| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.1523941982912775, | |
| "eval_steps": 50, | |
| "global_step": 1450, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 3e-05, | |
| "loss": 2.0927, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 3e-05, | |
| "loss": 0.267, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1926, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1601, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1255, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_accuracy": 0.2330188679245283, | |
| "eval_f1_macro": 0.07731831394851697, | |
| "eval_f1_micro": 0.2330188679245283, | |
| "eval_loss": 0.24590551853179932, | |
| "eval_precision_macro": 0.09801465635567799, | |
| "eval_precision_micro": 0.2330188679245283, | |
| "eval_recall_macro": 0.0939265996231733, | |
| "eval_recall_micro": 0.2330188679245283, | |
| "eval_runtime": 67.1714, | |
| "eval_samples_per_second": 15.781, | |
| "eval_steps_per_second": 3.945, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6981, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1356, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0993, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1038, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1076, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_accuracy": 0.4075471698113208, | |
| "eval_f1_macro": 0.1681261191284492, | |
| "eval_f1_micro": 0.4075471698113208, | |
| "eval_loss": 0.14505280554294586, | |
| "eval_precision_macro": 0.19505399860785297, | |
| "eval_precision_micro": 0.4075471698113208, | |
| "eval_recall_macro": 0.18462138174503467, | |
| "eval_recall_micro": 0.4075471698113208, | |
| "eval_runtime": 67.1009, | |
| "eval_samples_per_second": 15.797, | |
| "eval_steps_per_second": 3.949, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 3e-05, | |
| "loss": 0.331, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0809, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0812, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0601, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 3e-05, | |
| "loss": 0.066, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_accuracy": 0.5386792452830189, | |
| "eval_f1_macro": 0.2780127225833117, | |
| "eval_f1_micro": 0.5386792452830189, | |
| "eval_loss": 0.10953618586063385, | |
| "eval_precision_macro": 0.3493311966119182, | |
| "eval_precision_micro": 0.5386792452830189, | |
| "eval_recall_macro": 0.2871523900319283, | |
| "eval_recall_micro": 0.5386792452830189, | |
| "eval_runtime": 67.0903, | |
| "eval_samples_per_second": 15.8, | |
| "eval_steps_per_second": 3.95, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2732, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0754, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0649, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0674, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0699, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_accuracy": 0.620754716981132, | |
| "eval_f1_macro": 0.3797608124202816, | |
| "eval_f1_micro": 0.620754716981132, | |
| "eval_loss": 0.09009388834238052, | |
| "eval_precision_macro": 0.3837197141355178, | |
| "eval_precision_micro": 0.620754716981132, | |
| "eval_recall_macro": 0.39915842112719735, | |
| "eval_recall_micro": 0.620754716981132, | |
| "eval_runtime": 67.0671, | |
| "eval_samples_per_second": 15.805, | |
| "eval_steps_per_second": 3.951, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1946, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0657, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0547, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0615, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3e-05, | |
| "loss": 0.066, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.6103773584905661, | |
| "eval_f1_macro": 0.41348516498355786, | |
| "eval_f1_micro": 0.6103773584905661, | |
| "eval_loss": 0.08832413703203201, | |
| "eval_precision_macro": 0.45439839834135715, | |
| "eval_precision_micro": 0.6103773584905661, | |
| "eval_recall_macro": 0.4312111435526721, | |
| "eval_recall_micro": 0.6103773584905661, | |
| "eval_runtime": 66.9345, | |
| "eval_samples_per_second": 15.836, | |
| "eval_steps_per_second": 3.959, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1494, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0655, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 3e-05, | |
| "loss": 0.06, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0616, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0452, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_accuracy": 0.6877358490566038, | |
| "eval_f1_macro": 0.5091555575082085, | |
| "eval_f1_micro": 0.6877358490566038, | |
| "eval_loss": 0.08789286762475967, | |
| "eval_precision_macro": 0.5649217974276287, | |
| "eval_precision_micro": 0.6877358490566038, | |
