Chungyezun
loss decreased
16838da
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.9125329935378175,
"global_step": 2000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.07,
"learning_rate": 9.757045675413022e-05,
"loss": 1.7333,
"step": 50
},
{
"epoch": 0.15,
"learning_rate": 9.514091350826045e-05,
"loss": 1.3819,
"step": 100
},
{
"epoch": 0.22,
"learning_rate": 9.271137026239067e-05,
"loss": 1.3078,
"step": 150
},
{
"epoch": 0.29,
"learning_rate": 9.02818270165209e-05,
"loss": 1.3028,
"step": 200
},
{
"epoch": 0.36,
"learning_rate": 8.785228377065113e-05,
"loss": 1.2742,
"step": 250
},
{
"epoch": 0.44,
"learning_rate": 8.542274052478134e-05,
"loss": 1.2594,
"step": 300
},
{
"epoch": 0.51,
"learning_rate": 8.299319727891157e-05,
"loss": 1.2554,
"step": 350
},
{
"epoch": 0.58,
"learning_rate": 8.056365403304179e-05,
"loss": 1.2412,
"step": 400
},
{
"epoch": 0.66,
"learning_rate": 7.8134110787172e-05,
"loss": 1.2448,
"step": 450
},
{
"epoch": 0.73,
"learning_rate": 7.570456754130224e-05,
"loss": 1.2325,
"step": 500
},
{
"epoch": 0.8,
"learning_rate": 7.327502429543247e-05,
"loss": 1.2233,
"step": 550
},
{
"epoch": 0.87,
"learning_rate": 7.08454810495627e-05,
"loss": 1.2326,
"step": 600
},
{
"epoch": 0.95,
"learning_rate": 6.841593780369291e-05,
"loss": 1.2112,
"step": 650
},
{
"epoch": 1.02,
"learning_rate": 6.598639455782313e-05,
"loss": 1.2144,
"step": 700
},
{
"epoch": 1.09,
"learning_rate": 6.355685131195336e-05,
"loss": 1.1948,
"step": 750
},
{
"epoch": 1.17,
"learning_rate": 6.112730806608357e-05,
"loss": 1.2022,
"step": 800
},
{
"epoch": 1.24,
"learning_rate": 5.8697764820213804e-05,
"loss": 1.2028,
"step": 850
},
{
"epoch": 1.31,
"learning_rate": 5.626822157434403e-05,
"loss": 1.1978,
"step": 900
},
{
"epoch": 1.38,
"learning_rate": 5.3838678328474244e-05,
"loss": 1.1879,
"step": 950
},
{
"epoch": 1.46,
"learning_rate": 5.1409135082604474e-05,
"loss": 1.188,
"step": 1000
},
{
"epoch": 1.53,
"learning_rate": 4.89795918367347e-05,
"loss": 1.1836,
"step": 1050
},
{
"epoch": 1.6,
"learning_rate": 4.655004859086492e-05,
"loss": 1.1794,
"step": 1100
},
{
"epoch": 1.67,
"learning_rate": 4.412050534499514e-05,
"loss": 1.1815,
"step": 1150
},
{
"epoch": 1.75,
"learning_rate": 4.1690962099125366e-05,
"loss": 1.1856,
"step": 1200
},
{
"epoch": 1.82,
"learning_rate": 3.926141885325559e-05,
"loss": 1.1747,
"step": 1250
},
{
"epoch": 1.89,
"learning_rate": 3.683187560738581e-05,
"loss": 1.1736,
"step": 1300
},
{
"epoch": 1.97,
"learning_rate": 3.4402332361516035e-05,
"loss": 1.1899,
"step": 1350
},
{
"epoch": 2.04,
"learning_rate": 3.1972789115646265e-05,
"loss": 1.1726,
"step": 1400
},
{
"epoch": 2.11,
"learning_rate": 2.954324586977648e-05,
"loss": 1.1544,
"step": 1450
},
{
"epoch": 2.18,
"learning_rate": 2.7113702623906705e-05,
"loss": 1.1568,
"step": 1500
},
{
"epoch": 2.26,
"learning_rate": 2.468415937803693e-05,
"loss": 1.1684,
"step": 1550
},
{
"epoch": 2.33,
"learning_rate": 2.225461613216715e-05,
"loss": 1.1629,
"step": 1600
},
{
"epoch": 2.4,
"learning_rate": 1.9825072886297377e-05,
"loss": 1.1558,
"step": 1650
},
{
"epoch": 2.48,
"learning_rate": 1.73955296404276e-05,
"loss": 1.1608,
"step": 1700
},
{
"epoch": 2.55,
"learning_rate": 1.4965986394557824e-05,
"loss": 1.155,
"step": 1750
},
{
"epoch": 2.62,
"learning_rate": 1.2536443148688048e-05,
"loss": 1.153,
"step": 1800
},
{
"epoch": 2.69,
"learning_rate": 1.010689990281827e-05,
"loss": 1.1597,
"step": 1850
},
{
"epoch": 2.77,
"learning_rate": 7.677356656948493e-06,
"loss": 1.1517,
"step": 1900
},
{
"epoch": 2.84,
"learning_rate": 5.247813411078718e-06,
"loss": 1.1488,
"step": 1950
},
{
"epoch": 2.91,
"learning_rate": 2.818270165208941e-06,
"loss": 1.1602,
"step": 2000
}
],
"max_steps": 2058,
"num_train_epochs": 3,
"total_flos": 7.611953410899763e+17,
"trial_name": null,
"trial_params": null
}