| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 50.0, | |
| "eval_steps": 500, | |
| "global_step": 17300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 4.855491329479769e-05, | |
| "loss": 1.6594, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 4.710982658959538e-05, | |
| "loss": 0.9215, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 4.566473988439307e-05, | |
| "loss": 0.6453, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "learning_rate": 4.421965317919075e-05, | |
| "loss": 0.4758, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.23, | |
| "learning_rate": 4.2774566473988445e-05, | |
| "loss": 0.3575, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.67, | |
| "learning_rate": 4.132947976878613e-05, | |
| "loss": 0.3063, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 10.12, | |
| "learning_rate": 3.988439306358382e-05, | |
| "loss": 0.31, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 11.56, | |
| "learning_rate": 3.84393063583815e-05, | |
| "loss": 0.3325, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 3.699421965317919e-05, | |
| "loss": 0.414, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 14.45, | |
| "learning_rate": 3.554913294797688e-05, | |
| "loss": 0.4969, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 15.9, | |
| "learning_rate": 3.410404624277457e-05, | |
| "loss": 0.668, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 17.34, | |
| "learning_rate": 3.265895953757225e-05, | |
| "loss": 0.7799, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 18.79, | |
| "learning_rate": 3.1213872832369946e-05, | |
| "loss": 0.8613, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 20.23, | |
| "learning_rate": 2.9768786127167632e-05, | |
| "loss": 0.8462, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 21.68, | |
| "learning_rate": 2.832369942196532e-05, | |
| "loss": 0.7924, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 23.12, | |
| "learning_rate": 2.6878612716763007e-05, | |
| "loss": 0.7011, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 24.57, | |
| "learning_rate": 2.5433526011560693e-05, | |
| "loss": 0.5652, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 26.01, | |
| "learning_rate": 2.3988439306358382e-05, | |
| "loss": 0.4699, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 27.46, | |
| "learning_rate": 2.254335260115607e-05, | |
| "loss": 0.3833, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 28.9, | |
| "learning_rate": 2.1098265895953757e-05, | |
| "loss": 0.3351, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 30.35, | |
| "learning_rate": 1.9653179190751446e-05, | |
| "loss": 0.2929, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 31.79, | |
| "learning_rate": 1.8208092485549132e-05, | |
| "loss": 0.2636, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 33.24, | |
| "learning_rate": 1.676300578034682e-05, | |
| "loss": 0.2358, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 34.68, | |
| "learning_rate": 1.531791907514451e-05, | |
| "loss": 0.2153, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "learning_rate": 1.3872832369942197e-05, | |
| "loss": 0.1957, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 37.57, | |
| "learning_rate": 1.2427745664739884e-05, | |
| "loss": 0.1784, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 39.02, | |
| "learning_rate": 1.0982658959537573e-05, | |
| "loss": 0.169, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 40.46, | |
| "learning_rate": 9.53757225433526e-06, | |
| "loss": 0.1565, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 41.91, | |
| "learning_rate": 8.092485549132949e-06, | |
| "loss": 0.1508, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 43.35, | |
| "learning_rate": 6.647398843930635e-06, | |
| "loss": 0.1412, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 44.8, | |
| "learning_rate": 5.202312138728324e-06, | |
| "loss": 0.1378, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 46.24, | |
| "learning_rate": 3.757225433526012e-06, | |
| "loss": 0.1329, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 47.69, | |
| "learning_rate": 2.3121387283236993e-06, | |
| "loss": 0.131, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 49.13, | |
| "learning_rate": 8.670520231213873e-07, | |
| "loss": 0.1268, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 17300, | |
| "total_flos": 4.337225635212288e+17, | |
| "train_loss": 0.4313709532456591, | |
| "train_runtime": 11982.9489, | |
| "train_samples_per_second": 369.375, | |
| "train_steps_per_second": 1.444 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 17300, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 4.337225635212288e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |