{ "best_metric": null, "best_model_checkpoint": null, "epoch": 2.0, "global_step": 10950, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.02, "learning_rate": 2.9726027397260275e-05, "loss": 0.2472, "step": 100 }, { "epoch": 0.04, "learning_rate": 2.945205479452055e-05, "loss": 0.179, "step": 200 }, { "epoch": 0.05, "learning_rate": 2.9178082191780824e-05, "loss": 0.169, "step": 300 }, { "epoch": 0.07, "learning_rate": 2.8904109589041095e-05, "loss": 0.1476, "step": 400 }, { "epoch": 0.09, "learning_rate": 2.863013698630137e-05, "loss": 0.1419, "step": 500 }, { "epoch": 0.11, "learning_rate": 2.8356164383561644e-05, "loss": 0.1517, "step": 600 }, { "epoch": 0.13, "learning_rate": 2.808219178082192e-05, "loss": 0.1466, "step": 700 }, { "epoch": 0.15, "learning_rate": 2.7808219178082193e-05, "loss": 0.1398, "step": 800 }, { "epoch": 0.16, "learning_rate": 2.7534246575342468e-05, "loss": 0.1409, "step": 900 }, { "epoch": 0.18, "learning_rate": 2.726027397260274e-05, "loss": 0.1381, "step": 1000 }, { "epoch": 0.2, "learning_rate": 2.6986301369863014e-05, "loss": 0.1349, "step": 1100 }, { "epoch": 0.22, "learning_rate": 2.6712328767123288e-05, "loss": 0.1375, "step": 1200 }, { "epoch": 0.24, "learning_rate": 2.6438356164383563e-05, "loss": 0.1397, "step": 1300 }, { "epoch": 0.26, "learning_rate": 2.6164383561643837e-05, "loss": 0.1354, "step": 1400 }, { "epoch": 0.27, "learning_rate": 2.589041095890411e-05, "loss": 0.1288, "step": 1500 }, { "epoch": 0.29, "learning_rate": 2.5616438356164386e-05, "loss": 0.1216, "step": 1600 }, { "epoch": 0.31, "learning_rate": 2.5342465753424657e-05, "loss": 0.1282, "step": 1700 }, { "epoch": 0.33, "learning_rate": 2.5068493150684932e-05, "loss": 0.1302, "step": 1800 }, { "epoch": 0.35, "learning_rate": 2.4794520547945206e-05, "loss": 0.1289, "step": 1900 }, { "epoch": 0.37, "learning_rate": 2.452054794520548e-05, "loss": 0.1209, "step": 2000 }, { "epoch": 0.38, "learning_rate": 2.4246575342465755e-05, "loss": 0.1253, "step": 2100 }, { "epoch": 0.4, "learning_rate": 2.397260273972603e-05, "loss": 0.1268, "step": 2200 }, { "epoch": 0.42, "learning_rate": 2.36986301369863e-05, "loss": 0.1234, "step": 2300 }, { "epoch": 0.44, "learning_rate": 2.3424657534246575e-05, "loss": 0.1225, "step": 2400 }, { "epoch": 0.46, "learning_rate": 2.315068493150685e-05, "loss": 0.1148, "step": 2500 }, { "epoch": 0.47, "learning_rate": 2.2876712328767124e-05, "loss": 0.1228, "step": 2600 }, { "epoch": 0.49, "learning_rate": 2.26027397260274e-05, "loss": 0.1141, "step": 2700 }, { "epoch": 0.51, "learning_rate": 2.2328767123287673e-05, "loss": 0.1142, "step": 2800 }, { "epoch": 0.53, "learning_rate": 2.2054794520547945e-05, "loss": 0.11, "step": 2900 }, { "epoch": 0.55, "learning_rate": 2.178082191780822e-05, "loss": 0.1142, "step": 3000 }, { "epoch": 0.57, "learning_rate": 2.1506849315068494e-05, "loss": 0.1118, "step": 3100 }, { "epoch": 0.58, "learning_rate": 2.1232876712328768e-05, "loss": 0.1126, "step": 3200 }, { "epoch": 0.6, "learning_rate": 2.0958904109589043e-05, "loss": 0.1147, "step": 3300 }, { "epoch": 0.62, "learning_rate": 2.0684931506849317e-05, "loss": 0.1134, "step": 3400 }, { "epoch": 0.64, "learning_rate": 2.041095890410959e-05, "loss": 0.1139, "step": 3500 }, { "epoch": 0.66, "learning_rate": 2.0136986301369863e-05, "loss": 0.1136, "step": 3600 }, { "epoch": 0.68, "learning_rate": 1.9863013698630137e-05, "loss": 0.1034, "step": 3700 }, { "epoch": 0.69, "learning_rate": 