8.9K skills · sorted by popularity
Use SAM to separate objects from images by marking points or drawing boxes, without training a model for each task. Covers automatic region generation, refining selections, preparing annotation data, and exporting models for deployment.
Build question answering and document retrieval workflows with DSPy by defining inputs, outputs, and reusable components. Use training examples and evaluation metrics to automatically optimize prompts and compare results before and after optimization.
Use the hf command-line tool to find, download, and upload models and datasets on Hugging Face Hub. The guide also covers repository management, hosted computing resources, local cache maintenance, and automation options.
Helps you use Unsloth for language model fine-tuning with LoRA and QLoRA, methods that adapt models with fewer trainable parameters. Supports API questions, implementation, and debugging using reference material drawn from official documentation.
Find language models on Hugging Face Hub and run them locally with llama.cpp. The skill helps compare GGUF model files, choose compressed versions that fit available memory, and prepare commands for a local chat server or command-line session.
Set up vLLM to serve large language models through an OpenAI-compatible API or process prompts in batches. The skill covers deployment, performance monitoring, and quantization, which reduces model memory requirements.
Generate model responses in defined formats for data extraction and classification. Use JSON schemas, Pydantic data models, regular expressions, or fixed choices to produce structured results that other programs can parse and validate.
Get guidance for fine-tuning large language models with Axolotl, adapting them through additional training. Use YAML configuration examples, dataset references, and training methods such as LoRA to set up training and investigate problems.
Guide language model training with TRL using example answers, preferred response pairs, or reward scores. The skill covers data preparation, method selection, reward model training, evaluation, and adjustments for memory limits or unstable training.
Build agent memory with Pinecone by storing searchable representations of conversations and documents. Retrieve relevant past context for answers, organize memory by session or user, and prototype semantic search that finds information by meaning.
Evaluate large language models on shared tests of knowledge, reasoning, and code generation with lm-evaluation-harness. Save scores and individual predictions, compare models in tables, and track performance across saved stages of model training.
Improve prompts, regex patterns, SQL queries, or small code snippets through repeated scoring and revision. The skill helps define measurable tests, generate new candidates with an AI model, and inspect results to find better-performing versions.
Set up large language model pretraining with TorchTitan and PyTorch. Configure training across multiple GPUs, combine four ways of splitting the workload, enable Float8 calculations on H100 hardware, and manage checkpoints for saving and resuming training.
Set up and optimize large language model inference, or response generation, on NVIDIA GPUs. Use quantization, batch processing, and multiple GPUs to improve serving performance, and expose models through a chat completion API.
Track machine learning experiments in W&B, record training settings and metrics, and compare results in dashboards. Automate searches for training settings and keep version histories for models and datasets.
Extract website content using direct HTTP requests, JavaScript-capable browsers, or stealth browsing with Cloudflare handling. Use selectors to collect specific data, crawl linked pages, and export results as Markdown, text, HTML, or JSON.
Explore concepts learned by language models using sparse autoencoders, which separate internal activity into a small set of features. Analyze pretrained models, train custom autoencoders, assess reconstruction quality, and experiment with feature based output adjustments.
Fine-tune large language models with LoRA and QLoRA, methods that train small additions to a base model while limiting GPU memory use. Configure training, save task-specific adapters, switch between them, and merge them for deployment.
Helps organize PyTorch model training with LightningModule and Trainer, including validation, logging, model saving, and early stopping. It also guides distributed training across multiple GPUs and settings for reduced-precision computation.
Set up reinforcement learning, training guided by response rewards, for large language models. Use Megatron-LM and SGLang to prepare data, configure training and response generation, and build asynchronous or multi-turn experiments.
Implement and debug large-model training with PyTorch FSDP, which splits training state across devices. Consult documentation and code patterns for sharding choices, mixed precision, and checkpoints when building or reviewing distributed training code.
Use LLaVA, a model that understands images and text, to answer questions about pictures, write image descriptions, and build image chat applications. Includes guidance for setup, ongoing conversations, and model memory options.
Build document search with Chroma, a database that stores text as numerical representations of meaning. Save and update documents, filter by metadata, and connect retrieval to AI applications using LangChain or LlamaIndex.
Integrate Flash Attention into Transformer models to speed up attention, the calculation that relates input elements, while reducing GPU memory use. Covers PyTorch integration, performance measurement, output comparison, and a separate H100 optimization path.