travis-scott / trainer_state.json
AlekseyKorshuk's picture
huggingartists
077312f
raw
history blame
4.66 kB
{
"best_metric": 3.3389594554901123,
"best_model_checkpoint": "output/travis-scott/checkpoint-174",
"epoch": 1.0,
"global_step": 174,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 0.0001369206553164936,
"loss": 3.9488,
"step": 5
},
{
"epoch": 0.06,
"learning_rate": 0.00013608489629373538,
"loss": 3.7969,
"step": 10
},
{
"epoch": 0.09,
"learning_rate": 0.00013469952948681868,
"loss": 3.6955,
"step": 15
},
{
"epoch": 0.11,
"learning_rate": 0.00013277583754449623,
"loss": 3.6337,
"step": 20
},
{
"epoch": 0.14,
"learning_rate": 0.00013032948732148497,
"loss": 3.5633,
"step": 25
},
{
"epoch": 0.17,
"learning_rate": 0.0001273804022850966,
"loss": 3.4878,
"step": 30
},
{
"epoch": 0.2,
"learning_rate": 0.0001239526002553348,
"loss": 3.6424,
"step": 35
},
{
"epoch": 0.23,
"learning_rate": 0.0001200739977999292,
"loss": 3.5744,
"step": 40
},
{
"epoch": 0.26,
"learning_rate": 0.00011577618287734484,
"loss": 3.4553,
"step": 45
},
{
"epoch": 0.29,
"learning_rate": 0.0001110941575793986,
"loss": 3.5316,
"step": 50
},
{
"epoch": 0.32,
"learning_rate": 0.00010606605306862761,
"loss": 3.4538,
"step": 55
},
{
"epoch": 0.34,
"learning_rate": 0.00010073281903200561,
"loss": 3.2755,
"step": 60
},
{
"epoch": 0.37,
"learning_rate": 9.513789018014557e-05,
"loss": 3.2518,
"step": 65
},
{
"epoch": 0.4,
"learning_rate": 8.932683250807268e-05,
"loss": 3.4163,
"step": 70
},
{
"epoch": 0.43,
"learning_rate": 8.334697219847626e-05,
"loss": 3.3031,
"step": 75
},
{
"epoch": 0.46,
"learning_rate": 7.724701018971185e-05,
"loss": 3.2611,
"step": 80
},
{
"epoch": 0.49,
"learning_rate": 7.107662554757363e-05,
"loss": 3.3722,
"step": 85
},
{
"epoch": 0.52,
"learning_rate": 6.488607087104036e-05,
"loss": 3.3329,
"step": 90
},
{
"epoch": 0.55,
"learning_rate": 5.872576302707542e-05,
"loss": 3.3299,
"step": 95
},
{
"epoch": 0.57,
"learning_rate": 5.264587254760233e-05,
"loss": 3.3178,
"step": 100
},
{
"epoch": 0.6,
"learning_rate": 4.6695915032671784e-05,
"loss": 3.4824,
"step": 105
},
{
"epoch": 0.63,
"learning_rate": 4.092434788749914e-05,
"loss": 3.412,
"step": 110
},
{
"epoch": 0.66,
"learning_rate": 3.537817567760932e-05,
"loss": 3.121,
"step": 115
},
{
"epoch": 0.69,
"learning_rate": 3.0102567316140575e-05,
"loss": 3.3823,
"step": 120
},
{
"epoch": 0.72,
"learning_rate": 2.5140488200994943e-05,
"loss": 3.217,
"step": 125
},
{
"epoch": 0.75,
"learning_rate": 2.0532350297768717e-05,
"loss": 3.2794,
"step": 130
},
{
"epoch": 0.78,
"learning_rate": 1.6315683018244145e-05,
"loss": 3.232,
"step": 135
},
{
"epoch": 0.8,
"learning_rate": 1.2524827574860022e-05,
"loss": 3.2429,
"step": 140
},
{
"epoch": 0.83,
"learning_rate": 9.190657300387505e-06,
"loss": 3.2955,
"step": 145
},
{
"epoch": 0.86,
"learning_rate": 6.340326210572357e-06,
"loss": 3.2237,
"step": 150
},
{
"epoch": 0.89,
"learning_rate": 3.9970478574895515e-06,
"loss": 3.2312,
"step": 155
},
{
"epoch": 0.92,
"learning_rate": 2.179906274664157e-06,
"loss": 3.3031,
"step": 160
},
{
"epoch": 0.95,
"learning_rate": 9.037005536513067e-07,
"loss": 3.4051,
"step": 165
},
{
"epoch": 0.98,
"learning_rate": 1.7882431786864473e-07,
"loss": 3.4258,
"step": 170
},
{
"epoch": 1.0,
"eval_loss": 3.3389594554901123,
"eval_runtime": 11.911,
"eval_samples_per_second": 20.989,
"eval_steps_per_second": 2.687,
"step": 174
}
],
"max_steps": 174,
"num_train_epochs": 1,
"total_flos": 181597962240000.0,
"trial_name": null,
"trial_params": null
}