| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "global_step": 6220, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9522821307182312, | |
| "eval_loss": 0.12891167402267456, | |
| "eval_runtime": 9.8242, | |
| "eval_samples_per_second": 147.188, | |
| "eval_steps_per_second": 18.424, | |
| "step": 311 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 3.106972231499722e-05, | |
| "loss": 0.2799, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9612724781036377, | |
| "eval_loss": 0.11741334199905396, | |
| "eval_runtime": 9.9246, | |
| "eval_samples_per_second": 145.698, | |
| "eval_steps_per_second": 18.237, | |
| "step": 622 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9688796401023865, | |
| "eval_loss": 0.1398700326681137, | |
| "eval_runtime": 9.777, | |
| "eval_samples_per_second": 147.898, | |
| "eval_steps_per_second": 18.513, | |
| "step": 933 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 2.8353837497252705e-05, | |
| "loss": 0.0986, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9640387296676636, | |
| "eval_loss": 0.19078733026981354, | |
| "eval_runtime": 10.086, | |
| "eval_samples_per_second": 143.368, | |
| "eval_steps_per_second": 17.946, | |
| "step": 1244 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 2.563795267950819e-05, | |
| "loss": 0.0467, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9598893523216248, | |
| "eval_loss": 0.2066243439912796, | |
| "eval_runtime": 9.9772, | |
| "eval_samples_per_second": 144.93, | |
| "eval_steps_per_second": 18.141, | |
| "step": 1555 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9619640111923218, | |
| "eval_loss": 0.23871752619743347, | |
| "eval_runtime": 9.7438, | |
| "eval_samples_per_second": 148.402, | |
| "eval_steps_per_second": 18.576, | |
| "step": 1866 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "learning_rate": 2.292206786176368e-05, | |
| "loss": 0.0299, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9640387296676636, | |
| "eval_loss": 0.2593800723552704, | |
| "eval_runtime": 9.8624, | |
| "eval_samples_per_second": 146.617, | |
| "eval_steps_per_second": 18.352, | |
| "step": 2177 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9681881070137024, | |
| "eval_loss": 0.21497485041618347, | |
| "eval_runtime": 9.6756, | |
| "eval_samples_per_second": 149.448, | |
| "eval_steps_per_second": 18.707, | |
| "step": 2488 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 2.020618304401917e-05, | |
| "loss": 0.0174, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9702627658843994, | |
| "eval_loss": 0.21150082349777222, | |
| "eval_runtime": 10.0341, | |
| "eval_samples_per_second": 144.108, | |
| "eval_steps_per_second": 18.038, | |
| "step": 2799 | |
| }, | |
| { | |
| "epoch": 9.65, | |
| "learning_rate": 1.749029822627466e-05, | |
| "loss": 0.0133, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9661133885383606, | |
| "eval_loss": 0.2336471825838089, | |
| "eval_runtime": 9.6376, | |
| "eval_samples_per_second": 150.037, | |
| "eval_steps_per_second": 18.781, | |
| "step": 3110 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9640387296676636, | |
| "eval_loss": 0.26857590675354004, | |
| "eval_runtime": 10.0374, | |
| "eval_samples_per_second": 144.061, | |
| "eval_steps_per_second": 18.033, | |
| "step": 3421 | |
| }, | |
| { | |
| "epoch": 11.25, | |
| "learning_rate": 1.4774413408530146e-05, | |
| "loss": 0.0065, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9647302627563477, | |
| "eval_loss": 0.29211708903312683, | |
| "eval_runtime": 9.8275, | |
| "eval_samples_per_second": 147.138, | |
| "eval_steps_per_second": 18.418, | |
| "step": 3732 | |
| }, | |
| { | |
| "epoch": 12.86, | |
| "learning_rate": 1.2058528590785633e-05, | |
| "loss": 0.006, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9633471369743347, | |
| "eval_loss": 0.30721694231033325, | |
| "eval_runtime": 9.9209, | |
| "eval_samples_per_second": 145.753, | |
| "eval_steps_per_second": 18.244, | |
| "step": 4043 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9674965143203735, | |
| "eval_loss": 0.30118855834007263, | |
| "eval_runtime": 9.8913, | |
| "eval_samples_per_second": 146.189, | |
| "eval_steps_per_second": 18.299, | |
| "step": 4354 | |
| }, | |
| { | |
| "epoch": 14.47, | |
| "learning_rate": 9.342643773041121e-06, | |
| "loss": 0.0068, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9647302627563477, | |
| "eval_loss": 0.29607123136520386, | |
| "eval_runtime": 9.8325, | |
| "eval_samples_per_second": 147.063, | |
| "eval_steps_per_second": 18.408, | |
| "step": 4665 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9633471369743347, | |
| "eval_loss": 0.29950061440467834, | |
| "eval_runtime": 9.895, | |
| "eval_samples_per_second": 146.134, | |
| "eval_steps_per_second": 18.292, | |
| "step": 4976 | |
| }, | |
| { | |
| "epoch": 16.08, | |
| "learning_rate": 6.626758955296609e-06, | |
| "loss": 0.0062, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.9674965143203735, | |
| "eval_loss": 0.2802109718322754, | |
| "eval_runtime": 9.7901, | |
| "eval_samples_per_second": 147.7, | |
| "eval_steps_per_second": 18.488, | |
| "step": 5287 | |
| }, | |
| { | |
| "epoch": 17.68, | |
| "learning_rate": 3.910874137552097e-06, | |
| "loss": 0.0069, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9688796401023865, | |
| "eval_loss": 0.2822701930999756, | |
| "eval_runtime": 9.8995, | |
| "eval_samples_per_second": 146.068, | |
| "eval_steps_per_second": 18.284, | |
| "step": 5598 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.9681881070137024, | |
| "eval_loss": 0.27318963408470154, | |
| "eval_runtime": 9.6783, | |
| "eval_samples_per_second": 149.407, | |
| "eval_steps_per_second": 18.702, | |
| "step": 5909 | |
| }, | |
| { | |
| "epoch": 19.29, | |
| "learning_rate": 1.1949893198075852e-06, | |
| "loss": 0.0041, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9681881070137024, | |
| "eval_loss": 0.27601131796836853, | |
| "eval_runtime": 9.9846, | |
| "eval_samples_per_second": 144.824, | |
| "eval_steps_per_second": 18.128, | |
| "step": 6220 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 6220, | |
| "total_flos": 6546379231641600.0, | |
| "train_loss": 0.042026480821551235, | |
| "train_runtime": 4438.5906, | |
| "train_samples_per_second": 22.422, | |
| "train_steps_per_second": 1.401 | |
| } | |
| ], | |
| "max_steps": 6220, | |
| "num_train_epochs": 20, | |
| "total_flos": 6546379231641600.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |