| { | |
| "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": 4.279976844787598, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.564, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7142857142857143, | |
| "eval_f1": 0.6246286393345217, | |
| "eval_loss": 0.520961344242096, | |
| "eval_precision": 0.6432360742705571, | |
| "eval_recall": 0.6178396072013094, | |
| "eval_runtime": 1.7251, | |
| "eval_samples_per_second": 231.293, | |
| "eval_steps_per_second": 28.984, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.9077770709991455, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5007, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7518796992481203, | |
| "eval_f1": 0.7122952431589911, | |
| "eval_loss": 0.4797453284263611, | |
| "eval_precision": 0.7062235989862011, | |
| "eval_recall": 0.7219494453537008, | |
| "eval_runtime": 1.727, | |
| "eval_samples_per_second": 231.043, | |
| "eval_steps_per_second": 28.953, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.219292640686035, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.428, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8245614035087719, | |
| "eval_f1": 0.7933804817423211, | |
| "eval_loss": 0.39089536666870117, | |
| "eval_precision": 0.7873726262158407, | |
| "eval_recall": 0.800872885979269, | |
| "eval_runtime": 1.7391, | |
| "eval_samples_per_second": 229.423, | |
| "eval_steps_per_second": 28.75, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 5.223276615142822, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3751, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8471177944862155, | |
| "eval_f1": 0.8151164115613962, | |
| "eval_loss": 0.34778618812561035, | |
| "eval_precision": 0.8159193371512123, | |
| "eval_recall": 0.8143298781596654, | |
| "eval_runtime": 1.7324, | |
| "eval_samples_per_second": 230.322, | |
| "eval_steps_per_second": 28.862, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 5.300344467163086, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.339, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8546365914786967, | |
| "eval_f1": 0.8280070158749033, | |
| "eval_loss": 0.33690375089645386, | |
| "eval_precision": 0.8223795620437956, | |
| "eval_recall": 0.8346517548645208, | |
| "eval_runtime": 1.7412, | |
| "eval_samples_per_second": 229.147, | |
| "eval_steps_per_second": 28.715, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 6.1557698249816895, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.3096, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.844327731092437, | |
| "eval_loss": 0.3206353187561035, | |
| "eval_precision": 0.8411320530352577, | |
| "eval_recall": 0.8477905073649754, | |
| "eval_runtime": 1.7364, | |
| "eval_samples_per_second": 229.787, | |
| "eval_steps_per_second": 28.795, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.458906412124634, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2931, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8306935047100303, | |
| "eval_loss": 0.31404000520706177, | |
| "eval_precision": 0.8372758729160114, | |
| "eval_recall": 0.8249681760320058, | |
| "eval_runtime": 1.74, | |
| "eval_samples_per_second": 229.312, | |
| "eval_steps_per_second": 28.736, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.627723217010498, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2765, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8461962888779714, | |
| "eval_loss": 0.30450862646102905, | |
| "eval_precision": 0.8453465227094517, | |
| "eval_recall": 0.8470631023822512, | |
| "eval_runtime": 1.739, | |
| "eval_samples_per_second": 229.443, | |
| "eval_steps_per_second": 28.752, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 3.7235734462738037, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2637, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8556004584112431, | |
| "eval_loss": 0.3003471791744232, | |
| "eval_precision": 0.8538865546218487, | |
| "eval_recall": 0.85738316057465, | |
| "eval_runtime": 1.7315, | |
| "eval_samples_per_second": 230.44, | |
| "eval_steps_per_second": 28.877, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 5.832447052001953, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2601, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8548827059465357, | |
| "eval_loss": 0.2909964919090271, | |
| "eval_precision": 0.8548827059465357, | |
| "eval_recall": 0.8548827059465357, | |
| "eval_runtime": 1.7405, | |
| "eval_samples_per_second": 229.242, | |
| "eval_steps_per_second": 28.727, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 12.791139602661133, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.2547, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8897243107769424, | |
| "eval_f1": 0.8641933287950987, | |
| "eval_loss": 0.28499358892440796, | |
| "eval_precision": 0.872603606453654, | |
| "eval_recall": 0.8569739952718676, | |
| "eval_runtime": 1.7358, | |
| "eval_samples_per_second": 229.86, | |
| "eval_steps_per_second": 28.804, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 9.72183895111084, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2426, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8696722245432793, | |
| "eval_loss": 0.279752641916275, | |
| "eval_precision": 0.8706135006701596, | |
| "eval_recall": 0.8687488634297145, | |
| "eval_runtime": 1.7375, | |
| "eval_samples_per_second": 229.634, | |
| "eval_steps_per_second": 28.776, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 2.6714916229248047, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.2319, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8661961395983623, | |
| "eval_loss": 0.28113481402397156, | |
| "eval_precision": 0.8784532165625604, | |
| "eval_recall": 0.8562465902891435, | |
| "eval_runtime": 1.7421, | |
| "eval_samples_per_second": 229.035, | |
| "eval_steps_per_second": 28.701, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 7.962621688842773, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.2359, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8609292598654301, | |
| "eval_loss": 0.2719880938529968, | |
| "eval_precision": 0.8609292598654301, | |
| "eval_recall": 0.8609292598654301, | |
| "eval_runtime": 1.7325, | |
| "eval_samples_per_second": 230.305, | |
| "eval_steps_per_second": 28.86, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 2.756800413131714, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2229, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_f1": 0.8736504011098378, | |
| "eval_loss": 0.2721886932849884, | |
| "eval_precision": 0.8718487394957983, | |
| "eval_recall": 0.8755228223313329, | |
| "eval_runtime": 1.7352, | |
| "eval_samples_per_second": 229.939, | |
| "eval_steps_per_second": 28.814, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 3.3028435707092285, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2218, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8649563392675828, | |
| "eval_loss": 0.2730589210987091, | |
| "eval_precision": 0.8623655913978494, | |
| "eval_recall": 0.8677032187670486, | |
| "eval_runtime": 1.7334, | |
| "eval_samples_per_second": 230.181, | |
| "eval_steps_per_second": 28.845, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 1.6590169668197632, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.2174, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8696722245432793, | |
| "eval_loss": 0.27383190393447876, | |
| "eval_precision": 0.8706135006701596, | |
| "eval_recall": 0.8687488634297145, | |
| "eval_runtime": 1.738, | |
| "eval_samples_per_second": 229.575, | |
| "eval_steps_per_second": 28.769, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 3.2556533813476562, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2165, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8703223612108386, | |
| "eval_loss": 0.27394846081733704, | |
| "eval_precision": 0.8694131129742446, | |
| "eval_recall": 0.8712493180578287, | |
| "eval_runtime": 1.738, | |
| "eval_samples_per_second": 229.571, | |
| "eval_steps_per_second": 28.768, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 7.895025253295898, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2153, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8972431077694235, | |
| "eval_f1": 0.8751002084335417, | |
| "eval_loss": 0.2726791203022003, | |
| "eval_precision": 0.8780701754385964, | |
| "eval_recall": 0.8722949627204946, | |
| "eval_runtime": 1.7412, | |
| "eval_samples_per_second": 229.156, | |
| "eval_steps_per_second": 28.716, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 4.127798080444336, | |
| "learning_rate": 0.0, | |
| "loss": 0.2159, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_f1": 0.8717238211879976, | |
| "eval_loss": 0.2725962698459625, | |
| "eval_precision": 0.8757194133300328, | |
| "eval_recall": 0.8680214584469903, | |
| "eval_runtime": 1.7378, | |
| "eval_samples_per_second": 229.603, | |
| "eval_steps_per_second": 28.772, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7635661149264000.0, | |
| "train_loss": 0.29423871978384547, | |
| "train_runtime": 626.8339, | |
| "train_samples_per_second": 116.075, | |
| "train_steps_per_second": 3.893 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7635661149264000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |