Fine-tune language models with validated data.
What it does
Creates data-preparation, training, and evaluation code for adapting language models to specific tasks.
When to use it
Use when fine-tuning language models, training custom models, or adapting foundation models for specific tasks.
How to use it
Give it training data and a fine-tuning request; it produces dataset-preparation and evaluation scripts, training configuration, and a design rationale.
What you provide
- Your files
Uses
LL
Latest review
What you need · 7
Uses Python to run the dataset-validation checkpoint.
Selects the training method according to GPU memory, using QLoRA when memory is constrained.
The main LoRA training example imports datasets to load training data.
Access · 1
This skill
meta-llama/Llama-3-8B
Read
Loads the base Llama model