Llama-8b-MI1 / train_results.json
Teng Xiao
TX
dd97401
{
"epoch": 0.998691442030882,
"total_flos": 0.0,
"train_loss": 8.9737869758526,
"train_runtime": 8198.1649,
"train_samples": 61135,
"train_samples_per_second": 7.457,
"train_steps_per_second": 0.058
}