| { | |
| "best_metric": 0.9173800019578199, | |
| "best_model_checkpoint": "./output//roberta-large_ipc1_E_5_32_5e-6_0.01_0.06_07-08-22_05-38/checkpoint-42000", | |
| "epoch": 0.8361998517645717, | |
| "global_step": 44000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 6.334726973267452e-07, | |
| "loss": 0.3829, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_accuracy": 0.7333631854505374, | |
| "eval_f1": 0.8072349594700899, | |
| "eval_loss": 0.1738697588443756, | |
| "eval_roc_auc": 0.8691996632382517, | |
| "eval_runtime": 2556.9417, | |
| "eval_samples_per_second": 34.063, | |
| "eval_steps_per_second": 2.129, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.2669453946534904e-06, | |
| "loss": 0.152, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_accuracy": 0.8074768072012491, | |
| "eval_f1": 0.8585274784241828, | |
| "eval_loss": 0.12472031265497208, | |
| "eval_roc_auc": 0.9038230127982044, | |
| "eval_runtime": 2557.151, | |
| "eval_samples_per_second": 34.06, | |
| "eval_steps_per_second": 2.129, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.900418091980236e-06, | |
| "loss": 0.1207, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_accuracy": 0.8247680720124919, | |
| "eval_f1": 0.8746601834456846, | |
| "eval_loss": 0.10768114030361176, | |
| "eval_roc_auc": 0.9167191830505206, | |
| "eval_runtime": 2559.7707, | |
| "eval_samples_per_second": 34.025, | |
| "eval_steps_per_second": 2.127, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 2.5338907893069808e-06, | |
| "loss": 0.1068, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_accuracy": 0.8348489023606136, | |
| "eval_f1": 0.8825789946684457, | |
| "eval_loss": 0.09896202385425568, | |
| "eval_roc_auc": 0.9219814574886337, | |
| "eval_runtime": 2560.9139, | |
| "eval_samples_per_second": 34.01, | |
| "eval_steps_per_second": 2.126, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 3.1673634866337266e-06, | |
| "loss": 0.1011, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_accuracy": 0.8379948562505741, | |
| "eval_f1": 0.8862482315394654, | |
| "eval_loss": 0.0956236720085144, | |
| "eval_roc_auc": 0.9259124171373225, | |
| "eval_runtime": 2559.6669, | |
| "eval_samples_per_second": 34.026, | |
| "eval_steps_per_second": 2.127, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 3.800836183960472e-06, | |
| "loss": 0.0947, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eval_accuracy": 0.8484890236061358, | |
| "eval_f1": 0.8952724037759265, | |
| "eval_loss": 0.08779484033584595, | |
| "eval_roc_auc": 0.9306411069461392, | |
| "eval_runtime": 2561.9326, | |
| "eval_samples_per_second": 33.996, | |
| "eval_steps_per_second": 2.125, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.434308881287217e-06, | |
| "loss": 0.0915, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_accuracy": 0.849901258381556, | |
| "eval_f1": 0.8959137400012078, | |
| "eval_loss": 0.08673737943172455, | |
| "eval_roc_auc": 0.9310974491642802, | |
| "eval_runtime": 2561.6246, | |
| "eval_samples_per_second": 34.0, | |
| "eval_steps_per_second": 2.125, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 4.995673428787469e-06, | |
| "loss": 0.0878, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_accuracy": 0.8544020391292367, | |
| "eval_f1": 0.8991452427312628, | |
| "eval_loss": 0.08379313349723816, | |
| "eval_roc_auc": 0.9323214390537408, | |
| "eval_runtime": 2562.3097, | |
| "eval_samples_per_second": 33.991, | |
| "eval_steps_per_second": 2.125, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.9552381838105365e-06, | |
| "loss": 0.0865, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_accuracy": 0.8521631303389363, | |
| "eval_f1": 0.8989961118555455, | |
| "eval_loss": 0.08386027067899704, | |
| "eval_roc_auc": 0.9327737350993932, | |
| "eval_runtime": 2560.8339, | |
| "eval_samples_per_second": 34.011, | |
| "eval_steps_per_second": 2.126, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 4.914802938833605e-06, | |
| "loss": 0.0835, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_accuracy": 0.8534605492789565, | |
| "eval_f1": 0.9011981436136274, | |
| "eval_loss": 0.08270147442817688, | |
| "eval_roc_auc": 0.9364532450443123, | |
| "eval_runtime": 2562.5704, | |
| "eval_samples_per_second": 33.988, | |
| "eval_steps_per_second": 2.124, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 4.874367693856674e-06, | |
| "loss": 0.0813, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_accuracy": 0.8624391476072379, | |
| "eval_f1": 0.9062087330430322, | |
| "eval_loss": 0.07851053774356842, | |
| "eval_roc_auc": 0.9391581637644638, | |
| "eval_runtime": 2562.0501, | |
| "eval_samples_per_second": 33.995, | |
| "eval_steps_per_second": 2.125, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.8339324488797415e-06, | |
| "loss": 0.0799, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_accuracy": 0.8644139799761184, | |
| "eval_f1": 0.9090770321103419, | |
| "eval_loss": 0.07563477754592896, | |
| "eval_roc_auc": 0.9411536483338017, | |
| "eval_runtime": 2560.4534, | |
| "eval_samples_per_second": 34.016, | |
| "eval_steps_per_second": 2.126, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.79349720390281e-06, | |
| "loss": 0.0792, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_accuracy": 0.8651832460732984, | |
| "eval_f1": 0.9101030478163677, | |
| "eval_loss": 0.07443315535783768, | |
| "eval_roc_auc": 0.9417119100729429, | |
| "eval_runtime": 2562.0577, | |
| "eval_samples_per_second": 33.995, | |
| "eval_steps_per_second": 2.125, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.753061958925879e-06, | |
| "loss": 0.0764, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_accuracy": 0.8657573252502985, | |
| "eval_f1": 0.9101674732621782, | |
| "eval_loss": 0.07442668080329895, | |
| "eval_roc_auc": 0.9411051257108755, | |
| "eval_runtime": 2562.7135, | |
| "eval_samples_per_second": 33.986, | |
| "eval_steps_per_second": 2.124, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.7126267139489465e-06, | |
| "loss": 0.0758, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_accuracy": 0.8675369706989988, | |
| "eval_f1": 0.9117887285597605, | |
| "eval_loss": 0.0741124227643013, | |
| "eval_roc_auc": 0.9431533957100788, | |
| "eval_runtime": 2561.7593, | |
| "eval_samples_per_second": 33.999, | |
| "eval_steps_per_second": 2.125, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 4.672191468972015e-06, | |
| "loss": 0.075, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_accuracy": 0.8696840268209791, | |
| "eval_f1": 0.9111988941522289, | |
| "eval_loss": 0.07489842176437378, | |
| "eval_roc_auc": 0.941613439232639, | |
| "eval_runtime": 2562.3874, | |
| "eval_samples_per_second": 33.99, | |
| "eval_steps_per_second": 2.125, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 4.631756223995084e-06, | |
| "loss": 0.0744, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_accuracy": 0.8696610636538992, | |
| "eval_f1": 0.9132626153335907, | |
| "eval_loss": 0.07311609387397766, | |
| "eval_roc_auc": 0.944537581534918, | |
| "eval_runtime": 2561.9943, | |
| "eval_samples_per_second": 33.995, | |
| "eval_steps_per_second": 2.125, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 4.5913209790181515e-06, | |
| "loss": 0.0741, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_accuracy": 0.8714177459355195, | |
| "eval_f1": 0.9147231870702489, | |
| "eval_loss": 0.07061254978179932, | |
| "eval_roc_auc": 0.9443614210381402, | |
| "eval_runtime": 2562.603, | |
| "eval_samples_per_second": 33.987, | |
| "eval_steps_per_second": 2.124, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 4.55088573404122e-06, | |
| "loss": 0.0726, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_accuracy": 0.8722903462845596, | |
| "eval_f1": 0.9149092652418596, | |
| "eval_loss": 0.07059279829263687, | |
| "eval_roc_auc": 0.9447903799628505, | |
| "eval_runtime": 2562.4912, | |
| "eval_samples_per_second": 33.989, | |
| "eval_steps_per_second": 2.124, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 4.510450489064289e-06, | |
| "loss": 0.0728, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_accuracy": 0.8742766602369799, | |
| "eval_f1": 0.9169329073482427, | |
| "eval_loss": 0.06948799639940262, | |
| "eval_roc_auc": 0.9469121100055765, | |
| "eval_runtime": 2562.6751, | |
| "eval_samples_per_second": 33.986, | |
| "eval_steps_per_second": 2.124, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.470015244087356e-06, | |
| "loss": 0.071, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_accuracy": 0.87463258932672, | |
| "eval_f1": 0.9173800019578199, | |
| "eval_loss": 0.06844224780797958, | |
| "eval_roc_auc": 0.9468508740850688, | |
| "eval_runtime": 2572.6023, | |
| "eval_samples_per_second": 33.855, | |
| "eval_steps_per_second": 2.116, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 4.429579999110425e-06, | |
| "loss": 0.0713, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_accuracy": 0.8746555524937999, | |
| "eval_f1": 0.9166339527174387, | |
| "eval_loss": 0.06880520284175873, | |
| "eval_roc_auc": 0.9455051967809519, | |
| "eval_runtime": 2562.466, | |
| "eval_samples_per_second": 33.989, | |
| "eval_steps_per_second": 2.125, | |
| "step": 44000 | |
| } | |
| ], | |
| "max_steps": 263095, | |
| "num_train_epochs": 5, | |
| "total_flos": 1.312181527117824e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |