| { | |
| "best_metric": 62.27396685546466, | |
| "best_model_checkpoint": "outputs/bitfit/t5-base/superglue-multirc/checkpoint-2400", | |
| "epoch": 3.0, | |
| "global_step": 2556, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23, | |
| "eval_average_metrics": 58.45389071733288, | |
| "eval_em": 49.88986784140969, | |
| "eval_f1": 67.01791359325607, | |
| "eval_loss": 0.21679049730300903, | |
| "eval_runtime": 23.0565, | |
| "eval_samples_per_second": 105.133, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_average_metrics": 61.27063054507724, | |
| "eval_em": 51.98237885462555, | |
| "eval_f1": 70.55888223552893, | |
| "eval_loss": 0.2198609709739685, | |
| "eval_runtime": 23.4946, | |
| "eval_samples_per_second": 103.173, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.0002413145539906103, | |
| "loss": 0.2582, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_average_metrics": 59.16978670550705, | |
| "eval_em": 49.779735682819386, | |
| "eval_f1": 68.55983772819472, | |
| "eval_loss": 0.20913197100162506, | |
| "eval_runtime": 23.7161, | |
| "eval_samples_per_second": 102.209, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_average_metrics": 61.270356888999785, | |
| "eval_em": 51.43171806167401, | |
| "eval_f1": 71.10899571632555, | |
| "eval_loss": 0.20762182772159576, | |
| "eval_runtime": 26.3745, | |
| "eval_samples_per_second": 91.907, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.00018262910798122063, | |
| "loss": 0.1865, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_average_metrics": 61.45286279488806, | |
| "eval_em": 51.87224669603524, | |
| "eval_f1": 71.03347889374089, | |
| "eval_loss": 0.2089201658964157, | |
| "eval_runtime": 24.8833, | |
| "eval_samples_per_second": 97.415, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "eval_average_metrics": 61.79826184816632, | |
| "eval_em": 52.202643171806166, | |
| "eval_f1": 71.39388052452648, | |
| "eval_loss": 0.20319344103336334, | |
| "eval_runtime": 25.4876, | |
| "eval_samples_per_second": 95.105, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "eval_average_metrics": 62.13396621755072, | |
| "eval_em": 52.53303964757709, | |
| "eval_f1": 71.73489278752436, | |
| "eval_loss": 0.21759574115276337, | |
| "eval_runtime": 26.3526, | |
| "eval_samples_per_second": 91.983, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 0.00012394366197183098, | |
| "loss": 0.1745, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "eval_average_metrics": 62.159955413411296, | |
| "eval_em": 52.09251101321586, | |
| "eval_f1": 72.22739981360672, | |
| "eval_loss": 0.21236789226531982, | |
| "eval_runtime": 22.8791, | |
| "eval_samples_per_second": 105.948, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_average_metrics": 61.48864179183695, | |
| "eval_em": 51.651982378854626, | |
| "eval_f1": 71.32530120481927, | |
| "eval_loss": 0.2192794233560562, | |
| "eval_runtime": 23.8863, | |
| "eval_samples_per_second": 101.481, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 6.525821596244131e-05, | |
| "loss": 0.1692, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "eval_average_metrics": 62.06655934464116, | |
| "eval_em": 52.75330396475771, | |
| "eval_f1": 71.37981472452462, | |
| "eval_loss": 0.2168177217245102, | |
| "eval_runtime": 23.5217, | |
| "eval_samples_per_second": 103.054, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "eval_average_metrics": 62.263646102553764, | |
| "eval_em": 52.31277533039648, | |
| "eval_f1": 72.21451687471105, | |
| "eval_loss": 0.22297202050685883, | |
| "eval_runtime": 22.3978, | |
| "eval_samples_per_second": 108.225, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "eval_average_metrics": 62.27396685546466, | |
| "eval_em": 52.6431718061674, | |
| "eval_f1": 71.90476190476191, | |
| "eval_loss": 0.22205480933189392, | |
| "eval_runtime": 23.6577, | |
| "eval_samples_per_second": 102.461, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 6.5727699530516425e-06, | |
| "loss": 0.1589, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 2556, | |
| "total_flos": 2.511166406964019e+16, | |
| "train_loss": 0.18877839930180654, | |
| "train_runtime": 1478.3404, | |
| "train_samples_per_second": 55.284, | |
| "train_steps_per_second": 1.729 | |
| } | |
| ], | |
| "max_steps": 2556, | |
| "num_train_epochs": 3, | |
| "total_flos": 2.511166406964019e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |