tmnam20's picture
Training in progress, step 6000, checkpoint
599720a
raw
history blame
2.8 kB
{
"best_metric": 1.0983413457870483,
"best_model_checkpoint": "/kaggle/output/checkpoint-6000",
"epoch": 0.24445893089960888,
"eval_steps": 1000,
"global_step": 6000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 2.7777777777777777e-11,
"loss": 1.0788,
"step": 1
},
{
"epoch": 0.04,
"learning_rate": 2.7750000000000004e-08,
"loss": 1.2063,
"step": 1000
},
{
"epoch": 0.04,
"eval_accuracy": 0.32894211576846305,
"eval_loss": 1.109653115272522,
"eval_runtime": 50.6057,
"eval_samples_per_second": 99.001,
"eval_steps_per_second": 12.39,
"step": 1000
},
{
"epoch": 0.08,
"learning_rate": 5.5527777777777784e-08,
"loss": 1.1293,
"step": 2000
},
{
"epoch": 0.08,
"eval_accuracy": 0.32994011976047904,
"eval_loss": 1.1056402921676636,
"eval_runtime": 50.7169,
"eval_samples_per_second": 98.784,
"eval_steps_per_second": 12.363,
"step": 2000
},
{
"epoch": 0.12,
"learning_rate": 8.327777777777778e-08,
"loss": 1.1263,
"step": 3000
},
{
"epoch": 0.12,
"eval_accuracy": 0.3347305389221557,
"eval_loss": 1.1063188314437866,
"eval_runtime": 50.8081,
"eval_samples_per_second": 98.606,
"eval_steps_per_second": 12.341,
"step": 3000
},
{
"epoch": 0.16,
"learning_rate": 1.1105555555555557e-07,
"loss": 1.1211,
"step": 4000
},
{
"epoch": 0.16,
"eval_accuracy": 0.34291417165668664,
"eval_loss": 1.1010773181915283,
"eval_runtime": 50.6469,
"eval_samples_per_second": 98.92,
"eval_steps_per_second": 12.38,
"step": 4000
},
{
"epoch": 0.2,
"learning_rate": 1.3883333333333335e-07,
"loss": 1.1176,
"step": 5000
},
{
"epoch": 0.2,
"eval_accuracy": 0.3395209580838323,
"eval_loss": 1.099412441253662,
"eval_runtime": 50.7266,
"eval_samples_per_second": 98.765,
"eval_steps_per_second": 12.36,
"step": 5000
},
{
"epoch": 0.24,
"learning_rate": 1.6658333333333335e-07,
"loss": 1.1165,
"step": 6000
},
{
"epoch": 0.24,
"eval_accuracy": 0.3596806387225549,
"eval_loss": 1.0983413457870483,
"eval_runtime": 50.7493,
"eval_samples_per_second": 98.721,
"eval_steps_per_second": 12.355,
"step": 6000
}
],
"logging_steps": 1000,
"max_steps": 10000000,
"num_train_epochs": 408,
"save_steps": 1000,
"total_flos": 1.2542130782208e+16,
"trial_name": null,
"trial_params": null
}