Sohaibsoussi's picture
Training in progress, step 2500
04afc11 verified
raw
history blame
2.03 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 4.716981132075472,
"eval_steps": 500,
"global_step": 1500,
"is_hyper_param_search": true,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_accuracy": 0.6780645161290323,
"eval_loss": 0.40519121289253235,
"eval_runtime": 2.0436,
"eval_samples_per_second": 1516.949,
"eval_steps_per_second": 31.807,
"step": 318
},
{
"epoch": 1.57,
"grad_norm": 0.8578532934188843,
"learning_rate": 1.650593990216632e-05,
"loss": 0.6413,
"step": 500
},
{
"epoch": 2.0,
"eval_accuracy": 0.8396774193548387,
"eval_loss": 0.14012566208839417,
"eval_runtime": 2.0827,
"eval_samples_per_second": 1488.438,
"eval_steps_per_second": 31.209,
"step": 636
},
{
"epoch": 3.0,
"eval_accuracy": 0.8964516129032258,
"eval_loss": 0.07267240434885025,
"eval_runtime": 2.1252,
"eval_samples_per_second": 1458.715,
"eval_steps_per_second": 30.586,
"step": 954
},
{
"epoch": 3.14,
"grad_norm": 0.5715540051460266,
"learning_rate": 1.3011879804332637e-05,
"loss": 0.1635,
"step": 1000
},
{
"epoch": 4.0,
"eval_accuracy": 0.9145161290322581,
"eval_loss": 0.05143408849835396,
"eval_runtime": 2.2923,
"eval_samples_per_second": 1352.353,
"eval_steps_per_second": 28.356,
"step": 1272
},
{
"epoch": 4.72,
"grad_norm": 0.5522882342338562,
"learning_rate": 9.517819706498952e-06,
"loss": 0.0817,
"step": 1500
}
],
"logging_steps": 500,
"max_steps": 2862,
"num_input_tokens_seen": 0,
"num_train_epochs": 9,
"save_steps": 500,
"total_flos": 391368939443328.0,
"train_batch_size": 48,
"trial_name": null,
"trial_params": {
"alpha": 0.9764301931204143,
"num_train_epochs": 9,
"temperature": 2
}
}