albert-base-v2_squad / trainer_state.json
Palak Goenka
Add files
6257fbf
raw
history blame
4.62 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"global_step": 16620,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.09,
"learning_rate": 2.909747292418773e-05,
"loss": 1.4916,
"step": 500
},
{
"epoch": 0.18,
"learning_rate": 2.819494584837545e-05,
"loss": 1.143,
"step": 1000
},
{
"epoch": 0.27,
"learning_rate": 2.729241877256318e-05,
"loss": 1.0765,
"step": 1500
},
{
"epoch": 0.36,
"learning_rate": 2.6389891696750903e-05,
"loss": 1.0025,
"step": 2000
},
{
"epoch": 0.45,
"learning_rate": 2.548736462093863e-05,
"loss": 1.0014,
"step": 2500
},
{
"epoch": 0.54,
"learning_rate": 2.4584837545126353e-05,
"loss": 0.9314,
"step": 3000
},
{
"epoch": 0.63,
"learning_rate": 2.368231046931408e-05,
"loss": 0.9617,
"step": 3500
},
{
"epoch": 0.72,
"learning_rate": 2.2779783393501805e-05,
"loss": 0.9426,
"step": 4000
},
{
"epoch": 0.81,
"learning_rate": 2.1877256317689534e-05,
"loss": 0.9326,
"step": 4500
},
{
"epoch": 0.9,
"learning_rate": 2.0974729241877255e-05,
"loss": 0.8923,
"step": 5000
},
{
"epoch": 0.99,
"learning_rate": 2.0072202166064983e-05,
"loss": 0.8866,
"step": 5500
},
{
"epoch": 1.08,
"learning_rate": 1.9169675090252708e-05,
"loss": 0.7289,
"step": 6000
},
{
"epoch": 1.17,
"learning_rate": 1.8267148014440436e-05,
"loss": 0.6916,
"step": 6500
},
{
"epoch": 1.26,
"learning_rate": 1.7364620938628157e-05,
"loss": 0.6857,
"step": 7000
},
{
"epoch": 1.35,
"learning_rate": 1.6462093862815885e-05,
"loss": 0.6954,
"step": 7500
},
{
"epoch": 1.44,
"learning_rate": 1.555956678700361e-05,
"loss": 0.6555,
"step": 8000
},
{
"epoch": 1.53,
"learning_rate": 1.4657039711191336e-05,
"loss": 0.6923,
"step": 8500
},
{
"epoch": 1.62,
"learning_rate": 1.3754512635379063e-05,
"loss": 0.6944,
"step": 9000
},
{
"epoch": 1.71,
"learning_rate": 1.2851985559566788e-05,
"loss": 0.6636,
"step": 9500
},
{
"epoch": 1.81,
"learning_rate": 1.1949458483754514e-05,
"loss": 0.6424,
"step": 10000
},
{
"epoch": 1.9,
"learning_rate": 1.1046931407942239e-05,
"loss": 0.6906,
"step": 10500
},
{
"epoch": 1.99,
"learning_rate": 1.0144404332129965e-05,
"loss": 0.6409,
"step": 11000
},
{
"epoch": 2.08,
"learning_rate": 9.24187725631769e-06,
"loss": 0.5255,
"step": 11500
},
{
"epoch": 2.17,
"learning_rate": 8.339350180505416e-06,
"loss": 0.4748,
"step": 12000
},
{
"epoch": 2.26,
"learning_rate": 7.436823104693141e-06,
"loss": 0.4609,
"step": 12500
},
{
"epoch": 2.35,
"learning_rate": 6.534296028880867e-06,
"loss": 0.473,
"step": 13000
},
{
"epoch": 2.44,
"learning_rate": 5.631768953068592e-06,
"loss": 0.4867,
"step": 13500
},
{
"epoch": 2.53,
"learning_rate": 4.729241877256318e-06,
"loss": 0.4564,
"step": 14000
},
{
"epoch": 2.62,
"learning_rate": 3.826714801444043e-06,
"loss": 0.45,
"step": 14500
},
{
"epoch": 2.71,
"learning_rate": 2.924187725631769e-06,
"loss": 0.4649,
"step": 15000
},
{
"epoch": 2.8,
"learning_rate": 2.0216606498194946e-06,
"loss": 0.4557,
"step": 15500
},
{
"epoch": 2.89,
"learning_rate": 1.1191335740072204e-06,
"loss": 0.4464,
"step": 16000
},
{
"epoch": 2.98,
"learning_rate": 2.1660649819494586e-07,
"loss": 0.4505,
"step": 16500
},
{
"epoch": 3.0,
"step": 16620,
"total_flos": 4404294320329728.0,
"train_loss": 0.7218501507590394,
"train_runtime": 2515.671,
"train_samples_per_second": 105.703,
"train_steps_per_second": 6.607
}
],
"max_steps": 16620,
"num_train_epochs": 3,
"total_flos": 4404294320329728.0,
"trial_name": null,
"trial_params": null
}