ramdhanfirdaus's picture
Training in progress, step 800, checkpoint
ee79f93
raw
history blame
2.92 kB
{
"best_metric": 1.6834267377853394,
"best_model_checkpoint": "./outputs/checkpoint-800",
"epoch": 0.5828779599271403,
"eval_steps": 100,
"global_step": 800,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 2.1831,
"step": 100
},
{
"epoch": 0.07,
"eval_loss": 2.0129995346069336,
"eval_runtime": 143.7126,
"eval_samples_per_second": 43.657,
"eval_steps_per_second": 5.462,
"step": 100
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 1.9629,
"step": 200
},
{
"epoch": 0.15,
"eval_loss": 1.9196887016296387,
"eval_runtime": 143.7285,
"eval_samples_per_second": 43.652,
"eval_steps_per_second": 5.462,
"step": 200
},
{
"epoch": 0.22,
"learning_rate": 0.0002,
"loss": 1.8881,
"step": 300
},
{
"epoch": 0.22,
"eval_loss": 1.8589884042739868,
"eval_runtime": 143.706,
"eval_samples_per_second": 43.659,
"eval_steps_per_second": 5.463,
"step": 300
},
{
"epoch": 0.29,
"learning_rate": 0.0002,
"loss": 1.8368,
"step": 400
},
{
"epoch": 0.29,
"eval_loss": 1.813078761100769,
"eval_runtime": 143.7443,
"eval_samples_per_second": 43.647,
"eval_steps_per_second": 5.461,
"step": 400
},
{
"epoch": 0.36,
"learning_rate": 0.0002,
"loss": 1.7849,
"step": 500
},
{
"epoch": 0.36,
"eval_loss": 1.7718570232391357,
"eval_runtime": 143.7117,
"eval_samples_per_second": 43.657,
"eval_steps_per_second": 5.462,
"step": 500
},
{
"epoch": 0.44,
"learning_rate": 0.0002,
"loss": 1.7567,
"step": 600
},
{
"epoch": 0.44,
"eval_loss": 1.742559790611267,
"eval_runtime": 143.6984,
"eval_samples_per_second": 43.661,
"eval_steps_per_second": 5.463,
"step": 600
},
{
"epoch": 0.51,
"learning_rate": 0.0002,
"loss": 1.718,
"step": 700
},
{
"epoch": 0.51,
"eval_loss": 1.7121007442474365,
"eval_runtime": 143.6682,
"eval_samples_per_second": 43.67,
"eval_steps_per_second": 5.464,
"step": 700
},
{
"epoch": 0.58,
"learning_rate": 0.0002,
"loss": 1.6976,
"step": 800
},
{
"epoch": 0.58,
"eval_loss": 1.6834267377853394,
"eval_runtime": 143.6959,
"eval_samples_per_second": 43.662,
"eval_steps_per_second": 5.463,
"step": 800
}
],
"logging_steps": 100,
"max_steps": 4116,
"num_train_epochs": 3,
"save_steps": 100,
"total_flos": 4.770964776085094e+16,
"trial_name": null,
"trial_params": null
}