| "eval_recall_macro": 0.513496327466451, | |
| "eval_recall_micro": 0.6877358490566038, | |
| "eval_runtime": 67.1619, | |
| "eval_samples_per_second": 15.783, | |
| "eval_steps_per_second": 3.946, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1535, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0731, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 3e-05, | |
| "loss": 0.044, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 3e-05, | |
| "loss": 0.053, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0545, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_accuracy": 0.6764150943396227, | |
| "eval_f1_macro": 0.503999030020007, | |
| "eval_f1_micro": 0.6764150943396227, | |
| "eval_loss": 0.07607663422822952, | |
| "eval_precision_macro": 0.5194445629359009, | |
| "eval_precision_micro": 0.6764150943396227, | |
| "eval_recall_macro": 0.5287937722322651, | |
| "eval_recall_micro": 0.6764150943396227, | |
| "eval_runtime": 67.2353, | |
| "eval_samples_per_second": 15.766, | |
| "eval_steps_per_second": 3.941, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1543, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0609, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0479, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0532, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0647, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_accuracy": 0.7339622641509433, | |
| "eval_f1_macro": 0.5492932704438783, | |
| "eval_f1_micro": 0.7339622641509433, | |
| "eval_loss": 0.06653406471014023, | |
| "eval_precision_macro": 0.6193164476598846, | |
| "eval_precision_micro": 0.7339622641509433, | |
| "eval_recall_macro": 0.5252411264940735, | |
| "eval_recall_micro": 0.7339622641509433, | |
| "eval_runtime": 67.4334, | |
| "eval_samples_per_second": 15.719, | |
| "eval_steps_per_second": 3.93, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 3e-05, | |
| "loss": 0.104, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0458, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0552, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0512, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 3e-05, | |
| "loss": 0.056, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_accuracy": 0.7396226415094339, | |
| "eval_f1_macro": 0.5671730153967399, | |
| "eval_f1_micro": 0.7396226415094339, | |
| "eval_loss": 0.05136344954371452, | |
| "eval_precision_macro": 0.6096698581228938, | |
| "eval_precision_micro": 0.7396226415094339, | |
| "eval_recall_macro": 0.5767264087198709, | |
| "eval_recall_micro": 0.7396226415094339, | |
| "eval_runtime": 66.8962, | |
| "eval_samples_per_second": 15.845, | |
| "eval_steps_per_second": 3.961, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0773, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0474, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0405, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0461, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0513, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_accuracy": 0.7613207547169811, | |
| "eval_f1_macro": 0.601977568492687, | |
| "eval_f1_micro": 0.761320754716981, | |
| "eval_loss": 0.047934673726558685, | |
| "eval_precision_macro": 0.638418606498986, | |
| "eval_precision_micro": 0.7613207547169811, | |
| "eval_recall_macro": 0.6145296570629574, | |
| "eval_recall_micro": 0.7613207547169811, | |
| "eval_runtime": 66.9411, | |
| "eval_samples_per_second": 15.835, | |
| "eval_steps_per_second": 3.959, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0788, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0495, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0552, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0415, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0501, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_accuracy": 0.7509433962264151, | |
| "eval_f1_macro": 0.6074975120648255, | |
| "eval_f1_micro": 0.7509433962264151, | |
| "eval_loss": 0.05019384250044823, | |
| "eval_precision_macro": 0.624502704252128, | |
| "eval_precision_micro": 0.7509433962264151, | |
| "eval_recall_macro": 0.6167049341328479, | |
| "eval_recall_micro": 0.7509433962264151, | |
| "eval_runtime": 67.3498, | |
| "eval_samples_per_second": 15.739, | |
| "eval_steps_per_second": 3.935, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0633, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0484, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0418, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0524, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0533, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_accuracy": 0.7641509433962265, | |
| "eval_f1_macro": 0.607265930345707, | |
| "eval_f1_micro": 0.7641509433962265, | |
| "eval_loss": 0.048058342188596725, | |
| "eval_precision_macro": 0.6499724898555727, | |
| "eval_precision_micro": 0.7641509433962265, | |
| "eval_recall_macro": 0.6139175086252339, | |
| "eval_recall_micro": 0.7641509433962265, | |
| "eval_runtime": 66.897, | |
| "eval_samples_per_second": 15.845, | |
| "eval_steps_per_second": 3.961, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0418, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0482, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0458, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0432, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0462, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "eval_accuracy": 0.7481132075471698, | |
| "eval_f1_macro": 0.5679477471859753, | |
| "eval_f1_micro": 0.7481132075471698, | |
| "eval_loss": 0.047320980578660965, | |
| "eval_precision_macro": 0.5941670973495327, | |
| "eval_precision_micro": 0.7481132075471698, | |
| "eval_recall_macro": 0.5739727328111488, | |
| "eval_recall_micro": 0.7481132075471698, | |
| "eval_runtime": 67.2106, | |
| "eval_samples_per_second": 15.771, | |
| "eval_steps_per_second": 3.943, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0668, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0501, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0366, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0374, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0496, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_accuracy": 0.7971698113207547, | |
| "eval_f1_macro": 0.6517694520426227, | |
| "eval_f1_micro": 0.7971698113207546, | |
| "eval_loss": 0.04193812981247902, | |
| "eval_precision_macro": 0.6678204026981202, | |
| "eval_precision_micro": 0.7971698113207547, | |
| "eval_recall_macro": 0.6480125227888868, | |
| "eval_recall_micro": 0.7971698113207547, | |
| "eval_runtime": 67.3982, | |
| "eval_samples_per_second": 15.727, | |
| "eval_steps_per_second": 3.932, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0649, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0447, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0442, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 3e-05, | |
| "loss": 0.037, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0614, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_accuracy": 0.7773584905660378, | |
| "eval_f1_macro": 0.6308119664331103, | |
| "eval_f1_micro": 0.7773584905660378, | |
| "eval_loss": 0.04885416477918625, | |
| "eval_precision_macro": 0.6677975283624125, | |
| "eval_precision_micro": 0.7773584905660378, | |
| "eval_recall_macro": 0.6360471775658058, | |
| "eval_recall_micro": 0.7773584905660378, | |
| "eval_runtime": 67.7832, | |
| "eval_samples_per_second": 15.638, | |
| "eval_steps_per_second": 3.91, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0649, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0426, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0347, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0414, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0468, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_accuracy": 0.7830188679245284, | |
| "eval_f1_macro": 0.6493890925237205, | |
| "eval_f1_micro": 0.7830188679245284, | |
| "eval_loss": 0.044340912252664566, | |
| "eval_precision_macro": 0.6435014283226803, | |
| "eval_precision_micro": 0.7830188679245284, | |
| "eval_recall_macro": 0.6816157451405587, | |
| "eval_recall_micro": 0.7830188679245284, | |
| "eval_runtime": 67.2351, | |
| "eval_samples_per_second": 15.766, | |
| "eval_steps_per_second": 3.941, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 3e-05, | |
| "loss": 0.052, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0414, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0342, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0451, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0477, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_accuracy": 0.7971698113207547, | |
| "eval_f1_macro": 0.6662808099368048, | |
| "eval_f1_micro": 0.7971698113207546, | |
| "eval_loss": 0.041995830833911896, | |
| "eval_precision_macro": 0.7040157648486967, | |
| "eval_precision_micro": 0.7971698113207547, | |
| "eval_recall_macro": 0.6567342355863813, | |
| "eval_recall_micro": 0.7971698113207547, | |
| "eval_runtime": 67.3249, | |
| "eval_samples_per_second": 15.745, | |
| "eval_steps_per_second": 3.936, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0468, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0461, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0436, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0369, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0519, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_accuracy": 0.7632075471698113, | |
| "eval_f1_macro": 0.6291599323302522, | |
| "eval_f1_micro": 0.7632075471698113, | |
| "eval_loss": 0.04627140238881111, | |
| "eval_precision_macro": 0.6519385252086033, | |
| "eval_precision_micro": 0.7632075471698113, | |
| "eval_recall_macro": 0.6290591814696965, | |
| "eval_recall_micro": 0.7632075471698113, | |
| "eval_runtime": 67.0228, | |
| "eval_samples_per_second": 15.816, | |
| "eval_steps_per_second": 3.954, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0543, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0426, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0421, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0338, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0453, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_accuracy": 0.780188679245283, | |
| "eval_f1_macro": 0.6564187596520696, | |
| "eval_f1_micro": 0.780188679245283, | |
| "eval_loss": 0.042860858142375946, | |
| "eval_precision_macro": 0.67574812222591, | |
| "eval_precision_micro": 0.780188679245283, | |
| "eval_recall_macro": 0.6697872775950671, | |
| "eval_recall_micro": 0.780188679245283, | |
| "eval_runtime": 67.3483, | |
| "eval_samples_per_second": 15.739, | |
| "eval_steps_per_second": 3.935, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0554, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0397, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0407, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0361, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0452, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_accuracy": 0.7377358490566037, | |
| "eval_f1_macro": 0.6049285124615932, | |
| "eval_f1_micro": 0.7377358490566037, | |
| "eval_loss": 0.047125279903411865, | |
| "eval_precision_macro": 0.6181852032037266, | |
| "eval_precision_micro": 0.7377358490566037, | |
| "eval_recall_macro": 0.6300074429793591, | |
| "eval_recall_micro": 0.7377358490566037, | |
| "eval_runtime": 66.8035, | |
| "eval_samples_per_second": 15.867, | |
| "eval_steps_per_second": 3.967, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0482, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0379, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0403, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0471, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0367, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "eval_accuracy": 0.7981132075471699, | |
| "eval_f1_macro": 0.6800660818700823, | |
| "eval_f1_micro": 0.79811320754717, | |
| "eval_loss": 0.03875497728586197, | |
| "eval_precision_macro": 0.6856812225733196, | |
| "eval_precision_micro": 0.7981132075471699, | |
| "eval_recall_macro": 0.6992476720564776, | |
| "eval_recall_micro": 0.7981132075471699, | |
| "eval_runtime": 66.8444, | |
| "eval_samples_per_second": 15.858, | |
| "eval_steps_per_second": 3.964, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0351, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0479, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0421, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0406, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0377, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_accuracy": 0.8, | |
| "eval_f1_macro": 0.6590911576508658, | |
| "eval_f1_micro": 0.8000000000000002, | |
| "eval_loss": 0.03815627098083496, | |
| "eval_precision_macro": 0.6636349851737382, | |
| "eval_precision_micro": 0.8, | |
| "eval_recall_macro": 0.6697553358712118, | |
| "eval_recall_micro": 0.8, | |
| "eval_runtime": 66.9434, | |
| "eval_samples_per_second": 15.834, | |
| "eval_steps_per_second": 3.959, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0365, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0353, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0388, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0358, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0429, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "eval_accuracy": 0.7952830188679245, | |
| "eval_f1_macro": 0.6465609013784224, | |
| "eval_f1_micro": 0.7952830188679245, | |
| "eval_loss": 0.03976297378540039, | |
| "eval_precision_macro": 0.6923924758215005, | |
| "eval_precision_micro": 0.7952830188679245, | |
| "eval_recall_macro": 0.6441492192889419, | |
| "eval_recall_micro": 0.7952830188679245, | |
| "eval_runtime": 67.1705, | |
| "eval_samples_per_second": 15.781, | |
| "eval_steps_per_second": 3.945, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0461, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0434, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0524, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0362, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0451, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_accuracy": 0.7943396226415095, | |
| "eval_f1_macro": 0.6535399936575059, | |
| "eval_f1_micro": 0.7943396226415095, | |
| "eval_loss": 0.037755727767944336, | |
| "eval_precision_macro": 0.6712905678869693, | |
