初始化项目,由ModelHub XC社区提供模型

Model: xd2010/Qwen1.5-MOE-sft-math7k-densemixer
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-05-15 21:29:47 +08:00
commit 8b6e8f053d
22 changed files with 156960 additions and 0 deletions

8
train_results.json Normal file
View File

@@ -0,0 +1,8 @@
{
"total_flos": 20312064000.0,
"train_loss": 0.8405675888061523,
"train_runtime": 27.3869,
"train_samples": 6851,
"train_samples_per_second": 1.168,
"train_steps_per_second": 0.037
}