| { |
| "best_metric": 0.13911749422550201, |
| "best_model_checkpoint": "limbxy_pose/checkpoint-1780", |
| "epoch": 20.0, |
| "eval_steps": 500, |
| "global_step": 1780, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.2808988764044944, |
| "grad_norm": 4.863247798176859e-14, |
| "learning_rate": 5e-06, |
| "loss": 0.3282, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.5617977528089888, |
| "grad_norm": 80268.6796875, |
| "learning_rate": 1e-05, |
| "loss": 0.3192, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.8426966292134831, |
| "grad_norm": 1568863.75, |
| "learning_rate": 1.5e-05, |
| "loss": 0.2116, |
| "step": 75 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.17222893238067627, |
| "eval_rmse": 0.4150047302246094, |
| "eval_runtime": 9.1184, |
| "eval_samples_per_second": 109.668, |
| "eval_steps_per_second": 1.755, |
| "step": 89 |
| }, |
| { |
| "epoch": 1.1235955056179776, |
| "grad_norm": 1656748.25, |
| "learning_rate": 2e-05, |
| "loss": 0.1818, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.404494382022472, |
| "grad_norm": 1260438.5, |
| "learning_rate": 2.5e-05, |
| "loss": 0.1681, |
| "step": 125 |
| }, |
| { |
| "epoch": 1.6853932584269664, |
| "grad_norm": 3767623.0, |
| "learning_rate": 3e-05, |
| "loss": 0.1607, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.9662921348314608, |
| "grad_norm": 718866.8125, |
| "learning_rate": 3.5e-05, |
| "loss": 0.154, |
| "step": 175 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.16393612325191498, |
| "eval_rmse": 0.4048902690410614, |
| "eval_runtime": 8.9983, |
| "eval_samples_per_second": 111.132, |
| "eval_steps_per_second": 1.778, |
| "step": 178 |
| }, |
| { |
| "epoch": 2.247191011235955, |
| "grad_norm": 1494838.0, |
| "learning_rate": 4e-05, |
| "loss": 0.1776, |
| "step": 200 |
| }, |
| { |
| "epoch": 2.5280898876404496, |
| "grad_norm": 182619.0, |
| "learning_rate": 4.5e-05, |
| "loss": 0.1685, |
| "step": 225 |
| }, |
| { |
| "epoch": 2.808988764044944, |
| "grad_norm": 568053.375, |
| "learning_rate": 5e-05, |
| "loss": 0.1667, |
| "step": 250 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.14878585934638977, |
| "eval_rmse": 0.3857277035713196, |
| "eval_runtime": 9.1378, |
| "eval_samples_per_second": 109.436, |
| "eval_steps_per_second": 1.751, |
| "step": 267 |
| }, |
| { |
| "epoch": 3.0898876404494384, |
| "grad_norm": 234025.9375, |
| "learning_rate": 4.918300653594771e-05, |
| "loss": 0.1701, |
| "step": 275 |
| }, |
| { |
| "epoch": 3.370786516853933, |
| "grad_norm": 1965555.75, |
| "learning_rate": 4.8366013071895424e-05, |
| "loss": 0.1514, |
| "step": 300 |
| }, |
| { |
| "epoch": 3.6516853932584272, |
| "grad_norm": 851152.4375, |
| "learning_rate": 4.7549019607843135e-05, |
| "loss": 0.1577, |
| "step": 325 |
| }, |
| { |
| "epoch": 3.932584269662921, |
| "grad_norm": 746618.4375, |
| "learning_rate": 4.673202614379085e-05, |
| "loss": 0.1487, |
| "step": 350 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.15289434790611267, |
| "eval_rmse": 0.3910170793533325, |
| "eval_runtime": 8.8946, |
| "eval_samples_per_second": 112.428, |
| "eval_steps_per_second": 1.799, |
| "step": 356 |
| }, |
| { |
| "epoch": 4.213483146067416, |
| "grad_norm": 1243115.125, |
| "learning_rate": 4.5915032679738564e-05, |
| "loss": 0.1504, |
| "step": 375 |
| }, |
| { |
| "epoch": 4.49438202247191, |
| "grad_norm": 1589738.0, |
| "learning_rate": 4.5098039215686275e-05, |
| "loss": 0.1497, |
| "step": 400 |
| }, |
| { |
| "epoch": 4.775280898876405, |
| "grad_norm": 2689456.25, |
| "learning_rate": 4.4281045751633986e-05, |
| "loss": 0.153, |
| "step": 425 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.1420373022556305, |
| "eval_rmse": 0.37687838077545166, |
| "eval_runtime": 9.1741, |
| "eval_samples_per_second": 109.002, |
| "eval_steps_per_second": 1.744, |
| "step": 445 |
| }, |
| { |
| "epoch": 5.056179775280899, |
| "grad_norm": 1829392.75, |
| "learning_rate": 4.3464052287581704e-05, |
| "loss": 0.148, |
| "step": 450 |
| }, |
| { |
| "epoch": 5.337078651685394, |
| "grad_norm": 1871118.375, |
| "learning_rate": 4.2647058823529415e-05, |
| "loss": 0.1563, |
| "step": 475 |
| }, |
| { |
| "epoch": 5.617977528089888, |
| "grad_norm": 407695.96875, |
| "learning_rate": 4.1830065359477126e-05, |
| "loss": 0.1614, |
| "step": 500 |
| }, |
| { |
| "epoch": 5.898876404494382, |
| "grad_norm": 2876208.0, |
| "learning_rate": 4.101307189542484e-05, |
| "loss": 0.1516, |
| "step": 525 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 0.1412646770477295, |
| "eval_rmse": 0.37585192918777466, |
| "eval_runtime": 9.4039, |
| "eval_samples_per_second": 106.339, |
| "eval_steps_per_second": 1.701, |
| "step": 534 |
| }, |
| { |
| "epoch": 6.179775280898877, |
| "grad_norm": 1888194.125, |
| "learning_rate": 4.0196078431372555e-05, |
| "loss": 0.1484, |
| "step": 550 |
| }, |
| { |
| "epoch": 6.460674157303371, |
| "grad_norm": 2303002.75, |
| "learning_rate": 3.9379084967320266e-05, |
| "loss": 0.1544, |
| "step": 575 |
| }, |
| { |
| "epoch": 6.741573033707866, |
| "grad_norm": 3461513.0, |
| "learning_rate": 3.8562091503267977e-05, |
| "loss": 0.1492, |
| "step": 600 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 0.14228275418281555, |
| "eval_rmse": 0.3772038519382477, |
| "eval_runtime": 9.0867, |
| "eval_samples_per_second": 110.051, |
| "eval_steps_per_second": 1.761, |
| "step": 623 |
| }, |
| { |
| "epoch": 7.022471910112359, |
| "grad_norm": 1115296.0, |
| "learning_rate": 3.774509803921569e-05, |
| "loss": 0.1452, |
| "step": 625 |
| }, |
| { |
| "epoch": 7.303370786516854, |
| "grad_norm": 951385.5, |
| "learning_rate": 3.6928104575163405e-05, |
| "loss": 0.1481, |
| "step": 650 |
| }, |
| { |
| "epoch": 7.584269662921348, |
| "grad_norm": 1324887.875, |
| "learning_rate": 3.611111111111111e-05, |
| "loss": 0.1436, |
| "step": 675 |
| }, |
| { |
| "epoch": 7.865168539325842, |
| "grad_norm": 1049904.875, |
| "learning_rate": 3.529411764705883e-05, |
| "loss": 0.1472, |
| "step": 700 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 0.14021900296211243, |
| "eval_rmse": 0.3744582533836365, |
| "eval_runtime": 9.2311, |
| "eval_samples_per_second": 108.33, |
| "eval_steps_per_second": 1.733, |
| "step": 712 |
| }, |
| { |
| "epoch": 8.146067415730338, |
| "grad_norm": 2260510.0, |
| "learning_rate": 3.447712418300654e-05, |
| "loss": 0.1457, |
| "step": 725 |
| }, |
| { |
| "epoch": 8.426966292134832, |
| "grad_norm": 866467.125, |
| "learning_rate": 3.366013071895425e-05, |
| "loss": 0.1407, |
| "step": 750 |
| }, |
| { |
| "epoch": 8.707865168539326, |
| "grad_norm": 1763516.375, |
| "learning_rate": 3.284313725490196e-05, |
| "loss": 0.1428, |
| "step": 775 |
| }, |
| { |
| "epoch": 8.98876404494382, |
| "grad_norm": 1886526.5, |
| "learning_rate": 3.202614379084967e-05, |
| "loss": 0.1507, |
| "step": 800 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 0.14308789372444153, |
| "eval_rmse": 0.37826961278915405, |
| "eval_runtime": 9.0525, |
| "eval_samples_per_second": 110.466, |
| "eval_steps_per_second": 1.767, |
| "step": 801 |
| }, |
| { |
| "epoch": 9.269662921348315, |
| "grad_norm": 268722.6875, |
| "learning_rate": 3.120915032679739e-05, |
| "loss": 0.1452, |
| "step": 825 |
| }, |
| { |
| "epoch": 9.55056179775281, |
| "grad_norm": 35291.34375, |
| "learning_rate": 3.0392156862745097e-05, |
| "loss": 0.1457, |
| "step": 850 |
| }, |
| { |
| "epoch": 9.831460674157304, |
| "grad_norm": 3213099.0, |
| "learning_rate": 2.957516339869281e-05, |
| "loss": 0.1494, |
| "step": 875 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 0.13986632227897644, |
| "eval_rmse": 0.3739870488643646, |
| "eval_runtime": 9.3282, |
| "eval_samples_per_second": 107.201, |
| "eval_steps_per_second": 1.715, |
| "step": 890 |
| }, |
| { |
| "epoch": 10.112359550561798, |
| "grad_norm": 364516.15625, |
| "learning_rate": 2.8758169934640522e-05, |
| "loss": 0.1496, |
| "step": 900 |
| }, |
| { |
| "epoch": 10.393258426966293, |
| "grad_norm": 646683.375, |
| "learning_rate": 2.7941176470588236e-05, |
| "loss": 0.1519, |
| "step": 925 |
| }, |
| { |
| "epoch": 10.674157303370787, |
| "grad_norm": 157647.40625, |
| "learning_rate": 2.7124183006535947e-05, |
| "loss": 0.1564, |
| "step": 950 |
| }, |
| { |
| "epoch": 10.955056179775282, |
| "grad_norm": 436840.03125, |
| "learning_rate": 2.630718954248366e-05, |
| "loss": 0.1483, |
| "step": 975 |
| }, |
| { |
| "epoch": 11.0, |
| "eval_loss": 0.1415373831987381, |
| "eval_rmse": 0.37621453404426575, |
| "eval_runtime": 9.0431, |
| "eval_samples_per_second": 110.581, |
| "eval_steps_per_second": 1.769, |
| "step": 979 |
| }, |
| { |
| "epoch": 11.235955056179776, |
| "grad_norm": 1514792.0, |
| "learning_rate": 2.5490196078431373e-05, |
| "loss": 0.1415, |
| "step": 1000 |
| }, |
| { |
| "epoch": 11.51685393258427, |
| "grad_norm": 611788.6875, |
| "learning_rate": 2.4673202614379087e-05, |
| "loss": 0.1462, |
| "step": 1025 |
| }, |
| { |
| "epoch": 11.797752808988765, |
| "grad_norm": 1177116.5, |
| "learning_rate": 2.38562091503268e-05, |
| "loss": 0.1447, |
| "step": 1050 |
| }, |
| { |
| "epoch": 12.0, |
| "eval_loss": 0.14167071878910065, |
| "eval_rmse": 0.3763917088508606, |
| "eval_runtime": 9.3028, |
| "eval_samples_per_second": 107.495, |
| "eval_steps_per_second": 1.72, |
| "step": 1068 |
| }, |
| { |
| "epoch": 12.07865168539326, |
| "grad_norm": 1233414.0, |
| "learning_rate": 2.303921568627451e-05, |
| "loss": 0.1433, |
| "step": 1075 |
| }, |
| { |
| "epoch": 12.359550561797754, |
| "grad_norm": 323718.25, |
| "learning_rate": 2.2222222222222223e-05, |
| "loss": 0.1409, |
| "step": 1100 |
| }, |
| { |
| "epoch": 12.640449438202246, |
