metadata
language: es
license: cc-by-sa-4.0
datasets:
- wikipedia
- cc100
widget:
- text: Yo soy
- text: Me llamo
GPT2 Spanish base model (Uncased)
Model architecture
This model uses GPT2 base setttings except vocabulary size.
Tokenizer
Using BPE tokenizer with vocabulary size 50,000.
Training Data
- wiki40b/es (Spanish Wikipedia)
- Subset of CC-100/es : Monolingual Datasets from Web Crawl Data
Usage
from transformers import pipeline
generator = pipeline('text-generation', model='ClassCat/gpt2-base-spanish')
generator("Yo soy ", max_length=50, num_return_sequences=5)