goldfish-models / wuu_hani_full

huggingface.co
Total runs: 36
24-hour runs: 0
7-day runs: 2
30-day runs: 23
Model's Last Updated: August 27 2024
text-generation

Introduction of wuu_hani_full

Model Details of wuu_hani_full

wuu_hani_full

Goldfish is a suite of monolingual language models trained for 350 languages. This model is the Wu Chinese (Han script) model trained on 24MB of data (all our data in the language), after accounting for an estimated byte premium of 0.70; content-matched text in Wu Chinese takes on average 0.70x 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: wuu_hani is an individual language code. It is not contained in any macrolanguage codes contained in Goldfish (for script hani).

All training and hyperparameter details are in our paper, Goldfish: Monolingual Language Models for 350 Languages (Chang et al., 2024) .

Training code and sample usage: https://github.com/tylerachang/goldfish

Sample usage also in this Google Colab: link

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. Details for this model specifically:

  • Architecture: gpt2
  • Parameters: 124770816
  • Maximum sequence length: 512 tokens
  • Training text data (raw): 17.17MB
  • Training text data (byte premium scaled): 24.535MB
  • Training tokens: 4112384 (x10 epochs)
  • Vocabulary size: 50000
  • Compute cost: 2.098697601024e+16 FLOPs or ~2.0 NVIDIA A6000 GPU hours

Training datasets (percentages prior to deduplication):

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},
}

Runs of goldfish-models wuu_hani_full on huggingface.co

36
Total runs
0
24-hour runs
0
3-day runs
2
7-day runs
23
30-day runs

More Information About wuu_hani_full huggingface.co Model

More wuu_hani_full license Visit here:

https://choosealicense.com/licenses/apache-2.0

wuu_hani_full huggingface.co

wuu_hani_full huggingface.co is an AI model on huggingface.co that provides wuu_hani_full's model effect (), which can be used instantly with this goldfish-models wuu_hani_full model. huggingface.co supports a free trial of the wuu_hani_full model, and also provides paid use of the wuu_hani_full. Support call wuu_hani_full model through api, including Node.js, Python, http.

goldfish-models wuu_hani_full online free

wuu_hani_full huggingface.co is an online trial and call api platform, which integrates wuu_hani_full's modeling effects, including api services, and provides a free online trial of wuu_hani_full, you can try wuu_hani_full online for free by clicking the link below.

goldfish-models wuu_hani_full online free url in huggingface.co:

https://huggingface.co/goldfish-models/wuu_hani_full

wuu_hani_full install

wuu_hani_full is an open source model from GitHub that offers a free installation service, and any user can find wuu_hani_full on GitHub to install. At the same time, huggingface.co provides the effect of wuu_hani_full install, users can directly use wuu_hani_full installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

wuu_hani_full install url in huggingface.co:

https://huggingface.co/goldfish-models/wuu_hani_full

Url of wuu_hani_full

Provider of wuu_hani_full huggingface.co

goldfish-models
ORGANIZATIONS

Other API from goldfish-models