Files
Qwen2.5-1.5B-GRPO-v5/train_results.json

8 lines
189 B
JSON
Raw Normal View History

{
"total_flos": 0.0,
"train_loss": 0.0,
"train_runtime": 1.9027,
"train_samples": 93733,
"train_samples_per_second": 197054.689,
"train_steps_per_second": 8209.473
}