| "grad_norm": 1150212.0, |
| "learning_rate": 2.1405228758169934e-05, |
| "loss": 0.1427, |
| "step": 1125 |
| }, |
| { |
| "epoch": 12.921348314606742, |
| "grad_norm": 845101.5625, |
| "learning_rate": 2.058823529411765e-05, |
| "loss": 0.1562, |
| "step": 1150 |
| }, |
| { |
| "epoch": 13.0, |
| "eval_loss": 0.14507122337818146, |
| "eval_rmse": 0.380882203578949, |
| "eval_runtime": 8.886, |
| "eval_samples_per_second": 112.536, |
| "eval_steps_per_second": 1.801, |
| "step": 1157 |
| }, |
| { |
| "epoch": 13.202247191011235, |
| "grad_norm": 833533.0, |
| "learning_rate": 1.977124183006536e-05, |
| "loss": 0.1502, |
| "step": 1175 |
| }, |
| { |
| "epoch": 13.48314606741573, |
| "grad_norm": 2828633.0, |
| "learning_rate": 1.895424836601307e-05, |
| "loss": 0.1541, |
| "step": 1200 |
| }, |
| { |
| "epoch": 13.764044943820224, |
| "grad_norm": 195191.921875, |
| "learning_rate": 1.8137254901960785e-05, |
| "loss": 0.1444, |
| "step": 1225 |
| }, |
| { |
| "epoch": 14.0, |
| "eval_loss": 0.14213800430297852, |
| "eval_rmse": 0.3770119547843933, |
| "eval_runtime": 9.1277, |
| "eval_samples_per_second": 109.556, |
| "eval_steps_per_second": 1.753, |
| "step": 1246 |
| }, |
| { |
| "epoch": 14.044943820224718, |
| "grad_norm": 976718.9375, |
| "learning_rate": 1.7320261437908496e-05, |
| "loss": 0.1433, |
| "step": 1250 |
| }, |
| { |
| "epoch": 14.325842696629213, |
| "grad_norm": 240414.078125, |
| "learning_rate": 1.650326797385621e-05, |
| "loss": 0.1411, |
| "step": 1275 |
| }, |
| { |
| "epoch": 14.606741573033707, |
| "grad_norm": 1395570.375, |
| "learning_rate": 1.568627450980392e-05, |
| "loss": 0.1427, |
| "step": 1300 |
| }, |
| { |
| "epoch": 14.887640449438202, |
| "grad_norm": 69238.7109375, |
| "learning_rate": 1.4869281045751634e-05, |
| "loss": 0.1474, |
| "step": 1325 |
| }, |
| { |
| "epoch": 15.0, |
| "eval_loss": 0.13990294933319092, |
| "eval_rmse": 0.3740360140800476, |
| "eval_runtime": 9.2196, |
| "eval_samples_per_second": 108.465, |
| "eval_steps_per_second": 1.735, |
| "step": 1335 |
| }, |
| { |
| "epoch": 15.168539325842696, |
| "grad_norm": 1121152.75, |
| "learning_rate": 1.4052287581699347e-05, |
| "loss": 0.141, |
| "step": 1350 |
| }, |
| { |
| "epoch": 15.44943820224719, |
| "grad_norm": 321656.96875, |
| "learning_rate": 1.323529411764706e-05, |
| "loss": 0.1439, |
| "step": 1375 |
| }, |
| { |
| "epoch": 15.730337078651685, |
| "grad_norm": 148007.78125, |
| "learning_rate": 1.2418300653594772e-05, |
| "loss": 0.1414, |
| "step": 1400 |
| }, |
| { |
| "epoch": 16.0, |
| "eval_loss": 0.14314845204353333, |
| "eval_rmse": 0.3783496618270874, |
| "eval_runtime": 9.1061, |
| "eval_samples_per_second": 109.817, |
| "eval_steps_per_second": 1.757, |
| "step": 1424 |
| }, |
| { |
| "epoch": 16.01123595505618, |
| "grad_norm": 1851991.5, |
| "learning_rate": 1.1601307189542485e-05, |
| "loss": 0.1454, |
| "step": 1425 |
| }, |
| { |
| "epoch": 16.292134831460675, |
| "grad_norm": 42276.51171875, |
| "learning_rate": 1.0784313725490197e-05, |
| "loss": 0.1427, |
| "step": 1450 |
| }, |
| { |
| "epoch": 16.573033707865168, |
| "grad_norm": 302381.03125, |
| "learning_rate": 9.96732026143791e-06, |
| "loss": 0.1446, |
| "step": 1475 |
| }, |
| { |
| "epoch": 16.853932584269664, |
| "grad_norm": 51647.3046875, |
| "learning_rate": 9.150326797385621e-06, |
| "loss": 0.1411, |
| "step": 1500 |
| }, |
| { |
| "epoch": 17.0, |
| "eval_loss": 0.13933230936527252, |
| "eval_rmse": 0.37327244877815247, |
| "eval_runtime": 9.0522, |
| "eval_samples_per_second": 110.471, |
| "eval_steps_per_second": 1.768, |
| "step": 1513 |
| }, |
| { |
| "epoch": 17.134831460674157, |
| "grad_norm": 385612.03125, |
| "learning_rate": 8.333333333333334e-06, |
| "loss": 0.1421, |
| "step": 1525 |
| }, |
| { |
| "epoch": 17.415730337078653, |
| "grad_norm": 76190.3203125, |
| "learning_rate": 7.5163398692810456e-06, |
| "loss": 0.1426, |
| "step": 1550 |
| }, |
| { |
| "epoch": 17.696629213483146, |
| "grad_norm": 1091171.375, |
| "learning_rate": 6.699346405228758e-06, |
| "loss": 0.1391, |
| "step": 1575 |
| }, |
| { |
| "epoch": 17.97752808988764, |
| "grad_norm": 344962.09375, |
| "learning_rate": 5.882352941176471e-06, |
| "loss": 0.1394, |
| "step": 1600 |
| }, |
| { |
| "epoch": 18.0, |
| "eval_loss": 0.13929913938045502, |
| "eval_rmse": 0.3732280135154724, |
| "eval_runtime": 9.0911, |
| "eval_samples_per_second": 109.998, |
| "eval_steps_per_second": 1.76, |
| "step": 1602 |
| }, |
| { |
| "epoch": 18.258426966292134, |
| "grad_norm": 226082.828125, |
| "learning_rate": 5.065359477124184e-06, |
| "loss": 0.1391, |
| "step": 1625 |
| }, |
| { |
| "epoch": 18.53932584269663, |
| "grad_norm": 3716.295654296875, |
| "learning_rate": 4.2483660130718954e-06, |
| "loss": 0.1385, |
| "step": 1650 |
| }, |
| { |
| "epoch": 18.820224719101123, |
| "grad_norm": 57442.78515625, |
| "learning_rate": 3.431372549019608e-06, |
| "loss": 0.1427, |
| "step": 1675 |
| }, |
| { |
| "epoch": 19.0, |
| "eval_loss": 0.13926273584365845, |
| "eval_rmse": 0.37317925691604614, |
| "eval_runtime": 9.1349, |
| "eval_samples_per_second": 109.471, |
| "eval_steps_per_second": 1.752, |
| "step": 1691 |
| }, |
| { |
| "epoch": 19.10112359550562, |
| "grad_norm": 204073.96875, |
| "learning_rate": 2.6143790849673204e-06, |
| "loss": 0.1406, |
| "step": 1700 |
| }, |
| { |
| "epoch": 19.382022471910112, |
| "grad_norm": 7520.126953125, |
| "learning_rate": 1.7973856209150326e-06, |
| "loss": 0.1404, |
| "step": 1725 |
| }, |
| { |
| "epoch": 19.662921348314608, |
| "grad_norm": 312753.4375, |
| "learning_rate": 9.80392156862745e-07, |
| "loss": 0.1402, |
| "step": 1750 |
| }, |
| { |
| "epoch": 19.9438202247191, |
| "grad_norm": 339669.75, |
| "learning_rate": 1.6339869281045752e-07, |
| "loss": 0.139, |
| "step": 1775 |
| }, |
| { |
| "epoch": 20.0, |
| "eval_loss": 0.13911749422550201, |
| "eval_rmse": 0.3729845881462097, |
| "eval_runtime": 9.2284, |
| "eval_samples_per_second": 108.361, |
| "eval_steps_per_second": 1.734, |
| "step": 1780 |
| }, |
| { |
| "epoch": 20.0, |
| "step": 1780, |
| "total_flos": 0.0, |
| "train_loss": 0.15459485087501867, |
| "train_runtime": 3248.8153, |
| "train_samples_per_second": 34.856, |
| "train_steps_per_second": 0.548 |
| } |
| ], |
| "logging_steps": 25, |
| "max_steps": 1780, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 20, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 0.0, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
