|
{ |
|
"best_metric": 89.27591554336156, |
|
"best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/mbert/mbert-base-finetuned-parsing-ud-English-EWT/checkpoint-5000", |
|
"epoch": 19.132653061224488, |
|
"global_step": 7500, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.26, |
|
"learning_rate": 7.840000000000001e-05, |
|
"loss": 3.0745, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.51, |
|
"learning_rate": 7.947382550335571e-05, |
|
"loss": 0.8754, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.77, |
|
"learning_rate": 7.893691275167786e-05, |
|
"loss": 0.6847, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"learning_rate": 7.840000000000001e-05, |
|
"loss": 0.5794, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 1.28, |
|
"learning_rate": 7.786308724832216e-05, |
|
"loss": 0.424, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 1.28, |
|
"eval_las": 86.52033878086604, |
|
"eval_loss": 0.5685478448867798, |
|
"eval_runtime": 12.8416, |
|
"eval_samples_per_second": 155.822, |
|
"eval_steps_per_second": 19.546, |
|
"eval_uas": 89.4031571831882, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 1.53, |
|
"learning_rate": 7.73261744966443e-05, |
|
"loss": 0.4108, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 1.79, |
|
"learning_rate": 7.678926174496645e-05, |
|
"loss": 0.4139, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 2.04, |
|
"learning_rate": 7.62523489932886e-05, |
|
"loss": 0.3663, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 2.3, |
|
"learning_rate": 7.571543624161075e-05, |
|
"loss": 0.2592, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 2.55, |
|
"learning_rate": 7.51785234899329e-05, |
|
"loss": 0.2636, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 2.55, |
|
"eval_las": 87.8523996978011, |
|
"eval_loss": 0.5544716715812683, |
|
"eval_runtime": 12.834, |
|
"eval_samples_per_second": 155.914, |
|
"eval_steps_per_second": 19.557, |
|
"eval_uas": 90.59604755656288, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 2.81, |
|
"learning_rate": 7.464161073825505e-05, |
|
"loss": 0.2582, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 3.06, |
|
"learning_rate": 7.410469798657718e-05, |
|
"loss": 0.2466, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 3.32, |
|
"learning_rate": 7.357315436241611e-05, |
|
"loss": 0.1731, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 3.57, |
|
"learning_rate": 7.303624161073826e-05, |
|
"loss": 0.1932, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 3.83, |
|
"learning_rate": 7.249932885906041e-05, |
|
"loss": 0.1935, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 3.83, |
|
"eval_las": 88.85045131019126, |
|
"eval_loss": 0.5856050252914429, |
|
"eval_runtime": 12.8153, |
|
"eval_samples_per_second": 156.141, |
|
"eval_steps_per_second": 19.586, |
|
"eval_uas": 91.49469163783849, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 4.08, |
|
"learning_rate": 7.196241610738256e-05, |
|
"loss": 0.1652, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 4.34, |
|
"learning_rate": 7.142550335570471e-05, |
|
"loss": 0.1374, |
|
"step": 1700 |
|
}, |
|
{ |
|
"epoch": 4.59, |
|
"learning_rate": 7.088859060402686e-05, |
|
"loss": 0.1356, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 4.85, |
|
"learning_rate": 7.0351677852349e-05, |
|
"loss": 0.1431, |
|
"step": 1900 |
|
}, |
|
{ |
|
"epoch": 5.1, |
|
"learning_rate": 6.981476510067114e-05, |
|
"loss": 0.1221, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 5.1, |
|
"eval_las": 88.68742295916338, |
|
"eval_loss": 0.7061353325843811, |
|
"eval_runtime": 12.9124, |
|
"eval_samples_per_second": 154.967, |
|
"eval_steps_per_second": 19.439, |
|
"eval_uas": 91.10103781462485, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 5.36, |
|
"learning_rate": 6.927785234899329e-05, |
|
"loss": 0.1071, |
|
"step": 2100 |
|
}, |
|
{ |
|
"epoch": 5.61, |
|
"learning_rate": 6.874093959731543e-05, |
|
"loss": 0.1046, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 5.87, |
|
"learning_rate": 6.820402684563758e-05, |
|
"loss": 0.1113, |
|
"step": 2300 |
|
}, |
|
{ |
|
"epoch": 6.12, |
|
"learning_rate": 6.766711409395973e-05, |
|
"loss": 0.103, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 6.38, |
|
"learning_rate": 6.713020134228188e-05, |
|
"loss": 0.0852, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 6.38, |
|
"eval_las": 88.96974034752873, |
|
"eval_loss": 0.775122344493866, |
|
"eval_runtime": 12.9063, |
|
"eval_samples_per_second": 155.04, |
|
"eval_steps_per_second": 19.448, |
|
"eval_uas": 91.53047834903973, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 6.63, |
|
"learning_rate": 6.659328859060403e-05, |
|
"loss": 0.0804, |
|
"step": 2600 |
|
}, |
|
{ |
|
"epoch": 6.89, |
|
"learning_rate": 6.605637583892618e-05, |
|
"loss": 0.0855, |
|
"step": 2700 |
|
}, |
|
{ |
|
"epoch": 7.14, |
|
"learning_rate": 6.551946308724832e-05, |
|
"loss": 0.0762, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 7.4, |
|
"learning_rate": 6.498255033557047e-05, |
|
"loss": 0.0682, |
|
"step": 2900 |
|
}, |
|
{ |
|
"epoch": 7.65, |
|
"learning_rate": 6.444563758389262e-05, |
|
"loss": 0.0739, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 7.65, |
|
"eval_las": 88.96576404628415, |
|
"eval_loss": 0.8953888416290283, |
|
"eval_runtime": 12.8724, |
|
"eval_samples_per_second": 155.449, |
|
"eval_steps_per_second": 19.499, |
|
"eval_uas": 91.44299972165891, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 7.91, |
|
"learning_rate": 6.390872483221477e-05, |
|
"loss": 0.0693, |
|
"step": 3100 |
|
}, |
|
{ |
|
"epoch": 8.16, |
|
"learning_rate": 6.337181208053692e-05, |
|
"loss": 0.0661, |
|
"step": 3200 |
|
}, |
|
{ |
|
"epoch": 8.42, |
|
"learning_rate": 6.283489932885907e-05, |
|
"loss": 0.0642, |
|
"step": 3300 |
|
}, |
|
{ |
|
"epoch": 8.67, |
|
"learning_rate": 6.229798657718121e-05, |
|
"loss": 0.0607, |
|
"step": 3400 |
|
}, |
|
{ |
|
"epoch": 8.93, |
|
"learning_rate": 6.176644295302013e-05, |
|
"loss": 0.0668, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 8.93, |
|
"eval_las": 88.93792993757206, |
|
"eval_loss": 0.8302858471870422, |
|
"eval_runtime": 12.9054, |
|
"eval_samples_per_second": 155.052, |
|
"eval_steps_per_second": 19.449, |
|
"eval_uas": 91.45890492663725, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 9.18, |
|
"learning_rate": 6.122953020134228e-05, |
|
"loss": 0.0553, |
|
"step": 3600 |
|
}, |
|
{ |
|
"epoch": 9.44, |
|
"learning_rate": 6.069261744966444e-05, |
|
"loss": 0.0546, |
|
"step": 3700 |
|
}, |
|
{ |
|
"epoch": 9.69, |
|
"learning_rate": 6.0155704697986585e-05, |
|
"loss": 0.0554, |
|
"step": 3800 |
|
}, |
|
{ |
|
"epoch": 9.95, |
|
"learning_rate": 5.9618791946308734e-05, |
|
"loss": 0.0552, |
|
"step": 3900 |
|
}, |
|
{ |
|
"epoch": 10.2, |
|
"learning_rate": 5.9081879194630875e-05, |
|
"loss": 0.0512, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 10.2, |
|
"eval_las": 88.94985884130581, |
|
"eval_loss": 0.9160601496696472, |
|
"eval_runtime": 12.8984, |
|
"eval_samples_per_second": 155.135, |
|
"eval_steps_per_second": 19.46, |
|
"eval_uas": 91.40721301045768, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 10.46, |
|
"learning_rate": 5.854496644295302e-05, |
|
"loss": 0.0504, |
|
"step": 4100 |
|
}, |
|
{ |
|
"epoch": 10.71, |
|
"learning_rate": 5.800805369127517e-05, |
|
"loss": 0.0495, |
|
"step": 4200 |
|
}, |
|
{ |
|
"epoch": 10.97, |
|
"learning_rate": 5.747114093959732e-05, |
|
"loss": 0.044, |
|
"step": 4300 |
|
}, |
|
{ |
|
"epoch": 11.22, |
|
"learning_rate": 5.693422818791947e-05, |
|
"loss": 0.0402, |
|
"step": 4400 |
|
}, |
|
{ |
|
"epoch": 11.48, |
|
"learning_rate": 5.6397315436241616e-05, |
|
"loss": 0.0378, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 11.48, |
|
"eval_las": 89.11686349357826, |
|
"eval_loss": 1.0157352685928345, |
|
"eval_runtime": 12.8902, |
|
"eval_samples_per_second": 155.234, |
|
"eval_steps_per_second": 19.472, |
|
"eval_uas": 91.51854944530598, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 11.73, |
|
"learning_rate": 5.5860402684563764e-05, |
|
"loss": 0.044, |
|
"step": 4600 |
|
}, |
|
{ |
|
"epoch": 11.99, |
|
"learning_rate": 5.532348993288591e-05, |
|
"loss": 0.0423, |
|
"step": 4700 |
|
}, |
|
{ |
|
"epoch": 12.24, |
|
"learning_rate": 5.478657718120806e-05, |
|
"loss": 0.0372, |
|
"step": 4800 |
|
}, |
|
{ |
|
"epoch": 12.5, |
|
"learning_rate": 5.424966442953021e-05, |
|
"loss": 0.0367, |
|
"step": 4900 |
|
}, |
|
{ |
|
"epoch": 12.76, |
|
"learning_rate": 5.371275167785236e-05, |
|
"loss": 0.0354, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 12.76, |
|
"eval_las": 89.27591554336156, |
|
"eval_loss": 0.9186694025993347, |
|
"eval_runtime": 12.8914, |
|
"eval_samples_per_second": 155.22, |
|
"eval_steps_per_second": 19.47, |
|
"eval_uas": 91.59409916895304, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 13.01, |
|
"learning_rate": 5.31758389261745e-05, |
|
"loss": 0.0379, |
|
"step": 5100 |
|
}, |
|
{ |
|
"epoch": 13.27, |
|
"learning_rate": 5.263892617449665e-05, |
|
"loss": 0.0299, |
|
"step": 5200 |
|
}, |
|
{ |
|
"epoch": 13.52, |
|
"learning_rate": 5.2102013422818795e-05, |
|
"loss": 0.0329, |
|
"step": 5300 |
|
}, |
|
{ |
|
"epoch": 13.78, |
|
"learning_rate": 5.1565100671140944e-05, |
|
"loss": 0.032, |
|
"step": 5400 |
|
}, |
|
{ |
|
"epoch": 14.03, |
|
"learning_rate": 5.102818791946309e-05, |
|
"loss": 0.0287, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 14.03, |
|
"eval_las": 88.96178774503957, |
|
"eval_loss": 1.0049980878829956, |
|
"eval_runtime": 12.88, |
|
"eval_samples_per_second": 155.357, |
|
"eval_steps_per_second": 19.488, |
|
"eval_uas": 91.26406616565271, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 14.29, |
|
"learning_rate": 5.049127516778524e-05, |
|
"loss": 0.0271, |
|
"step": 5600 |
|
}, |
|
{ |
|
"epoch": 14.54, |
|
"learning_rate": 4.995436241610739e-05, |
|
"loss": 0.0268, |
|
"step": 5700 |
|
}, |
|
{ |
|
"epoch": 14.8, |
|
"learning_rate": 4.941744966442954e-05, |
|
"loss": 0.0305, |
|
"step": 5800 |
|
}, |
|
{ |
|
"epoch": 15.05, |
|
"learning_rate": 4.8880536912751685e-05, |
|
"loss": 0.026, |
|
"step": 5900 |
|
}, |
|
{ |
|
"epoch": 15.31, |
|
"learning_rate": 4.834362416107383e-05, |
|
"loss": 0.0211, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 15.31, |
|
"eval_las": 89.10891089108911, |
|
"eval_loss": 1.1387187242507935, |
|
"eval_runtime": 12.8891, |
|
"eval_samples_per_second": 155.247, |
|
"eval_steps_per_second": 19.474, |
|
"eval_uas": 91.48276273410474, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 15.56, |
|
"learning_rate": 4.780671140939598e-05, |
|
"loss": 0.0252, |
|
"step": 6100 |
|
}, |
|
{ |
|
"epoch": 15.82, |
|
"learning_rate": 4.726979865771813e-05, |
|
"loss": 0.0259, |
|
"step": 6200 |
|
}, |
|
{ |
|
"epoch": 16.07, |
|
"learning_rate": 4.673288590604027e-05, |
|
"loss": 0.0277, |
|
"step": 6300 |
|
}, |
|
{ |
|
"epoch": 16.33, |
|
"learning_rate": 4.619597315436242e-05, |
|
"loss": 0.0205, |
|
"step": 6400 |
|
}, |
|
{ |
|
"epoch": 16.58, |
|
"learning_rate": 4.565906040268457e-05, |
|
"loss": 0.0248, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 16.58, |
|
"eval_las": 89.20831842220367, |
|
"eval_loss": 1.041021704673767, |
|
"eval_runtime": 12.8807, |
|
"eval_samples_per_second": 155.348, |
|
"eval_steps_per_second": 19.486, |
|
"eval_uas": 91.59807547019763, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 16.84, |
|
"learning_rate": 4.5122147651006716e-05, |
|
"loss": 0.0242, |
|
"step": 6600 |
|
}, |
|
{ |
|
"epoch": 17.09, |
|
"learning_rate": 4.4585234899328864e-05, |
|
"loss": 0.0252, |
|
"step": 6700 |
|
}, |
|
{ |
|
"epoch": 17.35, |
|
"learning_rate": 4.404832214765101e-05, |
|
"loss": 0.0212, |
|
"step": 6800 |
|
}, |
|
{ |
|
"epoch": 17.6, |
|
"learning_rate": 4.351140939597316e-05, |
|
"loss": 0.0221, |
|
"step": 6900 |
|
}, |
|
{ |
|
"epoch": 17.86, |
|
"learning_rate": 4.297986577181209e-05, |
|
"loss": 0.0245, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 17.86, |
|
"eval_las": 89.0651715773987, |
|
"eval_loss": 1.0336178541183472, |
|
"eval_runtime": 12.8848, |
|
"eval_samples_per_second": 155.299, |
|
"eval_steps_per_second": 19.48, |
|
"eval_uas": 91.42709451668058, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 18.11, |
|
"learning_rate": 4.244295302013423e-05, |
|
"loss": 0.0169, |
|
"step": 7100 |
|
}, |
|
{ |
|
"epoch": 18.37, |
|
"learning_rate": 4.190604026845638e-05, |
|
"loss": 0.0188, |
|
"step": 7200 |
|
}, |
|
{ |
|
"epoch": 18.62, |
|
"learning_rate": 4.1369127516778525e-05, |
|
"loss": 0.0193, |
|
"step": 7300 |
|
}, |
|
{ |
|
"epoch": 18.88, |
|
"learning_rate": 4.0832214765100673e-05, |
|
"loss": 0.0197, |
|
"step": 7400 |
|
}, |
|
{ |
|
"epoch": 19.13, |
|
"learning_rate": 4.029530201342282e-05, |
|
"loss": 0.016, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 19.13, |
|
"eval_las": 89.12481609606743, |
|
"eval_loss": 1.1899999380111694, |
|
"eval_runtime": 12.8943, |
|
"eval_samples_per_second": 155.185, |
|
"eval_steps_per_second": 19.466, |
|
"eval_uas": 91.38733150423477, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 19.13, |
|
"step": 7500, |
|
"total_flos": 4.007127707268403e+16, |
|
"train_loss": 0.15542098660469056, |
|
"train_runtime": 4224.7831, |
|
"train_samples_per_second": 113.615, |
|
"train_steps_per_second": 3.55 |
|
} |
|
], |
|
"max_steps": 15000, |
|
"num_train_epochs": 39, |
|
"total_flos": 4.007127707268403e+16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|