| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "global_step": 14584, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 2.897147558968733e-05, | |
| "loss": 3.8538, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 2.794295117937466e-05, | |
| "loss": 2.9886, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 2.6914426769061988e-05, | |
| "loss": 2.685, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 2.5885902358749317e-05, | |
| "loss": 2.4911, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 2.4857377948436646e-05, | |
| "loss": 2.2862, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 2.3828853538123975e-05, | |
| "loss": 2.14, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 2.28003291278113e-05, | |
| "loss": 2.122, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 2.177180471749863e-05, | |
| "loss": 2.0886, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 2.074328030718596e-05, | |
| "loss": 2.0471, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.9714755896873284e-05, | |
| "loss": 1.9984, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 1.8686231486560613e-05, | |
| "loss": 2.0018, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.7657707076247942e-05, | |
| "loss": 1.9633, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.662918266593527e-05, | |
| "loss": 1.9129, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 1.56006582556226e-05, | |
| "loss": 1.9094, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 1.4572133845309929e-05, | |
| "loss": 1.8448, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1.3543609434997258e-05, | |
| "loss": 1.7691, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 1.2515085024684585e-05, | |
| "loss": 1.7561, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 1.1486560614371914e-05, | |
| "loss": 1.7437, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.0458036204059243e-05, | |
| "loss": 1.7462, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 9.429511793746572e-06, | |
| "loss": 1.6874, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 8.400987383433901e-06, | |
| "loss": 1.7223, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 7.372462973121229e-06, | |
| "loss": 1.7071, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 6.343938562808558e-06, | |
| "loss": 1.7036, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 5.315414152495887e-06, | |
| "loss": 1.6991, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 4.286889742183214e-06, | |
| "loss": 1.6523, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 3.258365331870543e-06, | |
| "loss": 1.6678, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 2.229840921557872e-06, | |
| "loss": 1.6783, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 1.2013165112452003e-06, | |
| "loss": 1.6777, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 1.727921009325288e-07, | |
| "loss": 1.6829, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 14584, | |
| "total_flos": 1842702589181952.0, | |
| "train_loss": 2.0058399489209733, | |
| "train_runtime": 1498.6781, | |
| "train_samples_per_second": 116.775, | |
| "train_steps_per_second": 9.731 | |
| } | |
| ], | |
| "max_steps": 14584, | |
| "num_train_epochs": 2, | |
| "total_flos": 1842702589181952.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |