gpt2_cx-cs_00000-00019_50k / trainer_state.json
jonasknobloch's picture
Initial commit
7ef86d3
raw
history blame
73.1 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"eval_steps": 10000,
"global_step": 268932,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 4.990703969776747e-05,
"loss": 8.1971,
"step": 500
},
{
"epoch": 0.0,
"learning_rate": 4.9814079395534934e-05,
"loss": 7.3697,
"step": 1000
},
{
"epoch": 0.01,
"learning_rate": 4.97211190933024e-05,
"loss": 6.9478,
"step": 1500
},
{
"epoch": 0.01,
"learning_rate": 4.962815879106986e-05,
"loss": 6.5908,
"step": 2000
},
{
"epoch": 0.01,
"learning_rate": 4.953519848883733e-05,
"loss": 6.3102,
"step": 2500
},
{
"epoch": 0.01,
"learning_rate": 4.94422381866048e-05,
"loss": 6.0716,
"step": 3000
},
{
"epoch": 0.01,
"learning_rate": 4.9349277884372256e-05,
"loss": 5.859,
"step": 3500
},
{
"epoch": 0.01,
"learning_rate": 4.925631758213973e-05,
"loss": 5.6554,
"step": 4000
},
{
"epoch": 0.02,
"learning_rate": 4.916335727990719e-05,
"loss": 5.4972,
"step": 4500
},
{
"epoch": 0.02,
"learning_rate": 4.907039697767465e-05,
"loss": 5.3606,
"step": 5000
},
{
"epoch": 0.02,
"learning_rate": 4.8977436675442126e-05,
"loss": 5.2518,
"step": 5500
},
{
"epoch": 0.02,
"learning_rate": 4.8884476373209585e-05,
"loss": 5.1562,
"step": 6000
},
{
"epoch": 0.02,
"learning_rate": 4.879151607097705e-05,
"loss": 5.0784,
"step": 6500
},
{
"epoch": 0.03,
"learning_rate": 4.8698555768744516e-05,
"loss": 5.0087,
"step": 7000
},
{
"epoch": 0.03,
"learning_rate": 4.860559546651198e-05,
"loss": 4.9465,
"step": 7500
},
{
"epoch": 0.03,
"learning_rate": 4.851263516427945e-05,
"loss": 4.8951,
"step": 8000
},
{
"epoch": 0.03,
"learning_rate": 4.841967486204691e-05,
"loss": 4.8451,
"step": 8500
},
{
"epoch": 0.03,
"learning_rate": 4.832671455981438e-05,
"loss": 4.8033,
"step": 9000
},
{
"epoch": 0.04,
"learning_rate": 4.8233754257581845e-05,
"loss": 4.772,
"step": 9500
},
{
"epoch": 0.04,
"learning_rate": 4.814079395534931e-05,
"loss": 4.729,
"step": 10000
},
{
"epoch": 0.04,
"eval_accuracy": 0.2836232206800241,
"eval_loss": 4.607733249664307,
"eval_runtime": 5046.5458,
"eval_samples_per_second": 89.595,
"eval_steps_per_second": 1.4,
"step": 10000
},
{
"epoch": 0.04,
"learning_rate": 4.8047833653116776e-05,
"loss": 4.698,
"step": 10500
},
{
"epoch": 0.04,
"learning_rate": 4.7954873350884235e-05,
"loss": 4.6666,
"step": 11000
},
{
"epoch": 0.04,
"learning_rate": 4.786191304865171e-05,
"loss": 4.6425,
"step": 11500
},
{
"epoch": 0.04,
"learning_rate": 4.7768952746419174e-05,
"loss": 4.6117,
"step": 12000
},
{
"epoch": 0.05,
"learning_rate": 4.767599244418663e-05,
"loss": 4.5848,
"step": 12500
},
{
"epoch": 0.05,
"learning_rate": 4.7583032141954105e-05,
"loss": 4.5665,
"step": 13000
},
{
"epoch": 0.05,
"learning_rate": 4.7490071839721564e-05,
"loss": 4.5407,
"step": 13500
},
{
"epoch": 0.05,
"learning_rate": 4.739711153748903e-05,
"loss": 4.5278,
"step": 14000
},
{
"epoch": 0.05,
"learning_rate": 4.73041512352565e-05,
"loss": 4.501,
"step": 14500
},
{
"epoch": 0.06,
"learning_rate": 4.721119093302396e-05,
"loss": 4.4807,
"step": 15000
},
{
"epoch": 0.06,
"learning_rate": 4.711823063079143e-05,
"loss": 4.468,
"step": 15500
},
{
"epoch": 0.06,
"learning_rate": 4.702527032855889e-05,
"loss": 4.4464,
"step": 16000
},
{
"epoch": 0.06,
"learning_rate": 4.693231002632636e-05,
"loss": 4.4371,
"step": 16500
},
{
"epoch": 0.06,
"learning_rate": 4.6839349724093824e-05,
"loss": 4.4197,
"step": 17000
},
{
"epoch": 0.07,
"learning_rate": 4.674638942186129e-05,
"loss": 4.4046,
"step": 17500
},
{
"epoch": 0.07,
"learning_rate": 4.6653429119628756e-05,
"loss": 4.3892,
"step": 18000
},
{
"epoch": 0.07,
"learning_rate": 4.656046881739622e-05,
"loss": 4.3779,
"step": 18500
},
{
"epoch": 0.07,
"learning_rate": 4.646750851516369e-05,
"loss": 4.3627,
"step": 19000
},
{
"epoch": 0.07,
"learning_rate": 4.637454821293115e-05,
"loss": 4.3501,
"step": 19500
},
{
"epoch": 0.07,
"learning_rate": 4.628158791069861e-05,
"loss": 4.3383,
"step": 20000
},
{
"epoch": 0.07,
"eval_accuracy": 0.31616356516397504,
"eval_loss": 4.231759548187256,
"eval_runtime": 5043.1658,
"eval_samples_per_second": 89.655,
"eval_steps_per_second": 1.401,
"step": 20000
},
{
"epoch": 0.08,
"learning_rate": 4.6188627608466085e-05,
"loss": 4.3304,
"step": 20500
},
{
"epoch": 0.08,
"learning_rate": 4.609566730623355e-05,
"loss": 4.3221,
"step": 21000
},
{
"epoch": 0.08,
"learning_rate": 4.600270700400101e-05,
"loss": 4.3029,
"step": 21500
},
{
"epoch": 0.08,
"learning_rate": 4.590974670176848e-05,
"loss": 4.2954,
"step": 22000
},
{
"epoch": 0.08,
"learning_rate": 4.581678639953594e-05,
"loss": 4.2818,
"step": 22500
},
{
"epoch": 0.09,
"learning_rate": 4.5723826097303407e-05,
"loss": 4.2812,
"step": 23000
},
{
"epoch": 0.09,
"learning_rate": 4.563086579507088e-05,
"loss": 4.2749,
"step": 23500
},
{
"epoch": 0.09,
"learning_rate": 4.553790549283834e-05,
"loss": 4.2555,
"step": 24000
},
{
"epoch": 0.09,
"learning_rate": 4.5444945190605804e-05,
"loss": 4.2467,
"step": 24500
},
{
"epoch": 0.09,
"learning_rate": 4.535198488837327e-05,
"loss": 4.242,
"step": 25000
},
{
"epoch": 0.09,
"learning_rate": 4.5259024586140735e-05,
"loss": 4.2395,
"step": 25500
},
{
"epoch": 0.1,
"learning_rate": 4.51660642839082e-05,
"loss": 4.2208,
"step": 26000
},
{
"epoch": 0.1,
"learning_rate": 4.507310398167567e-05,
"loss": 4.2181,
"step": 26500
},
{
"epoch": 0.1,
"learning_rate": 4.498014367944313e-05,
"loss": 4.2108,
"step": 27000
},
{
"epoch": 0.1,
"learning_rate": 4.48871833772106e-05,
"loss": 4.2053,
"step": 27500
},
{
"epoch": 0.1,
"learning_rate": 4.4794223074978064e-05,
"loss": 4.1924,
"step": 28000
},
{
"epoch": 0.11,
"learning_rate": 4.470126277274553e-05,
"loss": 4.187,
"step": 28500
},
{
"epoch": 0.11,
"learning_rate": 4.460830247051299e-05,
"loss": 4.1839,
"step": 29000
},
{
"epoch": 0.11,
"learning_rate": 4.451534216828046e-05,
"loss": 4.1739,
"step": 29500
},
{
"epoch": 0.11,
"learning_rate": 4.442238186604793e-05,
"loss": 4.1706,
"step": 30000
},
{
"epoch": 0.11,
"eval_accuracy": 0.33159489024981786,
"eval_loss": 4.065094947814941,
"eval_runtime": 5049.5785,
"eval_samples_per_second": 89.541,
"eval_steps_per_second": 1.399,
"step": 30000
},
{
"epoch": 0.11,
"learning_rate": 4.432942156381539e-05,
"loss": 4.1606,
"step": 30500
},
{
"epoch": 0.12,
"learning_rate": 4.423646126158286e-05,
"loss": 4.1613,
"step": 31000
},
{
"epoch": 0.12,
"learning_rate": 4.414350095935032e-05,
"loss": 4.1533,
"step": 31500
},
{
"epoch": 0.12,
"learning_rate": 4.405054065711779e-05,
"loss": 4.1449,
"step": 32000
},
{
"epoch": 0.12,
"learning_rate": 4.3957580354885256e-05,
"loss": 4.1376,
"step": 32500
},
{
"epoch": 0.12,
"learning_rate": 4.3864620052652715e-05,
"loss": 4.1307,
"step": 33000
},
{
"epoch": 0.12,
"learning_rate": 4.377165975042019e-05,
"loss": 4.1225,
"step": 33500
},
{
"epoch": 0.13,
"learning_rate": 4.3678699448187646e-05,
"loss": 4.1147,
"step": 34000
},
{
"epoch": 0.13,
"learning_rate": 4.358573914595511e-05,
"loss": 4.1113,
"step": 34500
},
{
"epoch": 0.13,
"learning_rate": 4.3492778843722584e-05,
"loss": 4.1075,
"step": 35000
},
{
"epoch": 0.13,
"learning_rate": 4.3399818541490043e-05,
"loss": 4.1023,
"step": 35500
},
{
"epoch": 0.13,
"learning_rate": 4.330685823925751e-05,
"loss": 4.0986,
"step": 36000
},
{
"epoch": 0.14,
"learning_rate": 4.3213897937024975e-05,
"loss": 4.0933,
"step": 36500
},
{
"epoch": 0.14,
"learning_rate": 4.312093763479244e-05,
"loss": 4.0899,
"step": 37000
},
{
"epoch": 0.14,
"learning_rate": 4.3027977332559906e-05,
"loss": 4.0842,
"step": 37500
},
{
"epoch": 0.14,
"learning_rate": 4.293501703032737e-05,
"loss": 4.0783,
"step": 38000
},
{
"epoch": 0.14,
"learning_rate": 4.284205672809484e-05,
"loss": 4.0721,
"step": 38500
},
{
"epoch": 0.15,
"learning_rate": 4.2749096425862304e-05,
"loss": 4.0692,
"step": 39000
},
{
"epoch": 0.15,
"learning_rate": 4.265613612362977e-05,
"loss": 4.0613,
"step": 39500
},
{
"epoch": 0.15,
"learning_rate": 4.2563175821397235e-05,
"loss": 4.0594,
"step": 40000
},
{
"epoch": 0.15,
"eval_accuracy": 0.34162108157696924,
"eval_loss": 3.9599251747131348,
"eval_runtime": 5055.4019,
"eval_samples_per_second": 89.438,
"eval_steps_per_second": 1.398,
"step": 40000
},
{
"epoch": 0.15,
"learning_rate": 4.2470215519164694e-05,
"loss": 4.0552,
"step": 40500
},
{
"epoch": 0.15,
"learning_rate": 4.237725521693217e-05,
"loss": 4.0538,
"step": 41000
},
{
"epoch": 0.15,
"learning_rate": 4.228429491469963e-05,
"loss": 4.049,
"step": 41500
},
{
"epoch": 0.16,
"learning_rate": 4.219133461246709e-05,
"loss": 4.04,
"step": 42000
},
{
"epoch": 0.16,
"learning_rate": 4.2098374310234564e-05,
"loss": 4.0417,
"step": 42500
},
{
"epoch": 0.16,
"learning_rate": 4.200541400800202e-05,
"loss": 4.0398,
"step": 43000
},
{
"epoch": 0.16,
"learning_rate": 4.191245370576949e-05,
"loss": 4.0348,
"step": 43500
},
{
"epoch": 0.16,
"learning_rate": 4.181949340353696e-05,
"loss": 4.0271,
"step": 44000
},
{
"epoch": 0.17,
"learning_rate": 4.172653310130442e-05,
"loss": 4.0187,
"step": 44500
},
{
"epoch": 0.17,
"learning_rate": 4.1633572799071886e-05,
"loss": 4.0195,
"step": 45000
},
{
"epoch": 0.17,
"learning_rate": 4.154061249683935e-05,
"loss": 4.0134,
"step": 45500
},
{
"epoch": 0.17,
"learning_rate": 4.144765219460682e-05,
"loss": 4.012,
"step": 46000
},
{
"epoch": 0.17,
"learning_rate": 4.135469189237428e-05,
"loss": 4.0103,
"step": 46500
},
{
"epoch": 0.17,
"learning_rate": 4.126173159014175e-05,
"loss": 4.0061,
"step": 47000
},
{
"epoch": 0.18,
"learning_rate": 4.1168771287909215e-05,
"loss": 3.9976,
"step": 47500
},
{
"epoch": 0.18,
"learning_rate": 4.107581098567668e-05,
"loss": 4.0012,
"step": 48000
},
{
"epoch": 0.18,
"learning_rate": 4.0982850683444146e-05,
"loss": 4.0002,
"step": 48500
},
{
"epoch": 0.18,
"learning_rate": 4.088989038121161e-05,
"loss": 3.9899,
"step": 49000
},
{
"epoch": 0.18,
"learning_rate": 4.079693007897907e-05,
"loss": 3.9881,
"step": 49500
},
{
"epoch": 0.19,
"learning_rate": 4.070396977674654e-05,
"loss": 3.9842,
"step": 50000
},
{
"epoch": 0.19,
"eval_accuracy": 0.3486961893069126,
"eval_loss": 3.8825323581695557,
"eval_runtime": 5041.9196,
"eval_samples_per_second": 89.677,
"eval_steps_per_second": 1.401,
"step": 50000
},
{
"epoch": 0.19,
"learning_rate": 4.0611009474514e-05,
"loss": 3.9765,
"step": 50500
},
{
"epoch": 0.19,
"learning_rate": 4.051804917228147e-05,
"loss": 3.974,
"step": 51000
},
{
"epoch": 0.19,
"learning_rate": 4.042508887004894e-05,
"loss": 3.9809,
"step": 51500
},
{
"epoch": 0.19,
"learning_rate": 4.03321285678164e-05,
"loss": 3.9751,
"step": 52000
},
{
"epoch": 0.2,
"learning_rate": 4.0239168265583865e-05,
"loss": 3.9714,
"step": 52500
},
{
"epoch": 0.2,
"learning_rate": 4.014620796335133e-05,
"loss": 3.967,
"step": 53000
},
{
"epoch": 0.2,
"learning_rate": 4.00532476611188e-05,
"loss": 3.9642,
"step": 53500
},
{
"epoch": 0.2,
"learning_rate": 3.996028735888626e-05,
"loss": 3.9586,
"step": 54000
},
{
"epoch": 0.2,
"learning_rate": 3.986732705665373e-05,
"loss": 3.9591,
"step": 54500
},
{
"epoch": 0.2,
"learning_rate": 3.9774366754421194e-05,
"loss": 3.9532,
"step": 55000
},
{
"epoch": 0.21,
"learning_rate": 3.968140645218866e-05,
"loss": 3.9432,
"step": 55500
},
{
"epoch": 0.21,
"learning_rate": 3.9588446149956126e-05,
"loss": 3.9482,
"step": 56000
},
{
"epoch": 0.21,
"learning_rate": 3.949548584772359e-05,
"loss": 3.9414,
"step": 56500
},
{
"epoch": 0.21,
"learning_rate": 3.940252554549105e-05,
"loss": 3.9397,
"step": 57000
},
{
"epoch": 0.21,
"learning_rate": 3.930956524325852e-05,
"loss": 3.9329,
"step": 57500
},
{
"epoch": 0.22,
"learning_rate": 3.921660494102599e-05,
"loss": 3.9361,
"step": 58000
},
{
"epoch": 0.22,
"learning_rate": 3.912364463879345e-05,
"loss": 3.9322,
"step": 58500
},
{
"epoch": 0.22,
"learning_rate": 3.903068433656092e-05,
"loss": 3.929,
"step": 59000
},
{
"epoch": 0.22,
"learning_rate": 3.893772403432838e-05,
"loss": 3.9308,
"step": 59500
},
{
"epoch": 0.22,
"learning_rate": 3.8844763732095845e-05,
"loss": 3.9298,
"step": 60000
},
{
"epoch": 0.22,
"eval_accuracy": 0.3545327865706106,
"eval_loss": 3.824397087097168,
"eval_runtime": 5028.3225,
"eval_samples_per_second": 89.92,
"eval_steps_per_second": 1.405,
"step": 60000
},
{
"epoch": 0.22,
"learning_rate": 3.875180342986332e-05,
"loss": 3.9259,
"step": 60500
},
{
"epoch": 0.23,
"learning_rate": 3.8658843127630776e-05,
"loss": 3.9188,
"step": 61000
},
{
"epoch": 0.23,
"learning_rate": 3.856588282539824e-05,
"loss": 3.9109,
"step": 61500
},
{
"epoch": 0.23,
"learning_rate": 3.847292252316571e-05,
"loss": 3.9193,
"step": 62000
},
{
"epoch": 0.23,
"learning_rate": 3.8379962220933174e-05,
"loss": 3.9078,
"step": 62500
},
{
"epoch": 0.23,
"learning_rate": 3.828700191870064e-05,
"loss": 3.913,
"step": 63000
},
{
"epoch": 0.24,
"learning_rate": 3.8194041616468105e-05,
"loss": 3.9015,
"step": 63500
},
{
"epoch": 0.24,
"learning_rate": 3.810108131423557e-05,
"loss": 3.9075,
"step": 64000
},
{
"epoch": 0.24,
"learning_rate": 3.8008121012003037e-05,
"loss": 3.9038,
"step": 64500
},
{
"epoch": 0.24,
"learning_rate": 3.79151607097705e-05,
"loss": 3.8996,
"step": 65000
},
{
"epoch": 0.24,
"learning_rate": 3.782220040753797e-05,
"loss": 3.8987,
"step": 65500
},
{
"epoch": 0.25,
"learning_rate": 3.772924010530543e-05,
"loss": 3.8999,
"step": 66000
},
{
"epoch": 0.25,
"learning_rate": 3.76362798030729e-05,
"loss": 3.8889,
"step": 66500
},
{
"epoch": 0.25,
"learning_rate": 3.7543319500840365e-05,
"loss": 3.8927,
"step": 67000
},
{
"epoch": 0.25,
"learning_rate": 3.7450359198607824e-05,
"loss": 3.8974,
"step": 67500
},
{
"epoch": 0.25,
"learning_rate": 3.73573988963753e-05,
"loss": 3.8895,
"step": 68000
},
{
"epoch": 0.25,
"learning_rate": 3.7264438594142756e-05,
"loss": 3.8908,
"step": 68500
},
{
"epoch": 0.26,
"learning_rate": 3.717147829191022e-05,
"loss": 3.8816,
"step": 69000
},
{
"epoch": 0.26,
"learning_rate": 3.7078517989677694e-05,
"loss": 3.8718,
"step": 69500
},
{
"epoch": 0.26,
"learning_rate": 3.698555768744515e-05,
"loss": 3.8777,
"step": 70000
},
{
"epoch": 0.26,
"eval_accuracy": 0.35923910731713965,
"eval_loss": 3.7790777683258057,
"eval_runtime": 5041.8181,
"eval_samples_per_second": 89.679,
"eval_steps_per_second": 1.401,
"step": 70000
},
{
"epoch": 0.26,
"learning_rate": 3.689259738521262e-05,
"loss": 3.8821,
"step": 70500
},
{
"epoch": 0.26,
"learning_rate": 3.6799637082980085e-05,
"loss": 3.8743,
"step": 71000
},
{
"epoch": 0.27,
"learning_rate": 3.670667678074755e-05,
"loss": 3.8735,
"step": 71500
},
{
"epoch": 0.27,
"learning_rate": 3.6613716478515016e-05,
"loss": 3.8722,
"step": 72000
},
{
"epoch": 0.27,
"learning_rate": 3.652075617628248e-05,
"loss": 3.8644,
"step": 72500
},
{
"epoch": 0.27,
"learning_rate": 3.642779587404995e-05,
"loss": 3.867,
"step": 73000
},
{
"epoch": 0.27,
"learning_rate": 3.633483557181741e-05,
"loss": 3.8666,
"step": 73500
},
{
"epoch": 0.28,
"learning_rate": 3.624187526958488e-05,
"loss": 3.8615,
"step": 74000
},
{
"epoch": 0.28,
"learning_rate": 3.6148914967352345e-05,
"loss": 3.8662,
"step": 74500
},
{
"epoch": 0.28,
"learning_rate": 3.6055954665119804e-05,
"loss": 3.8645,
"step": 75000
},
{
"epoch": 0.28,
"learning_rate": 3.5962994362887276e-05,
"loss": 3.8559,
"step": 75500
},
{
"epoch": 0.28,
"learning_rate": 3.587003406065474e-05,
"loss": 3.8536,
"step": 76000
},
{
"epoch": 0.28,
"learning_rate": 3.57770737584222e-05,
"loss": 3.8527,
"step": 76500
},
{
"epoch": 0.29,
"learning_rate": 3.5684113456189673e-05,
"loss": 3.848,
"step": 77000
},
{
"epoch": 0.29,
"learning_rate": 3.559115315395713e-05,
"loss": 3.8495,
"step": 77500
},
{
"epoch": 0.29,
"learning_rate": 3.54981928517246e-05,
"loss": 3.8523,
"step": 78000
},
{
"epoch": 0.29,
"learning_rate": 3.540523254949207e-05,
"loss": 3.8505,
"step": 78500
},
{
"epoch": 0.29,
"learning_rate": 3.531227224725953e-05,
"loss": 3.8439,
"step": 79000
},
{
"epoch": 0.3,
"learning_rate": 3.5219311945026995e-05,
"loss": 3.8387,
"step": 79500
},
{
"epoch": 0.3,
"learning_rate": 3.512635164279446e-05,
"loss": 3.8455,
"step": 80000
},
{
"epoch": 0.3,
"eval_accuracy": 0.36290677735789373,
"eval_loss": 3.7436022758483887,
"eval_runtime": 5030.6589,
"eval_samples_per_second": 89.878,
"eval_steps_per_second": 1.404,
"step": 80000
},
{
"epoch": 0.3,
"learning_rate": 3.503339134056193e-05,
"loss": 3.8383,
"step": 80500
},
{
"epoch": 0.3,
"learning_rate": 3.494043103832939e-05,
"loss": 3.8445,
"step": 81000
},
{
"epoch": 0.3,
"learning_rate": 3.484747073609686e-05,
"loss": 3.8439,
"step": 81500
},
{
"epoch": 0.3,
"learning_rate": 3.4754510433864324e-05,
"loss": 3.8357,
"step": 82000
},
{
"epoch": 0.31,
"learning_rate": 3.466155013163179e-05,
"loss": 3.8362,
"step": 82500
},
{
"epoch": 0.31,
"learning_rate": 3.4568589829399256e-05,
"loss": 3.8311,
"step": 83000
},
{
"epoch": 0.31,
"learning_rate": 3.447562952716672e-05,
"loss": 3.8263,
"step": 83500
},
{
"epoch": 0.31,
"learning_rate": 3.438266922493418e-05,
"loss": 3.8328,
"step": 84000
},
{
"epoch": 0.31,
"learning_rate": 3.428970892270165e-05,
"loss": 3.8303,
"step": 84500
},
{
"epoch": 0.32,
"learning_rate": 3.419674862046912e-05,
"loss": 3.8202,
"step": 85000
},
{
"epoch": 0.32,
"learning_rate": 3.410378831823658e-05,
"loss": 3.8243,
"step": 85500
},
{
"epoch": 0.32,
"learning_rate": 3.401082801600405e-05,
"loss": 3.8284,
"step": 86000
},
{
"epoch": 0.32,
"learning_rate": 3.391786771377151e-05,
"loss": 3.8215,
"step": 86500
},
{
"epoch": 0.32,
"learning_rate": 3.3824907411538975e-05,
"loss": 3.8179,
"step": 87000
},
{
"epoch": 0.33,
"learning_rate": 3.373194710930645e-05,
"loss": 3.821,
"step": 87500
},
{
"epoch": 0.33,
"learning_rate": 3.3638986807073906e-05,
"loss": 3.816,
"step": 88000
},
{
"epoch": 0.33,
"learning_rate": 3.354602650484138e-05,
"loss": 3.8129,
"step": 88500
},
{
"epoch": 0.33,
"learning_rate": 3.345306620260884e-05,
"loss": 3.8203,
"step": 89000
},
{
"epoch": 0.33,
"learning_rate": 3.3360105900376304e-05,
"loss": 3.818,
"step": 89500
},
{
"epoch": 0.33,
"learning_rate": 3.3267145598143776e-05,
"loss": 3.8104,
"step": 90000
},
{
"epoch": 0.33,
"eval_accuracy": 0.36599874258539994,
"eval_loss": 3.7120308876037598,
"eval_runtime": 5023.4047,
"eval_samples_per_second": 90.008,
"eval_steps_per_second": 1.406,
"step": 90000
},
{
"epoch": 0.34,
"learning_rate": 3.3174185295911235e-05,
"loss": 3.8046,
"step": 90500
},
{
"epoch": 0.34,
"learning_rate": 3.30812249936787e-05,
"loss": 3.8116,
"step": 91000
},
{
"epoch": 0.34,
"learning_rate": 3.298826469144617e-05,
"loss": 3.8063,
"step": 91500
},
{
"epoch": 0.34,
"learning_rate": 3.289530438921363e-05,
"loss": 3.8029,
"step": 92000
},
{
"epoch": 0.34,
"learning_rate": 3.28023440869811e-05,
"loss": 3.8023,
"step": 92500
},
{
"epoch": 0.35,
"learning_rate": 3.2709383784748564e-05,
"loss": 3.799,
"step": 93000
},
{
"epoch": 0.35,
"learning_rate": 3.261642348251603e-05,
"loss": 3.7974,
"step": 93500
},
{
"epoch": 0.35,
"learning_rate": 3.2523463180283495e-05,
"loss": 3.7981,
"step": 94000
},
{
"epoch": 0.35,
"learning_rate": 3.243050287805096e-05,
"loss": 3.7979,
"step": 94500
},
{
"epoch": 0.35,
"learning_rate": 3.233754257581843e-05,
"loss": 3.7971,
"step": 95000
},
{
"epoch": 0.36,
"learning_rate": 3.2244582273585886e-05,
"loss": 3.7956,
"step": 95500
},
{
"epoch": 0.36,
"learning_rate": 3.215162197135336e-05,
"loss": 3.7966,
"step": 96000
},
{
"epoch": 0.36,
"learning_rate": 3.2058661669120824e-05,
"loss": 3.7903,
"step": 96500
},
{
"epoch": 0.36,
"learning_rate": 3.196570136688828e-05,
"loss": 3.7944,
"step": 97000
},
{
"epoch": 0.36,
"learning_rate": 3.1872741064655756e-05,
"loss": 3.7942,
"step": 97500
},
{
"epoch": 0.36,
"learning_rate": 3.1779780762423215e-05,
"loss": 3.7871,
"step": 98000
},
{
"epoch": 0.37,
"learning_rate": 3.168682046019068e-05,
"loss": 3.7912,
"step": 98500
},
{
"epoch": 0.37,
"learning_rate": 3.1593860157958146e-05,
"loss": 3.7909,
"step": 99000
},
{
"epoch": 0.37,
"learning_rate": 3.150089985572561e-05,
"loss": 3.7798,
"step": 99500
},
{
"epoch": 0.37,
"learning_rate": 3.140793955349308e-05,
"loss": 3.7908,
"step": 100000
},
{
"epoch": 0.37,
"eval_accuracy": 0.36866443948556843,
"eval_loss": 3.686194896697998,
"eval_runtime": 5036.4153,
"eval_samples_per_second": 89.775,
"eval_steps_per_second": 1.403,
"step": 100000
},
{
"epoch": 0.37,
"learning_rate": 3.131497925126054e-05,
"loss": 3.7867,
"step": 100500
},
{
"epoch": 0.38,
"learning_rate": 3.122201894902801e-05,
"loss": 3.7815,
"step": 101000
},
{
"epoch": 0.38,
"learning_rate": 3.1129058646795475e-05,
"loss": 3.781,
"step": 101500
},
{
"epoch": 0.38,
"learning_rate": 3.103609834456294e-05,
"loss": 3.7821,
"step": 102000
},
{
"epoch": 0.38,
"learning_rate": 3.0943138042330406e-05,
"loss": 3.7731,
"step": 102500
},
{
"epoch": 0.38,
"learning_rate": 3.0850177740097865e-05,
"loss": 3.7803,
"step": 103000
},
{
"epoch": 0.38,
"learning_rate": 3.075721743786534e-05,
"loss": 3.7747,
"step": 103500
},
{
"epoch": 0.39,
"learning_rate": 3.0664257135632804e-05,
"loss": 3.776,
"step": 104000
},
{
"epoch": 0.39,
"learning_rate": 3.057129683340026e-05,
"loss": 3.7718,
"step": 104500
},
{
"epoch": 0.39,
"learning_rate": 3.0478336531167735e-05,
"loss": 3.7762,
"step": 105000
},
{
"epoch": 0.39,
"learning_rate": 3.0385376228935197e-05,
"loss": 3.7731,
"step": 105500
},
{
"epoch": 0.39,
"learning_rate": 3.029241592670266e-05,
"loss": 3.7678,
"step": 106000
},
{
"epoch": 0.4,
"learning_rate": 3.019945562447013e-05,
"loss": 3.7706,
"step": 106500
},
{
"epoch": 0.4,
"learning_rate": 3.0106495322237595e-05,
"loss": 3.7676,
"step": 107000
},
{
"epoch": 0.4,
"learning_rate": 3.0013535020005057e-05,
"loss": 3.7703,
"step": 107500
},
{
"epoch": 0.4,
"learning_rate": 2.9920574717772526e-05,
"loss": 3.7657,
"step": 108000
},
{
"epoch": 0.4,
"learning_rate": 2.982761441553999e-05,
"loss": 3.7717,
"step": 108500
},
{
"epoch": 0.41,
"learning_rate": 2.9734654113307454e-05,
"loss": 3.7669,
"step": 109000
},
{
"epoch": 0.41,
"learning_rate": 2.9641693811074923e-05,
"loss": 3.7598,
"step": 109500
},
{
"epoch": 0.41,
"learning_rate": 2.9548733508842386e-05,
"loss": 3.7613,
"step": 110000
},
{
"epoch": 0.41,
"eval_accuracy": 0.37119206097292273,
"eval_loss": 3.6628386974334717,
"eval_runtime": 5026.6773,
"eval_samples_per_second": 89.949,
"eval_steps_per_second": 1.406,
"step": 110000
},
{
"epoch": 0.41,
"learning_rate": 2.9455773206609848e-05,
"loss": 3.7575,
"step": 110500
},
{
"epoch": 0.41,
"learning_rate": 2.9362812904377317e-05,
"loss": 3.7598,
"step": 111000
},
{
"epoch": 0.41,
"learning_rate": 2.9269852602144783e-05,
"loss": 3.7473,
"step": 111500
},
{
"epoch": 0.42,
"learning_rate": 2.9176892299912245e-05,
"loss": 3.7578,
"step": 112000
},
{
"epoch": 0.42,
"learning_rate": 2.9083931997679715e-05,
"loss": 3.7512,
"step": 112500
},
{
"epoch": 0.42,
"learning_rate": 2.8990971695447177e-05,
"loss": 3.7517,
"step": 113000
},
{
"epoch": 0.42,
"learning_rate": 2.8898011393214643e-05,
"loss": 3.7585,
"step": 113500
},
{
"epoch": 0.42,
"learning_rate": 2.8805051090982112e-05,
"loss": 3.7595,
"step": 114000
},
{
"epoch": 0.43,
"learning_rate": 2.8712090788749574e-05,
"loss": 3.7493,
"step": 114500
},
{
"epoch": 0.43,
"learning_rate": 2.8619130486517037e-05,
"loss": 3.7554,
"step": 115000
},
{
"epoch": 0.43,
"learning_rate": 2.8526170184284506e-05,
"loss": 3.7515,
"step": 115500
},
{
"epoch": 0.43,
"learning_rate": 2.843320988205197e-05,
"loss": 3.7483,
"step": 116000
},
{
"epoch": 0.43,
"learning_rate": 2.8340249579819434e-05,
"loss": 3.7577,
"step": 116500
},
{
"epoch": 0.44,
"learning_rate": 2.8247289277586903e-05,
"loss": 3.753,
"step": 117000
},
{
"epoch": 0.44,
"learning_rate": 2.8154328975354365e-05,
"loss": 3.7472,
"step": 117500
},
{
"epoch": 0.44,
"learning_rate": 2.806136867312183e-05,
"loss": 3.742,
"step": 118000
},
{
"epoch": 0.44,
"learning_rate": 2.79684083708893e-05,
"loss": 3.7445,
"step": 118500
},
{
"epoch": 0.44,
"learning_rate": 2.7875448068656763e-05,
"loss": 3.7493,
"step": 119000
},
{
"epoch": 0.44,
"learning_rate": 2.7782487766424225e-05,
"loss": 3.7427,
"step": 119500
},
{
"epoch": 0.45,
"learning_rate": 2.7689527464191694e-05,
"loss": 3.7492,
"step": 120000
},
{
"epoch": 0.45,
"eval_accuracy": 0.3731269306324982,
"eval_loss": 3.643425464630127,
"eval_runtime": 5034.359,
"eval_samples_per_second": 89.812,
"eval_steps_per_second": 1.403,
"step": 120000
},
{
"epoch": 0.45,
"learning_rate": 2.759656716195916e-05,
"loss": 3.7424,
"step": 120500
},
{
"epoch": 0.45,
"learning_rate": 2.7503606859726622e-05,
"loss": 3.7326,
"step": 121000
},
{
"epoch": 0.45,
"learning_rate": 2.741064655749409e-05,
"loss": 3.7368,
"step": 121500
},
{
"epoch": 0.45,
"learning_rate": 2.7317686255261554e-05,
"loss": 3.7387,
"step": 122000
},
{
"epoch": 0.46,
"learning_rate": 2.722472595302902e-05,
"loss": 3.7392,
"step": 122500
},
{
"epoch": 0.46,
"learning_rate": 2.713176565079649e-05,
"loss": 3.7371,
"step": 123000
},
{
"epoch": 0.46,
"learning_rate": 2.703880534856395e-05,
"loss": 3.7302,
"step": 123500
},
{
"epoch": 0.46,
"learning_rate": 2.6945845046331413e-05,
"loss": 3.7336,
"step": 124000
},
{
"epoch": 0.46,
"learning_rate": 2.6852884744098882e-05,
"loss": 3.7401,
"step": 124500
},
{
"epoch": 0.46,
"learning_rate": 2.6759924441866345e-05,
"loss": 3.7264,
"step": 125000
},
{
"epoch": 0.47,
"learning_rate": 2.666696413963381e-05,
"loss": 3.7328,
"step": 125500
},
{
"epoch": 0.47,
"learning_rate": 2.657400383740128e-05,
"loss": 3.7239,
"step": 126000
},
{
"epoch": 0.47,
"learning_rate": 2.6481043535168742e-05,
"loss": 3.727,
"step": 126500
},
{
"epoch": 0.47,
"learning_rate": 2.6388083232936204e-05,
"loss": 3.7448,
"step": 127000
},
{
"epoch": 0.47,
"learning_rate": 2.6295122930703673e-05,
"loss": 3.7296,
"step": 127500
},
{
"epoch": 0.48,
"learning_rate": 2.620216262847114e-05,
"loss": 3.7226,
"step": 128000
},
{
"epoch": 0.48,
"learning_rate": 2.61092023262386e-05,
"loss": 3.7289,
"step": 128500
},
{
"epoch": 0.48,
"learning_rate": 2.601624202400607e-05,
"loss": 3.7117,
"step": 129000
},
{
"epoch": 0.48,
"learning_rate": 2.5923281721773533e-05,
"loss": 3.7261,
"step": 129500
},
{
"epoch": 0.48,
"learning_rate": 2.5830321419541e-05,
"loss": 3.7228,
"step": 130000
},
{
"epoch": 0.48,
"eval_accuracy": 0.3750882777539584,
"eval_loss": 3.6245925426483154,
"eval_runtime": 5034.4807,
"eval_samples_per_second": 89.81,
"eval_steps_per_second": 1.403,
"step": 130000
},
{
"epoch": 0.49,
"learning_rate": 2.5737361117308468e-05,
"loss": 3.7316,
"step": 130500
},
{
"epoch": 0.49,
"learning_rate": 2.564440081507593e-05,
"loss": 3.7187,
"step": 131000
},
{
"epoch": 0.49,
"learning_rate": 2.5551440512843393e-05,
"loss": 3.7147,
"step": 131500
},
{
"epoch": 0.49,
"learning_rate": 2.5458480210610862e-05,
"loss": 3.7185,
"step": 132000
},
{
"epoch": 0.49,
"learning_rate": 2.5365519908378328e-05,
"loss": 3.7186,
"step": 132500
},
{
"epoch": 0.49,
"learning_rate": 2.527255960614579e-05,
"loss": 3.7204,
"step": 133000
},
{
"epoch": 0.5,
"learning_rate": 2.517959930391326e-05,
"loss": 3.7154,
"step": 133500
},
{
"epoch": 0.5,
"learning_rate": 2.508663900168072e-05,
"loss": 3.7149,
"step": 134000
},
{
"epoch": 0.5,
"learning_rate": 2.499367869944819e-05,
"loss": 3.7138,
"step": 134500
},
{
"epoch": 0.5,
"learning_rate": 2.4900718397215656e-05,
"loss": 3.7194,
"step": 135000
},
{
"epoch": 0.5,
"learning_rate": 2.480775809498312e-05,
"loss": 3.7161,
"step": 135500
},
{
"epoch": 0.51,
"learning_rate": 2.4714797792750584e-05,
"loss": 3.7113,
"step": 136000
},
{
"epoch": 0.51,
"learning_rate": 2.462183749051805e-05,
"loss": 3.7124,
"step": 136500
},
{
"epoch": 0.51,
"learning_rate": 2.4528877188285516e-05,
"loss": 3.7113,
"step": 137000
},
{
"epoch": 0.51,
"learning_rate": 2.443591688605298e-05,
"loss": 3.7079,
"step": 137500
},
{
"epoch": 0.51,
"learning_rate": 2.4342956583820447e-05,
"loss": 3.7111,
"step": 138000
},
{
"epoch": 0.52,
"learning_rate": 2.424999628158791e-05,
"loss": 3.7081,
"step": 138500
},
{
"epoch": 0.52,
"learning_rate": 2.415703597935538e-05,
"loss": 3.7061,
"step": 139000
},
{
"epoch": 0.52,
"learning_rate": 2.4064075677122845e-05,
"loss": 3.7066,
"step": 139500
},
{
"epoch": 0.52,
"learning_rate": 2.3971115374890307e-05,
"loss": 3.7127,
"step": 140000
},
{
"epoch": 0.52,
"eval_accuracy": 0.3766884162574383,
"eval_loss": 3.609013319015503,
"eval_runtime": 5024.1177,
"eval_samples_per_second": 89.995,
"eval_steps_per_second": 1.406,
"step": 140000
},
{
"epoch": 0.52,
"learning_rate": 2.3878155072657773e-05,
"loss": 3.7094,
"step": 140500
},
{
"epoch": 0.52,
"learning_rate": 2.378519477042524e-05,
"loss": 3.6994,
"step": 141000
},
{
"epoch": 0.53,
"learning_rate": 2.3692234468192704e-05,
"loss": 3.7066,
"step": 141500
},
{
"epoch": 0.53,
"learning_rate": 2.359927416596017e-05,
"loss": 3.705,
"step": 142000
},
{
"epoch": 0.53,
"learning_rate": 2.3506313863727636e-05,
"loss": 3.7026,
"step": 142500
},
{
"epoch": 0.53,
"learning_rate": 2.3413353561495098e-05,
"loss": 3.7035,
"step": 143000
},
{
"epoch": 0.53,
"learning_rate": 2.3320393259262567e-05,
"loss": 3.7052,
"step": 143500
},
{
"epoch": 0.54,
"learning_rate": 2.3227432957030033e-05,
"loss": 3.7018,
"step": 144000
},
{
"epoch": 0.54,
"learning_rate": 2.3134472654797495e-05,
"loss": 3.7042,
"step": 144500
},
{
"epoch": 0.54,
"learning_rate": 2.304151235256496e-05,
"loss": 3.7033,
"step": 145000
},
{
"epoch": 0.54,
"learning_rate": 2.2948552050332427e-05,
"loss": 3.7057,
"step": 145500
},
{
"epoch": 0.54,
"learning_rate": 2.2855591748099893e-05,
"loss": 3.7018,
"step": 146000
},
{
"epoch": 0.54,
"learning_rate": 2.276263144586736e-05,
"loss": 3.7007,
"step": 146500
},
{
"epoch": 0.55,
"learning_rate": 2.2669671143634824e-05,
"loss": 3.6892,
"step": 147000
},
{
"epoch": 0.55,
"learning_rate": 2.2576710841402286e-05,
"loss": 3.7029,
"step": 147500
},
{
"epoch": 0.55,
"learning_rate": 2.2483750539169756e-05,
"loss": 3.6992,
"step": 148000
},
{
"epoch": 0.55,
"learning_rate": 2.239079023693722e-05,
"loss": 3.6961,
"step": 148500
},
{
"epoch": 0.55,
"learning_rate": 2.2297829934704684e-05,
"loss": 3.6916,
"step": 149000
},
{
"epoch": 0.56,
"learning_rate": 2.220486963247215e-05,
"loss": 3.6983,
"step": 149500
},
{
"epoch": 0.56,
"learning_rate": 2.2111909330239615e-05,
"loss": 3.694,
"step": 150000
},
{
"epoch": 0.56,
"eval_accuracy": 0.3782522490519747,
"eval_loss": 3.5962026119232178,
"eval_runtime": 5027.3439,
"eval_samples_per_second": 89.937,
"eval_steps_per_second": 1.405,
"step": 150000
},
{
"epoch": 0.56,
"learning_rate": 2.201894902800708e-05,
"loss": 3.6916,
"step": 150500
},
{
"epoch": 0.56,
"learning_rate": 2.1925988725774547e-05,
"loss": 3.69,
"step": 151000
},
{
"epoch": 0.56,
"learning_rate": 2.1833028423542012e-05,
"loss": 3.6858,
"step": 151500
},
{
"epoch": 0.57,
"learning_rate": 2.1740068121309475e-05,
"loss": 3.7009,
"step": 152000
},
{
"epoch": 0.57,
"learning_rate": 2.164710781907694e-05,
"loss": 3.6886,
"step": 152500
},
{
"epoch": 0.57,
"learning_rate": 2.155414751684441e-05,
"loss": 3.6869,
"step": 153000
},
{
"epoch": 0.57,
"learning_rate": 2.1461187214611872e-05,
"loss": 3.6898,
"step": 153500
},
{
"epoch": 0.57,
"learning_rate": 2.1368226912379338e-05,
"loss": 3.695,
"step": 154000
},
{
"epoch": 0.57,
"learning_rate": 2.1275266610146804e-05,
"loss": 3.6882,
"step": 154500
},
{
"epoch": 0.58,
"learning_rate": 2.118230630791427e-05,
"loss": 3.6909,
"step": 155000
},
{
"epoch": 0.58,
"learning_rate": 2.1089346005681735e-05,
"loss": 3.6862,
"step": 155500
},
{
"epoch": 0.58,
"learning_rate": 2.09963857034492e-05,
"loss": 3.6878,
"step": 156000
},
{
"epoch": 0.58,
"learning_rate": 2.0903425401216663e-05,
"loss": 3.6882,
"step": 156500
},
{
"epoch": 0.58,
"learning_rate": 2.081046509898413e-05,
"loss": 3.6841,
"step": 157000
},
{
"epoch": 0.59,
"learning_rate": 2.0717504796751598e-05,
"loss": 3.6828,
"step": 157500
},
{
"epoch": 0.59,
"learning_rate": 2.062454449451906e-05,
"loss": 3.6819,
"step": 158000
},
{
"epoch": 0.59,
"learning_rate": 2.0531584192286526e-05,
"loss": 3.6854,
"step": 158500
},
{
"epoch": 0.59,
"learning_rate": 2.0438623890053992e-05,
"loss": 3.6759,
"step": 159000
},
{
"epoch": 0.59,
"learning_rate": 2.0345663587821458e-05,
"loss": 3.6811,
"step": 159500
},
{
"epoch": 0.59,
"learning_rate": 2.0252703285588923e-05,
"loss": 3.6871,
"step": 160000
},
{
"epoch": 0.59,
"eval_accuracy": 0.37971800692359575,
"eval_loss": 3.583056688308716,
"eval_runtime": 5027.862,
"eval_samples_per_second": 89.928,
"eval_steps_per_second": 1.405,
"step": 160000
},
{
"epoch": 0.6,
"learning_rate": 2.015974298335639e-05,
"loss": 3.6786,
"step": 160500
},
{
"epoch": 0.6,
"learning_rate": 2.0066782681123855e-05,
"loss": 3.6793,
"step": 161000
},
{
"epoch": 0.6,
"learning_rate": 1.9973822378891317e-05,
"loss": 3.6791,
"step": 161500
},
{
"epoch": 0.6,
"learning_rate": 1.9880862076658786e-05,
"loss": 3.6806,
"step": 162000
},
{
"epoch": 0.6,
"learning_rate": 1.9787901774426252e-05,
"loss": 3.6757,
"step": 162500
},
{
"epoch": 0.61,
"learning_rate": 1.9694941472193715e-05,
"loss": 3.6772,
"step": 163000
},
{
"epoch": 0.61,
"learning_rate": 1.960198116996118e-05,
"loss": 3.6803,
"step": 163500
},
{
"epoch": 0.61,
"learning_rate": 1.9509020867728646e-05,
"loss": 3.674,
"step": 164000
},
{
"epoch": 0.61,
"learning_rate": 1.9416060565496112e-05,
"loss": 3.6761,
"step": 164500
},
{
"epoch": 0.61,
"learning_rate": 1.9323100263263578e-05,
"loss": 3.6718,
"step": 165000
},
{
"epoch": 0.62,
"learning_rate": 1.9230139961031043e-05,
"loss": 3.675,
"step": 165500
},
{
"epoch": 0.62,
"learning_rate": 1.9137179658798506e-05,
"loss": 3.6734,
"step": 166000
},
{
"epoch": 0.62,
"learning_rate": 1.9044219356565975e-05,
"loss": 3.6716,
"step": 166500
},
{
"epoch": 0.62,
"learning_rate": 1.895125905433344e-05,
"loss": 3.6724,
"step": 167000
},
{
"epoch": 0.62,
"learning_rate": 1.8858298752100903e-05,
"loss": 3.6742,
"step": 167500
},
{
"epoch": 0.62,
"learning_rate": 1.876533844986837e-05,
"loss": 3.6758,
"step": 168000
},
{
"epoch": 0.63,
"learning_rate": 1.8672378147635834e-05,
"loss": 3.6704,
"step": 168500
},
{
"epoch": 0.63,
"learning_rate": 1.85794178454033e-05,
"loss": 3.6657,
"step": 169000
},
{
"epoch": 0.63,
"learning_rate": 1.8486457543170766e-05,
"loss": 3.6662,
"step": 169500
},
{
"epoch": 0.63,
"learning_rate": 1.839349724093823e-05,
"loss": 3.6784,
"step": 170000
},
{
"epoch": 0.63,
"eval_accuracy": 0.3810095955450411,
"eval_loss": 3.5707802772521973,
"eval_runtime": 5019.3993,
"eval_samples_per_second": 90.08,
"eval_steps_per_second": 1.408,
"step": 170000
},
{
"epoch": 0.63,
"learning_rate": 1.8300536938705694e-05,
"loss": 3.672,
"step": 170500
},
{
"epoch": 0.64,
"learning_rate": 1.8207576636473163e-05,
"loss": 3.6694,
"step": 171000
},
{
"epoch": 0.64,
"learning_rate": 1.811461633424063e-05,
"loss": 3.6686,
"step": 171500
},
{
"epoch": 0.64,
"learning_rate": 1.802165603200809e-05,
"loss": 3.6712,
"step": 172000
},
{
"epoch": 0.64,
"learning_rate": 1.7928695729775557e-05,
"loss": 3.6743,
"step": 172500
},
{
"epoch": 0.64,
"learning_rate": 1.7835735427543023e-05,
"loss": 3.6632,
"step": 173000
},
{
"epoch": 0.65,
"learning_rate": 1.774277512531049e-05,
"loss": 3.6609,
"step": 173500
},
{
"epoch": 0.65,
"learning_rate": 1.7649814823077954e-05,
"loss": 3.668,
"step": 174000
},
{
"epoch": 0.65,
"learning_rate": 1.755685452084542e-05,
"loss": 3.6703,
"step": 174500
},
{
"epoch": 0.65,
"learning_rate": 1.7463894218612882e-05,
"loss": 3.6671,
"step": 175000
},
{
"epoch": 0.65,
"learning_rate": 1.7370933916380348e-05,
"loss": 3.66,
"step": 175500
},
{
"epoch": 0.65,
"learning_rate": 1.7277973614147817e-05,
"loss": 3.6648,
"step": 176000
},
{
"epoch": 0.66,
"learning_rate": 1.718501331191528e-05,
"loss": 3.6681,
"step": 176500
},
{
"epoch": 0.66,
"learning_rate": 1.7092053009682745e-05,
"loss": 3.6685,
"step": 177000
},
{
"epoch": 0.66,
"learning_rate": 1.699909270745021e-05,
"loss": 3.6669,
"step": 177500
},
{
"epoch": 0.66,
"learning_rate": 1.6906132405217677e-05,
"loss": 3.6691,
"step": 178000
},
{
"epoch": 0.66,
"learning_rate": 1.6813172102985143e-05,
"loss": 3.6575,
"step": 178500
},
{
"epoch": 0.67,
"learning_rate": 1.6720211800752608e-05,
"loss": 3.6688,
"step": 179000
},
{
"epoch": 0.67,
"learning_rate": 1.662725149852007e-05,
"loss": 3.6558,
"step": 179500
},
{
"epoch": 0.67,
"learning_rate": 1.6534291196287536e-05,
"loss": 3.6606,
"step": 180000
},
{
"epoch": 0.67,
"eval_accuracy": 0.38233432261147465,
"eval_loss": 3.559264898300171,
"eval_runtime": 5018.0904,
"eval_samples_per_second": 90.103,
"eval_steps_per_second": 1.408,
"step": 180000
},
{
"epoch": 0.67,
"learning_rate": 1.6441330894055006e-05,
"loss": 3.6577,
"step": 180500
},
{
"epoch": 0.67,
"learning_rate": 1.6348370591822468e-05,
"loss": 3.6628,
"step": 181000
},
{
"epoch": 0.67,
"learning_rate": 1.6255410289589934e-05,
"loss": 3.6661,
"step": 181500
},
{
"epoch": 0.68,
"learning_rate": 1.61624499873574e-05,
"loss": 3.6582,
"step": 182000
},
{
"epoch": 0.68,
"learning_rate": 1.6069489685124865e-05,
"loss": 3.6554,
"step": 182500
},
{
"epoch": 0.68,
"learning_rate": 1.597652938289233e-05,
"loss": 3.6577,
"step": 183000
},
{
"epoch": 0.68,
"learning_rate": 1.5883569080659797e-05,
"loss": 3.6586,
"step": 183500
},
{
"epoch": 0.68,
"learning_rate": 1.579060877842726e-05,
"loss": 3.6565,
"step": 184000
},
{
"epoch": 0.69,
"learning_rate": 1.5697648476194725e-05,
"loss": 3.6623,
"step": 184500
},
{
"epoch": 0.69,
"learning_rate": 1.5604688173962194e-05,
"loss": 3.6617,
"step": 185000
},
{
"epoch": 0.69,
"learning_rate": 1.5511727871729656e-05,
"loss": 3.6555,
"step": 185500
},
{
"epoch": 0.69,
"learning_rate": 1.5418767569497122e-05,
"loss": 3.6524,
"step": 186000
},
{
"epoch": 0.69,
"learning_rate": 1.5325807267264588e-05,
"loss": 3.6602,
"step": 186500
},
{
"epoch": 0.7,
"learning_rate": 1.5232846965032052e-05,
"loss": 3.6563,
"step": 187000
},
{
"epoch": 0.7,
"learning_rate": 1.5139886662799518e-05,
"loss": 3.6551,
"step": 187500
},
{
"epoch": 0.7,
"learning_rate": 1.5046926360566985e-05,
"loss": 3.6567,
"step": 188000
},
{
"epoch": 0.7,
"learning_rate": 1.4953966058334447e-05,
"loss": 3.6475,
"step": 188500
},
{
"epoch": 0.7,
"learning_rate": 1.4861005756101915e-05,
"loss": 3.654,
"step": 189000
},
{
"epoch": 0.7,
"learning_rate": 1.476804545386938e-05,
"loss": 3.6559,
"step": 189500
},
{
"epoch": 0.71,
"learning_rate": 1.4675085151636846e-05,
"loss": 3.646,
"step": 190000
},
{
"epoch": 0.71,
"eval_accuracy": 0.3834675999220702,
"eval_loss": 3.549065351486206,
"eval_runtime": 5019.1074,
"eval_samples_per_second": 90.085,
"eval_steps_per_second": 1.408,
"step": 190000
},
{
"epoch": 0.71,
"learning_rate": 1.458212484940431e-05,
"loss": 3.6544,
"step": 190500
},
{
"epoch": 0.71,
"learning_rate": 1.4489164547171776e-05,
"loss": 3.6484,
"step": 191000
},
{
"epoch": 0.71,
"learning_rate": 1.4396204244939244e-05,
"loss": 3.6467,
"step": 191500
},
{
"epoch": 0.71,
"learning_rate": 1.4303243942706706e-05,
"loss": 3.6558,
"step": 192000
},
{
"epoch": 0.72,
"learning_rate": 1.4210283640474173e-05,
"loss": 3.6461,
"step": 192500
},
{
"epoch": 0.72,
"learning_rate": 1.4117323338241639e-05,
"loss": 3.6532,
"step": 193000
},
{
"epoch": 0.72,
"learning_rate": 1.4024363036009103e-05,
"loss": 3.6551,
"step": 193500
},
{
"epoch": 0.72,
"learning_rate": 1.3931402733776569e-05,
"loss": 3.6476,
"step": 194000
},
{
"epoch": 0.72,
"learning_rate": 1.3838442431544035e-05,
"loss": 3.6466,
"step": 194500
},
{
"epoch": 0.73,
"learning_rate": 1.3745482129311499e-05,
"loss": 3.6518,
"step": 195000
},
{
"epoch": 0.73,
"learning_rate": 1.3652521827078964e-05,
"loss": 3.6413,
"step": 195500
},
{
"epoch": 0.73,
"learning_rate": 1.3559561524846432e-05,
"loss": 3.6432,
"step": 196000
},
{
"epoch": 0.73,
"learning_rate": 1.3466601222613894e-05,
"loss": 3.6464,
"step": 196500
},
{
"epoch": 0.73,
"learning_rate": 1.3373640920381362e-05,
"loss": 3.644,
"step": 197000
},
{
"epoch": 0.73,
"learning_rate": 1.3280680618148827e-05,
"loss": 3.6496,
"step": 197500
},
{
"epoch": 0.74,
"learning_rate": 1.3187720315916292e-05,
"loss": 3.6407,
"step": 198000
},
{
"epoch": 0.74,
"learning_rate": 1.3094760013683757e-05,
"loss": 3.6412,
"step": 198500
},
{
"epoch": 0.74,
"learning_rate": 1.3001799711451223e-05,
"loss": 3.646,
"step": 199000
},
{
"epoch": 0.74,
"learning_rate": 1.2908839409218687e-05,
"loss": 3.6432,
"step": 199500
},
{
"epoch": 0.74,
"learning_rate": 1.2815879106986153e-05,
"loss": 3.6453,
"step": 200000
},
{
"epoch": 0.74,
"eval_accuracy": 0.38430268095273507,
"eval_loss": 3.5410099029541016,
"eval_runtime": 5035.174,
"eval_samples_per_second": 89.797,
"eval_steps_per_second": 1.403,
"step": 200000
},
{
"epoch": 0.75,
"learning_rate": 1.272291880475362e-05,
"loss": 3.6374,
"step": 200500
},
{
"epoch": 0.75,
"learning_rate": 1.2629958502521083e-05,
"loss": 3.6464,
"step": 201000
},
{
"epoch": 0.75,
"learning_rate": 1.253699820028855e-05,
"loss": 3.6488,
"step": 201500
},
{
"epoch": 0.75,
"learning_rate": 1.2444037898056014e-05,
"loss": 3.6405,
"step": 202000
},
{
"epoch": 0.75,
"learning_rate": 1.235107759582348e-05,
"loss": 3.6354,
"step": 202500
},
{
"epoch": 0.75,
"learning_rate": 1.2258117293590946e-05,
"loss": 3.6386,
"step": 203000
},
{
"epoch": 0.76,
"learning_rate": 1.216515699135841e-05,
"loss": 3.6349,
"step": 203500
},
{
"epoch": 0.76,
"learning_rate": 1.2072196689125877e-05,
"loss": 3.6377,
"step": 204000
},
{
"epoch": 0.76,
"learning_rate": 1.1979236386893341e-05,
"loss": 3.6374,
"step": 204500
},
{
"epoch": 0.76,
"learning_rate": 1.1886276084660807e-05,
"loss": 3.6403,
"step": 205000
},
{
"epoch": 0.76,
"learning_rate": 1.1793315782428273e-05,
"loss": 3.6397,
"step": 205500
},
{
"epoch": 0.77,
"learning_rate": 1.1700355480195738e-05,
"loss": 3.6388,
"step": 206000
},
{
"epoch": 0.77,
"learning_rate": 1.1607395177963202e-05,
"loss": 3.6413,
"step": 206500
},
{
"epoch": 0.77,
"learning_rate": 1.1514434875730668e-05,
"loss": 3.6353,
"step": 207000
},
{
"epoch": 0.77,
"learning_rate": 1.1421474573498134e-05,
"loss": 3.6395,
"step": 207500
},
{
"epoch": 0.77,
"learning_rate": 1.1328514271265598e-05,
"loss": 3.6378,
"step": 208000
},
{
"epoch": 0.78,
"learning_rate": 1.1235553969033065e-05,
"loss": 3.6394,
"step": 208500
},
{
"epoch": 0.78,
"learning_rate": 1.114259366680053e-05,
"loss": 3.6402,
"step": 209000
},
{
"epoch": 0.78,
"learning_rate": 1.1049633364567995e-05,
"loss": 3.6421,
"step": 209500
},
{
"epoch": 0.78,
"learning_rate": 1.0956673062335461e-05,
"loss": 3.6393,
"step": 210000
},
{
"epoch": 0.78,
"eval_accuracy": 0.38505967217174975,
"eval_loss": 3.5341877937316895,
"eval_runtime": 5017.9224,
"eval_samples_per_second": 90.106,
"eval_steps_per_second": 1.408,
"step": 210000
},
{
"epoch": 0.78,
"learning_rate": 1.0863712760102927e-05,
"loss": 3.6379,
"step": 210500
},
{
"epoch": 0.78,
"learning_rate": 1.0770752457870393e-05,
"loss": 3.6319,
"step": 211000
},
{
"epoch": 0.79,
"learning_rate": 1.0677792155637857e-05,
"loss": 3.6318,
"step": 211500
},
{
"epoch": 0.79,
"learning_rate": 1.0584831853405322e-05,
"loss": 3.6345,
"step": 212000
},
{
"epoch": 0.79,
"learning_rate": 1.0491871551172788e-05,
"loss": 3.6346,
"step": 212500
},
{
"epoch": 0.79,
"learning_rate": 1.0398911248940254e-05,
"loss": 3.6304,
"step": 213000
},
{
"epoch": 0.79,
"learning_rate": 1.0305950946707718e-05,
"loss": 3.6273,
"step": 213500
},
{
"epoch": 0.8,
"learning_rate": 1.0212990644475184e-05,
"loss": 3.6346,
"step": 214000
},
{
"epoch": 0.8,
"learning_rate": 1.012003034224265e-05,
"loss": 3.6329,
"step": 214500
},
{
"epoch": 0.8,
"learning_rate": 1.0027070040010113e-05,
"loss": 3.6349,
"step": 215000
},
{
"epoch": 0.8,
"learning_rate": 9.934109737777581e-06,
"loss": 3.6369,
"step": 215500
},
{
"epoch": 0.8,
"learning_rate": 9.841149435545045e-06,
"loss": 3.6272,
"step": 216000
},
{
"epoch": 0.81,
"learning_rate": 9.74818913331251e-06,
"loss": 3.6283,
"step": 216500
},
{
"epoch": 0.81,
"learning_rate": 9.655228831079976e-06,
"loss": 3.6353,
"step": 217000
},
{
"epoch": 0.81,
"learning_rate": 9.562268528847442e-06,
"loss": 3.6251,
"step": 217500
},
{
"epoch": 0.81,
"learning_rate": 9.469308226614906e-06,
"loss": 3.6345,
"step": 218000
},
{
"epoch": 0.81,
"learning_rate": 9.376347924382372e-06,
"loss": 3.6256,
"step": 218500
},
{
"epoch": 0.81,
"learning_rate": 9.283387622149838e-06,
"loss": 3.6259,
"step": 219000
},
{
"epoch": 0.82,
"learning_rate": 9.190427319917302e-06,
"loss": 3.6288,
"step": 219500
},
{
"epoch": 0.82,
"learning_rate": 9.09746701768477e-06,
"loss": 3.6207,
"step": 220000
},
{
"epoch": 0.82,
"eval_accuracy": 0.3857340442835604,
"eval_loss": 3.5280263423919678,
"eval_runtime": 5028.331,
"eval_samples_per_second": 89.919,
"eval_steps_per_second": 1.405,
"step": 220000
},
{
"epoch": 0.82,
"learning_rate": 9.004506715452233e-06,
"loss": 3.6311,
"step": 220500
},
{
"epoch": 0.82,
"learning_rate": 8.911546413219699e-06,
"loss": 3.6254,
"step": 221000
},
{
"epoch": 0.82,
"learning_rate": 8.818586110987165e-06,
"loss": 3.6307,
"step": 221500
},
{
"epoch": 0.83,
"learning_rate": 8.72562580875463e-06,
"loss": 3.629,
"step": 222000
},
{
"epoch": 0.83,
"learning_rate": 8.632665506522095e-06,
"loss": 3.6264,
"step": 222500
},
{
"epoch": 0.83,
"learning_rate": 8.53970520428956e-06,
"loss": 3.6215,
"step": 223000
},
{
"epoch": 0.83,
"learning_rate": 8.446744902057026e-06,
"loss": 3.6181,
"step": 223500
},
{
"epoch": 0.83,
"learning_rate": 8.35378459982449e-06,
"loss": 3.6253,
"step": 224000
},
{
"epoch": 0.83,
"learning_rate": 8.260824297591958e-06,
"loss": 3.6204,
"step": 224500
},
{
"epoch": 0.84,
"learning_rate": 8.167863995359422e-06,
"loss": 3.6264,
"step": 225000
},
{
"epoch": 0.84,
"learning_rate": 8.074903693126887e-06,
"loss": 3.6258,
"step": 225500
},
{
"epoch": 0.84,
"learning_rate": 7.981943390894353e-06,
"loss": 3.6279,
"step": 226000
},
{
"epoch": 0.84,
"learning_rate": 7.888983088661817e-06,
"loss": 3.6294,
"step": 226500
},
{
"epoch": 0.84,
"learning_rate": 7.796022786429285e-06,
"loss": 3.621,
"step": 227000
},
{
"epoch": 0.85,
"learning_rate": 7.703062484196749e-06,
"loss": 3.6315,
"step": 227500
},
{
"epoch": 0.85,
"learning_rate": 7.610102181964214e-06,
"loss": 3.6199,
"step": 228000
},
{
"epoch": 0.85,
"learning_rate": 7.51714187973168e-06,
"loss": 3.6182,
"step": 228500
},
{
"epoch": 0.85,
"learning_rate": 7.424181577499145e-06,
"loss": 3.6196,
"step": 229000
},
{
"epoch": 0.85,
"learning_rate": 7.33122127526661e-06,
"loss": 3.6216,
"step": 229500
},
{
"epoch": 0.86,
"learning_rate": 7.2382609730340766e-06,
"loss": 3.6288,
"step": 230000
},
{
"epoch": 0.86,
"eval_accuracy": 0.386489052989915,
"eval_loss": 3.5217700004577637,
"eval_runtime": 5019.3728,
"eval_samples_per_second": 90.08,
"eval_steps_per_second": 1.408,
"step": 230000
},
{
"epoch": 0.86,
"learning_rate": 7.1453006708015415e-06,
"loss": 3.6201,
"step": 230500
},
{
"epoch": 0.86,
"learning_rate": 7.052340368569006e-06,
"loss": 3.6216,
"step": 231000
},
{
"epoch": 0.86,
"learning_rate": 6.959380066336472e-06,
"loss": 3.6214,
"step": 231500
},
{
"epoch": 0.86,
"learning_rate": 6.866419764103937e-06,
"loss": 3.6208,
"step": 232000
},
{
"epoch": 0.86,
"learning_rate": 6.773459461871402e-06,
"loss": 3.6116,
"step": 232500
},
{
"epoch": 0.87,
"learning_rate": 6.6804991596388685e-06,
"loss": 3.6187,
"step": 233000
},
{
"epoch": 0.87,
"learning_rate": 6.587538857406333e-06,
"loss": 3.6179,
"step": 233500
},
{
"epoch": 0.87,
"learning_rate": 6.494578555173798e-06,
"loss": 3.6112,
"step": 234000
},
{
"epoch": 0.87,
"learning_rate": 6.401618252941265e-06,
"loss": 3.6226,
"step": 234500
},
{
"epoch": 0.87,
"learning_rate": 6.30865795070873e-06,
"loss": 3.6148,
"step": 235000
},
{
"epoch": 0.88,
"learning_rate": 6.2156976484761956e-06,
"loss": 3.6107,
"step": 235500
},
{
"epoch": 0.88,
"learning_rate": 6.1227373462436605e-06,
"loss": 3.6129,
"step": 236000
},
{
"epoch": 0.88,
"learning_rate": 6.029777044011125e-06,
"loss": 3.6178,
"step": 236500
},
{
"epoch": 0.88,
"learning_rate": 5.936816741778591e-06,
"loss": 3.624,
"step": 237000
},
{
"epoch": 0.88,
"learning_rate": 5.843856439546057e-06,
"loss": 3.6232,
"step": 237500
},
{
"epoch": 0.88,
"learning_rate": 5.750896137313522e-06,
"loss": 3.6185,
"step": 238000
},
{
"epoch": 0.89,
"learning_rate": 5.6579358350809875e-06,
"loss": 3.6195,
"step": 238500
},
{
"epoch": 0.89,
"learning_rate": 5.564975532848453e-06,
"loss": 3.6236,
"step": 239000
},
{
"epoch": 0.89,
"learning_rate": 5.472015230615918e-06,
"loss": 3.6091,
"step": 239500
},
{
"epoch": 0.89,
"learning_rate": 5.379054928383383e-06,
"loss": 3.6176,
"step": 240000
},
{
"epoch": 0.89,
"eval_accuracy": 0.3872213352261681,
"eval_loss": 3.5150856971740723,
"eval_runtime": 5028.7528,
"eval_samples_per_second": 89.912,
"eval_steps_per_second": 1.405,
"step": 240000
},
{
"epoch": 0.89,
"learning_rate": 5.286094626150849e-06,
"loss": 3.6167,
"step": 240500
},
{
"epoch": 0.9,
"learning_rate": 5.193134323918314e-06,
"loss": 3.62,
"step": 241000
},
{
"epoch": 0.9,
"learning_rate": 5.1001740216857795e-06,
"loss": 3.612,
"step": 241500
},
{
"epoch": 0.9,
"learning_rate": 5.007213719453245e-06,
"loss": 3.614,
"step": 242000
},
{
"epoch": 0.9,
"learning_rate": 4.91425341722071e-06,
"loss": 3.6181,
"step": 242500
},
{
"epoch": 0.9,
"learning_rate": 4.821293114988176e-06,
"loss": 3.6146,
"step": 243000
},
{
"epoch": 0.91,
"learning_rate": 4.728332812755641e-06,
"loss": 3.6145,
"step": 243500
},
{
"epoch": 0.91,
"learning_rate": 4.6353725105231065e-06,
"loss": 3.6156,
"step": 244000
},
{
"epoch": 0.91,
"learning_rate": 4.542412208290571e-06,
"loss": 3.6149,
"step": 244500
},
{
"epoch": 0.91,
"learning_rate": 4.449451906058037e-06,
"loss": 3.613,
"step": 245000
},
{
"epoch": 0.91,
"learning_rate": 4.356491603825503e-06,
"loss": 3.6158,
"step": 245500
},
{
"epoch": 0.91,
"learning_rate": 4.263531301592968e-06,
"loss": 3.6231,
"step": 246000
},
{
"epoch": 0.92,
"learning_rate": 4.1705709993604336e-06,
"loss": 3.6094,
"step": 246500
},
{
"epoch": 0.92,
"learning_rate": 4.077610697127899e-06,
"loss": 3.6135,
"step": 247000
},
{
"epoch": 0.92,
"learning_rate": 3.984650394895364e-06,
"loss": 3.6115,
"step": 247500
},
{
"epoch": 0.92,
"learning_rate": 3.891690092662829e-06,
"loss": 3.6093,
"step": 248000
},
{
"epoch": 0.92,
"learning_rate": 3.798729790430295e-06,
"loss": 3.6147,
"step": 248500
},
{
"epoch": 0.93,
"learning_rate": 3.7057694881977598e-06,
"loss": 3.6174,
"step": 249000
},
{
"epoch": 0.93,
"learning_rate": 3.6128091859652255e-06,
"loss": 3.6175,
"step": 249500
},
{
"epoch": 0.93,
"learning_rate": 3.5198488837326913e-06,
"loss": 3.6099,
"step": 250000
},
{
"epoch": 0.93,
"eval_accuracy": 0.3877557812917544,
"eval_loss": 3.510763168334961,
"eval_runtime": 5030.6308,
"eval_samples_per_second": 89.878,
"eval_steps_per_second": 1.404,
"step": 250000
},
{
"epoch": 0.93,
"learning_rate": 3.4268885815001566e-06,
"loss": 3.6052,
"step": 250500
},
{
"epoch": 0.93,
"learning_rate": 3.3339282792676215e-06,
"loss": 3.6146,
"step": 251000
},
{
"epoch": 0.94,
"learning_rate": 3.2409679770350872e-06,
"loss": 3.6118,
"step": 251500
},
{
"epoch": 0.94,
"learning_rate": 3.1480076748025526e-06,
"loss": 3.6099,
"step": 252000
},
{
"epoch": 0.94,
"learning_rate": 3.055047372570018e-06,
"loss": 3.6089,
"step": 252500
},
{
"epoch": 0.94,
"learning_rate": 2.9620870703374832e-06,
"loss": 3.6042,
"step": 253000
},
{
"epoch": 0.94,
"learning_rate": 2.8691267681049485e-06,
"loss": 3.6121,
"step": 253500
},
{
"epoch": 0.94,
"learning_rate": 2.7761664658724143e-06,
"loss": 3.6041,
"step": 254000
},
{
"epoch": 0.95,
"learning_rate": 2.683206163639879e-06,
"loss": 3.6135,
"step": 254500
},
{
"epoch": 0.95,
"learning_rate": 2.5902458614073445e-06,
"loss": 3.6043,
"step": 255000
},
{
"epoch": 0.95,
"learning_rate": 2.4972855591748103e-06,
"loss": 3.6029,
"step": 255500
},
{
"epoch": 0.95,
"learning_rate": 2.4043252569422756e-06,
"loss": 3.6112,
"step": 256000
},
{
"epoch": 0.95,
"learning_rate": 2.311364954709741e-06,
"loss": 3.6021,
"step": 256500
},
{
"epoch": 0.96,
"learning_rate": 2.2184046524772062e-06,
"loss": 3.6078,
"step": 257000
},
{
"epoch": 0.96,
"learning_rate": 2.1254443502446716e-06,
"loss": 3.6037,
"step": 257500
},
{
"epoch": 0.96,
"learning_rate": 2.032484048012137e-06,
"loss": 3.6054,
"step": 258000
},
{
"epoch": 0.96,
"learning_rate": 1.9395237457796022e-06,
"loss": 3.6063,
"step": 258500
},
{
"epoch": 0.96,
"learning_rate": 1.8465634435470675e-06,
"loss": 3.6049,
"step": 259000
},
{
"epoch": 0.96,
"learning_rate": 1.753603141314533e-06,
"loss": 3.6056,
"step": 259500
},
{
"epoch": 0.97,
"learning_rate": 1.6606428390819984e-06,
"loss": 3.6093,
"step": 260000
},
{
"epoch": 0.97,
"eval_accuracy": 0.3880700733260521,
"eval_loss": 3.507866144180298,
"eval_runtime": 5028.8015,
"eval_samples_per_second": 89.911,
"eval_steps_per_second": 1.405,
"step": 260000
},
{
"epoch": 0.97,
"learning_rate": 1.567682536849464e-06,
"loss": 3.6062,
"step": 260500
},
{
"epoch": 0.97,
"learning_rate": 1.4747222346169293e-06,
"loss": 3.6027,
"step": 261000
},
{
"epoch": 0.97,
"learning_rate": 1.3817619323843946e-06,
"loss": 3.6005,
"step": 261500
},
{
"epoch": 0.97,
"learning_rate": 1.28880163015186e-06,
"loss": 3.6119,
"step": 262000
},
{
"epoch": 0.98,
"learning_rate": 1.1958413279193255e-06,
"loss": 3.6036,
"step": 262500
},
{
"epoch": 0.98,
"learning_rate": 1.1028810256867908e-06,
"loss": 3.6089,
"step": 263000
},
{
"epoch": 0.98,
"learning_rate": 1.009920723454256e-06,
"loss": 3.6052,
"step": 263500
},
{
"epoch": 0.98,
"learning_rate": 9.169604212217216e-07,
"loss": 3.6063,
"step": 264000
},
{
"epoch": 0.98,
"learning_rate": 8.240001189891869e-07,
"loss": 3.6115,
"step": 264500
},
{
"epoch": 0.99,
"learning_rate": 7.310398167566523e-07,
"loss": 3.6035,
"step": 265000
},
{
"epoch": 0.99,
"learning_rate": 6.380795145241177e-07,
"loss": 3.5975,
"step": 265500
},
{
"epoch": 0.99,
"learning_rate": 5.45119212291583e-07,
"loss": 3.6047,
"step": 266000
},
{
"epoch": 0.99,
"learning_rate": 4.521589100590484e-07,
"loss": 3.6039,
"step": 266500
},
{
"epoch": 0.99,
"learning_rate": 3.5919860782651375e-07,
"loss": 3.6111,
"step": 267000
},
{
"epoch": 0.99,
"learning_rate": 2.6623830559397913e-07,
"loss": 3.6062,
"step": 267500
},
{
"epoch": 1.0,
"learning_rate": 1.7327800336144456e-07,
"loss": 3.6008,
"step": 268000
},
{
"epoch": 1.0,
"learning_rate": 8.031770112890991e-08,
"loss": 3.6071,
"step": 268500
},
{
"epoch": 1.0,
"step": 268932,
"total_flos": 4.497266479988736e+18,
"train_loss": 3.8604874901614594,
"train_runtime": 258151.997,
"train_samples_per_second": 33.336,
"train_steps_per_second": 1.042
}
],
"logging_steps": 500,
"max_steps": 268932,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 10000,
"total_flos": 4.497266479988736e+18,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}