Files
Qwen3-Codeforces-GRPO/train_results.json

8 lines
181 B
JSON
Raw Permalink Normal View History

{
"total_flos": 0.0,
"train_loss": 0.0,
"train_runtime": 1.5617,
"train_samples": 16676,
"train_samples_per_second": 10.245,
"train_steps_per_second": 0.64
}