[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78SWBJ6KHC0ACN3M2JZN11V":3,"skill-same:01M78SWBJ6KHC0ACN3M2JZN11V":66},{"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},"01M78SWBJ6KHC0ACN3M2JZN11V","model-architecture",null,"Instantiate and debug the attention-is-all-you-need-pytorch Transformer architecture, masks, attention layers, feed-forward blocks, weight sharing, tensor shapes, and scheduled optimizer.","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\u002Fmodel-architecture\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: model-architecture\ndescription: \"Instantiate and debug the attention-is-all-you-need-pytorch\n  Transformer architecture, masks, attention layers, feed-forward blocks, weight\n  sharing, tensor shapes, and scheduled optimizer.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# Model Architecture\n\nUse this sub-skill when you need to construct, inspect, or debug the repository's\nTransformer model classes rather than preprocess data, run training CLIs, or\nperform checkpoint translation.\n\n## Use This For\n\n- Building `Transformer`, `Encoder`, `Decoder`, `EncoderLayer`, `DecoderLayer`,\n  `MultiHeadAttention`, `PositionwiseFeedForward`, `ScaledDotProductAttention`,\n  `PositionalEncoding`, or `ScheduledOptim` instances.\n- Checking source\u002Ftarget mask shapes, boolean semantics, causal target masks,\n  device placement, attention tensor shapes, and flattened model logits.\n- Choosing safe tiny-model settings for unit tests, especially weight-sharing\n  flags and `d_model == d_word_vec`.\n- Debugging architecture assertions, projection\u002Fembedding scaling, incompatible\n  shared vocabularies, and Noam-style warmup scheduling.\n\n## Route Elsewhere\n\n- Dataset fields, vocab building, BPE, spaCy, or pickle schemas belong to the\n  data-preparation sub-skill.\n- Training command-line options, logging, checkpoints, label smoothing, and loss\n  handling belong to the training sub-skill.\n- Checkpoint loading, `Translator`, beam search, and decoding workflows belong\n  to the translation sub-skill; this sub-skill only covers class-level model\n  shapes that those workflows rely on.\n\n## Primary References\n\n- Start with [references\u002Fapi-reference.md](references\u002Fapi-reference.md) for\n  constructor signatures, forward contracts, return shapes, and optimizer API.\n- Use [references\u002Farchitecture-notes.md](references\u002Farchitecture-notes.md) for\n  end-to-end tensor flow, mask broadcasting, tiny-model recipes, and weight\n  sharing rules.\n- Use [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md) for common\n  assertion, mask, CUDA\u002Fdevice, flattened-output, sharing, and scheduler\n  failures.\n\n## Smoke Check\n\nRun the bundled architecture check from any current directory by pointing it at\nan explicit repository checkout:\n\n```bash\npython scripts\u002Farchitecture_smoke_check.py --repo-root \u002Fpath\u002Fto\u002Fattention-is-all-you-need-pytorch\n```\n\nUseful variants:\n\n```bash\npython scripts\u002Farchitecture_smoke_check.py --repo-root \u002Fpath\u002Fto\u002Frepo --device cpu\npython scripts\u002Farchitecture_smoke_check.py --repo-root \u002Fpath\u002Fto\u002Frepo --device cuda\npython scripts\u002Farchitecture_smoke_check.py --repo-root \u002Fpath\u002Fto\u002Frepo --skip-negative-checks\n```\n\nThe script constructs a tiny deterministic Transformer with sharing disabled,\nchecks pad and causal masks, verifies multi-head attention broadcasting, probes\n`ScheduledOptim` warmup\u002Fdecay behavior, and optionally confirms that known-bad\narchitecture settings fail for the expected reasons.\n\n## Fast Operating Pattern\n\n1. Keep `d_model` equal to `d_word_vec`; the model asserts this after parameter\n   initialization because residual connections require a common hidden width.\n2. If source and target vocabularies are not the same shared vocabulary with the\n   same token ids, set `emb_src_trg_weight_sharing=False`. For tiny tests with\n   different vocab sizes, also set `trg_emb_prj_weight_sharing=False` unless you\n   specifically need target embedding\u002Fprojection sharing.\n3. Use `scale_emb_or_prj` only as one of `emb`, `prj`, or `none`; scaling is\n   applied only when target embedding\u002Fprojection sharing is enabled.\n4. Treat masks as boolean keep-masks: `True` means a key position is visible and\n   `False` means it is masked before softmax.\n5. Remember that `Transformer.forward(src_seq, trg_seq)` returns flattened\n   logits with shape `[batch * target_length, n_trg_vocab]`, not\n   `[batch, target_length, n_trg_vocab]`.\n","# Model Architecture\n\nUse this sub-skill when you need to construct, inspect, or debug the repository's\nTransformer model classes rather than preprocess data, run training CLIs, or\nperform checkpoint translation.\n\n## Use This For\n\n- Building `Transformer`, `Encoder`, `Decoder`, `EncoderLayer`, `DecoderLayer`,\n  `MultiHeadAttention`, `PositionwiseFeedForward`, `ScaledDotProductAttention`,\n  `PositionalEncoding`, or `ScheduledOptim` instances.\n- Checking source\u002Ftarget mask shapes, boolean semantics, causal target masks,\n  device placement, attention tensor shapes, and flattened model logits.\n- Choosing safe tiny-model settings for unit tests, especially weight-sharing\n  flags and `d_model == d_word_vec`.\n- Debugging architecture assertions, projection\u002Fembedding scaling, incompatible\n  shared vocabularies, and Noam-style warmup scheduling.\n\n## Route Elsewhere\n\n- Dataset fields, vocab building, BPE, spaCy, or pickle schemas belong to the\n  data-preparation sub-skill.\n- Training command-line options, logging, checkpoints, label smoothing, and loss\n  handling belong to the training sub-skill.\n- Checkpoint loading, `Translator`, beam search, and decoding workflows belong\n  to the translation sub-skill; this sub-skill only covers class-level model\n  shapes that those workflows rely on.\n\n## Primary References\n\n- Start with [references\u002Fapi-reference.md](references\u002Fapi-reference.md) for\n  constructor signatures, forward contracts, return shapes, and optimizer API.\n- Use [references\u002Farchitecture-notes.md](references\u002Farchitecture-notes.md) for\n  end-to-end tensor flow, mask broadcasting, tiny-model recipes, and weight\n  sharing rules.\n- Use [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md) for common\n  assertion, mask, CUDA\u002Fdevice, flattened-output, sharing, and scheduler\n  failures.\n\n## Smoke Check\n\nRun the bundled architecture check from any current directory by pointing it at\nan explicit repository checkout:\n\n```bash\npython scripts\u002Farchitecture_smoke_check.py --repo-root \u002Fpath\u002Fto\u002Fattention-is-all-you-need-pytorch\n```\n\nUseful variants:\n\n```bash\npython scripts\u002Farchitecture_smoke_check.py --repo-root \u002Fpath\u002Fto\u002Frepo --device cpu\npython scripts\u002Farchitecture_smoke_check.py --repo-root \u002Fpath\u002Fto\u002Frepo --device cuda\npython scripts\u002Farchitecture_smoke_check.py --repo-root \u002Fpath\u002Fto\u002Frepo --skip-negative-checks\n```\n\nThe script constructs a tiny deterministic Transformer with sharing disabled,\nchecks pad and causal masks, verifies multi-head attention broadcasting, probes\n`ScheduledOptim` warmup\u002Fdecay behavior, and optionally confirms that known-bad\narchitecture settings fail for the expected reasons.\n\n## Fast Operating Pattern\n\n1. Keep `d_model` equal to `d_word_vec`; the model asserts this after parameter\n   initialization because residual connections require a common hidden width.\n2. If source and target vocabularies are not the same shared vocabulary with the\n   same token ids, set `emb_src_trg_weight_sharing=False`. For tiny tests with\n   different vocab sizes, also set `trg_emb_prj_weight_sharing=False` unless you\n   specifically need target embedding\u002Fprojection sharing.\n3. Use `scale_emb_or_prj` only as one of `emb`, `prj`, or `none`; scaling is\n   applied only when target embedding\u002Fprojection sharing is enabled.\n4. Treat masks as boolean keep-masks: `True` means a key position is visible and\n   `False` means it is masked before softmax.\n5. Remember that `Transformer.forward(src_seq, trg_seq)` returns flattened\n   logits with shape `[batch * target_length, n_trg_vocab]`, not\n   `[batch, target_length, n_trg_vocab]`.\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\u002Fmodel-architecture",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 model-architecture",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill model-architecture",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill model-architecture","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\u002Fmodel-architecture\u002FSKILL.md",[46,52,56,60],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fapi-reference.md","REFERENCE","参考文档",9028,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fattention-is-all-you-need-pytorch\u002Fsub-skills\u002Fmodel-architecture\u002Freferences\u002Fapi-reference.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Farchitecture-notes.md",8153,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fattention-is-all-you-need-pytorch\u002Fsub-skills\u002Fmodel-architecture\u002Freferences\u002Farchitecture-notes.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Ftroubleshooting.md",8872,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fattention-is-all-you-need-pytorch\u002Fsub-skills\u002Fmodel-architecture\u002Freferences\u002Ftroubleshooting.md",{"path":61,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":64,"sourceHtmlUrl":65},"scripts\u002Farchitecture_smoke_check.py","SCRIPT","脚本",11789,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fattention-is-all-you-need-pytorch\u002Fsub-skills\u002Fmodel-architecture\u002Fscripts\u002Farchitecture_smoke_check.py",[67,75,82,89,96,103],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"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":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"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":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"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":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"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":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"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":104,"name":105,"nameZh":6,"description":106,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":107,"categories":108,"keywords":109,"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",[],[]]