--- license: apache-2.0 language: - kmr - kur datasets: - allenai/nllb - cis-lmu/Glot500 - sil-ai/bloom-lm library_name: transformers pipeline_tag: text-generation tags: - goldfish - arxiv:2408.10441 --- # kmr_latn_5mb Goldfish is a suite of monolingual language models trained for 350 languages. This model is the <b>Northern Kurdish</b> (Latin script) model trained on 5MB of data, after accounting for an estimated byte premium of 1.03; content-matched text in Northern Kurdish takes on average 1.03x 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: kmr_latn is an [individual language](https://iso639-3.sil.org/code_tables/639/data) code. Macrolanguage code kur_latn (Kurdish) 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): 5.17MB * Training text data (byte premium scaled): 5.005MB * Training tokens: 1204736 (x10 epochs) * Vocabulary size: 50000 * Compute cost: 910420287160320.0 FLOPs or ~0.1 NVIDIA A6000 GPU hours Training datasets (percentages prior to deduplication): * 98.85869%: [NLLB (CommonCrawl and ParaCrawl)](https://huggingface.co/datasets/allenai/nllb) * 1.14131%: [Glot500](https://huggingface.co/datasets/cis-lmu/Glot500), including [BLOOM](https://huggingface.co/datasets/sil-ai/bloom-lm), [TICO](https://tico-19.github.io/) ## 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}, } ```