| { | |
| "best_metric": 80.30581039755351, | |
| "best_model_checkpoint": "outputs/bitfit/t5-base/superglue-boolq/checkpoint-2600", | |
| "epoch": 20.0, | |
| "global_step": 5900, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.68, | |
| "eval_accuracy": 74.6788990825688, | |
| "eval_average_metrics": 74.6788990825688, | |
| "eval_loss": 0.21743574738502502, | |
| "eval_runtime": 16.9831, | |
| "eval_samples_per_second": 96.272, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_accuracy": 75.71865443425077, | |
| "eval_average_metrics": 75.71865443425077, | |
| "eval_loss": 0.21506452560424805, | |
| "eval_runtime": 14.6109, | |
| "eval_samples_per_second": 111.903, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00027457627118644066, | |
| "loss": 0.374, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "eval_accuracy": 77.18654434250764, | |
| "eval_average_metrics": 77.18654434250764, | |
| "eval_loss": 0.20255930721759796, | |
| "eval_runtime": 17.1645, | |
| "eval_samples_per_second": 95.254, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "eval_accuracy": 76.5137614678899, | |
| "eval_average_metrics": 76.5137614678899, | |
| "eval_loss": 0.220754012465477, | |
| "eval_runtime": 17.1929, | |
| "eval_samples_per_second": 95.097, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 0.00024915254237288135, | |
| "loss": 0.2114, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "eval_accuracy": 78.1651376146789, | |
| "eval_average_metrics": 78.1651376146789, | |
| "eval_loss": 0.19822736084461212, | |
| "eval_runtime": 14.5865, | |
| "eval_samples_per_second": 112.09, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "eval_accuracy": 78.71559633027523, | |
| "eval_average_metrics": 78.71559633027523, | |
| "eval_loss": 0.19769148528575897, | |
| "eval_runtime": 17.1866, | |
| "eval_samples_per_second": 95.132, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "eval_accuracy": 78.2262996941896, | |
| "eval_average_metrics": 78.2262996941896, | |
| "eval_loss": 0.1974276602268219, | |
| "eval_runtime": 17.1722, | |
| "eval_samples_per_second": 95.212, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 0.000223728813559322, | |
| "loss": 0.1997, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.42, | |
| "eval_accuracy": 78.71559633027523, | |
| "eval_average_metrics": 78.71559633027523, | |
| "eval_loss": 0.19954562187194824, | |
| "eval_runtime": 14.5679, | |
| "eval_samples_per_second": 112.233, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "eval_accuracy": 78.2262996941896, | |
| "eval_average_metrics": 78.2262996941896, | |
| "eval_loss": 0.20299072563648224, | |
| "eval_runtime": 17.2624, | |
| "eval_samples_per_second": 94.714, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 0.0001983050847457627, | |
| "loss": 0.1934, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "eval_accuracy": 79.63302752293579, | |
| "eval_average_metrics": 79.63302752293579, | |
| "eval_loss": 0.19578830897808075, | |
| "eval_runtime": 17.2808, | |
| "eval_samples_per_second": 94.614, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.46, | |
| "eval_accuracy": 79.26605504587157, | |
| "eval_average_metrics": 79.26605504587157, | |
| "eval_loss": 0.1937599629163742, | |
| "eval_runtime": 15.3794, | |
| "eval_samples_per_second": 106.311, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 8.14, | |
| "eval_accuracy": 77.92048929663609, | |
| "eval_average_metrics": 77.92048929663609, | |
| "eval_loss": 0.2019716054201126, | |
| "eval_runtime": 17.5057, | |
| "eval_samples_per_second": 93.398, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "learning_rate": 0.0001728813559322034, | |
| "loss": 0.1907, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.81, | |
| "eval_accuracy": 80.30581039755351, | |
| "eval_average_metrics": 80.30581039755351, | |
| "eval_loss": 0.19154316186904907, | |
| "eval_runtime": 17.3782, | |
| "eval_samples_per_second": 94.083, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 9.49, | |
| "eval_accuracy": 79.93883792048929, | |
| "eval_average_metrics": 79.93883792048929, | |
| "eval_loss": 0.19677455723285675, | |
| "eval_runtime": 15.9661, | |
| "eval_samples_per_second": 102.404, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 10.17, | |
| "learning_rate": 0.00014745762711864405, | |
| "loss": 0.183, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 10.17, | |
| "eval_accuracy": 79.44954128440367, | |
| "eval_average_metrics": 79.44954128440367, | |
| "eval_loss": 0.19117017090320587, | |
| "eval_runtime": 17.2749, | |
| "eval_samples_per_second": 94.646, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 10.85, | |
| "eval_accuracy": 78.10397553516819, | |
| "eval_average_metrics": 78.10397553516819, | |
| "eval_loss": 0.20411182940006256, | |
| "eval_runtime": 17.1181, | |
| "eval_samples_per_second": 95.513, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 11.53, | |
| "eval_accuracy": 79.02140672782875, | |
| "eval_average_metrics": 79.02140672782875, | |
| "eval_loss": 0.194900244474411, | |
| "eval_runtime": 17.1955, | |
| "eval_samples_per_second": 95.083, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "learning_rate": 0.00012203389830508474, | |
| "loss": 0.181, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 12.2, | |
| "eval_accuracy": 79.38837920489297, | |
| "eval_average_metrics": 79.38837920489297, | |
| "eval_loss": 0.20091596245765686, | |
| "eval_runtime": 16.9905, | |
| "eval_samples_per_second": 96.23, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 12.88, | |
| "eval_accuracy": 79.81651376146789, | |
| "eval_average_metrics": 79.81651376146789, | |
| "eval_loss": 0.18894420564174652, | |
| "eval_runtime": 17.2706, | |
| "eval_samples_per_second": 94.669, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 13.56, | |
| "learning_rate": 9.661016949152541e-05, | |
| "loss": 0.1786, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 13.56, | |
| "eval_accuracy": 78.77675840978593, | |
| "eval_average_metrics": 78.77675840978593, | |
| "eval_loss": 0.20160046219825745, | |
| "eval_runtime": 17.0941, | |
| "eval_samples_per_second": 95.647, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 14.24, | |
| "eval_accuracy": 79.57186544342507, | |
| "eval_average_metrics": 79.57186544342507, | |
| "eval_loss": 0.19864365458488464, | |
| "eval_runtime": 17.0536, | |
| "eval_samples_per_second": 95.874, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 14.92, | |
| "eval_accuracy": 79.51070336391437, | |
| "eval_average_metrics": 79.51070336391437, | |
| "eval_loss": 0.19150203466415405, | |
| "eval_runtime": 17.2063, | |
| "eval_samples_per_second": 95.024, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 15.25, | |
| "learning_rate": 7.11864406779661e-05, | |
| "loss": 0.1769, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 15.59, | |
| "eval_accuracy": 78.77675840978593, | |
| "eval_average_metrics": 78.77675840978593, | |
| "eval_loss": 0.19904659688472748, | |
| "eval_runtime": 17.1806, | |
| "eval_samples_per_second": 95.165, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 16.27, | |
| "eval_accuracy": 79.20489296636084, | |
| "eval_average_metrics": 79.20489296636084, | |
| "eval_loss": 0.19741013646125793, | |
| "eval_runtime": 17.2538, | |
| "eval_samples_per_second": 94.762, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 16.95, | |
| "learning_rate": 4.576271186440678e-05, | |
| "loss": 0.1741, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 16.95, | |
| "eval_accuracy": 79.51070336391437, | |
| "eval_average_metrics": 79.51070336391437, | |
| "eval_loss": 0.19429509341716766, | |
| "eval_runtime": 17.4899, | |
| "eval_samples_per_second": 93.483, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 17.63, | |
| "eval_accuracy": 78.89908256880734, | |
| "eval_average_metrics": 78.89908256880734, | |
| "eval_loss": 0.20053960382938385, | |
| "eval_runtime": 17.2615, | |
| "eval_samples_per_second": 94.719, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 18.31, | |
| "eval_accuracy": 79.32721712538226, | |
| "eval_average_metrics": 79.32721712538226, | |
| "eval_loss": 0.1975349634885788, | |
| "eval_runtime": 17.2849, | |
| "eval_samples_per_second": 94.591, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 18.64, | |
| "learning_rate": 2.0338983050847455e-05, | |
| "loss": 0.1717, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "eval_accuracy": 78.77675840978593, | |
| "eval_average_metrics": 78.77675840978593, | |
| "eval_loss": 0.20098499953746796, | |
| "eval_runtime": 17.3626, | |
| "eval_samples_per_second": 94.168, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 19.66, | |
| "eval_accuracy": 79.20489296636084, | |
| "eval_average_metrics": 79.20489296636084, | |
| "eval_loss": 0.19755637645721436, | |
| "eval_runtime": 17.3825, | |
| "eval_samples_per_second": 94.06, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 5900, | |
| "total_flos": 5.74047486286578e+16, | |
| "train_loss": 0.20100380073159427, | |
| "train_runtime": 3071.6757, | |
| "train_samples_per_second": 61.38, | |
| "train_steps_per_second": 1.921 | |
| } | |
| ], | |
| "max_steps": 5900, | |
| "num_train_epochs": 20, | |
| "total_flos": 5.74047486286578e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |