Modelpublic

ai4bharat/romansetu-cpt-native-200m

sourceHugging Facellama2updated 2y agoView on Hugging Face

0likes73downloads

Model Card

RomanSetu

This was trained as part of the paper RomanSetu: Efficiently unlocking multilingual capabilities of Large Language Models via Romanization. The codebase used to train and evaluate this model can be found at https://github.com/AI4Bharat/romansetu.

Usage

python3

from transformers import AutoTokenizer, AutoModelForCausalLM
model_path = "ai4bharat/romansetu-cpt-native-200m"
tokenizer = AutoTokenizer.from_pretrained(model_path)
model = AutoModelForCausalLM.from_pretrained(model_path)