jonathanagustin's picture
Model save
4a88fe3
raw
history blame
7.47 kB
{
"best_metric": 1.3331981897354126,
"best_model_checkpoint": "/content/drive/My Drive/Colab Notebooks/aai520-project/checkpoints/distilbert-finetuned-uncased/checkpoint-1600",
"epoch": 4.0,
"eval_steps": 100,
"global_step": 2040,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.39,
"learning_rate": 1.607843137254902e-05,
"loss": 3.6437,
"step": 100
},
{
"epoch": 0.39,
"eval_loss": 2.1780340671539307,
"eval_runtime": 8.4412,
"eval_samples_per_second": 1417.927,
"eval_steps_per_second": 11.136,
"step": 100
},
{
"epoch": 0.78,
"learning_rate": 1.215686274509804e-05,
"loss": 2.1596,
"step": 200
},
{
"epoch": 0.78,
"eval_loss": 1.6557185649871826,
"eval_runtime": 8.3581,
"eval_samples_per_second": 1432.022,
"eval_steps_per_second": 11.247,
"step": 200
},
{
"epoch": 1.18,
"learning_rate": 8.23529411764706e-06,
"loss": 1.8138,
"step": 300
},
{
"epoch": 1.18,
"eval_loss": 1.5682721138000488,
"eval_runtime": 8.4191,
"eval_samples_per_second": 1421.643,
"eval_steps_per_second": 11.165,
"step": 300
},
{
"epoch": 1.57,
"learning_rate": 4.313725490196079e-06,
"loss": 1.6987,
"step": 400
},
{
"epoch": 1.57,
"eval_loss": 1.5075817108154297,
"eval_runtime": 8.3595,
"eval_samples_per_second": 1431.779,
"eval_steps_per_second": 11.245,
"step": 400
},
{
"epoch": 1.96,
"learning_rate": 3.921568627450981e-07,
"loss": 1.6586,
"step": 500
},
{
"epoch": 1.96,
"eval_loss": 1.535025715827942,
"eval_runtime": 8.4289,
"eval_samples_per_second": 1419.991,
"eval_steps_per_second": 11.152,
"step": 500
},
{
"epoch": 1.18,
"learning_rate": 8.23529411764706e-06,
"loss": 1.5957,
"step": 600
},
{
"epoch": 1.18,
"eval_loss": 1.4431041479110718,
"eval_runtime": 8.5594,
"eval_samples_per_second": 1398.342,
"eval_steps_per_second": 21.964,
"step": 600
},
{
"epoch": 1.37,
"learning_rate": 6.274509803921569e-06,
"loss": 1.5825,
"step": 700
},
{
"epoch": 1.37,
"eval_loss": 1.495514154434204,
"eval_runtime": 8.5285,
"eval_samples_per_second": 1403.419,
"eval_steps_per_second": 22.044,
"step": 700
},
{
"epoch": 1.57,
"learning_rate": 4.313725490196079e-06,
"loss": 1.5523,
"step": 800
},
{
"epoch": 1.57,
"eval_loss": 1.4444433450698853,
"eval_runtime": 8.5112,
"eval_samples_per_second": 1406.272,
"eval_steps_per_second": 22.089,
"step": 800
},
{
"epoch": 1.76,
"learning_rate": 2.3529411764705885e-06,
"loss": 1.5346,
"step": 900
},
{
"epoch": 1.76,
"eval_loss": 1.393009066581726,
"eval_runtime": 8.5265,
"eval_samples_per_second": 1403.737,
"eval_steps_per_second": 22.049,
"step": 900
},
{
"epoch": 1.96,
"learning_rate": 3.921568627450981e-07,
"loss": 1.5098,
"step": 1000
},
{
"epoch": 1.96,
"eval_loss": 1.4284701347351074,
"eval_runtime": 8.5994,
"eval_samples_per_second": 1391.84,
"eval_steps_per_second": 21.862,
"step": 1000
},
{
"epoch": 2.16,
"learning_rate": 9.215686274509804e-06,
"loss": 1.4632,
"step": 1100
},
{
"epoch": 2.16,
"eval_loss": 1.3630493879318237,
"eval_runtime": 8.4807,
"eval_samples_per_second": 1411.328,
"eval_steps_per_second": 22.168,
"step": 1100
},
{
"epoch": 2.35,
"learning_rate": 8.23529411764706e-06,
"loss": 1.4468,
"step": 1200
},
{
"epoch": 2.35,
"eval_loss": 1.370953917503357,
"eval_runtime": 8.5147,
"eval_samples_per_second": 1405.685,
"eval_steps_per_second": 22.079,
"step": 1200
},
{
"epoch": 2.55,
"learning_rate": 7.2549019607843145e-06,
"loss": 1.4343,
"step": 1300
},
{
"epoch": 2.55,
"eval_loss": 1.3422259092330933,
"eval_runtime": 8.4859,
"eval_samples_per_second": 1410.461,
"eval_steps_per_second": 22.154,
"step": 1300
},
{
"epoch": 2.75,
"learning_rate": 6.274509803921569e-06,
"loss": 1.4225,
"step": 1400
},
{
"epoch": 2.75,
"eval_loss": 1.397080659866333,
"eval_runtime": 8.4725,
"eval_samples_per_second": 1412.689,
"eval_steps_per_second": 22.189,
"step": 1400
},
{
"epoch": 2.94,
"learning_rate": 5.294117647058824e-06,
"loss": 1.408,
"step": 1500
},
{
"epoch": 2.94,
"eval_loss": 1.435463547706604,
"eval_runtime": 8.4775,
"eval_samples_per_second": 1411.85,
"eval_steps_per_second": 22.176,
"step": 1500
},
{
"epoch": 3.14,
"learning_rate": 4.313725490196079e-06,
"loss": 1.3609,
"step": 1600
},
{
"epoch": 3.14,
"eval_loss": 1.3331981897354126,
"eval_runtime": 8.4786,
"eval_samples_per_second": 1411.679,
"eval_steps_per_second": 22.174,
"step": 1600
},
{
"epoch": 3.33,
"learning_rate": 3.3333333333333333e-06,
"loss": 1.3398,
"step": 1700
},
{
"epoch": 3.33,
"eval_loss": 1.3791619539260864,
"eval_runtime": 8.4678,
"eval_samples_per_second": 1413.466,
"eval_steps_per_second": 22.202,
"step": 1700
},
{
"epoch": 3.53,
"learning_rate": 2.3529411764705885e-06,
"loss": 1.3224,
"step": 1800
},
{
"epoch": 3.53,
"eval_loss": 1.41716730594635,
"eval_runtime": 8.4259,
"eval_samples_per_second": 1420.506,
"eval_steps_per_second": 22.312,
"step": 1800
},
{
"epoch": 3.73,
"learning_rate": 1.3725490196078434e-06,
"loss": 1.3152,
"step": 1900
},
{
"epoch": 3.73,
"eval_loss": 1.3955893516540527,
"eval_runtime": 8.444,
"eval_samples_per_second": 1417.453,
"eval_steps_per_second": 22.264,
"step": 1900
},
{
"epoch": 3.92,
"learning_rate": 3.921568627450981e-07,
"loss": 1.3141,
"step": 2000
},
{
"epoch": 3.92,
"eval_loss": 1.3748189210891724,
"eval_runtime": 8.4509,
"eval_samples_per_second": 1416.303,
"eval_steps_per_second": 22.246,
"step": 2000
},
{
"epoch": 4.0,
"step": 2040,
"total_flos": 8.491863563129856e+16,
"train_loss": 0.025466402839211855,
"train_runtime": 27.6855,
"train_samples_per_second": 18855.037,
"train_steps_per_second": 73.685
},
{
"epoch": 4.0,
"eval_loss": 1.3331981897354126,
"eval_runtime": 8.4623,
"eval_samples_per_second": 1414.395,
"eval_steps_per_second": 22.216,
"step": 2040
}
],
"logging_steps": 100,
"max_steps": 2040,
"num_train_epochs": 4,
"save_steps": 100,
"total_flos": 8.491863563129856e+16,
"trial_name": null,
"trial_params": null
}