ds_no_offload_liger_packing_zero2 / train_results.json
sedrickkeh's picture
End of training
425bd53 verified
raw
history blame contribute delete
206 Bytes
{
"epoch": 3.0,
"total_flos": 1.2010811790553252e+18,
"train_loss": 1.01401486992836,
"train_runtime": 1429.1641,
"train_samples_per_second": 1.207,
"train_steps_per_second": 0.013
}