| { | |
| "best_metric": 94.51452754944249, | |
| "best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/mbert/mbert-base-finetuned-parsing-ud-Japanese-GSD/checkpoint-10000", | |
| "epoch": 56.56108597285068, | |
| "global_step": 12500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 7.76e-05, | |
| "loss": 2.8808, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 7.947919463087248e-05, | |
| "loss": 0.5888, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 7.894228187919463e-05, | |
| "loss": 0.387, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 7.840536912751678e-05, | |
| "loss": 0.3305, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 7.786845637583893e-05, | |
| "loss": 0.2643, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "eval_las": 92.13803206641165, | |
| "eval_loss": 0.37336885929107666, | |
| "eval_runtime": 3.7431, | |
| "eval_samples_per_second": 135.449, | |
| "eval_steps_per_second": 17.098, | |
| "eval_uas": 94.02620655977863, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 7.733154362416108e-05, | |
| "loss": 0.22, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 7.679463087248322e-05, | |
| "loss": 0.1981, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 7.625771812080537e-05, | |
| "loss": 0.1498, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 7.572080536912752e-05, | |
| "loss": 0.1522, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 7.518389261744967e-05, | |
| "loss": 0.1105, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "eval_las": 93.34255717424921, | |
| "eval_loss": 0.3645283281803131, | |
| "eval_runtime": 3.7438, | |
| "eval_samples_per_second": 135.422, | |
| "eval_steps_per_second": 17.095, | |
| "eval_uas": 94.83193619272402, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "learning_rate": 7.464697986577182e-05, | |
| "loss": 0.1191, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 7.411006711409397e-05, | |
| "loss": 0.0871, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "learning_rate": 7.357315436241611e-05, | |
| "loss": 0.09, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "learning_rate": 7.303624161073826e-05, | |
| "loss": 0.0724, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "learning_rate": 7.249932885906041e-05, | |
| "loss": 0.0729, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "eval_las": 93.60299503540327, | |
| "eval_loss": 0.42577120661735535, | |
| "eval_runtime": 3.7422, | |
| "eval_samples_per_second": 135.482, | |
| "eval_steps_per_second": 17.102, | |
| "eval_uas": 94.98657117278424, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.24, | |
| "learning_rate": 7.196241610738256e-05, | |
| "loss": 0.069, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 7.142550335570471e-05, | |
| "loss": 0.0616, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 8.14, | |
| "learning_rate": 7.088859060402686e-05, | |
| "loss": 0.0579, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 8.6, | |
| "learning_rate": 7.0351677852349e-05, | |
| "loss": 0.0512, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 6.981476510067114e-05, | |
| "loss": 0.05, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "eval_las": 93.67624318385285, | |
| "eval_loss": 0.5269723534584045, | |
| "eval_runtime": 3.7443, | |
| "eval_samples_per_second": 135.405, | |
| "eval_steps_per_second": 17.093, | |
| "eval_uas": 95.07609668755596, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "learning_rate": 6.927785234899329e-05, | |
| "loss": 0.0453, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "learning_rate": 6.874093959731543e-05, | |
| "loss": 0.0459, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 10.41, | |
| "learning_rate": 6.820402684563758e-05, | |
| "loss": 0.0399, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 10.86, | |
| "learning_rate": 6.766711409395973e-05, | |
| "loss": 0.04, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 11.31, | |
| "learning_rate": 6.713020134228188e-05, | |
| "loss": 0.0353, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 11.31, | |
| "eval_las": 93.99365182713437, | |
| "eval_loss": 0.5733376145362854, | |
| "eval_runtime": 3.7349, | |
| "eval_samples_per_second": 135.745, | |
| "eval_steps_per_second": 17.136, | |
| "eval_uas": 95.26328640026044, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 11.76, | |
| "learning_rate": 6.659328859060403e-05, | |
| "loss": 0.0378, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 12.22, | |
| "learning_rate": 6.605637583892618e-05, | |
| "loss": 0.0355, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 12.67, | |
| "learning_rate": 6.551946308724832e-05, | |
| "loss": 0.0324, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 13.12, | |
| "learning_rate": 6.498255033557047e-05, | |
| "loss": 0.0316, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 13.57, | |
| "learning_rate": 6.444563758389262e-05, | |
| "loss": 0.0307, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 13.57, | |
| "eval_las": 93.83901684707413, | |
| "eval_loss": 0.5204811096191406, | |
| "eval_runtime": 3.7373, | |
| "eval_samples_per_second": 135.659, | |
| "eval_steps_per_second": 17.125, | |
| "eval_uas": 95.23073166761617, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "learning_rate": 6.390872483221477e-05, | |
| "loss": 0.0293, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 14.48, | |
| "learning_rate": 6.337181208053692e-05, | |
| "loss": 0.0271, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 14.93, | |
| "learning_rate": 6.283489932885907e-05, | |
| "loss": 0.0286, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 6.229798657718121e-05, | |
| "loss": 0.0232, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 15.84, | |
| "learning_rate": 6.176107382550336e-05, | |
| "loss": 0.0251, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 15.84, | |
| "eval_las": 93.86343289655734, | |
| "eval_loss": 0.58524090051651, | |
| "eval_runtime": 3.7398, | |
| "eval_samples_per_second": 135.568, | |
| "eval_steps_per_second": 17.113, | |
| "eval_uas": 95.19817693497193, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 16.29, | |
| "learning_rate": 6.122416107382551e-05, | |
| "loss": 0.023, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 16.74, | |
| "learning_rate": 6.068724832214766e-05, | |
| "loss": 0.0224, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 17.19, | |
| "learning_rate": 6.015033557046981e-05, | |
| "loss": 0.0202, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 17.65, | |
| "learning_rate": 5.9613422818791955e-05, | |
| "loss": 0.0223, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 18.1, | |
| "learning_rate": 5.90765100671141e-05, | |
| "loss": 0.0208, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 18.1, | |
| "eval_las": 93.76576869862457, | |
| "eval_loss": 0.6501343250274658, | |
| "eval_runtime": 3.7407, | |
| "eval_samples_per_second": 135.537, | |
| "eval_steps_per_second": 17.109, | |
| "eval_uas": 95.01912590542851, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 18.55, | |
| "learning_rate": 5.853959731543625e-05, | |
| "loss": 0.019, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 5.80026845637584e-05, | |
| "loss": 0.0191, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 19.46, | |
| "learning_rate": 5.7465771812080534e-05, | |
| "loss": 0.0166, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 19.91, | |
| "learning_rate": 5.692885906040268e-05, | |
| "loss": 0.0175, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 20.36, | |
| "learning_rate": 5.639194630872483e-05, | |
| "loss": 0.0183, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 20.36, | |
| "eval_las": 94.2378123219663, | |
| "eval_loss": 0.5928093194961548, | |
| "eval_runtime": 3.7386, | |
| "eval_samples_per_second": 135.611, | |
| "eval_steps_per_second": 17.119, | |
| "eval_uas": 95.46675347928705, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 20.81, | |
| "learning_rate": 5.5855033557046986e-05, | |
| "loss": 0.0162, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 21.27, | |
| "learning_rate": 5.5318120805369134e-05, | |
| "loss": 0.0158, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 21.72, | |
| "learning_rate": 5.478120805369128e-05, | |
| "loss": 0.0176, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 22.17, | |
| "learning_rate": 5.424429530201343e-05, | |
| "loss": 0.0136, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 22.62, | |
| "learning_rate": 5.370738255033558e-05, | |
| "loss": 0.0146, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 22.62, | |
| "eval_las": 93.96109709449011, | |
| "eval_loss": 0.6554161906242371, | |
| "eval_runtime": 3.737, | |
| "eval_samples_per_second": 135.671, | |
| "eval_steps_per_second": 17.126, | |
| "eval_uas": 95.27956376658257, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 23.08, | |
| "learning_rate": 5.317046979865773e-05, | |
| "loss": 0.0132, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 23.53, | |
| "learning_rate": 5.2633557046979875e-05, | |
| "loss": 0.0137, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 23.98, | |
| "learning_rate": 5.2096644295302024e-05, | |
| "loss": 0.0163, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 24.43, | |
| "learning_rate": 5.155973154362417e-05, | |
| "loss": 0.0115, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 24.89, | |
| "learning_rate": 5.1022818791946307e-05, | |
| "loss": 0.0129, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 24.89, | |
| "eval_las": 94.2622283714495, | |
| "eval_loss": 0.657804548740387, | |
| "eval_runtime": 3.7329, | |
| "eval_samples_per_second": 135.821, | |
| "eval_steps_per_second": 17.145, | |
| "eval_uas": 95.51558557825344, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 25.34, | |
| "learning_rate": 5.0485906040268455e-05, | |
| "loss": 0.0133, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 25.79, | |
| "learning_rate": 4.99489932885906e-05, | |
| "loss": 0.0118, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 26.24, | |
| "learning_rate": 4.941208053691275e-05, | |
| "loss": 0.0123, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 26.7, | |
| "learning_rate": 4.88751677852349e-05, | |
| "loss": 0.0094, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 27.15, | |
| "learning_rate": 4.833825503355705e-05, | |
| "loss": 0.0099, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 27.15, | |
| "eval_las": 94.1238707577114, | |
| "eval_loss": 0.6814541220664978, | |
| "eval_runtime": 3.7329, | |
| "eval_samples_per_second": 135.82, | |
| "eval_steps_per_second": 17.145, | |
| "eval_uas": 95.31211849922683, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 27.6, | |
| "learning_rate": 4.7801342281879196e-05, | |
| "loss": 0.0112, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 28.05, | |
| "learning_rate": 4.726442953020135e-05, | |
| "loss": 0.01, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 28.51, | |
| "learning_rate": 4.67275167785235e-05, | |
| "loss": 0.0096, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 28.96, | |
| "learning_rate": 4.619060402684565e-05, | |
| "loss": 0.0116, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 29.41, | |
| "learning_rate": 4.5653691275167796e-05, | |
| "loss": 0.0081, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 29.41, | |
| "eval_las": 94.34361520306015, | |
| "eval_loss": 0.6585590839385986, | |
| "eval_runtime": 3.7358, | |
| "eval_samples_per_second": 135.713, | |
| "eval_steps_per_second": 17.131, | |
| "eval_uas": 95.5969724098641, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 29.86, | |
| "learning_rate": 4.511677852348993e-05, | |
| "loss": 0.0082, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 30.32, | |
| "learning_rate": 4.4585234899328864e-05, | |
| "loss": 0.0081, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 30.77, | |
| "learning_rate": 4.404832214765101e-05, | |
| "loss": 0.0088, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 31.22, | |
| "learning_rate": 4.351140939597316e-05, | |
| "loss": 0.009, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 31.67, | |
| "learning_rate": 4.297449664429531e-05, | |
| "loss": 0.0088, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 31.67, | |
| "eval_las": 94.32733783673801, | |
| "eval_loss": 0.6393760442733765, | |
| "eval_runtime": 3.7286, | |
| "eval_samples_per_second": 135.974, | |
| "eval_steps_per_second": 17.164, | |
| "eval_uas": 95.46675347928705, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 32.13, | |
| "learning_rate": 4.243758389261746e-05, | |
| "loss": 0.008, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 32.58, | |
| "learning_rate": 4.1900671140939605e-05, | |
| "loss": 0.0065, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 33.03, | |
| "learning_rate": 4.1363758389261754e-05, | |
| "loss": 0.0066, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 33.48, | |
| "learning_rate": 4.0826845637583895e-05, | |
| "loss": 0.0052, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 33.94, | |
| "learning_rate": 4.028993288590604e-05, | |
| "loss": 0.0057, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 33.94, | |
| "eval_las": 94.34361520306015, | |
| "eval_loss": 0.7660797834396362, | |
| "eval_runtime": 3.7339, | |
| "eval_samples_per_second": 135.781, | |
| "eval_steps_per_second": 17.14, | |
| "eval_uas": 95.51558557825344, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 34.39, | |
| "learning_rate": 3.975302013422819e-05, | |
| "loss": 0.0065, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 34.84, | |
| "learning_rate": 3.921610738255034e-05, | |
| "loss": 0.0067, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 35.29, | |
| "learning_rate": 3.867919463087249e-05, | |
| "loss": 0.0064, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 35.75, | |
| "learning_rate": 3.8142281879194636e-05, | |
| "loss": 0.0077, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 36.2, | |
| "learning_rate": 3.7605369127516784e-05, | |
| "loss": 0.0054, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 36.2, | |
| "eval_las": 94.29478310409377, | |
| "eval_loss": 0.7726401686668396, | |
| "eval_runtime": 3.7388, | |
| "eval_samples_per_second": 135.605, | |
| "eval_steps_per_second": 17.118, | |
| "eval_uas": 95.45861479612599, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 36.65, | |
| "learning_rate": 3.706845637583893e-05, | |
| "loss": 0.0064, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 37.1, | |
| "learning_rate": 3.653154362416108e-05, | |
| "loss": 0.0058, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 37.56, | |
| "learning_rate": 3.599463087248322e-05, | |
| "loss": 0.006, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 38.01, | |
| "learning_rate": 3.545771812080537e-05, | |
| "loss": 0.0061, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 38.46, | |
| "learning_rate": 3.492080536912752e-05, | |
| "loss": 0.0044, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 38.46, | |
| "eval_las": 94.43314071783185, | |
| "eval_loss": 0.7485831379890442, | |
| "eval_runtime": 3.7354, | |
| "eval_samples_per_second": 135.729, | |
| "eval_steps_per_second": 17.133, | |
| "eval_uas": 95.57255636038089, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 38.91, | |
| "learning_rate": 3.438389261744967e-05, | |
| "loss": 0.0043, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 39.37, | |
| "learning_rate": 3.3846979865771815e-05, | |
| "loss": 0.0033, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 39.82, | |
| "learning_rate": 3.3310067114093964e-05, | |
| "loss": 0.004, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 40.27, | |
| "learning_rate": 3.277315436241611e-05, | |
| "loss": 0.0049, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 40.72, | |
| "learning_rate": 3.223624161073826e-05, | |
| "loss": 0.0038, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 40.72, | |
| "eval_las": 94.30292178725482, | |
| "eval_loss": 0.7324322462081909, | |
| "eval_runtime": 3.7333, | |
| "eval_samples_per_second": 135.804, | |
| "eval_steps_per_second": 17.143, | |
| "eval_uas": 95.50744689509237, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 41.18, | |
| "learning_rate": 3.169932885906041e-05, | |
| "loss": 0.0042, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 41.63, | |
| "learning_rate": 3.1162416107382557e-05, | |
| "loss": 0.0042, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 42.08, | |
| "learning_rate": 3.0625503355704705e-05, | |
| "loss": 0.0033, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 42.53, | |
| "learning_rate": 3.0088590604026846e-05, | |
| "loss": 0.0031, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 42.99, | |
| "learning_rate": 2.9551677852348995e-05, | |
| "loss": 0.0035, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 42.99, | |
| "eval_las": 94.24595100512737, | |
| "eval_loss": 0.8114320635795593, | |
| "eval_runtime": 3.7335, | |
| "eval_samples_per_second": 135.799, | |
| "eval_steps_per_second": 17.142, | |
| "eval_uas": 95.42606006348173, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 43.44, | |
| "learning_rate": 2.9014765100671143e-05, | |
| "loss": 0.0027, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 43.89, | |
| "learning_rate": 2.848322147651007e-05, | |
| "loss": 0.0041, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 44.34, | |
| "learning_rate": 2.7946308724832218e-05, | |
| "loss": 0.0029, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 44.8, | |
| "learning_rate": 2.7409395973154366e-05, | |
| "loss": 0.0036, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 45.25, | |
| "learning_rate": 2.6872483221476514e-05, | |
| "loss": 0.0034, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 45.25, | |
| "eval_las": 94.51452754944249, | |
| "eval_loss": 0.7882774472236633, | |
| "eval_runtime": 3.7243, | |
| "eval_samples_per_second": 136.135, | |
| "eval_steps_per_second": 17.185, | |
| "eval_uas": 95.56441767721982, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 45.7, | |
| "learning_rate": 2.6335570469798663e-05, | |
| "loss": 0.0035, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 46.15, | |
| "learning_rate": 2.5798657718120804e-05, | |
| "loss": 0.0027, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 46.61, | |
| "learning_rate": 2.5261744966442952e-05, | |
| "loss": 0.0023, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 47.06, | |
| "learning_rate": 2.47248322147651e-05, | |
| "loss": 0.0024, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 47.51, | |
| "learning_rate": 2.4187919463087252e-05, | |
| "loss": 0.0025, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 47.51, | |
| "eval_las": 94.44941808415398, | |
| "eval_loss": 0.8562086224555969, | |
| "eval_runtime": 3.7341, | |
| "eval_samples_per_second": 135.776, | |
| "eval_steps_per_second": 17.139, | |
| "eval_uas": 95.5237242614145, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 47.96, | |
| "learning_rate": 2.36510067114094e-05, | |
| "loss": 0.0032, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 48.42, | |
| "learning_rate": 2.311409395973155e-05, | |
| "loss": 0.002, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 48.87, | |
| "learning_rate": 2.257718120805369e-05, | |
| "loss": 0.0017, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 49.32, | |
| "learning_rate": 2.204026845637584e-05, | |
| "loss": 0.0015, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 49.77, | |
| "learning_rate": 2.1503355704697987e-05, | |
| "loss": 0.0013, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 49.77, | |
| "eval_las": 94.51452754944249, | |
| "eval_loss": 0.8181952834129333, | |
| "eval_runtime": 3.7391, | |
| "eval_samples_per_second": 135.594, | |
| "eval_steps_per_second": 17.116, | |
| "eval_uas": 95.64580450883047, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 50.23, | |
| "learning_rate": 2.0966442953020135e-05, | |
| "loss": 0.0015, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 50.68, | |
| "learning_rate": 2.0429530201342283e-05, | |
| "loss": 0.0016, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 51.13, | |
| "learning_rate": 1.989261744966443e-05, | |
| "loss": 0.0014, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 51.58, | |
| "learning_rate": 1.935570469798658e-05, | |
| "loss": 0.0012, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 52.04, | |
| "learning_rate": 1.8818791946308724e-05, | |
| "loss": 0.0013, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 52.04, | |
| "eval_las": 94.3761699357044, | |
| "eval_loss": 0.8620641231536865, | |
| "eval_runtime": 3.734, | |
| "eval_samples_per_second": 135.778, | |
| "eval_steps_per_second": 17.14, | |
| "eval_uas": 95.47489216244811, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 52.49, | |
| "learning_rate": 1.8281879194630873e-05, | |
| "loss": 0.0012, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 52.94, | |
| "learning_rate": 1.774496644295302e-05, | |
| "loss": 0.0014, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 53.39, | |
| "learning_rate": 1.720805369127517e-05, | |
| "loss": 0.0005, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 53.85, | |
| "learning_rate": 1.6671140939597317e-05, | |
| "loss": 0.0022, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 54.3, | |
| "learning_rate": 1.6139597315436244e-05, | |
| "loss": 0.0007, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 54.3, | |
| "eval_las": 94.4250020346708, | |
| "eval_loss": 0.891247570514679, | |
| "eval_runtime": 3.7334, | |
| "eval_samples_per_second": 135.801, | |
| "eval_steps_per_second": 17.143, | |
| "eval_uas": 95.5481403108977, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 54.75, | |
| "learning_rate": 1.560268456375839e-05, | |
| "loss": 0.0009, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 55.2, | |
| "learning_rate": 1.5065771812080539e-05, | |
| "loss": 0.0008, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 55.66, | |
| "learning_rate": 1.4528859060402685e-05, | |
| "loss": 0.0005, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 56.11, | |
| "learning_rate": 1.3991946308724834e-05, | |
| "loss": 0.0007, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 56.56, | |
| "learning_rate": 1.3455033557046982e-05, | |
| "loss": 0.0009, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 56.56, | |
| "eval_las": 94.44941808415398, | |
| "eval_loss": 0.9227140545845032, | |
| "eval_runtime": 3.7344, | |
| "eval_samples_per_second": 135.765, | |
| "eval_steps_per_second": 17.138, | |
| "eval_uas": 95.5237242614145, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 56.56, | |
| "step": 12500, | |
| "total_flos": 6.65850338806825e+16, | |
| "train_loss": 0.05763526131868362, | |
| "train_runtime": 7165.6139, | |
| "train_samples_per_second": 66.987, | |
| "train_steps_per_second": 2.093 | |
| } | |
| ], | |
| "max_steps": 15000, | |
| "num_train_epochs": 68, | |
| "total_flos": 6.65850338806825e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |