| { | |
| "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": 4.92190408706665, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5593, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.731829573934837, | |
| "eval_f1": 0.6531872202500386, | |
| "eval_loss": 0.5129385590553284, | |
| "eval_precision": 0.6696820563659538, | |
| "eval_recall": 0.6452536824877251, | |
| "eval_runtime": 1.7112, | |
| "eval_samples_per_second": 233.172, | |
| "eval_steps_per_second": 29.22, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.805196285247803, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.481, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7343358395989975, | |
| "eval_f1": 0.7054243048977545, | |
| "eval_loss": 0.4831171929836273, | |
| "eval_precision": 0.6993341053850608, | |
| "eval_recall": 0.7295417348608838, | |
| "eval_runtime": 1.7107, | |
| "eval_samples_per_second": 233.24, | |
| "eval_steps_per_second": 29.228, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.332752227783203, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4234, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8220551378446115, | |
| "eval_f1": 0.7739917826798037, | |
| "eval_loss": 0.3973781168460846, | |
| "eval_precision": 0.7925961082107262, | |
| "eval_recall": 0.7615930169121659, | |
| "eval_runtime": 1.7083, | |
| "eval_samples_per_second": 233.57, | |
| "eval_steps_per_second": 29.269, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.6373705863952637, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3701, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8395989974937343, | |
| "eval_f1": 0.7991821327461466, | |
| "eval_loss": 0.3780345618724823, | |
| "eval_precision": 0.8127623983206507, | |
| "eval_recall": 0.7890070921985816, | |
| "eval_runtime": 1.708, | |
| "eval_samples_per_second": 233.603, | |
| "eval_steps_per_second": 29.274, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 8.633732795715332, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.3499, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8471177944862155, | |
| "eval_f1": 0.819530641752864, | |
| "eval_loss": 0.3612293004989624, | |
| "eval_precision": 0.8134920634920635, | |
| "eval_recall": 0.8268321513002364, | |
| "eval_runtime": 1.706, | |
| "eval_samples_per_second": 233.875, | |
| "eval_steps_per_second": 29.308, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.133713960647583, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.3165, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8270676691729323, | |
| "eval_f1": 0.8072288313015957, | |
| "eval_loss": 0.3760314881801605, | |
| "eval_precision": 0.7952690166975882, | |
| "eval_recall": 0.8376523004182579, | |
| "eval_runtime": 1.7076, | |
| "eval_samples_per_second": 233.655, | |
| "eval_steps_per_second": 29.28, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.4740520715713501, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2968, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8419946387230413, | |
| "eval_loss": 0.33418920636177063, | |
| "eval_precision": 0.8437691365584814, | |
| "eval_recall": 0.8402891434806329, | |
| "eval_runtime": 1.708, | |
| "eval_samples_per_second": 233.608, | |
| "eval_steps_per_second": 29.274, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 5.327579021453857, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2812, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8350789627607721, | |
| "eval_loss": 0.3310687243938446, | |
| "eval_precision": 0.8463358876939919, | |
| "eval_recall": 0.8260138206946717, | |
| "eval_runtime": 1.707, | |
| "eval_samples_per_second": 233.739, | |
| "eval_steps_per_second": 29.291, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 8.512028694152832, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2682, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8454251965513313, | |
| "eval_loss": 0.32693037390708923, | |
| "eval_precision": 0.8463049835506276, | |
| "eval_recall": 0.8445626477541371, | |
| "eval_runtime": 1.7093, | |
| "eval_samples_per_second": 233.423, | |
| "eval_steps_per_second": 29.251, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 5.219644546508789, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2596, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8541488972828073, | |
| "eval_loss": 0.3144831955432892, | |
| "eval_precision": 0.8559859154929578, | |
| "eval_recall": 0.8523822513184216, | |
| "eval_runtime": 1.7104, | |
| "eval_samples_per_second": 233.278, | |
