{ "best_metric": null, "best_model_checkpoint": null, "epoch": 27.0, "global_step": 101034, "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": 10.5048, "step": 1 }, { "epoch": 0.13, "learning_rate": 2.5e-06, "loss": 9.5536, "step": 500 }, { "epoch": 0.27, "learning_rate": 5e-06, "loss": 7.9704, "step": 1000 }, { "epoch": 0.4, "learning_rate": 7.5e-06, "loss": 6.9597, "step": 1500 }, { "epoch": 0.53, "learning_rate": 1e-05, "loss": 6.6577, "step": 2000 }, { "epoch": 0.67, "learning_rate": 1.25e-05, "loss": 6.4897, "step": 2500 }, { "epoch": 0.8, "learning_rate": 1.5e-05, "loss": 6.3759, "step": 3000 }, { "epoch": 0.94, "learning_rate": 1.75e-05, "loss": 6.2799, "step": 3500 }, { "epoch": 1.07, "learning_rate": 2e-05, "loss": 6.2039, "step": 4000 }, { "epoch": 1.2, "learning_rate": 2.25e-05, "loss": 6.1376, "step": 4500 }, { "epoch": 1.34, "learning_rate": 2.5e-05, "loss": 6.0826, "step": 5000 }, { "epoch": 1.47, "learning_rate": 2.7500000000000004e-05, "loss": 6.0309, "step": 5500 }, { "epoch": 1.6, "learning_rate": 3e-05, "loss": 5.9909, "step": 6000 }, { "epoch": 1.74, "learning_rate": 3.2500000000000004e-05, "loss": 5.9553, "step": 6500 }, { "epoch": 1.87, "learning_rate": 3.5e-05, "loss": 5.919, "step": 7000 }, { "epoch": 2.0, "learning_rate": 3.7500000000000003e-05, "loss": 5.8937, "step": 7500 }, { "epoch": 2.14, "learning_rate": 4e-05, "loss": 5.8604, "step": 8000 }, { "epoch": 2.27, "learning_rate": 4.2495e-05, "loss": 5.8383, "step": 8500 }, { "epoch": 2.41, "learning_rate": 4.4995000000000005e-05, "loss": 5.8167, "step": 9000 }, { "epoch": 2.54, "learning_rate": 4.7495e-05, "loss": 5.7994, "step": 9500 }, { "epoch": 2.67, "learning_rate": 4.9995000000000005e-05, "loss": 5.7835, "step": 10000 }, { "epoch": 2.81, "learning_rate": 4.998325262308313e-05, "loss": 5.7681, "step": 10500 }, { "epoch": 2.94, "learning_rate": 4.996643798762443e-05, "loss": 5.753, "step": 11000 }, { "epoch": 3.07, "learning_rate": 4.994962335216573e-05, "loss": 5.7391, "step": 11500 }, { "epoch": 3.21, "learning_rate": 4.993280871670703e-05, "loss": 5.7271, "step": 12000 }, { "epoch": 3.34, "learning_rate": 4.991602771051924e-05, "loss": 5.7125, "step": 12500 }, { "epoch": 3.47, "learning_rate": 4.9899213075060537e-05, "loss": 5.7069, "step": 13000 }, { "epoch": 3.61, "learning_rate": 4.988239843960183e-05, "loss": 5.6974, "step": 13500 }, { "epoch": 3.74, "learning_rate": 4.986558380414313e-05, "loss": 5.687, "step": 14000 }, { "epoch": 3.87, "learning_rate": 4.984880279795534e-05, "loss": 5.6798, "step": 14500 }, { "epoch": 4.01, "learning_rate": 4.983198816249664e-05, "loss": 5.6725, "step": 15000 }, { "epoch": 4.14, "learning_rate": 4.9815207156308854e-05, "loss": 5.6667, "step": 15500 }, { "epoch": 4.28, "learning_rate": 4.979839252085015e-05, "loss": 5.6581, "step": 16000 }, { "epoch": 4.41, "learning_rate": 4.9781577885391446e-05, "loss": 5.6536, "step": 16500 }, { "epoch": 4.54, "learning_rate": 4.9764763249932745e-05, "loss": 5.6439, "step": 17000 }, { "epoch": 4.68, "learning_rate": 4.974794861447404e-05, "loss": 5.6382, "step": 17500 }, { "epoch": 4.81, "learning_rate": 4.9731167608286254e-05, "loss": 5.6337, "step": 18000 }, { "epoch": 4.94, "learning_rate": 4.9714352972827547e-05, "loss": 5.6271, "step": 18500 }, { "epoch": 5.08, "learning_rate": 4.9697538337368846e-05, "loss": 5.6235, "step": 19000 }, { "epoch": 5.21, "learning_rate": 4.9680723701910145e-05, "loss": 5.6189, "step": 19500 }, { "epoch": 5.34, "learning_rate": 4.966394269572236e-05, "loss": 5.6165, "step": 20000 }, { "epoch": 5.48, "learning_rate": 4.9647128060263654e-05, "loss": 5.6132, "step": 20500 }, { "epoch": 5.61, "learning_rate": 4.9630313424804953e-05, "loss": 5.6078, "step": 21000 }, { "epoch": 5.75, "learning_rate": 4.9613498789346246e-05, "loss": 5.6022, "step": 21500 }, { "epoch": 5.88, "learning_rate": 4.9596684153887545e-05, "loss": 5.5965, "step": 22000 }, { "epoch": 6.01, "learning_rate": 4.9579903147699755e-05, "loss": 5.5932, "step": 22500 }, { "epoch": 6.15, "learning_rate": 4.9563088512241054e-05, "loss": 5.5905, "step": 23000 }, { "epoch": 6.28, "learning_rate": 4.9546273876782353e-05, "loss": 5.5882, "step": 23500 }, { "epoch": 6.41, "learning_rate": 4.952949287059457e-05, "loss": 5.5835, "step": 24000 }, { "epoch": 6.55, "learning_rate": 4.951267823513586e-05, "loss": 5.5795, "step": 24500 }, { "epoch": 6.68, "learning_rate": 4.949586359967716e-05, "loss": 5.5758, "step": 25000 }, { "epoch": 6.81, "learning_rate": 4.947904896421846e-05, "loss": 5.5759, "step": 25500 }, { "epoch": 6.95, "learning_rate": 4.9462234328759754e-05, "loss": 5.5725, "step": 26000 }, { "epoch": 7.08, "learning_rate": 4.944541969330105e-05, "loss": 5.5683, "step": 26500 }, { "epoch": 7.22, "learning_rate": 4.9428605057842345e-05, "loss": 5.5647, "step": 27000 }, { "epoch": 7.35, "learning_rate": 4.9411790422383644e-05, "loss": 5.5634, "step": 27500 }, { "epoch": 7.48, "learning_rate": 4.939500941619586e-05, "loss": 5.56, "step": 28000 }, { "epoch": 7.62, "learning_rate": 4.937822841000808e-05, "loss": 5.5573, "step": 28500 }, { "epoch": 7.75, "learning_rate": 4.936141377454937e-05, "loss": 5.5541, "step": 29000 }, { "epoch": 7.88, "learning_rate": 4.934459913909067e-05, "loss": 5.551, "step": 29500 }, { "epoch": 8.02, "learning_rate": 4.932778450363196e-05, "loss": 5.5479, "step": 30000 }, { "epoch": 8.15, "learning_rate": 4.931096986817326e-05, "loss": 5.5441, "step": 30500 }, { "epoch": 8.28, "learning_rate": 4.9294155232714554e-05, "loss": 5.5424, "step": 31000 }, { "epoch": 8.42, "learning_rate": 4.927737422652677e-05, "loss": 5.5381, "step": 31500 }, { "epoch": 8.55, "learning_rate": 4.926055959106807e-05, "loss": 5.5362, "step": 32000 }, { "epoch": 8.69, "learning_rate": 4.924374495560937e-05, "loss": 5.5352, "step": 32500 }, { "epoch": 8.82, "learning_rate": 4.922693032015066e-05, "loss": 5.5319, "step": 33000 }, { "epoch": 8.95, "learning_rate": 4.921011568469196e-05, "loss": 5.5338, "step": 33500 }, { "epoch": 9.09, "learning_rate": 4.919330104923325e-05, "loss": 5.5267, "step": 34000 }, { "epoch": 9.22, "learning_rate": 4.917652004304547e-05, "loss": 5.5255, "step": 34500 }, { "epoch": 9.35, "learning_rate": 4.915970540758676e-05, "loss": 5.5245, "step": 35000 }, { "epoch": 9.49, "learning_rate": 4.914289077212806e-05, "loss": 5.5187, "step": 35500 }, { "epoch": 9.62, "learning_rate": 4.912607613666936e-05, "loss": 5.5183, "step": 36000 }, { "epoch": 9.75, "learning_rate": 4.910929513048158e-05, "loss": 5.5174, "step": 36500 }, { "epoch": 9.89, "learning_rate": 4.909248049502287e-05, "loss": 5.5174, "step": 37000 }, { "epoch": 10.02, "learning_rate": 4.907566585956417e-05, "loss": 5.5139, "step": 37500 }, { "epoch": 10.15, "learning_rate": 4.905885122410546e-05, "loss": 5.5066, "step": 38000 }, { "epoch": 10.29, "learning_rate": 4.904203658864676e-05, "loss": 5.5077, "step": 38500 }, { "epoch": 10.42, "learning_rate": 4.902522195318805e-05, "loss": 5.5018, "step": 39000 }, { "epoch": 10.56, "learning_rate": 4.900844094700027e-05, "loss": 5.4997, "step": 39500 }, { "epoch": 10.69, "learning_rate": 4.899162631154157e-05, "loss": 5.3193, "step": 40000 }, { "epoch": 10.82, "learning_rate": 4.897481167608287e-05, "loss": 5.0833, "step": 40500 }, { "epoch": 10.96, "learning_rate": 4.895799704062416e-05, "loss": 4.8992, "step": 41000 }, { "epoch": 11.09, "learning_rate": 4.894118240516546e-05, "loss": 4.7196, "step": 41500 }, { "epoch": 11.22, "learning_rate": 4.892436776970675e-05, "loss": 4.552, "step": 42000 }, { "epoch": 11.36, "learning_rate": 4.890755313424805e-05, "loss": 4.3827, "step": 42500 }, { "epoch": 11.49, "learning_rate": 4.889073849878935e-05, "loss": 4.2208, "step": 43000 }, { "epoch": 11.62, "learning_rate": 4.887392386333064e-05, "loss": 4.0618, "step": 43500 }, { "epoch": 11.76, "learning_rate": 4.885710922787194e-05, "loss": 3.8934, "step": 44000 }, { "epoch": 11.89, "learning_rate": 4.884029459241324e-05, "loss": 3.6578, "step": 44500 }, { "epoch": 12.03, "learning_rate": 4.882347995695454e-05, "loss": 3.1504, "step": 45000 }, { "epoch": 12.16, "learning_rate": 4.880669895076675e-05, "loss": 2.52, "step": 45500 }, { "epoch": 12.29, "learning_rate": 4.878991794457896e-05, "loss": 2.2575, "step": 46000 }, { "epoch": 12.43, "learning_rate": 4.877310330912026e-05, "loss": 2.1152, "step": 46500 }, { "epoch": 12.56, "learning_rate": 4.875632230293247e-05, "loss": 2.0167, "step": 47000 }, { "epoch": 12.69, "learning_rate": 4.873950766747377e-05, "loss": 1.9382, "step": 47500 }, { "epoch": 12.83, "learning_rate": 4.872269303201507e-05, "loss": 1.8744, "step": 48000 }, { "epoch": 12.96, "learning_rate": 4.870587839655637e-05, "loss": 1.8223, "step": 48500 }, { "epoch": 13.09, "learning_rate": 4.868906376109767e-05, "loss": 1.773, "step": 49000 }, { "epoch": 13.23, "learning_rate": 4.867224912563896e-05, "loss": 1.7323, "step": 49500 }, { "epoch": 13.36, "learning_rate": 4.8655468119451176e-05, "loss": 1.6973, "step": 50000 }, { "epoch": 13.5, "learning_rate": 4.863865348399247e-05, "loss": 1.6658, "step": 50500 }, { "epoch": 13.63, "learning_rate": 4.862183884853377e-05, "loss": 1.6321, "step": 51000 }, { "epoch": 13.76, "learning_rate": 4.860502421307506e-05, "loss": 1.6099, "step": 51500 }, { "epoch": 13.9, "learning_rate": 4.858820957761636e-05, "loss": 1.5828, "step": 52000 }, { "epoch": 14.03, "learning_rate": 4.857139494215766e-05, "loss": 1.5617, "step": 52500 }, { "epoch": 14.16, "learning_rate": 4.855458030669896e-05, "loss": 1.539, "step": 53000 }, { "epoch": 14.3, "learning_rate": 4.853779930051117e-05, "loss": 1.5203, "step": 53500 }, { "epoch": 14.43, "learning_rate": 4.852098466505247e-05, "loss": 1.5006, "step": 54000 }, { "epoch": 14.56, "learning_rate": 4.850417002959376e-05, "loss": 1.4819, "step": 54500 }, { "epoch": 14.7, "learning_rate": 4.848735539413506e-05, "loss": 1.4656, "step": 55000 }, { "epoch": 14.83, "learning_rate": 4.847054075867635e-05, "loss": 1.4481, "step": 55500 }, { "epoch": 14.97, "learning_rate": 4.845372612321765e-05, "loss": 1.4359, "step": 56000 }, { "epoch": 15.1, "learning_rate": 4.843691148775894e-05, "loss": 1.4202, "step": 56500 }, { "epoch": 15.23, "learning_rate": 4.842009685230024e-05, "loss": 1.4077, "step": 57000 }, { "epoch": 15.37, "learning_rate": 4.840328221684154e-05, "loss": 1.3939, "step": 57500 }, { "epoch": 15.5, "learning_rate": 4.838650121065376e-05, "loss": 1.3803, "step": 58000 }, { "epoch": 15.63, "learning_rate": 4.836968657519505e-05, "loss": 1.37, "step": 58500 }, { "epoch": 15.77, "learning_rate": 4.835287193973635e-05, "loss": 1.3594, "step": 59000 }, { "epoch": 15.9, "learning_rate": 4.833605730427764e-05, "loss": 1.3471, "step": 59500 }, { "epoch": 16.03, "learning_rate": 4.831924266881894e-05, "loss": 1.3378, "step": 60000 }, { "epoch": 16.17, "learning_rate": 4.8302428033360234e-05, "loss": 1.3292, "step": 60500 }, { "epoch": 16.3, "learning_rate": 4.828564702717246e-05, "loss": 1.317, "step": 61000 }, { "epoch": 16.44, "learning_rate": 4.826883239171375e-05, "loss": 1.3105, "step": 61500 }, { "epoch": 16.57, "learning_rate": 4.825201775625505e-05, "loss": 1.3016, "step": 62000 }, { "epoch": 16.7, "learning_rate": 4.823520312079635e-05, "loss": 1.2924, "step": 62500 }, { "epoch": 16.84, "learning_rate": 4.821838848533764e-05, "loss": 1.2859, "step": 63000 }, { "epoch": 16.97, "learning_rate": 4.820160747914985e-05, "loss": 1.2765, "step": 63500 }, { "epoch": 17.1, "learning_rate": 4.818482647296207e-05, "loss": 1.2693, "step": 64000 }, { "epoch": 17.24, "learning_rate": 4.8168011837503366e-05, "loss": 1.26, "step": 64500 }, { "epoch": 17.37, "learning_rate": 4.8151197202044665e-05, "loss": 1.2559, "step": 65000 }, { "epoch": 17.5, "learning_rate": 4.813438256658596e-05, "loss": 1.2461, "step": 65500 }, { "epoch": 17.64, "learning_rate": 4.811756793112726e-05, "loss": 1.2371, "step": 66000 }, { "epoch": 17.77, "learning_rate": 4.8100753295668556e-05, "loss": 1.2331, "step": 66500 }, { "epoch": 17.9, "learning_rate": 4.808393866020985e-05, "loss": 1.2268, "step": 67000 }, { "epoch": 18.04, "learning_rate": 4.8067157654022065e-05, "loss": 1.2191, "step": 67500 }, { "epoch": 18.17, "learning_rate": 4.805034301856336e-05, "loss": 1.2137, "step": 68000 }, { "epoch": 18.31, "learning_rate": 4.803352838310466e-05, "loss": 1.2077, "step": 68500 }, { "epoch": 18.44, "learning_rate": 4.801671374764595e-05, "loss": 1.2037, "step": 69000 }, { "epoch": 18.57, "learning_rate": 4.799993274145817e-05, "loss": 1.1989, "step": 69500 }, { "epoch": 18.71, "learning_rate": 4.7983118105999466e-05, "loss": 1.191, "step": 70000 }, { "epoch": 18.84, "learning_rate": 4.7966303470540765e-05, "loss": 1.1843, "step": 70500 }, { "epoch": 18.97, "learning_rate": 4.794948883508206e-05, "loss": 1.1815, "step": 71000 }, { "epoch": 19.11, "learning_rate": 4.7932707828894274e-05, "loss": 1.1736, "step": 71500 }, { "epoch": 19.24, "learning_rate": 4.7915893193435566e-05, "loss": 1.1709, "step": 72000 }, { "epoch": 19.37, "learning_rate": 4.7899078557976866e-05, "loss": 1.1651, "step": 72500 }, { "epoch": 19.51, "learning_rate": 4.788226392251816e-05, "loss": 1.1602, "step": 73000 }, { "epoch": 19.64, "learning_rate": 4.786548291633038e-05, "loss": 1.1588, "step": 73500 }, { "epoch": 19.78, "learning_rate": 4.7848668280871674e-05, "loss": 1.1525, "step": 74000 }, { "epoch": 19.91, "learning_rate": 4.783185364541297e-05, "loss": 1.1486, "step": 74500 }, { "epoch": 20.04, "learning_rate": 4.7815039009954266e-05, "loss": 1.1437, "step": 75000 }, { "epoch": 20.18, "learning_rate": 4.779825800376648e-05, "loss": 1.1383, "step": 75500 }, { "epoch": 20.31, "learning_rate": 4.7781443368307775e-05, "loss": 1.1339, "step": 76000 }, { "epoch": 20.44, "learning_rate": 4.776466236211999e-05, "loss": 1.1319, "step": 76500 }, { "epoch": 20.58, "learning_rate": 4.774784772666129e-05, "loss": 1.1266, "step": 77000 }, { "epoch": 20.71, "learning_rate": 4.773103309120259e-05, "loss": 1.1244, "step": 77500 }, { "epoch": 20.84, "learning_rate": 4.771421845574388e-05, "loss": 1.1191, "step": 78000 }, { "epoch": 20.98, "learning_rate": 4.769740382028518e-05, "loss": 1.1156, "step": 78500 }, { "epoch": 21.11, "learning_rate": 4.768062281409739e-05, "loss": 1.1107, "step": 79000 }, { "epoch": 21.25, "learning_rate": 4.766380817863869e-05, "loss": 1.1068, "step": 79500 }, { "epoch": 21.38, "learning_rate": 4.764699354317998e-05, "loss": 1.1029, "step": 80000 }, { "epoch": 21.51, "learning_rate": 4.763017890772128e-05, "loss": 1.0994, "step": 80500 }, { "epoch": 21.65, "learning_rate": 4.761336427226258e-05, "loss": 1.0979, "step": 81000 }, { "epoch": 21.78, "learning_rate": 4.759654963680388e-05, "loss": 1.0913, "step": 81500 }, { "epoch": 21.91, "learning_rate": 4.757976863061609e-05, "loss": 1.09, "step": 82000 }, { "epoch": 22.05, "learning_rate": 4.756295399515739e-05, "loss": 1.0872, "step": 82500 }, { "epoch": 22.18, "learning_rate": 4.754613935969868e-05, "loss": 1.0837, "step": 83000 }, { "epoch": 22.31, "learning_rate": 4.752932472423998e-05, "loss": 1.0798, "step": 83500 }, { "epoch": 22.45, "learning_rate": 4.7512510088781274e-05, "loss": 1.0777, "step": 84000 }, { "epoch": 22.58, "learning_rate": 4.749569545332257e-05, "loss": 1.0765, "step": 84500 }, { "epoch": 22.72, "learning_rate": 4.7478880817863866e-05, "loss": 