gpt2-ts_cx-cs_00000-00019_50k / trainer_state.json
jonasknobloch's picture
Initial commit
faccc14
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"eval_steps": 10000,
"global_step": 274994,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 4.990908892557656e-05,
"loss": 8.099,
"step": 500
},
{
"epoch": 0.0,
"learning_rate": 4.981817785115312e-05,
"loss": 7.2102,
"step": 1000
},
{
"epoch": 0.01,
"learning_rate": 4.9727266776729675e-05,
"loss": 6.762,
"step": 1500
},
{
"epoch": 0.01,
"learning_rate": 4.9636355702306236e-05,
"loss": 6.4126,
"step": 2000
},
{
"epoch": 0.01,
"learning_rate": 4.954544462788279e-05,
"loss": 6.1281,
"step": 2500
},
{
"epoch": 0.01,
"learning_rate": 4.945453355345935e-05,
"loss": 5.8943,
"step": 3000
},
{
"epoch": 0.01,
"learning_rate": 4.936362247903591e-05,
"loss": 5.6943,
"step": 3500
},
{
"epoch": 0.01,
"learning_rate": 4.9272711404612464e-05,
"loss": 5.5168,
"step": 4000
},
{
"epoch": 0.02,
"learning_rate": 4.9181800330189025e-05,
"loss": 5.3501,
"step": 4500
},
{
"epoch": 0.02,
"learning_rate": 4.909088925576558e-05,
"loss": 5.2239,
"step": 5000
},
{
"epoch": 0.02,
"learning_rate": 4.899997818134214e-05,
"loss": 5.1144,
"step": 5500
},
{
"epoch": 0.02,
"learning_rate": 4.89090671069187e-05,
"loss": 5.0334,
"step": 6000
},
{
"epoch": 0.02,
"learning_rate": 4.881815603249525e-05,
"loss": 4.9547,
"step": 6500
},
{
"epoch": 0.03,
"learning_rate": 4.8727244958071815e-05,
"loss": 4.8913,
"step": 7000
},
{
"epoch": 0.03,
"learning_rate": 4.863633388364837e-05,
"loss": 4.8409,
"step": 7500
},
{
"epoch": 0.03,
"learning_rate": 4.854542280922493e-05,
"loss": 4.7906,
"step": 8000
},
{
"epoch": 0.03,
"learning_rate": 4.845451173480149e-05,
"loss": 4.7443,
"step": 8500
},
{
"epoch": 0.03,
"learning_rate": 4.836360066037804e-05,
"loss": 4.705,
"step": 9000
},
{
"epoch": 0.03,
"learning_rate": 4.8272689585954604e-05,
"loss": 4.6654,
"step": 9500
},
{
"epoch": 0.04,
"learning_rate": 4.818177851153116e-05,
"loss": 4.6338,
"step": 10000
},
{
"epoch": 0.04,
"eval_accuracy": 0.29678543332895735,
"eval_loss": 4.5132622718811035,
"eval_runtime": 5196.2601,
"eval_samples_per_second": 88.97,
"eval_steps_per_second": 1.39,
"step": 10000
},
{
"epoch": 0.04,
"learning_rate": 4.809086743710772e-05,
"loss": 4.6023,
"step": 10500
},
{
"epoch": 0.04,
"learning_rate": 4.7999956362684276e-05,
"loss": 4.5757,
"step": 11000
},
{
"epoch": 0.04,
"learning_rate": 4.790904528826084e-05,
"loss": 4.5454,
"step": 11500
},
{
"epoch": 0.04,
"learning_rate": 4.781813421383739e-05,
"loss": 4.517,
"step": 12000
},
{
"epoch": 0.05,
"learning_rate": 4.772722313941395e-05,
"loss": 4.4973,
"step": 12500
},
{
"epoch": 0.05,
"learning_rate": 4.763631206499051e-05,
"loss": 4.4756,
"step": 13000
},
{
"epoch": 0.05,
"learning_rate": 4.7545400990567065e-05,
"loss": 4.4538,
"step": 13500
},
{
"epoch": 0.05,
"learning_rate": 4.745448991614363e-05,
"loss": 4.4379,
"step": 14000
},
{
"epoch": 0.05,
"learning_rate": 4.736357884172019e-05,
"loss": 4.4163,
"step": 14500
},
{
"epoch": 0.05,
"learning_rate": 4.7272667767296744e-05,
"loss": 4.397,
"step": 15000
},
{
"epoch": 0.06,
"learning_rate": 4.7181756692873306e-05,
"loss": 4.3884,
"step": 15500
},
{
"epoch": 0.06,
"learning_rate": 4.709084561844986e-05,
"loss": 4.3621,
"step": 16000
},
{
"epoch": 0.06,
"learning_rate": 4.699993454402642e-05,
"loss": 4.3503,
"step": 16500
},
{
"epoch": 0.06,
"learning_rate": 4.690902346960298e-05,
"loss": 4.3406,
"step": 17000
},
{
"epoch": 0.06,
"learning_rate": 4.681811239517953e-05,
"loss": 4.3222,
"step": 17500
},
{
"epoch": 0.07,
"learning_rate": 4.6727201320756095e-05,
"loss": 4.3134,
"step": 18000
},
{
"epoch": 0.07,
"learning_rate": 4.663629024633265e-05,
"loss": 4.2956,
"step": 18500
},
{
"epoch": 0.07,
"learning_rate": 4.654537917190921e-05,
"loss": 4.2854,
"step": 19000
},
{
"epoch": 0.07,
"learning_rate": 4.645446809748577e-05,
"loss": 4.2743,
"step": 19500
},
{
"epoch": 0.07,
"learning_rate": 4.636355702306233e-05,
"loss": 4.2588,
"step": 20000
},
{
"epoch": 0.07,
"eval_accuracy": 0.3283652796999879,
"eval_loss": 4.153079032897949,
"eval_runtime": 5192.4762,
"eval_samples_per_second": 89.035,
"eval_steps_per_second": 1.391,
"step": 20000
},
{
"epoch": 0.07,
"learning_rate": 4.6272645948638884e-05,
"loss": 4.2551,
"step": 20500
},
{
"epoch": 0.08,
"learning_rate": 4.618173487421544e-05,
"loss": 4.2418,
"step": 21000
},
{
"epoch": 0.08,
"learning_rate": 4.6090823799792e-05,
"loss": 4.2276,
"step": 21500
},
{
"epoch": 0.08,
"learning_rate": 4.5999912725368556e-05,
"loss": 4.2163,
"step": 22000
},
{
"epoch": 0.08,
"learning_rate": 4.590900165094512e-05,
"loss": 4.2055,
"step": 22500
},
{
"epoch": 0.08,
"learning_rate": 4.581809057652167e-05,
"loss": 4.1955,
"step": 23000
},
{
"epoch": 0.09,
"learning_rate": 4.572717950209823e-05,
"loss": 4.1867,
"step": 23500
},
{
"epoch": 0.09,
"learning_rate": 4.563626842767479e-05,
"loss": 4.1839,
"step": 24000
},
{
"epoch": 0.09,
"learning_rate": 4.5545357353251345e-05,
"loss": 4.1728,
"step": 24500
},
{
"epoch": 0.09,
"learning_rate": 4.545444627882791e-05,
"loss": 4.1605,
"step": 25000
},
{
"epoch": 0.09,
"learning_rate": 4.536353520440446e-05,
"loss": 4.158,
"step": 25500
},
{
"epoch": 0.09,
"learning_rate": 4.527262412998102e-05,
"loss": 4.1553,
"step": 26000
},
{
"epoch": 0.1,
"learning_rate": 4.518171305555758e-05,
"loss": 4.1388,
"step": 26500
},
{
"epoch": 0.1,
"learning_rate": 4.5090801981134134e-05,
"loss": 4.1304,
"step": 27000
},
{
"epoch": 0.1,
"learning_rate": 4.4999890906710696e-05,
"loss": 4.1231,
"step": 27500
},
{
"epoch": 0.1,
"learning_rate": 4.490897983228725e-05,
"loss": 4.1182,
"step": 28000
},
{
"epoch": 0.1,
"learning_rate": 4.481806875786381e-05,
"loss": 4.1205,
"step": 28500
},
{
"epoch": 0.11,
"learning_rate": 4.472715768344037e-05,
"loss": 4.1053,
"step": 29000
},
{
"epoch": 0.11,
"learning_rate": 4.463624660901692e-05,
"loss": 4.101,
"step": 29500
},
{
"epoch": 0.11,
"learning_rate": 4.4545335534593485e-05,
"loss": 4.0955,
"step": 30000
},
{
"epoch": 0.11,
"eval_accuracy": 0.34322272271152204,
"eval_loss": 3.990593433380127,
"eval_runtime": 5193.4093,
"eval_samples_per_second": 89.019,
"eval_steps_per_second": 1.391,
"step": 30000
},
{
"epoch": 0.11,
"learning_rate": 4.445442446017004e-05,
"loss": 4.0833,
"step": 30500
},
{
"epoch": 0.11,
"learning_rate": 4.43635133857466e-05,
"loss": 4.0821,
"step": 31000
},
{
"epoch": 0.11,
"learning_rate": 4.427260231132316e-05,
"loss": 4.0703,
"step": 31500
},
{
"epoch": 0.12,
"learning_rate": 4.418169123689971e-05,
"loss": 4.0654,
"step": 32000
},
{
"epoch": 0.12,
"learning_rate": 4.4090780162476274e-05,
"loss": 4.063,
"step": 32500
},
{
"epoch": 0.12,
"learning_rate": 4.399986908805283e-05,
"loss": 4.0545,
"step": 33000
},
{
"epoch": 0.12,
"learning_rate": 4.390895801362939e-05,
"loss": 4.0507,
"step": 33500
},
{
"epoch": 0.12,
"learning_rate": 4.3818046939205946e-05,
"loss": 4.0434,
"step": 34000
},
{
"epoch": 0.13,
"learning_rate": 4.372713586478251e-05,
"loss": 4.0374,
"step": 34500
},
{
"epoch": 0.13,
"learning_rate": 4.3636224790359063e-05,
"loss": 4.0303,
"step": 35000
},
{
"epoch": 0.13,
"learning_rate": 4.354531371593562e-05,
"loss": 4.0336,
"step": 35500
},
{
"epoch": 0.13,
"learning_rate": 4.345440264151218e-05,
"loss": 4.0259,
"step": 36000
},
{
"epoch": 0.13,
"learning_rate": 4.3363491567088736e-05,
"loss": 4.0219,
"step": 36500
},
{
"epoch": 0.13,
"learning_rate": 4.32725804926653e-05,
"loss": 4.0159,
"step": 37000
},
{
"epoch": 0.14,
"learning_rate": 4.318166941824185e-05,
"loss": 4.0035,
"step": 37500
},
{
"epoch": 0.14,
"learning_rate": 4.309075834381841e-05,
"loss": 4.0016,
"step": 38000
},
{
"epoch": 0.14,
"learning_rate": 4.299984726939497e-05,
"loss": 3.9975,
"step": 38500
},
{
"epoch": 0.14,
"learning_rate": 4.2908936194971525e-05,
"loss": 3.998,
"step": 39000
},
{
"epoch": 0.14,
"learning_rate": 4.2818025120548087e-05,
"loss": 3.9885,
"step": 39500
},
{
"epoch": 0.15,
"learning_rate": 4.272711404612464e-05,
"loss": 3.9884,
"step": 40000
},
{
"epoch": 0.15,
"eval_accuracy": 0.35303792801557193,
"eval_loss": 3.886583089828491,
"eval_runtime": 5193.0489,
"eval_samples_per_second": 89.025,
"eval_steps_per_second": 1.391,
"step": 40000
},
{
"epoch": 0.15,
"learning_rate": 4.26362029717012e-05,
"loss": 3.9807,
"step": 40500
},
{
"epoch": 0.15,
"learning_rate": 4.254529189727776e-05,
"loss": 3.9839,
"step": 41000
},
{
"epoch": 0.15,
"learning_rate": 4.245438082285432e-05,
"loss": 3.9719,
"step": 41500
},
{
"epoch": 0.15,
"learning_rate": 4.2363469748430876e-05,
"loss": 3.9641,
"step": 42000
},
{
"epoch": 0.15,
"learning_rate": 4.227255867400744e-05,
"loss": 3.9639,
"step": 42500
},
{
"epoch": 0.16,
"learning_rate": 4.218164759958399e-05,
"loss": 3.9692,
"step": 43000
},
{
"epoch": 0.16,
"learning_rate": 4.2090736525160555e-05,
"loss": 3.9574,
"step": 43500
},
{
"epoch": 0.16,
"learning_rate": 4.199982545073711e-05,
"loss": 3.9617,
"step": 44000
},
{
"epoch": 0.16,
"learning_rate": 4.190891437631367e-05,
"loss": 3.9504,
"step": 44500
},
{
"epoch": 0.16,
"learning_rate": 4.181800330189023e-05,
"loss": 3.9451,
"step": 45000
},
{
"epoch": 0.17,
"learning_rate": 4.172709222746679e-05,
"loss": 3.9377,
"step": 45500
},
{
"epoch": 0.17,
"learning_rate": 4.1636181153043344e-05,
"loss": 3.9423,
"step": 46000
},
{
"epoch": 0.17,
"learning_rate": 4.15452700786199e-05,
"loss": 3.935,
"step": 46500
},
{
"epoch": 0.17,
"learning_rate": 4.145435900419646e-05,
"loss": 3.9314,
"step": 47000
},
{
"epoch": 0.17,
"learning_rate": 4.1363447929773016e-05,
"loss": 3.9329,
"step": 47500
},
{
"epoch": 0.17,
"learning_rate": 4.127253685534958e-05,
"loss": 3.9271,
"step": 48000
},
{
"epoch": 0.18,
"learning_rate": 4.118162578092613e-05,
"loss": 3.9203,
"step": 48500
},
{
"epoch": 0.18,
"learning_rate": 4.1090714706502695e-05,
"loss": 3.9244,
"step": 49000
},
{
"epoch": 0.18,
"learning_rate": 4.099980363207925e-05,
"loss": 3.9084,
"step": 49500
},
{
"epoch": 0.18,
"learning_rate": 4.0908892557655805e-05,
"loss": 3.914,
"step": 50000
},
{
"epoch": 0.18,
"eval_accuracy": 0.36014139406297696,
"eval_loss": 3.814393997192383,
"eval_runtime": 5187.7852,
"eval_samples_per_second": 89.115,
"eval_steps_per_second": 1.393,
"step": 50000
},
{
"epoch": 0.18,
"learning_rate": 4.081798148323237e-05,
"loss": 3.9057,
"step": 50500
},
{
"epoch": 0.19,
"learning_rate": 4.072707040880892e-05,
"loss": 3.915,
"step": 51000
},
{
"epoch": 0.19,
"learning_rate": 4.0636159334385484e-05,
"loss": 3.9058,
"step": 51500
},
{
"epoch": 0.19,
"learning_rate": 4.054524825996204e-05,
"loss": 3.8986,
"step": 52000
},
{
"epoch": 0.19,
"learning_rate": 4.0454337185538594e-05,
"loss": 3.8985,
"step": 52500
},
{
"epoch": 0.19,
"learning_rate": 4.0363426111115156e-05,
"loss": 3.898,
"step": 53000
},
{
"epoch": 0.19,
"learning_rate": 4.027251503669171e-05,
"loss": 3.8909,
"step": 53500
},
{
"epoch": 0.2,
"learning_rate": 4.018160396226827e-05,
"loss": 3.8841,
"step": 54000
},
{
"epoch": 0.2,
"learning_rate": 4.009069288784483e-05,
"loss": 3.8852,
"step": 54500
},
{
"epoch": 0.2,
"learning_rate": 3.999978181342138e-05,
"loss": 3.88,
"step": 55000
},
{
"epoch": 0.2,
"learning_rate": 3.9908870738997945e-05,
"loss": 3.8836,
"step": 55500
},
{
"epoch": 0.2,
"learning_rate": 3.98179596645745e-05,
"loss": 3.874,
"step": 56000
},
{
"epoch": 0.21,
"learning_rate": 3.972704859015106e-05,
"loss": 3.8757,
"step": 56500
},
{
"epoch": 0.21,
"learning_rate": 3.963613751572762e-05,
"loss": 3.8762,
"step": 57000
},
{
"epoch": 0.21,
"learning_rate": 3.954522644130418e-05,
"loss": 3.8681,
"step": 57500
},
{
"epoch": 0.21,
"learning_rate": 3.9454315366880734e-05,
"loss": 3.875,
"step": 58000
},
{
"epoch": 0.21,
"learning_rate": 3.936340429245729e-05,
"loss": 3.8619,
"step": 58500
},
{
"epoch": 0.21,
"learning_rate": 3.927249321803385e-05,
"loss": 3.8595,
"step": 59000
},
{
"epoch": 0.22,
"learning_rate": 3.9181582143610406e-05,
"loss": 3.8633,
"step": 59500
},
{
"epoch": 0.22,
"learning_rate": 3.909067106918697e-05,
"loss": 3.8563,
"step": 60000
},
{
"epoch": 0.22,
"eval_accuracy": 0.3656418963030083,
"eval_loss": 3.7592036724090576,
"eval_runtime": 5193.0699,
"eval_samples_per_second": 89.025,
"eval_steps_per_second": 1.391,
"step": 60000
},
{
"epoch": 0.22,
"learning_rate": 3.899975999476352e-05,
"loss": 3.8497,
"step": 60500
},
{
"epoch": 0.22,
"learning_rate": 3.890884892034008e-05,
"loss": 3.8458,
"step": 61000
},
{
"epoch": 0.22,
"learning_rate": 3.881793784591664e-05,
"loss": 3.8475,
"step": 61500
},
{
"epoch": 0.23,
"learning_rate": 3.8727026771493195e-05,
"loss": 3.8524,
"step": 62000
},
{
"epoch": 0.23,
"learning_rate": 3.863611569706976e-05,
"loss": 3.8529,
"step": 62500
},
{
"epoch": 0.23,
"learning_rate": 3.854520462264631e-05,
"loss": 3.8483,
"step": 63000
},
{
"epoch": 0.23,
"learning_rate": 3.845429354822287e-05,
"loss": 3.8458,
"step": 63500
},
{
"epoch": 0.23,
"learning_rate": 3.836338247379943e-05,
"loss": 3.8403,
"step": 64000
},
{
"epoch": 0.23,
"learning_rate": 3.8272471399375984e-05,
"loss": 3.8351,
"step": 64500
},
{
"epoch": 0.24,
"learning_rate": 3.8181560324952546e-05,
"loss": 3.828,
"step": 65000
},
{
"epoch": 0.24,
"learning_rate": 3.80906492505291e-05,
"loss": 3.8338,
"step": 65500
},
{
"epoch": 0.24,
"learning_rate": 3.799973817610566e-05,
"loss": 3.8235,
"step": 66000
},
{
"epoch": 0.24,
"learning_rate": 3.790882710168222e-05,
"loss": 3.8338,
"step": 66500
},
{
"epoch": 0.24,
"learning_rate": 3.7817916027258774e-05,
"loss": 3.8247,
"step": 67000
},
{
"epoch": 0.25,
"learning_rate": 3.7727004952835335e-05,
"loss": 3.8199,
"step": 67500
},
{
"epoch": 0.25,
"learning_rate": 3.763609387841189e-05,
"loss": 3.8179,
"step": 68000
},
{
"epoch": 0.25,
"learning_rate": 3.754518280398845e-05,
"loss": 3.8142,
"step": 68500
},
{
"epoch": 0.25,
"learning_rate": 3.745427172956501e-05,
"loss": 3.8153,
"step": 69000
},
{
"epoch": 0.25,
"learning_rate": 3.736336065514157e-05,
"loss": 3.8175,
"step": 69500
},
{
"epoch": 0.25,
"learning_rate": 3.7272449580718124e-05,
"loss": 3.8136,
"step": 70000
},
{
"epoch": 0.25,
"eval_accuracy": 0.3701075745828096,
"eval_loss": 3.713653087615967,
"eval_runtime": 5189.9064,
"eval_samples_per_second": 89.079,
"eval_steps_per_second": 1.392,
"step": 70000
},
{
"epoch": 0.26,
"learning_rate": 3.7181538506294686e-05,
"loss": 3.8074,
"step": 70500
},
{
"epoch": 0.26,
"learning_rate": 3.709062743187124e-05,
"loss": 3.8123,
"step": 71000
},
{
"epoch": 0.26,
"learning_rate": 3.69997163574478e-05,
"loss": 3.8035,
"step": 71500
},
{
"epoch": 0.26,
"learning_rate": 3.690880528302436e-05,
"loss": 3.8004,
"step": 72000
},
{
"epoch": 0.26,
"learning_rate": 3.681789420860092e-05,
"loss": 3.8008,
"step": 72500
},
{
"epoch": 0.27,
"learning_rate": 3.6726983134177475e-05,
"loss": 3.7991,
"step": 73000
},
{
"epoch": 0.27,
"learning_rate": 3.663607205975404e-05,
"loss": 3.8076,
"step": 73500
},
{
"epoch": 0.27,
"learning_rate": 3.654516098533059e-05,
"loss": 3.796,
"step": 74000
},
{
"epoch": 0.27,
"learning_rate": 3.6454249910907154e-05,
"loss": 3.7911,
"step": 74500
},
{
"epoch": 0.27,
"learning_rate": 3.636333883648371e-05,
"loss": 3.7931,
"step": 75000
},
{
"epoch": 0.27,
"learning_rate": 3.6272427762060265e-05,
"loss": 3.7883,
"step": 75500
},
{
"epoch": 0.28,
"learning_rate": 3.6181516687636826e-05,
"loss": 3.7893,
"step": 76000
},
{
"epoch": 0.28,
"learning_rate": 3.609060561321338e-05,
"loss": 3.7877,
"step": 76500
},
{
"epoch": 0.28,
"learning_rate": 3.5999694538789943e-05,
"loss": 3.7863,
"step": 77000
},
{
"epoch": 0.28,
"learning_rate": 3.59087834643665e-05,
"loss": 3.7826,
"step": 77500
},
{
"epoch": 0.28,
"learning_rate": 3.5817872389943054e-05,
"loss": 3.7878,
"step": 78000
},
{
"epoch": 0.29,
"learning_rate": 3.5726961315519616e-05,
"loss": 3.7856,
"step": 78500
},
{
"epoch": 0.29,
"learning_rate": 3.563605024109617e-05,
"loss": 3.7817,
"step": 79000
},
{
"epoch": 0.29,
"learning_rate": 3.554513916667273e-05,
"loss": 3.7754,
"step": 79500
},
{
"epoch": 0.29,
"learning_rate": 3.545422809224929e-05,
"loss": 3.7762,
"step": 80000
},
{
"epoch": 0.29,
"eval_accuracy": 0.37404703542213524,
"eval_loss": 3.6766245365142822,
"eval_runtime": 5182.779,
"eval_samples_per_second": 89.202,
"eval_steps_per_second": 1.394,
"step": 80000
},
{
"epoch": 0.29,
"learning_rate": 3.536331701782585e-05,
"loss": 3.7748,
"step": 80500
},
{
"epoch": 0.29,
"learning_rate": 3.5272405943402405e-05,
"loss": 3.7659,
"step": 81000
},
{
"epoch": 0.3,
"learning_rate": 3.518149486897896e-05,
"loss": 3.7766,
"step": 81500
},
{
"epoch": 0.3,
"learning_rate": 3.509058379455552e-05,
"loss": 3.7767,
"step": 82000
},
{
"epoch": 0.3,
"learning_rate": 3.499967272013208e-05,
"loss": 3.7671,
"step": 82500
},
{
"epoch": 0.3,
"learning_rate": 3.490876164570864e-05,
"loss": 3.7676,
"step": 83000
},
{
"epoch": 0.3,
"learning_rate": 3.4817850571285194e-05,
"loss": 3.7649,
"step": 83500
},
{
"epoch": 0.31,
"learning_rate": 3.472693949686175e-05,
"loss": 3.7634,
"step": 84000
},
{
"epoch": 0.31,
"learning_rate": 3.463602842243831e-05,
"loss": 3.7662,
"step": 84500
},
{
"epoch": 0.31,
"learning_rate": 3.4545117348014866e-05,
"loss": 3.7587,
"step": 85000
},
{
"epoch": 0.31,
"learning_rate": 3.445420627359143e-05,
"loss": 3.7557,
"step": 85500
},
{
"epoch": 0.31,
"learning_rate": 3.436329519916798e-05,
"loss": 3.7531,
"step": 86000
},
{
"epoch": 0.31,
"learning_rate": 3.4272384124744545e-05,
"loss": 3.7589,
"step": 86500
},
{
"epoch": 0.32,
"learning_rate": 3.41814730503211e-05,
"loss": 3.7513,
"step": 87000
},
{
"epoch": 0.32,
"learning_rate": 3.4090561975897655e-05,
"loss": 3.7595,
"step": 87500
},
{
"epoch": 0.32,
"learning_rate": 3.399965090147422e-05,
"loss": 3.7532,
"step": 88000
},
{
"epoch": 0.32,
"learning_rate": 3.390873982705077e-05,
"loss": 3.753,
"step": 88500
},
{
"epoch": 0.32,
"learning_rate": 3.3817828752627334e-05,
"loss": 3.7432,
"step": 89000
},
{
"epoch": 0.33,
"learning_rate": 3.372691767820389e-05,
"loss": 3.7441,
"step": 89500
},
{
"epoch": 0.33,
"learning_rate": 3.3636006603780444e-05,
"loss": 3.7481,
"step": 90000
},
{
"epoch": 0.33,
"eval_accuracy": 0.3772804989980488,
"eval_loss": 3.646758794784546,
"eval_runtime": 5190.3572,
"eval_samples_per_second": 89.071,
"eval_steps_per_second": 1.392,
"step": 90000
},
{
"epoch": 0.33,
"learning_rate": 3.3545095529357006e-05,
"loss": 3.7435,
"step": 90500
},
{
"epoch": 0.33,
"learning_rate": 3.345418445493356e-05,
"loss": 3.7478,
"step": 91000
},
{
"epoch": 0.33,
"learning_rate": 3.336327338051012e-05,
"loss": 3.7444,
"step": 91500
},
{
"epoch": 0.33,
"learning_rate": 3.327236230608668e-05,
"loss": 3.7496,
"step": 92000
},
{
"epoch": 0.34,
"learning_rate": 3.318145123166323e-05,
"loss": 3.7387,
"step": 92500
},
{
"epoch": 0.34,
"learning_rate": 3.3090540157239795e-05,
"loss": 3.7361,
"step": 93000
},
{
"epoch": 0.34,
"learning_rate": 3.299962908281635e-05,
"loss": 3.732,
"step": 93500
},
{
"epoch": 0.34,
"learning_rate": 3.290871800839291e-05,
"loss": 3.7308,
"step": 94000
},
{
"epoch": 0.34,
"learning_rate": 3.281780693396947e-05,
"loss": 3.7348,
"step": 94500
},
{
"epoch": 0.35,
"learning_rate": 3.272689585954603e-05,
"loss": 3.7356,
"step": 95000
},
{
"epoch": 0.35,
"learning_rate": 3.2635984785122584e-05,
"loss": 3.7335,
"step": 95500
},
{
"epoch": 0.35,
"learning_rate": 3.254507371069914e-05,
"loss": 3.733,
"step": 96000
},
{
"epoch": 0.35,
"learning_rate": 3.24541626362757e-05,
"loss": 3.7295,
"step": 96500
},
{
"epoch": 0.35,
"learning_rate": 3.2363251561852256e-05,
"loss": 3.7281,
"step": 97000
},
{
"epoch": 0.35,
"learning_rate": 3.227234048742882e-05,
"loss": 3.7294,
"step": 97500
},
{
"epoch": 0.36,
"learning_rate": 3.218142941300537e-05,
"loss": 3.7181,
"step": 98000
},
{
"epoch": 0.36,
"learning_rate": 3.2090518338581935e-05,
"loss": 3.7196,
"step": 98500
},
{
"epoch": 0.36,
"learning_rate": 3.199960726415849e-05,
"loss": 3.7254,
"step": 99000
},
{
"epoch": 0.36,
"learning_rate": 3.190869618973505e-05,
"loss": 3.7199,
"step": 99500
},
{
"epoch": 0.36,
"learning_rate": 3.181778511531161e-05,
"loss": 3.7199,
"step": 100000
},
{
"epoch": 0.36,
"eval_accuracy": 0.3800426605894803,
"eval_loss": 3.6193912029266357,
"eval_runtime": 5183.2235,
"eval_samples_per_second": 89.194,
"eval_steps_per_second": 1.394,
"step": 100000
},
{
"epoch": 0.37,
"learning_rate": 3.172687404088817e-05,
"loss": 3.7263,
"step": 100500
},
{
"epoch": 0.37,
"learning_rate": 3.1635962966464724e-05,
"loss": 3.7218,
"step": 101000
},
{
"epoch": 0.37,
"learning_rate": 3.1545051892041286e-05,
"loss": 3.7115,
"step": 101500
},
{
"epoch": 0.37,
"learning_rate": 3.145414081761784e-05,
"loss": 3.7159,
"step": 102000
},
{
"epoch": 0.37,
"learning_rate": 3.13632297431944e-05,
"loss": 3.7107,
"step": 102500
},
{
"epoch": 0.37,
"learning_rate": 3.127231866877096e-05,
"loss": 3.7126,
"step": 103000
},
{
"epoch": 0.38,
"learning_rate": 3.118140759434752e-05,
"loss": 3.7106,
"step": 103500
},
{
"epoch": 0.38,
"learning_rate": 3.1090496519924075e-05,
"loss": 3.7136,
"step": 104000
},
{
"epoch": 0.38,
"learning_rate": 3.099958544550063e-05,
"loss": 3.7057,
"step": 104500
},
{
"epoch": 0.38,
"learning_rate": 3.090867437107719e-05,
"loss": 3.7126,
"step": 105000
},
{
"epoch": 0.38,
"learning_rate": 3.081776329665375e-05,
"loss": 3.7075,
"step": 105500
},
{
"epoch": 0.39,
"learning_rate": 3.072685222223031e-05,
"loss": 3.7041,
"step": 106000
},
{
"epoch": 0.39,
"learning_rate": 3.0635941147806864e-05,
"loss": 3.7002,
"step": 106500
},
{
"epoch": 0.39,
"learning_rate": 3.054503007338342e-05,
"loss": 3.6996,
"step": 107000
},
{
"epoch": 0.39,
"learning_rate": 3.0454118998959978e-05,
"loss": 3.7001,
"step": 107500
},
{
"epoch": 0.39,
"learning_rate": 3.0363207924536537e-05,
"loss": 3.6985,
"step": 108000
},
{
"epoch": 0.39,
"learning_rate": 3.0272296850113095e-05,
"loss": 3.6977,
"step": 108500
},
{
"epoch": 0.4,
"learning_rate": 3.0181385775689654e-05,
"loss": 3.6997,
"step": 109000
},
{
"epoch": 0.4,
"learning_rate": 3.0090474701266212e-05,
"loss": 3.6931,
"step": 109500
},
{
"epoch": 0.4,
"learning_rate": 2.999956362684277e-05,
"loss": 3.6886,
"step": 110000
},
{
"epoch": 0.4,
"eval_accuracy": 0.3823600835290051,
"eval_loss": 3.596660852432251,
"eval_runtime": 5184.5518,
"eval_samples_per_second": 89.171,
"eval_steps_per_second": 1.393,
"step": 110000
},
{
"epoch": 0.4,
"learning_rate": 2.9908652552419326e-05,
"loss": 3.6943,
"step": 110500
},
{
"epoch": 0.4,
"learning_rate": 2.9817741477995888e-05,
"loss": 3.6929,
"step": 111000
},
{
"epoch": 0.41,
"learning_rate": 2.9726830403572443e-05,
"loss": 3.6972,
"step": 111500
},
{
"epoch": 0.41,
"learning_rate": 2.9635919329149005e-05,
"loss": 3.6917,
"step": 112000
},
{
"epoch": 0.41,
"learning_rate": 2.954500825472556e-05,
"loss": 3.6865,
"step": 112500
},
{
"epoch": 0.41,
"learning_rate": 2.9454097180302115e-05,
"loss": 3.6944,
"step": 113000
},
{
"epoch": 0.41,
"learning_rate": 2.9363186105878677e-05,
"loss": 3.682,
"step": 113500
},
{
"epoch": 0.41,
"learning_rate": 2.9272275031455232e-05,
"loss": 3.693,
"step": 114000
},
{
"epoch": 0.42,
"learning_rate": 2.9181363957031794e-05,
"loss": 3.6839,
"step": 114500
},
{
"epoch": 0.42,
"learning_rate": 2.909045288260835e-05,
"loss": 3.6838,
"step": 115000
},
{
"epoch": 0.42,
"learning_rate": 2.8999541808184904e-05,
"loss": 3.6873,
"step": 115500
},
{
"epoch": 0.42,
"learning_rate": 2.8908630733761466e-05,
"loss": 3.6841,
"step": 116000
},
{
"epoch": 0.42,
"learning_rate": 2.881771965933802e-05,
"loss": 3.675,
"step": 116500
},
{
"epoch": 0.43,
"learning_rate": 2.8726808584914583e-05,
"loss": 3.6874,
"step": 117000
},
{
"epoch": 0.43,
"learning_rate": 2.8635897510491138e-05,
"loss": 3.6798,
"step": 117500
},
{
"epoch": 0.43,
"learning_rate": 2.85449864360677e-05,
"loss": 3.6808,
"step": 118000
},
{
"epoch": 0.43,
"learning_rate": 2.8454075361644255e-05,
"loss": 3.68,
"step": 118500
},
{
"epoch": 0.43,
"learning_rate": 2.836316428722081e-05,
"loss": 3.6735,
"step": 119000
},
{
"epoch": 0.43,
"learning_rate": 2.8272253212797372e-05,
"loss": 3.673,
"step": 119500
},
{
"epoch": 0.44,
"learning_rate": 2.8181342138373927e-05,
"loss": 3.677,
"step": 120000
},
{
"epoch": 0.44,
"eval_accuracy": 0.38427083988271826,
"eval_loss": 3.578906297683716,
"eval_runtime": 5195.9268,
"eval_samples_per_second": 88.976,
"eval_steps_per_second": 1.39,
"step": 120000
},
{
"epoch": 0.44,
"learning_rate": 2.809043106395049e-05,
"loss": 3.6737,
"step": 120500
},
{
"epoch": 0.44,
"learning_rate": 2.7999519989527044e-05,
"loss": 3.6648,
"step": 121000
},
{
"epoch": 0.44,
"learning_rate": 2.7908608915103602e-05,
"loss": 3.6685,
"step": 121500
},
{
"epoch": 0.44,
"learning_rate": 2.781769784068016e-05,
"loss": 3.6699,
"step": 122000
},
{
"epoch": 0.45,
"learning_rate": 2.772678676625672e-05,
"loss": 3.6733,
"step": 122500
},
{
"epoch": 0.45,
"learning_rate": 2.7635875691833278e-05,
"loss": 3.6709,
"step": 123000
},
{
"epoch": 0.45,
"learning_rate": 2.7544964617409836e-05,
"loss": 3.6721,
"step": 123500
},
{
"epoch": 0.45,
"learning_rate": 2.745405354298639e-05,
"loss": 3.6635,
"step": 124000
},
{
"epoch": 0.45,
"learning_rate": 2.7363142468562953e-05,
"loss": 3.6677,
"step": 124500
},
{
"epoch": 0.45,
"learning_rate": 2.727223139413951e-05,
"loss": 3.6657,
"step": 125000
},
{
"epoch": 0.46,
"learning_rate": 2.718132031971607e-05,
"loss": 3.6624,
"step": 125500
},
{
"epoch": 0.46,
"learning_rate": 2.7090409245292626e-05,
"loss": 3.6656,
"step": 126000
},
{
"epoch": 0.46,
"learning_rate": 2.6999498170869187e-05,
"loss": 3.6654,
"step": 126500
},
{
"epoch": 0.46,
"learning_rate": 2.6908587096445743e-05,
"loss": 3.6641,
"step": 127000
},
{
"epoch": 0.46,
"learning_rate": 2.6817676022022298e-05,
"loss": 3.6579,
"step": 127500
},
{
"epoch": 0.47,
"learning_rate": 2.672676494759886e-05,
"loss": 3.6616,
"step": 128000
},
{
"epoch": 0.47,
"learning_rate": 2.6635853873175415e-05,
"loss": 3.6612,
"step": 128500
},
{
"epoch": 0.47,
"learning_rate": 2.6544942798751977e-05,
"loss": 3.6632,
"step": 129000
},
{
"epoch": 0.47,
"learning_rate": 2.645403172432853e-05,
"loss": 3.657,
"step": 129500
},
{
"epoch": 0.47,
"learning_rate": 2.6363120649905087e-05,
"loss": 3.6611,
"step": 130000
},
{
"epoch": 0.47,
"eval_accuracy": 0.38627043105731984,
"eval_loss": 3.5599796772003174,
"eval_runtime": 5193.7894,
"eval_samples_per_second": 89.012,
"eval_steps_per_second": 1.391,
"step": 130000
},
{
"epoch": 0.47,
"learning_rate": 2.627220957548165e-05,
"loss": 3.6568,
"step": 130500
},
{
"epoch": 0.48,
"learning_rate": 2.6181298501058204e-05,
"loss": 3.6501,
"step": 131000
},
{
"epoch": 0.48,
"learning_rate": 2.6090387426634766e-05,
"loss": 3.6523,
"step": 131500
},
{
"epoch": 0.48,
"learning_rate": 2.599947635221132e-05,
"loss": 3.6604,
"step": 132000
},
{
"epoch": 0.48,
"learning_rate": 2.5908565277787883e-05,
"loss": 3.6545,
"step": 132500
},
{
"epoch": 0.48,
"learning_rate": 2.5817654203364438e-05,
"loss": 3.652,
"step": 133000
},
{
"epoch": 0.49,
"learning_rate": 2.5726743128940993e-05,
"loss": 3.6503,
"step": 133500
},
{
"epoch": 0.49,
"learning_rate": 2.5635832054517555e-05,
"loss": 3.6562,
"step": 134000
},
{
"epoch": 0.49,
"learning_rate": 2.554492098009411e-05,
"loss": 3.6497,
"step": 134500
},
{
"epoch": 0.49,
"learning_rate": 2.5454009905670672e-05,
"loss": 3.6527,
"step": 135000
},
{
"epoch": 0.49,
"learning_rate": 2.5363098831247227e-05,
"loss": 3.651,
"step": 135500
},
{
"epoch": 0.49,
"learning_rate": 2.5272187756823785e-05,
"loss": 3.6489,
"step": 136000
},
{
"epoch": 0.5,
"learning_rate": 2.5181276682400344e-05,
"loss": 3.6379,
"step": 136500
},
{
"epoch": 0.5,
"learning_rate": 2.5090365607976902e-05,
"loss": 3.6407,
"step": 137000
},
{
"epoch": 0.5,
"learning_rate": 2.499945453355346e-05,
"loss": 3.6423,
"step": 137500
},
{
"epoch": 0.5,
"learning_rate": 2.490854345913002e-05,
"loss": 3.6444,
"step": 138000
},
{
"epoch": 0.5,
"learning_rate": 2.4817632384706578e-05,
"loss": 3.643,
"step": 138500
},
{
"epoch": 0.51,
"learning_rate": 2.4726721310283136e-05,
"loss": 3.6412,
"step": 139000
},
{
"epoch": 0.51,
"learning_rate": 2.4635810235859695e-05,
"loss": 3.6382,
"step": 139500
},
{
"epoch": 0.51,
"learning_rate": 2.454489916143625e-05,
"loss": 3.6442,
"step": 140000
},
{
"epoch": 0.51,
"eval_accuracy": 0.3879397112192979,
"eval_loss": 3.5442757606506348,
"eval_runtime": 5193.2385,
"eval_samples_per_second": 89.022,
"eval_steps_per_second": 1.391,
"step": 140000
},
{
"epoch": 0.51,
"learning_rate": 2.445398808701281e-05,
"loss": 3.6382,
"step": 140500
},
{
"epoch": 0.51,
"learning_rate": 2.4363077012589367e-05,
"loss": 3.6415,
"step": 141000
},
{
"epoch": 0.51,
"learning_rate": 2.4272165938165925e-05,
"loss": 3.6385,
"step": 141500
},
{
"epoch": 0.52,
"learning_rate": 2.4181254863742484e-05,
"loss": 3.6377,
"step": 142000
},
{
"epoch": 0.52,
"learning_rate": 2.4090343789319042e-05,
"loss": 3.6403,
"step": 142500
},
{
"epoch": 0.52,
"learning_rate": 2.3999432714895598e-05,
"loss": 3.6393,
"step": 143000
},
{
"epoch": 0.52,
"learning_rate": 2.3908521640472156e-05,
"loss": 3.6345,
"step": 143500
},
{
"epoch": 0.52,
"learning_rate": 2.3817610566048715e-05,
"loss": 3.6379,
"step": 144000
},
{
"epoch": 0.53,
"learning_rate": 2.3726699491625273e-05,
"loss": 3.642,
"step": 144500
},
{
"epoch": 0.53,
"learning_rate": 2.363578841720183e-05,
"loss": 3.6306,
"step": 145000
},
{
"epoch": 0.53,
"learning_rate": 2.354487734277839e-05,
"loss": 3.6331,
"step": 145500
},
{
"epoch": 0.53,
"learning_rate": 2.3453966268354945e-05,
"loss": 3.6313,
"step": 146000
},
{
"epoch": 0.53,
"learning_rate": 2.3363055193931504e-05,
"loss": 3.6298,
"step": 146500
},
{
"epoch": 0.53,
"learning_rate": 2.3272144119508062e-05,
"loss": 3.6309,
"step": 147000
},
{
"epoch": 0.54,
"learning_rate": 2.318123304508462e-05,
"loss": 3.6316,
"step": 147500
},
{
"epoch": 0.54,
"learning_rate": 2.309032197066118e-05,
"loss": 3.637,
"step": 148000
},
{
"epoch": 0.54,
"learning_rate": 2.2999410896237734e-05,
"loss": 3.6294,
"step": 148500
},
{
"epoch": 0.54,
"learning_rate": 2.2908499821814293e-05,
"loss": 3.6295,
"step": 149000
},
{
"epoch": 0.54,
"learning_rate": 2.281758874739085e-05,
"loss": 3.6265,
"step": 149500
},
{
"epoch": 0.55,
"learning_rate": 2.272667767296741e-05,
"loss": 3.6285,
"step": 150000
},
{
"epoch": 0.55,
"eval_accuracy": 0.38935745059803717,
"eval_loss": 3.5313050746917725,
"eval_runtime": 5185.4328,
"eval_samples_per_second": 89.156,
"eval_steps_per_second": 1.393,
"step": 150000
},
{
"epoch": 0.55,
"learning_rate": 2.2635766598543968e-05,
"loss": 3.6235,
"step": 150500
},
{
"epoch": 0.55,
"learning_rate": 2.2544855524120527e-05,
"loss": 3.6299,
"step": 151000
},
{
"epoch": 0.55,
"learning_rate": 2.2453944449697085e-05,
"loss": 3.6261,
"step": 151500
},
{
"epoch": 0.55,
"learning_rate": 2.2363033375273644e-05,
"loss": 3.626,
"step": 152000
},
{
"epoch": 0.55,
"learning_rate": 2.2272122300850202e-05,
"loss": 3.6231,
"step": 152500
},
{
"epoch": 0.56,
"learning_rate": 2.218121122642676e-05,
"loss": 3.6222,
"step": 153000
},
{
"epoch": 0.56,
"learning_rate": 2.209030015200332e-05,
"loss": 3.6189,
"step": 153500
},
{
"epoch": 0.56,
"learning_rate": 2.1999389077579878e-05,
"loss": 3.6239,
"step": 154000
},
{
"epoch": 0.56,
"learning_rate": 2.1908478003156433e-05,
"loss": 3.6231,
"step": 154500
},
{
"epoch": 0.56,
"learning_rate": 2.181756692873299e-05,
"loss": 3.6192,
"step": 155000
},
{
"epoch": 0.57,
"learning_rate": 2.172665585430955e-05,
"loss": 3.6203,
"step": 155500
},
{
"epoch": 0.57,
"learning_rate": 2.163574477988611e-05,
"loss": 3.62,
"step": 156000
},
{
"epoch": 0.57,
"learning_rate": 2.1544833705462667e-05,
"loss": 3.6183,
"step": 156500
},
{
"epoch": 0.57,
"learning_rate": 2.1453922631039225e-05,
"loss": 3.62,
"step": 157000
},
{
"epoch": 0.57,
"learning_rate": 2.136301155661578e-05,
"loss": 3.6114,
"step": 157500
},
{
"epoch": 0.57,
"learning_rate": 2.127210048219234e-05,
"loss": 3.6203,
"step": 158000
},
{
"epoch": 0.58,
"learning_rate": 2.1181189407768898e-05,
"loss": 3.617,
"step": 158500
},
{
"epoch": 0.58,
"learning_rate": 2.1090278333345456e-05,
"loss": 3.6135,
"step": 159000
},
{
"epoch": 0.58,
"learning_rate": 2.0999367258922015e-05,
"loss": 3.6167,
"step": 159500
},
{
"epoch": 0.58,
"learning_rate": 2.0908456184498573e-05,
"loss": 3.6126,
"step": 160000
},
{
"epoch": 0.58,
"eval_accuracy": 0.390987320272403,
"eval_loss": 3.517598867416382,
"eval_runtime": 5192.2207,
"eval_samples_per_second": 89.039,
"eval_steps_per_second": 1.391,
"step": 160000
},
{
"epoch": 0.58,
"learning_rate": 2.0817545110075128e-05,
"loss": 3.6135,
"step": 160500
},
{
"epoch": 0.59,
"learning_rate": 2.0726634035651687e-05,
"loss": 3.6206,
"step": 161000
},
{
"epoch": 0.59,
"learning_rate": 2.0635722961228245e-05,
"loss": 3.6114,
"step": 161500
},
{
"epoch": 0.59,
"learning_rate": 2.0544811886804804e-05,
"loss": 3.6137,
"step": 162000
},
{
"epoch": 0.59,
"learning_rate": 2.0453900812381362e-05,
"loss": 3.6132,
"step": 162500
},
{
"epoch": 0.59,
"learning_rate": 2.0362989737957917e-05,
"loss": 3.608,
"step": 163000
},
{
"epoch": 0.59,
"learning_rate": 2.0272078663534476e-05,
"loss": 3.6167,
"step": 163500
},
{
"epoch": 0.6,
"learning_rate": 2.0181167589111034e-05,
"loss": 3.6111,
"step": 164000
},
{
"epoch": 0.6,
"learning_rate": 2.0090256514687593e-05,
"loss": 3.61,
"step": 164500
},
{
"epoch": 0.6,
"learning_rate": 1.999934544026415e-05,
"loss": 3.6119,
"step": 165000
},
{
"epoch": 0.6,
"learning_rate": 1.990843436584071e-05,
"loss": 3.608,
"step": 165500
},
{
"epoch": 0.6,
"learning_rate": 1.9817523291417268e-05,
"loss": 3.6096,
"step": 166000
},
{
"epoch": 0.61,
"learning_rate": 1.9726612216993827e-05,
"loss": 3.6036,
"step": 166500
},
{
"epoch": 0.61,
"learning_rate": 1.9635701142570385e-05,
"loss": 3.5971,
"step": 167000
},
{
"epoch": 0.61,
"learning_rate": 1.9544790068146944e-05,
"loss": 3.6046,
"step": 167500
},
{
"epoch": 0.61,
"learning_rate": 1.9453878993723502e-05,
"loss": 3.6058,
"step": 168000
},
{
"epoch": 0.61,
"learning_rate": 1.936296791930006e-05,
"loss": 3.6085,
"step": 168500
},
{
"epoch": 0.61,
"learning_rate": 1.9272056844876616e-05,
"loss": 3.6052,
"step": 169000
},
{
"epoch": 0.62,
"learning_rate": 1.9181145770453174e-05,
"loss": 3.6022,
"step": 169500
},
{
"epoch": 0.62,
"learning_rate": 1.9090234696029733e-05,
"loss": 3.6051,
"step": 170000
},
{
"epoch": 0.62,
"eval_accuracy": 0.39213097714310974,
"eval_loss": 3.506274938583374,
"eval_runtime": 5194.6737,
"eval_samples_per_second": 88.997,
"eval_steps_per_second": 1.391,
"step": 170000
},
{
"epoch": 0.62,
"learning_rate": 1.899932362160629e-05,
"loss": 3.5994,
"step": 170500
},
{
"epoch": 0.62,
"learning_rate": 1.890841254718285e-05,
"loss": 3.6081,
"step": 171000
},
{
"epoch": 0.62,
"learning_rate": 1.881750147275941e-05,
"loss": 3.5989,
"step": 171500
},
{
"epoch": 0.63,
"learning_rate": 1.8726590398335963e-05,
"loss": 3.6068,
"step": 172000
},
{
"epoch": 0.63,
"learning_rate": 1.8635679323912522e-05,
"loss": 3.6018,
"step": 172500
},
{
"epoch": 0.63,
"learning_rate": 1.854476824948908e-05,
"loss": 3.5997,
"step": 173000
},
{
"epoch": 0.63,
"learning_rate": 1.845385717506564e-05,
"loss": 3.5935,
"step": 173500
},
{
"epoch": 0.63,
"learning_rate": 1.8362946100642197e-05,
"loss": 3.5985,
"step": 174000
},
{
"epoch": 0.63,
"learning_rate": 1.8272035026218753e-05,
"loss": 3.5942,
"step": 174500
},
{
"epoch": 0.64,
"learning_rate": 1.818112395179531e-05,
"loss": 3.6007,
"step": 175000
},
{
"epoch": 0.64,
"learning_rate": 1.809021287737187e-05,
"loss": 3.5943,
"step": 175500
},
{
"epoch": 0.64,
"learning_rate": 1.7999301802948428e-05,
"loss": 3.6018,
"step": 176000
},
{
"epoch": 0.64,
"learning_rate": 1.7908390728524987e-05,
"loss": 3.599,
"step": 176500
},
{
"epoch": 0.64,
"learning_rate": 1.7817479654101545e-05,
"loss": 3.5966,
"step": 177000
},
{
"epoch": 0.65,
"learning_rate": 1.77265685796781e-05,
"loss": 3.5952,
"step": 177500
},
{
"epoch": 0.65,
"learning_rate": 1.763565750525466e-05,
"loss": 3.5965,
"step": 178000
},
{
"epoch": 0.65,
"learning_rate": 1.7544746430831217e-05,
"loss": 3.5881,
"step": 178500
},
{
"epoch": 0.65,
"learning_rate": 1.7453835356407776e-05,
"loss": 3.5889,
"step": 179000
},
{
"epoch": 0.65,
"learning_rate": 1.7362924281984334e-05,
"loss": 3.5926,
"step": 179500
},
{
"epoch": 0.65,
"learning_rate": 1.7272013207560893e-05,
"loss": 3.5946,
"step": 180000
},
{
"epoch": 0.65,
"eval_accuracy": 0.3933301859072139,
"eval_loss": 3.4956858158111572,
"eval_runtime": 5179.6133,
"eval_samples_per_second": 89.256,
"eval_steps_per_second": 1.395,
"step": 180000
},
{
"epoch": 0.66,
"learning_rate": 1.718110213313745e-05,
"loss": 3.5924,
"step": 180500
},
{
"epoch": 0.66,
"learning_rate": 1.709019105871401e-05,
"loss": 3.5934,
"step": 181000
},
{
"epoch": 0.66,
"learning_rate": 1.6999279984290568e-05,
"loss": 3.5931,
"step": 181500
},
{
"epoch": 0.66,
"learning_rate": 1.6908368909867127e-05,
"loss": 3.5896,
"step": 182000
},
{
"epoch": 0.66,
"learning_rate": 1.6817457835443685e-05,
"loss": 3.5905,
"step": 182500
},
{
"epoch": 0.67,
"learning_rate": 1.6726546761020244e-05,
"loss": 3.5917,
"step": 183000
},
{
"epoch": 0.67,
"learning_rate": 1.66356356865968e-05,
"loss": 3.598,
"step": 183500
},
{
"epoch": 0.67,
"learning_rate": 1.6544724612173357e-05,
"loss": 3.5893,
"step": 184000
},
{
"epoch": 0.67,
"learning_rate": 1.6453813537749916e-05,
"loss": 3.5871,
"step": 184500
},
{
"epoch": 0.67,
"learning_rate": 1.6362902463326474e-05,
"loss": 3.5854,
"step": 185000
},
{
"epoch": 0.67,
"learning_rate": 1.6271991388903033e-05,
"loss": 3.5876,
"step": 185500
},
{
"epoch": 0.68,
"learning_rate": 1.6181080314479588e-05,
"loss": 3.5837,
"step": 186000
},
{
"epoch": 0.68,
"learning_rate": 1.6090169240056146e-05,
"loss": 3.5847,
"step": 186500
},
{
"epoch": 0.68,
"learning_rate": 1.5999258165632705e-05,
"loss": 3.5847,
"step": 187000
},
{
"epoch": 0.68,
"learning_rate": 1.5908347091209263e-05,
"loss": 3.5766,
"step": 187500
},
{
"epoch": 0.68,
"learning_rate": 1.5817436016785822e-05,
"loss": 3.5861,
"step": 188000
},
{
"epoch": 0.69,
"learning_rate": 1.572652494236238e-05,
"loss": 3.5847,
"step": 188500
},
{
"epoch": 0.69,
"learning_rate": 1.5635613867938935e-05,
"loss": 3.5857,
"step": 189000
},
{
"epoch": 0.69,
"learning_rate": 1.5544702793515494e-05,
"loss": 3.5822,
"step": 189500
},
{
"epoch": 0.69,
"learning_rate": 1.5453791719092052e-05,
"loss": 3.5883,
"step": 190000
},
{
"epoch": 0.69,
"eval_accuracy": 0.39440587506912217,
"eval_loss": 3.4857895374298096,
"eval_runtime": 5191.8412,
"eval_samples_per_second": 89.046,
"eval_steps_per_second": 1.391,
"step": 190000
},
{
"epoch": 0.69,
"learning_rate": 1.536288064466861e-05,
"loss": 3.5789,
"step": 190500
},
{
"epoch": 0.69,
"learning_rate": 1.527196957024517e-05,
"loss": 3.5762,
"step": 191000
},
{
"epoch": 0.7,
"learning_rate": 1.518105849582173e-05,
"loss": 3.5819,
"step": 191500
},
{
"epoch": 0.7,
"learning_rate": 1.5090147421398285e-05,
"loss": 3.5859,
"step": 192000
},
{
"epoch": 0.7,
"learning_rate": 1.4999236346974843e-05,
"loss": 3.5851,
"step": 192500
},
{
"epoch": 0.7,
"learning_rate": 1.4908325272551402e-05,
"loss": 3.5807,
"step": 193000
},
{
"epoch": 0.7,
"learning_rate": 1.481741419812796e-05,
"loss": 3.5793,
"step": 193500
},
{
"epoch": 0.71,
"learning_rate": 1.4726503123704519e-05,
"loss": 3.5748,
"step": 194000
},
{
"epoch": 0.71,
"learning_rate": 1.4635592049281077e-05,
"loss": 3.5808,
"step": 194500
},
{
"epoch": 0.71,
"learning_rate": 1.4544680974857632e-05,
"loss": 3.583,
"step": 195000
},
{
"epoch": 0.71,
"learning_rate": 1.4453769900434191e-05,
"loss": 3.5786,
"step": 195500
},
{
"epoch": 0.71,
"learning_rate": 1.436285882601075e-05,
"loss": 3.5755,
"step": 196000
},
{
"epoch": 0.71,
"learning_rate": 1.4271947751587308e-05,
"loss": 3.5708,
"step": 196500
},
{
"epoch": 0.72,
"learning_rate": 1.4181036677163866e-05,
"loss": 3.5847,
"step": 197000
},
{
"epoch": 0.72,
"learning_rate": 1.4090125602740425e-05,
"loss": 3.5762,
"step": 197500
},
{
"epoch": 0.72,
"learning_rate": 1.3999214528316982e-05,
"loss": 3.5861,
"step": 198000
},
{
"epoch": 0.72,
"learning_rate": 1.390830345389354e-05,
"loss": 3.5804,
"step": 198500
},
{
"epoch": 0.72,
"learning_rate": 1.3817392379470099e-05,
"loss": 3.5792,
"step": 199000
},
{
"epoch": 0.73,
"learning_rate": 1.3726481305046657e-05,
"loss": 3.5685,
"step": 199500
},
{
"epoch": 0.73,
"learning_rate": 1.3635570230623216e-05,
"loss": 3.5789,
"step": 200000
},
{
"epoch": 0.73,
"eval_accuracy": 0.3951073601816376,
"eval_loss": 3.478769302368164,
"eval_runtime": 5201.4054,
"eval_samples_per_second": 88.882,
"eval_steps_per_second": 1.389,
"step": 200000
},
{
"epoch": 0.73,
"learning_rate": 1.354465915619977e-05,
"loss": 3.5745,
"step": 200500
},
{
"epoch": 0.73,
"learning_rate": 1.345374808177633e-05,
"loss": 3.5689,
"step": 201000
},
{
"epoch": 0.73,
"learning_rate": 1.3362837007352888e-05,
"loss": 3.574,
"step": 201500
},
{
"epoch": 0.73,
"learning_rate": 1.3271925932929446e-05,
"loss": 3.5697,
"step": 202000
},
{
"epoch": 0.74,
"learning_rate": 1.3181014858506005e-05,
"loss": 3.5731,
"step": 202500
},
{
"epoch": 0.74,
"learning_rate": 1.3090103784082563e-05,
"loss": 3.5712,
"step": 203000
},
{
"epoch": 0.74,
"learning_rate": 1.299919270965912e-05,
"loss": 3.5726,
"step": 203500
},
{
"epoch": 0.74,
"learning_rate": 1.2908281635235679e-05,
"loss": 3.5719,
"step": 204000
},
{
"epoch": 0.74,
"learning_rate": 1.2817370560812237e-05,
"loss": 3.574,
"step": 204500
},
{
"epoch": 0.75,
"learning_rate": 1.2726459486388796e-05,
"loss": 3.5714,
"step": 205000
},
{
"epoch": 0.75,
"learning_rate": 1.2635548411965354e-05,
"loss": 3.5694,
"step": 205500
},
{
"epoch": 0.75,
"learning_rate": 1.2544637337541913e-05,
"loss": 3.5721,
"step": 206000
},
{
"epoch": 0.75,
"learning_rate": 1.245372626311847e-05,
"loss": 3.5706,
"step": 206500
},
{
"epoch": 0.75,
"learning_rate": 1.2362815188695026e-05,
"loss": 3.5741,
"step": 207000
},
{
"epoch": 0.75,
"learning_rate": 1.2271904114271585e-05,
"loss": 3.5713,
"step": 207500
},
{
"epoch": 0.76,
"learning_rate": 1.2180993039848143e-05,
"loss": 3.5611,
"step": 208000
},
{
"epoch": 0.76,
"learning_rate": 1.20900819654247e-05,
"loss": 3.5684,
"step": 208500
},
{
"epoch": 0.76,
"learning_rate": 1.1999170891001258e-05,
"loss": 3.5723,
"step": 209000
},
{
"epoch": 0.76,
"learning_rate": 1.1908259816577817e-05,
"loss": 3.5708,
"step": 209500
},
{
"epoch": 0.76,
"learning_rate": 1.1817348742154374e-05,
"loss": 3.5693,
"step": 210000
},
{
"epoch": 0.76,
"eval_accuracy": 0.39628516900233696,
"eval_loss": 3.4701921939849854,
"eval_runtime": 5197.3649,
"eval_samples_per_second": 88.951,
"eval_steps_per_second": 1.39,
"step": 210000
},
{
"epoch": 0.77,
"learning_rate": 1.1726437667730932e-05,
"loss": 3.5625,
"step": 210500
},
{
"epoch": 0.77,
"learning_rate": 1.163552659330749e-05,
"loss": 3.5685,
"step": 211000
},
{
"epoch": 0.77,
"learning_rate": 1.154461551888405e-05,
"loss": 3.5689,
"step": 211500
},
{
"epoch": 0.77,
"learning_rate": 1.1453704444460608e-05,
"loss": 3.5671,
"step": 212000
},
{
"epoch": 0.77,
"learning_rate": 1.1362793370037166e-05,
"loss": 3.5691,
"step": 212500
},
{
"epoch": 0.77,
"learning_rate": 1.1271882295613723e-05,
"loss": 3.5599,
"step": 213000
},
{
"epoch": 0.78,
"learning_rate": 1.1180971221190282e-05,
"loss": 3.5622,
"step": 213500
},
{
"epoch": 0.78,
"learning_rate": 1.109006014676684e-05,
"loss": 3.5694,
"step": 214000
},
{
"epoch": 0.78,
"learning_rate": 1.0999149072343397e-05,
"loss": 3.556,
"step": 214500
},
{
"epoch": 0.78,
"learning_rate": 1.0908237997919955e-05,
"loss": 3.5662,
"step": 215000
},
{
"epoch": 0.78,
"learning_rate": 1.0817326923496514e-05,
"loss": 3.5645,
"step": 215500
},
{
"epoch": 0.79,
"learning_rate": 1.072641584907307e-05,
"loss": 3.5655,
"step": 216000
},
{
"epoch": 0.79,
"learning_rate": 1.063550477464963e-05,
"loss": 3.5647,
"step": 216500
},
{
"epoch": 0.79,
"learning_rate": 1.0544593700226186e-05,
"loss": 3.5643,
"step": 217000
},
{
"epoch": 0.79,
"learning_rate": 1.0453682625802744e-05,
"loss": 3.5631,
"step": 217500
},
{
"epoch": 0.79,
"learning_rate": 1.0362771551379303e-05,
"loss": 3.5602,
"step": 218000
},
{
"epoch": 0.79,
"learning_rate": 1.0271860476955861e-05,
"loss": 3.5539,
"step": 218500
},
{
"epoch": 0.8,
"learning_rate": 1.018094940253242e-05,
"loss": 3.5632,
"step": 219000
},
{
"epoch": 0.8,
"learning_rate": 1.0090038328108978e-05,
"loss": 3.5624,
"step": 219500
},
{
"epoch": 0.8,
"learning_rate": 9.999127253685535e-06,
"loss": 3.5584,
"step": 220000
},
{
"epoch": 0.8,
"eval_accuracy": 0.39701544180672643,
"eval_loss": 3.463245391845703,
"eval_runtime": 5178.7616,
"eval_samples_per_second": 89.271,
"eval_steps_per_second": 1.395,
"step": 220000
},
{
"epoch": 0.8,
"learning_rate": 9.908216179262094e-06,
"loss": 3.5657,
"step": 220500
},
{
"epoch": 0.8,
"learning_rate": 9.817305104838652e-06,
"loss": 3.5575,
"step": 221000
},
{
"epoch": 0.81,
"learning_rate": 9.726394030415209e-06,
"loss": 3.5637,
"step": 221500
},
{
"epoch": 0.81,
"learning_rate": 9.635482955991768e-06,
"loss": 3.5602,
"step": 222000
},
{
"epoch": 0.81,
"learning_rate": 9.544571881568326e-06,
"loss": 3.5491,
"step": 222500
},
{
"epoch": 0.81,
"learning_rate": 9.453660807144883e-06,
"loss": 3.5554,
"step": 223000
},
{
"epoch": 0.81,
"learning_rate": 9.362749732721441e-06,
"loss": 3.5639,
"step": 223500
},
{
"epoch": 0.81,
"learning_rate": 9.271838658298e-06,
"loss": 3.5607,
"step": 224000
},
{
"epoch": 0.82,
"learning_rate": 9.180927583874557e-06,
"loss": 3.5572,
"step": 224500
},
{
"epoch": 0.82,
"learning_rate": 9.090016509451115e-06,
"loss": 3.5548,
"step": 225000
},
{
"epoch": 0.82,
"learning_rate": 8.999105435027674e-06,
"loss": 3.5591,
"step": 225500
},
{
"epoch": 0.82,
"learning_rate": 8.908194360604232e-06,
"loss": 3.5558,
"step": 226000
},
{
"epoch": 0.82,
"learning_rate": 8.81728328618079e-06,
"loss": 3.5497,
"step": 226500
},
{
"epoch": 0.83,
"learning_rate": 8.72637221175735e-06,
"loss": 3.5581,
"step": 227000
},
{
"epoch": 0.83,
"learning_rate": 8.635461137333906e-06,
"loss": 3.5537,
"step": 227500
},
{
"epoch": 0.83,
"learning_rate": 8.544550062910465e-06,
"loss": 3.5489,
"step": 228000
},
{
"epoch": 0.83,
"learning_rate": 8.453638988487023e-06,
"loss": 3.5609,
"step": 228500
},
{
"epoch": 0.83,
"learning_rate": 8.36272791406358e-06,
"loss": 3.5526,
"step": 229000
},
{
"epoch": 0.83,
"learning_rate": 8.271816839640138e-06,
"loss": 3.5605,
"step": 229500
},
{
"epoch": 0.84,
"learning_rate": 8.180905765216695e-06,
"loss": 3.5546,
"step": 230000
},
{
"epoch": 0.84,
"eval_accuracy": 0.397713683414333,
"eval_loss": 3.4574100971221924,
"eval_runtime": 5191.4742,
"eval_samples_per_second": 89.052,
"eval_steps_per_second": 1.392,
"step": 230000
},
{
"epoch": 0.84,
"learning_rate": 8.089994690793254e-06,
"loss": 3.5595,
"step": 230500
},
{
"epoch": 0.84,
"learning_rate": 7.999083616369812e-06,
"loss": 3.5594,
"step": 231000
},
{
"epoch": 0.84,
"learning_rate": 7.908172541946369e-06,
"loss": 3.5508,
"step": 231500
},
{
"epoch": 0.84,
"learning_rate": 7.817261467522927e-06,
"loss": 3.551,
"step": 232000
},
{
"epoch": 0.85,
"learning_rate": 7.726350393099486e-06,
"loss": 3.5544,
"step": 232500
},
{
"epoch": 0.85,
"learning_rate": 7.635439318676044e-06,
"loss": 3.5524,
"step": 233000
},
{
"epoch": 0.85,
"learning_rate": 7.544528244252602e-06,
"loss": 3.5538,
"step": 233500
},
{
"epoch": 0.85,
"learning_rate": 7.4536171698291606e-06,
"loss": 3.5495,
"step": 234000
},
{
"epoch": 0.85,
"learning_rate": 7.362706095405718e-06,
"loss": 3.5589,
"step": 234500
},
{
"epoch": 0.85,
"learning_rate": 7.271795020982277e-06,
"loss": 3.547,
"step": 235000
},
{
"epoch": 0.86,
"learning_rate": 7.180883946558835e-06,
"loss": 3.5511,
"step": 235500
},
{
"epoch": 0.86,
"learning_rate": 7.089972872135392e-06,
"loss": 3.5431,
"step": 236000
},
{
"epoch": 0.86,
"learning_rate": 6.9990617977119505e-06,
"loss": 3.5518,
"step": 236500
},
{
"epoch": 0.86,
"learning_rate": 6.908150723288509e-06,
"loss": 3.5509,
"step": 237000
},
{
"epoch": 0.86,
"learning_rate": 6.817239648865066e-06,
"loss": 3.5529,
"step": 237500
},
{
"epoch": 0.87,
"learning_rate": 6.726328574441624e-06,
"loss": 3.5526,
"step": 238000
},
{
"epoch": 0.87,
"learning_rate": 6.635417500018183e-06,
"loss": 3.5515,
"step": 238500
},
{
"epoch": 0.87,
"learning_rate": 6.5445064255947405e-06,
"loss": 3.5505,
"step": 239000
},
{
"epoch": 0.87,
"learning_rate": 6.453595351171299e-06,
"loss": 3.5495,
"step": 239500
},
{
"epoch": 0.87,
"learning_rate": 6.3626842767478575e-06,
"loss": 3.5434,
"step": 240000
},
{
"epoch": 0.87,
"eval_accuracy": 0.3982921056372081,
"eval_loss": 3.4519968032836914,
"eval_runtime": 5192.4597,
"eval_samples_per_second": 89.035,
"eval_steps_per_second": 1.391,
"step": 240000
},
{
"epoch": 0.87,
"learning_rate": 6.271773202324414e-06,
"loss": 3.5523,
"step": 240500
},
{
"epoch": 0.88,
"learning_rate": 6.180862127900973e-06,
"loss": 3.5486,
"step": 241000
},
{
"epoch": 0.88,
"learning_rate": 6.0899510534775304e-06,
"loss": 3.5477,
"step": 241500
},
{
"epoch": 0.88,
"learning_rate": 5.999039979054089e-06,
"loss": 3.5445,
"step": 242000
},
{
"epoch": 0.88,
"learning_rate": 5.908128904630647e-06,
"loss": 3.5472,
"step": 242500
},
{
"epoch": 0.88,
"learning_rate": 5.817217830207205e-06,
"loss": 3.5486,
"step": 243000
},
{
"epoch": 0.89,
"learning_rate": 5.726306755783763e-06,
"loss": 3.5445,
"step": 243500
},
{
"epoch": 0.89,
"learning_rate": 5.635395681360321e-06,
"loss": 3.5467,
"step": 244000
},
{
"epoch": 0.89,
"learning_rate": 5.544484606936879e-06,
"loss": 3.5434,
"step": 244500
},
{
"epoch": 0.89,
"learning_rate": 5.4535735325134365e-06,
"loss": 3.5456,
"step": 245000
},
{
"epoch": 0.89,
"learning_rate": 5.362662458089995e-06,
"loss": 3.5507,
"step": 245500
},
{
"epoch": 0.89,
"learning_rate": 5.2717513836665535e-06,
"loss": 3.5457,
"step": 246000
},
{
"epoch": 0.9,
"learning_rate": 5.180840309243111e-06,
"loss": 3.5449,
"step": 246500
},
{
"epoch": 0.9,
"learning_rate": 5.089929234819669e-06,
"loss": 3.5462,
"step": 247000
},
{
"epoch": 0.9,
"learning_rate": 4.999018160396227e-06,
"loss": 3.544,
"step": 247500
},
{
"epoch": 0.9,
"learning_rate": 4.908107085972785e-06,
"loss": 3.5418,
"step": 248000
},
{
"epoch": 0.9,
"learning_rate": 4.817196011549343e-06,
"loss": 3.541,
"step": 248500
},
{
"epoch": 0.91,
"learning_rate": 4.726284937125901e-06,
"loss": 3.5427,
"step": 249000
},
{
"epoch": 0.91,
"learning_rate": 4.63537386270246e-06,
"loss": 3.5448,
"step": 249500
},
{
"epoch": 0.91,
"learning_rate": 4.544462788279017e-06,
"loss": 3.5447,
"step": 250000
},
{
"epoch": 0.91,
"eval_accuracy": 0.3988207398482906,
"eval_loss": 3.447261095046997,
"eval_runtime": 5198.0091,
"eval_samples_per_second": 88.94,
"eval_steps_per_second": 1.39,
"step": 250000
},
{
"epoch": 0.91,
"learning_rate": 4.453551713855575e-06,
"loss": 3.5431,
"step": 250500
},
{
"epoch": 0.91,
"learning_rate": 4.3626406394321334e-06,
"loss": 3.5399,
"step": 251000
},
{
"epoch": 0.91,
"learning_rate": 4.271729565008691e-06,
"loss": 3.5408,
"step": 251500
},
{
"epoch": 0.92,
"learning_rate": 4.180818490585249e-06,
"loss": 3.5471,
"step": 252000
},
{
"epoch": 0.92,
"learning_rate": 4.089907416161807e-06,
"loss": 3.547,
"step": 252500
},
{
"epoch": 0.92,
"learning_rate": 3.998996341738366e-06,
"loss": 3.5483,
"step": 253000
},
{
"epoch": 0.92,
"learning_rate": 3.908085267314923e-06,
"loss": 3.5353,
"step": 253500
},
{
"epoch": 0.92,
"learning_rate": 3.817174192891482e-06,
"loss": 3.5471,
"step": 254000
},
{
"epoch": 0.93,
"learning_rate": 3.7262631184680395e-06,
"loss": 3.5406,
"step": 254500
},
{
"epoch": 0.93,
"learning_rate": 3.635352044044597e-06,
"loss": 3.536,
"step": 255000
},
{
"epoch": 0.93,
"learning_rate": 3.5444409696211557e-06,
"loss": 3.5426,
"step": 255500
},
{
"epoch": 0.93,
"learning_rate": 3.4535298951977138e-06,
"loss": 3.544,
"step": 256000
},
{
"epoch": 0.93,
"learning_rate": 3.3626188207742714e-06,
"loss": 3.5396,
"step": 256500
},
{
"epoch": 0.93,
"learning_rate": 3.2717077463508295e-06,
"loss": 3.5416,
"step": 257000
},
{
"epoch": 0.94,
"learning_rate": 3.180796671927388e-06,
"loss": 3.5408,
"step": 257500
},
{
"epoch": 0.94,
"learning_rate": 3.0898855975039456e-06,
"loss": 3.5375,
"step": 258000
},
{
"epoch": 0.94,
"learning_rate": 2.9989745230805037e-06,
"loss": 3.5451,
"step": 258500
},
{
"epoch": 0.94,
"learning_rate": 2.908063448657062e-06,
"loss": 3.5397,
"step": 259000
},
{
"epoch": 0.94,
"learning_rate": 2.81715237423362e-06,
"loss": 3.5356,
"step": 259500
},
{
"epoch": 0.95,
"learning_rate": 2.726241299810178e-06,
"loss": 3.5353,
"step": 260000
},
{
"epoch": 0.95,
"eval_accuracy": 0.39932546219691223,
"eval_loss": 3.442690849304199,
"eval_runtime": 5201.218,
"eval_samples_per_second": 88.885,
"eval_steps_per_second": 1.389,
"step": 260000
},
{
"epoch": 0.95,
"learning_rate": 2.6353302253867356e-06,
"loss": 3.5345,
"step": 260500
},
{
"epoch": 0.95,
"learning_rate": 2.544419150963294e-06,
"loss": 3.5408,
"step": 261000
},
{
"epoch": 0.95,
"learning_rate": 2.4535080765398518e-06,
"loss": 3.5344,
"step": 261500
},
{
"epoch": 0.95,
"learning_rate": 2.36259700211641e-06,
"loss": 3.5391,
"step": 262000
},
{
"epoch": 0.95,
"learning_rate": 2.271685927692968e-06,
"loss": 3.5359,
"step": 262500
},
{
"epoch": 0.96,
"learning_rate": 2.180774853269526e-06,
"loss": 3.5373,
"step": 263000
},
{
"epoch": 0.96,
"learning_rate": 2.089863778846084e-06,
"loss": 3.5374,
"step": 263500
},
{
"epoch": 0.96,
"learning_rate": 1.998952704422642e-06,
"loss": 3.5366,
"step": 264000
},
{
"epoch": 0.96,
"learning_rate": 1.9080416299992e-06,
"loss": 3.535,
"step": 264500
},
{
"epoch": 0.96,
"learning_rate": 1.8171305555757583e-06,
"loss": 3.5332,
"step": 265000
},
{
"epoch": 0.97,
"learning_rate": 1.726219481152316e-06,
"loss": 3.5395,
"step": 265500
},
{
"epoch": 0.97,
"learning_rate": 1.6353084067288742e-06,
"loss": 3.5311,
"step": 266000
},
{
"epoch": 0.97,
"learning_rate": 1.544397332305432e-06,
"loss": 3.5383,
"step": 266500
},
{
"epoch": 0.97,
"learning_rate": 1.4534862578819902e-06,
"loss": 3.5306,
"step": 267000
},
{
"epoch": 0.97,
"learning_rate": 1.3625751834585482e-06,
"loss": 3.5382,
"step": 267500
},
{
"epoch": 0.97,
"learning_rate": 1.2716641090351063e-06,
"loss": 3.5391,
"step": 268000
},
{
"epoch": 0.98,
"learning_rate": 1.1807530346116644e-06,
"loss": 3.5469,
"step": 268500
},
{
"epoch": 0.98,
"learning_rate": 1.0898419601882222e-06,
"loss": 3.5453,
"step": 269000
},
{
"epoch": 0.98,
"learning_rate": 9.989308857647805e-07,
"loss": 3.5361,
"step": 269500
},
{
"epoch": 0.98,
"learning_rate": 9.080198113413384e-07,
"loss": 3.5382,
"step": 270000
},
{
"epoch": 0.98,
"eval_accuracy": 0.39966931600545597,
"eval_loss": 3.4401557445526123,
"eval_runtime": 5192.2106,
"eval_samples_per_second": 89.04,
"eval_steps_per_second": 1.391,
"step": 270000
},
{
"epoch": 0.98,
"learning_rate": 8.171087369178965e-07,
"loss": 3.5344,
"step": 270500
},
{
"epoch": 0.99,
"learning_rate": 7.261976624944544e-07,
"loss": 3.5399,
"step": 271000
},
{
"epoch": 0.99,
"learning_rate": 6.352865880710125e-07,
"loss": 3.5356,
"step": 271500
},
{
"epoch": 0.99,
"learning_rate": 5.443755136475705e-07,
"loss": 3.5372,
"step": 272000
},
{
"epoch": 0.99,
"learning_rate": 4.5346443922412856e-07,
"loss": 3.5386,
"step": 272500
},
{
"epoch": 0.99,
"learning_rate": 3.6255336480068653e-07,
"loss": 3.5466,
"step": 273000
},
{
"epoch": 0.99,
"learning_rate": 2.716422903772446e-07,
"loss": 3.535,
"step": 273500
},
{
"epoch": 1.0,
"learning_rate": 1.8073121595380266e-07,
"loss": 3.5324,
"step": 274000
},
{
"epoch": 1.0,
"learning_rate": 8.982014153036067e-08,
"loss": 3.5376,
"step": 274500
},
{
"epoch": 1.0,
"step": 274994,
"total_flos": 4.598623226953728e+18,
"train_loss": 3.7837416634722314,
"train_runtime": 282476.6171,
"train_samples_per_second": 31.152,
"train_steps_per_second": 0.974
}
],
"logging_steps": 500,
"max_steps": 274994,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 10000,
"total_flos": 4.598623226953728e+18,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}