roberta-medium / trainer_state.json
Yuexiang Zhai
push
39298c3
raw
history blame
7.68 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 21.693934737413,
"global_step": 30000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.36,
"learning_rate": 0.0002777777777777778,
"loss": 7.7201,
"step": 500
},
{
"epoch": 0.72,
"learning_rate": 0.0005555555555555556,
"loss": 6.4428,
"step": 1000
},
{
"epoch": 1.08,
"learning_rate": 0.0008333333333333334,
"loss": 5.9072,
"step": 1500
},
{
"epoch": 1.45,
"learning_rate": 0.0009929078014184398,
"loss": 3.7123,
"step": 2000
},
{
"epoch": 1.81,
"learning_rate": 0.000975177304964539,
"loss": 2.6271,
"step": 2500
},
{
"epoch": 2.17,
"learning_rate": 0.0009574468085106384,
"loss": 2.2823,
"step": 3000
},
{
"epoch": 2.53,
"learning_rate": 0.0009397163120567376,
"loss": 2.1231,
"step": 3500
},
{
"epoch": 2.89,
"learning_rate": 0.0009219858156028368,
"loss": 2.0297,
"step": 4000
},
{
"epoch": 3.25,
"learning_rate": 0.0009042553191489362,
"loss": 1.9652,
"step": 4500
},
{
"epoch": 3.62,
"learning_rate": 0.0008865248226950354,
"loss": 1.9186,
"step": 5000
},
{
"epoch": 3.98,
"learning_rate": 0.0008687943262411348,
"loss": 1.8837,
"step": 5500
},
{
"epoch": 4.34,
"learning_rate": 0.000851063829787234,
"loss": 1.8546,
"step": 6000
},
{
"epoch": 4.7,
"learning_rate": 0.0008333333333333334,
"loss": 1.8322,
"step": 6500
},
{
"epoch": 5.06,
"learning_rate": 0.0008156028368794326,
"loss": 1.8122,
"step": 7000
},
{
"epoch": 5.42,
"learning_rate": 0.0007978723404255319,
"loss": 1.7951,
"step": 7500
},
{
"epoch": 5.79,
"learning_rate": 0.0007801418439716312,
"loss": 1.7814,
"step": 8000
},
{
"epoch": 6.15,
"learning_rate": 0.0007624113475177306,
"loss": 1.7679,
"step": 8500
},
{
"epoch": 6.51,
"learning_rate": 0.0007446808510638298,
"loss": 1.756,
"step": 9000
},
{
"epoch": 6.87,
"learning_rate": 0.0007269503546099291,
"loss": 1.7466,
"step": 9500
},
{
"epoch": 7.23,
"learning_rate": 0.0007092198581560284,
"loss": 1.7355,
"step": 10000
},
{
"epoch": 7.59,
"learning_rate": 0.0006914893617021278,
"loss": 1.7268,
"step": 10500
},
{
"epoch": 7.95,
"learning_rate": 0.0006737588652482269,
"loss": 1.7197,
"step": 11000
},
{
"epoch": 8.32,
"learning_rate": 0.0006560283687943263,
"loss": 1.7111,
"step": 11500
},
{
"epoch": 8.68,
"learning_rate": 0.0006382978723404256,
"loss": 1.7043,
"step": 12000
},
{
"epoch": 9.04,
"learning_rate": 0.0006205673758865247,
"loss": 1.6979,
"step": 12500
},
{
"epoch": 9.4,
"learning_rate": 0.0006028368794326241,
"loss": 1.6909,
"step": 13000
},
{
"epoch": 9.76,
"learning_rate": 0.0005851063829787234,
"loss": 1.6856,
"step": 13500
},
{
"epoch": 10.12,
"learning_rate": 0.0005673758865248228,
"loss": 1.6796,
"step": 14000
},
{
"epoch": 10.49,
"learning_rate": 0.0005496453900709219,
"loss": 1.6744,
"step": 14500
},
{
"epoch": 10.85,
"learning_rate": 0.0005319148936170213,
"loss": 1.6699,
"step": 15000
},
{
"epoch": 11.21,
"learning_rate": 0.0005141843971631206,
"loss": 1.6648,
"step": 15500
},
{
"epoch": 11.57,
"learning_rate": 0.0004964539007092199,
"loss": 1.6595,
"step": 16000
},
{
"epoch": 11.93,
"learning_rate": 0.0004787234042553192,
"loss": 1.6553,
"step": 16500
},
{
"epoch": 12.29,
"learning_rate": 0.0004609929078014184,
"loss": 1.6511,
"step": 17000
},
{
"epoch": 12.65,
"learning_rate": 0.0004432624113475177,
"loss": 1.6466,
"step": 17500
},
{
"epoch": 13.02,
"learning_rate": 0.000425531914893617,
"loss": 1.6431,
"step": 18000
},
{
"epoch": 13.38,
"learning_rate": 0.0004078014184397163,
"loss": 1.6386,
"step": 18500
},
{
"epoch": 13.74,
"learning_rate": 0.0003900709219858156,
"loss": 1.6353,
"step": 19000
},
{
"epoch": 14.1,
"learning_rate": 0.0003723404255319149,
"loss": 1.6307,
"step": 19500
},
{
"epoch": 14.46,
"learning_rate": 0.0003546099290780142,
"loss": 1.6279,
"step": 20000
},
{
"epoch": 14.82,
"learning_rate": 0.00033687943262411345,
"loss": 1.6241,
"step": 20500
},
{
"epoch": 15.19,
"learning_rate": 0.0003191489361702128,
"loss": 1.6215,
"step": 21000
},
{
"epoch": 15.55,
"learning_rate": 0.00030141843971631205,
"loss": 1.6178,
"step": 21500
},
{
"epoch": 15.91,
"learning_rate": 0.0002836879432624114,
"loss": 1.6149,
"step": 22000
},
{
"epoch": 16.27,
"learning_rate": 0.00026595744680851064,
"loss": 1.6118,
"step": 22500
},
{
"epoch": 16.63,
"learning_rate": 0.00024822695035460994,
"loss": 1.6086,
"step": 23000
},
{
"epoch": 16.99,
"learning_rate": 0.0002304964539007092,
"loss": 1.6064,
"step": 23500
},
{
"epoch": 17.36,
"learning_rate": 0.0002127659574468085,
"loss": 1.6028,
"step": 24000
},
{
"epoch": 17.72,
"learning_rate": 0.0001950354609929078,
"loss": 1.5994,
"step": 24500
},
{
"epoch": 18.08,
"learning_rate": 0.0001773049645390071,
"loss": 1.5971,
"step": 25000
},
{
"epoch": 18.44,
"learning_rate": 0.0001595744680851064,
"loss": 1.5938,
"step": 25500
},
{
"epoch": 18.8,
"learning_rate": 0.0001418439716312057,
"loss": 1.5922,
"step": 26000
},
{
"epoch": 19.16,
"learning_rate": 0.00012411347517730497,
"loss": 1.589,
"step": 26500
},
{
"epoch": 19.52,
"learning_rate": 0.00010638297872340425,
"loss": 1.5867,
"step": 27000
},
{
"epoch": 19.89,
"learning_rate": 8.865248226950355e-05,
"loss": 1.5839,
"step": 27500
},
{
"epoch": 20.25,
"learning_rate": 7.092198581560285e-05,
"loss": 1.5811,
"step": 28000
},
{
"epoch": 20.61,
"learning_rate": 5.319148936170213e-05,
"loss": 1.5795,
"step": 28500
},
{
"epoch": 20.97,
"learning_rate": 3.5460992907801425e-05,
"loss": 1.577,
"step": 29000
},
{
"epoch": 21.33,
"learning_rate": 1.7730496453900712e-05,
"loss": 1.575,
"step": 29500
},
{
"epoch": 21.69,
"learning_rate": 0.0,
"loss": 1.5733,
"step": 30000
}
],
"max_steps": 30000,
"num_train_epochs": 22,
"total_flos": 1.9277555263183258e+19,
"trial_name": null,
"trial_params": null
}