goldfish-models / gsw_latn_full

huggingface.co
Total runs: 57
24-hour runs: 0
7-day runs: 2
30-day runs: 50
Model's Last Updated: August 27 2024
text-generation

Introduction of gsw_latn_full

Model Details of gsw_latn_full

gsw_latn_full

Goldfish is a suite of monolingual language models trained for 350 languages. This model is the Swiss German (Latin script) model trained on 128MB of data (all our data in the language), after accounting for an estimated byte premium of 1.14; content-matched text in Swiss German takes on average 1.14x 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: gsw_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): 147.17MB
  • Training text data (byte premium scaled): 128.815MB
  • Training tokens: 38605824 (x10 epochs)
  • Vocabulary size: 50000
  • Compute cost: 1.97044502003712e+17 FLOPs or ~18.6 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 gsw_latn_full on huggingface.co

57
Total runs
0
24-hour runs
1
3-day runs
2
7-day runs
50
30-day runs

More Information About gsw_latn_full huggingface.co Model

More gsw_latn_full license Visit here:

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

gsw_latn_full huggingface.co

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

goldfish-models gsw_latn_full online free

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

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

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

gsw_latn_full install

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

gsw_latn_full install url in huggingface.co:

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

Url of gsw_latn_full

Provider of gsw_latn_full huggingface.co

goldfish-models
ORGANIZATIONS

Other API from goldfish-models