| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.999700449326011, | |
| "global_step": 15021, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.8338992077757805e-05, | |
| "loss": 2.2639, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 4.667465548232475e-05, | |
| "loss": 1.7816, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 4.501031888689169e-05, | |
| "loss": 1.6597, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.334598229145863e-05, | |
| "loss": 1.5932, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.168164569602557e-05, | |
| "loss": 1.5107, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.0017309100592506e-05, | |
| "loss": 1.4919, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 3.8352972505159445e-05, | |
| "loss": 1.4724, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 3.6688635909726384e-05, | |
| "loss": 1.43, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 3.502429931429332e-05, | |
| "loss": 1.4131, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3.336329139205113e-05, | |
| "loss": 1.3983, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 3.169895479661807e-05, | |
| "loss": 1.3639, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 3.003461820118501e-05, | |
| "loss": 1.3449, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 2.8370281605751946e-05, | |
| "loss": 1.3259, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 2.6709273683509755e-05, | |
| "loss": 1.3239, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 2.504493708807669e-05, | |
| "loss": 1.3131, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 2.3380600492643633e-05, | |
| "loss": 1.3052, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 2.1716263897210572e-05, | |
| "loss": 1.2947, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 2.0058584648159245e-05, | |
| "loss": 1.2927, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 1.8394248052726184e-05, | |
| "loss": 1.2736, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.6729911457293123e-05, | |
| "loss": 1.265, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 1.5065574861860063e-05, | |
| "loss": 1.2467, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 1.3401238266427002e-05, | |
| "loss": 1.2387, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 1.1736901670993943e-05, | |
| "loss": 1.2283, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 1.0072565075560882e-05, | |
| "loss": 1.2271, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 8.408228480127821e-06, | |
| "loss": 1.2382, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 6.7438918846947615e-06, | |
| "loss": 1.2197, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 5.0795552892617005e-06, | |
| "loss": 1.217, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 3.4152186938286403e-06, | |
| "loss": 1.2096, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 1.7508820983955797e-06, | |
| "loss": 1.2048, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.654550296251915e-08, | |
| "loss": 1.2008, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 15021, | |
| "total_flos": 1.1987319801621504e+17, | |
| "train_loss": 1.378044397486057, | |
| "train_runtime": 6661.4989, | |
| "train_samples_per_second": 288.654, | |
| "train_steps_per_second": 2.255 | |
| } | |
| ], | |
| "max_steps": 15021, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.1987319801621504e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |