mobilebert_qqp / trainer_state.json
Alireza1044's picture
End of training
12eda73
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.5,
"global_step": 14215,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.09,
"learning_rate": 4.8241294407316215e-05,
"loss": 39415.716,
"step": 500
},
{
"epoch": 0.18,
"learning_rate": 4.6482588814632434e-05,
"loss": 0.3762,
"step": 1000
},
{
"epoch": 0.26,
"learning_rate": 4.472388322194865e-05,
"loss": 0.4107,
"step": 1500
},
{
"epoch": 0.35,
"learning_rate": 4.2965177629264866e-05,
"loss": 0.3225,
"step": 2000
},
{
"epoch": 0.44,
"learning_rate": 4.120647203658108e-05,
"loss": 0.3177,
"step": 2500
},
{
"epoch": 0.53,
"learning_rate": 3.944776644389729e-05,
"loss": 0.2994,
"step": 3000
},
{
"epoch": 0.62,
"learning_rate": 3.768906085121351e-05,
"loss": 0.3909,
"step": 3500
},
{
"epoch": 0.7,
"learning_rate": 3.5930355258529723e-05,
"loss": 0.2921,
"step": 4000
},
{
"epoch": 0.79,
"learning_rate": 3.417164966584594e-05,
"loss": 0.279,
"step": 4500
},
{
"epoch": 0.88,
"learning_rate": 3.2412944073162155e-05,
"loss": 0.2796,
"step": 5000
},
{
"epoch": 0.97,
"learning_rate": 3.065423848047837e-05,
"loss": 0.2694,
"step": 5500
},
{
"epoch": 1.06,
"learning_rate": 2.8895532887794584e-05,
"loss": 0.2447,
"step": 6000
},
{
"epoch": 1.14,
"learning_rate": 2.71368272951108e-05,
"loss": 0.234,
"step": 6500
},
{
"epoch": 1.23,
"learning_rate": 2.5378121702427016e-05,
"loss": 0.2273,
"step": 7000
},
{
"epoch": 1.32,
"learning_rate": 2.3619416109743232e-05,
"loss": 0.2558,
"step": 7500
},
{
"epoch": 1.41,
"learning_rate": 2.1860710517059448e-05,
"loss": 0.2255,
"step": 8000
},
{
"epoch": 1.49,
"learning_rate": 2.010200492437566e-05,
"loss": 0.2263,
"step": 8500
},
{
"epoch": 1.58,
"learning_rate": 1.8343299331691876e-05,
"loss": 0.2825,
"step": 9000
},
{
"epoch": 1.67,
"learning_rate": 1.6584593739008092e-05,
"loss": 0.2196,
"step": 9500
},
{
"epoch": 1.76,
"learning_rate": 1.4825888146324307e-05,
"loss": 0.2178,
"step": 10000
},
{
"epoch": 1.85,
"learning_rate": 1.3067182553640523e-05,
"loss": 0.2175,
"step": 10500
},
{
"epoch": 1.93,
"learning_rate": 1.1308476960956737e-05,
"loss": 0.2177,
"step": 11000
},
{
"epoch": 2.02,
"learning_rate": 9.549771368272951e-06,
"loss": 0.2053,
"step": 11500
},
{
"epoch": 2.11,
"learning_rate": 7.791065775589167e-06,
"loss": 0.3079,
"step": 12000
},
{
"epoch": 2.2,
"learning_rate": 6.032360182905382e-06,
"loss": 0.1813,
"step": 12500
},
{
"epoch": 2.29,
"learning_rate": 4.2736545902215974e-06,
"loss": 0.1804,
"step": 13000
},
{
"epoch": 2.37,
"learning_rate": 2.5149489975378126e-06,
"loss": 0.1771,
"step": 13500
},
{
"epoch": 2.46,
"learning_rate": 7.562434048540275e-07,
"loss": 0.1697,
"step": 14000
},
{
"epoch": 2.5,
"step": 14215,
"total_flos": 1.426060746003456e+16,
"train_loss": 1386.6627225032541,
"train_runtime": 8097.4527,
"train_samples_per_second": 112.333,
"train_steps_per_second": 1.755
}
],
"max_steps": 14215,
"num_train_epochs": 3,
"total_flos": 1.426060746003456e+16,
"trial_name": null,
"trial_params": null
}