|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 3.0, |
|
"eval_steps": 500, |
|
"global_step": 3792, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.0, |
|
"learning_rate": 0.0, |
|
"loss": 4.2141, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"learning_rate": 2.4750000000000004e-06, |
|
"loss": 2.7204, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.79, |
|
"learning_rate": 4.975000000000001e-06, |
|
"loss": 1.7191, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 1.19, |
|
"learning_rate": 7.4750000000000004e-06, |
|
"loss": 1.5429, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 1.58, |
|
"learning_rate": 9.975e-06, |
|
"loss": 1.4304, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 1.98, |
|
"learning_rate": 1.2475e-05, |
|
"loss": 1.3428, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 2.37, |
|
"learning_rate": 1.4975e-05, |
|
"loss": 1.2759, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 2.77, |
|
"learning_rate": 1.7475e-05, |
|
"loss": 1.2186, |
|
"step": 3500 |
|
}, |
|
{ |
|
"before_init_mem_cpu": 1261252608, |
|
"before_init_mem_gpu": 0, |
|
"epoch": 3.0, |
|
"init_mem_cpu_alloc_delta": 835670016, |
|
"init_mem_cpu_peaked_delta": 91971584, |
|
"init_mem_gpu_alloc_delta": 443396608, |
|
"init_mem_gpu_peaked_delta": 0, |
|
"step": 3792, |
|
"total_flos": 2.555110221659767e+18, |
|
"train_loss": 1.5746790189531785, |
|
"train_mem_cpu_alloc_delta": 1865678848, |
|
"train_mem_cpu_peaked_delta": 333721600, |
|
"train_mem_gpu_alloc_delta": 1350910464, |
|
"train_mem_gpu_peaked_delta": 21529119744, |
|
"train_runtime": 8836.9123, |
|
"train_samples_per_second": 1098.03, |
|
"train_steps_per_second": 0.429 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 3792, |
|
"num_train_epochs": 3, |
|
"save_steps": 500, |
|
"total_flos": 2.555110221659767e+18, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|