{ "best_metric": null, "best_model_checkpoint": null, "epoch": 29.0, "global_step": 98252, "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": 10.5588, "step": 1 }, { "epoch": 0.15, "learning_rate": 2.5e-06, "loss": 9.4906, "step": 500 }, { "epoch": 0.3, "learning_rate": 5e-06, "loss": 7.9572, "step": 1000 }, { "epoch": 0.44, "learning_rate": 7.5e-06, "loss": 7.0139, "step": 1500 }, { "epoch": 0.59, "learning_rate": 1e-05, "loss": 6.7454, "step": 2000 }, { "epoch": 0.74, "learning_rate": 1.25e-05, "loss": 6.5877, "step": 2500 }, { "epoch": 0.89, "learning_rate": 1.5e-05, "loss": 6.4748, "step": 3000 }, { "epoch": 1.03, "learning_rate": 1.75e-05, "loss": 6.3744, "step": 3500 }, { "epoch": 1.18, "learning_rate": 2e-05, "loss": 6.2984, "step": 4000 }, { "epoch": 1.33, "learning_rate": 2.25e-05, "loss": 6.2341, "step": 4500 }, { "epoch": 1.48, "learning_rate": 2.5e-05, "loss": 6.1767, "step": 5000 }, { "epoch": 1.62, "learning_rate": 2.7500000000000004e-05, "loss": 6.1274, "step": 5500 }, { "epoch": 1.77, "learning_rate": 3e-05, "loss": 6.0823, "step": 6000 }, { "epoch": 1.92, "learning_rate": 3.2500000000000004e-05, "loss": 6.0421, "step": 6500 }, { "epoch": 2.07, "learning_rate": 3.5e-05, "loss": 6.0017, "step": 7000 }, { "epoch": 2.21, "learning_rate": 3.7500000000000003e-05, "loss": 5.9722, "step": 7500 }, { "epoch": 2.36, "learning_rate": 4e-05, "loss": 5.9385, "step": 8000 }, { "epoch": 2.51, "learning_rate": 4.2495e-05, "loss": 5.9107, "step": 8500 }, { "epoch": 2.66, "learning_rate": 4.4995000000000005e-05, "loss": 5.8896, "step": 9000 }, { "epoch": 2.8, "learning_rate": 4.7495e-05, "loss": 5.8665, "step": 9500 }, { "epoch": 2.95, "learning_rate": 4.9995000000000005e-05, "loss": 5.8475, "step": 10000 }, { "epoch": 3.1, "learning_rate": 4.998148974130241e-05, "loss": 5.8261, "step": 10500 }, { "epoch": 3.25, "learning_rate": 4.9962905144216476e-05, "loss": 5.8092, "step": 11000 }, { "epoch": 3.39, "learning_rate": 4.994432054713054e-05, "loss": 5.7955, "step": 11500 }, { "epoch": 3.54, "learning_rate": 4.9925735950044606e-05, "loss": 5.7808, "step": 12000 }, { "epoch": 3.69, "learning_rate": 4.990718852215284e-05, "loss": 5.7663, "step": 12500 }, { "epoch": 3.84, "learning_rate": 4.9888603925066905e-05, "loss": 5.758, "step": 13000 }, { "epoch": 3.98, "learning_rate": 4.987001932798097e-05, "loss": 5.7457, "step": 13500 }, { "epoch": 4.13, "learning_rate": 4.9851434730895035e-05, "loss": 5.7345, "step": 14000 }, { "epoch": 4.28, "learning_rate": 4.98328501338091e-05, "loss": 5.7287, "step": 14500 }, { "epoch": 4.43, "learning_rate": 4.981430270591734e-05, "loss": 5.7177, "step": 15000 }, { "epoch": 4.57, "learning_rate": 4.97957181088314e-05, "loss": 5.7086, "step": 15500 }, { "epoch": 4.72, "learning_rate": 4.9777133511745463e-05, "loss": 5.7048, "step": 16000 }, { "epoch": 4.87, "learning_rate": 4.975854891465953e-05, "loss": 5.6978, "step": 16500 }, { "epoch": 5.02, "learning_rate": 4.974000148676777e-05, "loss": 5.6939, "step": 17000 }, { "epoch": 5.17, "learning_rate": 4.9721416889681834e-05, "loss": 5.6829, "step": 17500 }, { "epoch": 5.31, "learning_rate": 4.97028322925959e-05, "loss": 5.6763, "step": 18000 }, { "epoch": 5.46, "learning_rate": 4.9684247695509964e-05, "loss": 5.6745, "step": 18500 }, { "epoch": 5.61, "learning_rate": 4.96657002676182e-05, "loss": 5.6688, "step": 19000 }, { "epoch": 5.76, "learning_rate": 4.964711567053226e-05, "loss": 5.6605, "step": 19500 }, { "epoch": 5.9, "learning_rate": 4.96285682426405e-05, "loss": 5.6596, "step": 20000 }, { "epoch": 6.05, "learning_rate": 4.960998364555456e-05, "loss": 5.6507, "step": 20500 }, { "epoch": 6.2, "learning_rate": 4.959139904846863e-05, "loss": 5.6504, "step": 21000 }, { "epoch": 6.35, "learning_rate": 4.95728144513827e-05, "loss": 5.6443, "step": 21500 }, { "epoch": 6.49, "learning_rate": 4.955426702349094e-05, "loss": 5.641, "step": 22000 }, { "epoch": 6.64, "learning_rate": 4.9535682426405e-05, "loss": 5.6369, "step": 22500 }, { "epoch": 6.79, "learning_rate": 4.951709782931906e-05, "loss": 5.6307, "step": 23000 }, { "epoch": 6.94, "learning_rate": 4.9498513232233127e-05, "loss": 5.6267, "step": 23500 }, { "epoch": 7.08, "learning_rate": 4.947992863514719e-05, "loss": 5.6207, "step": 24000 }, { "epoch": 7.23, "learning_rate": 4.946134403806126e-05, "loss": 5.6199, "step": 24500 }, { "epoch": 7.38, "learning_rate": 4.94427966101695e-05, "loss": 5.6176, "step": 25000 }, { "epoch": 7.53, "learning_rate": 4.942421201308356e-05, "loss": 5.6159, "step": 25500 }, { "epoch": 7.67, "learning_rate": 4.940562741599763e-05, "loss": 5.6105, "step": 26000 }, { "epoch": 7.82, "learning_rate": 4.9387042818911685e-05, "loss": 5.6074, "step": 26500 }, { "epoch": 7.97, "learning_rate": 4.936845822182575e-05, "loss": 5.6082, "step": 27000 }, { "epoch": 8.12, "learning_rate": 4.9349873624739815e-05, "loss": 5.6027, "step": 27500 }, { "epoch": 8.26, "learning_rate": 4.9331326196848056e-05, "loss": 5.598, "step": 28000 }, { "epoch": 8.41, "learning_rate": 4.931274159976212e-05, "loss": 5.5981, "step": 28500 }, { "epoch": 8.56, "learning_rate": 4.9294157002676186e-05, "loss": 5.5927, "step": 29000 }, { "epoch": 8.71, "learning_rate": 4.927557240559025e-05, "loss": 5.5912, "step": 29500 }, { "epoch": 8.85, "learning_rate": 4.925698780850431e-05, "loss": 5.5881, "step": 30000 }, { "epoch": 9.0, "learning_rate": 4.9238403211418374e-05, "loss": 5.5843, "step": 30500 }, { "epoch": 9.15, "learning_rate": 4.921981861433244e-05, "loss": 5.584, "step": 31000 }, { "epoch": 9.3, "learning_rate": 4.9201234017246504e-05, "loss": 5.5809, "step": 31500 }, { "epoch": 9.45, "learning_rate": 4.918272375854891e-05, "loss": 5.5765, "step": 32000 }, { "epoch": 9.59, "learning_rate": 4.916413916146298e-05, "loss": 5.575, "step": 32500 }, { "epoch": 9.74, "learning_rate": 4.914555456437704e-05, "loss": 5.5707, "step": 33000 }, { "epoch": 9.89, "learning_rate": 4.912696996729111e-05, "loss": 