goldfish-models / tam_taml_100mb

huggingface.co
Total runs: 48
24-hour runs: 0
7-day runs: 16
30-day runs: 38
Model's Last Updated: August 27 2024
text-generation

Introduction of tam_taml_100mb

Model Details of tam_taml_100mb

tam_taml_100mb

Goldfish is a suite of monolingual language models trained for 350 languages. This model is the Tamil (Tamil script) model trained on 100MB of data, after accounting for an estimated byte premium of 2.73; content-matched text in Tamil takes on average 2.73x 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: tam_taml is an individual language code. It is not contained in any macrolanguage codes contained in Goldfish (for script taml).

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): 272.91MB
  • Training text data (byte premium scaled): 100.005MB
  • Training tokens: 20064256 (x10 epochs)
  • Vocabulary size: 50000
  • Compute cost: 1.02445289570304e+17 FLOPs or ~9.7 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 tam_taml_100mb on huggingface.co

48
Total runs
0
24-hour runs
3
3-day runs
16
7-day runs
38
30-day runs

More Information About tam_taml_100mb huggingface.co Model

More tam_taml_100mb license Visit here:

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

tam_taml_100mb huggingface.co

tam_taml_100mb huggingface.co is an AI model on huggingface.co that provides tam_taml_100mb's model effect (), which can be used instantly with this goldfish-models tam_taml_100mb model. huggingface.co supports a free trial of the tam_taml_100mb model, and also provides paid use of the tam_taml_100mb. Support call tam_taml_100mb model through api, including Node.js, Python, http.

goldfish-models tam_taml_100mb online free

tam_taml_100mb huggingface.co is an online trial and call api platform, which integrates tam_taml_100mb's modeling effects, including api services, and provides a free online trial of tam_taml_100mb, you can try tam_taml_100mb online for free by clicking the link below.

goldfish-models tam_taml_100mb online free url in huggingface.co:

https://huggingface.co/goldfish-models/tam_taml_100mb

tam_taml_100mb install

tam_taml_100mb is an open source model from GitHub that offers a free installation service, and any user can find tam_taml_100mb on GitHub to install. At the same time, huggingface.co provides the effect of tam_taml_100mb install, users can directly use tam_taml_100mb installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

tam_taml_100mb install url in huggingface.co:

https://huggingface.co/goldfish-models/tam_taml_100mb

Url of tam_taml_100mb

Provider of tam_taml_100mb huggingface.co

goldfish-models
ORGANIZATIONS

Other API from goldfish-models