bart-base-detox / checkpoint-2295 /trainer_state.json
ChrisZeng's picture
Training in progress, epoch 16
a9676e8
{
"best_metric": 0.18199852108955383,
"best_model_checkpoint": "outputs/bart-base-detox/checkpoint-1620",
"epoch": 16.995391705069125,
"global_step": 2295,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"learning_rate": 9e-06,
"loss": 0.5633,
"step": 135
},
{
"epoch": 1.0,
"eval_loss": 0.25235414505004883,
"eval_runtime": 12.2006,
"eval_samples_per_second": 254.085,
"eval_steps_per_second": 31.802,
"step": 135
},
{
"epoch": 2.0,
"learning_rate": 8.000000000000001e-06,
"loss": 0.2589,
"step": 270
},
{
"epoch": 2.0,
"eval_loss": 0.21927106380462646,
"eval_runtime": 11.2792,
"eval_samples_per_second": 274.842,
"eval_steps_per_second": 34.4,
"step": 270
},
{
"epoch": 3.0,
"learning_rate": 7e-06,
"loss": 0.2307,
"step": 405
},
{
"epoch": 3.0,
"eval_loss": 0.19929151237010956,
"eval_runtime": 11.2313,
"eval_samples_per_second": 276.016,
"eval_steps_per_second": 34.546,
"step": 405
},
{
"epoch": 4.0,
"learning_rate": 6e-06,
"loss": 0.2171,
"step": 540
},
{
"epoch": 4.0,
"eval_loss": 0.20016591250896454,
"eval_runtime": 11.0923,
"eval_samples_per_second": 279.473,
"eval_steps_per_second": 34.979,
"step": 540
},
{
"epoch": 5.0,
"learning_rate": 5e-06,
"loss": 0.2027,
"step": 675
},
{
"epoch": 5.0,
"eval_loss": 0.19366399943828583,
"eval_runtime": 11.2232,
"eval_samples_per_second": 276.213,
"eval_steps_per_second": 34.571,
"step": 675
},
{
"epoch": 6.0,
"learning_rate": 4.000000000000001e-06,
"loss": 0.1946,
"step": 810
},
{
"epoch": 6.0,
"eval_loss": 0.19715721905231476,
"eval_runtime": 11.2739,
"eval_samples_per_second": 274.973,
"eval_steps_per_second": 34.416,
"step": 810
},
{
"epoch": 7.0,
"learning_rate": 3e-06,
"loss": 0.1874,
"step": 945
},
{
"epoch": 7.0,
"eval_loss": 0.19172348082065582,
"eval_runtime": 11.1804,
"eval_samples_per_second": 277.27,
"eval_steps_per_second": 34.704,
"step": 945
},
{
"epoch": 8.0,
"learning_rate": 2.0000000000000003e-06,
"loss": 0.1853,
"step": 1080
},
{
"epoch": 8.0,
"eval_loss": 0.1867838054895401,
"eval_runtime": 11.2543,
"eval_samples_per_second": 275.451,
"eval_steps_per_second": 34.476,
"step": 1080
},
{
"epoch": 9.0,
"learning_rate": 1.0000000000000002e-06,
"loss": 0.1811,
"step": 1215
},
{
"epoch": 9.0,
"eval_loss": 0.18896569311618805,
"eval_runtime": 11.6639,
"eval_samples_per_second": 265.777,
"eval_steps_per_second": 33.265,
"step": 1215
},
{
"epoch": 10.0,
"learning_rate": 0.0,
"loss": 0.1776,
"step": 1350
},
{
"epoch": 10.0,
"eval_loss": 0.18712429702281952,
"eval_runtime": 11.9321,
"eval_samples_per_second": 259.803,
"eval_steps_per_second": 32.517,
"step": 1350
},
{
"epoch": 11.0,
"learning_rate": 4.5e-06,
"loss": 0.1798,
"step": 1485
},
{
"epoch": 11.0,
"eval_loss": 0.18577729165554047,
"eval_runtime": 11.5191,
"eval_samples_per_second": 269.117,
"eval_steps_per_second": 33.683,
"step": 1485
},
{
"epoch": 12.0,
"learning_rate": 4.000000000000001e-06,
"loss": 0.1745,
"step": 1620
},
{
"epoch": 12.0,
"eval_loss": 0.18199852108955383,
"eval_runtime": 11.54,
"eval_samples_per_second": 268.63,
"eval_steps_per_second": 33.622,
"step": 1620
},
{
"epoch": 13.0,
"learning_rate": 3.5e-06,
"loss": 0.1689,
"step": 1755
},
{
"epoch": 13.0,
"eval_loss": 0.18266141414642334,
"eval_runtime": 11.4865,
"eval_samples_per_second": 269.881,
"eval_steps_per_second": 33.779,
"step": 1755
},
{
"epoch": 14.0,
"learning_rate": 3e-06,
"loss": 0.1707,
"step": 1890
},
{
"epoch": 14.0,
"eval_loss": 0.18427973985671997,
"eval_runtime": 11.7088,
"eval_samples_per_second": 264.757,
"eval_steps_per_second": 33.137,
"step": 1890
},
{
"epoch": 15.0,
"learning_rate": 2.5e-06,
"loss": 0.1658,
"step": 2025
},
{
"epoch": 15.0,
"eval_loss": 0.18342117965221405,
"eval_runtime": 11.3728,
"eval_samples_per_second": 272.58,
"eval_steps_per_second": 34.116,
"step": 2025
},
{
"epoch": 16.0,
"learning_rate": 2.0000000000000003e-06,
"loss": 0.1647,
"step": 2160
},
{
"epoch": 16.0,
"eval_loss": 0.18202748894691467,
"eval_runtime": 11.7571,
"eval_samples_per_second": 263.669,
"eval_steps_per_second": 33.001,
"step": 2160
},
{
"epoch": 17.0,
"learning_rate": 1.5e-06,
"loss": 0.1645,
"step": 2295
},
{
"epoch": 17.0,
"eval_loss": 0.18365143239498138,
"eval_runtime": 11.3384,
"eval_samples_per_second": 273.408,
"eval_steps_per_second": 34.22,
"step": 2295
}
],
"max_steps": 2700,
"num_train_epochs": 20,
"total_flos": 9142374164152320.0,
"trial_name": null,
"trial_params": null
}