ML Ops
Machine learning operations
gptq
by majiayu000
Post-training 4-bit quantization for LLMs with minimal accuracy loss. Use for deploying large models (70B, 405B) on consumer GPUs, when you need 4× memory reduction with <2% perplexity degradation, or for faster inference (3-4× speedup) vs FP16. Integrates with transformers and PEFT for QLoRA fine-tuning.
verl-rl-training
by majiayu000
Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL). Use when implementing RLHF, GRPO, PPO, or other RL algorithms for LLM post-training at scale with flexible infrastructure backends.
sparse-autoencoder-training
by majiayu000
Provides guidance for training and analyzing Sparse Autoencoders (SAEs) using SAELens to decompose neural network activations into interpretable features. Use when discovering interpretable features, analyzing superposition, or studying monosemantic representations in language models.
rwkv-architecture
by majiayu000
RNN+Transformer hybrid with O(n) inference. Linear time, infinite context, no KV cache. Train like GPT (parallel), infer like RNN (sequential). Linux Foundation AI project. Production at Windows, Office, NeMo. RWKV-7 (March 2025). Models up to 14B parameters.
nanogpt
by majiayu000
Educational GPT implementation in ~300 lines. Reproduces GPT-2 (124M) on OpenWebText. Clean, hackable code for learning transformers. By Andrej Karpathy. Perfect for understanding GPT architecture from scratch. Train on Shakespeare (CPU) or OpenWebText (multi-GPU).
pyvene-interventions
by majiayu000
Provides guidance for performing causal interventions on PyTorch models using pyvene's declarative intervention framework. Use when conducting causal tracing, activation patching, interchange intervention training, or testing causal hypotheses about model behavior.
evaluator
by SimHacker
Independent assessment without debate context — adversarial loop prevents gaming
3d-web-experience
by benjaminasterA
"Expert in building 3D experiences for the web - Three.js, React Three Fiber, Spline, WebGL, and interactive 3D scenes. Covers product configurators, 3D portfolios, immersive websites, and bringing ..."
wait
by elie222
Pause execution for a user-specified duration
step-by-step
by elie222
Execute tasks one step at a time with user confirmation
thinking-model-router
by tjboudreaux
When unsure which thinking skill fits, map domain and problem type, then return NONE or one primary skill by default (at most three complementary).
thinking-model-combination
by tjboudreaux
When one mental model leaves a material blind spot on a multi-domain or high-stakes problem, sequence complementary models with named roles and a conflict rule.
adding-models
by letta-ai
Guide for adding new LLM models to Letta Code. Use when the user wants to add support for a new model, needs to know valid model handles, or wants to update the model configuration. Covers models.json configuration, CI test matrix, and handle validation.
azure-openai-2025
by JosiahSiegel
Azure OpenAI Service 2025 models including GPT-5, GPT-4.1, reasoning models, and Azure AI Foundry integration
dbs-diagnosis
by Pluviobyte
dontbesilent 商业模式诊断。两种模式：问诊（消解你的问题）和体检（拆解你的商业模式）。 触发方式：/dbs-diagnosis、/问诊、「帮我看看商业模式」「诊断一下我的业务」「我有个商业问题」 Business model diagnosis using dontbesilent's ontological framework. Two modes: consultation (dissolve your question) and checkup (analyze your business model). Trigger: /dbs-diagnosis, "diagnose my business model", "I have a business question"
security-awareness-training
by Ed1s0nZ
安全意识培训的专业技能和方法论
runwayml
by runwayml
Generate AI videos, images, and audio with Runway API. Use when generating video from images, text-to-video, video-to-video, character performance, text-to-image, text-to-speech, sound effects, or voice processing with Runway.
hermes-atropos-environments
by NousResearch
Build, test, and debug Hermes Agent RL environments for Atropos training. Covers the HermesAgentBaseEnv interface, reward functions, agent loop integration, evaluation with tools, wandb logging, and the three CLI modes (serve/process/evaluate). Use when creating, reviewing, or fixing RL environments in the hermes-agent repo.
result-integrator
by GongLingRui
整合多情点分析结果为综合报告，通过去重、分类、排序、总结生成高质量分析。适用于整合多分析源、生成统一报告
character-relationships
by GongLingRui
分析故事人物关系，识别类型、特点及发展变化。适用于深度理解人物关系网络、分析关系对情节推动作用、为情节设计提供关系支撑
web-search
by GongLingRui
使用联网搜索获取最新市场信息和行业动态。适用于获取竖屏短剧市场趋势、热门题材分析、成功案例研究
knowledge-query
by GongLingRui
专注于竖屏短剧知识库查询和信息检索，提供专业知识服务。适用于查询剧本桥段、高能情节、创作技巧、商业运营等专业知识
series-analyzer
by GongLingRui
分析已播剧集，包括信息获取、拉片分析、五元素分析及联网搜索。适用于深度分析剧集创作技法、学习情节设计
story-type-analyzer
by GongLingRui
分析故事题材类型，提炼创意元素与故事特色。适用于识别题材定位、提炼创意、分析市场受众