| |
|
| | --- |
| | license: apache-2.0 |
| | language: |
| | - nso |
| | datasets: |
| | - allenai/nllb |
| | - allenai/MADLAD-400 |
| | - cis-lmu/Glot500 |
| | - sil-ai/bloom-lm |
| | - statmt/cc100 |
| | - legacy-datasets/wikipedia |
| | library_name: transformers |
| | pipeline_tag: text-generation |
| | tags: |
| | - goldfish |
| | - arxiv:2408.10441 |
| | --- |
| | |
| | # nso_latn_10mb |
| |
|
| | Goldfish is a suite of monolingual language models trained for 350 languages. |
| | This model is the <b>Pedi</b> (Latin script) model trained on 10MB of data, after accounting for an estimated byte premium of 1.12; content-matched text in Pedi takes on average 1.12x as many UTF-8 bytes to encode as English. |
| | The Goldfish models are trained primarily for comparability across languages and for low-resource languages; Goldfish performance for high-resource languages is not designed to be comparable with modern large language models (LLMs). |
| |
|
| | Note: nso_latn is an [individual language](https://iso639-3.sil.org/code_tables/639/data) code. It is not contained in any macrolanguage codes contained in Goldfish (for script latn). |
| | |
| | All training and hyperparameter details are in our paper, [Goldfish: Monolingual Language Models for 350 Languages (Chang et al., 2024)](https://www.arxiv.org/abs/2408.10441). |
| | |
| | Training code and sample usage: https://github.com/tylerachang/goldfish |
| | |
| | Sample usage also in this Google Colab: [link](https://colab.research.google.com/drive/1rHFpnQsyXJ32ONwCosWZ7frjOYjbGCXG?usp=sharing) |
| | |
| | ## Model details: |
| | |
| | To access all Goldfish model details programmatically, see https://github.com/tylerachang/goldfish/blob/main/model_details.json. |
| | All models are trained with a [CLS] (same as [BOS]) token prepended, and a [SEP] (same as [EOS]) token separating sequences. |
| | For best results, make sure that [CLS] is prepended to your input sequence (see sample usage linked above)! |
| | Details for this model specifically: |
| |
|
| | * Architecture: gpt2 |
| | * Parameters: 39087104 |
| | * Maximum sequence length: 512 tokens |
| | * Training text data (raw): 11.16MB |
| | * Training text data (byte premium scaled): 10.005MB |
| | * Training tokens: 3006464 (x10 epochs) |
| | * Vocabulary size: 50000 |
| | * Compute cost: 2274791679590400.0 FLOPs or ~0.2 NVIDIA A6000 GPU hours |
| |
|
| | Training datasets (percentages prior to deduplication): |
| | * 46.87464%: [NLLB (CommonCrawl and ParaCrawl)](https://huggingface.co/datasets/allenai/nllb) |
| | * 28.49201%: [MADLAD-400 (CommonCrawl)](https://huggingface.co/datasets/allenai/MADLAD-400) |
| | * 22.91655%: [Glot500](https://huggingface.co/datasets/cis-lmu/Glot500), including [BLOOM](https://huggingface.co/datasets/sil-ai/bloom-lm), [CC100](https://huggingface.co/datasets/statmt/cc100), [CORP.NCHLT](https://repo.sadilar.org/handle/20.500.12185/7), [Wortschatz Leipzig Data](https://wortschatz.uni-leipzig.de/en/download), [Wikipedia Hugging Face](https://huggingface.co/datasets/legacy-datasets/wikipedia) |
| | * 1.71679%: [Wikipedia 2023/08](https://dumps.wikimedia.org/) |
| |
|
| |
|
| | ## Citation |
| |
|
| | If you use this model, please cite: |
| |
|
| | ``` |
| | @article{chang-etal-2024-goldfish, |
| | title={Goldfish: Monolingual Language Models for 350 Languages}, |
| | author={Chang, Tyler A. and Arnett, Catherine and Tu, Zhuowen and Bergen, Benjamin K.}, |
| | journal={Preprint}, |
| | year={2024}, |
| | url={https://www.arxiv.org/abs/2408.10441}, |
| | } |
| | ``` |
| |
|
