[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M797VM4F7N5JN0Y5MXGSFSWD":3,"skill-same:01M797VM4F7N5JN0Y5MXGSFSWD":54},{"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":23,"commitSha":24,"skillDirectory":25,"forksCount":26,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":29,"install":30,"resources":43},"01M797VM4F7N5JN0Y5MXGSFSWD","rail_representation_loss",null,"Compute RAIL-KD layerwise and concatenated intermediate representation losses from pooled hidden states.","VectorSpaceLab\u002FAREX-Skill","VectorSpaceLab","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F180637102?v=4","01M7538C5WPSJGWF7KHR4EFZ9F","01M7538C6R2GFP53NPXTCZNBRT",351,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPaperBench\u002Fskills\u002Fadaptive-pruning\u002Fskill\u002Frail_kd_layer_mapping\u002Frail_representation_loss\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: rail_representation_loss\ndescription: Compute RAIL-KD layerwise and concatenated intermediate representation losses from pooled hidden states.\n---\n\n# RAIL Representation Loss\n\nUse this skill when a distillation experiment must implement the RAIL-KD intermediate representation objective. It is appropriate after a teacher-layer mapping has selected exactly one teacher layer for each student layer. Do not use it for attention distribution matching, MiniLM relation losses, or layer-search logic.\n\n## Inputs\n\n- Teacher hidden states: nested numeric lists shaped `[m][tokens][teacher_dim]`.\n- Student hidden states: nested numeric lists shaped `[m][tokens][student_dim]`.\n- Projection matrices for teacher and student vectors.\n- Variant: `layerwise` for RAIL-KDl or `concatenated` for RAIL-KDc.\n- Optional layer weights for layerwise mode.\n\n## Outputs\n\n- Scalar intermediate loss.\n- Diagnostics containing pooled vectors, normalized vectors, and per-layer distances when applicable.\n\n## Workflow\n\n1. Mean-pool each selected hidden-state layer across tokens.\n2. Project teacher and student vectors to the same dimension.\n3. L2-normalize projected vectors with a small epsilon guard.\n4. For layerwise mode, compute weighted squared distances and sum them.\n5. For concatenated mode, concatenate pooled vectors first, project once, normalize, then compute one squared distance.\n6. Return diagnostics for recovery mechanism checks.\n\n## Validation\n\nRun:\n\n```bash\npython scripts\u002Frail_loss.py --demo\npython -m pytest tests\n```\n\n## Limitations\n\nThis skill uses small pure-Python numeric lists for portability. Production transformer code should pass actual tensors through equivalent pooling, projection, normalization, and distance operations while preserving the same contracts.\n","# RAIL Representation Loss\n\nUse this skill when a distillation experiment must implement the RAIL-KD intermediate representation objective. It is appropriate after a teacher-layer mapping has selected exactly one teacher layer for each student layer. Do not use it for attention distribution matching, MiniLM relation losses, or layer-search logic.\n\n## Inputs\n\n- Teacher hidden states: nested numeric lists shaped `[m][tokens][teacher_dim]`.\n- Student hidden states: nested numeric lists shaped `[m][tokens][student_dim]`.\n- Projection matrices for teacher and student vectors.\n- Variant: `layerwise` for RAIL-KDl or `concatenated` for RAIL-KDc.\n- Optional layer weights for layerwise mode.\n\n## Outputs\n\n- Scalar intermediate loss.\n- Diagnostics containing pooled vectors, normalized vectors, and per-layer distances when applicable.\n\n## Workflow\n\n1. Mean-pool each selected hidden-state layer across tokens.\n2. Project teacher and student vectors to the same dimension.\n3. L2-normalize projected vectors with a small epsilon guard.\n4. For layerwise mode, compute weighted squared distances and sum them.\n5. For concatenated mode, concatenate pooled vectors first, project once, normalize, then compute one squared distance.\n6. Return diagnostics for recovery mechanism checks.\n\n## Validation\n\nRun:\n\n```bash\npython scripts\u002Frail_loss.py --demo\npython -m pytest tests\n```\n\n## Limitations\n\nThis skill uses small pure-Python numeric lists for portability. Production transformer code should pass actual tensors through equivalent pooling, projection, normalization, and distance operations while preserving the same contracts.\n",{"name":5,"description":7},"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Ftask-oriented\u002FPaperBench\u002Fskills\u002Fadaptive-pruning\u002Fskill\u002Frail_kd_layer_mapping\u002Frail_representation_loss",27,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab","2026-09-03 10:25:21",{"source":8,"skillName":5,"commands":31,"prompt":41,"zipAvailable":42},[32,35,38],{"tool":33,"command":34},"npx","npx skills add VectorSpaceLab\u002FAREX-Skill --skill rail_representation_loss",{"tool":36,"command":37},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill rail_representation_loss",{"tool":39,"command":40},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill rail_representation_loss","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPaperBench\u002Fskills\u002Fadaptive-pruning\u002Fskill\u002Frail_kd_layer_mapping\u002Frail_representation_loss\u002FSKILL.md",true,[44,50],{"path":45,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":48,"sourceHtmlUrl":49},"scripts\u002Frail_loss.py","SCRIPT","脚本",4059,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPaperBench\u002Fskills\u002Fadaptive-pruning\u002Fskill\u002Frail_kd_layer_mapping\u002Frail_representation_loss\u002Fscripts\u002Frail_loss.py",{"path":51,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":52,"sourceHtmlUrl":53},"tests\u002Ftest_rail_loss.py",1209,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPaperBench\u002Fskills\u002Fadaptive-pruning\u002Fskill\u002Frail_kd_layer_mapping\u002Frail_representation_loss\u002Ftests\u002Ftest_rail_loss.py",[55,62,69,76,83,90],{"skillNo":56,"name":57,"nameZh":6,"description":58,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":59,"categories":60,"keywords":61,"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.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-triton-opt\u002FSKILL.md",[],[],{"skillNo":63,"name":64,"nameZh":6,"description":65,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":66,"categories":67,"keywords":68,"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":70,"name":71,"nameZh":6,"description":72,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"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":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"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":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"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":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"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",[],[]]