childes-bert / trainer_state.json
smeylan's picture
initial commit
b9255bc
raw
history blame
5.82 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"global_step": 22170,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.07,
"learning_rate": 4.8872350022553004e-05,
"loss": 1.7015,
"step": 500
},
{
"epoch": 0.14,
"learning_rate": 4.7744700045106e-05,
"loss": 1.5752,
"step": 1000
},
{
"epoch": 0.2,
"learning_rate": 4.6617050067659e-05,
"loss": 1.5199,
"step": 1500
},
{
"epoch": 0.27,
"learning_rate": 4.5489400090211996e-05,
"loss": 1.4862,
"step": 2000
},
{
"epoch": 0.34,
"learning_rate": 4.4361750112765e-05,
"loss": 1.4695,
"step": 2500
},
{
"epoch": 0.41,
"learning_rate": 4.3234100135318e-05,
"loss": 1.44,
"step": 3000
},
{
"epoch": 0.47,
"learning_rate": 4.2106450157870994e-05,
"loss": 1.4353,
"step": 3500
},
{
"epoch": 0.54,
"learning_rate": 4.0978800180423996e-05,
"loss": 1.4165,
"step": 4000
},
{
"epoch": 0.61,
"learning_rate": 3.9851150202977e-05,
"loss": 1.4017,
"step": 4500
},
{
"epoch": 0.68,
"learning_rate": 3.872350022553e-05,
"loss": 1.3854,
"step": 5000
},
{
"epoch": 0.74,
"learning_rate": 3.7595850248083e-05,
"loss": 1.3809,
"step": 5500
},
{
"epoch": 0.81,
"learning_rate": 3.6468200270635996e-05,
"loss": 1.3731,
"step": 6000
},
{
"epoch": 0.88,
"learning_rate": 3.5340550293189e-05,
"loss": 1.3641,
"step": 6500
},
{
"epoch": 0.95,
"learning_rate": 3.4212900315742e-05,
"loss": 1.3546,
"step": 7000
},
{
"epoch": 1.01,
"learning_rate": 3.3085250338294994e-05,
"loss": 1.3382,
"step": 7500
},
{
"epoch": 1.08,
"learning_rate": 3.1957600360847996e-05,
"loss": 1.3303,
"step": 8000
},
{
"epoch": 1.15,
"learning_rate": 3.0829950383401e-05,
"loss": 1.3253,
"step": 8500
},
{
"epoch": 1.22,
"learning_rate": 2.9702300405953993e-05,
"loss": 1.3242,
"step": 9000
},
{
"epoch": 1.29,
"learning_rate": 2.8574650428506994e-05,
"loss": 1.307,
"step": 9500
},
{
"epoch": 1.35,
"learning_rate": 2.744700045105999e-05,
"loss": 1.3029,
"step": 10000
},
{
"epoch": 1.42,
"learning_rate": 2.631935047361299e-05,
"loss": 1.3085,
"step": 10500
},
{
"epoch": 1.49,
"learning_rate": 2.5191700496165993e-05,
"loss": 1.2841,
"step": 11000
},
{
"epoch": 1.56,
"learning_rate": 2.406405051871899e-05,
"loss": 1.2921,
"step": 11500
},
{
"epoch": 1.62,
"learning_rate": 2.293640054127199e-05,
"loss": 1.2855,
"step": 12000
},
{
"epoch": 1.69,
"learning_rate": 2.1808750563824988e-05,
"loss": 1.2759,
"step": 12500
},
{
"epoch": 1.76,
"learning_rate": 2.068110058637799e-05,
"loss": 1.26,
"step": 13000
},
{
"epoch": 1.83,
"learning_rate": 1.955345060893099e-05,
"loss": 1.2742,
"step": 13500
},
{
"epoch": 1.89,
"learning_rate": 1.842580063148399e-05,
"loss": 1.2761,
"step": 14000
},
{
"epoch": 1.96,
"learning_rate": 1.7298150654036988e-05,
"loss": 1.2549,
"step": 14500
},
{
"epoch": 2.03,
"learning_rate": 1.6170500676589986e-05,
"loss": 1.2552,
"step": 15000
},
{
"epoch": 2.1,
"learning_rate": 1.5042850699142988e-05,
"loss": 1.2356,
"step": 15500
},
{
"epoch": 2.17,
"learning_rate": 1.3915200721695986e-05,
"loss": 1.239,
"step": 16000
},
{
"epoch": 2.23,
"learning_rate": 1.2787550744248986e-05,
"loss": 1.2378,
"step": 16500
},
{
"epoch": 2.3,
"learning_rate": 1.1659900766801986e-05,
"loss": 1.2431,
"step": 17000
},
{
"epoch": 2.37,
"learning_rate": 1.0532250789354985e-05,
"loss": 1.2202,
"step": 17500
},
{
"epoch": 2.44,
"learning_rate": 9.404600811907985e-06,
"loss": 1.232,
"step": 18000
},
{
"epoch": 2.5,
"learning_rate": 8.276950834460983e-06,
"loss": 1.2274,
"step": 18500
},
{
"epoch": 2.57,
"learning_rate": 7.149300857013984e-06,
"loss": 1.2233,
"step": 19000
},
{
"epoch": 2.64,
"learning_rate": 6.021650879566982e-06,
"loss": 1.2234,
"step": 19500
},
{
"epoch": 2.71,
"learning_rate": 4.894000902119982e-06,
"loss": 1.2216,
"step": 20000
},
{
"epoch": 2.77,
"learning_rate": 3.7663509246729816e-06,
"loss": 1.2164,
"step": 20500
},
{
"epoch": 2.84,
"learning_rate": 2.638700947225981e-06,
"loss": 1.2203,
"step": 21000
},
{
"epoch": 2.91,
"learning_rate": 1.5110509697789808e-06,
"loss": 1.2105,
"step": 21500
},
{
"epoch": 2.98,
"learning_rate": 3.8340099233198017e-07,
"loss": 1.2186,
"step": 22000
},
{
"epoch": 3.0,
"step": 22170,
"train_runtime": 19825.6129,
"train_samples_per_second": 1.118
}
],
"max_steps": 22170,
"num_train_epochs": 3,
"total_flos": 59663810085304320,
"trial_name": null,
"trial_params": null
}