{ "best_metric": null, "best_model_checkpoint": null, "epoch": 10.0, "eval_steps": 500, "global_step": 430750, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.01, "learning_rate": 4.994196169471852e-05, "loss": 0.4992, "step": 500 }, { "epoch": 0.02, "learning_rate": 4.988392338943703e-05, "loss": 0.4491, "step": 1000 }, { "epoch": 0.03, "learning_rate": 4.982588508415554e-05, "loss": 0.4336, "step": 1500 }, { "epoch": 0.05, "learning_rate": 4.976784677887406e-05, "loss": 0.4251, "step": 2000 }, { "epoch": 0.06, "learning_rate": 4.970980847359257e-05, "loss": 0.4192, "step": 2500 }, { "epoch": 0.07, "learning_rate": 4.965177016831109e-05, "loss": 0.4137, "step": 3000 }, { "epoch": 0.08, "learning_rate": 4.95937318630296e-05, "loss": 0.4099, "step": 3500 }, { "epoch": 0.09, "learning_rate": 4.953569355774811e-05, "loss": 0.4068, "step": 4000 }, { "epoch": 0.1, "learning_rate": 4.947765525246663e-05, "loss": 0.4041, "step": 4500 }, { "epoch": 0.12, "learning_rate": 4.941961694718514e-05, "loss": 0.4017, "step": 5000 }, { "epoch": 0.13, "learning_rate": 4.936157864190366e-05, "loss": 0.3992, "step": 5500 }, { "epoch": 0.14, "learning_rate": 4.930354033662217e-05, "loss": 0.3976, "step": 6000 }, { "epoch": 0.15, "learning_rate": 4.924550203134069e-05, "loss": 0.3958, "step": 6500 }, { "epoch": 0.16, "learning_rate": 4.91874637260592e-05, "loss": 0.3942, "step": 7000 }, { "epoch": 0.17, "learning_rate": 4.912942542077771e-05, "loss": 0.393, "step": 7500 }, { "epoch": 0.19, "learning_rate": 4.907138711549623e-05, "loss": 0.3915, "step": 8000 }, { "epoch": 0.2, "learning_rate": 4.901334881021474e-05, "loss": 0.3904, "step": 8500 }, { "epoch": 0.21, "learning_rate": 4.895531050493326e-05, "loss": 0.3894, "step": 9000 }, { "epoch": 0.22, "learning_rate": 4.889727219965177e-05, "loss": 0.388, "step": 9500 }, { "epoch": 0.23, "learning_rate": 4.883923389437029e-05, "loss": 0.387, "step": 10000 }, { "epoch": 0.24, "learning_rate": 4.87811955890888e-05, "loss": 0.386, "step": 10500 }, { "epoch": 0.26, "learning_rate": 4.872315728380731e-05, "loss": 0.385, "step": 11000 }, { "epoch": 0.27, "learning_rate": 4.866511897852583e-05, "loss": 0.384, "step": 11500 }, { "epoch": 0.28, "learning_rate": 4.860708067324434e-05, "loss": 0.3833, "step": 12000 }, { "epoch": 0.29, "learning_rate": 4.854904236796286e-05, "loss": 0.3829, "step": 12500 }, { "epoch": 0.3, "learning_rate": 4.849100406268137e-05, "loss": 0.3819, "step": 13000 }, { "epoch": 0.31, "learning_rate": 4.8432965757399886e-05, "loss": 0.3811, "step": 13500 }, { "epoch": 0.33, "learning_rate": 4.83749274521184e-05, "loss": 0.3805, "step": 14000 }, { "epoch": 0.34, "learning_rate": 4.831688914683691e-05, "loss": 0.3799, "step": 14500 }, { "epoch": 0.35, "learning_rate": 4.825885084155543e-05, "loss": 0.379, "step": 15000 }, { "epoch": 0.36, "learning_rate": 4.820081253627394e-05, "loss": 0.3787, "step": 15500 }, { "epoch": 0.37, "learning_rate": 4.814277423099246e-05, "loss": 0.3779, "step": 16000 }, { "epoch": 0.38, "learning_rate": 4.808473592571097e-05, "loss": 0.3776, "step": 16500 }, { "epoch": 0.39, "learning_rate": 4.8026697620429486e-05, "loss": 0.3766, "step": 17000 }, { "epoch": 0.41, "learning_rate": 4.7968659315148e-05, "loss": 0.3759, "step": 17500 }, { "epoch": 0.42, "learning_rate": 4.791062100986651e-05, "loss": 0.376, "step": 18000 }, { "epoch": 0.43, "learning_rate": 4.785258270458503e-05, "loss": 0.3751, "step": 18500 }, { "epoch": 0.44, "learning_rate": 4.779454439930354e-05, "loss": 0.375, "step": 19000 }, { "epoch": 0.45, "learning_rate": 4.7736506094022056e-05, "loss": 0.3748, "step": 19500 }, { "epoch": 0.46, "learning_rate": 4.767846778874057e-05, "loss": 0.3738, "step": 20000 }, { "epoch": 0.48, "learning_rate": 4.7620429483459086e-05, "loss": 0.3739, "step": 20500 }, { "epoch": 0.49, "learning_rate": 4.75623911781776e-05, "loss": 0.3727, "step": 21000 }, { "epoch": 0.5, "learning_rate": 4.750435287289611e-05, "loss": 0.3726, "step": 21500 }, { "epoch": 0.51, "learning_rate": 4.744631456761463e-05, "loss": 0.3719, "step": 22000 }, { "epoch": 0.52, "learning_rate": 4.738827626233314e-05, "loss": 0.372, "step": 22500 }, { "epoch": 0.53, "learning_rate": 4.7330237957051655e-05, "loss": 0.3714, "step": 23000 }, { "epoch": 0.55, "learning_rate": 4.727219965177017e-05, "loss": 0.3714, "step": 23500 }, { "epoch": 0.56, "learning_rate": 4.7214161346488686e-05, "loss": 0.3706, "step": 24000 }, { "epoch": 0.57, "learning_rate": 4.71561230412072e-05, "loss": 0.37, "step": 24500 }, { "epoch": 0.58, "learning_rate": 4.709808473592571e-05, "loss": 0.37, "step": 25000 }, { "epoch": 0.59, "learning_rate": 4.704004643064423e-05, "loss": 0.3698, "step": 25500 }, { "epoch": 0.6, "learning_rate": 4.698200812536274e-05, "loss": 0.3695, "step": 26000 }, { "epoch": 0.62, "learning_rate": 4.6923969820081255e-05, "loss": 0.3692, "step": 26500 }, { "epoch": 0.63, "learning_rate": 4.686593151479977e-05, "loss": 0.3686, "step": 27000 }, { "epoch": 0.64, "learning_rate": 4.6807893209518286e-05, "loss": 0.3682, "step": 27500 }, { "epoch": 0.65, "learning_rate": 4.67498549042368e-05, "loss": 0.3682, "step": 28000 }, { "epoch": 0.66, "learning_rate": 4.669181659895531e-05, "loss": 0.3677, "step": 28500 }, { "epoch": 0.67, "learning_rate": 4.6633778293673825e-05, "loss": 0.3673, "step": 29000 }, { "epoch": 0.68, "learning_rate": 4.657573998839234e-05, "loss": 0.367, "step": 29500 }, { "epoch": 0.7, "learning_rate": 4.6517701683110855e-05, "loss": 0.3664, "step": 30000 }, { "epoch": 0.71, "learning_rate": 4.645966337782937e-05, "loss": 0.3665, "step": 30500 }, { "epoch": 0.72, "learning_rate": 4.6401625072547886e-05, "loss": 0.3664, "step": 31000 }, { "epoch": 0.73, "learning_rate": 4.63435867672664e-05, "loss": 0.366, "step": 31500 }, { "epoch": 0.74, "learning_rate": 4.628554846198491e-05, "loss": 0.3654, "step": 32000 }, { "epoch": 0.75, "learning_rate": 4.6227510156703424e-05, "loss": 0.3652, "step": 32500 }, { "epoch": 0.77, "learning_rate": 4.616947185142194e-05, "loss": 0.3654, "step": 33000 }, { "epoch": 0.78, "learning_rate": 4.6111433546140455e-05, "loss": 0.365, "step": 33500 }, { "epoch": 0.79, "learning_rate": 4.605339524085897e-05, "loss": 0.3642, "step": 34000 }, { "epoch": 0.8, "learning_rate": 4.5995356935577485e-05, "loss": 0.3643, "step": 34500 }, { "epoch": 0.81, "learning_rate": 4.5937318630296e-05, "loss": 0.3644, "step": 35000 }, { "epoch": 0.82, "learning_rate": 4.587928032501451e-05, "loss": 0.364, "step": 35500 }, { "epoch": 0.84, "learning_rate": 4.5821242019733024e-05, "loss": 0.3638, "step": 36000 }, { "epoch": 0.85, "learning_rate": 4.576320371445154e-05, "loss": 0.3633, "step": 36500 }, { "epoch": 0.86, "learning_rate": 4.5705165409170055e-05, "loss": 0.3636, "step": 37000 }, { "epoch": 0.87, "learning_rate": 4.564712710388857e-05, "loss": 0.3629, "step": 37500 }, { "epoch": 0.88, "learning_rate": 4.5589088798607085e-05, "loss": 0.3629, "step": 38000 }, { "epoch": 0.89, "learning_rate": 4.5531050493325594e-05, "loss": 0.3625, "step": 38500 }, { "epoch": 0.91, "learning_rate": 4.547301218804411e-05, "loss": 0.3629, "step": 39000 }, { "epoch": 0.92, "learning_rate": 4.5414973882762624e-05, "loss": 0.362, "step": 39500 }, { "epoch": 0.93, "learning_rate": 4.535693557748114e-05, "loss": 0.3619, "step": 40000 }, { "epoch": 0.94, "learning_rate": 4.5298897272199654e-05, "loss": 0.3619, "step": 40500 }, { "epoch": 0.95, "learning_rate": 4.524085896691817e-05, "loss": 0.3612, "step": 41000 }, { "epoch": 0.96, "learning_rate": 4.5182820661636685e-05, "loss": 0.3616, "step": 41500 }, { "epoch": 0.98, "learning_rate": 4.5124782356355193e-05, "loss": 0.3608, "step": 42000 }, { "epoch": 0.99, "learning_rate": 4.506674405107371e-05, "loss": 0.361, "step": 42500 }, { "epoch": 1.0, "learning_rate": 4.5008705745792224e-05, "loss": 0.3608, "step": 43000 }, { "epoch": 1.0, "eval_accuracy": 0.8883345013697975, "eval_loss": 0.35087689757347107, "eval_runtime": 2032.4427, "eval_samples_per_second": 188.179, "eval_steps_per_second": 2.352, "step": 43075 }, { "epoch": 1.01, "learning_rate": 4.495066744051074e-05, "loss": 0.3596, "step": 43500 }, { "epoch": 1.02, "learning_rate": 4.4892629135229254e-05, "loss": 0.3592, "step": 44000 }, { "epoch": 1.03, "learning_rate": 4.483459082994777e-05, "loss": 0.3594, "step": 44500 }, { "epoch": 1.04, "learning_rate": 4.4776552524666285e-05, "loss": 0.3594, "step": 45000 }, { "epoch": 1.06, "learning_rate": 4.471851421938479e-05, "loss": 0.3592, "step": 45500 }, { "epoch": 1.07, "learning_rate": 4.466047591410331e-05, "loss": 0.359, "step": 46000 }, { "epoch": 1.08, "learning_rate": 4.4602437608821824e-05, "loss": 0.3586, "step": 46500 }, { "epoch": 1.09, "learning_rate": 4.454439930354034e-05, "loss": 0.3586, "step": 47000 }, { "epoch": 1.1, "learning_rate": 4.4486360998258854e-05, "loss": 0.3583, "step": 47500 }, { "epoch": 1.11, "learning_rate": 4.442832269297737e-05, "loss": 0.3584, "step": 48000 }, { "epoch": 1.13, "learning_rate": 4.4370284387695885e-05, "loss": 0.3582, "step": 48500 }, { "epoch": 1.14, "learning_rate": 4.431224608241439e-05, "loss": 0.3579, "step": 49000 }, { "epoch": 1.15, "learning_rate": 4.425420777713291e-05, "loss": 0.3581, "step": 49500 }, { "epoch": 1.16, "learning_rate": 4.4196169471851423e-05, "loss": 0.3575, "step": 50000 }, { "epoch": 1.17, "learning_rate": 4.413813116656994e-05, "loss": 0.3575, "step": 50500 }, { "epoch": 1.18, "learning_rate": 4.4080092861288454e-05, "loss": 0.3573, "step": 51000 }, { "epoch": 1.2, "learning_rate": 4.402205455600696e-05, "loss": 0.3574, "step": 51500 }, { "epoch": 1.21, "learning_rate": 4.3964016250725484e-05, "loss": 0.357, "step": 52000 }, { "epoch": 1.22, "learning_rate": 4.390597794544399e-05, "loss": 0.357, "step": 52500 }, { "epoch": 1.23, "learning_rate": 4.384793964016251e-05, "loss": 0.3568, "step": 53000 }, { "epoch": 1.24, "learning_rate": 4.378990133488102e-05, "loss": 0.3569, "step": 53500 }, { "epoch": 1.25, "learning_rate": 4.373186302959954e-05, "loss": 0.3569, "step": 54000 }, { "epoch": 1.27, "learning_rate": 4.3673824724318054e-05, "loss": 0.3564, "step": 54500 }, { "epoch": 1.28, "learning_rate": 4.361578641903656e-05, "loss": 0.3562, "step": 55000 }, { "epoch": 1.29, "learning_rate": 4.3557748113755084e-05, "loss": 0.3562, "step": 55500 }, { "epoch": 1.3, "learning_rate": 4.349970980847359e-05, "loss": 0.3559, "step": 56000 }, { "epoch": 1.31, "learning_rate": 4.344167150319211e-05, "loss": 0.3555, "step": 56500 }, { "epoch": 1.32, "learning_rate": 4.338363319791062e-05, "loss": 0.3554, "step": 57000 }, { "epoch": 1.33, "learning_rate": 4.332559489262914e-05, "loss": 0.3552, "step": 57500 }, { "epoch": 1.35, "learning_rate": 4.3267556587347654e-05, "loss": 0.3553, "step": 58000 }, { "epoch": 1.36, "learning_rate": 4.320951828206616e-05, "loss": 0.3552, "step": 58500 }, { "epoch": 1.37, "learning_rate": 4.3151479976784684e-05, "loss": 0.3551, "step": 59000 }, { "epoch": 1.38, "learning_rate": 4.309344167150319e-05, "loss": 0.3554, "step": 59500 }, { "epoch": 1.39, "learning_rate": 4.303540336622171e-05, "loss": 0.3548, "step": 60000 }, { "epoch": 1.4, "learning_rate": 4.297736506094022e-05, "loss": 0.3547, "step": 60500 }, { "epoch": 1.42, "learning_rate": 4.291932675565873e-05, "loss": 0.3548, "step": 61000 }, { "epoch": 1.43, "learning_rate": 4.286128845037725e-05, "loss": 0.3545, "step": 61500 }, { "epoch": 1.44, "learning_rate": 4.280325014509576e-05, "loss": 0.3543, "step": 62000 }, { "epoch": 1.45, "learning_rate": 4.2745211839814284e-05, "loss": 0.3542, "step": 62500 }, { "epoch": 1.46, "learning_rate": 4.268717353453279e-05, "loss": 0.3544, "step": 63000 }, { "epoch": 1.47, "learning_rate": 4.262913522925131e-05, "loss": 0.3542, "step": 63500 }, { "epoch": 1.49, "learning_rate": 4.257109692396982e-05, "loss": 0.3544, "step": 64000 }, { "epoch": 1.5, "learning_rate": 4.251305861868833e-05, "loss": 0.3541, "step": 64500 }, { "epoch": 1.51, "learning_rate": 4.245502031340685e-05, "loss": 0.354, "step": 65000 }, { "epoch": 1.52, "learning_rate": 4.239698200812536e-05, "loss": 0.3538, "step": 65500 }, { "epoch": 1.53, "learning_rate": 4.2338943702843884e-05, "loss": 0.3536, "step": 66000 }, { "epoch": 1.54, "learning_rate": 4.228090539756239e-05, "loss": 0.3532, "step": 66500 }, { "epoch": 1.56, "learning_rate": 4.222286709228091e-05, "loss": 0.3533, "step": 67000 }, { "epoch": 1.57, "learning_rate": 4.216482878699942e-05, "loss": 0.3535, "step": 67500 }, { "epoch": 1.58, "learning_rate": 4.210679048171793e-05, "loss": 0.3536, "step": 68000 }, { "epoch": 1.59, "learning_rate": 4.204875217643645e-05, "loss": 0.353, "step": 68500 }, { "epoch": 1.6, "learning_rate": 4.199071387115496e-05, "loss": 0.3528, "step": 69000 }, { "epoch": 1.61, "learning_rate": 4.1932675565873483e-05, "loss": 0.3524, "step": 69500 }, { "epoch": 1.63, "learning_rate": 4.187463726059199e-05, "loss": 0.3524, "step": 70000 }, { "epoch": 1.64, "learning_rate": 4.181659895531051e-05, "loss": 0.3523, "step": 70500 }, { "epoch": 1.65, "learning_rate": 4.175856065002902e-05, "loss": 0.3524, "step": 71000 }, { "epoch": 1.66, "learning_rate": 4.170052234474753e-05, "loss": 0.3524, "step": 71500 }, { "epoch": 1.67, "learning_rate": 4.164248403946605e-05, "loss": 0.3523, "step": 72000 }, { "epoch": 1.68, "learning_rate": 4.158444573418456e-05, "loss": 0.3524, "step": 72500 }, { "epoch": 1.69, "learning_rate": 4.152640742890308e-05, "loss": 0.3518, "step": 73000 }, { "epoch": 1.71, "learning_rate": 4.146836912362159e-05, "loss": 0.3519, "step": 73500 }, { "epoch": 1.72, "learning_rate": 4.141033081834011e-05, "loss": 0.3516, "step": 74000 }, { "epoch": 1.73, "learning_rate": 4.135229251305862e-05, "loss": 0.3511, "step": 74500 }, { "epoch": 1.74, "learning_rate": 4.129425420777713e-05, "loss": 0.3515, "step": 75000 }, { "epoch": 1.75, "learning_rate": 4.123621590249565e-05, "loss": 0.3517, "step": 75500 }, { "epoch": 1.76, "learning_rate": 4.117817759721416e-05, "loss": 0.3515, "step": 76000 }, { "epoch": 1.78, "learning_rate": 4.112013929193268e-05, "loss": 0.3513, "step": 76500 }, { "epoch": 1.79, "learning_rate": 4.106210098665119e-05, "loss": 0.3509, "step": 77000 }, { "epoch": 1.8, "learning_rate": 4.100406268136971e-05, "loss": 0.3512, "step": 77500 }, { "epoch": 1.81, "learning_rate": 4.094602437608822e-05, "loss": 0.351, "step": 78000 }, { "epoch": 1.82, "learning_rate": 4.088798607080673e-05, "loss": 0.3511, "step": 78500 }, { "epoch": 1.83, "learning_rate": 4.082994776552525e-05, "loss": 0.3507, "step": 79000 }, { "epoch": 1.85, "learning_rate": 4.077190946024376e-05, "loss": 0.3506, "step": 79500 }, { "epoch": 1.86, "learning_rate": 4.071387115496228e-05, "loss": 0.3507, "step": 80000 }, { "epoch": 1.87, "learning_rate": 4.065583284968079e-05, "loss": 0.3508, "step": 80500 }, { "epoch": 1.88, "learning_rate": 4.0597794544399307e-05, "loss": 0.3501, "step": 81000 }, { "epoch": 1.89, "learning_rate": 4.053975623911782e-05, "loss": 0.3507, "step": 81500 }, { "epoch": 1.9, "learning_rate": 4.048171793383633e-05, "loss": 0.3505, "step": 82000 }, { "epoch": 1.92, "learning_rate": 4.042367962855485e-05, "loss": 0.3501, "step": 82500 }, { "epoch": 1.93, "learning_rate": 4.036564132327336e-05, "loss": 0.35, "step": 83000 }, { "epoch": 1.94, "learning_rate": 4.0307603017991876e-05, "loss": 0.3498, "step": 83500 }, { "epoch": 1.95, "learning_rate": 4.024956471271039e-05, "loss": 0.3499, "step": 84000 }, { "epoch": 1.96, "learning_rate": 4.0191526407428906e-05, "loss": 0.3496, "step": 84500 }, { "epoch": 1.97, "learning_rate": 4.013348810214742e-05, "loss": 0.35, "step": 85000 }, { "epoch": 1.98, "learning_rate": 4.007544979686593e-05, "loss": 0.3496, "step": 85500 }, { "epoch": 2.0, "learning_rate": 4.001741149158445e-05, "loss": 0.3495, "step": 86000 }, { "epoch": 2.0, "eval_accuracy": 0.8910229909772815, "eval_loss": 0.3408386707305908, "eval_runtime": 1994.5363, "eval_samples_per_second": 191.756, "eval_steps_per_second": 2.397, "step": 86150 }, { "epoch": 2.01, "learning_rate": 3.995937318630296e-05, "loss": 0.3486, "step": 86500 }, { "epoch": 2.02, "learning_rate": 3.9901334881021476e-05, "loss": 0.3487, "step": 87000 }, { "epoch": 2.03, "learning_rate": 3.984329657573999e-05, "loss": 0.3486, "step": 87500 }, { "epoch": 2.04, "learning_rate": 3.9785258270458506e-05, "loss": 0.3488, "step": 88000 }, { "epoch": 2.05, "learning_rate": 3.972721996517702e-05, "loss": 0.3482, "step": 88500 }, { "epoch": 2.07, "learning_rate": 3.966918165989553e-05, "loss": 0.3481, "step": 89000 }, { "epoch": 2.08, "learning_rate": 3.961114335461405e-05, "loss": 0.3482, "step": 89500 }, { "epoch": 2.09, "learning_rate": 3.955310504933256e-05, "loss": 0.3482, "step": 90000 }, { "epoch": 2.1, "learning_rate": 3.9495066744051076e-05, "loss": 0.3479, "step": 90500 }, { "epoch": 2.11, "learning_rate": 3.943702843876959e-05, "loss": 0.348, "step": 91000 }, { "epoch": 2.12, "learning_rate": 3.9378990133488106e-05, "loss": 0.3478, "step": 91500 }, { "epoch": 2.14, "learning_rate": 3.932095182820662e-05, "loss": 0.3477, "step": 92000 }, { "epoch": 2.15, "learning_rate": 3.926291352292513e-05, "loss": 0.3476, "step": 92500 }, { "epoch": 2.16, "learning_rate": 3.9204875217643645e-05, "loss": 0.3477, "step": 93000 }, { "epoch": 2.17, "learning_rate": 3.914683691236216e-05, "loss": 0.3478, "step": 93500 }, { "epoch": 2.18, "learning_rate": 3.9088798607080675e-05, "loss": 0.3473, "step": 94000 }, { "epoch": 2.19, "learning_rate": 3.903076030179919e-05, "loss": 0.3476, "step": 94500 }, { "epoch": 2.21, "learning_rate": 3.8972721996517706e-05, "loss": 0.3477, "step": 95000 }, { "epoch": 2.22, "learning_rate": 3.891468369123622e-05, "loss": 0.3473, "step": 95500 }, { "epoch": 2.23, "learning_rate": 3.885664538595473e-05, "loss": 0.3475, "step": 96000 }, { "epoch": 2.24, "learning_rate": 3.8798607080673245e-05, "loss": 0.3471, "step": 96500 }, { "epoch": 2.25, "learning_rate": 3.874056877539176e-05, "loss": 0.3471, "step": 97000 }, { "epoch": 2.26, "learning_rate": 3.8682530470110275e-05, "loss": 0.3475, "step": 97500 }, { "epoch": 2.28, "learning_rate": 3.862449216482879e-05, "loss": 0.3471, "step": 98000 }, { "epoch": 2.29, "learning_rate": 3.8566453859547306e-05, "loss": 0.347, "step": 98500 }, { "epoch": 2.3, "learning_rate": 3.850841555426582e-05, "loss": 0.3471, "step": 99000 }, { "epoch": 2.31, "learning_rate": 3.845037724898433e-05, "loss": 0.3469, "step": 99500 }, { "epoch": 2.32, "learning_rate": 3.8392338943702845e-05, "loss": 0.3469, "step": 100000 }, { "epoch": 2.33, "learning_rate": 3.833430063842136e-05, "loss": 0.3467, "step": 100500 }, { "epoch": 2.34, "learning_rate": 3.8276262333139875e-05, "loss": 0.3463, "step": 101000 }, { "epoch": 2.36, "learning_rate": 3.821822402785839e-05, "loss": 0.3463, "step": 101500 }, { "epoch": 2.37, "learning_rate": 3.8160185722576905e-05, "loss": 0.3465, "step": 102000 }, { "epoch": 2.38, "learning_rate": 3.8102147417295414e-05, "loss": 0.3462, "step": 102500 }, { "epoch": 2.39, "learning_rate": 3.804410911201393e-05, "loss": 0.3462, "step": 103000 }, { "epoch": 2.4, "learning_rate": 3.7986070806732444e-05, "loss": 0.3462, "step": 103500 }, { "epoch": 2.41, "learning_rate": 3.792803250145096e-05, "loss": 0.3463, "step": 104000 }, { "epoch": 2.43, "learning_rate": 3.7869994196169475e-05, "loss": 0.3464, "step": 104500 }, { "epoch": 2.44, "learning_rate": 3.781195589088799e-05, "loss": 0.3464, "step": 105000 }, { "epoch": 2.45, "learning_rate": 3.7753917585606505e-05, "loss": 0.346, "step": 105500 }, { "epoch": 2.46, "learning_rate": 3.7695879280325014e-05, "loss": 0.3459, "step": 106000 }, { "epoch": 2.47, "learning_rate": 3.763784097504353e-05, "loss": 0.3456, "step": 106500 }, { "epoch": 2.48, "learning_rate": 3.7579802669762044e-05, "loss": 0.346, "step": 107000 }, { "epoch": 2.5, "learning_rate": 3.752176436448056e-05, "loss": 0.3462, "step": 107500 }, { "epoch": 2.51, "learning_rate": 3.7463726059199075e-05, "loss": 0.346, "step": 108000 }, { "epoch": 2.52, "learning_rate": 3.740568775391759e-05, "loss": 0.3456, "step": 108500 }, { "epoch": 2.53, "learning_rate": 3.73476494486361e-05, "loss": 0.3456, "step": 109000 }, { "epoch": 2.54, "learning_rate": 3.7289611143354614e-05, "loss": 0.3457, "step": 109500 }, { "epoch": 2.55, "learning_rate": 3.723157283807313e-05, "loss": 0.3454, "step": 110000 }, { "epoch": 2.57, "learning_rate": 3.7173534532791644e-05, "loss": 0.3453, "step": 110500 }, { "epoch": 2.58, "learning_rate": 3.711549622751016e-05, "loss": 0.3452, "step": 111000 }, { "epoch": 2.59, "learning_rate": 3.7057457922228674e-05, "loss": 0.3456, "step": 111500 }, { "epoch": 2.6, "learning_rate": 3.699941961694719e-05, "loss": 0.3455, "step": 112000 }, { "epoch": 2.61, "learning_rate": 3.69413813116657e-05, "loss": 0.3455, "step": 112500 }, { "epoch": 2.62, "learning_rate": 3.688334300638421e-05, "loss": 0.3454, "step": 113000 }, { "epoch": 2.63, "learning_rate": 3.682530470110273e-05, "loss": 0.3453, "step": 113500 }, { "epoch": 2.65, "learning_rate": 3.6767266395821244e-05, "loss": 0.3453, "step": 114000 }, { "epoch": 2.66, "learning_rate": 3.670922809053976e-05, "loss": 0.3448, "step": 114500 }, { "epoch": 2.67, "learning_rate": 3.6651189785258274e-05, "loss": 0.3451, "step": 115000 }, { "epoch": 2.68, "learning_rate": 3.659315147997678e-05, "loss": 0.3452, "step": 115500 }, { "epoch": 2.69, "learning_rate": 3.65351131746953e-05, "loss": 0.3448, "step": 116000 }, { "epoch": 2.7, "learning_rate": 3.647707486941381e-05, "loss": 0.3445, "step": 116500 }, { "epoch": 2.72, "learning_rate": 3.641903656413233e-05, "loss": 0.3443, "step": 117000 }, { "epoch": 2.73, "learning_rate": 3.6360998258850844e-05, "loss": 0.3445, "step": 117500 }, { "epoch": 2.74, "learning_rate": 3.630295995356936e-05, "loss": 0.3445, "step": 118000 }, { "epoch": 2.75, "learning_rate": 3.6244921648287874e-05, "loss": 0.345, "step": 118500 }, { "epoch": 2.76, "learning_rate": 3.618688334300638e-05, "loss": 0.3443, "step": 119000 }, { "epoch": 2.77, "learning_rate": 3.61288450377249e-05, "loss": 0.3445, "step": 119500 }, { "epoch": 2.79, "learning_rate": 3.607080673244341e-05, "loss": 0.3446, "step": 120000 }, { "epoch": 2.8, "learning_rate": 3.601276842716193e-05, "loss": 0.3443, "step": 120500 }, { "epoch": 2.81, "learning_rate": 3.5954730121880443e-05, "loss": 0.3442, "step": 121000 }, { "epoch": 2.82, "learning_rate": 3.589669181659896e-05, "loss": 0.3443, "step": 121500 }, { "epoch": 2.83, "learning_rate": 3.5838653511317474e-05, "loss": 0.3444, "step": 122000 }, { "epoch": 2.84, "learning_rate": 3.578061520603598e-05, "loss": 0.3438, "step": 122500 }, { "epoch": 2.86, "learning_rate": 3.57225769007545e-05, "loss": 0.3444, "step": 123000 }, { "epoch": 2.87, "learning_rate": 3.566453859547301e-05, "loss": 0.3442, "step": 123500 }, { "epoch": 2.88, "learning_rate": 3.560650029019153e-05, "loss": 0.3436, "step": 124000 }, { "epoch": 2.89, "learning_rate": 3.554846198491004e-05, "loss": 0.3441, "step": 124500 }, { "epoch": 2.9, "learning_rate": 3.549042367962855e-05, "loss": 0.3438, "step": 125000 }, { "epoch": 2.91, "learning_rate": 3.5432385374347074e-05, "loss": 0.3438, "step": 125500 }, { "epoch": 2.93, "learning_rate": 3.537434706906558e-05, "loss": 0.3438, "step": 126000 }, { "epoch": 2.94, "learning_rate": 3.53163087637841e-05, "loss": 0.3437, "step": 126500 }, { "epoch": 2.95, "learning_rate": 3.525827045850261e-05, "loss": 0.3438, "step": 127000 }, { "epoch": 2.96, "learning_rate": 3.520023215322113e-05, "loss": 0.3438, "step": 127500 }, { "epoch": 2.97, "learning_rate": 3.514219384793964e-05, "loss": 0.3434, "step": 128000 }, { "epoch": 2.98, "learning_rate": 3.508415554265815e-05, "loss": 0.3434, "step": 128500 }, { "epoch": 2.99, "learning_rate": 3.5026117237376674e-05, "loss": 0.3435, "step": 129000 }, { "epoch": 3.0, "eval_accuracy": 0.8924505609685972, "eval_loss": 0.3355477452278137, "eval_runtime": 1982.819, "eval_samples_per_second": 192.889, "eval_steps_per_second": 2.411, "step": 129225 }, { "epoch": 3.01, "learning_rate": 3.496807893209518e-05, "loss": 0.3429, "step": 129500 }, { "epoch": 3.02, "learning_rate": 3.49100406268137e-05, "loss": 0.3425, "step": 130000 }, { "epoch": 3.03, "learning_rate": 3.485200232153221e-05, "loss": 0.3425, "step": 130500 }, { "epoch": 3.04, "learning_rate": 3.479396401625073e-05, "loss": 0.3424, "step": 131000 }, { "epoch": 3.05, "learning_rate": 3.473592571096924e-05, "loss": 0.3426, "step": 131500 }, { "epoch": 3.06, "learning_rate": 3.467788740568775e-05, "loss": 0.3425, "step": 132000 }, { "epoch": 3.08, "learning_rate": 3.461984910040627e-05, "loss": 0.3421, "step": 132500 }, { "epoch": 3.09, "learning_rate": 3.456181079512478e-05, "loss": 0.3422, "step": 133000 }, { "epoch": 3.1, "learning_rate": 3.45037724898433e-05, "loss": 0.3426, "step": 133500 }, { "epoch": 3.11, "learning_rate": 3.444573418456181e-05, "loss": 0.3423, "step": 134000 }, { "epoch": 3.12, "learning_rate": 3.438769587928033e-05, "loss": 0.3424, "step": 134500 }, { "epoch": 3.13, "learning_rate": 3.432965757399884e-05, "loss": 0.3423, "step": 135000 }, { "epoch": 3.15, "learning_rate": 3.427161926871735e-05, "loss": 0.3422, "step": 135500 }, { "epoch": 3.16, "learning_rate": 3.421358096343587e-05, "loss": 0.342, "step": 136000 }, { "epoch": 3.17, "learning_rate": 3.415554265815438e-05, "loss": 0.342, "step": 136500 }, { "epoch": 3.18, "learning_rate": 3.40975043528729e-05, "loss": 0.342, "step": 137000 }, { "epoch": 3.19, "learning_rate": 3.403946604759141e-05, "loss": 0.342, "step": 137500 }, { "epoch": 3.2, "learning_rate": 3.398142774230992e-05, "loss": 0.3416, "step": 138000 }, { "epoch": 3.22, "learning_rate": 3.392338943702844e-05, "loss": 0.342, "step": 138500 }, { "epoch": 3.23, "learning_rate": 3.386535113174695e-05, "loss": 0.342, "step": 139000 }, { "epoch": 3.24, "learning_rate": 3.380731282646547e-05, "loss": 0.3419, "step": 139500 }, { "epoch": 3.25, "learning_rate": 3.374927452118398e-05, "loss": 0.3421, "step": 140000 }, { "epoch": 3.26, "learning_rate": 3.36912362159025e-05, "loss": 0.3415, "step": 140500 }, { "epoch": 3.27, "learning_rate": 3.363319791062101e-05, "loss": 0.3417, "step": 141000 }, { "epoch": 3.28, "learning_rate": 3.357515960533952e-05, "loss": 0.3417, "step": 141500 }, { "epoch": 3.3, "learning_rate": 3.351712130005804e-05, "loss": 0.3413, "step": 142000 }, { "epoch": 3.31, "learning_rate": 3.345908299477655e-05, "loss": 0.3414, "step": 142500 }, { "epoch": 3.32, "learning_rate": 3.340104468949507e-05, "loss": 0.3415, "step": 143000 }, { "epoch": 3.33, "learning_rate": 3.334300638421358e-05, "loss": 0.3411, "step": 143500 }, { "epoch": 3.34, "learning_rate": 3.3284968078932096e-05, "loss": 0.3414, "step": 144000 }, { "epoch": 3.35, "learning_rate": 3.322692977365061e-05, "loss": 0.3415, "step": 144500 }, { "epoch": 3.37, "learning_rate": 3.316889146836912e-05, "loss": 0.3412, "step": 145000 }, { "epoch": 3.38, "learning_rate": 3.311085316308764e-05, "loss": 0.3415, "step": 145500 }, { "epoch": 3.39, "learning_rate": 3.305281485780615e-05, "loss": 0.3415, "step": 146000 }, { "epoch": 3.4, "learning_rate": 3.299477655252467e-05, "loss": 0.3417, "step": 146500 }, { "epoch": 3.41, "learning_rate": 3.293673824724318e-05, "loss": 0.3412, "step": 147000 }, { "epoch": 3.42, "learning_rate": 3.2878699941961696e-05, "loss": 0.3413, "step": 147500 }, { "epoch": 3.44, "learning_rate": 3.282066163668021e-05, "loss": 0.3413, "step": 148000 }, { "epoch": 3.45, "learning_rate": 3.276262333139872e-05, "loss": 0.3414, "step": 148500 }, { "epoch": 3.46, "learning_rate": 3.270458502611724e-05, "loss": 0.341, "step": 149000 }, { "epoch": 3.47, "learning_rate": 3.264654672083575e-05, "loss": 0.3408, "step": 149500 }, { "epoch": 3.48, "learning_rate": 3.258850841555427e-05, "loss": 0.3412, "step": 150000 }, { "epoch": 3.49, "learning_rate": 3.253047011027278e-05, "loss": 0.3409, "step": 150500 }, { "epoch": 3.51, "learning_rate": 3.2472431804991296e-05, "loss": 0.3412, "step": 151000 }, { "epoch": 3.52, "learning_rate": 3.241439349970981e-05, "loss": 0.3408, "step": 151500 }, { "epoch": 3.53, "learning_rate": 3.235635519442832e-05, "loss": 0.341, "step": 152000 }, { "epoch": 3.54, "learning_rate": 3.229831688914684e-05, "loss": 0.3407, "step": 152500 }, { "epoch": 3.55, "learning_rate": 3.224027858386535e-05, "loss": 0.3408, "step": 153000 }, { "epoch": 3.56, "learning_rate": 3.218224027858387e-05, "loss": 0.3407, "step": 153500 }, { "epoch": 3.58, "learning_rate": 3.212420197330238e-05, "loss": 0.3404, "step": 154000 }, { "epoch": 3.59, "learning_rate": 3.2066163668020896e-05, "loss": 0.3407, "step": 154500 }, { "epoch": 3.6, "learning_rate": 3.200812536273941e-05, "loss": 0.3404, "step": 155000 }, { "epoch": 3.61, "learning_rate": 3.195008705745792e-05, "loss": 0.3403, "step": 155500 }, { "epoch": 3.62, "learning_rate": 3.189204875217644e-05, "loss": 0.3404, "step": 156000 }, { "epoch": 3.63, "learning_rate": 3.183401044689495e-05, "loss": 0.3407, "step": 156500 }, { "epoch": 3.64, "learning_rate": 3.1775972141613465e-05, "loss": 0.3403, "step": 157000 }, { "epoch": 3.66, "learning_rate": 3.171793383633198e-05, "loss": 0.3403, "step": 157500 }, { "epoch": 3.67, "learning_rate": 3.1659895531050496e-05, "loss": 0.3401, "step": 158000 }, { "epoch": 3.68, "learning_rate": 3.160185722576901e-05, "loss": 0.3399, "step": 158500 }, { "epoch": 3.69, "learning_rate": 3.154381892048752e-05, "loss": 0.3403, "step": 159000 }, { "epoch": 3.7, "learning_rate": 3.148578061520604e-05, "loss": 0.3403, "step": 159500 }, { "epoch": 3.71, "learning_rate": 3.142774230992455e-05, "loss": 0.3404, "step": 160000 }, { "epoch": 3.73, "learning_rate": 3.1369704004643065e-05, "loss": 0.3401, "step": 160500 }, { "epoch": 3.74, "learning_rate": 3.131166569936158e-05, "loss": 0.3401, "step": 161000 }, { "epoch": 3.75, "learning_rate": 3.1253627394080096e-05, "loss": 0.3401, "step": 161500 }, { "epoch": 3.76, "learning_rate": 3.119558908879861e-05, "loss": 0.3401, "step": 162000 }, { "epoch": 3.77, "learning_rate": 3.113755078351712e-05, "loss": 0.34, "step": 162500 }, { "epoch": 3.78, "learning_rate": 3.107951247823564e-05, "loss": 0.3397, "step": 163000 }, { "epoch": 3.8, "learning_rate": 3.102147417295415e-05, "loss": 0.3399, "step": 163500 }, { "epoch": 3.81, "learning_rate": 3.0963435867672665e-05, "loss": 0.34, "step": 164000 }, { "epoch": 3.82, "learning_rate": 3.090539756239118e-05, "loss": 0.3401, "step": 164500 }, { "epoch": 3.83, "learning_rate": 3.0847359257109695e-05, "loss": 0.3397, "step": 165000 }, { "epoch": 3.84, "learning_rate": 3.078932095182821e-05, "loss": 0.3398, "step": 165500 }, { "epoch": 3.85, "learning_rate": 3.073128264654672e-05, "loss": 0.3399, "step": 166000 }, { "epoch": 3.87, "learning_rate": 3.067324434126524e-05, "loss": 0.3401, "step": 166500 }, { "epoch": 3.88, "learning_rate": 3.061520603598375e-05, "loss": 0.3399, "step": 167000 }, { "epoch": 3.89, "learning_rate": 3.0557167730702265e-05, "loss": 0.3399, "step": 167500 }, { "epoch": 3.9, "learning_rate": 3.049912942542078e-05, "loss": 0.3397, "step": 168000 }, { "epoch": 3.91, "learning_rate": 3.0441091120139292e-05, "loss": 0.3396, "step": 168500 }, { "epoch": 3.92, "learning_rate": 3.038305281485781e-05, "loss": 0.3397, "step": 169000 }, { "epoch": 3.93, "learning_rate": 3.0325014509576322e-05, "loss": 0.3395, "step": 169500 }, { "epoch": 3.95, "learning_rate": 3.0266976204294834e-05, "loss": 0.339, "step": 170000 }, { "epoch": 3.96, "learning_rate": 3.020893789901335e-05, "loss": 0.3393, "step": 170500 }, { "epoch": 3.97, "learning_rate": 3.015089959373186e-05, "loss": 0.3397, "step": 171000 }, { "epoch": 3.98, "learning_rate": 3.009286128845038e-05, "loss": 0.3391, "step": 171500 }, { "epoch": 3.99, "learning_rate": 3.003482298316889e-05, "loss": 0.3394, "step": 172000 }, { "epoch": 4.0, "eval_accuracy": 0.8933373896351878, "eval_loss": 0.3321678936481476, "eval_runtime": 1989.6768, "eval_samples_per_second": 192.224, "eval_steps_per_second": 2.403, "step": 172300 }, { "epoch": 4.0, "learning_rate": 2.997678467788741e-05, "loss": 0.3389, "step": 172500 }, { "epoch": 4.02, "learning_rate": 2.9918746372605922e-05, "loss": 0.3383, "step": 173000 }, { "epoch": 4.03, "learning_rate": 2.9860708067324434e-05, "loss": 0.3384, "step": 173500 }, { "epoch": 4.04, "learning_rate": 2.980266976204295e-05, "loss": 0.3387, "step": 174000 }, { "epoch": 4.05, "learning_rate": 2.974463145676146e-05, "loss": 0.3383, "step": 174500 }, { "epoch": 4.06, "learning_rate": 2.968659315147998e-05, "loss": 0.3385, "step": 175000 }, { "epoch": 4.07, "learning_rate": 2.962855484619849e-05, "loss": 0.3382, "step": 175500 }, { "epoch": 4.09, "learning_rate": 2.957051654091701e-05, "loss": 0.3383, "step": 176000 }, { "epoch": 4.1, "learning_rate": 2.9512478235635522e-05, "loss": 0.3387, "step": 176500 }, { "epoch": 4.11, "learning_rate": 2.9454439930354034e-05, "loss": 0.3382, "step": 177000 }, { "epoch": 4.12, "learning_rate": 2.939640162507255e-05, "loss": 0.3386, "step": 177500 }, { "epoch": 4.13, "learning_rate": 2.933836331979106e-05, "loss": 0.3381, "step": 178000 }, { "epoch": 4.14, "learning_rate": 2.928032501450958e-05, "loss": 0.338, "step": 178500 }, { "epoch": 4.16, "learning_rate": 2.922228670922809e-05, "loss": 0.3383, "step": 179000 }, { "epoch": 4.17, "learning_rate": 2.9164248403946603e-05, "loss": 0.3383, "step": 179500 }, { "epoch": 4.18, "learning_rate": 2.9106210098665122e-05, "loss": 0.3385, "step": 180000 }, { "epoch": 4.19, "learning_rate": 2.9048171793383634e-05, "loss": 0.3386, "step": 180500 }, { "epoch": 4.2, "learning_rate": 2.899013348810215e-05, "loss": 0.3385, "step": 181000 }, { "epoch": 4.21, "learning_rate": 2.893209518282066e-05, "loss": 0.3383, "step": 181500 }, { "epoch": 4.23, "learning_rate": 2.887405687753918e-05, "loss": 0.3384, "step": 182000 }, { "epoch": 4.24, "learning_rate": 2.881601857225769e-05, "loss": 0.338, "step": 182500 }, { "epoch": 4.25, "learning_rate": 2.8757980266976203e-05, "loss": 0.3382, "step": 183000 }, { "epoch": 4.26, "learning_rate": 2.869994196169472e-05, "loss": 0.3382, "step": 183500 }, { "epoch": 4.27, "learning_rate": 2.8641903656413233e-05, "loss": 0.3382, "step": 184000 }, { "epoch": 4.28, "learning_rate": 2.858386535113175e-05, "loss": 0.3376, "step": 184500 }, { "epoch": 4.29, "learning_rate": 2.852582704585026e-05, "loss": 0.338, "step": 185000 }, { "epoch": 4.31, "learning_rate": 2.846778874056878e-05, "loss": 0.3379, "step": 185500 }, { "epoch": 4.32, "learning_rate": 2.840975043528729e-05, "loss": 0.3377, "step": 186000 }, { "epoch": 4.33, "learning_rate": 2.8351712130005803e-05, "loss": 0.3382, "step": 186500 }, { "epoch": 4.34, "learning_rate": 2.829367382472432e-05, "loss": 0.3375, "step": 187000 }, { "epoch": 4.35, "learning_rate": 2.8235635519442833e-05, "loss": 0.338, "step": 187500 }, { "epoch": 4.36, "learning_rate": 2.817759721416135e-05, "loss": 0.3377, "step": 188000 }, { "epoch": 4.38, "learning_rate": 2.811955890887986e-05, "loss": 0.3375, "step": 188500 }, { "epoch": 4.39, "learning_rate": 2.8061520603598372e-05, "loss": 0.3373, "step": 189000 }, { "epoch": 4.4, "learning_rate": 2.800348229831689e-05, "loss": 0.3378, "step": 189500 }, { "epoch": 4.41, "learning_rate": 2.7945443993035403e-05, "loss": 0.3375, "step": 190000 }, { "epoch": 4.42, "learning_rate": 2.788740568775392e-05, "loss": 0.3375, "step": 190500 }, { "epoch": 4.43, "learning_rate": 2.7829367382472433e-05, "loss": 0.3379, "step": 191000 }, { "epoch": 4.45, "learning_rate": 2.7771329077190948e-05, "loss": 0.3375, "step": 191500 }, { "epoch": 4.46, "learning_rate": 2.771329077190946e-05, "loss": 0.3377, "step": 192000 }, { "epoch": 4.47, "learning_rate": 2.7655252466627972e-05, "loss": 0.3377, "step": 192500 }, { "epoch": 4.48, "learning_rate": 2.759721416134649e-05, "loss": 0.3379, "step": 193000 }, { "epoch": 4.49, "learning_rate": 2.7539175856065002e-05, "loss": 0.3374, "step": 193500 }, { "epoch": 4.5, "learning_rate": 2.748113755078352e-05, "loss": 0.3375, "step": 194000 }, { "epoch": 4.52, "learning_rate": 2.7423099245502033e-05, "loss": 0.3372, "step": 194500 }, { "epoch": 4.53, "learning_rate": 2.7365060940220548e-05, "loss": 0.3378, "step": 195000 }, { "epoch": 4.54, "learning_rate": 2.730702263493906e-05, "loss": 0.3375, "step": 195500 }, { "epoch": 4.55, "learning_rate": 2.7248984329657572e-05, "loss": 0.3372, "step": 196000 }, { "epoch": 4.56, "learning_rate": 2.719094602437609e-05, "loss": 0.3374, "step": 196500 }, { "epoch": 4.57, "learning_rate": 2.7132907719094602e-05, "loss": 0.3371, "step": 197000 }, { "epoch": 4.59, "learning_rate": 2.707486941381312e-05, "loss": 0.3373, "step": 197500 }, { "epoch": 4.6, "learning_rate": 2.7016831108531633e-05, "loss": 0.3372, "step": 198000 }, { "epoch": 4.61, "learning_rate": 2.6958792803250148e-05, "loss": 0.3372, "step": 198500 }, { "epoch": 4.62, "learning_rate": 2.690075449796866e-05, "loss": 0.3375, "step": 199000 }, { "epoch": 4.63, "learning_rate": 2.684271619268717e-05, "loss": 0.3371, "step": 199500 }, { "epoch": 4.64, "learning_rate": 2.678467788740569e-05, "loss": 0.3372, "step": 200000 }, { "epoch": 4.65, "learning_rate": 2.6726639582124202e-05, "loss": 0.3369, "step": 200500 }, { "epoch": 4.67, "learning_rate": 2.666860127684272e-05, "loss": 0.337, "step": 201000 }, { "epoch": 4.68, "learning_rate": 2.6610562971561232e-05, "loss": 0.3372, "step": 201500 }, { "epoch": 4.69, "learning_rate": 2.6552524666279744e-05, "loss": 0.337, "step": 202000 }, { "epoch": 4.7, "learning_rate": 2.649448636099826e-05, "loss": 0.3372, "step": 202500 }, { "epoch": 4.71, "learning_rate": 2.643644805571677e-05, "loss": 0.3373, "step": 203000 }, { "epoch": 4.72, "learning_rate": 2.637840975043529e-05, "loss": 0.3368, "step": 203500 }, { "epoch": 4.74, "learning_rate": 2.6320371445153802e-05, "loss": 0.3368, "step": 204000 }, { "epoch": 4.75, "learning_rate": 2.626233313987232e-05, "loss": 0.3372, "step": 204500 }, { "epoch": 4.76, "learning_rate": 2.6204294834590832e-05, "loss": 0.3369, "step": 205000 }, { "epoch": 4.77, "learning_rate": 2.6146256529309344e-05, "loss": 0.3368, "step": 205500 }, { "epoch": 4.78, "learning_rate": 2.608821822402786e-05, "loss": 0.337, "step": 206000 }, { "epoch": 4.79, "learning_rate": 2.603017991874637e-05, "loss": 0.3371, "step": 206500 }, { "epoch": 4.81, "learning_rate": 2.597214161346489e-05, "loss": 0.3371, "step": 207000 }, { "epoch": 4.82, "learning_rate": 2.59141033081834e-05, "loss": 0.3367, "step": 207500 }, { "epoch": 4.83, "learning_rate": 2.585606500290192e-05, "loss": 0.3368, "step": 208000 }, { "epoch": 4.84, "learning_rate": 2.5798026697620432e-05, "loss": 0.337, "step": 208500 }, { "epoch": 4.85, "learning_rate": 2.5739988392338944e-05, "loss": 0.3366, "step": 209000 }, { "epoch": 4.86, "learning_rate": 2.568195008705746e-05, "loss": 0.3369, "step": 209500 }, { "epoch": 4.88, "learning_rate": 2.562391178177597e-05, "loss": 0.3367, "step": 210000 }, { "epoch": 4.89, "learning_rate": 2.556587347649449e-05, "loss": 0.3363, "step": 210500 }, { "epoch": 4.9, "learning_rate": 2.5507835171213e-05, "loss": 0.3368, "step": 211000 }, { "epoch": 4.91, "learning_rate": 2.5449796865931513e-05, "loss": 0.3367, "step": 211500 }, { "epoch": 4.92, "learning_rate": 2.5391758560650032e-05, "loss": 0.3365, "step": 212000 }, { "epoch": 4.93, "learning_rate": 2.5333720255368544e-05, "loss": 0.3366, "step": 212500 }, { "epoch": 4.94, "learning_rate": 2.527568195008706e-05, "loss": 0.3364, "step": 213000 }, { "epoch": 4.96, "learning_rate": 2.521764364480557e-05, "loss": 0.3365, "step": 213500 }, { "epoch": 4.97, "learning_rate": 2.515960533952409e-05, "loss": 0.3364, "step": 214000 }, { "epoch": 4.98, "learning_rate": 2.51015670342426e-05, "loss": 0.3363, "step": 214500 }, { "epoch": 4.99, "learning_rate": 2.5043528728961113e-05, "loss": 0.3367, "step": 215000 }, { "epoch": 5.0, "eval_accuracy": 0.8939737566541442, "eval_loss": 0.32981738448143005, "eval_runtime": 1986.7948, "eval_samples_per_second": 192.503, "eval_steps_per_second": 2.406, "step": 215375 }, { "epoch": 5.0, "learning_rate": 2.4985490423679632e-05, "loss": 0.336, "step": 215500 }, { "epoch": 5.01, "learning_rate": 2.4927452118398144e-05, "loss": 0.3354, "step": 216000 }, { "epoch": 5.03, "learning_rate": 2.486941381311666e-05, "loss": 0.336, "step": 216500 }, { "epoch": 5.04, "learning_rate": 2.481137550783517e-05, "loss": 0.3356, "step": 217000 }, { "epoch": 5.05, "learning_rate": 2.4753337202553686e-05, "loss": 0.3358, "step": 217500 }, { "epoch": 5.06, "learning_rate": 2.46952988972722e-05, "loss": 0.3358, "step": 218000 }, { "epoch": 5.07, "learning_rate": 2.4637260591990716e-05, "loss": 0.3355, "step": 218500 }, { "epoch": 5.08, "learning_rate": 2.457922228670923e-05, "loss": 0.3356, "step": 219000 }, { "epoch": 5.1, "learning_rate": 2.4521183981427743e-05, "loss": 0.3355, "step": 219500 }, { "epoch": 5.11, "learning_rate": 2.4463145676146255e-05, "loss": 0.3359, "step": 220000 }, { "epoch": 5.12, "learning_rate": 2.440510737086477e-05, "loss": 0.3355, "step": 220500 }, { "epoch": 5.13, "learning_rate": 2.4347069065583286e-05, "loss": 0.3354, "step": 221000 }, { "epoch": 5.14, "learning_rate": 2.42890307603018e-05, "loss": 0.3357, "step": 221500 }, { "epoch": 5.15, "learning_rate": 2.4230992455020316e-05, "loss": 0.3353, "step": 222000 }, { "epoch": 5.17, "learning_rate": 2.417295414973883e-05, "loss": 0.3356, "step": 222500 }, { "epoch": 5.18, "learning_rate": 2.4114915844457343e-05, "loss": 0.3352, "step": 223000 }, { "epoch": 5.19, "learning_rate": 2.4056877539175855e-05, "loss": 0.335, "step": 223500 }, { "epoch": 5.2, "learning_rate": 2.399883923389437e-05, "loss": 0.3351, "step": 224000 }, { "epoch": 5.21, "learning_rate": 2.3940800928612885e-05, "loss": 0.3353, "step": 224500 }, { "epoch": 5.22, "learning_rate": 2.38827626233314e-05, "loss": 0.3353, "step": 225000 }, { "epoch": 5.24, "learning_rate": 2.3824724318049916e-05, "loss": 0.3355, "step": 225500 }, { "epoch": 5.25, "learning_rate": 2.3766686012768428e-05, "loss": 0.3357, "step": 226000 }, { "epoch": 5.26, "learning_rate": 2.3708647707486943e-05, "loss": 0.3351, "step": 226500 }, { "epoch": 5.27, "learning_rate": 2.3650609402205455e-05, "loss": 0.3355, "step": 227000 }, { "epoch": 5.28, "learning_rate": 2.359257109692397e-05, "loss": 0.335, "step": 227500 }, { "epoch": 5.29, "learning_rate": 2.3534532791642485e-05, "loss": 0.3356, "step": 228000 }, { "epoch": 5.3, "learning_rate": 2.3476494486361e-05, "loss": 0.3353, "step": 228500 }, { "epoch": 5.32, "learning_rate": 2.3418456181079512e-05, "loss": 0.3354, "step": 229000 }, { "epoch": 5.33, "learning_rate": 2.3360417875798028e-05, "loss": 0.3351, "step": 229500 }, { "epoch": 5.34, "learning_rate": 2.3302379570516543e-05, "loss": 0.3353, "step": 230000 }, { "epoch": 5.35, "learning_rate": 2.3244341265235055e-05, "loss": 0.3349, "step": 230500 }, { "epoch": 5.36, "learning_rate": 2.318630295995357e-05, "loss": 0.3351, "step": 231000 }, { "epoch": 5.37, "learning_rate": 2.3128264654672085e-05, "loss": 0.3357, "step": 231500 }, { "epoch": 5.39, "learning_rate": 2.30702263493906e-05, "loss": 0.3354, "step": 232000 }, { "epoch": 5.4, "learning_rate": 2.3012188044109112e-05, "loss": 0.3351, "step": 232500 }, { "epoch": 5.41, "learning_rate": 2.2954149738827627e-05, "loss": 0.3352, "step": 233000 }, { "epoch": 5.42, "learning_rate": 2.2896111433546143e-05, "loss": 0.3354, "step": 233500 }, { "epoch": 5.43, "learning_rate": 2.2838073128264654e-05, "loss": 0.335, "step": 234000 }, { "epoch": 5.44, "learning_rate": 2.278003482298317e-05, "loss": 0.3352, "step": 234500 }, { "epoch": 5.46, "learning_rate": 2.2721996517701685e-05, "loss": 0.3347, "step": 235000 }, { "epoch": 5.47, "learning_rate": 2.2663958212420197e-05, "loss": 0.3351, "step": 235500 }, { "epoch": 5.48, "learning_rate": 2.2605919907138712e-05, "loss": 0.3349, "step": 236000 }, { "epoch": 5.49, "learning_rate": 2.2547881601857227e-05, "loss": 0.3348, "step": 236500 }, { "epoch": 5.5, "learning_rate": 2.2489843296575742e-05, "loss": 0.3348, "step": 237000 }, { "epoch": 5.51, "learning_rate": 2.2431804991294254e-05, "loss": 0.3349, "step": 237500 }, { "epoch": 5.53, "learning_rate": 2.237376668601277e-05, "loss": 0.3345, "step": 238000 }, { "epoch": 5.54, "learning_rate": 2.2315728380731285e-05, "loss": 0.3345, "step": 238500 }, { "epoch": 5.55, "learning_rate": 2.2257690075449797e-05, "loss": 0.3348, "step": 239000 }, { "epoch": 5.56, "learning_rate": 2.2199651770168312e-05, "loss": 0.3344, "step": 239500 }, { "epoch": 5.57, "learning_rate": 2.2141613464886827e-05, "loss": 0.3349, "step": 240000 }, { "epoch": 5.58, "learning_rate": 2.2083575159605342e-05, "loss": 0.3348, "step": 240500 }, { "epoch": 5.59, "learning_rate": 2.2025536854323854e-05, "loss": 0.3347, "step": 241000 }, { "epoch": 5.61, "learning_rate": 2.196749854904237e-05, "loss": 0.3351, "step": 241500 }, { "epoch": 5.62, "learning_rate": 2.190946024376088e-05, "loss": 0.3347, "step": 242000 }, { "epoch": 5.63, "learning_rate": 2.1851421938479396e-05, "loss": 0.3346, "step": 242500 }, { "epoch": 5.64, "learning_rate": 2.179338363319791e-05, "loss": 0.3348, "step": 243000 }, { "epoch": 5.65, "learning_rate": 2.1735345327916427e-05, "loss": 0.335, "step": 243500 }, { "epoch": 5.66, "learning_rate": 2.1677307022634942e-05, "loss": 0.3346, "step": 244000 }, { "epoch": 5.68, "learning_rate": 2.1619268717353454e-05, "loss": 0.3346, "step": 244500 }, { "epoch": 5.69, "learning_rate": 2.1561230412071966e-05, "loss": 0.3346, "step": 245000 }, { "epoch": 5.7, "learning_rate": 2.150319210679048e-05, "loss": 0.3346, "step": 245500 }, { "epoch": 5.71, "learning_rate": 2.1445153801508996e-05, "loss": 0.3344, "step": 246000 }, { "epoch": 5.72, "learning_rate": 2.138711549622751e-05, "loss": 0.3346, "step": 246500 }, { "epoch": 5.73, "learning_rate": 2.1329077190946027e-05, "loss": 0.3345, "step": 247000 }, { "epoch": 5.75, "learning_rate": 2.1271038885664542e-05, "loss": 0.3351, "step": 247500 }, { "epoch": 5.76, "learning_rate": 2.1213000580383054e-05, "loss": 0.3348, "step": 248000 }, { "epoch": 5.77, "learning_rate": 2.1154962275101566e-05, "loss": 0.3343, "step": 248500 }, { "epoch": 5.78, "learning_rate": 2.109692396982008e-05, "loss": 0.3345, "step": 249000 }, { "epoch": 5.79, "learning_rate": 2.1038885664538596e-05, "loss": 0.3345, "step": 249500 }, { "epoch": 5.8, "learning_rate": 2.098084735925711e-05, "loss": 0.3344, "step": 250000 }, { "epoch": 5.82, "learning_rate": 2.0922809053975626e-05, "loss": 0.3344, "step": 250500 }, { "epoch": 5.83, "learning_rate": 2.086477074869414e-05, "loss": 0.3347, "step": 251000 }, { "epoch": 5.84, "learning_rate": 2.0806732443412654e-05, "loss": 0.3345, "step": 251500 }, { "epoch": 5.85, "learning_rate": 2.0748694138131165e-05, "loss": 0.3345, "step": 252000 }, { "epoch": 5.86, "learning_rate": 2.069065583284968e-05, "loss": 0.3343, "step": 252500 }, { "epoch": 5.87, "learning_rate": 2.0632617527568196e-05, "loss": 0.3342, "step": 253000 }, { "epoch": 5.89, "learning_rate": 2.057457922228671e-05, "loss": 0.3344, "step": 253500 }, { "epoch": 5.9, "learning_rate": 2.0516540917005226e-05, "loss": 0.3344, "step": 254000 }, { "epoch": 5.91, "learning_rate": 2.045850261172374e-05, "loss": 0.3347, "step": 254500 }, { "epoch": 5.92, "learning_rate": 2.0400464306442253e-05, "loss": 0.3341, "step": 255000 }, { "epoch": 5.93, "learning_rate": 2.0342426001160765e-05, "loss": 0.3344, "step": 255500 }, { "epoch": 5.94, "learning_rate": 2.028438769587928e-05, "loss": 0.3347, "step": 256000 }, { "epoch": 5.95, "learning_rate": 2.0226349390597796e-05, "loss": 0.3343, "step": 256500 }, { "epoch": 5.97, "learning_rate": 2.016831108531631e-05, "loss": 0.3343, "step": 257000 }, { "epoch": 5.98, "learning_rate": 2.0110272780034826e-05, "loss": 0.3342, "step": 257500 }, { "epoch": 5.99, "learning_rate": 2.0052234474753338e-05, "loss": 0.3343, "step": 258000 }, { "epoch": 6.0, "eval_accuracy": 0.8944719783406293, "eval_loss": 0.32793664932250977, "eval_runtime": 1984.9744, "eval_samples_per_second": 192.68, "eval_steps_per_second": 2.409, "step": 258450 }, { "epoch": 6.0, "learning_rate": 1.9994196169471853e-05, "loss": 0.3343, "step": 258500 }, { "epoch": 6.01, "learning_rate": 1.9936157864190365e-05, "loss": 0.3336, "step": 259000 }, { "epoch": 6.02, "learning_rate": 1.987811955890888e-05, "loss": 0.3334, "step": 259500 }, { "epoch": 6.04, "learning_rate": 1.9820081253627395e-05, "loss": 0.3335, "step": 260000 }, { "epoch": 6.05, "learning_rate": 1.976204294834591e-05, "loss": 0.3335, "step": 260500 }, { "epoch": 6.06, "learning_rate": 1.9704004643064426e-05, "loss": 0.3334, "step": 261000 }, { "epoch": 6.07, "learning_rate": 1.9645966337782938e-05, "loss": 0.3329, "step": 261500 }, { "epoch": 6.08, "learning_rate": 1.9587928032501453e-05, "loss": 0.3334, "step": 262000 }, { "epoch": 6.09, "learning_rate": 1.9529889727219965e-05, "loss": 0.3333, "step": 262500 }, { "epoch": 6.11, "learning_rate": 1.947185142193848e-05, "loss": 0.3334, "step": 263000 }, { "epoch": 6.12, "learning_rate": 1.9413813116656995e-05, "loss": 0.3334, "step": 263500 }, { "epoch": 6.13, "learning_rate": 1.935577481137551e-05, "loss": 0.3337, "step": 264000 }, { "epoch": 6.14, "learning_rate": 1.9297736506094022e-05, "loss": 0.3334, "step": 264500 }, { "epoch": 6.15, "learning_rate": 1.9239698200812538e-05, "loss": 0.3332, "step": 265000 }, { "epoch": 6.16, "learning_rate": 1.9181659895531053e-05, "loss": 0.3331, "step": 265500 }, { "epoch": 6.18, "learning_rate": 1.9123621590249565e-05, "loss": 0.3335, "step": 266000 }, { "epoch": 6.19, "learning_rate": 1.906558328496808e-05, "loss": 0.3333, "step": 266500 }, { "epoch": 6.2, "learning_rate": 1.9007544979686595e-05, "loss": 0.3335, "step": 267000 }, { "epoch": 6.21, "learning_rate": 1.8949506674405107e-05, "loss": 0.3335, "step": 267500 }, { "epoch": 6.22, "learning_rate": 1.8891468369123622e-05, "loss": 0.3333, "step": 268000 }, { "epoch": 6.23, "learning_rate": 1.8833430063842137e-05, "loss": 0.333, "step": 268500 }, { "epoch": 6.24, "learning_rate": 1.8775391758560653e-05, "loss": 0.3332, "step": 269000 }, { "epoch": 6.26, "learning_rate": 1.8717353453279164e-05, "loss": 0.3334, "step": 269500 }, { "epoch": 6.27, "learning_rate": 1.865931514799768e-05, "loss": 0.3334, "step": 270000 }, { "epoch": 6.28, "learning_rate": 1.8601276842716195e-05, "loss": 0.3334, "step": 270500 }, { "epoch": 6.29, "learning_rate": 1.8543238537434707e-05, "loss": 0.3334, "step": 271000 }, { "epoch": 6.3, "learning_rate": 1.8485200232153222e-05, "loss": 0.3331, "step": 271500 }, { "epoch": 6.31, "learning_rate": 1.8427161926871737e-05, "loss": 0.3332, "step": 272000 }, { "epoch": 6.33, "learning_rate": 1.836912362159025e-05, "loss": 0.3333, "step": 272500 }, { "epoch": 6.34, "learning_rate": 1.8311085316308764e-05, "loss": 0.3332, "step": 273000 }, { "epoch": 6.35, "learning_rate": 1.825304701102728e-05, "loss": 0.3333, "step": 273500 }, { "epoch": 6.36, "learning_rate": 1.819500870574579e-05, "loss": 0.3332, "step": 274000 }, { "epoch": 6.37, "learning_rate": 1.8136970400464307e-05, "loss": 0.3332, "step": 274500 }, { "epoch": 6.38, "learning_rate": 1.8078932095182822e-05, "loss": 0.3333, "step": 275000 }, { "epoch": 6.4, "learning_rate": 1.8020893789901337e-05, "loss": 0.3333, "step": 275500 }, { "epoch": 6.41, "learning_rate": 1.796285548461985e-05, "loss": 0.3331, "step": 276000 }, { "epoch": 6.42, "learning_rate": 1.7904817179338364e-05, "loss": 0.3333, "step": 276500 }, { "epoch": 6.43, "learning_rate": 1.784677887405688e-05, "loss": 0.3331, "step": 277000 }, { "epoch": 6.44, "learning_rate": 1.778874056877539e-05, "loss": 0.3331, "step": 277500 }, { "epoch": 6.45, "learning_rate": 1.7730702263493906e-05, "loss": 0.3331, "step": 278000 }, { "epoch": 6.47, "learning_rate": 1.767266395821242e-05, "loss": 0.3331, "step": 278500 }, { "epoch": 6.48, "learning_rate": 1.7614625652930937e-05, "loss": 0.3331, "step": 279000 }, { "epoch": 6.49, "learning_rate": 1.755658734764945e-05, "loss": 0.3332, "step": 279500 }, { "epoch": 6.5, "learning_rate": 1.7498549042367964e-05, "loss": 0.333, "step": 280000 }, { "epoch": 6.51, "learning_rate": 1.7440510737086476e-05, "loss": 0.3329, "step": 280500 }, { "epoch": 6.52, "learning_rate": 1.738247243180499e-05, "loss": 0.3326, "step": 281000 }, { "epoch": 6.54, "learning_rate": 1.7324434126523506e-05, "loss": 0.333, "step": 281500 }, { "epoch": 6.55, "learning_rate": 1.726639582124202e-05, "loss": 0.3329, "step": 282000 }, { "epoch": 6.56, "learning_rate": 1.7208357515960537e-05, "loss": 0.3329, "step": 282500 }, { "epoch": 6.57, "learning_rate": 1.715031921067905e-05, "loss": 0.333, "step": 283000 }, { "epoch": 6.58, "learning_rate": 1.709228090539756e-05, "loss": 0.3333, "step": 283500 }, { "epoch": 6.59, "learning_rate": 1.7034242600116076e-05, "loss": 0.3326, "step": 284000 }, { "epoch": 6.6, "learning_rate": 1.697620429483459e-05, "loss": 0.3329, "step": 284500 }, { "epoch": 6.62, "learning_rate": 1.6918165989553106e-05, "loss": 0.3326, "step": 285000 }, { "epoch": 6.63, "learning_rate": 1.686012768427162e-05, "loss": 0.3326, "step": 285500 }, { "epoch": 6.64, "learning_rate": 1.6802089378990136e-05, "loss": 0.333, "step": 286000 }, { "epoch": 6.65, "learning_rate": 1.6744051073708648e-05, "loss": 0.3326, "step": 286500 }, { "epoch": 6.66, "learning_rate": 1.668601276842716e-05, "loss": 0.3329, "step": 287000 }, { "epoch": 6.67, "learning_rate": 1.6627974463145675e-05, "loss": 0.3327, "step": 287500 }, { "epoch": 6.69, "learning_rate": 1.656993615786419e-05, "loss": 0.3326, "step": 288000 }, { "epoch": 6.7, "learning_rate": 1.6511897852582706e-05, "loss": 0.3329, "step": 288500 }, { "epoch": 6.71, "learning_rate": 1.645385954730122e-05, "loss": 0.333, "step": 289000 }, { "epoch": 6.72, "learning_rate": 1.6395821242019736e-05, "loss": 0.3327, "step": 289500 }, { "epoch": 6.73, "learning_rate": 1.6337782936738248e-05, "loss": 0.3328, "step": 290000 }, { "epoch": 6.74, "learning_rate": 1.627974463145676e-05, "loss": 0.3325, "step": 290500 }, { "epoch": 6.76, "learning_rate": 1.6221706326175275e-05, "loss": 0.3326, "step": 291000 }, { "epoch": 6.77, "learning_rate": 1.616366802089379e-05, "loss": 0.3327, "step": 291500 }, { "epoch": 6.78, "learning_rate": 1.6105629715612306e-05, "loss": 0.3327, "step": 292000 }, { "epoch": 6.79, "learning_rate": 1.604759141033082e-05, "loss": 0.3329, "step": 292500 }, { "epoch": 6.8, "learning_rate": 1.5989553105049336e-05, "loss": 0.3326, "step": 293000 }, { "epoch": 6.81, "learning_rate": 1.5931514799767848e-05, "loss": 0.3326, "step": 293500 }, { "epoch": 6.83, "learning_rate": 1.587347649448636e-05, "loss": 0.3331, "step": 294000 }, { "epoch": 6.84, "learning_rate": 1.5815438189204875e-05, "loss": 0.3328, "step": 294500 }, { "epoch": 6.85, "learning_rate": 1.575739988392339e-05, "loss": 0.3327, "step": 295000 }, { "epoch": 6.86, "learning_rate": 1.5699361578641905e-05, "loss": 0.3326, "step": 295500 }, { "epoch": 6.87, "learning_rate": 1.564132327336042e-05, "loss": 0.3326, "step": 296000 }, { "epoch": 6.88, "learning_rate": 1.5583284968078933e-05, "loss": 0.3324, "step": 296500 }, { "epoch": 6.89, "learning_rate": 1.5525246662797448e-05, "loss": 0.3327, "step": 297000 }, { "epoch": 6.91, "learning_rate": 1.546720835751596e-05, "loss": 0.3326, "step": 297500 }, { "epoch": 6.92, "learning_rate": 1.5409170052234475e-05, "loss": 0.3322, "step": 298000 }, { "epoch": 6.93, "learning_rate": 1.535113174695299e-05, "loss": 0.3325, "step": 298500 }, { "epoch": 6.94, "learning_rate": 1.5293093441671505e-05, "loss": 0.3326, "step": 299000 }, { "epoch": 6.95, "learning_rate": 1.5235055136390017e-05, "loss": 0.3326, "step": 299500 }, { "epoch": 6.96, "learning_rate": 1.517701683110853e-05, "loss": 0.3325, "step": 300000 }, { "epoch": 6.98, "learning_rate": 1.5118978525827046e-05, "loss": 0.3325, "step": 300500 }, { "epoch": 6.99, "learning_rate": 1.5060940220545561e-05, "loss": 0.3323, "step": 301000 }, { "epoch": 7.0, "learning_rate": 1.5002901915264075e-05, "loss": 0.3326, "step": 301500 }, { "epoch": 7.0, "eval_accuracy": 0.8947962788449951, "eval_loss": 0.3267107605934143, "eval_runtime": 1985.0991, "eval_samples_per_second": 192.667, "eval_steps_per_second": 2.408, "step": 301525 }, { "epoch": 7.01, "learning_rate": 1.494486360998259e-05, "loss": 0.3319, "step": 302000 }, { "epoch": 7.02, "learning_rate": 1.4886825304701105e-05, "loss": 0.3318, "step": 302500 }, { "epoch": 7.03, "learning_rate": 1.4828786999419617e-05, "loss": 0.3321, "step": 303000 }, { "epoch": 7.05, "learning_rate": 1.477074869413813e-05, "loss": 0.3315, "step": 303500 }, { "epoch": 7.06, "learning_rate": 1.4712710388856646e-05, "loss": 0.3317, "step": 304000 }, { "epoch": 7.07, "learning_rate": 1.4654672083575161e-05, "loss": 0.3321, "step": 304500 }, { "epoch": 7.08, "learning_rate": 1.4596633778293674e-05, "loss": 0.3317, "step": 305000 }, { "epoch": 7.09, "learning_rate": 1.453859547301219e-05, "loss": 0.3318, "step": 305500 }, { "epoch": 7.1, "learning_rate": 1.4480557167730702e-05, "loss": 0.3319, "step": 306000 }, { "epoch": 7.12, "learning_rate": 1.4422518862449217e-05, "loss": 0.3319, "step": 306500 }, { "epoch": 7.13, "learning_rate": 1.436448055716773e-05, "loss": 0.3316, "step": 307000 }, { "epoch": 7.14, "learning_rate": 1.4306442251886245e-05, "loss": 0.3318, "step": 307500 }, { "epoch": 7.15, "learning_rate": 1.424840394660476e-05, "loss": 0.3318, "step": 308000 }, { "epoch": 7.16, "learning_rate": 1.4190365641323274e-05, "loss": 0.3316, "step": 308500 }, { "epoch": 7.17, "learning_rate": 1.413232733604179e-05, "loss": 0.3319, "step": 309000 }, { "epoch": 7.19, "learning_rate": 1.4074289030760301e-05, "loss": 0.3319, "step": 309500 }, { "epoch": 7.2, "learning_rate": 1.4016250725478817e-05, "loss": 0.3316, "step": 310000 }, { "epoch": 7.21, "learning_rate": 1.395821242019733e-05, "loss": 0.332, "step": 310500 }, { "epoch": 7.22, "learning_rate": 1.3900174114915845e-05, "loss": 0.3314, "step": 311000 }, { "epoch": 7.23, "learning_rate": 1.384213580963436e-05, "loss": 0.332, "step": 311500 }, { "epoch": 7.24, "learning_rate": 1.3784097504352874e-05, "loss": 0.332, "step": 312000 }, { "epoch": 7.25, "learning_rate": 1.3726059199071386e-05, "loss": 0.3318, "step": 312500 }, { "epoch": 7.27, "learning_rate": 1.3668020893789901e-05, "loss": 0.3318, "step": 313000 }, { "epoch": 7.28, "learning_rate": 1.3609982588508416e-05, "loss": 0.3317, "step": 313500 }, { "epoch": 7.29, "learning_rate": 1.355194428322693e-05, "loss": 0.3317, "step": 314000 }, { "epoch": 7.3, "learning_rate": 1.3493905977945445e-05, "loss": 0.3318, "step": 314500 }, { "epoch": 7.31, "learning_rate": 1.343586767266396e-05, "loss": 0.3312, "step": 315000 }, { "epoch": 7.32, "learning_rate": 1.3377829367382472e-05, "loss": 0.3318, "step": 315500 }, { "epoch": 7.34, "learning_rate": 1.3319791062100986e-05, "loss": 0.3316, "step": 316000 }, { "epoch": 7.35, "learning_rate": 1.3261752756819501e-05, "loss": 0.3319, "step": 316500 }, { "epoch": 7.36, "learning_rate": 1.3203714451538016e-05, "loss": 0.3317, "step": 317000 }, { "epoch": 7.37, "learning_rate": 1.314567614625653e-05, "loss": 0.3314, "step": 317500 }, { "epoch": 7.38, "learning_rate": 1.3087637840975045e-05, "loss": 0.3316, "step": 318000 }, { "epoch": 7.39, "learning_rate": 1.302959953569356e-05, "loss": 0.3318, "step": 318500 }, { "epoch": 7.41, "learning_rate": 1.2971561230412072e-05, "loss": 0.3318, "step": 319000 }, { "epoch": 7.42, "learning_rate": 1.2913522925130586e-05, "loss": 0.3317, "step": 319500 }, { "epoch": 7.43, "learning_rate": 1.28554846198491e-05, "loss": 0.3317, "step": 320000 }, { "epoch": 7.44, "learning_rate": 1.2797446314567616e-05, "loss": 0.3318, "step": 320500 }, { "epoch": 7.45, "learning_rate": 1.273940800928613e-05, "loss": 0.3317, "step": 321000 }, { "epoch": 7.46, "learning_rate": 1.2681369704004645e-05, "loss": 0.3314, "step": 321500 }, { "epoch": 7.48, "learning_rate": 1.2623331398723157e-05, "loss": 0.3313, "step": 322000 }, { "epoch": 7.49, "learning_rate": 1.2565293093441672e-05, "loss": 0.3318, "step": 322500 }, { "epoch": 7.5, "learning_rate": 1.2507254788160185e-05, "loss": 0.3315, "step": 323000 }, { "epoch": 7.51, "learning_rate": 1.24492164828787e-05, "loss": 0.3311, "step": 323500 }, { "epoch": 7.52, "learning_rate": 1.2391178177597216e-05, "loss": 0.3315, "step": 324000 }, { "epoch": 7.53, "learning_rate": 1.2333139872315728e-05, "loss": 0.3317, "step": 324500 }, { "epoch": 7.54, "learning_rate": 1.2275101567034243e-05, "loss": 0.3313, "step": 325000 }, { "epoch": 7.56, "learning_rate": 1.2217063261752758e-05, "loss": 0.3314, "step": 325500 }, { "epoch": 7.57, "learning_rate": 1.2159024956471272e-05, "loss": 0.3315, "step": 326000 }, { "epoch": 7.58, "learning_rate": 1.2100986651189785e-05, "loss": 0.3314, "step": 326500 }, { "epoch": 7.59, "learning_rate": 1.20429483459083e-05, "loss": 0.3314, "step": 327000 }, { "epoch": 7.6, "learning_rate": 1.1984910040626814e-05, "loss": 0.3314, "step": 327500 }, { "epoch": 7.61, "learning_rate": 1.1926871735345327e-05, "loss": 0.3314, "step": 328000 }, { "epoch": 7.63, "learning_rate": 1.1868833430063843e-05, "loss": 0.3316, "step": 328500 }, { "epoch": 7.64, "learning_rate": 1.1810795124782358e-05, "loss": 0.3315, "step": 329000 }, { "epoch": 7.65, "learning_rate": 1.1752756819500871e-05, "loss": 0.3311, "step": 329500 }, { "epoch": 7.66, "learning_rate": 1.1694718514219385e-05, "loss": 0.3314, "step": 330000 }, { "epoch": 7.67, "learning_rate": 1.16366802089379e-05, "loss": 0.3314, "step": 330500 }, { "epoch": 7.68, "learning_rate": 1.1578641903656414e-05, "loss": 0.3313, "step": 331000 }, { "epoch": 7.7, "learning_rate": 1.1520603598374927e-05, "loss": 0.331, "step": 331500 }, { "epoch": 7.71, "learning_rate": 1.1462565293093443e-05, "loss": 0.3315, "step": 332000 }, { "epoch": 7.72, "learning_rate": 1.1404526987811956e-05, "loss": 0.3314, "step": 332500 }, { "epoch": 7.73, "learning_rate": 1.1346488682530471e-05, "loss": 0.3315, "step": 333000 }, { "epoch": 7.74, "learning_rate": 1.1288450377248985e-05, "loss": 0.3314, "step": 333500 }, { "epoch": 7.75, "learning_rate": 1.1230412071967498e-05, "loss": 0.3312, "step": 334000 }, { "epoch": 7.77, "learning_rate": 1.1172373766686014e-05, "loss": 0.3315, "step": 334500 }, { "epoch": 7.78, "learning_rate": 1.1114335461404527e-05, "loss": 0.3313, "step": 335000 }, { "epoch": 7.79, "learning_rate": 1.105629715612304e-05, "loss": 0.3311, "step": 335500 }, { "epoch": 7.8, "learning_rate": 1.0998258850841556e-05, "loss": 0.3313, "step": 336000 }, { "epoch": 7.81, "learning_rate": 1.0940220545560071e-05, "loss": 0.3316, "step": 336500 }, { "epoch": 7.82, "learning_rate": 1.0882182240278585e-05, "loss": 0.3314, "step": 337000 }, { "epoch": 7.84, "learning_rate": 1.0824143934997098e-05, "loss": 0.3314, "step": 337500 }, { "epoch": 7.85, "learning_rate": 1.0766105629715613e-05, "loss": 0.3311, "step": 338000 }, { "epoch": 7.86, "learning_rate": 1.0708067324434127e-05, "loss": 0.3311, "step": 338500 }, { "epoch": 7.87, "learning_rate": 1.065002901915264e-05, "loss": 0.331, "step": 339000 }, { "epoch": 7.88, "learning_rate": 1.0591990713871156e-05, "loss": 0.3311, "step": 339500 }, { "epoch": 7.89, "learning_rate": 1.0533952408589671e-05, "loss": 0.3314, "step": 340000 }, { "epoch": 7.9, "learning_rate": 1.0475914103308183e-05, "loss": 0.3312, "step": 340500 }, { "epoch": 7.92, "learning_rate": 1.0417875798026698e-05, "loss": 0.3311, "step": 341000 }, { "epoch": 7.93, "learning_rate": 1.0359837492745213e-05, "loss": 0.3311, "step": 341500 }, { "epoch": 7.94, "learning_rate": 1.0301799187463727e-05, "loss": 0.3314, "step": 342000 }, { "epoch": 7.95, "learning_rate": 1.024376088218224e-05, "loss": 0.3312, "step": 342500 }, { "epoch": 7.96, "learning_rate": 1.0185722576900755e-05, "loss": 0.3313, "step": 343000 }, { "epoch": 7.97, "learning_rate": 1.0127684271619269e-05, "loss": 0.3309, "step": 343500 }, { "epoch": 7.99, "learning_rate": 1.0069645966337783e-05, "loss": 0.3311, "step": 344000 }, { "epoch": 8.0, "learning_rate": 1.0011607661056298e-05, "loss": 0.3311, "step": 344500 }, { "epoch": 8.0, "eval_accuracy": 0.8951035305605985, "eval_loss": 0.32558220624923706, "eval_runtime": 1982.3976, "eval_samples_per_second": 192.93, "eval_steps_per_second": 2.412, "step": 344600 }, { "epoch": 8.01, "learning_rate": 9.953569355774813e-06, "loss": 0.3307, "step": 345000 }, { "epoch": 8.02, "learning_rate": 9.895531050493327e-06, "loss": 0.3303, "step": 345500 }, { "epoch": 8.03, "learning_rate": 9.83749274521184e-06, "loss": 0.3309, "step": 346000 }, { "epoch": 8.04, "learning_rate": 9.779454439930355e-06, "loss": 0.3306, "step": 346500 }, { "epoch": 8.06, "learning_rate": 9.721416134648869e-06, "loss": 0.3305, "step": 347000 }, { "epoch": 8.07, "learning_rate": 9.663377829367382e-06, "loss": 0.3306, "step": 347500 }, { "epoch": 8.08, "learning_rate": 9.605339524085898e-06, "loss": 0.3308, "step": 348000 }, { "epoch": 8.09, "learning_rate": 9.547301218804411e-06, "loss": 0.3306, "step": 348500 }, { "epoch": 8.1, "learning_rate": 9.489262913522926e-06, "loss": 0.3307, "step": 349000 }, { "epoch": 8.11, "learning_rate": 9.43122460824144e-06, "loss": 0.3307, "step": 349500 }, { "epoch": 8.13, "learning_rate": 9.373186302959953e-06, "loss": 0.3302, "step": 350000 }, { "epoch": 8.14, "learning_rate": 9.315147997678469e-06, "loss": 0.3308, "step": 350500 }, { "epoch": 8.15, "learning_rate": 9.257109692396982e-06, "loss": 0.3304, "step": 351000 }, { "epoch": 8.16, "learning_rate": 9.199071387115496e-06, "loss": 0.3305, "step": 351500 }, { "epoch": 8.17, "learning_rate": 9.141033081834011e-06, "loss": 0.3305, "step": 352000 }, { "epoch": 8.18, "learning_rate": 9.082994776552524e-06, "loss": 0.3305, "step": 352500 }, { "epoch": 8.2, "learning_rate": 9.02495647127104e-06, "loss": 0.3306, "step": 353000 }, { "epoch": 8.21, "learning_rate": 8.966918165989553e-06, "loss": 0.3308, "step": 353500 }, { "epoch": 8.22, "learning_rate": 8.908879860708068e-06, "loss": 0.3304, "step": 354000 }, { "epoch": 8.23, "learning_rate": 8.850841555426582e-06, "loss": 0.331, "step": 354500 }, { "epoch": 8.24, "learning_rate": 8.792803250145096e-06, "loss": 0.3305, "step": 355000 }, { "epoch": 8.25, "learning_rate": 8.73476494486361e-06, "loss": 0.3307, "step": 355500 }, { "epoch": 8.26, "learning_rate": 8.676726639582124e-06, "loss": 0.3309, "step": 356000 }, { "epoch": 8.28, "learning_rate": 8.618688334300638e-06, "loss": 0.3304, "step": 356500 }, { "epoch": 8.29, "learning_rate": 8.560650029019153e-06, "loss": 0.3304, "step": 357000 }, { "epoch": 8.3, "learning_rate": 8.502611723737668e-06, "loss": 0.3306, "step": 357500 }, { "epoch": 8.31, "learning_rate": 8.44457341845618e-06, "loss": 0.3304, "step": 358000 }, { "epoch": 8.32, "learning_rate": 8.386535113174695e-06, "loss": 0.3305, "step": 358500 }, { "epoch": 8.33, "learning_rate": 8.32849680789321e-06, "loss": 0.3308, "step": 359000 }, { "epoch": 8.35, "learning_rate": 8.270458502611724e-06, "loss": 0.3303, "step": 359500 }, { "epoch": 8.36, "learning_rate": 8.212420197330238e-06, "loss": 0.3305, "step": 360000 }, { "epoch": 8.37, "learning_rate": 8.154381892048753e-06, "loss": 0.3308, "step": 360500 }, { "epoch": 8.38, "learning_rate": 8.096343586767268e-06, "loss": 0.3303, "step": 361000 }, { "epoch": 8.39, "learning_rate": 8.03830528148578e-06, "loss": 0.3306, "step": 361500 }, { "epoch": 8.4, "learning_rate": 7.980266976204295e-06, "loss": 0.3305, "step": 362000 }, { "epoch": 8.42, "learning_rate": 7.92222867092281e-06, "loss": 0.3301, "step": 362500 }, { "epoch": 8.43, "learning_rate": 7.864190365641324e-06, "loss": 0.3303, "step": 363000 }, { "epoch": 8.44, "learning_rate": 7.806152060359837e-06, "loss": 0.33, "step": 363500 }, { "epoch": 8.45, "learning_rate": 7.748113755078353e-06, "loss": 0.3302, "step": 364000 }, { "epoch": 8.46, "learning_rate": 7.690075449796866e-06, "loss": 0.3304, "step": 364500 }, { "epoch": 8.47, "learning_rate": 7.63203714451538e-06, "loss": 0.3305, "step": 365000 }, { "epoch": 8.49, "learning_rate": 7.573998839233895e-06, "loss": 0.3303, "step": 365500 }, { "epoch": 8.5, "learning_rate": 7.5159605339524085e-06, "loss": 0.3303, "step": 366000 }, { "epoch": 8.51, "learning_rate": 7.457922228670923e-06, "loss": 0.3303, "step": 366500 }, { "epoch": 8.52, "learning_rate": 7.399883923389437e-06, "loss": 0.3302, "step": 367000 }, { "epoch": 8.53, "learning_rate": 7.341845618107951e-06, "loss": 0.3307, "step": 367500 }, { "epoch": 8.54, "learning_rate": 7.283807312826465e-06, "loss": 0.3303, "step": 368000 }, { "epoch": 8.55, "learning_rate": 7.22576900754498e-06, "loss": 0.3303, "step": 368500 }, { "epoch": 8.57, "learning_rate": 7.167730702263495e-06, "loss": 0.3304, "step": 369000 }, { "epoch": 8.58, "learning_rate": 7.109692396982008e-06, "loss": 0.3303, "step": 369500 }, { "epoch": 8.59, "learning_rate": 7.051654091700523e-06, "loss": 0.3303, "step": 370000 }, { "epoch": 8.6, "learning_rate": 6.993615786419037e-06, "loss": 0.3303, "step": 370500 }, { "epoch": 8.61, "learning_rate": 6.935577481137551e-06, "loss": 0.3302, "step": 371000 }, { "epoch": 8.62, "learning_rate": 6.877539175856065e-06, "loss": 0.3305, "step": 371500 }, { "epoch": 8.64, "learning_rate": 6.81950087057458e-06, "loss": 0.3302, "step": 372000 }, { "epoch": 8.65, "learning_rate": 6.761462565293093e-06, "loss": 0.3304, "step": 372500 }, { "epoch": 8.66, "learning_rate": 6.703424260011608e-06, "loss": 0.3301, "step": 373000 }, { "epoch": 8.67, "learning_rate": 6.6453859547301225e-06, "loss": 0.3303, "step": 373500 }, { "epoch": 8.68, "learning_rate": 6.587347649448636e-06, "loss": 0.3304, "step": 374000 }, { "epoch": 8.69, "learning_rate": 6.5293093441671504e-06, "loss": 0.3304, "step": 374500 }, { "epoch": 8.71, "learning_rate": 6.471271038885665e-06, "loss": 0.3302, "step": 375000 }, { "epoch": 8.72, "learning_rate": 6.41323273360418e-06, "loss": 0.3302, "step": 375500 }, { "epoch": 8.73, "learning_rate": 6.355194428322693e-06, "loss": 0.3305, "step": 376000 }, { "epoch": 8.74, "learning_rate": 6.297156123041208e-06, "loss": 0.3304, "step": 376500 }, { "epoch": 8.75, "learning_rate": 6.2391178177597215e-06, "loss": 0.3304, "step": 377000 }, { "epoch": 8.76, "learning_rate": 6.181079512478236e-06, "loss": 0.3305, "step": 377500 }, { "epoch": 8.78, "learning_rate": 6.12304120719675e-06, "loss": 0.3298, "step": 378000 }, { "epoch": 8.79, "learning_rate": 6.065002901915264e-06, "loss": 0.3308, "step": 378500 }, { "epoch": 8.8, "learning_rate": 6.006964596633779e-06, "loss": 0.3303, "step": 379000 }, { "epoch": 8.81, "learning_rate": 5.9489262913522925e-06, "loss": 0.3301, "step": 379500 }, { "epoch": 8.82, "learning_rate": 5.890887986070807e-06, "loss": 0.3302, "step": 380000 }, { "epoch": 8.83, "learning_rate": 5.832849680789321e-06, "loss": 0.3306, "step": 380500 }, { "epoch": 8.85, "learning_rate": 5.774811375507836e-06, "loss": 0.3306, "step": 381000 }, { "epoch": 8.86, "learning_rate": 5.71677307022635e-06, "loss": 0.3304, "step": 381500 }, { "epoch": 8.87, "learning_rate": 5.658734764944864e-06, "loss": 0.33, "step": 382000 }, { "epoch": 8.88, "learning_rate": 5.600696459663378e-06, "loss": 0.3303, "step": 382500 }, { "epoch": 8.89, "learning_rate": 5.542658154381892e-06, "loss": 0.3301, "step": 383000 }, { "epoch": 8.9, "learning_rate": 5.484619849100407e-06, "loss": 0.3303, "step": 383500 }, { "epoch": 8.91, "learning_rate": 5.42658154381892e-06, "loss": 0.3305, "step": 384000 }, { "epoch": 8.93, "learning_rate": 5.3685432385374355e-06, "loss": 0.3299, "step": 384500 }, { "epoch": 8.94, "learning_rate": 5.310504933255949e-06, "loss": 0.3305, "step": 385000 }, { "epoch": 8.95, "learning_rate": 5.252466627974463e-06, "loss": 0.3304, "step": 385500 }, { "epoch": 8.96, "learning_rate": 5.194428322692978e-06, "loss": 0.33, "step": 386000 }, { "epoch": 8.97, "learning_rate": 5.136390017411491e-06, "loss": 0.3304, "step": 386500 }, { "epoch": 8.98, "learning_rate": 5.0783517121300065e-06, "loss": 0.3301, "step": 387000 }, { "epoch": 9.0, "learning_rate": 5.02031340684852e-06, "loss": 0.33, "step": 387500 }, { "epoch": 9.0, "eval_accuracy": 0.8952824353922539, "eval_loss": 0.3249286413192749, "eval_runtime": 1998.0692, "eval_samples_per_second": 191.417, "eval_steps_per_second": 2.393, "step": 387675 }, { "epoch": 9.01, "learning_rate": 4.9622751015670345e-06, "loss": 0.33, "step": 388000 }, { "epoch": 9.02, "learning_rate": 4.904236796285549e-06, "loss": 0.3294, "step": 388500 }, { "epoch": 9.03, "learning_rate": 4.846198491004063e-06, "loss": 0.3299, "step": 389000 }, { "epoch": 9.04, "learning_rate": 4.788160185722578e-06, "loss": 0.3298, "step": 389500 }, { "epoch": 9.05, "learning_rate": 4.730121880441091e-06, "loss": 0.3296, "step": 390000 }, { "epoch": 9.07, "learning_rate": 4.6720835751596055e-06, "loss": 0.3296, "step": 390500 }, { "epoch": 9.08, "learning_rate": 4.61404526987812e-06, "loss": 0.3297, "step": 391000 }, { "epoch": 9.09, "learning_rate": 4.556006964596634e-06, "loss": 0.3294, "step": 391500 }, { "epoch": 9.1, "learning_rate": 4.497968659315148e-06, "loss": 0.3301, "step": 392000 }, { "epoch": 9.11, "learning_rate": 4.439930354033662e-06, "loss": 0.3295, "step": 392500 }, { "epoch": 9.12, "learning_rate": 4.3818920487521766e-06, "loss": 0.3299, "step": 393000 }, { "epoch": 9.14, "learning_rate": 4.323853743470691e-06, "loss": 0.3298, "step": 393500 }, { "epoch": 9.15, "learning_rate": 4.265815438189205e-06, "loss": 0.3294, "step": 394000 }, { "epoch": 9.16, "learning_rate": 4.207777132907719e-06, "loss": 0.3297, "step": 394500 }, { "epoch": 9.17, "learning_rate": 4.149738827626234e-06, "loss": 0.3297, "step": 395000 }, { "epoch": 9.18, "learning_rate": 4.091700522344748e-06, "loss": 0.3297, "step": 395500 }, { "epoch": 9.19, "learning_rate": 4.033662217063262e-06, "loss": 0.3296, "step": 396000 }, { "epoch": 9.2, "learning_rate": 3.975623911781776e-06, "loss": 0.3299, "step": 396500 }, { "epoch": 9.22, "learning_rate": 3.91758560650029e-06, "loss": 0.3295, "step": 397000 }, { "epoch": 9.23, "learning_rate": 3.859547301218805e-06, "loss": 0.3299, "step": 397500 }, { "epoch": 9.24, "learning_rate": 3.8015089959373187e-06, "loss": 0.33, "step": 398000 }, { "epoch": 9.25, "learning_rate": 3.7434706906558326e-06, "loss": 0.3296, "step": 398500 }, { "epoch": 9.26, "learning_rate": 3.6854323853743474e-06, "loss": 0.3295, "step": 399000 }, { "epoch": 9.27, "learning_rate": 3.6273940800928614e-06, "loss": 0.3295, "step": 399500 }, { "epoch": 9.29, "learning_rate": 3.5693557748113754e-06, "loss": 0.3295, "step": 400000 }, { "epoch": 9.3, "learning_rate": 3.51131746952989e-06, "loss": 0.3296, "step": 400500 }, { "epoch": 9.31, "learning_rate": 3.453279164248404e-06, "loss": 0.3299, "step": 401000 }, { "epoch": 9.32, "learning_rate": 3.3952408589669185e-06, "loss": 0.3294, "step": 401500 }, { "epoch": 9.33, "learning_rate": 3.3372025536854324e-06, "loss": 0.33, "step": 402000 }, { "epoch": 9.34, "learning_rate": 3.2791642484039464e-06, "loss": 0.3296, "step": 402500 }, { "epoch": 9.36, "learning_rate": 3.221125943122461e-06, "loss": 0.3298, "step": 403000 }, { "epoch": 9.37, "learning_rate": 3.163087637840975e-06, "loss": 0.3299, "step": 403500 }, { "epoch": 9.38, "learning_rate": 3.1050493325594895e-06, "loss": 0.3299, "step": 404000 }, { "epoch": 9.39, "learning_rate": 3.047011027278004e-06, "loss": 0.3299, "step": 404500 }, { "epoch": 9.4, "learning_rate": 2.988972721996518e-06, "loss": 0.3295, "step": 405000 }, { "epoch": 9.41, "learning_rate": 2.930934416715032e-06, "loss": 0.3293, "step": 405500 }, { "epoch": 9.43, "learning_rate": 2.8728961114335462e-06, "loss": 0.3294, "step": 406000 }, { "epoch": 9.44, "learning_rate": 2.8148578061520606e-06, "loss": 0.3295, "step": 406500 }, { "epoch": 9.45, "learning_rate": 2.7568195008705746e-06, "loss": 0.3298, "step": 407000 }, { "epoch": 9.46, "learning_rate": 2.698781195589089e-06, "loss": 0.3298, "step": 407500 }, { "epoch": 9.47, "learning_rate": 2.6407428903076033e-06, "loss": 0.3296, "step": 408000 }, { "epoch": 9.48, "learning_rate": 2.5827045850261177e-06, "loss": 0.3298, "step": 408500 }, { "epoch": 9.5, "learning_rate": 2.5246662797446317e-06, "loss": 0.3293, "step": 409000 }, { "epoch": 9.51, "learning_rate": 2.4666279744631456e-06, "loss": 0.3296, "step": 409500 }, { "epoch": 9.52, "learning_rate": 2.40858966918166e-06, "loss": 0.3293, "step": 410000 }, { "epoch": 9.53, "learning_rate": 2.3505513639001744e-06, "loss": 0.3298, "step": 410500 }, { "epoch": 9.54, "learning_rate": 2.2925130586186883e-06, "loss": 0.3296, "step": 411000 }, { "epoch": 9.55, "learning_rate": 2.2344747533372027e-06, "loss": 0.3297, "step": 411500 }, { "epoch": 9.56, "learning_rate": 2.176436448055717e-06, "loss": 0.3294, "step": 412000 }, { "epoch": 9.58, "learning_rate": 2.118398142774231e-06, "loss": 0.3299, "step": 412500 }, { "epoch": 9.59, "learning_rate": 2.060359837492745e-06, "loss": 0.3296, "step": 413000 }, { "epoch": 9.6, "learning_rate": 2.0023215322112594e-06, "loss": 0.3297, "step": 413500 }, { "epoch": 9.61, "learning_rate": 1.9442832269297738e-06, "loss": 0.3294, "step": 414000 }, { "epoch": 9.62, "learning_rate": 1.8862449216482881e-06, "loss": 0.3296, "step": 414500 }, { "epoch": 9.63, "learning_rate": 1.828206616366802e-06, "loss": 0.3297, "step": 415000 }, { "epoch": 9.65, "learning_rate": 1.7701683110853163e-06, "loss": 0.3298, "step": 415500 }, { "epoch": 9.66, "learning_rate": 1.7121300058038307e-06, "loss": 0.3291, "step": 416000 }, { "epoch": 9.67, "learning_rate": 1.654091700522345e-06, "loss": 0.3299, "step": 416500 }, { "epoch": 9.68, "learning_rate": 1.596053395240859e-06, "loss": 0.3295, "step": 417000 }, { "epoch": 9.69, "learning_rate": 1.5380150899593732e-06, "loss": 0.3298, "step": 417500 }, { "epoch": 9.7, "learning_rate": 1.4799767846778875e-06, "loss": 0.3297, "step": 418000 }, { "epoch": 9.72, "learning_rate": 1.4219384793964017e-06, "loss": 0.3298, "step": 418500 }, { "epoch": 9.73, "learning_rate": 1.3639001741149159e-06, "loss": 0.3295, "step": 419000 }, { "epoch": 9.74, "learning_rate": 1.30586186883343e-06, "loss": 0.3299, "step": 419500 }, { "epoch": 9.75, "learning_rate": 1.2478235635519444e-06, "loss": 0.3294, "step": 420000 }, { "epoch": 9.76, "learning_rate": 1.1897852582704586e-06, "loss": 0.3298, "step": 420500 }, { "epoch": 9.77, "learning_rate": 1.1317469529889728e-06, "loss": 0.3299, "step": 421000 }, { "epoch": 9.79, "learning_rate": 1.073708647707487e-06, "loss": 0.3297, "step": 421500 }, { "epoch": 9.8, "learning_rate": 1.0156703424260013e-06, "loss": 0.3299, "step": 422000 }, { "epoch": 9.81, "learning_rate": 9.576320371445155e-07, "loss": 0.3297, "step": 422500 }, { "epoch": 9.82, "learning_rate": 8.995937318630296e-07, "loss": 0.3295, "step": 423000 }, { "epoch": 9.83, "learning_rate": 8.415554265815438e-07, "loss": 0.3294, "step": 423500 }, { "epoch": 9.84, "learning_rate": 7.835171213000581e-07, "loss": 0.3296, "step": 424000 }, { "epoch": 9.85, "learning_rate": 7.254788160185723e-07, "loss": 0.3292, "step": 424500 }, { "epoch": 9.87, "learning_rate": 6.674405107370864e-07, "loss": 0.3296, "step": 425000 }, { "epoch": 9.88, "learning_rate": 6.094022054556007e-07, "loss": 0.3293, "step": 425500 }, { "epoch": 9.89, "learning_rate": 5.513639001741149e-07, "loss": 0.3295, "step": 426000 }, { "epoch": 9.9, "learning_rate": 4.933255948926291e-07, "loss": 0.3295, "step": 426500 }, { "epoch": 9.91, "learning_rate": 4.352872896111433e-07, "loss": 0.3294, "step": 427000 }, { "epoch": 9.92, "learning_rate": 3.772489843296576e-07, "loss": 0.3296, "step": 427500 }, { "epoch": 9.94, "learning_rate": 3.192106790481718e-07, "loss": 0.3299, "step": 428000 }, { "epoch": 9.95, "learning_rate": 2.6117237376668603e-07, "loss": 0.3297, "step": 428500 }, { "epoch": 9.96, "learning_rate": 2.0313406848520025e-07, "loss": 0.3294, "step": 429000 }, { "epoch": 9.97, "learning_rate": 1.4509576320371445e-07, "loss": 0.3293, "step": 429500 }, { "epoch": 9.98, "learning_rate": 8.705745792222868e-08, "loss": 0.3297, "step": 430000 }, { "epoch": 9.99, "learning_rate": 2.9019152640742893e-08, "loss": 0.3296, "step": 430500 }, { "epoch": 10.0, "eval_accuracy": 0.8953714759150758, "eval_loss": 0.3246208131313324, "eval_runtime": 1996.1692, "eval_samples_per_second": 191.599, "eval_steps_per_second": 2.395, "step": 430750 }, { "epoch": 10.0, "step": 430750, "total_flos": 9.004123682884289e+18, "train_loss": 0.34171454371066545, "train_runtime": 354822.8145, "train_samples_per_second": 97.119, "train_steps_per_second": 1.214 } ], "logging_steps": 500, "max_steps": 430750, "num_input_tokens_seen": 0, "num_train_epochs": 10, "save_steps": 500, "total_flos": 9.004123682884289e+18, "train_batch_size": 16, "trial_name": null, "trial_params": null }