| "eval_precision_micro": 0.7943396226415095, | |
| "eval_recall_macro": 0.6537773538776073, | |
| "eval_recall_micro": 0.7943396226415095, | |
| "eval_runtime": 66.9611, | |
| "eval_samples_per_second": 15.83, | |
| "eval_steps_per_second": 3.958, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0456, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0455, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0409, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 3e-05, | |
| "loss": 0.037, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0347, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.7839622641509434, | |
| "eval_f1_macro": 0.6330944207402169, | |
| "eval_f1_micro": 0.7839622641509434, | |
| "eval_loss": 0.041340529918670654, | |
| "eval_precision_macro": 0.6735372413807635, | |
| "eval_precision_micro": 0.7839622641509434, | |
| "eval_recall_macro": 0.6450299050285588, | |
| "eval_recall_micro": 0.7839622641509434, | |
| "eval_runtime": 66.9053, | |
| "eval_samples_per_second": 15.843, | |
| "eval_steps_per_second": 3.961, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0421, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 3e-05, | |
| "loss": 0.041, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 3e-05, | |
| "loss": 0.033, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 3e-05, | |
| "loss": 0.036, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0378, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "eval_accuracy": 0.8047169811320755, | |
| "eval_f1_macro": 0.6488791804614907, | |
| "eval_f1_micro": 0.8047169811320755, | |
| "eval_loss": 0.037683386355638504, | |
| "eval_precision_macro": 0.7109359814450084, | |
| "eval_precision_micro": 0.8047169811320755, | |
| "eval_recall_macro": 0.6387082579227776, | |
| "eval_recall_micro": 0.8047169811320755, | |
| "eval_runtime": 67.3206, | |
| "eval_samples_per_second": 15.746, | |
| "eval_steps_per_second": 3.936, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0343, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0321, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 3e-05, | |
| "loss": 0.031, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 3e-05, | |
| "loss": 0.039, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0357, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_accuracy": 0.8028301886792453, | |
| "eval_f1_macro": 0.6648963473667772, | |
| "eval_f1_micro": 0.8028301886792453, | |
| "eval_loss": 0.03860827535390854, | |
| "eval_precision_macro": 0.6898539099210392, | |
| "eval_precision_micro": 0.8028301886792453, | |
| "eval_recall_macro": 0.6558796396655843, | |
| "eval_recall_micro": 0.8028301886792453, | |
| "eval_runtime": 67.0656, | |
| "eval_samples_per_second": 15.805, | |
| "eval_steps_per_second": 3.951, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0445, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0375, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0375, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0333, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0418, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_accuracy": 0.7962264150943397, | |
| "eval_f1_macro": 0.6910242491250081, | |
| "eval_f1_micro": 0.7962264150943396, | |
| "eval_loss": 0.0368194542825222, | |
| "eval_precision_macro": 0.7114033533579757, | |
| "eval_precision_micro": 0.7962264150943397, | |
| "eval_recall_macro": 0.6942176996685531, | |
| "eval_recall_micro": 0.7962264150943397, | |
| "eval_runtime": 66.8832, | |
| "eval_samples_per_second": 15.849, | |
| "eval_steps_per_second": 3.962, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0414, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0357, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0272, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0323, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0293, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_accuracy": 0.8141509433962264, | |
| "eval_f1_macro": 0.7097996478763092, | |
| "eval_f1_micro": 0.8141509433962264, | |
| "eval_loss": 0.035770244896411896, | |
| "eval_precision_macro": 0.7222302630120379, | |
| "eval_precision_micro": 0.8141509433962264, | |
| "eval_recall_macro": 0.7125706602249756, | |
| "eval_recall_micro": 0.8141509433962264, | |
| "eval_runtime": 67.0694, | |
| "eval_samples_per_second": 15.805, | |
| "eval_steps_per_second": 3.951, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "step": 1450, | |
| "total_flos": 3.612646182806976e+17, | |
| "train_loss": 0.07879953698865298, | |
| "train_runtime": 5948.326, | |
| "train_samples_per_second": 3.9, | |
| "train_steps_per_second": 0.244 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1450, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 250, | |
| "total_flos": 3.612646182806976e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |