{ "best_metric": null, "best_model_checkpoint": null, "epoch": 20.0, "eval_steps": 500, "global_step": 16920, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0, "learning_rate": 5.91016548463357e-08, "loss": 0.0002, "step": 1 }, { "epoch": 0.1, "learning_rate": 5.023640661938535e-06, "loss": 0.0009, "step": 85 }, { "epoch": 0.2, "learning_rate": 1.004728132387707e-05, "loss": 0.0012, "step": 170 }, { "epoch": 0.3, "learning_rate": 1.5070921985815603e-05, "loss": 0.0012, "step": 255 }, { "epoch": 0.4, "learning_rate": 2.009456264775414e-05, "loss": 0.0007, "step": 340 }, { "epoch": 0.5, "learning_rate": 2.5118203309692672e-05, "loss": 0.0012, "step": 425 }, { "epoch": 0.6, "learning_rate": 3.0141843971631207e-05, "loss": 0.0014, "step": 510 }, { "epoch": 0.7, "learning_rate": 3.516548463356974e-05, "loss": 0.0016, "step": 595 }, { "epoch": 0.8, "learning_rate": 4.018912529550828e-05, "loss": 0.0018, "step": 680 }, { "epoch": 0.9, "learning_rate": 4.5212765957446815e-05, "loss": 0.0013, "step": 765 }, { "epoch": 1.0, "eval_loss": 0.005299737676978111, "eval_max_distance": 3, "eval_mean_distance": 0, "eval_runtime": 34.0947, "eval_samples_per_second": 7.626, "eval_steps_per_second": 0.528, "step": 846 }, { "epoch": 1.0, "learning_rate": 5.0236406619385343e-05, "loss": 0.001, "step": 850 }, { "epoch": 1.11, "learning_rate": 5.526004728132388e-05, "loss": 0.0008, "step": 935 }, { "epoch": 1.21, "learning_rate": 6.0283687943262414e-05, "loss": 0.0012, "step": 1020 }, { "epoch": 1.31, "learning_rate": 6.530732860520094e-05, "loss": 0.0015, "step": 1105 }, { "epoch": 1.41, "learning_rate": 7.033096926713948e-05, "loss": 0.0019, "step": 1190 }, { "epoch": 1.51, "learning_rate": 7.535460992907801e-05, "loss": 0.001, "step": 1275 }, { "epoch": 1.61, "learning_rate": 8.037825059101656e-05, "loss": 0.0012, "step": 1360 }, { "epoch": 1.71, "learning_rate": 8.54018912529551e-05, "loss": 0.0007, "step": 1445 }, { "epoch": 1.81, "learning_rate": 9.042553191489363e-05, "loss": 0.0008, "step": 1530 }, { "epoch": 1.91, "learning_rate": 9.544917257683215e-05, "loss": 0.0011, "step": 1615 }, { "epoch": 2.0, "eval_loss": 0.004299119580537081, "eval_max_distance": 3, "eval_mean_distance": 0, "eval_runtime": 30.4314, "eval_samples_per_second": 8.544, "eval_steps_per_second": 0.591, "step": 1692 }, { "epoch": 2.01, "learning_rate": 9.994746519569215e-05, "loss": 0.0007, "step": 1700 }, { "epoch": 2.11, "learning_rate": 9.93892828999212e-05, "loss": 0.001, "step": 1785 }, { "epoch": 2.21, "learning_rate": 9.883110060415024e-05, "loss": 0.0008, "step": 1870 }, { "epoch": 2.31, "learning_rate": 9.82729183083793e-05, "loss": 0.0015, "step": 1955 }, { "epoch": 2.41, "learning_rate": 9.771473601260836e-05, "loss": 0.0011, "step": 2040 }, { "epoch": 2.51, "learning_rate": 9.715655371683741e-05, "loss": 0.0011, "step": 2125 }, { "epoch": 2.61, "learning_rate": 9.659837142106646e-05, "loss": 0.0008, "step": 2210 }, { "epoch": 2.71, "learning_rate": 9.604018912529551e-05, "loss": 0.0008, "step": 2295 }, { "epoch": 2.81, "learning_rate": 9.548200682952457e-05, "loss": 0.0009, "step": 2380 }, { "epoch": 2.91, "learning_rate": 9.492382453375361e-05, "loss": 0.0009, "step": 2465 }, { "epoch": 3.0, "eval_loss": 0.004937924910336733, "eval_max_distance": 3, "eval_mean_distance": 0, "eval_runtime": 30.2916, "eval_samples_per_second": 8.583, "eval_steps_per_second": 0.594, "step": 2538 }, { "epoch": 3.01, "learning_rate": 9.436564223798267e-05, "loss": 0.0013, "step": 2550 }, { "epoch": 3.11, "learning_rate": 9.380745994221172e-05, "loss": 0.0008, "step": 2635 }, { "epoch": 3.22, "learning_rate": 9.324927764644077e-05, "loss": 0.0006, "step": 2720 }, { "epoch": 3.32, "learning_rate": 9.269109535066982e-05, "loss": 0.0009, "step": 2805 }, { "epoch": 3.42, "learning_rate": 9.213291305489888e-05, "loss": 0.0007, "step": 2890 }, { "epoch": 3.52, "learning_rate": 9.157473075912792e-05, "loss": 0.0007, "step": 2975 }, { "epoch": 3.62, "learning_rate": 9.101654846335698e-05, "loss": 0.001, "step": 3060 }, { "epoch": 3.72, "learning_rate": 9.045836616758603e-05, "loss": 0.0007, "step": 3145 }, { "epoch": 3.82, "learning_rate": 8.990018387181508e-05, "loss": 0.0009, "step": 3230 }, { "epoch": 3.92, "learning_rate": 8.934200157604413e-05, "loss": 0.0008, "step": 3315 }, { "epoch": 4.0, "eval_loss": 0.004295715596526861, "eval_max_distance": 3, "eval_mean_distance": 0, "eval_runtime": 30.309, "eval_samples_per_second": 8.578, "eval_steps_per_second": 0.594, "step": 3384 }, { "epoch": 4.02, "learning_rate": 8.878381928027319e-05, "loss": 0.0009, "step": 3400 }, { "epoch": 4.12, "learning_rate": 8.822563698450223e-05, "loss": 0.0007, "step": 3485 }, { "epoch": 4.22, "learning_rate": 8.766745468873129e-05, "loss": 0.0005, "step": 3570 }, { "epoch": 4.32, "learning_rate": 8.710927239296034e-05, "loss": 0.0004, "step": 3655 }, { "epoch": 4.42, "learning_rate": 8.655109009718939e-05, "loss": 0.0009, "step": 3740 }, { "epoch": 4.52, "learning_rate": 8.599290780141844e-05, "loss": 0.0009, "step": 3825 }, { "epoch": 4.62, "learning_rate": 8.54347255056475e-05, "loss": 0.0005, "step": 3910 }, { "epoch": 4.72, "learning_rate": 8.487654320987654e-05, "loss": 0.0006, "step": 3995 }, { "epoch": 4.82, "learning_rate": 8.43183609141056e-05, "loss": 0.0004, "step": 4080 }, { "epoch": 4.92, "learning_rate": 8.376017861833465e-05, "loss": 0.0005, "step": 4165 }, { "epoch": 5.0, "eval_loss": 0.004825934302061796, "eval_max_distance": 3, "eval_mean_distance": 0, "eval_runtime": 31.3644, "eval_samples_per_second": 8.29, "eval_steps_per_second": 0.574, "step": 4230 }, { "epoch": 5.02, "learning_rate": 8.320199632256371e-05, "loss": 0.0007, "step": 4250 }, { "epoch": 5.12, "learning_rate": 8.264381402679275e-05, "loss": 0.0007, "step": 4335 }, { "epoch": 5.22, "learning_rate": 8.208563173102181e-05, "loss": 0.0006, "step": 4420 }, { "epoch": 5.33, "learning_rate": 8.152744943525086e-05, "loss": 0.0008, "step": 4505 }, { "epoch": 5.43, "learning_rate": 8.09692671394799e-05, "loss": 0.0009, "step": 4590 }, { "epoch": 5.53, "learning_rate": 8.041108484370896e-05, "loss": 0.0007, "step": 4675 }, { "epoch": 5.63, "learning_rate": 7.985290254793802e-05, "loss": 0.0007, "step": 4760 }, { "epoch": 5.73, "learning_rate": 7.929472025216706e-05, "loss": 0.0004, "step": 4845 }, { "epoch": 5.83, "learning_rate": 7.873653795639612e-05, "loss": 0.0007, "step": 4930 }, { "epoch": 5.93, "learning_rate": 7.817835566062517e-05, "loss": 0.0006, "step": 5015 }, { "epoch": 6.0, "eval_loss": 0.0023884631227701902, "eval_max_distance": 3, "eval_mean_distance": 0, "eval_runtime": 30.4064, "eval_samples_per_second": 8.551, "eval_steps_per_second": 0.592, "step": 5076 }, { "epoch": 6.03, "learning_rate": 7.762017336485422e-05, "loss": 0.0005, "step": 5100 }, { "epoch": 6.13, "learning_rate": 7.706199106908327e-05, "loss": 0.0006, "step": 5185 }, { "epoch": 6.23, "learning_rate": 7.650380877331233e-05, "loss": 0.0005, "step": 5270 }, { "epoch": 6.33, "learning_rate": 7.594562647754137e-05, "loss": 0.0003, "step": 5355 }, { "epoch": 6.43, "learning_rate": 7.538744418177043e-05, "loss": 0.0007, "step": 5440 }, { "epoch": 6.53, "learning_rate": 7.482926188599948e-05, "loss": 0.0008, "step": 5525 }, { "epoch": 6.63, "learning_rate": 7.427107959022853e-05, "loss": 0.0004, "step": 5610 }, { "epoch": 6.73, "learning_rate": 7.371289729445758e-05, "loss": 0.0009, "step": 5695 }, { "epoch": 6.83, "learning_rate": 7.315471499868664e-05, "loss": 0.0008, "step": 5780 }, { "epoch": 6.93, "learning_rate": 7.259653270291568e-05, "loss": 0.0005, "step": 5865 }, { "epoch": 7.0, "eval_loss": 0.002862530527636409, "eval_max_distance": 3, "eval_mean_distance": 0, "eval_runtime": 30.524, "eval_samples_per_second": 8.518, "eval_steps_per_second": 0.59, "step": 5922 }, { "epoch": 7.03, "learning_rate": 7.203835040714474e-05, "loss": 0.0007, "step": 5950 }, { "epoch": 7.13, "learning_rate": 7.14801681113738e-05, "loss": 0.0006, "step": 6035 }, { "epoch": 7.23, "learning_rate": 7.092198581560284e-05, "loss": 0.0005, "step": 6120 }, { "epoch": 7.33, "learning_rate": 7.036380351983189e-05, "loss": 0.0004, "step": 6205 }, { "epoch": 7.43, "learning_rate": 6.980562122406095e-05, "loss": 0.0007, "step": 6290 }, { "epoch": 7.54, "learning_rate": 6.924743892829e-05, "loss": 0.0006, "step": 6375 }, { "epoch": 7.64, "learning_rate": 6.868925663251905e-05, "loss": 0.0005, "step": 6460 }, { "epoch": 7.74, "learning_rate": 6.81310743367481e-05, "loss": 0.0007, "step": 6545 }, { "epoch": 7.84, "learning_rate": 6.757289204097716e-05, "loss": 0.0012, "step": 6630 }, { "epoch": 7.94, "learning_rate": 6.70147097452062e-05, "loss": 0.0008, "step": 6715 }, { "epoch": 8.0, "eval_loss": 0.006056753918528557, "eval_max_distance": 3, "eval_mean_distance": 0, "eval_runtime": 30.8852, "eval_samples_per_second": 8.418, "eval_steps_per_second": 0.583, "step": 6768 }, { "epoch": 8.04, "learning_rate": 6.645652744943526e-05, "loss": 0.0008, "step": 6800 }, { "epoch": 8.14, "learning_rate": 6.589834515366431e-05, "loss": 0.0007, "step": 6885 }, { "epoch": 8.24, "learning_rate": 6.534016285789336e-05, "loss": 0.0005, "step": 6970 }, { "epoch": 8.34, "learning_rate": 6.478198056212241e-05, "loss": 0.0006, "step": 7055 }, { "epoch": 8.44, "learning_rate": 6.422379826635147e-05, "loss": 0.0007, "step": 7140 }, { "epoch": 8.54, "learning_rate": 6.366561597058051e-05, "loss": 0.0004, "step": 7225 }, { "epoch": 8.64, "learning_rate": 6.310743367480957e-05, "loss": 0.0006, "step": 7310 }, { "epoch": 8.74, "learning_rate": 6.254925137903862e-05, "loss": 0.0004, "step": 7395 }, { "epoch": 8.84, "learning_rate": 6.199106908326767e-05, "loss": 0.0005, "step": 7480 }, { "epoch": 8.94, "learning_rate": 6.143288678749672e-05, "loss": 0.0006, "step": 7565 }, { "epoch": 9.0, "eval_loss": 0.005007054656744003, "eval_max_distance": 3, "eval_mean_distance": 0, "eval_runtime": 29.9284, "eval_samples_per_second": 8.687, "eval_steps_per_second": 0.601, "step": 7614 }, { "epoch": 9.04, "learning_rate": 6.087470449172577e-05, "loss": 0.0006, "step": 7650 }, { "epoch": 9.14, "learning_rate": 6.031652219595483e-05, "loss": 0.0007, "step": 7735 }, { "epoch": 9.24, "learning_rate": 5.975833990018388e-05, "loss": 0.0005, "step": 7820 }, { "epoch": 9.34, "learning_rate": 5.9200157604412935e-05, "loss": 0.0006, "step": 7905 }, { "epoch": 9.44, "learning_rate": 5.8641975308641984e-05, "loss": 0.0006, "step": 7990 }, { "epoch": 9.54, "learning_rate": 5.8083793012871034e-05, "loss": 0.0004, "step": 8075 }, { "epoch": 9.65, "learning_rate": 5.752561071710009e-05, "loss": 0.0004, "step": 8160 }, { "epoch": 9.75, "learning_rate": 5.696742842132914e-05, "loss": 0.0004, "step": 8245 }, { "epoch": 9.85, "learning_rate": 5.640924612555819e-05, "loss": 0.0004, "step": 8330 }, { "epoch": 9.95, "learning_rate": 5.585106382978723e-05, "loss": 0.0006, "step": 8415 }, { "epoch": 10.0, "eval_loss": 0.009199680760502815, "eval_max_distance": 3, "eval_mean_distance": 0, "eval_runtime": 29.8643, "eval_samples_per_second": 8.706, "eval_steps_per_second": 0.603, "step": 8460 }, { "epoch": 10.05, "learning_rate": 5.529288153401628e-05, "loss": 0.0005, "step": 8500 }, { "epoch": 10.15, "learning_rate": 5.473469923824534e-05, "loss": 0.0006, "step": 8585 }, { "epoch": 10.25, "learning_rate": 5.4176516942474386e-05, "loss": 0.0005, "step": 8670 }, { "epoch": 10.35, "learning_rate": 5.3618334646703436e-05, "loss": 0.0004, "step": 8755 }, { "epoch": 10.45, "learning_rate": 5.306015235093249e-05, "loss": 0.0004, "step": 8840 }, { "epoch": 10.55, "learning_rate": 5.250197005516154e-05, "loss": 0.0004, "step": 8925 }, { "epoch": 10.65, "learning_rate": 5.19437877593906e-05, "loss": 0.0006, "step": 9010 }, { "epoch": 10.75, "learning_rate": 5.138560546361965e-05, "loss": 0.0005, "step": 9095 }, { "epoch": 10.85, "learning_rate": 5.0827423167848696e-05, "loss": 0.0005, "step": 9180 }, { "epoch": 10.95, "learning_rate": 5.026924087207775e-05, "loss": 0.0004, "step": 9265 }, { "epoch": 11.0, "eval_loss": 0.006612914614379406, "eval_max_distance": 3, "eval_mean_distance": 0, "eval_runtime": 30.0706, "eval_samples_per_second": 8.646, "eval_steps_per_second": 0.599, "step": 9306 }, { "epoch": 11.05, "learning_rate": 4.971105857630681e-05, "loss": 0.0005, "step": 9350 }, { "epoch": 11.15, "learning_rate": 4.915287628053586e-05, "loss": 0.0005, "step": 9435 }, { "epoch": 11.25, "learning_rate": 4.8594693984764914e-05, "loss": 0.0006, "step": 9520 }, { "epoch": 11.35, "learning_rate": 4.8036511688993963e-05, "loss": 0.0004, "step": 9605 }, { "epoch": 11.45, "learning_rate": 4.747832939322301e-05, "loss": 0.0006, "step": 9690 }, { "epoch": 11.55, "learning_rate": 4.692014709745207e-05, "loss": 0.0004, "step": 9775 }, { "epoch": 11.65, "learning_rate": 4.636196480168112e-05, "loss": 0.0005, "step": 9860 }, { "epoch": 11.76, "learning_rate": 4.580378250591017e-05, "loss": 0.0005, "step": 9945 }, { "epoch": 11.86, "learning_rate": 4.5245600210139224e-05, "loss": 0.0006, "step": 10030 }, { "epoch": 11.96, "learning_rate": 4.4687417914368273e-05, "loss": 0.0004, "step": 10115 }, { "epoch": 12.0, "eval_loss": 0.006491700652986765, "eval_max_distance": 3, "eval_mean_distance": 0, "eval_runtime": 29.5704, "eval_samples_per_second": 8.793, "eval_steps_per_second": 0.609, "step": 10152 }, { "epoch": 12.06, "learning_rate": 4.412923561859732e-05, "loss": 0.0004, "step": 10200 }, { "epoch": 12.16, "learning_rate": 4.357105332282638e-05, "loss": 0.0004, "step": 10285 }, { "epoch": 12.26, "learning_rate": 4.301287102705542e-05, "loss": 0.0003, "step": 10370 }, { "epoch": 12.36, "learning_rate": 4.245468873128448e-05, "loss": 0.0004, "step": 10455 }, { "epoch": 12.46, "learning_rate": 4.189650643551353e-05, "loss": 0.0003, "step": 10540 }, { "epoch": 12.56, "learning_rate": 4.133832413974258e-05, "loss": 0.0004, "step": 10625 }, { "epoch": 12.66, "learning_rate": 4.078014184397163e-05, "loss": 0.0003, "step": 10710 }, { "epoch": 12.76, "learning_rate": 4.022195954820068e-05, "loss": 0.0004, "step": 10795 }, { "epoch": 12.86, "learning_rate": 3.966377725242973e-05, "loss": 0.0005, "step": 10880 }, { "epoch": 12.96, "learning_rate": 3.910559495665879e-05, "loss": 0.0003, "step": 10965 }, { "epoch": 13.0, "eval_loss": 0.006619863212108612, "eval_max_distance": 3, "eval_mean_distance": 0, "eval_runtime": 29.8838, "eval_samples_per_second": 8.7, "eval_steps_per_second": 0.602, "step": 10998 }, { "epoch": 13.06, "learning_rate": 3.854741266088784e-05, "loss": 0.0004, "step": 11050 }, { "epoch": 13.16, "learning_rate": 3.7989230365116893e-05, "loss": 0.0003, "step": 11135 }, { "epoch": 13.26, "learning_rate": 3.743104806934594e-05, "loss": 0.0003, "step": 11220 }, { "epoch": 13.36, "learning_rate": 3.687286577357499e-05, "loss": 0.0004, "step": 11305 }, { "epoch": 13.46, "learning_rate": 3.631468347780405e-05, "loss": 0.0003, "step": 11390 }, { "epoch": 13.56, "learning_rate": 3.57565011820331e-05, "loss": 0.0003, "step": 11475 }, { "epoch": 13.66, "learning_rate": 3.519831888626215e-05, "loss": 0.0004, "step": 11560 }, { "epoch": 13.76, "learning_rate": 3.4640136590491203e-05, "loss": 0.0005, "step": 11645 }, { "epoch": 13.87, "learning_rate": 3.408195429472025e-05, "loss": 0.0004, "step": 11730 }, { "epoch": 13.97, "learning_rate": 3.35237719989493e-05, "loss": 0.0005, "step": 11815 }, { "epoch": 14.0, "eval_loss": 0.006221045274287462, "eval_max_distance": 3, "eval_mean_distance": 0, "eval_runtime": 30.0954, "eval_samples_per_second": 8.639, "eval_steps_per_second": 0.598, "step": 11844 }, { "epoch": 14.07, "learning_rate": 3.296558970317836e-05, "loss": 0.0005, "step": 11900 }, { "epoch": 14.17, "learning_rate": 3.240740740740741e-05, "loss": 0.0003, "step": 11985 }, { "epoch": 14.27, "learning_rate": 3.184922511163646e-05, "loss": 0.0005, "step": 12070 }, { "epoch": 14.37, "learning_rate": 3.1291042815865513e-05, "loss": 0.0004, "step": 12155 }, { "epoch": 14.47, "learning_rate": 3.073286052009456e-05, "loss": 0.0005, "step": 12240 }, { "epoch": 14.57, "learning_rate": 3.0174678224323616e-05, "loss": 0.0004, "step": 12325 }, { "epoch": 14.67, "learning_rate": 2.961649592855267e-05, "loss": 0.0005, "step": 12410 }, { "epoch": 14.77, "learning_rate": 2.905831363278172e-05, "loss": 0.0005, "step": 12495 }, { "epoch": 14.87, "learning_rate": 2.850013133701077e-05, "loss": 0.0006, "step": 12580 }, { "epoch": 14.97, "learning_rate": 2.7941949041239823e-05, "loss": 0.0003, "step": 12665 }, { "epoch": 15.0, "eval_loss": 0.006079429294914007, "eval_max_distance": 3, "eval_mean_distance": 0, "eval_runtime": 29.8562, "eval_samples_per_second": 8.708, "eval_steps_per_second": 0.603, "step": 12690 }, { "epoch": 15.07, "learning_rate": 2.7383766745468876e-05, "loss": 0.0004, "step": 12750 }, { "epoch": 15.17, "learning_rate": 2.6825584449697926e-05, "loss": 0.0004, "step": 12835 }, { "epoch": 15.27, "learning_rate": 2.626740215392698e-05, "loss": 0.0007, "step": 12920 }, { "epoch": 15.37, "learning_rate": 2.570921985815603e-05, "loss": 0.0003, "step": 13005 }, { "epoch": 15.47, "learning_rate": 2.5151037562385084e-05, "loss": 0.0005, "step": 13090 }, { "epoch": 15.57, "learning_rate": 2.4592855266614133e-05, "loss": 0.0004, "step": 13175 }, { "epoch": 15.67, "learning_rate": 2.4034672970843186e-05, "loss": 0.0005, "step": 13260 }, { "epoch": 15.77, "learning_rate": 2.3476490675072236e-05, "loss": 0.0004, "step": 13345 }, { "epoch": 15.87, "learning_rate": 2.291830837930129e-05, "loss": 0.0002, "step": 13430 }, { "epoch": 15.98, "learning_rate": 2.2360126083530338e-05, "loss": 0.0004, "step": 13515 }, { "epoch": 16.0, "eval_loss": 0.005943103693425655, "eval_max_distance": 3, "eval_mean_distance": 0, "eval_runtime": 29.5405, "eval_samples_per_second": 8.801, "eval_steps_per_second": 0.609, "step": 13536 }, { "epoch": 16.08, "learning_rate": 2.180194378775939e-05, "loss": 0.0005, "step": 13600 }, { "epoch": 16.18, "learning_rate": 2.1243761491988443e-05, "loss": 0.0003, "step": 13685 }, { "epoch": 16.28, "learning_rate": 2.0685579196217493e-05, "loss": 0.0005, "step": 13770 }, { "epoch": 16.38, "learning_rate": 2.0127396900446546e-05, "loss": 0.0003, "step": 13855 }, { "epoch": 16.48, "learning_rate": 1.95692146046756e-05, "loss": 0.0004, "step": 13940 }, { "epoch": 16.58, "learning_rate": 1.901103230890465e-05, "loss": 0.0003, "step": 14025 }, { "epoch": 16.68, "learning_rate": 1.84528500131337e-05, "loss": 0.0004, "step": 14110 }, { "epoch": 16.78, "learning_rate": 1.7894667717362753e-05, "loss": 0.0004, "step": 14195 }, { "epoch": 16.88, "learning_rate": 1.7336485421591806e-05, "loss": 0.0004, "step": 14280 }, { "epoch": 16.98, "learning_rate": 1.677830312582086e-05, "loss": 0.0003, "step": 14365 }, { "epoch": 17.0, "eval_loss": 0.006018976215273142, "eval_max_distance": 3, "eval_mean_distance": 0, "eval_runtime": 29.8665, "eval_samples_per_second": 8.705, "eval_steps_per_second": 0.603, "step": 14382 }, { "epoch": 17.08, "learning_rate": 1.622012083004991e-05, "loss": 0.0004, "step": 14450 }, { "epoch": 17.18, "learning_rate": 1.566193853427896e-05, "loss": 0.0002, "step": 14535 }, { "epoch": 17.28, "learning_rate": 1.5103756238508012e-05, "loss": 0.0003, "step": 14620 }, { "epoch": 17.38, "learning_rate": 1.4545573942737065e-05, "loss": 0.0005, "step": 14705 }, { "epoch": 17.48, "learning_rate": 1.3987391646966116e-05, "loss": 0.0008, "step": 14790 }, { "epoch": 17.58, "learning_rate": 1.3429209351195169e-05, "loss": 0.0003, "step": 14875 }, { "epoch": 17.68, "learning_rate": 1.287102705542422e-05, "loss": 0.0003, "step": 14960 }, { "epoch": 17.78, "learning_rate": 1.2312844759653271e-05, "loss": 0.0005, "step": 15045 }, { "epoch": 17.88, "learning_rate": 1.1754662463882322e-05, "loss": 0.0003, "step": 15130 }, { "epoch": 17.98, "learning_rate": 1.1196480168111375e-05, "loss": 0.0004, "step": 15215 }, { "epoch": 18.0, "eval_loss": 0.006091571878641844, "eval_max_distance": 3, "eval_mean_distance": 0, "eval_runtime": 29.7889, "eval_samples_per_second": 8.728, "eval_steps_per_second": 0.604, "step": 15228 }, { "epoch": 18.09, "learning_rate": 1.0638297872340426e-05, "loss": 0.0003, "step": 15300 }, { "epoch": 18.19, "learning_rate": 1.0080115576569477e-05, "loss": 0.0004, "step": 15385 }, { "epoch": 18.29, "learning_rate": 9.52193328079853e-06, "loss": 0.0004, "step": 15470 }, { "epoch": 18.39, "learning_rate": 8.963750985027581e-06, "loss": 0.0003, "step": 15555 }, { "epoch": 18.49, "learning_rate": 8.405568689256634e-06, "loss": 0.0003, "step": 15640 }, { "epoch": 18.59, "learning_rate": 7.847386393485685e-06, "loss": 0.0004, "step": 15725 }, { "epoch": 18.69, "learning_rate": 7.289204097714736e-06, "loss": 0.0004, "step": 15810 }, { "epoch": 18.79, "learning_rate": 6.731021801943787e-06, "loss": 0.0004, "step": 15895 }, { "epoch": 18.89, "learning_rate": 6.172839506172839e-06, "loss": 0.0004, "step": 15980 }, { "epoch": 18.99, "learning_rate": 5.614657210401891e-06, "loss": 0.0003, "step": 16065 }, { "epoch": 19.0, "eval_loss": 0.006117388140410185, "eval_max_distance": 3, "eval_mean_distance": 0, "eval_runtime": 29.5322, "eval_samples_per_second": 8.804, "eval_steps_per_second": 0.61, "step": 16074 }, { "epoch": 19.09, "learning_rate": 5.056474914630943e-06, "loss": 0.0003, "step": 16150 }, { "epoch": 19.19, "learning_rate": 4.498292618859995e-06, "loss": 0.0004, "step": 16235 }, { "epoch": 19.29, "learning_rate": 3.940110323089047e-06, "loss": 0.0002, "step": 16320 }, { "epoch": 19.39, "learning_rate": 3.3819280273180985e-06, "loss": 0.0003, "step": 16405 }, { "epoch": 19.49, "learning_rate": 2.82374573154715e-06, "loss": 0.0003, "step": 16490 }, { "epoch": 19.59, "learning_rate": 2.265563435776202e-06, "loss": 0.0002, "step": 16575 }, { "epoch": 19.69, "learning_rate": 1.7073811400052534e-06, "loss": 0.0004, "step": 16660 }, { "epoch": 19.79, "learning_rate": 1.1491988442343054e-06, "loss": 0.0003, "step": 16745 }, { "epoch": 19.89, "learning_rate": 5.91016548463357e-07, "loss": 0.0003, "step": 16830 }, { "epoch": 19.99, "learning_rate": 3.2834252692408726e-08, "loss": 0.0005, "step": 16915 }, { "epoch": 20.0, "eval_loss": 0.00614793598651886, "eval_max_distance": 3, "eval_mean_distance": 0, "eval_runtime": 29.6354, "eval_samples_per_second": 8.773, "eval_steps_per_second": 0.607, "step": 16920 }, { "epoch": 20.0, "step": 16920, "total_flos": 4933712379248640.0, "train_loss": 0.0005908621804599912, "train_runtime": 1909.7013, "train_samples_per_second": 132.796, "train_steps_per_second": 8.86 } ], "logging_steps": 85, "max_steps": 16920, "num_train_epochs": 20, "save_steps": 170, "total_flos": 4933712379248640.0, "trial_name": null, "trial_params": null }