Mistral-Nemo-Baseline-SFT / train_results.json
pbevan11's picture
Model save
3bc137e verified
raw
history blame contribute delete
243 Bytes
{
"epoch": 2.8363636363636364,
"total_flos": 50553107251200.0,
"train_loss": 1.237262059480716,
"train_runtime": 1937.0641,
"train_samples": 18999,
"train_samples_per_second": 5.391,
"train_steps_per_second": 0.02
}