skills.rest
scooter-lacroix
scooter-lacroixCOMMUNITY
Explore all skills by this author →
Website𝕏@scooter_lacroixGitHub

llm-tuning-patterns

Configure LLM parameters like max_tokens and temperature for specific tasks.

8|1|Updated Jan 4, 2026
Search Tags:#prompt engineering#code generation#llm#parameters#theorem proving#tuning
One-click install
npx skills add https://github.com/scooter-lacroix/Maestro --skill llm-tuning-patterns-scooter-lacroix
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: llm-tuning-patterns
Source: https://github.com/scooter-lacroix/Maestro/tree/main/maestro/skills/llm-tuning-patterns
Command: npx skills add https://github.com/scooter-lacroix/Maestro --skill llm-tuning-patterns-scooter-lacroix

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill addresses the challenge of selecting optimal Large Language Model (LLM) parameters for different types of tasks, ensuring better performance and more predictable outcomes.

Core Features & Use Cases

  • Task-Specific Configurations: Provides evidence-based LLM parameter settings for theorem proving, code generation, and creative tasks.
  • Pattern Guidance: Offers clear recommendations for max_tokens and temperature based on research.
  • Use Case: When tasked with generating formal proofs in Lean 4, use the theorem proving parameters to maximize the LLM's effectiveness.

Quick Start

Apply the recommended LLM parameters for theorem proving tasks.

Frequently Asked Questions about llm-tuning-patterns

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
What are the best LLM parameters for theorem proving in Lean 4?▼

Optimal LLM parameters for theorem proving involve specific max_tokens and temperature settings that maximize the model's effectiveness in generating formal proofs. This Skill provides evidence-based configurations tailored for formal verification tasks.

How do I configure temperature and max_tokens for code generation tasks?▼

Configuring LLM parameters for code generation requires balancing temperature and max_tokens to ensure predictable syntax and logical correctness. This Skill offers clear, research-based recommendations for these exact settings.

Why does my LLM output degrade when I use default parameters for specialized tasks?▼

Using default LLM parameters for specialized tasks often leads to suboptimal performance because settings are not tailored to task complexity. This Skill identifies anti-patterns to avoid and provides task-specific parameter tuning guidance.

What is the difference between LLM parameter tuning for creative exploration versus formal proofs?▼

LLM parameter tuning differs significantly between creative exploration and formal proofs; creative tasks benefit from higher temperature for novelty, while theorem proving requires conservative settings for logical precision. This Skill outlines these distinct configurations.

When do I need to adjust LLM parameters instead of using prompt engineering?▼

Adjusting LLM parameters is necessary when prompt engineering alone cannot control output variability, token limits, or determinism required for specific tasks. This Skill provides parameter patterns to complement prompt design.

Related Skills

Explore More Software Engineering Skills→
yonatangrossyonatangross
217

llm-integration

Integrate LLMs with function calling, SSE streaming, Ollama, and LoRA fine-tuning.

Community
Advanced
xaionaroxaionaro

proof-driven-development

Prove algorithm correctness in Lean 4 before implementing in the target language.

Community
Advanced
NagyViktNagyVikt
2

prompt-optimizer

Compress natural language prompts into structured, token-efficient instructions.

Community
Intermediate
xchang1121xchang1121
6

hint-mode

Extract parameter range constraints from task descriptions into SPACE_CONFIG for tuning.

Community
Advanced