592e-05, | |
| "loss": 0.7268, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 1.271655144367629e-05, | |
| "loss": 0.7516, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 1.1699877999186661e-05, | |
| "loss": 0.7257, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 1.0683204554697033e-05, | |
| "loss": 0.7033, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 9.666531110207402e-06, | |
| "loss": 0.7459, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 8.649857665717772e-06, | |
| "loss": 0.7182, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 7.633184221228141e-06, | |
| "loss": 0.7289, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 6.616510776738511e-06, | |
| "loss": 0.7087, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 5.5998373322488825e-06, | |
| "loss": 0.7312, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 4.583163887759252e-06, | |
| "loss": 0.6932, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 3.566490443269622e-06, | |
| "loss": 0.6984, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 2.549816998779992e-06, | |
| "loss": 0.709, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 1.533143554290362e-06, | |
| "loss": 0.722, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 5.164701098007319e-07, | |
| "loss": 0.7122, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 14754, | |
| "total_flos": 3.4696551139946496e+16, | |
| "train_loss": 0.9924759711614223, | |
| "train_runtime": 6803.2023, | |
| "train_samples_per_second": 26.024, | |
| "train_steps_per_second": 2.169 | |
| } | |
| ], | |
| "max_steps": 14754, | |
| "num_train_epochs": 2, | |
| "total_flos": 3.4696551139946496e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |