goldfish-models / eng_latn_100mb

huggingface.co
Total runs: 534
24-hour runs: 0
7-day runs: 129
30-day runs: 428
Model's Last Updated: August 27 2024
text-generation

Introduction of eng_latn_100mb

Model Details of eng_latn_100mb

eng_latn_100mb

Goldfish is a suite of monolingual language models trained for 350 languages. This model is the English (Latin script) model trained on 100MB of data, after accounting for an estimated byte premium of 1.00; content-matched text in English 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: eng_latn is an individual language code. It is not contained in any macrolanguage codes contained in Goldfish (for script latn).

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): 100.00MB
  • Training text data (byte premium scaled): 100.005MB
  • Training tokens: 21390848 (x10 epochs)
  • Vocabulary size: 50000
  • Compute cost: 1.092070047744e+17 FLOPs or ~10.3 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 eng_latn_100mb on huggingface.co

534
Total runs
0
24-hour runs
69
3-day runs
129
7-day runs
428
30-day runs

More Information About eng_latn_100mb huggingface.co Model

More eng_latn_100mb license Visit here:

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

eng_latn_100mb huggingface.co

eng_latn_100mb huggingface.co is an AI model on huggingface.co that provides eng_latn_100mb's model effect (), which can be used instantly with this goldfish-models eng_latn_100mb model. huggingface.co supports a free trial of the eng_latn_100mb model, and also provides paid use of the eng_latn_100mb. Support call eng_latn_100mb model through api, including Node.js, Python, http.

goldfish-models eng_latn_100mb online free

eng_latn_100mb huggingface.co is an online trial and call api platform, which integrates eng_latn_100mb's modeling effects, including api services, and provides a free online trial of eng_latn_100mb, you can try eng_latn_100mb online for free by clicking the link below.

goldfish-models eng_latn_100mb online free url in huggingface.co:

https://huggingface.co/goldfish-models/eng_latn_100mb

eng_latn_100mb install

eng_latn_100mb is an open source model from GitHub that offers a free installation service, and any user can find eng_latn_100mb on GitHub to install. At the same time, huggingface.co provides the effect of eng_latn_100mb install, users can directly use eng_latn_100mb installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

eng_latn_100mb install url in huggingface.co:

https://huggingface.co/goldfish-models/eng_latn_100mb

Url of eng_latn_100mb

Provider of eng_latn_100mb huggingface.co

goldfish-models
ORGANIZATIONS

Other API from goldfish-models