[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78SWFP8SWYP25TM5CM95VZB":3,"skill-same:01M78SWFP8SWYP25TM5CM95VZB":70},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":26,"commitSha":27,"skillDirectory":28,"forksCount":29,"repositoryHtmlUrl":30,"repositoryHomepageUrl":6,"ownerHtmlUrl":31,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":32,"install":33,"resources":45},"01M78SWFP8SWYP25TM5CM95VZB","translation",null,"Run and troubleshoot checkpoint translation with translate.py and transformer.Translator beam search.","VectorSpaceLab\u002FAREX-Skill","VectorSpaceLab","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F180637102?v=4","01M7538C5WPSJGWF7KHR4EFZ9F","01M7538C6R2GFP53NPXTCZNBRT",351,"MIT","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fattention-is-all-you-need-pytorch\u002Fsub-skills\u002Ftranslation\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: translation\ndescription: \"Run and troubleshoot checkpoint translation with translate.py and\n  transformer.Translator beam search.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# Translation Sub-skill\n\nUse this sub-skill when a future task needs to run, validate, or troubleshoot\ncheckpoint-based translation for this repository. It covers the `translate.py`\ncommand-line flow, the `load_model` checkpoint contract, and programmatic beam\nsearch with `transformer.Translator`.\n\n## Route here for\n\n- Validating that a training checkpoint has the `settings` and `model` objects\n  required by translation before launching a full decode job.\n- Checking that a preprocessing pickle contains the `test` split plus source and\n  target vocab fields expected by `translate.py`.\n- Running the repository translation CLI on CPU or CUDA and interpreting its\n  output file.\n- Calling `Translator.translate_sentence(...)` directly for one already-tokenized\n  source sentence.\n- Debugging beam-size, max-length, token-index, batch-size-one, and device\n  mismatches in translation.\n\n## Do not route here for\n\n- Creating preprocessing pickles or BPE files; use the data-preparation skill.\n- Producing checkpoints, choosing training flags, or optimizer settings; use the\n  training skill.\n- Explaining encoder\u002Fdecoder internals, attention masks, or parameter shapes in\n  depth; use the model-architecture skill.\n\n## Start with these references\n\n1. Read [references\u002Fcli-reference.md](references\u002Fcli-reference.md) for the\n   `translate.py` CLI, checkpoint\u002Fdata-pickle schemas, device selection, and\n   output semantics.\n2. Read [references\u002Fapi-reference.md](references\u002Fapi-reference.md) for\n   programmatic `Transformer` construction from a checkpoint and\n   `Translator.translate_sentence` beam search usage.\n3. Read [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md) when a\n   checkpoint, pickle, vocabulary, CUDA, OOV, or BPE issue appears.\n\n## Bundled helpers\n\n- `scripts\u002Finspect_checkpoint.py` safely gates unpickling behind an explicit\n  trust flag, then validates checkpoint and optional data-pickle schemas.\n- `scripts\u002Ftranslation_smoke_check.py` builds a tiny deterministic Transformer\n  and exercises `Translator` beam search without requiring a trained checkpoint\n  or dataset.\n\nTypical preflight sequence:\n\n```bash\npython scripts\u002Finspect_checkpoint.py \\\n  --checkpoint trained.chkpt --data-pkl m30k_deen_shr.pkl --trust-inputs\npython scripts\u002Ftranslation_smoke_check.py \\\n  --repo-root \u002Fpath\u002Fto\u002Frepo --device cpu\n```\n\n## Operating reminders\n\n- `translate.py` loads the checkpoint with `torch.load(..., map_location=device)`\n  and then reconstructs `Transformer` from `checkpoint['settings']`; incompatible\n  settings are found only at load time unless you preflight them.\n- The CLI sets `opt.cuda = not opt.no_cuda`; on a CPU-only host include\n  `-no_cuda` or model\u002Fdevice transfer will fail.\n- `Translator.translate_sentence` asserts batch size one. For multiple examples,\n  iterate one sentence at a time or implement a separate batched decoder.\n- Output lines are target-vocabulary token strings with `\u003Cs>` and `\u003C\u002Fs>` removed\n  by simple string replacement. BPE post-decoding is marked TODO\u002Fnot ready in the\n  repository flow.\n","# Translation Sub-skill\n\nUse this sub-skill when a future task needs to run, validate, or troubleshoot\ncheckpoint-based translation for this repository. It covers the `translate.py`\ncommand-line flow, the `load_model` checkpoint contract, and programmatic beam\nsearch with `transformer.Translator`.\n\n## Route here for\n\n- Validating that a training checkpoint has the `settings` and `model` objects\n  required by translation before launching a full decode job.\n- Checking that a preprocessing pickle contains the `test` split plus source and\n  target vocab fields expected by `translate.py`.\n- Running the repository translation CLI on CPU or CUDA and interpreting its\n  output file.\n- Calling `Translator.translate_sentence(...)` directly for one already-tokenized\n  source sentence.\n- Debugging beam-size, max-length, token-index, batch-size-one, and device\n  mismatches in translation.\n\n## Do not route here for\n\n- Creating preprocessing pickles or BPE files; use the data-preparation skill.\n- Producing checkpoints, choosing training flags, or optimizer settings; use the\n  training skill.\n- Explaining encoder\u002Fdecoder internals, attention masks, or parameter shapes in\n  depth; use the model-architecture skill.\n\n## Start with these references\n\n1. Read [references\u002Fcli-reference.md](references\u002Fcli-reference.md) for the\n   `translate.py` CLI, checkpoint\u002Fdata-pickle schemas, device selection, and\n   output semantics.\n2. Read [references\u002Fapi-reference.md](references\u002Fapi-reference.md) for\n   programmatic `Transformer` construction from a checkpoint and\n   `Translator.translate_sentence` beam search usage.\n3. Read [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md) when a\n   checkpoint, pickle, vocabulary, CUDA, OOV, or BPE issue appears.\n\n## Bundled helpers\n\n- `scripts\u002Finspect_checkpoint.py` safely gates unpickling behind an explicit\n  trust flag, then validates checkpoint and optional data-pickle schemas.\n- `scripts\u002Ftranslation_smoke_check.py` builds a tiny deterministic Transformer\n  and exercises `Translator` beam search without requiring a trained checkpoint\n  or dataset.\n\nTypical preflight sequence:\n\n```bash\npython scripts\u002Finspect_checkpoint.py \\\n  --checkpoint trained.chkpt --data-pkl m30k_deen_shr.pkl --trust-inputs\npython scripts\u002Ftranslation_smoke_check.py \\\n  --repo-root \u002Fpath\u002Fto\u002Frepo --device cpu\n```\n\n## Operating reminders\n\n- `translate.py` loads the checkpoint with `torch.load(..., map_location=device)`\n  and then reconstructs `Transformer` from `checkpoint['settings']`; incompatible\n  settings are found only at load time unless you preflight them.\n- The CLI sets `opt.cuda = not opt.no_cuda`; on a CPU-only host include\n  `-no_cuda` or model\u002Fdevice transfer will fail.\n- `Translator.translate_sentence` asserts batch size one. For multiple examples,\n  iterate one sentence at a time or implement a separate batched decoder.\n- Output lines are target-vocabulary token strings with `\u003Cs>` and `\u003C\u002Fs>` removed\n  by simple string replacement. BPE post-decoding is marked TODO\u002Fnot ready in the\n  repository flow.\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Fattention-is-all-you-need-pytorch\u002Fsub-skills\u002Ftranslation",27,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab","2026-09-03 10:25:21",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":25},[35,38,41],{"tool":36,"command":37},"npx","npx skills add VectorSpaceLab\u002FAREX-Skill --skill translation",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill translation",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill translation","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fattention-is-all-you-need-pytorch\u002Fsub-skills\u002Ftranslation\u002FSKILL.md",[46,52,56,60,66],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fapi-reference.md","REFERENCE","参考文档",4638,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fattention-is-all-you-need-pytorch\u002Fsub-skills\u002Ftranslation\u002Freferences\u002Fapi-reference.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Fcli-reference.md",5561,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fattention-is-all-you-need-pytorch\u002Fsub-skills\u002Ftranslation\u002Freferences\u002Fcli-reference.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Ftroubleshooting.md",4070,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fattention-is-all-you-need-pytorch\u002Fsub-skills\u002Ftranslation\u002Freferences\u002Ftroubleshooting.md",{"path":61,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":64,"sourceHtmlUrl":65},"scripts\u002Finspect_checkpoint.py","SCRIPT","脚本",16254,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fattention-is-all-you-need-pytorch\u002Fsub-skills\u002Ftranslation\u002Fscripts\u002Finspect_checkpoint.py",{"path":67,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":68,"sourceHtmlUrl":69},"scripts\u002Ftranslation_smoke_check.py",6676,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fattention-is-all-you-need-pytorch\u002Fsub-skills\u002Ftranslation\u002Fscripts\u002Ftranslation_smoke_check.py",[71,79,86,93,100,107],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":18},"01M799NYSH1R3RVAHSM2D88Y84","passnet-triton-opt","Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block\u002Fgrid\u002Fwarp tuning, autotune policy, launch overhead, and per-op numeric recipes to pass the dtype baseline tolerances. Use when a pass matches and is correct but speedup below expected, or when correctness fails by small numeric margins.","Apache-2.0","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-triton-opt\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":18},"01M799NS5APEC0N1FZ7XZVW6BN","passnet-solve","END-TO-END playbook for solving one PassNet sample: analyze the computation graphs, decide the optimization strategy (what to fuse, what to replace, what to leave alone), drive the iteration loop, and maximize the sample score. This is the ENTRY skill — invoke it first for any PassNet optimization task; it tells you when to use passnet-pattern-fusion, passnet-triton-opt and passnet-feedback.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-solve\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":18},"01M799NNY9BKSM21V5R7QVTBJX","passnet-skill","PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under .\u002Fpass_dir\u002F, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-skill\u002FSKILL.md",[],[],{"skillNo":94,"name":95,"nameZh":6,"description":96,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":99,"publishedAt":18},"01M799NHH6NP7Z1RK77KGXAN6S","passnet-pattern-fusion","Author PassNet pass files: write patterns that actually MATCH the FX graph, pick fusion regions (which ops to absorb into one Triton kernel), and structure multi-pass submissions with the mandatory shared-dispatch architecture. Use when creating or fixing pattern()\u002Freplacement_args()\u002Freplacement_func() files, when a pass \"failed to match\", or when deciding how to fuse multiple kernels\u002Fops.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-pattern-fusion\u002FSKILL.md",[],[],{"skillNo":101,"name":102,"nameZh":6,"description":103,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":104,"categories":105,"keywords":106,"publishedAt":18},"01M799ND2J4WNC2PGAZ0A06NDF","passnet-orchestrate","Round\u002Fbatch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep\u002Frevert decisions. For solving one concrete sample end to end, use passnet-solve as the entry skill; consult this skill only when you need broader planning or a specific decision gate.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-orchestrate\u002FSKILL.md",[],[],{"skillNo":108,"name":109,"nameZh":6,"description":110,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":111,"categories":112,"keywords":113,"publishedAt":18},"01M799N8SBTMY02ZDG2AZF5ZWE","passnet-feedback","Fast, deterministic feedback for PassNet work: pre-flight pattern\u002Fmatch verification WITHOUT burning a GPU evaluation, per-node bottleneck analysis of a sample's graphs, and eval-log parsing into per-variant status + estimated score + failure classification. Use BEFORE every GPU evaluation (check_pattern), at the START of a sample (analyze_graph), and AFTER every evaluation (parse_eval_log).","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-feedback\u002FSKILL.md",[],[]]