--- license: apache-2.0 language: - ltg - lav datasets: - allenai/MADLAD-400 - allenai/nllb library_name: transformers pipeline_tag: text-generation tags: - goldfish - arxiv:2408.10441 --- # ltg_latn_5mb Goldfish is a suite of monolingual language models trained for 350 languages. This model is the <b>Latgalian</b> (Latin script) model trained on 5MB of data, after accounting for an estimated byte premium of 1.00; content-matched text in Latgalian takes on average 1.00x 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: ltg_latn is an [individual language](https://iso639-3.sil.org/code_tables/639/data) code. Macrolanguage code lav_latn (Latvian) is included in Goldfish. Consider using that model depending on your use case. 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): 5.01MB * Training text data (byte premium scaled): 5.005MB * Training tokens: 1123840 (x10 epochs) * Vocabulary size: 50000 * Compute cost: 849908968980480.0 FLOPs or ~0.1 NVIDIA A6000 GPU hours Training datasets (percentages prior to deduplication): * 65.03249%: [MADLAD-400 (CommonCrawl)](https://huggingface.co/datasets/allenai/MADLAD-400) * 31.87600%: [NLLB (CommonCrawl and ParaCrawl)](https://huggingface.co/datasets/allenai/nllb) * 3.09151%: [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}, } ```