---
license: apache-2.0
---



# Rationalyst (with rationales extracted from reasoning datasets)

This model is a distilled version of the [LLaMa-3-Instruct-8B](https://huggingface.co/bert-base-uncased). It was
introduced in RATIONALYST: Supervising Reasoning via Self-Supervised Rationale Extraction. The code for the rationale extraction, model training and 
inference can be found [here](https://github.com/JHU-CLSP/reasoning_world_model).