LLama-8b-MI-6e-7 / train_results.json
tengxiao1
TX
57ac55d
raw
history blame
232 Bytes
{
"epoch": 0.998691442030882,
"total_flos": 0.0,
"train_loss": 2.0152689970764226,
"train_runtime": 14277.4481,
"train_samples": 61135,
"train_samples_per_second": 4.282,
"train_steps_per_second": 0.033
}