初始化项目,由ModelHub XC社区提供模型

Model: promotion/ronpo-qwen3-8b-fair-ht-mnpo-safety-s42
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-07-18 11:51:09 +08:00
commit 502a45182e
18 changed files with 5148 additions and 0 deletions

9
all_results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"epoch": 0.8589835361488904,
"total_flos": 0.0,
"train_loss": 0.36845981503112446,
"train_runtime": 7265.877,
"train_samples": 16764,
"train_samples_per_second": 1.982,
"train_steps_per_second": 0.124
}