MLOps
模型部署、评估与运维
nnsight-remote-interpretability
Orchestra-Research
Provides guidance for interpreting and manipulating neural network internals using nnsight with optional NDIF remote execution. Use when needing to run interpretability experiments on massive models (70B+) without local GPU resources, or when working with any PyTorch architecture.
sparse-autoencoder-training
Orchestra-Research
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.
lambda-labs-gpu-cloud
Orchestra-Research
Reserved and on-demand GPU cloud instances for ML training and inference. Use when you need dedicated GPU instances with simple SSH access, persistent filesystems, or high-performance multi-node clusters for large-scale training.
huggingface-tokenizers
Orchestra-Research
Fast tokenizers optimized for research and production. Rust-based implementation tokenizes 1GB in <20 seconds. Supports BPE, WordPiece, and Unigram algorithms. Train custom vocabularies, track alignments, handle padding/truncation. Integrates seamlessly with transformers. Use when you need high-performance tokenization or custom tokenizer training.
llama-cpp
Orchestra-Research
Runs LLM inference on CPU, Apple Silicon, and consumer GPUs without NVIDIA hardware. Use for edge deployment, M1/M2/M3 Macs, AMD/Intel GPUs, or when CUDA is unavailable. Supports GGUF quantization (1.5-8 bit) for reduced memory and 4-10× speedup vs PyTorch on CPU.
nanogpt
Orchestra-Research
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).
evaluating-code-models
Orchestra-Research
Evaluates code generation models across HumanEval, MBPP, MultiPL-E, and 15+ benchmarks with pass@k metrics. Use when benchmarking code models, comparing coding abilities, testing multi-language support, or measuring code generation quality. Industry standard from BigCode Project used by HuggingFace leaderboards.
pyvene-interventions
Orchestra-Research
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.
evaluating-llms-harness
Orchestra-Research
Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.
unsloth
Orchestra-Research
Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization
gguf-quantization
Orchestra-Research
GGUF format and llama.cpp quantization for efficient CPU/GPU inference. Use when deploying models on consumer hardware, Apple Silicon, or when needing flexible quantization from 2-8 bit without GPU requirements.
transformer-lens-interpretability
Orchestra-Research
Provides guidance for mechanistic interpretability research using TransformerLens to inspect and manipulate transformer internals via HookPoints and activation caching. Use when reverse-engineering model algorithms, studying attention patterns, or performing activation patching experiments.
grpo-rl-training
Orchestra-Research
Expert guidance for GRPO/RL fine-tuning with TRL for reasoning and task-specific model training
sentencepiece
Orchestra-Research
Language-independent tokenizer treating text as raw Unicode. Supports BPE and Unigram algorithms. Fast (50k sentences/sec), lightweight (6MB memory), deterministic vocabulary. Used by T5, ALBERT, XLNet, mBART. Train on raw text without pre-tokenization. Use when you need multilingual support, CJK languages, or reproducible tokenization.
rwkv-architecture
Orchestra-Research
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.
distributed-llm-pretraining-torchtitan
Orchestra-Research
Provides PyTorch-native distributed LLM pretraining using torchtitan with 4D parallelism (FSDP2, TP, PP, CP). Use when pretraining Llama 3.1, DeepSeek V3, or custom models at scale from 8 to 512+ GPUs with Float8, torch.compile, and distributed checkpointing.
async-io-model
tursodatabase
Explanations of common asynchronous patterns used in tursodb. Involves IOResult, state machines, re-entrancy pitfalls, CompletionGroup. Always use these patterns in core when doing anything IO
Build Your Pipecat Skill
majiayu000
"Create your Pipecat skill from official documentation, then learn to improve it throughout the chapter"
Build Your Pipecat Skill
majiayu000
"Create your Pipecat skill from official documentation, then learn to improve it throughout the chapter"
model-selection
majiayu000
Automatically applies when choosing LLM models and providers. Ensures proper model comparison, provider selection, cost optimization, fallback patterns, and multi-model strategies.
flow-nexus-neural
majiayu000
Train and deploy neural networks in distributed E2B sandboxes with Flow Nexus
model-selection
majiayu000
Automatically applies when choosing LLM models and providers. Ensures proper model comparison, provider selection, cost optimization, fallback patterns, and multi-model strategies.
higress-auto-router
alibaba
"Configure automatic model routing using the get-ai-gateway.sh CLI tool for Higress AI Gateway. Use when: (1) User wants to configure automatic model routing, (2) User mentions 'route to', 'switch model', 'use model when', 'auto routing', (3) User describes scenarios that should trigger specific models, (4) User wants to add, list, or remove routing rules."
swiftui-view-refactor
steipete
Refactor and review SwiftUI view files for consistent structure, dependency injection, and Observation usage. Use when asked to clean up a SwiftUI view’s layout/ordering, handle view models safely (non-optional when possible), or standardize how dependencies and @Observable state are initialized and passed.