del split func for fused_experts

[xxyux]

Before submitting

• Lint code. If there are lint issues, please format the code first.

# Install and register `pre-commit` in the project folder
pip install pre-commit && pre-commit install

# Process previous code files separately
pre-commit run --file XXXX.py

• Add test cases into tests folder. If there are codecov issues, please add tests cases first.

PR types

Performance optimization

PR changes

APIs

Description

Removed the slicing function for fused experts and replaced it with a batched method to compute the muon optimizer, improving performance without compromising accuracy.

[paddle-bot]

Thanks for your contribution!

[codecov-commenter]

Codecov Report

❌ Patch coverage is 0% with 6 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (develop@79332b4). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
paddleformers/transformers/glm_moe_dsa/modeling.py	0.00%	3 Missing ⚠️
paddleformers/transformers/minimax_m2/modeling.py	0.00%	3 Missing ⚠️

❌ Your patch status has failed because the patch coverage (0.00%) is below the target coverage (75.00%). You can increase the patch coverage or adjust the target coverage.

Additional details and impacted files

@@            Coverage Diff             @@
##             develop    #4501   +/-   ##
==========================================
  Coverage           ?   46.46%           
==========================================
  Files              ?      475           
  Lines              ?    90580           
  Branches           ?        0           
==========================================
  Hits               ?    42089           
  Misses             ?    48491           
  Partials           ?        0           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[Paddle-CI-Bot]

PaddleFormers Log Analysis

Run #26155592839 · Attempt 1

日志分析报告

流水线名称	问题标签	修复建议
CI_ILUVATAR	环境问题（container failed）	自定义容器执行失败，请联系 CI 维护人员检查 iluvatar self-hosted runner 的容器配置与资源状态。

---

失败的测试 case：

无具体测试 case 执行到——Job 在 PR 合并/脚本执行阶段即已崩溃，未进入任何模型测试步骤。

根本原因分析：

日志在执行以下步骤时中止：

git fetch origin pull/4501/head:pr/4501
git merge --no-ff pr/4501

随后连续出现两次：

##[error]Error: failed to run script step: [object Object]
##[error]Process completed with exit code 1.
##[error]Executing the custom container implementation failed. Please contact your self hosted runner administrator.

这是典型的 容器环境问题，报错来自 /home/runner/k8s/index.js（自定义 K8s 容器执行层），并非 PR 代码本身引入的 Bug。PR #4501（分支 batched_ns，触发人 xxyux）的代码变更并未被执行。

修复建议：

1. 联系 CI 维护人员：错误信息明确提示 "Please contact your self hosted runner administrator"，需排查 iluvatar-gpu-2-nczzk-runner-rlsjt 节点上 K8s 容器运行时状态。
2. 直接 Rerun：若为偶发容器调度问题，可先尝试触发 Rerun，观察是否稳定复现。
3. 排查方向：检查 iluvatar runner 的 Docker/容器配置、K8s Pod 资源限制、及 /home/runner/k8s/index.js 的执行权限与依赖是否完整。

---

_{🔄 每次 Re-run 后自动更新}

[From00]

LGTM