---
license: apache-2.0
language:
  - en
base_model:
  - OpenSciLM/OpenScholar_Retriever
---

neophyte-retriever is a continued pre-trained version of [OpenSciLM/OpenScholar_Retriever](https://huggingface.co/OpenSciLM/OpenScholar_Retriever) for scientific literature synthesis.

### Model Description

- **Developed by:** Bio Protocol
- **Model type:** a masked language model.
- **Language(s) (NLP):** English
- **Date cutoff:** The pre-training data is mixture of [neophyte-faiss-index-v1](https://huggingface.co/datasets/bio-protocol/neophyte-faiss-index-v1), [peS2o](https://huggingface.co/datasets/allenai/peS2o), [CCNews](https://huggingface.co/datasets/vblagoje/cc_news) and [Proofpile2](https://huggingface.co/datasets/EleutherAI/proof-pile-2).