| "eval_steps_per_second": 29.233, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 6.501763820648193, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.2464, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8377439939939939, | |
| "eval_loss": 0.31378793716430664, | |
| "eval_precision": 0.8503401360544218, | |
| "eval_recall": 0.8277868703400618, | |
| "eval_runtime": 1.7091, | |
| "eval_samples_per_second": 233.452, | |
| "eval_steps_per_second": 29.255, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 12.380316734313965, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2415, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8564658408408408, | |
| "eval_loss": 0.31256482005119324, | |
| "eval_precision": 0.8697278911564625, | |
| "eval_recall": 0.8459265320967448, | |
| "eval_runtime": 1.71, | |
| "eval_samples_per_second": 233.331, | |
| "eval_steps_per_second": 29.239, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 4.9941325187683105, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.2354, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8521068445832446, | |
| "eval_loss": 0.3136414587497711, | |
| "eval_precision": 0.8693800752624282, | |
| "eval_recall": 0.8391525731951264, | |
| "eval_runtime": 1.7127, | |
| "eval_samples_per_second": 232.961, | |
| "eval_steps_per_second": 29.193, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 12.018842697143555, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.2303, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8510304659498208, | |
| "eval_loss": 0.3171806037425995, | |
| "eval_precision": 0.8463237893248498, | |
| "eval_recall": 0.8563375159119839, | |
| "eval_runtime": 1.711, | |
| "eval_samples_per_second": 233.193, | |
| "eval_steps_per_second": 29.222, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 3.3128843307495117, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2172, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8537492688633261, | |
| "eval_loss": 0.3120233416557312, | |
| "eval_precision": 0.8656062850151329, | |
| "eval_recall": 0.8441534824513548, | |
| "eval_runtime": 1.7092, | |
| "eval_samples_per_second": 233.436, | |
| "eval_steps_per_second": 29.253, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 5.912248611450195, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2159, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8357422474382676, | |
| "eval_loss": 0.31162190437316895, | |
| "eval_precision": 0.8319228265372551, | |
| "eval_recall": 0.8399709038006911, | |
| "eval_runtime": 1.7092, | |
| "eval_samples_per_second": 233.439, | |
| "eval_steps_per_second": 29.253, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 8.530476570129395, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.2192, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8556621579112929, | |
| "eval_loss": 0.31227871775627136, | |
| "eval_precision": 0.871654421411703, | |
| "eval_recall": 0.8434260774686306, | |
| "eval_runtime": 1.7098, | |
| "eval_samples_per_second": 233.357, | |
| "eval_steps_per_second": 29.243, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 7.491436958312988, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2124, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.839868600986979, | |
| "eval_loss": 0.31498095393180847, | |
| "eval_precision": 0.834029197080292, | |
| "eval_recall": 0.8467448627023095, | |
| "eval_runtime": 1.7079, | |
| "eval_samples_per_second": 233.62, | |
| "eval_steps_per_second": 29.276, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 5.7076263427734375, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2077, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8526315789473684, | |
| "eval_loss": 0.3084200918674469, | |
| "eval_precision": 0.8585304054054055, | |
| "eval_recall": 0.8473813420621932, | |
| "eval_runtime": 1.7079, | |
| "eval_samples_per_second": 233.62, | |
| "eval_steps_per_second": 29.276, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 6.0706095695495605, | |
| "learning_rate": 0.0, | |
| "loss": 0.205, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8533986527862829, | |
| "eval_loss": 0.3077850043773651, | |
| "eval_precision": 0.8572003218020917, | |
| "eval_recall": 0.8498817966903074, | |
| "eval_runtime": 1.7073, | |
| "eval_samples_per_second": 233.696, | |
| "eval_steps_per_second": 29.285, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7662265464912000.0, | |
| "train_loss": 0.2918538148285913, | |
| "train_runtime": 616.0393, | |
| "train_samples_per_second": 118.109, | |
| "train_steps_per_second": 3.961 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7662265464912000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |