Files
DeepSeek-R1-Distill-Qwen-0.…/all_results.json
ModelHub XC ff5738198f 初始化项目,由ModelHub XC社区提供模型
Model: Blancy/DeepSeek-R1-Distill-Qwen-0.5B-GRPO
Source: Original Platform
2026-06-23 11:31:18 +08:00

8 lines
199 B
JSON

{
"total_flos": 0.0,
"train_loss": 0.02214584331980233,
"train_runtime": 5149.8835,
"train_samples": 1000,
"train_samples_per_second": 0.194,
"train_steps_per_second": 0.016
}