| { | |
| "best_metric": 0.8226299694189603, | |
| "best_model_checkpoint": "checkpoints/boolq-roberta/checkpoint-5605", | |
| "epoch": 19.0, | |
| "global_step": 5605, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6217125382262997, | |
| "eval_loss": 1.2398264408111572, | |
| "eval_runtime": 8.8922, | |
| "eval_samples_per_second": 367.737, | |
| "eval_steps_per_second": 45.995, | |
| "step": 295 | |
| }, | |
| { | |
| "best_epoch": 0, | |
| "best_eval_accuracy": 0.6217125382262997, | |
| "epoch": 1.0, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.006406779661016949, | |
| "loss": 0.8402, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6422018348623854, | |
| "eval_loss": 0.6416757702827454, | |
| "eval_runtime": 8.9446, | |
| "eval_samples_per_second": 365.585, | |
| "eval_steps_per_second": 45.726, | |
| "step": 590 | |
| }, | |
| { | |
| "best_epoch": 1, | |
| "best_eval_accuracy": 0.6422018348623854, | |
| "epoch": 2.0, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7186544342507645, | |
| "eval_loss": 0.5749825239181519, | |
| "eval_runtime": 8.9497, | |
| "eval_samples_per_second": 365.375, | |
| "eval_steps_per_second": 45.7, | |
| "step": 885 | |
| }, | |
| { | |
| "best_epoch": 2, | |
| "best_eval_accuracy": 0.7186544342507645, | |
| "epoch": 3.0, | |
| "step": 885 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 0.005813559322033898, | |
| "loss": 0.6921, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6801223241590214, | |
| "eval_loss": 0.6661111116409302, | |
| "eval_runtime": 8.9486, | |
| "eval_samples_per_second": 365.422, | |
| "eval_steps_per_second": 45.706, | |
| "step": 1180 | |
| }, | |
| { | |
| "best_epoch": 2, | |
| "best_eval_accuracy": 0.7186544342507645, | |
| "epoch": 4.0, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7642201834862385, | |
| "eval_loss": 0.4996618628501892, | |
| "eval_runtime": 8.9587, | |
| "eval_samples_per_second": 365.007, | |
| "eval_steps_per_second": 45.654, | |
| "step": 1475 | |
| }, | |
| { | |
| "best_epoch": 4, | |
| "best_eval_accuracy": 0.7642201834862385, | |
| "epoch": 5.0, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 0.005220338983050848, | |
| "loss": 0.5652, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7896024464831805, | |
| "eval_loss": 0.5553064346313477, | |
| "eval_runtime": 8.9376, | |
| "eval_samples_per_second": 365.869, | |
| "eval_steps_per_second": 45.762, | |
| "step": 1770 | |
| }, | |
| { | |
| "best_epoch": 5, | |
| "best_eval_accuracy": 0.7896024464831805, | |
| "epoch": 6.0, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 0.004627118644067797, | |
| "loss": 0.516, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8107033639143731, | |
| "eval_loss": 0.43402931094169617, | |
| "eval_runtime": 8.9391, | |
| "eval_samples_per_second": 365.811, | |
| "eval_steps_per_second": 45.754, | |
| "step": 2065 | |
| }, | |
| { | |
| "best_epoch": 6, | |
| "best_eval_accuracy": 0.8107033639143731, | |
| "epoch": 7.0, | |
| "step": 2065 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7577981651376147, | |
| "eval_loss": 0.5680531859397888, | |
| "eval_runtime": 8.9453, | |
| "eval_samples_per_second": 365.553, | |
| "eval_steps_per_second": 45.722, | |
| "step": 2360 | |
| }, | |
| { | |
| "best_epoch": 6, | |
| "best_eval_accuracy": 0.8107033639143731, | |
| "epoch": 8.0, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "learning_rate": 0.004033898305084746, | |
| "loss": 0.4934, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7944954128440367, | |
| "eval_loss": 0.4493260383605957, | |
| "eval_runtime": 8.9507, | |
| "eval_samples_per_second": 365.334, | |
| "eval_steps_per_second": 45.695, | |
| "step": 2655 | |
| }, | |
| { | |
| "best_epoch": 6, | |
| "best_eval_accuracy": 0.8107033639143731, | |
| "epoch": 9.0, | |
| "step": 2655 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8048929663608563, | |
| "eval_loss": 0.4496236741542816, | |
| "eval_runtime": 8.9349, | |
| "eval_samples_per_second": 365.983, | |
| "eval_steps_per_second": 45.776, | |
| "step": 2950 | |
| }, | |
| { | |
| "best_epoch": 6, | |
| "best_eval_accuracy": 0.8107033639143731, | |