1.9589041095890412e-05, "loss": 0.1063, "step": 3800 }, { "epoch": 0.71, "learning_rate": 1.9315068493150686e-05, "loss": 0.1075, "step": 3900 }, { "epoch": 0.73, "learning_rate": 1.904109589041096e-05, "loss": 0.1076, "step": 4000 }, { "epoch": 0.75, "learning_rate": 1.8767123287671235e-05, "loss": 0.1057, "step": 4100 }, { "epoch": 0.77, "learning_rate": 1.8493150684931506e-05, "loss": 0.1016, "step": 4200 }, { "epoch": 0.79, "learning_rate": 1.821917808219178e-05, "loss": 0.1058, "step": 4300 }, { "epoch": 0.8, "learning_rate": 1.7945205479452055e-05, "loss": 0.1099, "step": 4400 }, { "epoch": 0.82, "learning_rate": 1.767123287671233e-05, "loss": 0.1082, "step": 4500 }, { "epoch": 0.84, "learning_rate": 1.7397260273972604e-05, "loss": 0.1009, "step": 4600 }, { "epoch": 0.86, "learning_rate": 1.712328767123288e-05, "loss": 0.1076, "step": 4700 }, { "epoch": 0.88, "learning_rate": 1.6849315068493153e-05, "loss": 0.106, "step": 4800 }, { "epoch": 0.89, "learning_rate": 1.6575342465753425e-05, "loss": 0.103, "step": 4900 }, { "epoch": 0.91, "learning_rate": 1.63013698630137e-05, "loss": 0.1111, "step": 5000 }, { "epoch": 0.93, "learning_rate": 1.6027397260273974e-05, "loss": 0.1039, "step": 5100 }, { "epoch": 0.95, "learning_rate": 1.5753424657534248e-05, "loss": 0.1004, "step": 5200 }, { "epoch": 0.97, "learning_rate": 1.5479452054794523e-05, "loss": 0.1039, "step": 5300 }, { "epoch": 0.99, "learning_rate": 1.5205479452054795e-05, "loss": 0.1053, "step": 5400 }, { "epoch": 1.0, "eval_exact_match": 73.0085146641438, "eval_f1": 83.09249441563516, "step": 5475 }, { "epoch": 1.0, "learning_rate": 1.4931506849315068e-05, "loss": 0.0943, "step": 5500 }, { "epoch": 1.02, "learning_rate": 1.4657534246575343e-05, "loss": 0.0767, "step": 5600 }, { "epoch": 1.04, "learning_rate": 1.4383561643835615e-05, "loss": 0.0762, "step": 5700 }, { "epoch": 1.06, "learning_rate": 1.410958904109589e-05, "loss": 0.0759, "step": 5800 }, { "epoch": 1.08, "learning_rate": 1.3835616438356164e-05, "loss": 0.0801, "step": 5900 }, { "epoch": 1.1, "learning_rate": 1.3561643835616437e-05, "loss": 0.0719, "step": 6000 }, { "epoch": 1.11, "learning_rate": 1.3287671232876712e-05, "loss": 0.0754, "step": 6100 }, { "epoch": 1.13, "learning_rate": 1.3013698630136986e-05, "loss": 0.0781, "step": 6200 }, { "epoch": 1.15, "learning_rate": 1.2739726027397259e-05, "loss": 0.0823, "step": 6300 }, { "epoch": 1.17, "learning_rate": 1.2465753424657534e-05, "loss": 0.0827, "step": 6400 }, { "epoch": 1.19, "learning_rate": 1.2191780821917808e-05, "loss": 0.0718, "step": 6500 }, { "epoch": 1.21, "learning_rate": 1.1917808219178083e-05, "loss": 0.0755, "step": 6600 }, { "epoch": 1.22, "learning_rate": 1.1643835616438355e-05, "loss": 0.0807, "step": 6700 }, { "epoch": 1.24, "learning_rate": 1.136986301369863e-05, "loss": 0.083, "step": 6800 }, { "epoch": 1.26, "learning_rate": 1.1095890410958904e-05, "loss": 0.0741, "step": 6900 }, { "epoch": 1.28, "learning_rate": 1.0821917808219177e-05, "loss": 0.0782, "step": 7000 }, { "epoch": 1.3, "learning_rate": 1.0547945205479452e-05, "loss": 0.0761, "step": 7100 }, { "epoch": 1.32, "learning_rate": 1.0273972602739726e-05, "loss": 0.0739, "step": 7200 }, { "epoch": 1.33, "learning_rate": 9.999999999999999e-06, "loss": 0.0763, "step": 7300 }, { "epoch": 1.35, "learning_rate": 9.726027397260274e-06, "loss": 0.0796, "step": 7400 }, { "epoch": 1.37, "learning_rate": 9.452054794520548e-06, "loss": 0.08, "step": 7500 }, { "epoch": 1.39, "learning_rate": 9.178082191780821e-06, "loss": 0.0768, "step": 7600 }, { "epoch": 1.41, "learning_rate": 8.904109589041095e-06, "loss": 0.0766, "step": 7700 }, { "epoch": 1.42, "learning_rate": 8.63013698630137e-06, "loss": 0.0739, "step": 7800 }, { "epoch": 1.44, "learning_rate": 8.356164383561643e-06, "loss": 0.0718, "step": 7900 }, { "epoch": 1.46, "learning_rate": 8.082191780821917e-06, "loss": 0.0737, "step": 8000 }, { "epoch": 1.48, "learning_rate": 7.808219178082192e-06, "loss": 0.0703, "step": 8100 }, { "epoch": 1.5, "learning_rate": 7.5342465753424655e-06, "loss": 0.0723, "step": 8200 }, { "epoch": 1.52, "learning_rate": 7.260273972602739e-06, "loss": 0.0747, "step": 8300 }, { "epoch": 1.53, "learning_rate": 6.986301369863014e-06, "loss": 0.0787, "step": 8400 }, { "epoch": 1.55, "learning_rate": 6.712328767123287e-06, "loss": 0.0704, "step": 8500 }, { "epoch": 1.57, "learning_rate": 6.438356164383562e-06, "loss": 0.0703, "step": 8600 }, { "epoch": 1.59, "learning_rate": 6.1643835616438354e-06, "loss": 0.0688, "step": 8700 }, { "epoch": 1.61, "learning_rate": 5.890410958904109e-06, "loss": 0.0762, "step": 8800 }, { "epoch": 1.63, "learning_rate": 5.616438356164384e-06, "loss": 0.0723, "step": 8900 }, { "epoch": 1.64, "learning_rate": 5.342465753424657e-06, "loss": 0.073, "step": 9000 }, { "epoch": 1.66, "learning_rate": 5.068493150684931e-06, "loss": 0.0725, "step": 9100 }, { "epoch": 1.68, "learning_rate": 4.7945205479452054e-06, "loss": 0.0719, "step": 9200 }, { "epoch": 1.7, "learning_rate": 4.520547945205479e-06, "loss": 0.0674, "step": 9300 }, { "epoch": 1.72, "learning_rate": 4.246575342465754e-06, "loss": 0.0667, "step": 9400 }, { "epoch": 1.74, "learning_rate": 3.972602739726027e-06, "loss": 0.0722, "step": 9500 }, { "epoch": 1.75, "learning_rate": 3.6986301369863014e-06, "loss": 0.071, "step": 9600 }, { "epoch": 1.77, "learning_rate": 3.4246575342465754e-06, "loss": 0.074, "step": 9700 }, { "epoch": 1.79, "learning_rate": 3.150684931506849e-06, "loss": 0.0696, "step": 9800 }, { "epoch": 1.81, "learning_rate": 2.876712328767123e-06, "loss": 0.0699, "step": 9900 }, { "epoch": 1.83, "learning_rate": 2.6027397260273973e-06, "loss": 0.0734, "step": 10000 }, { "epoch": 1.84, "learning_rate": 2.3287671232876713e-06, "loss": 0.0687, "step": 10100 }, { "epoch": 1.86, "learning_rate": 2.054794520547945e-06, "loss": 0.0729, "step": 10200 }, { "epoch": 1.88, "learning_rate": 1.780821917808219e-06, "loss": 0.0703, "step": 10300 }, { "epoch": 1.9, "learning_rate": 1.5068493150684932e-06, "loss": 0.0734, "step": 10400 }, { "epoch": 1.92, "learning_rate": 1.232876712328767e-06, "loss": 0.0728, "step": 10500 }, { "epoch": 1.94, "learning_rate": 9.589041095890411e-07, "loss": 0.073, "step": 10600 }, { "epoch": 1.95, "learning_rate": 6.84931506849315e-07, "loss": 0.0694, "step": 10700 }, { "epoch": 1.97, "learning_rate": 4.1095890410958903e-07, "loss": 0.0675, "step": 10800 }, { "epoch": 1.99, "learning_rate": 1.36986301369863e-07, "loss": 0.0709, "step": 10900 }, { "epoch": 2.0, "eval_exact_match": 67.9848628192999, "eval_f1": 79.76062168781495, "step": 10950 }, { "epoch": 2.0, "step": 10950, "total_flos": 5.341230400536576e+16, "train_loss": 0.09864264518702956, "train_runtime": 9855.5808, "train_samples_per_second": 17.777, "train_steps_per_second": 1.111 } ], "max_steps": 10950, "num_train_epochs": 2, "total_flos": 5.341230400536576e+16, "trial_name": null, "trial_params": null }