| { | |
| "best_metric": 82.41013261020537, | |
| "best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/mbert/mbert-base-finetuned-parsing-ud-Arabic-PADT/checkpoint-3000", | |
| "epoch": 28.94736842105263, | |
| "global_step": 5500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 7.76e-05, | |
| "loss": 3.3094, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 7.947919463087248e-05, | |
| "loss": 1.0202, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 7.894228187919463e-05, | |
| "loss": 0.7642, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 7.840536912751678e-05, | |
| "loss": 0.697, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 7.786845637583893e-05, | |
| "loss": 0.5389, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "eval_las": 81.76527001554284, | |
| "eval_loss": 0.8606711626052856, | |
| "eval_runtime": 7.3104, | |
| "eval_samples_per_second": 124.343, | |
| "eval_steps_per_second": 15.594, | |
| "eval_uas": 86.67614669797281, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 7.733154362416108e-05, | |
| "loss": 0.5056, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 7.679463087248322e-05, | |
| "loss": 0.408, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 7.625771812080537e-05, | |
| "loss": 0.3727, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 7.572080536912752e-05, | |
| "loss": 0.312, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "learning_rate": 7.518389261744967e-05, | |
| "loss": 0.2764, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "eval_las": 81.93723337411951, | |
| "eval_loss": 1.1254280805587769, | |
| "eval_runtime": 7.2968, | |
| "eval_samples_per_second": 124.575, | |
| "eval_steps_per_second": 15.623, | |
| "eval_uas": 87.01676642746122, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 7.464697986577182e-05, | |
| "loss": 0.2391, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "learning_rate": 7.411006711409397e-05, | |
| "loss": 0.2091, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.84, | |
| "learning_rate": 7.357315436241611e-05, | |
| "loss": 0.1907, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 7.303624161073826e-05, | |
| "loss": 0.1597, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "learning_rate": 7.249932885906041e-05, | |
| "loss": 0.1574, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "eval_las": 81.97691722609875, | |
| "eval_loss": 1.3033808469772339, | |
| "eval_runtime": 7.3011, | |
| "eval_samples_per_second": 124.502, | |
| "eval_steps_per_second": 15.614, | |
| "eval_uas": 86.96385462482225, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 8.42, | |
| "learning_rate": 7.196241610738256e-05, | |
| "loss": 0.132, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 8.95, | |
| "learning_rate": 7.142550335570471e-05, | |
| "loss": 0.1349, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "learning_rate": 7.088859060402686e-05, | |
| "loss": 0.1038, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 7.0351677852349e-05, | |
| "loss": 0.117, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 10.53, | |
| "learning_rate": 6.981476510067114e-05, | |
| "loss": 0.0903, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 10.53, | |
| "eval_las": 82.0099871027481, | |
| "eval_loss": 1.4602577686309814, | |
| "eval_runtime": 7.3008, | |
| "eval_samples_per_second": 124.507, | |
| "eval_steps_per_second": 15.615, | |
| "eval_uas": 87.14573894639372, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 6.927785234899329e-05, | |
| "loss": 0.1005, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 11.58, | |
| "learning_rate": 6.874093959731543e-05, | |
| "loss": 0.0844, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 12.11, | |
| "learning_rate": 6.820402684563758e-05, | |
| "loss": 0.0867, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 12.63, | |
| "learning_rate": 6.766711409395973e-05, | |
| "loss": 0.0742, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 13.16, | |
| "learning_rate": 6.713020134228188e-05, | |
| "loss": 0.0739, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 13.16, | |
| "eval_las": 81.9537683124442, | |
| "eval_loss": 1.6436030864715576, | |
| "eval_runtime": 7.3127, | |
| "eval_samples_per_second": 124.304, | |
| "eval_steps_per_second": 15.589, | |
| "eval_uas": 87.02668739045603, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 13.68, | |
| "learning_rate": 6.659328859060403e-05, | |
| "loss": 0.0676, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 14.21, | |
| "learning_rate": 6.605637583892618e-05, | |
| "loss": 0.0669, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 14.74, | |
| "learning_rate": 6.551946308724832e-05, | |
| "loss": 0.0645, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 15.26, | |
| "learning_rate": 6.498255033557047e-05, | |
| "loss": 0.0588, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 15.79, | |
| "learning_rate": 6.444563758389262e-05, | |
| "loss": 0.0564, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 15.79, | |
| "eval_las": 82.41013261020537, | |
| "eval_loss": 1.7569572925567627, | |
| "eval_runtime": 7.2931, | |
| "eval_samples_per_second": 124.639, | |
| "eval_steps_per_second": 15.631, | |
| "eval_uas": 87.33754423096002, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 16.32, | |
| "learning_rate": 6.390872483221477e-05, | |
| "loss": 0.0517, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 16.84, | |
| "learning_rate": 6.337181208053692e-05, | |
| "loss": 0.0542, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 17.37, | |
| "learning_rate": 6.283489932885907e-05, | |
| "loss": 0.0473, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 17.89, | |
| "learning_rate": 6.229798657718121e-05, | |
| "loss": 0.0494, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 18.42, | |
| "learning_rate": 6.176107382550336e-05, | |
| "loss": 0.0452, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 18.42, | |
| "eval_las": 81.9537683124442, | |
| "eval_loss": 1.7844057083129883, | |
| "eval_runtime": 7.3028, | |
| "eval_samples_per_second": 124.473, | |
| "eval_steps_per_second": 15.61, | |
| "eval_uas": 86.84149608121962, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 18.95, | |
| "learning_rate": 6.122416107382551e-05, | |
| "loss": 0.0473, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 19.47, | |
| "learning_rate": 6.068724832214766e-05, | |
| "loss": 0.0402, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 6.015033557046981e-05, | |
| "loss": 0.0426, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 20.53, | |
| "learning_rate": 5.9613422818791955e-05, | |
| "loss": 0.0375, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 21.05, | |
| "learning_rate": 5.90765100671141e-05, | |
| "loss": 0.0402, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 21.05, | |
| "eval_las": 82.2844670789378, | |
| "eval_loss": 1.867564082145691, | |
| "eval_runtime": 7.2792, | |
| "eval_samples_per_second": 124.877, | |
| "eval_steps_per_second": 15.661, | |
| "eval_uas": 87.14573894639372, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 21.58, | |
| "learning_rate": 5.853959731543625e-05, | |
| "loss": 0.037, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 22.11, | |
| "learning_rate": 5.80026845637584e-05, | |
| "loss": 0.0369, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 22.63, | |
| "learning_rate": 5.7465771812080534e-05, | |
| "loss": 0.0348, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 23.16, | |
| "learning_rate": 5.692885906040268e-05, | |
| "loss": 0.0359, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 23.68, | |
| "learning_rate": 5.639194630872483e-05, | |
| "loss": 0.0329, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 23.68, | |
| "eval_las": 82.05297794239227, | |
| "eval_loss": 1.9352997541427612, | |
| "eval_runtime": 7.2823, | |
| "eval_samples_per_second": 124.824, | |
| "eval_steps_per_second": 15.654, | |
| "eval_uas": 87.08952015608982, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 24.21, | |
| "learning_rate": 5.5855033557046986e-05, | |
| "loss": 0.0311, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 24.74, | |
| "learning_rate": 5.532348993288591e-05, | |
| "loss": 0.0309, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 25.26, | |
| "learning_rate": 5.478657718120806e-05, | |
| "loss": 0.0302, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 25.79, | |
| "learning_rate": 5.424966442953021e-05, | |
| "loss": 0.0281, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 26.32, | |
| "learning_rate": 5.371275167785236e-05, | |
| "loss": 0.0286, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 26.32, | |
| "eval_las": 82.0463639670624, | |
| "eval_loss": 1.9560511112213135, | |
| "eval_runtime": 7.3048, | |
| "eval_samples_per_second": 124.438, | |
| "eval_steps_per_second": 15.606, | |
| "eval_uas": 87.08952015608982, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 26.84, | |
| "learning_rate": 5.31758389261745e-05, | |
| "loss": 0.027, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 27.37, | |
| "learning_rate": 5.263892617449665e-05, | |
| "loss": 0.0265, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 27.89, | |
| "learning_rate": 5.2102013422818795e-05, | |
| "loss": 0.0268, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 28.42, | |
| "learning_rate": 5.1565100671140944e-05, | |
| "loss": 0.0251, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 28.95, | |
| "learning_rate": 5.102818791946309e-05, | |
| "loss": 0.0245, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 28.95, | |
| "eval_las": 82.21832732563908, | |
| "eval_loss": 2.1567718982696533, | |
| "eval_runtime": 7.2888, | |
| "eval_samples_per_second": 124.712, | |
| "eval_steps_per_second": 15.64, | |
| "eval_uas": 87.03660835345084, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 28.95, | |
| "step": 5500, | |
| "total_flos": 2.936455296878592e+16, | |
| "train_loss": 0.20698133672367441, | |
| "train_runtime": 3488.4911, | |
| "train_samples_per_second": 137.595, | |
| "train_steps_per_second": 4.3 | |
| } | |
| ], | |
| "max_steps": 15000, | |
| "num_train_epochs": 79, | |
| "total_flos": 2.936455296878592e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |