VBART Model Card
Model Description
VBART is the first sequence-to-sequence LLM pre-trained on Turkish corpora from scratch on a large scale. It was pre-trained by VNGRS in February 2023.
The model is capable of conditional text generation tasks such as text summarization, paraphrasing, and title generation when fine-tuned.
It outperforms its multilingual counterparts, albeit being much smaller than other implementations.
This repository contains pre-trained TensorFlow and Safetensors weights of VBART-Medium-Base.
- Developed by: VNGRS-AI
- Model type: Transformer encoder-decoder based on mBART architecture
- Language(s) (NLP): Turkish
- License: CC BY-NC-SA 4.0
- Paper: arXiv
Training Details
Training Data
The base model is pre-trained on vngrs-web-corpus. It is curated by cleaning and filtering Turkish parts of OSCAR-2201 and mC4 datasets. These datasets consist of documents of unstructured web crawl data. More information about the dataset can be found on their respective pages. Data is filtered using a set of heuristics and certain rules, explained in the appendix of our paper.
Limitations
This model is the pre-trained base model and is capable of masked language modeling. Its purpose is to serve as the base model to be fine-tuned for downstream tasks.
Training Procedure
Pre-trained for a total of 63B tokens.
Hardware
- GPUs: 8 x Nvidia A100-80 GB
Software
- TensorFlow
Hyperparameters
Pretraining
- Training regime: fp16 mixed precision
- Training objective: Span masking (using mask lengths sampled from Poisson distribution λ=3.5, masking 30% of tokens)
- Optimizer : Adam optimizer (β1 = 0.9, β2 = 0.98, Ɛ = 1e-6)
- Scheduler: Custom scheduler from the original Transformers paper (20,000 warm-up steps)
- Dropout: 0.1
- Initial Learning rate: 5e-6
- Training tokens: 63B
Citation
@article{turker2024vbart,
title={VBART: The Turkish LLM},
author={Turker, Meliksah and Ari, Erdi and Han, Aydin},
journal={arXiv preprint arXiv:2403.01308},
year={2024}
}