1.0731, "step": 85000 }, { "epoch": 22.85, "learning_rate": 4.746209981167609e-05, "loss": 1.0673, "step": 85500 }, { "epoch": 22.98, "learning_rate": 4.744528517621738e-05, "loss": 1.0655, "step": 86000 }, { "epoch": 23.12, "learning_rate": 4.742847054075868e-05, "loss": 1.0623, "step": 86500 }, { "epoch": 23.25, "learning_rate": 4.741165590529997e-05, "loss": 1.0599, "step": 87000 }, { "epoch": 23.38, "learning_rate": 4.739487489911219e-05, "loss": 1.0587, "step": 87500 }, { "epoch": 23.52, "learning_rate": 4.737806026365348e-05, "loss": 1.0544, "step": 88000 }, { "epoch": 23.65, "learning_rate": 4.736124562819478e-05, "loss": 1.0527, "step": 88500 }, { "epoch": 23.78, "learning_rate": 4.7344430992736074e-05, "loss": 1.0489, "step": 89000 }, { "epoch": 23.92, "learning_rate": 4.73276499865483e-05, "loss": 1.0463, "step": 89500 }, { "epoch": 24.05, "learning_rate": 4.731083535108959e-05, "loss": 1.0447, "step": 90000 }, { "epoch": 24.18, "learning_rate": 4.729402071563089e-05, "loss": 1.0401, "step": 90500 }, { "epoch": 24.32, "learning_rate": 4.727720608017218e-05, "loss": 1.04, "step": 91000 }, { "epoch": 24.45, "learning_rate": 4.726039144471348e-05, "loss": 1.0371, "step": 91500 }, { "epoch": 24.59, "learning_rate": 4.724364406779661e-05, "loss": 1.0343, "step": 92000 }, { "epoch": 24.72, "learning_rate": 4.722682943233791e-05, "loss": 1.0336, "step": 92500 }, { "epoch": 24.85, "learning_rate": 4.721001479687921e-05, "loss": 1.0307, "step": 93000 }, { "epoch": 24.99, "learning_rate": 4.7193200161420506e-05, "loss": 1.0294, "step": 93500 }, { "epoch": 25.12, "learning_rate": 4.71763855259618e-05, "loss": 1.0243, "step": 94000 }, { "epoch": 25.25, "learning_rate": 4.71595708905031e-05, "loss": 1.0225, "step": 94500 }, { "epoch": 25.39, "learning_rate": 4.714275625504439e-05, "loss": 1.0205, "step": 95000 }, { "epoch": 25.52, "learning_rate": 4.712594161958569e-05, "loss": 1.0201, "step": 95500 }, { "epoch": 25.65, "learning_rate": 4.710912698412699e-05, "loss": 1.0171, "step": 96000 }, { "epoch": 25.79, "learning_rate": 4.7092379607210116e-05, "loss": 1.0139, "step": 96500 }, { "epoch": 25.92, "learning_rate": 4.7075564971751415e-05, "loss": 1.0131, "step": 97000 }, { "epoch": 26.06, "learning_rate": 4.7058750336292714e-05, "loss": 1.0114, "step": 97500 }, { "epoch": 26.19, "learning_rate": 4.704193570083401e-05, "loss": 1.0082, "step": 98000 }, { "epoch": 26.32, "learning_rate": 4.7025121065375306e-05, "loss": 1.0081, "step": 98500 }, { "epoch": 26.46, "learning_rate": 4.7008306429916605e-05, "loss": 1.0042, "step": 99000 }, { "epoch": 26.59, "learning_rate": 4.69914917944579e-05, "loss": 1.0034, "step": 99500 }, { "epoch": 26.72, "learning_rate": 4.697471078827011e-05, "loss": 1.0016, "step": 100000 }, { "epoch": 26.86, "learning_rate": 4.695789615281141e-05, "loss": 0.9991, "step": 100500 }, { "epoch": 26.99, "learning_rate": 4.6941115146623624e-05, "loss": 0.9963, "step": 101000 } ], "max_steps": 1496800, "num_train_epochs": 400, "total_flos": 2.722996359502024e+19, "trial_name": null, "trial_params": null }