5.5738, "step": 33500 }, { "epoch": 10.04, "learning_rate": 4.910838537020518e-05, "loss": 5.5686, "step": 34000 }, { "epoch": 10.18, "learning_rate": 4.9089800773119245e-05, "loss": 5.5673, "step": 34500 }, { "epoch": 10.33, "learning_rate": 4.907121617603331e-05, "loss": 5.5328, "step": 35000 }, { "epoch": 10.48, "learning_rate": 4.9052631578947375e-05, "loss": 5.2919, "step": 35500 }, { "epoch": 10.63, "learning_rate": 4.9034046981861434e-05, "loss": 5.1132, "step": 36000 }, { "epoch": 10.77, "learning_rate": 4.9015499553969674e-05, "loss": 4.9545, "step": 36500 }, { "epoch": 10.92, "learning_rate": 4.899691495688374e-05, "loss": 4.7923, "step": 37000 }, { "epoch": 11.07, "learning_rate": 4.8978330359797804e-05, "loss": 4.6472, "step": 37500 }, { "epoch": 11.22, "learning_rate": 4.895974576271187e-05, "loss": 4.508, "step": 38000 }, { "epoch": 11.36, "learning_rate": 4.89411983348201e-05, "loss": 4.3801, "step": 38500 }, { "epoch": 11.51, "learning_rate": 4.892261373773417e-05, "loss": 4.2399, "step": 39000 }, { "epoch": 11.66, "learning_rate": 4.890402914064823e-05, "loss": 4.0865, "step": 39500 }, { "epoch": 11.81, "learning_rate": 4.88854445435623e-05, "loss": 3.7505, "step": 40000 }, { "epoch": 11.95, "learning_rate": 4.886689711567054e-05, "loss": 3.1766, "step": 40500 }, { "epoch": 12.1, "learning_rate": 4.8848312518584596e-05, "loss": 2.7324, "step": 41000 }, { "epoch": 12.25, "learning_rate": 4.8829765090692836e-05, "loss": 2.5016, "step": 41500 }, { "epoch": 12.4, "learning_rate": 4.88111804936069e-05, "loss": 2.352, "step": 42000 }, { "epoch": 12.54, "learning_rate": 4.8792595896520966e-05, "loss": 2.247, "step": 42500 }, { "epoch": 12.69, "learning_rate": 4.877401129943503e-05, "loss": 2.1678, "step": 43000 }, { "epoch": 12.84, "learning_rate": 4.87554267023491e-05, "loss": 2.095, "step": 43500 }, { "epoch": 12.99, "learning_rate": 4.873687927445733e-05, "loss": 2.0451, "step": 44000 }, { "epoch": 13.13, "learning_rate": 4.8718294677371395e-05, "loss": 1.992, "step": 44500 }, { "epoch": 13.28, "learning_rate": 4.869971008028546e-05, "loss": 1.9523, "step": 45000 }, { "epoch": 13.43, "learning_rate": 4.8681125483199525e-05, "loss": 1.9121, "step": 45500 }, { "epoch": 13.58, "learning_rate": 4.866254088611359e-05, "loss": 1.8785, "step": 46000 }, { "epoch": 13.72, "learning_rate": 4.8643993458221824e-05, "loss": 1.8475, "step": 46500 }, { "epoch": 13.87, "learning_rate": 4.862540886113589e-05, "loss": 1.8158, "step": 47000 }, { "epoch": 14.02, "learning_rate": 4.860682426404996e-05, "loss": 1.7896, "step": 47500 }, { "epoch": 14.17, "learning_rate": 4.8588239666964026e-05, "loss": 1.7639, "step": 48000 }, { "epoch": 14.32, "learning_rate": 4.856965506987809e-05, "loss": 1.7434, "step": 48500 }, { "epoch": 14.46, "learning_rate": 4.8551107641986324e-05, "loss": 1.7184, "step": 49000 }, { "epoch": 14.61, "learning_rate": 4.853252304490039e-05, "loss": 1.7003, "step": 49500 }, { "epoch": 14.76, "learning_rate": 4.8513938447814454e-05, "loss": 1.6781, "step": 50000 }, { "epoch": 14.91, "learning_rate": 4.849535385072852e-05, "loss": 1.6588, "step": 50500 }, { "epoch": 15.05, "learning_rate": 4.847680642283676e-05, "loss": 1.6461, "step": 51000 }, { "epoch": 15.2, "learning_rate": 4.845822182575082e-05, "loss": 1.627, "step": 51500 }, { "epoch": 15.35, "learning_rate": 4.843963722866488e-05, "loss": 1.6137, "step": 52000 }, { "epoch": 15.5, "learning_rate": 4.842105263157895e-05, "loss": 1.5991, "step": 52500 }, { "epoch": 15.64, "learning_rate": 4.840246803449301e-05, "loss": 1.5868, "step": 53000 }, { "epoch": 15.79, "learning_rate": 4.8383920606601253e-05, "loss": 1.5735, "step": 53500 }, { "epoch": 15.94, "learning_rate": 4.836533600951532e-05, "loss": 1.5621, "step": 54000 }, { "epoch": 16.09, "learning_rate": 4.8346751412429384e-05, "loss": 1.546, "step": 54500 }, { "epoch": 16.23, "learning_rate": 4.832816681534345e-05, "loss": 1.5339, "step": 55000 }, { "epoch": 16.38, "learning_rate": 4.830961938745168e-05, "loss": 1.5267, "step": 55500 }, { "epoch": 16.53, "learning_rate": 4.829103479036575e-05, "loss": 1.5158, "step": 56000 }, { "epoch": 16.68, "learning_rate": 4.827245019327981e-05, "loss": 1.5052, "step": 56500 }, { "epoch": 16.82, "learning_rate": 4.8253902765388046e-05, "loss": 1.4964, "step": 57000 }, { "epoch": 16.97, "learning_rate": 4.823531816830211e-05, "loss": 1.4859, "step": 57500 }, { "epoch": 17.12, "learning_rate": 4.8216733571216176e-05, "loss": 1.4758, "step": 58000 }, { "epoch": 17.27, "learning_rate": 4.819814897413024e-05, "loss": 1.4669, "step": 58500 }, { "epoch": 17.41, "learning_rate": 4.8179564377044306e-05, "loss": 1.4597, "step": 59000 }, { "epoch": 17.56, "learning_rate": 4.8161016949152546e-05, "loss": 1.4498, "step": 59500 }, { "epoch": 17.71, "learning_rate": 4.8142432352066604e-05, "loss": 1.4422, "step": 60000 }, { "epoch": 17.86, "learning_rate": 4.812384775498067e-05, "loss": 1.4347, "step": 60500 }, { "epoch": 18.0, "learning_rate": 4.8105263157894735e-05, "loss": 1.4253, "step": 61000 }, { "epoch": 18.15, "learning_rate": 4.8086678560808806e-05, "loss": 1.4161, "step": 61500 }, { "epoch": 18.3, "learning_rate": 4.806813113291704e-05, "loss": 1.4107, "step": 62000 }, { "epoch": 18.45, "learning_rate": 4.8049546535831105e-05, "loss": 1.4052, "step": 62500 }, { "epoch": 18.6, "learning_rate": 4.803096193874517e-05, "loss": 1.3986, "step": 63000 }, { "epoch": 18.74, "learning_rate": 4.8012377341659235e-05, "loss": 1.3891, "step": 63500 }, { "epoch": 18.89, "learning_rate": 4.7993829913767475e-05, "loss": 1.3862, "step": 64000 }, { "epoch": 19.04, "learning_rate": 4.797524531668154e-05, "loss": 1.3764, "step": 64500 }, { "epoch": 19.19, "learning_rate": 4.7956660719595605e-05, "loss": 1.3687, "step": 65000 }, { "epoch": 19.33, "learning_rate": 4.793807612250967e-05, "loss": 1.3658, "step": 65500 }, { "epoch": 19.48, "learning_rate": 4.791949152542373e-05, "loss": 1.3577, "step": 66000 }, { "epoch": 19.63, "learning_rate": 4.790094409753197e-05, "loss": 1.3526, "step": 66500 }, { "epoch": 19.78, "learning_rate": 4.7882359500446034e-05, "loss": 1.3471, "step": 67000 }, { "epoch": 19.92, "learning_rate": 4.78637749033601e-05, "loss": 1.3417, "step": 67500 }, { "epoch": 20.07, "learning_rate": 4.7845190306274164e-05, "loss": 1.3349, "step": 68000 }, { "epoch": 20.22, "learning_rate": 4.782668004757657e-05, "loss": 1.3294, "step": 68500 }, { "epoch": 20.37, "learning_rate": 4.780809545049064e-05, "loss": 1.3268, "step": 69000 }, { "epoch": 20.51, "learning_rate": 4.77895108534047e-05, "loss": 1.3201, "step": 69500 }, { "epoch": 20.66, "learning_rate": 4.777092625631877e-05, "loss": 1.3142, "step": 70000 }, { "epoch": 20.81, "learning_rate": 4.775234165923283e-05, "loss": 1.3085, "step": 70500 }, { "epoch": 20.96, "learning_rate": 4.773375706214689e-05, "loss": 1.3043, "step": 71000 }, { "epoch": 21.1, "learning_rate": 4.7715172465060956e-05, "loss": 1.2965, "step": 71500 }, { "epoch": 21.25, "learning_rate": 4.76966250371692e-05, "loss": 1.2928, "step": 72000 }, { "epoch": 21.4, "learning_rate": 4.767804044008326e-05, "loss": 1.2882, "step": 72500 }, { "epoch": 21.55, "learning_rate": 4.765945584299733e-05, "loss": 1.2852, "step": 73000 }, { "epoch": 21.69, "learning_rate": 4.764087124591139e-05, "loss": 1.2786, "step": 73500 }, { "epoch": 21.84, "learning_rate": 4.762228664882546e-05, "loss": 1.2759, "step": 74000 }, { "epoch": 21.99, "learning_rate": 4.7603702051739515e-05, "loss": 1.2722, "step": 74500 }, { "epoch": 22.14, "learning_rate": 4.758511745465358e-05, "loss": 1.2661, "step": 75000 }, { "epoch": 22.28, "learning_rate": 4.756657002676182e-05, "loss": 1.2615, "step": 75500 }, { "epoch": 22.43, "learning_rate": 4.7547985429675886e-05, "loss": 1.2571, "step": 76000 }, { "epoch": 22.58, "learning_rate": 4.752940083258996e-05, "loss": 1.2531, "step": 76500 }, { "epoch": 22.73, "learning_rate": 4.7510816235504016e-05, "loss": 1.2517, "step": 77000 }, { "epoch": 22.87, "learning_rate": 4.749223163841808e-05, "loss": 1.2477, "step": 77500 }, { "epoch": 23.02, "learning_rate": 4.7473647041332146e-05, "loss": 1.2422, "step": 78000 }, { "epoch": 23.17, "learning_rate": 4.745506244424621e-05, "loss": 1.2389, "step": 78500 }, { "epoch": 23.32, "learning_rate": 4.7436477847160276e-05, "loss": 1.2372, "step": 79000 }, { "epoch": 23.47, "learning_rate": 4.7417930419268516e-05, "loss": 1.2314, "step": 79500 }, { "epoch": 23.61, "learning_rate": 4.739938299137675e-05, "loss": 1.2282, "step": 80000 }, { "epoch": 23.76, "learning_rate": 4.7380798394290815e-05, "loss": 1.2249, "step": 80500 }, { "epoch": 23.91, "learning_rate": 4.736221379720488e-05, "loss": 1.2211, "step": 81000 }, { "epoch": 24.06, "learning_rate": 4.7343629200118945e-05, "loss": 1.22, "step": 81500 }, { "epoch": 24.2, "learning_rate": 4.732508177222718e-05, "loss": 1.2131, "step": 82000 }, { "epoch": 24.35, "learning_rate": 4.730649717514124e-05, "loss": 1.2093, "step": 82500 }, { "epoch": 24.5, "learning_rate": 4.728791257805531e-05, "loss": 1.2078, "step": 83000 }, { "epoch": 24.65, "learning_rate": 4.7269327980969373e-05, "loss": 1.2069, "step": 83500 }, { "epoch": 24.79, "learning_rate": 4.725074338388344e-05, "loss": 1.2037, "step": 84000 }, { "epoch": 24.94, "learning_rate": 4.7232158786797504e-05, "loss": 1.2007, "step": 84500 }, { "epoch": 25.09, "learning_rate": 4.7213611358905744e-05, "loss": 1.1953, "step": 85000 }, { "epoch": 25.24, "learning_rate": 4.71950267618198e-05, "loss": 1.193, "step": 85500 }, { "epoch": 25.38, "learning_rate": 4.717644216473387e-05, "loss": 1.1923, "step": 86000 }, { "epoch": 25.53, "learning_rate": 4.715785756764793e-05, "loss": 1.1889, "step": 86500 }, { "epoch": 25.68, "learning_rate": 4.713931013975617e-05, "loss": 1.1844, "step": 87000 }, { "epoch": 25.83, "learning_rate": 4.712072554267024e-05, "loss": 1.1819, "step": 87500 }, { "epoch": 25.97, "learning_rate": 4.71021409455843e-05, "loss": 1.1791, "step": 88000 }, { "epoch": 26.12, "learning_rate": 4.708355634849837e-05, "loss": 1.1767, "step": 88500 }, { "epoch": 26.27, "learning_rate": 4.7064971751412426e-05, "loss": 1.1728, "step": 89000 }, { "epoch": 26.42, "learning_rate": 4.70463871543265e-05, "loss": 1.1702, "step": 89500 }, { "epoch": 26.56, "learning_rate": 4.702780255724056e-05, "loss": 1.1686, "step": 90000 }, { "epoch": 26.71, "learning_rate": 4.70092551293488e-05, "loss": 1.1677, "step": 90500 }, { "epoch": 26.86, "learning_rate": 4.699067053226287e-05, "loss": 1.1649, "step": 91000 }, { "epoch": 27.01, "learning_rate": 4.6972085935176926e-05, "loss": 1.1625, "step": 91500 }, { "epoch": 27.15, "learning_rate": 4.695350133809099e-05, "loss": 1.1587, "step": 92000 }, { "epoch": 27.3, "learning_rate": 4.6934916741005057e-05, "loss": 1.1565, "step": 92500 }, { "epoch": 27.45, "learning_rate": 4.691633214391912e-05, "loss": 1.156, "step": 93000 }, { "epoch": 27.6, "learning_rate": 4.689774754683319e-05, "loss": 1.1515, "step": 93500 }, { "epoch": 27.74, "learning_rate": 4.687920011894143e-05, "loss": 1.1487, "step": 94000 }, { "epoch": 27.89, "learning_rate": 4.686061552185549e-05, "loss": 1.1495, "step": 94500 }, { "epoch": 28.04, "learning_rate": 4.684203092476955e-05, "loss": 1.1444, "step": 95000 }, { "epoch": 28.19, "learning_rate": 4.6823446327683615e-05, "loss": 1.1424, "step": 95500 }, { "epoch": 28.34, "learning_rate": 4.680486173059768e-05, "loss": 1.1396, "step": 96000 }, { "epoch": 28.48, "learning_rate": 4.6786277133511746e-05, "loss": 1.1387, "step": 96500 }, { "epoch": 28.63, "learning_rate": 4.676769253642581e-05, "loss": 1.1351, "step": 97000 }, { "epoch": 28.78, "learning_rate": 4.6749107939339876e-05, "loss": 1.134, "step": 97500 }, { "epoch": 28.93, "learning_rate": 4.6730560511448116e-05, "loss": 1.1338, "step": 98000 } ], "max_steps": 1355200, "num_train_epochs": 400, "total_flos": 2.648101887798988e+19, "trial_name": null, "trial_params": null }