Files
math-1b-sft-numinamath-bs51…/train_results.json

8 lines
208 B
JSON
Raw Permalink Normal View History

{
"epoch": 1.0,
"total_flos": 9.42165996526266e+18,
"train_loss": 0.5865233610470143,
"train_runtime": 4862.2389,
"train_samples_per_second": 176.769,
"train_steps_per_second": 0.345
}