| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9999496263957686, | |
| "global_step": 14888, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 4.8320795271359486e-05, | |
| "loss": 1.1362, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 4.664159054271897e-05, | |
| "loss": 0.893, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.4962385814078453e-05, | |
| "loss": 0.8577, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.328318108543794e-05, | |
| "loss": 0.844, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.160397635679743e-05, | |
| "loss": 0.8234, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.992477162815691e-05, | |
| "loss": 0.8012, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 3.824556689951639e-05, | |
| "loss": 0.7857, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 3.656636217087587e-05, | |
| "loss": 0.7729, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 3.4887157442235356e-05, | |
| "loss": 0.7692, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 3.320795271359484e-05, | |
| "loss": 0.7535, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 3.152874798495433e-05, | |
| "loss": 0.7449, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2.984954325631381e-05, | |
| "loss": 0.7479, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 2.8170338527673297e-05, | |
| "loss": 0.7434, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 2.649113379903278e-05, | |
| "loss": 0.7316, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 2.4811929070392265e-05, | |
| "loss": 0.7168, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 2.3132724341751745e-05, | |
| "loss": 0.7216, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 2.1453519613111232e-05, | |
| "loss": 0.7154, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 1.9774314884470716e-05, | |
| "loss": 0.707, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 1.80951101558302e-05, | |
| "loss": 0.7104, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1.6415905427189683e-05, | |
| "loss": 0.7074, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.4736700698549168e-05, | |
| "loss": 0.6972, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 1.305749596990865e-05, | |
| "loss": 0.6836, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 1.1378291241268136e-05, | |
| "loss": 0.6849, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 9.69908651262762e-06, | |
| "loss": 0.6836, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 8.019881783987105e-06, | |
| "loss": 0.6795, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 6.3406770553465885e-06, | |
| "loss": 0.6776, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.661472326706072e-06, | |
| "loss": 0.6816, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 2.9822675980655563e-06, | |
| "loss": 0.6736, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.3030628694250402e-06, | |
| "loss": 0.6719, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 14888, | |
| "total_flos": 1.4932161644834488e+18, | |
| "train_loss": 0.7501088408357927, | |
| "train_runtime": 33342.7498, | |
| "train_samples_per_second": 3.572, | |
| "train_steps_per_second": 0.447 | |
| } | |
| ], | |
| "max_steps": 14888, | |
| "num_train_epochs": 1, | |
| "total_flos": 1.4932161644834488e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |