| { | |
| "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": 5.168577671051025, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5663, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7293233082706767, | |
| "eval_f1": 0.6572164948453608, | |
| "eval_loss": 0.5216349363327026, | |
| "eval_precision": 0.6677272727272727, | |
| "eval_recall": 0.6509819967266777, | |
| "eval_runtime": 4.6427, | |
| "eval_samples_per_second": 85.941, | |
| "eval_steps_per_second": 10.769, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.071716785430908, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5149, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7243107769423559, | |
| "eval_f1": 0.681013431013431, | |
| "eval_loss": 0.5133972764015198, | |
| "eval_precision": 0.6758417508417509, | |
| "eval_recall": 0.6899436261138389, | |
| "eval_runtime": 5.1009, | |
| "eval_samples_per_second": 78.221, | |
| "eval_steps_per_second": 9.802, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.960922956466675, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4925, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7568922305764411, | |
| "eval_f1": 0.7014049083067808, | |
| "eval_loss": 0.48209038376808167, | |
| "eval_precision": 0.70548914753067, | |
| "eval_recall": 0.6979905437352246, | |
| "eval_runtime": 5.0731, | |
| "eval_samples_per_second": 78.65, | |
| "eval_steps_per_second": 9.856, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.4121310710906982, | |
| "learning_rate": 4e-05, | |
| "loss": 0.4608, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7644110275689223, | |
| "eval_f1": 0.7114035087719299, | |
| "eval_loss": 0.46544694900512695, | |
| "eval_precision": 0.7149962462462462, | |
| "eval_recall": 0.7083106019276232, | |
| "eval_runtime": 5.1025, | |
| "eval_samples_per_second": 78.197, | |
| "eval_steps_per_second": 9.799, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.662163734436035, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.4493, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7568922305764411, | |
| "eval_f1": 0.7193095695688436, | |
| "eval_loss": 0.46004626154899597, | |
| "eval_precision": 0.7126202191903377, | |
| "eval_recall": 0.7304964539007093, | |
| "eval_runtime": 5.0734, | |
| "eval_samples_per_second": 78.645, | |
| "eval_steps_per_second": 9.855, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.3096864223480225, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.4257, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7869674185463659, | |
| "eval_f1": 0.7369400814426992, | |
| "eval_loss": 0.4306720495223999, | |
| "eval_precision": 0.7433180603283696, | |
| "eval_recall": 0.7317694126204765, | |
| "eval_runtime": 5.1042, | |
| "eval_samples_per_second": 78.17, | |
| "eval_steps_per_second": 9.796, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.605820894241333, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.4178, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7969924812030075, | |
| "eval_f1": 0.7580931280454488, | |
| "eval_loss": 0.4180676341056824, | |
| "eval_precision": 0.7551784340415459, | |
| "eval_recall": 0.7613657028550646, | |
| "eval_runtime": 5.0532, | |
| "eval_samples_per_second": 78.959, | |
| "eval_steps_per_second": 9.895, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 5.129829406738281, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3977, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8070175438596491, | |
| "eval_f1": 0.7616986620127981, | |
| "eval_loss": 0.3971773684024811, | |
| "eval_precision": 0.7687094310805651, | |
| "eval_recall": 0.7559556282960538, | |
| "eval_runtime": 5.047, | |
| "eval_samples_per_second": 79.057, | |
| "eval_steps_per_second": 9.907, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 5.125335693359375, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.3946, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8145363408521303, | |
| "eval_f1": 0.7715978711553932, | |
| "eval_loss": 0.3936789035797119, | |
| "eval_precision": 0.7779341980385954, | |
| "eval_recall": 0.7662756864884525, | |
| "eval_runtime": 5.0565, | |
| "eval_samples_per_second": 78.909, | |
| "eval_steps_per_second": 9.888, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 9.01931095123291, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.3762, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8245614035087719, | |
| "eval_f1": 0.7738095238095237, | |
| "eval_loss": 0.38743722438812256, | |
| "eval_precision": 0.7995467004005903, | |
| "eval_recall": 0.7583651573013275, | |
| "eval_runtime": 5.1497, | |
| "eval_samples_per_second": 77.481, | |
| "eval_steps_per_second": 9.709, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 4.317044734954834, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.3727, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8320802005012531, | |
| "eval_f1": 0.7915149151881459, | |
| "eval_loss": 0.3787190616130829, | |
| "eval_precision": 0.801371627277996, | |
| "eval_recall": 0.7836879432624113, | |
| "eval_runtime": 5.0708, | |
| "eval_samples_per_second": 78.687, | |
| "eval_steps_per_second": 9.86, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 9.624013900756836, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3626, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8370927318295739, | |
| "eval_f1": 0.7999105055663995, | |
| "eval_loss": 0.37497109174728394, | |
| "eval_precision": 0.8058980811575966, | |
| "eval_recall": 0.794735406437534, | |
| "eval_runtime": 5.0526, | |
| "eval_samples_per_second": 78.97, | |
| "eval_steps_per_second": 9.896, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 4.469719886779785, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.359, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8295739348370927, | |
| "eval_f1": 0.7802721088435374, | |
| "eval_loss": 0.37280407547950745, | |
| "eval_precision": 0.8065570314147164, | |
| "eval_recall": 0.7644117112202218, | |
| "eval_runtime": 5.0466, | |
| "eval_samples_per_second": 79.063, | |
| "eval_steps_per_second": 9.908, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 6.554767608642578, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.3488, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8295739348370927, | |
| "eval_f1": 0.7816021636937343, | |
| "eval_loss": 0.370919793844223, | |
| "eval_precision": 0.8049342105263158, | |
| "eval_recall": 0.7669121658483361, | |
| "eval_runtime": 5.0529, | |
| "eval_samples_per_second": 78.965, | |
| "eval_steps_per_second": 9.895, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.5811327695846558, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.3445, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_f1": 0.8050261780104713, | |
| "eval_loss": 0.3667002022266388, | |
| "eval_precision": 0.8131443298969072, | |
| "eval_recall": 0.7982815057283142, | |
| "eval_runtime": 5.0467, | |
| "eval_samples_per_second": 79.061, | |
| "eval_steps_per_second": 9.907, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 5.522835731506348, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3344, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_f1": 0.8039617859231819, | |
| "eval_loss": 0.3655822277069092, | |
| "eval_precision": 0.8142185588254234, | |
| "eval_recall": 0.7957810511002, | |
| "eval_runtime": 5.05, | |
| "eval_samples_per_second": 79.01, | |
| "eval_steps_per_second": 9.901, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 4.577361583709717, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.3339, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8395989974937343, | |
| "eval_f1": 0.7991821327461466, | |
| "eval_loss": 0.36540165543556213, | |
| "eval_precision": 0.8127623983206507, | |
| "eval_recall": 0.7890070921985816, | |
| "eval_runtime": 5.0524, | |
| "eval_samples_per_second": 78.973, | |
| "eval_steps_per_second": 9.896, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 2.5822412967681885, | |
| "learning_rate": 5e-06, | |
| "loss": 0.3357, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_f1": 0.8028733423778791, | |
| "eval_loss": 0.36383578181266785, | |
| "eval_precision": 0.8154009126466754, | |
| "eval_recall": 0.7932805964720859, | |
| "eval_runtime": 5.0893, | |
| "eval_samples_per_second": 78.399, | |
| "eval_steps_per_second": 9.824, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 4.345070838928223, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.3357, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_f1": 0.8028733423778791, | |
| "eval_loss": 0.36459243297576904, | |
| "eval_precision": 0.8154009126466754, | |
| "eval_recall": 0.7932805964720859, | |
| "eval_runtime": 5.0505, | |
| "eval_samples_per_second": 79.002, | |
| "eval_steps_per_second": 9.9, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 6.21327018737793, | |
| "learning_rate": 0.0, | |
| "loss": 0.3359, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8446115288220551, | |
| "eval_f1": 0.8054576910978295, | |
| "eval_loss": 0.3638042211532593, | |
| "eval_precision": 0.8193059564418788, | |
| "eval_recall": 0.795053646117476, | |
| "eval_runtime": 5.0736, | |
| "eval_samples_per_second": 78.643, | |
| "eval_steps_per_second": 9.855, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7590599775312000.0, | |
| "train_loss": 0.39794730983796667, | |
| "train_runtime": 1936.5815, | |
| "train_samples_per_second": 37.571, | |
| "train_steps_per_second": 1.26 | |
| } | |
| ], | |
| "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 | |
| } | |