| { | |
| "best_metric": 0.8528747611603266, | |
| "best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/bert/bert-base-finetuned-pos-ud-Vietnamese-VTB/checkpoint-4500", | |
| "epoch": 159.0909090909091, | |
| "global_step": 7000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 4.9e-05, | |
| "loss": 1.5268, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 4.967114093959731e-05, | |
| "loss": 0.4858, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.82, | |
| "learning_rate": 4.933557046979866e-05, | |
| "loss": 0.2763, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.1559, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 11.36, | |
| "learning_rate": 4.8664429530201344e-05, | |
| "loss": 0.0886, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 11.36, | |
| "eval_accuracy": 0.8447976376585027, | |
| "eval_loss": 0.760968804359436, | |
| "eval_runtime": 4.1015, | |
| "eval_samples_per_second": 195.051, | |
| "eval_steps_per_second": 24.381, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 13.64, | |
| "learning_rate": 4.8328859060402684e-05, | |
| "loss": 0.0614, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 15.91, | |
| "learning_rate": 4.799328859060403e-05, | |
| "loss": 0.0452, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "learning_rate": 4.7657718120805376e-05, | |
| "loss": 0.0346, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 20.45, | |
| "learning_rate": 4.7322147651006715e-05, | |
| "loss": 0.0298, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 22.73, | |
| "learning_rate": 4.698657718120806e-05, | |
| "loss": 0.0227, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 22.73, | |
| "eval_accuracy": 0.8438422789647386, | |
| "eval_loss": 0.9447396993637085, | |
| "eval_runtime": 4.0661, | |
| "eval_samples_per_second": 196.749, | |
| "eval_steps_per_second": 24.594, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 4.66510067114094e-05, | |
| "loss": 0.0204, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 27.27, | |
| "learning_rate": 4.631543624161074e-05, | |
| "loss": 0.0176, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 29.55, | |
| "learning_rate": 4.597986577181208e-05, | |
| "loss": 0.0157, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 31.82, | |
| "learning_rate": 4.5644295302013425e-05, | |
| "loss": 0.0133, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 34.09, | |
| "learning_rate": 4.5308724832214764e-05, | |
| "loss": 0.0137, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 34.09, | |
| "eval_accuracy": 0.8472294597880841, | |
| "eval_loss": 1.0613317489624023, | |
| "eval_runtime": 4.0723, | |
| "eval_samples_per_second": 196.45, | |
| "eval_steps_per_second": 24.556, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 36.36, | |
| "learning_rate": 4.497315436241611e-05, | |
| "loss": 0.0144, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 38.64, | |
| "learning_rate": 4.463758389261745e-05, | |
| "loss": 0.0118, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 40.91, | |
| "learning_rate": 4.4302013422818796e-05, | |
| "loss": 0.0117, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 43.18, | |
| "learning_rate": 4.3966442953020135e-05, | |
| "loss": 0.01, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 45.45, | |
| "learning_rate": 4.363087248322148e-05, | |
| "loss": 0.0085, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 45.45, | |
| "eval_accuracy": 0.8465346534653465, | |
| "eval_loss": 1.073764681816101, | |
| "eval_runtime": 4.0894, | |
| "eval_samples_per_second": 195.629, | |
| "eval_steps_per_second": 24.454, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 47.73, | |
| "learning_rate": 4.329530201342282e-05, | |
| "loss": 0.0105, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 4.2959731543624166e-05, | |
| "loss": 0.0087, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 52.27, | |
| "learning_rate": 4.2624161073825505e-05, | |
| "loss": 0.0076, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 54.55, | |
| "learning_rate": 4.2288590604026845e-05, | |
| "loss": 0.0079, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 56.82, | |
| "learning_rate": 4.195302013422819e-05, | |
| "loss": 0.0087, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 56.82, | |
| "eval_accuracy": 0.8443633837067918, | |
| "eval_loss": 1.1314551830291748, | |
| "eval_runtime": 4.0535, | |
| "eval_samples_per_second": 197.361, | |
| "eval_steps_per_second": 24.67, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 59.09, | |
| "learning_rate": 4.161744966442953e-05, | |
| "loss": 0.0077, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 61.36, | |
| "learning_rate": 4.1281879194630876e-05, | |
| "loss": 0.0081, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 63.64, | |
| "learning_rate": 4.0946308724832215e-05, | |
| "loss": 0.007, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 65.91, | |
| "learning_rate": 4.061073825503356e-05, | |
| "loss": 0.007, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 68.18, | |
| "learning_rate": 4.02751677852349e-05, | |
| "loss": 0.0048, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 68.18, | |
| "eval_accuracy": 0.8502692374500608, | |
| "eval_loss": 1.1457550525665283, | |
| "eval_runtime": 4.065, | |
| "eval_samples_per_second": 196.803, | |
| "eval_steps_per_second": 24.6, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 70.45, | |
| "learning_rate": 3.993959731543625e-05, | |
| "loss": 0.006, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 72.73, | |
| "learning_rate": 3.9604026845637586e-05, | |
| "loss": 0.0047, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "learning_rate": 3.926845637583893e-05, | |
| "loss": 0.0066, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 77.27, | |
| "learning_rate": 3.893288590604027e-05, | |
| "loss": 0.0046, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 79.55, | |
| "learning_rate": 3.859731543624161e-05, | |
| "loss": 0.005, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 79.55, | |
| "eval_accuracy": 0.8461872503039778, | |
| "eval_loss": 1.1730798482894897, | |
| "eval_runtime": 4.0626, | |
| "eval_samples_per_second": 196.919, | |
| "eval_steps_per_second": 24.615, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 81.82, | |
| "learning_rate": 3.826174496644295e-05, | |
| "loss": 0.0055, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 84.09, | |
| "learning_rate": 3.7926174496644296e-05, | |
| "loss": 0.0052, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 86.36, | |
| "learning_rate": 3.7590604026845635e-05, | |
| "loss": 0.0053, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 88.64, | |
| "learning_rate": 3.725503355704698e-05, | |
| "loss": 0.0028, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 90.91, | |
| "learning_rate": 3.691946308724833e-05, | |
| "loss": 0.0045, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 90.91, | |
| "eval_accuracy": 0.8475768629494529, | |
| "eval_loss": 1.1992286443710327, | |
| "eval_runtime": 4.0548, | |
| "eval_samples_per_second": 197.296, | |
| "eval_steps_per_second": 24.662, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 93.18, | |
| "learning_rate": 3.6583892617449666e-05, | |
| "loss": 0.0041, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 95.45, | |
| "learning_rate": 3.624832214765101e-05, | |
| "loss": 0.0035, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 97.73, | |
| "learning_rate": 3.591275167785235e-05, | |
| "loss": 0.0045, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 3.55771812080537e-05, | |
| "loss": 0.0053, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 102.27, | |
| "learning_rate": 3.524161073825504e-05, | |
| "loss": 0.0047, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 102.27, | |
| "eval_accuracy": 0.8528747611603266, | |
| "eval_loss": 1.158167839050293, | |
| "eval_runtime": 4.0568, | |
| "eval_samples_per_second": 197.201, | |
| "eval_steps_per_second": 24.65, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 104.55, | |
| "learning_rate": 3.4906040268456376e-05, | |
| "loss": 0.0028, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 106.82, | |
| "learning_rate": 3.4570469798657716e-05, | |
| "loss": 0.0025, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 109.09, | |
| "learning_rate": 3.423489932885906e-05, | |
| "loss": 0.0048, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 111.36, | |
| "learning_rate": 3.38993288590604e-05, | |
| "loss": 0.0036, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 113.64, | |
| "learning_rate": 3.356375838926175e-05, | |
| "loss": 0.0037, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 113.64, | |
| "eval_accuracy": 0.8517457008858781, | |
| "eval_loss": 1.2208179235458374, | |
| "eval_runtime": 4.0691, | |
| "eval_samples_per_second": 196.603, | |
| "eval_steps_per_second": 24.575, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 115.91, | |
| "learning_rate": 3.3228187919463086e-05, | |
| "loss": 0.005, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 118.18, | |
| "learning_rate": 3.289261744966443e-05, | |
| "loss": 0.0038, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 120.45, | |
| "learning_rate": 3.255704697986577e-05, | |
| "loss": 0.0037, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 122.73, | |
| "learning_rate": 3.222147651006712e-05, | |
| "loss": 0.0043, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 125.0, | |
| "learning_rate": 3.188590604026846e-05, | |
| "loss": 0.0032, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 125.0, | |
| "eval_accuracy": 0.847663713739795, | |
| "eval_loss": 1.2347321510314941, | |
| "eval_runtime": 4.0676, | |
| "eval_samples_per_second": 196.677, | |
| "eval_steps_per_second": 24.585, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 127.27, | |
| "learning_rate": 3.15503355704698e-05, | |
| "loss": 0.0024, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 129.55, | |
| "learning_rate": 3.121476510067114e-05, | |
| "loss": 0.003, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 131.82, | |
| "learning_rate": 3.087919463087248e-05, | |
| "loss": 0.0016, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 134.09, | |
| "learning_rate": 3.054362416107383e-05, | |
| "loss": 0.0026, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 136.36, | |
| "learning_rate": 3.0208053691275167e-05, | |
| "loss": 0.002, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 136.36, | |
| "eval_accuracy": 0.8496612819176654, | |
| "eval_loss": 1.2792454957962036, | |
| "eval_runtime": 4.062, | |
| "eval_samples_per_second": 196.947, | |
| "eval_steps_per_second": 24.618, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 138.64, | |
| "learning_rate": 2.987248322147651e-05, | |
| "loss": 0.0022, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 140.91, | |
| "learning_rate": 2.9536912751677852e-05, | |
| "loss": 0.002, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 143.18, | |
| "learning_rate": 2.9201342281879195e-05, | |
| "loss": 0.0024, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 145.45, | |
| "learning_rate": 2.8865771812080537e-05, | |
| "loss": 0.003, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 147.73, | |
| "learning_rate": 2.853020134228188e-05, | |
| "loss": 0.0017, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 147.73, | |
| "eval_accuracy": 0.8494007295466389, | |
| "eval_loss": 1.2664971351623535, | |
| "eval_runtime": 4.0627, | |
| "eval_samples_per_second": 196.913, | |
| "eval_steps_per_second": 24.614, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 150.0, | |
| "learning_rate": 2.8194630872483223e-05, | |
| "loss": 0.002, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 152.27, | |
| "learning_rate": 2.785906040268457e-05, | |
| "loss": 0.0017, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 154.55, | |
| "learning_rate": 2.752348993288591e-05, | |
| "loss": 0.0023, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 156.82, | |
| "learning_rate": 2.7187919463087247e-05, | |
| "loss": 0.002, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 159.09, | |
| "learning_rate": 2.685234899328859e-05, | |
| "loss": 0.0014, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 159.09, | |
| "eval_accuracy": 0.8520062532569046, | |
| "eval_loss": 1.2847464084625244, | |
| "eval_runtime": 4.0924, | |
| "eval_samples_per_second": 195.486, | |
| "eval_steps_per_second": 24.436, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 159.09, | |
| "step": 7000, | |
| "total_flos": 2.9102998258397184e+16, | |
| "train_loss": 0.04406098906695843, | |
| "train_runtime": 1822.8763, | |
| "train_samples_per_second": 263.32, | |
| "train_steps_per_second": 8.229 | |
| } | |
| ], | |
| "max_steps": 15000, | |
| "num_train_epochs": 341, | |
| "total_flos": 2.9102998258397184e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |