| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 15.0, | |
| "eval_steps": 500, | |
| "global_step": 343740, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 2.909175539652063e-09, | |
| "loss": 0.0085, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 5.000872752661896e-06, | |
| "loss": 0.0066, | |
| "step": 1719 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 1.0001745505323792e-05, | |
| "loss": 0.0061, | |
| "step": 3438 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1.5002618257985687e-05, | |
| "loss": 0.0059, | |
| "step": 5157 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 2.0003491010647585e-05, | |
| "loss": 0.0056, | |
| "step": 6876 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 2.500436376330948e-05, | |
| "loss": 0.0052, | |
| "step": 8595 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 3.0005236515971374e-05, | |
| "loss": 0.0052, | |
| "step": 10314 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 3.500610926863327e-05, | |
| "loss": 0.0049, | |
| "step": 12033 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.000698202129517e-05, | |
| "loss": 0.0049, | |
| "step": 13752 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 4.5007854773957064e-05, | |
| "loss": 0.0052, | |
| "step": 15471 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 5.000872752661896e-05, | |
| "loss": 0.0051, | |
| "step": 17190 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 5.500960027928086e-05, | |
| "loss": 0.0052, | |
| "step": 18909 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 6.001047303194275e-05, | |
| "loss": 0.0048, | |
| "step": 20628 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 6.501134578460465e-05, | |
| "loss": 0.0052, | |
| "step": 22347 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.027106985449790955, | |
| "eval_max_distance": 9, | |
| "eval_mean_distance": 0, | |
| "eval_runtime": 14.4629, | |
| "eval_samples_per_second": 17.355, | |
| "eval_steps_per_second": 1.175, | |
| "step": 22916 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 7.001221853726654e-05, | |
| "loss": 0.0045, | |
| "step": 24066 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 7.501309128992844e-05, | |
| "loss": 0.0044, | |
| "step": 25785 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 8.001396404259034e-05, | |
| "loss": 0.0048, | |
| "step": 27504 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 8.501483679525223e-05, | |
| "loss": 0.0045, | |
| "step": 29223 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 9.001570954791413e-05, | |
| "loss": 0.0044, | |
| "step": 30942 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 9.501658230057602e-05, | |
| "loss": 0.0047, | |
| "step": 32661 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 9.999806054964024e-05, | |
| "loss": 0.005, | |
| "step": 34380 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 9.944240802156669e-05, | |
| "loss": 0.0049, | |
| "step": 36099 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 9.888675549349314e-05, | |
| "loss": 0.0053, | |
| "step": 37818 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 9.83311029654196e-05, | |
| "loss": 0.0047, | |
| "step": 39537 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 9.777545043734606e-05, | |
| "loss": 0.005, | |
| "step": 41256 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 9.721979790927251e-05, | |
| "loss": 0.0048, | |
| "step": 42975 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 9.666414538119898e-05, | |
| "loss": 0.0051, | |
| "step": 44694 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.026126669719815254, | |
| "eval_max_distance": 8, | |
| "eval_mean_distance": 0, | |
| "eval_runtime": 13.5883, | |
| "eval_samples_per_second": 18.472, | |
| "eval_steps_per_second": 1.251, | |
| "step": 45832 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 9.610849285312543e-05, | |
| "loss": 0.0047, | |
| "step": 46413 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 9.555284032505189e-05, | |
| "loss": 0.004, | |
| "step": 48132 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 9.499718779697834e-05, | |
| "loss": 0.004, | |
| "step": 49851 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 9.44415352689048e-05, | |
| "loss": 0.0042, | |
| "step": 51570 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 9.388588274083125e-05, | |
| "loss": 0.004, | |
| "step": 53289 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 9.333023021275771e-05, | |
| "loss": 0.0043, | |
| "step": 55008 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 9.277457768468416e-05, | |
| "loss": 0.0042, | |
| "step": 56727 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 9.221892515661063e-05, | |
| "loss": 0.004, | |
| "step": 58446 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 9.166327262853708e-05, | |
| "loss": 0.0045, | |
| "step": 60165 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 9.110762010046352e-05, | |
| "loss": 0.0044, | |
| "step": 61884 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 9.055196757238999e-05, | |
| "loss": 0.0044, | |
| "step": 63603 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 8.999631504431645e-05, | |
| "loss": 0.0044, | |
| "step": 65322 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 8.94406625162429e-05, | |
| "loss": 0.0043, | |
| "step": 67041 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.03130079433321953, | |
| "eval_max_distance": 8, | |
| "eval_mean_distance": 0, | |
| "eval_runtime": 13.4308, | |
| "eval_samples_per_second": 18.688, | |
| "eval_steps_per_second": 1.266, | |
| "step": 68748 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.888500998816935e-05, | |
| "loss": 0.0046, | |
| "step": 68760 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 8.832935746009581e-05, | |
| "loss": 0.0036, | |
| "step": 70479 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 8.777370493202228e-05, | |
| "loss": 0.0036, | |
| "step": 72198 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 8.721805240394872e-05, | |
| "loss": 0.0038, | |
| "step": 73917 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 8.666239987587517e-05, | |
| "loss": 0.0036, | |
| "step": 75636 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 8.610674734780164e-05, | |
| "loss": 0.0038, | |
| "step": 77355 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 8.55510948197281e-05, | |
| "loss": 0.0038, | |
| "step": 79074 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 8.499544229165455e-05, | |
| "loss": 0.0038, | |
| "step": 80793 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 8.4439789763581e-05, | |
| "loss": 0.004, | |
| "step": 82512 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 8.388413723550746e-05, | |
| "loss": 0.0037, | |
| "step": 84231 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 8.332848470743392e-05, | |
| "loss": 0.0038, | |
| "step": 85950 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 8.277283217936037e-05, | |
| "loss": 0.0039, | |
| "step": 87669 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 8.221717965128682e-05, | |
| "loss": 0.0039, | |
| "step": 89388 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 8.166152712321329e-05, | |
| "loss": 0.0041, | |
| "step": 91107 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.02780107595026493, | |
| "eval_max_distance": 10, | |
| "eval_mean_distance": 0, | |
| "eval_runtime": 13.351, | |
| "eval_samples_per_second": 18.8, | |
| "eval_steps_per_second": 1.273, | |
| "step": 91664 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 8.110587459513974e-05, | |
| "loss": 0.0037, | |
| "step": 92826 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 8.05502220670662e-05, | |
| "loss": 0.0032, | |
| "step": 94545 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "learning_rate": 7.999456953899266e-05, | |
| "loss": 0.0034, | |
| "step": 96264 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 7.943891701091911e-05, | |
| "loss": 0.0034, | |
| "step": 97983 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 7.888326448284556e-05, | |
| "loss": 0.0035, | |
| "step": 99702 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "learning_rate": 7.832761195477202e-05, | |
| "loss": 0.0034, | |
| "step": 101421 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 7.777195942669849e-05, | |
| "loss": 0.0036, | |
| "step": 103140 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 7.721630689862494e-05, | |
| "loss": 0.0035, | |
| "step": 104859 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 7.666065437055139e-05, | |
| "loss": 0.0034, | |
| "step": 106578 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 7.610500184247783e-05, | |
| "loss": 0.0034, | |
| "step": 108297 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 7.554934931440431e-05, | |
| "loss": 0.0036, | |
| "step": 110016 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 7.499369678633076e-05, | |
| "loss": 0.0034, | |
| "step": 111735 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 7.443804425825721e-05, | |
| "loss": 0.0037, | |
| "step": 113454 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.028013188391923904, | |
| "eval_max_distance": 8, | |
| "eval_mean_distance": 0, | |
| "eval_runtime": 13.4584, | |
| "eval_samples_per_second": 18.65, | |
| "eval_steps_per_second": 1.263, | |
| "step": 114580 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 7.388239173018366e-05, | |
| "loss": 0.0033, | |
| "step": 115173 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "learning_rate": 7.332673920211012e-05, | |
| "loss": 0.0031, | |
| "step": 116892 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "learning_rate": 7.277108667403659e-05, | |
| "loss": 0.0031, | |
| "step": 118611 | |
| }, | |
| { | |
| "epoch": 5.25, | |
| "learning_rate": 7.221543414596303e-05, | |
| "loss": 0.0032, | |
| "step": 120330 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 7.16597816178895e-05, | |
| "loss": 0.0031, | |
| "step": 122049 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "learning_rate": 7.110412908981595e-05, | |
| "loss": 0.0032, | |
| "step": 123768 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "learning_rate": 7.054847656174241e-05, | |
| "loss": 0.0031, | |
| "step": 125487 | |
| }, | |
| { | |
| "epoch": 5.55, | |
| "learning_rate": 6.999282403366886e-05, | |
| "loss": 0.0032, | |
| "step": 127206 | |
| }, | |
| { | |
| "epoch": 5.63, | |
| "learning_rate": 6.943717150559532e-05, | |
| "loss": 0.0032, | |
| "step": 128925 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 6.888151897752177e-05, | |
| "loss": 0.0032, | |
| "step": 130644 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "learning_rate": 6.832586644944823e-05, | |
| "loss": 0.0031, | |
| "step": 132363 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 6.777021392137468e-05, | |
| "loss": 0.0031, | |
| "step": 134082 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 6.721456139330115e-05, | |
| "loss": 0.0032, | |
| "step": 135801 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.028835317119956017, | |
| "eval_max_distance": 8, | |
| "eval_mean_distance": 0, | |
| "eval_runtime": 13.4137, | |
| "eval_samples_per_second": 18.712, | |
| "eval_steps_per_second": 1.267, | |
| "step": 137496 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 6.66589088652276e-05, | |
| "loss": 0.0033, | |
| "step": 137520 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 6.610325633715405e-05, | |
| "loss": 0.0028, | |
| "step": 139239 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 6.554760380908051e-05, | |
| "loss": 0.0026, | |
| "step": 140958 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "learning_rate": 6.499195128100697e-05, | |
| "loss": 0.0027, | |
| "step": 142677 | |
| }, | |
| { | |
| "epoch": 6.3, | |
| "learning_rate": 6.443629875293342e-05, | |
| "loss": 0.0029, | |
| "step": 144396 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "learning_rate": 6.388064622485987e-05, | |
| "loss": 0.0029, | |
| "step": 146115 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 6.332499369678633e-05, | |
| "loss": 0.0029, | |
| "step": 147834 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "learning_rate": 6.27693411687128e-05, | |
| "loss": 0.0028, | |
| "step": 149553 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "learning_rate": 6.221368864063925e-05, | |
| "loss": 0.0029, | |
| "step": 151272 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "learning_rate": 6.16580361125657e-05, | |
| "loss": 0.0029, | |
| "step": 152991 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "learning_rate": 6.110238358449216e-05, | |
| "loss": 0.0029, | |
| "step": 154710 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 6.0546731056418614e-05, | |
| "loss": 0.0028, | |
| "step": 156429 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "learning_rate": 5.999107852834507e-05, | |
| "loss": 0.0029, | |
| "step": 158148 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "learning_rate": 5.943542600027152e-05, | |
| "loss": 0.003, | |
| "step": 159867 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.030847659334540367, | |
| "eval_max_distance": 8, | |
| "eval_mean_distance": 0, | |
| "eval_runtime": 13.4895, | |
| "eval_samples_per_second": 18.607, | |
| "eval_steps_per_second": 1.26, | |
| "step": 160412 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 5.887977347219798e-05, | |
| "loss": 0.0027, | |
| "step": 161586 | |
| }, | |
| { | |
| "epoch": 7.13, | |
| "learning_rate": 5.832412094412444e-05, | |
| "loss": 0.0025, | |
| "step": 163305 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 5.7768468416050895e-05, | |
| "loss": 0.0026, | |
| "step": 165024 | |
| }, | |
| { | |
| "epoch": 7.28, | |
| "learning_rate": 5.7212815887977344e-05, | |
| "loss": 0.0027, | |
| "step": 166743 | |
| }, | |
| { | |
| "epoch": 7.35, | |
| "learning_rate": 5.665716335990381e-05, | |
| "loss": 0.0025, | |
| "step": 168462 | |
| }, | |
| { | |
| "epoch": 7.43, | |
| "learning_rate": 5.610151083183026e-05, | |
| "loss": 0.0026, | |
| "step": 170181 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "learning_rate": 5.554585830375671e-05, | |
| "loss": 0.0026, | |
| "step": 171900 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "learning_rate": 5.499020577568318e-05, | |
| "loss": 0.0026, | |
| "step": 173619 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "learning_rate": 5.443455324760963e-05, | |
| "loss": 0.0026, | |
| "step": 175338 | |
| }, | |
| { | |
| "epoch": 7.73, | |
| "learning_rate": 5.387890071953609e-05, | |
| "loss": 0.0025, | |
| "step": 177057 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "learning_rate": 5.332324819146254e-05, | |
| "loss": 0.0027, | |
| "step": 178776 | |
| }, | |
| { | |
| "epoch": 7.88, | |
| "learning_rate": 5.2767595663389e-05, | |
| "loss": 0.0028, | |
| "step": 180495 | |
| }, | |
| { | |
| "epoch": 7.95, | |
| "learning_rate": 5.2211943135315456e-05, | |
| "loss": 0.0025, | |
| "step": 182214 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.03048335202038288, | |
| "eval_max_distance": 8, | |
| "eval_mean_distance": 0, | |
| "eval_runtime": 13.5077, | |
| "eval_samples_per_second": 18.582, | |
| "eval_steps_per_second": 1.259, | |
| "step": 183328 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 5.1656290607241906e-05, | |
| "loss": 0.0026, | |
| "step": 183933 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "learning_rate": 5.110063807916836e-05, | |
| "loss": 0.0023, | |
| "step": 185652 | |
| }, | |
| { | |
| "epoch": 8.18, | |
| "learning_rate": 5.0544985551094825e-05, | |
| "loss": 0.0024, | |
| "step": 187371 | |
| }, | |
| { | |
| "epoch": 8.25, | |
| "learning_rate": 4.998933302302128e-05, | |
| "loss": 0.0023, | |
| "step": 189090 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 4.943368049494773e-05, | |
| "loss": 0.0024, | |
| "step": 190809 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "learning_rate": 4.887802796687419e-05, | |
| "loss": 0.0024, | |
| "step": 192528 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "learning_rate": 4.832237543880065e-05, | |
| "loss": 0.0025, | |
| "step": 194247 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 4.77667229107271e-05, | |
| "loss": 0.0024, | |
| "step": 195966 | |
| }, | |
| { | |
| "epoch": 8.63, | |
| "learning_rate": 4.721107038265356e-05, | |
| "loss": 0.0026, | |
| "step": 197685 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "learning_rate": 4.665541785458001e-05, | |
| "loss": 0.0023, | |
| "step": 199404 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "learning_rate": 4.6099765326506474e-05, | |
| "loss": 0.0024, | |
| "step": 201123 | |
| }, | |
| { | |
| "epoch": 8.85, | |
| "learning_rate": 4.5544112798432924e-05, | |
| "loss": 0.0024, | |
| "step": 202842 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "learning_rate": 4.4988460270359386e-05, | |
| "loss": 0.0025, | |
| "step": 204561 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.030335595831274986, | |
| "eval_max_distance": 8, | |
| "eval_mean_distance": 0, | |
| "eval_runtime": 13.4109, | |
| "eval_samples_per_second": 18.716, | |
| "eval_steps_per_second": 1.268, | |
| "step": 206244 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 4.4432807742285836e-05, | |
| "loss": 0.0026, | |
| "step": 206280 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "learning_rate": 4.38771552142123e-05, | |
| "loss": 0.0021, | |
| "step": 207999 | |
| }, | |
| { | |
| "epoch": 9.15, | |
| "learning_rate": 4.332150268613875e-05, | |
| "loss": 0.0022, | |
| "step": 209718 | |
| }, | |
| { | |
| "epoch": 9.23, | |
| "learning_rate": 4.2765850158065204e-05, | |
| "loss": 0.0022, | |
| "step": 211437 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "learning_rate": 4.221019762999166e-05, | |
| "loss": 0.0023, | |
| "step": 213156 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "learning_rate": 4.165454510191812e-05, | |
| "loss": 0.0023, | |
| "step": 214875 | |
| }, | |
| { | |
| "epoch": 9.45, | |
| "learning_rate": 4.109889257384457e-05, | |
| "loss": 0.0023, | |
| "step": 216594 | |
| }, | |
| { | |
| "epoch": 9.53, | |
| "learning_rate": 4.054324004577103e-05, | |
| "loss": 0.0023, | |
| "step": 218313 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 3.998758751769749e-05, | |
| "loss": 0.0024, | |
| "step": 220032 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "learning_rate": 3.943193498962394e-05, | |
| "loss": 0.0024, | |
| "step": 221751 | |
| }, | |
| { | |
| "epoch": 9.75, | |
| "learning_rate": 3.88762824615504e-05, | |
| "loss": 0.0022, | |
| "step": 223470 | |
| }, | |
| { | |
| "epoch": 9.83, | |
| "learning_rate": 3.8320629933476854e-05, | |
| "loss": 0.0023, | |
| "step": 225189 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "learning_rate": 3.776497740540331e-05, | |
| "loss": 0.0024, | |
| "step": 226908 | |
| }, | |
| { | |
| "epoch": 9.98, | |
| "learning_rate": 3.7209324877329766e-05, | |
| "loss": 0.0023, | |
| "step": 228627 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.034065987914800644, | |
| "eval_max_distance": 8, | |
| "eval_mean_distance": 0, | |
| "eval_runtime": 13.4726, | |
| "eval_samples_per_second": 18.63, | |
| "eval_steps_per_second": 1.262, | |
| "step": 229160 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "learning_rate": 3.665367234925622e-05, | |
| "loss": 0.0021, | |
| "step": 230346 | |
| }, | |
| { | |
| "epoch": 10.13, | |
| "learning_rate": 3.609801982118268e-05, | |
| "loss": 0.0021, | |
| "step": 232065 | |
| }, | |
| { | |
| "epoch": 10.2, | |
| "learning_rate": 3.5542367293109135e-05, | |
| "loss": 0.0021, | |
| "step": 233784 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "learning_rate": 3.498671476503559e-05, | |
| "loss": 0.0022, | |
| "step": 235503 | |
| }, | |
| { | |
| "epoch": 10.35, | |
| "learning_rate": 3.443106223696205e-05, | |
| "loss": 0.0021, | |
| "step": 237222 | |
| }, | |
| { | |
| "epoch": 10.43, | |
| "learning_rate": 3.38754097088885e-05, | |
| "loss": 0.0022, | |
| "step": 238941 | |
| }, | |
| { | |
| "epoch": 10.5, | |
| "learning_rate": 3.331975718081496e-05, | |
| "loss": 0.0021, | |
| "step": 240660 | |
| }, | |
| { | |
| "epoch": 10.58, | |
| "learning_rate": 3.276410465274141e-05, | |
| "loss": 0.0021, | |
| "step": 242379 | |
| }, | |
| { | |
| "epoch": 10.65, | |
| "learning_rate": 3.220845212466787e-05, | |
| "loss": 0.0021, | |
| "step": 244098 | |
| }, | |
| { | |
| "epoch": 10.73, | |
| "learning_rate": 3.165279959659433e-05, | |
| "loss": 0.0022, | |
| "step": 245817 | |
| }, | |
| { | |
| "epoch": 10.8, | |
| "learning_rate": 3.1097147068520784e-05, | |
| "loss": 0.0022, | |
| "step": 247536 | |
| }, | |
| { | |
| "epoch": 10.88, | |
| "learning_rate": 3.054149454044724e-05, | |
| "loss": 0.002, | |
| "step": 249255 | |
| }, | |
| { | |
| "epoch": 10.95, | |
| "learning_rate": 2.9985842012373693e-05, | |
| "loss": 0.0022, | |
| "step": 250974 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.03288768604397774, | |
| "eval_max_distance": 8, | |
| "eval_mean_distance": 0, | |
| "eval_runtime": 13.3832, | |
| "eval_samples_per_second": 18.755, | |
| "eval_steps_per_second": 1.27, | |
| "step": 252076 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 2.9430189484300152e-05, | |
| "loss": 0.0022, | |
| "step": 252693 | |
| }, | |
| { | |
| "epoch": 11.1, | |
| "learning_rate": 2.8874536956226605e-05, | |
| "loss": 0.002, | |
| "step": 254412 | |
| }, | |
| { | |
| "epoch": 11.18, | |
| "learning_rate": 2.8318884428153065e-05, | |
| "loss": 0.002, | |
| "step": 256131 | |
| }, | |
| { | |
| "epoch": 11.25, | |
| "learning_rate": 2.7763231900079517e-05, | |
| "loss": 0.002, | |
| "step": 257850 | |
| }, | |
| { | |
| "epoch": 11.33, | |
| "learning_rate": 2.7207579372005977e-05, | |
| "loss": 0.0021, | |
| "step": 259569 | |
| }, | |
| { | |
| "epoch": 11.4, | |
| "learning_rate": 2.665192684393243e-05, | |
| "loss": 0.0019, | |
| "step": 261288 | |
| }, | |
| { | |
| "epoch": 11.48, | |
| "learning_rate": 2.6096274315858886e-05, | |
| "loss": 0.002, | |
| "step": 263007 | |
| }, | |
| { | |
| "epoch": 11.55, | |
| "learning_rate": 2.5540621787785342e-05, | |
| "loss": 0.002, | |
| "step": 264726 | |
| }, | |
| { | |
| "epoch": 11.63, | |
| "learning_rate": 2.4984969259711798e-05, | |
| "loss": 0.002, | |
| "step": 266445 | |
| }, | |
| { | |
| "epoch": 11.7, | |
| "learning_rate": 2.4429316731638254e-05, | |
| "loss": 0.0019, | |
| "step": 268164 | |
| }, | |
| { | |
| "epoch": 11.78, | |
| "learning_rate": 2.387366420356471e-05, | |
| "loss": 0.0021, | |
| "step": 269883 | |
| }, | |
| { | |
| "epoch": 11.85, | |
| "learning_rate": 2.3318011675491167e-05, | |
| "loss": 0.0021, | |
| "step": 271602 | |
| }, | |
| { | |
| "epoch": 11.93, | |
| "learning_rate": 2.2762359147417623e-05, | |
| "loss": 0.0019, | |
| "step": 273321 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.03355114161968231, | |
| "eval_max_distance": 8, | |
| "eval_mean_distance": 0, | |
| "eval_runtime": 13.4567, | |
| "eval_samples_per_second": 18.652, | |
| "eval_steps_per_second": 1.263, | |
| "step": 274992 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 2.220670661934408e-05, | |
| "loss": 0.0021, | |
| "step": 275040 | |
| }, | |
| { | |
| "epoch": 12.08, | |
| "learning_rate": 2.1651054091270535e-05, | |
| "loss": 0.002, | |
| "step": 276759 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "learning_rate": 2.109540156319699e-05, | |
| "loss": 0.002, | |
| "step": 278478 | |
| }, | |
| { | |
| "epoch": 12.23, | |
| "learning_rate": 2.0539749035123444e-05, | |
| "loss": 0.0018, | |
| "step": 280197 | |
| }, | |
| { | |
| "epoch": 12.3, | |
| "learning_rate": 1.9984096507049904e-05, | |
| "loss": 0.002, | |
| "step": 281916 | |
| }, | |
| { | |
| "epoch": 12.38, | |
| "learning_rate": 1.942844397897636e-05, | |
| "loss": 0.0019, | |
| "step": 283635 | |
| }, | |
| { | |
| "epoch": 12.45, | |
| "learning_rate": 1.8872791450902816e-05, | |
| "loss": 0.0018, | |
| "step": 285354 | |
| }, | |
| { | |
| "epoch": 12.53, | |
| "learning_rate": 1.8317138922829272e-05, | |
| "loss": 0.0019, | |
| "step": 287073 | |
| }, | |
| { | |
| "epoch": 12.6, | |
| "learning_rate": 1.776148639475573e-05, | |
| "loss": 0.0018, | |
| "step": 288792 | |
| }, | |
| { | |
| "epoch": 12.68, | |
| "learning_rate": 1.7205833866682185e-05, | |
| "loss": 0.0019, | |
| "step": 290511 | |
| }, | |
| { | |
| "epoch": 12.75, | |
| "learning_rate": 1.665018133860864e-05, | |
| "loss": 0.0019, | |
| "step": 292230 | |
| }, | |
| { | |
| "epoch": 12.83, | |
| "learning_rate": 1.6094528810535094e-05, | |
| "loss": 0.0021, | |
| "step": 293949 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 1.553887628246155e-05, | |
| "loss": 0.0019, | |
| "step": 295668 | |
| }, | |
| { | |
| "epoch": 12.98, | |
| "learning_rate": 1.4983223754388006e-05, | |
| "loss": 0.002, | |
| "step": 297387 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.035788267850875854, | |
| "eval_max_distance": 8, | |
| "eval_mean_distance": 0, | |
| "eval_runtime": 13.4958, | |
| "eval_samples_per_second": 18.598, | |
| "eval_steps_per_second": 1.26, | |
| "step": 297908 | |
| }, | |
| { | |
| "epoch": 13.05, | |
| "learning_rate": 1.4427571226314462e-05, | |
| "loss": 0.0019, | |
| "step": 299106 | |
| }, | |
| { | |
| "epoch": 13.13, | |
| "learning_rate": 1.3871918698240918e-05, | |
| "loss": 0.0018, | |
| "step": 300825 | |
| }, | |
| { | |
| "epoch": 13.2, | |
| "learning_rate": 1.3316266170167374e-05, | |
| "loss": 0.0018, | |
| "step": 302544 | |
| }, | |
| { | |
| "epoch": 13.28, | |
| "learning_rate": 1.276061364209383e-05, | |
| "loss": 0.0018, | |
| "step": 304263 | |
| }, | |
| { | |
| "epoch": 13.35, | |
| "learning_rate": 1.2204961114020287e-05, | |
| "loss": 0.0018, | |
| "step": 305982 | |
| }, | |
| { | |
| "epoch": 13.43, | |
| "learning_rate": 1.1649308585946743e-05, | |
| "loss": 0.0018, | |
| "step": 307701 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "learning_rate": 1.1093656057873199e-05, | |
| "loss": 0.0019, | |
| "step": 309420 | |
| }, | |
| { | |
| "epoch": 13.58, | |
| "learning_rate": 1.0538003529799655e-05, | |
| "loss": 0.0018, | |
| "step": 311139 | |
| }, | |
| { | |
| "epoch": 13.65, | |
| "learning_rate": 9.982351001726111e-06, | |
| "loss": 0.0017, | |
| "step": 312858 | |
| }, | |
| { | |
| "epoch": 13.73, | |
| "learning_rate": 9.426698473652567e-06, | |
| "loss": 0.0018, | |
| "step": 314577 | |
| }, | |
| { | |
| "epoch": 13.8, | |
| "learning_rate": 8.871045945579024e-06, | |
| "loss": 0.002, | |
| "step": 316296 | |
| }, | |
| { | |
| "epoch": 13.88, | |
| "learning_rate": 8.31539341750548e-06, | |
| "loss": 0.0018, | |
| "step": 318015 | |
| }, | |
| { | |
| "epoch": 13.95, | |
| "learning_rate": 7.759740889431934e-06, | |
| "loss": 0.0018, | |
| "step": 319734 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.03550655022263527, | |
| "eval_max_distance": 8, | |
| "eval_mean_distance": 0, | |
| "eval_runtime": 13.4713, | |
| "eval_samples_per_second": 18.632, | |
| "eval_steps_per_second": 1.262, | |
| "step": 320824 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "learning_rate": 7.204088361358391e-06, | |
| "loss": 0.0017, | |
| "step": 321453 | |
| }, | |
| { | |
| "epoch": 14.1, | |
| "learning_rate": 6.648435833284847e-06, | |
| "loss": 0.0018, | |
| "step": 323172 | |
| }, | |
| { | |
| "epoch": 14.18, | |
| "learning_rate": 6.092783305211304e-06, | |
| "loss": 0.0018, | |
| "step": 324891 | |
| }, | |
| { | |
| "epoch": 14.25, | |
| "learning_rate": 5.537130777137759e-06, | |
| "loss": 0.0017, | |
| "step": 326610 | |
| }, | |
| { | |
| "epoch": 14.33, | |
| "learning_rate": 4.981478249064216e-06, | |
| "loss": 0.0018, | |
| "step": 328329 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "learning_rate": 4.425825720990671e-06, | |
| "loss": 0.0018, | |
| "step": 330048 | |
| }, | |
| { | |
| "epoch": 14.48, | |
| "learning_rate": 3.8701731929171274e-06, | |
| "loss": 0.0018, | |
| "step": 331767 | |
| }, | |
| { | |
| "epoch": 14.55, | |
| "learning_rate": 3.3145206648435836e-06, | |
| "loss": 0.0019, | |
| "step": 333486 | |
| }, | |
| { | |
| "epoch": 14.63, | |
| "learning_rate": 2.7588681367700398e-06, | |
| "loss": 0.0017, | |
| "step": 335205 | |
| }, | |
| { | |
| "epoch": 14.7, | |
| "learning_rate": 2.2032156086964955e-06, | |
| "loss": 0.0018, | |
| "step": 336924 | |
| }, | |
| { | |
| "epoch": 14.78, | |
| "learning_rate": 1.6475630806229517e-06, | |
| "loss": 0.0017, | |
| "step": 338643 | |
| }, | |
| { | |
| "epoch": 14.85, | |
| "learning_rate": 1.0919105525494076e-06, | |
| "loss": 0.0018, | |
| "step": 340362 | |
| }, | |
| { | |
| "epoch": 14.93, | |
| "learning_rate": 5.362580244758636e-07, | |
| "loss": 0.0019, | |
| "step": 342081 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.03661360964179039, | |
| "eval_max_distance": 8, | |
| "eval_mean_distance": 0, | |
| "eval_runtime": 13.3536, | |
| "eval_samples_per_second": 18.796, | |
| "eval_steps_per_second": 1.273, | |
| "step": 343740 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 343740, | |
| "total_flos": 8.727792619277722e+16, | |
| "train_loss": 0.0029792904397642345, | |
| "train_runtime": 24306.5697, | |
| "train_samples_per_second": 212.119, | |
| "train_steps_per_second": 14.142 | |
| } | |
| ], | |
| "logging_steps": 1719, | |
| "max_steps": 343740, | |
| "num_train_epochs": 15, | |
| "save_steps": 3438, | |
| "total_flos": 8.727792619277722e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |