fix: resolve NPU OOM with default training config

[curnane-lab]

Motivation

The default NPU training examples for Qwen3.5-4B DFlash use --num-anchors values (512) that cause out-of-memory errors on common 64GB Ascend NPU cards such as 910B(A2 node) and 910C(A3 node). This PR lowers the default to a value that fits within the available device memory while keeping the examples runnable out-of-the-box.

Modifications

• examples/run_qwen3.5_4b_dflash_online_npu.sh
  • Changed --num-anchors from 512 to 186
• examples/run_qwen3.5_4b_domino_online_npu.sh
  • Changed --num-anchors from 16 to 186

Both scripts now use the same --num-anchors 186 default, which avoids OOM on 64GB NPU devices.

Related Issues

N/A

Accuracy Test

Not applicable — this change only adjusts a training hyper-parameter default in example launch scripts. No model architecture or kernel code is modified.

Benchmark & Profiling

Not applicable — the change reduces memory usage for the default NPU example configuration.

Checklist

• Format your code according to the Code Formatting with Pre-Commit.
• Add unit tests as outlined in the Running Unit Tests.
• Update documentation / docstrings / example tutorials as needed, according to Writing Documentation.
• Provide throughput / latency benchmark results and accuracy evaluation results as needed, according to Benchmark and Profiling and Accuracy Results.
• For reviewers: If you haven't made any contributions to this PR and are only assisting with merging the main branch, please remove yourself as a co-author when merging the PR.
• Please feel free to join our Slack channel at https://sgl-fru7574.slack.com/archives/C09784E3EN6 to discuss your PR.

[gemini-code-assist]

Code Review

This pull request updates the --num-anchors parameter to 186 in both the run_qwen3.5_4b_dflash_online_npu.sh and run_qwen3.5_4b_domino_online_npu.sh example scripts. There are no review comments, so I have no feedback to provide.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

[FrankLeeeee]

I think 512 is the default setting for these two methods? I would like to keep them intact if possible, but you have a good point, perhaps we should raise some logs to alert the users to lower this value if the gpu memory is low.