All skills
fine-tuning-expert
Write LoRA/QLoRA fine-tuning code, dataset validation scripts, and evaluation scripts for adapting LLMs.
- Part of Jeffallan/claude-skills
- 4.6k installs
- Machine Learning · Fine-tuning and quantization
When to use it
You have a base model and a task in mind but still need training data formatted, adapter settings chosen, and a way to check the result against the base model.
For example
You say
Use fine-tuning-expert to fine-tune Llama-3-8B on my customer-support JSONL so it answers refund questions in our tone, using LoRA on a single 24GB GPU.
You get
- Dataset validation script
- LoRA training config
- Evaluation script
How it works
- Chooses LoRA, QLoRA, or full fine-tuning based on GPU memory and model size.
- Covers the full path from data checks through merging adapters and quantizing for deployment.
More Fine-tuning and quantization skills
See all- unsloth25Fine-tunes models 2-5x faster with LoRA or QLoRA while using less VRAM.
- axolotl21Fine-tunes language models with Axolotl using YAML configs for LoRA, DPO and GRPO.
- codonfm-finetune15Fine-tunes public CodonFM Encodon checkpoints on labeled coding-sequence data for regression or classification.
- kermt-pretrain-scratch12Pretrains a brand-new KERMT model from scratch on your molecule corpus.
Also in this pack
- golang-pro19.8kWrites concurrent Go services with goroutines, channels, gRPC or REST, and profiles performance.
- php-pro16.4kBuilds modern PHP applications with strict typing, Laravel or Symfony, tests and REST or GraphQL APIs.
- typescript-pro15kBuilds advanced TypeScript types and end-to-end type-safe APIs with tRPC.
- flutter-expert14.9kBuilds cross-platform Flutter apps with widgets, Riverpod or Bloc state, navigation and performance tuning.