Instructions to use allenai/PRIMERA-multixscience with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use allenai/PRIMERA-multixscience with Transformers:
# Load model directly from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("allenai/PRIMERA-multixscience") model = AutoModelForSeq2SeqLM.from_pretrained("allenai/PRIMERA-multixscience") - Notebooks
- Google Colab
- Kaggle
Add trigram block
#3
by johngiorgi - opened
After spending a ton of time trying to reproduce PRIMERA results on Multi-X-Science in HuggingFace, I've found that I get the closest when a trigram block is applied, so I am adding "no_repeat_ngram_size": 3 to the config.
johngiorgi changed pull request status to merged