{
"best_metric": 1.046238899230957,
"best_model_checkpoint": "/kaggle/output/checkpoint-30000",
"epoch": 1.263037809647979,
"eval_steps": 1000,
"global_step": 31000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 2.7777777777777777e-11,
"loss": 1.2358,
"step": 1
},
{
"epoch": 0.04,
"learning_rate": 2.7750000000000004e-08,
"loss": 1.1519,
"step": 1000
},
{
"epoch": 0.04,
"eval_accuracy": 0.3201596806387226,
"eval_loss": 1.1120171546936035,
"eval_runtime": 29.027,
"eval_samples_per_second": 172.598,
"eval_steps_per_second": 21.601,
"step": 1000
},
{
"epoch": 0.08,
"learning_rate": 5.5527777777777784e-08,
"loss": 1.128,
"step": 2000
},
{
"epoch": 0.08,
"eval_accuracy": 0.31796407185628744,
"eval_loss": 1.108055830001831,
"eval_runtime": 28.9944,
"eval_samples_per_second": 172.792,
"eval_steps_per_second": 21.625,
"step": 2000
},
{
"epoch": 0.12,
"learning_rate": 8.327777777777778e-08,
"loss": 1.122,
"step": 3000
},
{
"epoch": 0.12,
"eval_accuracy": 0.3345309381237525,
"eval_loss": 1.1074285507202148,
"eval_runtime": 28.9901,
"eval_samples_per_second": 172.818,
"eval_steps_per_second": 21.628,
"step": 3000
},
{
"epoch": 0.16,
"learning_rate": 1.1105555555555557e-07,
"loss": 1.1196,
"step": 4000
},
{
"epoch": 0.16,
"eval_accuracy": 0.3337325349301397,
"eval_loss": 1.1045424938201904,
"eval_runtime": 28.9791,
"eval_samples_per_second": 172.883,
"eval_steps_per_second": 21.636,
"step": 4000
},
{
"epoch": 0.2,
"learning_rate": 1.3880555555555558e-07,
"loss": 1.1169,
"step": 5000
},
{
"epoch": 0.2,
"eval_accuracy": 0.3397205588822355,
"eval_loss": 1.1033179759979248,
"eval_runtime": 29.0295,
"eval_samples_per_second": 172.583,
"eval_steps_per_second": 21.599,
"step": 5000
},
{
"epoch": 0.24,
"learning_rate": 1.6658333333333335e-07,
"loss": 1.118,
"step": 6000
},
{
"epoch": 0.24,
"eval_accuracy": 0.3401197604790419,
"eval_loss": 1.1030455827713013,
"eval_runtime": 28.9748,
"eval_samples_per_second": 172.909,
"eval_steps_per_second": 21.639,
"step": 6000
},
{
"epoch": 0.29,
"learning_rate": 1.9433333333333334e-07,
"loss": 1.1135,
"step": 7000
},
{
"epoch": 0.29,
"eval_accuracy": 0.35568862275449104,
"eval_loss": 1.0998502969741821,
"eval_runtime": 28.9964,
"eval_samples_per_second": 172.78,
"eval_steps_per_second": 21.623,
"step": 7000
},
{
"epoch": 0.33,
"learning_rate": 2.2211111111111114e-07,
"loss": 1.1136,
"step": 8000
},
{
"epoch": 0.33,
"eval_accuracy": 0.34510978043912177,
"eval_loss": 1.098555564880371,
"eval_runtime": 28.983,
"eval_samples_per_second": 172.86,
"eval_steps_per_second": 21.633,
"step": 8000
},
{
"epoch": 0.37,
"learning_rate": 2.4986111111111113e-07,
"loss": 1.11,
"step": 9000
},
{
"epoch": 0.37,
"eval_accuracy": 0.3411177644710579,
"eval_loss": 1.1035741567611694,
"eval_runtime": 28.9758,
"eval_samples_per_second": 172.903,
"eval_steps_per_second": 21.639,
"step": 9000
},
{
"epoch": 0.41,
"learning_rate": 2.776388888888889e-07,
"loss": 1.1076,
"step": 10000
},
{
"epoch": 0.41,
"eval_accuracy": 0.35788423153692617,
"eval_loss": 1.0970662832260132,
"eval_runtime": 29.0262,
"eval_samples_per_second": 172.603,
"eval_steps_per_second": 21.601,
"step": 10000
},
{
"epoch": 0.45,
"learning_rate": 3.0541666666666667e-07,
"loss": 1.1067,
"step": 11000
},
{
"epoch": 0.45,
"eval_accuracy": 0.37425149700598803,
"eval_loss": 1.094608187675476,
"eval_runtime": 28.8738,
"eval_samples_per_second": 173.513,
"eval_steps_per_second": 21.715,
"step": 11000
},
{
"epoch": 0.49,
"learning_rate": 3.3319444444444444e-07,
"loss": 1.1032,
"step": 12000
},
{
"epoch": 0.49,
"eval_accuracy": 0.37924151696606784,
"eval_loss": 1.092067837715149,
"eval_runtime": 28.9514,
"eval_samples_per_second": 173.049,
"eval_steps_per_second": 21.657,
"step": 12000
},
{
"epoch": 0.53,
"learning_rate": 3.6094444444444446e-07,
"loss": 1.1052,
"step": 13000
},
{
"epoch": 0.53,
"eval_accuracy": 0.3435129740518962,
"eval_loss": 1.0962508916854858,
"eval_runtime": 29.2203,
"eval_samples_per_second": 171.456,
"eval_steps_per_second": 21.458,
"step": 13000
},
{
"epoch": 0.57,
"learning_rate": 3.8872222222222223e-07,
"loss": 1.1015,
"step": 14000
},
{
"epoch": 0.57,
"eval_accuracy": 0.36806387225548903,
"eval_loss": 1.0941067934036255,
"eval_runtime": 29.2088,
"eval_samples_per_second": 171.524,
"eval_steps_per_second": 21.466,
"step": 14000
},
{
"epoch": 0.61,
"learning_rate": 4.1650000000000006e-07,
"loss": 1.0986,
"step": 15000
},
{
"epoch": 0.61,
"eval_accuracy": 0.3718562874251497,
"eval_loss": 1.0906143188476562,
"eval_runtime": 28.9463,
"eval_samples_per_second": 173.079,
"eval_steps_per_second": 21.661,
"step": 15000
},
{
"epoch": 0.65,
"learning_rate": 4.4425e-07,
"loss": 1.0984,
"step": 16000
},
{
"epoch": 0.65,
"eval_accuracy": 0.38183632734530937,
"eval_loss": 1.0862348079681396,
"eval_runtime": 28.9909,
"eval_samples_per_second": 172.813,
"eval_steps_per_second": 21.627,
"step": 16000
},
{
"epoch": 0.69,
"learning_rate": 4.7202777777777785e-07,
"loss": 1.0959,
"step": 17000
},
{
"epoch": 0.69,
"eval_accuracy": 0.40079840319361276,
"eval_loss": 1.0846956968307495,
"eval_runtime": 28.9606,
"eval_samples_per_second": 172.994,
"eval_steps_per_second": 21.65,
"step": 17000
},
{
"epoch": 0.73,
"learning_rate": 4.998055555555556e-07,
"loss": 1.0965,
"step": 18000
},
{
"epoch": 0.73,
"eval_accuracy": 0.4043912175648703,
"eval_loss": 1.086153507232666,
"eval_runtime": 28.8411,
"eval_samples_per_second": 173.71,
"eval_steps_per_second": 21.74,
"step": 18000
},
{
"epoch": 0.77,
"learning_rate": 5.275555555555556e-07,
"loss": 1.0966,
"step": 19000
},
{
"epoch": 0.77,
"eval_accuracy": 0.4041916167664671,
"eval_loss": 1.0824861526489258,
"eval_runtime": 28.9652,
"eval_samples_per_second": 172.966,
"eval_steps_per_second": 21.647,
"step": 19000
},
{
"epoch": 0.81,
"learning_rate": 5.553333333333334e-07,
"loss": 1.0943,
"step": 20000
},
{
"epoch": 0.81,
"eval_accuracy": 0.4103792415169661,
"eval_loss": 1.0817394256591797,
"eval_runtime": 28.9548,
"eval_samples_per_second": 173.028,
"eval_steps_per_second": 21.654,
"step": 20000
},
{
"epoch": 0.86,
"learning_rate": 5.830833333333334e-07,
"loss": 1.0898,
"step": 21000
},
{
"epoch": 0.86,
"eval_accuracy": 0.41357285429141716,
"eval_loss": 1.0764833688735962,
"eval_runtime": 28.9932,
"eval_samples_per_second": 172.799,
"eval_steps_per_second": 21.626,
"step": 21000
},
{
"epoch": 0.9,
"learning_rate": 6.108611111111111e-07,
"loss": 1.0877,
"step": 22000
},
{
"epoch": 0.9,
"eval_accuracy": 0.4293413173652695,
"eval_loss": 1.0701513290405273,
"eval_runtime": 28.9628,
"eval_samples_per_second": 172.98,
"eval_steps_per_second": 21.648,
"step": 22000
},
{
"epoch": 0.94,
"learning_rate": 6.386388888888889e-07,
"loss": 1.0826,
"step": 23000
},
{
"epoch": 0.94,
"eval_accuracy": 0.4285429141716567,
"eval_loss": 1.0636069774627686,
"eval_runtime": 28.9704,
"eval_samples_per_second": 172.935,
"eval_steps_per_second": 21.643,
"step": 23000
},
{
"epoch": 0.98,
"learning_rate": 6.664166666666667e-07,
"loss": 1.0781,
"step": 24000
},
{
"epoch": 0.98,
"eval_accuracy": 0.43213572854291415,
"eval_loss": 1.059762954711914,
"eval_runtime": 29.0265,
"eval_samples_per_second": 172.601,
"eval_steps_per_second": 21.601,
"step": 24000
},
{
"epoch": 1.02,
"learning_rate": 6.941666666666667e-07,
"loss": 1.0728,
"step": 25000
},
{
"epoch": 1.02,
"eval_accuracy": 0.4373253493013972,
"eval_loss": 1.056991696357727,
"eval_runtime": 28.9761,
"eval_samples_per_second": 172.901,
"eval_steps_per_second": 21.639,
"step": 25000
},
{
"epoch": 1.06,
"learning_rate": 7.219444444444444e-07,
"loss": 1.0719,
"step": 26000
},
{
"epoch": 1.06,
"eval_accuracy": 0.43193612774451096,
"eval_loss": 1.0573900938034058,
"eval_runtime": 28.9853,
"eval_samples_per_second": 172.846,
"eval_steps_per_second": 21.632,
"step": 26000
},
{
"epoch": 1.1,
"learning_rate": 7.496944444444444e-07,
"loss": 1.0683,
"step": 27000
},
{
"epoch": 1.1,
"eval_accuracy": 0.4305389221556886,
"eval_loss": 1.0567399263381958,
"eval_runtime": 28.9611,
"eval_samples_per_second": 172.991,
"eval_steps_per_second": 21.65,
"step": 27000
},
{
"epoch": 1.14,
"learning_rate": 7.774722222222223e-07,
"loss": 1.0669,
"step": 28000
},
{
"epoch": 1.14,
"eval_accuracy": 0.4297405189620758,
"eval_loss": 1.05367910861969,
"eval_runtime": 28.9913,
"eval_samples_per_second": 172.81,
"eval_steps_per_second": 21.627,
"step": 28000
},
{
"epoch": 1.18,
"learning_rate": 8.052222222222223e-07,
"loss": 1.0629,
"step": 29000
},
{
"epoch": 1.18,
"eval_accuracy": 0.4303393213572854,
"eval_loss": 1.0497413873672485,
"eval_runtime": 28.9605,
"eval_samples_per_second": 172.994,
"eval_steps_per_second": 21.65,
"step": 29000
},
{
"epoch": 1.22,
"learning_rate": 8.330000000000001e-07,
"loss": 1.0614,
"step": 30000
},
{
"epoch": 1.22,
"eval_accuracy": 0.43173652694610776,
"eval_loss": 1.046238899230957,
"eval_runtime": 28.9896,
"eval_samples_per_second": 172.821,
"eval_steps_per_second": 21.628,
"step": 30000
},
{
"epoch": 1.26,
"learning_rate": 8.607500000000001e-07,
"loss": 1.0669,
"step": 31000
},
{
"epoch": 1.26,
"eval_accuracy": 0.43233532934131735,
"eval_loss": 1.0541218519210815,
"eval_runtime": 28.9432,
"eval_samples_per_second": 173.098,
"eval_steps_per_second": 21.663,
"step": 31000
}
],
"logging_steps": 1000,
"max_steps": 10000000,
"num_train_epochs": 408,
"save_steps": 1000,
"total_flos": 3.240125152808141e+16,
"trial_name": null,
"trial_params": null
}