peft-fine-tuning

21.8k

davila7

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train less than 1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.

prioritization-effort-impact

119

nicepkg

Use when ranking backlogs, deciding what to do first based on effort vs impact (quick wins vs big bets), prioritizing feature roadmaps, triaging bugs or technical debt, allocating resources across initiatives, identifying low-hanging fruit, evaluating strategic options with 2x2 matrix, or when user mentions prioritization, quick wins, effort-impact matrix, high-impact low-effort, big bets, or asks "what should we do first?".

lora

14

itsmostafa

Parameter-efficient fine-tuning with Low-Rank Adaptation (LoRA). Use when fine-tuning large language models with limited GPU memory, creating task-specific adapters, or when you need to train multiple specialized models from a single base.

Test Design

10

shabaraba

Automatically design comprehensive end-to-end (E2E) test cases for newly implemented vibing.nvim features. Use immediately after completing feature implementation (Phase 5.4) and before running E2E tests. Generates test scenarios covering happy paths, error cases, edge cases, and integration points, with a prioritized risk ranking (Critical / High / Medium / Low) and ready-to-use test code templates following e2e_helper.lua conventions.

compare-attempts

10

brazil-bench

This SOP compares evaluated brazil-bench attempts across multiple dimensions to produce a ranked leaderboard and detailed comparison summary. It supports up to 10 attempts in the "Top 10" format, automatically pruning lower-ranked entries when more are added.