mistral20pGrad / train_results.json
terry69's picture
Model save
4c4879e verified
raw
history blame
239 Bytes
{
"epoch": 1.0,
"total_flos": 2.1031592360023163e+18,
"train_loss": 0.7368072108388535,
"train_runtime": 156724.9428,
"train_samples": 103932,
"train_samples_per_second": 0.663,
"train_steps_per_second": 0.003
}