InternLM
- 20 skills
- 0 followers
- 17 hours ago last updated
- ▌ Model Patch · internlm bundleMerge a base HF model dir with an extra HF model dir by index diff — take tensors only present in extra, append them to a new output dir in HF-standard shard layout. Base wins on overlap.
- ▌ Add Hf Model · internlmAdd support for a HuggingFace model in XTuner so it can be trained with the full set of parallelism strategies and optimization switches. The user can provide a HuggingFace Hub repo id, a local model directory, or a model family already supported by `transformers`. This skill walks through the full path: locating the reference implementation, classifying the model, splitting new code across the model / module / ops layers, implementing the XTuner model class and config, registering the entry point, validating bitwise numerical parity with HuggingFace, writing regression tests, and enabling the training optimizations (EP/SP, micro-batch, torch.compile, fp8, activation offload).
- ▌ Sphinx Debug · internlmUse when `make html` or `sphinx-build` fails with `Extension error (sphinx.ext.autosummary)`, `ImportExceptionGroup`, or other Sphinx autosummary / autodoc import failures. Provides automated pdb diagnosis for `autodoc_mock_imports` issues.
- ▌ Xtuner Trace · internlm bundleUse when the user wants to instrument, inspect, or debug XTuner traces to reconstruct a sample/request execution path, follow cross-process call chains, or identify latency hotspots, bottleneck stages, and abnormal paths from span data.
- ▌ Model Normalize · internlm bundleNormalize an HF model checkpoint — quantize bf16/fp16 weights to per-block FP8, and/or repack engine-named safetensors into HF-standard ~4GB shards.
- ▌ Add Chat Template · internlm bundleAdd or audit an XTuner chat-template implementation and its loss mask from an official Hugging Face model repository or local model directory. Use when supporting a new model under xtuner/v1/data_proto/messages, checking an existing implementation against tokenizer or processor apply_chat_template, deciding which assistant output and stop/EOS tokens receive loss, adding regression tests, or validating that exported training behavior matches vLLM and SGLang inference.
- ▌ Check Hf Config Save · internlm bundleImplement a missing XTuner Hugging Face config export from the official HF model when possible, then validate it against the installed Transformers round-trip and versioned inference-engine field contracts. Use when adding a model, when hf_config is missing or returns None, when changing from_hf/hf_config/save_hf, when reviewing config.json differences, or when debugging vLLM/SGLang checkpoint-load or inference failures. Also use for requests named check_hf_config_save.
- ▌ Xtuner Custom Judger · internlmUse when creating, modifying, or reviewing XTuner RL judgers under xtuner/v1/rl/judger or configs that build custom judgers. Enforces the preferred Judger payload contract and the BaseJudger/ComposedJudger constraints.
- ▌ Xtuner Sync Supported Models · internlm bundleSynchronize xtuner's supported model documentation (docs/en/pretrain_sft/advanced_tutorial/model.md and docs/zh_cn/pretrain_sft/advanced_tutorial/model.md) with the actual Config classes defined under xtuner/v1/model/. Use when (1) new TransformerConfig, MoEConfig, or BaseComposeConfig subclasses are added, removed, or renamed in xtuner/v1/model/, (2) existing model configs change their inheritance hierarchy, scale, or HuggingFace counterpart, or (3) a code review or user request points out that model.md is out of sync with the codebase.
- ▌ Cross Validated Search · internlm bundleOpenClaw skill for source-backed web search, page reading, and evidence-aware claim checking. Use it to verify factual answers with live search results and explicit source handling.
- ▌
- ▌
- ▌ Docker Build · internlmBuild a CUDA 13.0 or CUDA 12.8 LMDeploy Docker image and push it to the inner registry.
- ▌ Ddgs Search · internlm bundleFree multi-engine web search via ddgs CLI (DuckDuckGo, Google, Bing, Brave, Yandex, Yahoo, Wikipedia) + arXiv API search. No API keys required. Use when user needs web search, research paper discovery, or when other skills need a search backend. Drop-in replacement for web-search-plus.
- ▌ 03 Task1 · internlmCoordinates meeting scheduling via Gmail and Calendar APIs. Use when: scheduling a meeting, checking participant availability, sending coordination emails, creating calendar events, or notifying organizers of confirmed bookings.
- ▌ 03 Task2 · internlmExtracts action items and pending tasks from Slack messages. Use when: reviewing unread messages for todos, finding deadlines or requests directed at the user, or producing a task summary from recent Slack activity.
- ▌ 03 Task3 · internlmAnalyzes Slack messages to assess deal or project feasibility. Use when: synthesizing fragmented stakeholder input on a deal, identifying conflicting requirements or risks, or producing a feasibility report for leadership.
- ▌ 03 Task4 · internlmReads recent Slack messages to compile an accurate project status report and saves it as a draft for review before sending to a client. Use when: preparing a client-facing update, reconciling conflicting or revised numbers from multiple sources, or drafting a message that requires human review before sending.
- ▌ 03 Task5 · internlmTriages a support inbox, prioritizes urgent issues, routes each to the right internal team member, and drafts any customer-facing replies for review. Use when: catching up on a backlog of support messages, producing an escalation report, or routing issues to appropriate owners.
- ▌ 03 Task6 · internlm整合多部门 Slack 消息，交叉核对数据，识别依赖关系与风险，为高管起草董事会级别的执行摘要草稿。适用场景：跨部门项目状态汇总、董事会会议准备、需要数据对账和风险识别的高管报告。