170 lines
4.0 KiB
JSON
170 lines
4.0 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 5.0,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 90,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"epoch": 0.2888086642599278,
|
||
|
|
"grad_norm": 0.694216302872654,
|
||
|
|
"learning_rate": 1.7777777777777777e-05,
|
||
|
|
"loss": 0.6462,
|
||
|
|
"step": 5
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5776173285198556,
|
||
|
|
"grad_norm": 0.837614155906322,
|
||
|
|
"learning_rate": 4e-05,
|
||
|
|
"loss": 0.6855,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8664259927797834,
|
||
|
|
"grad_norm": 0.7964079287374712,
|
||
|
|
"learning_rate": 3.9625106212547696e-05,
|
||
|
|
"loss": 0.7169,
|
||
|
|
"step": 15
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.1155234657039712,
|
||
|
|
"grad_norm": 0.9598295257498393,
|
||
|
|
"learning_rate": 3.8514479385377813e-05,
|
||
|
|
"loss": 0.637,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.404332129963899,
|
||
|
|
"grad_norm": 0.7944514703083934,
|
||
|
|
"learning_rate": 3.6709756228258735e-05,
|
||
|
|
"loss": 0.4803,
|
||
|
|
"step": 25
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.6931407942238268,
|
||
|
|
"grad_norm": 0.6518082836692719,
|
||
|
|
"learning_rate": 3.4278594691157985e-05,
|
||
|
|
"loss": 0.4857,
|
||
|
|
"step": 30
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.9819494584837545,
|
||
|
|
"grad_norm": 0.6886973647996735,
|
||
|
|
"learning_rate": 3.1312137509730776e-05,
|
||
|
|
"loss": 0.4416,
|
||
|
|
"step": 35
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.2310469314079424,
|
||
|
|
"grad_norm": 0.6635042415823782,
|
||
|
|
"learning_rate": 2.792159532078314e-05,
|
||
|
|
"loss": 0.3906,
|
||
|
|
"step": 40
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.51985559566787,
|
||
|
|
"grad_norm": 0.6315628131041415,
|
||
|
|
"learning_rate": 2.423407744458822e-05,
|
||
|
|
"loss": 0.3395,
|
||
|
|
"step": 45
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.808664259927798,
|
||
|
|
"grad_norm": 0.5243150154226285,
|
||
|
|
"learning_rate": 2.038782663543649e-05,
|
||
|
|
"loss": 0.322,
|
||
|
|
"step": 50
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.0577617328519855,
|
||
|
|
"grad_norm": 0.9433727073495656,
|
||
|
|
"learning_rate": 1.6527036446661396e-05,
|
||
|
|
"loss": 0.2934,
|
||
|
|
"step": 55
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.3465703971119134,
|
||
|
|
"grad_norm": 0.4645384807088082,
|
||
|
|
"learning_rate": 1.2796445503905797e-05,
|
||
|
|
"loss": 0.2552,
|
||
|
|
"step": 60
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.6353790613718413,
|
||
|
|
"grad_norm": 0.3922743659029623,
|
||
|
|
"learning_rate": 9.33591134396618e-06,
|
||
|
|
"loss": 0.2543,
|
||
|
|
"step": 65
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.9241877256317688,
|
||
|
|
"grad_norm": 0.3518437547019525,
|
||
|
|
"learning_rate": 6.275167242625331e-06,
|
||
|
|
"loss": 0.2196,
|
||
|
|
"step": 70
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 4.1732851985559565,
|
||
|
|
"grad_norm": 0.33805015628047197,
|
||
|
|
"learning_rate": 3.7289585947406504e-06,
|
||
|
|
"loss": 0.2177,
|
||
|
|
"step": 75
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 4.462093862815885,
|
||
|
|
"grad_norm": 0.27964209039699955,
|
||
|
|
"learning_rate": 1.792741180677069e-06,
|
||
|
|
"loss": 0.2231,
|
||
|
|
"step": 80
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 4.750902527075812,
|
||
|
|
"grad_norm": 0.2753850482905527,
|
||
|
|
"learning_rate": 5.391025884035239e-07,
|
||
|
|
"loss": 0.1894,
|
||
|
|
"step": 85
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 5.0,
|
||
|
|
"grad_norm": 0.27245810015793187,
|
||
|
|
"learning_rate": 1.5040949915399173e-08,
|
||
|
|
"loss": 0.209,
|
||
|
|
"step": 90
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 5.0,
|
||
|
|
"step": 90,
|
||
|
|
"total_flos": 274293220966400.0,
|
||
|
|
"train_loss": 0.3892772012286716,
|
||
|
|
"train_runtime": 9724.0454,
|
||
|
|
"train_samples_per_second": 1.139,
|
||
|
|
"train_steps_per_second": 0.009
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 5,
|
||
|
|
"max_steps": 90,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 5,
|
||
|
|
"save_steps": 100,
|
||
|
|
"stateful_callbacks": {
|
||
|
|
"TrainerControl": {
|
||
|
|
"args": {
|
||
|
|
"should_epoch_stop": false,
|
||
|
|
"should_evaluate": false,
|
||
|
|
"should_log": false,
|
||
|
|
"should_save": false,
|
||
|
|
"should_training_stop": false
|
||
|
|
},
|
||
|
|
"attributes": {}
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"total_flos": 274293220966400.0,
|
||
|
|
"train_batch_size": 1,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|