| "epoch": 10.0, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 10.17, | |
| "learning_rate": 0.0034406779661016952, | |
| "loss": 0.4653, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8140672782874617, | |
| "eval_loss": 0.435769259929657, | |
| "eval_runtime": 8.9447, | |
| "eval_samples_per_second": 365.58, | |
| "eval_steps_per_second": 45.725, | |
| "step": 3245 | |
| }, | |
| { | |
| "best_epoch": 10, | |
| "best_eval_accuracy": 0.8140672782874617, | |
| "epoch": 11.0, | |
| "step": 3245 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "learning_rate": 0.002847457627118644, | |
| "loss": 0.4084, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8180428134556575, | |
| "eval_loss": 0.41644611954689026, | |
| "eval_runtime": 8.9366, | |
| "eval_samples_per_second": 365.912, | |
| "eval_steps_per_second": 45.767, | |
| "step": 3540 | |
| }, | |
| { | |
| "best_epoch": 11, | |
| "best_eval_accuracy": 0.8180428134556575, | |
| "epoch": 12.0, | |
| "step": 3540 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8103975535168195, | |
| "eval_loss": 0.452688992023468, | |
| "eval_runtime": 8.9757, | |
| "eval_samples_per_second": 364.318, | |
| "eval_steps_per_second": 45.568, | |
| "step": 3835 | |
| }, | |
| { | |
| "best_epoch": 11, | |
| "best_eval_accuracy": 0.8180428134556575, | |
| "epoch": 13.0, | |
| "step": 3835 | |
| }, | |
| { | |
| "epoch": 13.56, | |
| "learning_rate": 0.002254237288135593, | |
| "loss": 0.3903, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.818348623853211, | |
| "eval_loss": 0.44870346784591675, | |
| "eval_runtime": 8.9507, | |
| "eval_samples_per_second": 365.335, | |
| "eval_steps_per_second": 45.695, | |
| "step": 4130 | |
| }, | |
| { | |
| "best_epoch": 13, | |
| "best_eval_accuracy": 0.818348623853211, | |
| "epoch": 14.0, | |
| "step": 4130 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8155963302752294, | |
| "eval_loss": 0.4322122633457184, | |
| "eval_runtime": 8.9612, | |
| "eval_samples_per_second": 364.908, | |
| "eval_steps_per_second": 45.641, | |
| "step": 4425 | |
| }, | |
| { | |
| "best_epoch": 13, | |
| "best_eval_accuracy": 0.818348623853211, | |
| "epoch": 15.0, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 15.25, | |
| "learning_rate": 0.0016610169491525426, | |
| "loss": 0.3761, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8211009174311926, | |
| "eval_loss": 0.4511786997318268, | |
| "eval_runtime": 8.9443, | |
| "eval_samples_per_second": 365.595, | |
| "eval_steps_per_second": 45.727, | |
| "step": 4720 | |
| }, | |
| { | |
| "best_epoch": 15, | |
| "best_eval_accuracy": 0.8211009174311926, | |
| "epoch": 16.0, | |
| "step": 4720 | |
| }, | |
| { | |
| "epoch": 16.95, | |
| "learning_rate": 0.0010677966101694915, | |
| "loss": 0.3515, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8122324159021407, | |
| "eval_loss": 0.4329771399497986, | |
| "eval_runtime": 8.9372, | |
| "eval_samples_per_second": 365.885, | |
| "eval_steps_per_second": 45.764, | |
| "step": 5015 | |
| }, | |
| { | |
| "best_epoch": 15, | |
| "best_eval_accuracy": 0.8211009174311926, | |
| "epoch": 17.0, | |
| "step": 5015 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8171253822629969, | |
| "eval_loss": 0.4531969130039215, | |
| "eval_runtime": 8.9346, | |
| "eval_samples_per_second": 365.994, | |
| "eval_steps_per_second": 45.777, | |
| "step": 5310 | |
| }, | |
| { | |
| "best_epoch": 15, | |
| "best_eval_accuracy": 0.8211009174311926, | |
| "epoch": 18.0, | |
| "step": 5310 | |
| }, | |
| { | |
| "epoch": 18.64, | |
| "learning_rate": 0.0004745762711864407, | |
| "loss": 0.3487, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8226299694189603, | |
| "eval_loss": 0.45660164952278137, | |
| "eval_runtime": 8.9479, | |
| "eval_samples_per_second": 365.448, | |
| "eval_steps_per_second": 45.709, | |
| "step": 5605 | |
| }, | |
| { | |
| "best_epoch": 18, | |
| "best_eval_accuracy": 0.8226299694189603, | |
| "epoch": 19.0, | |
| "step": 5605 | |
| } | |
| ], | |
| "max_steps": 5900, | |
| "num_train_epochs": 20, | |
| "total_flos": 4.173025556994202e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |