mistral5pGrad / train_results.json
terry69's picture
Model save
176fa6b verified
raw
history blame contribute delete
239 Bytes
{
"epoch": 1.0,
"total_flos": 2.1410486778152878e+18,
"train_loss": 0.7449803688549643,
"train_runtime": 152626.4688,
"train_samples": 103932,
"train_samples_per_second": 0.681,
"train_steps_per_second": 0.003
}