mistral20pGrad_1 / train_results.json
terry69's picture
Model save
4efd986 verified
raw
history blame
253 Bytes
{
"epoch": 0.9990766389658357,
"total_flos": 2.2228397020277637e+18,
"train_loss": 0.7336494325489742,
"train_runtime": 26329.8311,
"train_samples": 103932,
"train_samples_per_second": 3.947,
"train_steps_per_second": 0.021
}