Llama-8b-MI01 / train_results.json
Teng Xiao
TX
9484d16
{
"epoch": 0.998691442030882,
"total_flos": 0.0,
"train_loss": 1.7629729861733299,
"train_runtime": 8203.4319,
"train_samples": 61135,
"train_samples_per_second": 7.452,
"train_steps_per_second": 0.058
}