Skip to content

Dataset · bertin-project

Spanish translation of Stanford Alpaca instructions

bertin-project/alpaca-spanish

Between 10 and 100 thousand Spanish instruction-response pairs translated using GPT-3.5-turbo for language model fine-tuning.

What's inside

Contains Spanish texts representing a translation of the cleaned version of the Stanford Alpaca dataset (alpaca_data_cleaned.json). The corpus contains between 10 and 100 thousand entries translated using OpenAI gpt-3.5-turbo (an early version was translated via NLLB 1.3B). Due to the use of the OpenAI model, the license prohibits using this data to create models that compete with OpenAI.

What you can do with it

  • SFT fine-tuning of LLMs and LoRA adapters for Spanish instruction understanding
  • Testing text generation and command execution quality in Spanish
Tasks
Text generation
Languages
Spanish
Size
10K–100K rows
Downloads this month
1K
Likes
38

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("bertin-project/alpaca-spanish")