{ "best_metric": null, "best_model_checkpoint": null, "epoch": 27.0, "global_step": 95877, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0, "learning_rate": 5e-09, "loss": 10.5552, "step": 1 }, { "epoch": 0.14, "learning_rate": 2.5e-06, "loss": 9.5602, "step": 500 }, { "epoch": 0.28, "learning_rate": 5e-06, "loss": 7.9946, "step": 1000 }, { "epoch": 0.42, "learning_rate": 7.5e-06, "loss": 6.9937, "step": 1500 }, { "epoch": 0.56, "learning_rate": 1e-05, "loss": 6.7102, "step": 2000 }, { "epoch": 0.7, "learning_rate": 1.25e-05, "loss": 6.549, "step": 2500 }, { "epoch": 0.84, "learning_rate": 1.5e-05, "loss": 6.4354, "step": 3000 }, { "epoch": 0.99, "learning_rate": 1.75e-05, "loss": 6.3387, "step": 3500 }, { "epoch": 1.13, "learning_rate": 2e-05, "loss": 6.2644, "step": 4000 }, { "epoch": 1.27, "learning_rate": 2.25e-05, "loss": 6.1987, "step": 4500 }, { "epoch": 1.41, "learning_rate": 2.5e-05, "loss": 6.1412, "step": 5000 }, { "epoch": 1.55, "learning_rate": 2.7500000000000004e-05, "loss": 6.0888, "step": 5500 }, { "epoch": 1.69, "learning_rate": 3e-05, "loss": 6.0466, "step": 6000 }, { "epoch": 1.83, "learning_rate": 3.2500000000000004e-05, "loss": 6.0064, "step": 6500 }, { "epoch": 1.97, "learning_rate": 3.5e-05, "loss": 5.9739, "step": 7000 }, { "epoch": 2.11, "learning_rate": 3.7500000000000003e-05, "loss": 5.9407, "step": 7500 }, { "epoch": 2.25, "learning_rate": 4e-05, "loss": 5.9138, "step": 8000 }, { "epoch": 2.39, "learning_rate": 4.2495e-05, "loss": 5.8914, "step": 8500 }, { "epoch": 2.53, "learning_rate": 4.4995000000000005e-05, "loss": 5.8642, "step": 9000 }, { "epoch": 2.68, "learning_rate": 4.7495e-05, "loss": 5.8451, "step": 9500 }, { "epoch": 2.82, "learning_rate": 4.9995000000000005e-05, "loss": 5.8283, "step": 10000 }, { "epoch": 2.96, "learning_rate": 4.998234543391946e-05, "loss": 5.8097, "step": 10500 }, { "epoch": 3.1, "learning_rate": 4.996465541690301e-05, "loss": 5.793, "step": 11000 }, { "epoch": 3.24, "learning_rate": 4.9946929948950656e-05, "loss": 5.7748, "step": 11500 }, { "epoch": 3.38, "learning_rate": 4.99292044809983e-05, "loss": 5.7613, "step": 12000 }, { "epoch": 3.52, "learning_rate": 4.991147901304595e-05, "loss": 5.7526, "step": 12500 }, { "epoch": 3.66, "learning_rate": 4.989375354509359e-05, "loss": 5.7426, "step": 13000 }, { "epoch": 3.8, "learning_rate": 4.987602807714124e-05, "loss": 5.7323, "step": 13500 }, { "epoch": 3.94, "learning_rate": 4.985830260918888e-05, "loss": 5.7168, "step": 14000 }, { "epoch": 4.08, "learning_rate": 4.984057714123653e-05, "loss": 5.7093, "step": 14500 }, { "epoch": 4.22, "learning_rate": 4.9822851673284175e-05, "loss": 5.7007, "step": 15000 }, { "epoch": 4.36, "learning_rate": 4.9805161656267726e-05, "loss": 5.6922, "step": 15500 }, { "epoch": 4.51, "learning_rate": 4.9787436188315376e-05, "loss": 5.6871, "step": 16000 }, { "epoch": 4.65, "learning_rate": 4.9769710720363025e-05, "loss": 