| { | |
| "best_metric": 80.61162079510703, | |
| "best_model_checkpoint": "outputs/adapter/superglue-boolq/checkpoint-3835", | |
| "epoch": 20.0, | |
| "global_step": 5900, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 62.62996941896024, | |
| "eval_average_metrics": 62.62996941896024, | |
| "eval_loss": 0.2789694368839264, | |
| "eval_runtime": 4.4788, | |
| "eval_samples_per_second": 365.053, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00027457627118644066, | |
| "loss": 0.3259, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 68.31804281345566, | |
| "eval_average_metrics": 68.31804281345566, | |
| "eval_loss": 0.2598051428794861, | |
| "eval_runtime": 4.7399, | |
| "eval_samples_per_second": 344.945, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 74.6788990825688, | |
| "eval_average_metrics": 74.6788990825688, | |
| "eval_loss": 0.22243507206439972, | |
| "eval_runtime": 4.7583, | |
| "eval_samples_per_second": 343.612, | |
| "step": 885 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 0.00024915254237288135, | |
| "loss": 0.2455, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 74.6788990825688, | |
| "eval_average_metrics": 74.6788990825688, | |
| "eval_loss": 0.21567188203334808, | |
| "eval_runtime": 4.7503, | |
| "eval_samples_per_second": 344.185, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 79.32721712538226, | |
| "eval_average_metrics": 79.32721712538226, | |
| "eval_loss": 0.2073381394147873, | |
| "eval_runtime": 4.7337, | |
| "eval_samples_per_second": 345.395, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 0.000223728813559322, | |
| "loss": 0.1872, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 78.71559633027523, | |
| "eval_average_metrics": 78.71559633027523, | |
| "eval_loss": 0.207603320479393, | |
| "eval_runtime": 4.7412, | |
| "eval_samples_per_second": 344.851, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 0.0001983050847457627, | |
| "loss": 0.134, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 80.0, | |
| "eval_average_metrics": 80.0, | |
| "eval_loss": 0.22447817027568817, | |
| "eval_runtime": 4.7578, | |
| "eval_samples_per_second": 343.643, | |
| "step": 2065 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 78.83792048929664, | |
| "eval_average_metrics": 78.83792048929664, | |
| "eval_loss": 0.2642129361629486, | |
| "eval_runtime": 4.7785, | |
| "eval_samples_per_second": 342.156, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "learning_rate": 0.0001728813559322034, | |
| "loss": 0.0978, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 80.1223241590214, | |
| "eval_average_metrics": 80.1223241590214, | |
| "eval_loss": 0.26512306928634644, | |
| "eval_runtime": 4.7549, | |
| "eval_samples_per_second": 343.857, | |
| "step": 2655 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 79.69418960244649, | |
| "eval_average_metrics": 79.69418960244649, | |
| "eval_loss": 0.2741082012653351, | |
| "eval_runtime": 4.749, | |
| "eval_samples_per_second": 344.283, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 10.17, | |
| "learning_rate": 0.00014745762711864405, | |
| "loss": 0.0743, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 80.42813455657493, | |
| "eval_average_metrics": 80.42813455657493, | |
| "eval_loss": 0.31817376613616943, | |
| "eval_runtime": 4.7525, | |
| "eval_samples_per_second": 344.028, | |
| "step": 3245 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "learning_rate": 0.00012203389830508474, | |
| "loss": 0.0541, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 79.81651376146789, | |
| "eval_average_metrics": 79.81651376146789, | |
| "eval_loss": 0.33367282152175903, | |
| "eval_runtime": 4.7584, | |
| "eval_samples_per_second": 343.602, | |
| "step": 3540 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 80.61162079510703, | |
| "eval_average_metrics": 80.61162079510703, | |
| "eval_loss": 0.3976394534111023, | |
| "eval_runtime": 4.7841, | |
| "eval_samples_per_second": 341.758, | |
| "step": 3835 | |
| }, | |
| { | |
| "epoch": 13.56, | |
| "learning_rate": 9.661016949152541e-05, | |
| "loss": 0.0399, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 79.8776758409786, | |
| "eval_average_metrics": 79.8776758409786, | |
| "eval_loss": 0.41730642318725586, | |
| "eval_runtime": 4.7467, | |
| "eval_samples_per_second": 344.451, | |
| "step": 4130 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 80.18348623853211, | |
| "eval_average_metrics": 80.18348623853211, | |
| "eval_loss": 0.4511031210422516, | |
| "eval_runtime": 4.7544, | |
| "eval_samples_per_second": 343.89, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 15.25, | |
| "learning_rate": 7.11864406779661e-05, | |
| "loss": 0.0303, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 79.63302752293579, | |
| "eval_average_metrics": 79.63302752293579, | |
| "eval_loss": 0.4779604375362396, | |
| "eval_runtime": 4.7435, | |
| "eval_samples_per_second": 344.682, | |
| "step": 4720 | |
| }, | |
| { | |
| "epoch": 16.95, | |
| "learning_rate": 4.576271186440678e-05, | |
| "loss": 0.0234, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 80.18348623853211, | |
| "eval_average_metrics": 80.18348623853211, | |
| "eval_loss": 0.5195048451423645, | |
| "eval_runtime": 4.7472, | |
| "eval_samples_per_second": 344.414, | |
| "step": 5015 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 79.69418960244649, | |
| "eval_average_metrics": 79.69418960244649, | |
| "eval_loss": 0.5308667421340942, | |
| "eval_runtime": 4.7851, | |
| "eval_samples_per_second": 341.683, | |
| "step": 5310 | |
| }, | |
| { | |
| "epoch": 18.64, | |
| "learning_rate": 2.0338983050847455e-05, | |
| "loss": 0.019, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 80.06116207951071, | |
| "eval_average_metrics": 80.06116207951071, | |
| "eval_loss": 0.5453023910522461, | |
| "eval_runtime": 4.7413, | |
| "eval_samples_per_second": 344.844, | |
| "step": 5605 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 80.36697247706422, | |
| "eval_average_metrics": 80.36697247706422, | |
| "eval_loss": 0.5537198185920715, | |
| "eval_runtime": 4.7547, | |
| "eval_samples_per_second": 343.869, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 5900, | |
| "total_flos": 5.80271991662592e+16, | |
| "train_loss": 0.1054745042930215, | |
| "train_runtime": 1312.1786, | |
| "train_samples_per_second": 143.685, | |
| "train_steps_per_second": 4.496 | |
| } | |
| ], | |
| "max_steps": 5900, | |
| "num_train_epochs": 20, | |
| "total_flos": 5.80271991662592e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |