初始化项目,由ModelHub XC社区提供模型

Model: lightblue/DeepSeek-R1-Distill-Qwen-7B-Japanese
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-06-24 17:08:18 +08:00
commit 9a7835e1f2
20 changed files with 1558 additions and 0 deletions

8
train_results.json Normal file
View File

@@ -0,0 +1,8 @@
{
"epoch": 1.0,
"total_flos": 7.03161217014825e+16,
"train_loss": 0.50457603516786,
"train_runtime": 362.0803,
"train_samples_per_second": 1.016,
"train_steps_per_second": 0.127
}