Files
ronpo-qwen3-8b-fair-ronpo-k…/all_results.json

9 lines
232 B
JSON
Raw Permalink Normal View History

{
"epoch": 0.2860548271752086,
"total_flos": 0.0,
"train_loss": 0.4462373371918996,
"train_runtime": 7238.9345,
"train_samples": 50340,
"train_samples_per_second": 1.989,
"train_steps_per_second": 0.124
}