|
|
| --- |
| license: apache-2.0 |
| language: |
| - amh |
| datasets: |
| - cis-lmu/Glot500 |
| - castorini/afriberta-corpus |
| - allenai/MADLAD-400 |
| - allenai/nllb |
| - oscar-corpus/OSCAR-2109 |
| library_name: transformers |
| pipeline_tag: text-generation |
| tags: |
| - goldfish |
| - arxiv:2408.10441 |
| --- |
| |
| # amh_ethi_10mb |
|
|
| Goldfish is a suite of monolingual language models trained for 350 languages. |
| This model is the <b>Amharic</b> (Ge'ez script) model trained on 10MB of data, after accounting for an estimated byte premium of 1.72; content-matched text in Amharic takes on average 1.72x 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: amh_ethi 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 ethi). |
| |
| 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): 17.21MB |
| * Training text data (byte premium scaled): 10.005MB |
| * Training tokens: 2212864 (x10 epochs) |
| * Vocabulary size: 50000 |
| * Compute cost: 1672080355491840.0 FLOPs or ~0.2 NVIDIA A6000 GPU hours |
|
|
| Training datasets (percentages prior to deduplication): |
| * 35.99036%: [Glot500](https://huggingface.co/datasets/cis-lmu/Glot500), including [AfriBERTa](https://huggingface.co/datasets/castorini/afriberta-corpus), [AfroMAFT](https://zenodo.org/record/6990611#.Y0-yU-xBw-Q), [CCNet](https://github.com/facebookresearch/cc_net), [Earthlings](https://publicdata.canterbury.ac.nz/Research/Geocorpus/CCGLU_v5.0/), [HornMT](https://github.com/asmelashteka/HornMT), [OSCAR](https://oscar-project.org/), [Parallel Corpora for Ethiopian Languages](https://github.com/AAUThematic4LT/Parallel-Corpora-for-Ethiopian-Languages), [TICO](https://tico-19.github.io/), [W2C](https://lindat.mff.cuni.cz/repository/xmlui/handle/11858/00-097C-0000-0022-6133-9) |
| * 33.89884%: [MADLAD-400 (CommonCrawl)](https://huggingface.co/datasets/allenai/MADLAD-400) |
| * 23.09942%: [NLLB (CommonCrawl and ParaCrawl)](https://huggingface.co/datasets/allenai/nllb) |
| * 6.48807%: [OSCAR 2021/09](https://huggingface.co/datasets/oscar-corpus/OSCAR-2109) |
| * 0.48971%: [Wikipedia 2023/08](https://dumps.wikimedia.org/) |
| * 0.03361%: [eBible](https://ebible.org/find/) |
|
|
|
|
| ## 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}, |
| } |
| ``` |
|
|
