---
license: apache-2.0
language:
- asm
datasets:
- cis-lmu/Glot500
- statmt/cc100
- legacy-datasets/wikipedia
- allenai/MADLAD-400
- allenai/nllb
- oscar-corpus/OSCAR-2109
library_name: transformers
pipeline_tag: text-generation
tags:
- goldfish
- arxiv:2408.10441
---

# asm_beng_full

Goldfish is a suite of monolingual language models trained for 350 languages.
This model is the <b>Assamese</b> (Bengali script) model trained on 348MB of data (all our data in the language), after accounting for an estimated byte premium of 2.53; content-matched text in Assamese takes on average 2.53x 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: asm_beng 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 beng).

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: 124770816
* Maximum sequence length: 512 tokens
* Training text data (raw): 881.38MB
* Training text data (byte premium scaled): 348.885MB
* Training tokens: 77216256 (x10 epochs)
* Vocabulary size: 50000
* Compute cost: 3.94054774751232e+17 FLOPs or ~37.3 NVIDIA A6000 GPU hours

Training datasets (percentages prior to deduplication):
* 44.77996%: [Glot500](https://huggingface.co/datasets/cis-lmu/Glot500), including [AI4Bharat](https://ai4bharat.org/), [Anuvaad](https://github.com/project-anuvaad/anuvaad-parallel-corpus), [CC100](https://huggingface.co/datasets/statmt/cc100), [CCNet](https://github.com/facebookresearch/cc_net), [Earthlings](https://publicdata.canterbury.ac.nz/Research/Geocorpus/CCGLU_v5.0/), [Indiccorp](https://ai4bharat.iitm.ac.in/corpora), [Wortschatz Leipzig Data](https://wortschatz.uni-leipzig.de/en/download), [OSCAR](https://oscar-project.org/), [Wikipedia Hugging Face](https://huggingface.co/datasets/legacy-datasets/wikipedia), [WikiMatrix](https://github.com/facebookresearch/LASER/tree/main/tasks/WikiMatrix)
* 25.83555%: [MADLAD-400 (CommonCrawl)](https://huggingface.co/datasets/allenai/MADLAD-400)
* 18.99529%: [NLLB (CommonCrawl and ParaCrawl)](https://huggingface.co/datasets/allenai/nllb)
* 5.32777%: [OSCAR 2021/09](https://huggingface.co/datasets/oscar-corpus/OSCAR-2109)
* 4.47875%: [Wikipedia 2023/08](https://dumps.wikimedia.org/)
* 0.58268%: [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},
}
```