elonmusk / trainer_state.json
Frikallo's picture
End of training
06f5f90
raw
history blame
5.5 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"global_step": 209,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.02,
"learning_rate": 0.0001339177033492823,
"loss": 4.9076,
"step": 5
},
{
"epoch": 0.05,
"learning_rate": 0.0001306354066985646,
"loss": 4.477,
"step": 10
},
{
"epoch": 0.07,
"learning_rate": 0.0001273531100478469,
"loss": 4.4198,
"step": 15
},
{
"epoch": 0.1,
"learning_rate": 0.0001240708133971292,
"loss": 4.2271,
"step": 20
},
{
"epoch": 0.12,
"learning_rate": 0.00012078851674641149,
"loss": 4.0551,
"step": 25
},
{
"epoch": 0.14,
"learning_rate": 0.00011750622009569378,
"loss": 3.9819,
"step": 30
},
{
"epoch": 0.17,
"learning_rate": 0.00011422392344497608,
"loss": 3.9541,
"step": 35
},
{
"epoch": 0.19,
"learning_rate": 0.00011094162679425838,
"loss": 3.9238,
"step": 40
},
{
"epoch": 0.22,
"learning_rate": 0.00010765933014354067,
"loss": 3.888,
"step": 45
},
{
"epoch": 0.24,
"learning_rate": 0.00010437703349282297,
"loss": 3.6435,
"step": 50
},
{
"epoch": 0.26,
"learning_rate": 0.00010109473684210525,
"loss": 3.548,
"step": 55
},
{
"epoch": 0.29,
"learning_rate": 9.781244019138755e-05,
"loss": 3.6575,
"step": 60
},
{
"epoch": 0.31,
"learning_rate": 9.453014354066986e-05,
"loss": 3.7024,
"step": 65
},
{
"epoch": 0.33,
"learning_rate": 9.124784688995216e-05,
"loss": 3.5278,
"step": 70
},
{
"epoch": 0.36,
"learning_rate": 8.796555023923445e-05,
"loss": 3.4996,
"step": 75
},
{
"epoch": 0.38,
"learning_rate": 8.468325358851675e-05,
"loss": 3.5006,
"step": 80
},
{
"epoch": 0.41,
"learning_rate": 8.140095693779905e-05,
"loss": 3.4716,
"step": 85
},
{
"epoch": 0.43,
"learning_rate": 7.811866028708133e-05,
"loss": 3.3409,
"step": 90
},
{
"epoch": 0.45,
"learning_rate": 7.483636363636363e-05,
"loss": 3.3467,
"step": 95
},
{
"epoch": 0.48,
"learning_rate": 7.155406698564593e-05,
"loss": 3.3855,
"step": 100
},
{
"epoch": 0.5,
"learning_rate": 6.827177033492822e-05,
"loss": 3.3672,
"step": 105
},
{
"epoch": 0.53,
"learning_rate": 6.498947368421052e-05,
"loss": 3.2343,
"step": 110
},
{
"epoch": 0.55,
"learning_rate": 6.170717703349283e-05,
"loss": 3.3405,
"step": 115
},
{
"epoch": 0.57,
"learning_rate": 5.842488038277512e-05,
"loss": 3.2181,
"step": 120
},
{
"epoch": 0.6,
"learning_rate": 5.5142583732057416e-05,
"loss": 3.2851,
"step": 125
},
{
"epoch": 0.62,
"learning_rate": 5.1860287081339706e-05,
"loss": 3.2362,
"step": 130
},
{
"epoch": 0.65,
"learning_rate": 4.857799043062201e-05,
"loss": 3.1771,
"step": 135
},
{
"epoch": 0.67,
"learning_rate": 4.529569377990431e-05,
"loss": 3.2146,
"step": 140
},
{
"epoch": 0.69,
"learning_rate": 4.2013397129186604e-05,
"loss": 3.2209,
"step": 145
},
{
"epoch": 0.72,
"learning_rate": 3.8731100478468894e-05,
"loss": 3.2406,
"step": 150
},
{
"epoch": 0.74,
"learning_rate": 3.54488038277512e-05,
"loss": 3.0664,
"step": 155
},
{
"epoch": 0.77,
"learning_rate": 3.2166507177033495e-05,
"loss": 3.0473,
"step": 160
},
{
"epoch": 0.79,
"learning_rate": 2.8884210526315788e-05,
"loss": 3.2034,
"step": 165
},
{
"epoch": 0.81,
"learning_rate": 2.5601913875598085e-05,
"loss": 3.1364,
"step": 170
},
{
"epoch": 0.84,
"learning_rate": 2.2319617224880382e-05,
"loss": 3.0634,
"step": 175
},
{
"epoch": 0.86,
"learning_rate": 1.903732057416268e-05,
"loss": 3.1028,
"step": 180
},
{
"epoch": 0.89,
"learning_rate": 1.5755023923444976e-05,
"loss": 3.0197,
"step": 185
},
{
"epoch": 0.91,
"learning_rate": 1.2472727272727273e-05,
"loss": 3.0413,
"step": 190
},
{
"epoch": 0.93,
"learning_rate": 9.19043062200957e-06,
"loss": 3.0188,
"step": 195
},
{
"epoch": 0.96,
"learning_rate": 5.908133971291866e-06,
"loss": 2.9653,
"step": 200
},
{
"epoch": 0.98,
"learning_rate": 2.6258373205741627e-06,
"loss": 3.063,
"step": 205
},
{
"epoch": 1.0,
"step": 209,
"total_flos": 109220069376000.0,
"train_loss": 3.4709447650818164,
"train_runtime": 56.0951,
"train_samples_per_second": 3.726,
"train_steps_per_second": 3.726
}
],
"max_steps": 209,
"num_train_epochs": 1,
"total_flos": 109220069376000.0,
"trial_name": null,
"trial_params": null
}