5.6804, "step": 16500 }, { "epoch": 4.79, "learning_rate": 4.975198525241067e-05, "loss": 5.6747, "step": 17000 }, { "epoch": 4.93, "learning_rate": 4.973429523539422e-05, "loss": 5.6666, "step": 17500 }, { "epoch": 5.07, "learning_rate": 4.971656976744186e-05, "loss": 5.662, "step": 18000 }, { "epoch": 5.21, "learning_rate": 4.969884429948951e-05, "loss": 5.6527, "step": 18500 }, { "epoch": 5.35, "learning_rate": 4.9681118831537154e-05, "loss": 5.6505, "step": 19000 }, { "epoch": 5.49, "learning_rate": 4.9663428814520705e-05, "loss": 5.6443, "step": 19500 }, { "epoch": 5.63, "learning_rate": 4.964570334656835e-05, "loss": 5.642, "step": 20000 }, { "epoch": 5.77, "learning_rate": 4.9628013329551906e-05, "loss": 5.6377, "step": 20500 }, { "epoch": 5.91, "learning_rate": 4.961028786159955e-05, "loss": 5.6315, "step": 21000 }, { "epoch": 6.05, "learning_rate": 4.95925623936472e-05, "loss": 5.6261, "step": 21500 }, { "epoch": 6.2, "learning_rate": 4.957483692569484e-05, "loss": 5.6202, "step": 22000 }, { "epoch": 6.34, "learning_rate": 4.955711145774249e-05, "loss": 5.615, "step": 22500 }, { "epoch": 6.48, "learning_rate": 4.953938598979013e-05, "loss": 5.6091, "step": 23000 }, { "epoch": 6.62, "learning_rate": 4.952166052183778e-05, "loss": 5.6078, "step": 23500 }, { "epoch": 6.76, "learning_rate": 4.950397050482133e-05, "loss": 5.6059, "step": 24000 }, { "epoch": 6.9, "learning_rate": 4.9486245036868976e-05, "loss": 5.5979, "step": 24500 }, { "epoch": 7.04, "learning_rate": 4.946851956891662e-05, "loss": 5.5988, "step": 25000 }, { "epoch": 7.18, "learning_rate": 4.945079410096427e-05, "loss": 5.595, "step": 25500 }, { "epoch": 7.32, "learning_rate": 4.943306863301191e-05, "loss": 5.5909, "step": 26000 }, { "epoch": 7.46, "learning_rate": 4.941534316505956e-05, "loss": 5.5905, "step": 26500 }, { "epoch": 7.6, "learning_rate": 4.9397617697107204e-05, "loss": 5.584, "step": 27000 }, { "epoch": 7.74, "learning_rate": 4.937989222915485e-05, "loss": 5.5788, "step": 27500 }, { "epoch": 7.89, "learning_rate": 4.9362237663074306e-05, "loss": 5.5768, "step": 28000 }, { "epoch": 8.03, "learning_rate": 4.9344512195121955e-05, "loss": 5.5757, "step": 28500 }, { "epoch": 8.17, "learning_rate": 4.93267867271696e-05, "loss": 5.57, "step": 29000 }, { "epoch": 8.31, "learning_rate": 4.930906125921724e-05, "loss": 5.5683, "step": 29500 }, { "epoch": 8.45, "learning_rate": 4.929133579126489e-05, "loss": 5.5653, "step": 30000 }, { "epoch": 8.59, "learning_rate": 4.927361032331253e-05, "loss": 5.5612, "step": 30500 }, { "epoch": 8.73, "learning_rate": 4.925588485536018e-05, "loss": 5.5606, "step": 31000 }, { "epoch": 8.87, "learning_rate": 4.9238159387407825e-05, "loss": 5.5517, "step": 31500 }, { "epoch": 9.01, "learning_rate": 4.922046937039138e-05, "loss": 5.5098, "step": 32000 }, { "epoch": 9.15, "learning_rate": 4.9202743902439026e-05, "loss": 5.2495, "step": 32500 }, { "epoch": 9.29, "learning_rate": 4.9185018434486675e-05, "loss": 5.0503, "step": 33000 }, { "epoch": 9.43, "learning_rate": 4.916729296653432e-05, "loss": 4.8823, "step": 33500 }, { "epoch": 9.57, "learning_rate": 4.914956749858197e-05, "loss": 4.7268, "step": 34000 }, { "epoch": 9.72, "learning_rate": 4.913187748156551e-05, "loss": 4.5778, "step": 34500 }, { "epoch": 9.86, "learning_rate": 4.911415201361316e-05, "loss": 4.4355, "step": 35000 }, { "epoch": 10.0, "learning_rate": 4.9096426545660804e-05, "loss": 4.2972, "step": 35500 }, { "epoch": 10.14, "learning_rate": 4.907873652864436e-05, "loss": 4.1631, "step": 36000 }, { "epoch": 10.28, "learning_rate": 4.9061011060692005e-05, "loss": 4.0378, "step": 36500 }, { "epoch": 10.42, "learning_rate": 4.9043285592739654e-05, "loss": 3.9145, "step": 37000 }, { "epoch": 10.56, "learning_rate": 4.90255601247873e-05, "loss": 3.7891, "step": 37500 }, { "epoch": 10.7, "learning_rate": 4.9007834656834946e-05, "loss": 3.6041, "step": 38000 }, { "epoch": 10.84, "learning_rate": 4.899010918888259e-05, "loss": 3.2511, "step": 38500 }, { "epoch": 10.98, "learning_rate": 4.897238372093024e-05, "loss": 2.7629, "step": 39000 }, { "epoch": 11.12, "learning_rate": 4.895469370391378e-05, "loss": 2.5565, "step": 39500 }, { "epoch": 11.26, "learning_rate": 4.893696823596143e-05, "loss": 2.4138, "step": 40000 }, { "epoch": 11.41, "learning_rate": 4.8919242768009075e-05, "loss": 2.2866, "step": 40500 }, { "epoch": 11.55, "learning_rate": 4.8901517300056725e-05, "loss": 2.1777, "step": 41000 }, { "epoch": 11.69, "learning_rate": 4.8883827283040276e-05, "loss": 2.0846, "step": 41500 }, { "epoch": 11.83, "learning_rate": 4.8866101815087925e-05, "loss": 2.0126, "step": 42000 }, { "epoch": 11.97, "learning_rate": 4.884837634713557e-05, "loss": 1.9535, "step": 42500 }, { "epoch": 12.11, "learning_rate": 4.883065087918322e-05, "loss": 1.9016, "step": 43000 }, { "epoch": 12.25, "learning_rate": 4.881292541123086e-05, "loss": 1.8548, "step": 43500 }, { "epoch": 12.39, "learning_rate": 4.879523539421441e-05, "loss": 1.8197, "step": 44000 }, { "epoch": 12.53, "learning_rate": 4.8777509926262054e-05, "loss": 1.7838, "step": 44500 }, { "epoch": 12.67, "learning_rate": 4.8759784458309704e-05, "loss": 1.7528, "step": 45000 }, { "epoch": 12.81, "learning_rate": 4.8742058990357346e-05, "loss": 1.7196, "step": 45500 }, { "epoch": 12.95, "learning_rate": 4.872433352240499e-05, "loss": 1.6943, "step": 46000 }, { "epoch": 13.09, "learning_rate": 4.870664350538855e-05, "loss": 1.6715, "step": 46500 }, { "epoch": 13.24, "learning_rate": 4.868891803743619e-05, "loss": 1.6442, "step": 47000 }, { "epoch": 13.38, "learning_rate": 4.867119256948384e-05, "loss": 1.6238, "step": 47500 }, { "epoch": 13.52, "learning_rate": 4.865346710153148e-05, "loss": 1.6014, "step": 48000 }, { "epoch": 13.66, "learning_rate": 4.863574163357913e-05, "loss": 1.581, "step": 48500 }, { "epoch": 13.8, "learning_rate": 4.8618016165626774e-05, "loss": 1.5615, "step": 49000 }, { "epoch": 13.94, "learning_rate": 4.8600326148610325e-05, "loss": 1.5459, "step": 49500 }, { "epoch": 14.08, "learning_rate": 4.858260068065797e-05, "loss": 1.5299, "step": 50000 }, { "epoch": 14.22, "learning_rate": 4.856487521270562e-05, "loss": 1.5138, "step": 50500 }, { "epoch": 14.36, "learning_rate": 4.854714974475326e-05, "loss": 1.4984, "step": 51000 }, { "epoch": 14.5, "learning_rate": 4.852945972773681e-05, "loss": 1.4886, "step": 51500 }, { "epoch": 14.64, "learning_rate": 4.851173425978446e-05, "loss": 1.472, "step": 52000 }, { "epoch": 14.78, "learning_rate": 4.849400879183211e-05, "loss": 1.4619, "step": 52500 }, { "epoch": 14.93, "learning_rate": 4.847628332387975e-05, "loss": 1.4486, "step": 53000 }, { "epoch": 15.07, "learning_rate": 4.84585578559274e-05, "loss": 1.4368, "step": 53500 }, { "epoch": 15.21, "learning_rate": 4.844086783891095e-05, "loss": 1.4239, "step": 54000 }, { "epoch": 15.35, "learning_rate": 4.8423142370958596e-05, "loss": 1.4122, "step": 54500 }, { "epoch": 15.49, "learning_rate": 4.840541690300624e-05, "loss": 1.4043, "step": 55000 }, { "epoch": 15.63, "learning_rate": 4.838769143505389e-05, "loss": 1.3943, "step": 55500 }, { "epoch": 15.77, "learning_rate": 4.836996596710153e-05, "loss": 1.3843, "step": 56000 }, { "epoch": 15.91, "learning_rate": 4.835224049914918e-05, "loss": 1.3764, "step": 56500 }, { "epoch": 16.05, "learning_rate": 4.8334515031196824e-05, "loss": 1.3662, "step": 57000 }, { "epoch": 16.19, "learning_rate": 4.831678956324447e-05, "loss": 1.357, "step": 57500 }, { "epoch": 16.33, "learning_rate": 4.8299170448099834e-05, "loss": 1.349, "step": 58000 }, { "epoch": 16.47, "learning_rate": 4.828144498014748e-05, "loss": 1.3404, "step": 58500 }, { "epoch": 16.62, "learning_rate": 4.8263719512195126e-05, "loss": 1.3355, "step": 59000 }, { "epoch": 16.76, "learning_rate": 4.824599404424277e-05, "loss": 1.3258, "step": 59500 }, { "epoch": 16.9, "learning_rate": 4.822826857629041e-05, "loss": 1.3188, "step": 60000 }, { "epoch": 17.04, "learning_rate": 4.821054310833806e-05, "loss": 1.3101, "step": 60500 }, { "epoch": 17.18, "learning_rate": 4.8192817640385704e-05, "loss": 1.3034, "step": 61000 }, { "epoch": 17.32, "learning_rate": 4.8175092172433354e-05, "loss": 1.2974, "step": 61500 }, { "epoch": 17.46, "learning_rate": 4.8157366704480996e-05, "loss": 1.2897, "step": 62000 }, { "epoch": 17.6, "learning_rate": 4.8139676687464554e-05, "loss": 1.2832, "step": 62500 }, { "epoch": 17.74, "learning_rate": 4.81219512195122e-05, "loss": 1.2764, "step": 63000 }, { "epoch": 17.88, "learning_rate": 4.8104225751559846e-05, "loss": 1.2702, "step": 63500 }, { "epoch": 18.02, "learning_rate": 4.808650028360749e-05, "loss": 1.2651, "step": 64000 }, { "epoch": 18.16, "learning_rate": 4.806877481565514e-05, "loss": 1.2572, "step": 64500 }, { "epoch": 18.3, "learning_rate": 4.805108479863868e-05, "loss": 1.2547, "step": 65000 }, { "epoch": 18.45, "learning_rate": 4.803335933068633e-05, "loss": 1.2476, "step": 65500 }, { "epoch": 18.59, "learning_rate": 4.8015633862733975e-05, "loss": 1.243, "step": 66000 }, { "epoch": 18.73, "learning_rate": 4.7997908394781625e-05, "loss": 1.2364, "step": 66500 }, { "epoch": 18.87, "learning_rate": 4.798018292682927e-05, "loss": 1.2312, "step": 67000 }, { "epoch": 19.01, "learning_rate": 4.7962492909812825e-05, "loss": 1.2246, "step": 67500 }, { "epoch": 19.15, "learning_rate": 4.794476744186047e-05, "loss": 1.2205, "step": 68000 }, { "epoch": 19.29, "learning_rate": 4.792704197390812e-05, "loss": 1.2162, "step": 68500 }, { "epoch": 19.43, "learning_rate": 4.790931650595576e-05, "loss": 1.2122, "step": 69000 }, { "epoch": 19.57, "learning_rate": 4.789162648893931e-05, "loss": 1.2071, "step": 69500 }, { "epoch": 19.71, "learning_rate": 4.7873901020986954e-05, "loss": 1.2001, "step": 70000 }, { "epoch": 19.85, "learning_rate": 4.7856175553034604e-05, "loss": 1.1961, "step": 70500 }, { "epoch": 19.99, "learning_rate": 4.7838450085082246e-05, "loss": 1.191, "step": 71000 }, { "epoch": 20.14, "learning_rate": 4.78207600680658e-05, "loss": 1.1853, "step": 71500 }, { "epoch": 20.28, "learning_rate": 4.780303460011345e-05, "loss": 1.1842, "step": 72000 }, { "epoch": 20.42, "learning_rate": 4.7785309132161097e-05, "loss": 1.1771, "step": 72500 }, { "epoch": 20.56, "learning_rate": 4.776758366420874e-05, "loss": 1.1728, "step": 73000 }, { "epoch": 20.7, "learning_rate": 4.774985819625638e-05, "loss": 1.1709, "step": 73500 }, { "epoch": 20.84, "learning_rate": 4.773213272830403e-05, "loss": 1.1664, "step": 74000 }, { "epoch": 20.98, "learning_rate": 4.771444271128758e-05, "loss": 1.1615, "step": 74500 }, { "epoch": 21.12, "learning_rate": 4.7696717243335225e-05, "loss": 1.1567, "step": 75000 }, { "epoch": 21.26, "learning_rate": 4.767899177538287e-05, "loss": 1.1554, "step": 75500 }, { "epoch": 21.4, "learning_rate": 4.766126630743052e-05, "loss": 1.1511, "step": 76000 }, { "epoch": 21.54, "learning_rate": 4.764354083947816e-05, "loss": 1.146, "step": 76500 }, { "epoch": 21.68, "learning_rate": 4.762581537152581e-05, "loss": 1.1418, "step": 77000 }, { "epoch": 21.82, "learning_rate": 4.760808990357345e-05, "loss": 1.1388, "step": 77500 }, { "epoch": 21.97, "learning_rate": 4.75903644356211e-05, "loss": 1.1344, "step": 78000 }, { "epoch": 22.11, "learning_rate": 4.757267441860465e-05, "loss": 1.1315, "step": 78500 }, { "epoch": 22.25, "learning_rate": 4.75549489506523e-05, "loss": 1.1258, "step": 79000 }, { "epoch": 22.39, "learning_rate": 4.7537223482699945e-05, "loss": 1.1206, "step": 79500 }, { "epoch": 22.53, "learning_rate": 4.7519498014747595e-05, "loss": 1.1188, "step": 80000 }, { "epoch": 22.67, "learning_rate": 4.750184344866705e-05, "loss": 1.1194, "step": 80500 }, { "epoch": 22.81, "learning_rate": 4.748411798071469e-05, "loss": 1.1127, "step": 81000 }, { "epoch": 22.95, "learning_rate": 4.746639251276234e-05, "loss": 1.1116, "step": 81500 }, { "epoch": 23.09, "learning_rate": 4.744866704480998e-05, "loss": 1.1082, "step": 82000 }, { "epoch": 23.23, "learning_rate": 4.743094157685763e-05, "loss": 1.1043, "step": 82500 }, { "epoch": 23.37, "learning_rate": 4.741321610890528e-05, "loss": 1.1018, "step": 83000 }, { "epoch": 23.51, "learning_rate": 4.7395490640952924e-05, "loss": 1.0988, "step": 83500 }, { "epoch": 23.66, "learning_rate": 4.7377765173000574e-05, "loss": 1.0939, "step": 84000 }, { "epoch": 23.8, "learning_rate": 4.736007515598412e-05, "loss": 1.0936, "step": 84500 }, { "epoch": 23.94, "learning_rate": 4.734234968803177e-05, "loss": 1.09, "step": 85000 }, { "epoch": 24.08, "learning_rate": 4.732462422007941e-05, "loss": 1.0868, "step": 85500 }, { "epoch": 24.22, "learning_rate": 4.730689875212706e-05, "loss": 1.0855, "step": 86000 }, { "epoch": 24.36, "learning_rate": 4.72891732841747e-05, "loss": 1.0812, "step": 86500 }, { "epoch": 24.5, "learning_rate": 4.727144781622235e-05, "loss": 1.0798, "step": 87000 }, { "epoch": 24.64, "learning_rate": 4.7253722348269995e-05, "loss": 1.0772, "step": 87500 }, { "epoch": 24.78, "learning_rate": 4.7236032331253546e-05, "loss": 1.0763, "step": 88000 }, { "epoch": 24.92, "learning_rate": 4.7218306863301195e-05, "loss": 1.0732, "step": 88500 }, { "epoch": 25.06, "learning_rate": 4.7200581395348845e-05, "loss": 1.0699, "step": 89000 }, { "epoch": 25.2, "learning_rate": 4.718289137833239e-05, "loss": 1.0659, "step": 89500 }, { "epoch": 25.34, "learning_rate": 4.716516591038004e-05, "loss": 1.0668, "step": 90000 }, { "epoch": 25.49, "learning_rate": 4.714744044242768e-05, "loss": 1.0615, "step": 90500 }, { "epoch": 25.63, "learning_rate": 4.712971497447533e-05, "loss": 1.0594, "step": 91000 }, { "epoch": 25.77, "learning_rate": 4.7111989506522974e-05, "loss": 1.0612, "step": 91500 }, { "epoch": 25.91, "learning_rate": 4.7094299489506525e-05, "loss": 1.0545, "step": 92000 }, { "epoch": 26.05, "learning_rate": 4.707657402155417e-05, "loss": 1.0543, "step": 92500 }, { "epoch": 26.19, "learning_rate": 4.705884855360182e-05, "loss": 1.051, "step": 93000 }, { "epoch": 26.33, "learning_rate": 4.704112308564947e-05, "loss": 1.05, "step": 93500 }, { "epoch": 26.47, "learning_rate": 4.702339761769711e-05, "loss": 1.0478, "step": 94000 }, { "epoch": 26.61, "learning_rate": 4.700567214974476e-05, "loss": 1.0448, "step": 94500 }, { "epoch": 26.75, "learning_rate": 4.69879466817924e-05, "loss": 1.0433, "step": 95000 }, { "epoch": 26.89, "learning_rate": 4.697022121384005e-05, "loss": 1.0401, "step": 95500 } ], "max_steps": 1420400, "num_train_epochs": 400, "total_flos": 2.583546946148054e+19, "trial_name": null, "trial_params": null }