| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 5.8857197761535645, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.563, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7243107769423559, | |
| "eval_f1": 0.6586454703832753, | |
| "eval_loss": 0.5137906670570374, | |
| "eval_precision": 0.6636154141595185, | |
| "eval_recall": 0.65493726132024, | |
| "eval_runtime": 5.1569, | |
| "eval_samples_per_second": 77.372, | |
| "eval_steps_per_second": 9.696, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.8023645877838135, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.509, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7167919799498746, | |
| "eval_f1": 0.68198649992594, | |
| "eval_loss": 0.5057324767112732, | |
| "eval_precision": 0.6763453815261045, | |
| "eval_recall": 0.6996272049463539, | |
| "eval_runtime": 5.0504, | |
| "eval_samples_per_second": 79.004, | |
| "eval_steps_per_second": 9.9, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.055728912353516, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4924, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7393483709273183, | |
| "eval_f1": 0.6901433691756272, | |
| "eval_loss": 0.4707716703414917, | |
| "eval_precision": 0.6876546482856133, | |
| "eval_recall": 0.6930805601018367, | |
| "eval_runtime": 5.0475, | |
| "eval_samples_per_second": 79.049, | |
| "eval_steps_per_second": 9.906, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.744323253631592, | |
| "learning_rate": 4e-05, | |
| "loss": 0.468, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7844611528822055, | |
| "eval_f1": 0.7286143625434989, | |
| "eval_loss": 0.4378769099712372, | |
| "eval_precision": 0.741162203468669, | |
| "eval_recall": 0.7199945444626296, | |
| "eval_runtime": 5.0868, | |
| "eval_samples_per_second": 78.438, | |
| "eval_steps_per_second": 9.829, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.431583285331726, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.4495, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7593984962406015, | |
| "eval_f1": 0.7313131313131314, | |
| "eval_loss": 0.44655734300613403, | |
| "eval_precision": 0.7233381157340986, | |
| "eval_recall": 0.7547735951991271, | |
| "eval_runtime": 5.1176, | |
| "eval_samples_per_second": 77.966, | |
| "eval_steps_per_second": 9.77, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.092684745788574, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.4334, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8270676691729323, | |
| "eval_f1": 0.7887122892379951, | |
| "eval_loss": 0.40413278341293335, | |
| "eval_precision": 0.7926829268292683, | |
| "eval_recall": 0.7851427532278596, | |
| "eval_runtime": 5.0433, | |
| "eval_samples_per_second": 79.114, | |
| "eval_steps_per_second": 9.914, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.259237289428711, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.415, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7994987468671679, | |
| "eval_f1": 0.765982404692082, | |
| "eval_loss": 0.4057486653327942, | |
| "eval_precision": 0.7590435228437963, | |
| "eval_recall": 0.7756410256410257, | |
| "eval_runtime": 5.0592, | |
| "eval_samples_per_second": 78.867, | |
| "eval_steps_per_second": 9.883, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.503615379333496, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3974, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8320802005012531, | |
| "eval_f1": 0.7958954625621293, | |
| "eval_loss": 0.3851749002933502, | |
| "eval_precision": 0.7982456140350878, | |
| "eval_recall": 0.7936897617748682, | |
| "eval_runtime": 5.1017, | |
| "eval_samples_per_second": 78.209, | |
| "eval_steps_per_second": 9.801, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 3.705313205718994, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.3849, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8245614035087719, | |
| "eval_f1": 0.7894173351830629, | |
| "eval_loss": 0.3829491138458252, | |
| "eval_precision": 0.7880252100840336, | |
| "eval_recall": 0.7908710674668122, | |
| "eval_runtime": 5.058, | |
| "eval_samples_per_second": 78.884, | |
| "eval_steps_per_second": 9.885, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 4.997393608093262, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.3771, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8395989974937343, | |
| "eval_f1": 0.8065102745953809, | |
| "eval_loss": 0.3785531520843506, | |
| "eval_precision": 0.8065102745953809, | |
| "eval_recall": 0.8065102745953809, | |
| "eval_runtime": 5.0614, | |
| "eval_samples_per_second": 78.831, | |
| "eval_steps_per_second": 9.879, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 3.873344659805298, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.3633, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8295739348370927, | |
| "eval_f1": 0.7992838965496833, | |
| "eval_loss": 0.3843457102775574, | |
| "eval_precision": 0.7931478693839741, | |
| "eval_recall": 0.8069194398981633, | |
| "eval_runtime": 5.0605, | |
| "eval_samples_per_second": 78.846, | |
| "eval_steps_per_second": 9.88, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 10.848413467407227, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3591, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8295739348370927, | |
| "eval_f1": 0.7992838965496833, | |
| "eval_loss": 0.3832751214504242, | |
| "eval_precision": 0.7931478693839741, | |
| "eval_recall": 0.8069194398981633, | |
| "eval_runtime": 5.0639, | |
| "eval_samples_per_second": 78.793, | |
| "eval_steps_per_second": 9.874, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.711672306060791, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.354, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8395989974937343, | |
| "eval_f1": 0.8065102745953809, | |
| "eval_loss": 0.3705191910266876, | |
| "eval_precision": 0.8065102745953809, | |
| "eval_recall": 0.8065102745953809, | |
| "eval_runtime": 5.0892, | |
| "eval_samples_per_second": 78.401, | |
| "eval_steps_per_second": 9.825, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 8.37232494354248, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.3451, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8370927318295739, | |
| "eval_f1": 0.8049369344976196, | |
| "eval_loss": 0.37085026502609253, | |
| "eval_precision": 0.8027777777777778, | |
| "eval_recall": 0.8072376795781051, | |
| "eval_runtime": 5.0497, | |
| "eval_samples_per_second": 79.014, | |
| "eval_steps_per_second": 9.902, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.1326128244400024, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.3403, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8320802005012531, | |
| "eval_f1": 0.8026866442779643, | |
| "eval_loss": 0.3732873201370239, | |
| "eval_precision": 0.795995733394834, | |
| "eval_recall": 0.8111929441716675, | |
| "eval_runtime": 5.0568, | |
| "eval_samples_per_second": 78.903, | |
| "eval_steps_per_second": 9.888, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 6.462644100189209, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3282, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8345864661654135, | |
| "eval_f1": 0.8060710498409331, | |
| "eval_loss": 0.3714647889137268, | |
| "eval_precision": 0.7988372093023256, | |
| "eval_recall": 0.8154664484451719, | |
| "eval_runtime": 5.1224, | |
| "eval_samples_per_second": 77.894, | |
| "eval_steps_per_second": 9.761, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 3.0335631370544434, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.3286, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8320802005012531, | |
| "eval_f1": 0.7999041923338897, | |
| "eval_loss": 0.36644819378852844, | |
| "eval_precision": 0.7965023376930815, | |
| "eval_recall": 0.803691580287325, | |
| "eval_runtime": 5.0626, | |
| "eval_samples_per_second": 78.813, | |
| "eval_steps_per_second": 9.876, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 2.8270199298858643, | |
| "learning_rate": 5e-06, | |
| "loss": 0.3348, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8270676691729323, | |
| "eval_f1": 0.7949075143216848, | |
| "eval_loss": 0.3669916093349457, | |
| "eval_precision": 0.7904105571847508, | |
| "eval_recall": 0.8001454809965449, | |
| "eval_runtime": 5.0702, | |
| "eval_samples_per_second": 78.695, | |
| "eval_steps_per_second": 9.861, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 4.275764465332031, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.325, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8320802005012531, | |
| "eval_f1": 0.801779557335113, | |
| "eval_loss": 0.36691346764564514, | |
| "eval_precision": 0.7960927960927962, | |
| "eval_recall": 0.8086924895435534, | |
| "eval_runtime": 5.0526, | |
| "eval_samples_per_second": 78.969, | |
| "eval_steps_per_second": 9.896, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 6.228915691375732, | |
| "learning_rate": 0.0, | |
| "loss": 0.3266, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8320802005012531, | |
| "eval_f1": 0.801779557335113, | |
| "eval_loss": 0.3671587407588959, | |
| "eval_precision": 0.7960927960927962, | |
| "eval_recall": 0.8086924895435534, | |
| "eval_runtime": 5.0972, | |
| "eval_samples_per_second": 78.279, | |
| "eval_steps_per_second": 9.809, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7590599775312000.0, | |
| "train_loss": 0.3947384412171411, | |
| "train_runtime": 1956.7085, | |
| "train_samples_per_second": 37.185, | |
| "train_steps_per_second": 1.247 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7590599775312000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |