guanghulab/.github/ISSUE_TEMPLATE/ask-zhuyuan-training.md
Guanghu Domestic Migration d1e47f4565
Some checks are pending
自动更新代码和重启 / update-and-restart (push) Waiting to run
CI检查 + 自动部署 / check (push) Waiting to run
CI检查 + 自动部署 / deploy (push) Blocked by required conditions
重启聊天服务 / restart (push) Waiting to run
chore: import sanitized domestic snapshot for REPO-002
Source snapshot: ca48d3ddf926d79aa138306164169baf764bb829
2026-07-17 15:54:41 +08:00

1.2 KiB
Raw Blame History

name, about, title, labels, assignees
name about title labels assignees
🔥 问铸渊副将(训练相关) 训练正不正常?卡哪了?下一步该做什么?副将会读 data/training/state.json 后回答 [训练]
deputy-message-board
training

🔥 训练状态咨询

冰朔的问题:


📊 副将会自动读取的真相源

提交后副将会读以下数据后再回答:

  • data/training/state.json — 训练阶段、step、loss、GPU 指标、最近事件时间线
  • data/deputy-status.json — 副将自身健康
  • brain/fast-wake.json — 系统快速唤醒
  • 当前训练机: 119.45.160.137 (zy-gpu-train · V100×4)

💡 副将由 .github/workflows/deputy-message-board.yml 自动唤醒, 触发条件: Issue 标签包含 deputy-message-board。 LLM 多模型降级链: deepseek → qwen → moonshot → glm-4。

📜 国作登字-2026-A-00037559 · TCS-0002∞ 冰朔