jonasknobloch's picture
Initial commit
384ca66
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"eval_steps": 10000,
"global_step": 319678,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 4.992179630753446e-05,
"loss": 6.4174,
"step": 500
},
{
"epoch": 0.0,
"learning_rate": 4.984359261506892e-05,
"loss": 5.3568,
"step": 1000
},
{
"epoch": 0.0,
"learning_rate": 4.9765388922603374e-05,
"loss": 5.0234,
"step": 1500
},
{
"epoch": 0.01,
"learning_rate": 4.968718523013783e-05,
"loss": 4.7982,
"step": 2000
},
{
"epoch": 0.01,
"learning_rate": 4.960898153767228e-05,
"loss": 4.5929,
"step": 2500
},
{
"epoch": 0.01,
"learning_rate": 4.9530777845206745e-05,
"loss": 4.4101,
"step": 3000
},
{
"epoch": 0.01,
"learning_rate": 4.94525741527412e-05,
"loss": 4.2672,
"step": 3500
},
{
"epoch": 0.01,
"learning_rate": 4.9374370460275654e-05,
"loss": 4.1536,
"step": 4000
},
{
"epoch": 0.01,
"learning_rate": 4.929616676781011e-05,
"loss": 4.0694,
"step": 4500
},
{
"epoch": 0.02,
"learning_rate": 4.921796307534457e-05,
"loss": 3.9953,
"step": 5000
},
{
"epoch": 0.02,
"learning_rate": 4.9139759382879026e-05,
"loss": 3.9374,
"step": 5500
},
{
"epoch": 0.02,
"learning_rate": 4.906155569041348e-05,
"loss": 3.8853,
"step": 6000
},
{
"epoch": 0.02,
"learning_rate": 4.8983351997947935e-05,
"loss": 3.8492,
"step": 6500
},
{
"epoch": 0.02,
"learning_rate": 4.89051483054824e-05,
"loss": 3.8114,
"step": 7000
},
{
"epoch": 0.02,
"learning_rate": 4.882694461301685e-05,
"loss": 3.7802,
"step": 7500
},
{
"epoch": 0.03,
"learning_rate": 4.8748740920551307e-05,
"loss": 3.7541,
"step": 8000
},
{
"epoch": 0.03,
"learning_rate": 4.867053722808577e-05,
"loss": 3.7294,
"step": 8500
},
{
"epoch": 0.03,
"learning_rate": 4.859233353562022e-05,
"loss": 3.704,
"step": 9000
},
{
"epoch": 0.03,
"learning_rate": 4.851412984315468e-05,
"loss": 3.6819,
"step": 9500
},
{
"epoch": 0.03,
"learning_rate": 4.843592615068913e-05,
"loss": 3.6569,
"step": 10000
},
{
"epoch": 0.03,
"eval_accuracy": 0.35023928642253593,
"eval_loss": 3.576359748840332,
"eval_runtime": 5980.3073,
"eval_samples_per_second": 90.014,
"eval_steps_per_second": 1.407,
"step": 10000
},
{
"epoch": 0.03,
"learning_rate": 4.8357722458223594e-05,
"loss": 3.6327,
"step": 10500
},
{
"epoch": 0.03,
"learning_rate": 4.827951876575805e-05,
"loss": 3.6243,
"step": 11000
},
{
"epoch": 0.04,
"learning_rate": 4.8201315073292504e-05,
"loss": 3.6073,
"step": 11500
},
{
"epoch": 0.04,
"learning_rate": 4.812311138082696e-05,
"loss": 3.5984,
"step": 12000
},
{
"epoch": 0.04,
"learning_rate": 4.804490768836142e-05,
"loss": 3.5788,
"step": 12500
},
{
"epoch": 0.04,
"learning_rate": 4.7966703995895875e-05,
"loss": 3.565,
"step": 13000
},
{
"epoch": 0.04,
"learning_rate": 4.788850030343033e-05,
"loss": 3.5536,
"step": 13500
},
{
"epoch": 0.04,
"learning_rate": 4.7810296610964785e-05,
"loss": 3.5403,
"step": 14000
},
{
"epoch": 0.05,
"learning_rate": 4.7732092918499246e-05,
"loss": 3.5292,
"step": 14500
},
{
"epoch": 0.05,
"learning_rate": 4.76538892260337e-05,
"loss": 3.5258,
"step": 15000
},
{
"epoch": 0.05,
"learning_rate": 4.7575685533568156e-05,
"loss": 3.5033,
"step": 15500
},
{
"epoch": 0.05,
"learning_rate": 4.749748184110261e-05,
"loss": 3.4912,
"step": 16000
},
{
"epoch": 0.05,
"learning_rate": 4.741927814863707e-05,
"loss": 3.4873,
"step": 16500
},
{
"epoch": 0.05,
"learning_rate": 4.734107445617153e-05,
"loss": 3.4821,
"step": 17000
},
{
"epoch": 0.05,
"learning_rate": 4.726287076370598e-05,
"loss": 3.4656,
"step": 17500
},
{
"epoch": 0.06,
"learning_rate": 4.7184667071240443e-05,
"loss": 3.4606,
"step": 18000
},
{
"epoch": 0.06,
"learning_rate": 4.71064633787749e-05,
"loss": 3.4492,
"step": 18500
},
{
"epoch": 0.06,
"learning_rate": 4.702825968630935e-05,
"loss": 3.4429,
"step": 19000
},
{
"epoch": 0.06,
"learning_rate": 4.695005599384381e-05,
"loss": 3.4329,
"step": 19500
},
{
"epoch": 0.06,
"learning_rate": 4.687185230137827e-05,
"loss": 3.4317,
"step": 20000
},
{
"epoch": 0.06,
"eval_accuracy": 0.37270295864352604,
"eval_loss": 3.358124256134033,
"eval_runtime": 5996.148,
"eval_samples_per_second": 89.776,
"eval_steps_per_second": 1.403,
"step": 20000
},
{
"epoch": 0.06,
"learning_rate": 4.6793648608912724e-05,
"loss": 3.4168,
"step": 20500
},
{
"epoch": 0.07,
"learning_rate": 4.671544491644718e-05,
"loss": 3.4053,
"step": 21000
},
{
"epoch": 0.07,
"learning_rate": 4.6637241223981634e-05,
"loss": 3.4086,
"step": 21500
},
{
"epoch": 0.07,
"learning_rate": 4.655903753151609e-05,
"loss": 3.4018,
"step": 22000
},
{
"epoch": 0.07,
"learning_rate": 4.648083383905055e-05,
"loss": 3.3929,
"step": 22500
},
{
"epoch": 0.07,
"learning_rate": 4.6402630146585005e-05,
"loss": 3.3869,
"step": 23000
},
{
"epoch": 0.07,
"learning_rate": 4.632442645411946e-05,
"loss": 3.3841,
"step": 23500
},
{
"epoch": 0.08,
"learning_rate": 4.6246222761653915e-05,
"loss": 3.3783,
"step": 24000
},
{
"epoch": 0.08,
"learning_rate": 4.6168019069188376e-05,
"loss": 3.3731,
"step": 24500
},
{
"epoch": 0.08,
"learning_rate": 4.608981537672283e-05,
"loss": 3.3619,
"step": 25000
},
{
"epoch": 0.08,
"learning_rate": 4.6011611684257286e-05,
"loss": 3.361,
"step": 25500
},
{
"epoch": 0.08,
"learning_rate": 4.593340799179174e-05,
"loss": 3.3522,
"step": 26000
},
{
"epoch": 0.08,
"learning_rate": 4.5855204299326196e-05,
"loss": 3.3518,
"step": 26500
},
{
"epoch": 0.08,
"learning_rate": 4.577700060686066e-05,
"loss": 3.3413,
"step": 27000
},
{
"epoch": 0.09,
"learning_rate": 4.569879691439511e-05,
"loss": 3.3311,
"step": 27500
},
{
"epoch": 0.09,
"learning_rate": 4.562059322192957e-05,
"loss": 3.3358,
"step": 28000
},
{
"epoch": 0.09,
"learning_rate": 4.554238952946402e-05,
"loss": 3.3264,
"step": 28500
},
{
"epoch": 0.09,
"learning_rate": 4.5464185836998476e-05,
"loss": 3.3236,
"step": 29000
},
{
"epoch": 0.09,
"learning_rate": 4.538598214453294e-05,
"loss": 3.3148,
"step": 29500
},
{
"epoch": 0.09,
"learning_rate": 4.530777845206739e-05,
"loss": 3.3161,
"step": 30000
},
{
"epoch": 0.09,
"eval_accuracy": 0.3848112637589684,
"eval_loss": 3.2447316646575928,
"eval_runtime": 6006.5513,
"eval_samples_per_second": 89.621,
"eval_steps_per_second": 1.4,
"step": 30000
},
{
"epoch": 0.1,
"learning_rate": 4.522957475960185e-05,
"loss": 3.3048,
"step": 30500
},
{
"epoch": 0.1,
"learning_rate": 4.51513710671363e-05,
"loss": 3.3062,
"step": 31000
},
{
"epoch": 0.1,
"learning_rate": 4.5073167374670764e-05,
"loss": 3.3039,
"step": 31500
},
{
"epoch": 0.1,
"learning_rate": 4.499496368220522e-05,
"loss": 3.3031,
"step": 32000
},
{
"epoch": 0.1,
"learning_rate": 4.4916759989739674e-05,
"loss": 3.2953,
"step": 32500
},
{
"epoch": 0.1,
"learning_rate": 4.4838556297274135e-05,
"loss": 3.2892,
"step": 33000
},
{
"epoch": 0.1,
"learning_rate": 4.476035260480859e-05,
"loss": 3.2839,
"step": 33500
},
{
"epoch": 0.11,
"learning_rate": 4.4682148912343045e-05,
"loss": 3.281,
"step": 34000
},
{
"epoch": 0.11,
"learning_rate": 4.46039452198775e-05,
"loss": 3.2831,
"step": 34500
},
{
"epoch": 0.11,
"learning_rate": 4.452574152741196e-05,
"loss": 3.2739,
"step": 35000
},
{
"epoch": 0.11,
"learning_rate": 4.4447537834946416e-05,
"loss": 3.2801,
"step": 35500
},
{
"epoch": 0.11,
"learning_rate": 4.436933414248087e-05,
"loss": 3.2715,
"step": 36000
},
{
"epoch": 0.11,
"learning_rate": 4.4291130450015326e-05,
"loss": 3.2685,
"step": 36500
},
{
"epoch": 0.12,
"learning_rate": 4.421292675754979e-05,
"loss": 3.2653,
"step": 37000
},
{
"epoch": 0.12,
"learning_rate": 4.413472306508424e-05,
"loss": 3.2563,
"step": 37500
},
{
"epoch": 0.12,
"learning_rate": 4.40565193726187e-05,
"loss": 3.2579,
"step": 38000
},
{
"epoch": 0.12,
"learning_rate": 4.397831568015315e-05,
"loss": 3.2497,
"step": 38500
},
{
"epoch": 0.12,
"learning_rate": 4.390011198768761e-05,
"loss": 3.2465,
"step": 39000
},
{
"epoch": 0.12,
"learning_rate": 4.382190829522207e-05,
"loss": 3.2437,
"step": 39500
},
{
"epoch": 0.13,
"learning_rate": 4.374370460275652e-05,
"loss": 3.2463,
"step": 40000
},
{
"epoch": 0.13,
"eval_accuracy": 0.39235570262450464,
"eval_loss": 3.176072835922241,
"eval_runtime": 6008.1585,
"eval_samples_per_second": 89.597,
"eval_steps_per_second": 1.4,
"step": 40000
},
{
"epoch": 0.13,
"learning_rate": 4.3665500910290985e-05,
"loss": 3.2465,
"step": 40500
},
{
"epoch": 0.13,
"learning_rate": 4.358729721782544e-05,
"loss": 3.2423,
"step": 41000
},
{
"epoch": 0.13,
"learning_rate": 4.3509093525359894e-05,
"loss": 3.2381,
"step": 41500
},
{
"epoch": 0.13,
"learning_rate": 4.343088983289435e-05,
"loss": 3.2327,
"step": 42000
},
{
"epoch": 0.13,
"learning_rate": 4.335268614042881e-05,
"loss": 3.2307,
"step": 42500
},
{
"epoch": 0.13,
"learning_rate": 4.3274482447963265e-05,
"loss": 3.2348,
"step": 43000
},
{
"epoch": 0.14,
"learning_rate": 4.319627875549772e-05,
"loss": 3.2245,
"step": 43500
},
{
"epoch": 0.14,
"learning_rate": 4.3118075063032175e-05,
"loss": 3.2262,
"step": 44000
},
{
"epoch": 0.14,
"learning_rate": 4.3039871370566637e-05,
"loss": 3.2169,
"step": 44500
},
{
"epoch": 0.14,
"learning_rate": 4.296166767810109e-05,
"loss": 3.2157,
"step": 45000
},
{
"epoch": 0.14,
"learning_rate": 4.2883463985635546e-05,
"loss": 3.2128,
"step": 45500
},
{
"epoch": 0.14,
"learning_rate": 4.280526029317e-05,
"loss": 3.2125,
"step": 46000
},
{
"epoch": 0.15,
"learning_rate": 4.272705660070446e-05,
"loss": 3.2079,
"step": 46500
},
{
"epoch": 0.15,
"learning_rate": 4.264885290823892e-05,
"loss": 3.2102,
"step": 47000
},
{
"epoch": 0.15,
"learning_rate": 4.257064921577337e-05,
"loss": 3.2079,
"step": 47500
},
{
"epoch": 0.15,
"learning_rate": 4.249244552330783e-05,
"loss": 3.2032,
"step": 48000
},
{
"epoch": 0.15,
"learning_rate": 4.241424183084229e-05,
"loss": 3.2009,
"step": 48500
},
{
"epoch": 0.15,
"learning_rate": 4.2336038138376743e-05,
"loss": 3.2049,
"step": 49000
},
{
"epoch": 0.15,
"learning_rate": 4.22578344459112e-05,
"loss": 3.1951,
"step": 49500
},
{
"epoch": 0.16,
"learning_rate": 4.217963075344566e-05,
"loss": 3.1897,
"step": 50000
},
{
"epoch": 0.16,
"eval_accuracy": 0.39768394188345635,
"eval_loss": 3.127730369567871,
"eval_runtime": 6028.4791,
"eval_samples_per_second": 89.295,
"eval_steps_per_second": 1.395,
"step": 50000
},
{
"epoch": 0.16,
"learning_rate": 4.2101427060980115e-05,
"loss": 3.1863,
"step": 50500
},
{
"epoch": 0.16,
"learning_rate": 4.202322336851457e-05,
"loss": 3.1942,
"step": 51000
},
{
"epoch": 0.16,
"learning_rate": 4.1945019676049024e-05,
"loss": 3.192,
"step": 51500
},
{
"epoch": 0.16,
"learning_rate": 4.1866815983583486e-05,
"loss": 3.1954,
"step": 52000
},
{
"epoch": 0.16,
"learning_rate": 4.178861229111794e-05,
"loss": 3.1879,
"step": 52500
},
{
"epoch": 0.17,
"learning_rate": 4.1710408598652395e-05,
"loss": 3.1788,
"step": 53000
},
{
"epoch": 0.17,
"learning_rate": 4.163220490618685e-05,
"loss": 3.1776,
"step": 53500
},
{
"epoch": 0.17,
"learning_rate": 4.155400121372131e-05,
"loss": 3.1808,
"step": 54000
},
{
"epoch": 0.17,
"learning_rate": 4.147579752125577e-05,
"loss": 3.1724,
"step": 54500
},
{
"epoch": 0.17,
"learning_rate": 4.139759382879022e-05,
"loss": 3.1746,
"step": 55000
},
{
"epoch": 0.17,
"learning_rate": 4.1319390136324676e-05,
"loss": 3.1694,
"step": 55500
},
{
"epoch": 0.18,
"learning_rate": 4.124118644385914e-05,
"loss": 3.1685,
"step": 56000
},
{
"epoch": 0.18,
"learning_rate": 4.116298275139359e-05,
"loss": 3.1661,
"step": 56500
},
{
"epoch": 0.18,
"learning_rate": 4.108477905892805e-05,
"loss": 3.1644,
"step": 57000
},
{
"epoch": 0.18,
"learning_rate": 4.100657536646251e-05,
"loss": 3.1659,
"step": 57500
},
{
"epoch": 0.18,
"learning_rate": 4.0928371673996964e-05,
"loss": 3.1625,
"step": 58000
},
{
"epoch": 0.18,
"learning_rate": 4.085016798153142e-05,
"loss": 3.1587,
"step": 58500
},
{
"epoch": 0.18,
"learning_rate": 4.0771964289065874e-05,
"loss": 3.1563,
"step": 59000
},
{
"epoch": 0.19,
"learning_rate": 4.0693760596600335e-05,
"loss": 3.155,
"step": 59500
},
{
"epoch": 0.19,
"learning_rate": 4.061555690413479e-05,
"loss": 3.152,
"step": 60000
},
{
"epoch": 0.19,
"eval_accuracy": 0.40218440985366416,
"eval_loss": 3.091034412384033,
"eval_runtime": 5997.8965,
"eval_samples_per_second": 89.75,
"eval_steps_per_second": 1.402,
"step": 60000
},
{
"epoch": 0.19,
"learning_rate": 4.0537353211669245e-05,
"loss": 3.1516,
"step": 60500
},
{
"epoch": 0.19,
"learning_rate": 4.04591495192037e-05,
"loss": 3.1454,
"step": 61000
},
{
"epoch": 0.19,
"learning_rate": 4.038094582673816e-05,
"loss": 3.1488,
"step": 61500
},
{
"epoch": 0.19,
"learning_rate": 4.0302742134272616e-05,
"loss": 3.1497,
"step": 62000
},
{
"epoch": 0.2,
"learning_rate": 4.022453844180707e-05,
"loss": 3.1484,
"step": 62500
},
{
"epoch": 0.2,
"learning_rate": 4.0146334749341526e-05,
"loss": 3.1441,
"step": 63000
},
{
"epoch": 0.2,
"learning_rate": 4.006813105687599e-05,
"loss": 3.1412,
"step": 63500
},
{
"epoch": 0.2,
"learning_rate": 3.998992736441044e-05,
"loss": 3.1416,
"step": 64000
},
{
"epoch": 0.2,
"learning_rate": 3.99117236719449e-05,
"loss": 3.1436,
"step": 64500
},
{
"epoch": 0.2,
"learning_rate": 3.983351997947936e-05,
"loss": 3.1409,
"step": 65000
},
{
"epoch": 0.2,
"learning_rate": 3.975531628701381e-05,
"loss": 3.1392,
"step": 65500
},
{
"epoch": 0.21,
"learning_rate": 3.967711259454827e-05,
"loss": 3.1381,
"step": 66000
},
{
"epoch": 0.21,
"learning_rate": 3.959890890208272e-05,
"loss": 3.1345,
"step": 66500
},
{
"epoch": 0.21,
"learning_rate": 3.9520705209617184e-05,
"loss": 3.131,
"step": 67000
},
{
"epoch": 0.21,
"learning_rate": 3.944250151715164e-05,
"loss": 3.1339,
"step": 67500
},
{
"epoch": 0.21,
"learning_rate": 3.9364297824686094e-05,
"loss": 3.1295,
"step": 68000
},
{
"epoch": 0.21,
"learning_rate": 3.928609413222055e-05,
"loss": 3.128,
"step": 68500
},
{
"epoch": 0.22,
"learning_rate": 3.920789043975501e-05,
"loss": 3.1308,
"step": 69000
},
{
"epoch": 0.22,
"learning_rate": 3.9129686747289465e-05,
"loss": 3.1157,
"step": 69500
},
{
"epoch": 0.22,
"learning_rate": 3.905148305482392e-05,
"loss": 3.1341,
"step": 70000
},
{
"epoch": 0.22,
"eval_accuracy": 0.4060193438823364,
"eval_loss": 3.057492256164551,
"eval_runtime": 6003.3195,
"eval_samples_per_second": 89.669,
"eval_steps_per_second": 1.401,
"step": 70000
},
{
"epoch": 0.22,
"learning_rate": 3.8973279362358375e-05,
"loss": 3.1319,
"step": 70500
},
{
"epoch": 0.22,
"learning_rate": 3.8895075669892836e-05,
"loss": 3.1217,
"step": 71000
},
{
"epoch": 0.22,
"learning_rate": 3.881687197742729e-05,
"loss": 3.1223,
"step": 71500
},
{
"epoch": 0.23,
"learning_rate": 3.8738668284961746e-05,
"loss": 3.125,
"step": 72000
},
{
"epoch": 0.23,
"learning_rate": 3.86604645924962e-05,
"loss": 3.1223,
"step": 72500
},
{
"epoch": 0.23,
"learning_rate": 3.8582260900030656e-05,
"loss": 3.1109,
"step": 73000
},
{
"epoch": 0.23,
"learning_rate": 3.850405720756512e-05,
"loss": 3.1138,
"step": 73500
},
{
"epoch": 0.23,
"learning_rate": 3.842585351509957e-05,
"loss": 3.1168,
"step": 74000
},
{
"epoch": 0.23,
"learning_rate": 3.834764982263403e-05,
"loss": 3.1137,
"step": 74500
},
{
"epoch": 0.23,
"learning_rate": 3.826944613016848e-05,
"loss": 3.1085,
"step": 75000
},
{
"epoch": 0.24,
"learning_rate": 3.8191242437702937e-05,
"loss": 3.1145,
"step": 75500
},
{
"epoch": 0.24,
"learning_rate": 3.81130387452374e-05,
"loss": 3.1045,
"step": 76000
},
{
"epoch": 0.24,
"learning_rate": 3.803483505277185e-05,
"loss": 3.1078,
"step": 76500
},
{
"epoch": 0.24,
"learning_rate": 3.795663136030631e-05,
"loss": 3.1054,
"step": 77000
},
{
"epoch": 0.24,
"learning_rate": 3.787842766784076e-05,
"loss": 3.1047,
"step": 77500
},
{
"epoch": 0.24,
"learning_rate": 3.780022397537522e-05,
"loss": 3.1038,
"step": 78000
},
{
"epoch": 0.25,
"learning_rate": 3.772202028290968e-05,
"loss": 3.1034,
"step": 78500
},
{
"epoch": 0.25,
"learning_rate": 3.7643816590444134e-05,
"loss": 3.0953,
"step": 79000
},
{
"epoch": 0.25,
"learning_rate": 3.756561289797859e-05,
"loss": 3.1006,
"step": 79500
},
{
"epoch": 0.25,
"learning_rate": 3.7487409205513043e-05,
"loss": 3.1006,
"step": 80000
},
{
"epoch": 0.25,
"eval_accuracy": 0.4084469897262718,
"eval_loss": 3.0362584590911865,
"eval_runtime": 5986.8174,
"eval_samples_per_second": 89.916,
"eval_steps_per_second": 1.405,
"step": 80000
},
{
"epoch": 0.25,
"learning_rate": 3.7409205513047505e-05,
"loss": 3.0934,
"step": 80500
},
{
"epoch": 0.25,
"learning_rate": 3.733100182058196e-05,
"loss": 3.1024,
"step": 81000
},
{
"epoch": 0.25,
"learning_rate": 3.7252798128116415e-05,
"loss": 3.0957,
"step": 81500
},
{
"epoch": 0.26,
"learning_rate": 3.7174594435650876e-05,
"loss": 3.1003,
"step": 82000
},
{
"epoch": 0.26,
"learning_rate": 3.709639074318533e-05,
"loss": 3.0987,
"step": 82500
},
{
"epoch": 0.26,
"learning_rate": 3.7018187050719786e-05,
"loss": 3.0952,
"step": 83000
},
{
"epoch": 0.26,
"learning_rate": 3.693998335825424e-05,
"loss": 3.0971,
"step": 83500
},
{
"epoch": 0.26,
"learning_rate": 3.68617796657887e-05,
"loss": 3.0974,
"step": 84000
},
{
"epoch": 0.26,
"learning_rate": 3.678357597332316e-05,
"loss": 3.0896,
"step": 84500
},
{
"epoch": 0.27,
"learning_rate": 3.670537228085761e-05,
"loss": 3.0862,
"step": 85000
},
{
"epoch": 0.27,
"learning_rate": 3.662716858839207e-05,
"loss": 3.0809,
"step": 85500
},
{
"epoch": 0.27,
"learning_rate": 3.654896489592653e-05,
"loss": 3.088,
"step": 86000
},
{
"epoch": 0.27,
"learning_rate": 3.647076120346098e-05,
"loss": 3.0818,
"step": 86500
},
{
"epoch": 0.27,
"learning_rate": 3.639255751099544e-05,
"loss": 3.0715,
"step": 87000
},
{
"epoch": 0.27,
"learning_rate": 3.631435381852989e-05,
"loss": 3.0869,
"step": 87500
},
{
"epoch": 0.28,
"learning_rate": 3.6236150126064354e-05,
"loss": 3.0856,
"step": 88000
},
{
"epoch": 0.28,
"learning_rate": 3.615794643359881e-05,
"loss": 3.0812,
"step": 88500
},
{
"epoch": 0.28,
"learning_rate": 3.6079742741133264e-05,
"loss": 3.0768,
"step": 89000
},
{
"epoch": 0.28,
"learning_rate": 3.6001539048667726e-05,
"loss": 3.0823,
"step": 89500
},
{
"epoch": 0.28,
"learning_rate": 3.592333535620218e-05,
"loss": 3.0806,
"step": 90000
},
{
"epoch": 0.28,
"eval_accuracy": 0.41145725219591245,
"eval_loss": 3.011796236038208,
"eval_runtime": 5969.2265,
"eval_samples_per_second": 90.181,
"eval_steps_per_second": 1.409,
"step": 90000
},
{
"epoch": 0.28,
"learning_rate": 3.5845131663736635e-05,
"loss": 3.0726,
"step": 90500
},
{
"epoch": 0.28,
"learning_rate": 3.576692797127109e-05,
"loss": 3.0725,
"step": 91000
},
{
"epoch": 0.29,
"learning_rate": 3.568872427880555e-05,
"loss": 3.0681,
"step": 91500
},
{
"epoch": 0.29,
"learning_rate": 3.5610520586340006e-05,
"loss": 3.0711,
"step": 92000
},
{
"epoch": 0.29,
"learning_rate": 3.553231689387446e-05,
"loss": 3.0798,
"step": 92500
},
{
"epoch": 0.29,
"learning_rate": 3.5454113201408916e-05,
"loss": 3.07,
"step": 93000
},
{
"epoch": 0.29,
"learning_rate": 3.537590950894338e-05,
"loss": 3.0729,
"step": 93500
},
{
"epoch": 0.29,
"learning_rate": 3.529770581647783e-05,
"loss": 3.0668,
"step": 94000
},
{
"epoch": 0.3,
"learning_rate": 3.521950212401229e-05,
"loss": 3.0722,
"step": 94500
},
{
"epoch": 0.3,
"learning_rate": 3.514129843154674e-05,
"loss": 3.0681,
"step": 95000
},
{
"epoch": 0.3,
"learning_rate": 3.5063094739081204e-05,
"loss": 3.0712,
"step": 95500
},
{
"epoch": 0.3,
"learning_rate": 3.498489104661566e-05,
"loss": 3.0684,
"step": 96000
},
{
"epoch": 0.3,
"learning_rate": 3.490668735415011e-05,
"loss": 3.0683,
"step": 96500
},
{
"epoch": 0.3,
"learning_rate": 3.4828483661684575e-05,
"loss": 3.0669,
"step": 97000
},
{
"epoch": 0.3,
"learning_rate": 3.475027996921903e-05,
"loss": 3.0587,
"step": 97500
},
{
"epoch": 0.31,
"learning_rate": 3.4672076276753484e-05,
"loss": 3.0624,
"step": 98000
},
{
"epoch": 0.31,
"learning_rate": 3.459387258428794e-05,
"loss": 3.0589,
"step": 98500
},
{
"epoch": 0.31,
"learning_rate": 3.45156688918224e-05,
"loss": 3.056,
"step": 99000
},
{
"epoch": 0.31,
"learning_rate": 3.4437465199356856e-05,
"loss": 3.0616,
"step": 99500
},
{
"epoch": 0.31,
"learning_rate": 3.435926150689131e-05,
"loss": 3.0555,
"step": 100000
},
{
"epoch": 0.31,
"eval_accuracy": 0.41383883246085473,
"eval_loss": 2.9919075965881348,
"eval_runtime": 5973.2974,
"eval_samples_per_second": 90.12,
"eval_steps_per_second": 1.408,
"step": 100000
},
{
"epoch": 0.31,
"learning_rate": 3.4281057814425765e-05,
"loss": 3.0601,
"step": 100500
},
{
"epoch": 0.32,
"learning_rate": 3.420285412196023e-05,
"loss": 3.0567,
"step": 101000
},
{
"epoch": 0.32,
"learning_rate": 3.412465042949468e-05,
"loss": 3.0644,
"step": 101500
},
{
"epoch": 0.32,
"learning_rate": 3.4046446737029137e-05,
"loss": 3.0515,
"step": 102000
},
{
"epoch": 0.32,
"learning_rate": 3.396824304456359e-05,
"loss": 3.0567,
"step": 102500
},
{
"epoch": 0.32,
"learning_rate": 3.389003935209805e-05,
"loss": 3.0588,
"step": 103000
},
{
"epoch": 0.32,
"learning_rate": 3.381183565963251e-05,
"loss": 3.0492,
"step": 103500
},
{
"epoch": 0.33,
"learning_rate": 3.373363196716696e-05,
"loss": 3.0471,
"step": 104000
},
{
"epoch": 0.33,
"learning_rate": 3.3655428274701424e-05,
"loss": 3.0544,
"step": 104500
},
{
"epoch": 0.33,
"learning_rate": 3.357722458223588e-05,
"loss": 3.0546,
"step": 105000
},
{
"epoch": 0.33,
"learning_rate": 3.3499020889770334e-05,
"loss": 3.0469,
"step": 105500
},
{
"epoch": 0.33,
"learning_rate": 3.342081719730479e-05,
"loss": 3.0514,
"step": 106000
},
{
"epoch": 0.33,
"learning_rate": 3.334261350483925e-05,
"loss": 3.0494,
"step": 106500
},
{
"epoch": 0.33,
"learning_rate": 3.3264409812373705e-05,
"loss": 3.0522,
"step": 107000
},
{
"epoch": 0.34,
"learning_rate": 3.318620611990816e-05,
"loss": 3.0447,
"step": 107500
},
{
"epoch": 0.34,
"learning_rate": 3.3108002427442615e-05,
"loss": 3.0454,
"step": 108000
},
{
"epoch": 0.34,
"learning_rate": 3.3029798734977076e-05,
"loss": 3.0495,
"step": 108500
},
{
"epoch": 0.34,
"learning_rate": 3.295159504251153e-05,
"loss": 3.0431,
"step": 109000
},
{
"epoch": 0.34,
"learning_rate": 3.2873391350045986e-05,
"loss": 3.038,
"step": 109500
},
{
"epoch": 0.34,
"learning_rate": 3.279518765758044e-05,
"loss": 3.038,
"step": 110000
},
{
"epoch": 0.34,
"eval_accuracy": 0.41561401879437854,
"eval_loss": 2.9786083698272705,
"eval_runtime": 6001.372,
"eval_samples_per_second": 89.698,
"eval_steps_per_second": 1.402,
"step": 110000
},
{
"epoch": 0.35,
"learning_rate": 3.27169839651149e-05,
"loss": 3.0403,
"step": 110500
},
{
"epoch": 0.35,
"learning_rate": 3.263878027264936e-05,
"loss": 3.0437,
"step": 111000
},
{
"epoch": 0.35,
"learning_rate": 3.256057658018381e-05,
"loss": 3.0337,
"step": 111500
},
{
"epoch": 0.35,
"learning_rate": 3.248237288771827e-05,
"loss": 3.0308,
"step": 112000
},
{
"epoch": 0.35,
"learning_rate": 3.240416919525273e-05,
"loss": 3.0356,
"step": 112500
},
{
"epoch": 0.35,
"learning_rate": 3.232596550278718e-05,
"loss": 3.0378,
"step": 113000
},
{
"epoch": 0.36,
"learning_rate": 3.224776181032164e-05,
"loss": 3.0383,
"step": 113500
},
{
"epoch": 0.36,
"learning_rate": 3.21695581178561e-05,
"loss": 3.0391,
"step": 114000
},
{
"epoch": 0.36,
"learning_rate": 3.2091354425390554e-05,
"loss": 3.0335,
"step": 114500
},
{
"epoch": 0.36,
"learning_rate": 3.201315073292501e-05,
"loss": 3.0356,
"step": 115000
},
{
"epoch": 0.36,
"learning_rate": 3.1934947040459464e-05,
"loss": 3.0354,
"step": 115500
},
{
"epoch": 0.36,
"learning_rate": 3.1856743347993925e-05,
"loss": 3.0346,
"step": 116000
},
{
"epoch": 0.36,
"learning_rate": 3.177853965552838e-05,
"loss": 3.0311,
"step": 116500
},
{
"epoch": 0.37,
"learning_rate": 3.1700335963062835e-05,
"loss": 3.0334,
"step": 117000
},
{
"epoch": 0.37,
"learning_rate": 3.162213227059729e-05,
"loss": 3.0225,
"step": 117500
},
{
"epoch": 0.37,
"learning_rate": 3.154392857813175e-05,
"loss": 3.0252,
"step": 118000
},
{
"epoch": 0.37,
"learning_rate": 3.1465724885666206e-05,
"loss": 3.0335,
"step": 118500
},
{
"epoch": 0.37,
"learning_rate": 3.138752119320066e-05,
"loss": 3.03,
"step": 119000
},
{
"epoch": 0.37,
"learning_rate": 3.1309317500735116e-05,
"loss": 3.0303,
"step": 119500
},
{
"epoch": 0.38,
"learning_rate": 3.123111380826958e-05,
"loss": 3.0291,
"step": 120000
},
{
"epoch": 0.38,
"eval_accuracy": 0.41712760392004566,
"eval_loss": 2.9651412963867188,
"eval_runtime": 5971.0497,
"eval_samples_per_second": 90.154,
"eval_steps_per_second": 1.409,
"step": 120000
},
{
"epoch": 0.38,
"learning_rate": 3.115291011580403e-05,
"loss": 3.0349,
"step": 120500
},
{
"epoch": 0.38,
"learning_rate": 3.107470642333849e-05,
"loss": 3.0318,
"step": 121000
},
{
"epoch": 0.38,
"learning_rate": 3.099650273087294e-05,
"loss": 3.0156,
"step": 121500
},
{
"epoch": 0.38,
"learning_rate": 3.09182990384074e-05,
"loss": 3.0226,
"step": 122000
},
{
"epoch": 0.38,
"learning_rate": 3.084009534594186e-05,
"loss": 3.0269,
"step": 122500
},
{
"epoch": 0.38,
"learning_rate": 3.076189165347631e-05,
"loss": 3.0239,
"step": 123000
},
{
"epoch": 0.39,
"learning_rate": 3.068368796101077e-05,
"loss": 3.0242,
"step": 123500
},
{
"epoch": 0.39,
"learning_rate": 3.060548426854522e-05,
"loss": 3.0267,
"step": 124000
},
{
"epoch": 0.39,
"learning_rate": 3.052728057607968e-05,
"loss": 3.0167,
"step": 124500
},
{
"epoch": 0.39,
"learning_rate": 3.0449076883614136e-05,
"loss": 3.0179,
"step": 125000
},
{
"epoch": 0.39,
"learning_rate": 3.0370873191148597e-05,
"loss": 3.0145,
"step": 125500
},
{
"epoch": 0.39,
"learning_rate": 3.0292669498683052e-05,
"loss": 3.0234,
"step": 126000
},
{
"epoch": 0.4,
"learning_rate": 3.0214465806217507e-05,
"loss": 3.0254,
"step": 126500
},
{
"epoch": 0.4,
"learning_rate": 3.0136262113751962e-05,
"loss": 3.0209,
"step": 127000
},
{
"epoch": 0.4,
"learning_rate": 3.0058058421286423e-05,
"loss": 3.0188,
"step": 127500
},
{
"epoch": 0.4,
"learning_rate": 2.9979854728820878e-05,
"loss": 3.0142,
"step": 128000
},
{
"epoch": 0.4,
"learning_rate": 2.9901651036355333e-05,
"loss": 3.0214,
"step": 128500
},
{
"epoch": 0.4,
"learning_rate": 2.9823447343889795e-05,
"loss": 3.0254,
"step": 129000
},
{
"epoch": 0.41,
"learning_rate": 2.974524365142425e-05,
"loss": 3.0214,
"step": 129500
},
{
"epoch": 0.41,
"learning_rate": 2.9667039958958704e-05,
"loss": 3.0182,
"step": 130000
},
{
"epoch": 0.41,
"eval_accuracy": 0.4190574389830463,
"eval_loss": 2.949920415878296,
"eval_runtime": 5975.917,
"eval_samples_per_second": 90.08,
"eval_steps_per_second": 1.408,
"step": 130000
},
{
"epoch": 0.41,
"learning_rate": 2.958883626649316e-05,
"loss": 3.0109,
"step": 130500
},
{
"epoch": 0.41,
"learning_rate": 2.951063257402762e-05,
"loss": 3.0117,
"step": 131000
},
{
"epoch": 0.41,
"learning_rate": 2.9432428881562075e-05,
"loss": 3.0183,
"step": 131500
},
{
"epoch": 0.41,
"learning_rate": 2.935422518909653e-05,
"loss": 3.0173,
"step": 132000
},
{
"epoch": 0.41,
"learning_rate": 2.9276021496630985e-05,
"loss": 3.0113,
"step": 132500
},
{
"epoch": 0.42,
"learning_rate": 2.9197817804165443e-05,
"loss": 3.0125,
"step": 133000
},
{
"epoch": 0.42,
"learning_rate": 2.91196141116999e-05,
"loss": 3.0126,
"step": 133500
},
{
"epoch": 0.42,
"learning_rate": 2.9041410419234356e-05,
"loss": 3.0149,
"step": 134000
},
{
"epoch": 0.42,
"learning_rate": 2.896320672676881e-05,
"loss": 3.0148,
"step": 134500
},
{
"epoch": 0.42,
"learning_rate": 2.888500303430327e-05,
"loss": 3.0172,
"step": 135000
},
{
"epoch": 0.42,
"learning_rate": 2.8806799341837724e-05,
"loss": 3.0114,
"step": 135500
},
{
"epoch": 0.43,
"learning_rate": 2.8728595649372182e-05,
"loss": 3.0123,
"step": 136000
},
{
"epoch": 0.43,
"learning_rate": 2.8650391956906637e-05,
"loss": 3.0065,
"step": 136500
},
{
"epoch": 0.43,
"learning_rate": 2.8572188264441095e-05,
"loss": 3.0085,
"step": 137000
},
{
"epoch": 0.43,
"learning_rate": 2.849398457197555e-05,
"loss": 3.0126,
"step": 137500
},
{
"epoch": 0.43,
"learning_rate": 2.8415780879510005e-05,
"loss": 3.0092,
"step": 138000
},
{
"epoch": 0.43,
"learning_rate": 2.8337577187044467e-05,
"loss": 3.0072,
"step": 138500
},
{
"epoch": 0.43,
"learning_rate": 2.825937349457892e-05,
"loss": 3.0088,
"step": 139000
},
{
"epoch": 0.44,
"learning_rate": 2.8181169802113376e-05,
"loss": 3.0117,
"step": 139500
},
{
"epoch": 0.44,
"learning_rate": 2.810296610964783e-05,
"loss": 3.0145,
"step": 140000
},
{
"epoch": 0.44,
"eval_accuracy": 0.42045388991709604,
"eval_loss": 2.93807053565979,
"eval_runtime": 6002.1748,
"eval_samples_per_second": 89.686,
"eval_steps_per_second": 1.401,
"step": 140000
},
{
"epoch": 0.44,
"learning_rate": 2.8024762417182293e-05,
"loss": 3.0022,
"step": 140500
},
{
"epoch": 0.44,
"learning_rate": 2.7946558724716747e-05,
"loss": 3.0051,
"step": 141000
},
{
"epoch": 0.44,
"learning_rate": 2.7868355032251202e-05,
"loss": 2.9975,
"step": 141500
},
{
"epoch": 0.44,
"learning_rate": 2.7790151339785657e-05,
"loss": 3.0062,
"step": 142000
},
{
"epoch": 0.45,
"learning_rate": 2.771194764732012e-05,
"loss": 3.0083,
"step": 142500
},
{
"epoch": 0.45,
"learning_rate": 2.7633743954854573e-05,
"loss": 3.0046,
"step": 143000
},
{
"epoch": 0.45,
"learning_rate": 2.7555540262389028e-05,
"loss": 3.0013,
"step": 143500
},
{
"epoch": 0.45,
"learning_rate": 2.7477336569923483e-05,
"loss": 2.9973,
"step": 144000
},
{
"epoch": 0.45,
"learning_rate": 2.7399132877457945e-05,
"loss": 2.9979,
"step": 144500
},
{
"epoch": 0.45,
"learning_rate": 2.73209291849924e-05,
"loss": 3.0,
"step": 145000
},
{
"epoch": 0.46,
"learning_rate": 2.7242725492526854e-05,
"loss": 3.0043,
"step": 145500
},
{
"epoch": 0.46,
"learning_rate": 2.7164521800061316e-05,
"loss": 2.9987,
"step": 146000
},
{
"epoch": 0.46,
"learning_rate": 2.708631810759577e-05,
"loss": 2.9951,
"step": 146500
},
{
"epoch": 0.46,
"learning_rate": 2.7008114415130226e-05,
"loss": 2.9972,
"step": 147000
},
{
"epoch": 0.46,
"learning_rate": 2.692991072266468e-05,
"loss": 2.9968,
"step": 147500
},
{
"epoch": 0.46,
"learning_rate": 2.6851707030199142e-05,
"loss": 2.995,
"step": 148000
},
{
"epoch": 0.46,
"learning_rate": 2.6773503337733597e-05,
"loss": 2.9944,
"step": 148500
},
{
"epoch": 0.47,
"learning_rate": 2.669529964526805e-05,
"loss": 2.9944,
"step": 149000
},
{
"epoch": 0.47,
"learning_rate": 2.6617095952802506e-05,
"loss": 2.9975,
"step": 149500
},
{
"epoch": 0.47,
"learning_rate": 2.6538892260336968e-05,
"loss": 2.9891,
"step": 150000
},
{
"epoch": 0.47,
"eval_accuracy": 0.42188237683918567,
"eval_loss": 2.9272468090057373,
"eval_runtime": 5980.3151,
"eval_samples_per_second": 90.014,
"eval_steps_per_second": 1.407,
"step": 150000
},
{
"epoch": 0.47,
"learning_rate": 2.6460688567871423e-05,
"loss": 3.007,
"step": 150500
},
{
"epoch": 0.47,
"learning_rate": 2.6382484875405878e-05,
"loss": 2.9883,
"step": 151000
},
{
"epoch": 0.47,
"learning_rate": 2.6304281182940332e-05,
"loss": 2.9949,
"step": 151500
},
{
"epoch": 0.48,
"learning_rate": 2.6226077490474794e-05,
"loss": 2.994,
"step": 152000
},
{
"epoch": 0.48,
"learning_rate": 2.614787379800925e-05,
"loss": 2.9923,
"step": 152500
},
{
"epoch": 0.48,
"learning_rate": 2.6069670105543704e-05,
"loss": 2.9863,
"step": 153000
},
{
"epoch": 0.48,
"learning_rate": 2.5991466413078165e-05,
"loss": 2.9943,
"step": 153500
},
{
"epoch": 0.48,
"learning_rate": 2.591326272061262e-05,
"loss": 2.9937,
"step": 154000
},
{
"epoch": 0.48,
"learning_rate": 2.5835059028147075e-05,
"loss": 2.9929,
"step": 154500
},
{
"epoch": 0.48,
"learning_rate": 2.575685533568153e-05,
"loss": 2.9884,
"step": 155000
},
{
"epoch": 0.49,
"learning_rate": 2.567865164321599e-05,
"loss": 2.9874,
"step": 155500
},
{
"epoch": 0.49,
"learning_rate": 2.5600447950750446e-05,
"loss": 2.9952,
"step": 156000
},
{
"epoch": 0.49,
"learning_rate": 2.55222442582849e-05,
"loss": 2.9878,
"step": 156500
},
{
"epoch": 0.49,
"learning_rate": 2.5444040565819356e-05,
"loss": 2.9886,
"step": 157000
},
{
"epoch": 0.49,
"learning_rate": 2.5365836873353814e-05,
"loss": 2.9845,
"step": 157500
},
{
"epoch": 0.49,
"learning_rate": 2.5287633180888272e-05,
"loss": 2.993,
"step": 158000
},
{
"epoch": 0.5,
"learning_rate": 2.5209429488422727e-05,
"loss": 2.9941,
"step": 158500
},
{
"epoch": 0.5,
"learning_rate": 2.513122579595718e-05,
"loss": 2.9907,
"step": 159000
},
{
"epoch": 0.5,
"learning_rate": 2.505302210349164e-05,
"loss": 2.9851,
"step": 159500
},
{
"epoch": 0.5,
"learning_rate": 2.4974818411026095e-05,
"loss": 2.9836,
"step": 160000
},
{
"epoch": 0.5,
"eval_accuracy": 0.4229763453615049,
"eval_loss": 2.919142723083496,
"eval_runtime": 5984.1342,
"eval_samples_per_second": 89.957,
"eval_steps_per_second": 1.406,
"step": 160000
},
{
"epoch": 0.5,
"learning_rate": 2.4896614718560553e-05,
"loss": 2.9821,
"step": 160500
},
{
"epoch": 0.5,
"learning_rate": 2.4818411026095008e-05,
"loss": 2.9832,
"step": 161000
},
{
"epoch": 0.51,
"learning_rate": 2.4740207333629466e-05,
"loss": 2.9875,
"step": 161500
},
{
"epoch": 0.51,
"learning_rate": 2.466200364116392e-05,
"loss": 2.9932,
"step": 162000
},
{
"epoch": 0.51,
"learning_rate": 2.458379994869838e-05,
"loss": 2.9803,
"step": 162500
},
{
"epoch": 0.51,
"learning_rate": 2.4505596256232834e-05,
"loss": 2.9877,
"step": 163000
},
{
"epoch": 0.51,
"learning_rate": 2.4427392563767292e-05,
"loss": 2.9899,
"step": 163500
},
{
"epoch": 0.51,
"learning_rate": 2.4349188871301747e-05,
"loss": 2.9806,
"step": 164000
},
{
"epoch": 0.51,
"learning_rate": 2.4270985178836205e-05,
"loss": 2.9802,
"step": 164500
},
{
"epoch": 0.52,
"learning_rate": 2.419278148637066e-05,
"loss": 2.9822,
"step": 165000
},
{
"epoch": 0.52,
"learning_rate": 2.4114577793905118e-05,
"loss": 2.9787,
"step": 165500
},
{
"epoch": 0.52,
"learning_rate": 2.4036374101439573e-05,
"loss": 2.9782,
"step": 166000
},
{
"epoch": 0.52,
"learning_rate": 2.395817040897403e-05,
"loss": 2.9797,
"step": 166500
},
{
"epoch": 0.52,
"learning_rate": 2.3879966716508486e-05,
"loss": 2.9812,
"step": 167000
},
{
"epoch": 0.52,
"learning_rate": 2.3801763024042944e-05,
"loss": 2.9824,
"step": 167500
},
{
"epoch": 0.53,
"learning_rate": 2.37235593315774e-05,
"loss": 2.9816,
"step": 168000
},
{
"epoch": 0.53,
"learning_rate": 2.3645355639111857e-05,
"loss": 2.9752,
"step": 168500
},
{
"epoch": 0.53,
"learning_rate": 2.3567151946646315e-05,
"loss": 2.9867,
"step": 169000
},
{
"epoch": 0.53,
"learning_rate": 2.348894825418077e-05,
"loss": 2.9804,
"step": 169500
},
{
"epoch": 0.53,
"learning_rate": 2.3410744561715228e-05,
"loss": 2.9717,
"step": 170000
},
{
"epoch": 0.53,
"eval_accuracy": 0.4241124752924122,
"eval_loss": 2.9102535247802734,
"eval_runtime": 5992.1438,
"eval_samples_per_second": 89.836,
"eval_steps_per_second": 1.404,
"step": 170000
},
{
"epoch": 0.53,
"learning_rate": 2.3332540869249683e-05,
"loss": 2.9745,
"step": 170500
},
{
"epoch": 0.53,
"learning_rate": 2.325433717678414e-05,
"loss": 2.9788,
"step": 171000
},
{
"epoch": 0.54,
"learning_rate": 2.3176133484318596e-05,
"loss": 2.9801,
"step": 171500
},
{
"epoch": 0.54,
"learning_rate": 2.3097929791853054e-05,
"loss": 2.9712,
"step": 172000
},
{
"epoch": 0.54,
"learning_rate": 2.301972609938751e-05,
"loss": 2.9767,
"step": 172500
},
{
"epoch": 0.54,
"learning_rate": 2.2941522406921967e-05,
"loss": 2.9699,
"step": 173000
},
{
"epoch": 0.54,
"learning_rate": 2.2863318714456422e-05,
"loss": 2.9698,
"step": 173500
},
{
"epoch": 0.54,
"learning_rate": 2.278511502199088e-05,
"loss": 2.9742,
"step": 174000
},
{
"epoch": 0.55,
"learning_rate": 2.2706911329525335e-05,
"loss": 2.9816,
"step": 174500
},
{
"epoch": 0.55,
"learning_rate": 2.2628707637059793e-05,
"loss": 2.9703,
"step": 175000
},
{
"epoch": 0.55,
"learning_rate": 2.2550503944594248e-05,
"loss": 2.9699,
"step": 175500
},
{
"epoch": 0.55,
"learning_rate": 2.2472300252128706e-05,
"loss": 2.9715,
"step": 176000
},
{
"epoch": 0.55,
"learning_rate": 2.239409655966316e-05,
"loss": 2.9706,
"step": 176500
},
{
"epoch": 0.55,
"learning_rate": 2.231589286719762e-05,
"loss": 2.9689,
"step": 177000
},
{
"epoch": 0.56,
"learning_rate": 2.2237689174732077e-05,
"loss": 2.9728,
"step": 177500
},
{
"epoch": 0.56,
"learning_rate": 2.2159485482266532e-05,
"loss": 2.9702,
"step": 178000
},
{
"epoch": 0.56,
"learning_rate": 2.208128178980099e-05,
"loss": 2.9703,
"step": 178500
},
{
"epoch": 0.56,
"learning_rate": 2.2003078097335445e-05,
"loss": 2.967,
"step": 179000
},
{
"epoch": 0.56,
"learning_rate": 2.1924874404869904e-05,
"loss": 2.9689,
"step": 179500
},
{
"epoch": 0.56,
"learning_rate": 2.184667071240436e-05,
"loss": 2.9651,
"step": 180000
},
{
"epoch": 0.56,
"eval_accuracy": 0.4249954188646057,
"eval_loss": 2.9038655757904053,
"eval_runtime": 5968.0597,
"eval_samples_per_second": 90.199,
"eval_steps_per_second": 1.41,
"step": 180000
},
{
"epoch": 0.56,
"learning_rate": 2.1768467019938817e-05,
"loss": 2.9695,
"step": 180500
},
{
"epoch": 0.57,
"learning_rate": 2.169026332747327e-05,
"loss": 2.9705,
"step": 181000
},
{
"epoch": 0.57,
"learning_rate": 2.161205963500773e-05,
"loss": 2.9677,
"step": 181500
},
{
"epoch": 0.57,
"learning_rate": 2.1533855942542184e-05,
"loss": 2.9684,
"step": 182000
},
{
"epoch": 0.57,
"learning_rate": 2.1455652250076643e-05,
"loss": 2.9638,
"step": 182500
},
{
"epoch": 0.57,
"learning_rate": 2.1377448557611097e-05,
"loss": 2.9618,
"step": 183000
},
{
"epoch": 0.57,
"learning_rate": 2.1299244865145556e-05,
"loss": 2.9713,
"step": 183500
},
{
"epoch": 0.58,
"learning_rate": 2.122104117268001e-05,
"loss": 2.9715,
"step": 184000
},
{
"epoch": 0.58,
"learning_rate": 2.1142837480214465e-05,
"loss": 2.963,
"step": 184500
},
{
"epoch": 0.58,
"learning_rate": 2.1064633787748923e-05,
"loss": 2.9695,
"step": 185000
},
{
"epoch": 0.58,
"learning_rate": 2.0986430095283378e-05,
"loss": 2.9593,
"step": 185500
},
{
"epoch": 0.58,
"learning_rate": 2.0908226402817836e-05,
"loss": 2.9695,
"step": 186000
},
{
"epoch": 0.58,
"learning_rate": 2.083002271035229e-05,
"loss": 2.9696,
"step": 186500
},
{
"epoch": 0.58,
"learning_rate": 2.075181901788675e-05,
"loss": 2.9624,
"step": 187000
},
{
"epoch": 0.59,
"learning_rate": 2.0673615325421204e-05,
"loss": 2.9576,
"step": 187500
},
{
"epoch": 0.59,
"learning_rate": 2.0595411632955662e-05,
"loss": 2.9655,
"step": 188000
},
{
"epoch": 0.59,
"learning_rate": 2.0517207940490117e-05,
"loss": 2.9667,
"step": 188500
},
{
"epoch": 0.59,
"learning_rate": 2.0439004248024575e-05,
"loss": 2.9581,
"step": 189000
},
{
"epoch": 0.59,
"learning_rate": 2.036080055555903e-05,
"loss": 2.9597,
"step": 189500
},
{
"epoch": 0.59,
"learning_rate": 2.028259686309349e-05,
"loss": 2.9615,
"step": 190000
},
{
"epoch": 0.59,
"eval_accuracy": 0.4257997033905501,
"eval_loss": 2.897066116333008,
"eval_runtime": 5974.8859,
"eval_samples_per_second": 90.096,
"eval_steps_per_second": 1.408,
"step": 190000
},
{
"epoch": 0.6,
"learning_rate": 2.0204393170627943e-05,
"loss": 2.9678,
"step": 190500
},
{
"epoch": 0.6,
"learning_rate": 2.01261894781624e-05,
"loss": 2.9542,
"step": 191000
},
{
"epoch": 0.6,
"learning_rate": 2.0047985785696856e-05,
"loss": 2.9594,
"step": 191500
},
{
"epoch": 0.6,
"learning_rate": 1.9969782093231314e-05,
"loss": 2.9661,
"step": 192000
},
{
"epoch": 0.6,
"learning_rate": 1.989157840076577e-05,
"loss": 2.9648,
"step": 192500
},
{
"epoch": 0.6,
"learning_rate": 1.9813374708300227e-05,
"loss": 2.9638,
"step": 193000
},
{
"epoch": 0.61,
"learning_rate": 1.9735171015834686e-05,
"loss": 2.9664,
"step": 193500
},
{
"epoch": 0.61,
"learning_rate": 1.965696732336914e-05,
"loss": 2.9585,
"step": 194000
},
{
"epoch": 0.61,
"learning_rate": 1.95787636309036e-05,
"loss": 2.9579,
"step": 194500
},
{
"epoch": 0.61,
"learning_rate": 1.9500559938438054e-05,
"loss": 2.9673,
"step": 195000
},
{
"epoch": 0.61,
"learning_rate": 1.9422356245972512e-05,
"loss": 2.953,
"step": 195500
},
{
"epoch": 0.61,
"learning_rate": 1.9344152553506967e-05,
"loss": 2.9615,
"step": 196000
},
{
"epoch": 0.61,
"learning_rate": 1.9265948861041425e-05,
"loss": 2.9599,
"step": 196500
},
{
"epoch": 0.62,
"learning_rate": 1.918774516857588e-05,
"loss": 2.9584,
"step": 197000
},
{
"epoch": 0.62,
"learning_rate": 1.9109541476110338e-05,
"loss": 2.9578,
"step": 197500
},
{
"epoch": 0.62,
"learning_rate": 1.9031337783644793e-05,
"loss": 2.9648,
"step": 198000
},
{
"epoch": 0.62,
"learning_rate": 1.895313409117925e-05,
"loss": 2.9576,
"step": 198500
},
{
"epoch": 0.62,
"learning_rate": 1.8874930398713706e-05,
"loss": 2.9555,
"step": 199000
},
{
"epoch": 0.62,
"learning_rate": 1.8796726706248164e-05,
"loss": 2.9612,
"step": 199500
},
{
"epoch": 0.63,
"learning_rate": 1.871852301378262e-05,
"loss": 2.9556,
"step": 200000
},
{
"epoch": 0.63,
"eval_accuracy": 0.42693775816826174,
"eval_loss": 2.8882317543029785,
"eval_runtime": 5989.3476,
"eval_samples_per_second": 89.878,
"eval_steps_per_second": 1.404,
"step": 200000
},
{
"epoch": 0.63,
"learning_rate": 1.8640319321317077e-05,
"loss": 2.957,
"step": 200500
},
{
"epoch": 0.63,
"learning_rate": 1.8562115628851535e-05,
"loss": 2.9636,
"step": 201000
},
{
"epoch": 0.63,
"learning_rate": 1.848391193638599e-05,
"loss": 2.9576,
"step": 201500
},
{
"epoch": 0.63,
"learning_rate": 1.8405708243920448e-05,
"loss": 2.9571,
"step": 202000
},
{
"epoch": 0.63,
"learning_rate": 1.8327504551454903e-05,
"loss": 2.9543,
"step": 202500
},
{
"epoch": 0.64,
"learning_rate": 1.824930085898936e-05,
"loss": 2.9551,
"step": 203000
},
{
"epoch": 0.64,
"learning_rate": 1.8171097166523816e-05,
"loss": 2.9454,
"step": 203500
},
{
"epoch": 0.64,
"learning_rate": 1.8092893474058274e-05,
"loss": 2.9542,
"step": 204000
},
{
"epoch": 0.64,
"learning_rate": 1.801468978159273e-05,
"loss": 2.9548,
"step": 204500
},
{
"epoch": 0.64,
"learning_rate": 1.7936486089127187e-05,
"loss": 2.9557,
"step": 205000
},
{
"epoch": 0.64,
"learning_rate": 1.7858282396661642e-05,
"loss": 2.9511,
"step": 205500
},
{
"epoch": 0.64,
"learning_rate": 1.77800787041961e-05,
"loss": 2.9551,
"step": 206000
},
{
"epoch": 0.65,
"learning_rate": 1.7701875011730555e-05,
"loss": 2.9415,
"step": 206500
},
{
"epoch": 0.65,
"learning_rate": 1.7623671319265013e-05,
"loss": 2.9552,
"step": 207000
},
{
"epoch": 0.65,
"learning_rate": 1.7545467626799468e-05,
"loss": 2.9497,
"step": 207500
},
{
"epoch": 0.65,
"learning_rate": 1.7467263934333926e-05,
"loss": 2.9488,
"step": 208000
},
{
"epoch": 0.65,
"learning_rate": 1.738906024186838e-05,
"loss": 2.9536,
"step": 208500
},
{
"epoch": 0.65,
"learning_rate": 1.731085654940284e-05,
"loss": 2.9552,
"step": 209000
},
{
"epoch": 0.66,
"learning_rate": 1.7232652856937294e-05,
"loss": 2.9523,
"step": 209500
},
{
"epoch": 0.66,
"learning_rate": 1.715444916447175e-05,
"loss": 2.9452,
"step": 210000
},
{
"epoch": 0.66,
"eval_accuracy": 0.4276823016234361,
"eval_loss": 2.8825109004974365,
"eval_runtime": 5967.0534,
"eval_samples_per_second": 90.214,
"eval_steps_per_second": 1.41,
"step": 210000
},
{
"epoch": 0.66,
"learning_rate": 1.7076245472006207e-05,
"loss": 2.95,
"step": 210500
},
{
"epoch": 0.66,
"learning_rate": 1.6998041779540662e-05,
"loss": 2.9454,
"step": 211000
},
{
"epoch": 0.66,
"learning_rate": 1.691983808707512e-05,
"loss": 2.9505,
"step": 211500
},
{
"epoch": 0.66,
"learning_rate": 1.6841634394609575e-05,
"loss": 2.9475,
"step": 212000
},
{
"epoch": 0.66,
"learning_rate": 1.6763430702144033e-05,
"loss": 2.9538,
"step": 212500
},
{
"epoch": 0.67,
"learning_rate": 1.6685227009678488e-05,
"loss": 2.9508,
"step": 213000
},
{
"epoch": 0.67,
"learning_rate": 1.6607023317212946e-05,
"loss": 2.9519,
"step": 213500
},
{
"epoch": 0.67,
"learning_rate": 1.65288196247474e-05,
"loss": 2.9447,
"step": 214000
},
{
"epoch": 0.67,
"learning_rate": 1.645061593228186e-05,
"loss": 2.9466,
"step": 214500
},
{
"epoch": 0.67,
"learning_rate": 1.6372412239816314e-05,
"loss": 2.9437,
"step": 215000
},
{
"epoch": 0.67,
"learning_rate": 1.6294208547350772e-05,
"loss": 2.9471,
"step": 215500
},
{
"epoch": 0.68,
"learning_rate": 1.6216004854885227e-05,
"loss": 2.952,
"step": 216000
},
{
"epoch": 0.68,
"learning_rate": 1.6137801162419685e-05,
"loss": 2.9465,
"step": 216500
},
{
"epoch": 0.68,
"learning_rate": 1.6059597469954143e-05,
"loss": 2.9418,
"step": 217000
},
{
"epoch": 0.68,
"learning_rate": 1.5981393777488598e-05,
"loss": 2.9459,
"step": 217500
},
{
"epoch": 0.68,
"learning_rate": 1.5903190085023056e-05,
"loss": 2.9403,
"step": 218000
},
{
"epoch": 0.68,
"learning_rate": 1.582498639255751e-05,
"loss": 2.9474,
"step": 218500
},
{
"epoch": 0.69,
"learning_rate": 1.574678270009197e-05,
"loss": 2.9403,
"step": 219000
},
{
"epoch": 0.69,
"learning_rate": 1.5668579007626424e-05,
"loss": 2.9438,
"step": 219500
},
{
"epoch": 0.69,
"learning_rate": 1.5590375315160882e-05,
"loss": 2.9412,
"step": 220000
},
{
"epoch": 0.69,
"eval_accuracy": 0.42842816160118896,
"eval_loss": 2.8766274452209473,
"eval_runtime": 5967.0165,
"eval_samples_per_second": 90.215,
"eval_steps_per_second": 1.41,
"step": 220000
},
{
"epoch": 0.69,
"learning_rate": 1.5512171622695337e-05,
"loss": 2.9422,
"step": 220500
},
{
"epoch": 0.69,
"learning_rate": 1.5433967930229795e-05,
"loss": 2.9483,
"step": 221000
},
{
"epoch": 0.69,
"learning_rate": 1.535576423776425e-05,
"loss": 2.9449,
"step": 221500
},
{
"epoch": 0.69,
"learning_rate": 1.5277560545298708e-05,
"loss": 2.9443,
"step": 222000
},
{
"epoch": 0.7,
"learning_rate": 1.5199356852833163e-05,
"loss": 2.9446,
"step": 222500
},
{
"epoch": 0.7,
"learning_rate": 1.5121153160367621e-05,
"loss": 2.9383,
"step": 223000
},
{
"epoch": 0.7,
"learning_rate": 1.5042949467902076e-05,
"loss": 2.9339,
"step": 223500
},
{
"epoch": 0.7,
"learning_rate": 1.4964745775436534e-05,
"loss": 2.9406,
"step": 224000
},
{
"epoch": 0.7,
"learning_rate": 1.4886542082970989e-05,
"loss": 2.9334,
"step": 224500
},
{
"epoch": 0.7,
"learning_rate": 1.4808338390505447e-05,
"loss": 2.9424,
"step": 225000
},
{
"epoch": 0.71,
"learning_rate": 1.4730134698039904e-05,
"loss": 2.9418,
"step": 225500
},
{
"epoch": 0.71,
"learning_rate": 1.465193100557436e-05,
"loss": 2.9401,
"step": 226000
},
{
"epoch": 0.71,
"learning_rate": 1.4573727313108817e-05,
"loss": 2.9335,
"step": 226500
},
{
"epoch": 0.71,
"learning_rate": 1.4495523620643273e-05,
"loss": 2.9357,
"step": 227000
},
{
"epoch": 0.71,
"learning_rate": 1.441731992817773e-05,
"loss": 2.9373,
"step": 227500
},
{
"epoch": 0.71,
"learning_rate": 1.4339116235712186e-05,
"loss": 2.9392,
"step": 228000
},
{
"epoch": 0.71,
"learning_rate": 1.4260912543246643e-05,
"loss": 2.937,
"step": 228500
},
{
"epoch": 0.72,
"learning_rate": 1.4182708850781098e-05,
"loss": 2.9427,
"step": 229000
},
{
"epoch": 0.72,
"learning_rate": 1.4104505158315556e-05,
"loss": 2.9348,
"step": 229500
},
{
"epoch": 0.72,
"learning_rate": 1.402630146585001e-05,
"loss": 2.9402,
"step": 230000
},
{
"epoch": 0.72,
"eval_accuracy": 0.42901379624141195,
"eval_loss": 2.872159719467163,
"eval_runtime": 6001.9708,
"eval_samples_per_second": 89.689,
"eval_steps_per_second": 1.402,
"step": 230000
},
{
"epoch": 0.72,
"learning_rate": 1.3948097773384469e-05,
"loss": 2.9363,
"step": 230500
},
{
"epoch": 0.72,
"learning_rate": 1.3869894080918924e-05,
"loss": 2.9372,
"step": 231000
},
{
"epoch": 0.72,
"learning_rate": 1.3791690388453382e-05,
"loss": 2.9384,
"step": 231500
},
{
"epoch": 0.73,
"learning_rate": 1.3713486695987837e-05,
"loss": 2.9363,
"step": 232000
},
{
"epoch": 0.73,
"learning_rate": 1.3635283003522295e-05,
"loss": 2.9368,
"step": 232500
},
{
"epoch": 0.73,
"learning_rate": 1.3557079311056753e-05,
"loss": 2.9436,
"step": 233000
},
{
"epoch": 0.73,
"learning_rate": 1.3478875618591208e-05,
"loss": 2.9427,
"step": 233500
},
{
"epoch": 0.73,
"learning_rate": 1.3400671926125666e-05,
"loss": 2.9359,
"step": 234000
},
{
"epoch": 0.73,
"learning_rate": 1.3322468233660121e-05,
"loss": 2.9371,
"step": 234500
},
{
"epoch": 0.74,
"learning_rate": 1.3244264541194579e-05,
"loss": 2.9314,
"step": 235000
},
{
"epoch": 0.74,
"learning_rate": 1.3166060848729034e-05,
"loss": 2.9339,
"step": 235500
},
{
"epoch": 0.74,
"learning_rate": 1.3087857156263492e-05,
"loss": 2.9305,
"step": 236000
},
{
"epoch": 0.74,
"learning_rate": 1.3009653463797947e-05,
"loss": 2.9341,
"step": 236500
},
{
"epoch": 0.74,
"learning_rate": 1.2931449771332405e-05,
"loss": 2.9356,
"step": 237000
},
{
"epoch": 0.74,
"learning_rate": 1.285324607886686e-05,
"loss": 2.9354,
"step": 237500
},
{
"epoch": 0.74,
"learning_rate": 1.2775042386401318e-05,
"loss": 2.9408,
"step": 238000
},
{
"epoch": 0.75,
"learning_rate": 1.2696838693935773e-05,
"loss": 2.9357,
"step": 238500
},
{
"epoch": 0.75,
"learning_rate": 1.2618635001470231e-05,
"loss": 2.9361,
"step": 239000
},
{
"epoch": 0.75,
"learning_rate": 1.2540431309004686e-05,
"loss": 2.9313,
"step": 239500
},
{
"epoch": 0.75,
"learning_rate": 1.2462227616539143e-05,
"loss": 2.9299,
"step": 240000
},
{
"epoch": 0.75,
"eval_accuracy": 0.42963944953623323,
"eval_loss": 2.867478847503662,
"eval_runtime": 5981.9362,
"eval_samples_per_second": 89.99,
"eval_steps_per_second": 1.406,
"step": 240000
},
{
"epoch": 0.75,
"learning_rate": 1.2384023924073599e-05,
"loss": 2.9345,
"step": 240500
},
{
"epoch": 0.75,
"learning_rate": 1.2305820231608056e-05,
"loss": 2.9319,
"step": 241000
},
{
"epoch": 0.76,
"learning_rate": 1.2227616539142512e-05,
"loss": 2.934,
"step": 241500
},
{
"epoch": 0.76,
"learning_rate": 1.2149412846676969e-05,
"loss": 2.9321,
"step": 242000
},
{
"epoch": 0.76,
"learning_rate": 1.2071209154211425e-05,
"loss": 2.9301,
"step": 242500
},
{
"epoch": 0.76,
"learning_rate": 1.1993005461745882e-05,
"loss": 2.934,
"step": 243000
},
{
"epoch": 0.76,
"learning_rate": 1.1914801769280338e-05,
"loss": 2.9358,
"step": 243500
},
{
"epoch": 0.76,
"learning_rate": 1.1836598076814795e-05,
"loss": 2.9328,
"step": 244000
},
{
"epoch": 0.76,
"learning_rate": 1.1758394384349251e-05,
"loss": 2.9332,
"step": 244500
},
{
"epoch": 0.77,
"learning_rate": 1.1680190691883708e-05,
"loss": 2.9288,
"step": 245000
},
{
"epoch": 0.77,
"learning_rate": 1.1601986999418166e-05,
"loss": 2.9357,
"step": 245500
},
{
"epoch": 0.77,
"learning_rate": 1.1523783306952622e-05,
"loss": 2.9231,
"step": 246000
},
{
"epoch": 0.77,
"learning_rate": 1.1445579614487079e-05,
"loss": 2.9373,
"step": 246500
},
{
"epoch": 0.77,
"learning_rate": 1.1367375922021535e-05,
"loss": 2.9313,
"step": 247000
},
{
"epoch": 0.77,
"learning_rate": 1.1289172229555992e-05,
"loss": 2.9398,
"step": 247500
},
{
"epoch": 0.78,
"learning_rate": 1.1210968537090448e-05,
"loss": 2.9304,
"step": 248000
},
{
"epoch": 0.78,
"learning_rate": 1.1132764844624905e-05,
"loss": 2.9308,
"step": 248500
},
{
"epoch": 0.78,
"learning_rate": 1.1054561152159361e-05,
"loss": 2.9292,
"step": 249000
},
{
"epoch": 0.78,
"learning_rate": 1.0976357459693818e-05,
"loss": 2.9204,
"step": 249500
},
{
"epoch": 0.78,
"learning_rate": 1.0898153767228274e-05,
"loss": 2.9302,
"step": 250000
},
{
"epoch": 0.78,
"eval_accuracy": 0.43037667130997825,
"eval_loss": 2.862294912338257,
"eval_runtime": 5973.5209,
"eval_samples_per_second": 90.116,
"eval_steps_per_second": 1.408,
"step": 250000
},
{
"epoch": 0.78,
"learning_rate": 1.081995007476273e-05,
"loss": 2.929,
"step": 250500
},
{
"epoch": 0.79,
"learning_rate": 1.0741746382297187e-05,
"loss": 2.9253,
"step": 251000
},
{
"epoch": 0.79,
"learning_rate": 1.0663542689831644e-05,
"loss": 2.9267,
"step": 251500
},
{
"epoch": 0.79,
"learning_rate": 1.05853389973661e-05,
"loss": 2.9289,
"step": 252000
},
{
"epoch": 0.79,
"learning_rate": 1.0507135304900557e-05,
"loss": 2.9273,
"step": 252500
},
{
"epoch": 0.79,
"learning_rate": 1.0428931612435013e-05,
"loss": 2.9326,
"step": 253000
},
{
"epoch": 0.79,
"learning_rate": 1.035072791996947e-05,
"loss": 2.934,
"step": 253500
},
{
"epoch": 0.79,
"learning_rate": 1.0272524227503926e-05,
"loss": 2.9263,
"step": 254000
},
{
"epoch": 0.8,
"learning_rate": 1.0194320535038383e-05,
"loss": 2.9304,
"step": 254500
},
{
"epoch": 0.8,
"learning_rate": 1.011611684257284e-05,
"loss": 2.9256,
"step": 255000
},
{
"epoch": 0.8,
"learning_rate": 1.0037913150107296e-05,
"loss": 2.9208,
"step": 255500
},
{
"epoch": 0.8,
"learning_rate": 9.959709457641752e-06,
"loss": 2.9256,
"step": 256000
},
{
"epoch": 0.8,
"learning_rate": 9.881505765176209e-06,
"loss": 2.9266,
"step": 256500
},
{
"epoch": 0.8,
"learning_rate": 9.803302072710665e-06,
"loss": 2.9238,
"step": 257000
},
{
"epoch": 0.81,
"learning_rate": 9.725098380245122e-06,
"loss": 2.924,
"step": 257500
},
{
"epoch": 0.81,
"learning_rate": 9.646894687779578e-06,
"loss": 2.9212,
"step": 258000
},
{
"epoch": 0.81,
"learning_rate": 9.568690995314035e-06,
"loss": 2.9224,
"step": 258500
},
{
"epoch": 0.81,
"learning_rate": 9.490487302848491e-06,
"loss": 2.9273,
"step": 259000
},
{
"epoch": 0.81,
"learning_rate": 9.412283610382948e-06,
"loss": 2.9249,
"step": 259500
},
{
"epoch": 0.81,
"learning_rate": 9.334079917917404e-06,
"loss": 2.9165,
"step": 260000
},
{
"epoch": 0.81,
"eval_accuracy": 0.4308058122732213,
"eval_loss": 2.858468532562256,
"eval_runtime": 5995.5219,
"eval_samples_per_second": 89.786,
"eval_steps_per_second": 1.403,
"step": 260000
},
{
"epoch": 0.81,
"learning_rate": 9.255876225451861e-06,
"loss": 2.9231,
"step": 260500
},
{
"epoch": 0.82,
"learning_rate": 9.177672532986317e-06,
"loss": 2.9306,
"step": 261000
},
{
"epoch": 0.82,
"learning_rate": 9.099468840520776e-06,
"loss": 2.9226,
"step": 261500
},
{
"epoch": 0.82,
"learning_rate": 9.021265148055232e-06,
"loss": 2.9248,
"step": 262000
},
{
"epoch": 0.82,
"learning_rate": 8.943061455589689e-06,
"loss": 2.9246,
"step": 262500
},
{
"epoch": 0.82,
"learning_rate": 8.864857763124145e-06,
"loss": 2.9206,
"step": 263000
},
{
"epoch": 0.82,
"learning_rate": 8.786654070658602e-06,
"loss": 2.9235,
"step": 263500
},
{
"epoch": 0.83,
"learning_rate": 8.708450378193058e-06,
"loss": 2.918,
"step": 264000
},
{
"epoch": 0.83,
"learning_rate": 8.630246685727513e-06,
"loss": 2.9265,
"step": 264500
},
{
"epoch": 0.83,
"learning_rate": 8.55204299326197e-06,
"loss": 2.924,
"step": 265000
},
{
"epoch": 0.83,
"learning_rate": 8.473839300796426e-06,
"loss": 2.9226,
"step": 265500
},
{
"epoch": 0.83,
"learning_rate": 8.395635608330883e-06,
"loss": 2.9215,
"step": 266000
},
{
"epoch": 0.83,
"learning_rate": 8.317431915865339e-06,
"loss": 2.9177,
"step": 266500
},
{
"epoch": 0.84,
"learning_rate": 8.239228223399796e-06,
"loss": 2.9188,
"step": 267000
},
{
"epoch": 0.84,
"learning_rate": 8.161024530934252e-06,
"loss": 2.9306,
"step": 267500
},
{
"epoch": 0.84,
"learning_rate": 8.082820838468709e-06,
"loss": 2.9158,
"step": 268000
},
{
"epoch": 0.84,
"learning_rate": 8.004617146003165e-06,
"loss": 2.9232,
"step": 268500
},
{
"epoch": 0.84,
"learning_rate": 7.926413453537622e-06,
"loss": 2.919,
"step": 269000
},
{
"epoch": 0.84,
"learning_rate": 7.84820976107208e-06,
"loss": 2.9246,
"step": 269500
},
{
"epoch": 0.84,
"learning_rate": 7.770006068606536e-06,
"loss": 2.915,
"step": 270000
},
{
"epoch": 0.84,
"eval_accuracy": 0.4314090792365293,
"eval_loss": 2.8536534309387207,
"eval_runtime": 5987.8082,
"eval_samples_per_second": 89.901,
"eval_steps_per_second": 1.405,
"step": 270000
},
{
"epoch": 0.85,
"learning_rate": 7.691802376140993e-06,
"loss": 2.9202,
"step": 270500
},
{
"epoch": 0.85,
"learning_rate": 7.613598683675449e-06,
"loss": 2.9281,
"step": 271000
},
{
"epoch": 0.85,
"learning_rate": 7.535394991209906e-06,
"loss": 2.921,
"step": 271500
},
{
"epoch": 0.85,
"learning_rate": 7.457191298744362e-06,
"loss": 2.9163,
"step": 272000
},
{
"epoch": 0.85,
"learning_rate": 7.378987606278819e-06,
"loss": 2.9255,
"step": 272500
},
{
"epoch": 0.85,
"learning_rate": 7.300783913813275e-06,
"loss": 2.9185,
"step": 273000
},
{
"epoch": 0.86,
"learning_rate": 7.222580221347732e-06,
"loss": 2.9219,
"step": 273500
},
{
"epoch": 0.86,
"learning_rate": 7.1443765288821875e-06,
"loss": 2.9281,
"step": 274000
},
{
"epoch": 0.86,
"learning_rate": 7.066172836416644e-06,
"loss": 2.9228,
"step": 274500
},
{
"epoch": 0.86,
"learning_rate": 6.9879691439511005e-06,
"loss": 2.9313,
"step": 275000
},
{
"epoch": 0.86,
"learning_rate": 6.909765451485557e-06,
"loss": 2.9184,
"step": 275500
},
{
"epoch": 0.86,
"learning_rate": 6.8315617590200135e-06,
"loss": 2.9111,
"step": 276000
},
{
"epoch": 0.86,
"learning_rate": 6.75335806655447e-06,
"loss": 2.9214,
"step": 276500
},
{
"epoch": 0.87,
"learning_rate": 6.6751543740889265e-06,
"loss": 2.9167,
"step": 277000
},
{
"epoch": 0.87,
"learning_rate": 6.596950681623385e-06,
"loss": 2.9143,
"step": 277500
},
{
"epoch": 0.87,
"learning_rate": 6.518746989157841e-06,
"loss": 2.9185,
"step": 278000
},
{
"epoch": 0.87,
"learning_rate": 6.440543296692298e-06,
"loss": 2.9263,
"step": 278500
},
{
"epoch": 0.87,
"learning_rate": 6.362339604226754e-06,
"loss": 2.9141,
"step": 279000
},
{
"epoch": 0.87,
"learning_rate": 6.28413591176121e-06,
"loss": 2.9197,
"step": 279500
},
{
"epoch": 0.88,
"learning_rate": 6.205932219295666e-06,
"loss": 2.92,
"step": 280000
},
{
"epoch": 0.88,
"eval_accuracy": 0.43186914122938036,
"eval_loss": 2.8506009578704834,
"eval_runtime": 5983.5515,
"eval_samples_per_second": 89.965,
"eval_steps_per_second": 1.406,
"step": 280000
},
{
"epoch": 0.88,
"learning_rate": 6.127728526830123e-06,
"loss": 2.9182,
"step": 280500
},
{
"epoch": 0.88,
"learning_rate": 6.0495248343645794e-06,
"loss": 2.92,
"step": 281000
},
{
"epoch": 0.88,
"learning_rate": 5.971321141899036e-06,
"loss": 2.9203,
"step": 281500
},
{
"epoch": 0.88,
"learning_rate": 5.8931174494334924e-06,
"loss": 2.9157,
"step": 282000
},
{
"epoch": 0.88,
"learning_rate": 5.814913756967949e-06,
"loss": 2.9169,
"step": 282500
},
{
"epoch": 0.89,
"learning_rate": 5.7367100645024055e-06,
"loss": 2.9154,
"step": 283000
},
{
"epoch": 0.89,
"learning_rate": 5.658506372036863e-06,
"loss": 2.9159,
"step": 283500
},
{
"epoch": 0.89,
"learning_rate": 5.580302679571319e-06,
"loss": 2.9122,
"step": 284000
},
{
"epoch": 0.89,
"learning_rate": 5.502098987105776e-06,
"loss": 2.9134,
"step": 284500
},
{
"epoch": 0.89,
"learning_rate": 5.423895294640232e-06,
"loss": 2.9108,
"step": 285000
},
{
"epoch": 0.89,
"learning_rate": 5.345691602174689e-06,
"loss": 2.915,
"step": 285500
},
{
"epoch": 0.89,
"learning_rate": 5.267487909709145e-06,
"loss": 2.9153,
"step": 286000
},
{
"epoch": 0.9,
"learning_rate": 5.189284217243601e-06,
"loss": 2.9169,
"step": 286500
},
{
"epoch": 0.9,
"learning_rate": 5.1110805247780575e-06,
"loss": 2.9172,
"step": 287000
},
{
"epoch": 0.9,
"learning_rate": 5.032876832312515e-06,
"loss": 2.9206,
"step": 287500
},
{
"epoch": 0.9,
"learning_rate": 4.954673139846971e-06,
"loss": 2.9141,
"step": 288000
},
{
"epoch": 0.9,
"learning_rate": 4.876469447381428e-06,
"loss": 2.9178,
"step": 288500
},
{
"epoch": 0.9,
"learning_rate": 4.798265754915884e-06,
"loss": 2.9089,
"step": 289000
},
{
"epoch": 0.91,
"learning_rate": 4.720062062450341e-06,
"loss": 2.9175,
"step": 289500
},
{
"epoch": 0.91,
"learning_rate": 4.641858369984797e-06,
"loss": 2.9186,
"step": 290000
},
{
"epoch": 0.91,
"eval_accuracy": 0.4320649544420721,
"eval_loss": 2.8484013080596924,
"eval_runtime": 5983.5065,
"eval_samples_per_second": 89.966,
"eval_steps_per_second": 1.406,
"step": 290000
},
{
"epoch": 0.91,
"learning_rate": 4.563654677519254e-06,
"loss": 2.9101,
"step": 290500
},
{
"epoch": 0.91,
"learning_rate": 4.48545098505371e-06,
"loss": 2.9148,
"step": 291000
},
{
"epoch": 0.91,
"learning_rate": 4.407247292588168e-06,
"loss": 2.9101,
"step": 291500
},
{
"epoch": 0.91,
"learning_rate": 4.3290436001226234e-06,
"loss": 2.9164,
"step": 292000
},
{
"epoch": 0.91,
"learning_rate": 4.25083990765708e-06,
"loss": 2.9132,
"step": 292500
},
{
"epoch": 0.92,
"learning_rate": 4.1726362151915364e-06,
"loss": 2.9142,
"step": 293000
},
{
"epoch": 0.92,
"learning_rate": 4.094432522725993e-06,
"loss": 2.9153,
"step": 293500
},
{
"epoch": 0.92,
"learning_rate": 4.0162288302604495e-06,
"loss": 2.9044,
"step": 294000
},
{
"epoch": 0.92,
"learning_rate": 3.938025137794906e-06,
"loss": 2.9113,
"step": 294500
},
{
"epoch": 0.92,
"learning_rate": 3.8598214453293625e-06,
"loss": 2.917,
"step": 295000
},
{
"epoch": 0.92,
"learning_rate": 3.78161775286382e-06,
"loss": 2.9144,
"step": 295500
},
{
"epoch": 0.93,
"learning_rate": 3.7034140603982763e-06,
"loss": 2.9171,
"step": 296000
},
{
"epoch": 0.93,
"learning_rate": 3.6252103679327324e-06,
"loss": 2.9087,
"step": 296500
},
{
"epoch": 0.93,
"learning_rate": 3.547006675467189e-06,
"loss": 2.9131,
"step": 297000
},
{
"epoch": 0.93,
"learning_rate": 3.4688029830016454e-06,
"loss": 2.9139,
"step": 297500
},
{
"epoch": 0.93,
"learning_rate": 3.390599290536102e-06,
"loss": 2.9134,
"step": 298000
},
{
"epoch": 0.93,
"learning_rate": 3.3123955980705584e-06,
"loss": 2.9051,
"step": 298500
},
{
"epoch": 0.94,
"learning_rate": 3.234191905605015e-06,
"loss": 2.9131,
"step": 299000
},
{
"epoch": 0.94,
"learning_rate": 3.155988213139472e-06,
"loss": 2.9101,
"step": 299500
},
{
"epoch": 0.94,
"learning_rate": 3.077784520673928e-06,
"loss": 2.9084,
"step": 300000
},
{
"epoch": 0.94,
"eval_accuracy": 0.43248307111762724,
"eval_loss": 2.8458340167999268,
"eval_runtime": 5966.4445,
"eval_samples_per_second": 90.223,
"eval_steps_per_second": 1.41,
"step": 300000
},
{
"epoch": 0.94,
"learning_rate": 2.999580828208385e-06,
"loss": 2.9091,
"step": 300500
},
{
"epoch": 0.94,
"learning_rate": 2.9213771357428414e-06,
"loss": 2.9053,
"step": 301000
},
{
"epoch": 0.94,
"learning_rate": 2.843173443277298e-06,
"loss": 2.9162,
"step": 301500
},
{
"epoch": 0.94,
"learning_rate": 2.7649697508117544e-06,
"loss": 2.9106,
"step": 302000
},
{
"epoch": 0.95,
"learning_rate": 2.6867660583462113e-06,
"loss": 2.9144,
"step": 302500
},
{
"epoch": 0.95,
"learning_rate": 2.6085623658806674e-06,
"loss": 2.9153,
"step": 303000
},
{
"epoch": 0.95,
"learning_rate": 2.530358673415124e-06,
"loss": 2.9175,
"step": 303500
},
{
"epoch": 0.95,
"learning_rate": 2.4521549809495804e-06,
"loss": 2.9068,
"step": 304000
},
{
"epoch": 0.95,
"learning_rate": 2.3739512884840374e-06,
"loss": 2.9149,
"step": 304500
},
{
"epoch": 0.95,
"learning_rate": 2.295747596018494e-06,
"loss": 2.911,
"step": 305000
},
{
"epoch": 0.96,
"learning_rate": 2.2175439035529504e-06,
"loss": 2.9049,
"step": 305500
},
{
"epoch": 0.96,
"learning_rate": 2.139340211087407e-06,
"loss": 2.9071,
"step": 306000
},
{
"epoch": 0.96,
"learning_rate": 2.0611365186218634e-06,
"loss": 2.9026,
"step": 306500
},
{
"epoch": 0.96,
"learning_rate": 1.98293282615632e-06,
"loss": 2.9093,
"step": 307000
},
{
"epoch": 0.96,
"learning_rate": 1.9047291336907764e-06,
"loss": 2.9199,
"step": 307500
},
{
"epoch": 0.96,
"learning_rate": 1.826525441225233e-06,
"loss": 2.9103,
"step": 308000
},
{
"epoch": 0.97,
"learning_rate": 1.7483217487596896e-06,
"loss": 2.9138,
"step": 308500
},
{
"epoch": 0.97,
"learning_rate": 1.6701180562941461e-06,
"loss": 2.9132,
"step": 309000
},
{
"epoch": 0.97,
"learning_rate": 1.5919143638286026e-06,
"loss": 2.9156,
"step": 309500
},
{
"epoch": 0.97,
"learning_rate": 1.5137106713630591e-06,
"loss": 2.9142,
"step": 310000
},
{
"epoch": 0.97,
"eval_accuracy": 0.43274495378893163,
"eval_loss": 2.843777894973755,
"eval_runtime": 5973.9023,
"eval_samples_per_second": 90.111,
"eval_steps_per_second": 1.408,
"step": 310000
},
{
"epoch": 0.97,
"learning_rate": 1.4355069788975157e-06,
"loss": 2.9082,
"step": 310500
},
{
"epoch": 0.97,
"learning_rate": 1.3573032864319724e-06,
"loss": 2.9102,
"step": 311000
},
{
"epoch": 0.97,
"learning_rate": 1.2790995939664287e-06,
"loss": 2.9079,
"step": 311500
},
{
"epoch": 0.98,
"learning_rate": 1.2008959015008854e-06,
"loss": 2.9063,
"step": 312000
},
{
"epoch": 0.98,
"learning_rate": 1.1226922090353419e-06,
"loss": 2.9061,
"step": 312500
},
{
"epoch": 0.98,
"learning_rate": 1.0444885165697984e-06,
"loss": 2.9097,
"step": 313000
},
{
"epoch": 0.98,
"learning_rate": 9.66284824104255e-07,
"loss": 2.91,
"step": 313500
},
{
"epoch": 0.98,
"learning_rate": 8.880811316387115e-07,
"loss": 2.8976,
"step": 314000
},
{
"epoch": 0.98,
"learning_rate": 8.09877439173168e-07,
"loss": 2.9136,
"step": 314500
},
{
"epoch": 0.99,
"learning_rate": 7.316737467076245e-07,
"loss": 2.9078,
"step": 315000
},
{
"epoch": 0.99,
"learning_rate": 6.534700542420811e-07,
"loss": 2.9078,
"step": 315500
},
{
"epoch": 0.99,
"learning_rate": 5.752663617765376e-07,
"loss": 2.9046,
"step": 316000
},
{
"epoch": 0.99,
"learning_rate": 4.970626693109942e-07,
"loss": 2.909,
"step": 316500
},
{
"epoch": 0.99,
"learning_rate": 4.188589768454507e-07,
"loss": 2.9017,
"step": 317000
},
{
"epoch": 0.99,
"learning_rate": 3.406552843799073e-07,
"loss": 2.9089,
"step": 317500
},
{
"epoch": 0.99,
"learning_rate": 2.6245159191436383e-07,
"loss": 2.9107,
"step": 318000
},
{
"epoch": 1.0,
"learning_rate": 1.842478994488204e-07,
"loss": 2.9075,
"step": 318500
},
{
"epoch": 1.0,
"learning_rate": 1.0604420698327694e-07,
"loss": 2.9116,
"step": 319000
},
{
"epoch": 1.0,
"learning_rate": 2.784051451773347e-08,
"loss": 2.9071,
"step": 319500
},
{
"epoch": 1.0,
"step": 319678,
"total_flos": 5.345861476810752e+18,
"train_loss": 3.072396659716232,
"train_runtime": 348731.7472,
"train_samples_per_second": 29.334,
"train_steps_per_second": 0.917
}
],
"logging_steps": 500,
"max_steps": 319678,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 10000,
"total_flos": 5.345861476810752e+18,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}